DOI: https://doi.org/10.52091/EVIK-2022/1-1-ENG

Received: August 2021 – Accepted: February 2022

¹ Mertcontrol Kft.

Keywords

mycotoxin, isotope dilution mass spectrometry (LC-IDMS), standardization, citrinin, validation, HorRat value, Horwitz-Thompson equation

2. Introduction

Legislation on natural (such as plant toxins) or artificial (such as residual substances) contaminants in foods is strictest in the European Union (EU) worldwide, regulating maximum allowable levels and limit values for contaminants in foods and feeds of plant and animal origin. Commission Regulation (EC) 1881/2006 [1] contains the so-called mycotoxin limit values in foods from byproducts of the secondary metabolism of molds in agricultural crops. The regulation is expanded constantly: while initially it only contained „classical” toxins such as deoxynivalenol (DON), aflatoxins (B1, G1, B2, G2, M1), fumonisins (B1 and B2) or patulin, by 2013 T-2, HT-2, and by 2016 citrinin were also included in the toxin regulation. The range of components is expanded constantly; the process is preceded by a scientific opinion formulated by the European Food Safety Authority (EFSA), as well as other impact studies. They take into account both the economic points of view of producers and the short- and long-term health risks of the toxins. Alternaria toxins are not yet regulated, the permissible limit values are expected in the 1-10 µg/kg range for ALT, AOH and AME, and in the 10-1,000 µg/kg range for TEA and TEN. The foodstuffs concerned are cereals (primarily wheat), tomato-based foods (tomato juice or puree) and products made from sunflower seeds and similar raw materials [2].

The EFSA report on Alternaria toxins titled „Scientific Opinion on the risks for animal and public health related to the presence of Alternaria toxins in feed and food” was published in 2011 [2], and it discusses their presence in various foods, human and animal health studies and their potential risks over 97 pages. A further goal of the report is to draw attention to future regulations and to the development of a uniform analytical method. Accordingly, the analysis of Alternaria toxins in wheat, tomato and sunflower seeds by liquid chromatography tandem mass spectrometry (LC-MS/MS) was published as a standardization procedure in the mycotoxin standardization tender announced by CEN in spring 2013.

3. Initial (intra-laboratory) analytical method

According to the basic requirement of the tender, the aspirant laboratory must have a valid accredited status according to standard ISO 17043, which applies to the organization of proficiency tests and means a well-defined test protocol that meets the analytical performance characteristics for single laboratory validation [3]. Lacking this, the laboratory must have a procedure previously certified by inter-laboratory validation. Due to its cost implications, the latter is a rarer case, but it is much more efficient in demonstrating the true reproducibility of the method than the requirements of standard ISO 17043, whereas the former validation only shows the in-laboratory reproducibility (intermediate precision) of the analysis.

The European Commission Joint Research Centre (JRC, Geel, Belgium) is a joint research center within the EU, which until 2017 included the EU Reference Laboratory for Mycotoxins (EU-RL for Mycotoxins). In 2013, an LC-MS/MS method was developed as an EU-RL method for wheat, tomato juice and sunflower seeds for the following five main Alternaria toxins (Figure 1): tenuazonic acid (TEA), altenuane (ALT), alternariol (AOH), tentoxin (TEN) and alternariol monomethyl ether (AME) [4]. Of the five toxins, TEA has the most different structure and physicochemical properties (chelating properties) from the other toxins [5]. Accordingly, previous literature has focused on the determination of TEA [5], or the other toxins [6], less attention has been paid to their simultaneous analysis. Our goal was a five-component simultaneous analysis, which was achieved by chemical derivatization. The structure of TEA contains an aldehyde functional group that is highly reactive with 2,4-dinitrophenylhydrazine (DNPH), and the physicochemical properties (e.g., Log P, octanol-water distribution) of the resulting TEA hydrazone are much closer to those of the other Alternaria toxins from a chromatography point of view [5]. In the derivatized form, it loses its chelating properties. DNPH reacts only with TEA among the target components (Figure 2), it does not interfere with the determination of the others [7]. The extraction procedure arrived at in the method was developed using an experimental design with a sample of rye naturally contaminated with the toxins. In addition to Alternaria toxins, citrinin was also included in the method. The main characteristics of the method developed in this way are the following [4], [7]:

• Analysis of six components (TEA, ALT, AOH, TEN, AME and citrinin);
• Matrices: cereals, tomato juice, peeled sunflower seeds;
• Sample weight for liquid samples: 1.0 g;
• Extraction solvent for liquid samples: 5 mL of methanol;
• Sample weight for solid samples: 2.0 g;
• Extraction solvent for solid samples:15 mL of methanol-water (70/30, v/v) mixture;
• Derivatizing agent: 0.58% DNPH in aqueous hydrochloric acid;
• Stop reagent: 5% (v/v) undecanal in methanol;
• Sample purification: polymer-based solid phase extraction (SPE);
• Sample evaporation and redissolution in methanol;

• Syringe filtration on PTFE filter;
• LC-MS/MS separation: acidic eluent, C-18 stationary phase and ESI negative ionization (Table 1);
• Syringe filtration on hydrophilic PTFE filter;
• Calibration: matrix-matched calibration without isotope-labelled internal standard.

In addition to the in-laboratory validation of the method, we also participated in an international proficiency test organized by the Bundesinstitut für Risikobewertung (BfR, Berlin, Germany) as a National Reference Laboratory (NRL) for the determination of the five Alternaria toxins in tomato juice. During the analysis, the five toxins had to be determined in three samples and a standard solution [7]. Our results are shown in Table 2. All reported values were acceptable, with Z-score values between -2 and +2. The results showed that the method recommended by us in the tender is suitable for the standardization of Alternaria toxins.

4. Modified analytical method

The analytical method proposed by the JRC was adopted by CEN and the mandate (mandate M/520) was given to the JRC in 2014. However, the working group (TC 275 WG 5 „Horizontal Methods for Food – Biotoxins”) did not support chemical derivatization in the method on the grounds that it is an additional and time-consuming step in the method, which may reduce its precision and should be avoided. The determination of citrinin could not be included in the method either, the analysis could only contain Alternaria toxins.

TEA can also be analyzed in its native form, but in this case HPLC separation has to be carried out with an alkaline eluent, requiring a stationary phase that is stable up to pH 9. The method has indeed become simpler without derivatization (Figure 3), but this has required significant modifications to maintain the accuracy of the procedure. In addition to being time-consuming, another disadvantage of derivatization was an increase in the noise level, as many matrix-forming compounds also react with DNPH, which can co-elute with the target components, increasing the noise in the MS/MS instrument. In the modified method, essentially the HPLC separation had to be optimized and an extraction medium had to be selected which ensured the best possible recovery from each matrix.

The main characteristics of the method developed in this way [8]:

• Analysis of five components (TEA, ALT, AOH, TEN and AME);
• Matrices: cereals, tomato juice, sunflower seeds;
• Sample weight for liquid samples: 2.0 g;
• Extraction solvent: 15 mL methanol/water/acetic acid (80/19/1, v/v/v);
• Sample purification: polymer-based solid phase extraction (SPE);
• Sample evaporation and redissolution in 100 µL of dimethyl sulfoxide and dilution with 900 µL of water;
• Syringe filtration on hydrophilic PTFE filter;
• LC-MS/MS separation: eluent with alkaline pH (pH 8.7), C-18 stationary phase and ESI negative ionization (Table 3);
• Calibration: matrix-matched calibration without isotope-labelled internal standard.

This modified method was accepted by the working group and, following its in-laboratory validation, the inter-laboratory validation of the analytical method could also begin in spring 2015.

5. Inter-laboratory validation of the method

The most important part of the standardization process is the inter-laboratory validation of the method, the main purpose of which is to check and evaluate the reproducibility of the analysis. To do this, the concentrations of the toxins must be determined in naturally contaminated (at low, medium and high levels) and spiked samples. To evaluate the concentration of a given component n a given sample, a minimum of eight independent values are required, however, only the results of two laboratories can be excluded [9]. It is advisable to involve at least fifteen laboratories in order to have an adequate number of results for each sample and component. This is so because, based on experience, about 2-3 laboratories do not report results, while some samples and their components do not always produce a sufficient number of reported results. This can occur mainly at low concentration levels because not all participants possess instruments with adequate sensitivity.

If, during the validation, the goal is to determine components that have long been analyzed (such as DON or aflatoxins), it is relatively easy to ask laboratories with the necessary experience for validation, based on their successful participation in previous proficiency tests. However, Alternaria toxins are still analyzed by very few laboratories to this day, so laboratories applying for the validation do not always have prior experience. For this reason, organization of a so-called pre-trial becomes necessary, in which the laboratories participating in the validation can master the method in advance. In this case, the pre-trial was performed with twenty-five laboratories, analyzing tomato juice samples [8], and only three of the participating laboratories had prior knowledge of Alternaria LC-MS/MS analysis. Of the twenty-five laboratories, only sixteen eventually participated in the final validation, because either they did not return any result or their results differed significantly from the consensus average.

During the final validation, the following samples were sent to the sixteen laboratories [8]:

• Cereals naturally contaminated with Alternaria toxins: wheat, triticale and sorghum;
• Tomato juice naturally contaminated with Alternaria toxins: 3 batches;
• Sunflower seeds naturally contaminated with Alternaria toxins: 2 batches of unpeeled seeds and 1 seed mixture, which was a mixture of peeled and unpeeled seeds;
• Participants received each sample under two codes (blind replicates) so that we could evaluate repeatability within the laboratory and to have more date available to analyze reproducibility;
• For the preparation of spiked samples, separate test samples were sent for each matrix, for which a standard solution mixture containing Alternaria toxins in unknown concentrations was also provided to the participating laboratories. Spiked samples were prepared by the laboratories according to the „spiking guide”;
• Blank samples for each matrix for matrix-matched calibration;
• In the case of sunflower seeds, the blank was peeled sunflower, because the unpeeled samples are high in TEA and therefore not suitable for calibration;
• The analytical standards of the target components and their standard solution mixture were also provided, so that all laboratories would use the same calibration solution, and no deviation would result from this;
• The homogeneity of the samples was checked according to the harmonized protocol before dispatch [10];
• Simultaneously with the sending of the samples, stability testing of the samples was initiated at different temperatures and for different durations.

Concentration levels required by CEN for validation: 1-10 µg/kg for ALT, AOH and AME, and 10-1,000 µg/kg for TEA and TEN. Recovery was assessed from the concentrations measured in the spiked samples, with spiking levels of 2 and 8 µg/kg for ALT, AOH and AME, and 50 and 200 µg/kg for TEA and TEN. These levels were unknown to participants.

Statistical evaluation of the results obtained (concentrations not corrected for recovery) focused mainly on reproducibility [9]. The reproducibility of the method is well characterized by the so-called HorRat value. The HorRat value is the quotient of the reproducibility of a given target component calculated for the given sample and the target reproducibility expected by the organizers. The latter reproducibility value (the „target reproducibility”) can be calculated from the Horwitz-Thompson equation: below 120 µg/kg it is uniformly 22%, while above this value the classical Horwitz relationship can be applied [11]. Based on the validation criteria, the HorRat value must be less than two; this condition was indeed met, except for TEA, in the case of the unpeeled sunflower samples. Table 4 shows the HorRat values calculated for TEA in the case of different sunflower samples. While in unpeeled sunflowers the calculated HorRat values were uniformly 2.4 regardless of the concentration [8], in the case of peeled samples, which contained much lower concentrations of TEA, the values were below two. The lower reproducibility observed during the analysis of unpeeled samples can be explained by the calibration and the matrix effect, which is a typical feature of LC-MS/MS-based measurements, and mainly effects the precision and accuracy of the method [11]. During the validation, a peeled sunflower sample was provided for calibration, because it contained a small amount of TEA contamination of natural origin, as opposed to unpeeled sunflower that was contaminated with high concentrations of TEA. The extracts of the unpeeled and peeled sunflower samples contain significantly different matrices, which can even be noticed by their color. Consequently, the calibration from the peeled sample could not compensate for the matrix effect in the unpeeled sunflower samples, so the detected concentrations were significantly affected by the matrix effect. The reason for this is that the endogenous constituents of unpeeled sunflower differ from those of peeled sunflower.

It is important to note that laboratories reported only the detected concentrations; the measured values were not corrected for recovery, in contrast to the usual procedure for conventional proficiency tests. Different laboratories used different instruments in which the matrix effect during the analysis of unpeeled sunflowers may have been different. Since the calibration recorded from the peeled sample did not adequately compensate for the matrix effect, there were large differences between the values measured by the participants. The same problem did not occur in the analysis of peeled sunflowers, because a similar degree of matrix effect may have occurred in the calibration and the test sample, due to the similarity of the samples. It is worth noting that the repeatability was also acceptable in the case of unpeeled samples (<20%). The reason for this is that repeated analysis of the same sample has the same matrix effect in the same instrument, so laboratories detected similar concentrations within the laboratory for duplicate samples, while inter-laboratory results were different due to the different matrix effects in the different instruments.

6. Final method with isotope dilution and its inter-laboratory validation

As the HorRat values were not below two for all components and samples during the validation, further development of the method became necessary. The reproducibility of LC-MS/MS methods can be greatly enhanced by isotope dilution (Isotope Dilution Mass Spectroscopy – IDMS), which compensates well for the matrix effect varying from sample to sample. In this case, a stable isotope-labeled analogue of the target compound is added to the sample as an internal standard (ISTD). the physicochemical properties of the internal standard are the same as those of the target component (a small difference in polarity may occur with deuterated standards), so the target compound and its isotopically labeled analogue ideally elute at the same retention time. As a result of the co-elution, the target component and its internal standard are subjected to the same direction and extent of matrix effect in the ion source, so the ratio of the responses (areas) of the target compound and the ISTD, the isotope ratio (IR), will be independent of the matrix effect.

The ISTD does not interfere with the signal of the target component, because it is detected at other m/z values that are sufficiently distant (preferably at least +3 Da) from the m/z value of the target component due to the isotope label.

This requires isotope-labeled ISTDs, which were not yet available in 2015, so we first used matrix-matched calibration. However, stable isotope-labeled ISTDs (labeled with ¹³C or deuterium) of Alternaria became commercially available in 2018 (TEA-¹³C2, ALT-d6, AOH-d3, TEN-d3 and AME-d3), making revalidation of the method possible using the IDMS technique.

After 2018, the JRC repeated the in-laboratory and inter-laboratory validation using the method supplemented with isotope-labeled ISTDs (Table 5). The concept was the same during the first and second validation, with the difference being that cereal-based samples only included wheat samples and tomato-based samples were tomato purees during the second procedure. In the case of TEA, HorRat values ranged from 0.40 to 0.66 with IDMS detection in unpeeled samples, while the value was 0.53 in peeled samples, which is significantly better than the values without ISTD (Table 4). As previously expected, isotope dilution greatly improved inter-laboratory reproducibility. During the validation, the expected precision could only be achieved with ISTDs, which is common in LC-MS quantitative studies. This is always due to matrix effect compensation.

7. Documentation

The full validation dossier was completed by 2020 [12], together with the draft standard. Review and revision of the draft standard will be completed soon and the proposed standard is expected to be adopted by CEN in the end of 2021 (the standard has been issued since the article was submitted: CEN EN 17521:2021 Foodstuffs - Determination of Alternaria toxins in tomato, wheat and sunflower seeds by SPE clean-up and HPLC-MS/MS. The Editor).

8. Deviation from the standard method

LC-MS/MS instruments from different vendors may vary significantly in terms of sensitivity. One of the main reasons for this is the ion source [11]. While the standard describes the use of ESI (Electrospray Ion Source), there is hardly any application in the literature where the ESI ion source of the instrument showed sufficient efficiency to achieve the desired detection limit, so the use of atmospheric pressure chemical ionization (APCI) became necessary [13]. Another possibility is when the instrument used is so sensitive that no solid phase purification or enrichment (Solid Phase Extraction – SPE) is required, but the extract of the sample can be injected directly into the device („dilute-and-shoot”) [11], [14]. The important feature of a standard is that all laboratories should be able to use the method described in it, so the application of SPE enrichment was unavoidable due to the low concentration levels and the complexity of the unpeeled sunflower seed samples.

If the first validation is successful, matrix-matched calibration would probably be recommended by the standard. However, with the advent of ISTDs, a group of laboratories would prefer to use IDMS later on. From this point of view, it is fortunate that IDMS has been introduced in the standard, which is simpler and more accurate, but the acquisition of ISTDs is more expensive. In the absence of ISTDs, standard addition (as a quantitative evaluation) can also be used to adequately compensate for the matrix effect, but this is time-consuming, because each sample must be prepared at least four or five times. Yet there are laboratories that use this type of evaluation.

9. Acknowledgment

I would like to give thanks Carlos Gonçalves for the successful completion of the standardization project.

10. References

[1] A Bizottság 1881/2006/EK rendelete (2006. december 19.) az élelmiszerekben előforduló egyes szennyező anyagok felső határértékeinek meghatározásáról. Az Európai Unió Hivatalos Lapja, L 364/5. (Hozzáférés: 2021.04.12)

[2] EFSA, European Food Safety Authority, (2011): Scientific Opinion on the risks for public and animal health related to the presence of citrinin in food and feed, EFSA J. 10 p. 1-82. DOI

[3] CEN/TR 16059, Food analysis. Performance criteria for single laboratory validated methods of analysis for the determination of mycotoxins.

[4] Tölgyesi, Á., Stroka, J. (2014): Report on the development of a method for the determination of Alternaria toxins and citrinin in wheat, tomato juice and sunflower seeds by liquid chromatography – tandem mass spectrometry. JRC Technical report (Hozzáférés: 2021.02.24)

[5] Asam, S., Liu, Y., Konitzer, K., Rychlik, M. (2011): Development of a stable isotope dilution assay for tenuazonic acid, J. Agr. Food Chem. 59 p. 2980–2987. DOI

[6] Lau, BP-Y, Scott, P.M., Lewis, D.A., Kanhere, S.R., Cleroux, C., Roscoe, V.A. (2003): Liquid chromatography–mass spectrometry and liquid chromatography–tandem mass spectrometry of the Alternaria mycotoxins alternariol and alternariol monomethyl ether in fruit juices and beverages. J Chromatogr A. 998 p. 119–131. DOI

[7] Tölgyesi, Á., Stroka, J., Tamosiunas, V., Zwickel, T. (2015): Simultaneous analysis of Alternaria toxins and citrinin in tomato: an optimised method using liquid chromatography-tandem mass spectrometry, J. Food Addit. Contam. 32 p.1512–1522. DOI

[8] Tölgyesi, Á., Stroka, J. (2016): Collaborative study report: Determination of Alternaria toxins in cereals, tomato juice and sunflower seeds by liquid chromatography tandem mass spectrometry, JRC Technical Report (Hozzáférés: 2021.03.14)

[9] Practical guide to ISO 5725-2:1994 — Accuracy (trueness and precision) of measurement methods and results — Part 2: Basic method for the determination of repeatability and reproducibility of a standard measurement method. Geneva, Switzerland.

[10] Thompson, M., Ellison, S.L.R., and Wood, R. (2006): The International Harmonised Protocol for the Proficiency Testing of Analytical Chemistry Laboratories. Pure Appl. Chem. 78(1):145–196.

[11] Tölgyesi, Á. (2021): Gyakorlati példák a folyadékkromatográfiával kapcsolt hármas kvadrupol rendszerű tandem tömegspektrometria élelmiszer-, bio- és textilanalitikai alkalmazására, Kromatográfus különszám, Gen-lab Kft., Budapest, Magyarország (Hozzáférés: 2021.02.07)

[12] Gonçalves, C., Tölgyesi, Á., Bouten, K., Robouch, P., Emons, H., Stroka, J. (2021): Determination of Alternaria toxins in tomato, wheat and sunflower seeds by SPE and LC-MS/MS – a method validation through a collaborative trial, J. AOAC Inter. 1-15. DOI

[13] Tölgyesi, Á., Kozma, L., Sharma, V.K. (2020): Determination of Alternaria toxins in sunflower oil by liquid chromatography isotope dilution tandem mass spectrometry, Molecules 25, 1685. DOI

[14] Tölgyesi, Á., Farkas, F., Bálint, M., McDonald, T.J., Sharma, V.K (2021): A dilute and shoot strategy for determining Alternaria toxins in tomato-based samples and in different flours using LC-IDMS separation, Molecules 26, 1017. DOI

DOI: https://doi.org/10.52091/EVIK-2021/4-4-ENG

Received: August 2021 – Accepted: November 2021

¹ National Food Chain Safety Office, Food Chain Safety Laboratory Directorate, National Analytical Reference Laboratory

Keywords

Acrylamide, carcinogenic, toxic compound, asparagine, Maillard reaction, capsule coffee, orol coffee, powdered coffee, robusta, arabica, effect of powdering processes, caffeinated and decaffeinated coffee, LC-MS/MS

2. Introduction

2.1. Acrylamide, its formation and effects

Acrylamide is an organic compound with the molecular formula C₃H₅NO. Its IUPAC name is prop-2-enamide. It is a low molecular weight, odorless, white solid which is highly soluble in water but also soluble in organic solvents. It is used in industry in the production of polyacrylamides, which are used as water-soluble thickeners and flocculants. It is a highly toxic compound therefore it is mainly handled in the form of an aqueous solution [2].

Acrylamide is a human neurotoxin, classified by the International Agency for Research on Cancer (IARC) as a Group 2A substance, i.e., as an agent which is probably carcinogenic to humans [1]. Acrylamide has been used in many industrial processes since the 1950s. An announcement was issued by the Swedish National Food Administration on April 24, 2002, about the discovery that it is formed as a byproduct in heat-treated foods with a high carbohydrate content [3] and can therefore be detected mainly in snack foods, potato chips, breads, cereal products and coffee. Following the discovery, more and more studies were launched to detect acrylamide content. An increasing number of researchers are looking for the answer to the question how it is formed in different foods.

Mottram et al. have conducted extensive studies on the formation of acrylamide from amino acids and reducing sugars during heat treatment as a result of the Maillard reaction. Asparagine, the amino acid most abundant in potatoes and cereals, has been found to contribute greatly to acrylamide formation. During baking and roasting, products of the Maillard reaction are responsible for the formation of flavor and aroma substances and the development of color. Strecker degradation of the amino acids also occurs at this time, during which amino acids are decarboxylated and then deaminated to form aldehydes. An outline of the process is shown in Figure 1 [4].

According to several studies, acrylamide is toxic because it forms adducts with compounds found in hemoglobin and also reacts with important functional proteins and DNA. Glycidamide, a metabolite of acrylamide, reacts similarly with hemoglobin as well [5].

The area most studied is related to the neurotoxic properties of acrylamide, since these can be observed in both humans and animals. Observations have been made in a variety of laboratory animals, including cats, rats, mice, rabbits and monkeys. After administration of 0.5 to 50 mg acrylamide/kg/day, limb movement disorders and muscle weakness could be observed in all animals [6].

2.2. Acrylamide in coffee

The acrylamide content of coffee is formed during roasting. In an extensive study, Guenther et al. found that it is produced in the highest amount (more than 7 mg/kg) during the initial stage of roasting, and then the amount decreases towards the end of the process. Towards the end of the roasting cycle, acrylamide is increasingly eliminated, with both physical and chemical losses [7].

Kinetic models and other experiments with isotopically labeled acrylamide have shown that more than 95% of the acrylamide formed is degraded during the entire roasting process. This means that the acrylamide content of lightly roasted coffees with a shorter roasting cycle is much higher than that of dark roasted beans [7].

The authors of the study also explained that green coffee beans contain very low concentrations of asparagine (0.2–1.0 g/kg), which is only negligibly higher in the case of Robusta species. Thus, it was found that the amount of asparagine and the acrylamide concentration showed a weak correlation, and even no correlation was found in Robusta beans. This is due to the fact that the rate of acrylamide loss far exceeds the rate of its formation [7].

Alves et al. studied how the acrylamide content in brewed espresso coffee changes, as in their opinion it most often enters consumers’ body in this form. Acrylamide is highly soluble in water, so it is extracted easily from coffee during brewing. The chemical properties of brewed coffee are influenced by many factors, such as the type of coffee (Arabica, Robusta, or a certain mixture of the two), the degree of roasting, or the amount of water used to make a given amount of coffee, which varies by individual taste. According to some studies the acrylamide content of different coffee beverages ranged from 2 to 25 μg/l [8].

3. Objective

The main objective of my work was to investigate the acrylamide content of different types of capsule coffees by HPLC-MS/MS measurement.

Based on literature data, it was assumed that the acrylamide content of beverages brewed from capsule coffees is higher than the acrylamide content of the ground coffee extracted from the capsules, as it dissolves easily in the water during brewing. The goal was to examine and confirm this with the measurements.

Another objective was to compare different coffee machines. Coffee machines have different parameters (e.g., temperature, pressure, amount of water used), which may affect the amount of acrylamide released from capsule coffees.

Literature data are also available showing how the roasting technology of coffee affects the acrylamide content in the final product. The acrylamide content of so-called light-roasted coffees, roasted for a shorter period of time, is higher than that of dark-roasted coffees, roasted for a longer period of time. This influencing factor was also checked.

Given that decaffeinated coffees are produced by a different technology, some of these types were also examined.

4. Materials and methods

4.1. Chemicals, equipment and instruments used

During my work, analytical grade chemicals, HPLC grade solvent (methanol, acetic acid (anhydrous), n-hexane) and distilled water were used, as well as the following: acrylamide and 10 µg/ml acrylamide-¹³C₃ as internal standard.

In addition to standard laboratory equipment, Biotage ISOLUTE^® Multimode 1g/6ml and Biotage ISOLUTE^® ENV+500mg/6ml SPE columns were used for sample preparation. For coffee brewing from capsule coffees, the following coffee machines were used: Nespresso Essenza Mini, Krups KP120H31, Tchibo Caffissimo and Martello Smart.

Instrumental analysis of the samples were performed on a Thermo Scientific™ Dionex UltiMate™ 3000 HPLC system with a Phenomenex Kinetex^® C18 2.6 µm 100 Å 150x4.6 mm column and a Thermo Scientific™ TSQ Quantis™ triple quadrupole MS detector.

4.2. Sample preparation

Sample preparation and the measurements were carried out as described in standard MSZ EN 16618:2015 Food analysis. Determination of acrylamide in food by liquid chromatography tandem mass spectrometry (LC-ESI-MS/MS).

The samples obtained from commercial sources were caffeinated (25 pcs) and decaffeinated (8 pcs) of capsule coffees from different manufacturers. Measurements were performed on both the ground coffee in the capsules and the brewed coffees. Table 1 shows the sample nos. of the examined coffees and the coffee machines used.

IBM SPSS Statistics software was used for the statistical evaluation of the results.

5. Results

5.1. Acrylamide content

Acrylamide content measurement results of the coffee samples are summarized in Table 2. Both the results measured in the ground coffees and the results of the corresponding brewed coffees are listed.

The results obtained were not in all cases in line with the reference level of 400 μg/kg for roasted coffee set out in Commission Regulation (EU) 2017/2158, as the acrylamide content of some caffeinated samples (nos. 13 and 33) exceeded this level. It is likely that the higher acrylamide level in the case of coffee sample no. 13 was due to the fact that the sample contained Robusta coffee with a higher intensity of acrylamide formation, according to the literature. The result of sample no. 33 can be explained by the fact that is was a hazelnut-flavored mixture. Given that a Robusta variety was added to the Arabica coffee variety, this may have been the reason for the higher results, to which the roasted hazelnut flavor could also have contributed. On average, the acrylamide content of the ground coffees was higher, or in some cases almost identical to the results of the brewed coffees. There were also samples in the case of which the brewed coffees contained more acrylamide than the ground coffees, but most of these values were within the 10% measurement uncertainty.

Based on my statistical (ANOVA) calculations, there was no significant difference between the measurement results of ground and brewed coffees at the 95% confidence level (p > 0.05).

5.2. Effect of brewing on acrylamide content

My objective was to investigate the extent to which the acrylamide content in ground and brewed coffees could vary depending on which coffee machine was used for brewing. Thus, I was looking to answer whether the coffee machines worked with different efficiencies. There was no significant difference between capsule and brewed coffees for any of the coffee machines (p > 0.05).

However, the results showed that, in the case of Martello type capsules, the measured values of both brewed and ground coffees were in a higher range than the results of the other types.

For the Martello type, this range was between 200 and 450 µg/kg (Figure 2), while for the other types (for example, for Nespresso, see Figure 3), typical values were between 100 and 250 µg/kg.

It was found that capsules made for Martello type coffee machines contained ground coffees that typically had a higher acrylamide content. The Martello type capsule coffees tested contained Robusta coffee or a mixture of Robusta and Arabica, which explains the higher acrylamide content, as Robusta-types coffees have higher acrylamide levels than Arabica varieties. One of the Martello type coffee capsules was roasted hazelnut flavored, which also may have contributed to the higher result.

Based on my measurement results, it can be stated that there was no significant difference between the effects of the different coffee machines. However, as the Martello type coffee capsules, on the whole, contains ground coffee with a higher acrylamide concentration compared to the other types, it caused a significant difference between the measurement results of the ground coffees in the capsules.

5.3. Effect of roasting on acrylamide content

It was also examined how different roasting levels affect the amount of the acrylamide formed. In Table 3, coffee samples are grouped according to roasting levels. The ground and brewed samples were marked with separate hues. Figure 4 shows the measured acrylamide amounts according to the different roasting levels. Light roasted samples typically yielded similar of higher results than dark roasted coffees. According to the literature, acrylamide levels of dark roasted coffees are lower than those of light roasted coffees, and this was confirmed by our results.

However, when performing statistical analyses, it was found that there was no significant difference in the amount of acrylamide formed between the results of either ground or brewed coffees at the 95% confidence level (p > 0.05).

The analyses were also performed for the different coffee machines, as the measured values of the coffees brewed with the different machines were typically in different ranges, so this grouping results in a more accurate comparison. However, there was no significant difference between the roasting levels this way either.

5.4. Results of caffeinated and decaffeinated coffees

Significantly different techniques are used for the production of decaffeinated coffees, therefore the acrylamide content results of caffeinated and decaffeinated coffees were also compared. The measured values of decaffeinated coffees were in a similar range as the values of caffeinated samples. It was found that there was no significant difference between the different types of samples in terms of acrylamide content.

To confirm this, ANOVA analyses were performed when examining the results of both ground and brewed coffees. At the 95% confidence level, there was no significant difference between the types in either case (p > 0,05).

6. Conclusions

Based on literature data, it was hypothesized that the acrylamide content of the beverages brewed from capsule coffees was higher than the acrylamide content of the ground coffee extracted from the capsules. Based on my measurements, it was found that the acrylamide content of the ground coffees was on average higher than or in some cases similar to the acrylamide levels of the brewed coffees. However, in some cases, brewed coffees did contain more acrylamide. Nevertheless, based on statistical calculations, the difference between the results was not significant. Based on these results, the claims in the literature could not be substantiated unequivocally.

Based on my results, it can be stated that there was no significant difference between the brewed coffees and their ground coffee counterparts in the case of any of the coffee machines in terms of the measured amount of acrylamide. It was found that the acrylamide levels of Robusta type coffees are higher than those of Arabica varieties. The Martello type capsule coffees contained Robusta coffee or a mixture of both, which may explain their higher acrylamide content.

Light roasted samples typically yielded similar or higher acrylamide content results than dark roasted coffees. There was no significant difference in the results of either ground or brewed coffees between the roasting levels.

It was found that there was no significant difference between caffeinated and decaffeinated coffee samples. The acrylamide content of coffee is not significantly affected by the decaffeination processes used.

7. References

[1] Acrylamide. (Hozzáférés: 2020. 01. 27.)

[2] Akrilamid. (Hozzáférés: 2020. 01. 27.)

[3] Löfstedt R. E. (2003): Science Communication and the Swedish Acrylamide ‘‘Alarm’’. Journal of Health Communication, 8 pp. 407–432. DOI

[4] Mottram, D. S., Wedzicha, B. L., Dodson, A. T. (2002): Acrylamide is formed in the Maillard reaction. NATURE, Vol. 419. DOI

[5] Sörgel, F., Weissenbacher, R., Kinzig-Schippers, M., Hofmann, A., Illauer, M., Skott, A., Landersdorfer, C. (2002): Acrylamide: increased concentrations in homemade food and first evidence of its variable absorption from food, variable metabolism and placental and breast milk transfer in humans. S. Karger AG, Basel 0009 3157/02/0486–0267. DOI

[6] Parzefall, W. (2008): Minireview on the toxicity of dietary acrylamide. Food and Chemical Toxicology 46 pp. 1360–1364. DOI

[7] Guenther, H., Anklam, E., Wenzl, T., Stadler, R. H. (2007): Acrylamide in coffee: review of progress in analysis, formation and level reduction. Food Additives & Contaminants, 24 Sup 1, pp. 60-70. DOI

[8] Alves, R. C., Soares, C., Casal, S., Fernandes, J.O., Oliveira, M. Beatriz P.P. (2010): Acrylamide in espresso coffee: influence of species, roast degree and brew length. Food Chemistry 119 pp. 929–934.DOI

DOI: https://doi.org/10.52091/EVIK-2021/3-2-ENG

Received: March 2021 – Accepted: June 2021

¹ University of Debrecen, Faculty of Economics and Business, Institute of Marketing and Commerce

Keywords

food industry developments, food nanotechnology, consumer acceptance, willingness to buy, food industrial use of titanium dioxide

2. Introduction – nanotechnology

One of the most dynamically developing disciplines today is the research of nanoscale materials. Research and application of nanotechnology is one of the great scientific, developmental and technical challenges of the 21^st century.

Nanotechnology means the production, of materials, devices and systems that use artificially formed nanoparticles, i.e., particles of material that do not exceed 100 nanometers in size [1]. Nanostructured materials are also found in nature (e.g., clays, zeolites), but can be produced artificially as well.

Many nanotechnological applications are known in practice. Examples include highly resistant materials used in construction; lightweight, elastic clothing and sports equipment made of materials resistant to physical stress; self-cleaning paints that protect buildings from, for example, the harmful effects of smog and other contaminants; nanosensors that enable efficient and economical quality control in the food industry; highly miniaturized electronic devices; antibacterial coatings for industrial equipment and household appliances; selective release and high bioavailability drugs; innovative tools for the remediation of contaminated soils and waters. However, in addition to the benefits, nanotechnology poses risks to the environment and human health that are difficult to assess. Scientific research, while still proving to be scarce, suggests that nanoparticles are more reactive and mobile than larger particles and can therefore be toxic to humans and the environment. Little is known about the fate of nanoparticles in the environment. In the human body, nanoparticles may be able to cross the cell membrane and reach internal organs. Some studies have shown that many types of nanoparticles cause greater oxidative stress at the cellular level, increasing the risk of degenerative diseases [1].

2.1. Nanotechnology in the food industry

Due to their special properties, the use of nanostructured materials can also be promising in many food applications [2].

Foods containing nanoparticles should be considered as novel foods under Regulation (EC) No 258/97, as foods or food ingredients produced by such technology were not consumed in significant quantities in the European Union before May 15, 1997; thus, their placing on the market is preceded by an authorization procedure accompanied by a rigorous safety assessment [2]. As part of the authorization process, EU regulation has recently required food ingredients derived from the use of nanotechnologies to undergo a safety assessment before they can be placed on the market, and only then can they be authorized [3]. Related to this, the term nanofood has emerged to refer to foods that are produced, processed or packaged using a nanotechnology technique or device, or to which a nanomaterial is added and/or is enriched with a nanomaterial [4].

Nanotechnologies aimed at improving food quality or safety can theoretically be diverse, but their practical application is still in its infancy. Since food nanotechnology is also a new field for food science, nanotechnology is also a major challenge for the food economy, including food security and safety, traceability, certain areas of food processing and packaging, some new opportunities for nutrient intake, longer food shelf life and many other aspects of consumer protection, from agricultural production to the consumers’ tables [2].

The use of nanoparticles in food processing can contribute to the improvement of nutritional quality, taste, color and stability or to increasing shelf life and, in the case of liquid foods, to the improvement of flow properties. An additional benefit of nanotechnology may be that it can contribute to the development of foods with lower fat, sugar and salt content, thereby reducing the incidence of food-related diseases [5].

Currently, these products are available in four categories:

• nanostructured food ingredients and substances, such as nano-titanium dioxide, which is used as an anti-caking agent or pigment;
• nanostructured delivery systems that improve the bioavailability of bioactive compounds in fortified foods and supplements;
• novel packaging materials designed to strengthen the protective function of the product;
• and the use of food contact materials for food processing and storage, such as nano-silver, which is used for its antimicrobial properties [6, 7, 8, 9].

Nanotechnology is currently considered to be the most widespread among food industrial commercial applications in the packaging process [2, 10]. Several types of use of nanomaterials in packaging materials can be distinguished. In the case of nanocomposites, advantageous properties (mechanical or functional, e.g., gastightness, temperature / humidity stability) are achieved by adding nanoparticles to the plastic.

A similar effect can be achieved with nanocoatings applied to the surface of the packaging material. Aluminum coatings applied with the help of vacuum are now widespread mainly in the packaging of snacks, confectionery and coffee. For example, if the thickness of the aluminum layer applied as a coating does not exceed 50 nm, the coating metal can be considered a nanomaterial [11]. In addition to the above, there are several applications that are still in the research phase [12, 13, 14, 15], such as newly developed food packaging capable of detecting the presence of pathogens and contaminants.

Although this technology offers consumers unprecedented benefits such as higher added value, longer shelf life and increased food safety, nanofoods also pose health, environmental, economic, social and political risks [16, 17]. According to Berekaa, despite the huge benefits that nanoparticles can bring to the food industry, the public is very concerned about their toxicity and potential negative environmental impact. Due to the health consequences of the nanoparticles entering the human body, their potential risks to human health need to be assessed without delay [5]. In his paper, Halliday points out that EU regulations on food and food packaging require a specific risk assessment before nanomaterials are placed on the market [18].

In the course of our research, it was examined to which extent the concept of nanotechnology in the food industry has spread in the public consciousness, i.e., presumably how many people are aware of this technology and its potential application in the food industry. Following this, it was assessed how receptive consumers were about the technology, how they saw its future, and whether they would be willing to buy nanofoods. In our work, the potential dangers of nanotechnology were analyzed, and also the areas in which they may occur, as well as how attitudes, consumer acceptance and willingness to buy change in the light of this.

2.1.1. Foods and packaging materials produced using nanotechnology – some examples [1]

2.1.1.1. Creamier ice cream with unchanged fat content

When making ice cream that is creamier than traditional ones, titanium dioxide consisting of nano-sized grains is added to the raw material of ice cream to increase its creaminess and improve its taste, while keeping its fat content the same as that of traditional ice creams. In its nano form, titanium dioxide is thought to be cytotoxic, however, no data have been found in the scientific literature on the mechanism of absorption of nano TiO2 from the intestinal tract.

2.1.1.2. Table salt and sugar that do not form lumps with moisture

Nano-sized particles of titanium dioxide are added to table salt and sugar as anti-caking agent. For toxicological aspects see Section 2.1.1.1.

2.1.1.3. Fruit juices enriched with bioactive molecules

Bioactive molecules such as phytosterols, vitamins and antioxidants are added to fruit juices by the way of nanoencapsulation to improve them. Nanoencapsulation is not known to have adverse health effects.

2.1.1.4. Bread enriched with omega-3 fatty acids

Omega-3 fatty acids are added to bread by nanoencapsulation; this way the unpleasant taste of the fatty acids is not felt, and thus the fortified bread retains its traditional taste. Nanoencapsulation is not known to have adverse health effects.

2.1.1.5. Plastic bottles for beer

Beer bottles with a modified composition are produced by adding a nanocomposite material containing clay particles. The purpose of clay-polymer nanocomposites is to minimize carbon dioxide loss and oxygen uptake to extend the shelf life of carbonated beverages. The toxicological effects of the nanolayer are unknown; it has not yet been demonstrated that nanoparticles can be released from the packaging material.

2.1.1.6. Antimicrobial food packaging for meat and other foods

Food packaging materials containing active nano-silver inhibit the growth of microbes and help to prevent possible bacterial contamination. Nanoparticle-sized silver is presumably cytotoxic. It has not yet been demonstrated that nanoparticles can be released from the packaging material.

3. Materials and methods

To answer the research questions, online questionnaire interviews involving 200 people were conducted. During the sampling, the snowball method was used, i.e., the selection of the sample was not random, but in this way we were able to reach a wide range of respondents. Under these conditions, the survey cannot be considered representative, the results obtained can only be applied to the actual respondents. Background variables of the questionnaire included gender, age, place of residence, education and average income.

In the course of the questionnaire survey, consumer attitudes towards nanotechnology in the food industry were assessed using 17 closed-ended questions. Then, in order to be able to analyze them in depth, two focus group studies were conducted. Consumers’ attitudes towards the topic were determined in advance by screening questions, based on which they were classified into one of the two focus groups. The first group included consumers who rejected nanotechnology based on the screening questions, while participants in the second group viewed this technology favorably. During the formation of the two groups, we sought to ensure that the consumers interviewed were included in the research in an equal distribution with regard to gender. In terms of age, people between the ages of 20 and 65 participated in the interviews.

Due to the pandemic situation at the time of the research, the two groups of eight people each were interviewed via an online platform.

At the beginning of the focus group interviews, participants were asked to briefly introduce themselves, and then two passages, taken from Sodano et al.’s communication and translated into Hungarian [1], were read aloud in the first half of the discussions. The first text introduced nanotechnology in general, while the second part described six products that had been made by some kind of nanotechnological process, but only the advantageous properties of the products have been emphasized in the description. The first half of the interview questions concerned the awareness and acceptance of nanotechnology in the food industry, but group members also had to answer questions related to the texts they had heard.

In the second half of the focus group discussion, the part of the text that highlights the potential risks and negative impacts associated with the technology and, thus, the products was read aloud. Following this, once again participants were asked questions, this time focusing on the risks, and it was examined how much their attitude towards the topic had changed.

4. Results and evaluation

In this chapter, the most important results of the primary research are presented, in the order they took place.

4.1. Results of the questionnaire survey

The first question of the questionnaire focused on factors considered important when purchasing food. This was important because, after this, the backbone of the research was the examination of the acceptance of nanotechnology in the food industry, taking into account the categories mentioned here. As can be seen from Figure 1, of the factors listed, taste was mentioned first, i.e., for 76.0% of the respondents taste was the most important consideration when purchasing or selecting a food. Based on the comparison with the background variables, it was revealed that men in the sample had a significantly (p=0.014) higher proportion (80.0%) who considered taste important than women (61.2%), and also that consumers who, according to their own statements, live in better-than-average financial conditions (live well on their income and can also save some money) also consider taste to be an important criterion when choosing (88.9%).

Slightly behind, high quality (68.5%) and price (63.5%) was second and third in terms of purchasing considerations. As had been expected previously, for these categories, 86.2% of those with a sound financial background rated high quality as an important aspect, while in terms of price, this proportion fell to 47.3%.

High food safety was considered important even less than one half of the respondents (47.0%), which may be due to the fact that they were not aware of the specific meaning of the term.

Respondents considered added value (for example, higher omega-3 fatty acid content) to be the least important aspect, with this factor ranking last of the listed ones with 17.5%. Only 20.0% of men and 16.4% of women consider this category when purchasing food. In terms of financial status, this criterion was least important for consumers with below-average income (7.0%).

In the following, the proportion of respondents with knowledge on nanotechnology in the food industry (spontaneous recall) was examined. The innovative and novel nature of the technology is also supported by the fact that only a quarter of respondents have heard of it.

When the four categories of nanotechnology currently available in the food industry were also listed [6, 7, 8, 9] (supported knowledge), only 62.0% of consumers still answered that they had not yet heard of the new technology in question (Figure 2). Of the entire sample, there was only one person who had heard of all the categories listed. Of the four categories, packaging materials made using nanotechnology were the best known (28.5%). 11.5% each of the participants in the survey have already heard of nanostructured food ingredients and materials, as well as the use of food contact nanomaterials, respectively. Respondents were least familiar with nanostructured delivery systems, the proportion in this case was not even 5.0%. Consumers who have heard of this category had some kind of college degree.

In the following, the acceptance of nanotechnology in the food industry was examined using the aspects listed in the first question that were considered important at the time of purchase. The results are shown in Table 1.

Based on the results obtained (for the sample), it can be said in general that the majority is open to the new technology if it has a beneficial effect on one of the properties of the food purchased. 71.9% of respondents would buy food made with nanotechnology if its organoleptic properties were better. Of the aspects considered important when buying food, taste finished first: 76% of respondents chose this factor. It should be noted that a greater willingness to buy due to more favorable sensory properties was an expected outcome. The older age group gave the highest proportion of affirmative answers to this characteristic (89.5%, p=0.047), and there was no significant relationship to the other background variables. In order to have a positive effect on the texture of foods, 68.8% of the consumers in the sample would buy a product made with a nanotechnological process. In the hope of better texture, 85.2% of respondents aged 56-65 would be open to buying products made with the new technology. A significant increase in the shelf life and use-by date of foods due to the nanotechnology process had an incentive effect on shopping for 62.5% of respondents. In the case of this question, a significantly higher proportion of women answered yes than men (women: 70.0%, men: 46.3%). 78.6% of the respondents to the questionnaire would buy food made with some kind of nanotechnology process if it increased food safety. 90% of the older age group and 78.6% of women were represented in the „yes” answers in this regard. 63.0% of respondents answered „yes” to the question of whether they would buy a food produced with nanotechnology development if it has added value such as a higher omega-3 fatty acid content. This represents an exceptionally high proportion considering that added value as a purchase criterion finished last in terms of importance with 17.5%. Thus, although it is typically not an important factor for the consumers in the sample that the food has some added value, they would still choose a product manufactured with nanotechnology that is richer in omega-3 fatty acids. Finally, 78.1% of respondents were open to food packaging produced with a new method that guarantees safer storage. In this case as well, women and those aged 55-65 had the highest proportion of „yes” answers.

Figure 3 illustrates how many percents more respondents would be willing to pay for a food that has been produced or modified using some kind of nanotechnology process. Typically, the additional cost consumers in the sample considered most acceptable was between 0% (i.e., they would not pay more at all for a product manufactured with this technology) and 5-10% (30.7% and 30.7% of respondents, respectively). 22.4% would pay 0-5% more and 15.6% would pay 10-20% more for this type of food. The proportion of respondents willing to assume an additional cost of more than 20% did not even reach one percent. Consumers who would be willing to pay 0-5% more for a product manufactured with nanotechnology are those who have a lower-than-average monthly net income, while respondents who say they live in better-than-average financial conditions would be willing to pay 5-10%, 10-15%, 15-20%, or even more than 20% more for such foods.

In our research, it was also addressed how respondents felt about the possible adverse consequences of nanotechnology in the food industry. Based on the results obtained, it was found that more than half of the questionnaire respondents (53.6%) believed that foods made with the nanotechnology process carry unknown hazards. In this case, in terms of proportions, men can be said to be the most skeptical, with 74.3% saying that nanotechnology in the food industry could pose a risk.

Figure 4 shows the probability of the occurrence of the different hazards in the opinion of the respondents in percentage distribution. 71.4% of respondents who consider the technology to be risky believed that foods made with the nanotechnology process pose mainly health risks. This was followed by environmental risks (56.3%). In this case, almost twice as many women believed that nanotechnology in the food industry could cause environmental damage (p=0.020). Consumers in the sample considered negative economic and social impacts to be the least probable. For these two categories, typically women were also in the significant majority (p=0.001). However, it can be said for all categories that respondents with higher education represented a higher proportion.

4.2. Results of the focus group studies

Since the main objective of our research was to examine nanotechnology in the food industry from a consumer perspective and to explore the expected rate of acceptance and possible rejection of the technology, after examining the quantitative results of the online questionnaire, it was considered appropriate to analyze the responses received in more depth using a qualitative method, therefore, focus group interviews were conducted to facilitate interpretation.

4.2.1. Results of the focus group study of people accepting nanotechnology in the food industry

Our discussion began with an association game designed to resolve any anxieties of the interviewees. Members of the group were asked to say positive and/or negative words and phrases that come to mind in connection with the topic. The following words were mentioned: innovation, invention, new opportunities, interesting, sci-fi, foods of the future, possible solutions to many problems.

The next question was whether they had already encountered any of the listed categories of nanotechnology applications or something similar (creamier ice cream with the same fat content; salt and sugar that do not form lumps with moisture; fruit juices enriched with bioactive molecules; bread enriched with omega-3 fatty acids; plastic bottles for beer; antimicrobial food packaging for meat and other foods). All of the respondents had already met soft drinks and beers packed in special PET bottles. Fruit juices enriched with various vitamins, minerals and antioxidants were mentioned by several people, and one person saw bread enriched with omega-3 fatty acids in a store while shopping (he didn’t remember exactly which store it was). In addition to the categories read aloud, they have seen eggs that contained excess omega-3 fatty acids, known various dietary supplements to which vitamins, minerals or antioxidants were added, and a participant had read on the internet about an intelligent packaging material that recognizes contaminants. He did not remember whether the packaging material had been made with nanotechnology in the food industry, but he believed that this category fit exactly into this topic.

Following this, those present were asked to express their views and evaluate how they perceived the six categories described above. Positive thoughts were associated with the products by everyone. They were thought to be useful in many ways, and it was thought to be a good idea to add such extra values to foods that allow people to get vitamins and other minerals without having to take separate capsules into their body. According to the participants, the facts that the use of nanotechnology can make food storage safer and increase shelf life can also be advantages. When asked if they would like to buy this type of food, all participants answered in the affirmative. One person stated that he was somewhat averse to nanotechnology-modified ice cream, while two people said the same thing about bread enriched with omega-3 fatty acids, but they could not specifically explain why.

This was followed by solving a task together, in which members of the group were asked to jointly establish an order for the six products based on which they considered to be the most sympathetic and which the least. The popularity of the products is illustrated by the data in Table 2.

The group unanimously agreed with the assumption that in the future we would encounter many of these or similar products on store shelves. It was thought that foods produced with nanotechnology were likely to become more widespread if the pace of food industry developments remained the same. One of our interviewees said that due to the overpopulation of the Earth and the constant decline of arable land, it will be necessary to deploy such tools in order to avoid an increasing rate of hunger and malnutrition, and to prevent people from suffering from the lack of certain nutrients. Everyone has accepted the vision that foods produced with such technology and other similar developments will become more popular and accessible, provided, of course, that they will be available at affordable prices. Intelligent food packaging that recognizes bacteria and contaminants has been found to be especially useful and practical.

According to them, basic foods (dairy products, pasta, flours, cereal flakes) could also be enriched with added values (vitamins, minerals, antioxidants).

In the second half of the focus group interview, the part of the texts was read aloud that described the potential risks involved in using the technology. Following this, it was assessed whether participants’ opinions, attitudes and willingness to buy changed as a result of what they had heard. The majority believed that if it were not safe to consume a product, it would ultimately not be able to be marketed. According to another opinion, while it sounded a little scary, and so he would think twice about buying this type of product, he still would not reject the technology.

Finally, participants were asked to reconsider, in light of the information they had learned, the order established above, as to which category they would be most likely to purchase. For better comparability, the order before describing potential hazards and the new order are listed in the same table. The results are shown in Table 3.

Although the final order was changed at several points, the opinions and willingness to buy of the group members did not change significantly after the exploration of possible dangers.

4.2.2. Results of the focus group study of people rejecting nanotechnology in the food industry

The study scenario in this case was the same as it was for the previous group. Presentation of the first part of the text was followed by an association game, the essence of which was that participants had to say adjectives and expressions, whether positive or negative, that came tom mind about nanotechnology. This time, compared to the interviews with the accepting group, the opinions (answers) were much more mixed: innovative, dangerous, bizarre, this is the future, foods made in a laboratory, unnatural. One of our interviewees also noted that these products were likely to be very expensive.

Of the six products made with nanotechnology in the food industry, half of the group had already encountered fruit juices enriched with bioactive molecules, and everyone was familiar with the special PET bottles. As similar products, sports drinks and dietary supplements fortified with vitamins and minerals were mentioned, which had already been encountered by them in retail trade, and one person had already read online about packaging materials made with nanotechnology, and another participant cited a scientific paper on artificial meat as an example.

Following this, once again, members of the group were asked to share their views on the six products which had been introduced at the beginning of the interview. Someone thought it was extremely scary to hear about these, while others thought that they would be very unhealthy for sure. Many people felt that it was unnecessary to enrich fruit juices with such substances when they were already full of vitamins anyway. The idea of bread enriched with omega-3 fatty acids was specifically thought to be „crazy”. One participant did not consider packaging to be a bad idea, and two of them also commented favorably on PET bottles.

When asked whether they would buy these products, the answer was clearly no. The group was less prone to rejection in the case of the PET bottles, with 4 people inclined to buy, and one person said the same about antimicrobial packaging.

Subsequently, the group rejecting nanotechnology also had to jointly establish an order for the six products, based on acceptability (in this case, we cannot speak of popularity, as the members of the group reject nanotechnology in the food industry). The results are shown in Table 4.

Regarding the vision for the future, participants believed that the trend of developments suggests that more and more products of this kind will be available commercially. There was also a remark in this regard that „the world is not moving in the right direction”. One person added that he was confident that we would stick to natural food sources. Several people agreed with the statement that if it is not the food industry that works with such technology, but the construction or textile industry, it may even be useful. When asked whether they would like more of these products to be available in the future, the group’s response was a clear and consistent no.

The final chapter of the focus group interview concentrated on the potential risks of nanotechnology. After discussing the potential dangers of nanotechnology with participants, their opinions were asked. Their position did not change much after what they had heard, since, as they said, they had not considered it to be a good idea, and it only strengthened their belief that such a technology could have negative consequences. The unanimous opinion of the group was that they would continue to not buy such products as they are sure that they are harmful not only to human health but also to the environment.

As a final task, participants were asked to, in possession of all the information, jointly establish a new, final order as to which category they would consider most acceptable and which least acceptable. Compared to the previous one, the order did not change much, and the result was as follows. The orders before and after the description of the risks (new order) are shown in Table 5.

5. Conclusions

Despite the fact that 74.5% of the respondents were not previously familiar with nanotechnology and its application possibilities, and almost half of the respondents believed that it involved some risk, the survey of knowledge of nanotechnology and the examination of consumers’ willingness to buy revealed that the degree of acceptance of the technology and the willingness to buy can be said to be very favorable. If, through this technology, food quality is expected to change in a positive direction, acceptance exceeds 60%.

The most important aspect when buying foods was taste, while added value finished last with 17.5%. Nevertheless, 63.0% of those who completed the questionnaire replied that they would buy a product made with a nanotechnology process if the product thus contained some kind of added value.

The focus group interview revealed that the group of acceptors, as expected, was extremely positive about the technology, and even after the description of the potential risks, neither their opinion, nor their willingness to buy typically changed.

Reaffirming Berekaa’s claim that the public is very concerned about toxicity and potential negative environmental effects [5], in the case of the group of rejectors, participants unanimously stated that the technology is extremely risky and dangerous to both the environment and humans. However, they also added that in their view and based on the trends, the proliferation of commercially available such products will be inevitable in the future. In their case it can be said that although they do not prefer the possibilities of using nanotechnology, their rejection was less pronounced for those categories of application of the technology that do not specifically change the properties of foods, but their peripherals (such as packaging).

In Chapter 3 of our paper, the statements of Sodano were already quoted, according to which the willingness to buy nanofoods for the six categories examined (creamier ice cream with the same fat content; salt and sugar that do not form lumps with moisture; fruit juices enriched with bioactive molecules; bread enriched with omega-3 fatty acids; plastic bottles for beer; antimicrobial food packaging for meat and other foods) depends to a large extent on the assessment of the perceived risks and benefits [1]. Our results obtained in the course of our research support this, as the willingness to buy of consumers who already had a positive attitude towards the technology is also very favorable, while rejectors showed the opposite consumer behavior.

6. Acknowledgment

This publication was prepared with the professional support of the New National Excellence Program of the Ministry of Innovation and Technology, code number ÚNKP-20-3-I-DE-404, financed from the National Research, Development and Innovation Fund.

7. References

[1] Sodano, V., Gorgitano, M.T., Verneau, F. (2015): Consumer acceptance of food nanotechnology in Italy. British Food Journal 118 (3) pp. 714-733

[2] Zentai A., Frecskáné Csáki K., Szeitzné Szabó M., Farkas J., Beczner J. (2014): Nanoanyagok felhasználása az élelmiszeriparban. Magyar Tudomány 175 (8) pp. 983-993

[3] Cubadda, F., Aureli, F., D Amato, M., Raggi, A., Mantovani, A. (2013): Nanomaterials in the food sector: new approaches for safety assessment. Rapporti ISTISAN 13/48.

[4] Joseph, T. and Morrison, M (2006): Nanoforum report: nanotechnology in agriculture and food. (Hozzáférés: 2014. 06. 12.).

[5] Berekaa, M. M. (2015): Nanotechnology in food industry; Advances in Food processing, Packaging and Food Safety. International Journal of Current Microbiology and Applied Sciences 4 (5) pp. 345-357

[6] Chaudhry, Q., Scotter, M., Blackburn, J., Ross, B., Boxall, A., Castle, L. y and Watkins, R. (2008): Applications and implications of nanotechnologies for the food sector. Food Additives and Contaminants 25 (3) pp. 241-258

[7] Cushen, M., Kerry, J., Morris, M., Cruz-Romero, M. and Cummins, E. (2012): Nanotechnologies in the food industry. Trends in Food Science & Technology 24 (1) pp. 30-46

[8] Weir, A., Westerhoff, P., Fabricius, L., Hristovski, K. and von Goetz, N. (2012): Titanium dioxide nanoparticles in food and personal care products. Environmental Science & Technology 46 (4) pp. 2242-2250 DOI

[9] Mura, S., Seddaiu, G., Bacchini, F., Roggero, P.P. and Greppi, G.F. (2013): Advances of nanotechnology in agro-environmental studies. Italian Journal of Agronomy 8 (18) pp. 127-140

[10] Chaudhry, Q., Castle, L., Watkins, R. (2010): Nanotechnologies in Food. Royal Society of Chemistry Publishers, Cambridge, UK.

[11] Bradley, E. L., Castle, L., Chaudhry, Q. (2011): Applications of Nanomaterials in Food Packaging with a Consideration of Opportunities for Developing Countries. Trends in Food Science & Technology 22 pp. 604-610

[12] Sozer, N. and Kokini, J.L. (2009): Nanotechnology and its applications in the food sector. Trends in Biotechnology, 27 (2) pp. 82-89.

[13] Neethirajan, S. and Jayas, D.S. (2011): Nanotechnology for the food and bioprocessing industries. Food and Bioprocess Technology 4 (1) pp. 39-47

[14] Cushen, M., Kerry, J., Morris, M., Cruz-Romero, M., Cummins, E. (2012): Nanotechnologies in the food industry. Trends in Food Science & Technology 24 (1) pp. 30-46

[15] Qureshi, M.A., Karthikeyan, S., Karthikeyan, P., Khan, P.A., Uprit, S. and Mishra, U.K. (2012): Application of nanotechnology in food and dairy processing: an overview. Pakistan Journal of Food Sciences 22 (1) pp. 23-31

[16] Cockburn, A., Bradford, R., Buck, N., Constable, A., Edwards, G., Haber, B., Hepburn, P., Howlett, J., Kampers, F., Klein, C., Radomski, M., Stamm, H., Wijnhoven, S. and Wildermann, T. (2012): Approaches to the safety assessment of engineered nanomaterials (ENM) in food. Food and Chemical Toxicology 50 (6) pp. 2224-2242

[17] Hubbs, A.F., Sargent, L.M., Porter, D.W., Sager, T.M., Chen, B.T., Frazer, D.G. and Battelli, L.A. (2013): Nanotechnology toxicologic pathology. Toxicologic Pathology 41 (2) pp. 395-409

[18] Halliday, J. (2007): EU Parliament votes for tougher additives regulation. FoodNavigator.com (Hozzáférés: 2014. 06. 12.).

DOI: https://doi.org/10.52091/JFI-2021/1-1-ENG

Received: October 2020 – Accepted: January 2021

¹ Department of Nutritional Science and Production Technology, Faculty of Agricultural and Environmental Sciences, Szent István University, Kaposvár Campus
² Department of Physics and Control, Faculty of Food Science, Szent István University, Budapest Campus
³ Adexgo Kft., Balatonfüred, Hungary
^* Corresponding author: bazar@agrilab.hu

Keywords

Fructose (fruit sugar), sugars, °Brix, NIR-spectroscopy (near-infrared), adverse physiological effects of fructose, metabolic disorder, cardio-vascular diseases, pre-treatment of spectra, valence vibration, harmonic vibration, statistical spectra analysis

2. Introduction

Food sweeteners have become the most widely used additives in the food processing industry, especially in the production of beverages and other products such as desserts and yoghurts. One of the oldest sweetener to have been documented in history is honey [1]. This, and some of the traditional sweeteners such as maple syrup, carob, and agave, consumed for decades are largely made up glucose, fructose, sucrose, minerals and other compounds [1]. Glucose is almost always present in foods and plays an essential role in the regulation of metabolism in human. It can be ingested either as free available sugar (glucose powder) or bonded in polymers, in the case of starch, dextrin, and maltodextrins. Glucose could also be bonded in disaccharides, like in the case fructose bond to glucose in sucrose [2].

For some time now, concerns about the form and levels of sweeteners used in the food industry, and the °Brix value of processed foods, have been topical due to the health implications of the consumers. This is mainly because of the risk of developing metabolic abnormality (diabetes) associated with high intake of sugar, especially sugar of high fructose content. Consumers are therefore becoming more conscious of what they consume, and will at times prefer a reduction of the caloric levels of processed foods, consequently reducing sugar intake [3].

High fructose intake was found to be associated with a high risk of metabolic syndrome [4], obesity, diabetes and an increase in blood triglyceride concentrations and insulin resistance compared with high glucose intake [5], [6], [7]. High risk of cardiovascular diseases and even malignant tumors in body tissues may be related to excessive fructose intake [8], and also dyslipidemia and kidney diseases [9].

Over the years, the application of near-infrared spectroscopy (NIR) to analyze the forms of sugar in food sweeteners, has been found to be easier, faster and cost-efficient [10] compared with tedious and reagent involving methods, such as gas chromatography (GC), high-performance liquid chromatography (HPLC) and enzymatic analysis [11], [12]. The HPLC is the most frequently used method for assessing free fructose, free glucose, sucrose, maltose, and lactose content [13].

The NIR spectral region is found between 800 to 2500 nm (12500–4000 cm−1) range, with absorptions representing overtones and combinations which are associated with –CH, –OH, –NH, and –SH functional groups [14]. In the case of glucose, 1st overtone of O–H stretching corresponds to absorption bands at 1195, 1385, 1520, 1590, 1730 nm, 1st overtone of O–H stretching of fructose and sucrose at 1433 nm, and O–H combination band of sucrose, glucose and fructose at 1928 nm [14].

Mono- and disaccharides, such as glucose, fructose, sucrose, lactose, were also analyzed in aqueous solutions [15]. Although the same molar concentrations of all the concerned sugars were dissolved, the mass that those represented differed considerably due to the differences in molecular weights of mono- and disaccharides. When quantifying sugars in mixtures, the molar concentration of the sugar solutions gave less accurate calibration models compared with those fitted on weight per volume concentration. Since the spectral information is mostly the light absorbance of chemical bonds during excitation, this information is more proportional with the number of chemical bonds and atoms in the aqueous solution, than with the number of molecules. Regression coefficient vectors of calibration models for each of the sugars also revealed the spectral regions holding the highest importance in the quantitative analysis of the sugars. Regression vectors of the 1100-1800 nm interval, associated with signals of O–H and C–H bonds, showed the significance of characteristic spectral regions of water and the dissolved sugars. The calibration on the concentration of the sugars within the mixtures showed accurate validation performance even at low concentration levels (0.0018 – 0.5243 g/cm3), R2CV of 0.841 and 0.961, SECV of 0.024 g/cm3 and 0.012 g/cm3 for glucose and fructose, respectively. This showed possible quantification of a specific sugar in a mixture of sugars in a solution using NIR spectroscopy [15].

In related studies [10], [16], [17], glucose, fructose and sucrose were quantified in different fruit juices using NIR, and accurate partial least square regression (PLSR) models were reported (R2 > 0.854, 0.963, 0.953 for glucose, fructose, sucrose, respectively). Good PLRS models were reported for predicting glucose within 900-2200 nm wavelength range [18], whereas the 900-1650 nm interval was reported to be good for the discrimination of organic sugar and conventional brown sugar using partial least squares discriminant analysis (PLS-DA) models [19]. In a study, the concentration of glucose in an aqueous mixture of glucose, albumin and phosphate was quantified using NIR and reported accurate PLRS models [20]. The possibility to predict the glucose, fructose and sucrose content in Morindae officinalis extracts utilizing NIR was also reported [21].

The Hungarian food industry is flooded with many sweeteners for food processing. However, there are three major sweeteners: K-syrup LDX and K-sweet F55, which are two commonly used isosugars, and D-sucrose. K-syrup LDX is a sweet, viscous, quickly crystallizing syrup often used in food and pharmaceutical industry as a raw material for fermentation. It contains high amount of glucose or dextrose (93%), and small amount of fructose (0.5%) and viscous liquid [22]. K-sweet F55, however, is a high caloric isosugar consisting of glucose and fructose, where the fructose content is higher (55%) than the glucose (45%) [23], and the third sweetener is D-sucrose or refined sugar, which is increasingly being replaced with K-syrup LDX and K-sweet F55.

This study aimed to determine the applicability of NIR spectroscopy to quantify glucose, fructose, sucrose content and °Brix of aqueous solutions of the widely used sweeteners, D-sucrose, K-syrup LDX, and K-sweet F55.

3. Materials and Methods

3.1. Sample preparation

Three kinds of sugars were used with brand names: D-sucrose (Carl Roth GmbH, Karlsruhe, Germany): 100% sucrose; K-Syrup LDX (KALL Ingredients Kft., Tiszapüspöki, Hungary): 93% glucose + 0.5% fructose + 6.5% water; K-Sweet F55 (KALL Ingredients Kft., Tiszapüspöki, Hungary): 45% glucose + 55% fructose. Aqueous solutions were prepared at 10 different concentrations for each of the three sugars, separately. A total sample of 30 samples was prepared, 100 ml of each.

3.2. Laboratory measurement

°Brix was measured with Hanna HI96801 Digital Refractometer, and recorded as reference for subsequent NIRS calibrations. Glucose, fructose and sucrose concentration of the respective sugar solutions was calculated based on the mass of sweetener added to the solutions and the percentages of the individual sugars within the sweeteners. The following formulas were used for the calculation of glucose and fructose in K-sweet F55 and K-syrup LDX solutions:

1. Glucose in K-syrup LDX solution = 93/100*amount of K-syrup in solution (g/100g)
2. Fructose in K-syrup LDX solution = 0.5/100*amount of K-syrup in solution (g/100g)
3. Glucose in K-Sweet F55 solution = 45/100*amount of K-sweet F55 in solution (g/100g)
4. Fructose in K-sweet F55 solution = 55/100* amount of K-sweet F55 in solution (g/100g)

Accordingly, each of the 30 samples was described with °Brix, and concentrations of total sugar, glucose, fructose and sucrose, as listed in Table 1.

D-sucrose: 100% sucrose; K-syrup LDX: 93% glucose+0.5% fructose; K-sweet F55: 45% glucose+ 55% fructose; SD: standard deviation; Max: maximum value; Min: minimum value

3.3. NIRS measurement

The samples were scanned at room temperature (25 °C) using a FOSS NIRSystems 6500 (FOSS NIRSystems, Inc, Laurel, MD, USA) spectrometer, operated with WinISI v1.5 software (InfraSoft International, Port Matilda, PS, USA). The scanning was done in transmission mode after measuring 1 ml sugar solution into a quartz cuvette having 1 mm pathlength. Two rounds of scanning of each sample were done randomly, and the subsequent sample was used to wash the cuvette three times between each sample scanning. Sixty spectral data were obtained and the spectra of the two rounds were averaged resulting in 30 spectra.

3.4. Spectral pre-processing and multivariate data analysis

The Unscrambler v9.7 (CAMO Software AS, Oslo, Norway) software was used for the analysis of the NIR data, while the MS Excel 2013 was used to calculate the descriptive statistics for the variables measured and calibrated for °Brix, glucose, fructose, and sucrose concentration.

For scatter correction of spectra, and to obtain accurate and robust calibration models, several spectral types of preprocessing were performed: standard normal variate (SNV), multiplicative scatter correction (MSC) and gap-segment second derivative (2nd order derivative, gap of 5 data points, segment of 5 data points).

Using multivariate data analyses, both the separation of the solutions prepared with different sweeteners and the calibration on the targeted quantitative parameters was performed. Principal component analysis (PCA) [24] was used to investigate the multidimensional pattern of the spectra data and to identify differences among the three groups of the sweetener solutions. The spectral data within the NIR range (1100-1850nm) were calibrated with the laboratory data as the reference, using partial least squares regression (PLSR) models [24]. The optimum number of latent variables (LV) used for the PLSR modelling was determined by full (leave-one-out) cross-validation, when in a 30-step iterative process each of the 30 samples was left out of the calibration once and was used for validating the model [24].

Evaluation of PLSR models was done by comparing the calibration statistics with that of the cross-validation. The determination coefficient of calibration (R2C) and cross-validation (R2CV), and the root mean square error of calibration (RMSEC) and cross-validation (RMSECV) were compared, where larger R2 value and smaller RMSE value represent the better model. During the model optimization processes, RMSECV values were minimized.

4. Results and discussions

The recorded raw spectra show the typical NIR absorption of water, with a major peak at 1450 nm, representing the 1st overtone region of O–H bond (Figure 1). The small peak around 1780 nm represents the 1st overtone of C–H bonds. The second derivative spectra were calculated with the gap-segment derivative function, where both gap and segment were set to 5 data points to avoid noise enhancement of the derivative function, still keeping the useful signals within the pretreated data.

The negative peaks of the 2nd derivative spectra (Figure 2) indicate the locations and relative amplitude of the original overlapping absorptions appearing as one in the raw spectra. This shows the well-described phenomenon that major peak of the raw spectrum at 1450 nm is formed by at least two underlying absorptions of water at 1416 nm and 1460 nm, representing less and more H-bonded water, respectively [15].

The applied spectral pretreatments (2nd derivative, or SNV, or MSC) did not allow visual differentiation of solutions with different sweeteners, while the gradual changes of the water absorption peaks indicated the effect of the increasing concentration of dissolved sugars on the structure of water [15].

Figure 3 shows the 3D plot of the PCA performed with 2^nd derivative spectra of all the 30 solutions. The solutions of the three types of sweeteners are indicated with different colors and numbers. The two plots show the same result from different angles, highlighting that 4th principal component (PC4) is responsible for the separation of K-Sweet F55 from D-sucrose and K-Syrup LDX, and PC2 is responsible for the separation of D-sucrose from K-Syrup LDX and K-Sweet F55. Thus, PC2, as new latent variable covering approximately 2% of the variance of the original NIR data, describes the difference between the disaccharide and monosaccharide solutions, while PC4, covering less than 1% of the variance of the original NIR data, describes the difference between the solutions of high fructose syrup and that of the other sweeteners. The combination of PC2 and PC4 describes the differences between glucose solutions and others.

Figure 4 shows the loading vectors of PC2 and PC4. The wavelength regions having the largest deviation from zero are the most responsible for score values of the principal components, thus, the assigned peaks indicate the absorptions causing the difference between the sugar solutions. The band assignments are in good harmony with previous findings [14,15], i.e. peaks in the 1300-1600 nm interval refer to the molecular changes of water caused by the dissolved sugars, while the peaks in the 1600-1850 nm interval represent characteristic C–H bands.

The results of the calibration models developed using PLS regression on the measured °Brix and calculated fructose, glucose, sucrose concentrations are presented in Table 2 and Figure 5.

LV: number of latent variables, R²_C: determination coefficient of calibration, RMSEC: root mean square error of calibration, R²_CV: determination coefficient of cross-validation, RMSECV: root mean square error of cross-validation, MSC: multiplicative scatter correction, SNV: standard normal variate, 2D5G5S: 2nd order derivative with 5-point gap and 5-point segment

The best results for °Brix were achieved with no spectral pretreatment. The RMSE of °Brix remained around 1 °Bx, which was almost third of the standard deviation of the measured reference values. The RMSE of the sugar concentrations was similarly low. The least accurate model was achieved for fructose, which is caused by the group of samples with fructose content below 0.05% - for these samples the model performed worse than in the higher concentration regions (Figure 5. (b)). Second derivative pretreatment gave the best result for glucose and sucrose, while the best models for fructose were achieved without pretreatment of the NIR spectra.

The calibration and cross-validation regression lines (Y-fit) of the best °Brix, fructose, glucose and sucrose models are shown in Figure 5. The black diagonal line shows the optimal Y-fit, while blue and red lines show the calibration and cross-validation Y-fits. The blue dots show the NIR predicted composition values of samples during calibration in the function of the laboratory reference values, and red dots show the NIR predicted values at cross-validation testing, again, in the function of the reference values measured. The closer the dots are to the regression line and the less the regression line deviates from the optimal Y-fit, the better the calibration model is. In most of the cases, the achieved Y-fits are hitting the optimum, meaning that the NIR predicted values are almost equal to the actual laboratory reference values. The calibration and cross-validation results of this study are in agreement with the previously cited results achieved with sugar solutions and fruit juices. These results confirm that, after a proper calibration process, NIR spectroscopy is a useful and effective tool for easy, rapid and accurate measurement of individual sugars in mixed solutions.

5. Conclusions

The results of this study performed with widely used sweeteners confirm the previously published findings that NIR spectroscopy is a useful and powerful technology to detect and quantify individual sugar types even in mixture solutions. Since NIR spectrometers have not only reached the portable size but have become extremely small as a fingernail-sized chip, the importance of this technology in everyday food qualification seems to be underestimated. Wide aspects of applications should be tested and used for monitoring products and warrant food safety and quality. Among these applications, checking and certifying the fructose content of beverages and foods would advantage consumers’ health, as this constituent has been proven to raise the risk of several diseases of modern times. NIR spectroscopy as secondary correlative analytical technology will likely remain to be unsuitable for detecting and quantifying fructose in a complex liquid of completely unknown composition, but may be suitable for indicating the excessive presence of fructose in a known liquid meant to be containing no or only a certain amount of fructose. The usability of NIR tools is limited and they should not be considered as subtituents of classical analytical methods, however, by rational use of opportunities, useful applications can be developed for practice.

6. References

[1] Edwards, C. H., Rossi, M., Corpe, C. P., Butterworth, P. J., & Ellis, P. R. (2016): The role of sugars and sweeteners in food, diet and health: Alternatives for the future. Trends in Food Science and Technology, 56, pp. 158-166.
https://doi.org/10.1016/j.tifs.2016.07.008

[2] White, E., McMahon, M., Walsh, M., Coffey, J. C., & O’Sullivan, L. (2014): Creating Biofidelic Phantom Anatomies of the Colorectal Region for Innovations in Colorectal Surgery. Proceedings of the International Symposium on Human Factors and Ergonomics in Health Care, 3 (1), pp. 277-282.
https://doi.org/10.1177/2327857914031045

[3] Gardner, C., Wylie-Rosett, J., Gidding, S. S., Steffen, L. M., Johnson, R. K., Reader, D., & Lichtenstein, A. H. (2012): Nonnutritive sweeteners: Current use and health perspectives: A scientific statement from the American heart association and the American diabetes association. Circulation, 126 (4), pp. 509-519.
https://doi.org/10.1161/CIR.0b013e31825c42ee

[4] Taskinen, M. R., Packard, C. J., & Borén, J. (2019): Dietary fructose and the metabolic syndrome. Nutrients, 11 (9), pp. 1-16.
https://doi.org/10.3390/nu11091987

[5] Bray, G. A. (2013): Energy and fructose from beverages sweetened with sugar or high-fructose corn syrup pose a health risk for some people. Advances in Nutrition, 4 (2), pp. 220-225.
https://doi.org/10.3945/an.112.002816

[6] Malik, V. S., & Hu, F. B. (2015): Fructose and Cardiometabolic Health What the Evidence from Sugar-Sweetened Beverages Tells Us. Journal of the American College of Cardiology, 66 (14), pp. 1615-1624.
https://doi.org/10.1016/j.jacc.2015.08.025

[7] Rizkalla, S. W. (2010): Health implications of fructose consumption: A review of recent data. Nutrition and Metabolism, 7, pp. 1-17.
https://doi.org/10.1186/1743-7075-7-82

[8] Biró, G. (2018): Human biological characteristics of fructose. Journal of Food Investigation, 64 (1), pp. 1908-1916.

[9] Collino, M. (2011): High dietary fructose intake: Sweet or bitter life? World Journal of Diabetes, 2 (6), pp. 77.
https://doi.org/10.4239/wjd.v2.i6.77

[10] Liu, Y., Ying, Y., Yu, H., & Fu, X. (2006): Comparison of the HPLC method and FT-NIR analysis for quantification of glucose, fructose, and sucrose in intact apple fruits. Journal of Agricultural and Food Chemistry, 54 (8), pp. 2810-2815.
https://doi.org/10.1021/jf052889e

[11] Giannoccaro, E., Wang, Y. J., & Chen, P. (2008): Comparison of two HPLC systems and an enzymatic method for quantification of soybean sugars. Food Chemistry, 106 (1), pp. 324-330.
https://doi.org/10.1016/j.foodchem.2007.04.065

[12] Yuan, H., Wu, Y., Liu, W., Liu, Y., Gao, X., Lin, J., & Zhao, Y. (2015): Mass spectrometry-based method to investigate the natural selectivity of sucrose as the sugar transport form for plants. Carbohydrate Research, 407, pp. 5-9.
https://doi.org/10.1016/j.carres.2015.01.011

[13] International Association of Analytical Chemistry, A. (1990): Official Methods of Analysis of the AOAC. Arlington VA

[14] López, M. G., García-González, A. S., & Franco-Robles, E. (2017): Carbohydrate Analysis by NIRSChemometrics. In K. G. Kyprianidis & S. Jan (Eds.), Developments in Near-Infrared Spectroscopy pp. 81-95
https://doi.org/10.5772/67208

[15] Bázár, G., Kovacs, Z., Tanaka, M., Furukawa, A., Nagai, A., Osawa, M., Itakura, Y., Sugiyama, H., Tsenkova, R. (2015): Water revealed as molecular mirror when measuring low concentrations of sugar with near infrared light. Analytica Chimica Acta, 896, pp. 52-62.
https://doi.org/10.1016/j.aca.2015.09.014

[16] Xie, L., Ye, X., Liu, D., & Ying, Y. (2009): Quantification of glucose, fructose and sucrose in bayberry juice by NIR and PLS. Food Chemistry, 114 (3), pp. 1135-1140.
https://doi.org/10.1016/j.foodchem.2008.10.076

[17] Rodriguez-Saona, L. E., Fry, F. S., McLaughlin, M. A., & Calvey, E. M. (2001): Rapid analysis of sugars in fruit juices by FT-NIR spectroscopy. Carbohydrate Research, 336 (1), pp. 63-74.
https://doi.org/10.1016/S0008-6215(01)00244-0

[18] Mekonnen, B. K., Yang, W., Hsieh, T. H., Liaw, S. K., & Yang, F. L. (2020): Accurate prediction of glucose concentration and identification of major contributing features from hardly distinguishable near-infrared spectroscopy. Biomedical Signal Processing and Control, 59, pp. 101923.
https://doi.org/10.1016/j.bspc.2020.101923

[19] De Oliveira, V. M. A. T., Baqueta, M. R., Março, P. H., & Valderrama, P. (2020): Authentication of organic sugars by NIR spectroscopy and partial least squares with discriminant analysis. Analytical Methods, 12, pp. 701-705.
https://doi.org/DOI: 10.1039/C9AY02025J

[20] Khadem, H., Eissa, M. R., Nemat, H., Alrezj, O., & Benaissa, M. (2020): Classification before regression for improving the accuracy of glucose quantification using absorption spectroscopy. Talanta, 211 (January), pp. 120740.
https://doi.org/10.1016/j.talanta.2020.120740

[21] Hao, Q., Zhou, J., Zhou, L., Kang, L., Nan, T., Yu, Y., & Guo, L. (2020): Prediction the contents of fructose, glucose, sucrose, fructo-oligosaccharides and iridoid glycosides in Morinda officinalis radix using near-infrared spectroscopy. Spectrochimica Acta - Part A: Molecular and Biomolecular Spectroscopy, pp. 234
https://doi.org/10.1016/j.saa.2020.118275

[22] KALL, I. (2020): K-syrup LDX. Retrieved from http://kallingredients.hu/en/products/2/14/k-syrup-ldx

[23] KALL, I. (2020): K-sweet F55. Retrieved from http://kallingredients.hu/en/products/2/12/k-sweet

[24] Naes, T., Isaksson, T., Fearn, T., & Davies, T. (2002): A user-friendly guide to multivariate calibration and classification. Chichester, UK: NIR Publications

DOI: https://doi.org/10.52091/JFI-2021/1-3-ENG

Received: 2020. September – Accepted: 2020. December

¹ Széchenyi István University, Faculty of Food and Agricultural Sciences, Department of Food Science, Mosonmagyaróvár, Hungary

Keywords

yogurt making, microwave irradiation, fruit drying, microbiological parameters, total viable count, yeast count, mold count, Escherichia coli, coliform

2. Introduction and literature review

Milk has been a mainstay in the human diet since the beginning of human history. Its useful ingredients have a beneficial effect on a person’s healthy physical and mental development. The ingredients of milk are physiologically beneficial, one of the outstanding features being its high calcium content, therefore it plays a role in the bone formation of developing organisms [1], and it also contains proteins that are important and easy to use for the body. Due to all these properties, dairy products can be considered as staple foods in the human diet. In the food industry, the milk of many farm animals (sheep, goats, cattle) is processed, but in Hungary cow’s milk is consumed in the largest amount.

Yogurt is a dairy product consumed all over the world. Nutrition science professionals believe that this sour milk product has a high nutritional value (a significant part of its lactose content is broken down during fermentation and it has a significant concentration of Ca++) and beneficial bioactive effects (prebiotic ingredients and probiotic bacteria). Natural yogurt is made by adding lactic acid bacteria that induce lactic acid fermentation in the culture medium during their basic physiological activities. Of all products manufactured from milk, yogurt is the most popular worldwide [2].

In the case of fruit yogurt, when dried fruit or dried pieces are added to the yogurt, the dried products tend to absorb some of the free water in the yogurt gel, thus helping to separate the whey of the product during storage [3]. It is also an advantage of adding fruit that, according to some studies [4], the addition of 10 v/v% of fruit significantly improves the physico-chemical properties of the product. The interior of healthy plant tissues does not contain microorganisms, so the primary microbiota of plant raw materials comes mainly from the soil, water, air and, occasionally, from insects or animals. Plant parts developing in the soil (tubers, roots) and in the vicinity of the soil are usually heavily contaminated, their microflora composition is practically identical to that of the soil. Microorganisms are present on fruit surfaces in the amount of roughly 10³ to 10⁵ CFU/g, a significant part of which are lactic acid and acetic acid bacteria. However, the largest part of the microbiota is made up of yeasts, the most common of which are Hensaniaspora, Torulaspora, Pichia, Saccharomyces, Candida and Rhodotorula species. Common spoilage microorganisms in fruits include Alternaria, Aspergillus, Fusarium, Monilia and Mucor species. Fruits are excellent culture media for molds, including many mycotoxin-producing ones. Contamination of the raw material and improper storage conditions often also allow the formation of toxic metabolites. For example, patulin, a toxic substance (mycotoxin) produced by Aspergillus and Penicillum fungi, can be detected in moldy fruits (mainly apples and pears) [5].

During the technological processing of fruits, cutting, slicing, chopping and peeling increase the likelihood of cross-contamination from other materials, tools and equipment at different stages of production. In addition, the increased availability of sugars and other nutrients in minimally processed fruits contributes to the change in the microbiota and increases its population [6, 7]. The main factors in the microbiota of the raw material are the hygiene of the surface of the materials used in the production and the processing equipment, as well as the hygiene of the production environment and the food handlers, which determine the microbiological quality and safety of the final product [8, 9, 10]. The authors of a study on minimally processed plant-based foods detected high total aerobic microorganism counts on food contact surfaces, especially on peelers, knives and cutting boards [10]. The same researchers also reported high levels of facultative anaerobic bacteria of the Enterobacteriaceae family on cutting tables and cutting boards [10]. Although washing and other decontamination procedures are used in the manufacturing processes of all processing plants, it is still difficult to achieve a significant reduction in microbial contamination [11]. Favorable conditions for the growth of microorganisms present in fruits and vegetables can also develop during the packaging and storage periods. Lehto et al. discovered a large number of aerobic microorganisms in surface sampling of vegetable processing plants on devices and equipment in contact with already cleaned vegetables, as well as in the air space of storage, processing and packaging rooms [10].

In the food industry, heat treatment processes are the most important determinants of food safety. Heat treatment of milk is necessary to guarantee its microbiological safety by killing pathogenic microorganisms in milk. Several heat treatment methods are used in the food industry. The efficiency of the heat treatment is ensured by strictly defined temperatures and holding times. In addition to the raw materials of the products, it is also important to ensure the appropriate microbiological properties of the additives. „Heat treatment is an operation related to the warming or heating of milk, cream, etc., the objective of which is to reduce the number of or destroy microorganisms” [12]. Heat treatment during milk processing is a general technological step aimed at improving the shelf life of milk by inactivating microorganisms and enzymes. The use of a raw material with a favorable microbiological condition can also improve the texture quality of certain dairy products, such as yogurt [13].

Microwave technology as a heat treatment process is primarily used in households. In the food industry, it can currently only be used reliably in certain areas. The reason for this is that heat transfer is uneven in microwave equipment, and underheated or overheated places develop in the product. In the case of liquids flowing in a pipe, e.g., when treating milk with microwave energy, this can be avoided [5]. where this technique can be used, it is an advantage, as the time of treatments applied to foods can be reduced, thus making the technique economical. In addition to cost-effectiveness, an additional advantage is that the directions of heat and material transport are the same, so that a dry crust that prevents flow is not formed [14]. Areas of application include drying, thawing of frozen meat, tempering, pasteurization, sterilization and prevention of food discoloration [15, 16].

Sterilizing and antimicrobial effects are also attributed to microwave radiation. In Pozar’s experiments [17], this effect could be achieved using a frequency of 2,450 MHz, and in some cases even using a frequency of 915 MHz. Radiation increases the shelf life of foods by killing the microbes in the food and/or inhibiting their growth.

The effect of microwave radiation on microbes has been investigated in a wide variety of foods and food raw materials, especially in meats. The spreading of microwave pasteurization [18] has been facilitated by the fact that its use in foods does not cause significant damage, as opposed to traditional heat transfer methods. The reason for this is the short heat treatment and irradiation times [19, 20].

Compared to the microwave treatment technology, drying is a more traditional method, the essence of which is the extraction of most of the water content from the fruit, less often, from the vegetable, by gentle heat transfer, which leaves behind an intensely flavored concentrate of significantly lower weight and size that the starting material. Thus, microscopic organisms that remain on the dried fruit lose their viability and ability to reproduce due to a lack of available water. Fresh fruits contain 90 to 95% water, which drops below 15% after drying. In this way, spoilage caused by bacteria and molds can be prevented while retaining certain nutrients, roughage and minerals, such as iron. Compared to fresh fruits, dried fruits contain a lot of carbohydrates, fiber and antioxidants, flavonoids, phenolic acids, carotenoids and vitamins in a concentrated form [21, 22].

The food industry produces yogurt products of various compositions, but the technological processes used in their production are almost the same until the inoculation of milk. Milk is usually inoculated with a 2 to 3% starter culture and incubated at 40 to 45 °C. In this temperature range, the desired final acidity is reached in 3 to 4 hours. If a lower temperature (30-37 °C) is used, the operation takes longer (7-8 hours), but in this case excessive acidification of the product can be prevented [23].

The bacterium Streptococcus thermophilus is primarily responsible for the taste, aroma and texture of the yogurt, and is in fact able to ferment pasteurized milk on its own into yogurt, however, in addition to Streptococcus, Lactobacillus bulgaricus is also used for the fermentation to produce the acids that from in the product. S. thermophilus and L. bulgaricus are usually used simultaneously, in a 1:1 ratio, when inoculating the milk (pH 6.6). Their proportion changes as the fermentation progresses [24].

3. Materials and methods

Manufacture of the products

For the manufacture of the products, raw, untreated milk and 2.8% fat UHT drinking milk (Mizo) were used, which were mixed in a 2:1 ratio before making the yogurts. Raw milk was pasteurized on a hot plate at 75 °C for 15 minutes to achieve adequate initial microbiological safety, and then the drinking milk was added. After heat treatment, the temperature of the milk was allowed to drop to 30 °C, it was inoculated with the amount of thermophilic yogurt culture (Lactobacillus delbrueckii subsp. bulgaricus and Streptococcus thermophilus, YF-L812, Chr. Hansen, France) according to the instructions for use, and then it was stirred for 15 minutes to ensure proper homogeneity. The inoculated milk was dispensed into 2 dl plastic yogurt cups and the cups were placed in a 43 °C thermostat (Binder, Germany), where they were allowed to curdle for 7 hours (pH 4.6). After acid curdling, the fruits treated with the following procedures were mixed into the yogurts.

In our experiment, with the exception of the control sample, two types of heat treatment procedures were used for the fruits, microwave irradiation and a conventional heat treatment method (drying).

The samples prepared in the course of the experiment:

• Sample 1: fruit yogurt with the addition of raw apples (Ny-A)
• Sample 2: fruit yogurt with the addition of raw bananas (Ny-B)
• Sample 3: fruit yogurt with the addition of dried apples (A-A)
• Sample 4: fruit yogurt with the addition of dried bananas (A-B)
• Sample 5: fruit yogurt with the addition of microwave-treated apples (MH-A)
• Sample 6: fruit yogurt with the addition of microwave-treated bananas (MH-B)

Raw fruits were processed in a clean, impeccable condition free of bruises and defective parts, at the optimum degree of ripeness.

For the microwave treatment, a MARS 5 (CEM Corporation, USA) microwave digestion oven was used. Diced raw apples and bananas were placed in the sample holder of the microwave apparatus and subjected to microwave irradiation. The energy transfer program on the device was set to heat the fruits to 55 °C when a power of 800 W with 100% efficiency and with a holding time of 10 minutes was applied, in a total of 15 minutes. Temperature detection and control was performed using a sensor (RTP 300) introduced into the sample holder.

The fruits used in the drying process were cut into pieces of equal size (0.5 cm), so that the drying time would be the same, and then they were placed in a drying apparatus for 24 hours at 55 °C. The dried and microwave-treated fruits were then ground. The knives and the grinder were treated with Mikrozid disinfectant before use. 5 grams each of the ground fruits were added to 150 ml of the already prepared yogurt samples. Until further analysis, the yogurt-dried fruit mixtures were stored in a refrigerator at 4 °C.

3.2. Product shelf-life analysis

The product were tested for 4 weeks in terms of shelf life. Analyses were carried out on days 0, 7, 14, 21 and 28. Microbiological properties (total viable count, yeast/mold count, E. coli/coliform count) were tested every week from the time of production. The experiment was carried out with 3 parallel measurements (n=3) on each sampling day, i.e., with 15 samples for each original sample (Ny-A; Ny-B, A-A; A-B; MH-A; MH-B), meaning that a total of 90 samples were processed.

A plate pouring method was used to grow the microorganisms. From food safety and technological hygiene point of view, a total viable count of 10⁵/cm³ is the critical limit for raw milk, because normal pasteurization procedures can still be used with sufficient efficiency at this microbe count.

The determination of the total viable count was carried out on a PC (Plate Count, Biolab) medium, with an incubation time of 72 hours at 30±1 °C [25]. By culturing on a selective medium prescribed in standard MSZ ISO 7954:1999 at 25 °C, yeasts and molds form colonies. YGC agar (Yeast Extract Glucose Chloramphenicol Agar, Biolab) was used for their detection, as prescribed by the standard. This selective medium is suitable for isolating and counting yeasts and filamentous fungi from milk and dairy products. Plates were incubated at 25±1 °C for 48 hours, after which the colonies developed on the plates were counted [26].

Co-determination of the coliform count and the E. coli count can be accomplished using CC agar (ChromoCULT Coliform Agar, Biolab). Differentiation between the colonies is aided by the fact that coliform colonies are salmon red, while the color of E. coli colonies ranges from dark blue to violet. Incubation parameters for E. coli/coliform were 24 hours and 35-37 °C [27].

Our measurement results were plotted using Microsoft Office Excel 2016®. During the evaluation of the microbiological results, microbe counts were displayed in a logarithmic form: the slope values of the lines fitted to each point characterize the exponential growth phase of the microorganisms.

4. Results and evaluation

4.1. Test results of the yogurt samples with apples

4.1.1. Total viable count

According to Figure 1, on days 0, 7 and 14 of the measurement, the total viable count showed almost the same results for the yogurt with dried apples and the yogurt with microwave-treated apples. The yogurt with raw apples already showed higher total viable count values in the second measurement (day 7) compared to the other two samples. Here we already saw a significant difference between the cell counts, which difference only increased over time (day 14). In the case of yogurts with microwave-treated apples and dried apples, the rates of increase in cell counts were approximately the same. This result is also supported by the slope values marked in Figure 1a.

4.1.2. Yeast/mold count

Based on Figure 2, it can be concluded that from the first measurement data to the last measurement result, the yogurts with microwave-treated apples and with dried apples showed significantly lower yeast counts than the yogurt with raw apples. There was no difference of the same order of magnitude between samples MH-A and A-A, however, on day 21 of the measurement, there was a clear, significant difference in favor of sample MH-A. Based on this, microwave heat treatment proved to be more effective in inhibiting the activity of yeasts, using the treatment settings applied by us.

In the case of yogurts with apples, it is clear that the microwave technology proved to be the best treatment for both the total viable count and the yeast/mold count. The yogurt with dried fruit exhibited similar cell counts and growth tendencies. However, at the end of the storage time, larger differences between the cell counts developed here. In terms of shelf life, the worst results were obtained for the samples with raw fruit. Differences of an order of magnitude were measured compared to the other two samples, there were significant differences (p≤0.05).

On day 21, at the third sampling time, with the exception of yogurts with microwave-treated apples, yogurts with raw and dried apples were spoiled. After the third measurement, in addition to the high yeast count, a significant mold count was also detected in the yogurts with raw or dried apples. In contrast, in yogurts with microwave-treated apples, no mold colonies could be detected after the third measurement.

For the mold count, under the current regulation [27], the level of compliance (m) is 10² CFU/cm³ for fermented milk, dairy products, sour dairy products, cottage cheese and cottage cheese products, while the rejection limit value (M) is 5*10³ CFU/cm³.

In terms of mold count, the presence of no molds was detected in yogurts with apples during the first two measurements, on days 0 and 7. On day 14 of the experiment, colonies of mold appeared in the samples with raw and dried apples, already with a value above the rejection limit as defined by the regulation in the case of raw apples (3*104 CFU/cm3). However, in the case of the product with dried apples, the number of mold colonies remained at an acceptable level (2.2*10² CFU/cm³) according to the relevant regulations [27].

4.2. Test results of the yogurt samples with bananas

4.2.1. Total viable count

In the case of yogurts with bananas, it was found that the two types of treatment procedures (drying, microwave) also have an effect on the microbial count. In the case of the microwave-treated sample, the increase in the total viable count was not significant until day 21 of the experiment compared to the initial TVC (Figure 3). On the other hand, the total viable count increased from week to week for the yogurts supplemented with raw or dried fruits. In addition, it was also found that the total viable count of the sample supplemented with dried fruit had the highest total viable count, and the most intense increase in the TVC was also observed in this sample. Already on day 7 of the storage experiment, there were significant differences between the test results of the samples, with an order of magnitude difference between sample A-B and samples Ny-B and MH-B. On day 14 of the experiment, orders of magnitude differences were observed between the date of all three samples. In terms of total viable count, the increase in TVC was the lowest in the case of sample MH-B.

4.2.2. Yeast/mold count

Based on the yeast count results (Figure 4), it was found that there was no significant difference between the yogurts with raw and microwave-treated bananas in terms of the colony counts of the samples and the growth trends of the microorganisms. However, yogurts with dried apple were characterized by a rapid increase in cell number, which also affected the organoleptic properties of the product. From day 7 of the experiment, there were already significant differences between samples MH-B and Ny-B and samples A-B and Ny-B. Microwave treatment proved to be the most effective in this case as well.

The evolution of the mold count during the shelf life was examined also in the case of yogurts with bananas. The results showed that during the first two measurements, on days 0 and 7, no mold colonies developed. However, on day 14 of the experiment, molds appeared in an amount of 1.4*10⁴ CFU/cm³ in the sample with dried bananas, a number which well exceeds the compliance limit value according to the regulation (10² CFU/cm³), moreover, it falls into the rejection category (5*10³ CFU/cm³). For the other two samples (raw and microwave-treated bananas), no molds were present on day 14. On day 21 of the experiment, molds also appeared in the yogurt supplemented with raw bananas in an amount of 3.0*10¹ CFU/cm³, which does not yet exceed the compliance limit value. Mold was still not detectable in sample MH-B. On day 28 of the experiment, the mold count of the yogurt with raw bananas also exceeded the rejection limit value by approximately 1 order of magnitude. No mold could be detected in sample MH-B even on day 28.

The results obtained for dried products suggests that the 24-hour drying with a gentle heat treatment did not sufficiently improve the microbiological condition of the materials used. It can be assumed that the hygienic condition of the air flowing through the drying apparatus was also inadequate. We believe that fruits prepared for yogurt products should only be dried in a room and equipment that has impeccable air, and have exhaust and adequate air filtration systems.

4.2.3. E. coli/coliform results of the yogurts

The bacterium Escherichia coli is the most important microbe in the normal intestinal flora, making it a natural component of the digestive system of all warm-blooded animals and humans. It can enter foods from fruits and vegetables if they had not been cleaned thoroughly enough, but it can also be found in raw milk or dairy products made from it.

Under current regulation [28], the compliance level is (m)<1/CFU/cm³ for fermented milk, dairy products, sour dairy products, cottage cheese and cottage cheese products, while the rejection limit value is (M)<10/CFU/cm³.

During the tests carried out on days 0 and 7 of the experiment, no E. coli bacteria were detected in any of the samples prepared by us. However, on day 21 (week 3) of the experiment, the bacterium became detectable in all yogurts except the samples with raw bananas and microwave-treated bananas. By week 4 of the experiment, E. coli also appeared in the yogurt with raw bananas. Thus, by the end of the study, only the yogurt supplemented with microwave-treated bananas met the legal requirements.

Coliform bacteria are found in wetlands, in soil and on the vegetation, and are usually present in large numbers in the feces of warm-blooded animals.

According to the relevant regulation (EüM decree 4/1998 (XI. 11.) – EÜM: former Ministry of Health Affairs), the compliance level is (m)<10 CFU/cm³ for fermented milk, dairy products, sour dairy products, cottage cheese and cottage cheese products, while the rejection limit value is (M)<102 CFU/cm³.

Coliform bacteria were detected in all samples on day 0 of the experiment, however, the compliance limit value was exceeded only by the results of the yogurt samples supplemented with dried fruits. After 1 week, however, coliforms could only be detected in the yogurt with dried bananas. It is likely that the decrease in the pH value of the yogurt prevented the bacteria from growing and surviving.

At week 3 of the experiment, coliform bacteria were detected in the samples supplemented with raw fruits, they were not present in the other samples. In our case, the samples supplemented with raw fruits reached the M value, so after 21 days the products were not suitable for human consumption.

During the microbiological studies, the determination of Salmonella and Staphylococcus aureus was also performed, as required by the regulation. These tests were negative in all cases.

Schnabel et al. infected raw fruits with seven microbial strains (including the E. coli bacterium also tested by us) with a cell count in the 10⁸ order of magnitude. The samples were then treated with microwave-assisted plasma, which reduced the cell count by 4 orders of magnitude already after 5 minutes of treatment. The treatment was performed under non-thermal conditions (at 30 °C), thus excluding the microbicidal effect of the temperature [29].

Picouet et al. showed that microwave treatments had a similar effect on the E. coli O157: H7 and total viable count values, i.e., a 1.01-1.16 log CFU g^-1 decrease was detected. The same treatment parameters greatly affected L. innocua, with population below the detection limit (10 CFU g^-1) in most cases. In apple puree samples, the total viable count remained stable during storage at 5 °C, with a slight increase on day 14 [30]. This trend was also observed during our own measurements. Our results confirm that the objective of our research was achieved, which was to verify the microbicidal and inhibitory effect of microwave treatment.

Our results are also supported by the fact that 5 to 25 seconds of microwave treatment (65 °C, 1200 W, 2.45 GHz) can reduce the Salmonella cell count in vegetables by 4 to 5 orders of magnitude, thus confirming the stronger microbicidal effect of microwave treatment compared to other ones [31].

5. Conclusions and recommendations

When using fruits as flavoring agents in yogurts, two types of heat treatment were applied to increase the shelf life of the products. The effect of the microwave treatment method on shelf life was characterized by the length of shelf life after the addition of conventional and untreated fruits to yogurt.

Based on the microbiological studies, it was found that microwave treatment was the most effective of the various heat treatments. Of the methods of heat treatment of fruits, microwave irradiation resulted in a lower total viable count compared to untreated fruits and drying technology.

Based on our microbiological results, we believe that the contamination or texture of the raw fruit fundamentally influences the effectiveness of the treatment. After microwave treatment of bananas, the presence of E. coli could not be detected in the yogurt by the end of our experiments (day 28), as opposed to the treated apples, where its presence was already detected on day 14. This may be noteworthy because the presence of E. coli was not detected in either case on the day the products were prepared. It can be assumed that microwave irradiation exerted a more intense germicidal effect in the softer texture of bananas than in apples which have a harder consistency.

It was found that the drying procedure is suitable for the production of microbiologically safe food if the microbiological condition of the air circulating in the equipment is also adequate.

Based o our results, it is hypothesized that microwave irradiation technology can be applied successfully to foods, in this case fruits, to inhibit microorganisms living inside and on the surface of fruits.

While in the case of yogurts flavored with apples, the microbiological characteristics of samples with raw fruit were worse, in the case of bananas, the drying technology proved to be the most unfavorable from a microbiological point of view. The most likely reason for this may be that, compared to apples, bananas contain on average three times more carbohydrates which became more concentrated as the result of drying. This high carbohydrate fruit mixed with the yogurt may have served as a culture medium for various microorganisms.

In order to determine whether or not the use of different microwave temperatures (other than 55 °C), power and treatment times would lead to better shelf-life results, further tests are required.

Of the heat treatment procedures, microwave may be suitable both for treating milk, thus reducing the number of microbes, and for reducing the total viable count of the flavoring agents (spices, fruits, vegetables) used.

6. Acknowledgment

This publication was supported by project no. EFOP-3.6.1-16-2016-00024 titled „Developments for intelligent specialization in cooperation between the University of Veterinary Medicine and the Faculty of Agricultural and Food Sciences of Széchenyi István University. The project was supported by the European Union and co-financed by the European Social Fund.

7. References

[1] Kukovics, S. (2009): A tej szerepe a humán táplálkozásban. Melánia Kiadó, Budapest.

[2] Coïsson, J.D., F. Travaglia, G. Piana, M. Capasso and M. Arlorio, (2005): Euterpe oleracea juice as a functional pigment for yogurt. Food Research International, 38 (8-9) pp. 893-897. https://doi.org/10.1016/j.foodres.2005.03.009

[3] Tamime, A.Y., Robinson. R.K. (1999): Yoghurt, Science and Technology. Cambridge, UK: Woodhead Publishing Limited. https://doi.org/10.1201/9780415876162

[4] Tarakci, Z., Kücüköner, E. (2003): Physical, chemical, microbiological and sensory characteristics of some fruit flavored yoghurt. Yüzüncü Yıl Üniversitesi Veteriner Fakültesi Dergisi 14 (2) pp. 10-14.

[5] Deák, T. (2006): Élelmiszer mikrobiológia. Mezőgazda Kiadó, Budapest.

[6] Oms-Oliu G., Rojas-Grau M.A., Gonzalez L.A., Varela P., Soliva-Fortuny R., Hernando M.I.H., Munuera I.P., Fiszman S., & Martin-Belloso, O. (2010): Recent approaches using chemical treatments to preserve quality of fresh-cut fruit: A review. Postharvest Biology and Technology 57 (3) pp. 139-148. https://doi.org/10.1016/j.postharvbio.2010.04.001

[7] Pasha, I., Saeed, F., Sultan, M.T., Khan, M. R., & Rohi, M. (2014): Recent developments in minimal processing: A tool to retain nutritional quality of food. Critical Reviews in Food Science and Nutrition 54 (3) pp. 340-351. https://doi.org/10.1080/10408398.2011.585254

[8] Abadias, M., Usall, J., Anguera, M., Solsona, C., & Viñas, I. (2008): Microbiological quality of fresh, minimally-processed fruit and vegetables, and sprouts from retail establishments. International Journal of Food Microbiology, 123 (1-2) pp. 121-129. https://doi.org/10.1016/j.ijfoodmicro.2007.12.013

[9] Johnston, L. M., Jaykus, L. A., Moll, D., Martinez, M. C., Anciso, J., Mora, B., & Moe, C. L. (2005): A field of study of the microbiological quality of fresh produce. Journal of Food Protection 68 (9) pp. 1840-1847. https://doi.org/10.4315/0362-028X-68.9.1840

[10] Lehto, M., Kuisma, R., Maatta, J., Kymalainen, H. R., & Maki, M. (2011). Hygienic level and surface contamination in fresh-cut vegetable production plants. Food Control 22 (3-4) pp. 469-475. https://doi.org/10.1016/j.foodcont.2010.09.029

[11] Beuchat, L. R. (1998): Progress in conventional methods for detection and enumeration of foodborne yeasts. Food Technology and Biotechnology 36 (4) pp. 267-272.

[12] Magyar Élelmiszerkönyv (Codex Alimentarius Hungaricus): 2-51 számú irányelv, Tej és tejtermékek, Dairy products

[13] Vasbinder, A. J.; C. G. De Kruif. (2003): Casein-whey protein interactions in heated milk: the influence of pH. International Dairy Journal 13 (8) pp. 669-677. https://doi.org/10.1016/S0958-6946(03)00120-1

[14] Berecz, L. (1999): Élelmiszerek száradási jellemzői, különös tekintettel az élesztőkre, PhD disszertáció, Pannon Agrártudományi Egyetem, Mezőgazdaságtudományi Kar, Mosonmagyaróvár

[15] Tang, Z. W., Mikhaylenko, G., Liu, F., Mah, J.H., Pandit, R., Younce, F., Tang, J.M., (2008): Microwave sterilization of sliced beef in gravy in 7-oz trays. Journal of Food Engineering 89 (4) pp. 375-383. https://doi.org/10.1016/j.jfoodeng.2008.04.025

[16] Venkatesh, M.S., Raghavan, G.S.V., (2004): An overview of microwave processing and dielectric properties of agri-food materials. Biosystems Engineering 88 (1) pp. 1-18. https://doi.org/10.1016/j.biosystemseng.2004.01.007

[17] Pozar, M.D. (1993): Microwave Engineering, Addison-Wesley Publishing Company.

[18] Rosenberg, U., Bogl, W. (1987): Microwave pasteurization, sterilization, blanching, and pest control in the food industry. Food Technology 41 (6) pp. 92-99.

[19] Lau, M.H., Tang, J. (2002): Pasteurization of pickled asparagus using 915 MHz microwaves. Journal of Food Engineering 51 (4) pp. 283-290. https://doi.org/10.1016/S0260-8774(01)00069-3

[20] Wang, Y., Wig, T.D., Tang, J., Hallberg, L.M. (2003): Dielectric properties of foods relevant to RF and microwave pasteurization and sterilization. Journal of Food Engineering 57 (3) pp. 257-268 https://doi.org/10.1016/S0260-8774(02)00306-0

[21] Bennett, L.E., Jegasothy, H., Konczak, I., Frank, D., Sudharmarajan, S., Clingeleffer, P.R. (2011): Total polyphenolics and anti-oxidant properties of selected dried fruits and relationships to drying conditions. Journal of Functional Foods 3 (2) pp. 115-124. https://doi.org/10.1016/j.jff.2011.03.005

[22] Donno, D., Beccaro, G.L., Mellano, M.G., Cerutti, A.K., & Bounous G. (2015): Goji berry fruit (Lycium spp.): antioxidant compound fingerprint and bioactivity evaluation. Journal of Functional Foods 18 (B) pp. 1070-1085. https://doi.org/10.1016/j.jff.2014.05.020

[23] Hassan, A.N., Ipsen, R., Janzen, T., Qvist, K.B., (2003): Microstructure and rheology of yogurt made with cultures differing only in their ability to produce exopolysaccharides. Journal of Dairy Science 86 (5) pp. 1632-1638. https://doi.org/10.3168/jds.S0022-0302(03)73748-5

[24] Huang, L., Lu, Z., Yuan, Y., Lu, F., Bie, X., (2006): Optimization of a protective medium for enhancing the viability of freeze-dried Lactobacillus delbrueckii subsp. bulgaricus based on response surface methodology. Journal of Industrial Microbiology and Biotechnology 33 (1) pp. 55-61. https://doi.org/10.1007/s10295-005-0041-8

[25] Magyar Szabványügyi Testület (MSzT) (2014): Az élelmiszerlánc mikrobiológiája. Horizontális módszer a mikroorganizmusok számlálására. 1. rész: Telepszámlálás 30 °C-on lemezöntéses módszerrel Magyar Szabvány MSZ EN ISO 4833-1:2014. Magyar Szabványügyi Testület, Budapest.

[26] Magyar Szabványügyi Testület (MSzT) (1999): Mikrobiológia. Általános útmutató élesztők és penészek számlálásához. Telepszámlálási technika 25 °C-on. Magyar Szabvány MSZ ISO 7954:1999. Magyar Szabványügyi Testület, Budapest.

[27] Magyar Szabványügyi Testület (MSzT) (2014): Az Escherichia coli és coliform baktériumok megszámlálása. 2. rész: A legvalószínűbb szám módszere (EN ISO 9308-2:2014). Angol Szabvány MSZ EN ISO 9308-2:2014. Magyar Szabványügyi Testület, Budapest.

[28] 4/1998. (XI. 11.) EüM rendelet: Az élelmiszerekben előforduló mikrobiológiai szennyeződések megengedhető mértékéről (hatályos: 2020. 04. 02.)

[29] Schnabel, U., Niquet , R., Schlüter, O., Gniffke, H.; Ehlbeck, J. (2014): Decontamination and sensory properties of microbiologically contaminated fresh fruits and vegetables by microwave plasma processed air (PPA). Journal of Food Processing and Preservation 29 (6) pp. 1745-4549. https://doi.org/10.1111/jfpp.12273

[30] Picouet, P. A., Landl, A., Abadias, M., Castellari, M., Viñas, I. (2009): Minimal processing of a Granny Smith apple purée by microwave heating, Innovative Food Science and Emerging Technologies 10 (2009) pp. 545-550. https://doi.org/10.1016/j.ifset.2009.05.007

[31] Vega, M., López, S., Malo, L., Morales, S. (2012): Inactivation of Salmonella typhimurium in fresh vegetables using water-assisted microwave heating. Food Control 26 (1) pp. 19-22. https://doi.org/10.1016/j.foodcont.2012.01.002

DOI: https://doi.org/10.52091/EVIK-2021/2-1-ENG

Received: February 2021 – Accepted: April 2021

^a University of Debrecen, Faculty of Agricultural and Food Sciences and Environmental Management
^b BIOMI Kft.
^c Pázmány Péter Catholic University
^d WESSLING Hungary Kft.
^e University of Veterinary Medicine
^* Corresponding author: horvath.brigitta920108@gmail.com

Keywords

food, Staphylococcus, antibiotic resistance, MALDI-TOF- MS, bacterial identification, food safety, human pathogen, nosocomial infection

2. Introduction and literature review

The number of nosocomial infections caused by antibiotic-resistant microorganisms has been increasing in all countries, thus posing a greater and greater challenge to the health care system [1, 2]. The situation is further exacerbated by the fact that antibiotic-resistant Staphylococcus species have already appeared not only in communities and health care, but also in intensive animal husbandry and thus in the food chain [3].

In Staphylococcus species, genes associated with antibiotic resistance and virulence are found in the mobile genetic elements (MGE), such as chromosome cassettes, pathogenicity islands, plasmids or transposons [4]. The mecA gene is responsible for methicillin resistance: the gene encodes a modified penicillin-binding protein that reduces the binding affinity of most beta-lactam antibiotics, such as penicillin and methicillin. The mecA gene is located on the Staphylococcus chromosome cassette (SSCmec), which is a group of MGE found only in Staphylococcus species [5]. The transfer mechanism of the mecA gene between Staphylococcus species is unknown, however, evidence supports horizontal gene transfer between coagulase-positive and coagulase-negative Staphylococcus species [6].

The presence of methicillin-resistant Staphylococcus aureus (MRSA) strains in the food chain has already been reported in several studies. Some of their authors examined strains isolated from food samples of animal origin, others investigated strains isolated from raw meat samples (pig, fish, poultry). In the Netherlands in 2009, 2,217 different food samples were analyzed, 12% of which contained MRSA strains [7], while in a Danish study 4.6% of 153 pork samples and 7.5% of imported pork samples were infected with MRSA strains [8]. MRSA strains have also been identified in Germany in raw milk, pork, turkey and broiler chicken [9]. In Hungary, 27 MRSA isolates were identified in the 595 individual milk samples of a dairy farm [10], while in another study only 4 strains out of the 626 S. aureus isolates of 42 farms proved to be methicillin-resistant [11]. However, beyond these examples, the presence of MRSA in foods from other categories and ready-to-eat foods has not been investigated so far. Molecular typing results of the strains have shown that many types of MRSA are present in the food chain in different countries [12], but the most common type is CC398, accounting for 85% of the MRSA strains isolated from farm animals in the EU and the EEA [13, 14, 7,15].

The presence of additional methicillin-resistant Staphylococcus (MRS) species in foods has been investigated by fewer studies. In Nigeria, 13 Staphylococcus species (S. xylosus, S. epidermidis, S. simulans) showed methicillin resistance out of 255 isolates from traditional foods [16]. In a study in Poland, out of 58 strains isolated from ready-to-eat foods, 33 Staphylococcus strains (S. epidermidis, S. simulans, S. xylosus, S. hycus, S. lentus, S. saprophyticus) showed resistance to at least one type of antibiotic [17].

In the European Union, testing for antibiotic resistance in Staphylococcus strains isolated from foods and farm animals is currently voluntary, so in 2016 only Germany, Switzerland, Denmark and Spain reported information related to this topic. The incidence of MRSA varied from country to country, but in the case of a comparison it should be taken into account that the studies were performed on strains isolated from different animal species, meats and meat products [18]. A small proportion of human infections can be traced back to MRSA strains of the CC398 type, and they are also mainly limited to occupational exposures, such as veterinary medicine and intensive animal husbandry. Nevertheless, the virulence factors detectable in CC398-type MRSA strains allow for high pathogenicity, so continuous revision is essential both in animals and in foods [19]. The need for surveillance is also justified by the possible presence of antibiotic resistance of other Staphylococcus species, which allows for the spread of resistance and poses a risk to consumer health.

A prerequisite for a successful surveillance system is the availability of a uniform, economically acceptable, rapid and reliable method for the species-level identification of the microorganisms and the determination of antibiotic resistance, a promising cornerstone of which could be matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS) based on peptide identification. In the early 2000s, several studies reported specific fragment ions that allow rapid identification of antibiotic-resistant Staphylococcus strains. The most often studied biomarker was the fragment ion with an m/z value of 2,414, the appearance of which in the mass spectrum correlates with the expression of psm-mec, characteristic of MRSA strains [20]. The applicability of the 2,414 m/z fragment ion for detection and discrimination has been demonstrated in several studies [21, 22].

In addition to the studies of the biomarkers of MRSA strains, methicillin resistance-specific fragment ion peaks of other Staphylococcus species have been analyzed in other studies. In an earlier article, two specific fragment ion values were determined: the ion fragment peak with an m/z value of 7,239, which is a biomarker of methicillin-resistant S. epidermidis, and the fragment ion peak with an m/z value of 9,674, which is the biomarker of methicillin-resistant S. haemolyticus [23].

The objective of our own experiments was to investigate the methicillin resistance of Staphylococcus strains isolated from foods using classical microbiological, molecular biological methods and the MALDI-TOF-MS technique, as well as the multi-locus sequence typing (MLST) of antibiotic-resistant strains for the epidemiological study of the strains.

3. Materials and methods

3.1. The isolates collected and culture conditions

77 Staphylococcus isolates, isolated according to the requirements of standard MSZ EN ISO 6888-1:2008, were analyzed in the Microbiological Laboratory of WESSLING Hungary Kft. in the period between August 2019 and September 2020. The isolates were grown on a Baird-Parker (Biokar, France) selective culture medium at 37 °C for 24±1 hour and colonies characteristic of Staphylococci were inoculated onto Columbia blood agar (Neogen, UK) (37 °C, 24±1 hours). In the course of culturing, 47 strains showed a positive coagulase reaction, while 30 isolates proved to be coagulase-negative, which was also confirmed by a latex agglutination rapid test (PASTOREX™ STAPH-PLUS). Isolates were collected from raw meat, meat products and ready-to-eat foods: poultry (n=14), beef (n=5), pork (n=42), game (n=1), fish (n=1), dairy product (n=3), ready-to-eat foods (n=3), vegetables (n=2) and dry pasta (n=6).

3.2. Identification of the isolates using the MALDI-TOF-MS technique

The 77 isolates collected were identified using a Bruker Microflex LT MALDI-TOF mass spectrometer and the MALDI BioTyper 3.1 (Bruker Daltonics) software. A formic acid suspension protocol was used in which a single colony was collected from the Columbia blood agar using a sterile loop, and then it was suspended in 40 µl formic acid. To the suspension was added 40 µl of acetonitrile, 1 μl of which was applied to one of the positions of the plate. After the solvent evaporated, 1 μl of α-HCCA (10 mg/ml α-Cyano-4-hydroxycinnamic acid) matrix solution was applied and the solvent was allowed to evaporate once more. 6 parallel measurements were performed for each sample.

To identify the isolates, the MALDI Biotyper 3.1 software was used, which compares the mass spectra obtained with the reference mass spectra in its database and calculates a compliance factor (score). In case of a log score value of 2.300 – 3.000, identity is highly probable. At a log score value this high, the species is considered to be identified. If the log score value is between 2.000 and 2.299, the identity is less certain, so in this case only the genus of the microorganism can be considered identified. When the log score value is between 1.700 and 1.999, even the identification of the genus cannot be considered sufficiently certain. If a log score value between 0.000 and 1.699 is returned by the evaluation software, identification should be considered unsuccessful. Identification of the coagulase-positive Staphylococcus strains included in the study was carried out earlier [24].

3.3. Antibiotic susceptibility testing

3.3.1. Investigation of the methicillin resistance specific peaks by the MALDI-TOF-MS method

The mass spectra obtained were exported to the flexAnalysis 3.4 software (Bruker Daltonics) and manual analysis and comparison of the mass spectra was performed. Smoothing of the mass spectra was carried out with the Savitzky–Golay filter, while baseline correction was performed using the TopHat algorithm. During the analysis the presence of methicillin resistant-specific fragment ion values was examined (Table 1).

3.3.2. Disk diffusion method

When examining the antibiotic resistance of the strains, the guidelines of the CLSI (Clinical and Laboratory Standards Institute, 2019) were followed [25]. A bacterial suspension equivalent to 0.5 McFarland unit was applied to the surface of a Mueller-Hinton agar (Oxoid, UK), and then Cefoxitin 30 μg disks were placed on the surface of the medium. The strains were incubated at 37 °C for 18 hours. In the case of MRSA strains, the reference range of the clearance zone was 6-19 mm volt, while for mecA negative species it was 20-32 mm.

3.3.3. Selective differentiation agar

In addition to the above, CHROMagar MRSAII selective differentiation medium (BD, UK) was used for the detection of methicillin-resistant Staphylococcus aureus species in the antibiotic resistance assays. Isolates were incubated at 37 °C for 24 – 48 hours under aerobic conditions. Strains that formed mauve colonies, morphologically similar to those of Staphylococci, were considered MSRA bacteria. The disk diffusion method (Cefoxitin 30 µg) and the MRSA CHROMagar assay were repeated twice for each strain isolated from a food (n=77). During the study, the reference MRSA strain ATCC 33591 was used as a positive control and the MRSA strain ATCC 29213 was used as a negative control.

3.3.4. mecA gene complex

Detection of the mecA gene was performed according to the protocol of the Danish National Food Institute (NFI), published in 2012 [26]. During the study, MRSA strain ATCC 43300 was used as a positive control and MRSA strain ATCC 29213 was used as a negative control. Genomic DNA was isolated from the bacteria and the mecA gene sequence was amplified using PCR. The primers used are listed in Table 2.

3.4. MLST study of methicillin-resistant Staphylococcus strains

According to the study of Thomas et al. [27], genomic DNA was isolated from the bacteria, and then the gene sequences specific for the 7 Staphylococcus aureus species were amplified using PCR (Table 3). The nucleotide sequences of the purified PCR products were determined and the sequence data were evaluated in the BioNumerics 7.6 software.

4. Results

4.1. Identification results of the isolated strains

In the MALDI-TOF-MS study, all coagulase-positive Staphylococcus (CPS) strains (n=47) were found to be S. aureus species (Tables 4 and 6). In the case of coagulase-negative Staphylococcus (CNS) strains, 8 different species (S. xylosus, S. saprophyticus, S. pasteuri, S. epidermidis, S. warneri, S. chromogenes, S. piscifermentans, S. haemolyticus) were identified (Tables 4 and 7). 30% of the CNS isolates (n=30) were found to be S. warneri species, while 23% were found to be S. pasteuri species. 70% of the S. aureus strains and all of the CNS strains were isolated from meat and meat products (Figures 1 and 2). In the case of S. aureus strains, 64% of the meats and meat products came from pigs, while this was true for 70% of the CNS strains. Within meat and meat products, the distribution of isolates coming from poultry and beef was nearly the same.

The mean identification log score values of the isolates and their standard deviations are summarized in Table 4. The mean identification log score value of S. aureus isolates exceeded 2.400. The lowest log score value was 2.304, and even in this case, identification can be considered safe. When examining CNS isolates, each species was identified with a log score value above 2.300 and the standard deviation did not exceed 0.1 in any of the cases.

4.2. Methicillin resistance results determined by the MALDI-TOF-MS technique

In the analysis of the mass spectra obtained from the isolates, 3 antibiotic resistance-specific peaks were examined. The m/z 2,414 peak is a protein product of the mecA gene [28], so the presence or absence of this peak was examined for all strains. The detectability of the m/z 7,239 peak was examined only in S. epidermidis species, while the presence/absence of the m/z 9,674 peak was examined only in S. haemolyticus species due to the species specificity of the peaks.

The m/z 2,414 peak was detected in two S. aureus strains out of 77 isolates, one of which came from goose liver (SA-17), while the other came from pork butt (SA-47). For the additional 75 isolates, this peak did not appear even at a low intensity (Table 5). In the analysis, mass spectra of the two S. aureus strains that proved to be methicillin-resistant were marked in red, while the mass spectra of the other S. aureus strains that did not have a methicillin resistance-specific peak were marked in black (Figures 3 and 4).

The m/z 7,239 peak could not be detected in any of the S. epidermidis strains (n=4), while the m/z 9,674 peak did not appear in any of the 4 S. haemolyticus species.

4.3. Results of the disk diffusion method and the MRSA CHROMagar selective differentiation agar

In the study of the 77 strains, the diameter of the clearance zone ranged from 23 to 29 mm in the case of 75 strains. The strain isolated from goose liver (SA-17) had a clearance zone with a diameter of 9 mm, while the strain isolated from pork butt (SA-47) had a clearance zone with a diameter of 17 mm, and the same strains also formed mauve-colored colonies on the MRSA CHROMagar selective differentiation medium (Tables 6 and 7).

4.4. mecA gene detection results

Based on the results of the MALDI-TOF-MS analyses, the disk diffusion method and the MRSA selective differentiation medium, it was found that the S. aureus strains isolated from goose liver and pork butt (SA-17, SA-47) carry methicillin resistance, and this was confirmed by the mecA gene responsible for PBP2a synthesis, which could also be detected in the two strains (Table 6).

4.5. MLST types of the MRSA strains

MLST typing of the two MRSA strains was also performed during the study, using data available on the PubMLST website (https://pubmlst.org/saureus/). Of the two MRSA strains, the BioNumerics 7.6 software could only assign the sequence type in the case of the strain isolated from pork butt. The strain isolated from goose liver belonged to a sequence type not yet known, while the strain isolated from pork butt belonged to sequence type 398 (Table 8).

5. Summary and conclusions

During the study, 77 Staphylococcus isolates were collected from various food matrices according to the methods described in standard MSZ EN ISO 6888-1:2008. Although the standard allows the separation of coagulase-positive and coagulase-negative Staphylococcus species, it does not allow species identification, which is significant because of the different virulence factors and pathogenicities of the species. Identification of the 47 coagulase positive and 30 coagulase-negative Staphylococcus strains isolated from foods was performed using the MALDI-TOF-MS technique, with high identification log score values. In addition, by analyzing the mass spectra obtained and based on methicillin resistant-specific ion fragment values determined in previous studies, methicillin resistance was found in two S. aureus strains, which was confirmed by the disk diffusion method, selective differentiation agar and the detection of the mecA gene. Thanks to the specific ion fragment values, the diagnostic time can be significantly reduced, which is not negligible from economic and therapeutic points of view. However, it should be taken into consideration that due to the high variability of MRSA strains, the sensitivity and specificity of these ion fragment values are not 100%, so confirmatory studies are required.

Multi-locus sequence typing (MLST) of the two MRSA strains isolated from foods was also carried out for the epidemiological study of the strains. The isolate coming from pork belonged to type ST398, which is the most common type of MRSA strain isolated from farm animals in the EU/EEA. However, taking into account the specific host adaptation capabilities of type ST398 strains, which allow them to adhere not only to pigs, but also to other animal species and the human body, contamination and infection may occur in a number of ways during the technological steps in food processing.

Given that 2 of the 47 S. aureus strains isolated from foods proved to be methicillin-resistant, this fact confirms the dangers posed by globally increasing antibiotic resistance, thus indicating the severity and urgency of the situation.

6. Acknowledgment

This study was carried out with the professional support of WESSLING Hungary Kft., BIOMI Kft. and the New National Excellence Program of the Ministry for Innovation and Technology, code no. ÚNKP-19-3-III.

7. Irodalom

[1] Böröcz, K. (2001): A magyarországi nosocomialis MRSA járványok tapasztalatai (1993-2000). Epidemiológiai Információs Hetilap. 8. (10-11).

[2] Böröcz, K. (2005): Az Országos tisztifőorvos állásfoglalása a methicillin-rezisztens Staphylococcus aureus (MRSA) törzsek által okozott, egészségügyi ellátással összefüggő fertőzések megelőzésükről és terjedésük megakadályozásáról. Epidemiológiai Információs Hetilap. 12. (5).

[3] Jungwhan, C., Kidon, S., Saeed, K. (2017): Methicillin-Resistant Staphylococcus aureus (MRSA) in Food-Producing and Companion Animals and Food Products. Frontiers in Staphylococcus aureus. Intechopen.

[4] Lim, D., Strynadka, N. C. J. (2002): Structural basis for the β-lactam resistance of PBP2a from metichillin-resistant Staphylococcus aureus. Nature Structural Biology. 9 (11), pp. 870-876. https://doi.org/10.1038/nsb858

[5] Ito, T., Katayama, Y., Hiramatsu, K. (1999): Cloning and nucleotide sequence determination of the entire mec DNA of pre-methicillin-resistant Staphylococcus aureus N315. Antimicrob Agents Chemother. 43 (6), pp. 1449-58. https://doi.org/10.1128/AAC.43.6.1449

[6] Wielders, C. L., Vriens, M. R., Brisse, S., De Graaf-Miltenburg, L. A., Troelstra, A., Fleer, A., Schmitz, F. J., Verhoef, J., Fluit, A. C. (2001): In-vivo transfer of mecA DNA to Staphylococcus aureus [corrected]. Lancet. 26; 357 (9269), pp. 1674-1675. https://doi.org/10.1016/S0140-6736(00)04832-7

[7] Köck, R., Harlizius, J., Bressan, N., Laerberg, R., Wieler, L. H., Witte, W., Deurenberg, R. H., Voss, A., Becker, K., Friedrich, A. W. (2009): Prevalence and Molecular Characteristics of Methicillin-Resistant Staphylococcus Aureus (MRSA) Among Pigs on German Farms and Import of Livestock-Related MRSA Into Hospitals. European Journal of Clinical Microbiology & Infectious Diseases. 28 (11), pp. 1375-1382. https://doi.org/10.1007/s10096-009-0795-4

[8] Agersø, Y., Hasman, H., Cavaco, L. M., Pedersen, K., Aarestrup, F. M. (2012): Study of methicillin-resistant Staphylococcus aureus (MRSA) in Danish pigs at slaughter and in imported retail meat reveals a novel MRSA type in slaughter pigs. Veterinary Microbiology. 157 (1-2), pp. 246-250. https://doi.org/10.1016/j.vetmic.2011.12.023

[9] Argudín, M. A., Tenhagen, B. A., Fetsch, A., Sachsenröder, J., Käsbohrer, A. (2011): Virulence and resistance determinants of German Staphylococcus aureus ST398 isolates from nonhuman sources. Applied and Environmental Microbiology. 77 (9), pp. 3052-3060 https://doi.org/10.1128/AEM.02260-10

[10] Juhász-Kaszanyitzky, E., Jánosi, S., Somogyi, P., Dán, A., Van Der, A., Graaf-Van Bloois, L. (2007): MRSA transmission between cows and humans. Emerging Infectious Diseases. 13 (4), pp. 630-632. https://doi.org/10.3201/eid1304.060833

[11] Albert, E., Sipos, R., Jánosi, Sz., Kovács, P., Kenéz, Á., Micsinai, A., Noszály, Zs., Biksi, I. (2020): Occurrence and characterisation of methicillin-resistant Staphylococcus aureus isolated from bovine milk in Hungary. Acta Veterinaria Hungarica. 68 (3) pp. 236-241. https://doi.org/10.1556/004.2020.00040

[12] Wendlandt, S., Schwarz, S., Silley, P. (2013): Methicillin-Resistant Staphylococcus aureus: A Food-Borne Pathogen? Annual Review of Food Science and Technology. 4, pp. 117-139. https://doi.org/10.1146/annurev-food-030212-182653

[13] Bosch, T., Verkade, E., Van Luit, M., Landman, F., Kluytmans, J., Schouls, L. M. (2015): Transmission andpersistence of livestock-associated methicillin-resistant Staphylococcus aureus among veterinarians and their household members. Applied and Environmental Microbiology. 81 (1), pp. 124-129. https://doi.org/10.1128/AEM.02803-14

[14] Fluit, A.C. (2012): Livestock-associated Staphylococcus aureus. Clinical Microbiology and Infection. 18 (8), pp. 735-744. https://doi.org/10.1111/j.1469-0691.2012.03846.x

[15] Verkade, E., Van Benthem, B., Den Bergh, M. K., Van Cleef, B., Van Rijen, M., Bosch, T., Kluytmans, J. (2013): Dynamics and determinants of Staphylococcus aureus carriage in livestock veterinarians: a prospective cohort study. Clinical Infectious Diseases. 57 (2), pp. 11-17. https://doi.org/10.1093/cid/cit228

[16] Fowoyo, P. T. - Ogunbanwo, S. T. (2017): Antimicrobial resistance in coagulase-negative staphylococci from Nigerian traditional fermented foods. Annals of Clinical Microbiology and Antimicrobials, 16 (4), pp. 2-7. https://doi.org/10.1186/s12941-017-0181-5

[17] Chaje, W., Zadernowska, A. C.-W., Nalepa, B., Sierpinska, M., - Łaniewska-Trokenheim, L. (2015): Coagulase-negative staphylococci (CoNS) isolated from ready-to-eat food of animal origin e Phenotypic and genotypic antibiotic resistance. Food Microbiology, 46, pp. 222-226. https://doi.org/10.1016/j.fm.2014.08.001

[18] European Food Safety Authority - EFSA (2018): The European Union summary report on antimicrobial resistance in zoonotic and indicator bacteria from humans, animals and food in 2016. EFSA journal. 16 (2), pp. 5182. https://doi.org/10.2903/j.efsa.2018.5182

[19] Aires-De-Sousa, M. (2016): Methicillin-resistant Staphylococcus aureus among animals: current overview. Clinical Microbiology and Infection. 23, pp. 373-380. https://doi.org/10.1016/j.cmi.2016.11.002

[20] Josten, M., Dischinger, J,. Szekat, C., Reif, M., Al-Sabti, N., Sahl, H. G., Parcina, M., Bekeredjian-Ding, I, Bierbaum, G. (2014): Identification of Agr-Positive Methicillin-Resistant Staphylococcus Aureus Harbouring the Class A Mec Complex by MALDI-TOF Mass Spectrometry. International Journal of Medical Microbiology. 304 (8), pp. 1018-1023. https://doi.org/10.1016/j.ijmm.2014.07.005

[21] Alksne, L., Makarova, S., Avsejenko, J., Cibrovska, A., Trofimova, J., Valciņa, O. (2020): Determination of methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis by MALDI-TOF-MS in clinical isolates from Latvia. Clinical Mass Spectrometry. 16, pp. 33-39. https://doi.org/10.1016/j.clinms.2020.03.001

[22] Pranada, A. B. - Bienia, M. - Kostrzewa, M. (2016): Optimization and Evaluation of MRSA Detection by Peak Analysis of MALDI-TOF Mass Spectra, in DGHM 2016 https://www.msacl.org/view_abstract/MSACL_2017_EU.php?id=305 Hozzáférés: 2021.01.08.

[23] Manukumar, H. M., Umesha, S. (2017): MALDI-TOF-MS based identification and molecular characterization of food associated methicillin-resistant Staphylococcus aureus. Scientific Reports. 7, 11414 pp. 1-16. https://doi.org/10.1038/s41598-017-11597-z

[24] Horvath, B., Peles, F., Szél, A., Sipos, R., Erős, Á., Albert, E., Micsinai, A. (2020): Molecular typing of foodborne coagulase-positive Staphylococcus isolates identified by MALDI-TOF-MS. Acta Alimentaria, An International Journal of Food Science. 49 (3), pp. 307-313. https://doi.org/10.1556/066.2020.49.3.9

[25] Clinical and Laboratory Standards Institute - CLSI (2019): Performance Standards for Antimicrobial Susceptibility Testing: Eighteenth Informational Supplement M100- S18. Wayne, PA, USA: CLSI.

[26] National Food Institute - NFI (2012): protocol for pcr amplification of meca, mecc (mecalga251), spa and pvl recommended by the eurl-ar 2st version. https://www.eurl-ar.eu/CustomerData/Files/Folders/21-protocols/279_pcr-spa-pvl-meca-mecc-sept12.pdf Hozzáférés: 2021.02.03.

[27] Thomas, J. C., Vargas, M. R., Miragaia, M., Peacock, S. J., Archer, G. L., Enright, M. (2007): Improved Multilocus Sequence Typing Scheme for Staphylococcus epidermidis. Journal of Clinical Microbiology. 45 (2), pp. 616-619. https://doi.org/10.1128/JCM.01934-06

[28] Josten, M., Dischinger, J,. Szekat, C., Reif, M., Al-Sabti, N., Sahl, H. G., Parcina, M., Bekeredjian-Ding, I, Bierbaum, G. (2014): Identification of Agr-Positive Methicillin-Resistant Staphylococcus Aureus Harbouring the Class A Mec Complex by MALDI-TOF Mass Spectrometry. International Journal of Medical Microbiology. 304 (8), pp. 1018-1023. https://doi.org/10.1016/j.ijmm.2014.07.005

DOI: https://doi.org/10.52091/EVIK-2021/2-2-ENG

Received: March 2021 – Accepted: May 2021

¹ National Food Chain Safety Office, System Management and Supervision Directorate
² University of Veterinary Medicine, Digital Food Chain Education, Research, Development and Innovation Institute
³ University of Debrecen, Doctoral School of Nutrition and Food Sciences

Keywords

deterministic exposure estimation, chronic aflatoxin M1 exposure, consumer groups at risk, carcinogenic effect, risk of liver cancer

2. Introduction

In our previous paper on mycotoxin contamination, we presented the mycotoxin contamination of foods and feeds, the legal regulation of their tolerable maximum concentrations, the limitations of sampling procedures, and the experiences of current domestic practice were analyzed [1]. Using the data of the food consumption survey conducted in 2009 and measurement results available from the period 2010-2018, the exposure of Hungarian consumers to DON and aflatoxin M1 was estimated. Based on our preliminary estimates, it was determined that for some consumers, exposure to aflatoxin M1 and DON may exceed the toxicological reference values from time to time due to the yearly variation in food contamination levels, which may pose a risk to human health.

In our present paper, the results of our calculations carried out by a deterministic method and using the latest analytical measurement results and the data of the Hungarian food consumption survey performed in 2018-2019 using the methodology uniformly applied in Europe are presented, which provides information on chronic AFM1 exposure of different consumer age groups.

Intensive research is induced by the expected spread of mycotoxin-producing fungi due to global warming, the increase in food and feed contamination levels caused by the toxins they produce and the health problems and economic damages attributed to them. The large number of reports on the results are made more manageable by research area by the regularly published review articles, such as those on the interaction of mycotoxin-producing Aspergillus species with soil microorganisms [2], on human physiological effects of mycotoxin exposure [3], on the application of biocontrol technologies to reduce aflatoxin contamination, on the effect of silage production technology and the microbiota on aflatoxin contamination [4], on the sources, occurrence and regulation of mycotoxin contamination [5, 6] and on its detection methods [7]. A special edition of the journal Frontiers in Microbiology, containing 22 of the latest research articles and summaries, has also been published in the form of a book [8].

In view of the above reviews, only literature publications closely related to the objective of our paper are summarized below.

2.1. Occurrence of aflatoxins

Aflatoxins and other mycotoxins that occur in raw agricultural products (mainly peanuts, maize, rice, nuts, figs, spices and dried fruits) and feeds enter the food chain and can be detected in milk [9, 10, 11, 12], eggs, meat and offal [6, 13]. Aflatoxicol has been detected in the liver, kidney and meat of broiler and laying hens [14, 15]. Compared to the concentration of AFB1 in the feed, >5700-, >4600- and >3800-fold concentrations were measured in the livers of hens, in egg yolk and in egg white, respectively [16].

In addition to the exposure of the animals to AFB1 (µg/kg body weight), the concentration of AFM1 entering the milk from the feed depends on a number of factors, such as the health status of the cow, its milk yield, the lactation period, etc. [17]. The transmission rate is higher in specimens with higher milk yields [18]. The results of several studies have been reported in the literature, according to which the rounded transmission rate varied from 0.35 to 6%. Lower transmission rates (0.08%-0.33%) were observed in sheep [19].

There is less research on the transmission of aflatoxin to the liver, meat and eggs, but these report significantly lower transmission rates compared to milk, making milk still the most significant source of aflatoxin among foods of animal origin [4, 20].

When a food contaminated with AFB1 is consumed, AFM1 is excreted in breast milk to a similar extent as in cow’s milk [21, 22, 23, 24, 25, 26]. Infants and young children who are fed formula or milk drinks based on cow’s milk may also be exposed to AFM1. The results of European surveys indicate much lower levels than African or Asian publications [27].

As with all mycotoxins, aflatoxins show significant annual fluctuations in their levels depending on the weather conditions affecting fungal growth and toxin production [28].

For its most recent risk assessment [29], EFSA used the results of aflatoxin M1 measurements reported by Member States after 2013. Statistical data for some of the major food categories are summarized in Table 1.

Comment:

N: number of measurement results; % LCD (left censored data): ratio of results below the detection/quantification limit; P95: 95th percentile; LB: lower bound – result of substitution with the lowest concentration value; UB: upper bound – result of substitution with the highest concentration value; Other: foods for infants and young children.

2.2. Health effects of aflatoxins

Aflatoxins (especially AFB1, AFG1 and AFM1) proved to be extremely potent carcinogenic, kidney and liver damaging, genotoxic, malformative, reproductive capacity decreasing, immunosuppressive and nervous system damaging compounds in all experimental animal species, such as fish, ducks, mice, rats and monkeys [30]. A recently published study showed that the spores of pathogenic fungi cause severe, fatal infections in various birds [31].

High levels of AFB1, both in humans and animals, can cause fast-action, acute poisoning, during which severe hepatitic failure can lead to death, however, human risk of this in developed countries is negligible. Of aflatoxins, aflatoxin B1 is the most potent carcinogenic and genotoxic compound, and it is the one most commonly found in foods and feeds. Most often, it causes hepatocellular carcinoma (HCC), which is why AFB1 has been classified as a Group 1 human carcinogen by the IARC. After consumption of feed contaminated with AFB1, its hydroxy metabolite, aflatoxin M1, which is also a carcinogenic compound, although with a toxicity that is about one tenth of that of AFB1, is excreted by dairy cows in milk [32, 33].

Aflatoxins are rapidly and extensively adsorbed in the small intestine and, once in the liver, the metabolism of aflatoxin is catalyzed by the cytochrome P450 enzyme system found there. AFB1, AFG1 and AFM1 are converted to a reactive electrophilic epoxide that is capable of covalently binding to both DNA and proteins. Glutathion S-transferases (GST) are able to form a conjugative link with the 8,9-exo epoxide of AFB1, which is no longer able to enter harmful reactions in the body, and is excreted through bile and the kidneys. Polymorphisms among individuals result in high variability in enzymatic processes, and thus sensitivity to aflatoxin also varies from individual to individual [30, 34, 35]. In optimal cases, most aflatoxin metabolites are excreted within a few days, however, they have been observed to be present in protein-bound form over a longer period of time (e.g., in the case of aflatoxin-albumin adducts), with a half-life of 30 to 60 days in peripheral circulation [36].

Aflatoxins also damage liver cells directly, as well as indirectly, by altering the expression of genes involved in lipid metabolism. Increased cholesterol, triglyceride and lipoprotein production can cause the disintegration of hepatocytes. Hepatocyte death may lead to acute hepatitis, which can result in liver failure and, in more severe cases death. The disrupted metabolism of hepatitis patients can lead to malnutrition, which indirectly contributes to a general decrease in the antioxidant capacity of hepatocytes, to a loss of liver tissue regeneration capacity and, ultimately, liver failure [3].

Based on the opinion of EFSA experts, a key point in the risk assessment of aflatoxins is the evaluation of the role these toxins play in the development of liver cancer. From this point of view, children are particularly sensitive to aflatoxins, because, due to their low body weight, have a higher intake of food per kg body weight, and the risk of developing liver cancer is also higher in individuals infected with the hepatitis B (or C) virus and in the elderly. In people living in areas where both hepatitis B virus (HBV) infection and aflatoxin exposure are common, hepatocellular carcinoma (HCC) samples show a mutation hotspot (G-T transformation) at codon 249 of the p53 gene, which mutation is considered to be a signature of aflatoxin-induced HCC [37]. The possible reason for this is that hepatitis infection of the liver alters the expression of genes encoding aflatoxin detoxification enzymes, resulting in, for example, the induction of CYP enzymes or a decreased GST activity, thereby preventing the body from adequately eliminating aflatoxins [35]. Due to the immunosuppressive effect of aflatoxins, elderly people with chronic diseases are at particular risk, because in their case the efficiency of cell-level repair mechanisms is inferior, so the elimination of aflatoxins is also less effective. It should be emphasized that aflatoxins are able to cross the placenta, so aflatoxin exposure of pregnant women can also endanger the fetus [38].

2.3. The effect of processing on the aflatoxin content in foods

The common feature of aflatoxins is that they are stable, resistant to processing and heat effects. As a consequence, their presence must also be taken into account in the case of processed foods. Certain processing steps, such as sorting, refining, grinding, cooking, baking, frying in oil, roasting, preservation, flocculating, alkaline cooking, nixtamalisation, extrusion and fermentation, can reduce the concentration of mycotoxins in crops and processed foods, but they are not adequate enough to eliminate all contaminants, so the role of prevention at the very beginning of the food chain is of paramount importance [20]. For example, in terms of AFM1 contamination, it is important to reduce the AFM1 contamination of feeds using pre- and post-harvest biotechnological methods as well as toxin binders [4, 17].

Of heat treatment processes, conventional cooking and baking have little effect on mycotoxin contamination, while methods performed at higher temperatures, or possibly using dry heat [39], are more efficient. The breakdown of mycotoxins is enhanced by the presence of sugars, e.g., glucose, during heat treatment [40].

During the wet milling of cereals, such as corn, aflatoxin is distributed among the milling fractions in the following proportions: soaking water: 39–42%, fiber: 30–38%, gluten: 13–17%, germ, 6–10% and starch: 1%. Thus, the total aflatoxin level in the processed products decreases with the proportion remaining in the soaking water. After the dry milling of corn, the groats, bran and flour fractions contain only 6 to 10% of the original aflatoxin content, with most of the aflatoxin entering the germ and husk fractions [20].

Contamination of rice with aflatoxin most often occurs due to improper harvest and storage conditions. Mycotoxins are found primarily in the rice husk and bran layers. Husked brown rice and white rice obtained by polishing are gradually less contaminated [41].

The various heat treatment processes, pasteurization and freezing do not have a significant effect on the aflatoxin content of milk and dairy products [42]. The reduction effect of some heat treatment processes on AFM1 expressed in numerical values are as follows: pasteurization: 7.6%-12.9%, boiling: 14.5-23.9% [43], UHT treatment: 32% [44].

Different physical and chemical methods have been used with good efficiency to reduce the AFM1 content of milk or other liquid products: microwave irradiation (52%) [43], membrane filtration (81%) [45], biofiltration (81%) [46] combination of centrifugation and filtration (83%) [45], ozone treatment [47], the use of adsorbents (85-90%) [48, 49].

Intensive research is underway on the use of microorganisms. Encouraging results for the reduction of AFM1 contamination in milk have been obtained using Saccharomyces cerevisiae (90-93%) [50], S. cerevisise + L. rhamnosus, L. delbrueckii spp. bulgaricus, B. lactis (100%) [49], the mixture of different yeasts (65-69%) [51], heat-treated L. plantarum (94,5%) [44], L. bulgaricus (58%) [52] and in yogurt using S. thermophilus, L. bulgaricus and L. plantrium strains [53]. It remains to be seen how (in the case of using live microbes) changes in organoleptic properties can be eliminated if non-conventional cultures are used, and how the lactic acid bacterium-AFM1 complex formed can be removed from the product [44].

3. Data used to estimate consumer exposure

Exposure (g/kg body weight or ng/kg body weight) is calculated by multiplying the amount of food consumed (g/kg body weight) and the contaminant concentrations measured in it (ng/kg). In the deterministic method, we multiply the mean (median), or sometimes an upper percentile (95.0, 97.5) value. This calculation results in a point estimate giving a specific value [54, 55]. A more subtle estimate is obtained by probabilistic methods [56, 57], in which the distribution of input data is taken into account and thus a distribution is also obtains for exposure. Care should be taken when considering test results below the limit of quantification (LOQ). If the proportion of samples below the LOQ is between 50 and 80%, a maximum likelihood estimate (MLE) gives the best results [58].

Whichever method is used for the estimation, it is important to take into account the uncertainty of each calculation step, their magnitude, and to evaluate the results obtained in light of their cumulative effect [59]. The calculated uncertainty interval includes the true value with a certain level of confidence, i.e., with a certain degree of certainty [60]. The amount of contaminant entering the consumer’s body (EDI) is compared to the toxicological reference value(s) to determine the expected health risk.

For both short-term and long-term exposure estimation, it is worth examining the consumer groups that are particularly affected by the consumption of the given food/contaminant combination, and comparing the exposure of average consumers and „large consumers” [61].

3.1. Reference values for exposure assessment

Depending on the specific properties of the contaminant, the reference value may be the Acceptable Daily Intake (ADI), the Provisionally Tolerable Weekly/Monthly Intake (PTW/MI) or the Acute Reference Dose (ARfD). For food contaminants, the reference value is usually the tolerable daily intake (TDI). The benchmark dose (BMD) is the smallest dose that is estimated from the fitted dose-response curve at which a preselected effect level (benchmark response – BMR) can be observed, usually an increase or decrease of 5 or 10% compared to the control group. The lower confidence value of the BMD is the BMDL [62]. In the case of aflatoxins, Margin of Exposure (MoE) analysis is used to characterize the risk, as no TDI or other toxicological reference value can be established. In such a case, the value of the BMDL, adjusted by the uncertainty factor, is compared to the estimated exposure. The risk attributed to a contaminant can also be expressed as the ratio of the exposure to other reference values, the Hazard Quotient (HQ) or the Hazard Index (HI), which is the sum of the hazard ratios of substances acting on the same target organ or organ system, usually used for cumulative estimates [63].

EFSA recommends the use of 4 μg/kg body weight/day as the BMDL10 value as a benchmark for AFM1 risk characterization [29]. The results obtained are considered to be of concern below 10,000, with an MoE of 10,000 or greater indicating little risk to public health.

To characterize the risk of AFM1, the safe dose recommended by Kuiper-Goodmann (0.2 ng/kg body weight/day) can also be used to calculate the hazard index (HI), which is a quotient of a tumor-causing dose in 50% of animals and a safety factor of 50,000 [64].

According to the 2018 calculations of the JECFA [20], with an average daily intake of 1 ng/kg body weight AFB1, the probability of developing liver cancer is on average 0.269 per 100,000 persons per year, with the upper limit of the 95% confidence interval of the estimate being 0.562/100,000 persons/year in HBsAg+ (positive for hepatitis B surface antigen) individuals. For HBsAg- (negative for hepatitis B surface antigen) individuals, the mean value was 0.017 cancers/year/100,000 persons, with the upper limit of the 95% confidence interval of the estimate being 0.049/100,000 persons/year. The estimated mean values for AFM1 are one order of magnitude lower: 0.027/100,000 persons for HBsAG+, and 0.002/100,000 persons for HbsAg- individuals [20].

JECFA estimated the risk of hepatocellular carcinoma (HCC) associated with aflatoxin exposure using Equation 1:

R_i = [(P_HBV+ × HBV+) + (P_HBV− × (1–HBV+))] x AF bevitel (1),

where R_i is the HCC risk for region i, HBV+ is the prevalence of chronic hepatitis B in the study population, P_HBV+ is the probability of developing liver cancer in this fraction of the population and P_HBV- is the probability of developing liver cancer in the rest of the population.

3.2. Food consumption data

The calculations were performed using data from two representative Hungarian food consumption surveys conducted 10 years apart. The three-day survey of 2009 provided food consumption data for 4,992 individuals for a total of 14,976 consumption days, processed by dietitians and broken down into raw materials for the characterization of food consumption habits [65]. The ratio of milk and dairy product consumption days is shown in Figure 1.

Of the 14,976 consumption days in the 2009 survey, the consumption frequency [%] of milk, sour cream and cream, cheese and kefir or yogurt was 75.2, 52.8, 46.3 and 19.1, respectively.

The 2018-2020 survey was conducted within the framework of EFSA’s Europe-wide EU MENU or “What’s on the table in Europe?” project, in accordance with the recommended, uniform methodology [66, 67]. Participating persons were selected from the households participating in the Hungarian Central Statistical Office Household Budget and Living Conditions survey. During the program, two consumption days of 2,657 individuals between the ages of 1 and 74 were recorded, with the help of dietitians. On the 5,314 consumption days, the consumption frequencies [%] of milk, sour cream and cream, cheese and kefir or yogurt were 96.8, 54, 60.6 and 24. The ratio of milk and dairy product consumption days is shown in Figure 2.

The distribution of consumers by age group is shown in Table 2.

Changes in the frequency of consumption of milk and various dairy products were compared using the milk and dairy product consumption days of the 2009 and 2018-2020 food consumption surveys. The numbers of consumption days of the different foods were compared to the total consumption days of the given survey (Figure 3). The frequencies of consumption of the different foods during the survey periods are characterized by the figure. Among the food categories studied, the consumption frequency of milk and milk-based desserts increased by more than 20%. The consumption frequency of cheeses shows an increase of 14%. The consumption frequencies of sour milk products (kefir, yogurt, sour cream), cream and flavored milks remained almost constant (with the former increasing slightly and the latter decreasing to a small degree). The consumption frequencies of condensed milk and milk powder has decreased significantly. Overall, it can be stated that the consumption frequencies of milk and dairy products has increased slightly over the last 10 years.

Based on the change in consumption frequencies over 10 years, an increase in aflatoxin exposure could be expected, however, this effect was offset by the change in the amounts consumed. The average consumption in milk equivalent, calculated with the median value of processing and enrichment factors, was 310.7 g/day in 2009, and this value decreased to 295.3 g/day in 2018-2020.

Very little data were available on AFM1 concentrations in processed dairy products, so a database of AFM1 processing and enrichment factors for sour milk products (e.g., kefir, yogurt, sour cream) and various cheeses (hard, semi-hard, soft and processed cheeses, fresh cheeses) was compiled on the basis of the latest literature data, and consumer exposure was calculated with the milk equivalent of the consumed quantities of these products.

3.3. Aflatoxin concentration data

AFM1 analytical data are partly derived from NÉBIH’s 2011-2020 Hungarian monitoring survey (1,288 data). 40% of the samples contained measurable amounts of AFM1. Most of the measurements were performed by and HPLC methods on samples taken from the milk of dairy farms or private producers and, to a small extent, from commercially available mixed milk. In addition to the large number of items exhibiting contamination below the LOQ (60%), there were also items with very high contamination compared to the average. Values above 100 ng/kg were: 110, 122, 141, 149, 150, 190, 238, 240, 252, 260, 292, 376, 513, 740 and 860 ng/kg, respectively. We were unable to check the correctness of the results, but we saw no reason to omit them either, so the full data set was used in our further calculations. Another 1,177 samples were analyzed by January 2021 within the framework of the joint project of the University of Debrecen and NÉBIH („Determining of the short- and long-term aflatoxin exposure of Hungarian consumers in the dairy product chain and establishing risk management measures”). In the latter case, milk samples taken directly from the transport tankers by the staff of the dairy company at the 9 dairy farms participating in the project were analyzed between 2019 and 2021 by the ELISA method in the laboratory of the Instrument Center of the University of Debrecen. In samples with concentrations above the 20 ng/kg „action level”, exact AFM1 concentrations were confirmed by HPLC in the laboratory of NÉBIH. The number of samples with concentrations above the LOQ was 672 (57.1%). In the case of samples with concentrations above 20 ng/kg, the dairy farm was notified and it was recommended that appropriate precautionary measures be taken. As a result of this intervention, it was possible to stop the increase in milk contamination, and the AFM1 contamination of the milk produced was kept below the 50 ng/kg level. Detailed results will be published in the final report of the project.

The number of milk samples examined, broken down by year, is shown in Figure 4.

To refine our estimate and to compensate for the large number of values below the LOQ, instead of the usual LOQ, LOQ=0 and LOQ/2 approximations, the values of concentration data below the LOQ were also taken into account with the values of data generated with the help of a distribution with an element number identical to that of the number of measurement results. To measurement results above the LOQ, different distributions were fitted using the GAMLSS and GAMLSS.dist packages of the R statistical software using maximum likelihood estimation, then we used the parameters describing the goodness of the fit (AIC – Akaike’s Information Criterion, BIC – Bayesian Information Criterion and Global Deviance) to select the distributions that gave the optimal fit. The adequacy of the fits was also evaluated by visual comparison of the histogram made from the data and the distribution obtained, as well as by examining the normality of the differences and using a Q-Q plot. The two best-fit distributions were the two-parameter lognormal (Figure 6) and the four-parameter Box-Cox t-distribution (BCT) (Figure 7), which is suitable for the modeling of slowly decaying, continuously distributed data with positive or negative distortion similar to those of aflatoxins [68, 69]. Exposure calculations were performed with a lognormal distribution generated with the assumption of LOQ=5. The selected distributions were then fitted to the entire AFM1 data set, and the evaluation was performed again. Given that a positive change was observed in the parameters describing the goodness of the fit, the distribution chosen were considered to be acceptable.

Descriptive statistics for AFM1 test results and the fitted distribution are summarized below.

The relative frequency distributions of NÉBIH and DE test results are shown in Figure 5.

With the exception of the outstanding NÉBIH measurement result values (indicated with blue-red asterisk) in the 10-15 ng/kg range, the frequency of AFM1 concentrations in the LOQ-70 ng/kg range was very similar in the two series of measurements and this justifies the joint evaluation of the measurement results. The relative frequency of samples containing AFM1 in concentrations above 70 ng/kg was <0.5% in the NÉBIH study.

A limiting factor in the risk assessment of aflatoxins was the lack of contamination data. According to the recommendation of EFSA [29], food categories for which the number of positive samples does not exceed 25 or for which the proportion of samples below the limit of quantification is greater than 80% should be excluded. In terms of AFM1 results, only the testing of milk met this criterion (Table 4), the number of tests for processed dairy products proved to be very small.

¹: Formula: infant formula and other baby food
²: Milk-based food

4. Consumer exposure

In the case of food consumption data, the Observed Individual Means (OIM) method recommended for long-term estimation was used. First, all milk and dairy product consumption data were converted to milk equivalent, using the enrichment and processing factors specific to the given food category (Equations 2 and 3).

The intake of foods e₁, ..., e_j expressed in g/kg body weight (B) on a given consumption day (n), expressed in milk equivalent is

where

m_e is the mass (g) of food e on consumption day n_i,

F is the processing (e.g., enrichment) factor characteristic of food e,

kg body weight is the body weight of the person belonging to the given consumption day,

where e is the AFM1 concentration in the milk used for the preparation of food e and is the AFM1 concentration in the processed food.

F_e is the value calculated from the min., med. and max. results obtained in the experiments.

By multiplying the amounts consumed in g/kg body weight/day by the average AFM1 concentration (ng/kg) calculated from the values of the fitted distribution functions, the exposure values for each consumption day were obtained (ng/kg body weight/day). The intake values of the 2 (2018-2020 survey) or 3 (2009 survey) consumption days of the participating persons were averaged. The results were aggregated by consumer age group and consumer exposure was calculated using the data from both food consumption surveys.

First, the effects of the minimum (F_min), median (F_med) and maximum (F_max) values of the processing factors on the result of the exposure estimation were examined. The calculation was performed with the fitted lognormal AFM1 mean data, as well as with the mean (EDI_átl) and 97.5 percentile (EDI_0,975) milk consumption values of the 2018-2020 survey. The results are summarized in Table 5. The table illustrates the differences between the mean calculated using the deterministic method and the 97.5 percentile results, based on the 2018-2020 (EU MENU) survey.

Taking into account the minimum-median-maximum values of the processing factors did not notably affect the results. There were significant differences in the mean values of the toddler age group when considering the minimum and median factors, therefore, the values calculated with the median of the processing and enrichment factors are used below to present the different exposure estimation results.

The exposure of the various age groups was calculated based on the P0.05, mean, median P0.975 percentile estimated daily intake values (EDI) of the 2009 food consumption survey, median processing factors and mean AFM1 concentration data. The exposures of the different consumer age groups were compared on the basis of the calculated EDI.

The estimated daily intake values of the age groups of the 2018-2020 survey show a trend similar to that of the 2009 data (Figure 9).

Comparing both the mean and the 97.5 percentile estimated daily intake values, it is clear that the exposure of each age group has been found mostly constant over the past 10 years. The only noticeable differences are in the mean values of the toddler age group and the 97.5 percentile values of the children age group, but the differences are not significant. The number of items in the toddler age group in the 2009 survey is very low (90 people) compared to the 2018-2020 survey (482 people). Values calculated with a smaller number of elements are burdened with a greater uncertainty.

4.1. Assessment of consumer exposure

Based on the obtained exposure values, the Margin of Exposure (MoE) approach (Equation 3), the hazard index (HI) (Equation 4) and the probability increase of liver cancer attributable to AFM1 intake were used to assess the risk of the Hungarian population. For the MoE method, the BMDL10 value for AFM1 of 4 μg/kg body weight/day was taken into consideration:

Consumer exposure is considered to be risky if the value of MoE is <10,000. The MoE values of exposure calculated by the deterministic estimation from the consumption data of the 2018-2020 food consumption survey are shown in Table 6.

The health risk threshold (10,000) is reached or approached only by the “large consumers” (97.5 percentile) of the toddler and children age groups. In the case of the other age groups, no significant risk can be identified using this risk characterization methodology.

For the calculation of the hazard index, the safe dose recommended by Kuiper-Goodmann [59] was used (0.2 ng/kg body weight):

When calculating the hazard index (HI), the degree of risk is directly proportional to the EDI values and is considered to be of concern when the value is 1 or higher. As an example, the results of deterministic estimates using the consumption data of the 2018-2020 food consumption survey by age group are shown in Table 7.

Note: HI values that pose a health risk are indicated by bold numbers

HI values calculated from the mean and 97.5 percentile values of the estimated daily intakes of the age groups indicate that the risk from exposure of the adolescent, adult and elderly age groups is not considered to be of concern. However, for toddlers and children, in the case of the 97.5 percentile values (large consumers), exposure is well above levels considered to be safe. One of the most important of the above results is the HI value of 1, characterizing the average intake of toddlers, as it suggests that a significant proportion of this age group is exposed to large amounts of AFM1, which is of great health concern.

Assuming a Hungarian hepatitis B prevalence of 0,7% [70], the incidence of liver cancer associated with aflatoxin was calculated using Equation 1. Calculations were also performed with the mean and upper 95% confidence values for the likelihood of developing liver cancer:

Átlag R_Mo= [(0,0269 × 0,007) + (0,0017× 0,993)] × EDI,

CI,.95 R_Mo= [(0,0562 × 0,007) + (0,0049× 0,993)] × EDI.

The values of hepatocellular carcinoma (HCCi) attributable to aflatoxin exposure derived from the mean and 97.5 percentile results of AFM1 exposure values calculated from the consumption data of the 2018-2020 food consumption survey by deterministic estimation (DET) (illness/100,000 persons/year) are summarized by age group in Table 8.

The risk of developing HCC is increased many times by aflatoxin exposure in the presence of chromic hepatitis B. As the prevalence of hepatitis B is low in Hungary (and in Europe in general), the increase in HCCi induced by aflatoxin is not high either. Although the numerical value of the estimated incidence of liver cancer proved to be very low, their relative values show in this case as well the high risk of “large consumers” of toddlers and children compared to the other age groups.

5. Situation assessment, recommendations

Chronic exposure to AFM1 calculated by the deterministic method and compared to various reference values consistently indicates that the exposure of children aged 1<3 is the highest in the studied age groups. The lowest exposure values are observed for the oldest age groups. Due to a lack of data, exposure in infants aged <1 could not be studied. However, the correlation is not directly between age and intake amounts, but between the change (typically increase) in body weight of increasingly older age groups and the intake amounts.

Given that the toxicity of aflatoxins primarily poses a health risk to developing organisms, special attention should be paid to reducing their exposure and keeping it to a minimum. However, it should be emphasized that the presence of carcinogenic compounds should be kept to a minimum in all age groups.

The body is burdened not only with the AFM1 contamination of breast milk and other milks or milk-based products, but also with the AFB1 taken with other foods and which is 10 times more toxic than AFM1. Since their mechanism of action is the same, the effects of aflatoxins and AFM1 add up. Therefore, we need to pay attention to the quality of our food and the storage conditions of products with open packaging. Products with a musty smell and traces of mold should not be consumed, even after cooking or baking.

Milks, analyzed during the annual monitoring inspections and exhibiting contamination levels that are 10 to 15 times higher than the maximum tolerable level allowed by the law are also marketed. Particularly at risk are those individuals who regularly consume milk from the same source where the animals are fed aflatoxin-contaminated feed.

To date, there is no routine and industry-wide large scale process that can reliably and completely eliminate the aflatoxin content of foods, so the focus remains on preventing contamination. This is a complex task that requires the involvement of all stakeholders in the food chain, starting with the application of good agricultural practices, and the proper preparation and management of arable land. This is followed by the selection of hybrids resistant to mold, and then a series of measures taken during the harvesting, transport and storage of crops that can prevent the growth of molds (setting appropriate temperature and humidity levels, sorting, peeling and physical treatment of the crops). Last but not least, the appropriate storage and treatment cereals, silage or other processed feed products intended for animal feed, checking their aflatoxin levels and their physical, chemical or biological detoxification, if necessary [4].

The success of prevention and the adequacy of milk shipments can also be checked at the level of dairy farms and dairy plants. With a sampling plan developed for the detection of the aflatoxin M1 content of raw milk and an early warning system, and by applying the 20 ng/kg action threshold already proven to be useful in Italy, an increase in the level of contamination can be predicted effectively. Based on the warning, the dairy farm can prevent the maximum tolerable AFM1 concentration (50 ng/kg) from being reached in accordance with local conditions, for example, by modifying the composition of the feed or by using toxin binders. This can reduce the use of contaminated milk batches in primary and secondary milk processing and, consequently, reduce consumer exposure [1, 10, 71].

It should also be noted that, due to the uncertainty of the detection, ELISA kits set to indicate an AFM1 concentration of 50 ng/kg may still classify batches of milk with contaminations of ≤ 65-70 ng/kg as adequate in 50% of the cases.

In order to protect infants and young children who are most exposed to AFM1 and who are also the most vulnerable, but also to protect the health of the entire population, it is recommended that the regulation of dairy plant inspections is amended in a way that ensures that if the AFM1 contamination of the milk delivered from a farm is ≥20 ng/kg, the plant is obligated to notify the dairy farm and NÉBIH, and thereafter to monitor the effectiveness of the dairy farm measures taken to reduce the contamination by daily monitoring of the contamination of the milk delivered from the farm.

It is also recommended that the warning level used in the self-inspection of dairy farms is set to 20 ng/kg instead of the current 50 ng/kg. ELISA kits for the detection of AFM1 at a concentration of 5-10 ng/kg are available for both dairy plant and producer monitoring, so there is no methodological obstacle to the establishment of a new warning threshold.

6. Acknowledgment

The authors would like to thank to Judit Sali and Katalin Csizmadia for providing the relevant data of the 2018-2020 NÉBIH consumption survey, the staff of the DE Instrument Center and NÉBIH for the AFM1 analysis of the milk samples, Attila Nagy and Gabriella Miklós of NÉBIH for their useful suggestions, Béla Béri, who participated in the DE-NÉBIH project, and to the managers and employees of Alföldi Tej and the dairy farms for their cooperation.

Our research program was supported by the National Research, Development and Innovation Fund in the 2018-1.2.1-NKP foundation system with the designation 2018-1.2.1-NKP-2018-00002 (AA, KK).

7. References

[1] Ambrus Á., Szenczi-Cseh, J., Griff, T., Kerekes K., Miklós G., Szigeti, T., Vásárhelyi, A. (2020): Élelmiszereink mikotoxin és növényvédőszer-maradék szennyezettségének élelmiszerbiztonsági megítélése 2. rész Mikotoxinok; Food safety assessment of the mycotoxin and pesticide residue contamination of our foods, Part 2. Mycotoxins. Journal of food Investigation, LXVI (2) pp. 2923-2949. 

[2] Pfliegler, W. P., Pócsi, I., Győri, Z., & Pusztahelyi, T. (2020): The Aspergilli and Their Mycotoxins: Metabolic Interactions With Plants and the Soil Biota. Frontiers in Microbiology, Vol. 10 2908. https://doi.org/doi:10.3389/fmicb.2019.02921

[3] Ráduly, Z., Szabó, L., Madar, A., Pócsi, I., & Csernoch, L. (2020): Toxicological and Medical Aspects of Aspergillus-Derived Mycotoxins Entering the Feed and Food Chain. Frontiers in Microbiology 10.https://doi.org/doi:10.3389/fmicb.2019.02908

[4] Peles, F., Sipos, P., Kovács, S., Győri, Z., Pócsi, I., Pusztahelyi, T. (2021): Biological Control and Mitigation of Aflatoxin Contamination in Commodities. Toxins 13 (104). https://doi.org/10.3390/toxins13020104

[5] Mahato, D.K., Lee, K.E., Kamle, M., Devi, S., Dewangan, K.N., Kumar, P, Kang, S.G. (2019): Aflatoxins in Food and Feed: An Overview on Prevalence, Detection and Control Strategies. Front. Microbiol. Vol. 10 (2266)

[6] Filazi, A., Tansel, U. (2019): Occurrence of Aflatoxins in Food (2013): in Mehdi Razzaghi, Abyaneh (Szerk.), Aflatoxins - Recent Advances and Future Prospects. doi: https://doi.org/10.5772/51031

[7] Mikló, G., Angeli, C., Ambrus, Á., Nagy, A., Kardos, V., Zentai, A., Kerekes, K., Farkas, Z., Józwiak, Á., Bartók, T. (2020): Detection of Aflatoxins in Different matrices and Food-Chain Positions. Front. Microbiol 11 (1916) https://doi.org/10.3389/fmicb.2020.01916

[8] Pócsi, I., Giacometti, F., Ambrus, Á. and Logrieco, A.F. (2020): Editorial: Aspergillus-Derived Mycotoxins in the Feed and Food Chain. Front. Microbiol 11 (606108). https://doi.org/10.3389/fmicb.2020.606108

[9] Martinez-Miranda, M. M., Rosero-Moreano, M., and Taborda-Ocampo, G. (2019): Occurrence, dietary exposure and risk assessment of aflatoxins in arepa, bread and rice. Food Control 98 pp. 359–366. https://doi.org/10.1016/j.foodcont.2018.11.046

[10] Serraino, A., Bonilauri, P., Kerekes, K., Farkas, Z., Giacometti, F., Canever, A., Zambrini, A.V., Ambrus, Á. (2019): Occurrence of Aflatoxin M1 in raw milk marketed in Italy: Exposure Assessment and Risk Characterization. Front. Microbiol. 10 (2516) https://doi.org/10.3389/fmicb.2019.02516

[11] Udovicki, B., Ilija Djekic, I., Eleni P., Kalogianni, E.P., Rajkovic, A. (2019): Exposure assessment and risk characterization of aflatoxin m1 intake through consumption of milk and yoghurt by student population in Serbia and Greece. Toxins 11 (4) pp. 205-216. https://doi.org/10.3390/toxins11040205

[12] Peles, F., Sipos, P., Győri, Z., Pfliegler, W.P., Giacometti, F., Serraino, A., Pagliuca, G., Gazzotti, T., Pócsi, I. (2019): Adverse Effects, Transformation and Channeling of Aflatoxins Into Food Raw Materials in Livestock. Front. Microbiol. 10 (2861) https://doi.org/10.3389/fmicb.2019.02861

[13] Campagnollo, F. B., Ganev, K. C., Khaneghah, A. M., Portela, J. B., Cruz, A. G., Granato, D., Corassin, C. H., Oliveira, C. A. F., Sant’Ana, A. S. (2016): The occurrence and effect of unit operations for dairy products processing on the fate of aflatoxin M1: A review. Food Control 68 pp. 310-329. doi: https://doi.org/10.1016/j.foodcont.2016.04.007

[14] Wolzak, A., Pearson, A. M., Coleman, T. H. (1986): Aflatoxin carry-over and clearance from tissues of laying hens. Food and Chemical Toxicology 24 pp. 37-41. https://doi.org/10.1016/0278-6915(86)90262-0

[15] Hussain, Z., Khan, M. Z., Khan, A., Javed, I., Saleemi, M. K., Mahmood, S., Asi, M. R. (2010): Residues of aflatoxin B1 in broiler meat: Effect of age and dietary aflatoxin B1 levels. Food and Chemical Toxicology 48 pp. 3304-3307. https://doi.org/10.1016/j.fct.2010.08.016

[16] Bintvihok, A., Thiengnin, S., Doi, K., & Kumagai, S. (2002): Residues of Aflatoxins in the Liver, Muscle and Eggs of Domestic Fowls. Journal of Veterinary Medical Science 64 (11) pp. 1037–1039. doi: https://doi.org/10.1292/jvms.64.1037

[17] Moran, C. A., Kettunen, H., Yiannikouris, A., Ojanperä, S., Pennala, E., Helander, I. M., & Apajalahti, J. (2013): A dairy cow model to assess aflatoxin transmission from feed into milk – Evaluating efficacy of the mycotoxin binder Mycosorb®. Journal of Applied Animal Nutrition, 2. doi: https://doi.org/10.1017/jan.2013.12

[18] Britzi, M., Friedman, S., Miron, J., Solomon, R., Cuneah, O., Shimshoni, J., Shlosberg, A. (2013): Carry-Over of Aflatoxin B1 to Aflatoxin M1 in High Yielding Israeli Cows in Mid- and Late-Lactation. Toxins 5(1), pp. 173–183. doi: https://doi.org/10.3390/toxins5010173

[19] Battacone, G., Nudda, A., Palomba, M., Pascale, M., Nicolussi, P., Pulina,G. (2005): Transfer of Aflatoxin B1 from Feed to Milk and from Milk to Curd and Whey in Dairy Sheep Fed Artificially Contaminated Concentrates. J. Dairy Sci. 88 (9) pp. 3063–3069.

[20] JECFA (2018): Aflatoxins. In: Safety evaluation of certain contaminants in food: prepared by the eighty-third meeting of the Joint FAO/WHO Expert Committee on Food Additives (JECFA). FAO JECFA Monographs, pp. 3-280. SBN (PDF) 978-92-4-069847-5

[21] Fakhri, Y., Ghorbani, R., Taghavi, M., Keramati, H., Amanidaz, N., Moradi, B., Nazari, S.H., Shariatifar, N., Khaneghah, A.M. (2019): Concentration and Prevalence of Aflatoxin M1 in Human Breast Milk in Iran: Systematic Review, Meta-Analysis, and Carcinogenic Risk Assessment: A Review. Journal of Food Protection 82 (5) pp. 785–795. https://doi.org/10.4315/0362-028X.JFP-18-367

[22] Fakhri, Y., Rahmani, J., Oliveira, C.A.F., Franco, L.T., Corassin, C.H., Saba, S., Rafique, J., Khaneghah, A.M. (2019): Aflatoxin M1 in human breast milk: a global systematic review, metaanalysis, and risk assessment study (Monte Carlo simulation). Trends in Food Science & Technology 88 (5) pp. 333-342. doi: https://doi.org/10.1016/j.tifs.2019.03.013

[23] Radonić, J. R., Kocić Tanackov, S. D., Mihajlović, I. J., Grujić, Z. S., Vojinović Miloradov, M. B., Škrinjar, M. M., & Turk Sekulić, M. M. (2017): Occurrence of aflatoxin M1 in human milk samples in Vojvodina, Serbia: Estimation of average daily intake by babies. Journal of Environmental Science and Health, Part B 52 (1) pp. 59-63. https://doi.org/10.1080/03601234.2016.1229454

[24] Kunter, I., Hürer, N., Gülcan, H. O., Öztürk, B., Dogan, I., Sahin, G. (2017): Assessment of Aflatoxin M1 and Heavy Metal Levels in Mothers Breast Milk in Famagusta, Cyprus. Biol Trace Elem Res. 175 pp. 42-49. doi: https://doi.org/10.1007/s12011-016-0750-z

[25] Valitutti, F., De Santis, B., Trovato, C.M., Montuori, M., Gatti, S., Oliva, S., Brera, C., Catassi, C. (2018): Assessment of Mycotoxin Exposure in Breastfeeding Mothers with Celiac Disease. Nutrients 10 (3) doi: https://doi.org/10.3390/nu10030336.

[26] Bogalho, F., Duarte, S., Cardoso, M., Almeida, A., Cabeças, R., Lino, C., Pena, A. (2018): Exposure assessment of Portuguese infants to Aflatoxin M1 in breast milk and maternal social-demographical and food consumption determinants, Food Control

[27] Csapó, J., Albert, C., Sipos, P. (2020): The aflatoxin content of milk and dairy products as well as breast milk and the possibilities of detoxification. Acta Universitatis Sapientiae, Alimentaria, 13 pp. 99-117. doi: https://doi.org/10.2478/ausal-2020-0006

[28] Trevisani, M., Farkas, Z., Serraino, A., Zambrini, A. V., Pizzamiglio, V., Giacometti, F. Ambrus, A. (2014): Analysis of industry-generated data. Part 1: a baseline for the development of a tool to assist the milk industry in designing sampling plans for controlling aflatoxin M1 in milk. Food Additives & Contaminants: Part A: Chemistry, Analysis, Control, Exposure & Risk Assessment 31 (7) pp. 1246-1256. https://doi.org/10.1080/19440049.2014.925587

[29] EFSA Panel on Contaminants in the Food Chain (2020): Scientific opinion - Risk assessment of aflatoxins in food. EFSA Journal 18 (e06040) pp. 1-112. https://doi.org/10.2903/j.efsa.2020.6040

[30] IARC (2012): Aflatoxins. Chemical Agents and Related Occupations. A review of Human Carcinogens. IARC monographs on the evaluation of carcinogenic risks to humans.

[31] Pascal, A., Risco-Castillo, V., Jouvion, G., Le Barzic, C. and Guillot, J. (2021): Aspergillosis in Wild Birds. J. Fungi 7 (3) p. 241; doi: https://doi.org/10.3390/jof7030241

[32] JECFA, (2001): Aflatoxin M1. In: Safety evaluation of certain mycotoxins in food. FAO Food and Nutrition Paper 74 pp. 1-102.

[33] WHO (2002): Evaluation of certain mycotoxins in food TRS 906-JECFA 56/8 WHO technical report series 906 https://apps.who.int/iris/bitstream/handle/10665/42448/WHO_TRS_906.pdf?sequence=1 (Hozzáférés: 2021.01.28.)

[34] Bedard, L. L., Massey, T. E. (2006): Aflatoxin B1-induced DNA damage and its repair. Cancer Letters, 241 (2) pp. 174-83. https://doi.org/10.1016/j.canlet.2005.11.018

[35] EFSA (2007): Opinion of the scientific panel on contaminants in the food chain (CONTAM) related to the potential increase of consumer health risk by a possible increase of the existing maximum levels for aflatoxins in almonds, hazelnuts and pistachios and derived products. EFSA Journal 5 446. https://doi.org/10.2903/j.efsa.2007.446

[36] Williams, J. H., Phillips, T. D., Jolly, P. E., Stiles, J. K., Jolly, C. M. & Aggarwal, D. (2004): Human aflatoxicosis in developing countries: a review of toxicology, exposure, potential health consequences, and interventions. The American Journal of Clinical Nutrition 80 pp. 1106-1122. doi: https://doi.org/10.1093/ajcn/80.5.1106

[37] Wang, J. S. & Groopman, J. D. (1999): DNA damage by mycotoxins. Mutation Research 424 (1-2) pp. 167-181. https://doi.org/10.1016/s0027-5107(99)00017-2

[38] Denning, D. W., Allen, R., Wilkinson, A. P. & Morgan, M. R. (1990): Transplacental transfer of aflatoxin in humans. Carcinogenesis 11 (6) pp. 1033-1035. https://doi.org/10.1093/carcin/11.6.1033

[39] Serrano-Niño, J. C., Cavazos-Garduño, A., Hernandez-Mendoza, A., Applegate, B., Ferruzzi, M. G., San Martin-González, M. F., García, H. S. (2013): Assessment of probiotic strains ability to reduce the bioaccessibility of aflatoxin M1 in artificially contaminated milk using an in vitro digestive model. Food Control 31 (1) pp. 202-207. doi: https://doi.org/10.1016/j.foodcont.2012.09.023

[40] Bullerman, L. B., Bianchini, A. (2014): Good Food-Processing Techniques: Stability of Mycotoxins in Processed Maize-Based Foods. In: LESLIE, J. F. (Szerk.) Mycotoxin Reduction in Grain Chains. Ames, Iowa, USA: Wiley Blackwell, John Wiley & Sons, Inc. p. 92-97 ISBN 978-0-8138-2083-5

[41] Ali, N. (2019): Aflatoxins in rice: worldwide occurrence and public health perspectives. Toxicology Reports. doi: https://doi.org/10.1016/j.toxrep.2019.11.007

[42] Prandini, A., Tansini, G., Sigolo, S., Filippi, L., Laporta, M., Piva, G. (2009): On the occurrence of aflatoxin M1 in milk and dairy products. Food and Chemical Toxicology, 47 (5) pp. 984-991. https://doi.org/10.1016/j.fct.2007.10.005.

[43] Yosef, T. A., Al-Julaifi, M. Z., Salah-El-Dein, W. M., Al-Rizqi, A. M. (2013): Assessment of Aflatoxin M1 Residues in Raw Cow Milk at Al- Riyadh Area with Reference to Some Detoxification Applications. Life Science Journal - Acta Zhengzhou University Overseas Edition, 10 pp. 3365-3369.

[44] Iqbal, S. Z., Jinap, S., Pirouz, A. A. & Faizal, A. R. A. (2015): Aflatoxin M-1 in milk and dairy products, occurrence and recent challenges: A review. Trends in Food Science and Technology, 46 pp. 110-119. https://doi.org/10.1016/j.tifs.2015.08.005

[45] Kuharic, Z., Jakopovic, Z., Canak, I., Frece, J., Bosnir, J., Pavlek, Z., Ivesic, M., Markov, K. (2018): Removing aflatoxin M1 from milk with native lactic acid bacteria, centrifugation, and filtration. Archives of Industrial Hygiene and Toxicology 69 (4) pp. 334-339. https://doi.org/10.2478/aiht-2018-69-3160

[46] Foroughi, M., Jamab, M. S., Keramat, J. & Foroughi, M. (2018): Immobilization of Saccharomyces cerevisiae on Perlite Beads for the Decontamination of Aflatoxin M1 in Milk. Journal of Food Science 83 (7) pp. 2008-2013. doi: https://doi.org/10.1111/1750-3841.14100

[47] Mohammadi, H., Mazloomi, S. M., Eskandari, M. H., Aminlari, M., Niakousari, M. (2017): The Effect of Ozone on Aflatoxin M1, Oxidative Stability, Carotenoid Content and the Microbial Count of Milk. Ozone: Science & Engineering 39(6) pp. 447-453. doi: https://doi.org/10.1080/01919512.2017.1329647

[48] Assaf, J. C., El Khoury, A., Atoui, A., Louka, N., Chokr, A. (2018): A novel technique for aflatoxin M1 detoxification using chitin or treated shrimp shells: in vitro effect of physical and kinetic parameters on the binding stability. Applied Microbiology & Biotechnology 102 pp. 6687-6697. doi: https://doi.org/10.1007/s00253-018-9124-0

[49] Womack, E. D., Sparks, D. L., Brown, A. E. (2016): Aflatoxin M-1 in milk and milk products: a short review. World Mycotoxin Journal 9 (2) pp. 305-315. doi: https://doi.org/10.3920/WMJ2014.1867

[50] Corassin, C. H., Bovo, F., Rosim, R. E., Oliveira, C. A. F. (2013): Efficiency of Saccharomyces cerevisiae and lactic acid bacteria strains to bind aflatoxin M-1 in UHT skim milk. Food Control, 31 (1) pp. 80-83. doi: https://doi.org/10.1016/j.foodcont.2012.09.033

[51] Kamyar, S., Movassaghghazani, M. (2017): Reduction of Aflatoxin M1 in Milk Using Kefir Starter. Iranian Journal of Toxicology 11 (6) pp. 27-31. doi: https://doi.org/10.29252/arakmu.11.6.27

[52] Rad, M. N., Razavilar, V., Anvar, S. A. A., Akbari-Adergani, B. (2018): Selected bio-physical factors affecting the efficiency of Bifidobacterium animalis lactis and Lactobacillus delbrueckii bulgaricus to degrade aflatoxin M-1 in artificially contaminated milk. Journal of Food Safety, 38 (4) (e12463) doi: https://doi.org/10.1111/jfs.12463

[53] Elsanhoty, R. M., Salam, S. A., Ramadan, M. F., Badr, F. H. (2014): Detoxification of aflatoxin M1 in yoghurt using probiotics and lactic acid bacteria. Food Control 43 pp. 129-134. doi: https://doi.org/10.1016/j.foodcont.2014.03.002

[54] Hamilton, D., Murray, B., Ambrus, Á., Baptista, G., Ohlin, B., Kovacicova, J. (1997): Optimum use of available residue data in the estimation of dietary intake of pesticides. Pure & Applied Chemistry 69 (6) pp. 1373-1410. doi: https://doi.org/10.1351/pac199769061373

[55] Zentai, A., Szeitzné Szabó, M., Mihucz, G., Szeli, N., Szabó, A., Kovács, M. (2019): Occurrence and risk assessment of fumonisin B1 and B2 mycotoxins in maize-based food products in Hungary. Toxins 11 (12) p. 709. https://doi.org/10.3390/toxins11120709

[56] Zentai, A., Sali, J. , Szabó, I.J., Szeitzné-Szabó, M., Ambrus, Á., Vásárhelyi, A. (2013): Factors affecting the estimated probabilistic acute exposure to captan from apple consumption. Food Additives & Contaminants: Part A 30 (5) pp. 833-842. doi: https://doi.org/10.1080/19440049.2013.794977

[57] Zentai, A., Kerekes, K, Szabó, I., Ambrus, Á. (2015): A fogyasztók növényvédőszermaradékokból származó expozíciójának finomítása, 1. rész. Élelmiszervizsgálati Közlemények LXI (3) pp. 681-719.

[58] EFSA (2010): Management of left-censored data in dietary exposure assessment of chemical substances. EFSA Journal 8 (3) p. 1557 doi: https://doi.org/10.2903/j.efsa.2010.1557

[59] Szenczi-Cseh, J. & Ambrus, A. (2017): Uncertainty of exposure assessment of consumers to pesticide residues derived from food consumed. Journal of Environmental Science and Health B, 52 (9) pp. 658-670. https://doi.org/10.1080/03601234.2017.1331671

[60] EFSA (2006): Opinion of the Scientific Committee related to Uncertainties in Dietary Exposure Assessment. EFSA Journal 438 pp. 1-54. doi: https://doi.org/10.2903/j.efsa.2007.438

[61] Delmaar, C., Heinemeyer, G., Jantunen, M., Schneide, K., Schümann, M. (2020): General Aspects of Exposure Evaluation. In: Heinemeyer, G. (Szerk.) The Practice of Consumer Exposure Assessment. Gewerbestrasse 11, 6330 Cham, Switzerland: Springer Nature Switzerland AG., pp. 55-155. ISBN 978-3-319-96148-4

[62] Gürtler, R. (2020): Hazard Assessment and Derivation of Health-Based Guidance Values. In: Heinemeyer, G. (Szerk.) The Practice of Consumer Exposure Assessment. Gewerbestrasse 11, 6330 Cham, Switzerland: Springer Nature Switzerland AG. pp. 253-254. ISBN 978-3-319-96148-4

[63] Sieke, C. (2020): Principles of Consumer Exposure Assessment for Pesticide Residues. In: Heinemeyer, G. (Szerk.) The Practice of Consumer Exposure Assessment. Gewerbestrasse 11, 6330 Cham, Switzerland: Springer Nature Switzerland AG. pp. 315-322. ISBN 978-3-319-96148-4

[64] Kuiper-Goodman, T. (1990): Uncertainties in the risk assessment of three mycotoxins: aflatoxin, ochratoxin, and zearalenone. Canadian Journal of Physiology and Pharmacology 68 pp. 1017-1024. https://doi.org/10.1139/y90-155

[65] Szeitzne Szabo, M., Bíró, L., Bíró, Gy., Sali, J. (2011): Dietary survey in Hungary, 2009. Part I. Macronutrients, alcohol, caffeine, fibre. Acta Alimentaria 40 (1) pp. 142-152. https://doi.org/10.1556/AAlim.40.2011.1.16

[66] Csizmadia, K., Larnsak, L., Pfaff, N., Sali, J., (2020): Hungarian national food consumption survey on adults. EFSA supporting publication 17 (12)EN‐1981. p. 26.

[67] Csizmadia, K., Larnsak, L., Pfaff, N., Sali, J. (2020): Hungarian national food consumption survey on toddlers and other children. EFSA supporting publication 17 (12)EN‐1982. p. 26. doi: https://doi.org/10.2903/sp.efsa.2020.EN‐1982

[68] Ferrari, S.L.P., Fumes, G. (2017): Box–Cox symmetric distributions and applications to nutritional data. AStA Adv. Stat. Anal. 101 pp. 321–344. doi: https://doi.org/10.1007/s10182-017-0291-6

[69] Rigby, R.A., Gillian, M. D. S., Heller, Z., De Bastiani, F. (2019): Continous three parameter distribution on (0,∞). Distributions for Modelling Location, Scale and Shape: Using GAMLSS in R. 1 ed.: Chapman and Hall/CRC. ISBN 9780367278847

[70] Horváth, G., Gerlei, Zs., Gervain, J., Lengyel, G., Makara, M., Pár, A., Rókusz, L., Szalay, F., Tornai, I., Werling, K., Hunyady, B. (2018): Diagnosis and treatment of chronic hepatitis B and D. National consensus guideline in Hungary from 22 September 2017. Orvosi Hetilap 159 (1) pp. 24-37. https://doi.org/10.1556/650.2018.31004

[71] Kerekes, K., Bonilauri, P., Serraino, A., Giacometti, F., Piva, S., Zambrini, V., Canever, A., Farkas, Z., Ambrus, A. (2016): An effective self-control strategy for the reduction of aflatoxin M1 content in milk and to decrease the exposure of consumers. Food Additives and Contaminants Part A Chemistry, Analysis, Control, Exposure & Risk Assessment 33 (12) pp. 1840-1849. https://doi.org/10.1080/19440049.2016.1241895

DOI: https://doi.org/10.52091/EVIK-2021/2-3-ENG

Submitted: July 2020 – Accepted: November 2020

¹ University of Debrecen, Faculty of Agricultural and Food Sciences and Environmental Management, Institute of Food Science
² University of Debrecen, Faculty of Agricultural and Food Sciences and Environmental Management, Institute of Animal Science, Biotechnology and Environmental Protection, not independent Faculty of Animal Husbandry
³ University of Debrecen, Doctoral School of Animal Science

Keywords

lactation number, lactation stage, cow’s milk, milk quantity, milk composition, microbiology

2. Introduction

Cow’s milk has a high nutritional value; it contains fats, proteins, carbohydrates, vitamins and minerals, among other things [1]. Examination of the composition of milk is a routine practice on dairy farms for monitoring the hygiene, nutritional and health aspects of dairy cows [2]. The composition of milk can be influenced by a number of factors, such as the number of lactations, the stage of lactation, the season and the feeding technology [3, 4, 5]. Thus, the composition of milk may vary during lactation, and as a result of the interactions of different environmental factors, there may be differences between the different dairy farms [6]. According to Dürr et al., the creation of databases in order to determine the causes and consequences of differences in milk yield and milk composition is of paramount importance. A database should also include records related to these parameters, as well as to lactation-related events and to the individual cows [7].

Due to its nutritional value, high water activity and neutral pH, milk serves as an excellent medium for various microorganisms, which may include pathogenic organisms such as Campylobacter jejuni, Salmonella spp., Staphylococcus aureus, Listeria monocytogenes, Yersinia enterocolitica, etc. [8, 9]. During the primary infection of milk, sick animals themselves are the sources of infection. In case of dairy animals suffering from a systemic disease accompanied by the spread of pathogens, the pathogens may be excreted in the milk. During secondary contamination, the contamination of milk is of environmental origin. Due to improper milking hygiene, milk can be contaminated by the faeces of the animals or the equipment used in milking (milking machines, milk lines, milk storage tanks), among other things [10]. As the milk enters the teat cistern and then the teat canal, it can become infected with various microorganisms, of environmental origin, so the bacterial infection of the first milk jets is remarkably high. At the beginning of milking, it is therefore advisable to separate the first jets of milk from milk that is milked subsequently, and to ensure that they are destroyed [11]. To reduce the number of bacteria in milk, heat treatment is most commonly used. The initial microbiological state of the milk is not only important from a food safety point of view, but can also affect the quality of the dairy products made from it [12].

Coliform bacteria can cause mastitis in dairy animals. Mastitis caused by coliform bacteria can reduce milk production in dairy animals, causing economic losses to dairy farms [13]. The presence of these bacteria in the environment is common, so their presence in food may indicate environmental contamination [14, 15].

Raw milk can be contaminated with S. aureus from a number of sources, such as the environment, milkers’ hands, milking equipment, and so on [16]. The economic damage due to mastitis caused by S. aureus comes from a decrease in the quantity and quality of the milk produced, an increase in the number of somatic cells measured in it, a reduction in the purchase price of lower quality milk, which also leads to a decrease in the turnover of dairy farmers [17]. Prevention is an effective means of controlling S. aureus infection. It can be prevented by adhering to appropriate housing and milking technologies, more frequent fly extermination, pre- and post-disinfection of the teats, the use of disposable udder wipers and regular laboratory testing [18].

External and internal factors can affect the composition of milk, as well as the microbiological state of raw milk. The latter is mostly determined by the hygienic condition of the surfaces that come into direct contact with the milk [19]. Research by Peles et al. found that different husbandry and milking methods at dairy farms with different cow numbers had influencing effects on the microbiological quality of milk [20]. Tessema sought to find a correlation between breed, the age of the animals, the number of lactations and the stage of lactation, as well as the likelihood of S. aureus occurrence.

In his study, he found a significant difference in the prevalence of S. aureus in milk for the two cattle species studied. S. aureus was found in higher proportion of crossbred cows and in older individuals [21]. Examining the milk of different varieties, Bytyqi et al. found no difference in the colony counts obtained [22].

Although fewer Hungarian studies have been conducted on the subject, several international publication have examined whether the number of lactations and the stage of lactation have an influence on the daily milk volume of the animals, the composition of the milk and its microbiological parameters. According to Tessema, there is a significant difference in the prevalence of S. aureus, depending on how many lactations the animals have been through. In his study, he found that S. aureus was more common in the milk of cows that had calved more than twice and produced a positive California Mastitis Test [21]. Tenhagen et al. also found that the incidence of S. aureus increases with the age of the animals [23]. This may be related to the fact that the milking machine may damage the teats during milking, allowing microorganisms to enter the udder from the environment [24]. Another possible reason is that the health status of dairy animals may deteriorate with advancing age, which may have an adverse effect on the somatic cell count of milk [25].

The objective of our study was to determine at a Hungarian large-scale dairy farm whether there is a difference in the daily milk yield of primiparous and multiparous cows, as well as cows at different stages of lactation, and also in the composition (fat and protein content) and microbiological parameters (somatic cell count, total plate count, coliform and S. aureus count) of primiparous and multiparous cows, as well as cows at different stages of lactation.

3. Materials and methods

3.1. Place and time of sampling

A dairy farm in Hajdú-Bihar county (Hungary) was the site of our studies. 440–450 Holstein-Friesian cows are milked on the farm. The farm uses deep litter livestock-keeping and monodiet feeding. Milking takes place in a milking parlor, and no post-disinfection is carried out after milking.

The data on the daily amount of milk, fat and protein content and somatic cell count used for the calculations are derived from the milking results, i.e., from the examination results of the milk samples collected monthly by the Állattenyésztési Teljesítményvizsgáló Kft. All milking results of 38 individuals between May 2015 and January 2020 were used in the calculations. During the calculations, data on the first lactation of the cows (n=387), data on the 2^nd to 5^th lactation of the cows (n=446), as well as data on the early (under 100 days; n=275), middle (100 to 200 days; n=249) and late (over 200 days; n=309) stages of the lactation of the cows were summarized for the abovementioned time period.

Microbiological tests were performed between May 2018 and October 2019. A total of 62 milk samples were taken from 38 randomly selected, clinically healthy individuals for the microbiological tests. According to which lactation cycle the animals were in and in which stage of lactation during the sampling, the samples taken from the individual cows were classified as follows: 26 samples were taken from 15 primiparous cows and 36 samples were taken from 23 multiparous (2 to 5 calvings) Holstein-Friesian cows. Of the total 62 samples taken, 23 were taken in the early stage of lactation, 21 samples in the middle stage of lactation and 18 samples were taken from cows in the late stage of lactation.

Following pre-disinfection of the teats, wiping them dry using paper towels and milking of the first milk jets, samples were taken from all four udder quarters into sealable sterile plastic sampling vessels with a capacity of 50 ml. The vessels were transported in coolers to the microbiology laboratory of the Institute of Food Science of the University of Debrecen within two hours of sampling. Samples were processed within 24 hours of sampling.

3.2. Microbiological tests

Preparation of the milk samples and the subsequent microbiological tests were performed according to the procedure described by Petróczki et al. [26]. Sample preparation was carried out according to standard MSZ EN ISO 6887-1:2017 [27], the samples were stored at 4 °C until the beginning of the analysis, and they were homogenized by shaking before processing. To prepare the dilution series, peptone water was used which was prepared by dissolving 8.5 g of sodium chloride (VWR International Kft., Hungary) and 1.0 g of peptone (Merck Kft., Hungary) in 1,000 ml of distilled water.

After weighing the appropriate amounts (9 ml each) into test tubes, the diluent was sterilized in a pressure cooker at 120 °C for 30 minutes, then it was cooled and the decimal dilution series was prepared.

The total plate count was determined according to standard MSZ EN ISO 4833-1:2014 [28], which prescribes the use of a tryptone-glucose-yeast agar (Plate Count Agar, PCA) culture medium (Biolab Zrt., Hungary) supplemented with milk powder. After performing the prescribed plate casting method, the plates were incubated at 30 °C for 72 hours.

The amount of coliform bacteria was determined by the plate casting method according to standard ISO 4832:2006 [29], using sterile crystal violet-bile-lactose agar (Violet Red Bile Lactose, VRBL, Biolab Zrt., Hungary). The incubation period was 24 hours at 30 °C.

The determination of S. aureus was performed according to standard MSZ EN ISO 6888-1:2008 [30] by the spreading method, for which Baird-Parker agar (BPA, Biolab Zrt., Hungary) supplemented with egg yolk-tellurite emulsion (LAB-KA Kft., Hungary). The incubation period was 48 hours at 37 °C. S. aureus was isolated from other Staphylococcus species using a latex agglutination test (Prolex Staph Xtra Kit, Ferol Kft., Hungary).

3.3. Statistical analysis

For the analysis of our experimental results, for the calculation of descriptive statistics, for the logarithmic transformation of the amount of microorganisms and for performing t-tests and analysis of variance, the SPSS v.22.0 [31] software was used.

In the case of lactation number, variables were compared using unpaired t-tests and non-parametric Mann-Whitney tests, while in the case of lactation stage, the comparison was performed by one factor analysis of variance and non-parametric Kruskal-Wallis tests. Since the total plate count, the coliform count and the somatic cell count were not found to be normally distributed variables in several cases, a logarithmic transformation was used for these parameters. During statistical analyses, a value of P<0.05 was considered to indicate a significant difference.

4. Results

4.1. Effect of lactation number on milk yield, raw milk composition and its microbiological parameters

Mean and standard deviation values of the daily milk yield of primiparous and multiparous cows, as well as those of the fat and protein content, somatic cell count, total plate count, coliform and S. aureus count of the milk produced by them are shown in Table 1. Cows selected for the study were classified into groups based on the number of lactations (and calvings) into groups of primiparous and multiparous individuals. The average milk yield was 25.67 kg/day for primiparous cows, while in the case of multiparous cows it was 31.04 kg/day. The difference is significant (P<0.05), which means that our experiments confirmed the findings of Bondan et al. and Yang et al., that the daily amount of milk produced by multiparous cows is higher than that of primiparous cows [5, 32]. In the case of crossbred Holstein-Friesian cows, Gurmessa and Melaku studied, inter alia, the effect of calving number on milk yield, but found no difference between the daily milk yield of primiparous and multiparous cows [33]. Pratap and his research group also found no difference in the average daily milk yield of primiparous and multiparous cows (6.43±1.39 and 5.89±2.37 l/day, respectively) [34].

The average fat content of milk during the first lactation of cows selected during our research was 3.74±0.40%, while the average fat content of milk samples taken during the second or later lactation was 3.75±0.36%. The difference was not found to be significant (P>0.05). Similarly to our own results, no difference was found between the average fat content of the milk of primiparous and multiparous crossbred Holstein-Friesian cows by Gurmessa and Melaku, or Pratap et al. [33, 34]. On the other hand, it was found by Bondan and his group that the number of lactations did have an effect on the fat content of milk in Holstein-Friesian cows. While the fat content was 3.47±0.67% in the case of cows lactating for the first time, it was 3.43±0.68% and 3.41±0.67% for cows lactating for the 2^nd or 3^rd time and for cows that had calved at least 4 times, respectively [5]. In their study in Sudan, Shuiep and his working group examined changes in the fat content of milk with the number of lactations in local and crossbred cows. In the case of local cows, cows in their fourth lactation period had a lower milk fat content (4.82±0.55%) than cows with fewer calvings (1:5.16±0.32; 2:5.22±0.34; 3:5.14±0.34). No difference was observed in the case of crossbred cows [6]. In contrast, Yang et al. found in their research that Holstein-Friesian cows lactating for the first time had a lower milk fat content (3.88%) [32]. Based on the varied results of our study and other literature references, we came to the conclusion that the fat content of milk may be influenced by other factors besides the lactation number.

During the first lactation of the Holstein-Friesian cows selected by us, the average protein content of the milk was 3.24±0.19%, while the average protein content of milk samples taken during the second or later lactation was 3.31±0.16%, with the difference being not significant (P>0.05). This is in line with the findings of Gurmessa and Melaku, as well as those of Pratap et al., as these authors did not find any difference between the average protein content of the milk of primiparous and multiparous cows [33, 34]. In contrast, the research group of Bondan found that the lactation number affected the protein content of milk in Holstein-Friesian cows. While the protein content was 3.24±0.37% in the case of cows lactating for the first time, it was 3.23±0.38% and 3.19±0.37% in the case of cows lactating for the 2^nd or 3^rd time and for cows that had calved at least 4 times, respectively [5]. Changes in the protein content of milk with the lactation number was also studied by the research group of Shuiep. In the case of local cows, the protein content of milk was lower in the case of cows in their fourth lactation period (3.67±0.19%), than in the case of cows with fewer calvings (1:4.01±0.11; 2:3.82±0.12; 3:3.84±0.12). No difference was observed in crossbred cows [6].

According to our results, the average somatic cell count [242.2×10³ (5.12±0.42 lg) cell/ml] in the milk of primiparous cows is less (P<0.05) then in the milk of multiparous cows [356.3×10³ (5.39±0.39 lg) cell/ml]. In none of the cases did the averages exceed the limit value [M=400.0×10³ (5.60 lg) cfu/ml] laid down in Regulation (EC) No 853/2004 [35]. The results obtained at the Hungarian dairy farm are in line with relevant literature data. According to Mikó et al., as the lactation number increases, the somatic cell count of milk may also increase [25]. This finding was also found to be true by Yang and his group [32]. Sheldrake et al. observed a significant relationship between the calving number and the somatic cell count. They found that with the increase in calving number, there was a smaller change in the udder quarters of healthy animals, however, in the case of udder quarters infected with S. aureus, the somatic cell count increased significantly [36]. The research team of Bondan found that as the lactation number of the Holstein cows studied increased, so did the somatic cell count in milk. While the average somatic cell count in primiparous cows was 4.83±1.73 lg cell/ml, it was 5.31±1.72 és 5.84±1.62 lg cell/ml in cows that had calved 2 or 3 times and cows that calved 4 or more times, respectively [5].

In the milk samples taken from Holstein-Friesian cows that only calved once, i.e., cows in their first lactation period, the average total plate count was 5.1×10³ (3.36±0.58 lg) cfu/ml, while it was 4.6×10³ (3.30±0.59 lg) cfu/ml in the milk samples taken from multiparous cows, however, the difference was not significant (P>0.05).

However, the average coliform count [1.1×10³ (1.35±1.20 lg) cfu/ml] in the milk of multiparous cows was more (P<0.05) than the average coliform count [1.1×10¹ (0.65±0.61) lg) cfu/ml] measured in the milk of primiparous cows. Tenhagen et al. also included clinically healthy cows in their study, which found that although coliform bacteria were found in higher quantities in the milk of multiparous cows, the difference was not significant [23].

S. aureus was present in only one of the 26 milk samples taken from primiparous cows, in an amount of 5.0×10¹ (1.70 lg) cfu/ml, while it could be detected in eight of the 36 milk samples taken from multiparous cows. The mean S. aureus count in these samples was 1.5×102 (1.92±0.56 lg) cfu/ml. The values do not exceed the limit value [M=5.0×10² (2.70 lg) cfu/ml] specified by EüM decree 4/1998 (XI. 11.) [37]. In the case of the individuals in whose milk S. aureus could be detected during the microbiological tests, the average somatic cell count was between 44.3×10³ (4.65 lg) cell/ml and 357.2×10³ (5.55 lg) cell/ml on the basis of milking data. In his study, Tessema also found that the prevalence of S. aureus was higher in the case of multiparous cows (i.e., who had calved more than twice) (who produced a positive California Mastitis Test) [21]. According to Tenhagen et al., the incidence of S. aureus increases with the age and lactation stage of the animals [23].

^{a, b} The values marked with different letters in the same rows of the table differ significantly (P<0.05)

4.2. Effect of lactation stage on milk yield, raw milk composition and its microbiological parameters

Mean and standard deviation values of the daily milk yield of cows in the early, middle and late stages of lactation, as well as those of the fat and protein content, somatic cell count, total plate count, coliform and S. aureus count of the milk produced by them are shown in Table 2. The cows selected for the study were classified into groups of early, middle and late lactation stage individuals, based in their stage of lactation. When examining the changes in the daily milk yield of the individuals with the stage of lactation, the finding in the literature that the daily milk yield decreases towards the end of lactation was confirmed. While the average daily milk yield of cows in the early stage of their lactation was 32.10±4.73 kg/day, that of cows in the middle stage of their lactation was 29.08±5.09 kg/day, while that of cows in the late stage of lactation was 23.36±3.63 kg/day. The difference was significant (P<0.05). Bondan and his group came to a similar conclusion in their study. The average milk yield between days 6 and 60 of the lactation of the Holstein-Friesian cows studied by them was 29.4±8.72 l/cow/day; for cows between days 61 and 120 of their lactation, it was 29.2±8.66 l/cow/day; for cows between lactation days 121 and 220, the milk yield was 26.2±8.01 l/cow/day, and at the end of the lactation (>220 days), it was 22.0±7.49 l/cow/day [5]. Gurmessa and Melaku, as well as Pratap et al. also found that the daily milk yield of the animals was higher at the beginning of lactation (6.81±1.45 l/day) than at the end of lactation (5.48±0.05 l/day). In their studies, the daily milk yield of crossbred Holstein-Friesian cows in the middle stage of their lactation was the highest (7.17±0.05 liter) [33, 34]. According to Auldist et al., the effect of lactation stage on milk yield (e.g., a decrease) may be due to a change in the number and activity of the secretory cells within the mammary gland because of physiological reasons [2].

The fat content of the milk of the individuals studied shows a change as we progress towards the end of lactation. The average fat content of cows in the early and middle stages of lactation was 3.65±0.43% and 3.59±0.41%, respectively, which were lower (P<0.05) than the fat content of the milk of cows in the late stage of lactation (3.99±0.47%). The increase in milk fat concentration at the end of lactation may be associated with a decrease in milk yield as lactation progresses, as a decrease in milk yield may have a concentrating effect on milk composition [2]. The fat content of milk also varies at the three stages of lactation according to the publication of Gurmessa and Melaku. For cows in the early or late stages of lactation, the average fat content of the milk (4.46±1.44% and 4.46±1.44%, respectively) was significantly higher than that of cows in the middle stage of lactation (3.70±0.89%) [33]. The publication of Bondan et al. states that the fat content of milk at the end of lactation (>200 days) is higher (3.55±0.67%) than at earlier stages of lactation. However, they also found that the measured average fat content (3.30±0.66%) was lowest between days 61 and 120 of the lactation of the cows. Between days 6 and 60, and between days 121 and 220, average fat contents of 3.40±0.65% and 3.40±0.66% were measured, respectively [5]. Shuiep et al. studied changes in the fat content of milk of local and crossbred cows in Sudan with lactation stage. In the case of the local breed, there was no difference in the fat content between the early (5.31±0.51%), middle (4.67±1.56%) and late (5.28±0.75%) stages of lactation. However, in the case of crossbred cows, the fat content was higher at the end of lactation (4.45±1.43%) than at the beginning of the lactation (3.41±1.09%) or in the middle stage of lactation (3.33±1.05%) [6].

Similar to the fat content, a change in protein content can also be observed as we progress towards the end of lactation. The average protein content measured at the beginning of lactation was 3.08±0.15%, it was 3.20±0.19% in the middle of lactation and 3.56±0.20% at the end of lactation, the difference being significant (P<0.05). Bondan et al. came to a similar conclusion: a higher protein content (3.41±0.36%) was measured at the end of lactation (>200 days) than at earlier stages of lactation. It was also found that the measured average protein content (3.03±0.31%) was the lowest between days 61 and 120 of the lactation of the cows. The measured average protein content between days 6 and 60 of the lactation and between days 121 and 220 were 3.05±0.36% and 3.18±0.32%, respectively [5]. Gurmessa and Melaku, as well as the research group of Pratap did not find any difference in protein content between cows in the early (3.55±1.43%), middle (3.17±0.15%) and late (3.33±0.16%) stages of their lactation [33, 34]. In the case of local and crossbred cows in Sudan, changes in milk protein content with the lactation stage were investigated by Shuiep et al. In the case of the local breed, the protein content of the milk of the animals was higher at the beginning (3.87±0.52%) and in the middle (3.91±0.18%) of lactation than at the end of lactation (3.67±0.17%). In the case of crossbred cows, there was no difference in protein content at the beginning (3.67±0.17%), middle (3.69±0.16%) and end (3.63±0.22%) of lactation [6].

The average somatic cell count in the early stage of lactation of the cows selected for our research was 195.1×10³ (5.07±0.43 lg) cell/ml, it was 370.6×10³ (5.28±0.50 lg) cell/ml in the middle of lactation, and it was 336.4×10³ (5.33±0.41 lg) cell/ml in the late stage of lactation. The somatic cell count was higher (P<0.05) in the milk of individuals in the late stage of lactation than in the case of cows at the beginning of lactation. As lactation progresses, somatic cell counts also show an increasing trend according to Bondan et al. While in the milk of Holstein-Friesian cows between days 6 and 60 of their lactation the average somatic cell count was 4.79±1.90 lg cell/ml, between days 61 and 120 it was 4.89±1.90 lg cell/ml, between days 121 and 220 it was 5.21±1.75 lg cell/ml, and in the case of lactations lasting more than 220 days, this parameter was 5.53±1.53 lg cell/ml in the milk of the cows studied [5].

Total plate counts were also determined during the different stages of lactation of Holstein-Friesian cows. At the beginning of lactation, the average total plate count was 6.8×10³ (3.42±0.67 lg) cfu/ml, in the middle of lactation it was 4.4×10³ (3.39±0.46 lg) cfu/ml, while at the end of lactation it was 2.7×10³ (3.13±0.56 lg) cfu/ml. No significant difference was found between the total plate count values obtained (P>0.05).

Regarding the number of coliform bacteria, the highest count [1.3×10³ (1.30±1.23 lg) cfu/ml] was measured in samples taken at the beginning of the lactation of the cows, while the lowest average coliform count [2.2×10¹ (0.76±0.79 lg) cfu/ml] was detected in samples taken in the middle stage of lactation. The average coliform count in the samples taken at the end of lactation was 1.50×10² (0.90±0.94 lg) cfu/ml. There was no significant difference between the results (P>0.05).

S. aureus was present in a total of 9 samples (14.52%) out of 62 individual milk samples, with an average count of 1.4×10² (1.89±0.53 lg) cfu/ml. In the six samples (9.68%) taken from cows in the early stage of lactation, the average S. aureus count was 1.2×10² (1.81±0.56 lg) cfu/ml, for the single animal in the middle of lactation (1.61%) the S. aureus count was 8.2×10¹ (1.91 lg) cfu/ml, while for the two cows at the end of lactation (3.23%) it was 2.4×10² (2.15±0.70 lg) cfu/ml.

^{a, b, c} The values marked with different letters in the same rows of the table differ significantly (P<0.05)

5. Conclusions

In the course of our research work in a Hungarian dairy farm, it was proven that, under large-scale husbandry conditions, for primiparous cows the average daily milk yield is significantly lower (P<0.05) than in the case of multiparous cows. This is probably due to the fact that cows in their first lactation need more amino acids and fat to develop their bodies compared to animals that have already undergone several lactation periods [38].

Regarding the fat and protein content of the milk, there was no significant difference between primiparous and multiparous cows. Since one can find in the literature findings that both support and contradict our results, we assume that the fat and protein contents of milk are also influenced by other factors (e.g., breed, season, etc.), but mapping out of these factors was not the objective of our study. For example, Shuiep et al. reported in their paper different results for two different cattle breeds when examining the changes in the fat and protein contents of milk with the lactation number [6].

Based on our measurement results, we found that the somatic cell count, as well as the coliform count and the S. aureus count were significantly higher (P<0.05) in the milk samples taken from multiparous cows compared to the milk samples taken from primiparous cows. The reason why microorganisms can be measured in larger amounts in milk samples taken from multiparous cows is probably that the teats may have been damaged during the lactations (due to, for example, the milking machine), which may have increased the chances of microorganisms entering the udder [24]. Another reason may be that as the age progressed or the number of lactations increased, the condition of the dairy animals may have deteriorated, which may have had an adverse effect, for example, on the somatic cell count of the milk [25].

It was also proven that, in the stages of lactation, a decreasing trend in daily milk yield can be observed, but the fat and protein contents show an increase. This is presumably due to the concentrating effect of the decreasing milk yield.

There was no difference in milk samples taken at different stages of lactation in terms of total plate count and coliform count, however, in the late stage of lactation, the somatic cell count showed an increase.

6. Acknowledgment

This publication was supported by the project EFOP-3.6.3-VEKOP-16-2017-00008. The project was supported by the European Union and co-financed by the European Social Fund.

We are grateful to the management and staff of the dairy farm for their helpful assistance.

7. Literature

[1] Hill B., Smythe B., Lindsay D., Shepherd J (2012): Microbiology of raw milk in New Zealand. International Journal of Food Microbiology 157 2 pp. 305-308. https://doi.org/10.1016/j.ijfoodmicro.2012.03.031

[2] Auldist M. J., Walsh B. J., Thomson N. A. (1998): Seasonal and lactational influences on bovine milk composition in New Zealand. Journal of Dairy Research 65 (3) pp. 401-411. https://doi.org/10.1017/S0022029998002970

[3] Heck J. M. L., van Valenberg H. J. F., Dijkstra J., van Hooijdonk A. C. M. (2009): Seasonal variation in the Dutch bovine raw milk composition. Journal of Dairy Science 92 (10) pp. 4745-4755. https://doi.org/10.3168/jds.2009-2146

[4] Lambertz C., Sanker C., Gauly M. (2014): Climatic effects on milk production traits and somatic cell score in lactating Holstein-Friesian cows in different housing systems. Journal of Dairy Science 97 (1) 3 pp. 19-329. https://doi.org/10.3168/jds.2013-7217

[5] Bondan C., Folchini J. A., Noro M., Quadros D. L., Machado K. M., González F. H. D. (2018): Milk composition of Holstein cows: a retrospective study. Ciência Rural 48 (12) pp. 1-8. https://doi.org/10.1590/0103-8478cr20180123

[6] Shuiep E. S., Eltaher H. A., El Zubeir I. E. M. (2016): Effect of Stage of Lactation and order of Parity on Milk Composition and Daily Milk Yield among Local and Crossbred Cows in South Darfur State, Sudan. SUST Journal of Agricultural and Veterinary Sciences (SJAVS) 17 (2) pp. 86-99.

[7] Dürr J. W., Ribas N. P., Costa C. N., Horst J. A., Bondan C. (2011): Milk recording as an indispensable procedure to assure milk quality. Revista Brasileira Zootecnia 40 pp. 76-81.

[8] Quigley L., O’sullivan O., Beresford T. P., Ross R. P., Fitzgerald G. F., Cotter P. D. (2011): Molecular approaches to analysing the microbial composition of raw milk and raw milk cheese. International Journal of Food Microbiology 150 (2-3) pp. 81-94. https://doi.org/10.1016/j.ijfoodmicro.2011.08.001

[9] Claeys W. I., Cardoen S., Daube G., De Block J., Dewettinck K., Dierick K., De Zutter L., Huyghebaert A., Imberechts H., Thiange P., Vandenplas Y., Herman L. (2013): Raw or heated cow milk composition: Review of risks and benefits. Food Control 31 (1) pp. 251-262. https://doi.org/10.1016/j.foodcont.2012.09.035

[10] Laczay P., Lehel J., Lányi K., László N. (2016): A nyers tejben potenciálisan jelen levő kórokozók közegészségügyi jelentősége. Magyar Állatorvosok Lapja 138 pp. 231-242.

[11] Laczay P. (2008): Élelmiszer-higiénia - Élelmiszerlánc-biztonság. Mezőgazda Kiadó, Budapest.

[12] Cilliers F. P., Gouws P. A., Koutchma T., Engelbrecht Y., Adriaanse C., Swart P. (2014): A microbiological, biochemical, and sensory characterisation of bovine milk treated by heat and ultraviolet (UV) light for manufacturing Cheddar cheese. Innovative Food Science & Emerging Technologies 23 pp. 94-106. https://doi.org/10.1016/j.ifset.2014.03.005

[13] Mbuk E. U., Kwaga J. K. P., Bale J. O. O., Boro L. A., Umoh J. U. (2016): Coliform organisms associated with milk of cows with mastitis and their sensitivity to commonly available antibiotics in Kaduna State, Nigeria. Journal of Veterinary Medicine and Animal Health 8 (12) pp. 228-236. https://doi.org/10.5897/JVMAH2016.0522

[14] Altalhi A. D., Hassan S. A. (2009): Bacterial quality of raw milk investigated by Escherichia coli and isolates analysis for specific virulence-gene markers. Food Control 20 (10) pp. 913-917. https://doi.org/10.1016/j.foodcont.2009.01.005

[15] Mhone T. A., Matope G., Saidi P. T. (2011): Aerobic bacterial, coliform, Escherichia coli and Staphylococcus aureus counts of raw and processed milk from selected smallholder dairy farms of Zimbabwe. International Journal of Food Microbiology 151 (2) pp. 223-228. https://doi.org/10.1016/j.ijfoodmicro.2011.08.028

[16] Markus G. (2001): A tejelő tehenek tőgygyulladása III. MezőHír. 9

[17] Ózsvári L., Fux A., Illés B. CS., Bíró O. (2003): A Staphylococcus aureus tőgygyulladás által okozott gazdasági veszteségek számszerűsítése egy nagyüzemi holstein-fríz tehenészetben. Magyar Állatorvosok Lapja 125 pp. 579-584.

[18] Rosengren Å., Fabricius A., Guss B., Sylvén S., Lindqvist R (2010): Occurrence of foodborne pathogens and characterization of Staphylococcus aureus in cheese produced on farm-dairies. International Journal of Food Microbiology 144 (2) pp. 263-269. https://doi.org/10.1016/j.ijfoodmicro.2010.10.004

[19] Anderson D., Dulmage D., McDougall M., Séguin G. (2003): General guidelines for effective dairy equipment cleaning. https://www.milk.org/Corporate/pdf/Farmers-UdderEquipmentCleaning.pdf (Hozzáférés: 21. 02. 2020.)

[20] Peles F., Máthéné Sz. Zs., Béri B., Szabó A. (2008): A tartástechnológia hatása a nyers tej mikrobiológiai állapotára. Agrártudományi Közlemények 31 pp. 67-75. https://doi.org/10.34101/actaagrar/31/3009

[21] Tessema F. (2016): Prevalence and Drug Resistance Patterns of Staphylococcus Aureus in Lactating Dairy Cow’s Milk in Wolayta Sodo, Ethiopia. EC Veterinary Science 2 (5) pp. 226-230.

[22] Bytyqi H., Vehapi I., Rexhepi S., Thaqi M., Sallahi D., Mehmeti I. (2013): Impact of Bacterial and Somatic Cells Content on Quality Fresh Milk in Small-Scale Dairy Farms in Kosovo. Food and Nutrition Sciences 4 (10) pp. 1014-1020. https://doi.org/10.4236/fns.2013.410132

[23] Tenhagen B. A., Köster G., Wallmann J., Heuwieser W. (2006): Prevalence of Mastitis Pathogens and Their Resistance Against Antimicrobial Agents in Dairy Cows in Brandenburg, Germany. Journal of Dairy Science 89 (7) pp. 2542-2551. https://doi.org/10.3168/jds.S0022-0302(06)72330-X

[24] Hamann J., Mein G. A., Wetzel S. (1993): Teat tissue reactions to milking: effects of vacuum level. Journal of Dairy Science 76 pp. 1040-1046. https://doi.org/10.3168/jds.S0022-0302(93)77432-9

[25] Mikó E., Baranyi A., Gráff M. (2015): Analysis of somatic cells in cow’s milk. Lucrări Ştiinţifice 17 (1) pp. 290-293.

[27] Magyar Szabványügyi Testület (MSzT) (2017): Élelmiszerek és takarmányok mikrobiológiája. A vizsgálati minták, az alapszuszpenzió és a decimális hígítások elkészítése mikrobiológiai vizsgálathoz. 1. rész: Az alapszuszpenzió és a decimális hígítások elkészítésének általános szabályai. Magyar szabvány MSZ EN ISO 6887-1:2017. Magyar Szabványügyi Testület, Budapest.

[26] Petróczki F. M., Tonamo T. A., Béri B., Peles F. (2019): The effect of breed and stage of lactation on the microbiological status of raw milk. Acta Agraria Debreceniensis 1 pp. 37-45. https://doi.org/10.34101/actaagrar/1/2367

[28] Magyar Szabványügyi Testület (MSzT) (2014): Az élelmiszerlánc mikrobiológiája. Horizontális módszer a mikroorganizmusok számlálására. 1. rész: Telepszámlálás 30 °C-on lemezöntés módszerrel. Magyar szabvány MSZ EN ISO 4833-1:2014. Magyar Szabványügyi Testület, Budapest.

[29] International Organization for Standardization (ISO) (2006): Microbiology of food and animal feeding stuffs - Horizontal method for the enumeration of coliforms - Colony-count technique. ISO 4832:2006

[30] Magyar Szabványügyi Testület (MSzT) (2008): Élelmiszerek és takarmányok mikrobiológiája. Horizontális módszer a koagulázpozitív sztafilokokkuszok (Staphylococcus aureus és más fajok) számának meghatározása. 1. rész: Baird-Parker-agar táptalajos eljárás. Magyar szabvány MSZ EN ISO 6888-1:2008. Magyar Szabványügyi Testület, Budapest.

[31] SPSS (2013): SPSS 22.0 for Windows. SPSS Inc., Chicago, IL, USA. Copyright © SPSS Inc., 1989-2013.

[32] Yang L., Yang Q., Yi M., Pang Z. H., Xiong B. H. (2013): Effects of seasonal change and parity on raw milk composition and related indices in Chinese Holstein cows in northern China. Journal of Dairy Science 96 (11) pp. 6863-6869. https://doi.org/10.3168/jds.2013-6846

[33] Gurmessa J., Melaku A. (2012): Effect of Lactation Stage, Pregnancy, Parity and Age on Yield and Major Components of Raw Milk in Bred Cross Holstein Friesian Cows. World Journal of Dairy & Food Sciences 7 (2) pp. 146-149.

[34] Pratap A., Verma D. K., Kumar P., & Singh A. (2014): Effect of Pregnancy, Lactation Stage, Parity and Age on Yield and Components of Raw Milk in Holstein Friesian Cows in organized Dairy form in Allahabad. IOSR Journal of Agriculture and Veterinary Science (IOSR-JAVS) 7 (2) pp. 112-115. https://doi.org/10.9790/2380-0721112115

[35] 853/2004/EK: Az Európai Parlament és a Tanács 853/2004/EK rendelete az állati eredetű élelmiszerek különleges higiéniai szabályainak megállapításáról

[36] Sheldrake R. F., Hoare R. J. T., McGregor G. D. (1983): Lactation Stage, Parity, and Infection Affecting Somatic Cells, Electrical Conductivity, and Serum Albumin in Milk. Journal of Dairy Science 66 pp. 542-547. https://doi.org/10.3168/jds.S0022-0302(83)81823-2

[37] 4/1998. (XI. 11.) EüM rendelet az élelmiszerekben előforduló mikrobiológiai szennyeződések megengedhető mértékéről

[38] Oltner R., Emanuelson M., Wiktorsson H. (1985): Urea concentrations in milk in relation to milk yield, live weight, lactation number and amount and composition of feed given to dairy cows. Livestock Production Science 12 (1) pp. 47-57. https://doi.org/10.1016/0301-6226(85)90039-9

DOI: https://doi.org/10.52091/EVIK-2021/2-4-ENG

Submitted: July 2020 – Accepted: December 2020

¹ Hungarian Dairy Research Institute Ltd, Mosonmagyaróvár
² Széchenyi István University, Wittmann Antal Multidisciplinary Doctoral School in Plant, Animal, and Food Sciences, Mosonmagyaróvár
³ Széchenyi István University, Faculty of Agricultural and Food Sciences, Department of Food Science, Mosonmagyaróvár

Keywords

nosocomial infection, Serratia species, Serratia marcescens, pathogen, prodigiosin, pigment, polymerase chain reaction (PCR), food diagnostics

2. Introduction and literature review

Nowadays, the impeccable quality and long shelf life of foods is a basic requirement of consumers. Accordingly, there is a growing demand for ever faster, more accurate and more reliable food diagnostic procedures. In this context, molecular diagnostic methods are gaining ground, for example in the rapid detection of pathogenic microorganisms. Polymerase chain reaction (PCR)-based diagnostic kits suitable for the identification of pathogenic microbes are produced by many manufacturers, and these are also used successfully in Hungarian food testing laboratories. These molecular biological tests are mainly suitable for the detection of microbes hose presence poses a high risk to public health (e.g., Escherichia coli, Salmonella Typhimurium, Listeria spp.). Less attention is paid to pathogens that are not required to be tested by law, such as Serratia species present in raw and pasteurized milk.

Serratia species are found in many places in our environment [1]. They are saprophytes or opportunistic pathogens [2], facultative anaerobic, biofilm-forming organisms [1, 3]. S. marcescens grows particularly well in phosphorus-containing environments (e.g., soaps, shampoos) and is also resistant to certain disinfectants [4, 5], so it can cause various nosocomial diseases [6, 7, 8]. Increasing antibiotic resistance of S. marcescens has also been reported in the literature [8, 9, 10]. The bacterium therefore survives and grows easily, so it may find its way into foods under inadequate hygienic conditions. Presumably, it can enter drinking milk as a result of violating hygiene rules, it can grow there and degrade the quality of food [1, 11, 12]. For some species, spoilage is indicated by a characteristic red hue.

In the case of the Hungarian dairy sector, accurate data are not available on the extent of the prevalence of Serratia species and S. marcescens, and on which species cause the infections and degrade milk quality. Nor is there a Hungarian survey on the extent of Serratia contamination of dairy farms. With the exception of a few publications, the available information on the exposure of the dairy industry to Serratia is also lacking at the international level. Such exceptions are a scientific article on the epidemic of mastitis caused by S. marcescens at Finnish diary farms [1], and an older report discussing the role played by pigment-forming Serratia species in mastitis [13].

The following Serratia species may be responsible for the red discoloration of milk: S. marcescens, S. rubidaea, S. plymuthica and S. nematodiphila (Table 1). According to their incidence, S. marcescens is of greater importance. Their characteristic pigment is the red prodigiosin, a water-insoluble secondary metabolite that is produced under specific environmental conditions [14, 15, 16, 17] (Figure 1). The typical red colonies appearing on the culture medium alone do not provide sufficient information to identify Serratia, as certain species of many other genera, not belonging to Enterobacteria, may also produce prodigiosin [14, 18].

There is currently no ISO standard for the detection of Serratia species in foods. In their 2006 book chapter [9], Grimont and Grimont discuss the characteristics of the genus Serratia, as well as aspects of their isolation and identification. However, identification by classical microbiological methods is rather cumbersome and often ineffective due to the inhibitory effect of the accompanying flora, despite the fact that the pink discoloration of the milk sample is clearly visible to the naked eye. Although culture media are available for the selective growth of the bacterium [47], in practice their use does not provide a satisfactory solution. In addition, conventional methods are time and labor intensive.

There are commercially available rapid methods for the determination of S. marcescens, for example the miniaturized test kit from bioMérieux called Rapid ID 32 E, which satisfies the requirements of standard ISO 7218 [48]. However, a colony growing on a culture medium is required to perform the test. Diagnostic tests based on the PCR method, as mentioned before, could provide a solution to overcome the difficulties of detection. At present, however, only the Genesig product of Primerdesign can be mentioned as a molecular diagnostic kit for the detection of S. marcescens [49].

The literature relevant for the food industry and, in particular, the dairy industry, is rather poor on the detection of Serratia species, including S. marcescens, by either endpoint PCR or real-time PCR methods. Hejazi et al. [50] carried out the serotyping of S. marcescens by the RAPD-PCR technique. Serological samples from patients in need of hospital care were used in their study. Iwaya et al. [6] also tested blood samples for S. marcescens strains using a real-time PCR method. Zhu et al. [51] performed molecular characterization of S. marcescens strains by RFLP and PCR methods, while Joyner et al. [2] detected S. marcescens strains in marine and other aquatic environmental samples (e.g., coral mucus, sponge pore water, sediment, sewage, wastewater and diluted wastewater) by real-time PCR. A study of Bussalleu and Althouse, published in 2018, reports a conventional endpoint PCR technique suitable for the identification of S. marcescens that effectively detects the presence of the microorganism in wild boar semen [52].

Our goal was the set up a classical PCR method suitable for the detection of S. marcescens in milk. The significance of our work lies in the fact that PCR-based methods described in the literature and the primers used were analyzed, then the procedure deemed appropriate was adopted to food hygiene analytical practice. In our experiments, qualitative determination of the possible S. marcescens contamination underlying the discoloration of factory, raw and pasteurized milk samples was performed.

3. Materials and methods

3.1. In silico studies

Based on the literature, three primer pairs (Table 2) were selected, which were evaluated by computer modeling, by so-called in silico analysis, as well as in vitro experiments in order to find the most suitable one for subsequent PCR assays.

In our in silico studies, the specificity of the a primer sequences was verified by comparison with a DNA database (NCBI BLAST) [54]. Comparison with the database allows for homology search (“blasting”). Following this, the suitability of the primers, i.e., whether a possible PCR reaction takes place with the selected genomes, was tested with a molecular biology software (SnapGene 5.1.5.) [55]. In the latter case, positive and negative control genomes were downloaded from the NCBI database, and then the SnapGene software was used, in an in silico way, to investigate whether the PCR reaction would take place with the primer pairs. The positive and negative controls used for reference purposes were whole chromosome genomes (Table 3).

* Primerek: A. Fpfs1 és Rpfs2; B. FluxS1 és RluxS2; C. Serratia2-for és Serratia2-rev.

Jelmagyarázat:

3.2. In vitro experimental studies

To confirm the results of the in silico studies, in vitro were performed in which the selected primer pairs were tested in laboratory PCR analyses on genomic DNA samples of selected strains of bacteria (several S. marcescens strains were used as positive control and Lactobacillus delbrueckii subsp. delbrueckii, Streptococcus thermophilus, Enterococcus faecalis and Micrococcus luteus were used as negative controls). The microorganisms were bacterial strains belonging to the collection of MTKI Kft. and coming from factory environment, determined by genetic identification.

When putting together the components required for the PCR reaction, 5.2 µL of PCR grade sterile water, 10 µL of DreamTaq Green 2× PCR Master Mix (Thermo Fisher Scientific, Waltham, Massachusetts, USA), 0.4 µL (10 pmol/µl) primer and 4 µL of isolated bacterial genomic DNA were used for each reaction. The negative control of the reactions was PCR grade sterile water. The program parameters of the PCR instrument (Mastercycler Nexus Gradient; Eppendorf International, Hamburg, Germany) were as follows: 95 °C for 1 minute, then for 40 cycles 95 °C for 15 seconds, 59.5 °C for 15 seconds, 72 °C for 10 seconds and, finally, 72 °C for 7 minutes [52].

For size separation of the DNA segments formed during the PCR reaction, a 10 µL sample was analyzed on a 2% agarose gel [TBE buffer (Tris-borate-EDTA) (10×), Thermo Fisher Scientific; Agarose DNA Pure Grade, VWR International, Debrecen, Hungary; ECO Safe Nucleic Acid Staining Solution 20.000×, Pacific Image Electronics, Torrance, California, USA]. The DNA size marker was the GeneRuler Low Range DNA Ladder (Thermo Fisher Scientific). Gel documentation was performed using the Gel Doc Universal Hood II gel documentation equipment and software (Bio-Rad, Hercules, California, USA).

3.3. analysis of raw and pasteurized milk samples

On the one hand, we used in our study factory raw and pasteurized milk samples in the case of which S. marcescens contamination was suspected due to their pink discoloration. On the other hand, factory raw and pasteurized milk samples that arrived at the laboratory together with the above samples but not exhibiting discoloration were also tested.

For the DNA digestion and purification process, the NucleoSpin Microbial DNA kit (Macherey-Nagel, Düren, Germany) was used according to the manufacturer’s instructions. The reaction tubes containing the eluted DNA were stored in a freezer at -20 °C.

Next, the suitability of DNA isolation and the amplifiability of the samples were checked by 16S rDNS polymerase chain reaction, using primers 27f (5’-AGAGTTGATCMTGGCTCAG-3’) and 1492r (5’-TACGGYTACCTTGTTACGACTT-3’). The total volume of the PCR reaction for 1 sample was 5.6 µL of PCR grade sterile water, 10 µL DreamTaq Green 2× PCR Master Mix, 0.2 µL (10 pmol/µl) of the primers and 4 µL of isolated bacterial genomic DNA. The negative control of the reactions was PCR grade sterile water. The program parameters of the PCR instrument were as follows: 95 °C for 4 minutes, then for 40 cycles 95 °C for 20 seconds, 54 °C for 30 seconds, 72 °C for 1 minute and, finally, 72 °C for 5 minutes.

For the separation of the DNA segments formed during the PCR reaction, a 5 µL sample was analyzed on a 1% agarose gel. The DNA size marker was the GeneRuler 1 kb Plus DNA Ladder (Thermo Fisher Scientific). The DNA sample tested was judged to be suitable for further PCR analysis if the length of the copies of the amplified DNA fragment was as expected (~1500 bp).

In the next step, samples were subjected to S. marcescens-specific PCR analysis and gel electrophoresis as described in subsection IN VITRO EXPERIMENTAL STUDIES. The results were evaluated on the basis of the presence/absence principle.

In order to check the suitability of the method, PCR results of the milk samples were compared with the few available API (bioMérieux, Budapest, Hungary) test results in a control test. The method was then used to detect the presence of S. marcescens in raw and pasteurized milks.

4. Results

In our in silico studies, when examining the homology of the primers, they showed similarity primarily to S. marcescens chromosome genomes. However, matches were also found in the case of S. rubidaea and S. nematodiphila strains and some non-Serratia species. These results were taken into account during the selection of reference genomes designed for our SnapGene software studies. The need for further investigation was justified by the fact that appropriate homology or the matching of the basis do not automatically mean that the PCR reaction will take place, because the direction of the primers, their melting temperature and the size of the PCR product formed are also critical, among other things.

In the SnapGene test, PCR reactions were predicted with the following parameters: our analyses were performed with at least 15 bases matching and the exclusion of single isolated mismatches. The minimum melting temperature was 50 °C and the maximum length of the fragment obtained as the result of the amplification was 3 kbp.

As shown in Table 3, when matched with the S. marcescens genomes, the primer pair Serratia2-for and Serratia2-rev showed amplification in all cases. The PCR reaction generally resulted in six or seven amplicons on the 16S rDNA sections. The adhesion site of the Fpfs1–Rpfs2 and FluxS1–RluxS2 primer pairs is located outside the 16S rDNA in most S. marcescens strains, but in some cases they did not show in silico amplification, so their sensitivity did not prove to be adequate. In the negative control genomes, the completion of a PCR reaction was predicted by the primer pair Serratia2-for and Serratia2-rev in some cases for certain S. rubidaea and S. nematodiphila strains. Using primers Fpfs1–Rpfs2, the PCR reaction would take place in the case of a S. nematodiphila strain. Primers FluxS1–RluxS2 did not predict the occurrence of a reaction on any of the selected negative control genomes (Table 3).

In S. marcescens genomes selected as positive controls in in vitro experiments, all three primer pairs gave signals according to the expected fragment size, and none gave a signal on the negative controls. The analysis carried out with the primer pair Serratia2-for and Serratia2-rev is shown in Figure 2. In the case of negative samples, the weak signals at around 50 bp are caused by the accumulation of the byproduct aspecific DNA fragments, primer dimers.

Based on the results of in silico analyses and in vitro studies, primers Serratia2-for and Serratia2-rev were considered to be suitable for further work, despite the fact that their specificity was not perfect. The decision was based on the probable frequency of occurrence of S. marcescens on the one hand and the importance of avoiding samples with false negative results on the other.

In order to check the suitability of the method that had been set up, factory milk samples were tested in a control study. Some of the milk samples (n=10) exhibited pink discoloration. Using our test method, nine samples were found to be positive for the microbe sought. We also had API test results for four of the samples. The four API-positive samples were also found to be positive in the PCR assay. The method was then used to detect S. marcescens in raw and pasteurized milks.

Some of the milk samples showed peach-pink discoloration (Figure 3), but it was not clear in many cases due to the pale or yellowish tint. A total of 60 samples were analyzed. Of these, 32 (53.3%) gave positive results and 28 (46.7%) gave negative results for the presence of S. marcescens.

Figure 4 shows the result of one of our assays, the separation by gel electrophoresis. It can be clearly seen that the positive control strain gave a positive signal, while the negative control sample gave a negative signal, and positive signals were obtained for three test samples. The weak signals appearing in the case of negative samples are again caused by the accumulation of primer dimers.

5. Discussion

When evaluating our results, it is important to take into account that the PCR analysis is a method suitable for the amplification and detection of the target DNA in the sample, based on which it is not possible to determine whether the amplified S. marcescens-specific DNA comes from viable, dead or so-called VBNC cells. In the VBNC (“viable but not culturable”) state, the cells are viable, metabolically active, but cannot be propagated by classical culture methods. This condition is reversible.

The objective of our work was to establish a classical PCR method for the detection of S. marcescens. Using the test procedure applied, qualitative determination of the S. marcescens contamination responsible for the discoloration of milk samples can be carried out.

Although the experiments presented here focused on the detection of pigment-producing S. marcescens, a future genus-level study could identify all 20 Serratia species (Table 1). The significance of the detection of other Serratia species is evidenced by the fact that, although the genus Pseudomonas is the main cause of the spoilage of chilled raw milk, the dangers of Serratia species in this respect are also known [56]. In addition to Pseudomonas strains, Serratia strains have also been identified in many cases as causes of milk spoilage. Members of the genus Serratia have been detected in dairy plants [3, 12], in raw milk samples stored at 4° C [56, 57, 58] and in milk containers [59]. It was noted by Grimont and Grimont [9] already a decade and a half ago that raw milk lots can occasionally be contaminated with Serratia species, and the species most often occurring in diary products are S. liquefaciens and S. grimesii.

The presence of psychotrophic Serratia species (e.g., S. liquefaciens) in raw milk can cause spoilage even after heat treatment. Baglinière et al. found that the thermally stable Ser2 protease produced by S. liquefaciens may be a significant factor in the destabilization of UHT milk [11, 60].

In conclusion, it can be stated that a genus-level study would be an interesting research project that would fill a gap, and which would allow the monitoring of raw milk in this respect, the wide detection of Serratia species. Presumably, the results would provide useful information not only to the stakeholders of the dairy economy and the dairy industry, but could also have an impact on Hungarian regulatory and monitoring practice.

6. References

[1] Friman, M.J., Eklund, M.H., Pitkälä, A.H., Rajala-Schultz, P.J., Rantala, M.H.J. (2019): Description of two Serratia marcescens associated mastitis outbreaks in Finnish dairy farms and a review of literature. Acta Veterinaria Scandinavica. 61, pp. 54. https://doi.org/10.1186/s13028-019-0488-7

[2] Joyner, J., Wanless, D., Sinigalliano, C.D., Lipp, E.K. (2014): Use of quantitative real-time PCR for direct detection of Serratia marcescens in marine and other aquatic environments. Applied and Environmental Microbiology. 80, pp. 1679-1683. https://doi.org/10.1128/AEM.02755-13

[3] Cleto, S., Matos, S., Kluskens, L., Vieira, M.J. (2012): Characterization of contaminants from a sanitized milk processing plant. PLoS ONE. 7(6), e40189. https://doi.org/10.1371/journal.pone.0040189

[4] Langsrud, S., Møretrø, T., Sundheim, G. (2003): Characterization of Serratia marcescens surviving in disinfecting footbaths. Journal of Applied Microbiology. 95, pp. 186-195. https://doi.org/10.1046/j.1365-2672.2003.01968.x

[5] Møretrø, T., Langsrud, S. (2017): Residential bacteria on surfaces in the food industry and their implications for food safety and quality. Comprehensive Reviews in Food Science and Food Safety. 16, pp. 1022-1041. https://doi.org/10.1111/1541-4337.12283

[6] Iwaya, A., Nakagawa, S., Iwakura, N., Taneike, I., Kurihara, M., Kuwano, T., Gondaira, F., Endo, M., Hatakeyama, K., Yamamoto, T. (2005): Rapid and quantitative detection of blood Serratia marcescens by a real-time PCR assay: Its clinical application and evaluation in a mouse infection model. FEMS Microbiology Letters. 248, pp. 163-170. https://doi.org/10.1016/j.femsle.2005.05.041

[7] Bayramoglu, G., Buruk, K., Dinc, U., Mutlu, M., Yilmaz, G., Aslan, Y. (2011): Investigation of an outbreak of Serratia marcescens in a neonatal intensive care unit. Journal of Microbiology, Immunology and Infection. 44, pp. 111-115. https://doi.org/10.1016/j.jmii.2010.02.002

[8] Moradigaravand, D., Boinett, C.J., Martin, V., Peacock, S.J., Parkhill, J. (2016): Recent independent emergence of multiple multidrug-resistant Serratia marcescens clones within the United Kingdom and Ireland. Genome Research. 26, pp. 1101-1109. https://doi.org/10.1101/gr.205245.116

[9] Grimont, F., Grimont, P.A.D. (2006): The genus Serratia. Prokaryotes. 6, pp. 219-244. https://doi.org/10.1007/0-387-30746-X_11

[10] Sandner-Miranda, L., Vinuesa, P., Cravioto, A., Morales-Espinosa, R. (2018): The genomic basis of intrinsic and acquired antibiotic resistance in the genus Serratia. Frontiers in Microbiology. 9, pp. 828. https://doi.org/10.3389/fmicb.2018.00828

[11] Baglinière, F., Tanguy, G., Salgado, R.L., Jardin, J., Rousseau, F., Robert, B., Harel-Oger, M., Dantas Vanetti, M.C., Gaucheron, F. (2017): Ser2 from Serratia liquefaciens L53: A new heat stable protease able to destabilize UHT milk during its storage. Food Chemistry. 229, pp. 104-110. https://doi.org/10.1016/j.foodchem.2017.02.054

[12] Salgado, C.A., Baglinière, F., Vanetti, M.C.D. (2020): Spoilage potential of a heat-stable lipase produced by Serratia liquefaciens isolated from cold raw milk. LWT - Food Science and Technology. 126, 109289. https://doi.org/10.1016/j.lwt.2020.109289

[13] Barnum, D.A., Thackeray, E.L., Fish, N.A. (1958): An outbreak of mastitis caused by Serratia marcescens. Canadian Journal of Comparative and Medical Veterinary Science. 22, pp. 392-395.

[14] Malik, K., Tokkas, J., Goyal, S. (2012): Microbial pigments: A review. International Journal of Microbial Resource Technology. 1 (4), pp. 361-365.

[15] Petersen, L.M., Tisa, L.S. (2013): Friend or foe? A review of the mechanisms that drive Serratia towards diverse lifestyles. Canadian Journal of Microbiology. 59, pp. 627-640. https://doi.org/10.1139/cjm-2013-0343

[16] Darshan, N., Manonmani, H.K. (2015): Prodigiosin and its potential applications. Journal of Food Science and Technology. 52, pp. 5393-5407. https://doi.org/10.1007/s13197-015-1740-4

[17] Srimathi, R., Priya, R., Nirmala, M., Malarvizhi, A. (2017): Isolation, identification, optimization of prodigiosin pigment produced by Serratia marcescens and its applications. International Journal of Latest Engineering and Management Research. 2 (9), pp. 11-21.

[18] Giri, A.V., Anandkumar, N., Muthukumaran, G., Pennathur, G. (2004): A novel medium for the enhanced cell growth and production of prodigiosin from Serratia marcescens isolated from soil. BMC Microbiology. 4, pp. 11. https://doi.org/10.1186/1471-2180-4-11

[19] Mahlen, S.D. (2011): Serratia infections: From military experiments to current practice. Clinical Microbiology Reviews. 24, pp. 755-791. https://doi.org/10.1128/CMR.00017-11

[20] Analyzer of Bio-resource Citations (2020): Microorganism Search for Paper, Patent, Genome and Nucleotic. http://abc.wfcc.info/index.jsp. Hozzáférés 2020.04.21.

[21] Birla Institute of Scientific Research, Bioinformatics Centre (2015): Database of Biochemical Tests of Pathogenic Enterobacteriaceae Family. https://bioinfo.bisr.res.in/cgi-bin/project/docter/serratia.cgi. Hozzáférés 2020.04.21.

[22] LPSN (2020): List of Prokaryotic Names with Standing in Nomenclature. https://lpsn.dsmz.de/genus/serratia. Hozzáférés 2020.04.21.

[23] Kämpfer, P., Glaeser, S.P. (2016): Serratia aquatilis sp. nov., isolated from drinking water systems. International Journal of Systematic and Evolutionary Microbiology. 66, pp. 407-413. https://doi.org/10.1099/ijsem.0.000731

[24] Grimont, P.A.D., Jackson, T.A., Ageron, E., Noonan, M.J. (1988): Serratia entomophila sp. nov. associated with amber disease in the New Zealand grass grub Costelytra zealandica. International Journal of Systematic Bacteriology. 38, pp. 1-6. https://doi.org/10.1099/00207713-38-1-1

[25] Grimont, P.A.D., Grimont, F., Starr, M.P. (1979): Serratia ficaria sp. nov., a bacterial species associated with Smyrna figs and the fig wasp Blastophaga psenes. Current Microbiology. 2, pp. 277-282. https://doi.org/10.1007/BF02602859

[26] Anahory, T., Darbas, H., Ongaro, O., Jean-Pierre, H., Mion, P. (1998): Serratia ficaria: A misidentified or unidentified rare cause of human infections in fig tree culture zones. Journal of Clinical Microbiology. 36, pp. 3266-3272. https://doi.org/10.1128/JCM.36.11.3266-3272.1998

[27] Gavini, F., Ferragut, C., Izard, D., Trinel, P.A., Leclerc, H., Lefebvre, B., Mossel, D.A.A. (1979): Serratia fonticola, a new species from water. International Journal of Systematic Bacteriology. 29, pp. 92-101. https://doi.org/10.1099/00207713-29-2-92

[28] Grimont, P.A.D., Grimont, F., Irino, K. (1982): Biochemical characterization of Serratia liquefaciens sensu stricto, Serratia proteamaculans, and Serratia grimesii sp. nov.. Current Microbiology. 7, pp. 69-74. https://doi.org/10.1007/BF01568416

[29] Hennessy, R.C., Dichmann, S.I., Martens, H.J., Zervas, A., Stougaard, P. (2020): Serratia inhibens sp. nov., a new antifungal species isolated from potato (Solanum tuberosum). International Journal of Systematic and Evolutionary Microbiology. 70, pp. 4204-4211. https://doi.org/10.1099/ijsem.0.004270

[30] Bizio, B. (1823): Lettera di Bartolomeo Bizio al chiarissimo canonico Angelo Bellani sopra il fenomeno della polenta porporina. Biblioteca Italiana, o sia Giornale di Letteratura, Scienze, e Arti (Anno VIII). 30, pp. 275-295.

[31] Williams, R.P., Gott, C.L., Qadri, S.M.H., Scott, R.H. (1971): Influence of temperature of incubation and type of growth medium on pigmentation in Serratia marcescens. Journal of Bacteriology. 106, pp. 438-443. https://doi.org/10.1128/JB.106.2.438-443.1971

[32] Wang, J., Zheng, M.L., Jiao, J.Y., Wang, W.J., Li, S., Xiao, M., Chen, C., Qu, P.H., Li, W.J. (2019): Serratia microhaemolytica sp. nov., isolated from an artificial lake in Southern China. Antonie Van Leeuwenhoek. 112, pp. 1447-1456. https://doi.org/10.1007/s10482-019-01273-9

[33] García-Fraile, P., Chudíčková, M., Benada, O., Pikula, J., Kolařík, M. (2015): Serratia myotis sp. nov. and Serratia vespertilionis sp. nov., isolated from bats hibernating in caves. International Journal of Systematic and Evolutionary Microbiology. 65, pp. 90-94. https://doi.org/10.1099/ijs.0.066407-0

[34] Zhang, C.X., Yang, S.Y., Xu, M.X., Sun, J., Liu, H., Liu, J.R., Liu, H., Kan, F., Sun, J., Lai, R., Zhang, K.Y. (2009): Serratia nematodiphila sp. nov., associated symbiotically with the entomopathogenic nematode Heterorhabditidoides chongmingensis (Rhabditida: Rhabditidae). International Journal of Systematic and Evolutionary Microbiology. 59, pp. 1603-1608. https://doi.org/10.1099/ijs.0.003871-0

[35] Grimont, P.A.D., Grimont, F., Richard, C., Davis, B.R., Steigerwalt, A.G., Brenner, D.J. (1978): Deoxyribonucleic acid relatedness between Serratia plymuthica and other Serratia species, with a description of Serratia odorifera sp. nov. (type strain: ICPB 3995). International Journal of Systematic Bacteriology. 28, pp. 453-463. https://doi.org/10.1099/00207713-28-4-453

[36] Van Houdt, R., Moons, P., Jansen, A., Vanoirbeek, K., Michiels, C.W. (2005): Genotypic and phenotypic characterization of a biofilm-forming Serratia plymuthica isolate from a raw vegetable processing line. FEMS Microbiology Letters. 246, pp. 265-272. https://doi.org/10.1016/j.femsle.2005.04.016

[37] Zhang, C.W., Zhang, J., Zhao, J.J., Zhao, X., Zhao, D.F., Yin, H.Q., Zhang, X.X. (2017): Serratia oryzae sp. nov., isolated from rice stems. International Journal of Systematic and Evolutionary Microbiology. 67, pp. 2928-2933. https://doi.org/10.1099/ijsem.0.002049

[38] Lehman, K.B., Neumann, R. (1896): Atlas und Grundriss der Bakeriologie und Lehrbuch der Speziellen Bakteriologischen Diagnostik, Volume 11st Ed. J.F. Lehmann, München.

[39] Breed, R.S., Murray, E.G.D., Hitchens, A.P. (Eds.) (1948): Bergey’s Manual of Determinative Bacteriology, 6th ed. Williams and Wilkins Co., Baltimore, MD, USA. pp. 1-1529.

[40] Paine, S.G., Stansfield, H. (1919): Studies in bacteriosis. III. A bacterial leaf spot disease of Peotea cynaroides, exhibiting a host reaction of possibly bacteriolytic nature. Annals of Applied Biology. 6, pp. 27-39. https://doi.org/10.1111/j.1744-7348.1919.tb05299.x

[41] Grimont, P.A.D., Grimont, F., Starr, M.P. (1978): Serratia proteamaculans (Paine and Stansfield) comb. nov., a senior subjective synonym of Serratia liquefaciens (Grimes and Hennerty) Bascomb et al. International Journal of Systematic Bacteriology. 28, pp. 503-510. https://doi.org/10.1099/00207713-28-4-503

[42] Ashelford, K.E., Fry, J.C., Bailey, M.J., Day, M.J. (2002): Characterization of Serratia isolates from soil, ecological implications and transfer of Serratia proteamaculans subsp. quinovora Grimont et al. 1982 to Serratia quinovorans corrig., sp. nov.. International Journal of Systematic and Evolutionary Microbiology. 52, pp. 2281-2289. https://doi.org/10.1099/00207713-52-6-2281

[43] Stapp, C. (1940): Bacterium rubidaeum nov. spec. Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene, Abt. II. 102, pp. 252-260.

[44] Ewing, W.H., Davis, B.R., Fife, M.A., Lessel, E.F. (1973): Biochemical characterization of Serratia liquefaciens (Grimes and Hennerty) Bascomb et al. (formerly Enterobacter liquefaciens) and Serratia rubidaea (Stapp) comb. nov. and designation of type and neotype strains. International Journal of Systematic Bacteriology. 23, pp. 217-225. https://doi.org/10.1099/00207713-23-3-217

[45] Sabri, A., Leroy, P., Haubruge, E., Hance, T., Frère, I., Destain, J., Thonart, P. (2011): Isolation, pure culture and characterization of Serratia symbiotica sp. nov., the R-type of secondary endosymbiont of the black bean aphid Aphis fabae. International Journal of Systematic and Evolutionary Microbiology. 61, pp. 2081-2088. https://doi.org/10.1099/ijs.0.024133-0

[46] Bhadra, B., Roy, P., Chakraborty, R. (2005): Serratia ureilytica sp. nov., a novel urea-utilizing species. International Journal of Systematic and Evolutionary Microbiology. 55, pp. 2155-2158. https://doi.org/10.1099/ijs.0.63674-0

[47] Starr, M.P., Grimont, P.A.D., Grimont, F., Starr, P.B. (1976): Caprylate-thallous agar medium for selectively isolating Serratia and its utility in the clinical laboratory. Journal of Clinical Microbiology. 4, pp. 270-276.

[48] BioMérieux (2015): API & ID 32 Identification Databases. https://www.biomerieux-diagnostics.com/sites/clinic/files/9308960-002-gb-b-apiweb-booklet.pdf. Hozzáférés 2020.04.02.

[49] Primerdesign (2019): Serratia marcescens Genesig kit. https://www.genesig.com/products/9405-serratia-marcescens. Hozzáférés 2020.04.02.

[50] Hejazi, A., Keane, C.T., Falkiner, F.R. (1997): The use of RAPD-PCR as a typing method for Serratia marcescens. Journal of Medical Microbiology. 46, pp. 913-919. https://doi.org/10.1099/00222615-46-11-913

[51] Zhu, H., Zhou, W.Y., Xu, M., Shen, Y.L., Wei, D.Z. (2007): Molecular characterization of Serratia marcescens strains by RFLP and sequencing of PCR-amplified 16S rDNA and 16S-23S rDNA intergenic spacer. Letters in Applied Microbiology. 45. pp. 174-178. https://doi.org/10.1111/j.1472-765X.2007.02166.x

[52] Bussalleu, E., Althouse, G.C. (2018): A PCR detection method for discerning Serratia marcescens in extended boar semen. Journal of Microbiological Methods. 151, pp. 106-110. https://doi.org/10.1016/j.mimet.2018.06.012

[53] Zhu, H., Sun, S.J., Dang, H.Y. (2008): PCR detection of Serratia spp. using primers targeting pfs and luxS genes involved in AI-2-dependent quorum sensing. Current Microbiology. 57, pp. 326-330. https://doi.org/10.1007/s00284-008-9197-6

[54] National Center for Biotechnology Information (2020): Search database. https://www.ncbi.nlm.nih.gov/. Hozzáférés 2020.03.20.

[55] Insightful Science (2020): SnapGene Software. https://www.snapgene.com/. Hozzáférés 2020.03.20.

[56] Machado, S.G., Baglinière, F., Marchand, S., Van Coillie, E., Vanetti, M.C.D., De Block, J., Heyndrick, M. (2017): The biodiversity of the microbiota producing heat-resistant enzymes responsible for spoilage in processed bovine milk and dairy products. Frontiers in Microbiology. 8, p. 302. https://doi.org/10.3389/fmicb.2017.00302

[57] Lafarge, V., Ogier, J.C., Girard, V., Maladen, V., Leveau, J.Y., Gruss, A., Delacroix-Buchet, A. (2004): Raw cow milk bacterial population shifts attributable to refrigeration. Applied and Environmental Microbiology. 70, pp. 5644-5650. https://doi.org/10.1128/AEM.70.9.5644-5650.2004

[58] Ribeiro Jr., J.C., de Oliveira, A.M., de G. Silva, F., Tamanini, R., de Oliveira, A.L.M., Beloti, V. (2018): The main spoilage-related psychrotrophic bacteria in refrigerated raw milk. Journal of Dairy Science. 101, pp. 75-83. https://doi.org/10.3168/jds.2017-13069

[59] Decimo, M., Morandi, S., Silvetti, T., Brasca, M. (2014): Characterization of gram-negative psychrotrophic bacteria isolated from Italian bulk tank milk. Journal of Food Science. 79, pp. 81-90. https://doi.org/10.1111/1750-3841.12645

[60] Baglinière, F., Salgado, R.L., Salgado, C.A., Dantas Vanetti, M.C. (2017): Biochemical characterization of an extracellular heat-stable protease from Serratia liquefaciens isolated from raw milk. Journal of Food Science. 82, pp. 952-959. https://doi.org/10.1111/1750-3841.13660