Aflatoxin M1 contamination in raw milk and its association with herd types in the ten provinces of Southern China

ABSTRACT:

The incidence of aflatoxin M1 (AFM1) contamination in raw milk and its association with herd types was investigated in the ten provinces of Southern China in 2018. A total of 797 raw milk samples were collected from Zhejiang, Jiangxi, Jiangsu, Hunan, Hubei, Hainan, Guangxi, Guangdong, Fujian, and Anhui provinces. The herd types were classified according to their characteristics. Among the 797 samples, 401 were owned by large-scale farms (herd size >500 cows, for dairy cow farming), 303 were owned by milk processing plants (herd size >500 cows, for dairy cow farming + milk processing), and 93 were owned by small farm cooperatives (herd size ≤500 cows, for dairy cow farming). The results revealed the presence of AFM1 in 94 of the 797 samples (11.80%). None of them was above 500 ng/L (Chinese and US limit standard), whereas 0.88% were above 50 ng/L (EU limit standard). The AFM1 concentrations in samples non-compliant with the EU standard were 52, 61, 88, 89.5, 164, 240, and 486 ng/L, respectively. It is important to note that samples exceeding the EU limit were mainly found in the area of Jiangsu and Guangxi provinces, and the herd type of milk processing plants. Hence, it is necessary to monitor and enforce the regulatory compliance of AFM1 levels in the raw milk from these specific areas and herd type.

Highlights

• A total of 797 raw milk samples from three herd types, were collected from ten provinces of Southern China in 2018.

• Ninety-four of the 797 raw milk samples (11.80%) were positive for AFM1, and none of them was above the Chinese and US legal limit, only 0.88% of samples were above the EU limit.

• All of the exceed samples defined by EU limit were found in the area of Jiangsu and Guangxi provinces and belonged to milk processing plants.

Keywords:

• Aflatoxin M1
• raw milk safety
• herd types
• Southern China

Introduction

Aflatoxins are poisonous carcinogenic and mutagenic compounds produced by certain fungi, which can contaminate agricultural products and thus represent a major food safety problem and a permanent risk for the food industry (Min et al. 2020). Aflatoxin M1 (AFM1) was classified by the International Agency for Research on Cancer (IARC) to its Group 1 toxin category in its 2012 monograph (IARC 2012) implying that it is carcinogenic to humans. AFM1 is found in the milk of nursing mothers and in lactating mammals that have ingested aflatoxin B1 (AFB1) contaminated food (Kos et al. 2014). The presence of AFM1 in milk poses a significant human health hazard and directly negatively impacts consumers due to its effects on immunosuppression, carcinogenicity, and teratogenesis (Bilandžić et al. 2015). It is worth noting that even consumption of milk contaminated with a low dose of AFM1 will impair the development of immune competence in infancy (Giovati et al. 2015). Currently, the detection of AFM1 in milk is usually performed by high-performance liquid chromatography (HPLC) or enzyme-linked immunosorbent assay (ELISA) methods. Due to the advantages of flexibility, simplicity, and high efficiency, the surveillance, and control of AFM1 in milk is frequently performed by ELISA test worldwide (Ketney et al. 2017). However, a high incidence of AFM1 in milk continues to be reported in many countries (Iqbal et al. 2015). Therefore, to protect consumers, most countries have set up maximum residue limits of AFM1 in milk and dairy products (Tomašević et al. 2015). In accordance with the European Union (EU) standards, the levels of AFM1 in milk should be less than 50 ng/L, whereas the United States (US) and China stipulate that the levels of AFM1 should not exceed 500 ng/L (Zheng et al. 2013a).

Generally, a hot and humid environment, insect infestation, and other factors will predispose crops to AFB1 contamination in the field before harvest (Jiang et al. 2021). Subsequently, unsuitable postharvest drying, storage, processing, and a lifetime of crops will induce AFB1 contamination (Min et al. 2021), leading to the presence of AFM1 in milk. Furthermore, the geographical factor and herd types might also influence the occurrence of AFM1 in milk, which will be further investigated in the present study.

Since the studies conducted to date have revealed that it is common to find milk samples that test positive for AFM1, further survey studies aiming at monitoring AFM1 are essential to ensure milk safety. Considering this issue, 797 raw milk samples were collected in 2018 from ten provinces in Southern China, including Zhejiang, Jiangxi, Jiangsu, Hunan, Hubei, Hainan, Guangxi, Guangdong, Fujian, and Anhui provinces, to determine the AFM1 levels in raw milk located in different regions, and to further verify the relationship between AFM1 levels and herd types (milk processing plants, large-scale farms, and small farm cooperatives).

Materials and methods

Ethical statement

The experiment was conducted ethically and responsibly and is in full compliance with all relevant codes of experimentation and legislation. It was approved and supervised by the Animal Care and Use Committee of Guangdong Academy of Agricultural Sciences (Guangzhou, China).

Sampling information

The sampling was performed according to the criteria of the convenience sampling method (Felicio et al. 2013), to obtain representative milk samples. The number of samples was determined considering the dairy herd size in various Chinese provinces, as mentioned by Li et al. (2017). Meanwhile, the selection of the herd types from which samples were collected was performed based on the proportion of milk processing plants, large-scale farms, and small farm cooperatives. The classifications of the herd types were based on their ownership (Zheng et al. 2013b). Raw milk from large-scale farms (herd size >500 cows) and small farm cooperatives (herd size ≤500 cows) would eventually transport to milk processing plants. Meanwhile, a portion of the raw milk collected by milk processing plants comes from their owned farms (herd size >500 cows). All the herd types consisted of Holstein cows and were fed the total mixed ration (TMR) diet (including silage maize, alfalfa, oat grass, concentrate, etc.). Overall, the above information of the geographical factor and herd types within the sample might strive to be typically representative of raw milk in the ten provinces of Southern China.

In this survey study, 797 raw milk samples were collected from March to June 2018, from Zhejiang (98), Jiangxi (61), Jiangsu (134), Hunan (29), Hubei (99), Hainan (2), Guangxi (116), Guangdong (137), Fujian (61), and Anhui (60) provinces, all located in Southern China. Of these 797 samples, 303 were owned by milk processing plants, 401 by large-scale farms, and 93 by small farm cooperatives. The raw milk samples were directly collected at milk tanks from milk processing plants, large-scale farms, and small farm cooperatives, respectively. All samples were stored at 4 °C, and detection was performed before the expiration date.

Detection of AFM1 levels

All raw milk samples were tested in duplicate. AFM1 levels were determined by ELISA using a commercial test kit (R-Biopharm AG, Darmstadt, Germany), which is generally used for the quantitative detection of AFM1 in milk and dairy products. The assay was conducted by the instructions provided by the manufacturer of the kit. Samples were thawed and then centrifuged at 3,500 g for 10 min, to remove milk fat. The supernatant was used as a testing sample. A 100 µL of the antibody solution was added to the bottom of each well and incubated. After washing three times, 100 µL of each standard (0, 5, 10, 20, 40, and 80 ng/L) and testing samples were added into individual wells and incubated for 30 min. After washing, 100 µL of the AFM1 enzyme-conjugated was added and incubated for 15 min in the dark. Then, after washing again, a 100 µL of substrate/chromogen solution was added and incubated for 15 min in the dark. Subsequently, a 100 µL of stop solution was added to each well. The value of the absorbance was measured at 450 nm, and the concentrations of AFM1 were calculated by formula. A sample was identified as positive if the concentration of AFM1 was above the minimum detection value of the test kit (5 ng/L). If the concentration of AFM1 was greater than 80 ng/L, this sample would be diluted with the diluent solution in the test kit and analysed again.

The validation parameters were examined and listed as follows: recovery was in the range of 86–112%, the limit of detection (LOD) = 3 ng/L, limit of quantification (LOQ) = 10 ng/L, and relative standard deviation values (RSD%) were less than 8.0%. The detection results of quality assurance and quality control in this study met the requirements of performance criteria and therefore qualified for further analysis (Tadesse et al. 2020).

Statistical analysis

Differences in concentrations of AFM1 in raw milk samples were statistically analysed with the SPSS software version 19.0 (IBM Corporation, Armonk, NY, USA) by nonparametric tests, using the Kruskal-Wallis one-way analysis of variance (ANOVA) test with stepwise step-down post-hoc comparisons. The significant difference was set at P < .05.

Results

AFM1 detection in raw milk in Southern China

The data of AFM1 contamination in 797 raw milk samples from three herd types in the ten provinces of Southern China are listed in Supplementary Table S1. The results of AFM1 contamination in raw milk in the ten provinces of Southern China are summarised in Table 1. Overall, 94 of the 797 samples (11.80%) were found to be AFM1-positive (>5 ng/L). No AFM1-positive samples were found in the provinces of Hunan, Hubei, Hainan, and Guangdong. The samples from the other six provinces (Zhejiang, Jiangxi, Jiangsu, Guangxi, Fujian, and Anhui) showed different levels of AFM1 contamination. In Zhejiang, Jiangxi, Guangxi, and Anhui provinces, eight (8.16%), five (8.20%), five (4.32%), and two (3.33%) samples were AFM1-positive, respectively. A higher incidence of AFM1 contamination was found in samples from Jiangsu (36.60%) and Fujian (41.00%) provinces. The concentrations of AFM1 in positive samples from Jiangxi and Guangxi provinces were significantly higher than those in samples from Jiangsu and Fujian provinces (P < .05), and the relatively lower AFM1 concentrations in positive samples were detected in Zhejiang and Anhui provinces (P < .05).

Table 1. Incidence and distribution information of Aflatoxin M1 (AFM1) in raw milk in the ten provinces of southern China.

Regions	Positive/total samples (%)	Distribution information of AFM1 levels			Max value ng/L	Mean ± SD ng/L
		≤50	51–500	>500
Zhejiang	8/98 (8.16%)	8	0	0	28.3	14.8 ± 9.1^C
Jiangxi	5/61 (8.20%)	5	0	0	47.6	42.3 ± 3.5^A
Jiangsu	49/134 (36.60%)	46	3	0	486	35.3 ± 66.8^B
Hunan	0/29 (0)	0	0	0	<5	–
Hubei	0/99 (0)	0	0	0	<5	–
Hainan	0/2 (0)	0	0	0	<5	–
Guangxi	5/116 (4.32%)	1	4	0	240	122.1 ± 81.4^A
Guangdong	0/137 (0)	0	0	0	<5	–
Fujian	25/61 (41.00%)	25	0	0	50	26.8 ± 9.5^B
Anhui	2/60 (3.33%)	2	0	0	8	7.0 ± 1.4^C
Total	94/797 (11.80%)	87	7	0	486	35.7 ± 55.7

Mean ± SD is calculated from the value of the positive samples. Within columns, the mean ± SD with different superscript letters indicate a significant difference (P < .05).

The maximum value of AFM1 in the 797 raw milk samples was 486 ng/L, which was detected in a sample from Jiangsu province. The mean concentration of AFM1 was 35.7 ± 55.7 ng/L. All of them were far below the standard of Chinese and US. Seven samples were above the EU limit standard of 50 ng/L, which were collected from Jiangsu (three samples) and Guangxi (four samples) provinces. The AFM1 concentrations in samples non-compliant with the EU standard were 52, 61, 88, 89.5, 164, 240, and 486 ng/L, respectively.

AFM1 contamination in raw milk and its association with herd types

The incidence of AFM1 in raw milk samples was 11.20%, 12.20%, and 11.80% in milk processing plants, large-scale farms, and small farm cooperatives, respectively (Table 2). The concentrations of AFM1 in positive samples from milk processing plants were significantly higher than those from large-scale farms (P < .05). The seven samples that exceeded the EU limit for AFM1 were found from milk processing plants.

Table 2. Aflatoxin M1 (AFM1) contamination in raw milk and its association with herd types.

Herd types	Positive/total samples (%)	Distribution information of AFM1 levels			Max value ng/L	Mean ± SD ng/L
		≤50	51–500	>500
Milk processing plants	34/303 (11.20%)	27	7	0	486	56.3 ± 88.5^A
Large-scale farms	49/401 (12.20%)	49	0	0	50	24.1 ± 11.6^B
Small farm cooperatives	11/93 (11.80%)	11	0	0	40	23.4 ± 9.8^AB

Mean ± SD is calculated from the value of the positive samples. Within columns, the mean ± SD with different superscript letters indicate a significant difference (P < .05).

The distribution information about AFM1 contamination in milk is presented by provinces and herd types in Figure 1. Eight samples collected from Zhejiang province and owned by large-scale farms were positive for AFM1. In Jiangxi province, four samples collected from milk processing plants and one sample collected from large-scale farms tested positive for AFM1. Twenty-five samples collected from milk processing plants, thirteen samples collected from large-scale farms, and eleven samples collected from small farm cooperatives were positive for AFM1 in Jiangsu province. Five samples from Guangxi province and owned by milk processing plants were positive for AFM1. Twenty-five samples from Fujian province and owned by large-scale farms were positive for AFM1. Two samples from Anhui province and owned by large-scale farms were positive for AFM1.

Figure 1. Distribution information of Aflatoxin M1 (AFM1)-positive samples by provinces and herd types. Seven samples from milk processing plants (three from Jiangsu and four from Guangxi), exceeded the EU regulatory limit of 50 ng/L.

Discussion

In recent years, several survey studies focussing on AFM1 contamination levels have been conducted in China to evaluate the safety of milk. Twelve raw milk samples gathered in Northeast China were tested for AFM1 contamination. Although all samples complied with Chinese regulations, the range of AFM1 concentrations was 160–500 ng/L (Pei et al. 2009). Moreover, authors reported a high risk of AFB1 contamination in corn, peanut, etc. from Zhejiang, Jiangsu, Hubei, Guangxi, Guangdong, and Fujian provinces (Southern China), so it is necessary to further estimate the status of milk safety in these areas, as indicated by the AFM1 concentrations in milk. In September 2010, 360 raw milk samples were collected and analysed from three herd types (milk processing plants, large-scale farms, and small farm cooperatives) based on their ownership, from Northern and Southern China (Zheng et al. 2013b). The results revealed that 78.1% of the samples were positive for AFM1, with levels ranging from 5 to 123 ng/L, which were far below the Chinese regulatory limit. However, 10% of them were above the EU limit standard. Furthermore, the occurrence and concentrations of AFM1 in raw milk samples from Southern China were generally higher than those from Northern China, although there was no significant difference among milk processing plants, large-scale farms, and small farm cooperatives. Subsequently, 1,550 raw milk samples from 2013 to 2015 were analysed for AFM1 levels in Western, Northeast, Northern, and Southern China (Zheng et al. 2017). Similar to a previous study (Zheng et al. 2013b), the occurrence and concentrations of AFM1 levels in Southern China were generally higher than those in the other three regions of China. The number of positive samples in Southern China were 53 (44.2%), 62 (25.8%), and 42 (21.0%) from 2013 to 2015, respectively, and the samples exceeding the EU AFM1 standard limit in the same time period were 39 (32.5%), 26 (10.8%), and 8 (4.0%), respectively. The conclusions pointed out that regulatory enforcement and prevention of AFM1 contamination in raw milk need to be implemented, especially in Southern China.

In this survey study, 11.80% of the raw milk samples collected from ten provinces of Southern China were positive for AFM1. All positive samples were below the Chinese and US limit standard. However, seven of the 797 raw milk samples (0.88%) exceeded the EU regulatory limit for AFM1. In recent years, regulations on AFM1 levels in milk have been improved in China. In 2010, 32.5% of 200 raw milk samples from ten main milk-producing areas in China tested positive for AFM1, and only 1.5% of them were non-compliant with the EU limit standard (Han et al. 2013). From November 2011 to September 2012, 59.7% of 72 raw milk samples from Shanghai city, Zhejiang, and Jiangsu provinces in Southern China tested positive for AFM1, and 23.6% were non-compliant with the EU limit standard, with a maximum concentration of 420 ng/L (Xiong et al. 2013). Guo et al. (2016) analysed 530 raw milk samples collected from Tangshan city in Northern China from 2012 to 2014 and found that 52.8% were positive for AFM1, with levels ranging from 10 to 200 ng/L. In this area, a significant reduction in the occurrence of AFM1 levels occurred from 2012 to 2014. A total of 1,550 raw milk samples were gathered in Western, Northeast, Northern, and Southern China from 2013 to 2015, and tested for AFM1 (Zheng et al. 2017). In 2013, 21.0% of 366 samples were contaminated with AFM1, and 11.7% of them were non-compliant with the EU limit standard. In 2014, 28.5% of 624 samples were found to be positive for AFM1, and 7.7% of them were non-compliant with the EU limit standard. In 2015, 14.1% of 560 samples were contaminated with AFM1, and only 1.8% of them were non-compliant with the EU limit standard. A large number of raw milk samples were gathered in Northeast and Northern China in 2016 (Li et al. 2017). Thereinto, 4.7% of 5,650 samples were positive for AFM1, and 1.1% of them exceeded the EU limit standard. In a subsequent study, Li et al. (2018) indicated that in 2016, 4.64% of 1,207 raw milk samples were positive for AFM1 in eleven provinces and one municipality located in Northeast, Northwest, Northern, and Central China, only 0.83% of samples exceeded the EU limit standard. Altogether, these studies showed that the implementation of regulatory measures to reduce the levels of AFM1 contamination in raw milk in China have been successful. In the present study, we have supplemented and improved the survey study data regarding AFM1 contamination in raw milk in Southern China. Of these, it is necessary to improve the regulatory enforcement for AFM1 in raw milk in the areas of Jiangsu and Guangxi provinces, as the maximum values of AFM1 were 486 and 240 ng/L in these two provinces, respectively.

The samples with the highest level of AFM1 concentration and all the samples exceeding the EU legal limit dated from milk processing plants. Hashemi (2016) indicated that the concentrations of AFM1 in raw milk samples from bulk tanks of farms (including large-scale farms and small farm cooperatives) were considerably lower than that in samples from milk collection centres (milk processing plants) in Southern Iran. Furthermore, a risk factor analysis study from Argentina also demonstrated that intensive commercial dairy farming (milk processing plants) is related to the contamination of AFM1 in raw milk (Malissiova et al. 2013). Thus, the prevalence of AFM1 contamination in raw milk samples might be a potential risk in milk processing plants. As a consequence, it is crucial to enforce the control of AFM1 levels in raw milk throughout the entire milk production chain, especially in milk processing plants, to minimise health risks associated with the presence of AFM1 in milk (Škrbić et al. 2014).

Conclusions

This study analysed the AFM1 levels in raw milk samples and its association with herd types in the ten provinces of Southern China. Overall, 94 of the 797 raw milk samples (11.80%) tested positive for AFM1. Nonetheless, none of them exceeded the Chinese and US regulatory limit, and seven samples (0.88%) were found to be above the EU limit standard of 50 ng/L. It is worth noting that all of the exceed samples defined by EU limit dated from Jiangsu and Guangxi provinces, and belonged to milk processing plants. Therefore, it is particularly important to implement continuous surveillance to reduce the levels of AFM1 contamination in raw milk, especially in the potential risk areas and herd type.

Acknowledgements

We sincerely thank Dr. Huaigu Yang (Sericultural & Agri-Food Research Institute, Guangdong Academy of Agricultural Sciences) for his suggestions about the Statistical Analysis.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.

Additional information

Funding

This study was supported by the Natural Science Foundation of Guangdong Province (2018A030313002), Special fund for scientific innovation strategy- construction of high level Academy of Agriculture Science (R2017YJ-YB3006, R2018PY-QF008, R2018QD-072, R2018QD-074), and Guangdong Modern Agro-industry Technology Research System (2020KJ114).