Mycobiota and aflatoxin B1 in poultry feeds

In this study, a total of 30 poultry (chicken and laying hens) feed samples
 collected from different poultry farms in Serbia in 2016 were tested for
 fungal and aflatoxin contamination. Using the plate count and standard
 mycological methods, total fungal counts and potentially toxigenic fungal
 genera were determined. Natural occurrence of aflatoxin B1 (AFB1) was
 detected by ELISA (enzyme-linked immune sorbent assay) method. The total
 fungal count was in the range from 1 x 102 (2 log CFU g-1) to 1.83 x 105 CFU
 g-1 (5.26 log CFU g-1). The majority of the chicken feeds (78.57%) had the
 total fungal count in the ranged from 1 x 102 to 4.8 x 104 CFU g-1, whereas
 in 68.75% of the laying hens feeds it was ranged from 5.3 x 104 to 1.83 x
 105 CFU g-1. In 21.43% of the chicken feeds fungal contamination reached the
 level above the regulation limits. Three potentially toxigenic fungal
 genera, Aspergillus, Fusarium, and Penicillium, have been identified. In the
 tested poultry feed samples, more samples contaminated with Aspergillus were
 determined compared to samples contaminated by Fusarium and Penicillium
 species. The AFB1 was detected in concentrations from 1.34 to 18.29?g kg-1,
 with an average of 4.47 and 4.56?g kg-1 in the chicken and laying hens feed
 samples, respectively. In 14.29% of the chicken feeds, the level of AFB1 was
 above the regulation limits. The obtained results confirmed the importance
 of continuous mycological and mycotoxicological control of poultry feed, as
 well as need to improve risk assessments of such contaminants along the food
 chain.


Introduction
The majority of cereals (maize, wheat, barley, ray and oats) commonly used as poultry feed may be contaminated with toxigenic fungi, mainly from genera Aspergillus, Fusarium and Penicillium which may produce poisonous secondary metabolites called mycotoxins. Thus mycotoxins can easily enter food chain via meat and meat products produced of animals fed with mycotoxin contaminated feed. Primarily the cereals contamination may appear in the field, where fungal spores are spread by the wind, rain, mechanical injuries or insects to the crops (Aliyu et al., 2016). The infection process can be further continued during the grain storage, due to the effect of abiotic and biotic factors (Krnjaja et al., 2015).
Aflatoxins (AFs) and ochratoxins have been the most common contaminants of poultry feed. Cereal kernels are a very suitable substrate for the development of Aspergillus species (Fareed et al., 2014). Aspergillus flavus and A. parasiticus are the main producers of aflatoxins. Among the different types of aflatoxins (B 1 , B 2 , G 1 , G 2 and M 1 ), aflatoxin B 1 (AFB 1 ) has been the most toxic (Babu et al., 2014). Consumption of poultry feed contaminated with AFs causes aflatoxicosis in animals and a severe economic losses in the poultry production. Aflatoxins have negative impact on important poultry production parameters such as feed intake, feed conversion, weight gain, etc. An immune response in poultry can also be reduced, which raises the risk to diseases (Fareed et al., 2014).
In order to avoid harmful effects of AFs on animal health, the European community set maximum permissible levels for AFB 1 to 20 µg kg -1 for complete and complementary poultry feed (except for young animals). The regulation limits for feeds of young animals have been set to 10 µg AFB 1 kg -1 (for complete feed) and 5 µg AFB 1 kg -1 (for complementary feed) (EC, 2003). In Serbia, according to the Regulation on the quality of feedstuffs (Službeni Glasnik RS, 4/2010, 113/2012, 27/2014, 25/2015, 39/2016, 54/2017, the maximum permissible levels in complete and complementary feeding stuffs have been set to 20 µg AFB 1 kg -1 for adult poultry, and 5 µg AFB 1 kg -1 for young poultry.
Since mycotoxins are inevitable contaminants of cereals as a main constituent of poultry feeds, the aim of this study was to determine the fungal contamination and aflatoxin presence in the samples of poultry feed collected from different farms in Serbia. These investigations are important in order to highlight the importance of quality control along the food and feed chains.

Materials and Methods
In this study, the mycological and mycotoxicological evaluation of 30 poultry feed samples (14 of chicken and 16 of laying hens feed) was performed. The group of the tested chicken feed samples was used for the feeding of the broilers and pullets. The samples were complete or complementary feed mixtures, collected from different poultry farms in Serbia in 2016. The samples of about 1 kg were stored for 2-3 days at 4°C, prior to analysis. The moisture content was determined using a laboratory moisture analyzer (OHAUS MB35, Parsippany, NJ, USA). The presence of fungal species was determined using the ISO 21527-2 method (2008).
Fungal species were identified according to fungal morphology and identification key of Watanabe (2002). The isolation frequency of potentially toxigenic fungi from genera Aspergillus, Fusarium, and Penicillium in the tested samples was calculated as the percentage of poultry feed samples contaminated with fungal species in relation to the total number of poultry feed samples.
The presence of aflatoxin B 1 (AFB 1 ) was detected by ELISA (enzymelinked immune sorbent assay) method according to the manufacturer's instructions Celer Tecna® ELISA kits. The absorbance was determined at a wavelength of 450 nm on an ELISA plate reader spectrophotometer (Biotek EL x 800TM, Winooski, VT, USA). The lower and upper detection limits of AFB 1 were 1 μg kg -1 and 40 μg kg -1 , respectively.
The SPSS software (IBM, Statistic 20) was used for data comparison of the tested parameters. The significance levels were determined by t-test and Pearson correlation coefficient.

Results
Total fungal counts in the tested poultry samples are shown in Table 1. In Serbia, according to the Regulation on the quality of feedstuffs (Službeni Glasnik RS, 4/2010, 113/2012, 27/2014, 25/2015, 39/2016, 54/2017, the acceptable limits for fungal contamination in plant origin feed mixtures has been set to 200,000 CFU g -1 (for adult animals) and 50,000 CFU g -1 (for young animals). Following this regulation, in 21.43% (3/14) of the chicken feeds the values were exceeded the permitted limits (Table 1). The mean moisture contents were 10.57% and 10.62% in the chicken and laying hens feed samples, respectively. A significantly higher fungal count was found in the laying hens feeds than in the chiken feeds (Table 2). The occurrence of potentially toxigenic fungal species from the Aspergillus genus was more common in the laying hens feeds (93.75% positive samples) than in the chicken feeds (85.71% positive samples). On average, the most number of Aspergillus spp. contaminated samples (89.73%) were established, followed by Fusarium spp. (79.47%) and Penicillium spp. (34.38%) contaminated samples (Table 3).  Table 4 shows the frequency, ranges and average concentrations of AFB 1 occurence in the tested poultry feed samples. A higher percentage of positive AFB 1 samples was detected in the laying hens feeds (100%) then in the chicken feeds (85.71%). Average concentrations of AFB 1 investigated in the chicken and laying hens feeds were 4.47 and 4.56 g kg -1 , respectively (Table 4). The level of AFB 1 which was above the regulation limit (5g kg -1 ) was recorded in two chicken feed samples (14.29%), whereas in all the laying hens feed samples, the levels of AFB 1 were under the permissible limit (20 g kg -1 ). According to data analyses, there was no significant positive correlations between the total fungal counts and the moisture contents (r = 0.39) and between the total fungal counts and the levels of AFB 1 (r = 0.41), while a statistically significant (P < 0.01) positive correlation was registered between the levels of AFB 1 and the moisture contents (r = 0.76) in the laying hens feeds. Further, there was no significant negative correlations between the total fungal counts and the moisture contents (r = -0.15) and the levels of AFB 1 (r = -0.24), while there was positive but not significant correlation between the levels of AFB 1 and the moisture contents (r = 0.31) in the chicken feeds.

Discussion
Fungi are ubiquitous plant pathogens that are common agents of foods and feedstuffs deterioration. Fungal and mycotoxin contamination of animal feed are the major threats to animal and human health worldwide.
In the tested poultry samples, the lower level of the total fungal counts was 1 x 10 2 whereas the highest level was 1.83 x 10 5 CFU g -1 . According to the Serbian Regulation the 21.43% of the chicken feeds exceeded the maximum permitted level set to provide food safety and quality assurance. The most of the samples were contaminated with Aspergillus spp. with average AFB 1 concentrations from 4.47 g kg -1 in the chicken feeds to 4.56 g kg -1 in the laying hens feeds. These results are similar to those of previous mycological investigations of poultry feed samples in Serbia (Krnjaja et al., 2010). However, according to the results of Cegielska- Radziejewska et al. (2013), the fungal count was below 1 x 10 4 CFU g -1 in feeds for broilers collected in Poland in 2010, with Aspergillus and Rhizopus as the most common genera. The same authors observed that fungal contamination in poultry feeds from western Poland in 2010 was much lower than in the period of 2006-2008 which accounted 10 4 -10 5 CFU g -1 . Additionally, in Argentina, Monge et al. (2013) established low values of total fungal counts (1 x 10 2 ) in pelleted poultry feed samples with relative high percentages (>40%) of Aspergillus flavus and A. parasiticus isolates. Similarily, total fungal count ranging from 10-10 6 CFU g -1 , and 43.5% of Aspergillus spp. isolates have been determined in poultry feed samples by Greco et al. (2014).
Feed ingredients such as cereals, sunflower, soybean, etc. are suitable for fungal development and mycotoxin contamination. The extreme high aflatoxin levels in maize crops has been recorded in Serbia during the summer of 2012 due to extreme high temperatures and low rainfalls which provoke the high incidence of Aspergillus species (Kos et al., 2012(Kos et al., , 2014Lević et al., 2013;Krnjaja et al., 2013). In the present study, 85.71% of the chicken feeds and 100% of the laying hens feeds were contaminated with AFB 1 . There were 14.29% of the chicken feeds with unacceptable concentrations of AFB 1 . Similarly, Parvathi et al. (2017) reported that aflatoxins have been the most common contaminants in different poultry feeds and feed ingredients collected in India. Furthermore, in Pakistan, Fareed et al. (2014) reported a higher incidence and contamination levels of aflatoxins then ochratoxin A (OTA) in local poultry feeds and feed ingredients.
The growth of Aspergillus species and aflatoxin biosynthesis in cereals and other feed crops are conditioned with suitable environmental factors such as temperature and relative humidity (Patel et al., 2015). In addition, water activity (a w ) and temperature of cereal grains are the main factors that influence the fungal growth and mycotoxin synthesis (Medina et al., 2017). In warm and humid areas, A. flavus and A. parasiticus as the main producers of aflatoxins have been dominant species on maize ears. Optimal conditions for their growth are defined with temperature of 35°C and a w = 0.95, whereas the higher value of water activity, a w = 0.99 and temperature of 33°C are necessary for aflatoxin production (Milani, 2013). It has also been reported that physical factors, such as moisture, relative humidity, temperature, and mechanical damage are critical for mycotoxin production (Bryden, 2012). Higher values of moisture content (20-25%) provide convenient conditions for fungal infections of crops prior to harvest (Magan 2006).
Proper field management practice and use of resistant cereals cultivars are particularly important in mycotoxin control. In addition, during harvest, as a first stage in the cereals production chain, regular and accurate moisture and temperature determination becomes the dominant control measure in the prevention of mycotoxin synthesis. The excessive moisture along the cereal production chain has been the most critical factor affecting the growth and proliferation of fungi, which further increases the risk of feedstuffs mycotoxin contamination (Kana et al., 2013). In this study, even the mean moisture content of the tested poultry feeds was relatively low (<11%), toxigenic fungi and AFB 1 were recorded in the most of the samples, confirming that contamination may occur not only during harvest but also during pre-harvest, incorrect storage and transportation conditions or during poultry feed processing (Binder et al., 2007). The positive correlations between the moisture content and the total fungal count and the levels of AFB 1 were detected which was in accordance with the observations of Greco et al. (2014).

Conclusion
In conclusion, it can be emphasized that the tested poultry feed samples collected in 2016 were mainly contaminated with toxigenic species from the genus Aspergillus, followed by Fusarium and Penicillium genera. In 21.43% of the chicken poultry feeds, the fungal contamination was above the maximum permitted values. The high percentage of positive AFB 1 samples has been registered, whereas in 14.29% of the chicken poultry feeds, the AFB 1 level was also above the regulation limit. These results confirm the necessity of continuous mycological and mycotoxicological control of feeds as the most important measure of control in feed and food safety strategy.

Acknowledgment
This work was supported by the Ministry of Education, Science and Technological Development, Republic of Serbia, projects TR-31023, TR-31033 and OI-46010.