Introduction

Maize (Zea mays L.) is the second most important cereal crop in Ethiopia next to teff (Eragrostis tef [Zucc.] Trotter), cultivated by more than 10 million smallholder farmers (CSA 2021). It is mainly used for domestic household consumption (88%), both as green and dry grain and the rest as animal feed and for sale in the local market (Abate et al. 2015). However, the crop is constrained by different biotic and abiotic factors of which plant pathogens are the major biotic constraints (Tolera et al. 2018).

Maize grains are highly susceptible to toxigenic fungi contamination at all stages of the value chain (from farm to fork) (Geary et al. 2016). Aspergillus, Fusarium, and Penicillium are major fungal genera commonly reported to colonize maize and maize products leading considerably to both quantitative and qualitative losses (Wu et al. 2014). Mycotoxins are secondary metabolites produced by fungi that can contaminate commodities (Leite et al. 2021). As reviewed by Chilaka et al. (2017), about 25% of food crops worldwide, including maize and maize-based products, are at risk to contamination by mycotoxins, which are responsible for considerable loss of the produce. So far, more than 400 mycotoxins are known to occur (Berthiller et al. 2007), varying from region to region depending on the prevailing climatic conditions (Medina et al. 2015), agronomic practices (Borras-Vallverdu et al. 2022), genetic factors, fungal activity, and storage conditions (Ferrigo et al. 2016). As such, they pose serious health risks to consumers (Udomkun et al. 2017; Sarrocco and Vannacci 2018). As a result, a few of these mycotoxins are regulated by several national, regional, and international governing bodies, including the European Commission (EC 2023).

Different studies have revealed the occurrence of toxigenic fungi and associated mycotoxins in maize grain in Ethiopia (Ayalew 2010; Dubale et al. 2014; Garba et al. 2018a b; Getachew et al. 2018; Worku et al. 2019; Yilma et al. 2019; Mohammed et al. 2022). However, these studies focused on stored maize, while fresh harvests were largely ignored. Therefore, the current study was carried out to assess mycotoxin contamination of maize sampled from subsistence farmers’ fields. To the best of our knowledge, this is the first comprehensive study of multi-mycotoxins in freshly harvested maize grown by subsistence farmers in Ethiopia.

Materials and methods

Description of the study area

Fifty-four maize grain samples were collected from farmers’ fields across three administrative zones and nine districts of southwestern part of Ethiopia (Fig. 1 and Table 1) at harvest during November to December of the 2020/21 cropping season. The study areas were selected as they are important maize-producing areas in Ethiopia.

Fig. 1
figure 1

Map of survey areas in southwestern Ethiopia

Full size image
Table 1 Geographic description of surveyed districts in southwestern Ethiopia
Full size table

Sample collection

A purposive multi-stage sampling method was used to select maize-producing zones, districts within zone, and peasant associations (PA) within districts. Potential maize-producing districts and PAs were selected by consulting zone and district Agricultural and Natural Resource Offices. Subsistence farmers’ fields within each PA were randomly selected for sampling, and 1 kg per grain sample was collected from each field. The collected fresh grains were homogenized by thoroughly mixing them repeatedly and kept separately in plastic bags at 4 °C for mycological and eventual mycotoxin analyses.

Mycological analyses

An agar plate method was used to determine the number of infected kernels and type of fungi from the maize samples. From each sample lot, 30 maize kernels were randomly selected and surface sterilized with 1% sodium hypochlorite for 1 min, rinsed three times in sterile distilled water for 30 s, and dried aseptically. From each surface disinfected sample, 10 randomly drawn kernels were placed aseptically per Petri plate (9 cm in diameter) containing potato dextrose agar (PDA) (Eur Pharm, Madrid Spain). The plates were incubated at 25 ± 2 °C for 7 days. Emerging fungal colonies were purified on new PDA and identified morphologically according to Barnett and Hunter (1987), Domsch et al. (1993), and Leslie and Summerell (2006) where applicable.

To confirm the identities of the fungal species, isolates of the Aspergillus genera were grown on Czapek Dox Agar (CZDA), Malt Extract Agar (MEA), and Czapek Yeast Extract Agar (CYA) at 25 °C for 7 days (Klich 2002). Fusarium genera were grown on Spezieller Nahrstoffarmer Agar (SNA) at 25 °C for 7 days exposed to a 12:12-h light/dark regime and identified to species level (Leslie and Summerell 2006). The two genera were isolated and identified to species level based on the micro- and macro-morphological characteristics of the standard identification keys (Klich 2002; Leslie and Summerell 2006; Pitt and Hocking 2009; Nyongesa et al. 2015).

The fungal incidence and frequency of occurrence on maize kernels were calculated as follows:

$$\mathrm{Fungi\;incidence\;on\;kernel\;}(\mathrm{\%})=\frac{\mathrm{Number\;of \;infected \;kernels}}{\mathrm{Total \;number\; of \;kernels}}\times 100$$
$$\mathrm{Frequency}\;\mathrm{of}\;\mathrm{occurrence}\;\%\;=\;\frac{\mathrm A}{\mathrm B}\times\;100$$

where A is the number of samples in which a particular fungus occurred and B is the total number of samples analyzed.

Mycotoxin extraction and analysis

LC gradient grade methanol and glacial acetic acid (p.a.) were purchased from Merck (Darmstadt, Germany), while LC gradient grade acetonitrile and ammonium acetate (MS grade) were purchased from VWR (Leuven, Belgium) and Sigma-Aldrich (Vienna, Austria), respectively. Water was successively purified by reverse osmosis with an ELGA PURELAB Ultra analytic system from Veolia Water (Bucks, UK).

The maize grain samples were thoroughly mixed and ground through a 1 mm sieve. Then, a homogenized 5 g flour from each sample was weighed into a 50 ml polypropylene tube, and metabolites of fungi were extracted for 90 min on a GFL 3017 rotary shaker (GFL, Burgwedel, Germany) using 20 ml of extraction solvent that consisted of acetonitrile/water/acetic acid (79:20:1, v/v/v). Extracts were diluted 1:1 (v/v) with a dilution solvent (acetonitrile/water/acetic acid, 20:79:1, v/v/v), and 5 μL of each diluted extract was used for further analysis.

Liquid chromatography-tandem mass spectrometric (LC-MS/MS) analysis was conducted for the simultaneous determination of multiple microbial metabolites following Sulyok et al. (2020). Briefly, a QTrap 5500 LC-MS/MS System (Applied Biosystem, Foster City, CA, USA) equipped with a Turbo Ion Spray coupled to a 1290 Series HPLC System (Agilent, Waldbronn, Germany) was used. Chromatographic separation was carried out at 25 °C on a Gemini C18 column (150 × 4.6 mm i.d., 5 μm particle size) and a C18 4 × 3 mm i.d. security guard cartridge (Phenomenex, Torrance, CA, US). ESI-MS/MS was performed in the time-scheduled multiple reaction monitoring (MRM) mode both in positive and negative polarities in two separate chromatographic runs per sample by scanning two fragmentation reactions per analyte. Compound-specific LC-MS/MS parameters can be seen in the supplementary table of Sulyok et al. (2020).

Confirmation of positive analyte identification was obtained by the acquisition of two MRMs per analyte (with the exception of moniliformin and 3-nitropropionic acid, which exhibited only one fragmentation), which yielded 4.0 identification points in accordance with the European Commission Decision No. 2002/657 (European Commission 2002). The LC retention time and the intensity ratio of the two MRM transitions also agreed with the related values of authentic standards within 0.03 min and 30% relative intensity ratio, respectively. Quantification was performed using external calibration based on serial dilution of a multi-analyte stock solution, and results were corrected for apparent recoveries (Sulyok et al. 2020). The limits of detection (LOD) and quantification (LOQ) were determined following the EURACHEM guide based on reputability data on the lowest spiking level. The method was scrutinized in a proficiency testing scheme organized by BIPEA (Bureau Interprofessionnel d’Etudes Analytiques), International Bureau for Analytical Studies, Gennevilliers, France. The current rate of 96% of the > 2100 submitted results is in the range of − 2 < z <  + 2.

Results and discussion

Major fungi associated with maize grains in the study areas

Results of the current study revealed wider distribution of fungi across the study areas, with fungal incidences ranging between 43 and 99%. The highest mean kernel incidence was recorded at Ilu Ababora zone 88.8% followed by Buno Bedele (81.3%) and Jimma (81%). Fungal species isolated from current samples were those belonging to Fusarium (100%), Aspergillus (60%), and Penicillium (58%) (Fig. 2). These results agree with Binyam and Girma (2016) and Garbaba et al. (2018a, b) who reported Aspergillus, Fusarium, and Penicillium as the most dominant genera infecting stored maize in Jimma area. Dubale et al. (2014) also reported Fusarium, Penicillium, and Aspergillus from maize stored under farm storage conditions within the same area. The same genera were also reported as important storage fungi in other parts of Africa (Ekwomadu et al. 2018).

Fig. 2
figure 2

Incidence of major toxigenic fungi in fresh maize grain samples from nine districts in southwestern Ethiopia

Full size image

At species level, Fusarium verticillioides was the most frequent (99%), followed by Fusarium graminearum (91%), Aspergillus niger (66%), and Aspergillus flavus (47%) (data not shown). Getachew et al. (2018) and Tsehaye et al. (2017) have also reported F. verticillioides as the major contaminant of maize grain in Ethiopia, while Yilma et al. (2019) identified A. flavus as the most frequently isolated species from stored maize kernel. Results suggest that contamination of maize grains by fungi starts right on the field (pre-harvest) and continues during storage (post-harvest).

Mycotoxins and other fungal metabolites in maize grains in the study areas

A total of 164 fungal metabolites were present in maize grain samples collected from southwestern Ethiopia at levels higher than the limits of detection, suggesting a 29% increase from the year 2015 (Getachew et al. 2018). This might have been caused by recent changes in environmental conditions favoring mycotoxin contamination. The fresh grain samples in the current study may also have higher moisture content than the stored grains creating favorable conditions for microbial activity.

The detected metabolites are grouped into 10 different categories. Penicillium metabolites were the most dominant ones representing 30% of the detected metabolites. They were followed by Fusarium metabolites that accounted for 27% of the metabolites contaminating the samples at levels higher than the limit of detection (LOD). Eighteen (11%) of the 164 metabolites were major mycotoxins and their derivatives. Unspecified metabolites, metabolites from other fungal genera, i.e., Alternaria, Ascochyta, and Trichoderma, and bacterial metabolites, were detected although they represented only lower proportion (1–8%) of the detected metabolites.

The incidence of mycotoxin contamination also differed across the metabolite groups (Fig. 3). Accordingly, Fusarium metabolites and unspecified metabolites occurred in all (100%) of the samples. They were followed by Penicillium metabolites that were detected in 96% of the samples, and major mycotoxins and derivatives and bacterial metabolites that were present in 92% of the samples each. Getachew et al. (2018), Mesfin et al. (2022), and Mohammed et al. (2023) reported up to 26% post-harvest maize grain contamination by Penicillium and Fusarium metabolites, with Penicillium metabolites being the most frequently detected. Penicillium metabolites were also recorded at high prevalence in stored sorghum grains from Ethiopia (Mohammed et al. 2022). To the best of our knowledge, this is the first report on the contamination of maize grains by bacterial metabolites in Ethiopia although bacterial metabolites were detected in other cereals like sorghum and millet (Chala et al. 2014; Mohammed et al. 2022).

Fig. 3
figure 3

Frequency of mycotoxin groups contaminating maize grains in southwestern Ethiopia

Full size image

The frequency of occurrence of mycotoxin groups varied across the sample collection zones (Table 2). Ten metabolite groups were detected in maize grains from Jimma and Illu Ababora zones, with Fusarium, bacteria, and unspecified metabolites being the most common. The highest frequency of major mycotoxin and their derivatives were recorded in Buno Bedele (100%) followed by Ilu Ababora zone (94%), while Fusarium and unspecified metabolites occurred in all maize grain samples across the three zones. Maize samples were collected from areas that represent different altitude ranges (1600 to above 2000 m.a.s.l.). Major mycotoxins and derivatives were detected at altitudes ≥ 1800 m.a.s.l., but their incidence varied across altitude gradients. Eighty percent of the samples were contaminated by major mycotoxins and derivatives in low altitude areas, while the incidence of these mycotoxins increased with altitudes and levels up at around 2200 m.a.s.l.

Table 2 Frequency (%) of mycotoxin groups across zones of southwestern Ethiopia (N = 18 samples/zone)
Full size table

Major mycotoxins and derivatives

A total of 18 major mycotoxins and derivatives contaminated the current maize grain samples (Table 3). All the maize grain samples analyzed in the current study were contaminated by one or more of these toxins. All but one (citrinin) of the major mycotoxins and derivatives were of Fusarium origin. Citrinin was present only in a sample from Dedesa district of Southwest Ethiopia, at concentration of 88.7 µg/kg. This metabolite can be produced by Aspergillus and Penicillium genera (Kamle et al. 2022). Current results are in line with the findings of the mycological analysis in which Fusarium spp. were among the most dominant fungi contaminating the grain samples. The present findings also agree with that of Getachew et al. (2018) and Tebele et al. (2020), who reported 100% grain contamination of stored maize with mycotoxins although current samples were fresh harvests. Worldwide food crops are contaminated by mycotoxins at a prevalence of up to 80% (Eskola et al. 2020).

Table 3 Major mycotoxins and derivatives detected in maize grain samples collected from southwestern Ethiopia (N = 54 samples)
Full size table

Zearalenone was the most prevalent major mycotoxin, occurring in 74% of the samples at average concentration of 154 µg/kg. Five maize grain samples contained zearalenone at levels ranging from 395 to 1310 µg/kg, which are above the maximum EU tolerable level (350 µg/kg) (European Commission 2007). Getachew et al. (2018) and Mohammed et al. (2023) also reported zearalenone to be the most prevalent in stored maize from Ethiopia with concentration up to 1656 and 3750 µg/kg, respectively, although at relatively higher concentrations than those in the current study. This is not surprising as those found in the current study were recovered from fresh harvests. This is because zearalenone contamination occurs both at harvest and during storage (Ropejko and Twaruzek 2021). Pleadin et al. (2012) also recorded high prevalence of same mycotoxin from harvested maize, but concentrations (max. 5.11 µg/kg) were much lower than those reported herein.

Nivalenol, another Fusarium mycotoxin, was the second most frequently detected mycotoxin with 63% prevalence and mean concentration of 1075 µg/kg recorded. Getachew et al. (2018) reported a 47% prevalence and maximum concentration of 793 µg/kg of this mycotoxin from stored maize in Ethiopia. Current results also relate to a report from Brazil in which the toxin was detected in 75.5% of samples at a mean concentration of 256 µg/kg (Oliveira et al. 2017). Nivalenol, nivalenol glucoside, and fusarenon-X metabolite were detected in current maize grain samples with high levels up to 17,300, 825, and 149 µg/kg, respectively.

In the current study, deoxynivalenol was detected in 31% of the samples at levels between 15.9 and 5140 µg/kg. Contamination of grains by this toxin was also reported by Xing et al. (2017) both in pre-nature drying maize (50.7 to 776.6 μg/kg) and post-nature drying maize (5.8 to 9843.3 μg/kg). Ayalew (2010) reported 5% stored maize grain contamination with levels of up to 700 µg/kg, while Getachew et al. (2018) and Mesfin et al. (2022) reported, respectively, 42 and 29% stored maize grain contamination by the same mycotoxin at levels of up to 221 µg/kg. The findings in these studies suggest a 6- to eightfold increase in DON contamination of maize grains over the past decade.

Four types of fumonisins (B1, B2, B3, and A1) were detected in the present study. Fumonisin B1 was the most frequently detected of this group occurring in 41% of maize grain samples at levels ranging from 9.46 to 6770 µg/kg. Similar prevalence of Fumonisin B1 was recorded in stored maize samples in Kenya (overall 38%) and Victoria region of Kenya (53%) (Kagot et al. 2022). Mohammed et al. (2022) reported FB1 in all maize grain samples from Ethiopia with levels ranging between 22.6 and 1058 µg/kg in stored maize. In North China, all 44 maize samples collected during pre- and post-nature drying periods were contaminated by FB1 with a mean level of 133 µg/kg (Xing et al. 2017). FB2, FB3, and FA1 were detected in 17–26% of samples at 385, 212, and 32.6 µg/kg, respectively. The present findings confirmed the widespread contamination of maize by Fusarium metabolites across locations and altitude gradients. Our findings agree with previous reports on these mycotoxins from Ethiopia (Getachew et al. 2018; Mesfin et al. 2022; Mohammed et al. 2023; Tsehaye et al. 2017; Worku et al. 2019). However, it should be noted that the findings reported in our study were from fresh harvests, while the previous studies were on stored maize.

Miscellaneous Fusarium metabolites

Bikaverin was the most prevalent Fusarium metabolite, occurring in 98% of the samples followed by moniliformin and aurofusarin (96% each) (Table 4) being recovered from samples at mean concentrations of 61.8, 1006, and 4262 µg/kg, respectively. Relatively high prevalence of these Fusarium metabolites was also reported from northern Serbia in two consecutive years (Radic et al. 2021). Moniliformin is considered a leading emerging toxin in South Africa with a high prevalence of 98% at a maximum contamination level of 1130 µg/kg (Ekwomadu et al. 2020). The presence of this mycotoxin at unusually higher levels in samples reported herein may confirm its increasing importance in Sub-Saharan Africa. A metabolite produced by Fusarium spp. (Nihei et al. 1998; Ezekiel et al. 2020), W493B was also present in 22% of the samples at higher levels (mean 162 µg/kg; max. 652 µg/kg).

Table 4 Fusarium metabolites other than major mycotoxins and derivatives contaminating maize grains in southwestern Ethiopia (N = 54 samples)
Full size table

Aspergillus metabolites

None of the current maize grain samples were contaminated by aflatoxins and ochratoxins. However, 21 metabolites from aflatoxin/sterigmatocystin pathway were detected in the current study (Table 5). Among them, averufin and sterigmatocystin were the most prevalent with 26 and 24% frequencies and mean concentrations of 0.46 and 2.49 µg/kg, respectively. However, aflatoxins and ochratoxins have been previously reported in stored maize at varying levels (Ayalew 2010; Chauhan et al. 2016; Getachew et al. 2018; Worku et al. 2019). The presence of metabolites from aflatoxin/sterigmatocystin pathway suggests the likely contamination of the grains with Aspergillus versicolor (Pitt 2014) or contamination by other fungal species that might have interfered with the biosynthesis of aflatoxins at the sterigmatocystin step. Additionally, the absence of aflatoxins in the current samples could have been caused by lower temperatures that typically prevail in Ethiopia during October–December.

Table 5 Aspergillus metabolites present in maize grain samples (N = 54 samples)
Full size table

Penicillium metabolites

Questiomycin derivative, pestalotin, and 7-hydroxypestalotin were the most prevalent Penicillium metabolites contaminating 94, 89, and 87% of the maize grain samples, occurring at mean concentrations of 185, 27.2, and 36.6 µg/kg, respectively (Table 6). The highest frequency of occurrence of those three metabolites on freshly harvested maize was in line with a previous study on post-harvest maize by Mohammed et al. (2023). However, their levels were about 30% higher than previously reported. Rugulovasine A was also detected at higher prevalence (74%) and concentration (mean 241 but up to 3420 µg/kg) compared to earlier reports of 33% prevalence and a maximum concentration of 1159 µg/kg (Getachew et al. 2018).

Table 6 Penicillium metabolites contaminating maize grains in southwestern Ethiopia. (N = 54 samples)
Full size table

Metabolites of other fungi

Maize grain samples analyzed in the current study were contaminated by metabolites produced by a diverse range of fungi in addition to those of the Fusarium, Penicillium, and Aspergillus genera (Table 7). Macrosporin, as an example, was the most frequently detected mycotoxin of Alternaria contaminating 56% of the samples at a mean level of 11.6 µg/kg. Trichoderma and Ascochyta metabolites were also detected but at a low frequency of 2–6%. Additional 12 metabolites were found contaminating the current maize sample at frequencies ranging from 1.85 to 53.7% (Table 7). These included radicicol and abscisic acid that contaminated 53.7 and 29.6% of the samples at mean concentrations of 143 and 973 µg/kg, respectively.

Table 7 Alternaria, Trichoderma, Ascochyta, and additional fungal metabolites detected in maize grain samples collected from southwestern Ethiopia (N = 54 samples)
Full size table

Conclusion

The present study shows that all the maize grains collected from farmers’ fields of southwestern Ethiopia were contaminated with a multitude of fungi and mycotoxins. The fungi genera Fusarium, Aspergillus, and Penicillium were the major contaminants of maize grain in the study area at harvest. Unlike previous studies that focused on post-harvest maize, the current work demonstrated the widespread contamination of maize at harvest stage. The presence of multiple fungi and associated mycotoxins, as reported at levels higher than international standards in this study, raises serious concerns about maize-based food and feed safety.

Therefore, special attention is needed to protect maize from these harmful compounds at pre- and post-harvest to ensure food security and safety in Ethiopia and enhance the export of maize grains from the country. There should be a concerted effort to quantitatively detect toxigenic fungi and associated mycotoxins across all stages of the maize value chain in the country. Additional studies should also be carried out to determine the biophysical factors that predispose maize to infection by toxigenic fungi and contamination with mycotoxins. The current work also demonstrates the need to create awareness on fungal and mycotoxin contamination along the maize value chain among the various actors. Additionally, sustainable monitoring and management strategies for mycotoxigenic fungi should be developed and put in place to improve food and feed safety and enhance trade. Harvest and post-harvest operations including timely harvesting, proper drying, sorting grains, and use of improved storage that prevents/minimizes contamination by toxigenic fungi and subsequent mycotoxin contamination should also be given due attention to improve grain quality and safety and protect consumers’ health and social resilience of local communities.