Aspergillus flavus is the main producer of aflatoxin B1, one of the most toxic contaminants of food and feed. With global warming, climate conditions have become favourable for aflatoxin contamination of agricultural products in several European countries, including Serbia. The infection of maize with A. flavus, and aflatoxin synthesis can be controlled and reduced by application of a biocontrol product based on non-toxigenic strains of A. flavus. Biological control relies on competition between atoxigenic and toxigenic strains. This is the most commonly used biological control mechanism of aflatoxin contamination in maize in countries where aflatoxins pose a significant threat. Mytoolbox Af01, a native atoxigenic A. flavus strain, was obtained from maize grown in Serbia and used to produce a biocontrol product that was applied in irrigated and non-irrigated Serbian fields during 2016 and 2017. The application of this biocontrol product reduced aflatoxin levels in maize kernels (51–83%). The biocontrol treatment had a highly significant effect of reducing total aflatoxin contamination by 73%. This study showed that aflatoxin contamination control in Serbian maize can be achieved through biological control methods using atoxigenic A. flavus strains.

Contamination of maize with aflatoxins and fumonisins is one of the major food safety concerns worldwide. Knowing the contamination in advance can help to reduce food safety risks and related health issues and economic losses. The current study aimed to develop forecasting models for the contamination of maize grown in Serbia with aflatoxins and fumonisins. An integrated modeling approach was used, linking mechanistic modeling with artificial intelligence, in particular Bayesian network (BN) modeling. Two of such combined models, i.e., the prediction model for aflatoxins (PREMA) and for fumonisins (PREFUM) in maize, were developed. Data used for developing PREMA were from 867 maize samples, collected in Serbia during the period from 2012 to 2018, of which 190 were also used for developing PREFUM. Both datasets were split randomly in a model training set and a model validation set. With corresponding geographical and meteorological data, the so-called risk indices for total aflatoxins and total fumonisins were calculated using existing mechanistic models. Subsequently, these risk indices were used as input variables for developing the BN models, together with the longitudes and latitudes of the sites at which the samples were collected and related weather data. PREMA and PREFUM were internally and externally validated, resulting in a prediction accuracy of PREMA of, respectively, 83 and 70%, and of PREFUM of 76% and 80%. The capability of PREMA and PREFUM for predicting aflatoxins and fumonisins contamination using data from the early maize growth stages only was explored as well, and promising results were obtained. The integrated approach combining two different modeling techniques, as developed in the current study, was able to overcome the obstacles of unbalanced data and deficiency of the datasets, which are often seen in historical observational data from the food safety domain. The models provide predictions for mycotoxin contamination at the field level; this information can assist stakeholders of the maize supply chain, including farmers, buyers/collectors, and food safety authorities, to take timely decisions for improved mycotoxin control. The developed models can be further validated by applying them into practice, and they can be extended to other European maize growing areas.

Maize is one of the leading export products in the Republic of Serbia. As a country where economic development depends on agriculture, maize production plays a critical role as a crop of strategic importance. Potential aflatoxin contamination of maize poses a risk to food and feed safety and tremendous economic losses. No aflatoxin contamination of maize samples harvested in 2019 and 2020 in different localities in the Republic of Serbia was detected by the Enzyme-Linked Immunosorbent Assay (ELISA) test and High-Performance Liquid Chromatography (HPLC) method. On the other hand, the Cluster Amplification Patterns (CAP) analyses of the isolated Aspergillus flavus strains from 2019 maize samples confirmed the presence of key biosynthesis genes responsible for aflatoxin production. Artificial inoculation and subsequent HPLC analysis of the inoculated maize samples confirmed the high capacity of the A. flavus strains for aflatoxin production, pointing to a high risk of contamination under favorable conditions. Prevention of aflatoxin contamination is primarily based on A. flavus control, where biocontrol agents play a significant role as sustainable disease management tools. In this study, antagonistic activity screening of the novel strains belonging to the Bacillus genus indicated superior suppression of A. flavus strains by two Bacillus strains isolated from the rhizosphere of Phaseolus vulgaris.

Soft rot and blackleg are common diseases affecting potato (Solanum tuberosum) production in Serbia. Pectinolytic plant pathogens belonging to the genera Pectobacterium cause soft rot and wilt diseases by plant cell wall degradation. These opportunistic phytopathogens lead to considerable economic losses in many potato-growing regions worldwide and are listed among top 10 plant pathogenic bacteria (Mansfield et al. 2012). Potato plants (cv. VR808) with symptoms of wilting, slow growth, stem blackening and tubers softening, were collected from a commercial potato field in Zobnatica (Serbia) in July 2019 and subjected to analysis. All symptoms occurred in the same field and the incidence of symptomatic plants was approximately 5%. Isolation was performed from 10 randomly chosen potato plant and tuber samples, expressing wilting and soft rot symptoms. Plant tissue was surface-disinfected and 1 cm length sections from the margins of lesions were macerated in sterile distilled water for 25 min and streaked on nutrient-agar medium. After 48 h of incubation at 26°C, predominant shiny, cream-colored, round colonies were obtained from all samples. Three representative isolates (MMZKVR1, MMZCVR2, and MMZKVR3) from independent samples were selected randomly and subjected to biochemical and pathogenicity tests. Isolates were gram-negative, nonfluorescent facultative anaerobes, exhibiting pectinolytic activity on potato tuber slices and hypersensitive response on tobacco leaves. They expressed catalase activity but did not express oxidase or acid phosphatase activity or produce indole. All strains grew at 37°C, in 5% NaCl, and reduced nitrate. Pathogenicity of the obtained isolates was tested on 3-week-old healthy potato plants (cv. VR808 and cv. Kiebitz) grown in commercial Baltic Tray Substrate (Hawita) in the greenhouse, as well as on potato tubers of the same varieties. Three potato plant stems per isolate were inoculated by the toothpick piercing method (Duarte et al. 2004) using bacterial suspension (approx. 1 × 108 CFU/ml). Inoculated plants were incubated under plastic bags in a greenhouse at 25 ± 2°C. Blackleg symptoms and stem wilting developed 48 hours after inoculation. No symptoms were observed on plants inoculated with sterile toothpicks dipped in sterile distilled water. The pathogen was re-isolated from symptomatic plants, fulfilling Koch's postulates and sequencing of 16S rDNA confirmed the originally isolated pathogen. Three potato tubers per isolate were inoculated by toothpicks dipped in bacterial suspension (approx. 1 × 108 CFU/ml). Inoculated tubers were placed in a sealed plastic container at 25 ± 2°C. Treatment with sterile distilled water was used as a negative control. Softening of the tissue around the inoculation point developed within 48 h from inoculation, and no symptoms developed on the control tubers. For molecular analyses, total DNA of the isolates was extracted using the DNeasy Plant Mini Kit (Qiagen). The isolates were not detected in diagnostic PCR assays using specific primers Br1F/L1R for the detection of P. brasiliense (Duarte et al. 2004) and primers EXPCCF/EXPCCR for P. catotovorum subsp. carotovorum (Kang et al. 2003). The 16S rDNA PCR amplification was performed using the universal PCR primer pair 27F/1492R (Fredriksson et al. 2013) and followed by Sanger sequencing (Macrogen Europe BV). The BLASTn analysis of sequences (GenBank Accession Numbers MZ048661, MZ048662, and MZ157274) revealed 100% query coverage and 100% identity to the sequences of Pectobacterium punjabense in NCBI (MT242589 and CP038498) isolated from potato in China and Pakistan (Sarfraz et al. 2018), respectively. All three obtained isolates were proposed to belong to Pectobacterium punjabense sp. nov. To further validate the identification, isolate MMZCVR2 of P. punjabense was selected for multilocus sequence analyses of 5 housekeeping genes (gyrA, recA, recN, rpoA and rpoS). The gyrA (MZ161817), recA (MZ161818), recN (MZ161819), rpoA (MZ161820) and rpoS (MZ161821) sequence analysis showed the highest nucleotide identity (99.44 to 100%) with P. punjabense strain SS95 (Sarfraz et al. 2018) previously deposited in NCBI GenBank database. To our knowledge, this is the first report of blackleg and soft rot caused by P. punjabense on potato in Serbia. Pectobacterium punjabense is a newly described species causing soft rot and blackleg disease in potato plants (Sarfraz et al. 2018). Its current geographic distribution is not well-described but important to know since soft rot bacteria are easily transported long distances in latently infected seed tubers and can cause significant economic losses in potato production worldwide.

Pectobacterium is a diverse genus which comprises of multiple destructive bacterial species which cause soft rot/blackleg/wilt disease complex in a wide variety of crops by employing high levels of virulence factors. During the 2018, 2019 and 2020 potato growing seasons, numerous outbreaks of bacterial wilt, stem blackleg and tuber soft rot were recorded, and symptomatic plant samples from ten localities in the Province of Vojvodina (Serbia) were collected and analysed. Bacterial soft-rot pathogens were detected in 63 samples using genus and species-specific primers. Through 16S rRNA Sanger sequencing of 19 representative isolates, the identity of P. brasiliense (73.7%), P. punjabense (15.8%), and P. carotovorum (10.5%) species were revealed. To further validate the identification, genotypic profiling of Pectobacterium strains using rep-PCR (ERIC, BOX, REP) was conducted for 25 selected isolates and the phylogenetic assessment based on four selected housekeeping genes (gyrA, recA, rpoA, and rpoS). Physiological and biochemical properties were analysed using basic microbiological tests and VITEK® 2 GN card, and pathogenicity was confirmed on cv. VR808 and cv. Desiree potato tubers and plants. This study confirmed the distinctiveness of the newly described P. punjabense in Serbia as well as the high diversity of Pectobacterium brasiliense and Pectobacterium carotovorum species in Serbia.