Pistachio nuts (Pistacia vera L.) are an economically important tree nut crop for many countries including the United States, Iran, Southern Europe and China. Of these, Iran and the United States are the major producers and exporters of pistachio nuts (FAOSTAT, 2020). Pistachio nut quality and safety has thus become an important factor because they are prone to infection by species from the Aspergillus section Flavi group, especially A. flavus pre-harvest, if the shells become split early, or due to poor post-harvest drying or storage practices (Kaminiaris et al., 2020). This can result in significant contamination with aflatoxins (AFs). This has resulted in many countries including the EU having strict legislative limits on the maximum contamination levels with aflatoxin B₁ (AFB₁) or total AFs allowable. Indeed, for a few years Iranian pistachio nuts were excluded for importation into EU countries because of batches consistently exceeding the prevailing legislative limits for AFs (Bui-Klimke et al., 2014). This required action, especially in the harvest and post-harvest phases of the chain to be improved to reduce the contamination levels with these carcinogenic mycotoxins to levels to meet the legislative maximum limits.

Pistachio nuts in various processed forms are very popular in the Middle East and have been imported into most countries including the Kingdom of Saudi Arabia (KSA) (Mozaffari, 2012). Examination of a range of such pistachio nuts from markets in KSA originating from various producers and processors showed that Aspergillus section Flavi strains was present in many of the samples as well as contamination with AFs (Nawar, 2008; Baazeem, 2018). In addition, in other countries including the United States, Iran, Spain and Greece A. flavus has often been the predominant species encountered in pistachio nuts (Ciegler, 1972; Mojtahedi et al., 1979; Fernane et al., 2010a,b; Varga et al., 2011).

There is interest in a better understanding of the key abiotic factors which affect infection and colonization of pistachio nuts pre-harvest during early shell splitting, and post-harvest during drying and storage (Georgiadou et al., 2012). For example, drought stress can lead to increased shell spliting and allow inoculum of A. flavus to colonize the nuts and contaminate them with AFs (Sommer et al., 1986; Doster and Michailides, 1995; Kaminiaris et al., 2020). In addition, harvested pistachio nuts are very hygroscopic, because of their high lipid content, and can re-absorb moisture from the atmosphere, especially during storage or transport that increases the water availability to levels that can become conducive to growth of such xerophilic fungi and AFs contamination. Generally, to maintain and conserve quality, pistachios need to be dried to <6–7% moisture content (= ≤0.70 water activity, a_w) and then stored without an increase in water availability during storage or transport and processing phases to avoid the risk of A. flavus colonization and contamination with AFs.

It is thus important to understand the relationship between interacting abiotic factors, especially of a_w and temperature on the colonization by A. flavus and contamination of the pistachio nuts with AFB₁. Some studies have examined germination, growth and AFB₁ production by single strains of A. flavus isolated in Spain on pistachio nut-based media and on stored pistachios (Marin et al., 2012; Aldars-García et al., 2015). Interestingly, having tested different storage m.c. conditions the studies suggested that storage at ≤10% was necessary to minimize AFB₁ contamination of pistachios to below the EU legislative levels. However, comparisons were not made between different strains of A. flavus isolated from raw or roasted, salted nuts. While there is a significant amount of data on A. flavus ecology and water relations in maize and peanuts, less is available on pistachio nuts (Giorni et al., 2008; Abdel-Hadi et al., 2012).

The objectives of this study were to examine the effect of interacting abiotic factors of a_w x temperature on growth and AFB₁ production by four different A. flavus strains isolated from pistachio nuts, on (a) pistachio nut-based media and (b) in stored raw pistachio nuts. The data was used to build boundary models for growth and AFB₁ production to identify the overall optimal and marginal conditions which represent the highest and lowest risks for growth and toxin contamination.

This study has examined both growth and AFB₁ production by a group of strains of A. flavus isolated from pistachio nuts. This has provided useful data in relation to relative colonization and AFB₁ production in pistachio nut-based media modified with both salt and a non-ionic solute, glycerol. This was to reflect both raw unsalted and roasted salted pistachio nuts to identify relative similarities and differences in terms of strain resilience and effects on toxin production. Changes in water availability, regardless of the solutes used had no significant differential effect on growth or indeed AFB₁ production. The contour maps for both these treatments in relation to temperature x a_w interacting abiotic factors were similar, for both optimum and marginal conditions, using 1 or all 4 A. flavus strains.

Previous studies by Aldars-García et al. (2017a) examined initial germination, growth and AFB₁ production on 3% pistachio extract medium. However, the studies were limited to 25°C and 0.85 and 0.87 a_w which were marginal for growth and toxin production. The time for germination, growth rates and time to visible colonies was estimated. They suggested that even at these a_w levels initiation of AFB₁ production could occur within 6 days. This suggests that at 30–35°C where germination rates, growth and toxin production are much more rapid, toxin contamination might occur even earlier. Perhaps a kinetic study at 30–35°C and these a_w levels would provide useful data to predict the risks of AFB₁ contamination. However, the present study has shown that toxin production rates were significantly higher on the milled PNA media than in raw pistachio nuts. Thus, care is needed in using such predictions of relative risks based on temperature x a_w conditions using in vitro studies. More data sets are required on relative AFB₁ production by A. flavus strains in stored pistachio nuts to obtain more accurate predictions of risks relative to the legislative limits (Aldars-García et al., 2017b).

In the present study, A. flavus strains were able to grow effectively when incubated at 25–35°C, over a range of a_w levels, both in vitro and in situ. However, the range of conditions for AFB₁ production was narrower with optimum production at 30°C in modified media, at 0.98 and 0.955 a_w; and at 35°C + 0.98 a_w by strain AB3 on raw pistachio nuts. Studies by Marin et al. (2012) examined the effect of 10–30% m.c. of pistachio nuts and storage temperatures of 10–42°C and inoculated with an A. flavus strain. They stored the pistachio nuts for longer periods of time (up to 60 days) and used kinetic and probability models to identify growth/no growth and toxin/no toxin conditions. The probability model showed good predictions for growth/no growth after 60 days incubation. Similarly, the probability for AFs presence was predicted in 89% of the cases. They thus suggested that the EU legislative limits would be exceeded if pistachio nuts were stored for up to one-month at 20°C if the m.c. was >10–12%. The present study included relevant a_w levels equivalent to approx. 10–30% m.c. We obtained little AFB₁ contamination at approx. ≥0.90 a_w. However, our studies were limited to 10 days and longer storage periods would have perhaps provided more information on the absolute minima for growth and toxin production. However, we were able to obtain data on different strains which suggest that the overall ecological behavior was relatively similar.

In addition, it is clear that colonization rates in vitro were generally more rapid than that observed in situ. In addition, AFB₁ production was significantly higher on the milled PNA media when compared with that produced in situ. This could be because of the better access to the nutrients in the milled pistachio nut-based media allowing the strains to grow faster and produce more toxins. In situ, the surface area is different and initial colonization and access to the nutrients may be slower or delayed. It was interesting that these strains isolated from pistachio nuts could all grow well at 35°C. Usually A. flavus strains from maize and peanuts are more sensitive to this temperature with optima closer to 28–30°C (Giorni et al., 2008; Abdel-Hadi et al., 2012; Bhatnagar et al., 2018). Thus, ecologically this may be important when examining intervention strategies to reduce AFs contamination of pistachio nuts during drying, storage and processing phases. In addition, the good tolerance of 35°C suggests resilience also in relation to climate-related more extreme abiotic factors (Medina et al., 2017).

Previous studies by Fernane et al. (2010a, b) showed that about 70% of A. flavus strains from a total of 204 were able to produce AFB₁ and AFB₂ on pistachio nuts. Aldars-García et al. (2015) used a predictive modeling approach to predict potential AFB₁ contamination of pistachio nuts based on environmental conditions. They found that using this approach it was possible to predict 70–81% of samples correctly with regard to growth and in 67–81% of the cases contamination with AFB₁ based on the models developed from their in vitro studies. Recently a mechanistic model has been developed using the available data on the life cycle of A. flavus ecologically as well as the pistachio development and the prevailing abiotic conditions (Kaminiaris et al., 2020). Ecological data on a wider number of strains isolated from pistachio nuts obtained in the present study might also be beneficial for the further development of the “AFLA-PISTACHIO” model for improving the accuracy and validation of evaluating the relative risks in different pistachio-producing countries.

This study has shown that optimum interacting abiotic conditions of temperature × a_w for growth and AFB₁ production by up to four strains of A. flavus isolated from pistachio nuts was very different from the ecology of strains colonizing maize and peanuts. It was interesting to note that there was no difference in growth or toxin production in PNA media modified with salt or the non-ionic solute glycerol. This also suggests that these strains have evolved to have better resilience to elevated temperatures although the water stress tolerance may well be similar. In addition, AFB₁ production on raw pistachio nuts was much lower than on milled pistachio nut media. The contour maps clearly show the optimum and marginal conditions of these two interacting abiotic factors on both growth and toxin production. It is recommended that the safe storage m.c. for pistachio nuts is <8–10% for medium term storage under ambient temperature conditions (20–25°C). However, they are very hygroscopic because of their fatty acid content. Thus, they can easily re-adsorb moisture during bulk transport through different climatic regions and poor storage, resulting in a deterioration in quality and an increase in toxin contamination.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AB carried out the research work. EG-C carried out the modeling work. AM and NM supervised the research and drafted the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.