1. Introduction
Maize is one of the most important cereals in the world. This crop is a regular host of toxigenic fungi infecting the ears, which can cause very high losses in crop yield. In contrary to wheat where F. graminearum is the leading toxigenic species nearly everywhere, the situation is more complex in maize, where at least two leading species exist from Fusarium and Aspergillus, with similar significance, but different amounts in different years.
Pathogens: Logrieco et al. [
1] mentioned 19
Fusarium spp. of maize in Europe. However, the most important species in most regions are
F. graminearum,
F. verticillioides. Therefore, their control is needed in every corn production area. In Hungary,
F. graminearum and
F. verticillioides are the most important [
2]. In drier years the latter species is dominant. In wet years, such as 1974, the number of species increased to 16, including
F. graminearum (30%),
F. verticillioides (27%),
F. culmorum (4%),
F. fusarioides (3%),
F. avenaceum (1%),
F. sporotrichioides (6%),
F. poae (2%),
F. semitectum (4%) and several others with 554 isolates. The dry year of 1975 saw a 28% occurrence of
F. graminearum, 69% of
F. verticillioides, and the rest made up 3% of the species represented by several entries of
n = 645 isolates. The case is similar in all countries where data from
Fusarium surveys exist (Mesterházy [
3],
Table 1). Because
F. moniliforme was reclassified to
F. verticillioides [
4], the position of
F. graminearum,
Gibberella zeae) remained. Many new
Fusarium spp. were described or reclassified, but these changes do not interfere with our two main causing agents. As more than 90% of the
F. graminearum isolates of wheat belong to
F. graminearum stricto senso [
5], we focused in this study on this specimen.
F. boothi, for example, produces nivalenol (2 isolates of the 29) [
5]. In 2007,
A. flavus occurred at a higher rate in Hungary, and aflatoxin-contaminated grain was also detected. In 2012, the aflatoxin contamination rate was high, and 2017 a smaller epidemic occurred. Due to global warming, the occurrence of aflatoxin in the field was predicted to reach significant rates in epidemic years; therefore, resistance to
A. flavus was also chosen for testing.
Environmental/weather conditions, toxigenic species and toxins. The three main pathogens mentioned need humid and moderately warm weather to infect silks and later the cobs. Thereafter,
F. graminearum needs humidity and warmth, but not very warm or hot weather,
F. verticillioides needs warmer and drier conditions, and
Aspergillus flavus needs the hottest conditions, especially for toxin accumulation [
3]. This means that in some years, only fumonisin, DON or aflatoxin occur; in another years, all combinations are possible at a high risk level. Experience throughout many decades shows that high toxin contamination is associated with outbreaks of significant epidemics. From this we may conclude that susceptibility relies somehow on disease epidemics and toxin contamination as summarized by Clements and White [
6], Mesterházy et al. [
3], Munkvold [
7], Reid et al. [
8]. For this reason, breeding for resistance has become a significant goal to decrease both disease severity and toxin contamination. Several authors such as Boling et al. [
9], King et al. [
10], Chaing et al. [
11], Cullen et al. [
12] mention resistance to disease (no toxins were measured). Others concentrate on the resistance to toxin, such as Brown et al. [
13] Bolduan et al. [
14], others look at genetic factors being specific in toxin contamination [
15,
16] but do not consider symptoms, and others investigate both disease severity and toxin contamination together [
17,
18,
19]. Menkir et al. [
20] reported on germplasms with resistance to aflatoxin contamination. Looking more closely at the data, we see that the resistance to toxin and disease do not overlap in every case. We do not know whether this is due to some additional metabolic pathways beside resistance. Complete agreement does not seem to be apparent between resistance and toxin accumulation. Since toxins have in most cases undergone strict regulation in human consumption, and restrictions in animal husbandry have also been suggested; the task now at hand is to achieve a low toxin contamination level. It is not clear what the real relationships are, but many independent data confirm a rather good and useful correlation between resistance to infection severity and toxin contamination; see Mesterhazy et al. [
3].
Economic significance. The harvested yield loss is regularly lower than the much larger problems caused by the toxins because farmers may suffer total financial loss when the yield contains toxins above certain regulated limits. Then, the grain can become unsuitable for food and feed. About one third (own estimation AM) of the total crop value was lost in 2014 in Hungary (about 330 million Euros), partly because of lowered prices because of higher toxin contamination, and partly due to losses in animal husbandry and additional costs of toxin binders, medication, etc. This epidemic again underlined the necessity of higher resistance levels in maize production. Nobody has exact data about losses in yield, value of the crop, or health costs in the human population and animal husbandry. However, we would be surprised if it would be globally lower than several billion dollars. Hungary and Serbia have had
Fusarium problems for a long time [
21,
22] and selection programs have started with moderate success. This, combined with the appearance of aflatoxin-producing species, especially
Aspergillus flavus, resulted in aflatoxin contamination above the EU limit in 2007, 2012 and at a lower degree in 2017. This alerted plant breeders and the milk industry in Northern Italy, Serbia, Slovenia, Croatia, Romania and in Hungary [
23,
24,
25,
26]. It seems that aflatoxin will not be a transitional problem we mentioned earlier. Battilani et al. [
27] indicated a very strong predicted increase in aflatoxin contamination for nearly all of France, the whole of the Carpathian Basin and the Balkan area when the average temperature increases by 2 °C. At an increase of 5 °C, nearly all of mainland Europe except Scandinavia will become moderately or heavily contaminated areas. For this reason, this
Aspergillus and aflatoxin problem should be taken seriously.
Breeding aspects of resistance. The most important fact is that at present, knowledge of complete resistance does not exist against any of the toxigenic species; what we have is a partial one at different degrees or none at all. In the past decades, most breeders favored natural infection. They were and are mostly convinced that this is the right way. They think that during the long years of variety breeding and testing, the probability is high enough to select the most resistant plants and hybrids. The variety registration and post registration were proved in the highly epidemic years; however, the disease pressure under natural infection is not enough to find the best ones—maybe it can be suitable to discard the most susceptible ones. We should add that the toxigenic species rapidly change from year to year. After a ‘F. graminearum’ year, we might have a ‘F. verticillioides’ or an ‘A. flavus’ year or any combination of them. As we do not have any proof that the resistance to different toxigenic fungi would be the same, the data speak against it, and natural infection does not allow a solid base for the breeding of more resistant hybrids. Therefore, most of the hybrids belong to the susceptible or very susceptible category. When we see the literature, in most cases, the breeding started only against the most important species. In best cases, two species are involved, but never more. It seems that the breeding and registration system could not adapt to the genetic background of the resistance.
In the cases of variety registration, only natural infection was considered in Hungary and elsewhere [
3]. Severe epidemics were rare, but 2010 showed a nation-wide epidemic. The maize post registration test, sown this year and performed in eight locations (
Figure 1) by the National Variety Office NEBIH, Hungary [
28] with the support of the Association of Cereal Growers, Hungary brought important results. The mean data for infected ears from the total are given as a percentage, and the ear coverage is also rated as a percentage. The correlation is close, indicating that a higher ratio of infected ears correlates well with the infected ear surface. The most resistant hybrid had a 26.3% visual rate and 7.5% coverage (
Figure 1, left panel). The toxin content was not measured, but such infection severity like this normally causes toxin levels to rise far above the toxin limit. When considering the most severe epidemic in Eszterágpuszta, the least infected hybrid had a 26% rate out of the total, and an 11% ear area showed visible infection. The maximum was 85% and 29.4%, respectively (
Figure 1, right panel). The conclusion is clear: very resistant hybrids do not exist in this group. Therefore, the breeding efforts of various breeding firms are by far not enough to give farmers a chance to produce healthy grain when they face an epidemic that might not occur every year. The situation is slowly changing, however, because farmers look for better hybrids in this respect. The reason is simple: they are pressured to keep toxin contamination levels under control. The other important conclusion is that in spite of the lack of the hybrids in the most resistant regions, we have a significant variability in resistance indicating a utilization of these differences. With the withdrawal of the susceptible hybrids from commercial production food safety, there would be a sharp increase in food and feed safety. Then, the breeding efforts could result in further significant improvement in food and feed safety.
For the reasons above, artificial inoculation methods should be introduced and applied at a much higher level than previously used in registration and breeding.
The artificial inoculations are normally conducted with a pure isolate or a mixture of isolates, but a mixture of different species can also be the case [
3]. In wheat, we have learned that the different isolates have different aggressiveness; therefore, gathering data during epidemics at different severity may help to produce more reliable resistance data [
29,
30,
31]. For this reason, we wanted to test our hypothesis (more isolates of the same species are used separately to have more reliable results) in maize. Statistically significant resistance differences to
Fusarium spp. [
3] and
A. flavus already exist [
13,
32,
33,
34,
35], and all fungal species should be treated together (
Fusarium spp.,
A. flavus) as they will remain important players also in the future.
A real answer to the question of the resistance level and resistance specificity or non-specificity cannot be expected from steadily changing natural infection and steadily changing fungal populations. Even though we have data about resistance correlations of toxigenic species in maize, from the toxigenic species, a maximum of two were involved, but not more. In several cases, a correlation was found between
F. graminearum and
F. verticillioides, and in other cases between
A. flavus and
F. verticillioides (see details in the review by Mesterhazy et al. [
3], but not in every case.
What this means for us as researchers is that we should provide reliable data for the farmers and breeders on resistance behavior of the hybrids to all fungi. The data we have now show that relationships between resistances to different toxigenic species might be present, but not necessarily so. Therefore, we should test the hybrids separately against the most important toxigenic fungi to observe their resistance, toxin contamination and estimate their safety risks to individual pathogens and consider them together.
In this paper, our objectives were as follows: (1) To test the resistance of Hungarian and Serbian maize hybrids against toxigenic fungi such as F. graminearum, F. culmorum, F. verticillioides and A. flavus. (2) To test toxin contamination following artificial inoculation. (3) To test the use of more isolates separately. (4) To suggest a screening methodology that allows the preference of multi toxigenic fungal resistance in maize hybrids and thereby significantly improve food and feed safety.