Peanut rust, a foliar disease caused by Puccinia arachidis Speg., is a widespread problem in countries with warm, tropical climates. The disease reduces peanut yield and quality and indirectly increases management costs (Subrahmanyam et al., 1997). Rust often co-occurs with early leaf spot (caused by Passalora arachidicola (Hori) U. Braun [syn. Cercospora arachidicola (S. Hori)]), and late leaf spot (caused by Nothopassalora personata (Berk. & M.A. Curtis) U. Braun, C. Nakash, Videira & Crous [syn. Cercosporidium personatum (Berk. & M. A. Curtis) Deighton]), which have been reported to cause up to 80% yield losses in India in the absence of fungicide control (Subrahmanyam et al., 1984). Rust is primarily a pathogen that afflicts tropical countries and has only sporadic outbreaks in the U.S. However, global climate change may increase the impact of this disease through increasing the frequency of tropical storms that carry rust inoculum into the U.S. and by expanding the range in which rust can overwinter (Power, 2014). Yield loss estimates due to rust alone are unavailable for the U.S. due to it being a localized issue in warm regions and its co-occurrence with leaf spots; however, losses due to damage caused by rust and early and late leaf spot and increased fungicide costs were approximately $32 million in Georgia in 2011 (Williams-Woodward, 2013). Crop rotations, eradicating volunteer peanut plants to reduce inoculum source, and allowing one-month fallow periods are cultural practices applied by farmers who do not have access to or cannot afford fungicides in countries such as India, Haiti, and Guyana (Subrahmanyam, 1997). Even for farmers that can use fungicides, there is still a likelihood that rust populations will develop resistance when exposed to frequent fungicide applications (Smith and Littrell, 1980). Therefore, rust resistant, high-yielding peanut cultivars are an important part of an integrated pest management strategy to control rust in tropical countries as well as the U.S.

A major limitation to breeding rust-resistant cultivars is that only moderate levels of resistance have been identified in A. hypogaea germplasm (Power et al., 2019). Fortunately, high resistance to rust has been identified in numerous wild Arachis species, including the readily usable A- and B-genome Arachis species, and resistance can be introgressed into cultivated peanut (Subrahmanyam et al., 1982, 1983; Pande and Rao, 2001; Fávero et al., 2009). Cultivated peanut is an allotetraploid (AABB; 2n=4x=40) species and the majority of wild Arachis species are diploid (2n=2x=20); the most efficient way to introgress genes from wild Arachis species into cultivated peanut is to cross A- and B-genome species to produce allotetraploid interspecific hybrids that are cross-compatible to peanut. Then, the wild Arachis derived allotetraploids are backcrossed to peanut cultivars. Markers linked to rust resistance allow quick introgression as well as pyramiding of multiple QTLs. Rust resistant QTLs have been identified in the A-genome species A. cardenasii GKP 10017 and the B-genome species A. magna K 30097. One major QTL from each species is being used in peanut breeding programs to introgress rust resistance into peanut cultivars (Khedikar et al., 2010; Sujay et al., 2012; Leal-Bertioli et al., 2015). For example, Gowda et al. (2002) released a rust resistant cultivated genotype ‘GPBD 4] (ssp. fastigiata var. vulgaris) with bunch growth habit, in which resistance was derived from A. cardenasii. GPBD 4 had a mean rust score of three on a scale of one to nine, in which one was equivalent to no disease and nine was equivalent to 80 to 100% disease (Gowda et al., 2002). While GPBD 4 still develops rust, it does so at far lower levels than most cultivated germplasm, and one rust resistance QTL derived from A. cardenasii but identified in a population derived from GPBD 4_, has been shown to improve yields by 56 to 96% in rust infected environments (Gowda et al., 2002; Varshney et al., 2014). While progress towards resistant rust cultivars is being made, more major rust resistance QTLs need to be identified for further pyramiding in peanut cultivars to strengthen resistance and more importantly, to increase resistance durability to rust population pressures. This study identified rust resistance in newly synthesized allotetraploids that are cross compatible with cultivated peanut. The long-term goal of this study is to create rust resistant peanut cultivars that can protect yields and decrease the need for fungicides in the U.S. and to increase yields in tropical, developing countries for farmers who cannot afford or access costly fungicides.

Significant genotypic effect on all rust resistance parameters was found for the 2017 and 2020 experiments (Table 2). Between both rust resistance experiments, 30 out of the 39 unique materials did not show rust symptoms and had no lesions at the end of each experiment (Table 3). In the 2017 experiment, the susceptible control, two wild Arachis species, Greg and Ipa, and three allotetraploids, IpaCor1, IpaDur1, and IpaVillo1, developed rust pustules. Greg and IpaDur1 only had one replication that developed rust pustules, so they were excluded from ANOVA analysis evaluating genotype effect on incubation time; however, these replications developed sporulating pustules, meaning they did not show a hypersensitive response. The susceptible cultivated control, Georgia Green, had the shortest incubation period of 14.33 d, while Ipa, IpaCor1, and IpaVillo1 had incubation periods within 15 to 16 d (Table 3). In the 2020 experiment, the susceptible control, one wild Arachis species, Ipa, and two allotetraploids, IpaCor2_2 and IpaDur4, developed rust pustules. The susceptible cultivated control 886 also had the shortest incubation period of 13.95 d, while Ipa, IpaCor2_2, and IpaDur4 had incubation periods between 14 and 15 d (Table 3).

In 2017 rust experiment, the susceptible cultivated control Georgia Green had the highest IA score, but Ipa, with an IA score of 1.42, was not significantly different from Georgia Green (Table 3). IpaCor1, IpaDur1, and IpaVillo1 had IA scores of 0.92 or less, which were not significantly different from the highly resistant genotypes that did not develop rust pustules (Table 3). In the 2020 experiment, Ipa had the highest IA score of 1.94, which was significantly greater than all tested genotypes except the susceptible control, 886, which had an IA score of 0.76.

As expected, the susceptible control had the highest IA, TLA, SLA in the 2017 experiment. However, Ipa and IpaCor1 had similar TLA and SLA scores compared to the susceptible control. In the 2020 experiment, Ipa demonstrated greater susceptibility to rust with a TLA and SLA score twice as much as the susceptible control, although this difference was not significant. IpaCor2_2 showed a TLA and SLA comparable to the susceptible control. While IpaCor2_2 demonstrated susceptibility in the 2020 experiment, IpaCor2^2x had no rust incidence in the 2017 experiment and IpaCor2_1 had no rust incidence in the 2017 experiment.

All the wild peanut-derived allotetraploid genotypes that were not produced with A. ipaënsis as a parent showed promise as sources for rust resistance for peanut breeding programs, since they were highly resistant to rust, with no evidence of disease development in any assay. Therefore, these nine interspecific hybrids, BatCor, BatDur1, BatDur2, BatSten1, GregSten, MagCard, MagDio, MagDur, and ValSten1 made from 13 unique Arachis accessions are recommended to peanut breeding programs for rust resistance introgression. BatSten1 and ValSten1 were deposited in the USDA-ARS National Plant Germplasm System (Fort Collins, CO) and in the USDA Plant Genetic Resources and Conservation Unit (Griffin, GA), and seed is available for research purposes (Bertioli et al., 2021; Chu et al., 2021). The other allotetraploids will be deposited for public use after sufficient disease and insect resistance characterization has been completed.

IpaCor2 performed variably in the two experiments. IpaCor2^2x and IpaCor2_1 had no rust incidence in the 2017 experiment, while IpaCor2_2 demonstrated susceptibility similar to the cultivated peanut control in the 2020 experiment. This variability in resistance of IpaCor2 may be due to a high level of heterogeneity in A. correntina 9530 complementing results Levinson et al. (2020) found, that different A. correntina 9530 plants and allotetraploid lines derived from this wild Arachis species accession exhibited various levels of resistance to fall armyworm in detached leaf assays. Three A. correntina 9530 plants were genotyped with the Affymetrix Axiom_Arachis2 SNP array (Clevenger et al., 2018; Korani et al., 2019) and 1,259 out of 5,342 total markers (23.5%) were found to be polymorphic (Levinson et al., 2020). This was a high level of heterogeneity when compared to A. ipaënsis and A. duranensis, which had 23 (0.4%) and 27 (0.5%) polymorphic markers, respectively (Levinson et al., 2020). The IpaCor2 plants tested in this study were made from crosses between A. ipaënsis and different A. correntina 9530 plants, so genetic difference between these plants may be due to accession heterogeneity or genetic segregation.

Of the wild Arachis species tested, only one species, Ipa, the B-genome progenitor of peanut, was susceptible to rust. In the 2020 experiment, Ipa showed higher susceptibility than the susceptible control, 886, with greater IA, TLA, and SLA scores than 886, although, these differences were not significant. This affirms previous rust bioassays that found most Arachis species, especially those in section Arachis, to be highly resistant to rust (Subrahmanyam et al., 1983; Fávero et al., 2009). Like this study, Pande and Rao (2001), Fávero et al. (2009), and Leal-Bertioli et al. (2015) found A. ipaënsis to be as susceptible or more susceptible than cultivated peanut. Fávero et al. (2009) also found sporulation on six accessions of A. stenosperma, two accessions of A. valida, and one accession of A. magna, which were not tested in this study. So far, high resistance to rust has not been identified in A. hypogaea germplasm (Power et al., 2019; Subrahmanyam et al., 1982; Pande and Rao, 2001; Fávero et al., 2009). The lack of high resistance to rust in cultivated germplasm was contributed in part by having a highly susceptible progenitor as well as the ploidy barrier between allotetraploid peanut and its diploid, highly resistant Arachis relatives.

This study builds upon previous reports by identifying rust resistant allotetraploids that are cross compatible to A. hypogaea and are therefore valuable to breeding programs, instead of focusing solely on diploid, wild Arachis species. A few studies have identified rust resistance QTLs derived from the A-genome species A. cardenasii GKP10017 and the B-genome species A. magna K 30097 (Khedikar et al., 2010; Sujay et al., 2012; Leal-Bertioli et al., 2015). Sujay et al. (2012) identified five rust resistance QTLs in the same linkage group that explained up to 63% to 83% phenotypic variation; these QTLs likely originated from A. cardenasii GKP 10017. The populations used in Sujay et al. (2012) both have A. hypogaea GPBD 4 as a parent, which has A. hypogaea ‘ICGV 86855] as a parent, which in turn originated from a cross between A. hypogaea and A. cardenasii GKP 10017 (Shirasawa et al., 2018). The QTL that explained the most rust phenotypic variation at 83% has been validated and introgressed into three cultivated varieties through marker-assisted backcrossing, improving yield by 56 to 96% in rust infected environments (Varshney et al., 2014). Leal-Bertioli et al. (2015) identified 13 rust resistance QTLs from A. magna K 30097. One of these QTLs contributed to four components of rust resistance, including IA, TLA, SLA, and incubation period, and explained up to 59% of phenotypic variation. Another resistance QTL explaining up to 35% of phenotypic variation was located in the same linkage group, just 25.4 to 33.1 cM away from the major QTL. These QTL identified by Leal-Bertioli et al. (2015) are distinct from those identified by Sujay et al. (2012) and can be pyramided into the same peanut cultivars to yield more effective and more robust resistance. All nine rust resistant allotetraploids identified in this study have unique Arachis parents compared to these two previous studies; therefore, any of these allotetraploids could be used to map new rust resistance QTL for further stacking of resistance QTLs in cultivated peanut. However, the greatest chance of identifying a major QTL would come from a mapping population using an allotetraploid that has a unique A- and B-genome species. For example, BatCor, BatDur1, BatDur2, BatSten1, GregSten, MagDio, and ValSten1 would be good candidates for future rust resistance QTL mapping.

One perceived limitation of this study was that both experiments were confined to in vitro bioassays using excised peanut leaves rather than having these in vitro experiments in addition to field evaluations. However, past rust field evaluations have been complicated by other fungal pathogens such as Colletotrichum spp., Leptosphaerulina crassiasca Sechet., early leaf spot, Myrothecium roridum Tode ex. Fr., and late leaf spot (Subrahmanyam et al., 1983; Sujay et al., 2012). Furthermore, while peanut rust is common in countries with warm, tropical climates, it only threatens U.S. peanut production every few years when brought by tropical storms (Bromfield, 1971; Subrahmanyam et al., 1985). When achieved, rust pressure is very inconsistent and fragmented within the same field, and results identifying resistant germplasm could easily result from avoidance of the pathogen instead of actual resistance. Lastly, the morphological differences, i.e., canopy structure, between wild Arachis species, allotetraploids, and cultivated peanut can add variations to pathogen pressure in the field. Despite these complications, Subrahmanyam et al. (1983) tested wild Arachis species and cultivated peanut germplasm in the field and in vitro and while complications in the field study from three other fungal pathogens were encountered, the reactions to rust were the same in the field and the in vitro experiments. Six of the wild Arachis accessions tested by Subrahmanyam et al. (1983) were also tested in this study and all were found to be highly resistant presenting no rust symptoms in both studies. Therefore, in vitro bioassays like this study have produced consistent results with previous studies, have similar results to field rust experiments, and have been used to map rust resistance QTL, indicating in vitro bioassays are sufficient for identifying rust resistant germplasm (Subrahmanyam et al., 1983; Khedikar et al., 2010; Sujay et al., 2012; Leal-Bertioli et al., 2015).

This study built upon previous reports by testing rust resistance in numerous synthetic allotetraploids and diploid wild Arachis species. Nine allotetraploids demonstrated high levels of resistance to rust, making them a better source of rust resistance than cultivated peanut germplasm, which has only been found to have moderate levels of resistance. These allotetraploids are cross-compatible with peanut cultivars and thus, available as a genetic resource for peanut breeders. A few of these unique allotetraploids will be used to map rust resistance QTLs so that they can be introgressed into peanut cultivars along with the previously identified rust resistance QTLs from A. cardenasii GKP 10017 and A. magna K 30097. The long-term goal of this study is to create rust resistant peanut cultivars that can protect yields in the U.S. and to increase yields in tropical, developing countries for farmers that cannot afford, or do not have access, to costly fungicides.