Introduction

The Columbia root-knot nematode, Meloidogyne chitwoodi Golden et al., is a plant parasitic nematode that can reproduce on roots and other underground tissue of a range of economically important crop plants, including potatoes, wheat, corn and alfalfa (Mojtahedi et al. 1988). In the United States, M. chitwoodi is most abundant in the Columbia Basin potato growing region of Oregon and Washington, but is also found in California, Idaho, Colorado, New Mexico, and Texas (Powers et al. 2005), as well as Utah (Griffin and Jensen 1997) and Nevada (Nyczepir et al. 1982). Outside of the United States, M. chitwoodi is found in Mexico, Argentina, Belgium, Germany, the Netherlands, Portugal, and South Africa (Powers et al. 2005). The species most closely related to M. chitwoodi is M. fallax Karssen (false Columbia root-knot nematode), which is found in the Netherlands, Australia, and New Zealand, but is not known to occur in potato production regions of the United States (Powers et al. 2005).

Meloidogyne chitwoodi emerges from eggs as second-stage juveniles (J2) after undergoing one molt within the egg (Mitkowski and Abawi 2003). Juvenile nematodes enter host roots and tubers, and establish giant cells, which support the developing nematodes (Mitkowski and Abawi 2003). In potato tubers, pimple-like bumps form at each infection site, dramatically reducing fresh-market appeal of tubers. Adult females lay eggs in the flesh of the tuber where pinhead-sized brown spots later develop. This is coupled with increased sugar concentrations in the tissue surrounding each infection site, resulting in browning of fried products thus making these tubers unsuitable for French fry and chip processing. The European Plant Protection Organization has listed M. chitwoodi as an A2 pest, recommending that infected plant materials be quarantined by member countries.

The predominant control methods for M. chitwoodi are chemical fumigants and non-fumigant nematicides. However, these chemicals have substantial negative aspects, including high cost and health and environmental hazards. Crop rotations offer some control, but their efficacy is reduced by the nematode’s wide host range and long persistence in soils. Considering the limited success of crop rotation, genetic resistance to M. chitwoodi in elite potato cultivars would be a valuable control tool for growers.

Two races of M. chitwoodi exist in the United States, races 1 and 2, and each infects a unique set of host plants (Santo and Pinkerton 1985). Both of these races can reproduce on a wide range of crops commonly grown in the Columbia Basin, including potatoes, corn, and wheat (Mojtahedi et al. 1988). A key difference between these races is that race 2 can reproduce on ‘Thor’ alfalfa, while race 1 cannot (Mojtahedi et al. 1994). Of these, race 1 was identified first and is more prevalent in the Columbia Basin, while race 2 is typically found when potatoes are grown in rotation with alfalfa (Mojtahedi et al. 1994).

Host genetic resistance to root-knot nematodes is thought to take advantage of a hypersensitive response (Castagnone-Sereno 2002; Williamson and Kumar 2006). The Mi gene in tomato confers resistance to M. javanica, M. incognita, and M. arenaria; it has been cloned, and found to be a member of the nucleotide binding leucine-rich repeat family of plant resistance genes (Milligan et al. 1998; Vos et al. 1998). In potato, two genes (RMc1(blb) and RMctuber(blb)) that confer resistance to M. chitwoodi are being employed in cultivar development efforts. Both genes were introgressed from the Solanum bulbocastanum clone SB22 (PI 275187) (Brown et al. 2009). RMc1(blb) confers root resistance to most isolates of race 1 of M. chitwoodi, with the exception of WAMCRoza, an isolate that was identified in experimental plots that had been planted repeatedly with clones carrying RMc1(blb) (Mojtahedi et al. 2007). In the clone CBP-233, a somatic hybrid between SB22 and the S. tuberosum clone R4 (PI 203900), necrotic tissue was observed to form in roots around nematode infection sites, suggesting that resistance from RMc1(blb) is expressed as a hypersensitive response. RMctuber(blb) confers tuber resistance to both race 1 and race 2 of M. chitwoodi. Although root and tuber resistance from SB22 have been successfully introgressed into elite potato germplasm, no cultivars with this resistance have been released.

In addition to root and tuber resistance from SB22, root resistance from S. hougasii (RMc1(hou)) and S. fendleri (RMc1(fen)) was partially introgressed into elite potato germplasm (Brown et al. 2006, Brown et al., 2014). Brown et al. (2014) suggested that both of these resistances were identical to RMc1(blb) based on genetic marker data and the close taxonomic relationship of S. bulbocastanum, S. hougasii, and S. fendleri. To the best of our knowledge, all clones with resistance introgressed from S. hougasii and S. fendleri have since been lost (C. Brown,Personal communication, January 30, 2018).

Resistance to cold temperature root-knot nematode species (M. chitwoodi, M. fallax, and M. hapla) is correlated with that of other cold-temperature species, but not with warm-temperature species (M. arenaria M. incognita, and M. javanica). The responses of wild potato clones to M. chitwoodi and M. fallax were particularly similar for clones derived from S. fendleri and S. hougasii (Janssen et al. 1997). This raises the possibility that RMc1(hou) and RMc1(fen) confer resistance to race 1 of M. chitwoodi and M. fallax, but not to race 2 of M. chitwoodi. Another study quickly selected for isolates that were virulent on S. fendleri clones and found that some of the virulent M. chitwoodi isolates were also virulent on other resistant clones from S. fendleri, S. bulbocastanum, S. hougasii, and S. stoloniferum Schltdl. (Janssen et al. 1998).

While resistance introgressed from SB22 may soon be present in clones that are suitable for release as cultivars in the Pacific Northwest, there is an urgent need to identify additional sources of resistance, especially to race 1 isolate WAMCRoza. The objective of this study was to identify new sources of resistance to M. chitwoodi for use in breeding, with an emphasis on identifying resistance to the M. chitwoodi isolate WAMCRoza.

Methods

Plant Material

Forty accessions representing nine wild potato species were obtained from NRSP-6 Potato Genebank and were evaluated for their response to M. chitwoodi (Table 1). When possible, these accessions were chosen to include species and germplasm in which resistance had previously been detected. Three checks were used: ‘Rutgers’ tomato, which is susceptible to all races of M. chitwoodi (Brown et al. 2014), ‘Vernema’ alfalfa (a differential check), which is resistant to race 1 but susceptible to race 2 (H. Mojtahedi, Personal Communication, January 30, 2018), and ‘Red Core Chantenay’ carrot (another differential check), which is susceptible to race 1 but resistant to race 2 (Mojtahedi et al. 1988). ‘Rutgers’ tomato was used as the susceptible check rather than long-day adapted cultivated potatoes as ‘Rutgers’ tomatoes are an excellent host to the nematode. While the cultivated potatoes are a good host, their tendency to form tubers under the long photoperiod present in greenhouses results in difficulties in accurately quantifying nematode reproduction from roots.

Table 1 Solanum accessions screened for resistance to Meloidogyne chitwoodi
Full size table

Isolates Used

For the initial screening, M. chitwoodi WAMCRoza (Mojtahedi et al., 2007), an isolate of race 1, distinguished by its ability to reproduce on roots of potato plants that have resistance to race 1 conferred by RMc1(blb), was used. In addition to WAMCRoza, WAMC1 (Pinkerton et al., 1987), an isolate representative of race 1, and WAMC27 (Santo and Pinkerton, 1985), an isolate representative of race 2, were used for replicate screening. All nematode isolates were maintained on ‘Rutgers’ tomato at USDA-ARS, Prosser as previously described (Brown et al., 1999). Eggs for each nematode isolate were extracted using the same methods used to extract eggs in the initial screening (see below). Egg concentrations were quantified using a nematode counting slide (Chalex, LLC, Park City, Utah, USA), then adjusted to 1,000 eggs/ml.

Initial Screening

In the initial screening, we attempted to test 10 seedlings from each of the 40 accessions, although for some accessions low germination and plant mortality resulted in fewer than 10 seedlings. In addition, 10 plants of ‘Rutgers’ tomato and four plants of ‘Vernema’ alfalfa were used as the susceptible and resistant checks, respectively. Approximately 20 seeds from each accession were soaked in a 0.1% gibberellic acid solution for 24 hours, then placed on a damp paper towel in a covered petri dish and kept moist for one week. The germinated seeds were transferred to 2.54 cm pots, filled with a sterilized mixture of 75% sand and 25% soil and fertilized with 2.0 g Osmocote 14-14-14 Flower and Vegetable Smart-Release Plant Food (The Scotts Company, Marysville, OH, USA) per liter of sand-soil mixture. Twenty-eight days after transplanting, the seedlings were transferred to 10 cm pots, filled with the same mixture of sand, soil, and fertilizer. During the second transplanting, the roots of each plant were inoculated using a pipette with 5,000 eggs of WAMCRoza suspended in water (5ml @ 1,000 eggs/ml). Plants were then grown in a greenhouse under 16 hours of artificial lighting per day; temperatures were kept at approximately 24 °C for 56 days, to allow the nematodes sufficient time to complete two generations in the plant roots. At the end of the 56-day period, potting soil was rinsed from the roots of each plant, and the roots were shaken at 90 rpm in a solution of 0.6% sodium hypochlorite for 4 min to release the eggs. The resulting solution was strained with a 0.841 mm sieve, over a 0.025 mm sieve to remove debris. The contents in the 0.025 mm sieve were rinsed into a bottle and quantified for egg concentration using a nematode counting slide (Chalex, LLC, Park City, Utah, USA). The initial screening was conducted in five batches to allow the timely harvest and quantification of eggs from each plant at the end of the trial. After the initial evaluation, further screenings were conducted with low levels of nematode reproduction in susceptible checks and were treated as extensions to the initial screening.

Clone Maintenance

At the end of the initial screening for resistance to WACRoza, we clonally propagated each selected seedling via shoot cuttings: 7 to 10 cm long shoots were collected and dipped in half the label recommended concentration of Dip ‘N’ Grow Rooting Concentrate (Dip ‘N’ Grow, Clackamas, OR, USA) to stimulate root formation. The treated shoots were planted in 10 cm pots for successful propagation. Stem segments from selected clonally propagated seedlings were collected and surface sterilized by soaking them in 70% ethanol for 1 minute, then in 0.6% sodium hypochlorite supplemented with three drops of TWEEN 20 (Sigma-Aldrich, St. Louis, MO, USA) per 100 ml sodium hypochlorite solution for 10 minutes, then soaked in sterile distilled water for 5 minutes. Surface sterilized stem segments were transferred to tissue culture and grown on MS-30 media (Murashige and Skoog 1962) for maintenance.

Replicated Evaluation

For the replicated evaluation, each clone identified as resistant in the initial screening was inoculated separately with WAMCRoza, WAMC1, and WAMC27. The transplantation, inoculation, and quantification of resistance for these isolates was carried out on subsequent days. Clones were replicated five times for each nematode isolate. Testing procedures were similar to those of the initial screening, with the following differences: plantlets from tissue cultured cuttings were placed in Greenhouse Mix #3 (Teufel Products Co., Hillsboro, OR, USA) and allowed to grow for 40-60 days before being transplanted into 2 L clay pots filled with a mixture of 84% sand, 10% silt and 6% clay, and fertilized with 2g Tree ‘N’ Vine 12-8-16 Agropell fertilizer (J.R. Simplot Company, Boise, ID, USA) per liter of sand-silt-clay and fertilizer mixture. Plants were inoculated five days after transplanting by pipetting 5,000 nematode eggs per plant into three holes made by inserting a pencil approximately 2.5 centimeters into the soil near the base of each plant.

Characterization of Resistant Accessions

To better characterize the accessions with resistance identified in our study, we planted additional seeds and evaluated 30 seedlings from each of the eight accessions for which we found at least one seedling with resistance to WAMCRoza. The seedlings were inoculated with WAMCRoza, and eggs were extracted and counted using the methods described for the initial screening.

Data Analysis

All data in this study were transformed using the following equation:

$$ \mathrm{value}={\log}_{10}\left[\left(\mathrm{no}.\mathrm{eggs}/50\right)+1\right] $$

Where “no. eggs” is the total number of nematode eggs extracted from the plant. Geometric mean reproduction values were made by back-transforming the average of the transformed reproduction values. This value was used to calculate a reproduction factor (Rf), defined as the number of eggs extracted divided by the number of eggs initially used to inoculate the plant. To match previous evaluations for M. chitwoodi resistance, clones were classified as hosts (Rf > 1.0, corresponding to susceptibility), poor hosts (1.0 > Rf > 0.1, corresponding to moderate resistance), or non-hosts (Rf < 0.1, corresponding to resistance). For the replicated evaluation, an analysis of variance (ANOVA) determined whether clones responded differently to the different nematode races. Tukey tests determined which clones were significantly different from others at α=0.05 using the R package “agricolae” (de Mendiburu 2017). For the clone PI545815sph-9mc, only one plant survived the replicated evaluation for isolate WAMC1, so it was excluded from statistical analysis for that isolate. All statistical tests were conducted using R version 3.2.3 (R Core Team 2005).

Relationship of Resistance to Geographic Origin

To investigate a relationship between M. chitwoodi resistance and geographic origin, accessions evaluated for M. chitwoodi resistance in this or earlier studies were plotted on maps of North, Central, and South America.

Results

Initial Screening

Fifteen clones from six accessions from S. hougasii and one clone each from S. bulbocastanum, and S. stenophyllidium were selected for replicated evaluation after initial screenings with M. chitwoodi race 1 WAMCRoza isolate. No resistance was found for any clones from S. andreanum, S. boliviense, S. brevicaule, S. guerreroese, S. stoloniferum and S. iopetalum used in this study. Accessions in the initial screening varied widely in their levels of nematode reproduction, with some accessions having close to no M. chitwoodi reproduction (Rf=0), and others approaching the susceptibility of ‘Rutgers’ tomato (Rf>10) (Supplementary file 1).

Replicated Evaluation

The ANOVA of the complete set of seventeen clones excluding checks in the replicated evaluation showed a significant interaction between clone and nematode isolate (p<0.001, Table 2), indicating that clones responded differently to different nematode isolates. Within each of the three nematode isolates, significant differences were found among the clones (Table 3), as expected. All selected clones displayed significantly greater resistance than ‘Rutgers’ tomato to all isolates of M. chitwoodi. However, in S. hougasii, only eight clones from three accessions were significantly more resistant to WAMCRoza than the ‘Red Core Chantenay’ carrot (the poor host check for race 1), and 11 clones from six accessions were significantly more resistant to MC1 than ‘Red Core Chantenay’ carrot (Table 3). For WAMC27, three clones from two accessions from S. hougasii and one clone from S. bulbocastanum (PIPI255518blb-4mc) have reproduction factor less than 0.1 and are considered more resistant than the other clones tested in the replicated trials. Reproduction factor in the replicated trials varied from 0.0 to 30, with mean highest Rf observed in ‘Rutgers’ tomato (Table 3, Supplementary file 2).

Table 2 ANOVA table showing the effect of Solanum clones, isolates, and clone × isolate interaction on Meloidogyne chitwoodi reproduction, using transformed reproduction values
Full size table
Table 3 Geometric means of reproduction factors (Rf = number of eggs extracted/initial number of eggs) and HSD tests for selected wild Solanum clones and three checks, against three M. chitwoodi isolates
Full size table

Characterization of Resistant Accessions

When additional seedlings from each resistant wild potato accession were evaluated, most of the seedlings tested were either poor hosts or non-hosts for WAMCRoza (Fig. 1). Solanum stenophyllidium accession PI 545815 was the exception In the case of PI 545815, the reproduction factors ranged from 0 to 3.6, with the majority of seedlings exhibiting intermediate levels of nematode reproduction. Additionally, one seedling from the S. hougasii accession PI 558402 had a reproduction factor of 1.1. The detection of additional resistant seedlings suggests resistance genes are present at a high frequency in these accessions.

Fig. 1
figure 1

Boxplot showing range of WAMCRoza reproduction factors in the eight wild potato accessions having at least one clone resistant to M. chitwoodi

Full size image

Geographical Mapping

For the accessions with recorded collection sites, the locations are presented on geographical maps of North and Central America, and of South America (Fig. 2, Supplementary File 3). Resistant and susceptible accessions are shown with different symbols, and accessions evaluated in previous M. chitwoodi resistance studies (Brown et al. 1989; Brown et al. 1991; Janssen et al. 1996; Brown and Mojtahedi 2004) are included. Although at least 46 accessions from South America have been tested, none have strong resistance to M. chitwoodi (Fig. 2b). Of the accessions from Mexico and the southwestern United States, resistance clustered in and around the Mexican states of Jalisco and Michoacán, which in part reflects the great number of S. hougasii accessions collected in this region. However, the same observation holds true in S. bulbocastanum, where accessions from this western region were more likely be resistant than accessions collected in the eastern part of the species’ range. The two resistant S. stenophyllidium accessions that have been identified originated an area just north of this region (Fig. 2a).

Fig. 2
figure 2

Distribution of collection sites of wild potato accessions. a Collection sites in Mexico, Guatemala, and the southern United States and b collection sites in Argentina, Bolivia, Columbia and Peru. Open circles indicate accessions with no detected resistance to Meloidogyne chitwoodi, stars indicate resistant accessions detected in this study, and diamonds indicate resistant accessions detected in previous studies

Full size image

Discussion

In this study, we identified strong resistance to all three isolates of M. chitwoodi. Fifteen of the 17 resistant clones were from six accessions from S. hougasii, while the other two clones were from S. bulbocastanum and S. stenophyllidium, respectively. The abundance of resistant clones from S. hougasii was evident at the initial screening, as S. hougasii accessions almost always had a lower mean reproduction factor than accessions from other species. The exception to this was the S. hougasii accession PI 161727, which was consistently susceptible to WAMCRoza. In addition to differences in genetic resistance between species, Solanum hougasii tended be more easily propagated from true potato seed and shoot cuttings than other species, which resulted in more resistant clones being maintained from each resistant accession.

The S. hougasii clones PI239424hou-2mc, PI239424hou-6mc, PI283107hou-5mc, and PI283107hou-9mc were non-hosts for each of the three isolates tested. While not directly tested, it appeared that S. hougasii clones from the accessions PI 161726, PI 558402, and PI 558422 were substantially more resistant to WAMC1 than WAMCRoza. This is consistent with the hypothesis that the resistance gene RMc1(hou) (originating from the S. hougasii accession PI 161726) is similar or identical to RMc1(blb) (Brown et al. 2014). However, each of the S. hougasii clones tested in the replicated evaluation had moderate to strong resistance to WAMCRoza, indicating that additional resistance genes independent of RMc1(blb) may be present in this species.

The S. bulbocastanum clone PI 255518blb-4mc was a poor host for WAMCRoza, and non-host for WAMC1 and WAMC27. The strong resistance to WAMC1 and WAMC27 of PI 255518blb-4mc was similar to that seen in S. bulbocastanum clone SB22. However, moderate resistance of PI255518blb-4mc to WAMCRoza suggests that it may have additional resistance not present in SB22.

The S. stenophyllidium clone PI545815sph-9mc was a non-host for WAMCRoza and a poor host for WAMC27. Like PI255518blb-4mc, the resistance of PI545815sph-9mc to WAMCRoza suggests that it holds resistance to race 1 independent of RMc1(blb). While S. stenophyllidium and S. bulbocastanum are in the same nuclear clade (Spooner et al. 2014), morphological differences between the two species suggest that they are not extremely similar and indicating that the resistance found in PI545815sph-9mc and PI255518blb-4mc may not share a common origin.

One recurring concern in breeding for CRKN resistance is the ability of M. chitwoodi to overcome host resistance in the field, as has been observed in the lab, through the selection of nematodes able to overcome resistance from S. hougasii (Janssen et al. 1998; Mojtahedi et al. 2007; Castagnone-Sereno 2002). One strategy to develop durable resistance is to focus efforts on genes that confer resistance to a wide range of isolates, as these genes likely target attributes that are more central to the nematode's pathogenicity and would likely be more difficult for the nematode to overcome. However, it will be necessary to cross susceptible parents and the selected S. hougasii clones that are resistant to multiple M. chitwoodi isolates to determine whether they carry any single genes that confer resistance to multiple isolates.

The main challenge with nematode screening is quantification of resistance relative to susceptible checks. In this study, we used ‘Rutgers’ tomato as the susceptible check to quantify the success of nematode screening. We did not use any commercially grown potato as susceptible check because most wild potatoes do not produce tubers under long day conditions; by contrast commercial potatoes do form tubers under such conditions.This would jeopardize quantification of nematode reproduction and thereby identification of resistant plants and the success of screening. While ‘Rutgers’ tomato was used to quantify the success of the screening, alfalfa and carrot were used as differential checks to confirm the races used in the study. In the initial inoculations, we selected clones that exhibited no- to low-reproduction rates relative to the susceptible check. Across all of our evaluations, nematode reproduction values varied widely, among plants of each of the checks and among plants of the same clone. We commonly observed a five-fold difference in reproduction among clones from a single accession. Difference in nematode reproduction values in the susceptible checks within each screening highlights the importance of replicated trials and appropriate statistical methodology for data analysis.

Based on the geographical mapping (Fig. 2), it appears that resistance to M. chitwoodi is clustered in and around the Mexican states of Jalisco and Michoacán. Therefore, we propose that accessions in this area are more likely to have resistance to M. chitwoodi, possibly because M. chitwoodi or a similar nematode species has been present for a long time and resistance evolved in the wild potato species. This hypothesis is supported by the observation that multiple types of resistance to M. chitwoodi are found in this area (including RMctuber(blb), RMc1(blb), root resistance to race 2 that was identified in this and earlier studies, and root resistance to WAMCRoza that was identified in this study), while to date no accessions with resistance to M. chitwoodi have been found in the entirety of South America. In addition, it is consistent with a report that resistance to Meloidogyne species in wild potato accessions is associated with geographic and climatic variables, including precipitation and temperature (Spooner et al. 2009). Thus, we recommend that future screening for resistance to M. chitwoodi focus on germplasm from this region.

Conclusion

We identified Solanum spp. clones from eight accessions with high levels of resistance to three key isolates of M. chitwoodi. Resistant accessions included six accessions from S. hougasii, one accession from S. bulbocastanum, and one accession from S. stenophyllidium. Of the 17 clones with resistance, PI239424hou-2mc, PI239424hou-6mc, PI283107hou-5mc, and PI283107hou-9mc were the only clones that were non-hosts for each of the three nematode isolates tested. We are in the process of introgressing these resistances into elite potato germplasm. For S. hougasii, it should be possible to cross directly to elite tetraploid potatoes (Brown et al. 1991), while resistance from S. bulbocastanum can be introgressed into elite potato germplasm through protoplast fusion (Austin et al. 1993). For both of these species, continued backcrossing with tetraploid cultivated potatoes after the initial hybridization will eventually result in a tetraploid potato (Brown et al. 2009, Haynes and Qu 2016). To the best of our knowledge, no efforts have been made to hybridize clones from S. stenophyllidium with cultivated potatoes, but the steps required for introgression would likely be similar to those for S. bulbocastanum. Segregation ratios after sexual recombination in the interspecific hybrids should provide information on the number and locations of the resistance genes in each selected clone. The information on the magnitude and breadth of resistance in these clones will aid in planning future efforts to transfer resistance genes to cultivated potato.