Dynamics of Meloidogyne incognita Virulence to Resistance Genes Rk and Rk² in Cowpea

Abstract

The virulence index of three Meloidogyne incognita field isolates to the resistance gene Rk in cowpea was 0%, 75%, and 120%, with the index measured as reproduction on resistant plants as a percentage of the reproduction on susceptible plants. Continuous culture of the 75% virulent isolate on susceptible tomato for more than 5 years (about 25 generations) resulted in virulence decline to about 4%. The rate of the decline in virulence was described by exponential decay, indicating the progressive loss of virulence on a susceptible host. The 120% virulent isolate declined to 90% virulence during five generations on susceptible cowpea. Following virulence decline, the two isolates were compared over 5 years in inoculated field microplots both separately and as a mixture on susceptible, gene Rk, and gene Rk² cowpea plants. At infestation of the plots, the two isolates were 1.2% and 92.0% virulent, respectively, to gene Rk and 0.2% and 8.1% virulent, respectively, to gene Rk². Virulence to gene Rk in the two isolates and in mixture increased under 5 years of continuous Rk cowpea plants to 129% to 172% and under Rk² cowpea plants to 113% to 139 % by year 5. Virulence to gene Rk² increased during continuous cropping with Rk cowpea plants to 42% to 47% and with Rk² cowpea plants to 22% to 48% by year 5. Selection of Rk²-virulence was slower in the isolate with low itt-virulence. The virulence to both genes Rk and Rk² in the mixed population was not different from that in the highly virulent isolate by year 5 of all cropping combinations. Selection of Rk²-virulence on plants with Rk, and vice versa, indicated at least partial overlap of gene specificity between Rk and Rk² with respect to selection of nematode virulence. This observation should be considered when resistance is used in cowpea rotations.

The root-knot nematode, Meloidogyne incognita, is a serious pathogen of cowpea (Vigna unguiculata) and is part of a disease complex with Fusarium wilt (Roberts et al., 1995). A dominant single gene (Rk) for resistance to M. incognita and other root-knot nematodes (Fery and Dukes, 1980) has been bred into several cowpea varieties that are grown as grain, fresh pod, and cover crops (Roberts et al., 1995; Ehlers et al., 2002; Hall et al., 2003). Although gene Rk is effective in reducing reproduction in most M. incognita populations, several cowpea fields in central California have M. incognita populations that reproduce on, and cause yield loss in, cowpea blackeye bean cultivars with gene Rk (Roberts and Matthews, 1995; Roberts et al., 1995). The cowpea fields with the virulent isolates have a history of growing resistant cowpea cultivars (Roberts et al., 1995, 1997), so it is likely that the virulent isolates resulted from selection on resistant cowpea plants. Field isolates virulent to Rk prompted a search for new cowpea resistance to M. incognita and M. javanica, and a stronger, broader-based resistance was discovered, based on a major gene, Rk² (Roberts et al., 1996). The Rk² gene confers a higher level of resistance to Rk-avirulent and Rk-virulent M. incognita isolates and also to M. javanica compared to gene Rk (Roberts et al., 1997). The Rk² gene was determined to be either allelic to gene Rk or a tightly linked locus within 0.l7cM of Rk (Roberts et al., 1996), and it is currently being bred into new cowpea cultivars for California and elsewhere.

Greenhouse stock cultures of the Rk-virulent M. incognita isolates maintained on susceptible tomato showed a decline in virulence to gene Rk in our standard tests. In addition, the original source of gene Rk² appeared to be slightly less resistant to Rk-virulent isolates than to Rk-avirulent isolates. Therefore, we assessed both the in-field stability of the Rk-virulence in an M. incognita isolate when cultured in the absence of gene Rk and whether cross-selection for virulence to the two genes would occur under simulated field conditions. In work related to the current study, we have made in-depth analyses of these Rk-(a)virulent M. incognita isolates, using isofemale lineages from the isolates to determine Rk-virulence selection and stability (Petrillo and Roberts, 2005a) and relative fitness components (Petrillo and Roberts, 2005b) under controlled greenhouse culture conditions.

The objectives of the research reported here were to determine the stability of field-detected virulence during long-term greenhouse culture on susceptible tomato and to determine the potential for virulence selection and cross-selection in response to two cowpea resistance genes, Rk and Rk², under field-microplot conditions. A preliminary report on the virulence of the M. incognita greenhouse cultures is available (Roberts and Matthews, 1995).

Materials and Methods

Plant material: Tomato (Lycopersicon esculentum ‘Tropic’) plants susceptible to M. incognita were used in long-term greenhouse culture experiments. Cowpea genotypes CB3, CB5, CB46, UCR430, H8-9, and 8685 were used to test for virulence and in the field plot experiments. CB5 and CB46 carry the single dominant gene Rk that confers resistance to avirulent M. incognita populations (Fery and Dukes, 1980) and here are termed Rk-plants. UCR430 carries the single dominant gene Rk² (Roberts et al., 1996) that confers resistance to both Rk-virulent and Rk-virulent M. incognita populations and here is termed the Rk²-implant. CB3, H8-9, and 8685 are not resistant to M. incognita and are termed susceptible plants.

Nematode isolates: An isolate of M. incognita race 3 was collected from a cotton field near Tipton, Tulare Co., California, and cultured for about 10 yr in the greenhouse on susceptible tomato Tropic. Two M. incognita race 1 isolates were collected 6 yr apart from a field site near Denair, Stanislaus Co., California, a location known to be infested with Rk-virulent M. incognita (Roberts et al., 1995) and where Rk-cowpea cultivars had been planted frequently. The two isolates were characterized as Rk-avirulent or Rk-virulent, respectively. The Rk-avirulent isolate was initially virulent when isolated from the field in 1989, with a virulence index on resistant cowpea of about 75%. As reported in Experiment 1, before the field plot study was initiated, the Rk-avirulent isolate declined to about 1 % virulence during 6 yr of continual culturing on susceptible tomato, representing ≥ 25 generations. For the purposes of comparison (Experiment 2), a sub-culture of this isolate was cultured on resistant CB46 cowpea for 1 yr. The Rk-virulent isolate, collected in 1995, declined in virulence index from 120% to 92% during 9 mon (five generations) of culture on susceptible cowpea prior to this research. The Rk-avirulent (1% virulence) and Rk-virulent (92% virulence) isolates are referred to as the “Avirulent” and “Virulent” isolates, respectively, as used in the experiments described here.

Greenhouse culture experiment 1: The original isolate of M. incognita race 1 collected from the cowpea field site near Denair, California, was assayed for virulence to gene Rk during continuous culturing on tomato Tropic plants. The initial inoculum was collected by extracting eggs from cowpea roots collected from the plant row at several locations within the field during late season. Eggs were recovered from nematode-infected roots by maceration of the total root system in 0.5% NaOCl solution (Hussey and Barker, 1973). Tomatoes were grown in 1-liter fiber pots containing steam-sterilized sand and were inoculated 3 wk after transplanting as young seedlings with a suspension of eggs and J2 in water to four holes around the root system. Tomato plants were provided slow-release fertilizer as needed, watered daily, and maintained at an ambient temperature of 22°C to 28°C. A minimum of 10 tomato plants were maintained as cultures, and eggs were extracted from roots for use as inoculum, as described, when plants were 20- to 25-wk-old. Ten new tomato plants were inoculated to maintain the stock culture, while the remaining inoculum was used for a virulence assay. Assays were also performed on inoculum from a subset of tomato culture plants at additional times during the 6-mon re-culturing intervals. Virulence to gene Rk was assayed on 22 occasions on CB46 and 17 occasions on CB5, during 2,061 d of culture on tomato.

All virulence assays were performed using a modified growth-pouch technique (Ehlers et al., 2000). Pouches consisted of a 12.5- by 15.0-cm plastic pouch containing a paper insert, folded into a trough near the top. Cowpea seeds were placed with the hymen facing down directly into the trough for germination and watered with 13 ml of tap water; the pouches were placed in a manila folder (two pouches per folder) held in an upright vertical position. Pouches were transferred to a controlled-environment growth chamber maintained at a constant temperature of 26.7°C ± 0.6°C and 16 hr light/d.

Reproduction bioassays for virulence were performed using eggs recovered from infected roots by macerating the root system in 0.5% NaOCl solution (Hussey and Barker, 1973) and hatched in a modified Baermann plate (Barker and Niblack, 1990) at 26°C ± 1.0°C. Eggs were allowed to hatch for 3 to 5 d prior to inoculation. Root systems were inoculated with a mean of 1,210 motile J2 12 to 14 d post-plant. Following inoculation, pouches were watered once or twice per day and fertilized with Hoagland's solution (Hoagland and Arnon, 1950) every 4 d. A complete randomized design was used with 10 replicates each of CB3 (susceptible), CB5, and CB46.

To determine reproduction, each pouch was inundated with 75 mg erioglaucine/L⁻¹ (Sigma Chemical Co., St. Louis, MO) solution to stain eggs masses, for at least 1 hr 28 d post-inoculation. Pouches were drained and root systems visibly evaluated for numbers of egg masses produced using an illuminated desk magnifier and light table. The virulence index was calculated as a percentage obtained by the number of egg masses produced on resistant plants divided by the number of egg masses produced on susceptible plants, times 100.

Greenhouse culture experiment 2: After 1,330 d in culture on tomato, the original Rk-virulent isolate was compared to its sub-culture maintained on resistant CB46 for 1 yr, and to the M. incognita race 3 isolate. Each culture was assayed for egg mass production and virulence as described in Experiment 1 with some modifications. Each isolate was tested on CB3, CB5, CB46, and UCR430. Root systems were inoculated with a mean of 1,516 motile J2. Each isolate x genotype combination was replicated 10 times, and inoculated plants in pouches were arranged in a randomized complete block design. The experiment was conducted twice.

Field microplot experiment: Small field microplots were established at the Agricultural Experiment Station, UC-Riverside, and were planted with H8-9, CB46, or UCR430 in each cowpea growing season (May through August) for 5 yr. Each plot was 0.64 m² and contained 7.2 m³ of loamy sand (87:2:11 sand:silt:clay) surrounded by a concrete barrier 1 m deep. Each plot was planted with the same cowpea genotype over the five seasons. Each plot was planted with 12 plants in late May and harvested in late August. Plots were maintained as a weed-free fallow after harvest until planting the next year. Each plot was infested with 160,000 eggs 3 wk after planting in the first year with either the Avirulent, the Virulent, or an 80,000:80,000 eggs combination of the Avirulent:Virulent isolates (mixed isolate). Each of the nine nematode x cowpea combinations was represented by a single plot arranged in a completely randomized design.

The Avirulent, Virulent, and mixed isolates in plots planted repeatedly with H8-9 are referred to as the AS, VS, and MS populations, respectively. The Avirulent, Virulent, and mixed isolates in plots planted repeatedly with CB46 are referred to as the AR, VR, and MR populations, respectively. The Avirulent, Virulent, and mixed isolates in plots planted repeatedly with UCR430 are referred to as the AR2, VR2, and MR2 populations, respectively.

Virulence to Rk and Rk² was assayed at infestation and at harvest in years 3, 4, and 5. All virulence assays were performed as described for the greenhouse culture experiments with the following modifications. Virulence assays were performed using eggs of the Avirulent and Virulent isolates at inoculation or sub-samples of the field plot populations collected from roots at harvest in years 3, 4, and 5. Roots systems in pouches were inoculated with 900 motile J2. The mixed isolate inoculum was not tested at the time of inoculation, and an estimate of its percent virulence to genes Rk and Rk² was calculated as follows: [(% Virulence of Avirulent isolate) (% Hatch of Avirulent isolate) + (% Virulence of Virulent isolate) (% Hatch of Virulent isolate)]/(Sum of % Hatch of Avirulent and Virulent isolates). Inoculum from the Avirulent and Virulent isolates used to inoculate the field plots in year 1 was assayed for virulence using 1,000 J2/pouch and cowpea 8685 as the susceptible genotype.

Statistical analysis: An analysis of variance (ANOVA) was performed using Fisher's Protected LSD to isolate significant paired differences in mean values. Statistical analysis of virulence data was performed on log₁₀ [(egg masses produced on resistant cowpea) + 1] − log₁₀ (mean egg masses produced on susceptible cowpea). All statistical analysis was performed using Minitab Release 13 statistical software.

Results

Greenhouse culture experiment 1: Virulence to gene Rk declined during continuous culture of the 75% Rk-virulent M. incognita isolate on susceptible tomato (Fig. 1). Periodic assays of virulence on both the Rk-carrying cowpea genotypes CB5 (Fig. 1A) and CB46 (Fig. 1B) revealed similar trends of decreasing Rk-virulence with time. Virulence to Rk declined from a range of 77% to 86% at isolation from the field to 4.5% on CB46 after 2,061 d on susceptible tomato (Fig. 1B). The virulence decline on both CB5 and CB46 fit exponential decay curves (Fig. 1A and 1B). There was no correlation of numbers of egg masses produced on susceptible CB3 plants with length of culture on tomato; a mean of 157.7 egg masses were produced per CB3 plant across all assays.

Fig. 1.

The effect of continuous culturing on susceptible tomato for several years on the virulence to resistance gene Rk in cowpea of a Meloidogyne incognita population isolated from a cowpea field showing breakdown of Rk-resistance. Percent virulence was determined on cowpea genotypes CB5 (A) and CB46 (B) carrying gene Rk, based on egg-mass production on these resistant genotypes as a proportion of that on susceptible cowpea CB3.

Greenhouse culture experiment 2: Analysis of variance indicated no significant effect of experiment, so the data from the two experiments were combined, and analyses are presented for the combined data. The natural avirulent M. incognita race 3 isolate produced a mean of < 1 egg mass/root system on Rk plants and no egg masses on Rk² plants (Table 1). The Rk-virulent isolate cultured on susceptible tomato for 1,330 d had a moderate level of virulence (average 13.2%) on both CB5 and CB46 plants. This isolate produced a mean of 2.7 egg masses/root system on Rk² plants (1.4%) (Table 1). In contrast, a sub-culture of this isolate maintained for 1 yr on Rk-cowpea plants produced similar numbers of egg masses per root system on susceptible CB3 and resistant (gene Rk) CB5 and CB46 plants, indicating a high level of virulence (88.6% and 119.8%, respectively) (Table 1). Egg mass production and virulence of this subculture were higher on CB5 plants than CB46 plants (P < 0.05) in both experiments (Table 1), and this subculture had a higher level of egg mass production and Rk²-virulence (7.4%) on UCR430 plants (P < 0.05) than the other two isolates.

Table 1.

Numbers of egg masses per root system and percent virulence to genes Rk and Rk² of Meloidogyne incognita isolates on susceptible and resistant cowpea genotypes.

Field microplot experiment: At the time of microplot infestation, virulence to Rk was higher (P < 0.05) in the Virulent isolate (92%) than in the Avirulent isolate (1.2%), and virulence to Rk² was higher (P < 0.05) in the Virulent isolate (8.1%) than in the Avirulent isolate (0.2%). The mixed population was not tested at infestation, but percent virulence was calculated to be 66% to Rk and 5.8% to Rk² from the virulence of the component isolates.

Differences (P < 0.05) in virulence to Rk (Table 2) and Rk² (Table 3) were observed at all assay dates among the populations. At each assay date, populations AS, VS, and MS from plots planted with susceptible cowpea were less virulent (P < 0.05) to Rk and Rk² compared to populations from plots planted with cowpeas carrying Rk or Rk². For example, population VS was less virulent to Rk and Rk² compared to populations VR or VR2.

Table 2.

Percentage virulence to gene Rk of Meloidogyne incognita populations in inoculated field plots planted with susceptible, Rk, or Rk² cowpea over 5 yr.

Table 3.

Percent virulence to gene Ris² of Meloidogyne populations in infested field plots planted with susceptible, Rk, or Rk² cowpea over five seasons.

Effects on virulence of susceptible cowpea plantings: Population AS was less virulent (P < 0.05) to Rk at all assay dates compared to populations VS and MS, ranging from 1.2% (at infestation) to 4.4% virulent over 5 yr (Table 2). Population VS declined in virulence to Rk from 92% at infestation to 26% in year 3, then increased to 58% in year 5. Population MS was similar to VS in virulence to Rk by years 4 and 5 (Table 2).

Population AS was highly avirulent to Rk² during 5 yr on susceptible cowpea (Table 3). Populations VS and MS retained a similar, low level of virulence to Rk² that was higher (P < 0.05) than that of AS in years 4 and 5 (Table 3).

Effects on virulence of Rk-gene cowpea plantings: Populations AR, VR, and MR increased in virulence to Rk from infestation to harvest in year 3 (Table 2). The three populations remained highly virulent to Rk in years 4 and 5, with virulence ranging from 122% to 190%. AR and VR did not differ in any year, whereas MR was more Rk-virulent (P < 0.05) than AR and VR in year 4 and than AR in year 5 (Table 2).

Virulence to Rk² increased in the AR, VR, and MR populations after 3 yr of Rk-cowpea cropping and remained at moderately high levels during years 4 (19%–60%) and 5 (42%–47%) (Table 3). AR was less Rk²-virulent (P < 0.05) than MR in years 3 and 4. Similarly, AR was less Rk²-virulent (P < 0.05) than VR in year 3 but not in year 4. MR was more Rk²-virulent (P< 0.05) than VR in years 3 and 4 (Table 3). The populations did not differ in Rk²-virulence in year 5 (Table 3).

Effects on virulence of Rk²-gene cowpea plantings: Populations AR2, VR2, and MR2 in plots planted with Rk² plants increased in virulence to Rk between infestation and harvest in year 3 (Table 2). Population AR2 was less Rk-virulent (P < 0.05) than VR2 in years 3 and 4 and less Rk-virulent (P < 0.05) than MR2 in year 3. After 5 yr of cropping to Rk²-plants, the three populations were highly Rk-virulent (113%–139%) and did not differ (Table 2). The Rk-virulence levels of the AR2, VR2, and MR2 populations were similar to those of the AR, VR, and MR populations from plots planted with Rk-cowpeas in years 4 and 5 (Table 2).

Population AR2 increased in virulence to Rk² between infestation and harvest in year 3 and increased again in year 4 and year 5 (Table 3). However, the Rk²-virulence of AR2 was lower (P < 0.05) than that of VR2 and MR2 in years 3, 4, and 5. Populations VR2 and MR2 did not differ on any assay date. By year 5, populations AR, VR, MR, VR2, and MR2 did not differ in Rk²-virulence (34% - 48%). Only population AR2 differed, being less virulent (22%, P < 0.05) than these other populations (Table 3).

Discussion

Virulence has been studied in other long-term culturing experiments with Meloidogyne populations, including virulence to gene Mi in tomato, and variable results have been obtained concerning selection and stability of virulence (Castagnone-Sereno et al., 1996). Selection for virulence in other nematodes also has been reported, e.g., in Globodera pallida to resistance in wild potato hybrids of Solanum vernei (Turner and Fleming, 2002), in Heterodera schachtii to resistance in sugar-beet derived from Beta procumbens (Mueller, 1992), and in H. avenae to resistance in oat (Lasserre et al., 1996).

The long-term culture experiment with M. incognita showed that virulence to a resistance gene can decline under standard culture of Meloidogyne on susceptible hosts, an expected consequence of stabilizing selection in the absence of the resistance gene (Caswell and Roberts, 1987). Similar observation of virulence loss during culture on susceptible host plants has been made for Meloidogyne populations virulent to resistant grapevines (M. V. McKenry, unpub. data). However, this type of virulence loss has not been clearly established in other nematode-host resistance interactions. In the case of Mi resistance in tomato, both stable maintenance of M. incognita virulence to Mi on susceptible tomato over many generations (Riggs and Winstead, 1959; Bost and Triantaphyllou, 1982; Castagnone-Sereno et al., 1993) and decline in virulence or loss of host range of virulent isolates have been reported (Jarquin-Barberena et al., 1991; Castagnone-Sereno et al., 1996).

Although the observed decrease in virulence was gradual and required several years in the present study, this finding has important implications for resistance deployment. Virulence decline in the absence of the resistance gene indicates that rotation to susceptible crops between resistant cowpea may reduce the virulence potential of the field population, assuming rotations of sufficient duration. However, the rapid directional selection of virulence by culture on resistant Rk-gene plants, as found in the sub-culture maintained for 1 yr on resistant cowpea, suggests that even one season of a resistant crop will result in increased virulence in the Meloidogyne population. The change in virulence frequency under greenhouse culture raises concerns for the use of cultures in experiments and screening for resistance. To maintain cultures of Rk-virulent lines for inoculum for screening sources of resistance effective against Rk-virulent populations, it is necessary to maintain cultures on resistant cowpea, such as the subculture maintained on Rk-cowpea in this study.

A detailed examination of Rk-virulence in these and other M. incognita isolates by isofemale line analysis revealed that the population was a mixture of virulent and avirulent lineages (Petrillo and Roberts, 2005a) and that some virulent lineages had a lower relative fitness than avirulent lineages (Petrillo and Roberts, 2005b). The differential fitness of the lineages provides an explanation for our observed changes in virulence in greenhouse cultures and the field plots. The substantial genetic variability within M. incognita isolates also may be involved. For example, a hierarchical analysis of mtDNA variable number tandem repeats of these same and other M. incognita isolates revealed only 7% diversity among the isolates, but 60% and 30% of the total genetic diversity occurred within individuals and within isolates, respectively (Whipple et al., 1998). The consistently higher egg mass production on CB5 plants than CB46 plants indicated that reproduction on resistant plants is influenced to some extent by factors other than the resistance alleles, although the overall virulence trends were the same for the two Rk-genotypes.

In greenhouse experiments, the isolate that had declined in Rk-virulence on tomato was quickly reselected for Rk-virulence by sub-culturing on resistant cowpea, presumably because of differential reproduction of the Rk-virulent lineages within the isolate. Those experiments also indicated that the Rk-virulent sub-culture was moderately virulent to gene Rk². This apparent co-virulence to genes Rk and Rk² was confirmed in the field plot study, wherein cross-selection of virulence to genes Rk and Rk² was observed in both directions. The AR, VR, and MR populations during five seasons on Rk-cowpea became moderately virulent to gene Rk². Likewise, in the reverse direction, all populations (AR2, VR2, MR2) exposed to Rk²-cowpea plantings for five seasons became highly virulent to gene Rk. This reciprocal cross-selection of virulence suggests that the Rk and Rk² genes share some commonality or overlap with respect to the signaling and recognition pathways involved in resistance. Gene Rk² is associated with Rk as an allele at the Rk locus or tightly linked to Rk within 0.17 map units (Roberts et al., 1996). Thus these genes have features of complex resistance loci, as found for other plant-pathogen resistance systems (Hulbert et al., 2001).

Rk² expresses a broader and stronger resistance than Rk, being effective against diverse M. incognita populations including ones virulent to Rk, and also against M. javanica (Roberts et al., 1996; Ehlers et al., 2002). Although cross-selection for virulence to Rk and Rk² was observed, it was not reciprocally uniform. By year 5, the levels of virulence to Rk² were much lower (34%–48%) than the virulence to Rk (113%–190%), regardless of whether virulence selection occurred on Rk-cowpea or Rk²-cowpea plantings. Nevertheless, cross-selection could be a problem for effective deployment of resistance using Rk²-cowpea cultivars, due to selection of virulence against both genes.

The mixed field study population was intended to simulate what we suspected and later confirmed (Petrillo and Roberts, 2005a), namely, that the Rk-virulent field populations were a natural mixture of avirulent and virulent lineages. The mixed population on susceptible plants retained virulence over five seasons and had a similar virulence profile to the virulent population (VS), suggesting that in-field virulence decline on susceptible hosts might be slower than what occurs in greenhouse cultures. On the Rk- and Rk²-cowpea plantings, the mixed population (MR, MR2) was indistinguishable from the virulent population (VR, VR2) in each case. In the field, the virulent population apparently had greater fitness on the resistant plants.

We have demonstrated the decline of virulence during greenhouse culture on susceptible host plants and that rapid selection for virulence is possible by culturing nematodes on resistant plants. In addition, cross-selection for virulence to genes Rk and Rk² in both directions was found, with important implications for nematode management based on deployment of these resistance genes in root-knot nematode-infested cowpea fields.