Reproduction of Meloidogyne arenaria race 2 on flue-cured tobacco with putative resistance derived from Nicotiana repanda

Both time and location of tests, and the interaction between test and genotype, had significant effects on results, so root-knot parasitism data from each trial were analyzed independently (p ≤ 0.0001 time and location; p ≤ 0.0181 interaction between test and genotype).

Significant differences in root galling were found among entries in all trials (Table 1). 81-R617A had significantly lower root galling than all other entries in the June to August trial conducted in Blackstone (Table 1). In the trial in Blacksburg from September to November, 81-R-617A exhibited significantly less root galling than all entries with the exception of STNCB 2-28 and BAG 29-15-3-32-1, both of which are homozygous for resistance genes Rk1 and Rk2 (Table 1). Similarly, in the trial in Blacksburg from June to August, galling was lowest on 81-R-617A and BAG 29-15-3-32-1 relative to T-15-1-1, which in this trial experienced the most root galling observed across all four trials (Table 1). Galling was significantly lower on 81-R-617A than on Hicks in three of the four trials. No entry consistently exhibited the highest levels of root galling in these trials (Table 1).

Significant differences in egg mass counts were found among entries in all trials (Table 2). Entry 81-R-617A had significantly fewer egg masses than all entries except for susceptible Hicks in the June to August trial in Blackstone, and fewer egg masses than all entries with the exception of BAG 29-15-3-32-1 in the test conducted in Blackstone from September to November (Table 2). In the trial conducted concurrently in Blacksburg, 81-R-617A had significantly fewer egg masses present than all entries with the exception of CC 13 and STNCB 2-28 (Table 2). In contrast, egg mass production was significantly less on Hicks and K326 than on STNCB 2-28, while egg production on 81-R-617A and BAG 29-15-3-32-1 was intermediate, along with T-15-1-1 and CC 13 in the trial conducted from June to August in Blacksburg (Table 2). Egg mass production was significantly lower on 81-R-617A than Hicks in both trials conducted from September to November (Table 2).

Significantly fewer eggs were recovered from 81-R-617A than K326 and CC 13 in the trial from June to August in Blackstone, while in the trials in Blackstone and Blacksburg from September to November, 81-R-617A experienced significantly less egg production than all other entries (Table 2). In the trial conducted from June to August in Blacksburg, significantly fewer eggs were recovered from K326 and T-15-1-1 than all entries except for 81-R-617A, which was intermediate in egg production in this trial (Table 2). Entries possessing both resistance genes almost never differed significantly in egg production from whatever entry experienced the greatest egg production in all four trials (Table 2). Egg production was significantly lower on 81-R-617A than Hicks in both trials conducted from September to November (Table 2).

Trends in reproductive indices were similar to those in egg production. Entry 81-R-617A had a reproductive index significantly less than all other entries except for STNCB 2-28 in the trial conducted in Blackstone from September to November, while in the concurrent trial conducted in Blacksburg, the reproductive index of 81-R-617A was significantly less than all other entries (Table 2). In the Blackstone trial at this time, the reproductive index of 81-R-617A was 0.0; reproduction was essentially undetectable, with only 17 eggs present per gram of root (Table 2). Hicks and K326 had reproductive indices significantly lower than entries CC 13 and STNCB 2-28 in the trial conducted in Blacksburg from June to August, and T-15-1-1 had a lower reproductive index than CC 13, while reproduction on 81-R-617Aand BAG 29-15-3-32-1 was intermediate (Table 2). No significant differences in reproductive indices were observed among entries in the trial in Blackstone from June to August (Table 2). 81-R-617A had significantly lower reproductive indices than Hicks in both trials conducted from September to November (Table 2).

The results of these trials suggest that entry 81-R-617A, which possesses traits derived from N. repanda, may exhibit resistance to M. arenaria race 2 that is greater than that imparted by resistance genes Rk1 and Rk2 alone or in combination. Despite some inconsistency in our results, galling was almost always numerically lowest on 81-R-617A, and in two trials this trend was significant relative to all other entries. In three of the four trials, galling on 81-R-617A was significantly lower than that on the susceptible control, and the same pattern significant with respect to reproduction in two trials. However, in one trial, susceptible entry Hicks and K326, which possesses Rk1 only, had a significantly lower reproductive index than that of entries CC 13 and STNCB 2-28 (which possess both Rk1 and Rk2), while the reproductive index of 81-R-617A was intermediate. In contrast to the observation of Ng’ambi et al. (1999a), we did not see root necrosis associated with reduced egg production in the highly susceptible entry, Hicks, or any other entry; in fact, egg numbers were high on susceptible entries in all four tests. It is unclear exactly what factors contributed to the considerably lower reproduction in both trials conducted from September to November. Inoculum was clearly viable in all four trials based on galling of the susceptible entry, and the inoculum appeared consistent with the initial population from trial to trial based on the morphology of perineal patterns on females recovered from select infested roots. Perhaps differences in daylength and fluctuation between day and night time ambient temperatures were more significant than anticipated.

Way (1994b) demonstrated that a male sterile hybrid, RK3, which is homozygous for Rk1 and heterozygous for Rk2, showed significant reductions in parasitism relative to a susceptible control. This is similar to our observation that entry CC 13, which possesses a similar combination of root-knot nematode resistance genes as RK3, experienced significantly less root-knot nematode parasitism and reproduction than the susceptible entry, but did not differ significantly from the two entries homozygous for both genes, STNCB 2-28 and BAG 29-15-3-32-1, with the exception of one trial in which BAG 29-15-3-32-1 exhibited significantly lower root galling that both CC 13 and STNCB 2-28. However, our results with M. arenaria race 2 appear to contrast with Way’s (1994a) observation that BAG line plants, of which STNCB 2-28 is a parent, are significantly more resistant to M. javanica than STNCB 2-28, and that BAG lines are more resistant than RK3. We did not detect significant differences in any metrics of reproduction among any of the entries possessing both resistance genes, regardless of zygosity, except in the aforementioned case of root galling.

Resistance to M. javanica in male sterile hybrids between N. repanda and N. longiflora or N. palmeri was reported by Schweppenhauser et al. (1963). Davis et al. (1988a) observed a high level of resistance to M. javanica and M. arenaria in a N. repanda line called 46-G, but did not indicate what race of M. arenaria was used in their trials. While we saw significant reductions in galling and reproduction on 81-R-617A relative to a susceptible entry and other entries lacking the N. repanda derived traits, considerable galling and egg production still occurred in several trials. Davis et al. (1988a) reported that the N. repanda entry in their study, 46-G, exhibited little or no gall development relative to standard cultivars; reproduction was inhibited completely, as inferred by the absence of juvenile nematodes in soil at the end of trials (Davis et al., 1988a). The authors also identified sources of resistance to both M. javanica and M. arenaria in breeding lines from other Nicotiana species, but noted that lines with favorable agronomic traits relative to standard cultivars lost this resistance, or tolerance in many cases, upon backcrossing with N. tabacum (Davis et al., 1988a). Bui et al. (1992) demonstrated that resistance to M. arenaria and tobacco mosaic virus could be integrated into N. tabacum from N. repanda using a form of protoplast fusion and N. sylvestris as a bridge species, but found that nematode resistance was considerably more difficult to retain in backcrosses than resistance to the virus, being present in only 25 of 270 backcrosses. Again, the authors of this study do not indicate what race of M. arenaria was used to evaluate root-knot nematode resistance (Bui et al., 1992). Ng’ambi et al. (1999b) observed a ‘moderate’ level of resistance to M. arenaria race 2 in 81-R-617A and a related breeding line, 81-RL-2K, in contrast to a ‘high level’ of resistance observed in a breeding line from South Africa, SA1214. In their trials, an average of over 7,000 M. arenaria race 2 eggs were present per gram of root for entry 81-R-617A, compared with fewer than 3,000 egg per gram of root on SA 1214, a considerable difference, but one that is not clearly identified as significant. Importantly, the authors note that these lines are the first N. repanda entries to be identified with resistance to M. arenaria race 2 that could be compatible with existing N. tabacum accessions for breeding and crop improvement (Ng’ambi et al., 1999b).

Ng’ambi et al. (1999a) demonstrated that resistance to M. arenaria race 1 in a breeding line related to 81-R-617A, 81-RL-2K, is conditioned by the same single-dominant gene which confers resistance to M. incognita races 1 and 3 in the commercial flue-cured tobacco cultivar Speight G 28. This gene, Rk1, is widely available in commercial flue-cured tobacco in the United States (Koenning et al., 1999). Rk1 confers resistance to root-knot nematodes via a hypersensitive response that inhibits nematode feeding site development and subsequent galling and reproduction (Ng’ambi et al., 1999b; Schneider, 1991) The presence of this gene explains the resistance to ‘M. arenaria’ populations not identified to race observed by other authors, which can be inferred to have been race 1 (Bui et al., 1992; Davis et al., 1988a). However, Rk1 does not impart resistance to M. arenaria race 2, and appears to have relatively little impact on M. javanica (Ng’ambi et al., 1999b), suggesting that a different system is responsible for the resistance to these nematodes observed in N. repanda and related breeding lines.

Another resistance gene, known as Rk2, provides considerably greater resistance to M. javanica than Rk1, and if the two genes are ‘stacked’ an even greater level of resistance is conferred (Ma et al. unpublished data; Ternouth et al., 1986). There is also evidence that the combination of both genes confers a significantly greater degree of resistance to a variant of M. incognita race 3 than either resistance gene alone. In the study in question, Rk1, rather than Rk2, was associated with greater reductions in nematode parasitism and reproduction (Pollok et al., 2016). The mechanism of resistance associated with Rk2 is unknown, although it has been speculated to be different than the hypersensitive response associated with Rk1 because Rk2 does not appear to consistently inhibit gall formation, but does reduce reproduction (Pollok et al., 2016).

It is unclear what gene or genes interact with Rk1 to confer the relatively high level of resistance to M. arenaria race 2 we observed in the line with N. repanda traits in our trials, which was equivalent to or greater than that exhibited by entries possessing both Rk1 and Rk2. The relatively drastic reductions in root galling we observed on 81-R-617A suggests the possibility of a hypersensitive response. In tomatoes, diminished nematode parasitism has been associated with the inhibition of penetration due to alterations in the composition of root exudates perceived by infective juveniles in the presence of arbuscular mycorrhizal fungi (Vos et al., 2012). Augmentations of root morphology have been observed in nematode resistant Prunus spp., wherein resistant cultivars feature a different arrangement of epidermal cells in root tips compared with susceptible cultivars (Ye et al., 2009). Pollok et al. (2016) speculated that Rk2 may simply slow down nematode feeding and development, or is involved in inhibition of a specific stage of egg mass development. It is possible that some combination of the aforementioned mechanisms is involved in resistance to root-knot nematodes in flue-cured tobacco entries like 81-R-617A that possess traits derived from N. repanda.

It would be of great value to determine what gene or genes are implicated in nematode resistance in N. repanda. As Ng’ambi et al. (1999a) observed, the genetic basis of resistance associated with Rk1 is ‘narrow’ and they cautioned that ‘repeated use of this resistance may select resistance-breaking biotypes.’ As Murphy et al. (1987) observed, tobacco cultivars within similar market classes are highly related. This presents potential challenges for breeders attempting to improve disease resistance and agronomic traits in cultivars in the future because of a limited genetic diversity within available germplasm (Lewis and Nicholson, 2007). Introducing new genetic sources of disease resistance to this pool of germ plasm from sources like N. repanda could help remedy some of these challenges. Additionally, moving forward, tobacco growers in the southeastern United States face a new, highly destructive, polyphagous nematode threat in the form of M. enterolobii. This nematode, which has recently been found on root-knot resistant sweet potato in South Carolina (Rutter et al., 2019) and is present in at least eight counties in North Carolina (Schwartz, 2019) is a quarantine-level pathogen (Thiessen, 2018) that, while not yet known to be directly impacting tobacco growers, presents a potentially serious threat, as this nematode is not controlled by available forms of resistance including Mi and Rk-mediated resistance (Ye, 2018). Perhaps exploiting N. repanda germplasm, which has long been a source of novel traits for cultivated tobaccos (Stavely et al., 1973) could present solutions to these and other issues facing tobacco breeders and growers alike.