Effect of Soil Temperature on Reproduction of Root-knot Nematodes in Flue-cured Tobacco with Homozygous Rk1 and/or Rk2 Resistance Genes

Population source: A population of root-knot nematodes was received from Clemson University in Clemson, South Carolina and identified as M. incognita race 3 based on perineal pattern and stylet morphology, esterase phenotypes, species-specific sequence-characterized-amplified-region (SCAR) primers (MiF/MiR, IncK14F/R, Rinc/Rinc), PCR-DNA sequences on rRNA 18S, ITS, 28S, D2/D3, histone and mitochondrial DNA COII-16S gene (W. M. Ye, pers. comm.), and three greenhouse differential host trials to identify to a host race level (Taylor and Sasser, 1978). The population was maintained on susceptible tomato (Solanum lycopersicum L.) cultivar ‘Rutgers’ in greenhouses at the Southern Piedmont Agricultural Research and Extension Center (SPAREC) near Blackstone, VA.

Greenhouse trials: Nine greenhouse trials were conducted in 2014 and 2015 to evaluate the influence of temperature on root-knot resistance gene efficacy in flue-cured tobacco. Among the nine trials, three were performed during the summer (June–July 2014), three were performed during the fall (September–October 2014), and were performed during the winter (December 2014–January 2015). All trials were conducted at the SPAREC and organized in a randomized complete block design. Six tobacco entries were evaluated in each trial: the cultivar Coker 371-Gold (C371G), susceptible to four common root-knot nematode species (M. incognita, M. arenaria, M. javanica, and M. hapla Chitwood, 1949); cultivars SC 72 and NC 95, homozygous for Rk1; the breeding line T-15-1-1, homozygous for Rk2; and the breeding lines STNCB-2-28 and NOD 8, homozygous for both Rk1 and Rk2. Plants were grown in adjacent open-top root zone cabinet growth chambers (Environmental Growth Chambers, Chagrin Falls, OH) set at one of the three soil temperatures so that nematode reproduction could be investigated at 25 ± 1.2°C, 30 ± 1.6°C, and 35 ± 2.3°C. Soil temperatures were taken with a digital stem thermometer between 8 a.m. and 9 a.m., and again between 4 p.m. and 5 p.m. each day, and the temperature range was averaged across all experiments (General Tools & Instruments LLC, Secaucus, NJ). The six entries were replicated four times in the three summer trials (one trial at each soil temperature) and six times in each of the three fall and three winter trials. Each seedling was transplanted into a 10-cm diameter clay pot filled with a 2:1 mixture of topsoil (53% sand, 40% silt, 7% clay; pH 5.5) and Profile® Greens Grade™ porous ceramic material (Profile Products, Buffalo Grove, IL) when seedlings had formed approximately four true leaves (approximately 5–10 cm tall). Each plant was inoculated with 3,000 root-knot nematode eggs one week after transplanting by pipetting 40 ml of egg suspension calibrated to contain 75 eggs/ml into two holes approximately 4 cm deep on either side of the plant. Plants were grown without supplemental lighting in summer and fall trials, but plants in the winter trials were grown with supplemental lights set to a 13:11 light:dark photoperiod using GE Plant and Aquarium Ecolux 40 W wide spectrum bulbs (General Electric, USA). Soil moisture was standardized between chambers based on a soil moisture deficit in a designated pot in each growth chamber using a soil tensiometer system (Irrometer Co. Inc., Riverside, CA).

Data on percent root galling, number of egg masses per gram of root, and number of eggs per gram of root were collected after 35 days. Soil was rinsed from roots and the fresh root system was weighed. Galled roots were separated from non-galled roots and percent root galling was calculated based on fresh weight of galled roots versus fresh weight of the entire root system. Galled roots were then recombined with non-galled roots and mixed. Mixed galled and healthy roots were then divided in half by weight, and half were stained with 0.15 g/L Phloxine B (Dickson and Struble, 1965) to define egg masses. Numbers of egg masses from three stained 1 g subsamples per plant were counted using a dissecting microscope at x10 to estimate the number of females. Eggs were bleach-extracted from the second half of the root system (Hussey and Barker, 1973). Extracted eggs were suspended in 500 ml of water and counted in two 10 ml aliquots using a compound microscope at x40 magnification. To assess nematode reproductive capability on each entry, reproductive index (P_f/P_i) was calculated by dividing the final number of eggs extracted per plant (P_f) by the number of eggs in the initial inoculum (P_i) (Sasser et al., 1984).

Statistical analysis: Data from each trial were analyzed separately by a one-way analysis of variance (ANOVA); means were compared using the Tukey-Kramer honest significant difference (HSD) test (P = 0.05). Percent galling, egg mass counts, and egg counts were transformed (log₁₀ (x + 1)) before ANOVA. Non-transformed means were presented. Data were analyzed using the Statistical Analysis System-JMP® Pro 11 (SAS Institute, Cary, NC).

Rk1 entry results: In all trials, root galling and numbers of egg masses and eggs per gram of root were significantly lower at 25°C in both entries possessing only Rk1 compared to the susceptible control, C371G (P ≤ 0.05; Figs. 1–3). Galling and numbers of egg masses and eggs were also reduced on both entries possessing Rk1 alone versus the susceptible control at 30°C in the summer and fall trials (P ≤ 0.05), but not in winter. In contrast, no differences in galling were noted among any of the entries at 30°C or 35°C in the winter test (Fig. 1C). Egg mass numbers were also similar for the Rk1 entries and the susceptible control at 35°C in the summer trial and at 30 and 35°C in the winter trial (Figs. 2A, 2C). Total numbers of eggs were lower at 25°C on Rk1 entries versus the susceptible control in all trials (P ≤ 0.05; Fig. 3). Total numbers of eggs were also similarly lower on both of the Rk1 entries relative to the susceptible control at 30°C in the summer and fall experiments (Figs. 3A, 3B). In the winter study, egg numbers were lower at 30°C on the Rk1 entry SC 72 versus the susceptible control but not on the Rk1 entry NC 95 (Fig. 3C). At 35°C, total egg numbers were lower on both Rk1 entries compared to the susceptible control in the fall trial (Figs. 3B). However, the total number of nematode eggs were not lower on the Rk1 entries compared to the susceptible control in the summer and winter trials, when nematode egg numbers appeared relatively low, even on the susceptible control (Figs. 3A, 3C). Nematode reproductive index was significantly lower on Rk1 entries compared to the susceptible entry at both 25°C and 30°C in all trials (P ≤ 0.05; Table 1). At 35°C, nematode reproductive index was significantly lower on the Rk1 entries versus the susceptible control only in the fall trial, but not in the summer or winter trials (Table 1).

Figure 1:

Figure 2:

Figure 3:

Similar trends were observed in galling in the summer and fall trials, and in reproduction in the fall trial at 35°C, but not in the winter trials (Fig. 1, Table 1). Significant differences were observed in numbers of egg masses and eggs among entries at 35°C in the fall trials, but not in the summer or winter experiments (Figs. 2, 3).

Rk2 entry results: Root galling in the summer trial was lower on all entries possessing the Rk2 gene compared to susceptible C371G when soil temperature was 25°C or 35°C (P ≤ 0.05), but not 30°C (Fig. 1). Rk2 did not reduce galling compared to C371G at any soil temperature in the fall and winter trials. At 25°C, fewer egg masses were observed per gram of root on T-15-1-1 versus the susceptible control in both fall and winter trials (P ≤ 0.05; Fig. 2). No differences were observed between the Rk2 entry and the susceptible control at 25°C in the summer trial, or at 30°C or 35°C in any trial. Total numbers of eggs per gram of root were similar for the Rk2 entry and the susceptible control at all three temperatures in all trials, with the exception that fewer eggs were counted for T-15-1-1 than for C371G at 25°C in the fall study (P ≤ 0.05; Fig. 3). While reproductive indices were significantly lower for Rk2 plants relative to the susceptible control at 25°C in all experiments (P ≤ 0.05), reproductive indices of these entries were more similar at 30°C and 35°C (Table 1).

Both galling and numbers of egg masses were significantly lower on the two Rk1 entries compared to the Rk2 entry at 25°C in all trials (P ≤ 0.05; Figs. 1, 2). At 30°C, these differences were significant only in the summer and fall experiments (P ≤ 0.05; Figs. 1, 2). At 35°C, both galling and egg mass numbers in the fall experiment were lower on the Rk1 entries compared to susceptible C371G, but not in the summer or fall trials (P ≤ 0.05; Figs. 1, 2). Total numbers of nematode eggs per gram of root were lower for Rk1 entries versus the Rk2 entry at 25°C in the winter trial (P ≤ 0.05; Fig. 3). Total nematode egg numbers in the fall 25°C test were lower for only one Rk1 entry, NC 95, compared to the Rk2 entry. Significantly fewer eggs were recovered from Rk1 entries compared to the Rk2 entry at 30°C in both summer and fall tests, but not in the winter test (P ≤ 0.05; Fig. 3). In contrast, differences in total egg numbers between the Rk1 entries and C371G at 35°C were only significant in the fall test, but not in the summer or winter trials (P ≤ 0.05; Fig. 3). Reproductive indices at 25°C were lower for Rk1 entries relative to the Rk2 entry in the fall and winter trials, but not in the fall trial (Table 1). Reproductive indices at 30°C were significantly lower on Rk1 entries compared to the Rk2 entry T-15-1-1 in all trials (P ≤ 0.05; Table 1). Reproductive indices at 35°C were significantly lower on Rk1 entries versus the entry with only Rk2 in the summer, fall, and winter trials. (Table 1).

Rk1Rk2 entry results: Galling remained significantly lower (P ≤ 0.05) for entries with both Rk1 and Rk2 than for the susceptible control at all temperatures in the summer and fall experiments. This also held true at 25°C in the winter trial, but galling was similar to that on all other entries, including the susceptible control, at 30°C and 35°C in the winter trial (Fig. 1). Galling was similar when Rk2 was combined with Rk1 in comparison to Rk1 alone at all temperatures in every trial, except that galling was lower on NOD 8 compared to SC 72 at 35°C in the summer trial (P ≤ 0.05; Fig. 1). Galling on entries with both Rk1 and Rk2, relative to Rk2 alone, was also consistently lower at 25°C across all experiments, and was also lower at 30°C for both entries possessing both resistance genes in the summer and fall trials, but not in the winter test. Possession of both Rk1 and Rk2 reduced galling at 35°C versus only Rk2 in the fall trial, but not in the winter trial. Galling was significantly lower for NOD 8 versus the Rk2-only entry in the summer trial but was similar for both STNCB-2-28 and the Rk2-only entry in that experiment (Fig. 1).

Mean egg mass numbers per g of root were always significantly lower for entries with both Rk1 and Rk2 than for the susceptible control and T-15-1-1 (with Rk2 alone) at 25°C (P ≤ 0.05), and at 30°C in summer and fall trials (P ≤ 0.05; Fig. 2). In the winter trial, mean egg mass numbers were significantly lower at 30°C on STNCB-2-28, but not for NOD 8, compared to the susceptible control (P ≤ 0.05; Fig. 2). Mean egg mass numbers were similar among all entries at 35°C in the summer and winter trials but were lower for entries with both Rk1 and Rk2 compared to the susceptible control and the entry with Rk1 alone in the fall study. At 25°C and 35°C, mean egg mass numbers were similar in all tests among entries possessing both Rk1 and Rk2 as well as those containing only Rk1. Mean egg mass numbers were similar at 30°C among entries with Rk1, with or without Rk2, in the fall trial, but were significantly lower (P ≤ 0.05) on NOD 8, but not STNCB-2-28 in the summer experiment. They were significantly lower on STNCB-2-28, but not NOD 8, in the winter study (Fig. 2).

Mean total numbers of nematode eggs produced per gram of feeder root were lower at 25°C for both Rk1Rk2 entries versus both the susceptible control and the entry possessing only Rk2 in all trials (Fig. 3; P ≤ 0.05). Mean numbers of eggs per g of root at 30°C were again lower in the fall experiment for both Rk1Rk2 entries compared to both the fully susceptible control and the entry possessing only Rk2 (P ≤ 0.05), but were similar among STNCB-2-2, the Rk2 entry, and the susceptible control in the summer trial. In the winter experiment, mean numbers of eggs/g root at 30°C were lower for the Rk1Rk2 entry STNCB-2-28 versus susceptible C 371G and the Rk2 entry T-15-1-1 (P ≤ 0.05), but similar for the other Rk1Rk2 entry NOD 8, and T-15-1-1. At 35°C, mean numbers of total eggs produced per gram of root were again significantly lower on both Rk1Rk2 entries compared to susceptible C371G and T-15-1-1, but no statistically significant differences were observed among any of the entries in the summer and winter studies, when very few nematode eggs were observed, even on the susceptible control C371G (Fig. 3).

Nematode egg production was similar among the Rk1Rk2 entries and the Rk1-only entries in all three trials at 25°C, and at 30°C in the fall experiment (Fig. 3). Total nematode eggs/g of root were low, but still intermediate for the Rk1Rk2 entry STNCB-2-28 versus all of the other entries except Rk1Rk2 entry NOD 8 at 30°C in the summer trial. Total nematode egg production/g of root at 30°C in the winter trial was lower on Rk1Rk2 entry STNCB-2-28 compared to the Rk1 entry NC 95 and the Rk1Rk2 entry NOD 8, but was similar for NOD 8 compared to both entries that possessed only Rk1 – SC 72 and NC 95. No differences were observed among entries that possessed Rk1 at 35°C in any of the three studies, whether or not Rk2 was also present. (Fig. 3).

Reproductive indices were lower versus the susceptible C371G on entries possessing only Rk1 at all temperatures, in all trials, except at 35°C in the winter screen, when numerical differences among entries were not statistically significant (P ≤ 0.05; Table 1). On the other hand, similar differences between fully susceptible C371G and the Rk2 entry T-15-1-1 were only significant at 25°C, but never at 30°C or 35°C. Reproductive indices were always numerically lower at 25°C for the two Rk1 entries compared to T-15-1-1. These differences were significant in the summer and winter experiments, but not in the study conducted during the fall. Similar differences were also observed in all trials conducted at 30°C, and in the fall trial at 35°C, but no significant differences were observed across all entries at 35°C in the summer and winter studies. Reproductive indices were similar among entries possessing Rk1 either alone or with Rk2 at all temperatures and in all experiments, with the exception of the fall trial at 30°C, when reproductive indices were higher for the Rk1Rk2 entry STNCB-2-29 than for the Rk1 entry NC 95. Reproductive indices were also always numerically lower at 25°C when Rk1 was combined with Rk2, although the difference in the fall study was not statistically significant (Table 1). Possession of both root-knot resistance genes versus only Rk2 was always associated with lower reproductive indices when investigations were performed at 30°C, but only in the fall test at 35°C (P ≤ 0.05; Table 1).

Results from these studies suggested that resistance to M. incognita race 3 declined at 30°C, and especially at 35°C, compared to 25°C. A drastic reduction in nematode reproduction at 25°C and 30°C was associated with Rk1 alone and Rk1Rk2 together compared to susceptible cultivar C371G. However, reproduction on both Rk1 and Rk1Rk2 appeared similar to that on C371G at 35°C. Additionally, Rk1Rk2 did not appear to be significantly more effective than Rk1 in suppressing reproduction of this population of M. incognita race 3. No consistent trends between Rk1 and Rk1Rk2 occurred with regard to production of egg masses or total nematode eggs, nor reproductive index, across temperatures, perhaps because there was so little parasitism on these cultivars and breeding lines.

The temperature-sensitive Rk1 resistance results corroborate those by Baum et al. (1995), who observed that M. incognita race 3 resistance began breaking down at 28°C, and further at 31°C and 35°C, in the Rk1-containing tobacco entry NC 95. Results of this study are also supported by results from other crops. Temperature-sensitive Meloidogyne resistance has been noted in tomato with the Mi resistance gene (Dropkin, 1969; El-Sappah et al., 2019), common bean (Omwega and Roberts, 1992), sweet potato (Jatala and Russell, 1972), grape (Vitis spp.) rootstocks (Ferris et al., 2013), and cotton (Gossypium hirsutum L.) (Carter, 1982).

Similarly, numbers of egg masses and eggs, and the reproductive index of plants with Rk2, were almost always significantly lower than those of the susceptible control at 25°C. However, this trend seemed to be lost at 30°C and 35°C when there were frequently no significant differences in reproduction between the Rk2 entry and the susceptible control. The Rk2 gene appeared to inhibit reproduction less effectively at 30°C and 35°C than at 25°C, although Rk2 seemed to provide only modest resistance to this population of M. incognita race 3, even at 25°C, especially compared to effects of Rk1 or Rk1Rk2.

Schweppenhauser (1968) observed higher M. javanica infection in resistant (Rk2) and susceptible lines in the summer, when approximate greenhouse air temperatures regularly reached 41°C, compared to winter infection, when temperatures were approximately 21°C to 35°C. Ng’ambi et al. (1999b) observed that the entry Okinawa did not provide any resistance to M. javanica in their study at greenhouse temperatures of 27 ± 2°C and 30 ± 3°C. In contrast, Fukudome and Kamigama (1982) found that galls caused by M. javanica were rare on Okinawa at 28°C, but at 35°C there were no differences in galling, invasion rate, and development compared with the susceptible entry. In Zimbabwe, increased M. javanica root invasion and galling on both resistant and susceptible tobacco was observed when soil temperature rose from 25°C to 30°C, but parasitism actually decreased when soil temperature was further increased to 35°C (Shepherd and Barker, 1990). Shepherd and Barker (1990) did not state whether or not M. javanica-resistant entries contained the Rk2 gene, although this seems likely. Despite some variation, this trend toward increased galling at higher temperatures appeared across all entries.

Despite reduced nematode reproduction, galling on the Rk2 entry in most trials was not significantly different from that of the susceptible entry at all temperatures evaluated. This observation was consistent with results from experiments performed at Virginia Tech in 2010 and 2011 examining M. javanica reproduction on the same flue-cured tobacco entries used in the current experiment (Ma et al. unpubl. data). However, Ng’ambi et al. (1999b) noted the amount of galling and reproduction were generally correlated. Consistent with our results, galling by M. incognita and M. javanica without a corresponding increase in egg masses was noted in Prunus rootstocks (Lu et al., 2000). The authors determined that nematodes “can infect and partially develop in resistant root systems, but cannot complete their life cycles.”

Because a reproductive index is calculated as the ratio of the final nematode population relative to the initial population (often calculated from the number of eggs recovered from the root system at the cessation of trials relative to the number of eggs delivered at inoculation), a reproductive index >1 suggests that a plant is not resistant to the nematode. Reproductive indices on the susceptible cultivar and the entry possessing only Rk2 were always >1 in this research and were generally higher in trials conducted at 30°C or 35°C versus at 25°C. Reproductive indices on entries possessing Rk1 alone, or with Rk2, were always <1 at 25°C but were almost always >1 at 30°C and 35°C, suggesting an inhibition of resistance. Although reproductive indices on Rk1 and Rk1Rk2 entries almost always increased from an index of <1 at 25°C to >1 at 30°C and 35°C, reproduction on these entries at these temperatures remained well below that on the susceptible cultivar and the Rk2 entry in most trials, suggesting that while resistance may diminish with increasing temperature, some effect persists.

A shorter average natural day length (11 hr 42 min) during the fall trial may have affected results compared to the summer (14 hr 25 min) and winter (13 hr) trials (U.S. Naval Observatory). Average natural day length during the winter trial (9 hr 48 min) was supplemented to 13 hr, which could have produced results more similar to those in the summer trial. The explanation for differences in egg mass and egg counts among trials is unclear. Meng et al. (2015) determined light quality had an effect on tobacco seedling root growth, which may have affected root growth in this experiment, but the extent to which this impacted nematode host-seeking and root penetration, as well as subsequent reproduction, is also not clear.

Meloidogyne incognita inhabits areas with an average annual temperature range of 18°C to 30°C, with an optimum warm-month temperature of 27°C (Eisenback et al., 1981). In this study, galling and egg mass numbers from the susceptible entry usually decreased as soil temperature increased from 25°C to 35°C, presumably because 35°C is above the optimal temperature range for this population. Soil temperatures in Virginia during the tobacco growing season are commonly above 30°C. Maximum daily soil temperatures logged at the Virginia Tech Southern Piedmont AREC were above 30°C for 8 days in May 2014, 15 days in June and July 2014, 11 days in August 2014, and six days in September 2014. May, June, and July also had maximum temperatures above 35°C for one, two, and three days, respectively. Soil temperatures were taken below grass; soil temperatures in tobacco fields were likely higher.

Fluctuation in soil temperatures might also influence efficacy of resistance to root-knot nematodes in tobacco. Tobacco plants resistant to tobacco mosaic virus (TMV) became susceptible at temperatures above 28°C, but the hypersensitive response returned when temperatures dropped, killing the plant (Malamy et al., 1992; Marathe et al., 2002; Zhang et al., 2009). Tomato plants resistant to M. incognita due to the Mi gene became susceptible after 48 hr at 34°C and remained susceptible for one to two days after temperatures were reduced to 27°C (Zacheo et al., 1995). Juveniles that had penetrated and started feeding when resistance was deactivated continued developing when temperatures decreased. Some entries of tomato (Ammati et al., 1986; Wang et al., 2013) and pepper (Capsicum annuum L.) (Thies and Fery, 1998; Djian-Caporalino et al., 1999, 2001) retained resistance to Meloidogyne species at high temperatures, even up to 32°C and 42°C in four pepper entries (Djian-Caporalino et al., 2001). Perhaps there are sources of temperature-stable root-knot nematode resistance in wild Nicotiana relatives that could mitigate some of these issues. However, at present, flue-cured tobacco growers will continue to face challenges associated with high and fluctuating soil temperatures complicating the management of root-knot nematodes.