Screening Cultivated Eggplant and Wild Relatives for Resistance to Bacterial Wilt ( Ralstonia solanacearum )

: Bacterial wilt, caused by Ralstonia solanacearum , is highly diverse and the identiﬁcation of new sources of resistance for the incorporation of multiple and complementary resistance genes in the same cultivar is the best strategy for durable and stable resistance. The objective of this study was to screen seven accessions of cultivated eggplant ( Solanum melongena L.) and 40 accessions from 12 wild relatives for resistance to two virulent R. solanacearum strains (Pss97 and Pss2016; phylotype I, race 1, biovar 3). The resistant or moderately resistant accessions were further evaluated with Pss97 in a second trial under high temperatures (and also with Pss2016 for S. anguivi accession VI050346). The resistant control EG203 was resistant to Pss97, but only moderately resistant to Pss2016. One accession of S. sisymbriifolium (SIS1) and two accessions of S. torvum (TOR2 and TOR3) were resistant or moderately resistant to Pss97 in both trials. Solanum anguivi VI050346, S. incanum accession MM577, and S. sisymbriifolium (SIS1 and SIS2) were resistant to Pss2016 in the ﬁrst trial. However, S. anguivi VI050346 was susceptible in the second trial. These results are important for breeding resistant rootstocks and cultivars that can be used to manage this endemic disease.


Introduction
Bacterial wilt (BW), caused by Ralstonia solanacearum, is one of the most economically important soil-borne diseases of tomato (Solanum lycopersicum L.) and eggplant (Solanum melongena L.) in the tropics and subtropics [1]. This pathogen enters plant roots through wounds and multiplies rapidly in the vascular system, so that the xylem elements are filled with bacterial cells which block the xylematic flow, leading to yellowing of foliage, general wilting, and eventually plant death [2]. It was first described by Smith (1896) in potato (Solanum tuberosum L.), tomato, and eggplant. Bacterial wilt causes great losses because of its severe symptoms, wide geographic distribution, and unusually broad host range, which includes more than 200 plant species belonging to 53 different families [3,4]. In addition, R. solanacearum can survive in soil for many years [5].
Ralstonia solanacearum grows well at a temperature range of 28-32 • C, but it is also found in cold weather in Europe and North America [6,7]. Phenotypically, R. solanacearum strains are divided into four phylotypes based on geographical regions: phylotype I strains originate from Asia and Africa, phylotype II from the Americas, phylotype III from Africa and the surrounding islands, and phylotype IV from Indonesia [8,9]. These phylotypes are able to infect eggplant and other important Solanaceae crops, such as potato, tomato, and pepper (Capsicum annuum L.) [10].

Material and Methods
The experimental investigation reported here was carried out at the greenhouses of the World Vegetable Center (WorldVeg), Taiwan.

Seeds and Plant Growth Conditions
Seeds of seven accessions of cultivated eggplant and 40 accessions of 12 wild relatives of eggplant were obtained from the genebanks of Universitat Politècnica de València (Spain) and the World Vegetable Center (WorldVeg) ( Table 1). Eggplant accessions EG203 and EG048 were used as resistant and susceptible checks, respectively. Before sowing, seeds were soaked in water for one day, and then another day in 500 ppm GA 3 (Gibberellic acid) to improve seed germination [27]. After these treatments, seeds were washed with water and directly sown in 3 inch diameter plastic pots containing a steam sterilized soil mixture (3:1:1:1 ratio of soil, rice hulls, sand, and compost) and moved to WorldVeg's greenhouse (16/8 h day/night). Temperature ranged from 23 • C to 36 • C and relative humidity from 81.5% to 84.1% during the evaluation trials. Seedlings were watered daily and fertilized weekly with an NPK (Nitrogen, Phosphorus, Potassium) 15-15-15 fertilizer. Four-week-old plants (four to six fully expanded true leaves) were tested for R. solanacearum resistance, as described below.

Pathogen and Resistance Assays in First Trial
Inoculations were conducted with two virulent R. solanacearum strains (Pss97 and Pss2016). The Pss97 strain was isolated from infected eggplants from Pingtung County of southern Taiwan in 1991, and belongs to the predominant virulence group in Taiwan. Pss2016 was isolated from infected tomatoes grafted on eggplant rootstocks from Yilan County of northern Taiwan in 2015. Both strains were identified as phylotype I, race 1, biovar 3, based on identification conducted through host range [28], biovar test [29][30][31], and molecular markers [8] at the Bacteriology unit of WorldVeg. Bacterial strains stored at −80 • C were cultured on a 2,3,5-triphenyl tetrazolium chloride-amended (TTC) medium [32] and incubated at 30 • C for two days. Then, several typical fluid white colonies with pink centers from TTC medium were transferred to 523 medium [33] and incubated at 30 • C overnight for multiplication. The bacterial mass from overnight cultures was transferred and suspended in water, and the concentration was adjusted to an optical density of 0.3 at 600 nm wavelength (about 10 8 cfu/mL).
Roots of accessions and checks were injured with a knife by cutting through the soil 1-2 cm away from the stem base before inoculation. The inoculum volume was determined as the ratio between bacterial suspension and potting mixture, to a proportion 1:10 (v/v). Hence, 30 mL of bacterial suspension (10 8 cfu/mL) was poured into each pot (3 inch) and the inoculated plants were kept in a plastic greenhouse [34]. Plants were watered in excess two times a day after inoculation to maintain the soil moisture high. Two plants per accession and check without inoculation were used as negative controls. Plants were arranged according to a randomized complete block design (RCBD), with three replications and eight plants per accession in each replication (24 plants per accession and resistant and susceptible checks). All accessions and check plants were kept in a greenhouse after inoculation (28.4 ± 2.0 • C, 16/8 h day/night, and humidity 81.5 ± 2.0 %) and the bacterial wilt severity was evaluated once a week for four weeks using wilting percentage (W%) and disease index (DI), based on a disease rating scale (0-5) (Figure 1), where 0 = no symptoms, 1 = one leaf partially wilted, 2 = two or three leaves wilted, 3 = all leaves wilted except the top two or three leaves, 4 = all leaves wilted, 5 = plant dead [35].

Evaluation of Resistant and Moderately Resistant Accessions in Second Trial
Accessions classified as R or MR, based on DI ≤ 40%, as described above, along with checks were re-evaluated using the same resistance screening protocol in the same season to confirm the resistance in accessions identified at WorldVeg's greenhouse under high temperatures (34.8 ± 2.0 • C, 16/8 h day/night, humidity 84.1 ± 2.0%). Due to limited seed numbers and low germination rates in accessions of S. incanum (MM577), S. sisymbriifolium (SIS1 and SIS2), and S. torvum (TOR2 and TOR3), these accessions were not re-evaluated for resistance to Pss97 in the second trial.

Results
The resistance reaction and category of seven accessions of cultivated eggplant and 40 accessions of wild relatives against R. solanacearum strains Pss97 and Pss2016 at four weeks of inoculation is presented in Table 2. The susceptible check (EG048) displayed the expected reactions to strains Pss97 and Pss2016. In this way, all EG048 plants wilted and died rapidly two and three weeks after inoculation by Pss97 and Pss2016, respectively. The resistant check (EG203) was resistant to Pss97, with less than 10% of W% and DI, and moderately resistant and moderately susceptible to Pss2016, with 50% and 62.5% of W% and 31.1% and 48.8% of DI in the two trials. None of the evaluated eggplant genotypes were immune or highly resistant to both strains. All accessions of cultivated eggplant were susceptible to Pss97 and Pss2016, with a range 75.8-100% of W% and 81.1-100% of DI, respectively. Of the 40 wild accessions screened for resistance to Pss97 in the first trial, two accessions-TOR2 and TOR3, of S. torvum-were resistant, with 16.7% and 12.5% of W% and 5.8% and 10.8% of DI, respectively. Two accessions-SIS1 and SIS2, of S. sisymbriifolium-were moderately resistant, with 33.3% and 37.5% of W% and 33.3% and 37.5% of DI, respectively. In the first trial, S. anguivi VI050346, S. incanum MM577, and two accessions-SIS1 and SIS2, of S. sisymbriifolium-were classified as resistant to Pss2016, with ranges of 9.2-23.3% for DI and 33.3-66.7% for W%. Solanum torvum TOR3 was classified as moderately resistant to Pss2016, with 32.2% of DI and 38.9% of W%, and S. anguivi ANG1 was moderately susceptible, with 95.8% of W% and 45.8% of DI. It is worth mentioning that all cultivated eggplant accessions and more than 85% of the wild relative accessions tested in the first trial were susceptible to both bacterial wilt strains (Figure 2), and wilt symptoms were appeared one or two weeks after inoculation.

Second Trial
Six out of 40 accessions of wild relatives of eggplant (VI050346, MM577, SIS1, SIS2, TOR2, and TOR3) that were identified as resistant or moderately resistant in the first trial (i.e., those that scored ≤40% for DI), along with the checks, were screened again in a second trial at high temperatures using the same R. solanacearum strains of Pss97. Solanum anguivi VI050346 was susceptible to both Pss97 and Pss2016. S. sisymbriifolium SIS1 was resistant to Pss97, with 17.5% of DI. Two accessions-TOR2 and TOR3, of S. torvum-were moderately resistant to Pss97 (Table 3). Table 3. Re-evaluation of the resistance of selected resistant and tolerant accessions against Ralstonia solanacearum strains Pss97 and Pss2016 at four weeks after inoculation in the second trial.

Taxa and Accession
Code a indicates the means of three replications of wilt percentage (W%) at fourth week after inoculation. Means followed by ± standard error (± SE). c indicates the means of three replications of disease index (DI) at fourth week after inoculation. Means followed by ± standard error (± SE). d Resistance category according to the DI at fourth week after inoculation. R = resistant (0-30%), MR = moderately resistant (>30-40%), MS = moderately susceptible (>40-50%), S = susceptible (>51%). ND indicates no data due to the limited number of seeds or low germination.

Discussion
Bacterial wilt, caused by R. solanacearum, has been ranked second in the list of the most scientifically and economically important bacterial pathogens [2,15]. Resistance to R. solanacearum has been reported in some tomato cultivars, such as Hawaii 7996, Hawaii 7997, and Hawaii 7998 [37], but these cultivars have not been widely accepted due to poor horticultural traits, such as small fruits, linked with bacterial wilt resistance [38]. Grafting susceptible tomato cultivars onto bacterial wilt-resistant eggplant rootstocks provides good control, especially during the hot-wet season, and can minimize problems caused by flooding [14,25]. This technology has been adopted by WorldVeg on a large scale in the Philippines, Vietnam, and Taiwan to control bacterial wilt in tomatoes. A number of bacterial wilt resistant eggplant rootstocks, such as EG203, EG195, EG190, and TS03, have been successfully developed and released by WorldVeg to manage bacterial wilt in tomato, but resistance levels are not stable under different environmental conditions. These reasons encouraged us to explore the germplasm for more stable sources of resistance to bacterial wilt in the cultivated eggplant genepool and in a representation of the highly genetically diverse eggplant wild relatives [39].
Resistance to R. solanacearum has been identified in previous studies in a number of accessions of cultivated eggplant (S. melongena) and in distant wild relatives, such as S. capsicoides, S. sisymbriifolium S. sessiliflorum, S. stramonifolium, S. virginianum, S. grandiflorum, S. hispidum, S. torvum, S. nigrum, S. americanum, and S. scabrum [39]. In our study, no immunity was found in the materials tested, however, two accessions of S. torvum (TOR2 and TOR3) and two accessions of S. sisymbriifolium (SIS1 and SIS2) were observed to be resistant or moderately resistant to one or both of the bacterial wilt strains tested. These results confirm earlier findings that found high levels of bacterial wilt resistance in these two wild species [25,40]. Both accessions belong to the tertiary genepool of eggplant and are therefore promising for introgression breeding in eggplant [39]. Rootstocks of S. torvum accessions have been used in several studies and were highly resistant to bacterial wilt, and resulted in a good fruit yield in the scion [41,42]. Solanum sisymbriifolium rootstocks showed resistance against bacterial wilt disease under sick plots in field conditions [40]. In addition, S. incanum MM577 was observed to be resistant to Pss2016, although these results should be confirmed. This species belongs to the secondary genepool of eggplant and can be crossed with cultivated eggplant [43]. In fact, lines of S. melongena with introgressions from S. incanum have been obtained recently [44]. This is the first report of bacterial wilt resistance in S. incanum, which has also been reported as resistant to Fusarium oxysporum f. sp. melongenae [42], another harmful soil-borne disease for Solanaceae crops.
Differences in bacterial wilt resistance levels between the first and second trials were evident in accessions of S. anguivi, S. sisymbriifolium (SIS2), and S. torvum. The susceptibility or reduction of resistance in the second trial second could be due to the higher temperatures observed in the second trial, which reached 36 • C. Similar results were found in tomato and tobacco (Nicotiana tabacum), where resistant cultivars become susceptible when exposed to temperatures above 28 • C [26,45]. In addition, soil moisture and soil temperature may have influenced the resistance reaction of the genotypes [46]. Although we did not study the mechanisms of resistance, the resistance present in identified genotypes could be due to a higher concentration of secondary metabolism, such as polyphenols and steroidal glycoalkaloids, that prevent bacterial movement into the vicinity of the plant system [16,47] All cultivated eggplant accessions and more than 85% of the wild relative accessions screened in our study were susceptible to bacterial wilt. The early wilt symptoms appeared one week after inoculation in most of susceptible accessions, and were completely wilted after two weeks. Similarly, a high incidence of bacterial wilt in tomato was observed 15 days after inoculation at the early stage of crop growth [48]. Also, other authors [46] found that most of the susceptible genotypes displayed a susceptible reaction in their early stages of growth (10 to 20 days after inoculation).

Conclusions
Among the eggplant genotypes tested in our experiments, high levels of resistance were detected in S. sisymbriifolium and S. torvum, for both strains. In addition, S. incanum (MM577) and S. anguivi (VI050346) displayed resistance to Pss2016. However, resistance in S. sisymbriifolium (SIS2) and S. anguivi (VI050346) might be decreased or broken down when temperature increases, as occurred in the second trial. Hence, an evaluation of bacterial wilt resistance under different environmental conditions would provide a better understanding of the resistance mechanisms of these sources and their potential interest for offering broad and stable resistance, which is required for the development of cultivars with durable resistance. Our results made it possible to identify some new sources of resistance to bacterial wilt in wild relatives of eggplant from very different origins. These materials may be of interest for the development of resistant rootstocks and/or cultivars that can be used to manage bacterial wilt in eggplant and also, when used as rootstocks, in tomato.  The project is managed by the Global Crop Diversity Trust with the Millennium Seed Bank of the Royal Botanic Gardens, Kew UK and implemented in partnership with national and international genebanks and plant breeding institutes around the world. For further information, go to the project website: http://www.cwrdiversity.org/. This work has also been funded in part by World Vegetable Center core funds from Republic of China (Taiwan), UK aid, United States Agency for International Development (USAID), Australian Centre for International Agricultural Research (ACIAR), Germany, Thailand, Philippines, Korea, and Japan. The authors declare no conflict of interests.

Conflicts of Interest:
The authors declare no conflict of interest.