Tomato Hybrid and Local Varieties Screened for Resistance to Bacterial Wilt Caused by Ralstonia solanacearum under Screen House and Field Conditions

Bacterial wilt is the most damaging disease of tomato in Benin. To select resistant varieties of tomato against bacterial wilt, 21 tomato varieties were evaluated twice under screen house and field conditions in a randomized complete block design with four replicates and 21 treatments which are the tested varieties. Three weeks old tomato seedlings were inoculated by crown flooding with 40 ml of a bacterial suspension of R. solanacearum strain CR-1103-2, phylotype I sequvar 14. At 28 days after inoculation, the control resistant varieties Hawaii 7997 and PADMA maintained their resistance and the known susceptible variety Tohounvi showed to be susceptible to the bacterial wilt. Among the hybrid tomato varieties, only Cobra 26 was moderately resistant. No resistant line was found among the tested local varieties. The hybrid varieties Buffalo, Petomech, Tropimech,


Introduction
Tomato (Lycopercicum esculentum) is one of the forty most cultivated vegetable species in the world. Its world production was estimated in 2019 at 180,766,329 tons [1]. In Benin, it is grown throughout the country [2], occupying an area of 37,648 ha [1] on which 274,700 tons of tomatoes are harvested [1]. Tomato is usually consumed by Beninese [3], due to its richness in vitamins and minerals that are no longer to be denied [4]. Its production occupies an important socio-economic activity in Benin [5] where the average yield is 7.29 t/ha while 35.97 t/ha is produced in the world in 2019 [1]. This low yield is mainly due to many factors including pests and diseases, particularly the bacterial wilt caused by the phytopathogenic bacterium Ralstonia solanacearum [6]. This bacterium causes yield losses of up to 72%. Bacterial wilt is known as economically important and most damaging disease of solanaceous plants [7].
Unfortunately, these different methods show limited effectiveness when not integrated. The most effective and worldwide reported management method is the use of resistant genes. According to Oussou et al. [16], disease management through the introduction of resistant varieties is the appropriate method that is less expensive for farmers and not aggressive to the environment. Thus, several resistant varieties have been selected worldwide. Wang et al. [17] developed the bacterial wilt resistant tomato variety Hawaii 7996 at the World Vegetable Center. In Korea, Kim et al. [18] selected four resistant and two moderately resistant accessions from 285 accessions tested from all continents. To detect resistant genes to be improved for rootstock, Ganiyu et al. [19] identified tomato varieties AVT09803, AVT00201, Tomachiva and Eyetom resistant to bacterial wilt. Laeshita and Arwiyanto [20], by evaluating bacterial wilt resistance of 16 tomato varieties, identified Servo, Kaliurang, Melinda, Amelia, Rewako varieties as moderately resistant. In Benin, few varieties resistant to bacterial wilt are available.
Research by Oussou et al. [16], when evaluating the behaviour of nine tomato varieties to bacterial wilt, identified two varieties, PADMA and Platinum as resistant. These two resistant varieties are not sufficient, especially for good production of tomato in contaminated areas. The choice of farmers is also limited regarding organoleptic criteria. Of the 36 tomato varieties grown in Benin [21], the status of only eight of them to bacterial wilt was known. The behavior of Be-nin local tomato varieties has not been reported except the susceptibility of variety Tounvi [6] [7] [16]. To increase the number of tomato resistant genes, this study aims to evaluate the behavior of other hybrid and local tomato cultivated varieties in Benin.

Tomato Varieties Tested
Twenty-one tomato varieties comprising 16 hybrids and 5 locals were tested, namely Sumo, Prado, Ninja, Jaguar, Anaya, Topaze, Cobra 26, Cobra34, Euclide, Savana, Kiara, Buffalo, Petomech, Tropimech, PADMA and Hawaii 7997 as hybrids varieties and Kêkêkô, Tounvi, Akikonkouinkouin, Agbotrui and Adjaa as local varieties. The hybrid varieties were obtained from registered seed "Bénin semence" distributors and the local varieties were provided by the Vegetable Program of National Agricultural Research Institute of Bénin (INRAB, Benin). The variety Hawaii 7997 was obtained from the World Vegetable Center, Cotonou. The local variety Tounvi was used as the known susceptible control [6] while PADMA [16] and Hawaii 7997 [22] were considered as resistant controls.

Inoculum Preparation
The virulent strain CR-1103-2 of phylotype I sequvar 14*, obtained from the experimental site of World Vegetable Center, Cotonou, was used in this study. It was mass-produced on SMSA culture medium [23] from a 48-hour old isolated colony. After 48 hours of incubation at 30˚C, the bacterial cream from the mass production was removed and suspended in sterile distilled water (SDW). The obtained stock solution was adjusted using a spectrophotometer to the concentration of 10 8 Colony Forming Unit per milliliter (CFU•ml −1 ).

Screen House Experimentation
The experiment was conducted in the screen house of the Laboratory of Crop Protection, INRAB (6˚24'35''N, 2˚19'56''E), Benin. The trial was set up using a Randomized Complete Block design with four replicates and 21 treatments that are the tomato varieties. The average air temperatures prevailing were 28.3˚C during the first experiment conducted from April to May (rainy period) and 31.2˚C during the second experiment conducted from August to September (dry period). The nursery was carried out for three weeks on a sterilized soil/manure substrate (2v:1v) for one hour at 80˚C. The young tomato plants of three weeks were root cut with hand and then transplanted into pots containing 0.5 kg of sterilized soil with manure substrate. There were 10 pots per variety per block. Each tomato seedling was inoculated by crown flooding with 40 ml of inoculum. Five plants per variety were inoculated with SDW to serve as control. The inoculated seedlings were monitored for 28 days after transplanting. Compost (10 g) was applied to each seedling 14 days after transplanting. Data were collected three times a week at 48-hour intervals from the day of transplanting.

Field Experimentation
Determination of the existing bacterial concentration in the soil at the experimental site To determine R. solanacearum concentration existing in soil before the experiment, rhizosphéric soils were collected three days before planting. From a composite of nine rhizospheric soil samples from each 6 m 2 10 grams were suspended in 100 ml of SDW. After stirring and decanting the mixture, 45 ml of the supernatant was taken, adjusted to 50 ml and centrifuged at 6000 rpm for 5 minutes. The supernatant was discarded and the pellet suspended in 5 ml. A series of four dilutions were made from this stock solution. Plates were inoculated with 100 µl per dilution and then incubated for 48 hours at 30˚C.

Colonization of Tomato Plants by Ralstonia solanacearum
The detection of bacteria in inoculated but not wilted plants was performed at 28 days after transplanting in screen house and field experiment. The fingerprinting technique described by [16] was used. Ten-cm length stems of unwilted plants were cut from the collar. Each stem was rinsed under the tap water, cleaned with 70% ethanol and then flamed with 90˚ ethanol. The base of each stem was refreshed with a sterile scalpel to make five fingerprints per plant on the SMSA medium. Inoculated plates were incubated at 30˚C for 48 hours.
Stem samples were also collected from inoculated and wilted plants to check and confirm presence of the bacteria as responsible for the disease symptoms observed after inoculation. A plant is considered colonized when R. solanacearum is isolated from that plant, whether it is wilted or not. The classification of varieties according to their resistance based on the IBW was carried out according to the [24] scale defined in Table 1.

Bacterial Colonization Index
The bacterial colonization index (BCI) was calculated as the percentage of The area under the bacterial wilt incidence progress curve The area under the bacterial wilt incidence progress curve (AUIbwPC) is a function of time. AUIbwPC = f(t). It was calculated according to the formula: bw is bacterial wilt and t i represents time at the period i [25].

Statistical Analysis
One-way analysis of variance (ANOVA) was used. It was applied to the Bacterial Colonization Index, the Bacterial Wilt Incidence and the AUIbwPC. To classify the means into a homogeneous group, the Student-Newman-Keuls test was used. The analyses were carried out under STATA statistical software. The virulent strain CR-1103-2 of phylotype I sequvar 14*, obtained from the experimental site of World Vegetable Center, Benin in Abomey-Calavi, was used in this study. It was mass-produced on SMSA culture medium [23] from a 48-hour old isolated colony. After 48 hours of incubation at 30˚C, the bacterial cream from the mass production was removed and suspended in sterile distilled water (SDW). The obtained stock solution was adjusted using a spectrophotometer to the concentration of 10 8 Colony Forming Unit per milliliter (CFU•ml −1 ).

Behavior of Tomato Varieties after Inoculation with Ralstonia solanacearum in the Screen House
Various responses were obtained from tomato varieties inoculated with R. solanacearum. This variation in response was observed in terms of bacterial wilt incidence, bacterial colonization index and area under the bacterial wilt incidence progress curve. The analyses of variance indicate a significant difference among varieties regarding IBW (P = 0.000) and AUIbwPC (P = 0.000). As for BCI, no significant difference (P = 0.1849) was observed among the varieties during the two experiments.
The variety Cobra 26 showed the lowest incidence of bacterial wilt during the two experiments (15% and 17.5%, respectively) which was not significantly different from that of the tested resistant varieties PADMA and Hawaii 7997 for which the IBW was 0%. The Kêkêfô, Tropimèche, Petomech, Topaze, Jaguar, Prado, Tounvi, Akikonkouin, Anaya, Agbotrui, Cobra 34, Kiara, Adjaa, Savana, Buffalo, Ninja, Sumo, and Euclid varieties were all susceptible with an IBW ranging from 60.71% to 97.50%. During the experiment 2, the IBW increased for 76.19% of the tested varieties ( Table 2).
The colonization test showed that all the tested varieties were colonized by R. solanacearum including the resistant check varieties. The varieties Tropimech, Kêkêfô, Petomech and Hawaii 7997 were 100% colonized. The bacterial colonization index increased during the experiment 2 compared to the experiment 1 for most varieties.
The area under the bacterial wilt incidence progress curve (AUIbwPC) showed a significant difference among varieties for the two experiments. The varieties Kêkêfô and Kiara showed the highest AUIbwPC, whether during the first or the experiment 2. As for the IBW and BCI, an increase in AUIbwPC was noted during the experiment 2 for 80.95% of the varieties. The increase of the IBW ranged from 4.76% to 42.5% for the susceptible varieties.
During the experiment 1, the lowest AUIbwPC (0 to 30) were recorded for the Hawaii 7997, PADMA and Cobra 26 varieties with a none significant difference ( Table 2). The same trend was noted for the three varieties except the AUIbwPC of Cobra 26 was slightly higher than those of Hawaii 7997 during the experiment 2 ( Table 2). The AUIbwPC was significantly high for other varieties.
In the screen house, considering the IBW, BCI and AUIbwPC of the experiment 2, the tomato varieties PADMA and Hawaii 7997 were highly resistant, the

Determination of the Bacterial Concentration in the Soil of the Experimental Site in Station
The mean concentration of bacteria in the soil of experimental site varied among blocks. It was on average 5.5 × 10 3 , 5.4 × 10 3 , 3.1 × 10 3 and 1.8 × 10 3 CFU•g −1 of soil for blocks 4, 3, 2 and 1, respectively.

Behavior of Tomato Varieties under Ralstonia solanacearum Pressure on Station
In the on-station experiment under screen house, no significant difference was obtained for the IBW (P = 0.1500). In the field experiment, the IBW showed significantly different among varieties (P = 0.000  In the same column, the means followed by the same letters are not significantly different according to the Student-Newmann-Keuls test with P = 0.05. E = Experiment. Values after the ± sign represent standard error. The AUIbwPC showed a similar trend to the IBW obtained. The lowest values were recorded for the PADMA, Hawaii 7997, Cobra 26 and Cobra 34 varieties and varied between 0 and 10.94 (experiment 1) and between 0 and 9.37 (experiment 2). The other varieties showed a significantly higher AUIbwPC ranging from 12.5 to 190 for experiment 1 and from 10.16 to 95.31 for experiment 2.

Discussion
Bacterial wilt of Solanaceae is one of the major constraints to tomato production in Benin [6]. In the current study, the behaviour of 21 hybrid and local tomato varieties was evaluated after artificial inoculation in the screen house and on a field naturally infested by R.  [27]. According to reports from Grimault et al. [28]; Monma et al. [29] and Oliveira et al. [30], this resistance trait might be due to single dominant or recessive genes.
The variety Cobra 34 showed an unstable state of resistance to bacterial wilt. With the field results, the variety Cobra 34 appeared to be resistant. The bacterial pressure induced by artificial inoculation demonstrated the real state of sus-ceptibility of this variety. This result showed that artificial inoculation of pathogens to crop assures a better breeding study for resistance. This result corroborates those of Sikirou and Wydra [31] who reported that the number of cowpea varieties previously selected as resistant in the field was reduced when artificially inoculated. On the other hand, the varieties Kêkêfô, Akikonkouin, Agbotrui, Adjaa, Sumo, Prado, Ninja, Jaguar, Anaya, Topaz, Euclid, Savana, Kiara, Buffalo, Petomech and Tropimech tested in the same conditions wilted under artificial and naturally infection conditions. This result might show that the bacterial concentration of about 10 3 CFU•ml −1 is enough to induce bacterial wilt in the field. Thus, the tomato variety Cobra 34 might not be considered as resistant since it wilted the most after artificial inoculation as compared to natural field inoculation.
In addition, the current finding showed that all the tested varieties, were colonized by R. solanacearum, regardless of their susceptibility or resistance level. These results might reveal that resistance of a variety to R. solanacearum does not prevent its penetration into the variety as indicated by Sang et al. [32]. These authors explained that varietal resistance does not depend on whether or not the plant is infested, but on its ability to prevent the pathogen from obstructing the sap-conducting vessels.
Our finding also showed that the IBW was low during the experiment 1 of screen house which was conducted during rainy season and high during the experiments 2 conducted when weather was dry. This weather condition during the experiment 2 seems to increase disease expression up to 76.19% and 90.47% of the tested varieties in the screen house and in the field, respectively. The increase in the temperatures by 2.9˚C and 4.4˚C in screen house and in the field respectively, might favour the disease development leading to the observed results. These might suggest that bacterial wilt is more damaging in hot conditions and may be a key factor favouring R. solanacearum virulence. These results are similar to those found by Caruso et al. [33] who concluded that R. solanacearum loses its virulence at low temperatures during their study on population variation of R. solanacearum biovar 2 in Spanish rivers at different seasons. In addition, the studies by van Overbeek et al. [34] on the storage temperatures of R. solanacearum biovar 2 strains demonstrated that low temperature induces virulence loss for R. solanacearum. Our results are also in line with those of Singh et al. [35] who evaluating the effect of temperature, cultivar and induced root wounding on virulence of R. solanacearum in greenhouse tomato crops, showed no wilt symptoms at 20˚C or lower. In addition, Bittner et al. [36], breeding tobacco varieties resistant against R. solanacearum, reported the highest incidences when temperatures were between 30˚C and 35˚C and no symptoms at temperatures less than or equal to 15˚C. Oussou et al. [16] also reported a high wilt rate of Solanum macrocarpum artificially inoculated with R. solanacearum during the dry period compared to the wet period.
The susceptibility behaviour of local and hybrid tomato varieties cultivated in Benin could result in a rapid expansion of the bacterial wilt with consequent re-duction of tomato yield in the contaminated areas where access to seeds of resistant varieties is difficult. The current study identified Cobra 26 as moderately resistant variety which is good alternative to control R. solanacearum wilt disease in the field. Efforts should continue to identify more resistant or moderately resistant varieties and evaluate other open-pollinated varieties to identify the resistant ones to R. solanacearum.

Conclusion
The response of tomato varieties to R. solanacearum stress varies from one variety to another. Out of the 21 tomato tested varieties, Cobra 26 was the only variety identified to be moderately resistant to bacterial wilt. The variety Buffalo was susceptible while the Kêkêfô, Akikonkouin, Tounvi, Agbotrui, Petomech, Savana, Ninja, Topaze, Jaguar, Sumo, Prado, Anaya, Kiara, Euclid, Cobra 34, Adja and Tropimech varieties were highly susceptible. This is the first report revealing Cobra 26 as moderately resistant variety identified during the current study. Resistant and moderately resistant tomato varieties to bacterial wilt should, when identified, be recommended to farmers and used to improve the open-pollinated varieties preferred by rural farmers.