Biological Control of Potato Bacterial Wilt (Ralstonia solanacearum) Using Selected Plant Extracts

Field experiments for growing seasons A (September 2020 – January 2021) and B (February–June 2021) were performed at the University of Rwanda Experimental Farm, Busogo Campus, Musanze District, Rwanda. This region is one of the main potato production areas in Rwanda (Ndungutse et al. 2019). Climatic and edaphic conditions in the study area during seasons A and B of field experiment were collected by Meteo Rwanda. Season A was a short rain (average rainfall – 708.7 mm), while season B was a long rain (average rainfall – 816 mm). In seasons A and B, the average ambient temperature ranged from 11.2 to 22.2 °C, soil temperature 20.4 and 20.3 °C, ambient relative humidity 75.2% and 75.5%, and soil moisture content in season A was 26.3% and 26.8% in season B.

Bacterial inoculum was prepared at the Plant Pathology Laboratory of the Rwanda Agriculture Board (RAB) Northern Zone, Musanze. Bacterial inoculum used in this experiment was isolated from diseased ‘Gikungu’ potatoes. ‘Gikungu’ isolate was chosen based on the virulence test carried out earlier under a greenhouse experiment (Mutimawurugo et al. 2019) as the most virulent among other inocula tested. After isolation of the bacterium, it was identified through a vascular flow technique, cultural and morphological characteristics on Kelman’s triphenyl tetrazolium chloride (TTC), and Casamino acid–peptone–glucose (CPG) culture media (Kelman 1954; Mihovilovich 2017), as well as on biovar and gram staining. The bacterium was then purified and stored in a sterile glycerol stock at −20 °C until use.

For three weeks, the collected aerial parts of tobacco (‘Virginia’), wild marigold (‘Gan Eden’), and bulb garlic (African garlic) plants were dried under ambient room temperature. This process was supplemented with oven drying at 40 °C for two days. Completely dried materials were ground into powdery and then soaked in water or methanol at the ratio of 1 : 4 and left to shake for three days on a rotary shaker (100 rpm at 27 °C). It was followed by filtration of this mixture with Whatman No. 42 filter paper and solvent evaporation using rotar vapor at 40 °C and 280 rpm until complete drying. Later, precipitates were diluted with 50 mg·mL⁻¹ with 1% dimethyl sulfoxide (DMSO) and stored at 4 °C until use (Mwitari et al. 2013). Methanol and water extracts were chosen because they inhibited the growth of bacteria at a higher level than chloroform extracts in previous in vitro screening experiments (Mutimawurugo et al. 2020).

The biological control efficacy (BCE) of plant extracts application frequencies against potato bacterial wilt was evaluated on susceptible potato cultivar – ‘Kirundo’ under field conditions. Certified and visually healthy potato tubers were obtained from the Kinigi RAB Station. Tubers were washed in running tap water, surface sterilized by using 70% ethanol for five minutes, and after rinsing in sterile distilled water, they were left to dry for one hour under room temperature conditions. Then, the tubers were soaked overnight in methanolic or water extracts of tobacco, wild marigold, and garlic previously diluted at 50 mg·mL⁻¹ with 1% DMSO or in 1% DMSO and 2% copper oxychloride used as negative and positive controls, respectively. The following day, the tubers were left standing under shade for a day to allow the sticking of active compounds. The tubers were then inoculated by soaking them overnight in bacterial suspension at a concentration of 10⁸ CFU mL⁻¹ and incubated for 24 hours at room temperature to enhance the sticking of the bacteria prior to planting. Treated tubers were grown under field conditions for three months. Before planting, the experimental plot was double dug to a depth of 20 cm, lined with a polyethylene sheet (to reduce bacterial cells spreading in the plot soil), and filled with the top soil mixed with NPK 17 : 17 : 17 (300 kg·ha⁻¹). The field experiments were laid out in a split-split plot design and three replications. Tobacco, wild marigold, and garlic were also revealed to be the most promising plant extracts in controlling R. solanacearum among ten test plant extracts in the same experiment. The concentration of 50 mg·mL⁻¹ was selected from a pilot greenhouse experiment, which was conducted to determine the effective dose of selected plant materials against potato bacterial wilt. In season A the preceding crop was beans, in season B it was maize. In both seasons, potato plants were grown at spacing of 75 cm (inter-row) and 30 cm (intra-row).

The experimental factors used in this in vivo test were three plant materials (tobacco, wild marigold, and garlic), two solvent (methanol and water) extracts, and three application frequencies (weekly, biweekly, and monthly). Plant extracts were applied by injecting three potato haulms per plant at the third leaf from the top with 50 μL of plant extracts at a concentration of 50 mg·mL⁻¹ every week, every two weeks, and every month, from 30 to 60 days after planting. Thus, application of extracts were made five, three, and two times. A positive control (2% copper oxychloride) was applied biweekly as recommended, whereas a negative control (1% DMSO) was applied at the same frequencies as plant extracts. This study used four main plots: three plots with plant extracts (tobacco, wild marigold, and garlic) and one plot of control (positive or negative). Each main plot with plant extracts was subdivided into two subplots (water and methanol as solvents). Each subplot in turn was split into three sub-subplots (application weekly, biweekly or monthly). Therefore, in each main plot with extracts six treatments were used (two solvent extracts × three application frequencies), which resulted in 18 treatments for three main plots of plant extracts. In each treatment, under field conditions, 48 plants (six treatments × eight plants) in total, per main plot per replication. In addition, each main plot was replicated three times in the field. In the case of controls, one plot was split into two subplots (negative and positive controls). The subplot of 1% DMSO that was used as a negative control was sub-divided into three sub-subplots (weekly, biweekly, and monthly application), while in the subplot of positive control 2% copper oxychloride was only applied biweekly. Therefore, four treatments were used in this main control plot. In each treatment, eight potato plants were also grown per replication and therefore, 32 plants in total in this main plot. Every treatment was also replicated three times in the field. Briefly, 22 treatments replicated three times were used during this field experiment.

The BCE of application frequencies of plant extracts against potato bacterial wilt was evaluated from 20 to 50 days after the first plant extract injection (50–80 days after planting) with an interval of 10 days. In these field experiments (seasons A and B 2021), sampling on disease incidence (DI) was done on the four plants per each experimental treatment in each replication. The plant was considered diseased when it showed the usual symptoms of bacterial wilt.

In addition, disease incidence – DI (%) on tubers was evaluated at harvesting time by the formula of Priou et al. (2001): DI=nN×100; DI = {n \over N} \times 100; where DI – disease incidence; n – number of infected plants; and N – total number of plants assessed.

From this experiment, biological control efficacy – BCE (%) of plant extracts against bacterial wilt was calculated according to Guo et al. (2004): BCE=DIcontrol−DItreatment groupDIcontrol×100. BCE = {{\left( {{DI_{control}} - {DI_{treatment\;group}}} \right)} \over {{DI_{control}}}} \times 100.

In addition, the efficacy of extracts application and frequencies in managing tuber PHI and PHL caused by bacterial wilt in storage was determined. Therefore, 18 healthy harvested tubers from the field were picked from each treatment (from three replications). Prior to storage, surface-sterilized tubers were submerged in plant extracts or controls and bacterial suspension at the same pattern as in seed preparation for field experiments prior to planting. After that, tubers were stored under ambient room temperature in polythene bags for 37 days, during which the observation of tuber PHI and total PHL caused by R. solanacearum was evaluated. This period of 37 days of storage depended on the health of the tubers because at 37 day, most of the tubers nontreated with extracts have already rotted, and there were no more samples to be used. Application frequencies of plant extracts were also achieved weekly, biweekly, and monthly (in the same pattern as in field experiments) by soaking healthy tubers in extracts or controls. The number of infected tubers and the total number of stored tubers were used to calculate the percentage of PHI and the percentage of total PHL, as described by Rahman et al. (2012). Each treatment was stored in three bags (considered as replications). Data collection in postharvest started from 7 to 37 days of storage with an interval of one week.

In the field experiments, the mean of DI of sampled plants from each treatment was calculated and used to calculate the BCE of each treatment in three replications starting from 20 to 50 days after the first injection of plant extracts. In the field experiments, a split-split plot design was considered, while the postharvest design was a factorial and completely randomized design. During postharvest, the same average was used to calculate the percentage of PHI and total PHL caused by potato bacterial wilt after 37 days of storage. The analysis of variance (ANOVA) was carried out using SAS software version 9 (TS M0) to determine the difference in BCE of methanol and water extracts from tobacco, wild marigold, and garlic species applied every week, every two weeks and every month at 50 mg·mL⁻¹ against bacterial wilt incidence in potato plants and tubers, PHI, and PHL. The treatment means were separated using Tukey’s test (p ≤ 0.05).

The BCE of methanol and water extracts from tobacco, wild marigold, and garlic, which were applied weekly, biweekly, and monthly, was determined in seasons A and B from 20 to 50 days after extract injection (DAI). During the whole observation period, there were no interactions between all these experimental factors in terms of BCE against the pathogen (Table 1). Therefore, the effect of these independent variables on DI and BCE were tested separately. Thus, the BCE of plant extracts against bacterial wilt in potato plants was evaluated separately in seasons A and B. Along both seasons, tobacco, wild marigold, and garlic extracts significantly reduced DI, thus resulting in higher BCE as compared to copper used as positive control (Fig. 1). Tobacco extract at the end of growth reduced DI by 56.5% and 70.4%, wild marigold extract by 50.8% and 65.6%, and garlic extract by 44.8% and 56.1% in seasons A and B, respectively, compared to the positive control. BCE had the same pattern in each term. The most effective was application of tobacco extract, than marigold, garlic, and copper. No big differences in BCE were observed between tobacco and marigold extracts, but between tobacco and garlic extracts usually differences were significant. In the season A, BCE values were lower due to less infection in the field.

Figure 1.

The BCE values presented the same patterns in both seasons regarding application frequency. Always the lowest value was obtained with copper application. The DI of potato plants in terms of application frequency was reduced at the end of storage by: 54.4% and 68.9% for weekly application, 53.6% and 64.9% for biweekly application, and 43.8% and 58.4% for monthly application, in seasons A and B, respectively, which was highly significant to the BCE value in copper. Weekly application was the most effective followed with biweekly and monthly, but no significant differences were noted between weekly and biweekly application (Fig. 2).

Figure 2.

No significant differences in BCE were obtained between solvent extracts, except at 30 (season A) and 40 DAI (seasons B), where methanol extract expressed higher BCE than water extract (Fig. 3). At the end of the observation period, methanol and water extracts reduced DI in potato plants by 52.7% and 66.0% (season A) and by 48.6% and 62.1% in season B, compared to the copper control.

Figure 3.

During the observation period, there was no interactions between all these experimental factors regarding the BCE of plant extracts against potato bacterial wilt on tubers. In both seasons, there were no significant differences between treatments regarding extracts (Fig. 4). All these extracts led to lower DI and higher BCE compared to copper. DI in potato tubers was 54.6% and 63.6% for tobacco, 57.0% and 60.1% for wild marigold, and 41.2% and 56.7% for garlic in seasons A and B, respectively. The frequency of weekly and biweekly application resulted in reduction in DI in tubers and a higher BCE value in both seasons compared to monthly application (Fig. 5). The BCE of the monthly application of extracts was several times higher than copper BCE in both seasons. Weekly application resulted in significantly higher DI values of potato tubers compared to copper – 59.9% and 70.8%, biweekly application – 56.9% and 62.3%, and monthly – 35.84% and 47.36% (Fig. 5). In season A, there was no significant difference in BCE between methanol and water extracts, while in season B, methanol extracts were more effective than water extracts, but both were significantly more effective than using copper (Fig. 6).

Figure 4.

Figure 5.

Figure 6.

Weekly, biweekly, and monthly application of methanol and water extracts from tobacco, wild marigold, and garlic were tested against bacterial PHI and PHL caused by bacterial inoculation before storage. In the seasons, there were no significant differences in PHI and PHL between treatments with different plant extracts. Application of plant extracts resulted in a lower PHI and PHL rates in tubers compared with treatments with copper and DMSO (Fig. 7). Plant extracts reduced the percentage of PHI twice in both seasons compared to copper and control. Similar proportions was obtained regarding PHL (Fig. 7).

Figure 7.

In both seasons, application frequency of plants extracts were equally effective in reducing of PHI and PHL compared with copper and in control. The weekly and biweekly extract application was more effective than monthly application and twice as effective as copper and the control (Fig. 8).

Figure 8.

The effect of solvent extracts on PHI and PHL showed no significant differences in both seasons, except for season B, in which the methanol extract was more effective (Fig. 9). Both, methanol and water extracts reduced PHI and PHL of tubers significantly compared with copper and DMSO.

Figure 9.