Abstract

In 2014, researchers in Uganda spotted signs typical bacterial leaf streak disease (Xanthomonas oryzae pv. oryzicola) in rice fields in Eastern Uganda. The disease was later confirmed to be bacterial leaf streak. In order to effectively plan for measures to manage this potentially devastating disease, it was imperative to score rice germplasm in Uganda for reaction to Xoc. Eighty four genotypes from the National Rice Improvement program were evaluated for their reaction to BLS using two Xoc isolates collected from Namulonge and Iganga. These were inoculated by the infiltration method using a needless syringe 30 days after planting. Data were collected on the streak length induced by BLS on the leaves 15 days after inoculation. The mean streak length per genotype was interpreted as; Resistant (R), 030mm. Genotypes showed signficant variability (P < 0.001) in their reaction to BLS. The Xoc isolates reacted signficantly differently (P=0.011) on the rice genotypes. For the Iganga isolate, 6 genotypes were resistant while 17 were moderately resistant. For the Namulonge isolate, 3 genotypes were resistant while 7 were moderately resistant. Three genotypes were resistant to both isolates. The observations ranged from highly resistant in Nerica1 to highly susceptible in Du 363. The resistant genotypes identified could be used as sources of genes for introgression into susceptible but agronomically desirable genotypes.

1. Introduction

Rice is both a food and cash crop in Uganda ¹. The crop is mostly grown by small scale farmers who cultivate both upland and lowland varieties with an estimated annual production of 191,000 metric tonnes ². This production is still low and like in many African countries it cannot meet the increasing local demand for rice. The average rice yield in Uganda is 2.5t/ha ² which compares poorly with China at 14t/ha ³. The low production of rice is due to many constraints which include diseases, pests like birds, changing weather patterns and soil fertility challenges ⁴.

In 2014, signs typical of bacterial leaf streak disease (Xanthomonas oryzae pv. oryzicola) were observed in rice fields in Uganda’s Eastern districts of Butaleja, Iganga and Namutumba with an incidence of 80, 40 and 30% respectively ⁵. The researchers went ahead to confirm the disease as bacterial leaf streak. In Asia, bacterial leaf streak is a serious constraint to rice production, causing yield losses of 10-20%, reaching 40 to 60% in severe cases ⁶. Bacterial leaf streak is considered so important in some parts of the world to the extent that in China and the United states, the disease is of quarantine importance ^{7, 8}. Bacterial leaf streak disease has a great potential to destroy rice and jeopardise food and income security in Uganda. Host resistance has been suggested as the most effective means of controlling the disease ^{7, 9}. In Uganda, no information was available on the reaction of the various rice varieties and lines to this disease, yet this is a prerequisite to the design of management measures. A study was therefore conducted to access Ugandan rice germplasm for reaction to bacterial leaf streak disease.

2. Methodology

The study was conducted in a screen house at the National Crops Resources Research Institute (NaCRRI) located in Namulonge; Kyadondo County, Wakiso district 30 Km North East of Kampala; at an elevation of 1,160 meters and coordinates 00°31^’30^”N32°36^’54^”E. NaCRRI climate is characterized by a bi-modal rainfall pattern averaging 1270 mm annually with a mean annual temperature of 22.2°C. The soils are sandy clay loams, with a pH of 4.9 to 5.0.

Eighty-four rice genotypes were evaluated for their reaction to two Xanthomonas oryzae pv. oryzicola (Xoc) isolates in a screen house at the National Crops Resources Research Institute (NaCRRI) in Namulonge, Wakiso district. These comprised of released commercial varieties, land races, crosses and advanced breeding lines in NARO’s rice breeding program. The Xoc isolates were collected from Namulonge (Nam Xoc) in Central Uganda and Iganga in Eastern Uganda (Iga Xoc) respectively. The rice genotypes were sown in a completely randomized design with a split plot treatment structure in two replicates. Six seeds were sown genotype at a spacing 10 cm within the rows and 10 cm between adjacent rows. The Xoc isolates were the main plot factors while the rice genotypes were the sub-plot factors. The study was repeated once.

Inoculum for the two isolates was prepared by growing bacterial cells on modified Wakimoto media (sucrose 20g/l, peptone 5 g/l, calcium nitrate 0.5 g, sodium phosphate 0.82 g/l and bactor agar 17 g/l) for 72 hours at 28°C. The bacterial cultures were then re-suspended in sterile distilled water and diluted to approx. 1 x 10⁸ c.f.u ml^-1 at an optical density of 0.35 and a wavelength of 600 nm ^{9, 10} using a spectrophotometer. This was then used to inoculate all the functioning leaves of each plant. Inoculation was done 30 days after planting using the infiltration method with a needless 10 ml syringe ^{9, 10, 11, 12}. Six plants were inoculated per genotype. Control, plants were similarly inoculated with sterile distilled water. Humid conditions were maintained around inoculated plants by constructing a chamber around the experiment using a translucent polythene sheet. This procedure was followed for Xoc BLS isolates. The experiment was repeated once.

Data were collected on the length of streaks (SL) induced by bacteria on the rice genotypes 15 days after inoculation. Measurements were taken on six randomly selected leaves per genotype using a 300 mm calibrated ruler. The reaction of plants was characterized using mean streak length per genotype in accordance to the scale used by ⁹ as; Resistant (R), 0<SL≤1mm, Moderately Resistant (MR), 1 < SL≤10mm, Moderately Susceptible (MS), 10< SL≤30 mm and Susceptible (S) SL>30mm. Mean streak length data were subjected to analysis of variance (ANOVA) using Genstat statistical software to further test for the signficance of observed differences in mean streak length between the genotypes and between the two isolates.

3. Results

Analysis of variance for mean streak length due to Xoc on the eighty four rice genotypes is given in Table 2. The mean streak length due to Xoc varied signficantly with both rice genotype (P<0.001) and Xoc isolate (P=0.011). The interaction effect of genotypes and Xoc isolates was also signficant (P<0.001).

On the rest of the genotypes, the first symptoms of BLS were observed 5 days after inoculation, manifesting as small water soaked streaks, progressing into translucent streaks expanding vertically along the leaf veins. At 7 days after inoculation, the Namulonge isolate formed yellow droplets of bacterial ooze along the streaks on varieties Du 363, Agoro, Jaribu, GSR IRLL 2015, IRBN (2014) Entry 15, IRLON (2014) Entry 38, IR Supa 1, IR Supa 6, Kafaci cross 1, Kafaci cross 14, IIRON 2 (2015) Plot 216, IURON (2015) Plot 4 and Aroma (2015) Plot 5. For the Iganga isolate, these signs were observed on the genotypes Du 363, Agoro, Jaribu, GSR IRLL 2015, IIRON (2015) Entry 26, Komboka, Aroma (2015) Plot 5, Namche 2, Moreberekan, IIRON 2 (2015) Plot 216, IIRON 2 (2015) Plot 224, IR Supa 6, IRBN (2014) Entry 15, IRLON (2014) Entry 38, Kafaci cross 14 and Kafaci cross 34 (Figure 1) 9 days after inoculation.

The mean streak length induced on the rice genotypes due to the two Xoc isolates is presented in Table 2. The mean streak length ranged from 0.5-43mm for the Iganga isolate and 0.4-76.3mm for the Iganga isolate. For both isolates, the least streak length were observed on Nerica 6 , IURON (2015) Plot 7, and Nerica 1 respectively. For both isolates, the longest streak was observed on DU 363. On the basis of streak length induced by the Iganga isolate, 16 genotypes were found to be susceptible (S), 6 resistant (R), 17 moderately resistant (MR) while 45 were moderately susceptible (MS). For the Namulonge isolate, 11 genotypes were found to be susceptible, 3 resistant, 7 moderately resistant while 63 were moderately susceptible.

Generally, mean streak length due to the Namulonge isolate was 19.8 mm which was signficantly (P = 0.011) higher than the Iganga isolate at 16.0 mm. The genotypes Agoro, Du 363, Jaribu, GSR IRLL 2015, IR Supa 6, Kafaci cross 14, IRBN (2014) Entry 15, Aroma Plot 5, IIRON 2 (2015) Plot 216 and IRLON (2014) entry 38 were found to be susceptible to both BLS isolates while Nerica 1, Nerica 6 and IURON (2015) Plot 7 were highly resistant to both isolates, producing a hypersensitive response 3 days after inoculation. The genotype IURON (2015) Plot 227, Kafaci cross 37, Namche 1 and Nerica 4 X WAC 116 were found to be moderately resistant to both BLS isolates while 42 genotypes were moderate susceptible to both (Table 3).

Twenty six genotypes produced varying reaction when separately inoculated with both isolates of the pathogen (Table 3). With the exception of Agoro, Du 363, Jaribu, GSR IRLL 2015, IR Supa 6, Kafaci cross 14, IRBN (2014) Entry 15, Aroma Plot 5, IIRON 2 (2015) Plot 216 and IRLON (2014) entry 38 which are susceptible to both isolates, all other genotypes susceptible (S) to the Iganga BLS isolate were found to be only moderately susceptible (MS) to the Namulonge isolate. The genotype IR Supa 1 which was only moderate susceptible to the Iganga isolate was found to be susceptible to the Namulonge isolate. The rest of the genotypes that exhibited moderate resistance (MR) to the Iganga isolate were moderately susceptible (MS) to the Namulonge isolate (Table 3).

4. Discussion

The analysis of variance for the reaction of rice genotypes to BLS revealed signficant (P<0.001) genotype and isolate (P=0.011) effects. The mean streak length induced by the Iganga BLS isolate on the rice leaves was 16.0 mm. This was signficantly (P=0.011) lower that the mean streak length induced by the Namulonge isolate (19.8 mm). ANOVA further revealed significant (P<0.001) interactions between rice genotypes and Xoc isolates. The reaction of rice genotypes to the two Xoc isolates was found to be normally distributed, ranging from highly resistant to highly susceptible with majority of the genotypes being moderately susceptible. Indeed, several authors have documented this general trend in the reaction of rice genotypes to BLS ^{7, 9, 13, 14, 15}. This reaction of genotypes to BLS suggests that resistance to the pathogen is quantitative in nature. Indeed, ⁷ and ¹⁴ suggested that resistance to BLS is polygenic. However, ¹⁶ identified a recessive BLS resistance gene bls1 in wild rice (Oryza rufipogon).

The hypersensitive reaction on the the resistant genotypes Nerica 1 and Nerica 6 is not unique. ¹⁷ while working with trangenic rice lines cloned with Rx01 observed a rapid hypersensitive reaction to Xoc. ⁹ while screenig rice genotypes for resistance to several Xoc isolates observed that the genotype FKR 14 consitently produced a hypersensitive reaction with all isolates. ¹⁵ found the rice varieties FKR19, FKR28, FKR43, NERICA 9, NERICA 12, NERICA 13 and NERICA-L-19 to be resistant to the African Xoc strains BAI6 and BAI1, producing a hypersensitive reaction (HR) within four days of inoculation. The hypersensitive reaction is associated with non-host resistance in which a plant species exhbits resistances to all genotypes in the pathogen species, often resulting into localised cell death at the site of infection ¹⁸. In this study, genotypes Nerica 1 and Nerica 6 exhbited non host resistance to the Namulonge Xoc isolate while IURON (2015) Plot 7, IURON (2015) Plot 110, E22, Rumbuka and Kafaci cross 3 exhibited non host resistance to the Iganga Xoc isolate. These genotypes can be said to be non hosts. Indeed ^{19, 20, 21} provided evidence that the hypersensitive reaction is part of non host resistance.

The fact that a genotype can react differently when seperately inoculated with two or more BLS isolates as observed in this study has also been widely reported. Whereas ¹⁵ found the rice varieties FKR19, FKR28, FKR43, NERICA 9, NERICA 12, NERICA 13 and NERICA-L-19 to produce a hypersenstive reaction when challenged with the African Xoc strains BAI6 and BAI1, these genotypes were all found to be suceptible to the Asian Xoc strains. Similarly, ⁹ found the rice genotypes FKR 14 and ITA 306 to be highly resistant to African Xoc strains but moderate susceptible and highly susceptible to the Philipine strain (BLS 256). They further observed that the genotype TN1 was highly susceptible to all Xoc strains tested and that whereas the Xoc strain BLS 256 induced large leisons on the genotype ITA, the isolates MA13 and MA11D induced small leisions on the same variety. Whereas ²² found the genotype Morebereckan to be one of the most resistant to Asian Xoc strains, this genotype was found to be moderately susceptible to the Namulonge BLS isolate and highly susceptible to the Iganga BLS isolate. In our study, this differential reaction of genotypes to the two Xoc isolates suggests that the two isolates could be genetically diverse and points to a potential large diversity in the pathogen population in the country. ^{22, 23} have indeed reported a large degree of genetic diversity among Xoc populations.

The Namulonge Xoc isolate was found to produce longer streaks than the Iganga Xoc isolate on all the rice genotypes that were susceptible to both isolates save for Kafaci cross 14, GSR IRLL (2015) and IRBN (2014) entry 15 where the Iganga isolate produced longer streaks. Long streaks are associated with high levels of susceptibility as a result of rapid multiplication of the pathogen ⁹ As obseved by ⁹ Wonni et al. (2015), genotypes with short streaks were resistant because they had the ability to curtail multiplication of bacterial cells in their tissues. Similar observations were made by ²⁴ while working with Xanthomonas campestris pv. oryzae.

5. Conclusion

The study indicated that the rice genotypes Nerica 1, Nerica 6 and IURON (2015) Plot 7 were resistant (R) to both the Namulonge and Iganga BLS isolates while the genotypes IURON (2015) Plot 227, Namche 1, Kafaci cross 37 and Nerica 4 x WAC 116 were moderately resistant (MR) to both pathogen isolates. The resistant genotypes Nerica 1, Nerica 6 and IURON (2015) Plot 7 could be used as sources of genes for introgression into susceptible but agronomically desirable genotypes. Screening of genotypes for resistance to BLS in Uganda should continue but with more geographically diverse isolates.

Acknowledgements

This reseach was conducted with financial, logistical and technical support from the Regional Universities Forum for Capacity building in Agriculture (RUFORUM), Makerere University and the National Crops Resources Research Institute (NaCRRI).

Statement of Competing Interests

The authors have no competing interests.

References