Characterizing the spatial and temporal spread of sweet potato mild mottle virus in Central Uganda

Abstract

Despite the fact that Uganda is the third biggest producer of sweet potato in the world, average yields are still very low. Viral diseases are among the major biotic causes of these low yields. Sweet potato virus disease caused by the synergistic interaction between Sweet potato chlorotic stunt virus (SPCSV) and Sweet potato feathery mottle virus (SPFMV) causes up to 90% yield losses. Sweet potato mild mottle virus (SPMMV) which is the third most important virus of sweet potato in Uganda (after SPCSV and SPFMV) is also involved in synergistic interactions with SPCSV to cause severe symptoms. However, the spread of SPMMV in time and space has not yet been characterized, which was the major objective of this study. To achieve this objective, a field experiment was laid out at Makerere University Agricultural Research Institute, Kabanyolo (MUARIK), in a completely randomized block design (CRBD) with two blocks. Each block had three plots, and each with a different level of SPMMV inoculum (level 1=4%, level 2=16% and level 3=40%). Temporal and spatial spread of SPMMV was then monitored in the plots at 10, 18, and 28 weeks after planting using NCM-ELISA protocols as described by Salazar and Jayasinghe (2001). Transmission studies were conducted in the screen house and in cages using whiteflies and mealybugs to test their efficiency in transmitting SPMMV. Whiteflies and mealybugs were reared independently on vines co-infected with SPMMV and SPCSV. After a rearing period of eight weeks, 200 whiteflies were captured and tested for SPMMV infectivity using NCM-ELISA protocols; whereas 100 whiteflies were transferred into separate cages containing Ipomoea setosa, and SPCSV-infected vines. Leaf samples were periodically tested for SPMMV after four weeks. Mealybugs were reared for one week, and then transferred onto SPCSV-infected vines. Then, leaf samples were collected at two-week interval after “inoculation feeding period” of one week.

In plots with inoculum level 1, 8%, 72% and 96% of the total plants were infected at 10, 18 and 28 weeks respectively; whereas for level 2, 46%, 81% and 97% of the total plants were infected after 10, 18, and 28 weeks respectively. For inoculum level 3, 61%, 91% and 99% of the plants were infected at 10, 18, and 28 weeks. SPMMV spread data were transformed and fitted to four growth models; - logistic, exponential, monomolecular and Gompertz. The four models were examined for goodness of fit, but the logistic model provided the best overall fit to all data. Coefficients of determination (R2) of 93.55%, 99.45% and 95.90% were obtained for plots with inoculum levels1, 2 and 3, respectively. Although SPMMV spread varied with initial inoculum level, there was no strong relationship (P=0.59) between SPMMV spread and initial inoculum level. Initial inoculum level was more critical in the first 10- 18 weeks, after which the spread became erratic. There was however a positive and strong relationship (P=0.002) between SPMMV spread with time, and between SPMMV spread and inoculum + time (P=0.001). This suggests that SPMMV spread was largely due to an interaction between initial inoculum and time. The change in rates of SPMMV spread were 0.4, 8.0 and 2.4; 3.0, 4.38 and 1.6 and 2.1, 3.75 and 0.8 for plots with inoculum level 1, 2 and 3, for 0-10, 10-18, and 18-28 weeks respectively. This shows that the rates of spread were highest between the 10th and 18th weeks. However, comparison of slopes (b-parameters) of SPMMV spread curves for the different inoculum levels showed that they were non significant, implying that SPMMV spread among plots was not significantly different.

Spatial pattern analysis using Clark and Evan’s (CE) distance based model indicated random distribution of the infected plants; (-1.37 and 7.13); (5.38 and 11.36); and (5.74 and 13.24) for plots with inoculum levels 1, 2 and 3 at 10 and 18 weeks respectively. The distribution of SPMMV infected plants at 28 weeks was regular. However, the CE values were more inclined towards the number of plants that were being infected with time than the actual pattern of SPMMV spread. This suggests that the putative vectors could be moving between plants erratically. Screen house results showed no spread of SPMMV under whiteflies alone. Cage trials showed that SPMMV virus was capable of infecting sweetpotato whiteflies as they tested positive for SPMMV, unlike the cassava whiteflies. But the virus could not be transmitted to other plants by either whiteflies or mealybugs.

In summary, the study showed that the temporal spread of SPMMV is more influenced by the time than amount of initial inoculum. The longer the sweetpotato crop is left in the field, the greater the spread of SPMMV virus if some inoculum is present and this would affect farmers who get planting materials from such old fields. The spread/ dispersal pattern of SPMMV based on the CE values would make it difficult to obtain healthy planting materials from a field which initially had some infected vines. The vector of SPMMV could not be confirmed in this study as the widely speculated whitefly was incompetent in transmitting the virus.

Key words: Inoculum, Spread, Transmission, SPMMV, NCM-ELISA.