Impact of Wheat streak mosaic virus and Triticum mosaic virus Coinfection ofWheat on Transmission Rates by Wheat Curl Mites

Oliveira-Hofman, C., Wegulo, S. N., Tatineni, S., and Hein, G. L. 2015. Impact ofWheat streak mosaic virus and Triticum mosaic virus coinfection of wheat on transmission rates by wheat curl mites. Plant Dis. 99:1170-1174. Wheat streak mosaic virus (WSMV) and Triticum mosaic virus (TriMV) are transmitted by the wheat curl mite (WCM, Aceria tosichella), and coinfections of wheat by these viruses are common in the field. Previous work has shown that mite genotypes vary in their ability to transmit TriMV. However, the degree to which coinfection of wheat modifies WCM vector competence has not been studied. The objective was to determine whether mite genotypes differed in virus transmission ability when feeding on wheat coinfected by WSMV and TriMV. First, WCM genotype type 2 was used to determine virus transmission rates from mock-, WSMV-, TriMV-, and coinfected wheat plants. Transmission rates were determined by using single-mite transfers from replicated source plants. Coinfection reduced WSMV transmission by type 2 WCM from 50 to 35.6%; however, coinfection increased TriMV transmission from 43.3 to 56.8%. Mite survival on single-mite transfer test plants indicates that the reduction in WSMV transmission may result from poor mite survival when TriMV is present. In a second study, two separate colonies of WCM genotype type 1 were tested to assess the impact of coinfection on transmission. Type 1 mites did not transmit TriMV from coinfected plants but the two colonies varied in transmission rates for WSMV (20.9 to 36.5%). Even though these changes in mite transmission rates are moderate, they help explain the high relative incidence of TriMV-positive plants that are coinfected with WSMV in field observations. These findings begin to demonstrate the complicated interactions found in this mite–virus complex. The wheat curl mite (WCM, Aceria tosichella Keifer) is the only known vector ofWheat streak mosaic virus (WSMV, genus Tritimovirus, family Potyviridae), Wheat mosaic virus (WMoV, also known as High Plains virus; tentative member of the genus Emaravirus) and Triticum mosaic virus (TriMV, genus Poacevirus, family Potyviridae) (15,16,21,25). These viruses are widespread across the Great Plains of the United States and cause significant yield losses to wheat in the region (2–4,26). Kansas’s disease reports estimate that the average annual loss due to the WCM-vectored virus complex was 1% through the past 20 years (1); however, severely affected fields often have 100% yield loss. In field surveys, WSMV is the most prevalent of these viruses, followed by WMoV and TriMV (2,3). Triple infection by the viruses can occur at low rates, but co-infections are more common (2,3). Single infections in the field are most likely to be WSMV. TriMV infections appear to be dependent on WSMV because 91% of TriMV-positive samples were coinfected with WSMV (3). WSMV and TriMV exhibit synergism when they coinfect wheat through increased titers of both viruses, greater symptom expression, and increased yield loss (4,23). Field populations of WCM are made up of different biotypes, strains, and genotypes. Biotypic differences in WCM response to resistant wheat lines have been observed (7–9). Host-specific strains of the WCM were shown when mites reared on various grass hosts could not survive on wheat and vice versa (20). Mites reared on western wheatgrass (Agropyron smithii Rydb.) transmittedWSMV at significantly lower rates than mites reared on wheat. Once these mites adapted to wheat, they transmitted WSMV at rates comparable with those of colonies that were always reared on wheat (6). More recently, Skoracka et al. (19) found multiple cryptic lineages of Aceria tosichella with diverse but distinct host ranges. This shows that A. tosichella is a genetically heterogeneous species complex. Two distinct WCM genotypes were found in Australia and named as type 1 and type 2 based on nuclear and mitochondrial DNA (5). Also, using nuclear and mitochondrial DNA, Hein et al. (10) were able to separate five mite colonies, originally isolated from collections made in South Dakota (SD), Montana (MT), Texas (TX), Kansas (KS), and Nebraska (NE), into two distinct groups that corresponded genetically to the type 1 (SD, MT, TX, and KS) and type 2 (NE) mites found in Australia (5). These two mite types also correspond to two genotypes that were collected from wheat (20). The state locations where these mites were collected are not representative of the genetic diversity present in the field. Mixed populations of type 1 and type 2 were found within fields and even within wheat heads collected in NE, KS, and MT (18). These five mite colonies (10) originated from the exact colonies used in making biotype comparisons (9) and in establishing differential transmission of WMoV by WCM (14). Thus, the type 1 and 2 mite genotypes differ in relation to biotypic and virus transmission characteristics. While mite types can vary in their ability to transmit viruses, there is also variation within types. In the United States, both type 1 and type 2 mite genotypes were shown to transmit WSMV at varying rates (14). However, in Australia, only type 2 mites were able to transmit WSMV (13). Type 2 mites (NE colony) transmitted WSMV at an average rate of 43 to 68%, depending on the vector’s phenological stage (18). WMoVwas transmitted by type 1 mites (KS, TX, SD, and MT colonies) at lower rates than type 2 (NE colony) mites (14). TriMVwas transmitted by single type 2 mites at a rate of 41% but not by type 1 mites. However, type 1 mites transmitted TriMV at a lower rate (2%) when allowing continuous movement of large numbers of mites from infected to uninfected plants (12). Coinfection of wheat with WCM-transmitted viruses may affect transmission rates of individual viruses. Low WMoV transmission rates on barley increased by type 1 mites (MT colony only) when coinfected withWSMV (14). Using an unknownmite source, Seifers et al. (16) obtained their highest TriMV transmission rate (21%) when single mites were transferred from source plants coinfected with WSMV. Corresponding author: Camila Oliveira-Hofman, [email protected] Accepted for publication 16 December 2014. http://dx.doi.org/10.1094/PDIS-08-14-0868-RE © 2015 The American Phytopathological Society 1170 Plant Disease /Vol. 99 No. 8 Given the increases in transmission rates for WMoV in the presence of WSMV and the uncertainty of transmission of TriMV from coinfected plants, a better understanding of the nature of these viral coinfections and their impact on transmission and epidemiology is needed. Because coinfection of wheat with these viruses readily occurs in nature (2,3), there is a need to evaluate WSMV and TriMV transmission in the presence of other WCM-transmitted viruses. The objective of this research was to determine how coinfection of wheat by WSMV and TriMV affects individual virus transmission by the WCM. Two studies were undertaken to (i) establish whether differential transmission of TriMV and WSMV by the type 2 WCM occurred from coinfected wheat and (ii) determine whether TriMVWSMV coinfection enhanced the ability of type 1 WCM to transmit TriMV. Materials and Methods Mite colony maintenance. Established aviruliferous colonies of type 1 (MT and SD) and type 2 (NE) WCMwere used. Mite colonies were maintained under artificial lights (cycle of 14 h of light and 10 h of darkness) in either a growth chamber or a colony roommaintained at approximately 22°C. Colonies were maintained on ‘Millennium’ wheat grown in 15-cm-diameter pots by regularly (approximately every 3 weeks) transferring mites to new wheat plants. Cylindrical cages were placed over each pot to prevent contamination. These cages contained two vents on opposite sides and an open top, all covered with Nytex screen (80-micron mesh opening; BioQuip Products). Transmission by type 2 WCM. This experiment compared the transmission of WSMV and TriMV by individual type 2 WCM from singleand coinfected wheat. Millenniumwheat was seeded in 4-cmdiameter cone-tainers (Stuewe & Sons Inc.) filled with autoclaved greenhouse soil. Plastic cylindrical cages (5 cm in diameter and 50 cm in height) with two to three Nytex vents were used to cover the cone-tainer plants. Three source plants (replicates) for each of four treatments were inoculated with sterilized water (mock), TriMV, WSMV, or WSMV + TriMV at 21 days after seeding. Crude sap of a 1:10 (wt/vol) ratio of infected tissue in sterilized water was extracted for each virus with a mortar and pestle. For single inoculations, 10 ml of the crude sap was combined with 10 ml of sterilized water. For coinoculations, 10 ml of each virus crude sap were combined. Thus, all inocula resulted in a 1:20 dilution. Plants to be inoculated were sprinkled with carborundum to allow scarring of the plant tissue and initiation of virus infection. Rub inoculation was performed by dipping the pestle in the inoculum and gently rubbing the entire length of the exposed leaves. Within a week after inoculation, 10 WCMwere placed on a pointmount triangle (card stock material, 11-mm height by 3-mm base) and carefully placed into the leaf axil of the newest leaf on each source plant. A mite transfer tool, made from a wood dowel with a single human eyelash attached, was used for mite transfers. Plants were then placed in a growth chamber (cycle of 14 h of light and 10 h of darkness) maintained at 27°C for 2 weeks. After 2 weeks, single mites were transferred from each source plant to each of ten 14-day old test plants (twoto three-leaf stage). Source plants were cut and viewed under the microscope and mites were picked up with the transfer tool. A test plant was placed on an adjacent microscope and one mite was transferred directly to the whorl of the newest leaf. Only large (adult or late nymph) mites exhibiting normal movement were transferred to test plants. Test plants were immediately covered with cages and left overnight to allow mite establishment. Test plants were then transferred to a growth chamber held at 27°C (cycle of 14 h of light and 10 h of darkness). After the single-mite transfers, approximately 0.15 to 0.2 g of plant tissue from each source plant was stored at −20°C and later tested for WSMV and TriMV via double-antibody sandwich enzymelinked immunosorbent assay (DAS-ELISA). Only test plants from sources testing positive for the respective virus treatment were included in the statistical analysis. Single-mite transfer test plants were harvested 21 to 24 days after infestation. Mite survival was determined by presence or absence for each of the test plants. At harvest, leaf pieces from test plants were sampled and stored at −20°C until assayed for WSMV and TriMV via DAS-ELISA. Because of the extensive labor involved in mite transfers, the number of replicates for each run was limited to 3; however, this was conducted four times for a total of 12 source plants (replicates) for each treatment. Type 1 versus type 2 transmission. The objective of this experiment was to determine whether WSMV + TriMV coinfection would influence virus transmission rates for type 1 WCM. Type 2 WCM was included as a comparison. Cone-tainer planting, inoculation procedures, and mite transfers were performed as described in the previous section. This experiment was conducted two times. The first run included three source plants each for a mock and WSMV + TriMV treatment for each of three WCM colonies tested: type 2 (NE) and type 1 (MT and SD). For the second run, the mock was eliminated to enable testing of six coinfected source plants. For each virus treatment, mites were transferred individually from each source plant to 10 separate test plants. For the mock treatment used in the first run, only five single-mite transfers were made for each source plant. Nine WSMV + TriMV coinfected source plants (90 test plants) were used for each colony, except the MT colony that only had 8 source plants (80 test plants) because one source plant tested negative for WSMV. Test plants were harvested 21 days after single-mite transfers in the first run. Due to advanced symptom development in the second run, test plants were harvested only 14 days post WCM transfers. Plants were cut at the soil level and inspected for mite survival. Leaf tissue (approximately 0.15 to 0.2 g) for each test plant was placed into a mesh bag and stored at −20°C until DAS-ELISA testing forWSMV and TriMV. Virus assay. DAS-ELISA for WSMV and TriMV was performed for all test plants. For each sample, approximately 0.15 to 0.2 g of plant tissue was added to a mesh bag (Agdia, Inc.). General extraction buffer (GEB) was added to the mesh bags at a 1:10 (wt/vol) ratio and then tissue was ground using a tissue homogenizer (Agdia Inc.). WSMV and TriMV tests were performed simultaneously, and leaf extract from the samemesh bag was used for both tests. ELISA plates (96-well Flat-Bottom Immuno Plate; Maxisorp, Nunc, Thermo Scientific Inc.) were coated with TriMV immunoglobulin G (IgG; 24) at 100 ml/well and 1:1000 (vol/vol) or WSMV capture antibody (Agdia Inc.) at 1:400 (vol/vol) in carbonate buffer, and stored overnight at 4°C. The following morning, plates were rinsed three times with phosphate-buffered saline with Tween (PBST). Extract (100 ml) of each sample was added to each of two wells of the WSMVand TriMV-IgG-coated plates and incubated for 1 h at 37°C. Plates were washed with 1× PBST. Rabbit anti-WSMV or TriMV IgG-ALP conjugate antibody diluted in GEB (100 ml) was added to the plates at 1:400 (vol/vol) for WSMV and 1:500 (vol/vol) for TriMV, and incubated for 1 h at 37°C. Plates were washed with 1× PBST. p-Nitrophenyl phosphate (100ml of 1 mg/ml) in 0.1M diethanolamine buffer, pH 9.8, was added to each well, and plates were incubated at room temperature in the dark for at least 1 h. Absorbance estimates at 405 nm were obtained with a Multiskan FC Spectrophotometer (Thermo Scientific Inc.). A sample was considered positive if absorbance value was at least two times higher than that of negative controls (buffer and healthy extract) (16,17). Data analysis. Data were analyzed using direct comparisons of transmission rates between WSMV and WSMV + TriMV and between TriMV and WSMV + TriMV treatments. PROC GLIMMIX (SAS, v. 9.3; SAS Institute Inc.) was used, specifying a binomial distribution for WSMV and TriMV presence because a plant was either positive or negative for each virus. Type III tests of analysis of variance and least significant differences for virus presence were used to generate differences between WSMV and TriMV transmission in singleand coinfected treatments; source plants were treated as random effects. Interactions of mite survival and treatment response were also tested by including survival–treatment interaction as a response in the model statement. Separate analyses of virus presence were performed for plants with surviving mites and for plants with no surviving mites. Treatment effects and interactions at P # 0.05 Plant Disease /August 2015 1171 were considered significant. Odds ratios were calculated by using PROC GLIMMIX (SAS, v. 9.3; SAS Institute Inc.) and compared the relative odds of WSMV transmission given its coinfection with TriMV, and TriMV transmission given its coinfection with WSMV.