Virus Vector Interactions

Aphids are among the most severe crop pests worldwide and their damage is largely mediated by virus transmission, as opposed to direct injury from sap-feeding. The vast majority (ca. 76%) of aphid-vectored viruses are transmitted in a nonpersistent manner. This means that aphids acquire the pathogen near-instantaneously upon probing an infected plant, but the virus remains stylet-borne and thus transmissible to healthy plants for a relatively brief period of time (i.e., minutes to hours). Notably, the process by which aphids identify suitable host-plants is ideal for inadvertently spreading nonpersistent viruses. Winged aphids cannot differentiate host vs. non-host plants in flight and haphazardly select plants upon which they engage in a series of shallow test probes of the epidermal cells. This behavior of repeatedly landing on and ‘tasting’ potential host-plants can result in an individual aphid sampling dozens of plants in search of a feeding site, moving virus from plant to plant in the process. Thus, vector species considered transient or non-colonists nevertheless play a central role in the epidemiology of nonpersistent viruses.

We have been studying the landscape epidemiology of four cucurbit viruses (Cucumber mosaic virus (CMV); Papaya ringspot virus - Type W (PRSV-W); Watermelon mosaic virus (WMV); and Zucchini yellow mosaic virus (ZYMV)), all of which are aphid-vectored (Fig. 1). Because of their nonpersistent mode of transmission, the majority of aphids in the Midwest can vector these viruses, albeit at varying levels of efficiency. This point is critical because only two species colonize and reproduce on the crop (green peach aphid, Myzus persicae and melon aphid, Aphis gossypii), the remainder are non-colonists, including the invasive soybean aphid, Aphis glycines, which has been implicated in CMV infection in other vegetable crops.

Figure 1. Numerous aphid species in the Midwest act as non-persistent vectors of cucurbit viruses.

Figure 2. Spatial scales over which vectors and/or viruses are predicted to move. The innermost white box is a vegetable field (‘local’) with x-marks denoting crop plants; the intermediate grey box represents the immediate ‘landscape’ (1-10 km) surrounding the focal crop field; and the outermost dotted box is the influence of regional inputs (figure not drawn to scale).

The objective of this project is to identify vector source populations and virus reservoirs contributing to epidemics at regional, landscape, and local scales (Fig. 2):

1. Regional

   Winged aphids are renowned for long-distance movements that span large geographical regions, primarily via passive dispersal on wind currents. Because of their small size and propensity for long-distance flight, aphids cannot be tracked using traditional methods such as mark-release-recapture. In collaboration with Andy Michel at Ohio State University, we are taking a population-level approach that assesses the genetic similarity of winged immigrants with various source populations, allowing us to infer movement. The question of whether aphids colonize from near or far has a major bearing on the success of management approaches that target vector sources. If vectors originate from local habitats (e.g., neighboring crop fields), for example, then disease epidemics can be predicted and managed based on crop and noncrop vegetation patterns in the immediate landscape.

2. Landscape

   We define ‘landscape’ as a 1-10 km radius surrounding a focal crop field (Fig. 3). Surprisingly few studies have quantified the impact of agricultural landscapes on plant viruses and their insect vectors. This is noteworthy because landscape ecology has been widely drawn upon to understand closely allied fields such as infectious disease risk in birds and mammals (e.g., Lyme disease) and natural enemy-pest interactions. A possible explanation for why plant disease epidemiology has not benefitted from landscape-level approaches is the widely-held notion that insect vectors such as aphids are merely long-distance migrants that function as “aerial plankton”. In other words, the perception is that aphids colonize from distant sources far outside of the local environment and thus landscape-scale analyses will offer no predictive power. In collaboration with Jeff Holland at Purdue, we are using GIS and land use data to test this assumption and uncover potentially important links between crop and noncrop habitats on viral epidemics.

3. Local

   We define ‘local’ as occurring within the crop field. In annual crops where the cultivated area is re-tilled each year, fields are unlikely to serve as a direct source of vectors due to the lack of stable and permanent vegetation. Weeds emerging from within the field, however, may very well act as virus reservoirs, leading to subsequent disease epidemics in the crop. We are quantifying the prevalence of cucurbit viruses in local weed reservoirs naturally occurring in-and-around crop fields, and evaluating their contribution to virus transmission.

Figure 3. Examples of farms embedded in (A) simple vs. (B) complex landscapes in Indiana. Both landscapes show land use based on National Land Cover Data (NLCD) within an approximately 6 km by 6 km window. Dominant land cover classes are: brown – cultivated crops; green – deciduous forest; beige – pasture and hay.