Spatial Epidemiology

Our West Nile virus (WNV) project began with research studying the spatial patterns of transmission across the landscape in Illinois. In 2002, Illinois led the nation in the number of human cases of WNV and the distribution of sick people were not evenly distributed. We found significant clusters of WNV in humans in certain areas in suburban Chicago (Figure 1).

WNV clusters
Figure 1. Distribution of clusters of human WNV cases in the Chicago region, 2002.

Further investigations into spatial epidemiology revealed particular landscape features that were associated with human cases of WNV. The residential locations where most of the humans cases occurred were developed in the 1940-1960 era, with moderate vegetation and population density. We predict that this type of landscape has the correct infrastructure to allow Culex pipiens mosquitoes to proliferate and the correct vegetation structure to sustain high densities of particular bird species. This recipe allows amplification of WNV between birds and mosquitoes and eventual spill-over into humans.

We recently completed a study investigating the influence of weather patterns on WNV transmission. The results indicate that temperature has the largest influence on WNV infection in Culex pipiens mosquitoes where hot temperature leads to high amounts of infection in mosquitoes. In general, lower precipitation promoted WNV activity in mosquitoes but the patterns were not consistent.


Transmission patterns

Our primary activities in the field involves capturing mosquitoes using a variety of traps and birds using mist-nets (Figure 2 and 3). Mosquitoes and bird blood are tested at Michigan State University for the presence of WNV and antibodies to WNV.

mosquito traps
Figure 2. Mosquito collection techniques; light traps, gravid traps, and a backpack aspirator.


bird work
Figure 3. Bird field work consisting of mist-netting to catch birds and processing birds which involves placing an aluminum leg band on the birds, taking a blood sample, and then releasing the bird.

The field epidemiological data collection allows us insight into the seasonal transmission patterns of WNV, which is characterized by a late-summer amplification (Figure 4). By testing bird blood for the presence of antibodies to WNV and for the virus, we were able to conclude that nestling birds are likely not involved in WNV transmission but fledgling birds (juvenile) are very important for the amplification process.

Figure 4. Culex spp. mosquito infection rate in our southwest suburban Chicago study region, 2005-2010.


Mosquito Host Selection

The mosquito blood meal analysis identifies the species of bird or mammal that a mosquito previously fed on (Figure 5). This molecular tool allows us to understand mosquito feeding patterns which helps implicate which species of mosquitoes are important for maintaining the bird-mosquito cycle of transmission and which ones 'bridge' WNV into humans. We have found evidence that Culex pipiens, the northern house mosquito, is perhaps responsible for both roles.

Blood meal analysis
Figure 5. Diagram showing steps in the mosquito blood meal analysis.

We also take this analysis a step further by encorporating bird availability to mosquitoes with the feeding patterns to identify if mosquitoes have a 'preference' or 'avoidance' of particular species. The only preferred species of bird by Culex pipiens was American robin while several species were avoided, such as house sparrow, common grackle, and European starling.

Finally, we combined information on the relative abundances of birds, the ability of birds to be a host of WNV, and mosquito feeding preferences to rank the contribution of different bird species to WNV transmission. We determined the most important bird for WNV transmission in our study region is the American robin, followed by blue jay, and then house finch.

The projects revolving around the blood meal analysis have been summarized in an article published in The SPOTLIGHT.



Avian Movement and Roosting Behavior

Given the importance of the American robin to WNV transmission, we are beginning a new project using radio telemetry to track robin movement and roosting behavior. This technique is achieved by mounting a small transmitter weighing less than 2 grams on the bird's tail feathers (Figure 6 and 7).


AMRO telemetry
Figure 6. Researchers Gabe Hamer and Bethany Krebs process an American robin at night in a phragmites marsh, the location of a large communal robin roost.

Figure 7. Graduate student Bethany Krebs mounts a transmitter on the tail feathers of an American robin.

The transmitter emits a signal and we use antennas to track the bird's daytime and nightime locations (Figure 8). When the robin molts its tail feathers, the transmitter falls off.

Figure 8. Map showing the daytime and nightime location of an American robin in our study region.

Mosquito Mark-Capture

The goal of this project will be to study dispersal and movement patterns of the primary WNV vector, Culex pipiens, the northern house mosquito.  This research complements our other field activities understanding avian host distribution and movement patterns and together, will give us unprecedented insights into the transmission ecology of this emerging infectious disease.

This mark-capture study will utilize breeding site labeling, which means naturally breeding mosquitoes will undergo amendment with stable isotopes.  The primary larval habitat for Culex spp. mosquitoes in the urban environment is catch basins, which is where we plan to perform the study (Figure 9).  We plan to amend catch basins with 15N-enriched potassium nitrate and 13C-enriched glucose which will allow the stable isotope to become fixed into structural body tissue in the mosquito.  The mosquitoes will emerge from catch basins as normal and adults will be captured using several types of traps in the study region.  Adult mosquitoes will be submitted to the UC-Davis Stable Isotope Facility for the analysis using isotope ratio mass spectrometry.

Figure 9. Culex pipiens larvae.


Molecular epidemiology

Once WNV has been identified in birds or mosqitoes, we are able to use molecular techniques to characterize the molecular variation of the virus. The results of this effort allow us to identify differences in the virus across space and time. The viral genetic diversity patterns demonstrate that transmission processes occur at fine spatial scales and that urbanization influences this process.