Molecular epidemiology, evolution, and transmission dynamics of raccoon rabies virus in Connecticut

Abstract

In North America, raccoon rabies virus (RRV) is a public health concern due to its potential for rapid spread, maintenance in wildlife, and impact on human and domesticated animal health. RRV is an endemic zoonotic pathogen throughout the eastern USA. In 1991, an outbreak of RRV in Fairfield County, Connecticut, spread through the state and eventually throughout the Northeast and into Canada. Factors that contribute to, or curb, RRV transmission should be explored and quantified to guide targeted rabies control efforts, including the size and location of buffer zones of vaccinated animals. However, population dynamics and potential underlying determinants of rabies virus diversity and circulation in Connecticut have not been fully studied. In this study, we aim to (i) investigate RRV source-sink dynamics between Connecticut and surrounding states and provinces, (ii) explore the impact of the Connecticut River as a natural barrier to transmission, and (iii) characterize the genomic diversity and transmission dynamics in Connecticut. Using RRV whole-genome sequences collected from various host species between 1990 and 2020, we performed comparative genetic and Bayesian phylodynamic analyses at multiple spatial scales. We analyzed 71 whole-genome sequences from Connecticut, including 21 recent RRV specimens collected at the Connecticut Veterinary Medical Diagnostic Laboratory that we sequenced for this study. Our analyses revealed evidence of RRV incursions over the US-Canada border, including bidirectional spread between Quebec and Vermont. Additionally, we highlighted the importance of Connecticut and New York in seeding RRV transmission in eastern North America, including two introduction events from New York to Connecticut that resulted in sustained local transmission. While RRV transmission does occur across the Housatonic and Connecticut Rivers, we demonstrated the distinct presence of spatial structuring in the phylogenetic trees and characterized the directionality of RRV migration. The significantly higher mean transition rates from locations east to west of the Connecticut River, compared to west to east, may be leveraged in directing interventions to fortify these natural barriers. Ultimately, the findings of these international, regional, and state analyses can inform targeted control programs, vaccination efforts, and enhanced surveillance at borders of key viral sources and sinks.

Keywords: comparative genomics; ecological barrie; phylodynamics; phylogeography; spatial epidemiology; wildlife disease.

Figure 1.

(a) Phylogeny of 304 sequences from eastern North America. Sequences were collected from New Brunswick, Ontario, Quebec, Connecticut, Maine, New York, and Vermont. (b) Map showing statistically supported transitions by location. Line thickness corresponds to BF values, ranging from strong to decisive support.

Figure 2.

Average Markov jump counts over time, North America. Upon examination of the average Markov jump counts over time, it appears that New York is a relatively low, but consistent source of RRV into Connecticut and Vermont during at least half of the study period. On the other hand, several sources exhibit spikes of RRV transmission during a short time interval, including Quebec to Vermont, Vermont to Quebec, and Maine to New Brunswick, which may suggest potential outbreaks.

Figure 3.

Phylogeny of 221 sequences from the Northeast. A time-scaled MCC tree was generated using 221 sequenced RRV samples from Maine, New York, Vermont, and eastern and western Connecticut. All sequences east of the Connecticut River (eastern Connecticut and Maine) fall into the second main clade. While the first clade is composed entirely of sequences west of the Connecticut River (New York, Vermont, and western Connecticut), the second clade demonstrates transmission between east and west locations. The third main clade is just four sequences from New York.

Figure 4.

Inferred statistically supported migration rates and patterns, Connecticut River. (a) Map showing statistically supported transitions by location. Line thickness corresponds to BF values, ranging from substantial to decisive support. (b) Density distribution of statistically supported mean transition rates between locations shows that transitions among locations west of the river and between east-to-west locations are significantly higher than transitions west to east or within locations east of the river. (c) Statistically supported mean migration rates per MCMC step of east-to-east transitions versus west-to-west transitions.

Figure 5.

Distribution of RRV variants in Connecticut. The geographical locations of samples analyzed in the phylogeny presented in Figure 1. The size of the circles corresponds to the number of sequences in each location.

Figure 6.

Phylogenetic tree of 71 Connecticut sequences. Clades are labeled, as well as marked with black diamonds at the node.

Figure 7.

Statistically supported transition rates, Connecticut. Map showing statistically supported transitions by location in Connecticut. Line thickness corresponds to BF values, ranging from substantial to decisive support.