Transmission of West Nile and other temperate mosquito-borne viruses peaks at intermediate environmental temperatures

Abstract

WORDS)Temperature is a key driver of mosquito-borne disease because it affects the physiology and life history traits of mosquitoes and the pathogens they transmit. These trait thermal responses are typically nonlinear and vary by species. However, the impact of temperature on many important temperate mosquito-borne diseases has never been quantified, and it is unclear if thermal responses of temperate mosquito-borne diseases differ systematically from the patterns found in better-studied tropical systems. These environment-transmission relationships are critical for understanding current seasonal and geographic distributions of disease and future shifts due to climate change. We used a mechanistic, trait-based approach to characterize temperature-dependent transmission of 10 vector-pathogen pairs in an overlapping network of mosquitoes (Culex pipiens, Cx. quinquefascsiatus, Cx. tarsalis, Aedes triseriatus, and others) and viruses (West Nile virus [WNV], Eastern and Western Equine Encephalitis viruses [EEEV and WEEV], St. Louis Encephalitis virus [SLEV], Sindbis virus [SINV], and Rift Valley Fever virus[RVFV]), most of which have predominantly temperate geographic distributions worldwide. We found that several of these temperate pathogens have lower thermal limits (8.6-19.0ºC) and optima (22.7-26.0ºC) that are cooler than those of tropical pathogens (lower thermal limits:16.2-22.8ºC; optima: 25.4-29.1ºC), as expected based on geography. However, their upper thermal limits (31.9-37.8ºC) were very similar to those of tropical pathogens (31.4-34.5ºC).Together, these patterns led to wider thermal breadths (NUMºC) for several of the temperate viruses compared to most tropical pathogens (NUMºC). Among temperate viruses, thermal limits for transmission varied more and were more uncertain than the optima. While the qualitative shapes of trait thermal responses generally matched prior studies, lifespan declined monotonically at temperatures above 14ºC for the temperate Culex vectors, in contrast to 4 unimodal responses observed within this temperature range for tropical species. We validated the mechanistic models with independent data on human disease cases. The incidence of WNV across US counties responded unimodally to temperature and peaked at 23.9ºC, closely matching model predictions (optima for WNV in North American vectors ranged from 23.9-25.2ºC).Additionally, data on the month of onset of cases of WNV, SLEV, and EEEV suggested that temperature is important for determining the seasonality of transmission. The thermal responses for parasite development rate, vector competence, biting rate, and lifespan, all of which varied among species and determined thermal limits in models, are a critical need for future empirical work. Because temperature effects on vector-borne disease transmission are pervasive and nonlinear, and climate regimes are changing rapidly, trait-based models are a key tool for predicting how temperature will affect mosquito-borne disease dynamics. 5

Thermal biology of mosquito‐borne disease

“…Finally, for West Nile virus, previous models predicted thermal optima up to 11°C higher than our predicted optima of 24-25°C (Table 3) (Kushmaro et al 2015;Paull et al 2017;Vogels et al 2017), which matched the unimodal thermal response of human neuroinvasive West Nile incidence that also peaked at 24°C ( Fig. S3; Shocket et al 2019).…”

“…quinquefasciatus and WNV, EEEV, WEEV, SLEV, and SINV), had R 0 thermal limits up to 32-35°C, similar to those of tropical diseases ( Fig. 2 ;Shocket et al 2019). Finally, the temperature response of R 0particularly the lower thermal limit and optimumvaried across multiple pathogens transmitted by a single mosquito species (e.g.…”

“…Colors designate different vectorpathogen systems, ordered by thermal optima for R 0 in the both panels. Abbreviations for all vectors and parasites are given in Table S1. traits, resulting in 16 temperature-dependent R 0 models (in some cases, we estimated separate R 0 models for different pathogens in the same vector or different vectors for the same pathogen) (Mordecai et al 2013(Mordecai et al , 2017Johnson et al 2015;Shocket et al 2018Shocket et al , 2019Tesla et al 2018). With these thermal performance curves, we compare temperature responses across vector-pathogen systems and estimate potential effects of climate change on disease transmission.…”

“…For WNV in Cx. pipiens, vector competence determined the lower limit (Shocket et al 2019). For WNV in Cx.…”

Can hot temperatures limit disease transmission? A test of mechanisms in a zooplankton–fungus system

“…Additionally, there is evidence for upper thermal constraints on disease in natural populations of the crab parasite (Gehman et al, ), mosquito‐borne pathogens (e.g. ) and fungi infecting grasshoppers (Carruthers, Larkin, Firstencel, & Feng, ), amphibians (Berger et al, ; Raffel, Michel, Sites, & Rohr, ) and bats (Langwig et al, ). However, temperature often covaries with other seasonal environmental factors, so causally linking temperature to observed patterns of disease is challenging (Altizer et al, ; Pascual & Dobson, ).…”