A Foundation for the Neural Control of Viral Fever

Abstract

Fever, an acute‐phase neurally‐regulated increase in body temperature, is thought to be an adaptive response to optimize host defenses against invading pathogens. However, fever can be an unpleasant symptom of infection and excessive fever (>41oC) can cause irreversible protein denaturation and result in tissue damage, seizure, coma and death. Common therapeutic approaches to fever include the use of non‐steroidal anti‐inflammatory drugs and cyclooxygenase inhibitors, however these approaches are ineffective for reducing prostaglandin‐independent fevers, can interfere with adaptive and innate immune defenses and are contraindicated for use in some viral infections, including SARS‐CoV2. Although research over the past couple of decades has defined the fundamental central nervous system (CNS) pathways necessary for the lipopolysaccharide model of bacterial fever, the CNS circuits mediating viral fevers are almost completely unknown. Double‐stranded RNA is an intermediary in viral replication and serves as a pathogen‐associated molecular pattern that elicits the acute phase responses to viral infection. Systemic administration of polyinosinic:polycytidylic acid (poly I:C), a synthetic double stranded RNA, mimics the acute phase responses to viral infection. Neurons in the rostral raphe pallidus area (rRPa) and the dorsomedial hypothalamus play important roles in the control of the sympathetic outflow to thermoeffectors, therefore we hypothesized that neurons in these areas would be essential for fever elicited by a model of viral fever, intravenous (iv) poly I:C. Female and male Sprague Dawley rats anesthetized with urethane and α‐chloralose (for BAT sympathetic nerve recording, n=3) or isoflurane (for cutaneous vasoconstriction experiments, n=3) received iv poly I:C (1mg/kg for experiments with sympathetic nerve recording, or 500μg/kg for experiments for cutaneous vasoconstriction). Systemically administered poly I:C increased BAT SNA (+601 ± 161% baseline), BAT temperature (TBAT; +3.1 ± 0.8°C), and expired CO2 (+0.9 ± 0.2%). Poly I:C also elicited CVC, as observed by reductions in paw or tail skin temperature (TPAW/TTail) (‐2.6 ± 1.2°C). A nanoinjection of the ionotropic glutamate receptor antagonists, AP5 and CNQX in the rRPa reversed poly I:C‐evoked increases in BAT SNA (‐596 ± 128% baseline), TBAT (‐1.3 ± 0.2°C), and expired CO2 (‐0.5 ± 0.1%). Nanoinjection of AP5 and CNQX into the DMH reversed the polyI:C‐evoked reduction in TPAW/TTail (+2.8 ± 1.1°C). These data suggest that the activity of neurons in the rRPa and the DMH are necessary for poly I:C‐induced activation of BAT thermogenesis and cutaneous vasoconstriction. These data define critical components of the neural circuits for poly I:C‐induced fever, highlight differences between the neural circuitry mediating bacterial fever and viral fever, and provide a foundation for elucidating additional brain mechanisms responsible for viral fever.