The sickness behaviour and CNS inflammatory mediator profile induced by systemic challenge of mice with synthetic double-stranded RNA (poly I:C)

Abstract

Poly inosinic:poly cytidylic acid (poly I:C) is a synthetic double-stranded RNA and is a ligand for the Toll like receptor-3. This receptor is involved in the innate immune response to viral infection and poly I:C has been used to mimic the acute phase of a viral infection. The effects of TLR3 activation on brain function have not been widely studied. In the current study we investigate the spectrum of sickness behavioural changes induced by poly I:C in C57BL/6 mice and the CNS expression of inflammatory mediators that may underlie this. Poly I:C, at doses of 2, 6 and 12 mg/kg, induced a dose–responsive sickness behaviour, decreasing locomotor activity, burrowing and body weight, and caused a mild hyperthermia at 6 h. The 12 mg/kg dose caused significant hypothermia at later times. The Remo400 remote Telemetry system proved a sensitive measure of this biphasic temperature response. The behavioural responses to poly I:C were not significantly blunted upon a second poly I:C challenge either 1 or 3 weeks later. Plasma concentrations of IL-6, TNF-α and IFN-β were markedly elevated and IL-1β was also detectable. Cytokine synthesis within the CNS, as determined by quantitative PCR, was dominated by IL-6, with lesser inductions of IL-1β, TNF-α and IFN-β and there was a clear activation of cyclooxygenase-2 at the brain endothelium. These findings demonstrate clear CNS effects of peripheral TLR3 stimulation and will be useful in studying aspects of the effects of systemic viral infection on brain function in both normal and pathological situations.

Introduction

Poly I:C is a synthetic double-stranded RNA that has been identified as a ligand for the Toll like receptor-3 (TLR3; Alexopoulou et al., 2001). Double-stranded RNA (dsRNA) is a product of viral replication and thus TLR3 is involved in the innate immune response to viral infection (Schulz et al., 2005). Most researchers use the synthetic dsRNA analog poly I:C to stimulate TLR3 since this high molecular weight compound is more potent than the smaller viral dsRNA fragments isolated from viral preparations (Fang et al., 1999). However it is clear that dsRNA detection only represents one part of the innate immune response to viral infection (Schroder and Bowie, 2005), and that there are other pathways through which cells respond to viral infection (Coccia et al., 2004, Edelmann et al., 2004, Li et al., 2005).

A number of researchers have used poly I:C to mimic the acute phase of a viral infection (Guha-Thakurta and Majde, 1997, Traynor et al., 2004) or as a model of chronic fatigue syndrome (Katafuchi et al., 2003, Katafuchi et al., 2005). Stimulation of TLR3 using dsRNA or poly I:C induces type one interferons α and β and many of the acute symptoms of viral infection have been attributed to these cytokines (see Majde, 2000 for review). Whether or not TLR3 activation is an essential part of the anti-viral response, the essential nature of Type I interferons in combating viral infection is undisputed (Muller et al., 1994) and these mediators have been shown to produce a spectrum of behavioural changes (Segall and Crnic, 1990, Crnic and Segall, 1992a, Crnic and Segall, 1992b). However, much like the TLR4 ligand LPS, stimulation of the relevant TLR is likely to produce a wider spectrum of behavioural changes than would any single cytokine induced as a result of its stimulation. To date a limited number of studies of poly I:C induced sickness behaviour have been conducted. It is clear that poly I:C does cause some symptoms of sickness: poly I:C has been shown to induce a fever in rats that is at least partially dependent on IL-1β (Fortier et al., 2004) and mouse studies have reported a biphasic temperature response accompanied by hypoactivity (Traynor et al., 2004). We know quite a lot about how sickness is induced by lipopolysaccharide (LPS) in mice (see Dantzer, 2004 for review) and many of the cytokines implicated in sickness behaviour responses to LPS have also been shown to be induced in vivo after systemic administration of poly I:C (Guha-Thakurta and Majde, 1997, Traynor et al., 2004). Evidence from in vitro studies suggest that a variety of different cell types can respond to poly I:C to synthesize these cytokines (Doyle et al., 2003, Olson and Miller, 2004, Re and Strominger, 2004, Sivori et al., 2004, Scumpia et al., 2005). However, the full spectrum of the sickness behaviour response after systemic poly I:C challenges has not been examined and the CNS expression of cytokines in response to these challenges remains virtually unexplored. A fuller understanding of the CNS consequences of systemic viral infection is desirable both in the study of sickness behaviour mechanisms per se and for the examination of interactions between viral infection and chronic neurodegeneration.

In the current study, we aimed to characterize the sickness behavioural and peripheral and CNS cytokine responses to a range of doses of poly I:C administered intraperitoneally (i.p.) in C57BL/6 mice. The sickness behaviour spectrum examined comprised core-body temperature, hypoactivity and anorexia as well as decreased performance of species-typical behaviours such as burrowing. We also used this experimental system to test the Remo400 remote telemetry system. This system allows remote monitoring of core body temperature using probes considerably smaller than currently available models, and capable of simultaneous monitoring of numerous animals in a single home cage. In addition we have investigated whether C57BL/6 mice show tolerance to the behavioural effects of poly I:C upon repeated challenge with this dsRNA. Aspects of the response to LPS are known to be down-regulated upon repeated LPS challenge but whether a similar phenomenon occurs with TLR3 has not, to our knowledge, been examined in vivo.

We also examined hypothalamic and hippocampal transcription of inflammatory genes in response to intraperitoneal poly I:C. We have assessed CNS transcription of IL-1β, IL-6, TNF-α, and IFN-β, using quantitative PCR and have examined the expression of COX-2 at the cerebral vascular endothelium by immunocytochemistry.