Repeated exposure to viral and bacterial danger signals in Parkin deficient mice induces inflammation but fails to trigger parkinsonism

Abstract

Experimental models of Parkinson's disease (PD) traditionally involve direct perturbations of the functioning of dopamine (DA) neurons. However, PD is increasingly recognized as multifactorial in origin and involving both cell-autonomous vulnerability factors and non-cell-autonomous triggers including inflammatory signals deriving from bacterial or viral infections. Notably, genes associated with early-onset familial PD as PRKN (Parkin) and PINK1, are ubiquitously expressed and have been implicated in immune regulation, further supporting a role for inflammation in disease etiology. In the present study, we tested the hypothesis that alternate Polyinosinic:polycytidylic acid (Poly(I:C)), mimicking viral infection, and lipopolysaccharides (LPS), mimicking bacterial infection, may induce sufficient brain inflammation to impair the DA system in Parkin-deficient mice and recapitulate some of the systemic features of PD. We developed a protocol with alternating Poly(I:C) and LPS, inducing acute peripheral inflammation in both genotypes, with consistent body weight loss and an increase of fecal lipocalin-2. Modest levels of neuroinflammation were present six months after the treatment, with higher levels of the microglial marker Iba1 in the dorsal striatum and reduced levels of the astrocyte marker GFAP in the mesencephalon, potentially indicating reduced neuronal support in the longer term. Finally, we find that this alternating Poly(I:C)/LPS treatment does not induce dopaminergic denervation or motor dysfunctions, even in Parkin KO mice. Taken together, our findings suggest that repeated short-term exposure to pathogen-derived signals is insufficient to induce DA system impairment in young adult Parkin deficient mice. Instead, induction of PD-like pathology may require a convergence of factors, including a prolonged or recurrent inflammatory burden, aging, genetic susceptibility, and potentially additional environmental or cellular stressors, that together create a pathogenic "perfect storm". Thus, in its current form, the paradigm developed in the present study should not be regarded as a fully representative model of PD. Rather, it provides proof of concept that temporally alternating inflammatory stimuli can shape non-motor symptom-like phenotypes and underscores the need for future animal model research to incorporate interacting risk factors, disease duration, and age-dependent vulnerability.