Human pegivirus alters brain and blood immune and transcriptomic profiles of patients with Parkinson's disease

Abstract

Parkinson's disease (PD) is a neurodegenerative disorder with both genetic and environmental factors contributing to pathogenesis. Viral infections are potential environmental triggers that influence PD pathology. Using ViroFind, an unbiased platform for whole virome sequencing, along with quantitative PCR (qPCR), we identified human pegivirus (HPgV) in 5 of 10 (50%) of PD brains, confirmed by IHC in 2 of 2 cases, suggesting an association with PD. All 14 age- and sex-matched controls were HPgV negative. HPgV-brain positive patients with PD showed increased neuropathology by Braak stage and Complexin-2 levels, while those positive in the blood had higher IGF-1 and lower pS65-ubiquitin, supporting disruption in metabolism or mitophagy in response to HPgV. RNA-Seq revealed altered immune signaling in HPgV-infected PD samples, including consistent suppression of IL-4 signaling in both the brain and blood. Longitudinal analysis of blood samples showed a genotype-dependent viral response, with HPgV titers correlating directly with IL-4 signaling in a LRRK2 genotype-dependent manner. YWHAB was a key hub gene in the LRRK2 genotypic response, which exhibited an altered relationship with immune-related factors, including NFKB1, ITPR2, and LRRK2 itself, in patients with PD who are positive for HPgV. These results suggest a role for HPgV in shaping PD pathology and highlight the complex interplay between viral infection, immunity, and neuropathogenesis.

Keywords: Genetic variation; Neurodegeneration; Neuroscience; Parkinson disease; Virology.

Figure 1. Viruses identified in PD and CT brain tissue by ViroFind.

Computed heatmap showing all viral taxa identified by ViroFind library preparation and pipeline with purple log₂ gradient scale indicating the number of raw viral reads. The frequency of each viral species from amygdala (AMG), posterior putamen (PPUT), and superior frontal cortex (SFC) from 10 individuals with PD (red) and 14 CT individuals (blue) as well as the raw mean read count on a log scale for both groups are shown. Human pegivirus, present only in individuals with PD, is outlined in green.

Figure 2. Representative IHC staining of HPgV in control and PD.

(A and B) Subcortical oligodendrocytes (black arrowheads) of 2 control individuals with no evidence of HPgV infection. (C) HPgV nuclear immunoreactivity in an oligodendrocyte (white arrow) within the subcortical white matter in a patient with PD. (D) Cytoplasmic HPgV immunostaining (white arrowhead) and nuclear HPgV immunoreactivity of an oligodendrocyte (white arrow) within the subcortical white matter in another patient with PD. Scale bar: 20 μm. HPgV, human pegivirus; PD, Parkinson’s disease.

Figure 3. Transcriptomic alterations in HPgV-infected PD brains.

(A and B) Top 25 pathways (A) and upstream regulators (B), which show enhancement (positive z scores, red) or suppression (negative z scores, blue), in HPgV^– PD brains as compared with CT (left column) or in HPgV⁺ PD brains as compared with CT (right column). Gray boxes indicate no significant pathway/regulator change between conditions. Ranked in order of absolute value, from lowest cumulative rank score to highest. Italicized pathways/regulators indicate an enhancement of the PD phenotype as determined by a more extreme z score in the HPgV⁺ as compared with HPgV^– analysis.

Figure 4. Transcriptomic alterations in HPgV-infected PD whole blood.

(A and B) Violin plots of DEGs by ANOVA (FUBP3 P_adj = 0.005; TSC22D3 P_adj = 0.012) in whole blood transcriptomes of patients with PD without and with HPgV infection when compared with matched controls. (*P = 0.05-0.01; **P = 0.009-0.001; ***P < 0.001). (C and D) Top 25 pathways (C) and upstream regulators (D), which show enhancement (positive z scores, red) or suppression (negative z scores, blue), in HPgV^– PD whole blood as compared with CT (left column) or in HPgV⁺ PD whole blood as compared with CT (right column). Ranked in order of absolute value, from lowest cumulative rank score to highest. Italicized pathways/regulators indicate an enhancement of the PD phenotype as determined by a more extreme z score in the HPgV⁺ as compared with HPgV^– analysis. Pathways/regulators that were also altered in brain tissue are marked with an asterisk.

Figure 5. HPgV persistence and transcriptomics associations in whole blood of patients with PD and controls over time.

(A) Line plot of mean HPgV titers (rPM, from 4 sequencing lanes) from all available longitudinal samples colored by diagnostic grouping (CT, blue; PD, red; SWEDD/prodromal, orange). (B) Clustered dendrogram of Euclidean distance between expression profiles (lower scores more similar) relative to HPgV titer shows genotypic separation between PD groups. (C) Heatmap of top upstream regulators that are most divergent between PD-WT and PD-LRRK2 patients, showing enhancement (positive z scores, red) or suppression (negative z scores, blue). Gray boxes indicate no significant pathway/regulator change between conditions. Ranked in order of S2N ratio, top 20 positive and negative markers are shown. Patients followed by “(C)” had a cleared HPgV infection before the final time point available. IL-4 previously identified as an upstream regulator in brain and blood is marked with an asterisk.

Figure 6. Differential gene expression and interaction network centered on YWHAB in PD-WT and PD-LRRK2 responses to HPgV infection.

(A) Venn diagram showing the pattern of genes differentiating PD-WT and PD-LRRK2 responses to HPgV infection. (B) Radially aligned gene interaction network centered on YWHAB. Blue edges highlight direct interactions with YWHAB. Select canonical pathways are overlaid to demonstrate functional relationships among the connected genes (e.g., regulation, coexpression, or protein binding).

Figure 7. YWHAB expression correlates to HPgV response in PD-WT and PD-LRRK2 in both brain and blood.

(A) Heatmap of summary z score for suppression (blue) or enrichment (red) for upstream regulators by Pearson’s correlation to YWHAB in brain and blood transcriptomes. Regulators that were previously identified as differentiating between the response to HPgV in PD-WT and PD-LRRK2 are shown to correlate with YWHAB expression in PD and Non-PD groups. Upstream regulators that were identified as enhanced by HPgV in PD-WT and repressed in PD-LRRK2 (above black line), as compared with those that were repressed by HPgV in PD-WT but enhanced in PD-LRRK2 (below black line). (B) Gene interaction network of the 7 genes that maintain the interaction network, which differentiates between the PD-WT and PD-LRRK2 response to HPgV and their relationship to “cell death of cortical neurons” (P = 4.07 × 10^–7). Nodes represent specific genes, and edges represent different types of functional interactions: Solid arrows indicate direct interaction, broken lines indicate an indirect interaction, blunted end indicates inhibition, and unended lines indicate nontargeting interactions. (C) Heatmap of group-based gene expression relative to YWHAB (left of black line) shows consistency of relationship between genes in all patient groups and tissue types, but they show disruption in relationship to HPgV in PD-WT and PD-LRRK2 from longitudinal analysis (right of black line)