Neuroproteomic Analysis after SARS-CoV-2 Infection Reveals Overrepresented Neurodegeneration Pathways and Disrupted Metabolic Pathways

Abstract

Besides respiratory illness, SARS-CoV-2, the causative agent of COVID-19, leads to neurological symptoms. The molecular mechanisms leading to neuropathology after SARS-CoV-2 infection are sparsely explored. SARS-CoV-2 enters human cells via different receptors, including ACE-2, TMPRSS2, and TMEM106B. In this study, we used a human-induced pluripotent stem cell-derived neuronal model, which expresses ACE-2, TMPRSS2, TMEM106B, and other possible SARS-CoV-2 receptors, to evaluate its susceptibility to SARS-CoV-2 infection. The neurons were exposed to SARS-CoV-2, followed by RT-qPCR, immunocytochemistry, and proteomic analyses of the infected neurons. Our findings showed that SARS-CoV-2 infects neurons at a lower rate than other human cells; however, the virus could not replicate or produce infectious virions in this neuronal model. Despite the aborted SARS-CoV-2 replication, the infected neuronal nuclei showed irregular morphology compared to other human cells. Since cytokine storm is a significant effect of SARS-CoV-2 infection in COVID-19 patients, in addition to the direct neuronal infection, the neurons were treated with pre-conditioned media from SARS-CoV-2-infected lung cells, and the neuroproteomic changes were investigated. The limited SARS-CoV-2 infection in the neurons and the neurons treated with the pre-conditioned media showed changes in the neuroproteomic profile, particularly affecting mitochondrial proteins and apoptotic and metabolic pathways, which may lead to the development of neurological complications. The findings from our study uncover a possible mechanism behind SARS-CoV-2-mediated neuropathology that might contribute to the lingering effects of the virus on the human brain.

Keywords: COVID-19; NeuroCOVID; SARS-CoV-2; apoptosis; iPSC-derived human neurons and astrocytes; mass spectrometry; metabolism; neurodegeneration; proteomics.

Figure 1

SARS-CoV-2 virus infects human iPSC-derived neurons. (A) iPSCs were differentiated into cortical-like glutamatergic neurons (i³Ns) and infected on Day 21, followed by immunocytochemistry analysis on Day 22. To assess infection in immature i³Ns, infection was carried out on Day 8 and immunocytochemistry on Day 9. Finally, for RT-qPCR analysis of viral replication, i³Ns were infected on Day 21, followed by analysis on Days 22, 23, and 24. (B) i³Ns showed infection 24 h post-infection, as shown by the white arrows. Inset shows that the infected nucleus (DAPI stain) looked fragmented compared to the uninfected nucleus. (C) Day 21 i³Ns showed more SARS-CoV-2 infected cells than Day 8 i³Ns. (D) RT-qPCR analysis of the E gene in infected i³Ns showed no virus replication in the i³Ns.

Figure 2

Proteomic analysis of SARS-CoV-2 infected human iPSC-derived neurons. (A) PCA plot shows the infected i³Ns segregated and clustered away from the uninfected i³Ns. (B) Mass spectrometric analysis of i³Ns infected directly with SARS-CoV-2 showed 13 upregulated and 10 downregulated proteins compared to uninfected i³Ns. (C) Heatmap showing differentially expressed proteins in infected (3 left columns) versus uninfected (3 right columns) i³Ns. Each column represents experimental replicates. Green text indicates downregulated proteins with > 2-fold change, while red text indicates upregulated proteins with > 2-fold change. (D) Mass spectrometric analysis of i³Ns treated with pre-conditioned SARS-CoV-2 infected Calu-3 media showed 14 upregulated and 7 downregulated proteins compared to i³Ns treated with uninfected Calu-3 media. (E) Heatmap showing differentially expressed proteins in i³Ns with infected (3 left columns) versus uninfected Calu-3 media (3 right columns). Each column represents experimental replicates. Green text indicates downregulated proteins with > 2-fold change, while red text indicates upregulated proteins with > 2-fold change. (F) Overlap of the two datasets (i³Ns directly infected versus i³Ns treated with pre-conditioned media compared to their respective controls) shows 1 protein (VGF) upregulated in both datasets.

Figure 3

Analysis of upregulated proteins in infected i³Ns. (A) Combining all the up and downregulated proteins in SARS-CoV-2 infected i³Ns (direct and treated with pre-conditioned media), 29 proteins were found to be associated with the synaptosome, while 9 mitochondrial and 2 lysosomal proteins were observed to be altered. (B) Protein–protein interaction of all upregulated proteins in infected i³Ns showed some interactions related to apoptosis, neurodegeneration pathways, and chemical carcinogenesis—reactive oxygen species. (C) Changes in biological processes for upregulated proteins in infected i³Ns. * indicates the top statistically significant processes. (D) Changes in molecular functions for upregulated proteins in infected i³Ns.

Figure 4

Analysis of downregulated proteins in infected i³Ns. (A) Protein–protein interaction of all downregulated proteins in infected i³Ns showed nominal interactions. (B) Metabolic processes were the most significant pathway to show change associated with downregulated proteins. (C) Changes in biological processes for downregulated proteins in infected i³Ns. * indicates the top statistically significant processes. (D) Changes in molecular functions for downregulated proteins in infected i³Ns.