Endogenous retroviruses are dysregulated in ALS

Abstract

Amyotrophic lateral sclerosis (ALS) is a universally fatal neurodegenerative disease with no cure. Human endogenous retroviruses (HERVs) have been implicated in its pathogenesis but their relevance to ALS is not fully understood. We examined bulk RNA-seq data from almost 2,000 ALS and unaffected control samples derived from the cortex and spinal cord. Using different methods of feature selection, including differential expression analysis and machine learning, we discovered that transcription of HERV-K loci 1q22 and 8p23.1 were significantly upregulated in the spinal cord of individuals with ALS. Additionally, we identified a subset of ALS patients with upregulated HERV-K expression in the cortex and spinal cord. We also found the expression of HERV-K loci 19q11 and 8p23.1 was correlated with protein coding genes previously implicated in ALS and dysregulated in ALS patients in this study. These results clarify the association of HERV-K and ALS and highlight specific genes in the pathobiology of late-stage ALS.

Keywords: Biocomputational method; Bioinformatics; Biological sciences; Health informatics; Medical informatics; Natural sciences.

Graphical abstract

Figure 1

ALS patients vs. unaffected control differential expression analysis (DEA) volcano and scatterplots of HERV-K features in the cortex (CTX) (A) Volcano plots (left) and scatterplots (right) for all samples, (B) female samples only, and (C) male samples only in the CTX. Features that were significantly associated with biological sex (A only) and did not have sufficient expression were removed prior to the analysis. For volcano plots, log2FC cut-off is >|2| and cut-off for p value is 10E-6. Feature plot of significantly differentially expressed (qval <0.05) HERV-K encoding features with partial or full-length Env-coding features are shown in red. Darker points indicate more significant differential expression.

Figure 2

ALS patient vs. unaffected control differential expression analysis (DEA) volcano and scatterplots of HERV-K features in the spinal cord (SC) (A) Volcano plots (left) and scatterplots (right) for all samples, (B) female samples only, and (C) male samples only in the SC. Features that were significantly associated with biological sex (A only) and did not have sufficient expression were removed prior to the analysis. For volcano plots, log2FC cut-off is >|2| and cut-off for p value is 10E-6. Feature plot of significantly differentially expressed (qval <0.05) HERV-K encoding features with partial or full-length Env-coding features in red. Darker points indicate more significant differential expression.

Figure 3

Ingenuity pathway analysis (IPA) comparison cortex (CTX) and spinal cord (SC) all patients vs. controls (A) IPA results for canonical pathway analysis and (B) upstream regulator analysis using DEA results comparing all ALS patients vs. controls in the CTX (right) and SC (left). Z score values were exported from IPA and are represented from low (blue) to high (orange). Gray values indicate that there was no clear direction of change. A Z score over 2 represents significant upregulation and lower than −2 represents significant downregulation. In general, the CTX and SC were opposites in terms of patterns of regulation among canonical pathways and upstream regulators in ALS patients relative to controls.

Figure 4

Optimum proportion of samples to include in high HERV-K group cortex (CTX) and spinal cord (SC) (A) Optimization was based on maximizing significance of q-value from Mann-Whitney test and effect size (difference in medians) in ALS patients (left) and unaffected controls (right) in CTX and (B) SC. All comparisons are based on the frequency metric of overall HERV-K expression. Orange line represents the Mann-Whitney significance level, while the blue line represents the difference in effect size of the corresponding ALS patient subgroup compared to all controls (or control subgroup compared to all ALS patients). The vertical red line indicates optimal proportion of samples to theoretically maximize differences in HERV-K expression. The optimum proportions were 0.2 in the ALS group and 0.3 in the control group.

Figure 5

Ingenuity Pathway Analysis (IPA) across cortex (CTX) and spinal cord (SC) High HERV-K-expressing ALS vs. rest of ALS patients and unaffected controls (A) IPA results for canonical pathway analysis and (B) upstream regulator analysis using DEA results comparing high HERV-K-expressing ALS patients vs. all controls in the CTX (left column), high HERV-K-expressing ALS patients vs. low HERV-K-expressing ALS patients in the CTX (middle left column), high HERV-K-expressing ALS patients vs. all controls in the SC (middle right column), and high HERV-K-expressing ALS patients vs. low HERV-K-expressing ALS patients in the SC (right column). Z score values were exported from IPA and are represented from low (blue) to high (orange). Gray values indicate that there was no clear direction of change. A Z score over 2 represents significant upregulation and lower than −2 represents significant downregulation. In general, there were similar expression patterns for canonical pathways across comparisons.