Blood Transcriptomes of SARS-CoV-2-Infected Kidney Transplant Recipients Associated with Immune Insufficiency Proportionate to Severity

Abstract

Background: Among patients with COVID-19, kidney transplant recipients (KTRs) have poor outcomes compared with non-KTRs. To provide insight into management of immunosuppression during acute illness, we studied immune signatures from the peripheral blood during and after COVID-19 infection from a multicenter KTR cohort.

Methods: We ascertained clinical data by chart review. A single sample of blood was collected for transcriptome analysis. Total RNA was poly-A selected and RNA was sequenced to evaluate transcriptome changes. We also measured cytokines and chemokines of serum samples collected during acute infection.

Results: A total of 64 patients with COVID-19 in KTRs were enrolled, including 31 with acute COVID-19 (<4 weeks from diagnosis) and 33 with post-acute COVID-19 (>4 weeks postdiagnosis). In the blood transcriptome of acute cases, we identified genes in positive or negative association with COVID-19 severity scores. Functional enrichment analyses showed upregulation of neutrophil and innate immune pathways but downregulation of T cell and adaptive immune activation pathways. This finding was independent of lymphocyte count, despite reduced immunosuppressant use in most KTRs. Compared with acute cases, post-acute cases showed "normalization" of these enriched pathways after 4 weeks, suggesting recovery of adaptive immune system activation despite reinstitution of immunosuppression. Analysis of the non-KTR cohort with COVID-19 showed significant overlap with KTRs in these functions. Serum inflammatory cytokines followed an opposite trend (i.e., increased with disease severity), indicating that blood lymphocytes are not the primary source.

Conclusions: The blood transcriptome of KTRs affected by COVID-19 shows decreases in T cell and adaptive immune activation pathways during acute disease that, despite reduced immunosuppressant use, associate with severity. These pathways show recovery after acute illness.

Keywords: COVID-19; SARS-CoV-2; immune deficiency; kidney transplantation; transcriptome.

Figure 1.

Flow diagram of study design. Out of 89 consented patients with documented evidence of SARS-CoV-2 infection from two transplant centers in New York, a total of 70 KTRs provided blood RNA. Blood was collected (approximately 2.5 ml) in a PAXgene tube for RNA extraction and transcriptome quantification. After quality control (QC), a total of 64 KTR transcriptomes remained for RNA sequencing (n=31 acute cases and n=33 post-acute cases). One patient in the post-acute group was diagnosed by antecedent COVID-19 IgG testing. The severity scores of included patients during acute COVID-19 illness were evaluated (World Health Organization severity score ranging from one to seven indicating escalating severity). Significantly, DEGs were identified from the blood transcriptomes of enrollees in the following analyses: DEGs associated ordinally with severity score and DEGs between post-acute and acute patients. External non-KTR datasets were downloaded to test validity of findings. Serum proteomics data were also generated, and proteins with expression correlated with severity scores or immunosuppression were identified.

Figure 2.

Transcriptome in KTRs with acute COVID-19 shows signature of neutrophil activation and deficient adaptive immune responses with increasing severity. (A) Volcano plot shows gene expression change associated with increasing severity score. The log2 (fold change) reflects the direction and magnitude of change in expression per unit increase in severity score (see definition in Table 1). DEGs were identified with nominal P≤0.01, with positively associated genes colored in red and negatively associated genes in blue. (B) Pathways enriched in DEGs with the proportion of positively and negatively associated genes overlapping each pathway colored in red and blue correspondingly. The pathways defined from different sources are labeled on the left and distinguished by horizontal lines. GO BP: Gene Ontology Biologic Process. (C) GSEA rank plot of representative pathways, in which genes were ranked by their log2 (fold change) from positive to negative association with increasing severity score. The short ticks along the horizontal line at 0 indicate the rank of the genes in the corresponding pathways. (D) The trend in expression change of representative genes with an increasing severity score for the neutrophil degranulation and antigen processing and presentation pathways. The scale of expression value is on log2 (CPM + 1); CPM: counts per million reads. (E) On the left panel, each Venn diagram (displayed in the form of bars) shows the overlap of enriched pathways of DEGs associated with COVID-19 severity in KTRs and non-KTRs. The overlap significance was evaluated with a hypergeometric test. On the right panel, the representative, overlapped pathways were listed in the dot plot with the size of the dot indicating the number of DEGs in each pathway and the color indicating the enrichment significance.

Figure 3.

Proteomics analyses in the serum of acute patients. Association of protein expression with severity score (A), steroid dosage (B), and CNI dosage (C). In (A) to (C), on the upper panel, the volcano plot shows the P value and log2 (fold change) of all of the 92 proteins in the assay, and on the lower panel, the heatmap shows the normalized expression value of significant proteins in the association analysis, where rows correspond to proteins and columns correspond to samples. (D) Overlap of proteins significantly associated with CNI, steroid dosage, and severity score. (E) Correlation of mRNA expression in blood and corresponding protein level in serum. Bar plot showing the Spearman correlation of all of the 92 proteins in the Olink panel with severity-associated proteins highlighted in red, and the scatter shows the protein and mRNA level of IL-6. Horizontal lines indicate a nominal P value threshold of 0.05.

Figure 4.

Transcriptome in post-acute versus acute KTRs suggests recovery of adaptive immune responses after acute disease. (A) Volcano plot shows gene expression change in post-acute patients as compared with acute patients. DEGs were identified with nominal P≤0.01, with upregulated genes colored in red and downregulated genes in blue. (B) Pathways enriched in DEGs with the proportion of up- and downregulated genes colored in red and blue correspondingly. The pathways defined from different sources are labeled on the left and distinguished by horizontal lines. GO BP, Gene Ontology Biologic Process. (C) GSEA rank plot of representative pathways, in which genes were ranked by their log2 (fold change) from up differential to down differential expression. The short ticks along the horizontal line at 0 indicate the rank of the genes in the corresponding pathways. (D) Expression change of representative genes in post-acute patients as compared with acute patients. The scale of expression value is on log2(CPM + 1); CPM: counts per million reads. (E) On the left panel, each Venn diagram (displayed in the form of bars) shows the overlap of pathways enriched in DEGs associated with severity score in acute patients and pathways enriched in DEGs in post-acute patients compared with acute patients. The overlap significance was evaluated with a hypergeometric test. On the right panel, the representative overlapped pathways were listed in the dot plot with the size of the dot indicating the number of DEGs in each pathway and the color indicating the enrichment significance.