Systems biological assessment of immunity to mild versus severe COVID-19 infection in humans

COVID-19 represents a global crisis, yet major knowledge gaps remain about human immunity to SARS-CoV-2. We analyzed immune responses in 76 COVID-19 patients and 69 healthy individuals from Hong Kong and Atlanta. In PBMCs of COVID-19 patients, there was reduced expression of HLA-DR and pro-inflammatory cytokines by myeloid cells, and impaired mTOR-signaling and IFN-α production by plasmacytoid DCs. In contrast, there were enhanced plasma levels of inflammatory mediators, including EN-RAGE, TNFSF14, and oncostatin-M, which correlated with disease severity and increased bacterial products in human plasma. Single-cell transcriptomics revealed no type-I IFN, reduced HLA-DR in myeloid cells of severe patients, and transient expression of IFN-stimulated genes. This was consistent with bulk PBMC transcriptomics, and transient, low plasma IFN-α levels during infection. These results reveal mechanisms and potential therapeutic targets for COVID-19

Results

Analysis of peripheral blood leukocytes from COVID-19 patients by mass cytometry COVID-19 infected patient samples and age- and sex- matched healthy controls were obtained from two independent cohorts, 1) from the Princess Margaret Hospital at Hong Kong University, and from 2) the Hope Clinic at Emory University in Atlanta, USA. Patient characteristics and the different assays performed are shown in Table 1. We used mass cytometry to assess immune responses to SARS-CoV-2 infection in 52 COVID-19 patients, that were confirmed positive for viral RNA by PCR, and 62 age and gender-matched healthy controls distributed between the two cohorts. To characterize immune cell phenotypes in PBMCs, we used a phospho-CyTOF panel that includes 22 cell surface markers and 12 intracellular markers against an assortment of kinases and phospho-specific epitopes of signaling molecules and H3K27ac, a marker of histone modification that drives epigenetic remodeling (12, 13) (table S1). The experimental strategy is described in Fig. 1A. The phospho-CyTOF identified 12 major subtypes of innate and adaptive immune cells in both cohorts, as represented in the tSNE plots (Fig. 1B). There was a striking increase in the frequency of plasmablast and effector CD8 T cells in all infected individuals (Fig. 1B) in both cohorts, as described recently (6, 8, 14). Of note, the kinetics of the CD8 effector T cell response was prolonged and continued to increase up to day 40 post onset of the symptoms

Discussion

We used a systems biology approach to determine host immune responses to COVID-19. Mass cytometry analysis of peripheral blood leukocytes from two independent cohorts, from Hong Kong and Atlanta revealed several common features of immune responses induced upon SARS-CoV-2 infection. There was a striking and protracted increase in the frequencies of plasmablasts and effector CD8 T cells in the peripheral blood consistent with recent studies (6, 8, 14). Of note, the effector T cells continued to increase up to day 40 post symptom onset. Studies have shown that SARS-CoV-2 infection induces exhaustion and apoptosis in T cells (30, 31). Whether the continuing effector CD8 T cell response reflects continuous exposure to antigen, and if the cells are exhausted needs further investigation.

In contrast to robust activation of B and T cells, we observed a significant decrease in the frequency of pDCs. Furthermore, mTOR signaling in pDCs was reduced significantly in COVID-19 infected individuals, as measured by decreased pS6 signaling by mass cytometry. These results suggest that pDCs, the major producers of type I IFNs are impaired in COVID-19 infection, consistent with studies in SARS-CoV infection (32). To determine whether the reduced mTOR signaling in pDCs resulted in impairment of type I IFN production, we stimulated cells in vitro with TLR ligands. Our results demonstrate that pDCs from COVID-19 infected patients are functionally impaired in their capacity to produce IFN-α in response to TLR stimulation. Taken together, the data suggest that COVID-19 causes an impaired type I IFN responses in the periphery. Administration of type I IFN has been proposed as a strategy for COVID-19 intervention (33); however, it must be noted that type I IFN signaling has been shown to elevate ACE2 expression (34) in lung cells, potentially leading to enhanced infection.

In addition to the impaired IFN-α production by pDCs, there was a marked diminution of the pro-inflammatory cytokines IL-6, TNF-α and IL-1β produced by monocytes and myeloid DCs, upon TLR stimulation (Fig. 2B). This was consistent with lack of or diminished expression of the genes encoding IL-6 and TNF in the CITE-seq analysis (Fig. 5C). These results suggest an impaired innate response in blood leukocytes of patients with COVID-19. This concept was further supported by the CyTOF and flow cytometry data showing decreased HLA-DR and CD86 expression, respectively, in myeloid cells (Fig. 5, D and E, and fig. S16). In order to obtain deeper insight into the mechanisms of host immunity to SARS-CoV-2, we performed CITE-seq single cell RNA sequencing and bulk RNA sequencing analysis in COVID-19 patients at various stages of clinical severity. Our data demonstrate that SARS-CoV-2 infection results in an early wave of IFN-α in the circulation that induces an ISG signature. While the ISG signature shows a strong temporal dependence in our datasets, we also find that the ISG signature is strongly induced in patients with moderate COVID-19 infection (Fig. 4G). Consistent with this, Hadjadj, et al. (5), report an enhanced expression of ISGs in moderate patients in comparison to patients with severe or critical disease. Taken together, these data suggest that SARS-CoV-2 infection induces an early, transient type I IFN production in the lung that induces ISGs in the peripheral blood, primarily in mild or moderate patients. Additionally, we observed reduced expression of genes encoding proinflammatory cytokines, as well as HLA-DR expression in myeloid cells, which was consistent with the CyTOF and flow cytometry data showing reduced HLA-DR and CD86 expression, respectively, in myeloid cells.

Our multiplex analysis of plasma cytokines revealed enhanced levels of several proinflammatory cytokines as observed previously (35), and a strong association of the inflammatory mediators EN-RAGE, TNFSF14 and OSM with the clinical severity of the disease. Importantly, the expression of genes encoding both TNFSF14 and OSM were down-regulated in PBMCs from COVID-19 patients with severe disease in the analysis of CITE-seq data (Fig. 5C), suggesting a tissue origin for these cytokines. The gene encoding EN-RAGE, was however, expressed at high levels in blood myeloid cells in patients with severe COVID-19 (Fig. 5, C to F), although it is also possible that EN-RAGE is also expressed in the lung. Of note, these three cytokines have been associated with lung inflammatory diseases. In particular, EN-RAGE has been shown to be expressed by CD14+ HLA-DRlo cells, the myeloid-derived suppressor cells, and is a marker of inflammation in severe sepsis (21, 25, 36) and its receptor, RAGE is highly expressed in type 1 alveolar cells in the lung (24). Strikingly, we observed that the classical monocytes and myeloid cells from severe COVID-19 patients in the single-cell RNAseq data expressed high levels of S100A12, the gene encoding EN-RAGE, but not the typical inflammatory molecules IL-6 and TNF-α. These data suggest that the pro-inflammatory cytokines observed in plasma likely originate from the cells in lung tissue rather than from peripheral blood cells. Taken together with the mass cytometry data, the plasma cytokine data may be utilized to construct an immunological profile that discriminates between severe versus moderate COVID-19 disease (fig. S20).

In summary, these results suggest that SARS-CoV-2 infection results in a spatial dichotomy in the innate immune response, characterized by suppression of peripheral innate immunity, in the face of proinflammatory responses reported in the lung (37). Furthermore, there is a temporal shift in the cytokine response from an early but transient type 1 IFN response to a proinflammatory response during the later and more severe stages, similar to that observed with other diseases such as influenza (38). Strikingly, there were enhanced levels of bacterial DNA and LPS in the plasma, which were positively correlated with the plasma levels of EN-RAGE, TNFSF14, OSM as well as IL-6, suggesting a role for bacterial products, perhaps of lung origin, in augmenting the production of inflammatory cytokines in severe COVID-19. The biological consequence of the impaired innate response in peripheral blood is unknown but may reflect a homeostatic mechanism to prevent rampant systemic hyperactivation, in the face of tissue inflammation. Finally, these results highlight molecules such as EN-RAGE or TNFSF14, and their receptors, which could represent attractive therapeutic target against COVID-19.

Reference & source information: https://science.sciencemag.org/