Reviewed by Martinez-Delgado, G as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn School of Medicine, Mount Sinai.
Targeted Immunosuppression Distinguishes COVID-19 from Influenza in Moderate and Severe Disease
Mudd et al.; medRxiv \[10.1101/2020.05.28.20115667\]
Despite similarities in presentation at onset, differences in the underlying immunopathology of SARS-CoV-2 infections and other respiratory infections, like influenza, remain largely unknown. In this pre-print, Mudd et al. performed single-cell RNA sequencing (scRNAseq) of PBMCs from COVID-19 patients and influenza patients, in order to delineate potential key differences between the aforementioned respiratory infections. Analyses were performed using a cohort of 79 COVID-19 patients (n=79; 35 of whom developed acute respiratory failure), 26 influenza patients (n=26; 7 of whom developed acute respiratory failure), and 15 healthy controls (n=15
First, plasma cytokine levels were evaluated in the 79 COVID-19 patients, 26 influenza patients, and 8 of the healthy controls. Cytokine analyses identified a reduced production of GM-CSF, IFN-ɣ. and IL-9 but a significant elevation of IL-6 and IL-8 across all COVID-19 patients, compared to influenza patients. In fact, certain chemokines and others were more up-regulated in influenza patients, as opposed to COVID-19 patients. The authors subsequently performed a computational assessment of whether certain groups (or modules) of cytokines were predictive of one of two infections.
Interestingly, two modules, containing G-CSF, IFN-ɣ, IL-2R, IL-6, IL-8, and MCP-1 (among several others), were inversely correlated with an increased likelihood of being SARS-CoV-2-positive. The authors observe that while higher generalized inflammation is characteristic of influenza patients, COVID-19 patients exhibit a marked elevation of a distinct subset of cytokines. Using intubation status and expiration as end-markers of disease severity, the authors found that IL1-RA and IL-6 were associated with COVID-19 disease severity and predictive of poor outcome, both with and without comparison to influenza patients. Collectively, these results suggest that only a selection of inflammatory cytokines are predictive of disease severity, while cytokine storm syndrome is not necessarily descriptive of all COVID-19 patients; these characterizations distinguish COVID-19 immunopathology from that of influenza.
PBMCs had been collected from 79 COVID-19 patients (n=79; 35 of whom developed acute respiratory failure), 26 influenza patients (n=26; 7 of whom developed acute respiratory failure), and 15 healthy controls (n=15). A comparison of the peripheral immune landscape identified several primary differences. Though both groups of patients exhibited pan-lymphopenia, generally, COVID-19 patients had more antibody-secreting plasmablasts and activated CD4+ T cells than influenza patients or controls. However, COVID-19 patients showed significantly reduced numbers of circulating monocytes, in line with previous reports that inflammatory monocytes are recruited to the lung and reduced in the periphery in COVID-19 patients. Notably, both these peripheral monocytes and CD4+ T cells in COVID-19 patients showed reduced HLA-DR expression, indicative of reduced activation.
A closer interrogation of potential immuno-regulatory cell types (as a compensatory response to the hyper-inflammation observed in COVID-19 patients) via scRNAseq (of 3 COVID-19 patients \[n=3\], 3 influenza patients \[n=3\], and 1 healthy control) revealed a significantly suppressed type I interferon (IFN) response among B cells, CD8+ T cells, regulatory T cells, plasmacytoid dendritic cells (pDCs), and especially among monocytes. In contrast, his pathway and its associated downstream cascades were enriched in influenza patients. Notably, pathways enriched in COVID-19 patients were glucocorticoid and metabolic stress pathways across multiple cell types, but most significantly in monocytes.
The limited patient sample size of the scRNAseq analysis should be noted.
Without additional clinical information, it is difficult to know whether relative time-points (at which blood samples were collected and cytokine analyses were performed) may have been different, so an analysis of patients at different stages of their disease course may be a confounding factor. Indeed, the authors make some reference to this potential limitation, in addition to age, in their linear regression models.
In addition, the authors use HLA-DR expression to evaluate myeloid cell activation; other markers should be used to validate the observation of reduced HLA-DR expression. This reduced activation phenotype, in combination with the fewer number of monocytes in the periphery and down-regulated IFN response, provides the basis for the authors’ conclusion that an overall suppressed monocyte response underscores COVID-19 immunopathology, when compared to the immune profile of influenza patients. However, it is important to consider the recruitment of the inflammatory subset of monocytes to the lung or other extrapulmonary organs as a reason for the reduced number in the vasculature.
Through a much needed comparison, Mudd et al. provide a closer look at the cellular differences between the immune response to COVID-19 and influenza. Using scRNAseq, the authors identify notable changes in monocyte transcriptional activity and number and in cytokine profiles that suggest potential associations with disease severity of COVID-19, but not influenza. In particular, the identification of a glucocorticoid response in monocytes is worth further investigation, given previous claims towards the use of immunosuppressive agents to treat COVID-19.
This review was undertaken by Matthew D. Park as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.
Sex differences in immune responses to SARS-CoV-2 that underlie disease outcomes
Takahashi et al.; medRxiv \[10.1101/2020.06.06.20123414\]
Sex-based differences in the immune response have been reported for various types of infections. There is a growing body of epidemiological evidence that supports the finding that men experience more severe COVID-19 disease than women do, but the immune mechanisms underscoring such a difference remain unknown.
Here, Takahashi et al. analyze PBMCs, plasma, and nasopharyngeal swabs or saliva from 93 mild-to-moderate COVID-19 patients (n=93), comprised of 48 women (n=48) and 45 men (n=45), to characterize potential sex-based differences in the immune response to SARS-CoV-2 infection. It is important to note that patients on hydroxychloroquine and Remdesivir were not excluded from a sub-cohort of patients (n=39) evaluated as baseline measures for untampered immune responses to SARS-CoV-2 (these patients were not treated prior to first sample collection). In a second sub-cohort, 54 patients were assessed longitudinally for an undisclosed amount of time. Samples from uninfected healthcare workers were used as controls.
Viral Load (nasopharyngeal or saliva samples)
No significant differences were identified between male and female patients. Still, median viral RNA was higher in male patients at first sample collection and generally throughout disease course.
Antibody production (plasma samples)
Anti-SARS-CoV-2 S1 protein-specific IgG and IgM antibodies were measured in the plasma of male and female patients. Though anti-S1-IgG antibodies were higher in female patients, compared to male patients, no significant differences could be identified either in the baseline cohort or in longitudinal patients.
Cytokine analysis (plasma samples)
Among baseline patients, who had not received immunomodulatory therapy prior to sample collection (except hydroxychloroquine), type I/II/III IFN levels were not significantly different between male and female patients. However, IL-8 was significantly higher in male than in female patients. Of note, among longitudinally evaluated patients, CCL5 levels were significantly higher in male than in female patients. CXCL10 levels show a similar trend, though this was not significant.
Immune cell landscape (PBMCs)
Both male and female patients exhibited a reduction among T cells and an increase in B cells. No significant differences in T cell subtypes (naïve, central/effector memory, follicular, regulatory) were observed between male and female patients. Of note, however, female patients showed (1) a significantly greater proportion of CD38+HLA-DR+ activated CD8+ T cells and (2) a concomitant enrichment of PD-1+TIM-3+ terminally differentiated T cells, compared to male patients. Otherwise, no other significant differences were identified between male and female patients.
The authors subsequently interrogated the peripheral myeloid compartment. Female patients showed a greater increase in CD14+CD16+ intermediate monocytes than male patients, while both patients exhibited a marked increase in total monocytes, compared to the controls. However, male patients showed higher levels of CD14loCD16+ non-classical monocytes than female patients and their uninfected, healthy counterparts. The authors noted that this enrichment of non-classical monocytes was correlated with CCL5 levels only in male patients.
Clinical outcomes were tracked for both male and female patients. Clinical scoring was used to separate each group into two sub-groups: patients that had remained stable throughout hospital stay (stabilized) and patients that had worsened since the first sample collection (deteriorated). Deteriorated male patients were significantly older than stabilized male patients; there was no significant difference in age between stabilized and deteriorated female patients. In terms of BMI, both deteriorated male and female patients tended to be higher in BMI than their respective stabilized counterparts. Interestingly, anti-S1-IgG antibodies were higher in stabilized female patients than their deteriorated counterparts, though this trend was not seen with male patients. Otherwise, no other significant differences in clinical parameters were observed.
Additional comparisons between deteriorated and stabilized patients of each sex revealed that certain innate cytokine mediators (TNFSF10 and IL-15) associated with worse outcome in female patients but not in male patients. In contrast, the proportion of CD38+HLA-DR+ activated CD8+ T cells was significantly reduced in deteriorated male patients compared to their stable counterparts, but this was not true for female patients. Indeed, poor CD8+ T cell activation and IFNɣ production were both negatively correlated with age in male patients, but not in female patients.
A significant number of patients were diagnosed with underlying chronic conditions that have been previously described to associate with poorer COVID-19 outcomes or with a compromised immune system.
Approximately two-thirds of each group (men and women) were treated with tocilizumab, and nearly a sixth of each group were treated with corticosteroids. While these patients were excluded from the baseline cohort, it is unclear whether or not these patients contributed to the second cohort that was longitudinally examined.
The mean age for patients is notably higher than the mean age for the HCW control group.
Duration of hospital stay was not considered, so it is unclear how quickly certain subsets of male and female patients deteriorated. This may be a confounding variable, or at the very least, the kinetics of disease course in male and female patients is a parameter that warrants investigation.
In summary, Takahashi et al. provide the first report-to-date that delineates immunological differences between male and female patients with mild-to-moderate COVID-19 disease during the initial stages of infection. For example, male patients deteriorate due to less robust T cell-mediated antiviral immunity, compared to their female counterparts. Several of the other findings substantiate previous reports, such as those of significant neutrophil chemotaxis in the lung of COVID-19 patients (and its association with poorer prognosis). This study, therefore, provides an important platform for additional inquiries into key signaling pathways and transcriptional programs that are differentially regulated between male and female COVID-19 patients by specific cell types (i.e. intermediate and non-classical monocytes, CD38+HLA-DR+ CD8+ T cells) identified in this report. These studies, alongside others, are warranted to better tailor therapies for male and female COVID-19 patients.
This review was undertaken by Matthew D. Park as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn School of Medicine, Mount Sinai.
Title: Rapid isolation and profiling of a diverse panel of human monoclonal antibodies targeting the SARS-CoV-2 spike protein
Keywords: Coronavirus; SARS-CoV-2; SARS-CoV; COVID-19; Monoclonal antibodies; Adaptive Immunity
Zost et al. describe the methodology used to efficiently generate a large library of highly-functional monoclonal antibodies directed against the SARS-CoV-2 spike (S) protein. Several different approaches were used to select the antibodies characterized in this study. Briefly, plasma or serum was obtained from four patients infected with SARS-CoV-2, and ELISA binding assays were used to confirm the presence of reactive antibodies to the prefusion ectodomain of either the SARS-CoV-2 or SARS-CoV S protein. Additional screens were used to assess the presence of antibodies capable of binding to the receptor binding domain (RBD) as well as the entire N-terminal domain (NTD) of the SARS-CoV-2 spike protein. The highest reactivity was seen in binding assays when the antigenic targets were the SARS-CoV-2 spike S2P ectodomain or RBD. SARS-CoV-2 S-specific class-switched memory B cells were isolated from peripheral blood mononuclear cells (PBMCs) via flow cytometry. The two patients whose blood was collected at later stages of convalescence displayed higher frequencies of antigen-specific memory B cells and greater levels of neutralizing antibodies. S2P ectodomain- and RBD-specific memory B cells sorted from these two patients PBMCs were pooled and cultured for one week in wells containing a feeder layer of cells expressing CD40L, IL-21, and BAFF. Approximately 50% of these cells were single-cell sequenced for antibody gene synthesis. The other half were placed in a Berkeley Lights Beacon Optofluidic instrument to further identify, select, and export antigen-reactive B cells prior to single cell antibody sequencing and cloning into immunoglobulin expression vectors. Both approaches yielded a combined total of 386 recombinant SARS-CoV-2 reactive human monoclonal antibodies. Subsequent ELISA and neutralization assays were used to separate these antibodies into five classes based on their cross-reactivity with SARS-CoV and the specific binding domains on the SARS-CoV-2 S protein. Bioinformatic analysis of the immunoglobulin sequences revealed a high degree of relatedness to the inferred unmutated ancestor immunoglobulin genes.
This study characterizes a robust repertoire of SARS-CoV-2 spike-specific antibodies. The authors begin to shed light on the binding sites of these antibodies by describing the domains on the spike protein to which these antibodies react. However, in order to more fully capture the mechanism of neutralization for the leading therapeutic candidates, it will be important to further characterize the specific epitopes and structural binding modes. This is especially important since many of the antibodies identified in this study will not directly interfere with the RBD/ACE2 interaction and therefore likely act through another mechanism such as destabilizing the spike prefusion conformation. Another interesting observation raised in this study is that, as seen with Ebola, patients do not possess a high frequency of memory B cells expressing neutralizing antibodies until later in convalescence. However, given the small number of patients in the study, a larger sample is needed to confirm this conclusion. While this study presents a comprehensive class of candidate antibodies for therapeutic development there is still much needed data describing the protective potential of these antibodies in animal models challenged with SARS-CoV-2, as the authors assert as well. Finally, as synergy has been observed in strong B cell response for other coronaviruses and the fact that antibody cocktails are an effective treatment platform to prevent mutation escape, it would be helpful to know whether specific combinations of these monoclonal antibodies enhance neutralization and in vivo protection.
In conclusion, this study presents a robust analysis of the specific B cell response to SARS-CoV-2 in a small number of individuals, and describes practical techniques to isolating a large and diverse panel of human monoclonal antibodies. In addition to revealing potential therapeutic antibody candidates for COVID-19, the authors provide additional information as to the complicated and inconsistent observations of antibody cross-reactivity and cross-neutralization in the context of SARS-CoV and SARS-CoV-2. Information on conserved and highly potent neutralizing targets of antibody responses will be critical down the road as we evaluate the immunogenicity of vaccine candidates. Meanwhile, the information in this study can be directly applied to the therapeutic antibody pipeline for SARS-CoV-2 and the methodologies described here can be adapted for similar emerging pathogens in the future.
Title: Inhibition of SARS-CoV-2 infection (previously 2019-nCoV) by a highly potent pan-coronavirus fusion inhibitor targeting its spike protein that harbors a high capacity to mediate membrane fusion
Keywords: SARS-CoV2, S protein, fusion inhibitor, hACE2
Members of the coronavirus family rely on fusion with the host’s cell membrane during infection. The spike (S) protein plays an essential role in this step (1). This work sheds light on the particularities of the S protein from SARS-CoV2 and compares it to previously existing coronaviral S proteins. SARS-CoV2 S protein showed stronger affinity for hACE2 compared to SARS-CoV. Previous work by this group had established the potential of a previous version of the peptide (EK1) in reducing the infectivity of several coronavirus by blocking the S protein interactions with the host cell (2). In this work, authors present an improved EK1C4 version of this peptide. In vitro experiments on human cells showed EK1C4-mediated reduction of the infectivity of SARS-CoV-2 and other coronavirus. Administration of EK1C4 in prophylaxis or shortly after exposure to the HCoV-OC43 coronavirus protected newborn mice from virus-induced lethality.
This work suggests that the furin cleavage site from CoV2-S could be related to its increased infectivity compared to CoV-S. However, a previous report suggested that the enhanced cell-to-cell fusion mediated by this furin cleavage might not necessarily translate into enhanced viral infectivity (3). The mutation of this site, or inhibition of furin in 293T cells expressing SARS-CoV or CoV-2 S proteins, could inform if this mechanism is actually involved in SARS-CoV2 pathogenesis, and potentially suggest additional routes for therapeutic intervention.
The in vivo model used to address the protective potential of EK1C4 is based on the coronavirus HCoV-OC43. Further work should aim to establish the potential of EK1C4 for prevention of SARS-CoV-2 replication in more relevant animal models (e.g. Rhesus Macaques or hACE2 mice). It is, however, acknowledged, that experimentation in vitro included both SARS-CoV2 and the same HCoV-OC43 used in the in vivo infection model.
Even though the safety of the unmodified EK1 peptide in mice had been previously examined (2), the potential side effects and pharmacokinetics/dynamics of EK1C4 in the in vivo setting are not addressed in this report.
The development of prophylactic drugs against viral infection is a paramount element of epidemic control, specially in the time before a vaccine can be readily available. Similar strategies have been followed in the past in response to threats of this nature (4,5). This work’s relevance is limited by the absence of relevant animal models of SARS-CoV-2 infection. Further pharmacological analyses would also be necessary to ensure the safety of EK1C4 in animal models before a potential safety analysis in human subjects.
1. Heald-Sargent T, Gallagher T. Ready, set, fuse! the coronavirus spike protein and acquisition of fusion competence. Vol. 4, Viruses. Multidisciplinary Digital Publishing Institute (MDPI); 2012. p. 557–80.
2. Xia S, Yan L, Xu W, Agrawal AS, Algaissi A, Tseng CTK, et al. A pan-coronavirus fusion inhibitor targeting the HR1 domain of human coronavirus spike. Sci Adv. 2019 Apr 1;5(4):eaav4580.
3. Follis KE, York J, Nunberg JH. Furin cleavage of the SARS coronavirus spike glycoprotein enhances cell-cell fusion but does not affect virion entry. Virology. 2006 Jul 5;350(2):358–69.
4. Jones JC, Turpin EA, Bultmann H, Brandt CR, Schultz-Cherry S. Inhibition of Influenza Virus Infection by a Novel Antiviral Peptide That Targets Viral Attachment to Cells. J Virol. 2006 Dec 15;80(24):11960–7.
5. Ahmed A, Siman-Tov G, Hall G, Bhalla N, Narayanan A. Human antimicrobial peptides as therapeutics for viral infections. Vol. 11, Viruses. MDPI AG; 2019.
Reduction of lymphocyte at early stage elevates severity and death risk of COVID-19 patients: a hospital-based case-cohort study
Keywords: COVID-19, lymphocyte, severity, organ injuries.
Main findings: The aim of this study was to assess an association between reduced blood lymphocyte counts at hospital admission and prognosis of COVID-19 patients (n=192). The authors found:
Patients with lymphopenia are more likely to progress to severe disease or succumb to COVID-19 (32.1% of COVID-19 patients with lymphocyte reduction died).
Reduction of lymphocytes mainly affects the elderly (> 70 years old).
Lymphocyte reduction is more prevalent in COVID-19 patients with cardiac disease and pulmonary disease, patients with increase in the chest CT score (key marker of lung injury) and a decrease in several respiratory function markers (PaCO2, SpO2, oxygenation index).
Limitations of the study: Reduced blood lymphocyte counts with aging have been known (https://www.medrxiv.org/content/10.1101/2020.03.08.20031229v2) https://onlinelibrary.wiley.com/doi/epdf/10.1111/sji.12413). Therefore, it is not unexpected that a larger fraction of COV ID-19 patients above 70 years old have lower lymphocytes counts. Since age has been reported to be a major factor that determines outcome for COVID-19, lymphocyte counts and prognosis should have been adjusted by age. Multivariate analysis to identify independent risk factors is lacking.
Relevance: Previous studies demonstrated that SARS-CoV-2 infection leads to a decrease of the T cell count. This study confirms these results and shows that lymphocyte reduction mainly affects the elderly. Lymphopenia was associated with disease severity as well as worse prognosis. Future studies need to address if lymphopenia is a negative predictive factor independent from age.
Review by Meriem Belabed as part of a project by students, postdocs and faculty at the Immunology Institute of the Icahn school of medicine, Mount Sinai.
COVID-19 Critical Illness Pathophysiology Driven by Diffuse Pulmonary Thrombi and Pulmonary Endothelial Dysfunction Responsive to Thrombolysis
Poor et al., medRxiv \[10.1101/2020.04.17.20057125\]
This report describes the use of systemic tissue plasminogen activator (tPA) to treat venous thromboembolism (VTE) seen in four critically ill COVID-19 patients with respiratory failure. These patients all exhibited gas exchange abnormalities, including shunt and dead-space ventilation, despite well-preserved lung mechanics. A pulmonary vascular etiology was suspected.
All four patients had elevated D-dimers and significant dead-space ventilation. All patients were also obese, and 3/4 patients were diabetic.
Not all patients exhibited an improvement in gas exchange or hemodynamics during the infusion, but some did demonstrate improvements in oxygenation after treatment. Two patients no longer required vasopressors or could be weaned off them, while one patient became hypoxemic and hypotensive and subsequently expired due to a cardiac arrest. Echocardiogram showed large biventricular thrombi.
In addition to the small sample size, all patients presented with chronic conditions that are conducive to an inflammatory state. It is unclear how this would have impacted the tPA therapy, but it is likely not representative of all patients who present with COVID-19-induced pneumonia. Moreover, each patient had received a different course of therapy prior to receiving the tPA infusion. One patient received hydroxychloroquine and ceftriaxone prior to tPA infusion, two patients required external ventilator support, and another patient received concurrent convalescent plasma therapy as part of a clinical trial. Each patient received an infusion of tPA at 2 mg/hour but for variable durations of time. One patient received an initial 50 mg infusion of tPA over two hours. 3/4 patients were also given norepinephrine to manage persistent, hypotensive shock. Of note, each patient was at a different stage of the disease; One patient showed cardiac abnormalities and no clots in transit on an echocardiogram, prior to tPA infusion.
The study describes emphasizes the importance of coagulopathies in COVID-19 and describes clinical outcomes for four severe, COVID-19 patients, who received tPA infusions to manage poor gas exchange. While the sample size is very limited and mixed benefits were observed, thrombolysis seems to warrant further investigation as a therapeutic for COVID-19-associated pneumonia that is characterized by D-dimer elevation and dead-space ventilation. All four patients had normal platelet levels, which may suggest that extrinsic triggers of the coagulation cascade are involved.
The authors suspect that endothelial dysfunction and injury contribute to the formation of pulmonary microthrombi, and these impair gas exchange. Pulmonary thrombus formation has also been reported by other groups; post-mortem analyses of 38 COVID-19 patients’ lungs showed diffuse alveolar disease and platelet-fibrin thrombi (Carsana et al., 2020). Inflammatory infiltrates were macrophages in the alveolar lumen and lymphocytes in the interstitial space (Carsana et al., 2020). Endothelial damage in COVID-19 patients has also been directly described, noting the presence of viral elements in the endothelium and inflammatory infiltrates within the intima (Varga et al., 2020). One hypothesis may be that the combination of circulating inflammatory monocytes (previously described to be enriched among PBMCs derived from COVID-19 patients) that express tissue factor, damaged endothelium, and complement elements that are also chemotactic for inflammatory cells may contribute to the overall pro-coagulative state described in COVID-19 patients.
Carsana, L., Sonzogni, A., Nasr, A., Rossi, R.S., Pellegrinelli, A., Zerbi, P., Rech, R., Colombo, R., Antinori, S., Corbellino, M., et al. (2020) Pulmonary post-mortem findings in a large series of COVID-19 cases from Northern Itality. medRxiv. 2020.04.19.20054262.
Varga, Z., Flammer, A.J., Steiger, P., Haberecker, M., Andermatt, R., Zinkernagal, A.S., Mehra, M.R., Schuepbach, R.A., Ruschitzka, F., Moch, H. (2020) Endothelial cell infection and endotheliitis in COVID-19. Lancet. 10.1016/S0140-6736(20)30937-5.
The study described in this review was conducted by physicians of the Divisions of Pulmonary, Critical Care, and Sleep Medicine, Cardiology, Nephrology, Surgery, and Neurosurgery and Neurology at the Icahn School of Medicine at Mount Sinai.
Reviewed by Matthew D. Park as part of a project by students, postdocs, and faculty at the Immunology Institute of the Icahn School of Medicine, Mount Sinai.