key: cord-1019263-g5owmgml authors: Tiwari-Heckler, Shilpa; Rauber, Conrad; Longhi, Maria Serena; Zörnig, Inka; Schnitzler, Paul; Jäger, Dirk; Giese, Thomas; Merle, Uta title: Dysregulated host response in SARS-CoV-2 induced critical illness date: 2021-01-18 journal: Open Forum Infect Dis DOI: 10.1093/ofid/ofab019 sha: 54315236123ac4003f062d7ed4d1f4325f16d890 doc_id: 1019263 cord_uid: g5owmgml BACKGROUND: Impaired immune response has been described to be the cause of the development of COVID-19 related respiratory failure. Further studies are needed to understand the immunopathogenesis and to enable an improved stratification of patients that are at risk for critical illness. METHODS: 32 severely ill hospitalized COVID19 patients were recruited in our center at the University Hospital Heidelberg. We performed a comprehensive analysis of immune phenotype, cytokine and chemokine profiling and leukocyte transcripts in severe COVID-19 patients comparing critically ill patients requiring mechanical ventilation and high flow oxygen therapy and non-critically ill patient receiving low flow oxygen therapy. RESULTS: Critically ill patients exhibited low levels of CD8 T cells and myeloid dendritic cells. We noted a pronounced CCR6 (+) TH17 phenotype in CD4 central memory cells and elevated circulating levels of IL-17 in the critical group. Gene expression of leukocytes derived from critically ill patients was characterized by an upregulation of proinflammatory cytokines and reduction of IFN-responsive genes upon stimulation with toll-like receptor 7/8 agonist. When correlating clinical improvement and immune kinetics, we found that CD8 T cell subsets and myeloid dendritic cells significantly increased after disconnection from the ventilator. CONCLUSION: Critical illness was characterized by a TH17-mediated response and dysfunctional IFN-associated response, indicating an impaired capacity to mount antiviral responses during SARS-CoV-2 severe infection. M a n u s c r i p t 4 BACKGROUND: To date, the COVID-2019 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has affected nearly 72 million individuals and led to 1,6 million deaths worldwide. According to World Health Organization (WHO), 15% of these cases are considered to be severe, requiring intensive care unit (ICU) admission and mechanical ventilation in some cases [1] . More than 600 clinical trials are currently conducted to find effective therapeutic strategies against COVID-19. However, the current standard of care still relies on supportive treatment for the severe cases. Therefore, investigation of immunopathology and immune response of COVID-19 related ARDS is warranted for the development of effective drugs to improve the outcomes [2] . Several publications have indicated that a dysfunctional immune response is associated with the severity of COVID-19. So far, lymphopenia has been considered the hallmark of the disease. It has been reported that CD3, CD4 and CD8 counts are correlated with in-hospital mortality, organ injury and severe pneumonia [2, 3] . Quin et al. showed higher neutrophillymphocyte ratio and increased level of pro-inflammatory markers in COVID-19 patients [4] . Further studies have described the so-called "cytokine storm", a strong indicator of the pathological host immune response [3] . In this study, we performed an in-depth immune phenotypic analysis of individual lymphocyte and myeloid cell populations obtained from severe COVID-19 patients, who were admitted to our intensive care unit at the University Hospital Heidelberg. A multiplex assay was run in order to establish a link between immune phenotype and cytokine and chemokine expression. Interferon (IFN)-associated response reflecting the ability of mounting an antiviral response was investigated in an ex vivo stimulation assay. All patients provided written informed consent for the study. Patient care and research were conducted in compliance with the Declaration of Helsinki. Experiments were approved by the Ethics Committee of the University Hospital of Heidelberg, Germany (S148/2020). Whole blood was collected and processed immediately for Flow Cytometry analysis. A c c e p t e d M a n u s c r i p t 5 Flow Cytometry: The phenotyping was performed in whole blood within 4 hours, as described recently [5] . Absolute cell count was measured using the Multitest™ 6-color TBNK reagent (BD Biosciences Blood serum was collected from each subject and was initially stored at -80°C for cytokine and chemokine profiling. For this purpose, multiplex analysis (Bio-Plex Pro Human Cytokine 48-PlexScreening Panel plus ICAM and VCAM, Biorad, Munich, Germany) was performed, as previously described [6] . Table 1 . Laboratory results at admission are presented in Table 2 . Clinical diagnosis of severe respiratory failure correlates with decreased CD4 and CD8 We first examined the absolute counts and frequency of peripheral blood derived lymphoid and myeloid cell populations from severe COVID-19 patients at day 0-2 after hospitalization. Immunophenotypic analysis revealed that numbers of CD8 and CD4 T cells were significantly decreased in critically ill COVID-19 patients, when compared to the noncritical group ( Figure 1A) . Notably, critically ill COVID-19 patients did not present with higher viral RNA load detected in nasopharyngeal swabs, when compared to non-critically ill cases (Suppl. Fig. 1A) . A c c e p t e d M a n u s c r i p t 7 Among the myeloid populations, dendritic cells (DCs) were significantly reduced in critically ill COVID-19 subjects, when compared to the non-critical group ( Figure 1B ). Monocyte and granulocyte counts did not differ between the cohorts (Suppl. Fig. 1B ). We next explored the different myeloid subsets in the circulation of COVID-19 patients. Myeloid DCs were defined as CD11c + CD123population, whereas plasmacytoid DCs were identified as CD11c -CD123 + subset. The frequency of myeloid DCs was diminished in the critical group, when compared to the non-critical group, while there was no difference in the frequency of plasmacytoid DCs ( Figure 1B ). As CD4 cell counts are significantly correlated with disease severity, we next sought to characterize CD4 T cells based on their memory function and T helper subset profile. We defined central and effector memory subsets based on their CD45-RA and CCR7 expression. CD4 central memory T cells (TCM) were marked as CD45-RA -CCR7 + cells, whereas CD4 effector memory T cells (TEM) were identified as CD45-RA -CCR7cells. Absolute counts of both CD4 memory T cells, particularly CD4 TCM, were significantly reduced in the critical group, when compared to the non-critical group ( Figure 2A ). To analyze the functional phenotype of CD4 helper cells, we subdivided this population based on the expression of CCR6 and CXCR3 cells. CD4 TCM from critically ill patients increased the proportion of CCR6 + cells, suggesting a TH17 phenotype ( Figure 2A ). Concomitantly, we found an increase of IL-17 and IL-6 in the serum of critically ill patients, when compared to non-critical cases ( Figure 2B ). Next, we assessed the ability of leukocytes derived from critically ill and non-critically ill COVID-19 patients to respond upon activation by the Toll-like receptor 7/8-agonist Resiquimod (R848), an imidazoquinoline compound. Toll-like receptors serve as pathogen sensors and are key regulators of innate and adaptive immune responses during viral infection. Exposure of leukocytes to R848 for 3 hours resulted in significant induction of proinflammatory mRNA levels, such as IL1B and IL18, in leukocytes of critically ill subjects, when compared to non-critically ill patients ( Figure 3A ). Concomitantly, we had found that A c c e p t e d M a n u s c r i p t 8 critically ill patients exhibited higher level of circulating pro-inflammatory cytokines, such as IL-1, and IL-18, when compared to non-critically ill patients (Suppl. Figure 2 A) . In regard to the chemokine profile, critically ill patients exhibited higher level of circulating chemokines, like Interferon-inducible CXCL10 and CCL7 compared to non-critically ill patients (Suppl. As activation of the TLR7/8-dependent signaling pathway triggers an antiviral IFNassociated response [7] , we evaluated IFN-responsive genes. We observed that MX1, IFIT1 and IFI44L transcripts, were downregulated in leukocytes derived from critically ill patients, when compared to non-critically ill subjects, postulating an impaired capacity to mount antiviral responses during SARS-CoV-2 severe infection ( Figure 3B ). To evaluate the immunological changes that occur during clinical improvement, we randomly selected 7 of our 15 critically ill patients and characterized the kinetics of immune subsets on the day of admission (time-point 1, T1) and on day 0/1 after disconnection from the ventilator (time-point 2, T2) for each subject. The time course of oxygen support is depicted in Figure 4A . Although case 5 was successfully disconnected from the ventilator, his clinical condition deteriorated 2 days later, and died 32 days after admission. All the other patients, depicted here, survived and were discharged alive. Analysis on immune kinetics demonstrated a significant increase in CD8 T cell and myeloid DC counts, when mechanical ventilation or high flow oxygen therapy was not needed ( Figure 4B ). Notably, no difference in TH17 phenotype, shown as frequency of CCR6+ cells within the CD4 central memory population, was detected between the two time points ( Figure 4C ). When considering CD8 memory cells, we found that the absolute counts of CD8 TEM cells significantly increased at time point 2, when compared to time point 1. When analyzing the frequency of CD45-RA -CCR7cells among the CD8 + lymphocytes, representing the effector memory phenotype, we noted a trend to increase in this cell subset at time point 2, when compared to time point 1 ( Figure 4D) . No difference was observed for the central memory phenotype, when measuring the proportion of CD45-RA -CCR7 + cells among the CD8 + lymphocytes derived from critically and non-critically ill patients (Suppl. Fig 3A) . We noted that clinical improvement was significantly associated with expansion of activated CD38 + HLA-DR + CD8 T cells ( Figure 4E ). In regard to CD4 T cells, our analysis showed an A c c e p t e d M a n u s c r i p t 9 increase in CD4 T cell counts at time point 2; as well as expansion of activated CD38 + HLA-DR + CD4 T cells that was associated with clinical improvement (Suppl. Fig 3B) . This study highlights the immunological features of peripheral blood cells derived from injury that was mediated through IL-17 [15] . These findings raise the possibility that TH17driven inflammation might be an important feature in the immunopathogenesis of COVID-19related ARDS. Targeting TH17 signaling with commercially available monoclonal antibodies might be a promising therapeutic option to treat critically ill patients in the future. When considering myeloid DCs, we observed decreased numbers and frequency of these cells in critically ill subjects, when compared to non-critically ill patients at day 0-2 after A c c e p t e d M a n u s c r i p t 10 hospitalization. Myeloid DCs are crucial for antigen presentation and are required for T cell priming to induce a robust antiviral response. A reduced migration of myeloid DCs to the lung result in delayed T cell kinetics in COVID-19, this leading to the pathogenic inflammatory exacerbation. This dysregulated response could explain the susceptibility to a severe outcome in elderly patients, due to the lack of function of antigen-presenting myeloid cells [16, 17] . Although our study sheds a light on a host response mediated by myeloid dendritic cells, further comprehensive analysis on this cell subset are needed to understand their role in the pathogenesis of COVID-19. Interestingly, we discovered that clinical improvement was associated with increased CD8 TEM and DC counts. Zammit and colleagues have shown that DCs are major driver of CD8 memory T cell activation, particularly during a tissue-specific infection of the respiratory tract We also found that clinical improvement resulted in expansion of activated CD38 + HLA-DR + [19, 20] . One of the first effective strategies of antiviral defense is represented by an IFN-mediated innate immunity. We performed gene expression analysis in order to evaluate the capacity of leukocytes from COVID-19 patients to respond to R848, an agonist for TLR7/8. While Animal models of SARS-CoV-2 and in vitro studies on respiratory tract cell lines with SARS-CoV-2 revealed a reduced type I/III IFN response, despite strong expression of IFN-inducible chemokines, such as CXCL10 and CXCL9 [21] . Furthermore, another mechanism, which might explain the high concentration of the pro-inflammatory cytokines IL-1 and IL-18, is through the inflammasome activation. As recently reported, SARS-CoV2 can activate the NLRP3 inflammasome pathway, which ultimately results in a cleavage of the inactive precursors and release of the active cytokines IL-1 and IL-18. Further studies are needed to elucidate the pathomechanism for NLRP3-inflammasome mediated ARDS. [22, 23] A c c e p t e d M a n u s c r i p t 11 Our study has some limitations. Firstly, we did not analyze the data based on days after onset of illness, but rather focused on day 0-2 after admission, as the time point of sudden deterioration marks a significant event in the clinical course of this fatal disease. As data on immune profiling at early time points before hospitalization are missing, the immunological changes that we have observed reflect the overall critical status of severe patients. Further longitudinal studies are needed to further dissect immune phenotypes driving disease progression. Secondly, due to the nature of emergency-scene medical care the blood sampling for research purposes was impaired in some cases, which might have led to selection bias for our immunological analysis. As reported by others, we confirmed that severity of the disease was associated with low counts of circulating CD4 and CD8 cells [3, 4, 24] . Notably, the viral RNA load did not differ between the two cohorts of patients, suggesting that virus load did not correlate with disease severity in our observed cohort. Although some authors hypothesize that a delayed T cell response is triggered by high viral loads in the lung, it remains controversial whether the virus persistence is actually responsible for enhancing tissue damage in the lungs [25] . In summary, our study suggests that COVID-19 related ARDS is characterized by a TH17driven inflammation and dysfunctional IFN-associated response reflecting an impaired capacity to mount antiviral responses. Moreover, respiratory improvement is correlated with dynamic change in CD8 T cell and dendritic cell response. A c c e p t e d M a n u s c r i p t 12 We would like to thank all the patients, their family and surrogates for contributing to this study. We also greatly thank all the caregivers from the "Gastro ICU" at the University Hospital Heidelberg for their support and their dedication to our patients. UM, TG and STH designed the study and acquired, analyzed and interpreted data. IZ, DJ, PS acquired and analyzed data. CR, MSL provided critical revision of the manuscript for important intellectual content. STH wrote the manuscript. All authors read the manuscript and approved the final version. A c c e p t e d M a n u s c r i p t 17 A. Absolute counts of peripheral blood derived CD8 and CD4 T cells in critically ill (N=14) and non-critically ill (N=10) patients at day 0-2 after admission. B. Absolute counts of circulating DCs in peripheral blood of critically ill and non-critically ill patients. Frequency of CD11c and CD123 in the DC population, identifying myeloid and plasmacytoid DC respectively. All bar figures represent mean ± SEM. P value calculated by Mann-Whitney U test. A. Cell counts of CD4 T central memory (TCM) and CD4 T effector memory (TEM) in critical (N=14) and non-critical groups (N=10). Frequency of CCR6 + CXCR3cells, identifying the TH17 cell subset, within the CD4 TCM population of critically ill and non-critically ill COVID19 cases. All bar figures represent in mean ± SEM. B. Serum concentration of circulating IL-17 and IL-6 levels in the critical and non-critical group (n=9 in each group). All bar figures represent mean ± SEM. P value calculated by Mann-Whitney U test. Human leukocytes derived from 8 individuals per group at day 0-2 after admission were used in the assay. A. Relative mRNA expression of IL1B and IL18 in response to TLR7/8 agonist R848. B. Response to R848, assessed by change in expression of the IFN-responsive genes, MX1, IFIT1 and IF44L, is shown. All bar figures represent mean ± SEM. Mann-Whitney U test was used to calculate p values. A. Time course of oxygen support is depicted for each of the 7 critically ill patients. Comparison are conducted between day of admission (T1) and day 0/1 after disconnection from the ventilator for each subject occurred. B. Changes in absolute counts of CD8 T cells and CD8 T effector memory cells at T1 and T2. Frequency of CD45-RA -CCR7cells, indicative of effector memory cell function, within the CD8 T cell subset at the two time points. C. Changes in the absolute count of CD38 + HLA-DR + CD8 T cells and frequency of CD38 + HLA-DR + cells within the CD8 T cell subset before and after disconnection from the ventilator. D. Absolute counts of myeloid DCs at the two time points. P values are generated by using Wilcoxon signed rank test. Clinical management of severe acute respiratory infection when novel coronavirus (nCoV) infection is suspected Immune response in COVID-19: addressing a pharmacological challenge by targeting pathways triggered by SARS-CoV-2 The trinity of COVID-19: immunity, inflammation and intervention Dysregulation of immune response in patients with COVID-19 in Wuhan, China Comprehensive flow cytometric reference intervals of leukocyte subsets from six study centers across Europe Localization and density of immune cells in the invasive margin of human colorectal cancer liver metastases are prognostic for response to chemotherapy Toll-like receptors in antiviral innate immunity SARS-CoV-2 Infections A Case for Targeting Th17 Cells and IL-17A Mikacenic and Mark M. Wurfel. 2020 IL-6: Regulator of Treg/Th17 balance The cytokine storm in COVID-19: An overview of the involvement of the chemokine/chemokine-receptor system Pathological findings of COVID-19 associated with acute respiratory distress syndrome Distinct immune response in two MERS-CoVinfected patients: Can we go from bench to bedside? PLoS One Cell host response to infection with novel human coronavirus EMC predicts potential antivirals and important differences with SARS coronavirus The ratio of Th17/Treg cells as a risk indicator in early acute respiratory distress syndrome IL-17 response mediates acute lung injury induced by the 2009 Pandemic Influenza A (H1N1) Virus Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals Age-related increases in PGD 2 expression impair respiratory DC migration, resulting in diminished T cell responses upon respiratory virus infection in mice Dendritic cells maximize the memory CD8 T cell response to infection Deep immune profiling of COVID-19 patients reveals patient heterogeneity and distinct immunotypes with implications for therapeutic interventions Breadth of concomitant immune responses prior to patient recovery: a case report of non-severe COVID-19 Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19 Inflammasomes are activated in response to SARS-CoV-2 infection and are associated with COVID-19 severity in patients Innate Immunity, Inflammasome Activation, Inflammatory Cell Death, and Cytokines. 2020 Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study The authors have declared that no conflict of interest exists.