Monocytopenia, monocyte morphological anomalies and hyperinflammation characterise severe COVID-19 in type 2 diabetes

Abstract

Early in the COVID-19 pandemic, type 2 diabetes (T2D) was marked as a risk factor for severe disease and mortality. Inflammation is central to the aetiology of both conditions where variations in immune responses can mitigate or aggravate disease course. Identifying at-risk groups based on immunoinflammatory signatures is valuable in directing personalised care and developing potential targets for precision therapy. This observational study characterised immunophenotypic variation associated with COVID-19 severity in T2D. Broad-spectrum immunophenotyping quantified 15 leucocyte populations in peripheral circulation from a cohort of 45 hospitalised COVID-19 patients with and without T2D. Lymphocytopenia and specific loss of cytotoxic CD8⁺ lymphocytes were associated with severe COVID-19 and requirement for intensive care in both non-diabetic and T2D patients. A morphological anomaly of increased monocyte size and monocytopenia restricted to classical CD14^Hi CD16^- monocytes was specifically associated with severe COVID-19 in patients with T2D requiring intensive care. Increased expression of inflammatory markers reminiscent of the type 1 interferon pathway (IL6, IL8, CCL2, INFB1) underlaid the immunophenotype associated with T2D. These immunophenotypic and hyperinflammatory changes may contribute to increased voracity of COVID-19 in T2D. These findings allow precise identification of T2D patients with severe COVID-19 as well as provide evidence that the type 1 interferon pathway may be an actionable therapeutic target for future studies.

Keywords: COVID-19; SARS-CoV-2; inflammation; monocyte; type 2 diabetes.

Figure 1. Type 2 diabetes is associated with decreased monocyte frequency and phenotypic alterations in COVID‐19 patients

1. A

   Flow cytometric quantification of lymphocytes, monocytes, B cells, natural killer (NK) cells, dendritic cells (DCs) and granulocytes in peripheral venous blood samples from non‐diabetic (ND) and type 2 diabetic (T2D) patients with COVID‐19.

2. B

   Quantification of monocyte subpopulation phenotypes as classical (CD14^Hi CD16⁻), intermediate (CD14^Hi CD16⁺) or non‐classical (CD14^Lo CD16⁺).

3. C

   Principal component analysis of lymphocyte and monocyte population frequencies in ND and T2D COVID‐19 patients and in T2D patients without COVID‐19.

Data information: Data are presented as mean ± SEM. Differences between groups were evaluated with unpaired t‐test. *p < 0.05. See also Appendix Fig S1 and Tables S1–S3. Sample size and exact P‐value in Appendix Table S5. Source data are available online for this figure.

Figure 2. Morphologically altered monocytes in type 2 diabetic COVID‐19 patients are associated with an aberrant inflammatory response and increased disease severity

1. A

   Quantification of CD14 expression and size (FSC) in monocyte populations from non‐diabetic (ND) and type 2 diabetic (T2D) COVID‐19 patients.

2. B

   Frequency of conventional (FSC^Lo) and large (FSC^Hi) monocytes in ND and T2D COVID‐19 patients.

3. C

   Expression of CD16, CD14 and HLA‐DR in FSC^Lo and FSC^Hi monocytes from COVID‐19 patients.

4. D

   Proportions of classical, intermediate and non‐classical monocytes within FSC^Lo and FSC^Hi monocytes from COVID‐19 patients.

5. E

   Correlative analyses of IL8 expression in peripheral blood mononuclear cells (PBMCs) to monocyte FSC.

6. F

   IL8, IL6 and CCL2 mRNA expression in PBMCs from ND and T2D COVID‐19 patients.

7. G

   Monocyte size quantified in ND and T2D COVID‐19 patients treated in general wards (Gen) or in the intensive care unit (ICU).

Data information: Data are presented as mean ± SEM. Differences between groups were evaluated with unpaired t‐test (except for (C) and (D) where paired by patient). Analysis of variance (ANOVA) or analysis of covariance (ANCOVA) was used for multiple group comparisons. *p < 0.05; **p < 0.01 and ***p < 0.001. For correlative analysis, Spearman's test was carried out calculating a 2‐tailed P‐value. See also Appendix Fig S2. Sample size and exact P‐value in Appendix Table S5.Source data are available online for this figure.

Figure 3. IRF5 expression is associated with monocyte activation and morphological adaptation in COVID‐19 patients

1. A

   Quantification of IRF5 and IFNB1 mRNA expression in peripheral blood mononuclear cells of non‐diabetic (ND) and type 2 diabetic (T2D) COVID‐19 patients.

2. B

   Histograms of IRF5 expression on different populations analysed by flow cytometry on venous blood samples.

3. C

   IRF5 median fluorescence intensity (MFI) in monocytes of ND and T2D patients with COVID‐19.

4. D

   IRF5 expression overlaid onto monocyte phenotypic gating.

5. E

   IRF5 expression (MFI) in monocyte subtypes from ND and T2D COVID‐19 patients.

6. F

   HLA‐DR and IRF5 expression in monocytes from COVID‐19 patients.

7. G

   IRF5 expression in monocytes of ND or T2D COVID‐19 patients admitted to the intensive care unit (ICU) or treated exclusively in general wards (Gen).

8. H

   IRF5 expression in monocytes of ND or T2D COVID‐19 patients admitted to the ICU or treated exclusively in Gen.

Data information: Data are presented as mean ± SEM. Differences between groups were evaluated with unpaired t‐test. Analysis of variance (ANOVA) was used for multiple group comparisons. *p < 0.05 and **p < 0.01. For correlative analysis, Spearman's test was carried out calculating a 2‐tailed P‐value. See also Appendix Fig S3. Sample size and exact P‐value in Appendix Table S5. Source data are available online for this figure.

Figure 4. Severe COVID‐19 is associated with loss of CD8⁺ lymphocytes and loss of classical monocytes in patients with T2D

1. A

   Pie charts of COVID‐19 hospitalisation that were exclusively treated in general wards (Gen) or that required intensive care (ICU). Proportions of patients admitted to the ICU that were non‐diabetic (ND) or with type 2 diabetes (T2D).

2. B

   Proportions of lymphocytes and monocytes in peripheral venous blood from T2D patients with COVID‐19 either treated in Gen or requiring ICU admission.

3. C

   Phenotypic analysis of lymphocyte and monocyte subpopulations in peripheral venous blood from T2D COVID‐19 patients treated in Gen or requiring ICU treatment.

4. D

   t‐SNE mapping of all cytometric acquired data with projections of population density, CD8 or IRF5 expression. Maps are of representative profiles from each group.

Data information: Data are presented as mean ± SEM. Differences between groups were evaluated with unpaired t‐test or non‐parametric Mann–Whitney U‐test. *p < 0.05. See also Appendix Fig S4. Sample size and exact P‐value in Appendix Table S5. Source data are available online for this figure.