Upregulation of interleukin-19 in saliva of patients with COVID-19

Abstract

Cytokines are major players in orchestrating inflammation, disease pathogenesis and severity during COVID-19 disease. However, the role of IL-19 in COVID-19 pathogenesis remains elusive. Herein, through the analysis of transcriptomic datasets of SARS-CoV-2 infected lung cells, nasopharyngeal swabs, and lung autopsies of COVID-19 patients, we report that expression levels of IL-19 and its receptor, IL-20R2, were upregulated following SARS-CoV-2 infection. Of 202 adult COVID-19 patients, IL-19 protein level was significantly higher in blood and saliva of asymptomatic patients compared to healthy controls when adjusted for patients' demographics (P < 0.001). Interestingly, high saliva IL-19 level was also associated with COVID-19 severity (P < 0.0001), need for mechanical ventilation (P = 0.002), and/or death (P = 0.010) within 29 days of admission, after adjusting for patients' demographics, diabetes mellitus comorbidity, and COVID-19 serum markers of severity such as D-dimer, C-reactive protein, and ferritin. Moreover, patients who received interferon beta during their hospital stay had lower plasma IL-19 concentrations (24 pg mL^-1) than those who received tocilizumab (39.2 pg mL^-1) or corticosteroids (42.5 pg mL^-1). Our findings indicate that high saliva IL-19 level was associated with COVID-19 infectivity and disease severity.

Figure 1

Flow chart of the study.

Figure 2

Higher IL-19 gene expression levels in lung and nasopharyngeal swabs of COVID-19 patients. (A–C) IL-19 mRNA levels in SARS-CoV-2 infected human airway epithelial cells [HAECs] (n = 3 infected HAECs vs. n = 3 mock-treated HAECs; GSE147507), nasopharyngeal swabs of COVID-19 patients (n = 430 COVID-19 patients vs n = 54 healthy controls; GSE152075); as well as lung autopsies of COVID-19 patients (n = 17) SARS-CoV-2 infected lung vs. n = 5 healthy lung biopsies; GSE150316). (D–F) IL-6 mRNA levels in SARS-CoV-2 infected HAECs, in nasopharyngeal swabs of COVID-19 patients as well as lung autopsies of COVID-19 patients. (G–L) IL-20R1 and IL-20R2 (IL-19 heterodimer receptors) mRNA levels in SARS-CoV-2 infected HAECs, in nasopharyngeal swabs as well as lung autopsies of COVID-19 patients. Comparison was done using unpaired t-test or Mann–Whitney U test, depending on the skewness of the data. ns = non-significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3

Higher IL-19 protein level in saliva of asymptomatic and severe COVID-19 patients. (A, B) IL-19 levels in plasma and saliva of COVID-19 patients with different severity (asymptomatic [n = 67], mild/moderate [n = 81], and severe [n = 54]), as well as healthy controls (n = 20). (C–E) Correlation of IL-19 saliva level with serum levels of D-dimer, CRP, and ferritin of these patients. (F, G) IL-6 levels in plasma and saliva of COVID-19 patients with different severity. (H–J) Correlation of IL-6 saliva level with serum levels of D-dimer, CRP, and ferritin of these patients. (K) Protein expression of p-STAT3 in saliva of COVID-19 patients (n = 5) and of healthy controls (n = 3). Blots were cut prior to hybridization with antibodies during immunoblotting. The full-length blots (uncut) of the replicates are represented in Supplementary Fig. 1. Statistical tests: Linear regression models adjusted for patient’s demographics factors (age, male sex, and body mass index), comorbidities (diabetes mellitus), and COVID-19 related severity serum markers (D-dimer, C-reactive protein, and ferritin); the Pearson correlation test; and Unpaired t-test or Mann–Whitney U test, depending on the skewness of the data. ns non-significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 4

Increased IL-19 level in saliva of severe COVID-19 patients associated with higher need for mechanical ventilation and/or death by day 29. Kaplan–Meier survival curves of the need for mechanical ventilation (A, B) and/or death (C, D), based on the IL-19 or IL-6 cytokine levels in saliva of patients with severe COVID-19 (n = 54). Statistical test: Cox proportional models adjusted for patient’s demographics factors (age, male sex, and body mass index), comorbidities (diabetes mellitus), and COVID-19 related severity serum markers (D-dimer, C-reactive protein, and ferritin), with significance indicated by P value of less than 0.05.

Figure 5

Expression of IL-19 during SARS-CoV-2 and other viral infections. (A) Upregulation of IL-19 in AECs infected with SARS-CoV-2 compared to other respiratory viral infections. (B) Upregulation of IL-19 in peripheral blood of infected patients with SARS-CoV-2 compared to other respiratory viral infections. LogFC was determined using adjusted LIMMA. Unpaired student t-test was used to compare between fold changes. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 6

Plasma interleukin (IL)-19 levels in COVID-19 patients on different treatment regimens. Level of IL-19 in the plasma of COVID-19 patients untreated or treated with corticosteroids, tocilizumab, or IFNβ during their course of hospital treatment. The data show a significant reduced IL-19 plasma level of IFNβ treated patient and a trend of reduction for glucocorticoids or tocilizumab treated COVID-19 patients. Statistical tests: Unpaired t-test or Mann–Whitney U test, depending on the skewness of data. NS nonsignificant. ***P < 0.001.