Figure 2.
Mild COVID-19 is characterized by high levels of IFN-III, but not IFN-I, in response to high viral loads in the upper airways
(A–M) Swabs from a cohort of SARS-CoV-2-positive hospitalized patients and ICU inpatients (HOSP, black dots; ICU, red dots; both HOSP and ICU, black lines and analyzed together) and home-isolated patients (HI, green dots and lines) were analyzed.
(A–H) IFNL1 (A), IFNL2,3 (B), IFNL4 (C), IFNB1 (D), IFNA2 (E), IFNA4 (F), IL1B (G), and IL6 (H) mRNA expression is plotted against mean viral RNA CT. Each dot represents a patient. Linear regression lines (continuous line) and 95% confidence interval (dashed line and shaded area) are depicted. Spearman correlation coefficients (r) and p value (p) are indicated in black and in green for “HOSP + ICU” and “HI” patients respectively.
(I) Mean viral RNA CT values are plotted against days from symptom onset (DFSO). Each dot represents a patient. Lines connect mean values for each range of DFSO.
(J) K-means clustering based on the expression of IFNA2, IFNB1 IFNL1, IFNL2,3, and IL1B was used to determine clusters 1–3 (cluster 1, n = 13; cluster 2, n = 12; cluster 3, n = 6). The color indicates the relative gene expression. Viral load tercile, age group, and severity are annotated. Viral load terciles (“+++,” “++,” and “+”) are defined by mean viral RNA CT (<20, >20 and <30, and >30). Age groups are defined as <70 or ≥70 years.
(K) IFNL1, IFNL2,3, IFNA2, IFNB1, and IL1B mRNA expression within clusters identified in (J). Each dot represents a patient. Violin plots are depicted.
(L) Percentage of patients with the indicated disease severity within clusters identified in (J).
(M) Odds ratio of patients in cluster 2 being hospitalized or admitted to the ICU relative to patients in cluster 3 (clusters identified in J). Symbols represent the odds ratio. Error bars represent the 95% confidence interval associated with the odds ratio.
(A–H and K) Expression is plotted as log2 (gene/GAPDH mRNA + 0.5 × gene-specific minimum). Statistics by test for difference between simple linear regression slopes: ns, not significant (p > 0.05); ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001 (A–H); two-way ANOVA: ns, not significant (p > 0.05); ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001 (I); or chi-square test for odds ratio: ns, not significant (p > 0.05); ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 (L). See also Figure S2 and Table S3.