SARS-CoV-2 infection predisposes patients to coinfection with Staphylococcus aureus

Abstract

Severe COVID-19 has been associated with coinfections with bacterial and fungal pathogens. Notably, patients with COVID-19 who develop Staphylococcus aureus bacteremia exhibit higher rates of mortality than those infected with either pathogen alone. To understand this clinical scenario, we collected and examined S. aureus blood and respiratory isolates from a hospital in New York City during the early phase of the pandemic from both SARS-CoV-2+ and SARS-CoV-2- patients. Whole genome sequencing of these S. aureus isolates revealed broad phylogenetic diversity in both patient groups, suggesting that SARS-CoV-2 coinfection was not associated with a particular S. aureus lineage. Phenotypic characterization of the contemporary collection of S. aureus isolates from SARS-CoV-2+ and SARS-CoV-2- patients revealed no notable differences in several virulence traits examined. However, we noted a trend toward overrepresentation of S. aureus bloodstream strains with low cytotoxicity in the SARS-CoV-2+ group. We observed that patients coinfected with SARS-CoV-2 and S. aureus were more likely to die during the acute phase of infection when the coinfecting S. aureus strain exhibited high or low cytotoxicity. To further investigate the relationship between SARS-CoV-2 and S. aureus infections, we developed a murine coinfection model. These studies revealed that infection with SARS-CoV-2 renders mice susceptible to subsequent superinfection with low cytotoxicity S. aureus. Thus, SARS-CoV-2 infection sensitizes the host to coinfections, including S. aureus isolates with low intrinsic virulence.

Importance: The COVID-19 pandemic has had an enormous impact on healthcare across the globe. Patients who were severely infected with SARS-CoV-2, the virus causing COVID-19, sometimes became infected with other pathogens, which is termed coinfection. If the coinfecting pathogen is the bacterium Staphylococcus aureus, there is an increased risk of patient death. We collected S. aureus strains that coinfected patients with SARS-CoV-2 to study the disease outcome caused by the interaction of these two important pathogens. We found that both in patients and in mice, coinfection with an S. aureus strain lacking toxicity resulted in more severe disease during the early phase of infection, compared with infection with either pathogen alone. Thus, SARS-CoV-2 infection can directly increase the severity of S. aureus infection.

Keywords: COVID; MRSA; SARS-CoV-2; agr; coinfection.

Fig 1

Establishment of a biorepository of S. aureus during the early COVID-19 pandemic (a) Survival curve of patients in our cohort with SARS-CoV-2 confirmed by PCR, with S. aureus blood or respiratory culture or with both infections. Time = 0 on the day of positive S. aureus culture or in the case of SARS-CoV-2 alone, on the day of SARS-CoV-2 PCR positivity. Data were analyzed by Wilcoxon-Breslow test. (b) Generation of the cohort of patients and isolates analyzed in this study. *P < 0.05

Fig 2

Phylogenetic analysis of S. aureus isolates. (a) The first isolate for each patient in the biorepository is included. Reference strains are labeled. Concentric circles represent CC/ST, mec type (only types II and IV were represented in this collection of isolates; if no mec type is indicated, the strain is mec negative), SARS-CoV-2 PCR result, and isolate source. (b) agr type and (c) agr mutations analyzed by SARS-CoV-2 PCR result. A disruptive agr variant was defined as a frameshift or stop gain. ** = reference strain CA-347. CC = clonal complex, ST = sequence type, “NA” = other or not categorized.

Fig 3

Analysis of cytotoxicity of S. aureus isolates in vitro. (a) Cytotoxicity z-score for each isolate. Cytotoxicity was analyzed by infecting PMNs from human donors and quantifying cell lysis via LDH release. PCR pos are isolates from SARS-CoV-2+ patients, and PCR− are isolates from SARS-CoV-2- patients. (b) As in panel a, but without the isolates that grew slowly and were not able to reach an OD₆₀₀ of 1 by the end of the overnight growth. The line in a and b indicates the cutoff used to define high and low cytotoxicity strains. (c) Cytotoxicity z-sores separated by the site of isolation. (d) Cytotoxicity visualized across the phylogenetic tree. (e) Cytotoxicity of isolates analyzed by 30-day mortality of the patients. (f) Cytotoxicity of isolates analyzed by community (CA) vs. healthcare (HA) acquisition. Error bars represent median with interquartile range. P > 0.05 for all plots, Mann-Whitney test (a-c), Kurksal-Wallis test (e and f).

Fig 4

Patient data analysis. (a) Proportions of patients with community-acquired (CA) and healthcare-acquired (HA) S. aureus infection categorized by SARS-CoV-2 PCR result. (b) Survival curve of patients in the four groups described in Table 1. CyTox = cytotoxicity. (c) The same data as in b, focusing on the first 20 days post-S. aureus infection. **P < 0.01, c, Wilcoxon-Breslow test.

Fig 5

SARS-CoV-2 increases susceptibility to S. aureus in mice. (a) Schematic of coinfection experimental design. Survival (b), weight loss (c), and disease score (d) of mice infected with the indicated pathogens (15–17 mice per group, 5 mice in the PBS group). Viral burdens in lungs as determined by plaque assay (e), and qPCR (f) at 4 days post-infection. (g) Bacterial burdens of S. aureus in the indicated organs by CFU at 4 days post-infection. (h) Number of mice that were colonized with S. aureus by 4 days post-infection in any organ (CFUs) or not (LOD) (h). Error bars represent mean +/- SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. b, Mantel-cox test, c, d, Mann-Whitney test, g, Student’s t-test, h, Fisher’s exact test. In panels c and d, * indicates a difference between the SARS-CoV-2 infected group and the coinfected group,and ^# indicates a difference between the S. aureus infected group and the coinfected group.