Human Blood Non-Classical/Classical Monocyte Cells Are Heterogeneously Presented in Severe COVID-19 and Correlate with Disease Activity

Coronavirus disease (COVID-19) is a complex pathological condition resulting from syndrome coronavirus 2 (SARS-CoV-2) virus, which still presents significant challenges to global public health. Although the underlying mechanisms that trigger the pathogenesis of COVID-19 remain elusive, severe COVID-19 has been associated with exaggerated immune-mediated inflammation [1]. Early and efficient activation of the immune system is crucial for controlling SARS-CoV-2, but the prolonged hyperinflammation characterized by excessive activation of mononuclear phagocytes (MNPs) may lead to progressive tissue damage [2].

Monocytes are blood-circulating, phagocytic innate immune cells, which form the part of the MNP system together with dendritic cells (DC) and macrophages [3]. Monocytes are a heterogeneous population of antigen presenting cells that express MHC class II molecules and can be divided into three subsets based on their respective expression level of CD14 and CD16: classical (CLs) (CD14⁺CD16⁻), non-classical (NCs) (CD14^dimCD16⁺), and intermediate (CD14⁺CD16⁺) [4, 5]. Recent studies indicate that CL migrate to the trachea and lung tissue, where they undergo differentiation into macrophages that play a crucial role in safeguarding pulmonary health and functionality. Conversely, NCs differentiate into macrophages within numerous blood vessels of the lung without infiltrating the lung parenchyma [6, 7].

Regarding COVID-19, it was reported a statistically significant increase in CLs and a decrease in NCs in COVID-19 patients compared to healthy controls [8]. In particular, there was a higher proportion of IL-6-secreting CD14⁺CD16⁺ inflammatory monocytes in COVID-19 patients compared to healthy individuals, which migrate to the lungs and induce subsequent lung damage [9]. Of note, SARS-CoV-2 may directly infect these monocytes as a certain population expresses ACE2 [10]. SARS-CoV-2-infected monocytes can produce large amounts of inflammatory mediators that contribute to local tissue inflammation and a dangerous systemic inflammatory response called cytokine storm [11, 12]. Nevertheless, the role of monocytes in COVID-19 disease is not fully understood and not very widely researched, especially in severe COVID-19. Thus, the current study was to complement the COVID-19 evaluation by integrating blood monocyte subsets, to investigate whether heterogeneous blood NC/CL cells correlate with development process and prognosis in severe COVID-19.

The study group consisted of 32 COVID-19 patients and 18 health controls from June 2023 to March 2024 in the department of respiratory and critical medicine of the First Affiliated Hospital of Soochow University. All COVID-19 patients were diagnosed according to the Guide of Diagnosis and Treatment of COVID-19 (10th edition, in Chinese) published by the National Health Commission of China. Moreover, all the included COVID-19 patients were infected by Omicron variants during this period based on the test statistics reported by the Center for Disease Control and Prevention in Suzhou. Those with active viral hepatitis, active tuberculosis, abnormal immune function, auto-immune diseases, immunosuppressive status, antibody, or antiviral therapy were excluded to mitigate the influence of other potential confounding variables [13]. The First Affiliated Hospital of Soochow University’s Ethics Committee gave its approval to this study and written informed consent was obtained from participants prior to the study.

The clinical information of all enrolled COVID-19 patients was collected from an electronic medical record and included the following: basic information such as age, gender and underlying disease; laboratory findings such as monocyte-lymphocyte ratio, D-dimer, lactate dehydrogenase, N-terminal B-type natriuretic peptide, hypersensitive C-reaction protein, cardiac troponin T; disease severity score such as sequential organ failure assessment and CURB-65 score; therapies; and therapeutic response.

Moreover, the COVID-19 patients were further classified as mild and severe illnesses based on World Health Organization (WHO) criteria [14]. Mild COVID-19 patients could be defined as symptomatic patients meeting at least one clinical sign of pneumonia (fever, cough, dyspnea, and/or fast breathing), without hypoxia (SpO₂ ≥ 90% on room air), and severe patients consisted of those with at least one clinical sign of pneumonia plus at least one of the following: respiratory rate > 30 breaths/min or SpO₂ < 90% on room air or a significant increase of over 50% in the size of the internal lesion in pulmonary imaging within 24–48 h by contrast. Progressive group was defined as the presence of radiographic progression, escalation of respiratory support, development of septic shock, or the occurrence of severe complications within 7 days of testing [15].

For flow cytometric analysis, 50 µL of peripheral blood and 1 µL of specific monoclonal antibodies were added to each cytometric tube for surface marker detection. Briefly, erythrocytes were lysed with OptiLyse C No-Wash Lysing Solution (Beckman Counter Life Science) for 10 min, then, cells were stained with fluorescently labeled antibodies for 30 min at 4°C in the dark, including PECF594-labeled anti-CD56 mAb (clone 5.1H11), PE-labeled anti-CD14 mAb (clone 63D3), FITC-labeled anti-CD16 mAb (clone 3G8), and AF647-labeled anti-ACE2 mAb (clone A20069I). After washing, cells were analyzed within 2 h. For each sample, a minimum of 100,000 events were collected using the Gallios^™ Flow Cytometer (Beckman Coulter, Inc.) and analyzed using FlowJo 10 software.

The cells were stimulated and blocked in 96 well plates for 8 h at 37°C in the presence of phorbol myristate acetate (PMA), ionomycin, and monensin. After further wash, the cells were suspended in PBS+2%FBS and stained for 20 min with PE-labeled anti-CD14 mAb (clone 63D3), AC7-labeled anti-CD16 mAb (clone 3G8) at 4°C. FITC-labeled anti-IL-1β mAb (clone JK1B-1), PC7-labeled IL-6 mAb (clone MQ2-13A5), and BV421-labeled anti-IFN-γ mAb (clone 4S.B3) were added after fixation and permeabilization for 30 min at 4°C. Finally, the cells were again washed twice and ready for analysis.

Statistical analysis was performed using GraphPad Prism 6 and SPSS 26. The ratios of NCs and CLs in this study were expressed as medians (25th to 75th percentile range). The Mann-Whitney U test, Fisher’s exact test, paired t test, and Kaplan-Meier Survival Analysis were used for analyzing all data. Statistical significance was set at p < 0.05.

We analyzed the blood NC/CL subsets in samples from COVID-19 patients (20 severe and 12 mild) and healthy controls (18 donors). The gating strategy of NC/CL subsets using flow cytometry was shown in Figure 1a. The CL subset with high expression of the CD14 cell surface receptor and no CD16 expression (CD14⁺⁺CD16⁻) and the NC subset otherwise with a low/negative level of CD14 expression and co-expression of the CD16 receptor (CD14^−/+CD16⁺⁺).

As shown in Figure 1b, the frequency of NC cells was lower in mild COVID-19 when compared to healthy controls (13.15% [9.08, 20.43] vs 21.95% [16.43, 28.95]; p = 0.038). In contrast, the frequencies of CL within monocytes were significantly higher in mild COVID-19 compared to healthy controls (83.95% [77.73, 88.00] vs 71.90% [66.23, 76.63]; p = 0.017). Of note, the frequency of NC and CL subsets was not substantially different in severe COVID patients (NCs 26.45% [8.63, 46.03]; CLs 63.80% [48.00, 86.55]; p both >0.05) compared with healthy controls and mild COVID patients.

Importantly, it was noticed that NC/CL subset was skewed toward NC^highCL^low and NC^lowCL^high cluster among the severe group. To be more specific, the prevalence of NCs among monocytes consistently excluded the presence of CL, which was confirmed by the uniform identification of individuals (Fig. 2a). Actually, NC^lowCL^high cluster in severe COVID-19 displayed comparable ratio of NCs and CLs to those in mild COVID-19, whereas NC^highCL^low cluster constituted heterogeneous population of severe COVID-19 in terms of blood NC/CL subsets (Fig. 2b–d).

A first clinical investigation of these skewed NCs/CLs was to perform other profiles between NC^highCL^low and NC^lowCL^high clusters. In detail, the distinct CUBR-65 score (4.0 [3.0, 4.0] vs. 2.0 [2.0, 2.8], p = 0.004) and sequential organ failure assessment score (6.5 [5.0, 8.0] vs. 3.5 [3.0, 4.8], p = 0.020) was found between these two clusters. Furthermore, the skewing of NC/CL subsets also resulted in a distinct laboratory finding in severe COVID-19, as determined by the monocyte to lymphocyte ratio (M/L, 1.23 [0.61, 1.57] vs. 0.54 [0.44, 0.74], p = 0.031), D-dimer (4.89 [1.92, 5.66] vs. 1.19 [0.63, 2.77], p = 0.039), and lactate dehydrogenase (357.85 [333.33, 433.63] vs. 240.5 [176.7, 298.9], p = 0.016) (Table 1). These data suggested skewed NCs/CLs could complement a deep understanding of severe COVID-19 and act as a novel biomarker reflecting systematic inflammation.

Then, we further determined whether the skewing in NC/CL subsets is associated with disease outcomes in severe COVID-19. Overall, patients were given homogenized treatment regarding antiviral, anti-fungal, glucocorticosteroid, and respiratory support. As shown in Figure 3, the NC/CL signature was consistently prognostic, demonstrating a higher 7-day disease progression rate (83.3% vs. 25%, p = 0.019) and a worse 28-day survival (p = 0.026) in NC^highCL^low cluster than that in NC^lowCL^high clusters.

Many of the cytokine and surface markers expressed on some or all of the monocytes reflect an activated state. As shown in Figure 4, IFN-γ (26.95% [13.00, 53.65] vs 75.85% [62.13, 85.30], p = 0.007) and IL-1β (18.50% [3.43, 43.83] vs. 91.55% [76.45, 95.93], p = 0.006) had statistically significant decreases in the NC versus CL comparison, whereas NCs displayed an increased capacity of IL-6 (25.05% [22.13, 35.40] vs. 6.56% [2.74, 11.66], p = 0.028). Nevertheless, we examined whether these NC/CL cells were associated with distinct ACE2 expression. It was indicated that CL subset expressed higher ACE2 than NC subset (22.75% [9.17, 32.50] vs. 1.06% [0.76, 6.27], p = 0.021). Overall, these data support the view that the simultaneously activated/suppressed NC/CL profile is heterogeneously presented in severe COVID-19, and correlates with disease outcome.

The Omicron variant of SARS-CoV-2 was characterized as high transmissibility, although it appears to be less pathogenic than previous strains [16]. Furthermore, owing to widespread vaccination efforts and the development of herd immunity, the ongoing COVID-19 epidemic exhibits periodic localized outbreaks rather than a large-scale surge in cases. According to the reports from the Center for Disease Control and Prevention in Suzhou, all the enrolled COVID-19 patients in this study experienced Omicron variant infection.

Recent studies have revealed an extensive and durable remodeling of immune cells during and after SARS-CoV-2 infection [17]. In our study, we analyzed the heterogeneous monocytes state of COVID-19 patients with varying degrees of severity and compared with healthy controls using flow cytometry. Accordingly, we identified an imbalance of monocytes, especially involvement of NC^highCL^low cluster, which was predictive of infection severity and short-term clinical therapeutic response.

First, it was characterized by an increase in CL subset and a decrease in NC subset in mild cases compared with healthy controls. It means that monocyte with an inflammatory, interferon-secreting, IL-1β-secreting phenotype is the dominant feature in mild disease. Other studies also found that the ratio of CL was higher in the early recovery stage of COVID-19 patients than the health controls [18], and the CL subset shifted toward a hyporesponsive phenotype [19]. Simultaneously, a distinct decrease in peripheral NCs was found in subjects with the most severe clinical status at admission [20]. Miao Li et al. [21] discovered that asymptomatic COVID-19 patients tended to maintain normal lymphocytes and reduce monocytes to inhibit monocyte-mediated inflammation, which was favorable to keep a stable and helpful immune response [22, 23]. These findings at least hinted that the activation of CL might play a role in dampening the inflammatory response and promoting tissue repair, but the differentiation of NC subset could indicate ongoing severe inflammatory lung damage.

Notably, the profile of NCs/CLs occurred in mild cases was observed in only a fraction of severe COVID-19 cases, which means significant heterogeneity regarding NC/CL differentiation. In various epidemic periods of COVID-19 strains, many researchers from diverse countries have also conducted studies on the alterations in grouping patterns of peripheral blood monocyte subsets among COVID-19 patients of different infection conditions. We have compiled a table to compare the findings reported by researchers across different periods (online suppl. material; for all online suppl. material, see https://doi.org/10.1159/000542652) [18‒20, 24‒26]. However, none of these studies have mentioned that NC/CL subsets could be skewed toward NC^highCL^low and NC^lowCL^high clusters among severe COVID-19 patients, which constituted a distinctive and significant finding of our study. In detail, the NC^highCL^low cluster in severe COVID-19 displayed a distinct clinical phenotype, implying a higher 7-day disease progression rate and a worse 28-day survival. Conversely, NC^lowCL^high was identified as a good biomarker for limited severity and a relatively favorable outcome.

As for the application of interleukin-blocking monoclonal antibodies, the available data in this study indicated that the secretion of IL-1β was primarily attributed to CL subset in monocytes. Moreover, considering that NC^lowCL^high was identified as a promising biomarker for limited severity and a relatively favorable outcome, we identified CL subset and its secreted IL-1β as protective factors in controlling the progress of Omicron variant infection. Another study also suggested that IL-1β was higher in convalescents plasma of Omicron variant infection patients than in healthy controls [13]. Therefore, our study did not appear to support the application of IL-1 blockade such as anakinra in patients with Omicron variant infection. Conversely, the increase of NC subset was proportional to the severity of disease and worse prognosis in Omicron variant infection. The excessive secretion of IL-6, which was mainly produced by NC subset in monocyte cells, seemed to contribute to the progression of infection. Thus, our study supported the indispensable role of IL-6 blockade in the management of patients with Omicron variant infection.

Although the exact role of the NC/CL axis in the pathogenesis and development of COVID-19 is still not completely clear, the heightened monocytes could be related to the severity and complexity of COVID-19 patients. Monocytes may be infected by SARS-CoV-2 through ACE2-dependent and ACE2-independent pathways [10, 27]. SARS-CoV-2 can effectively suppress the antiviral IFN response in monocytes. The SARS-CoV-2-infected monocytes can produce large amounts of numerous types of proinflammatory cytokines and chemokines, which contribute to local tissue inflammation and a dangerous systemic inflammatory response [28]. Although IFNs may be protective during the early stages of the disease, the extended IFN-γ production could eventually cause macrophage hyperactivation [29, 30]. The reprogramming of proinflammatory macrophages to anti-inflammatory macrophages may also be considered in COVID-19. Interestingly, published scRNA-sequencing datasets generated from severe COVID-19 patients demonstrate increased transcript reads of IL-1β and TNF-a from BALF macrophages [31], supporting the hypothesis that proinflammatory reprogramming of lung macrophages and/or precursor monocytes may drive prolonged and exacerbated COVID-19 symptomology.

The following limitations of this study should be acknowledged. These cohorts were all taken from a single medical center, which may hinder the generalization of the results to other populations. The number of COVID-19 patients and healthy controls was limited. Furthermore, the lack of serial measurements was very regrettable. Dynamic changes in the levels of monocyte subsets and functions of cytokines secreted by monocyte subsets remain targets for future studies, whereas experiments could be performed to enhance the understanding of monocyte subsets.

In summary, the promising finding of this study was that the blood NC/CL subset was skewed toward NC^highCL^low and NC^lowCL^high cluster among the severe COVID-19 patients, which supported a better understanding of the heterogeneous COVID-19 disease. These observations will also aid in optimizing evaluation in severe COVID-19, and developing strategies that target altered balance of NC/CL cell subsets.

We thank the patients, the nurses, and clinical staff who are providing care for the patient, and staff at the local and state health departments.

The First Affiliated Hospital of Soochow University’s Ethics Committee gave its approval to this study (2024-407). Written informed consent was obtained from participants prior to the study.

The authors declare that there is no conflict of interest.

The project was mainly supported by the Natural Science Foundation of Suzhou City Grant SYS2021034 (to C.C).

Q.Q. and C.C. conceived the idea and designed and supervised the study. D.Z. and Y.S. performed the experiment and drafted the manuscript. D.Z., S.J., D.Q., and Y.W. collected data, analyzed data, and performed statistical analysis.

Danhong Zhou and Yu Shen contributed equally and are regarded as co-first authors.

The datasets analyzed during the current study are not publicly available for the reason that the containing information could compromise the privacy of research participants but are available from the corresponding author (C.C.) on reasonable request.