Patterns of cytokine and chemokine expression in peripheral blood of patients with COVID-19 associated with disease severity

Abstract

Objectives

Cytokine dysregulation has been proposed as one of the main culprits for severe COVID-19 and poor prognosis. We examined the parallel presence of lymphopoietic, proinflammatory, Th1, Th2, regulatory cytokines, and chemokines in the serum of 47 patients with mild, moderate, and severe COVID-19 and evaluated the association between cytokine concentrations and disease severity.

Methods

A multiplex quantitative cytokine analysis ProcartaPlex™ immunoassay was applied, using the Luminex^TM 200X detection system (Invitrogen).

Results

The concentrations of twelve cytokines: IL-18, IFN-gamma, TNF-alpha; IL-21; IL-1alpha, IL-1beta, IL-6, IL-22; IL-10, IL-1RA; IL-7 and IFN-alpha were consistently elevated in the studied serum samples. All examined chemokines—Eotaxin, GRO-alpha, IL-8, IP-10, MCP-1, MIP-1alpha, MIP-1beta, SDF-1alpha, and RANTES, were detectable in all studied groups, confirming their importance in mediating the adaptive immune response regardless of disease severity. The serum concentrations of six mediators: IL-1beta, IL-6, IL-18, IL-10, IL-8, and IP-10, showed statistically significant differences among the groups with different disease severity. IL-6, IL-1beta, and IL-10 were more significantly elevated in severe cases while milder symptoms were associated with lower levels of IL-8 and IP-10.

Conclusion

Overall, the studied chemokines demonstrated an associated production in acute COVID-19 infection. A strong correlation was observed between the Th1 mediators IL-18 and IL-10 and the proinflammatory IL-6 in the severe COVID-19 group. Our results indicated that severe COVID-19 was characterized by a dysregulated cytokine pattern whereby the Th1 immune response is outweighed by the immunoregulatory response, while inhibitory signals cannot balance the hyperinflammatory response.

Introduction

In December 2019 in Wuhan, China, an outbreak of mysterious pneumonia was recognized and later the causative agent was identified as a novel type of coronavirus. The virus was named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the disease it caused—coronavirus disease 2019 (COVID-19). The rapid spread of illness worldwide and the high potential for further propagation of the problem prompted the WHO to declare COVID-19 as a global pandemic in March 2020. Infection with SARS-CoV-2 may be asymptomatic or present with symptoms ranging from mild or moderate to severe and even critical. The manifestations, clinical course and outcome of the disease are determined by multiple factors which can be virus-related (viral variant) or host-related (genetic, age-related, behavioral, underlying chronic illnesses, immune status, etc.). Thus, as a part of the effort to combat the COVID-19 pandemic and gain insight into the mechanism of the disease, many investigations have focused on the dynamics of the immune response.

Cytokines are soluble polypeptides with a central role in the activation and regulation of the host’s immune and inflammatory responses. Entry of virus through the respiratory tract is followed by its recognition and induction of several potent proinflammatory cytokines, including type I interferon (IFN-I), IL-1, IL-6, and TNF-alpha, as well as a variety of chemokines, which are a family of cytokines responsible for the recruitment of macrophages, neutrophils, and other important immune cells to the site of infection. This leads to expansion of the inflammatory process, development of the adaptive immune response and establishment of an antiviral state. Clinical symptoms include local and systemic signs of infection (cough, rhinorrhea, throat pain, dyspnea and tachypnea, chest pain, fever, fatigue, malaise, etc.) and in critical cases disease advances to respiratory failure, septic shock, multiple organ dysfunction, and death. Cytokine dysregulation has been proposed as one of the main culprits for severe COVID-19 and poor prognosis and the unique immune profile of these patients has been extensively studied in order to improve understanding and management of the disease.¹

The overexaggerated inflammatory response, termed cytokine release syndrome (CRS) or cytokine storm, is believed to be the substrate of severe disease manifestations as in cases of acute respiratory distress syndrome (ARDS) and multiorgan failure. In COVID-19 patients CRS is characterized by increased levels of certain cytokines (IL-1, IL-6, IL-18, IL-10, TNF-α, IL-8, IP-10, MCP-1, etc.) with some of these being more consistently detected and universally accepted as distinct features of SARS-CoV-2 infection.²

An imbalance in the initial antiviral response leads to distorted activation and progression of adaptive immunity response. A characteristic immune profile of COVID-19 patients with severe disease includes lymphocytopenia accompanied by high numbers of activated neutrophils (and therefore a high neutrophil-to-leukocyte ratio [NLR]). Neutrophils produce high quantities of NETs (neutrophil extracellular traps) which have been suggested to further aggravate the inflammatory state during COVID-19.^1,3,4 Cytokines are major contributing factors in these processes since they are responsible for most of the cell signaling and interactions.

The aim of this investigation is to examine the parallel presence of lymphopoietic, proinflammatory, Th1, Th2, regulatory cytokines, and chemokines in the serum of patients with mild, moderate, and severe COVID-19 and to evaluate the association between cytokine concentrations and disease severity.

Materials and methods

Patients and clinical samples

A group of 47 COVID-19 patients took part in this cross-sectional study, including 21 male and 26 female individuals with mean age 49.9 (±17,3) years. SARS CoV-2 infection was confirmed by RT-PCR performed on nasopharyngeal swab samples. Participants were assigned into three severity groups retrospectively: group 1/mild (n = 18), group 2/moderate (n = 19), and group 3/severe cases (n = 10). The criteria used for the definition of disease severity were in accordance with the WHO guideline.⁵ Patients with immunodeficiency and other diseases that may affect the immune response were excuded of the study. Studied serum samples were taken in the first week after onset of symptoms and were collected during the period June–July 2022.

Cytokines and chemokines assay

For multiplex quantitative cytokine analysis, ProcartaPlex™ immunoassay was applied, using the Luminex^TM 200X detection system (Invitrogen). Serum samples were tested with two commercially available panels: Cytokine 25-Plex Human ProcartaPlex™ Panel 1B, designed for detection of GM-CSF, IFN-gamma, IL-1beta, IL-2, IL-4, IL-5, IL-6, IL-12p70, IL-13, IL-18, TNF-alpha; IL-9, IL-10, IL-17A (CTLA-8), IL-21, IL-22, IL-23, IL-27, IFN-alpha, IL-1alpha, IL-1RA, IL-7, IL-15, IL-31, and TNF-beta and Chemokine 9-Plex Human ProcartaPlex™ Panel 1 for detection of Eotaxin (CCL11), GRO-alpha (CXCL1), IL-8 (CXCL8), IP-10 (CXCL10), MCP-1 (CCL2), MIP-1alpha (CCL3), MIP-1beta (CCL4), SDF-1alpha, and RANTES (CCL5) (Invitrogen, EPX250-12166–901, EPX090-12,187–901).

Both cytokines and chemokines panels, includes the key mediators that may operate during protective/dysregulated antiviral response as follows: all aspects of immune inflammation (leucopoesis, chemotaxis, migration, degranulation, pyrogenesis, superoxide production, angiogenesis, etc): GM-CSF, IL-6, IL-1alpha, IL-1beta, MIP-1alpha, MIP-1beta, IP-10, GRO-alpha, MCP-1, RANTES, SDF-1alpha, Eotaxin; different types of effector responses including Th1: IL-18 IL-12 IFN-gamma, TNF-alpha, IL-2, IL-15; Th2: IL-4, IL-5, IL-13 IL-5, IL-9, IL-13, IL-21; Th3: IL-17, IL-22, IL-23 cytokines; lymphopoesis and differentiation of T-cell memory: IL-7 IL-15; back-regulation of immune activation: IL-10, IL-1RA.

Samples were processed according to the manufacturer’s instructions and analyzed with ProcartaPlex^TM Analyst 1.0 Software.⁶ Serum concentrations are given as mean in pg/ml with a 95% CI (Confidence interval). Values above the detection limit were interpreted as presence of cytokines in the serum.

Ethical statements

Ethical approval for this study was obtained from the Institutional review board at NCIPD (approval number 4/17.02.2021). Written informed consent was obtained from all patients before the study.

Statistical analysis

The statistical analysis of data was performed in SPSS software using independent samples Kruskal–Wallis test. We evaluated the correlations between the cytokines overall in all groups of severity calculating the Spearman’s rank correlation coefficient. A p value of <.05 was considered as statistically significant. Graphs were made with GraphPad Prism 9 software.

Results

Serum levels of proinflammatory Th1, Th2, regulatory, and lymphopoietic cytokines

We examined several groups of cytokines related to antiviral immune response: the early inducers of immune inflammation: IFN-alpha, and GM-CSF; Th1 response mediators (IL-18, IFN-gamma, TNF-alpha, IL-2), Th2 response mediators (IL-4, IL-5, IL-9, IL-13, and IL-21), proinflammatory (IL-1alpha, IL-1beta, IL-6, IL-17A, and IL-22), regulatory (IL-10, IL-1RA, TNF-beta, and IL-27), and cytokines related to lymphopoiesis and Ly differentiation (IL-7 and IL-15). The concentrations of twelve cytokines: IL-18, IFN-gamma, TNF-alpha; IL-21; IL-1alpha, IL-1beta, IL-6, IL-22; IL-10, IL-1RA; IL-7 and IFN-alpha were consistently elevated in the studied serum samples. We considered them as important for the early immune response in COVID-19 infection. Another nine: IL-2, IL-5, IL-9, IL-13, IL-27; IL-15, IL-23, TNF-beta, and GM-CSF were found elevated only in single samples from the cohort group and their elevation was most likely incidental and not due to the COVID-19 infection. Four cytokines: IL-4, IL-17A, IL-12p70, and IL-31 were below the detection limit of the applied multiplex immunoassay in all tested samples (Table 1).

Table 1.

Mean serum levels with SD and 95% CI of the cytokines with consistently elevated concentrations. Results are presented separately for the mild (1), moderate (2), and severe (3) group of COVID-19 patients. Statistical differences were determined by the Kruskal–Wallis test. A p value < .05 was considered as statistically significant. (SD: Standard deviation, CI: Confidence interval).

Cytokine (pg/ml)	Severity COVID-19 groups:1-Mild 2-Moderate 3-Severe	p	Mean	Std. Deviation	95% Confidence Interval for Mean
					Lower Bound	Upper Bound
IFN-gamma	1	.066	4.51	4.66	2.70	6.32
	2		8.47	18.53	1.78	15.15
	3		12.56	19.30	2.96	22.16
IL-1beta	1	.019	0.66	0.80	0.35	0.98
	2		0.97	1.93	0.27	1.67
	3		3.85	6.50	0.62	7.08
IL-6	1	.003	10.83	3.75	9.37	12.28
	2		11.15	5.20	9.28	13.03
	3		215.07	606.12	-86.34	516.48
TNF-alpha	1	.174	0.68	0.60	0.45	0.91
	2		0.69	0.82	0.39	0.98
	3		6.94	20.32	-3.16	17.04
IL-18	1	.011	39.20	50.98	19.44	58.97
	2		26.31	32.68	14.53	38.09
	3		83.55	86.86	40.35	126.74
IL-10	1	.008	0.87	1.35	0.34	1.39
	2		1.67	3.95	0.24	3.09
	3		5.35	7.52	1.61	9.09
IL-21	1	.095	15.90	47.47	-2.51	34.31
	2		21.26	33.56	9.16	33.35
	3		18.37	32.46	2.22	34.51
IL-22	1	.873	7.17	23.66	-2.01	16.35
	2		18.26	66.22	-5.62	42.13
	3		39.87	125.98	-22.78	102.52
IFN-alpha	1	.782	3.43	7.96	-0.10	6.96
	2		3.55	6.63	1.03	6.07
	3		3.11	6.38	-0.06	6.29
IL-1alpha	1	.682	1.26	2.24	0.27	2.26
	2		1.66	5.09	-0.27	3.60
	3		5.49	11.94	-0.45	11.42
IL-1RA	1	.280	1162.99	2863.01	-106.40	2432.37
	2		564.75	1716.10	-88.02	1217.52
	3		1688.75	3109.51	142.42	3235.07
IL-7	1	.691	4.44	3.81	2.75	6.13
	2		5.24	5.40	3.18	7.29
	3		3.53	2.42	2.33	4.73
Eotaxin	1	.630	67.46	41.83	50.91	84.00
	2		54.06	32.82	42.23	65.90
	3		59.67	36.13	41.70	77.63
GRO-alpha	1	.116	8.99	8.31	5.70	12.28
	2		20.04	25.07	11.00	29.08
	3		15.02	10.35	9.87	20.17
IL-8	1	.003	47.61	145.77	-10.05	105.27
	2		200.22	516.00	14.18	386.26
	3		173.54	343.81	2.57	344.51
IP-10	1	>,001	94.91	105.91	53.01	136.80
	2		209.80	166.65	149.72	269.89
	3		275.24	226.49	162.61	387.87
MCP-1	1	.857	214.31	194.76	137.27	291.36
	2		266.74	283.48	164.53	368.94
	3		227.89	180.15	138.31	317.48
MIP-1alpha	1	.210	9.33	18.74	1.91	16.74
	2		15.30	38.96	1.25	29.34
	3		24.84	40.60	4.65	45.03
MIP-1beta	1	.130	118.63	122.96	69.99	167.27
	2		197.15	376.76	61.31	332.98
	3		179.76	151.99	104.18	255.35
SDF-1alpha	1	.874	382.68	81.61	350.39	414.96
	2		375.70	159.47	318.21	433.20
	3		395.08	87.79	351.42	438.73
RANTES	1	.692	72.76	26.28	62.37	83.16
	2		70.95	32.03	59.40	82.50
	3		81.64	40.54	61.48	101.80

IFN-alpha is an early alerting signal in case of viral infection that inhibits virus translation in infected cells before the instauration of adaptive effector mechanisms. Increased serum concentrations of IFN-alpha were found in all three groups of severity, but the differences in its expression were not statistically significant between the three groups. Elevated serum levels of the lymphopoietic IL-7 were found in most patients with mild, moderate, and severe COVID-19. All patients with moderate and severe course of COVID-19 had increased expression of IL-7. However, the difference between the three groups was not significant.

The serum levels of several important mediators of Th1 immune response—IL-18, IFN-gamma, and TNF-alpha, were elevated in all studied groups. The concentrations of IL-18 were significantly elevated in severe as compared to moderate cases (p < .05), Figure 1. IFN-gamma and TNF-alpha also showed a tendency for higher serum concentrations in patients with severe COVID-19, but the differences between the groups were not statistically significant. In contrast, the serum levels of IL-2 and IL-12p70 were below the limit of detection in all studied groups.

Figure 1.

Serum concentration of IL-18 in patients with mild, moderate and severe COVID-19. The individual values for patients with mild (group 1, open circle), moderate (group 2, gray circle), and severe (group 3, black circle) disease are shown. Means are denoted. Statistical differences were determined by the Kruskal–Wallis test (**p < .005; ns = not significant, p > .05).

The mediators of Th2 immune response, IL-5 and IL-9, were only slightly increased in single samples, while IL-4 was not detected at all. These results were expected, as the samples were taken in the very beginning of SARS-CoV-2 infection. The serum levels of IL-21 were increased in approximately one-third of the samples, with higher values in patients with moderate and severe disease that did not reach statistical significance.

Controlled immune inflammation is a prerequisite for induction of adaptive immune response. Increased serum levels of several pro-inflammatory cytokines: IL-1alpha, IL-1beta, IL-6, and IL-22, were established in patients from all studied groups with some association between the serum levels and the severity of COVID-19. IL-1beta and IL-6 production were significantly increased in severe cases and differentiated them from patients with moderate (p < .01) and mild (p < .05) disease, Figure 2. IL-17 A and IL-23, which are also involved in inflammatory immune response were not detectable in the studied cohort.

Figure 2.

Serum concentrations of the pro-inflammatory cytokines IL-6 (A) and IL-1beta (B) in the groups with mild, moderate, and severe COVID-19. The individual values for patients with mild (group 1, open circle), moderate (group 2, gray circle), and severe (group 3, black circle) disease are shown. Means are denoted. Statistical differences were determined by the Kruskal–Wallis test (**p < .005; *p < .05; ns = not significant, p > .05).

Back-regulation of immune inflammation after elimination of pathogen is required for successful instauration of immune memory. Our study revealed increased serum concentrations of both major anti-inflammatory cytokines IL-1RA and IL-10, mostly in patients with severe disease. The serum level of IL-10 was significantly higher in severe cases as compared to those with moderate (p < .05) and mild (p < .01) illness, Figure 3.

Figure 3.

Serum concentrations of IL-10 in the groups with mild, moderate and severe COVID-19. The individual values for patients with mild (group 1, open circle), moderate (group 2, gray circle), and severe (group 3, black circle) disease are shown. Means are denoted. The statistical differences were determined by the Kruskal–Wallis test (**p < .005; *p < .05; ns = not significant, p > .05).

Serum levels of chemokines

All examined chemokines—Eotaxin (CCL11), GRO-alpha (CXCL1), IL-8 (CXCL8), IP-10 (CXCL10), MCP-1 (CCL2), MIP-1alpha (CCL3), MIP-1beta (CCL4), SDF-1alpha, and RANTES (CCL5), were detectable in all studied groups, confirming their importance in mediating the adaptive immune response regardless of disease severity (Table1). The comparison between the mean levels of chemokines in the three groups revealed significant differences for only two molecules: IL-8 and IP-10. Both of them were significantly elevated in patients from the moderate and severe group as compared to the mild cases. Data is presented on Figures 4(a) and (b).

Figure 4.

Serum concentrations of the chemokines IL-8 (A) and IP-10 (B) in the groups with mild, moderate, and severe COVID-19. The individual values for patients with mild (group 1, open circle), moderate (group 2, gray circle), and severe (group 3, black circle) disease are shown. Means are denoted. Statistical differences were determined by the Kruskal–Wallis test (***p < .0005; **p < .005; *p < .05; ns = not significant, p > .05).

Correlations and patterns of cytokine and chemokine expression

Since none of the 33 examined mediator molecules could differentiate distinctly between the three examined groups, we looked whether specific patterns of cytokine/chemokine expression could be delineated in association with disease severity.

In order to trace a logical profile, we examined the correlations between all detectable mediators. (Figure 5) Overall, the studied chemokines demonstrated an associated production in acute COVID-19 infection. With the exception of CCL5, the expression of all chemokines correlated between each other. The strongest correlations were detected for MIP-1alpha and MIP-1beta (R = 0.89, p < .001); IL-8 and MIP-1alpha (R = 0.71, p < .001); and IL-8 and MIP-1beta (R = 0.69, p < .001). The same chemokines correlated moderately with IP-10 (R = 0.41, p < .001). No significant correlations were found between the examined chemokines and cytokines except for a weak correlation of GRO-alpha with IFN-gamma, and IL-1RA.

Figure 5.

Spearman correlation correologram for the detectable cytokines and chemokines. The strength of the correlation is represented by the color of the intersection square. Scale range is from strong positive correlation, R = 1.0 to strong negative correlation, R = - 1.0. The significant correlations are described in the text.

All detectable proinflammatory cytokines IL-1alpha, IL-1beta, IL-6, and IL-22 correlated in their expression. The most prominent correlations were established between the pro-inflammatory IL-6, and IL-1beta (R = 0.57, p < .001) on the one hand, and the inhibitory IL-10 (R = 0.50, p < .001), on the other hand. Further on, both IL-1beta and IL-6 correlated with the precursor of IFNgamma - IL-18. Logically, the mediators of Th1 response, IL-18 and IFN-gamma, correlated between each other (R = 0.54, p < .001). Overall these correlations showed that in the settings of more intensive inflammation, a stronger Th1 response might be expected, as well as stronger counter-balancing inhibitory signals

We further looked, whether the above correlations were reproduced in each of the severity groups. Actually, in mild cases no correlation was found between IL-6 and IL-10, or IL-18. In moderate cases a very weak correlation between IL-6 and IL-10 existed (R <0.35, p < .05). In contrast, a very strong correlation (R >0.50) was observed between Il-6 and IL-10 (R = 0.62, p < .01), as well as between Il-10 and IL-18 (R = 0.66, p < .1) in the severe COVID-19 cases. Notefully, this increased IL-18 expression was not associated with a higher level of the Th1 effector cytokine INFgamma. In fact, the ratio IFN-gamma/IL-10 was significantly decreased in severe COVID-19 cases: 1,8 vs 3,9 and 3,8 in mild and moderate cases (p < .05 for both). On the other hand, the ratio IL-6/IL-10 was comparable in the mild and moderate cases but increased in the severe group (0,2 vs 0,1 vs 0,7). Overall these data indicate a compromised cytokines balance.

A comparison between the expression of all detectable cytokines/chemokines in the mild, moderate, and severe groups is presented in Figure 6(a). A comparable associated elevation of chemokines was observed in all the three groups. However, mild cases were clearly distinguished by significantly lower concentrations of both IP-10 and IL-8.

Figure 6.

An expression patterns of separate cytokines and chemokines (A), and their ratios (B) in the serum of patients with mild (1, white bars), moderate (2, gray bars), and severe (3, black bars) COVID-19. The mean serum levels are presented as fold increase above the specific limit of detection for each cytokine/chemokine.

Further on, mild and moderate cases were characterized by a prominent associated elevation of the Th1 mediators IL-18 and FN-gamma and less prominent balanced elevation of the proinflammatory IL-1alpha and its inhibitor IL-1RA. Severe cases were clearly distinguished by a significant and correlated elevation of IL-18 in parallel with the proinflammatory molecules IL-6, IL-1alpha, IL-1beta and the inhibitory cytokines IL-10, IL-1RA, and IL-22. However, it was not accompanied by a comparable elevation of effector cytokines (IFN-gamma, TNF-alpha, or IL-21). The unbalanced release of cytokines in severe COVID-19 is further demonstrated by the ratios between the mean increase of IFN-gamma and IL-10; IL-6 and IL-10; and IL-1alpha and IL-1RA. As shown in Figure 6(b), severe cases were distinguished by a prevalence of the inhibitory IL-10 over the effector cytokines, and of the inflammatory over the regulatory ones.

Discussion

Our study investigated the presence of lymphopoietic, proinflammatory, Th1, Th2, regulatory cytokines, and chemokines in the serum of patients with mild, moderate, and severe COVID-19. We evaluated the association between cytokine concentrations and disease severity. The results from the study suggested that there was a statistically significant difference in the expression of six cytokines and chemokines—IL-1beta, IL-6, IL-18, IL-10, IL-8, and IP-10, in patients with different severity of COVID-19.

IL-8 and IP-10 were the only cytokines that differentiated mild cases from the moderate and severe cases. These chemokines have key roles in antiviral immunity and their production is induced by inflammation activators such as IL-1, TNF-alpha, LPS and, especially for IP-10, IFN-gamma. IL-8 is an important chemoattractant for neutrophils, leading to formation of NETs and induction of the oxidative burst. These stimulatory effects are further strengthened by IP-10 and GRO-alpha, both promoting neutrophil infiltration and activation. IP-10 is chemotactic for monocytes/macrophages, T lymphocytes, NK cells, and DCs. We observed a positive correlation between the serum levels of the two cytokines and a similar association was also noted by Chen.⁷ Both IP-10 and IL-8 were also correlated with increased concentrations of MIP-1beta and GRO-alpha which are responsible for the recruitment of various leukocytes to the site of inflammation and thus contribute additionally to tissue injury and more severe symptoms. IP-10 and IL-8 were also identified as major contributors to the development of pulmonary disease in infections caused by other coronaviruses (SARS-CoV and MERS-CoV).⁸

In addition, IL-8 stimulates contraction of the airway epithelium further advancing the inflammatory process, suppresses apoptosis of neutrophils, and inhibits IFN production. Thus, it is not surprising that higher levels of IL-8 were associated with development of more severe disease and ARDS.^2,9,10 IL-8 was superior in differentiating healthy and recovered from COVID-19 patients when compared to IL-6.¹¹ On the other hand, Gong¹² found no significant elevation of IL-8 in patients with mild and severe disease, while Chi¹³ found a significant difference only between healthy people and COVID-19 patients but not between groups with different severity.

In SARS-CoV-2 infection IP-10 is produced in high quantities by bronchiolar and alveolar epithelial cells. Migration of activated T cells to the lungs leads to tissue injury and in severe cases to ARDS.^2,14,15 Hayney¹⁶ have shown that IP-10 is a useful marker of disease severity, duration, and prognosis in acute respiratory infections, with higher levels correlating with detectable viral load. In COVID-19 patients IP-10 was proposed as a potential biomarker for poorer outcome.^6,15 In addition to recruiting various immune cells toward inflamed and infected tissues, IP-10 also promotes T cell apoptosis which, in conjunction with the increased activation of neutrophils, viral destruction of lymphocytes, redistribution and sequestration of lymphocytes in the lungs, contributes to the increased NLR characteristic of these patients.^8,10 Thus, in line and in addition to other published data, we propose both IL-8 and IL-10 as sensitive and early prognostic markers of COVID-19 progression.

Two proinflammatory molecules were significantly increased in the severe group compared to both the mild and the moderate group: IL-6 and IL-1beta. Severe cases were also differentiated by significantly increased levels of the inhibitory molecule IL-10. In addition, a significant correlation existed between the levels of IL-6 (a major promoter of inflammation and driver of the cytokine storm), IL-18 (an important activator of Th1 immune response), and IL-10 (a key immunoregulatory molecule with potent anti-inflammatory functions) in severe COVID-19 but in mild and moderate cases these correlations were weaker or absent. This parallel increase in severe cases was imbalanced since inhibitory signals overwhelmed Th1-induction, and inflammation overwhelmed regulatory signals.

IL-6 and IL-1beta are important factors of the early immune response that trigger the production of other proinflammatory cytokines and the induction of the acute-phase response. They potentiate the development of fever, lung inflammation, and fibrosis.^10,17 Both cytokines were shown to be potent promoters of coagulopathy and thrombotic events in the course of severe COVID-19.^2,7

Increased IL-6 levels have been consistently reported in severe COVID-19.^12,18-20 It has also been proposed as a prognostic marker for poor clinical outcome.^18,21 In a Meta-analysis of association of IL-6 and disease severity and mortality Jafrin finds that serum IL-6 levels are significantly higher in patients with severe infection than with mild.²² On the other hand, its prognostic value has been questioned by others.⁴

In our study, the serum levels of IL-6 correlated with other proinflammatory cytokines such as IL-1beta, IL-18, and IL-1alpha, which probably function cooperatively to maintain the pathologic inflammatory state. Positive correlation was also observed with the inhibitory cytokine IL-10, a back-regulator of immune inflammation, that may further contribute to impaired viral clearance and more severe disease in case of hyperactivation.

The observed association of higher IL-1beta production with disease severity, echoes data from previous studies.^20,23 Together with IL-18, it is a product of inflammasome activation after viral recognition and we found a positive correlation between the levels of the two cytokines. IL-1beta serves as a downstream indicator of inflammasome activation and pyroptosis. Pyroptosis has been reported to play a major role in the pathological processes of COVID-19.^2,24 In addition to promoting the secretion of other proinflammatory cytokines, IL-1beta is also capable of inducing its own production by binding to IL-1R on monocytes and macrophages leading to an ever-expanding inflammatory process.²⁵ Multiple clinical trials have shown the clinical benefit of IL-1R blockade with the biologic drug Anakinra. This has the bonus effect of inhibiting IL-1beta-induced IL-6 production, as well as the actions of IL-1alpha, another potent inflammatory protein.^25-28 However, not all have managed to find an association between increased expression of IL-1beta and severe disease.^9,29

IL-10 is a major anti-inflammatory cytokine and its main role is to regulate the immune processes in order to avoid excessive tissue damage. We found higher IL-10 concentrations in association with more severe disease, in agreement with other studies.^10,12-14,22 Elevated IL-10 levels in severe COVID-19 therefore could be an attempt to moderate the hyperactivated inflammatory response. However, overexpression of IL-10 leads to extreme immunosuppression and viral tolerance which further impedes disease resolution. In these cases, IL-10 has a detrimental rather than protective role. By contrast, Xu²⁹ did not observe any significant difference in IL-10 expression between patients with mild, moderate and severe disease and their results showed comparable serum concentrations in COVID-19 patients and healthy controls.

In our study the increased concentrations of IL-10 positively correlated with higher serum levels of proinflammatory cytokines such as IL-1beta, IL-6, and IL-18. However, the balance between inhibitory, proinflammatory and Th1 effector signals was not comparable in the three severity groups. The ratios IL-6/IL-10 and IFN-gamma/IL-10 showed that in severe cases the immunoregulatory response overrides the Th1 effector response but does not balance inflammation.

According to our results, IL-18 was significantly increased in the severe group compared to moderate cases. Previous studies have reported increased concentrations of IL-18 in more severe cases.^13,29,30 IL-18 is an important facilitator of cell-mediated immunity through activation of Th1 cells and NK cells. It also promotes the production of various cytokines including IFN-gamma—we found a positive correlation between the serum concentrations of the two molecules. In addition, IL-18 correlated with IL-1beta and IL-6 which corresponds with previous findings.²³ Its stimulatory effects on neutrophils, macrophages, and basophils further advance the immunopathological process. The abundance of IL-18 receptors in the lungs in combination with a dysregulated expression during disease point to its role as a potent mediator of lung injury and pulmonary fibrosis.^9,20,23 Indeed, Kerget³¹ found that higher IL-18 levels in COVID-19 patients were associated with macrophage activation syndrome (MAS) and ARDS.

Interestingly, we found no significant difference between the serum concentrations of IL-18 in mild and in severe cases. One possible explanation may be the concerted expression of its regulator IL-18BP (IL-18 binding protein). Most probably, in mild cases along with the elevation in IL-18 an adequate increase in the secretion of IL-18BP is observed, that keeps the former in check. Thus, only mild respiratory symptoms develop and unnecessary damage to the host organism is prevented. On the other hand, in severe disease the overproduction of IL-18 is not paralleled by elevation of IL-18BP and the resultant imbalance contributes to the hyperinflammatory state.²³

Cytokines which have been previously shown to correlate with disease severity in SARS-CoV-2 infection such as TNF-alpha, IFN-gamma, and MCP-1 were elevated in all three groups but did not show any significant differences. These results might be influenced by numerous factors including individual characteristics of the host response, genetic background, comorbidities, as well as the sampling time point. We also found production of IL-7, IL-1RA, IFN-alpha, IL-1alpha, IL-21, IL-22, eotaxin, GRO-alpha, MIP-1alpha, MIP-1beta, SDF-1alpha, and RANTES in COVID-19 patients from all groups regardless of disease severity suggesting a limited contribution to the development of lung inflammation and respiratory symptoms. Other cytokines (IL-9, IL-15, IL-23, IL-27, TNF-beta, IL-5, IL-2, IL-13, IL-4, IL-17A, IL-31, GM-CSF, and IL-12p70) were not detected in our study, or were only sporadically so. There are differing reports regarding the serum concentrations of these cytokines and their role in COVID-19 pathophysiology.^4,13,20 These discrepancies could be due to the dynamic changes in the serum cytokine profile at close moments of sampling. In addition, certain cytokines may only have local effects, while being undetectable in the systemic circulation.

Overall, in our study we have chosen to look at soluble mediators (cytokines and chemokines) because the combined changes of 25 cytokines and 9 chemokines related to migratory, effector, and regulatory functions at both innate and adaptive immunity level would give an integral picture of organism’s capacity to respond and overcome SARS-CoV-2 infection.

Limitations

As a major limitation of our study, we note the relatively small group of COVID-19 patients. Due to this reason we didn’t have the opportunity to perform a calculation and justification of the sample size included in the survey. This study was financed and performed as a part of a research project on the laboratory and clinical characteristics of SARS-CoV-2 with limited duration. Further studies on each severity group are necessary to thoroughly examine the expressed cytokine profile and its association with the severity of COVID-19.

Conclusion

The pathological processes during SARS-CoV-2 infection are inextricably linked to the concerted production and actions of various cytokines in the host organism. In this study we assessed the possible relationship between disease severity and the type and quantity of circulating cytokines, and compared our results with the published data. We tested 47 COVID-19 patients with mild, moderate, and severe symptoms for the presence of 33 cytokines, and chemokines and found IL-6, IL-1beta, and IL-10 to be more significantly elevated in severe cases while milder symptoms were associated with lower levels of IL-8 and IP-10. Our results indicated that severe COVID-19 was characterized by a dysregulated cytokine pattern whereby the Th1 immune response is outweighed by the immunoregulatory response, while inhibitory signals cannot balance the hyperinflammatory response.

ORCID iDs

Iva Trifonova https://orcid.org/0000-0002-7320-9548

Kim Ngoc https://orcid.org/0000-0001-8390-9970

Yana Todorova https://orcid.org/0000-0003-4708-6887