Proteases perform essential roles in many biological processes and the inhibition of enzymatic activity has been described as an important step, regulating numerous pathways [49]. In COVID-19, some enzymes are investigated as promising targets to prevent SARS-CoV-2 spread [50, 51]. Among the signaling pathways involved in viral infection, virus entry in the host cells mediated by spike protein, represents a critical step to disease development [52].
The spike protein (S) contains two distinct domains: S1, responsible for receptor binding to the host receptor and S2 domain, that mediates the membranes fusion with the host cells [4, 53, 54]. Specifically, the S1 receptor binding domain (RBD) of SARS-CoV-2 directly binds to peptidase domain (PD) of ACE2 [55], playing a key role in COVID-19 pathogenesis [7].
ACE2 presents two forms, full-length located on the cell membrane (mACE2) and the soluble form released into the circulation (sACE2) [56, 57]. The soluble form lacks the sequence of membrane-anchoring and usually circulates in low concentrations [58, 59], leading some authors to investigate the relationship between circulating ACE2 activity and COVID-19 disease severity [60–62]. The amount and distribution of ACE2 have been described as predictor factors of infectability and the poor outcome of COVID-19 [63]. Furthermore, the activity of circulating ACE2 is associated with disease severity and mortality [61].
Therefore, we first investigated circulating ACE2 enzymatic activity in serum of COVID-19 patients. Additionally, we examined the activity of ACE, DPPIV and PREP, enzymes involved in cardiovascular and renal diseases (frequently observed in patients infected with coronavirus), and inflammatory response. The activity of CAT L was also measured due to the importance of enzymatic activation of spike protein in virus entry in the host cells. We evaluated these enzymatic activities in 164 patients recruited for this study, separated in four groups: COVID-19 negative group (patients with COVID-19 symptoms with negative PCR test for COVID-19), mild, moderate and severe COVID-19 groups.
Regarding the comorbidities, the analyses demonstrated an increase of relative risk (RR) of experiencing severe forms of COVID-19 in patients with comorbidities, including metabolic disease (diabetes, obesity and dyslipidemia), hypertension, heart disease, chronic kidney disease and smoking comparing with patients without comorbidities. The comparison between mild and moderate groups showed the same profile observed in severe group, maintaining the increased probability of comorbidities occurrence (Fig. 1a). The analyses also showed a positive correlation between mortality rate and high levels of CRP (Fig. 1b) and D-dimer (Fig. 1c), corroborating previously published data [45, 64].
Subsequently, we evaluated ACE2 activity in the serum of COVID-19 groups. The results demonstrated elevated enzymatic activity in the moderate and severe groups when compared to mild and COVID-19 negative groups (Fig. 2). These results corroborate already published data showing the correlation between elevated levels of circulating ACE2 and COVID-19 severity [61, 63], suggesting that increased ACE2 may predispose to severe forms of COVID-19 disease [65]. The blockade of ACE2 and spike protein interaction has been described as a promisor target to the development of anti-COVID-19 drugs [66–68].
Cleavage of S1 domain is an important process to expose the fusion peptide, which is a critical mechanism for membranes fusion, allowing virus entry [69]. This cleavage is acid-dependent and is accomplished by several host proteases, including cathepsins [70]. Here, we measured CAT L catalytic activity, and as observed for ACE2, moderate and severe groups presented higher activity compared to mild and negative COVID-19 groups (Fig. 8). Data from literature demonstrated that the inhibition of CAT L using E64 and K777 irreversible inhibitors (in vitro) results in reduced virus replication [40, 71], suggesting a possible proteolytic activation mechanism of spike protein by CAT L [41, 54, 72].
Contrasting to ACE2 and CAT L, our findings show decreased activity of ACE in moderate and severe groups (Fig. 3). Reduced ACE activity might cause KKS imbalance, consequently generating BK accumulation and BK storm [73]. This maintenance of inflammatory response mediated by ACE has been associated to progression of COVID-19 [65, 74]. However, the correlation between ACE and COVID-19 severity is conflicting in the literature [75–77].
Moreover, we also investigated a possible correlation between ACE polymorphism (I/D) with COVID-19 severity. Our results show that ID genotype was the most prevalent among the groups investigated. The frequency of D allele was the most predominant, however, no statistical significance was observed between genotype or allele frequencies with COVID-19 severity (Table 2). The direct association of ACE I/D polymorphism with COVID-19 severity is also conflicting in the literature data [78–81]. Posteriorly, the correlation between ACE polymorphism (I/D) with ACE enzymatic activity was also investigated, since some authors associated DD genotype with increased ACE activity [82]. As expected, regarding the COVID-19 negative group, the DD genotype has increased ACE activity compared to ID + II genotypes, but this difference is not observed in the COVID-19 group (Fig. 4), corroborating a previous report [81]. When comparing specifically the genotypes, the analyses demonstrated that patients with COVID-19, both DD and ID + II genotypes present lower ACE activity, indicating that decreased ACE activity is only correlated with COVID-19 severity.
RAS dysregulation has been associated with the worsening of COVID-19 symptoms [21, 22, 83]. The binding of SARS-CoV-2 to ACE2 leads to the downregulation of the membrane-anchored enzyme, resulting in increased activity of the circulating form [14, 60]. Reduced levels of soluble ACE and increased circulating ACE2 activity promotes higher ACE2/ACE ratio. In contrast, reduced levels of membrane ACE2 (after attachment of coronavirus) and an increase of local ACE activity, results in reduced local ACE2/ACE ratio, favoring ACE axis in the system [21]. Thus, RAS dysregulation leads to the overactivation of the Ang II–angiotensin II type I receptor (AT1R) axis, which is characterized by a prominent vasoconstriction, triggering profibrotic and proinflammatory signalization in the lungs and other organs [19, 21]. Here, increased ACE2/ACE ratio was observed only in the severe group (Fig. 5), suggesting that this imbalance could influence the worsening of symptoms and the poor outcome of the disease.
In addition to the RAS impact in COVID-19, the inflammatory response plays a significant role in predicting the outcome of patients. Our analyses of DPPIV (Fig. 6) and PREP (Fig. 7) enzymes showed decreased activity of both peptidases in moderate and severe groups. Both PREP and ACE2 converts the pro-inflammatory ANG II into ANG 1–7, an anti-inflammatory peptide. Whereas ACE2 is more prominent in kidney and lung, PREP is the main peptidase that performs this conversion in the systemic circulation [84], suggesting a potential contribution for regulation of ANG II levels in COVID-19 [85]. The direct association between ANGII and inflammation is clearly recognized, since ANGII initiates inflammatory cascades and triggers the activation of many pro-inflammatory mediators, including reactive oxygen species (ROS), nuclear factor kappa B (NF-kB), CRP and others [86–89].
Previously published data demonstrated that in inflammatory diseases, the process of releasing DPPIV from the cell surface is inhibited, as observed in septic shock, atherosclerosis and COVID-19 [90, 91]. These findings corroborate our results, showing that patients with moderate and severe form of COVID-19 present lower activity of circulating DPPIV. These findings showing reduced circulating DPPIV in hospitalized patients with SARS-CoV-2 might help to comprehend the specific function of this enzyme in COVID-19 [92]. Accordingly, several studies investigated whether DPPIV inhibitors (DPPIVi) could affect the clinical course of COVID-19 disease [93–96]. DPPIVi have been widely used as an antidiabetic drug in the treatment of type 2 diabetes, a common comorbidity in patients with COVID-19 and a risk factor for the most severe condition of the disease (also observed in the present study) [95]. Meta-analyses studies suggest that the use of DPPIVi in patients with COVID-19 result in a reduction of mortality and clinical improvement, mostly in patients with type 2 diabetes, while some studies showed no effects of DPPIVi [92, 95, 96]. Other data also suggest a reduction in mortality and severity of COVID-19 in patients using the antidiabetic drug metformin and/or renin-angiotensin system blockers. The benefit of the DPPIVi is less pronounced when associated with these two drugs [95, 96]. However, the exact association between DPPIV, DPPIVi and coronavirus remains unclear.
Altogether, it is not surprising that the interaction of some enzymes with SARS-CoV-2 is an important regulatory factor for COVID-19 pathophysiology. Differential profile of some proteases has been reported in patients with COVID-19 and associated with clinical complications. In this context, the ACE-2 has a central importance in the pathogenesis, both as coronavirus receptor and acting in the post-infection phase [51]. Computational approaches demonstrated that DPPIV also interacts with the spike protein of SARS-CoV-2, suggesting that another enzyme might play a relevant role in virus entry [51]. The viral replication and activation/cleavage of the spike protein is also dependent of proteases, such as ACE-2 and CAT L [71, 72]. Mediated by CAT L, the virus achieves the cytoplasm and the infection is established, causing an extensive inflammatory response [41]. Then, the immune system presents a crucial contribution, as the cytokine storm leads to tissue damage, multiple organ failure and death [51]. Most of this damage is due to the activation of pro-inflammatory pathways, mainly pulmonary and renal, including increased levels of ANG II [84]. ACE acts directly in the conversion of ANG I to ANG II, while ACE2 and PREP convert ANG II to ANG 1–7, suggesting a potential dysregulation in the expression/shedding control of these enzymes and the renin-angiotensin system in COVID-19 patients [74, 85]. The present findings suggest a possible correlation between enzymatic activity and disease severity. Taken together, our data show that COVID-19 severity impacts the activity of different proteases in the blood and therefore the knowledge about this control might contribute to a better understanding of SARS-CoV-2 infection.