Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is known to impact the immune response,⁽¹⁾ and single-nucleotide polymorphisms (SNPs) in key genes related to host immunity and cell function contribute to the clinical variability observed in coronavirus disease 2019 (COVID-19).^(2-4) Notably, exacerbation of the immune response to SARS-CoV-2 involves the secretion of various immune mediators, which are crucial in determining the severity of COVID-19.^(2,3,5) While most individuals with COVID-19 recover without complications, a subset may prolonged symptoms or sequelae, often referred to as post-COVID-19 condition (PCC).⁽⁶⁾ According to the World Health Organization (WHO), PCC is a clinical condition characterized by the persistence of symptoms or the development of new symptoms three months after SARS-CoV-2 infection, lasting at least two months and not explained by an alternative diagnosis.⁽⁷⁾ The most common persistent symptoms of PCC have been identified, and several mechanisms to explain its pathophysiology, such as an association with a specific inflammatory profile, are currently under investigation.⁽⁸⁾

Refining the existing diagnostic approaches and exploring alternative factors contributing to the persistence of COVID-19 symptoms beyond the acute infection phase remain crucial areas of research. Host SNPs are thought to play a significant role in determining an individual's susceptibility or resistance to various viral infections, and numerous studies have investigated genetic polymorphisms in the context of COVID-19 to date.^(4,9-11) Certain studies have indicated that SNPs in genes involved in innate and adaptive immune responses, as well as in viral binding and host cell entry, are associated with the development and severity of COVID-19.^(12,13) Nevertheless, the specific SNPs that contribute to susceptibility, prognosis, and the persistence of symptoms in the post-acute COVID-19 phase remain to be fully elucidated,^(14-16) particularly given the variability in genetic profiles across different populations.^(9,10) The SNPs selected in this study were chosen based on their functional roles in the main immunological and physiological pathways relevant to COVID-19 pathogenesis. Variants in cytokine genes — such as TNF (rs1800629), IL10 (rs1800871, rs1800896), IL6 (rs1800795), and IL1A (rs1800587) — influence the expression of pro- and anti-inflammatory mediators involved in the acute response to viral infections and chronic inflammation.^(17-21) Polymorphisms in the angiotensin-converting enzyme genes - ACE (rs4646994) and ACE2 (rs2285666) - affect the expression of proteins essential for viral entry and regulation of the renin-angiotensin system, which has been implicated in disease severity.^(22-24) In addition, SNPs in the UMOD gene (rs12917707, rs13333226, rs4293393) modulate uromodulin levels, increase UMOD gene expression, and are associated with kidney function, a frequent site of COVID-19-related complications.^(25-27) Therefore, we hypothesize that these SNPs may influence the medium- and long-term pathophysiology of patients with COVID-19.

The aim of this study was to evaluate the influence of SNPs in different genes, including IL1A (rs1800587), IL6 (rs1800795), IL10 (rs1800896 and rs1800871), TNF (rs1800629), ACE (rs4646994), ACE2 (rs2285666) and UMOD (rs4293393, rs13333226 and rs12917707), on COVID-19 severity in a group of Brazilian patients during the first wave of the pandemic and in the post-acute phase of the disease, as well as to investigate associations with the development of the post-COVID-19 condition and its clinical manifestations.

Cohort 1 (acute COVID-19) comprised 112 patients hospitalized at our center with SARS-CoV-2 infection. The most common comorbidities were cardiovascular disease (59.5%), diabetes (37.5%), obesity (35.1%), and oncohematological disease (18.2%). The main complication during hospitalization was acute kidney injury (AKI), which affected 31 (27.7%) patients. Twenty-nine (25.9%) patients died during hospitalization. Regarding disease severity, 53 patients (47.2%) were classified as having severe COVID-19. This group exhibited a higher mean age compared with the moderate/mild group, along with a higher prevalence of diabetes. The demographic and clinical characteristics of cohort 1 are described in table 1.

The ACE rs4646994 ID genotype was more frequently observed in the severe COVID-19 group, as shown in table 2. In the model evaluating progression to severe COVID-19 (severe versus moderate/mild cases; adjusted for cardiovascular disease, diabetes, and age) the ID genotype was significantly associated with an increased risk of severe COVID-19 (OR=2.58; 95%CI=1.06-6.47; p=0.038), whereas the II genotype was associated with protection against severe disease (OR=0.37; 95%CI=0.14-0.90; p=0.032).

In females, the GG genotype of ACE2 rs2285666 was more frequent in severe cases (Table 2). Multivariate logistic regression confirmed that this genotype was associated with an increased risk risk of severe COVID-19 in females compared with non-severe cases (OR=4.20; 95%CI=1.22-16.36; p=0.028).

The IL1A rs1800587 AA genotype was more frequently observed among patients with mild disease, as described in table 2. In the model evaluating progression to severe COVID-19, the IL1A CT genotype was significantly associated with an increased risk of developing severe disease (OR=3.05; 95%CI=1.26- 7.71; p=0.015).

The TNF rs1800629 GA genotype was also more frequent in severe cases of COVID-19 (Table 2). When compared with non-severe cases in the multivariate logistic regression analysis, the GA genotype (OR=6.32, 95%CI=1.49-43.64; p=0.025) and the A allele (OR=4.15; 95%CI=1.16-19.66; p=0.042) were associated with an increased risk of developing severe disease.

According to ICU admission, we identified that the ACE2 rs2285666 showed statistically significant differences between groups, but only among women. In the multivariate logistic regression analysis, we observed that the ACE2 GG genotype was more frequent in women admitted to the ICU (p=0.021) (Table 3) with a significant risk effect (OR=7.41, 95%CI=2.08-31.68, p=0.003), whereas the GA genotype was more frequent in non-ICU patients (p=0.021) (OR=0.13, 95%CI=0.03-0.47, p=0.003) (Figure 1A). The G allele also showed a significant risk effect (OR=3.82, 95%CI=1.38-11.89, p=0.013) in the multivariate logistic regression analysis, whereas the A allele showed a protective effect (OR=0.26, 95%CI=0.08-0.72, p=0.013) for ICU admission in women (Figure 1A). No associations were observed for the other genes.

Next, we compared the frequency of SNPs in COVID-19 patients who required IMV and those who did not (non-IMV). Regarding IMV, we identified that IL1A rs1800587 and TNF rs1800629 showed statistically significant differences (Table 4). The IL1A CT genotype was more frequent in patients receiving IMV (p=0.016), whereas the TT genotype was more frequent in non-IMV patients (p=0.007). However, we found no difference in the multivariate logistic regression analysis for any IL1A genotypes or alleles. The polymorphic A allele of rs1800629 in TNF showed a risk effect (OR=4.29; 95%CI=1.16-20.54; p<0.001), whereas the G allele showed a protective effect (OR=0.23; 95%CI=0.05-0.85; p=0.039) (Figure 1B). In a dominant model of the A allele, the GA+AA genotypes showed a risk effect for IMV (OR=5.21; 95%CI=1.31-26.44; p=0.026).

Based on the impact of TNF, ACE2, and IL1A polymorphisms on specific outcomes in patients with acute COVID-19 (ICU admission and need for IMV), we analyzed these SNPs according to symptoms and PCC development. In this analysis, cohort 2 (n=107 patients) had a mean age of 54.70±15.18 years, and 73.8% were female. The most frequent comorbidities were hypertension (56.1%), dyslipidemia (47.7%), diabetes (30.8%), and depression (28.0%). The most commonly reported symptoms in the post-COVID-19 period were persistent cough (15.9%), dyspnea (16. 8%), shallow breathing (7.5%), and exertional fatigue (44.9%). The clinical and epidemiological data of the patients in the post-COVID-19 period are shown in the Table 1S, Supplementary Material.

The genotype frequencies of all the SNPs in cohort 2 did not deviate from the Hardy-Weinberg equilibrium. We then analyzed the SNPs associated with cough using regression analysis (Figure 1F), in which the TNF GA genotype showed a significant risk effect for the persistence of this symptom (OR=3.50, 95%CI=1.12-10.67, p=0.03), whereas the GG genotype showed a protective effect (OR=0.29, 95%CI=0.09-0.86, p=0.03). For the symptom of exertional fatigue (Figure 1E), we obtained similar results, the GG genotype showed protective effect (OR=0.38, 95%CI=0.13-0.98, p=0.049) whereas the GA genotype showed a risk effect (OR=2.62, 95%CI=1.01-7.91; p=0.049). For the other symptoms, no relevant associations were observed, corroborating the findings from the acute COVID-19 analysis and suggesting that the TNF rs1800629 polymorphism is associated with the course of the disease. In addition, the TT genotype of IL1A was identified as a risk factor for symptoms of dyspnea (Figure 1C) (OR =5.02; 95%CI 1.56-16.07; p =0.006) and shallow breathing (Figure 1D) (OR =6.61; 95%CI 1.41-31.30; p =0.013), suggesting a possible relationship between this SNP and the persistence of symptoms however no association was observed between with exertional fatigue and cough. ACE2 (rs2285666) was not statistically significant regarding the persistence of reported symptoms.

Regarding the haplotype analysis, complete linkage disequilibrium (D’ =1) was detected between IL10 rs1800871 and rs1800896. The allele combinations observed in our total sample were T-A (36.2%), C-G (34.0%), and T-G (29.8%). For UMOD polymorphisms, rs12917707 was in complete linkage disequilibrium with rs13333226 and rs4293393. UMOD rs13333226 and rs4293393 were almost in complete linkage disequilibrium (D’ =0.971). The UMOD haplotypes (rs12917707-rs13333226-rs4293393) showed a higher frequency of the G-A-A haplotype (73.0%), followed by T-G-G (15.8%), while the remaining haplotypes (G-A-G, G-G-A, G-G-G) accounted for 11.2% of the haplotype distribution in the total sample (Table 2S, Supplementary Material). We compared the haplotypes with clinical outcomes, including hospitalization, ICU admission, and IMV, however, no associations were observed. The complete results are provided in the Tables 3S, 4S and 5S, Supplementary Material.

The mechanisms involved in the pathogenesis of COVID-19 have already been widely explored. However, some aspects of the acute and post-acute phases of the disease remain poorly understood. As previously reported, variations in a single nucleotide in the genome sequence may explain individual clinical susceptibility to COVID-19.⁽¹⁴⁾ In this study we aimed to investigate whether SNPs in different genes, including IL1A (rs1800587), IL6 (rs1800795), IL10 (rs1800896 and rs1800871), TNF (rs1800629), ACE (rs4646994), ACE2 (rs2285666) and UMOD (rs4293393, rs13333226 and rs12917707) were associated with the COVID-19 severity during the acute phase or with the persistence or emergence of symptoms in the post-COVID phase in Brazilian patients.

ACE and its homolog, ACE2, are two essential enzymes responsible for generating bioactive peptides within the renin-angiotensin system (RAS).⁽³³⁾ Both ACE and ACE2 play pivotal roles in regulating inflammatory processes.⁽³⁴⁾ The I allele of ACE rs4646994 (I/D) has been associated with reduced circulating levels of ACE, potentially due to the inhibition of RNA polymerase II-mediated mRNA transcription or alternative splicing mechanisms that produce a truncated ACE protein lacking one of its active sites.^(35,36) In your study, we found a significant association between the DD genotype and protection against severe COVID-19. Yenmis et al.⁽³⁷⁾ also reported that the DD genotype and protective against severe SARS-CoV-2. However, previous studies have shown conflicting results,^(33,38) while others did not find this association.^(39,40) Alimoradi el al.⁽²²⁾ observed an association between ACE and susceptibility to COVID-19 infection, but not with disease severity. Therefore, further studies are needed to clarify the true impact of this polymorphism on COVID-19 severity.

ACE2 encodes the primary receptor for SARS-CoV-2 entry into host cells. The rs2285666 polymorphism in ACE2 is a splice region variant and has been shown to increase serum levels of ACE2,⁽⁴¹⁾ potentially facilitating viral entry and increasing disease susceptibility and severity.⁽⁴²⁾ In our study, the G allele of ACE2 (particularly the GG genotype) was associated with a higher risk of COVID-19 and ICU admission in women. The ACE2 rs2285666 polymorphism has previously been associated with hypertension.^(22,43) Therefore, it is important to note that we did not observe a difference in the frequency of hypertension between sexes (p=0.351). Möhlendick et al. reported that patients carrying the G allele or the GG genotype had a threefold increased risk of developing severe COVID-19.⁽⁴¹⁾ On the other hand, results from a recent meta-analysis did not confirm this association.^{(14, 38)}

Nevertheless, the variability in these results may be attributed to differences in clinical characteristics, epigenetic mechanisms regulating ACE2 receptor expression, or variations in other genes, such as those involved in immune response or coagulation pathways, which may influence disease prognosis.⁽⁴⁴⁾ Furthermore, it is important to note that we did not observe an association between this SNP and the persistence or development of long-term symptoms. Similarly, Fernández-de-Las-Peñas et al. found no significant association with symptom persistence in individuals who had previously been hospitalized, but only with the severity of the acute infection.⁽⁴⁵⁾

In the present study, we also identified two genes associated with with COVID-19 severity and the need for IMV as clinical outcomes, IL1A (rs1800587, CT genotype) and TNF (rs1800629, GA+AA genotypes). The rs1800587 polymorphism is located in the promoter region of the IL1A and has been associated with increased IL-1α expression, which may intensify the inflammatory response, and create a favorable environment for chronic inflammation.⁽⁴⁶⁾ When we performed the multivariate analysis controlling for potential confounders, we confirmed the significant association between IL1A rs1800587 and COVID-19 severity, but not with the need for IMV. To our knowledge, there are no studies investigating SNP in IL1A and the need for IMV in patients with COVID-19.⁽⁴⁷⁾ However, in our analysis of the post-COVID period, IL1A rs1800587 was significantly associated with dyspnea and shortened of breath. The literature has already described a relationship between IL1A rs1800587 and diseases caused by other coronaviruses.⁽⁴⁸⁾These findings highlight the importance of further investigating how genetic variations influence recovery and potential complications after COVID-19, as studies addressing the association of this SNP with the development of PCC remain scarce.

Lastly, rs1800629, located in the promoter region of the TNF gene, is associated with increased production of TNF-α and may exacerbate systemic inflammation.^(49,50) In the present study, the A-allele of this SNP was associated with a higher risk of severe COVID-19 and the need for IMV, while the G allele showed a protective effect. When analyzing genotypes, this association was further supported, as patients carrying the GG genotype showed a reduced risk of severe COVID-19 or requiring IMV. The TNF gene is located within the class III region of the major histocompatibility complex (MHC) on chromosome 6p21.3.⁽⁵¹⁾ TNF-α is well-known for its pro-inflammatory properties and, together with IL-1 and IL-6, plays an important role in the inflammatory response to COVID-19.^(52-54) Moreover, the rs1800629 has been associated with sepsis.^(55,56) A recent meta-analysis evaluating different TNF polymorphisms found a higher frequency of the AA and GA genotypes among patients who required IMV in COVID-19, however this difference was not statistically significant.⁽⁵⁶⁾ In a study conducted in Egypt in 2020, individuals carrying the A-allele (GA and AA genotypes) were more susceptible to COVID-19, and the AA genotype was associated with severe disease and the need for IMV.⁽⁹⁾ Importantly, it has been documented that the A allele of in the TNF promoter region (rs1800629) leads to increased transcriptional activity, resulting in elevated production of TNF-α by B and T cells.⁽⁵⁷⁾ Consistent with our findings, these results suggest that the association of the A allele with severe COVID-19 may be explained by a dysregulated TNF-α production.

We also observed an association between TNF rs1800629 and PCC. In the study of Fernández-de-Las-Peñas et al, no association between rs1800629 and post-COVID-19 symptoms was observed.⁽⁵⁸⁾ However, our findings suggest that this SNP influences both the acute phase and symptom persistence, particularly in the respiratory tract. This is supported by evidence linking TNF-α to asthma, acute respiratory distress syndrome, and chronic obstructive pulmonary disease, conditions in which TNF-α is directly involved in inflammatory processes affecting the alveoli and bronchi.⁽⁵⁹⁾ Furthermore, as reported by Saleh et al, the A allele and the GA genotype of TNF are directly associated with acute COVID-19, however, our findings suggest that they are also associated with symptom persistence in the post-COVID phase, as this allele leads to increased transcriptional activity and elevated production of TNF-α by B and T cells.^(49,57)

In this study, we did not aim to evaluate the combined effect of SNPs and comorbidities on COVID-19 severity, however, previous studies have demonstrated associations between genetic variations and comorbidities strongly related to COVID-19. Data from genome-wide association studies (GWAS) have identified genetic variants associated with COVID-19 severity in patients with asthma, obesity, and type 2 diabetes mellitus.^(60,61) Other case-control studies have reported associations between SNPs, and increased risk of hypertension, obesity, diabetes, and dyslipidemia in the context of COVID-19.^(62,63) Shared molecular mechanisms underlying the co-occurrence of COVID-19 and conditions such as osteoarthritis, obesity, and type 2 diabetes have also been investigated.^(64-67) Therefore, a combination of multiple genes and non-genetic factors is likely involved in COVID-19 severity and its associated comorbidities. Elucidating reliable markers of COVID-19 severity in patients with comorbidities is crucial for developing personalized treatments strategies.

This study has some limitations, mainly the relatively sample size. In addition, our cohort primarily consisted of patients with multiple comorbidities, as our center is a tertiary referral hospital. Furthermore, comparison with a healthy control group was not possible. Likewise, despite these limitations, our study identified associations between genes related to cytokines and renin-angiotensin-aldosterone system (RAAS) and COVID-19 severity, as well as the persistence of symptoms in the post-COVID-19 period.

Our findings indicate an important role for variations in ACE2, IL1A, and TNF in the risk of developing severe COVID-19, as well for IL1A and TNF in the persistence of symptoms in the post-COVID-19 phase among Brazilian patients. The A allele of TNF rs1800629 may represent a risk variant for disease severity, the need for IMV during the acute phase of COVID-19, and the persistence of cough and fatigue, suggesting a link between acute disease severity post-COVID-19 condition. The IL1A rs1800587 polymorphism was associated with disease severity during the acute phase and with the presence of dyspnea and shallow breathing after COVID-19. In contrast, ACE2 polymorphisms in women were associated with disease severity, but not with symptom persistence.