Gene Expression Profile of Immune Response Markers Associated with Long COVID and Its Clinical Aspects in a Cohort from Northern Brazil

Gene Expression Profile of Immune Response Markers Associated with Long COVID and Its Clinical Aspects in a Cohort from Northern Brazil | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 14 March 2025 V1 Latest version Share on Gene Expression Profile of Immune Response Markers Associated with Long COVID and Its Clinical Aspects in a Cohort from Northern Brazil Authors : Leonn Mendes Soares Pereira , Izailton de Souza e Souza , Wandrey Roberto dos Santos Brito , Erika Ferreira dos Santos , Flávia Póvoa da Costa , Kevin Matheus Lima de Sarges , Marcos Henrique Damasceno Cantanhede , … Show All … , Mioni Magalhães de Brito , Andréa Luciana Soares da Silva , Mauro de Meira Leite , Maria de Nazaré do Socorro de Almeida Viana , Fabíola Brasil Barbosa Rodrigues , Rosilene da Silva , Giselle Maria Rachid Viana , Tânia do Socorro Souza Chaves , Adriana de Oliveira Lameira Veríssimo , Mayara da Silva Carvalho , Daniele Freitas Henriques , Carla Pinheiro da Silva , Juliana Abreu Lima Nunes , Iran Barros Costa , Izaura Cayres-Vallinoto , Igor Brasil-Costa , Juarez Antonio Quaresma 0000-0002-6267-9966 , Andrea Nazare Monteiro Rangel da Silva , Maria Alice Queiroz 0000-0003-2014-2770 , Eduardo José Melo dos Santos , Luiz Fábio Magno Falcão , and Antonio Carlos Vallinoto 0000-0003-1135-6507 [email protected] Show Fewer Authors Info & Affiliations https://doi.org/10.22541/au.174195200.04677881/v1 489 views 236 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Current efforts are focused on the search for biomarkers associated with long COVID. In this study, we evaluated the gene expression of 15 biomarkers and their relationship with the clinical aspects of the condition. c-DNA samples from 15 patients with long COVID, 15 recovered patients (RWS), and 15 patients with symptomatic acute COVID-19 were analyzed. The relative expression of genes was determined by the 2-ΔΔCT method from real-time PCR. Sociodemographic and clinical data of interest were extracted from medical records. Of the 15 biomarkers, only the expression of TREX1, FOXP3, MYD88 and FASL was not associated with long COVID. The genes IRF7, IRF3, and IFI16 performed best as biomarkers of long COVID (AUC≥0.90, p≤0.05). With the exception of MDA5 and RIG-1 genes, the expression of the other eight genes was associated with the presence of comorbidities, medication use, and complaints of fever, abdominal pain, eye pain, and headache (H>9.0; p≤0.05). IRF3 expression was specifically associated with long COVID when compared to acute COVID (med.: 41.2; IQR: 116.20; p: 0.0036). Our results suggest that classical immune response genes are upregulated in long COVID and that certain clinical aspects of the disease may influence the expression profile of the studied genes. Gene Expression Profile of Immune Response Markers Associated with Long COVID and Its Clinical Aspects in a Cohort from Northern Brazil Leonn Mendes Soares Pereira 1 , Izailton de Souza e Souza 1 , Wandrey Roberto dos Santos Brito 1 , Erika Ferreira dos Santos 2 , Flávia Póvoa da Costa 2 , Kevin Matheus Lima de Sarges 2 , Marcos Henrique Damasceno Cantanhede 2 , Mioni Thieli Figueiredo Magalhães de Brito 2 , Andréa Luciana Soares da Silva 2 , Mauro de Meira Leite 2 , Maria de Nazaré do Socorro de Almeida Viana 2 , Fabíola Brasil Barbosa Rodrigues 2 , Rosilene da Silva 2 , Giselle Maria Rachid Viana 3 , Tânia do Socorro Souza Chaves 3 , Adriana de Oliveira Lameira Veríssimo 4 , Mayara da Silva Carvalho 4 , Daniele Freitas Henriques 5 , Carla Pinheiro da Silva 5 , Juliana Abreu Lima Nunes 6 , Iran Barros Costa 6 , Izaura Maria Vieira Cayres-Vallinoto 1 , Igor Brasil-Costa 6 , Juarez Antônio Simões Quaresma 7 , Andrea Nazare Monteiro Rangel da Silva 1 , Maria Alice Freitas Queiroz 1 , Eduardo José Melo dos Santos 2 , Luiz Fábio Magno Falcão 7 , Antonio Carlos Rosário Vallinoto 1 1. Laboratory of Virology, Institute of Biological Sciences, Federal University of Pará, Belém, Brazil; 2. Laboratory of Genetics of Complex Diseases, Institute of Biological Sciences, Federal University of Pará, Belém, Brazil; 3. Basic Research Laboratory on Malaria, Parasitology Section, Evandro Chagas Institute, Secretariat for Health and Environmental Surveillance, Brazilian Ministry of Health, Ananindeua, Brazil; 4. Adventist Hospital of Belém, Belém, Brazil; 5. Arbovirology and Hemorrhagic Fevers Section, Evandro Chagas Institute, Secretariat for Health and Environmental Surveillance, Ministry of Health, Ananindeua, Brazil; 6. Immunology Laboratory, Virology Section, Evandro Chagas Institute, Secretariat for Health and Environmental Surveillance, Brazilian Ministry of Health, Ananindeua, Brazil; 7. Center for Biological and Health Sciences, Pará State University, Belém, Brazil. * Correspondence: Antonio Vallinoto ( [email protected] ) Funding: The study was supported by the National Council for Scientific and Technological Development (CNPQ #401235/2020-3), (process 153000/2022-8 – CNPQ 25/2021) and (process 153643/2024-2 – CNPQ 58/2022); the Coordination for the Improvement of Higher Education Personnel - Brazil (CAPES), grant no. “Financial code 001, notice no. 13/2020”; Fundação Amazônia de Amparo a Estudos e Pesquisa do Pará (FAPESPA #005/2020 and #006/2020) and Secretaria de Estado de Ciência, Tecnologia e Educação Profissional e Tecnológica (#09/2021). Current efforts are focused on the search for biomarkers associated with long COVID. In this study, we evaluated the gene expression of 15 biomarkers and their relationship with the clinical aspects of the condition. c-DNA samples from 15 patients with long COVID, 15 recovered patients (RWS), and 15 patients with symptomatic acute COVID-19 were analyzed. The relative expression of genes was determined by the 2-ΔΔCT method from real-time PCR. Sociodemographic and clinical data of interest were extracted from medical records. Of the 15 biomarkers, only the expression of TREX1, FOXP3, MYD88 and FASL was not associated with long COVID. The genes IRF7, IRF3, and IFI16 performed best as biomarkers of long COVID (AUC≥0.90, p≤0.05). With the exception of MDA5 and RIG-1 genes, the expression of the other eight genes was associated with the presence of comorbidities, medication use, and complaints of fever, abdominal pain, eye pain, and headache (H>9.0; p≤0.05). IRF3 expression was specifically associated with long COVID when compared to acute COVID (med.: 41.2; IQR: 116.20; p: 0.0036). Our results suggest that classical immune response genes are upregulated in long COVID and that certain clinical aspects of the disease may influence the expression profile of the studied genes. Key words: Long COVID, gene expression, biomarkers. 1. Introduction Long COVID is characterized by a multisystemic and complex disease, the symptoms of which can last for more than three months after the onset of severe acute respiratory syndrome caused by the severe acute respiratory syndrome coronavirus 2 - SARS-CoV-2 [1]. In Brazil, it is estimated that around 4 million people have long COVID and, given the wide spread of the infection, the burden of the disease remains relevant, challenging the sustainability of research agendas and generating concerns about the demobilization of various stakeholders and the increased invisibility of the condition, particularly in contexts of high social vulnerability in the country [2]. Compiling the pathophysiological mechanisms that may be potentially related to long COVID is a hard and dynamic task, however, it is already recognized that this condition is due both to the maintenance of long-term tissue damage and to an unresolved chronic inflammatory state, which manifests itself mainly through neurological changes, cardiovascular damage, olfactory and taste dysfunctions, fatigue states and the impairment of the respiratory, musculoskeletal, gastrointestinal and hepatobiliary systems [3]. Due to the complexity of the condition, a diverse range of potential biomarkers have been investigated. In our previous studies, we showed the association of serum levels of the cytokines IFN-α, IL-17, IL-2, IL-4 and IL-10, as well as cGAS and STING gene expression with long COVID, although they were not associated with the symptoms or the socioepidemiological aspects analyzed [4,5]. In the present study, we proposed to evaluate, in a cohort from the northern region of Brazil, the expression of 15 genes that modulate the immune response, arranged in four functional groups: cytosolic sensors ( SAMHD-1, TREX-1, RIG-1, MDA-5 and IFI16 ), adaptor molecule and signaling transducer ( MYD88 and MAVS ), apoptotic signaling factors ( FAS and FASL ) and transcription factors ( FOXP3, IRF3, IRF7, RELA, RELB and NFKβ1 ). Additionally, a systematic approach to markers, taking into account plausible interactions between them, led us to better understand which signaling cascades are activated in COVID. In the present study, we observed that the expression of 11 genes was associated with long COVID and the clinical aspects related to this condition. 2. Material and Methods 2.1 Sampling and ethical aspects Peripheral blood samples were analyzed from 15 patients diagnosed with long COVID, 15 patients recovered and without sequelae of COVID19 (RWS) and 15 patients with COVID19 in its acute stage treated at specialized collection and testing centers and hospital wards of Hospital Adventista de Belém, Hospital Amazônia, Instituto Evandro Chagas and Universidade do Estado do Pará located in the metropolitan region of the city of Belém, Pará, Brazil between the years 2020 and 2022. Long COVID group was defined based on the following aspects: I- acute symptoms compatible with COVID-19 without attribution to other diseases, II- COVID-19 confirmed by viral detection through quantitative reverse transcription polymerase chain reaction or positive result on antibody testing for anti-SARS-CoV-2 IgG/IgM, and III- at least one symptom of long Covid related to COVID-19 for at least 4 weeks after the onset of symptoms (e.g. loss of smell or taste, muscle pain, headache, post exertion malaise, chronic cough, brain fog, heart palpitations, chest pain, shortness of breath, fatigue, dizziness, and tremors). RWS group was patients in the post-COVID stage, with positive serology for anti-SARS-CoV-2 IgG, negative serology for anti-SARS-CoV-2 IgM, negative detection of antigens or viral genetic material in the mucosa and absence of signs and symptoms indicative of post-COVID sequelae for at least three months from the period of resolution of the active infection. The group with symptomatic acute COVID-19 was defined as patients with positive detection of antigens or viral genetic material in the mucosa and positive serology for IgM anti-SARS-CoV-2. Pulmonary radiological findings suggestive of the disease were present in most individuals, as reported in our previous study [6]. These patients were classified according to the severity of the condition according to the management standards established by the World Health Organization [7], among which we observed that nine patients were in a severe condition and six were in a non-severe condition of the disease. None of the patients in this group had yet been immunized in the vaccination campaigns offered. These samples were collected exclusively between the years 2020 and 2021. In order to establish a reference pool for the execution of the gene expression method adopted in this study, we collected samples from 15 health professionals and/or laboratory technicians with a profile similar to the RWS group and who had their anti-SARS-CoV-2 immunization records updated according to the recommendations of the vaccination campaigns offered. Sociodemographic and clinical data regarding sex, age, symptoms, signs of primary immunodeficiency (PID), pulmonary radiological findings, smoking, presence of comorbidities and medications in use were extracted from medical records, anamnesis and questionnaires specific to the research project. This study was approved by the Research Ethics Committee of the Institute of Health Sciences of the Federal University of Pará (CAEE: 33470020.1001.0018), protocol no. 2.190.330. All individuals were informed about the research objectives and those who agreed to participate in the study signed the informed consent form in accordance with the guidelines of CNS Resolution No. 466/2012 and the Declaration of Helsinki. 2.2 RNA extraction and conversion to c-DNA RNA was extracted from blood samples using a TRIzol™ Plus RNA purification kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA) following the manufacturer’s recommendations. RNA concentration was determined using a BioDrop™ (Bio-Rad, Hercules, California, USA) following the manufacturer’s recommendations. RNA sample concentrations were standardized to 40 ng/µL. RNA was converted to cDNA using a cDNA reverse transcription kit with RNAse inhibitor (High-Capacity cDNA Reverse Transcription with RNAse Inhibitor-Applied Biosystems, Foster City, CA, USA), as published in previous studies [5]. 2.3 Gene expression The relative expression of target genes was determined using the 2-ΔΔCT analysis method from a real-time PCR platform, where ΔΔCT = ΔCT of the tested sample - ΔCT of the reference pool (Life Technologies, Carlsbad, California, USA), as published in previous studies [5]. The endogenous normalizing gene for the method was Glyceraldehyde-3-phosphate dehydrogenase (GAPDH). 2.4 Statistical analysis The comparison of sociodemographic and clinical data between the study groups was performed using Fisher’s exact test, G test and chi-square (χ 2 ) test, following the recommendations for application of each test. The median of the relative expression of the target genes was associated with the study groups using the Mann-Whitney test, in 2 by 2 comparisons, and by the Kruskal-Wallis analysis of variance, supplemented by Student-Newman-Keuls, in analyses with 3 or more groups. A heat plot showing the clustering of the normalized gene expression data between the groups was inferred, adopting the Ward d2 method as the distance type and the Canberra method as the aggregation type. The correlation between the medians was compared using the Spearman correlation test. We chose nonparametric tests due to the degree of normality of the data, which were calculated using the Lilliefors test. ROC curves were inferred to evaluate the performance of the studied genes in the classification and differentiation of long COVID and RWS cases through the sensitivity and specificity obtained from each gene expression. We used the Wilson/Brown method for curve inference. We applied multivariate correspondence analysis (MCA) to infer within the group with long COVID the presence of subgroups of individuals with similar profiles and the relationship between sociodemographic, clinical and gene expression variables with the subgroups formed; the quality of the representation of the variables was based on the calculation of the squared cosine (cos2). We subsequently tested the degree of dependence through simple and multiple linear regression between the relative expression of genes with the sociodemographic and clinical variables co-categorized on the same axis inferred in the MCA. We determined the degree of contribution of the studied gene expressions in accounting for principal component variability (PCA) in the group with long COVID and in the group with acute COVID. Heat maps and MCA and PCA analyses were created in RStudio 4.0.1 software [8] using the gplots, RColorBrewer and preprocessCore, FactoMineR and factoextra packages, respectively. Statistical analyses were performed using Bioestat 5.3 [9] and GraphPad Prism 5.0 (GraphPad Software, Boston, Massachusetts, USA) programs, with a significance level (α) of 5% (p≤0.05). 3. Results 3.1 Characterization of groups with long COVID and RWS In the RWS group, there was a smooth predominance of female individuals (53.33%); aged between 21 and 42 years (60%); without evidence of PID (100%); with normal pulmonary radiological findings (53.33%); non-smokers (86.67%); without comorbidities (73.33%); and who were not taking medications (66.67%). In the group with long COVID, there was a predominance of female individuals (66.67%); aged between 43 and 64 years (60%); without signs of PID (66.67%); with normal lung radiological findings (53.33%); non-smokers (73.33%); without comorbidities (80%); and who were not taking medications (73.33%). Regarding symptoms, the most frequent complaints were fatigue (100%) and myalgia (86.67%). All data on sociodemographic and clinical characteristics, their respective absolute count and percentage in each group are shown in Table 1. 3.2 Expression of 11 genes associated with the long COVID The relative expression of most genes was higher in the long COVID group. However, the expression of the TREX1 gene was higher in the RWS group and the expression of the FOXP3 , FASL and MYD88 genes was similar between the two groups (Figure 1A). In a group analysis, the IRF3 gene was the most expressed in the long COVID group (Figure 1B), while in the RWS group the TREX1 and MYD88 genes were the most expressed (Figure 1C). These data were reflected in the clustering pattern shown in the heat map, in which we observed a tendency for higher relative expression levels to cluster among individuals with long COVID (Figure 1D). The central measurement data and ranges of variation of relative gene expression in each group are shown in Supplementary Table 1. Therefore, in the long COVID group, gene expressions were positively correlated with each other. The FOXP3 gene was correlated only with the FASL gene (r: 0.50, p: 0.058), while the MYD88 gene was correlated with IFI16 (r: 0.70, p: 0.008) and IRF3 (r: 0.56, p: 0.028) genes (Figure 1E). Spearman’s coefficient (rs) data and the respective statistical significance (p value) are shown in Supplementary Table 2. The ROC curve models inferred for the 11 genes associated with long COVID were statistically significant (Figure 2). However, the area under the curve above 0.90 was observed only in the models of the genes IRF7 (AUC: 0.94, p < 0.0001) (Figure 2A), IRF3 (AUC: 0.92, p < 0.0001) (Figure 2B) and IFI16 (AUC: 0.91, p: 0.0001) (Figure 2C), which suggests that the high expression of these genes performed well as a biomarker of long COVID compared to RWS. For the IRF7 gene, when the relative expression of the sample was above 1, there was approximately 93% sensitivity and 80% specificity of being a long COVID sample. For the IRF3 gene, when the relative expression of the sample was above 1.5, there was approximately 93% sensitivity and 87% specificity of it being a long COVID sample. For the IFI16 gene, when the relative expression of the sample was above 0.3, there was approximately 93% sensitivity and 80% specificity of it being a long COVID sample. All statistical data regarding the inferred ROC curve models are shown in Supplementary Table 3. 3.3 Relative gene expression was associated with clinical features of long COVID In the proposed MCA model, the 15 individuals with long COVID were distributed along a two-dimensional four-quadrant plane. This model accounted for 60% of the data variability (Figure 3A). Among the co-categories of variables, the expression data of the MAVS , MDA5 and RELA genes had better representation quality (cos2) of quadrants 1 and 3 (Figure 3B). We tested the degree of dependence of gene expression with their respective co-categories of clustering on the same axis/quadrant of the MCA. With the exception of the expression of the MDA5 and RIG-1 genes, the other genes in both networks were inversely proportional to the use of medications and the presence of comorbidities, specifically for the expression of the RELB and FAS genes, in which the observed trend was also maintained in the multiple regression model. The models indicated that abdominal pain, fever, headache and ocular pain were inversely proportional to the expression of the genes SAMHD1 , RELB , FAS and IFI16 respectively. All statistical data related to the simple and multiple linear regression models for each data crossing are shown in tables 2 and 3. We describe the symptoms of the four patients with comorbidities/undergoing treatment (Figure 3C), however, only complaints of pain behind the eyes (3 of 4, 75%; p: 0.0330) and headache (4 of 4, 100%; p: 0.0769) were either associated or tendentiously associated with comorbidity/treatment (Figure 3D). We reinforced these findings through the Kruskal-Wallis test, in which we observed that: I- The expression of the RELB (Figure 3E, supplementary table 04), SAMHD1 (Figure 3F, supplementary table 04), FAS (Figure 4A, supplementary table 04) and IFI16 (Figure 4B, supplementary table 4) genes was higher in patients without complaints of abdominal pain, fever, headache and pain behind the eyes, respectively. II- The expression of the IRF3, IRF7, IFI16, RELA, MAVS, FAS, NFKβ, RELB and SAMHD1 genes was higher in patients without comorbidities/without therapy, when compared to patients with comorbidities/on therapy and the RWS groups (Figure 4C, supplementary table 4). We describe the classes of drugs in use and their respective active pharmaceutical ingredients (Figure 4D). We show the gene expression levels according to the drug regimen used and the types of alleged comorbidities (Figure 4E). Among the patients who were using combined therapeutic regimens, we did not observe a history of pharmacological interaction between the active ingredients of these drugs, according to the data source on the website “Drugs.com” (https://www.drugs.com/). We observed a tendency for patients under treatment for less than 6 months to maintain high expression of genes associated with the presence of comorbidity/therapy (Figure 4F). 3.4 Higher relative expression of the IRF3 gene in long COVID We initially compared the clinical characteristics of patients with long COVID and acute COVID. Among these, disease severity (60%; p: 0.0007), the presence of abnormal radiological findings (100%; p>0.0001), the presence of comorbidities (60%, p: 0.00341) and the use of medications (100%; p>0.0001) were more frequent in patients with acute COVID (Figure 5A-D). The relative expression of the genes IRF3 (med.: 41.2; IQR: 116.2; p: 0.0036) and IRF7 (med.: 14.2; IQR: 23.55; p: 0.0295) was higher in the group with long COVID than in the group with active COVID ((med.: 1.39; IQR: 9.85), (med.: 1.11; IQR: 11.75), respectively) (Figure 5F). We stratified the acute COVID group according to symptom severity to test the maintenance of the observed statistical significance. We noted that IRF3 gene expression remained higher in the long COVID group (med.: 41.2; IQR: 116.20) when compared to the severe acute COVID group (med.: 1.39; IQR: 16.97; p: 0.0312) and the non-severe acute COVID group (med.: 1.52; IQR: 7.00; p: 0.0483); however, IRF7 gene expression remained significant only when we compared the long COVID group (med.: 14.20; IQR: 23.55) with the non-severe acute COVID group (med.: 0.84; IQR: 0.41; p: 0.0364) (Figure 5G). We inferred correlation plots between variables with the principal components for the relative gene expression data in the groups with acute COVID (Figure 5H) and long COVID (Figure 5I). In both models, the representation of the principal components in two dimensions covered more than 50% of the data variability and all gene expressions were positively correlated. In the active COVID plot, the expression of the TREX1 gene had the smallest contribution to the formation of the principal components, while the MAVS , RELA and RIG-1 genes had the largest contributions to the principal components (Figure 5H). In the long COVID plot, the expression of the MYD88 gene had the smallest contribution to the composition of the principal components, while the expression of the IRF3 gene had the largest contribution (Figure 5I). 4. Discussion 4.1 Gene expression associated with long COVID In the present study, we observed that the expression of 11 of the 15 genes investigated was elevated in patients with long COVID. This is expected, since the expression of immune genes reflects the establishment of a prolonged pro-inflammatory environment after SARS-CoV-2 infection [10]. However, we observed that the genes FOXP3, FAS, MyD88 and TREX-1 were not associated with the condition. Since long COVID is characterized by systemic immunopathological mechanisms [11], the low relative expression of the FOXP3 gene observed is justifiable [12]. Dhawan and colleagues propose that the increased number of Tregs in COVID-19 may have deleterious effects by limiting the activity of effector cells. Furthermore, overexpressed FOXP3 gene may lead to excessive immunosuppressive activities [13]. Regarding markers of cellular apoptosis, we observed that, while FAS gene expression was associated with long COVID and correlated with gene expression of the other markers investigated, FASL gene expression did not show the same outcome. Studies show that low FasL gene expression is associated with less severe cases of COVID [14]; and in this context, we emphasize that in our long COVID group, all patients did not present severe symptoms and most did not claim to have comorbidities that could have a poor prognosis. The discordant relationship between FAS and FASL seems to be somewhat curious; however, in studies based on different models of oncogenic malignancies, it was observed that only FASL gene expression, but not FAS gene, was associated with more aggressive pathological stages [15]. In this context, it seems plausible to argue that in more severe profiles of COVID-19, FASL gene upregulation incites overstimulation of the Fas apoptosis pathway. However, in milder profiles, as in our long COVID cohort, only FAS is maintained, which by itself seems to be able to manage the activation of the pro-apoptotic pathway by other inducers or by Fas receptor oligomerization [16]. Although our group with long COVID did not express significant levels of the MYD88 gene, the expression of NFKβ, IRF3 and IRF7 genes, related to interferon maintenance, were associated with the condition. This suggests that in long COVID, the production of NFKβ and IFN does not appear to be dependent on MyD88. In contrast, our findings point to a correlation between the expression of these factors and the expression of the RIG-1 , MDA5 and MAVS genes, whose well-established interaction is associated with the modulation of NFKβ and IFN, and is proposed as a strategic signaling pathway for the advent of effective therapeutics against COVID-19 [17]. Among all genes evaluated, only TREX1 expression was associated with the non-development of long COVID. We understand that increased TREX1 expression may favor the response against SARS-CoV-2 through direct physiological mechanisms, since the factor can also act as a single-stranded RNA exonuclease [18], or indirect mechanisms, through a generalized antiviral pathway where infections induce stress or mitochondrial DNA release leading to the canonical activation of the TREX-1 pathway [19]. In both cases, TREX-1 modulation both restricts cellular viral load and negatively regulates inflammatory pathways, which in the context of long COVID, would limit the active immune state typical of the condition [3], thus promoting a post-COVID profile without sequelae. 4.2 Gene expression associated with clinical aspects of long COVID We observed that the presence of comorbidities and the use of medications were correlated factors with a greater degree of influence on the modulation of gene expressions. It is well established that chronic comorbidities predispose to pathogenic pro-inflammatory environments in COVID-19 [20], and in the more specific scenario of long COVID, it is particularly worrying due to the development of chronic sequelae that impair both the ability to respond to infections and exacerbate the symptoms of long COVID. In this sense, pharmacological management is foreseen as an intervention measure for comorbidities, although specific evidence is still needed in the management of long COVID [21]. As an example, we highlight that the patient with a history of depression used agomelatine and maintained low expression of some associated genes. The non-causal relationship of inflammation as a likely critical modifier of susceptibility to depression is debated; in fact, the expression of multiple cytokines in the condition has the potential to describe several aspects of depression and its management [22]. In long COVID, psychiatric disorders appear to be even more characteristic and overwhelming [23]. Thus, the administration of agomelatine favors the expected antidepressant effect and also achieves anti-inflammatory properties in neural tissue [24,25]. Naturally, this is what leads some therapeutic perspectives to propose it as a candidate for the treatment of neuroinflammation associated with depression due to long COVID (LCD) [26]. It is likely that the administration of this active principle may also be causing the modulation of immunological markers in other non-neural sites [27]. Our findings show that complaints of ocular pain and headache were associated with the presence of comorbidities/therapy, as well as with variations in the expression of the IFI16 and FAS genes, respectively. Studies suggest that both symptoms are associated with diverse etiologies and chronic systemic conditions that are aggravated by long COVID [28,29]. Therefore, the effects associated with symptoms may be the general reflection observed in the relationship between the presence of comorbidities, the use of medications and the expression of the genes evaluated. However, a direct relationship between symptoms and the expression of specific genes is intriguing. For the FAS gene, there is no conclusion about its association with headache, however, studies propose a relationship between migraine induction in guinea pigs and the observed pro-apoptotic activity [30], as well as the relationship between head trauma and the concentration of soluble Fas in humans [31]. For the IFI16 gene, the most coherent inference attributes it to the attempt to maintain the immune response against cases of herpetic stromal keratitis (HSK) due to herpesvirus reactivation [32]. More complex investigations that encompass the relationship between the presence of comorbidities, the use of medications and the associated symptom profile are essential to conclude the dynamics of these factors on the expression levels of immunological genes in long COVID. Low expression of the RELB and SAMHD1 genes was associated with fever and abdominal pain, respectively; however, these symptoms were not related to the presence of comorbidities or the use of medications that could justify the findings. In fact, low expression of the RELB gene is related to the regulation of the immune response, mainly of T cells, to states of limitation of acute inflammation and to the repression of innate immunity [33]. It is possible that patients with fever present low expression of the gene, since this symptom is common in infectious and inflammatory conditions [34]. Regarding the SAMHD1 gene, the relationship between its low expression and complaints of abdominal pain is not conclusive. On the contrary, studies suggest that in the abdominal region, SAMHD1 gene expression varies in a medium-high range in order to confer additional protection to the regions of entry of STDs [35], therefore, it is expected that peritoneal clinical changes are reflected in increased gene expression. In long COVID, abdominal pain accounts for 14% to 20% of gastrointestinal symptoms [36]. However, the multiple facets of pain symptoms in patients with long COVID are still under discussion, and some of these symptoms remain unexplained [37]. 4.3 IRF3 gene expression as a specific biomarker of long COVID Our data certainly indicate that immunological biomarkers are activated in long COVID, among which genes directly related to interferon maintenance were those that showed the best performance in distinguishing long COVID cases. Studies suggest that interferon regulation in long COVID can remain high even without direct stimuli and in a manner dependent on symptoms and patient age [38]. Interestingly, we observed that high expression of the IRF3 gene was the most specific marker of long COVID when compared to acute COVID, although the latter condition is characterized by more severe conditions. One of the molecular mechanisms that sustain long COVID is the persistence of SARS-CoV-2 in the cell nucleus by distinct means that can promote, under specific conditions, the formation of chimeric genes that, when expressed, can induce a chronic pro-inflammatory environment [39]. In this context, we suggest that the IRF3 signaling cascade, which is ubiquitously expressed [40], may be recruited to maintain type 1 interferon that is observed in patients with long COVID [5]. It is likely that while these mechanisms of persistence in the nucleoplasm are ongoing, patients with acute COVID may manifest low levels of interferon and its respective activation factors due to immune escape promoted directly by SARS-CoV-2 proteins [41], especially in the most severe cases of the disease [42]. It is worth noting that the expression of all PRRs related to the IRF3 pathway studied in the present study were similar between the different COVID conditions. However, we emphasize that although we did not address the direct analysis of TLR receptors, we indirectly showed that MYD88 gene expression was not associated with long COVID, which by inference suggests that only the TLR3 receptor may be related to the condition, since this receptor immediately signals to TRIF, but not to MyD88, in the cascade sequence. TRIF, in turn, signals to NAP1, which sequentially recruits the factors TRAF3, TBK1 and IKK-cc, which results in the phosphorylation and dimerization of IRF3 and subsequent production of interferon [43]. Indeed, it has been shown that systemic administration of exosomes carrying SARS-CoV-2 genetic material can increase TLR3 expression, and administration of a TLR3 inhibitor leads to a deficit in the immune response to the stimulus [44]. However, a specific analysis of the TLR3 pathway in long COVID is needed to ratify the discussion raised. Based on our findings, we conclude that the expression of classical immune response genes is elevated in long COVID when compared to the post-COVID state without sequelae. The presence of comorbidities, drug interventions and some specific symptoms are clinical components that contribute to the transcriptional dynamics of the genes studied. Mainly for the IRF3 gene, which has been suggested as a specific biomarker of long COVID, even in the face of more severe cases of acute COVID. We point out as the main limitation of the present study the low sample size of the groups, especially in the subgroups generated in the resulting analyses. However, with the volume of inferences generated, we collaborated by ratifying the thesis that a pro-inflammatory environment observed in different populations in the long COVID scenario is also sustained in a northern Brazilian cohort. Author Contributions ACRV, LFMF, IMVC-V and EJMS conceived of the project. LMSP and MAFQ wrote and reviewed the manuscript. LMSP and MAFQ performed the statistical analyses. ISS, WRSB, EFS, FPC, KMLS, MHDC, MTFMB, ALSS, MML, MNSAV, FBBR, RS, GMRV, TSSC, AOLV, MSC, DFH, CPS, JALN, IBaC, IBr-C, JASQ and ANMRS collected the biological samples and performed the laboratory analyses. All authors reviewed and approved the article. Acknowledgments We thank the patients who agreed to participate in this study. Ethics Statement This study was approved by the Ethics Committees for Research Involving Human Subjects of the Federal University of Pará (opinion no. 2.190.330) and the State University of Pará (opinion no. 4.252.664). All individuals were informed about the research objectives and those who agreed to participate in the study signed the informed consent form in accordance with the guidelines of CNS Resolution No. 466/2012 and the Declaration of Helsinki. Conflicts of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest. Data Availability Statement The datasets used and analyzed during the current investigation are available from the corresponding author upon request. References 1. H. E. Davis, L. McCorkell, J. M. Vogel, et al. Long COVID: major findings, mechanisms and recommendations. Nature Reviews Microbiology 21, no. 3 (2023):133-146. doi: 10.1038/s41579-022-00846-2. 2. A. N. Ramos Jr. Long COVID challenges in Brazil: an unfinished agenda for the Brazilian Unified National Health System. Caderno de Saude Publica 19, no. 40 (2024): e00008724. doi: 10.1590/0102-311XPT008724. 3. S. J. Yong. Long COVID or post-COVID-19 syndrome: putative pathophysiology, risk factors, and treatments. Infectious diseases (London) 53, no.10 (2021):737-754. doi: 10.1080/23744235.2021.1924397. 4. M. A. F. Queiroz, P. F. M. D. Neves, S. S. Lima, et al. Cytokine Profiles Associated With Acute COVID-19 and Long COVID-19 Syndrome. Frontiers in Cellular and Infection Microbiology 12, (2022):922422. doi: 10.3389/fcimb.2022.922422. 5. M. A. F. Queiroz, W. R. D. S. Brito, K. A. S. Pereira, et al. Severe COVID-19 and long COVID are associated with high expression of STING, cGAS and IFN-α. Scientific Reports 14, no. 1 (2024): 4974. doi: 10.1038/s41598-024-55696-0. 6. K. M. L. Sarges, F. P. da Costa, E. F. Santos, et al. Association of the IFNG +874T/A Polymorphism with Symptomatic COVID-19 Susceptibility. Viruses 16, no 4 (2024):650. https://doi.org/10.3390/v16040650. 7. World Health Organization. COVID-19 Clinical Management: Living Guidance, 25 January 2021. 2021. Available online: https://iris.who.int/handle/10665/338882 (accessed on 23 March 2024). 8. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/ (2022). 9. M. Ayres, J.R.M. Ayres, D.L. Ayres, et al. BioEstat 5.0: Aplicações Estatísticas nas Áreas de Ciências Biológicas e Médicas. Sociedade Civil Mamirauá . Brasília (2008): CNPq. 10. J. Y. Zhang, J. P. Whalley, J. C. Knight, et al. SARS-CoV-2 infection induces a long-lived pro-inflammatory transcriptional profile. Genome Medicine 15, no. 1 (2023):69. doi: 10.1186/s13073-023-01227-x. 11. M. Merad, C. A. Blish, F. Sallusto, et al. The immunology and immunopathology of COVID-19. Science 375, no. 6585:1122-1127. doi: 10.1126/science.abm8108. 12. L. V. Topchieva, O. V. Balan, A. V. Men’shenin. Quantitative Assay of SARS-CoV-2 RNA and Level of Proinflammatory Protein Gene Transcripts in Peripheral Blood Leukocytes after a Novel Coronavirus Infection. Bulletin of Experimental Biology and Medicine 173, no. 6 (2022):740-744. doi: 10.1007/s10517-022-05621-0. 13. M. Dhawan, A. A. Rabaan, S. Alwarthan, et al. Regulatory T Cells (Tregs) and COVID-19: Unveiling the Mechanisms, and Therapeutic Potentialities with a Special Focus on Long COVID. Vaccines (Basel) 11, no. 3 (2023):699. doi: 10.3390/vaccines11030699. 14. M. C. Albert, I. Uranga-Murillo, M. Arias, et al. Identification of FasL as a crucial host factor driving COVID-19 pathology and lethality. Cell Death & Differentiation 31, no. 5 (2024):544-557. doi: 10.1038/s41418-024-01278-6. 15. S. Kykalos, S. Mathaiou, A. J. Karayiannakis, et al. Tissue expression of the proteins fas and fas ligand in colorectal cancer and liver metastases. Journal of Gastrointestinal Cancer 43, no. 2 (2012):224-8. doi: 10.1007/s12029-011-9252-6. 16. F. Mollinedo and C. Gajate. FasL-Independent Activation of Fas. In: Madame Curie Bioscience Database (Internet). Austin (TX): Landes Bioscience; 2000-2013. Available from: https://www.ncbi.nlm.nih.gov/books/NBK6345/. 17. M. Wang, Y. Zhao, J. Liu, et al. SARS-CoV-2 modulation of RIG-I-MAVS signaling: Potential mechanisms of impairment on host antiviral immunity and therapeutic approaches. MedComm– Future Medicine 1, no. 2 (2022):e29. doi: 10.1002/mef2.29. 18. F. Yuan, T. Dutta, L. Wang, et al. Human DNA Exonuclease TREX1 Is Also an Exoribonuclease That Acts on Single-stranded RNA. Journal of Biological Chemistry 290, no. 21 (2015):13344-53. doi: 10.1074/jbc.M115.653915. 19. C. R. King, Y. Liu, K. A. Amato, et al. Pathogen-driven CRISPR screens identify TREX1 as a regulator of DNA self-sensing during influenza virus infection. Cell Host & Microbe 31, no. 9 (2023):1552-1567.e8. doi: 10.1016/j.chom.2023.08.001. 20. S. Chaudhari, S. D. Pereira, M. Asare-Warehene, et al. Comorbidities and inflammation associated with ovarian cancer and its influence on SARS-CoV-2 infection. Journal of Ovarian Research 14, no. 1 (2021):39. doi: 10.1186/s13048-021-00787-z. 21. R. Ashmawy, E. A. Hammouda, Y. A. El-Maradny, et al. Interplay between Comorbidities and Long COVID: Challenges and Multidisciplinary Approaches. Biomolecules 14, no. 7 (2024):835. doi: 10.3390/biom14070835. 22. S. Harsanyi, I. Kupcova, L. Danisovic, et al. Selected Biomarkers of Depression: What Are the Effects of Cytokines and Inflammation? International Journal of Molecular Sciences 24, no. 1 (2022):578. doi: 10.3390/ijms24010578. 23. J. B. Zawilska and K. Kuczyńska. Psychiatric and neurological complications of long COVID. Journal of Psychiatric Research 156 (2022):349-360. doi: 10.1016/j.jpsychires.2022.10.045. 24. R. Rebai, L. Jasmin, A. Boudah. Agomelatine effects on fat-enriched diet induced neuroinflammation and depression-like behavior in rats. Biomedicine & Pharmacotherapy 135 (2021):111246. doi: 10.1016/j.biopha.2021.111246. 25. W. Chumboatong, S. Khamchai, C. Tocharus, et al. Agomelatine Exerts an Anti-inflammatory Effect by Inhibiting Microglial Activation Through TLR4/NLRP3 Pathway in pMCAO Rats. Neurotoxicity Research 40, no. 1 (2022):259-266. doi: 10.1007/s12640-021-00447-6. 26. C. Fenton and A. Lee. Antidepressants with anti-inflammatory properties may be useful in long COVID depression. Drugs & Therapy Perspectives 39, no. 2 (2023):65-70. doi: 10.1007/s40267-022-00975-x. 27. N. M. Mahmoud, S. M. Elshazly, A. A. Hassan, et al. Agomelatine improves streptozotocin-induced diabetic nephropathy through melatonin receptors/SIRT1 signaling pathway. International Immunopharmacology 115 (2023):109646. doi: 10.1016/j.intimp.2022.109646. 28. M. McHarg, Y. Wang, M. Yakin, et al. Ocular Symptoms In COVID-19 Infection: A Survey Study. Research Square 12, no. 1 (2022):rs.3.rs-1703009. doi: 10.21203/rs.3.rs-1703009/v1. 29. C. Tana, E. Bentivegna, S. J. Cho, et al. Long COVID headache. The Journal of Headache and Pain 23, no. 1 (2022):93. doi: 10.1186/s10194-022-01450-8. 30. Y. Wang, Z. Shan, L. Zhang, et al. P2X7R/NLRP3 signaling pathway-mediated pyroptosis and neuroinflammation contributed to cognitive impairment in a mouse model of migraine. The Journal of Headache and Pain 23, no 1 (2022):75. doi: 10.1186/s10194-022-01442-8. 31. X. Zhang, Y. Chen, L. W. Jenkins, et al. Bench-to-bedside review: Apoptosis/programmed cell death triggered by traumatic brain injury. Critical Care 9, no. 1 (2005):66-75. doi: 10.1186/cc2950. 32. P. G. Coulon, N. Dhanushkodi, S. Prakash, et al. NLRP3, NLRP12, and IFI16 Inflammasomes Induction and Caspase-1 Activation Triggered by Virulent HSV-1 Strains Are Associated With Severe Corneal Inflammatory Herpetic Disease. Frontiers in Immunology 10 (2019):1631. doi: 10.3389/fimmu.2019.01631. 33. H. I. Navarro, Y. Liu, A. Fraser, et al. RelB-deficient autoinflammatory pathology presents as interferonopathy, but in mice is interferon-independent. Journal of Allergy and Clinical Immunology 152, no. 5 (2023):1261-1272. doi: 10.1016/j.jaci.2023.06.024. 34. A. Blomqvist and D. Engblom. Neural Mechanisms of Inflammation-Induced Fever. Neuroscientist 24, no. 4 (2018):381-399. doi: 10.1177/1073858418760481. 35. S. Schmidt, K. Schenkova, T. Adam, et al. SAMHD1’s protein expression profile in humans. Journal of Leukocyte Biology 98, no. 1 (2015):5-14. doi: 10.1189/jlb.4HI0714-338RR. 36. A. Choudhury, R. Tariq, A. Jena, et al. Gastrointestinal manifestations of long COVID: A systematic review and meta-analysis. Therapeutic Advances in Gastroenterology 15 (2022): 17562848221118403. doi: 10.1177/17562848221118403. 37. H. Hoshijima, T. Mihara, H. Seki, et al. Incidence of long-term post-acute sequelae of SARS-CoV-2 infection related to pain and other symptoms: A systematic review and meta-analysis. PLoS One 18, no. 11 (2023):e0250909. doi: 10.1371/journal.pone.0250909. 38. B. A. Krishna, E, Y. Lim, M. Metaxaki, et al. Spontaneous, persistent, T cell-dependent IFN-γ release in patients who progress to Long Covid. Science Advances 10, no. 8 (2024):eadi9379. doi: 10.1126/sciadv.adi9379. 39. L. Lafon-Hughes. Towards Understanding Long COVID: SARS-CoV-2 Strikes the Host Cell Nucleus. Pathogens 12, no. 6 (2023):806. doi: 10.3390/pathogens12060806. 40. C. A. Jefferies. Regulating IRFs in IFN Driven Disease. Frontiers in Immunology 10 (2019):325. doi: 10.3389/fimmu.2019.00325. 41. D. Blanco-Melo, B. E. Nilsson-Payant, W. C. Liu, et al. Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19. Cell 181, no. 5 (2020):1036-1045.e9. doi: 10.1016/j.cell.2020.04.026. 42. N. Smith, C. Possémé, V. Bondet, et al. Defective activation and regulation of type I interferon immunity is associated with increasing COVID-19 severity. Nature Communications 13, no. 1 (2022):7254. doi: 10.1038/s41467-022-34895-1. 43. K. Honda, A. Takaoka, T. Taniguchi. Type I interferon (corrected) gene induction by the interferon regulatory factor family of transcription factors. Immunity 25, no. 3 (2006):349-60. doi: 10.1016/j.immuni.2006.08.009. 44. L. Chen, R. Chen, M. Yao, et al. COVID-19 plasma exosomes promote proinflammatory immune responses in peripheral blood mononuclear cells. Scientific Reports 12, no. 1 (2022):21779. doi: 10.1038/s41598-022-26457-8. FIGURES LEGEND Figure 01- Expression of genes studied in long COVID. ( A ) Column graphs showing the median and interquartile range of expression of genes associated with long COVID when compared to the RWS group. *: p≤0.05; **: 0.050.0001; ***: p<0.0001. ( B ) Column graphs showing the median of genes expressed in long COVID and ( C ) in the RWS group grouped according to gene functionality. In red are cytosolic sensor genes; In blue are genes for signaling adaptor molecules and signal transducers; In green are transcription factor genes; In purple are apoptosis factor genes. ( D ) Heat-map graph showing the grouping of individuals with long COVID and RWS according to gene expression levels. The color legend depicts the most expressed genes (red) to the least expressed (green). ( E ) Correlation graph between the genes studied in the long COVID group. Larger spheres indicate a high correlation coefficient. Blue spheres indicate positive correlations, while red spheres indicate negative correlations. Figure 02- ROC curve: (A-L) ROC (Receiver Operating Characteristic) curve graphs with the description of the area under the curve (AUC) showing the performance of the proposed models in the expression of each gene as a biomarker of distinction between long COVID cases and RWS cases. Figure 03- Association of gene expression with clinical aspects of long COVID: (A) Dot-plot graph showing the dispersion of long COVID data in two dimensions according to the MCA model. The data were grouped into two opposite quadrants with approximately 60% representation of the total variance. ( B ) Graph showing the contribution of the variables to the dispersion of long COVID data in each dimension of the MCA model, based on the calculation of the squared cosine (cos2). The greater the contribution, the redder the highlight of the variable, according to the suggested color legend. Among the co-categories evaluated, the expression data of the MAVS, MDA5 and RELA genes had the best quality of representation of the dimensions. ( C ) Table describing the clinical characteristics of the four patients with long COVID who had comorbidities and were using medications. ( D ) Column graph showing the comparison of the frequency of long COVID symptoms between individuals with comorbidities/on therapy and those without comorbidities/without therapy. Complaints of eye pain and headache were more frequent in individuals with comorbidities. *: p<0.005; #: “p” at the significance threshold (p: 0.0769). ( E ) Column graph showing the breakdown of the median and interquartile range of RELB gene expression in individuals with long COVID with or without complaints of fever and in the RWS group. ( F ) Column graph showing the breakdown of the median and interquartile range of SAMHD1 gene expression in individuals with long COVID with or without complaints of abdominal pain and in the RWS group . *: p≤0.05; **: 0.050.0001; ***: p<0.0001. Figure 04- (Continued) Association of gene expression with clinical aspects of long COVID: (A) Column graph showing the median and interquartile range of FAS gene expression in individuals with long COVID with or without headache complaints and in the RWS group. ( B ) Column graph showing the median and interquartile range of IFI16 gene expression in individuals with long COVID with or without eye pain complaints and in the RWS group. ( C ) Column graph showing the median and interquartile range of nine gene expression in individuals with long COVID with or without comorbidities/undergoing treatment and in the RWS group. ( D ) Table describing the active ingredients and their respective pharmacological groups of individuals with long COVID undergoing continuous drug treatment. ( E ) Column graph showing the median and interquartile range of gene expression of individuals with long COVID according to the types of alleged comorbidities and the pharmaceutical groups of the respective treatments. ( F ) Column graph showing the breakdown of the median and interquartile range of gene expression of individuals with long COVID according to the treatment time of each individual. *: p≤0.05; **: 0.050.0001; ***: p<0.0001. Figure 05- Genes most expressed in long COVID compared to symptomatic acute COVID: (A) Whole-picture graphs showing the comparison of the frequency of severity, ( B ) abnormal radiological findings, ( C ) comorbidities and ( D ) use of continuous medications between individuals with long COVID and acute COVID. ( E-F ) Column graph with breakdown of the median and interquartile range of the genes expression. in individuals with long COVID and acute COVID. ( G ) Column graph with breakdown of the median and interquartile range of the expression of IRF3 and IRF7 genes in individuals with long COVID, mild acute COVID and severe acute COVID. ( H-I ) Correlation graphs between variables with the principal components for the gene expression data in the groups with acute COVID and long COVID. In both models, the representation of the principal components in two dimensions covered more than 50% of the data variability. The redder the highlight of the variables, the greater their contribution to the formation of the main components, according to the legend and suggested colors . *: p≤0.05. Supplementary Material File (table 1.docx) Download 19.87 KB File (table 2.docx) Download 25.39 KB File (table 3.docx) Download 24.87 KB Information & Authors Information Version history V1 Version 1 14 March 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords coronavirus immnopathology immune responses sars coronavirus virus classification Authors Affiliations Leonn Mendes Soares Pereira Federal University of Pará View all articles by this author Izailton de Souza e Souza Federal University of Pará View all articles by this author Wandrey Roberto dos Santos Brito Federal University of Pará View all articles by this author Erika Ferreira dos Santos Federal University of Pará View all articles by this author Flávia Póvoa da Costa Federal University of Pará View all articles by this author Kevin Matheus Lima de Sarges Federal University of Pará View all articles by this author Marcos Henrique Damasceno Cantanhede Federal University of Pará View all articles by this author Mioni Magalhães de Brito Federal University of Pará View all articles by this author Andréa Luciana Soares da Silva Federal University of Pará View all articles by this author Mauro de Meira Leite Federal University of Pará View all articles by this author Maria de Nazaré do Socorro de Almeida Viana Federal University of Pará View all articles by this author Fabíola Brasil Barbosa Rodrigues Federal University of Pará View all articles by this author Rosilene da Silva Federal University of Pará View all articles by this author Giselle Maria Rachid Viana Secretariat for Health and Environmental Surveillance View all articles by this author Tânia do Socorro Souza Chaves Secretariat for Health and Environmental Surveillance View all articles by this author Adriana de Oliveira Lameira Veríssimo Hospital Adventista de Belem View all articles by this author Mayara da Silva Carvalho Hospital Adventista de Belem View all articles by this author Daniele Freitas Henriques Secretariat for Health and Environmental Surveillance View all articles by this author Carla Pinheiro da Silva Secretariat for Health and Environmental Surveillance View all articles by this author Juliana Abreu Lima Nunes Secretariat for Health and Environmental Surveillance View all articles by this author Iran Barros Costa Secretariat for Health and Environmental Surveillance View all articles by this author Izaura Cayres-Vallinoto Federal University of Pará View all articles by this author Igor Brasil-Costa Secretariat for Health and Environmental Surveillance View all articles by this author Juarez Antonio Quaresma 0000-0002-6267-9966 Universidade do Estado do Para Centro de Ciencias Biologicas e da Saude View all articles by this author Andrea Nazare Monteiro Rangel da Silva Federal University of Pará View all articles by this author Maria Alice Queiroz 0000-0003-2014-2770 Federal University of Pará View all articles by this author Eduardo José Melo dos Santos Federal University of Pará View all articles by this author Luiz Fábio Magno Falcão Universidade do Estado do Para Centro de Ciencias Biologicas e da Saude View all articles by this author Antonio Carlos Vallinoto 0000-0003-1135-6507 [email protected] Federal University of Pará View all articles by this author Metrics & Citations Metrics Article Usage 489 views 236 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Leonn Mendes Soares Pereira, Izailton de Souza e Souza, Wandrey Roberto dos Santos Brito, et al. Gene Expression Profile of Immune Response Markers Associated with Long COVID and Its Clinical Aspects in a Cohort from Northern Brazil. Authorea . 14 March 2025. DOI: https://doi.org/10.22541/au.174195200.04677881/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174195200.04677881/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a00087bbebd4df88',t:'MTc3OTUwMjI4OQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();