Coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, is still active among the population. More than 775 million people across the world had been infected up until August 2024, according to the World Health Organization (WHO) (https://covid19.who.int/). However, there are still many gaps in our knowledge about the virus and the disease, as well as its relation to extrinsic and intrinsic host factors.

The clinical symptoms of COVID-19 can change throughout the course of the illness and, in more severe cases, may lead to acute respiratory distress syndrome. Furthermore, in some individuals, symptoms can persist for months or even years after the acute phase, a condition known as long COVID or PACS (post-acute COVID syndrome) (Scholkmann and May, 2023). One of the proposed mechanisms for long COVID is viral persistence in the body, manifesting either as replicating virus or as fragments of viral RNA and proteins (Al-Aly et al., 2024). It is known that the severe acute respiratory syndrome caused by SARS-CoV-2 has a mean incubation period of 6.57 days, defined as the time from infection to the onset of signs and symptoms, although the virus variant could influence the duration (Wu et al., 2022). An average of 18.29 days from the onset are required for viral RNA clearance from the upper respiratory tract (Okita et al., 2022), whereas viral load peaks at disease onset and shortly after the appearance of symptoms (Motoc et al., 2022). Nevertheless, the duration of viral RNA shedding is variable, and prolonged viral load does not necessarily mean infectiousness (Hong et al., 2020). Most patients with mild and moderate forms of COVID-19 are no longer infectious 10 days after the onset of symptoms (Motoc et al., 2022). However, some patients could test positive for SARS-CoV-2 by reverse transcriptase polymerase chain reaction (RT-PCR) in nasopharyngeal specimens for more than 4 weeks (Aldhaeefi et al., 2021; Long et al., 2021), a condition that is known in the bibliography as “prolonged or persistent positive RT-PCR testing”.

Several studies have shown substantial involvement of the gastrointestinal tract in COVID-­19, as well as the impact of the disease on gut function, including gut microbiome alterations (Chakraborty et al., 2022; Mancabelli et al., 2022; Lin et al., 2023; Rossini et al., 2024). In fact, fecal samples remained positive for SARS-CoV-2 longer than nasopharyngeal samples (Wu et al., 2020; Natarajan et al., 2022; Okita et al., 2022). Moreover, cumulative evidence has shown that gut dysbiosis may be linked to the severity of COVID-19 infection and may persist even after the disease has been resolved (Tian et al., 2021; Bucci et al., 2023; Rossini et al., 2024). COVID-­19 patients had significant alterations in fecal microbiomes compared with noninfected controls, characterized by the enrichment of opportunistic pathogens and the depletion of beneficial commensals, which may vary depending on the viral variant (Liu et al., 2022; Yokoyama et al., 2024). Furthermore, gut microbial imbalances have been found to affect acute inflammation well beyond the gut, including the lung, and has been associated with infection by various respiratory viruses, although the mechanisms underlying microbiome–gut–lung associations are still being explored (Sarkar et al., 2021).

Up until now, very few studies have investigated the factors involved in prolonged positive SARS-CoV-2 RT-PCR testing, especially in mild cases. Hoffman et al. (2021) described a correlation between comorbidities and the number of days required for the virus to clear in nonsevere COVID-19 patients (Hoffman et al., 2021). In addition, Motoc et al. (2022) associated old age, high BMI values, male gender, and cardiac, liver, and renal chronic disease with persistent positive RT-PCR testing in mild and moderate forms of COVID-19 (Motoc et al., 2022). Nevertheless, to the best of our knowledge, no studies have yet explored the relation between the gut microbiome and this condition, nor its relationship with other clinical or lifestyle host factors. Here, we conducted an in-depth analysis of the composition and functional profile of the gut microbiome of persistent SARS-CoV-2-positive RT-PCR subjects using shotgun metagenomics, to investigate the effects of this condition on the human gut bacteriome. As a secondary outcome, we sought to determine whether there was any association between clinical and individual patient characteristics, including dietary patterns, and the difficulty of SARS-CoV-2 clearance.

COVID-19 was able to affect gut composition and biodiversity by shifting it towards a state of dysbiosis (Hazan et al., 2022; Schult et al., 2022), an imbalance that could be maintained even months after the recovery (Tian et al., 2021; Ferreira-Junior et al., 2022). Patients with a dysbiotic gut microbiota have a higher risk of developing a more severe form of COVID-19, and key microbial markers associated with this disease has been suggested (Mancabelli et al., 2022). Moreover, gut dysbiosis has been linked to numerous health problems in long COVID-19/post-acute COVID-19 patients (Raj et al., 2024). Therefore, understanding how the gut microbiota may be altered and functionally influenced by SARS-CoV-2 and its pathogenic toxins is a major focus of current research. In our study, both the PCR+ and the control COVID-19 groups showed a gut dysbiosis compared to what has previously been described as a “healthy” state (Manor et al., 2020), which appears to be sustained over time following infection.

First, in the analysis of gut microbial diversity and composition, the PCR+ group showed remarkable differences in specific taxa, characterized by enrichment of potential pathogenic bacteria, such as Erysipelatoclostridium, Cloacibacillus, and Weissella. Some strains of Erysipelatoclostridium ramosum are capable of increasing host susceptibility to opportunistic bacterial invasion via translocation across the intestinal mucosa (Milosavljevic et al., 2021), whereas Cloacibacillus porcorum and Weissella confusa have been associated with different diseases and bacteremia (Domingo et al., 2015; Kamboj et al., 2015; Kodaka et al., 2022). Additionally, Blautia and Akkermasia were also increased in PCR+ subjects, although their contribution to health is controversial. For instance, it has been stated that increased levels of Akkermansia muciniphila may aggravate some metabolic and inflammatory pathologies (Xue et al., 2023). Some lactic acid bacteria were also highlighted in the PCR+ group, which may be favored by the proinflammatory environment resulting from the dysbiotic state and related acidic conditions (Rajamäki et al., 2013). SARS-CoV-2 seems to remain longer in feces than in the respiratory tract (Wu et al., 2020; Okita et al., 2022; Natarajan et al., 2022), which is not necessarily related to gastrointestinal symptoms and disease severity (Gaebler et al., 2021; Maeda et al., 2022). A longer duration of positive RT-PCR in nasal samples could cause a prolonged presence of the virus in the gut, which could extend the associated gut dysbiosis over time or hinder the reestablishment of gut microbiome homeostasis.

Conversely, the control group showed higher levels of commensals and taxa related to positive health effects, such as Roseburia and Slackia. Roseburia spp. constitute important denizens of the human gut microbiome that ferment complex polysaccharides to butyrate, with known beneficial effects on gut health (Hillman et al., 2020). In addition, Slackia members perform key metabolic functions for host homeostasis (Kim et al., 2020). Moreover, increased levels of some Bifidobacterium and Lactobacillus species in controls, also well-known for their health benefits and potential probiotic effects, should be highlighted (Chen et al., 2021; Rastogi and Singh, 2022). However, some opportunistic pathogen bacteria, like Veillonella dispar and Rothia mucilaginosa were increased in controls, in agreement with previous studies in COVID-19 patients (Tian et al., 2021), suggesting a long-term impact on the gut bacteriome. Rothia mucilaginosa is a normal component of the oral and respiratory tract microbiome, but the fecal enrichment of specific oral taxa, linked with increased oral–fecal microbial transmission, has frequently been regarded as a hallmark of disease (Schmidt et al., 2019). In our study, the observed effect might be attributed to the fact that both study groups experienced COVID-19, therefore the control group may not have fully restored their microbiota in post-infection recovery.

We further explored functional and metabolic differences in gut microbiome, revealing an enrichment in the PCR+ group of functional pathways involved in nucleotide, amino acid, and carbohydrate biosynthesis, suggesting a dysbiotic microbiome trying to recover functional balance. Zuo et al. (2021) stated that increased nucleotide and amino acid biosynthesis in gut microbiome functionality indicates potential enhanced production of bacterial cellular building blocks for macromolecules, and carbohydrates are structural components of bacterial cell wall (Zuo et al., 2021). The increased mannose biosynthesis could be linked to glycerol-manno-heptose synthesis, which is also a key compound of the cell wall and core region of lipopolysaccharide (LPS) of Gram-negative bacteria (Kneidinger et al., 2001). Rhamnose is also part of Gram-positive bacteria cell walls, being critical in cell wall structure, biogenesis, and pathogenesis (Mistou et al., 2016). Therefore, these functions, together with the catabolism of uncommon carbon sources such as myo-inositol, could be involved in promoting microbial growth, but they could also contribute to a greater harmful effect on the host’s health (Caputo et al., 2020). SCFA production in this group was also affected, suggesting that the PCR+ condition may also be associated with impaired capacity of the gut microbiome to produce these beneficial metabolites, in line with previously reported effects, including severe COVID-19 patients (Zhang et al., 2022). Taken together, these findings suggest that individuals who suffered from the COVID-19 with persistent positive RT-PCR have a different gut microbiome profile compared to those who tested positive for a normal length of time. Further studies are needed to determine the clinical significance and mechanisms of these differences.

It is noteworthy that the integration of the different datasets using mixOmics also showed differences between the PCR+ and control groups in all the datasets studied. When looking at the variables that contribute the most to the observed differences, the results of microbial taxa supported those previously observed in the differential abundance analysis. In addition to those previously mentioned, the integrative analysis also revealed other potential pathogenic taxa in the PCR+ group. Ruminococcus gnavus and Ruminococcus torques have been associated with inflammatory bowel disease and diabetes (Crost et al., 2023). Notably, higher levels of Ruminococcus gnavus have been reported in people with long COVID-19 compared with non-COVID-19 controls, with gut dysbiosis lasting 14 months (Davis et al., 2023). Moreover, Streptococcus mitis and Fretibacterium fastidiosum are part of the oral microbiota (Mitchell, 2011; Vartoukian et al., 2013), so their detection in the gut could be linked to imbalanced oral–gut translocation, related to different disorders, such as inflammatory bowel disease, among others (Abdelbary et al., 2022). Conversely, health-related Lactobacillus members were more abundant in controls. Given the above, this analysis confirms that the PCR+ group harbors higher levels of bacteria linked to negative health effects than controls, thus contributing to a greater dysbiotic state.

The PCR+ group was also associated with an animal protein-rich diet, and processed, precooked, oily, and high-sugar-content food. Previous research has linked a plant-based food diet or a high adherence to a Mediterranean diet with lower infection and mortality rates caused by SARS-CoV-2 (Vu et al., 2021; Perez-Araluce et al., 2022) and a reduced risk of contracting severe COVID-19 (Kim et al., 2021). Moreover, individuals with COVID-19 reported lower intakes of water, plant protein, seeds, legumes, garlic, dietary fiber, and various vitamins and minerals, as well as following diets that were higher on the dietary inflammation index (Jagielski et al., 2022). Indeed, plant-based food consumption may help improve dysbiotic gut microbiota in COVID-19 and long-COVID patients, potentially aiding disease management (Li et al., 2023), due to its high content of bioactive compounds, with anti-inflammatory, antibacterial, and antiviral properties (Campbell, 2023). Therefore, our findings suggest a potential protective role of healthy dietary patterns, and that an unhealthy diet could also increase the risk of developing the persistent positive RT-PCR condition.

Additionally, evidence suggests that the gut microbiome plays a key role in tackling viral infections by modulating antiviral innate and adaptive leucocyte function in the host immune system (Rossini et al., 2024). The cellular and humoral immune analysis confirmed that both groups show efficient immunity. Moreover, our results highlighted a correlation between persistent positive RT-PCR and basophil and segmented neutrophil levels, responsible for coordinating the adaptive immune response and crucial in the first line of cell-mediated defense against SARS-CoV-2. Increased neutrophil counts have been detected in blood as a feature of COVID-19, particularly in severe cases (Reusch et al., 2021). Moreover, the number of basophils is lower during acute disease than in the recovery phase, as it is a determinant of the antibody response to the virus (Murdaca et al., 2021). In the case of persistent positive RT-PCR, as the majority of the participants were asymptomatic or had mild symptoms, basophils and neutrophils could have helped in dealing with the prolonged shedding of the virus, and could have acted as antigen-presenting cells boosting humoral immune responses. General and SARS-CoV-2 IgG and IgM were correlated to this group, which could be a consequence of basophil actuation together with a longer exposure of the immune system to the virus. In addition, some liver injury markers like gamma GT, ALT, and prothrombin time were increased in the PCR+ group, probably due to the prolonged presence of the virus in the organism and a possible liver infection (Zhao et al., 2023). On the other hand, some parameters that were increased in controls have been associated with a better prognosis after SARS-CoV-2 infection. Vitamin B12 could act as a potential inhibitor against the virus and help balance immune responses during viral infections (Batista et al., 2022). Moreover, higher omega-3 levels, and specifically DHA, have been associated with a lower risk of testing positive for infection with SARS-CoV-2, hospitalization, and mortality (Calder, 2023). ARA could act as an endogenous antiviral compound and its deficiency could increase the susceptibility to COVID-19 (Hoxha, 2020). In addition, IgA controls the initial neutralizing antibody response to SARS-CoV-2 (Esmat et al., 2024); therefore, higher levels of IgA in the blood could be involved in neutralizing the virus and reducing the persistence of positive tests.

Finally, among all the datasets studied in the integrative analysis, relevant correlations were observed between some microbial taxa and food/diet components, highlighting their potential role in the PCR+ condition. Notably, animal protein (eggs) and oily and sugary food (pastries and soft drinks) were associated with specific taxa. Highlighting the relationships between diet and gut microbiome, and their repercussions in microbiome-host metabolic interactions is one of the main current challenges in numerous aspects of human biology, including metabolic health to infections and immune functions (Mancabelli et al., 2024). Previous studies have associated animal-derived food and a high-sugar high-fat diet with higher abundances of Ruminococcus spp. Ruminococcus gnavus and Blautia, among other taxa (Bolte et al., 2021; Lee et al., 2023). Moreover, Fusicatenibacter saccharivorans proportions seem to increase after the intake of processed food (Matsuyama et al., 2019), and a moderately high intake of protein has been linked to a decrease in Dielma and Dielma fastidiosa levels (Cuevas-Sierra et al., 2021). A possible explanation for this could be that the presence of SARS-CoV-2 in the gut, coupled to a diet with a high intake of animal protein, sugar, and saturated fats, could favor certain microbial taxa, disrupting the balance of microbial communities and potentially leading to a dysbiotic state.

This study has several limitations, primarily due to the relatively small sample size and the retrospective nature of the investigation. Furthermore, the collection of samples was occasionally delayed because of technical difficulties and local restrictions. A further limitation is that we did not determine fecal shedding of SARS-CoV-2 RNA or the presence of live viruses in the gut, as this could also influence the observed effects on the microbiome. For safety and practical reasons, it was necessary to process the fecal samples for virus inactivation. Nevertheless, this study offers new insights and corroborates existing knowledge. In particular, we shed new light on complex SARS-CoV-2-microbiome interactions highlighting the overlap with host conditions. Our findings illustrate the varying impacts of SARS-CoV-2 and COVID-19 infection on the intestinal microbiome and its metabolic functionality, contingent on the duration of viral persistence within the organism. Gut microbiome dysbiosis may persist for several months in individuals who have persistently positive RT-PCR results. Furthermore, our results indicated that certain blood markers and dietary habits are associated with prolonged positivity, suggesting the potential value of appropriate nutritional interventions as a part of a comprehensive care strategy for patients recovering from COVID-19. All in all, this study stands out as pioneer to integrate different omics approaches and datasets in a PCR persistent SARS-CoV-2 cohort and provides valuable data for deepening the understanding of the functional contribution of gut microbiota in infection and human health. Data presented here, when placed in the context of preliminary work, underscore the necessity of incorporating the gut microbiota into clinical trials to elucidate the biological mechanisms underlying disease, including viral persistence and the impact of disease on multiple organ systems.