Reader TTS

ABSTRACT Although mRNA SARS-CoV-2 vaccines are generally safe and effective, in certain immunocompromised individuals they can elicit poor immunogenic responses. Among these individuals,

people living with HIV (PLWH) have poor immunogenicity to several oral and parenteral vaccines. As the gut microbiome is known to affect vaccine immunogenicity, we investigated whether

baseline gut microbiota predicts immune responses to the BNT162b2 mRNA SARS-CoV-2 vaccine in healthy controls and PLWH after two doses of BNT162b2. Individuals with high spike IgG titers and

high spike-specific CD4+ T-cell responses against SARS-CoV-2 showed low α-diversity in the gut. Here, we investigated and presented initial evidence that the gut microbial composition

influences the response to BNT162b2 in PLWH. From our predictive models, _Bifidobacterium_ and _Faecalibacterium_ appeared to be microbial markers of individuals with higher spike IgG

titers, while _Cloacibacillus_ was associated with low spike IgG titers. We therefore propose that microbiome modulation could optimize immunogenicity of SARS-CoV-2 mRNA vaccines. SIMILAR

CONTENT BEING VIEWED BY OTHERS BASELINE GUT MICROBIOTA AND METABOLOME PREDICT DURABLE IMMUNOGENICITY TO SARS-COV-2 VACCINES Article Open access 25 September 2023 SHORT-CHAIN FATTY ACIDS PLAY

A KEY ROLE IN ANTIBODY RESPONSE TO SARS-COV-2 INFECTION IN PEOPLE LIVING WITH HIV Article Open access 28 December 2024 HUMAN IMMUNE AND GUT MICROBIAL PARAMETERS ASSOCIATED WITH

INTER-INDIVIDUAL VARIATIONS IN COVID-19 MRNA VACCINE-INDUCED IMMUNITY Article Open access 20 April 2023 INTRODUCTION The COVID-19 pandemic has caused more than 6 million deaths1. However,

vaccines against SARS-CoV-2 have changed the course of the pandemic by reducing the lethality of the disease2,3 and the incidence of the post-acute COVID-19 syndrome4. Two messenger RNA

(mRNA) vaccines (BNT162b2 from Pfizer-BioNTech and mRNA-1273 from Moderna) have reached an efficacy of upto 95% with minimal side effects in initial randomized clinical trials2,5. However,

the vaccines do not fully protect against reinfection, especially with new variants6, with boosters being essential for enhancing host immunity. Therefore, it is important to understand the

factors influencing the immunogenicity of SARS-CoV-2 vaccines, including their capacity for long-term protection against disease severity and death. Several known factors are essential for

the vaccine response, such as age, medications, disease-associated comorbidities, disorders of the immune system, and inflammatory conditions7,8. An additional factor reported to regulate

immunogenicity for several other oral and parenteral vaccines is the gut microbiota9. Without any bacterial stimulation, in fact, the mucosal immune system remains poorly developed, both

anatomically and functionally10. The crosstalk between the microbial communities and immune system is crucial in maintaining homeostasis and their mutualistic relationship. In fact, gut

dysbiosis has also been known to be associated with immunological disequilibrium11, which is often described as Th2 cell activation or T-regulatory cell deficiency11,12. Recent animal

studies have also shed new light on the association between the gut microbiome and the immune system. In vitro studies have in fact reported the role of _Bifidobacterium adolescentis_ in

reducing the adhesion of Norovirus to Caco-2 and HT-29 cells13, two colon epithelial cell lines. _Bifidobacterium_ has been shown to have broad immunomodulatory effects in humans. For

example, the cellular immunity in the elderly has been reported to be augmented by _Bifidobacterium_14,15. In addition, _Bifidobacterium animalis_ has been shown to increase NK cell activity

and polymorphonuclear cell functionality in old-aged individuals. In addition, the _Clostridium_ cluster XIVa and IV were shown to stimulate TGF-β for Treg cell activation16. In fact,

butyrate, a fatty acid of which Clostridia are known key producers, is pivotal in activating Treg cells and their anti-inflammatory functionalities17,18. Interestingly, many animal and

clinical studies have shown the role of the gut microbiome in vaccine immunogenicity19,20, such as the case of _Bifidobacterium longum_, which has been associated with antigen-specific

T-cell immunity against tuberculosis and polio vaccines21. Actinobacteria, the bacterial phylum to which _Bifidobacterium_ belongs, in fact, has been shown to have a positive correlation

with adaptive immunity to certain oral (i.e., polio) and systemic vaccines (i.e., BCG, tetanus toxoid, hepatitis B virus) in infants from Bangladesh21. Another study has correlated the gut

microbiome in infants with immune responses to the rotavirus vaccine22,23. Overall, the general mechanism with which the gut microbiome influences vaccine efficacy primarily involves its

ability to use bacterial-derived natural adjuvants to activate specific immune pathways responsible for both innate and adaptive immune responses24. One such case is the expression of

flagellin or LPS by the bacterial phylum Proteobacteria, leading to the activation of pattern-recognition receptors (PRRs) found on the antigen-presenting cells19. The concept of “endogenous

adjuvant potential” was further highlighted in a study where an inactivated influenza vaccine elicited lower antibody responses in germ-free or antibiotic-treated mice25,26. Treating these

mice with flagellated _Escherichia coli_ restored the normal antibody titers. Similar results were also obtained from analogously adjuvanted vaccines against poliovirus26. In addition,

activation of Nucleotide-binding oligomerization domain-containing protein 2 (Nod2) by gut commensals was observed to stimulate cholera toxin-mediated antigen-specific immune responses

through high levels of IL-1β during oral vaccination27,28. Other studies have shown the immunomodulatory property of bacterial flagellin, which further activated TLR5 for improved antibody

production24,26. Notably, differences in gut microbiome profiles have also been associated with health status, metabolic abnormalities, and aging, which are key elements in influencing

vaccine immunogenicity29. Studies have clearly shown a reduced efficacy of the Pfizer/BioNTech vaccine due to aging, obesity, and other comorbidities30,31,32. Elderly individuals with

underlying comorbidities are at utmost risk of COVID-19 infections and display low immunogenicity against COVID-19 vaccinations, since aging leads to reduced immunity and chronic

inflammation. A study compared BNT162b2 vaccine immunogenicity in two groups of patients aged <60 and >80 years, where the latter showed significantly lower spike antibody titers33.

Aside from the elderly, another COVID-19 risk group comprises individuals with underlying chronic diseases or disorders, which also affect the immunogenic responses to vaccines. Among them,

immunocompromised individuals with primary immunodeficiencies or secondary immunodeficiencies such as HIV, are particularly prone to SARS-CoV-2 infections and display less immunogenicity to

COVID-19 vaccines. Considering the significantly lower level of COVID-19 vaccine immunogenicity in immunocompromised individuals, such as people living with HIV (PLWH)34,35,36, we

hypothesized that the gut microbiome might influence the efficacy of COVID-19 vaccines in these individuals. Elucidating the potential role that the gut microbiome plays in COVID-19 vaccine

immunogenicity might thus offer the possibility to identify poor vaccine responders, as well as to develop potential microbiome-targeting therapies that may enhance vaccine responses.

Several clinical trials, in fact, are aiming to increase the efficacy of COVID-19 vaccines by manipulating the gut microbiome. One involves the use of 5-ALA-phosphate, which is known to

maintain gut homeostasis, to increase the immunogenic responses to COVID-19 vaccines37. Another trial involves the use of three _Bifidobacterium_ strains to increase the immunogenicity of

SARS-CoV-2 vaccines and reduce the side effects in elderly diabetic patients38. In the present work, we addressed the association between the gut microbiome and immune responses to the

BNT162b2 mRNA SARS-CoV-2 vaccine. The current study presents an association between the gut microbiome and humoral/cellular COVID-19 vaccine responses in PLWH and healthy controls (HC).

RESULTS PLWH AND ELDERLY INDIVIDUALS DISPLAYED LESS IMMUNOGENICITY TO THE BNT162B2 VACCINE All participants included were ≥18 years of age and without any prior history of SARS-CoV-2

infection, as determined by serological testing. PLWH were on antiretroviral therapy for an average of 10.8 years (33–66 years). Their median CD4+ T-cell count was 615 cells/mL and 86% of

PLWH had less than 50 copies/mL of HIV RNA. There were no significant differences in age, BMI, or number of comorbidities between PLWH and HC (Table 1). Two weeks after the second mRNA

vaccine dose (day 35) (Fig. 1a), the HC showed a significantly higher spike IgG titers as compared to the PLWH (_p_ = 0.0001) (Supplementary Fig. 1a). Gender, BMI, and baseline total IgG

levels did not affect the spike antibody titers in the whole cohort (Supplementary Fig. 1b–d). Furthermore, we observed significantly higher antibody titers in younger individuals (18–39

years) compared with middle-aged (40–59 years; _p_ = 0.003) and older participants (>60 years; _p_ < 0.0001) (Supplementary Fig. 1e). Age was therefore negatively correlated with spike

IgG (_p_ = 0.04). In addition, in PLWH, spike IgG levels were not affected by the CD4+ T-cell counts or CD4/CD8 ratio (Supplementary Fig. 1f, g). BACTERIAL DIVERSITY IS LOWER IN INDIVIDUALS

WITH HIGHER SPIKE IGG TITERS To better discriminate the different levels of vaccine immunogenicity among the whole cohort, individuals from both PLWH and HC group were divided into high and

low responders, based on the median spike IgG titer (1972 U/mL). Notably, high responders displayed significantly lower bacterial α-diversity (richness and evenness) than low responders

(Fig. 1b). All the α-diversity indices negatively correlated with spike IgG titers for the whole cohort (Fig. 1c, Observed _p_ = 0.05, Shannon _p_ = 0.016, Simpson _p_ = 0.01). In addition,

the phylogenetic diversity in high responders was reduced compared with low responders (Faith’s PD, _p_ = 0.02). However, there were no major changes in β-diversity shifts in the whole

cohort (Supplementary Fig. 2a), while NMDS2 scores differed significantly between high and low responders in the HC group (Supplementary Fig. 2b, _p_ = 0.0002). However, no significant

cluster differences were observed in the PLWH (Supplementary Fig. 2c). In addition to bacterial diversity, we also observed significant changes in the microbiota composition between

individuals with high and low spike IgG titers. A total of 258 bacterial genera were detected in the whole cohort. At the genus level, _Agathobacter_ (_p_ = 0.02), _Lachnopsira_ (_p_ = 

0.03), and Lachnospiraceae FCS020 group (_p_ = 0.03) were markers for high responders, according to linear discriminant analysis (LDA) scores (LDA > 2; _p_ < 0.05) (Fig. 1d).

_Butyricimonas_ (_p_ = 0.02), _Cloacibacillus_ (_p_ = 0.009), _Intestinimonas_ (_p_ = 0.02), Ruminococcaceae DTU089 (_p_ = 0.006), and _Paraprevotella_ (_p_ = 0.02) were significantly

enriched in low responders (LDA > 2; _p_ < 0.05) (Fig. 1d). Likewise, negative associations between α-diversity and antibody titers were observed within the HC (Supplementary Fig. 3a)

and the PLWH (Supplementary Fig. 3b) when individuals from these groups were divided into high and low responders. High responders with HC (Observed _p_ = 0.004, Shannon and Simpson _p_ <

 0.0001) and PLWH (Simpson _p_ = 0.034) displayed significantly reduced bacterial diversity than low responders within the respective groups (Fig. 2a, b). Enrichment of Bacteroidetes and

depletion of Firmicutes were observed in the higher responders of the HC (_p_ < 0.01). In addition, higher populations of _Bacteroides_ (_p_ < 0.05), _Sutterella_ (_p_ < 0.05),

Lachnospiraceae FCS020 group (_p_ < 0.05), and reduced numbers of _Alloprevotella_ (_p_ < 0.05), _Anaerofilum_ (_p_ < 0.05)_, Succinivibrio_ (_p_ < 0.05)_, Moryella_ (_p_ < 

0.01)_, Negativibacillus_ (_p_ < 0.05), and certain members of the Ruminococcaceae family at the genus level (_p_ < 0.05) were observed in the high responders within the HC (LDA > 

2.3) (Fig. 2c). Within the PLWH group, on the other hand, _Flavonifractor_, _Lachnospira_, and _Oscillibacter_ were increased in high responders (_p_ < 0.05, LDA > 2.6), whereas

_Butyricimonas_ and _Paraprevotella_ were depleted in low responders (_p_ < 0.05, LDA > 2) (Fig. 2c). Both the HC and the PLWH groups displayed an increased abundance of

_Hydrogenoanaerobacterium_, _Methanobrevibacter_, _Cloacibacillus_, and Ruminococcaceae DTU089 in individuals with low spike IgG titers. SPIKE-SPECIFIC CD4+ T-CELL RESPONSE SHOWS NEGATIVE

ASSOCIATION WITH Α-DIVERSITY CD4+ T-cell response to the spike protein of SARS-CoV-2 was evaluated in the HC (_n_ = 45) and PLWH (_n_ = 45) at day 35. We stratified all individuals (_n_ = 

90) into two groups based on the magnitude of their spike CD4+ T-cell response (median 0.36%). A significant decline in the α-diversity was found in individuals with higher levels of

spike-specific CD4+ T-cells (Fig. 3a) as compared to individuals with low levels of spike-specific CD4+ T-cells (Shannon _p_ = 0.045, Simpson _p_ = 0.025). Similar to spike IgG titers, we

also found a negative association between α-diversity and spike-specific CD4+ T-cell response (observed _p_ = 0.003, Shannon _p_ = 0.001, Simpson _p_ = 0.005) from linear regression analysis

(Supplementary Fig. 4). There were shifts in the β-diversity (_p_ = 0.01) with unique clustering patterns specific to each group (Supplementary Fig. 5). In addition, individuals with low

spike-specific CD4+ T cell response had more Firmicutes (_p_ = 0.005, LDA > 4.5) and less Bacteroidetes (_p_ = 0.005, LDA > 4) compared with high responders (Fig. 3b). Moreover,

Ruminococcaceae (_p_ = 0.02, LDA > 4.5), Erysipelotrichaceae (_p_ = 0.04, LDA > 3.5), and Akkermansiaceae (_p_ = 0.03, LDA > 3.5) were enriched in individuals with low CD4+ T cell

responses (Fig. 3b). Individuals eliciting a higher magnitude of CD4+ T cell response had an increased abundance of _Lachnospira_ (_p_ = 0.035, LDA > 3). In contrast, the low responders

showed an increased abundance of _Akkermansia_ (_p_ = 0.035, LDA > 3), _Fournierella_ (_p_ = 0.014, LDA > 3), and _Alistipes_ (_p_ = 0.029, LDA > 3.5) (Fig. 3b). Similarly to the

whole cohort, a consistent negative association between spike-specific CD4+ T cell response and α-diversity was also observed within the HC (Supplementary Fig. 6a, b) and PLWH (Supplementary

Fig. 6c, d). The low responders in the PLWH were enriched with Firmicutes (_p_ = 0.01, LDA > 4.5) and depleted of Bacteroidetes (_p_ = 0.046, LDA > 4.5). Erysipelotrichaceae (_p_ = 

0.01), Eggerthellaceae (_p_ = 0.006), and Succinivibrionaceae (_p_ = 0.036) were detected as signatures of low responders. Both the HC and PLWH had a high abundance of _Marvinbryantia_ in

communities with low CD4+ T cell response (_p_ < 0.05). Higher numbers of _Fournierella_ (_p_ = 0.03) were identified in low responders within the PLWH (Fig. 3c). On the other hand, the

high responders in the HC showed a higher abundance of Lactobacillaceae (_p_ = 0.01, LDA > 3) and a decline of Akkermansiaceae (_p_ = 0.02, LDA > 3.5). At the genus level, the high

responders in the HC displayed an enrichment of _Lactobacillus_ (_p_ = 0.014). Instead, _Akkermansia_ (_p_ = 0.017), _Ruminiclostridium_ (_p_ = 0.037), and _Hydrogenoanaerobacterium_ (_p_ = 

0.047) were more abundant in low spike-specific CD4+ T cell responders (Fig. 3c). Furthermore, we explored the correlation of bacterial taxa associated with spike CD4+ T-cell response and

spike IgG titers (Supplementary Table 1). In the whole cohort, _Sutterella_ (_p_ = 0.01), _Bifidobacterium_ (_p_ = 0.015), _Bacteroides_, _Lachnospira_, and _Lactobacillus_ showed possible

positive correlation, while _Escherichia-Shigella_ (_p_ = 0.015), _Marvinbryantia_ (_p_ = 0.002), Ruminococcaceae DTU089 (_p_ = 0.018), _Methanobrevibacter_ (_p_ = 0.028), and

_Cloacibacillus_ (_p_ = 0.015) showed negative correlation with antibody levels (Supplementary Fig. 7). GUT MICROBIOME DIVERSITY IS AFFECTED BY AGE A relative decline in diversity and

richness was observed in young adults (18–39 years, _n_ = 37) as compared to older individuals (>60 years, _n_ = 50) (Observed, _p_ < 0.001 and Shannon, Simpson, _p_ < 0.0001,

Supplementary Fig. 8a). Moreover, there was a significant positive association between age and α-diversity (Supplementary Fig. 8b, observed _p_ = 0.0002, Shannon _p_ = 0.00001, Simpson _p_ =

 0.0001). Similarly, a significant shift in β-diversity was observed among the different age groups (_p_ = 0.03) (data not shown). At the genus level, certain members of the Ruminococcaceae

family (_p_ < 0.05), _Butyricimonas_ (_p_ = 0.01), _Ruminiclostridium_ (_p_ = 0.005), _Hydrogenoanaerobacterium_ (_p_ = 0.005), _Fournierella_ (_p_ = 0.009), Christensenellaceae R_7 group

(_p_ = 0.007) and _Methanobrevibacter_ (_p_ = 0.007) showed increased abundance in older individuals (>60 years) (LDA > 2.5, Supplementary Fig. 8c). In young adults we observed an

enrichment of _Lachnospira_ (_p_ = 0.02), _Bacteroides_ (_p_ = 0.02), and _Agathobacter_ (_p_ = 0.009) which were also enriched in individuals with high spike IgG titers (LDA > 3). Within

the HC and PLWH groups, we observed a similar pattern of relative decrease in the α-diversity indices of young adults compared to the elderly and middle-aged (Supplementary Fig. 8d, e).

Predominantly, the Bacteroidetes population was significantly enriched in younger adults as compared to the older individuals (_p_ = 0.001), who harbored a higher abundance of Firmicutes

(_p_ = 0.006; LDA > 4.5) (Supplementary Fig. 8f). The Firmicutes/Bacteroidetes (F/B) ratio was highest in the elderly and lowest in the youths. GUT MICROBIOME PROFILES ARE AFFECTED BY

TOTAL IGG LEVELS AT BASELINE In addition, we studied whether total IgG levels at baseline were associated with specific microbiome profiles, since we observed a borderline significant

association with spike IgG levels in our cohort (_p_ = 0.09). Stratifying the individuals into those with high and low total IgG levels at baseline, we found a decline in the bacterial

diversity of the individuals with high total IgG levels (data not shown). The most differential genera among all samples from our analysis (individuals with high and low total IgG levels)

were _Anaerostipes_ (_p_ < 0.01), _Fournierella_ (_p_ < 0.01), _Mitsuokella_ (_p_ < 0.05), and _Lactobacillus_ (_p_ < 0.05) (Supplementary Fig. 9a). Interestingly, an increased

abundance of _Anaerostipes_ (_p_ < 0.01, LDA > 3), _Bacteroides_, and _Bifidobacterium_ were detected in the individuals with high levels of total IgG at baseline, and these genera

were also likely to be positively correlated with spike IgG titers (Supplementary Fig. 9b, c). GUT MICROBIOTA Α-DIVERSITY AND COMPOSITION ARE ASSOCIATED WITH SPIKE IGG LEVELS IRRESPECTIVE OF

AGE AND DISEASE STATUS We further integrated the cellular and humoral immune responses on day 35, with selected baseline factors, such as age and disease group, into a network and

correlated the microbial taxa, which were significantly associated with said clinical characteristics and immunological responses. Overall, from our predictive models, we observed a positive

association of spike IgG levels with _Lachnospira_, _Faecalibacterium_, and _Bifidobacterium_, which were also markers of the HC group (Fig. 4a). _Agathobacter_, _Lactobacillus,

Bacteroides_, and _Lachnospira_ positively correlated with both spike IgG levels and spike-specific CD4+ T-cell responses (Fig. 4a). The microbes positively linked with age

(_Hydrogenoanaerobacterium_, _Methanobrevibacter_, Ruminococcaceae DTU089, _Butyricimonas_) showed a negative correlation with both spike IgG levels and CD4+ T-cell responses. Spike IgG

levels and CD4+ T-cell responses were negatively associated with _Methanobrevibacter_, Ruminococcaceae DTU089, _Paraprevotella_, _Marvinbryantia_, _Cloacibacillus_, and _Succinivibrio_ (Fig.

4a). Furthermore, we carried out a DESeq2 analysis to model the absolute abundance of each microbial taxon associated with antibody levels irrespective of other clinical parameters (such as

age, gender, disease group) to determine the log2fold change (effect size estimate) for differential changes in microbial abundance as true absolute counts. We observed significant changes

in the abundance of several gut microbes directly associated with spike IgG. Thus, individuals with high spike IgG titers revealed enrichment of _Bifidobacterium_ (_p_ = 0.03),

_Faecalibacterium_ (_p_ = 0.03), _Blautia_ (_p_ = 0.03), _Catenibacterium_ (_p_ = 0.03), and _Hungatella_ (_p_ = 0.005) from this analysis (Fig. 4b). _Bifidobacterium_ was increased in

subjects with high spike IgG titers and high baseline total IgG. Conversely, individuals with low antibody titers showed a higher abundance of _Cloacibacillus_ (_p_ = 0.0001) (Fig. 4b). In

univariate analysis, the number of observed species, Shannon diversity, Simpson index, and age were all significantly associated with spike IgG levels (Table 2). To reveal the independent

association between α-diversity and humoral response to COVID-19 vaccination, we conducted multiple regression analysis. We found a significant association of bacterial abundance (Shannon

_p_ = 0.02, Simpson _p_ = 0.019) with spike IgG levels, independent of age (Table 2 and Supplementary Table 2). However, age showed a significant impact when the number of observed species

and age were correlated to spike IgG response (_p_ = 0.04). The results suggest that microbial abundance is independent of age and influences spike IgG levels. Nonetheless, the observed

richness shows age dependency to affect spike IgG levels, while richness estimates are unaffected by species abundance. Using multivariate analysis, we further analyzed the effect of other

baseline clinical factors, such as gender and total IgG, and observed that the α-diversity significantly affected the spike IgG response (Supplementary Table 3, observed _p_ = 0.02, Shannon

_p_ = 0.006, Simpson, _p_ = 0.005). Gender showed no effect on antibody titers, followed by age and IgG. In addition, the α-diversity significantly impacted the spike-specific CD4+ T-cell

responses in the cohort irrespective of age, gender, and total IgG (Supplementary Table 3, observed _p_ = 0.006, Shannon _p_ = 0.0009, Simpson, _p_ = 0.004). DISCUSSION SARS-CoV-2 mRNA

vaccines are highly immunogenic and effective in protecting from severe COVID-192. However, several factors, such as increasing age, underlying medical conditions, and concomitant

medications, are associated with lower vaccine efficacy7,8,39. Increasing evidence, from clinical cohorts, interventional studies, and animal models, show that gut microbiota plays a

significant role in modulating responses to vaccination9,40. In our study, we investigated whether the baseline gut microbial diversity and composition would affect the immunogenicity of

mRNA BNT162b2 SARS-CoV-2 vaccine in PLWH, as immunocompromised individuals possess lower immunogenic responses to these vaccines. Interestingly, α-diversity metrics were negatively

associated with spike IgG titers post-vaccination, both in the whole cohort and within the subgroups of HC and PLWH. In addition, we found that the elevated levels of spike-specific CD4+

T-cell responses were associated with reduced microbiome diversity, thus further validating the association of the microbiome with the immunogenicity of the BNT162b2 mRNA vaccine. These

results are in line with data presented by previous studies using vaccines against other viral diseases. A collection of studies reported that similar vaccines (both oral and parenteral)

elicit impaired immunogenicity in individuals living in low-income countries compared with people from high-income countries, due to their differences in gut microbiome profiles19. Moreover,

two studies on oral polio vaccine, in India and China, showed a low bacterial diversity in individuals with high vaccine efficacy41,42. Furthermore, the role of microbiota in modulating

T-cell responses has been shown in influenza infection43. Lastly, antibiotic treatment prior to H1N1 vaccination altered the microbial diversity, resulting in decreased influenza-specific

IgG1 and IgA antibody titers44. In addition to bacterial diversity, this study showed that the baseline microbiota composition could potentially influence the immunological effects of the

SARS-CoV-2 vaccines. We observed that certain bacterial genera at the baseline were associated with vaccine immunogenicity, measured by IgG spike and spike-specific CD4+ T-cell responses.

Interestingly, in the network analysis, _Agathobacter_, _Lactobacillus, Bacteroides_, and _Lachnospira_ were positively correlated with both spike IgG levels and spike-specific CD4+ T-cell

responses. These bacteria with known immunomodulatory properties could contribute to higher immune responses to vaccinations45,46,47. _Bacteroides_, in fact, have been shown to have a key

role in vaccine responses due to the induction of homeostatic immune priming. For instance, the LPS of _Bacteroides thetaiotaomicron_ acts as an adjuvant, enhancing the production of

hepatitis B virus antigen-specific antibodies48. In addition, another study revealed the association of different _Bacteroides_ strains with diverse response levels to rotavirus vaccine49.

Overall, _Bacteroides_ are resident commensals, acting as major polysaccharide degraders in the gut for butyrate production50. Conversely, _Methanobrevibacter_, Ruminococcaceae DTU089,

_Marvinbryantia_, _Cloacibacillus_, and _Succinivibrio_ were enriched in the baseline microbiome of individuals with low IgG titers and low spike-specific CD4+ T-cell responses. This is in

line with previous reports, as _Cloacibacillus_ has been previously associated with bacteremia51, _Succinivibrio_ was found in inflammatory conditions52 and previous studies have reported

the enrichment of _Marvinbryantia_ spp., which are SCFA-producers, in the elderly53, who experience a lower response to vaccinations, as observed in our study. Similarly, in our study,

_Marvinbryantia_ was observed to be less abundant in the younger group in the whole cohort and within the HC and PLWH groups. Moreover, our findings are also consistent with results from

Borgognone et al.54, in which lower α-diversity and gene richness were observed in PLWH, who showed better immunogenic response to an HIV vaccine. These individuals, dubbed viremic

controllers, were observed to have significantly higher Bacteroidales/Clostridiales ratio and lower Methanobacteriales than the non-controllers54. Our study also shows similar findings in

individuals with high spike IgG levels, with less abundance of bacterial groups belonging to Clostridiales and Methanobacteriales, such as Ruminococcaceae DTU089 and _Methanobrevibacter_,

respectively. The abundance of certain bacterial groups belonging to Clostridiales in low responders might be potentially associated with metabolic pathways related to methanogenesis and

carbohydrate synthesis. Notably, in the present work, _Bifidobacterium_ and _Faecalibacterium_ showed significant association with high spike IgG titers, irrespective of other factors, from

our predictive models. This is in line with earlier studies, where enrichment of _Bifidobacterium_ was correlated with CD4+ T-cell responses and higher antibody titers to parenteral and oral

vaccinations41,55. In addition, _B. adolescentis_ was also associated with a higher rate of neutralizing antibodies after CoronaVac immunization56. Current clinical trials are exploring the

effect of supplementation with different _Bifidobacterium_ species on immune responses to SARS-CoV-2 vaccine38. Interestingly, our findings on microbiome associations with mRNA vaccine

immunogenicity in the whole cohort were also consistent within the PLWH group. This is of note, since the gut microbial composition of PLWH is altered due to gut dysbiosis occurring during

the course of HIV-1 infection and further changed during antiretroviral therapy, as shown by our previous studies57,58. Interestingly, in our study, we observed different bacterial groups

from _Lachnospiraceae_ to be associated with either high or low spike IgG levels. Generally, the bacterial members belonging to the family _Lachnospiraceae_ are well-known producers of SCFAs

in colon mucosa-associated microbiota. _Lachnospiracea_ FSC020, which shows a positive association with spike IgG levels in our study, has been reported to have a potential positive

association with the production of acetate and propionate59 and to be linked to circulating lipid metabolites in blood60. In addition, some members of Lachnospiraceae family, such as

Lachnospiraceae UCG-003 can potentially protect against colon cancer by butyrate production61. While existing literature shows Lachnospiraceae UCG-008 to be negatively associated with age62

and saturated fatty acid intake63 and has decreased abundance in conditions like liver cirrhosis and hepatocellular carcinoma, our findings reveal the abundance of this taxon in low

responders to COVID-19 vaccine64. The key findings of this study link α-diversity with spike IgG responses, which in turn is also influenced by age. Although α-diversity has been known to

play a crucial role in association studies, we believe that the gut bacterial composition and its functional aspects play a much more important role in influencing the landscape of gut

balance. In this study, individuals with low α-diversity showed better immunogenic responses and an abundance of _Bifidobacterium_, which is known to have a protective function in the gut.

In this study, we also observed younger adults with low α-diversity to have better vaccine immunogenicity compared to elderly individuals with higher α-diversity. In addition, individuals

with low immunogenic responses were observed to have high α-diversity. This overall demonstrates a negative association between diversity and spike IgG levels and a positive association

between age and α-diversity. Higher α-diversity in elderly individuals has also been observed in a few other studies65,66,67. Moreover, when investigating the gut composition, we observed

bacterial genera specific to younger participants, such as _Lachnospira_, _Bacteroides_, and _Agathobacter_ to be positively associated with spike IgG levels. Similarly to the findings from

another study, we also detected enrichment of potential pathobionts in elderly individuals, which have higher α-diversity, such as an increase of Enterobacteriaceae which has been correlated

with frailty in elder individuals65. At the genus level, we also observed some microbial signatures of pathobionts such as _Escherichia-Shigella_, and _Enterococcus_ in the elderly group,

as corroborated by the literature66,67. Similarly to other studies65,66,67,68, our analyses also detected in this group an abundance of certain members of the Ruminococcaceae family,

_Butyricimonas_, Christensenellaceae R_7 group, _Akkermansia_, members of Erysipelotrichaceae and a reduction of _Faecalibacterium_, a pattern which has been linked to inflammatory

disorders69. Ruminococcaceae are well-known symbionts present in the human gut generating SCFA70 to enhance the protective functions of the intestinal epithelium and prevent infections from

opportunistic pathogens71. An enrichment of Ruminococcaceae diversity (Supplementary Fig. 10) might be associated with better metabolic plasticity and versatility of the gut in elderly

individuals65. Notably, while the elderly population has been reported to have a lower F/B ratio72, our study shows it to display a higher F/B ratio. This has also been observed in another

study cohort from Ukraine73. In fact, the LefSe data shows the abundance of Firmicutes in elderly and Bacteroidetes in younger individuals, both of which are the predominant bacterial phyla

colonizing the gut. As a matter of fact, certain key species within the immunomodulatory Gram-negative Bacteroidetes phylum have been reported to exert anti-inflammatory effects through

T-cell modulation74. Lastly, the elderly population showed an abundance of gut microbes, such as _Hydrogenoanaerobacterium_, _Methanobrevibacter, Negativibacillus_, which were also observed

in individuals with low spike IgG levels, which is in line with the observed negative association between age and spike IgG levels in our study. The mechanisms by which the gut microbiota

modulates immunological responses to vaccines are not yet fully understood. Several potential mechanisms, however, have been anticipated. Microbiota-derived flagellin, peptidoglycan, and

lipopolysaccharide can act as natural adjuvants to vaccination, recognized by PRRs19,75. Another important mechanism concerns gut integrity, which is crucial in regulating immune responses,

and is disrupted in a state of inflammation, malnutrition, or by antibiotic treatment76. The gut–barrier integrity is modulated by certain bacterial metabolites, which induce the expression

of tight junction proteins, thereby maintaining the epithelial integrity77. Microbiota-derived metabolites, such as SCFAs (acetate, butyrate, and propionate), tryptophan, and secondary bile

acids can, therefore, directly and indirectly alter immune responses19. They provide energy sources for enterocytes, strengthen the epithelial barrier, and can act as signaling molecules in

regulatory pathways of intestinal and systemic immunity. SCFAs also regulate T-cell metabolism and can augment B-cell metabolism, enhancing pathogen-specific antibody responses78,79. In the

present study, certain butyrate producers, such as _Lachnospira_, _Catenibacterium_, and _Faecalibacterium_, were enriched in individuals with higher immunogenic responses to the BNT162b2

vaccine, and metabolites derived from these bacteria might likely have an association with higher vaccine immunogenic responses. As a critical butyrate producer, _Faecalibacterium

prausnitzii_ has been previously added to probiotic formulations for restoring gut health80,81. Overall, we hypothesize that microbiota-derived metabolites, such as butyrate, indole, and

bile acids, might possibly have a strong potential to improve vaccine responses, and quantifying such metabolites could be an interesting topic for future research. The primary limitation of

the present study lies in its cross-sectional design, preventing definitive conclusions from being drawn due to the sample being collected at a single time point. Another constraint

pertains to the specific sequencing approach used on the samples. Employing 16S RNA sequencing, indeed, did not yield a high resolution of the microbiome, lacking species- and strain-level

information along with not being able to evaluate and understand their functional and metabolic potential. However, one of the strengths of the study is the substantial number of individuals

for each of the two groups that were enrolled in the clinical study. The participants were followed using a strict protocol, where we carefully selected the individuals who did not undergo

antibiotic treatment nor showed signs of prior or ongoing SARS-CoV-2 infection. Conclusively, we observed that the baseline gut microbiota diversity and abundance significantly affected

immunologic responses to SARS-CoV-2 vaccines, irrespective of age, gender, and total IgG. Although age had a significant association with both the microbial diversity and spike IgG levels,

results from multiple regression analyses showed that baseline microbial abundance was independent of age and had a dominant impact on modulating antibody response to the COVID-19 vaccine.

In summary, we describe novel findings on a strong association between the intestinal microbial diversity, composition, and immunogenicity of a mRNA SARS-CoV-2 vaccine. These findings may

influence the development of microbiota-centered therapies to optimize immunogenicity and durability of vaccination. For example, the augmentation of vaccination with concomitant treatment

with probiotics, prebiotics, or diet could be a scalable intervention for individuals with lower vaccine responses, like elderly or immunocompromised individuals, such as PLWH. The

feasibility of such therapies is considerable, as a systematic review of results from 26 interventional studies in humans using probiotics to enhance the efficacy of 17 different vaccines

revealed positive outcomes in half of the trials82. Furthermore, a recent preclinical study showed the effectiveness of synbiotics in enhancing immunogenicity of the cholera vaccine in mice

that were colonized with a poorly immunogenic infant’s microbiota83. In addition, the presented microbiome association with the vaccination outcome should warrant more careful use of

treatments that influence the gut microbiome, like antibiotics76. Overall, these results pave the way for future studies, which could lead to a deeper understanding of microbiota modulation

of vaccine responses in different populations and disease conditions. METHODS SARS-COV-2 COVAXID VACCINE COHORT AND SAMPLE COLLECTION An open-label, non-randomized clinical trial was

initiated during spring 2021 at the Karolinska University Hospital, Stockholm, Sweden, to investigate the safety and clinical efficacy of the mRNA BNT162b2 vaccine (Comirnaty®,

Pfizer/BioNTech) in healthy controls and immunocompromised patients (EudraCT no. 2021-000175-37, clinicaltrials.gov no. 2021-000175-37). The ethical permit was granted by the Swedish Ethical

Review Authority (ID 2021-00451), and all participants provided written informed consent (Table 1)84. The study cohort of that study included individuals belonging to six groups84; however,

the current study focused on two of these groups, PLWH (_n_ = 90) and healthy controls (HC) (_n_ = 90). Figure 1a displays the study design, with the fecal samples from the PLWH and HC

group being collected at baseline for DNA extraction before they had received two doses of the mRNA vaccine three weeks apart. The extracted DNA was further sent for 16S rRNA sequencing.

Humoral and cellular responses to SARS-CoV-2 vaccination were evaluated on day 35 after the 1st vaccine dose. Subjects with detectable baseline spike antibodies against SARS-CoV-2,

antibiotic treatment (3 months before vaccination), and those with missing spike IgG data at day 35 were excluded from further analysis (PLWH: _n_ = 22; HC: _n_ = 15). The fecal samples were

collected in RNA/DNA shield (Stratec, Germany). DNA was extracted using ZymoBIOMICS™ DNA Kit (Zymo Research, USA), according to the manufacturer’s specifications, and sequencing analysis

was performed on the MiSeq platform. DETECTION OF SPIKE IGG AND SPIKE CD4+ T-CELL RESPONSES Blood samples were collected at baseline and on day 35. Elecsys® Anti-SARS-CoV-2 S RBD (Roche

Diagnostics) was used to identify and quantify antibodies specific to SARS-CoV-2 spike protein in serum samples84. Spike-specific CD4+ T-cell responses were quantified using

activation-induced marker assays via up-regulation of CD69 and CD40L (CD154), as previously described85. Total IgG levels were analyzed by routine diagnostic methods with the Roche Elecsys

anti-SARS-CoV-2 S enzyme immunoassay, at the Clinical Immunology laboratory of Karolinska University Hospital84. For PLWH, CD4+ and CD8+ T-cell counts and HIV viral load (VL) were determined

by flow cytometry and Cobas Amplicor (Roche Molecular Systems Inc., USA), respectively86. MICROBIOME ANALYSIS Paired-end Illumina reads were checked for quality using FastQC87 and trimmed

using Cutadapt88. During the pre-processing step, primers, adapters, and low-quality (_Q_ < 30) reads were removed. The taxonomic classification and analysis of the trimmed reads were

performed using dada289 within Qiime290 in combination with SILVA database (SILVA v132)91. DADA2 was used for denoising, read pair merging and PCR chimera removal which reduces sequence

errors and dereplicates sequences (Supplementary Table 4). α-diversity of the samples was estimated using the R function _estimate_richness_ in R package phyloseq (v1.30.0)92 and visualized

using R package ggplot2 (v3.3.5)93. The diversity indices such as Observed, Shannon, and Simpson were performed to calculate the richness and diversity of the samples. The distances between

the samples were clustered based on the Bray-Curtis distance metrics and visualized using non-metric multidimensional scaling (NMDS) ordination plots, and the significance of the different

factors on the beta-diversity was calculated based on PERMANOVA using the vegan package (v2.5.7) (Adonis function). The relative abundance of the samples was calculated using Qiime2. Linear

discriminant analysis Effect Size (LEfSe) was employed to determine the significant microbial communities between the groups94. All the barplots were visualized using the R package ggplot2.

LefSe plot of spike CD4+ T-cell response and specific microbial biomarkers were visualized using GraPhlAn95. Multivariate regression analysis was used to predict the factors related to spike

IgG. α-diversity indices, age, gender, total baseline IgG, and BMI were used as covariates in the regression analysis (R package). The models were predicted using stepwise backward

selection, and _p_ values < 0.05 were considered significant. Furthermore, we performed multiple hypothesis testing using a false discovery rate (FDR). Multivariate analysis of variance

(MANOVA) was used to access the different patterns between the multiple dependent variables simultaneously and it was performed using the R function manova(). The Kruskal–Wallis test was

performed for the post hoc analysis for MANOVA. The OTUs belonging to Ruminococcaceae were segregated to assess the richness and diversity of Ruminococcaceae family between elderly and

younger individuals (Supplementary Fig. 10). CORRELATION AND NETWORK ANALYSIS Correlation analyses were based on the Spearman correlation method using R package psych (v2.2.3)96,97.

Benjamini–Hochberg method was used to adjust the _p_ values for the multiple testing. The results were visualized using the R package corrplot (v0.92)98. The input variables for the networks

were bacterial genus, age, spike CD4+ T-cell response, and the antibody level at day 35, visualized using Cytoscape (v3.6.1)99. Differential abundance analysis between high responders and

low responders was performed using R package DESeq2 (v1.26.0)100 and visualized using bubble plots. Bacterial taxa with _p_ values less than 0.05 were considered significant. REPORTING

SUMMARY Further information on research design is available in the Nature Research Reporting Summary linked to this article. DATA AVAILABILITY The metadata and raw 16S rRNA gene sequence

data generated and analyzed during this study are deposited at the NCBI SRA database (Project number: PRJNA902956). CODE AVAILABILITY All the codes are available at GitHub:

https://github.com/Asw614/Gut-microbiome-on-immunological-responses-to-COVID-vaccination.git. REFERENCES * World Health Organization (accessed 14 July 2023). https://covid19.who.int/. *

Polack, F. P. et al. Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine. _N. Engl. J. Med._ 383, 2603–2615 (2020). Article  CAS  PubMed  Google Scholar  * El Sahly, H. M. et al.

Efficacy of the mRNA-1273 SARS-CoV-2 vaccine at completion of blinded phase. _N. Engl. J. Med._ 385, 1774–1785 (2021). Article  CAS  PubMed  Google Scholar  * Antonelli, M. et al. Risk

factors and disease profile of post-vaccination SARS-CoV-2 infection in UK users of the COVID Symptom Study app: a prospective, community-based, nested, case-control study. _Lancet Infect.

Dis._ 22, 43–55 (2022). Article  CAS  PubMed  PubMed Central  Google Scholar  * Baden, L. R. et al. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine. _N. Engl. J. Med._ 384, 403–416

(2021). Article  CAS  PubMed  Google Scholar  * Carazo, S. et al. Estimated protection of prior SARS-CoV-2 infection against reinfection with the Omicron variant among messenger

RNA–vaccinated and nonvaccinated individuals in Quebec, Canada. _JAMA Netw. Open_ 5, e2236670 (2022). Article  PubMed  PubMed Central  Google Scholar  * Connors, J., Bell, M. R., Marcy, J.,

Kutzler, M. & Haddad, E. K. The impact of immuno-aging on SARS-CoV-2 vaccine development. _Geroscience_ 43, 31–51 (2021). Article  CAS  PubMed  PubMed Central  Google Scholar  * Soiza,

R. L., Scicluna, C. & Thomson, E. C. Efficacy and safety of COVID-19 vaccines in older people. _Age Ageing_ 50, 279–283 (2021). Article  PubMed  Google Scholar  * Lynn, D. J. &

Pulendran, B. The potential of the microbiota to influence vaccine responses. _J. Leukoc. Biol._ 103, 225–231 (2018). Article  CAS  PubMed  Google Scholar  * McDermott, A. J. &

Huffnagle, G. B. The microbiome and regulation of mucosal immunity. _Immunology_ 142, 24–31 (2014). Article  CAS  PubMed  PubMed Central  Google Scholar  * Baradaran Ghavami, S. et al.

Cross-talk between immune system and microbiota in COVID-19. _Expert Rev. Gastroenterol. Hepatol._ 15, 1281–1294 (2021). Article  CAS  PubMed  Google Scholar  * Mezouar, S. et al. Microbiome

and the immune system: from a healthy steady-state to allergy associated disruption. _Hum. Microbiome J._ 10, 11–20 (2018). Article  Google Scholar  * Li, D., Breiman, A., Le Pendu, J.

& Uyttendaele, M. Anti-viral effect of Bifidobacterium adolescentis against noroviruses. _Front. Microbiol._ 7, 864 (2016). PubMed  PubMed Central  Google Scholar  * Chiang, B.-L.,

Sheih, Y., Wang, L., Liao, C. & Gill, H. Enhancing immunity by dietary consumption of a probiotic lactic acid bacterium (_Bifidobacterium lactis_ HN019): optimization and definition of

cellular immune responses. _Eur. J. Clin. Nutr._ 54, 849–855 (2000). Article  CAS  PubMed  Google Scholar  * Gill, H. S., Rutherfurd, K. J., Cross, M. L. & Gopal, P. K. Enhancement of

immunity in the elderly by dietary supplementation with the probiotic _Bifidobacterium lactis_ HN019. _Am. J. Clin. Nutr._ 74, 833–839 (2001). Article  CAS  PubMed  Google Scholar  *

Atarashi, K. et al. Induction of colonic regulatory T cells by indigenous Clostridium species. _Science_ 331, 337–341 (2011). Article  CAS  PubMed  Google Scholar  * Chen, J. & Vitetta,

L. Inflammation-modulating effect of butyrate in the prevention of colon cancer by dietary fiber. _Clin. Colorectal Cancer_ 17, e541–e544 (2018). Article  PubMed  Google Scholar  * Furusawa,

Y. et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. _Nature_ 504, 446–450 (2013). Article  CAS  PubMed  Google Scholar  * Lynn, D. J.,

Benson, S. C., Lynn, M. A. & Pulendran, B. Modulation of immune responses to vaccination by the microbiota: implications and potential mechanisms. _Nat. Rev. Immunol._ 22, 33–46 (2022).

Article  CAS  PubMed  Google Scholar  * Vitetta, L., Saltzman, E. T., Thomsen, M., Nikov, T. & Hall, S. Adjuvant probiotics and the intestinal microbiome: enhancing vaccines and

immunotherapy outcomes. _Vaccines_ 5, 50 (2017). Article  PubMed  PubMed Central  Google Scholar  * Huda, M. N. et al. Stool microbiota and vaccine responses of infants. _Pediatrics_ 134,

e362–e372 (2014). Article  PubMed  PubMed Central  Google Scholar  * Harris, V. C. et al. Effect of antibiotic-mediated microbiome modulation on rotavirus vaccine immunogenicity: a human,

randomized-control proof-of-concept trial. _Cell Host Microbe_ 24, 197–207.e194 (2018). Article  CAS  PubMed  Google Scholar  * Harris, V. C. et al. Significant correlation between the

infant gut microbiome and rotavirus vaccine response in rural Ghana. _J. Infect. Dis._ 215, 34–41 (2017). Article  CAS  PubMed  Google Scholar  * Pabst, O. & Hornef, M. Gut microbiota: a

natural adjuvant for vaccination. _Immunity_ 41, 349–351 (2014). Article  CAS  PubMed  Google Scholar  * Nakaya, H. I. et al. Systems biology of vaccination for seasonal influenza in

humans. _Nat. Immunol._ 12, 786–795 (2011). Article  CAS  PubMed  PubMed Central  Google Scholar  * Oh, J. Z. et al. TLR5-mediated sensing of gut microbiota is necessary for antibody

responses to seasonal influenza vaccination. _Immunity_ 41, 478–492 (2014). Article  CAS  PubMed  PubMed Central  Google Scholar  * Kim, D. et al. Recognition of the microbiota by Nod2

contributes to the oral adjuvant activity of cholera toxin through the induction of interleukin‐1β. _Immunology_ 158, 219–229 (2019). Article  CAS  PubMed  PubMed Central  Google Scholar  *

Kim, D. et al. Nod2-mediated recognition of the microbiota is critical for mucosal adjuvant activity of cholera toxin. _Nat. Med._ 22, 524–530 (2016). Article  CAS  PubMed  PubMed Central 

Google Scholar  * Turroni, F. et al. Exploring the diversity of the bifidobacterial population in the human intestinal tract. _Appl. Environ. Microbiol._ 75, 1534–1545 (2009). Article  CAS 

PubMed  PubMed Central  Google Scholar  * Koff, W. C. et al. Development and deployment of COVID-19 vaccines for those most vulnerable. _Sci. Transl. Med._ 13, eabd1525 (2021). Article  CAS

  PubMed  Google Scholar  * Pellini, R. et al. Initial observations on age, gender, BMI and hypertension in antibody responses to SARS-CoV-2 BNT162b2 vaccine. _EClinicalMedicine_ 36, 100928

(2021). Article  PubMed  PubMed Central  Google Scholar  * Stefan, N., Birkenfeld, A. L. & Schulze, M. B. Global pandemics interconnected—obesity, impaired metabolic health and COVID-19.

_Nat. Rev. Endocrinol._ 17, 135–149 (2021). Article  CAS  PubMed  Google Scholar  * Müller, L. et al. Age-dependent immune response to the Biontech/Pfizer BNT162b2 coronavirus disease 2019

vaccination. _Clin. Infect. Dis._ 73, 2065–2072 (2021). Article  PubMed  Google Scholar  * Dandachi, D. et al. Characteristics, comorbidities, and outcomes in a multicenter registry of

patients with human immunodeficiency virus and coronavirus disease 2019. _Clin. Infect. Dis._ 73, e1964–e1972 (2021). Article  CAS  PubMed  Google Scholar  * Geretti, A. M. et al. Outcomes

of coronavirus disease 2019 (COVID-19) related hospitalization among people with human immunodeficiency virus (HIV) in the ISARIC World Health Organization (WHO) clinical characterization

protocol (UK): a prospective observational study. _Clin. Infect. Dis._ 73, e2095–e2106 (2021). Article  CAS  PubMed  Google Scholar  * Hoffmann, C. et al. Immune deficiency is a risk factor

for severe COVID‐19 in people living with HIV. _HIV Med._ 22, 372–378 (2021). Article  CAS  PubMed  Google Scholar  * Chang, M. et al. Changes of gut microbiota in pregnant sows induced by

5-Aminolevulinic acid. _Res. Vet. Sci._ 136, 57–65 (2021). Article  CAS  PubMed  Google Scholar  * Mak, J. W. Y. Modulation of gut microbiota to enhance health and immunity (accessed 2 July

2023). https://clinicaltrials.gov/ct2/show/NCT04884776 (2020). * Falahi, S. & Kenarkoohi, A. Host factors and vaccine efficacy: implications for COVID‐19 vaccines. _J. Med. Virol._ 94,

1330–1335 (2022). Article  CAS  PubMed  Google Scholar  * Chen, J., Vitetta, L., Henson, J. D. & Hall, S. The intestinal microbiota and improving the efficacy of COVID-19 vaccinations.

_J. Funct. Foods_ 87, 104850 (2021). Article  CAS  PubMed  PubMed Central  Google Scholar  * Zhao, T. et al. Influence of gut microbiota on mucosal IgA antibody response to the polio

vaccine. _npj Vaccines_ 5, 47 (2020). Article  CAS  PubMed  PubMed Central  Google Scholar  * Praharaj, I. et al. Influence of nonpolio enteroviruses and the bacterial gut microbiota on oral

poliovirus vaccine response: a study from South India. _J. Infect. Dis._ 219, 1178–1186 (2019). Article  CAS  PubMed  Google Scholar  * Ichinohe, T. et al. Microbiota regulates immune

defense against respiratory tract influenza A virus infection. _Proc. Natl Acad. Sci. USA._ 108, 5354–5359 (2011). Article  CAS  PubMed  PubMed Central  Google Scholar  * Hagan, T. et al.

Antibiotics-driven gut microbiome perturbation alters immunity to vaccines in humans. _Cell_ 178, 1313–1328.e1313 (2019). Article  CAS  PubMed  PubMed Central  Google Scholar  * Wells, J. M.

Immunomodulatory mechanisms of lactobacilli. _Microbial Cell Factories_ 1–15 (BioMed Central, 2011). * Wexler, H. M. Bacteroides: the good, the bad, and the nitty-gritty. _Clin. Microbiol.

Rev._ 20, 593–621 (2007). Article  CAS  PubMed  PubMed Central  Google Scholar  * Wright, E. K. et al. Microbial factors associated with postoperative Crohn’s disease recurrence. _J. Crohns

Colitis_ 11, 191–203 (2017). Article  PubMed  Google Scholar  * Chilton, P. M., Hadel, D. M., To, T. T., Mitchell, T. C. & Darveau, R. P. Adjuvant activity of naturally occurring

monophosphoryl lipopolysaccharide preparations from mucosa-associated bacteria. _Infect. Immun._ 81, 3317–3325 (2013). Article  CAS  PubMed  PubMed Central  Google Scholar  * Fix, J. et al.

Association between gut microbiome composition and rotavirus vaccine response among Nicaraguan infants. _Am. J. Tropical Med. Hyg._ 102, 213 (2020). Article  Google Scholar  * Zafar, H.

& Saier, M. H. Jr Gut Bacteroides species in health and disease. _Gut Microbes_ 13, 1848158 (2021). Article  PubMed  PubMed Central  Google Scholar  * Domingo, M.-C. et al.

Cloacibacillus sp., a potential human pathogen associated with bacteremia in Quebec and New Brunswick. _J. Clin. Microbiol._ 53, 3380–3383 (2015). Article  PubMed  PubMed Central  Google

Scholar  * Dong, T. S. et al. Gut microbiome profiles associated with steatosis severity in metabolic associated fatty liver disease. Hepatoma Res. 7,

https://doi.org/10.20517/2394-5079.2021.55 (2021). * Bian, G. et al. The gut microbiota of healthy aged Chinese is similar to that of the healthy young. _Msphere_ 2, e00327–00317 (2017).

Article  PubMed  PubMed Central  Google Scholar  * Borgognone, A. et al. Gut microbiome signatures linked to HIV-1 reservoir size and viremia control. _Microbiome_ 10, 59 (2022). Article 

CAS  PubMed  PubMed Central  Google Scholar  * Huda, M. N. et al. Bifidobacterium abundance in early infancy and vaccine response at 2 years of age. _Pediatrics_ 143, e20181489 (2019). * Ng,

S. C. et al. Gut microbiota composition is associated with SARS-CoV-2 vaccine immunogenicity and adverse events. _Gut_ 71, 1106–1116 (2022). Article  CAS  PubMed  Google Scholar  * Ray, S.

et al. Altered gut microbiome under antiretroviral therapy: Impact of efavirenz and zidovudine. _ACS Infect. Dis._ 7, 1104–1115 (2020). Article  PubMed  PubMed Central  Google Scholar  *

Nowak, P. et al. Gut microbiota diversity predicts immune status in HIV-1 infection. _Aids_ 29, 2409–2418 (2015). Article  CAS  PubMed  Google Scholar  * Ma, T. et al. Altered

mucosa-associated microbiota in the ileum and colon of neonatal calves in response to delayed first colostrum feeding. _J. Dairy Sci._ 102, 7073–7086 (2019). Article  CAS  PubMed  Google

Scholar  * Vojinovic, D. et al. Relationship between gut microbiota and circulating metabolites in population-based cohorts. _Nat. Commun._ 10, 5813 (2019). Article  CAS  PubMed  PubMed

Central  Google Scholar  * Gryaznova, M. et al. Dynamics of changes in the gut microbiota of healthy mice fed with lactic acid bacteria and bifidobacteria. _Microorganisms_ 10, 1020 (2022).

Article  CAS  PubMed  PubMed Central  Google Scholar  * Wei, Z.-Y. et al. Characterization of Changes and Driver Microbes in Gut Microbiota During Healthy Aging Using A Captive Monkey Model.

_Genomics, Proteomics Bioinformatics_ 20, 350–365 (2022). Article  PubMed  Google Scholar  * Matsumoto, N. et al. Relationship between nutrient intake and human gut microbiota in

monozygotic twins. _Medicina_ 57, 275 (2021). Article  PubMed  PubMed Central  Google Scholar  * Zhang, H. et al. Identification reproducible microbiota biomarkers for the diagnosis of

cirrhosis and hepatocellular carcinoma. _AMB Express_ 13, 1–14 (2023). Article  CAS  Google Scholar  * Tuikhar, N. et al. Comparative analysis of the gut microbiota in centenarians and young

adults shows a common signature across genotypically non-related populations. _Mech. Ageing Dev._ 179, 23–35 (2019). Article  PubMed  Google Scholar  * Kong, F. et al. Gut microbiota

signatures of longevity. _Curr. Biol._ 26, R832–R833 (2016). Article  CAS  PubMed  Google Scholar  * Yu, X. et al. Analysis of the intestinal microbial community structure of healthy and

long-living elderly residents in Gaotian Village of Liuyang City. _Appl. Microbiol. Biotechnol._ 99, 9085–9095 (2015). Article  CAS  PubMed  Google Scholar  * Biagi, E. et al. Through

ageing, and beyond: gut microbiota and inflammatory status in seniors and centenarians. _PLoS One_ 5, e10667 (2010). Article  PubMed  PubMed Central  Google Scholar  * van Tongeren, S. P.,

Slaets, J. P., Harmsen, H. J. & Welling, G. W. Fecal microbiota composition and frailty. _Appl. Environ. Microbiol._ 71, 6438–6442 (2005). Article  PubMed  PubMed Central  Google Scholar

  * Pryde, S. E., Duncan, S. H., Hold, G. L., Stewart, C. S. & Flint, H. J. The microbiology of butyrate formation in the human colon. _FEMS Microbiol. Lett._ 217, 133–139 (2002).

Article  CAS  PubMed  Google Scholar  * Chassard, C. & Bernalier‐Donadille, A. H2 and acetate transfers during xylan fermentation between a butyrate‐producing xylanolytic species and

hydrogenotrophic microorganisms from the human gut. _FEMS Microbiol. Lett._ 254, 116–122 (2006). Article  CAS  PubMed  Google Scholar  * Mariat, D. et al. The Firmicutes/Bacteroidetes ratio

of the human microbiota changes with age. _BMC Microbiol._ 9, 1–6 (2009). Article  Google Scholar  * Vaiserman, A. et al. Differences in the gut Firmicutes to Bacteroidetes ratio across age

groups in healthy Ukrainian population. _BMC Microbiol._ 20, 1–8 (2020). Article  Google Scholar  * Neff, C. P. et al. Diverse intestinal bacteria contain putative zwitterionic capsular

polysaccharides with anti-inflammatory properties. _Cell Host Microbe_ 20, 535–547 (2016). Article  CAS  PubMed  PubMed Central  Google Scholar  * Ashkar, A. A., Mossman, K. L., Coombes, B.

K., Gyles, C. L. & Mackenzie, R. FimH adhesin of type 1 fimbriae is a potent inducer of innate antimicrobial responses which requires TLR4 and type 1 interferon signalling. _PLoS

Pathog._ 4, e1000233 (2008). Article  PubMed  PubMed Central  Google Scholar  * Cheung, K.-S. et al. Association between recent usage of antibiotics and immunogenicity within six months

after COVID-19 vaccination. _Vaccines_ 10, 1122 (2022). Article  CAS  PubMed  PubMed Central  Google Scholar  * Scott, S. A., Fu, J. & Chang, P. V. Microbial tryptophan metabolites

regulate gut barrier function via the aryl hydrocarbon receptor. _Proc. Natl Acad. Sci. USA._ 117, 19376–19387 (2020). Article  CAS  PubMed  PubMed Central  Google Scholar  * Garrett, W. S.

Immune recognition of microbial metabolites. _Nat. Rev. Immunol._ 20, 91–92 (2020). Article  CAS  PubMed  Google Scholar  * Kim, M., Qie, Y., Park, J. & Kim, C. H. Gut microbial

metabolites fuel host antibody responses. _Cell Host Microbe_ 20, 202–214 (2016). Article  CAS  PubMed  PubMed Central  Google Scholar  * Gautier, T. et al. Next-generation probiotics and

their metabolites in COVID-19. _Microorganisms_ 9, 941 (2021). Article  CAS  PubMed  PubMed Central  Google Scholar  * O’Toole, P. W., Marchesi, J. R. & Hill, C. Next-generation

probiotics: the spectrum from probiotics to live biotherapeutics. _Nat. Microbiol._ 2, 1–6 (2017). Article  Google Scholar  * Zimmermann, P. & Curtis, N. The influence of probiotics on

vaccine responses–a systematic review. _Vaccine_ 36, 207–213 (2018). Article  CAS  PubMed  Google Scholar  * Di Luccia, B. et al. Combined prebiotic and microbial intervention improves oral

cholera vaccination responses in a mouse model of childhood undernutrition. _Cell Host Microbe_ 27, 899–908.e895 (2020). Article  PubMed  PubMed Central  Google Scholar  * Bergman, P. et al.

Safety and efficacy of the mRNA BNT162b2 vaccine against SARS-CoV-2 in five groups of immunocompromised patients and healthy controls in a prospective open-label clinical trial.

_EBioMedicine_ 74, 103705 (2021). * Gao, Y. et al. Ancestral SARS-CoV-2-specific T cells cross-recognize the Omicron variant. _Nat. Med._ 28, 472–476 (2022). Article  CAS  PubMed  PubMed

Central  Google Scholar  * Xu, X., Vesterbacka, J., Aleman, S. & Nowak, P. High seroconversion rate after vaccination with mRNA BNT162b2 vaccine against SARS-CoV-2 among people with

HIV–but HIV viremia matters? _AIDS_ 36, 479–481 (2022). Article  CAS  PubMed  Google Scholar  * Andrews, S. _Babraham Bioinformatics_. (Babraham Institute, Cambridge, United Kingdom, 2010).

Google Scholar  * Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. _EMBnet. J._ 17, 10–12 (2011). Article  Google Scholar  * Callahan, B. J. et al. DADA2:

High-resolution sample inference from Illumina amplicon data. _Nat. Methods_ 13, 581–583 (2016). Article  CAS  PubMed  PubMed Central  Google Scholar  * Bolyen, E. et al. Reproducible,

interactive, scalable and extensible microbiome data science using QIIME 2. _Nat. Biotechnol._ 37, 852–857 (2019). Article  CAS  PubMed  PubMed Central  Google Scholar  * Quast, C. et al.

The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. _Nucleic acids Res._ 41, D590–D596 (2012). Article  PubMed  PubMed Central  Google Scholar  *

McMurdie, P. J. & Holmes, S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. _PloS one_ 8, e61217 (2013). Article  CAS  PubMed 

PubMed Central  Google Scholar  * Ginestet, C. ggplot2: Elegant Graphics for Data Analysis. _Journal of the Royal Statistical Society_ _Series A: Statistics in Society_ 174, 245–246 (2011).

* Segata, N. et al. Metagenomic biomarker discovery and explanation. _Genome Biol._ 12, 1–18 (2011). Article  Google Scholar  * Asnicar, F., Weingart, G., Tickle, T. L., Huttenhower, C.

& Segata, N. Compact graphical representation of phylogenetic data and metadata with GraPhlAn. _PeerJ_ 3, e1029 (2015). Article  PubMed  PubMed Central  Google Scholar  * Revelle, W.

& Revelle, M. W. Package ‘psych’. _The Comprehensive R Archive Network_ 337, 338 (2015). * Revelle, W. _psych: Procedures for Psychological, Psychometric, and Personality Research._

Northwestern University, Evanston, Illinois. R package version 2.2.3 (2022). Retrieved from https://CRAN.R-project.org/package=psych. * Wei, T. et al. _package “corrplot”: Visualization of a

Correlation Matrix_. 2017. Version 0.84 (2021). * Shannon, P. et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. _Genome Res._ 13,

2498–2504 (2003). Article  CAS  PubMed  PubMed Central  Google Scholar  * Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with

DESeq2. _Genome Biol._ 15, 1–21 (2014). Article  Google Scholar  Download references ACKNOWLEDGEMENTS We thank all the participants enrolled in this study, and all the research and clinical

staff at Karolinska University Hospital. We would like to thank SciLifeLab National Genomics Infrastructure in Stockholm, and SNIC/Uppsala Multidisciplinary Center for Advanced Computational

Science for assistance with sequencing and access to the UPPMAX computational infrastructure. The authors also acknowledge the assistance and support from the Centre for Bioinformatics and

Biostatistics (CBB). The study was supported by grants from the Swedish Physicians Against AIDS research fund (FOa2021-0009; P.N., FOb2020-0019; S.R.), Stockholm County Council (SLL-KI for

P.N.; ALF nr 20190451), Karolinska Institute Research Foundation Grants (2020-02198; S.R.), Swedish Research Council (Dnr 2016-01675; A.S.), EuCARE project “European cohorts of patients and

schools to advance response to epidemics” funded by the EC under HORIZON-HLTH-2021-CORONA-01 Grant No. 101046016. FUNDING Open access funding provided by Karolinska Institute. AUTHOR

INFORMATION AUTHORS AND AFFILIATIONS * Department of Medicine Huddinge, Division of Infectious Diseases, Karolinska Institutet, Stockholm, Sweden Shilpa Ray, Aswathy Narayanan, Jan

Vesterbacka, Ola Blennow, Soo Aleman, Anders Sönnerborg & Piotr Nowak * Department of Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden Jan Vesterbacka, Ola Blennow,

 Soo Aleman, Anders Sönnerborg & Piotr Nowak * Department of Medicine Huddinge, Center for Infectious Medicine, Karolinska Institutet, Stockholm, Sweden Puran Chen, Yu Gao, Hans-Gustaf

Ljunggren & Marcus Buggert * Department of Dental Medicine, Karolinska Institutet, Stockholm, Sweden Giorgio Gabarrini & Margaret Sällberg Chen * National Bioinformatics

Infrastructure Sweden (NBIS), SciLifeLab, Department of Laboratory Medicine, Lund University, Lund, Sweden Lokeshwaran Manoharan * Department of Laboratory Medicine, Division of Clinical

Microbiology, ANA Futura, Karolinska Institutet, Stockholm, 141 52, Sweden Anders Sönnerborg Authors * Shilpa Ray View author publications You can also search for this author inPubMed Google

Scholar * Aswathy Narayanan View author publications You can also search for this author inPubMed Google Scholar * Jan Vesterbacka View author publications You can also search for this

author inPubMed Google Scholar * Ola Blennow View author publications You can also search for this author inPubMed Google Scholar * Puran Chen View author publications You can also search

for this author inPubMed Google Scholar * Yu Gao View author publications You can also search for this author inPubMed Google Scholar * Giorgio Gabarrini View author publications You can

also search for this author inPubMed Google Scholar * Hans-Gustaf Ljunggren View author publications You can also search for this author inPubMed Google Scholar * Marcus Buggert View author

publications You can also search for this author inPubMed Google Scholar * Lokeshwaran Manoharan View author publications You can also search for this author inPubMed Google Scholar *

Margaret Sällberg Chen View author publications You can also search for this author inPubMed Google Scholar * Soo Aleman View author publications You can also search for this author inPubMed

 Google Scholar * Anders Sönnerborg View author publications You can also search for this author inPubMed Google Scholar * Piotr Nowak View author publications You can also search for this

author inPubMed Google Scholar CONTRIBUTIONS Conception, planning, fund acquisition: P.N., S.A., M.B., H.-G.L., A.S. Coordinated the sample collection: P.N., J.V., O.B., P.C., S.A. Designed

the experiments: P.N., M.B. and S.R. Performed experiments and analyzed the data: S.R., A.N., G.G., Y.G., M.S.C., and L.M. Wrote the paper: S.R., A.N., and P.N. Reviewed and/or edited the

manuscript: all authors. CORRESPONDING AUTHOR Correspondence to Shilpa Ray. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. ETHICS APPROVAL AND CONSENT TO

PARTICIPATE The ethical permit was reviewed and approved by the Swedish Ethical Review Authority (ID 2021-00451), and all participants provided written informed consent. ADDITIONAL

INFORMATION PUBLISHER’S NOTE Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. SUPPLEMENTARY INFORMATION SUPPLEMENTARY

FIGURES SUPPLEMENTARY TABLES REPORTING SUMMARY RIGHTS AND PERMISSIONS OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use,

sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative

Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated

otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds

the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Reprints and

permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Ray, S., Narayanan, A., Vesterbacka, J. _et al._ Impact of the gut microbiome on immunological responses to COVID-19 vaccination in healthy

controls and people living with HIV. _npj Biofilms Microbiomes_ 9, 104 (2023). https://doi.org/10.1038/s41522-023-00461-w Download citation * Received: 05 May 2023 * Accepted: 20 November

2023 * Published: 20 December 2023 * DOI: https://doi.org/10.1038/s41522-023-00461-w SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get

shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative