Borgognone et al. Microbiome (2022) 10:59 https://doi.org/10.1186/s40168-022-01247-6 RESEARCH Open Access Gut microbiome signatures linked to HIV-1 reservoir size and viremia control Alessandra Borgognone1* , Marc Noguera‑Julian1,2,3, Bruna Oriol1,4, Laura Noël‑Romas5,6, Marta Ruiz‑Riol1,2, Yolanda Guillén7, Mariona Parera1, Maria Casadellà1, Clara Duran1,4, Maria C. Puertas1,2, Francesc Català‑Moll1, Marlon De Leon5, Samantha Knodel5,6, Kenzie Birse5,6, Christian Manzardo8, José M. Miró2,8, Bonaventura Clotet1,2,3,4,9,10, Javier Martinez‑Picado1,2,3,11, José Moltó2,9,10, Beatriz Mothe1,2,3,9,10, Adam Burgener5,6,12, Christian Brander1,2,3,11, Roger Paredes1,2,3,4,5,9,10* and the BCN02 Study Group Abstract Background: The potential role of the gut microbiome as a predictor of immune-mediated HIV-1 control in the absence of antiretroviral therapy (ART) is still unknown. In the BCN02 clinical trial, which combined the MVA.HIVconsv immunogen with the latency-reversing agent romidepsin in early-ART treated HIV-1 infected individuals, 23% (3/13) of participants showed sustained low-levels of plasma viremia during 32 weeks of a monitored ART pause (MAP). Here, we present a multi-omics analysis to identify compositional and functional gut microbiome patterns associated with HIV-1 control in the BCN02 trial. Results: Viremic controllers during the MAP (controllers) exhibited higher Bacteroidales/Clostridiales ratio and lower microbial gene richness before vaccination and throughout the study intervention when compared to non-control‑ lers. Longitudinal assessment indicated that the gut microbiome of controllers was enriched in pro-inflammatory bacteria and depleted in butyrate-producing bacteria and methanogenic archaea. Functional profiling also showed that metabolic pathways related to fatty acid and lipid biosynthesis were significantly increased in controllers. Fecal metaproteome analyses confirmed that baseline functional differences were mainly driven by Clostridiales. Par‑ ticipants with high baseline Bacteroidales/Clostridiales ratio had increased pre-existing immune activation-related transcripts. The Bacteroidales/Clostridiales ratio as well as host immune-activation signatures inversely correlated with HIV-1 reservoir size. Conclusions: The present proof-of-concept study suggests the Bacteroidales/Clostridiales ratio as a novel gut micro‑ biome signature associated with HIV-1 reservoir size and immune-mediated viral control after ART interruption. Keywords: HIV-1 post-treatment control, Gut microbiome, Therapeutic HIV-1 vaccine, Viral reservoir size, Microbiome-based predictive biomarker Background A major obstacle to HIV-1 cure is the persistence of viral reservoirs. This mainly refers to latently infected cells carrying transcriptionally silent, replication-competent *Correspondence: aborgognone@irsicaixa.es; rparedes@irsicaixa.es viruses which evade antiretroviral therapy (ART) as well 1 IrsiCaixa AIDS Research Institute, Hospital Universitari Germans Trias i Pujol, Barcelona, Catalonia, Spain as immune-mediated clearance [1–3]. The immune sys- 10 Department of Infectious Diseases Service, Germans Trias i Pujol tem is generally unable to contain HIV-1 replication University Hospital, Barcelona, Catalonia, Spain in the absence of ART [4]. However, up to 10–20% of Full list of author information is available at the end of the article © The Author(s) 2022. 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 licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence 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 licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Borgognone et al. Microbiome (2022) 10:59 Page 2 of 15 subjects that initiate ART within first weeks after HIV-1 reversing agent [21] in combination with a therapeutic acquisition may temporarily achieve HIV-1 viremia sup- HIV vaccine (MVA.HIVconsv) in a group of early-ART pression after ART interruption (ATI) [5]. Understand- treated HIV-1-infected individuals [22, 23]. During a ing the mechanisms behind immune-mediated viremia monitored ART interruption (MAP), 23% of individu- control after ATI is key to progress towards a functional als showed sustained viremia control up to 32 weeks of HIV cure. Broader and higher-magnitude CTL (cyto- follow-up. toxic T-lymphocyte) responses against less diverse HIV-1 Here, we aimed to identify salient compositional and epitopes [6, 7] in the context of favorable HLA class I functional gut microbiome patterns associated with con- genotypes [8] and smaller HIV-1 reservoir size [9] have trol of HIV-1 viremia after ART interruption in the “kick all been related to such post-treatment HIV-1 control. and kill” strategy used in the BCN02 study. There is indirect evidence that the gut microbiome might also contribute to immune-mediated control of Materials and methods HIV-1 replication [10, 11]. Vaccine-induced gut micro- Study design biome alterations, consisting in lower bacterial diver- This was a sub-study derived from the BCN02 clinical sity and negative correlation between richness and trial (NCT02616874). The BCN02 was a multicenter, ­CD14+DR- monocytes in colorectal intraepithelial lym- open-label, single-arm, phase I, proof-of-concept clini- phocytes, have been recently associated with HIV/SIV cal trial in which 15 HIV-1-infected individuals with sus- (SHIV) protection in a non-human primate challenge tained viral suppression who started ART within the first study after mucosal vaccination with HIV/SIV peptides, 6 months after HIV transmission were enrolled to evalu- modified vaccinia Ankara–SIV and HIV-gp120–CD4 ate the safety, tolerability, immunogenicity, and effect on fusion protein plus adjuvants through the oral route the viral reservoir of a kick and kill strategy consisting of [12]. In the HVTN 096 trial [13], where the impact of the combination of HIVconsv vaccines with romidepsin, the gut microbiota on HIV-specific immune response to given as a latency reversing agent (LRA) [20] (Additional a DNA-prime, poxvirus-boost strategy in human adults file 1: Fig. S1a). Romidepsin is a histone deacetylase was assessed, baseline and vaccine-induced gp41-reactive inhibitor (HDACi), developed as an anti-cancer drug, IgG titers were associated with different microbiota com- which has been shown to induce HIV-1 transcription munity structures, in terms of richness and composition both in vitro and in vivo [21, 24]. The HIVconsv immuno- [14]. In particular, co-occurring bacterial groups, such as gen was constructed by assembling 14 highly conserved Ruminococcaceae, Peptoniphilaceae, and Bacteroidaceae, regions derived from HIV-1 genes Gag, Pol, Vif, and Env were associated with vaccine-induced IgG response and alternating, for each domain, the consensus sequence of inversely correlated with pre-existing gp41 binding IgG the four major HIV-1 clades A, B, C, and D and deliv- antibodies, suggesting that the microbiome may influ- ered by non-replicating poxvirus MVA vector [20]. Fif- ence the immune response and vaccine immunogenicity teen individuals enrolled in the BCN02 trial (procedures [14]. Also, another study evidence has shown that HIV for recruitment and eligibility criteria are detailed else- vaccine-induced CD4+ T and B cell responses could be where [20]) were immunized with a first dose of MVA. imprinted by prior exposure to cross-reactive intesti- HIVconsv (MVA1, 2 × ­108 pfu intramuscularly), fol- nal microbiota-derived antigens [15]. Further evidence lowed by three weekly-doses of romidepsin ­(RMD1-2-3, 5 emerged from other studies in typhoid Ty21 [16], rota- mg/m2 BSA intravenously) and a second boost of MVA. virus [17] and oral polio virus, tetanus-toxoid, bacillus HIVconsv (MVA2, 2 × ­108 pfu intramuscularly) 4 weeks Calmette-Guérin, and hepatitis B immunization strate- after the last ­RMD3 infusion. To assess the ability for viral gies [18], in which specific gut microbiome signatures control after ART interruption, participants underwent a (Bifidobacterium, Streptococcus bovis, and Clostridiales, monitored antiviral pause (MAP), 8 weeks after the sec- respectively) positively correlated with vaccine-induced ond vaccination (MVA2), for a maximum of 32 weeks or immune response. In the absence of immune correlates until any ART resumption criteria were met (plasma viral of viral control, HIV cure trials usually incorporate an load > 2000 copies/ml, ­CD4+ cell counts < 500 cells/mm3 ART interruption phase to address the efficacy of a thera- and/or development of clinical symptoms related to an peutic intervention [19]. Data on the role of gut micro- acute retroviral syndrome [20]). The study was conducted biome composition in the responsiveness to a curative between February 2016 and October 2017 at two HIV-1 strategy and the relationship with viral control after ART units from university hospitals in Barcelona (Hospital interruption are lacking. The BCN02 study [20] was a Germans Trias i Pujol and Hospital Clínic) and a commu- single-arm, proof-of-concept “kick and kill” clinical trial nity center (BCN-Checkpoint, Barcelona). The microbi- evaluating the safety and the in vivo effects of the his- ome sub-study concept, design, and patient information tone deacetylase inhibitor romidepsin given as a latency were reviewed and approved by the institutional ethical Borgognone et al. Microbiome (2022) 10:59 Page 3 of 15 review board of the participating institutions (Reference Assay based on the Olink Inflammation Panel [32]. Cor- Nr AC-15-108-R) and by the Spanish Regulatory Author- relations between ‘omic’ datasets were computed using ities (EudraCT 2015-002300-84). Written informed con- Spearman’s correlation coefficients and integrative multi- sent was provided by all study participants in accordance omics analysis was assessed based on the mixOmics R to the principles expressed in the Declaration of Helsinki package [33]. A detailed description of wet-lab proce- and local personal data protection law (LOPD 15/1999). dures, bioinformatic methods and statistical analysis of metagenome, metaproteome, transcriptome, soluble Sample disposition and data analysis plasma markers and multi-omics data is available in the Fourteen participants from the BCN02 trial consented Additional file 1: Supplementary methods. to participate in the BCN02-microbiome study, 1 was excluded due to a protocol violation during MAP, and Results 13 were included for multi-omics analyses. Twelve from Patient characteristics the thirteen participants that finalized the “kick and kill” In this microbiome sub-study, we evaluated 13 par- intervention completed the MAP phase (n = 3 control- ticipants of the BCN02 study. Three had sustained low- lers and n = 9 non-controllers) and one subject (B07) did level HIV plasma viremia (< 2000 copies/ml) during 32 not enter the MAP period due to immune futility pre- weeks of MAP (viremic controllers), whereas 9 devel- defined criteria and absence of protective HLA class I oped HIV-1 RNA rebound (> 2000 copies/ml) during protective alleles associated with natural HIV-1 control MAP (non-controllers). One additional subject (B07) did (Additional file 1: Fig. S1b). Based on the gut microbiome not qualify for MAP due to pre-specified immune futil- similarity with non-controllers at study entry and over ity criteria and absence of protective HLA alleles, and the “kick and kill” intervention, the participant B07 was therefore, was also considered a non-controller in this included in the non-controller arm to increase the sta- microbiome study. (Additional file 1: Fig. S1b). Study par- tistical power in this microbiome sub-study. Fecal speci- ticipants were predominantly MSM (92%) of Caucasian mens were longitudinally collected at BCN02 during the ethnicity (92%), with median age of 42 years and median intervention period at study entry (pre-Vax), 1 week after body mass index of 22.9 kg/m2 (Table 1). Median base- 1st vaccination (MVA1), 1 week after R ­ MD3 (RMD) and line ­CD4+ T cell counts was 728 (416–1408) cells/mm3 4 weeks after 2nd vaccination (MVA2). Samples were also and median CD4/CD8 T cell ratio was 1.4 (0.97–1.9). All collected over the MAP period (from 4 to 34 weeks after subjects had been on integrase strand-transfer inhibitor- ART interruption) and 24 weeks after ART resumption based triple ART for >3 years, begun during the first 3 (Additional file 1: Fig. S1a). All samples were processed months after HIV-1 infection. Median baseline HIV-1 for shotgun metagenomics analysis. Taxonomical clas- proviral DNA was 140 copies/106 ­CD4+ T cells, being sification, microbial gene content and functional profil- numerically lower in controllers than in non-controllers ing were inferred using Metaphlan2 [25], IGC reference (65 vs 165 copies/106 ­CD4+ T cells, p = 0.29). catalog [26], and HUMAnN2 [27], respectively. Sequenc- ing analysis and quality control of metagenomics data are Baseline gut‑associated Bacteroidales/Clostridiales provided in the Additional file 1: Supplementary results. ratio and lower microbial gene richness discriminate To facilitate the interpretation, longitudinal time points between viremic controllers and non‑controllers were schematically grouped into three phases (Additional Viremic controllers had significantly higher Bacteroidales file 1: Fig. S2a). Fecal material, peripheral blood mono- levels than non-controllers at study entry (pre-Vax p = nuclear cells (PBMC), and plasma samples were also 0.007) and during all the intervention phase (MVA1 p = sampled at baseline to assess fecal metaproteome, host 0.049, RMD p = 0.049 and MVA2, p = 0.014) (Fig. 1a) transcriptome profiles and soluble inflammation bio- as well as lower baseline Clostridiales abundance (p = markers, respectively (Additional file 1: Fig. S2b). Micro- 0.014) (Fig. 1b). Accordingly, the Bacteroidales/Clostridi- bial proteins from fecal samples were measured by mass ales ratio remained significantly higher in controllers at spectrometry and protein identification performed using study entry and throughout the intervention (pre-Vax p Mascot search engine (v2.4, Matrix Science) and Scaffold = 0.007 and MVA2, p = 0.028) (Fig. 1c). Also, controllers Q+ software (v4.9.0, Proteome Software) [28]. PBMC were significantly depleted in archaeal members from transcriptomes were evaluated using RNA-sequenc- the methanogenic order Methanobacteriales (Additional ing and sequence reads aligned to the human reference file 1: Fig. S4). More detailed analyses at lower taxonomic genome by STAR v2.5.3a [29]. Read counts estimation level within the orders Bacteroidales and Clostridi- was inferred using RSEM v1.3.0 [30] and differential ales showed that controllers were mainly depleted in expression analysis performed by DESeq2 [31]. Plasma Clostridiales species, such as Eubacterium spp. and Sub- proteins were estimated using the Proximity Extension doligranulum spp., whereas the Bacteroidales species Borgognone et al. Microbiome (2022) 10:59 Page 4 of 15 Table 1 Study participant demographics and clinical characteristics Variable All participants (n = 13) Non-controllers (n = 10) Controllers (n = 3) Demographics Sex (M/F), n 12/1 9/1 3/0 Risk group (MSM/HTS), n 12/1 9/1 3/0 Ethnic group (Caucasian/Latin), n 12/1 9/1 3/0 Age (years) 42 (39–47) 43 (39–47) 34 (33–38) BMI (kg/m2) 22.9 (20.9–24) 22.3 (21.1–23.4) 24.3 (22.2–25) Treatment and clinical characteristics ART regimen, n (TDF_FTC_RAL/ABC_3TC_RAL/ 2/9/2 2/6/2 0/3/0 ABC_3TC_DTG) Viral reservoir (HIV-1 DNA cp/106 ­CD4+ T-cells) 140 (65–361) 165 (76.2–415.7) 65 (62.5–116.5)  ­CD4+ T cell (cells/mm3) 728 (648–1182) 839 (581.8–1293.8) 657 (652.5–814)  ­CD4+ T cell (%) 42.9 (42.2–49.3) 43.4 (42.3–48.1) 42.2 (38.4–48.1) CD4/CD8 T cell counts ratio 1.4 (1.2–1.6) 1.4 (1.2–1.5) 1.3 (1.1–1.6) Continuous data are presented using median, 25% and 75% interquartile range, unless otherwise described No statistically significant differences were observed (p ≤ 0.05; Wilcoxon rank-sum test) M Male, F Female, MSM Men who have sex with men, HTS Heterosexual, BMI Body mass index, ART​Antiretroviral therapy, cp Copies, TDF Tenofovir disoproxil fumarate, FTC Emtricitabine, RAL Raltegravir, ABC Abacavir, 3TC Lamivudine, DTG Dolutegravir Prevotella copri was significantly higher after the three together, these data showed that differences between romidepsin doses (Additional file 1: Fig. S5), being such controllers and non-controllers mainly emerged from trends maintained throughout the intervention (Addi- resident microbial communities, before any intervention tional file 1: Figs. S5 and S6). Viremic controllers also had was started in BCN02 study. Thus, subsequent analyses lower microbial gene richness than non-controllers at the were focused on characterizing further discriminant sig- study entry (p = 0.028) and MVA1 (p = 0.049), although natures at study entry. such differences lost statistical significance in the RMD and MVA2 assessments (Fig. 2a). Alpha diversity also Increased Bacteroidales/Clostridiales ratio in viremic remained numerically lower in controllers, but differ- controllers negatively correlated with longitudinal HIV‑1 ences were not statistically significant (Fig. 2b). Control- viral reservoir size lers exhibited lower beta-diversity, particularly already The Bacteroidales/Clostridiales ratio inversely and sig- at the study entry (p = 0.004; Additional file 1: Fig. S7), nificantly correlated with longitudinal total ­CD4+ T cell- and showed less intra-host longitudinal evolution than associated HIV-1 DNA measured at study entry (rho = non-controllers (p = 0.001) (Fig. 2c). Whereas the gut − 0.6, padj = 0.03) and over the intervention, whereas microbiome composition of controllers was signifi- an opposite trend was observed for gene richness (rho cantly different from that of non-controllers (p = 0.001), = 0.65, padj = 0.01 at study entry) (Fig. 3a). A similar no significant longitudinal differences were observed trend was observed for cell-associated (CA) HIV-1 RNA across time points (p = 0.815), suggesting that the com- (Bacteroidales/Clostridiales; rho = − 0.7, padj = 0.01 bined intervention did not significantly alter the gut- and gene richness; rho = 0.61, padj = 0.0 at study entry), microbiome composition (Fig. 2c). Of note, results did although stronger correlations were found at RMD and not change after removing B07 from the non-controller MVA2 (Fig. 3b). In both assessments, alpha-diversity arm (Additional file 1: Fig. S8). No differences in Bac- (Shannon index) exhibited weak positive correlation teroidales and Clostridiales abundances and microbial with the viral reservoir, being correlations not significant. diversity were observed during MAP and after ART re- Moreover, baseline ratio Bacteroidales/Clostridiales and initiation (Additional file 1: Supplementary results). Fur- gene richness showed a strong negative correlation (r ho thermore, longitudinal profiling of metabolic pathways = − 0.87, padj = 0.0001) (Fig. 3a, b), in line with trends associated with Bacteroidales, Clostridiales, and archaea observed in the microbiota characterization. In the lon- showed that controllers were mainly enriched in func- gitudinal comparison, controllers tended to displayed tions related to fatty acid biosynthesis, whereas functions lower viral reservoir size (Fig. 3c, d), although differences related to methanogenesis and carbohydrate biosynthe- were statistically significant only for CA HIV-1 RNA at sis were overrepresented in non-controllers (Additional RMD and MVA2 (p = 0.03) (Fig. 3d). An additional set file 1: Supplementary results and Figs. S9–S12). Taken of clinical and vaccine-response variables was screened Borgognone et al. Microbiome (2022) 10:59 Page 5 of 15 Fig. 1 Higher longitudinal Bacteroidales/Clostridiales ratio in viremic controllers. Relative abundance expressed as percentage of a Bacteroidales, b Clostridiales, and c their ratio in controllers (gray) and non-controllers (white) are represented by boxplots (left and right vertical panels) and line plots (middle vertical panels). In line plots, values for each subject are illustrated by white squares (non-controllers) and grey dots (viremic controllers). Boxplots show the median (horizontal black line) and interquartile range between the first and third quartiles (25th and 75th, respectively). Third vertical panels show non-controllers before ART interruption (pre-MAP, n = 7), last time point on MAP before ART resumption (last detect pVL, n = 4) and 24 weeks after ART resumption (post-MAP w4, n = 7). Abbreviations: MAP, monitored antiretroviral pause; pre-Vax, baseline (1 day before first MVA vaccination); MVA1, 1 week after first MVA vaccination; RMD, 1 week after third romidepsin infusion; MVA2, 4 weeks after second MVA vaccination. Unadjusted p values are shown. Benjamini–Hochberg multiple hypothesis correction for p values ≤ 0.05 are provided in Additional file 2: Dataset S7. for association with gut microbial signatures. Absolute richness (r ho = − 0.62, padj = 0.02), whereas a strong ­CD4+ T cell count before ART initiation was the only and inverse correlation was found between the Shannon factor significantly associated with the Bacteroidales/ index and CD4/CD8 ratio at BCN02 study entry (rho = Clostridiales ratio (rho = 0.65, padj = 0.01) and gene 0.9, padj = 2.83e−05) (Additional file 1: Fig. S13). Borgognone et al. Microbiome (2022) 10:59 Page 6 of 15 Distinct bacterial protein signatures associated Additional file 1: Fig. S14b-c), such as myeloperoxidase with viremia control (MPO), defensin alpha 1 and 4 (DEFA1, DEFA4), and Using baseline proteomic data (15,214 bacterial proteins neutrophil elastase (ELANE) (Additional file 3: Table S2) in total), 24 orders and 69 genera were quantified across were known to be implicated in immune response sign- samples (Additional file 2: Dataset S2). The abundance aling and regulation of inflammatory processes [34, 35]. of total Clostridiales and Bacteroidales orders derived Gene Ontology (GO) analysis confirmed that genes from proteome data was not different between groups upregulated in controllers were enriched in functions (Fig. 4a, b). However, several Clostridiales genera were related to immune system activation, such as neutrophil- decreased in controllers, i.e., Eubacterium (− 3.71%; mediate immunity, leukocyte degranulation and antimi- p = 0.03), Pseudoflavonifractor (− 0.49%; p = 0.049), crobial humoral response (Fig. 5c and Additional file 3: Oscillibacter (− 0.14%; p = 0.07), whereas Blautia was Table S3). Moreover, a group of inflammation-related increased (+ 5.02%; p = 0.03) (Fig. 4a). Unbiased hier- plasma proteins was significantly increased in controllers archical clustering based on Clostridiales genera showed at baseline (Additional file 1: Supplementary results and protein differences (p < 0.025) between groups (Fig. 4c). Fig. S15). Viremic controllers were enriched in bacterial proteins from Blautia and Ruminococcus, and depleted in proteins Integration analysis between Bacteroidales/Clostridiales derived from other Clostridiales such as Clostridium, ratio, host immune activation transcripts, bacterial Eubacterium, Coprococcus, Faecalibacterium, Oscil- proteins, and HIV‑1 reservoir size libacter, and Pseudoflavinofactor. Pathways associated The baseline Bacteroidales/Clostridiales ratio positively with Blautia included galactose, starch/sucrose, and gly- correlated (padj < 0.05) with differentially expressed genes oxylate/dicarboxylate metabolism as well as ribosome involved in inflammatory response and immune system activity (Fig. 4d). Butyrate and other short-chain fatty activation, including DEFA1, DEFA4, TOP1MT, CTSG, acid metabolism pathways were similar in both groups MPO, AZU1, ELANE (Fig. 6a, Spearman rho and adjusted (Fig. 4c, d). p values are given in Additional file 2: Dataset S5). Addi- tional correlation and enrichment analysis extended to Increased baseline immune activation and inflammatory the full set of host transcripts supported such observa- response transcripts in viremic controllers tion (Fig. 6b), also showing significant associations with Full-PBMC gene expression analysis detected a total the baseline viral reservoir (Additional file 1: Supple- of 27,426 transcripts at baseline, after filtering for low- mentary results and Fig. S16). In the integrated analysis expressed genes (Additional file 2: Dataset S3). Using of metagenomic, transcriptomic, and metaproteomic DESeq2 [30], a total of 31 differentially expressed genes data for the identification of discriminating signatures (DEGs) were identified ­(log2 FoldChange = 0 and padj between controllers and non-controllers, Bacteroidales < 0.1), of which 15 and 16 were upregulated in control- and Clostridiales were clearly separated through the com- lers and non-controllers, respectively (Fig. 5a and Addi- ponents (Additional file 1: Fig. S17a). While Bacteroidales tional file 3: Table S1). Hierarchical clustering based on clustered and positively associated with immune activa- transcriptional DEG profiles showed that controllers tion transcripts (MPO, AZU1, ELANE, TCN1, DEFA1, grouped together, while non-controllers separated into BPI, DEF4) as well as proteins from Ruminococcus, Blau- two distinct expression groups (Additional file 1: Fig. tia, and Prevotella, the order Clostridiales inversely cor- S14a). Upregulated genes in non-controllers included related with such features (Additional file 1: Fig. S17a-b). 11 transcripts with unknown function (Additional Multi-omics correlations at lower taxonomic scale, file 3: Table S2), which were excluded from downstream including viral reservoir data confirmed that Bacteroidales analyses. Upregulated genes in controllers (Fig. 5b and species (B. dorei and B. eggerthii) inversely correlated with (See figure on next page.) Fig. 2 Lower microbial diversity and richness in controllers. Longitudinal a microbial gene richness at 10 million (10 M filtered reads) down-sampling size and b alpha diversity based on Shannon index in viremic controllers (gray) and non-controllers (white). c Principal coordinates analysis (PCoA) of microbial diversity based on Bray-Curtis distances at pre-vaccination and during ‘kick and kill’ intervention. Proportion of variance explained by each principal coordinate axis is reported in the corresponding axis label. Subjects per each group are represented by squares (controllers) and circles (non-controllers). Each point stands for one subject, color coded by group and time point. The increase in purple (controllers) and blue (non-controllers) colors reflects sequential time points from baseline (pre-Vax) to the second vaccine administration (MVA2). Ellipses delineate the distribution of points per each group. Gray arrows link directional changes in bacterial abundance throughout the kick and kill intervention from baseline (pre-Vax). PERMANOVA statistical analysis of samples grouped by group, PatientID (patient internal identifier), and time point is shown on the top of the panel. Abbreviations: MAP, monitored antiretroviral pause; pre-Vax, baseline (1 day before first MVA vaccination); MVA1, 1 week after first MVA vaccination; RMD, 1 week after third romidepsin infusion; MVA2, 4 weeks after second MVA vaccination. Unadjusted p values are shown. Benjamini–Hochberg multiple hypothesis correction for p values ≤ 0.05 are provided in Additional file 2: Dataset S7 Borgognone et al. Microbiome (2022) 10:59 Page 7 of 15 Fig. 2 (See legend on previous page.) HIV-1 DNA levels, whereas members of Clostridiales (S. and host defense (Fig. 6c and Additional file 2: Dataset unclassified, D. formicigenerans, and E. siraeum) positively S6). These data together showed positive associations correlated with both HIV-1 DNA and CA HIV-1 RNA between Bacteroidales taxa and host transcripts related (Fig. 6c). In turn, viral reservoir size negatively correlated to immune system activation, and in turn negative corre- with genes involved in neutrophil-mediated immunity lation with the HIV-1 viral reservoir. Whereas, members Borgognone et al. Microbiome (2022) 10:59 Page 8 of 15 Fig. 3 Associations between HIV-1 reservoir size and gut microbial signatures. Spearman’s correlations between gut microbial signatures (ratio Bacteroidales/Clostridiales, gene richness, and alpha-diversity Shannon index) and longitudinal a HIV-1 DNA (HIV-1 DNA copies/106 ­CD4+ T cells) and b cell-associated (CA) HIV-1 RNA (HIV-1/TBP relative expression). Positive correlations are indicated in blue and negative correlations, in red. Color and size of the circles indicate the magnitude of the correlation. White asterisks indicate significant correlations (*p < 0.05; **p < 0.01; ***p < 0.001, Benjamini–Hochberg adjustment for multiple comparisons). Boxplots showing longitudinal comparison of c HIV-1 DNA and d CA HIV-1 RNA between controllers and non-controllers. Abbreviations: MAP, monitored antiretroviral pause; pre-Vax, baseline (1 day before first MVA vaccination); MVA1, 1 week after first MVA vaccination; RMD, 1 week after third romidepsin infusion; MVA2, 4 weeks after second MVA vaccination within the order Clostridiales showed the opposite trend. pause. Individuals with high Bacteroidales/Clostridiales Such observations emerged by analyzing either the ratio ratio showed gene expression signatures related to immune or individual taxa, further supporting potential pre-exist- activation, particularly neutrophil-mediated immunity and ing interactions between intestinal Bacteroidales species, antimicrobial humoral response, which negatively corre- host immune activation, and reservoir size in viremic lated with the viral reservoir size. Our findings largely arise controllers. from unsupervised analyses where many other signatures could have emerged, especially given the relatively low num- Discussion ber of individuals analyzed. However, they are internally In this proof-of-concept study, a longitudinal multi-‘omics’ coherent and consistent with a theoretical framework where analysis identified the Bacteroidales/Clostridiales ratio as increased inflammation might contribute to immune-medi- a novel gut microbiome signature associated with HIV-1 ated HIV-1 control. They also suggest a putative biomarker reservoir size and viremic control during a monitored ART for safer ART interruptions in HIV cure studies. Borgognone et al. Microbiome (2022) 10:59 Page 9 of 15 Fig. 4 Baseline metaproteomic signatures associated with HIV control after ART interruption. Comparison of a Clostridiales and b Bacteroidales abundances derived from gut metaproteome data in controllers and non-controllers at study entry. No differences in Clostridiales or Bacteroidales proteome at the order-level were observed. c Baseline levels of Clostridiales proteins distinguished controllers from non-controllers in a hierarchical clustering. d Functional annotation of Clostridiales bacterial proteins using KEGG gene ontology identified baseline differences in cellular metabolism pathways between groups. Abbreviations: C = controllers, NC = non-controllers, SC = spectral count. Unadjusted p values are shown. Benjamini–Hochberg multiple hypothesis correction for p values ≤ 0.05 are provided in Additional file 2: Dataset S7 The gut microbiome of controllers was enriched in pro- archaea. Microbial functional enrichment in lipid and inflammatory species, such as P. copri [36], and depleted fatty acid biosynthesis in controllers might be reflective in bacteria, traditionally associated with the maintenance of mechanisms of lipopolysaccharide biosynthesis and of gut homeostasis through production of SCFAs [37], production of inflammatory mediators [40] mediated by including R. intestinalis and Subdoligranulum spp. Lower members of Bacteroidales [41]. No discernible longitudi- microbial diversity and gene richness found in controllers nal variations were observed in the gut microbiome of were consistent with a previous work from our group in BCN02 participants, in line with previous evidence in people living with HIV [38], as well as other studies [39], oral typhoid immunization [16]. Of note, the gut micro- in which higher gene richness associated with increased biome in healthy population has been described generally levels of butyrate-producing bacteria and methanogenic resilient to perturbations [42]. Taken together, these Borgognone et al. Microbiome (2022) 10:59 Page 10 of 15 Fig. 5 Baseline functional enrichment in levels of immune activation and inflammatory response in viremic controllers. a Volcano plot of differentially expressed genes between controllers and non-controllers at baseline (pre-Vax) with adjusted p value < 0.1 (violet dots), adjusted p value < 0.05 (red dots) and log2 (foldchange) > 2 and adjusted p value < 0.05 (green dots). Gray-colored dots represent genes not displaying statistical significance (BH-adjusted p value > 0.1). The log2 fold change on the x-axis indicates the magnitude of change, while the −log10 (p-adjust) on the y-axis indicates the significance of each gene. b Violin plots showing relative expression levels (rlog, regularized log transformation) of differentially expressed genes with functional annotation. c Gene ontology (GO) enrichment analysis of upregulated genes in controllers. In the y-axis, only representative enriched GO terms (biological process) are reported (terms obtained after redundancy reduction by REVIGO). X-axis reports the percentage of genes in a given GO terms, expressed as ‘Gene ratio’. The color key from blue to red indicates low to high Bonferroni-adjusted log 10 p value. Dot sizes are based on the ’count’ (genes) associated to each GO term. Significantly enriched GO terms, number of genes associated with each GO term and adjusted p values are provided in Additional file 3: Table S3 observations would suggest a trend toward the mainte- transcriptome and inflammation-related plasma proteins nance of a relative stability in the gut microbiome com- profiling. It also emerged that the ratio Bacteroidales/ position upon vaccination. Baseline metaproteome Clostridiales inversely correlated with the viral reservoir analysis confirmed that functional differences between size in terms of HIV-1 DNA and CA HIV-1 RNA. controllers and non-controllers were mainly driven by Although controllers did not display significantly lower Clostridiales, which were actively producing microbial viral reservoir size compared to non-controllers, such proteins in both groups albeit in distinct functional con- associations were consistent with previous studies sug- texts. Further discriminating baseline signatures linked to gesting a role of low viral reservoir on ART interruption increased immune system activation and inflammatory outcomes [9]. The results of our independent analyses response in controllers emerged from PBMC (i.e., different omics approaches), as well as their Borgognone et al. Microbiome (2022) 10:59 Page 11 of 15 Fig. 6 Integrative analysis of baseline gut microbial signatures, immune activation-related transcripts, bacterial proteins and HIV-1 reservoir. a Spearman’s correlations between the ratio Bacteroidales/Clostridiales and DEGs (annotated transcripts). Color and size of the circles indicate the magnitude of the correlation. White asterisks indicate significant correlations (*p < 0.05; **p < 0.01; ***p < 0.001, Benjamini–Hochberg adjustment for multiple comparisons). b Network visualizing significant Spearman’s correlations between the ratio Bacteroidales/Clostridiales and transcripts involved in the enrichment analysis described in Additional file 3: Table S4. Transcripts are represented as vertices and border width is proportional to transcript expression (log2 |cpmTMM_w0 +1|) in controllers. Edge width indicates the magnitude of correlation. c Network showing Spearman’s correlation between viral reservoir (CA HIV-1 RNA and HIV-1 DNA), bacterial species within Bacteroidales and Clostrdiales, host transcripts correlated with the ratio Bacteroidales/Clostrdiales and bacterial proteins (p ≤ 0.025). Features are showed as vertices and colored by ‘omic’ dataset. Edge width indicates the magnitude of correlation coefficient. Protein-associated bacterial genera are reported in parentheses. In all panels, positive and negative correlations are indicated in blue and red, respectively. Abbreviations: DEGs, differentially expressed genes between controllers and non-controllers; R, Ruminococcus; Ps, Pseudoflavonifactor and pre-Vax, baseline timepoint (1 day before first MVA vaccination) integration suggest that baseline immune activation with the viral reservoir and, in turn, exert immunomodu- potentially associated with a microbial shift toward pro- latory effects on HIV-1 control during ART interruption inflammatory bacteria and lower viral reservoir may con- remains to be delineated. We speculate that a pre-exist- tribute to sustained post-ART interruption HIV-1 ing, altered balance of ‘beneficial’ gut microbial groups, control. While there is evidence suggesting a strong such as Clostridiales, and concomitant overabundance of impact of the gut microbiota composition on host pro-inflammatory bacteria would boost host immune immune system and inflammatory status [43], the mech- system activation, thus triggering a prompt control of anistic basis of how microbial communities may interact rebounding virus, as observed in controllers. In support Borgognone et al. Microbiome (2022) 10:59 Page 12 of 15 to this hypothesis, increased abundance of members limitations of non-parametric methods coupled with from Clostridiales were previously associated with neu- multiple testing correction [54], we used a reasonable trophilia and lower poliovirus and tetanus-hepatitis B threshold of 10% false-discovery rate (adjusted p value < vaccine response [18]. Moreover, baseline transcriptional 0.1) to prevent false negatives. Another fundamental pro-inflammatory and immune activation signatures issue to consider is that this sub-study was largely were suggested as potential predictors of increased influ- restricted to Caucasian/MSM population. Large-scale enza [44], systemic lupus erythematosus [45] and hepati- studies have revealed broad differences in the gut micro- tis B [46] vaccine-induced immune response, with weaker biome composition [55] and modes of HIV transmission responses in elderly [44, 46]. It is then reasonable to pos- [56] across distinct geographical populations, thus lead- tulate that immune activation prior vaccination together ing to dramatic variations in the characteristics of study with microbiome-associated factors may affect vaccine participants at study entry [57]. MSM sexual behavior outcomes. Results obtained in this exploratory study are itself was associated with Prevotella-rich/Bacteroides- indented to generate hypotheses and potentially contrib- poor microbiota profile and increased microbial diversity ute to establish a coherent framework for subsequent [58], suggesting that failure to control for this factor confirmatory studies. In this context, several studies have could confound data interpretation. Yet, most of the suggested plausible mechanisms by which the microbiota HIV-associated microbiome studies published to date, could modulate immune responses to vaccination in including this, have largely focused on single populations humans [47]). Despite causal links remain to be fully or individuals [57], posing the question on the generaliz- deciphered, the ability of the microbiota to directly inter- ability of the results in different population settings. act with immune cells in the gut as well as to regulate the Ensuring diversity among study participants (i.e., in eth- systemic availability of critical metabolites in distal loca- nic group, sexual behavior, or transmission mode) in tions is well-established [47]. A few studies also future microbiome studies, is then critical to both iden- attempted to investigate the exact role played by the gut tify population-specific features and to achieve broader- microbiota in HIV control. Despite gut dysbiosis persis- reaching biological conclusions. Bearing these limitations tence in HIV-1 infected patients after ART initiation in mind, our results should be interpreted with caution, (reduced gut microbial richness and depletion in anti- emphasizing the need of independent validation in rand- inflammatory bacteria [48]), microbial richness may be omized and placebo-controlled trials to assess potentially restored by prolonged ART treatment [49]. Also, elite unmeasured confounders and provide further perspec- controllers, who spontaneously control HIV replication tives on factors that might induce gut microbial shifts. without ART, showed higher gut microbial richness with Upcoming analyses in larger longitudinal trials, including a metabolic profile resembling that of HIV-uninfected the recently reported AELIX-002 trial [59], where fecal adults [50] and increased Th17 cells in the gut mucosa samples have been stored longitudinally, are expected to compared to ART-treated patients [51]. In line with our validate our results. These preliminary findings might results, a recent study showed enriched levels of Bacte- have important implications in the design of HIV-1 cure roidetes genera, such as Bacteroides and Prevotella, in intervention trials that include ART interruption. As pro- patients with sustained HIV control during ART inter- posed for other therapeutic areas [60], microbiome-asso- ruption, after receiving dendritic cells-based HIV-1 ciated predictive patterns could help to optimize patient immunization [52]. Moreover, additional evidence on stratification, thus resulting in more targeted studies and HIV virologic controllers receiving a TLR7 agonist before higher efficacy of HIV-1 interventions. In addition, if a ART interruption with baseline enrichment in P. copri given resident microbial community is to be defined that and negative association between abundance of Rumino- is indeed predictive of viral control during ART interrup- coccus gnavus and time to viral rebound suggested a tion, then modulating participants’ gut microbiota before potential impact of certain microbiome species on HIV immunization might potentially impact vaccine respon- persistence [53]. Together, such observations illustrate siveness and ultimately, clinical outcomes. While host- the exceptional challenge in deconvoluting the complex genetics and other vaccine-associated factors as baseline dynamics between specific microbial species or byprod- predictors are less amendable, the gut microbiome is ucts and HIV-1 control. Our study has several limita- potentially modifiable and even transferrable to another tions. Due to eligibility criteria in the parental BCN02 host. Strategies manipulating the gut microbiota compo- study, there was a limited sample size, and we were una- sition and relative by-products via prebiotics and/or pro- ble to include a control arm without the intervention. biotics administration [61] or microbiota engraftment Our considerations were therefore narrowed to three following fecal microbiota transplantation [62] are under individuals showing viremic control during ART inter- intense evaluation [63], albeit with several limitations. ruption. In addition, given the small sample size and Borgognone et al. Microbiome (2022) 10:59 Page 13 of 15 Conclusions biological process terms of host transcripts significantly correlated with In this exploratory study, we identified pre-existing gut the ratio Bacteroidales/Clostridiales. Table S5. Enriched GO biological microbial and immune activation signatures as potential process terms of transcripts (n=453) significantly correlated with the ratio Bacteroidales/Clostridiales and viral reservoir (CA HIV-1 RNA and HIV-1 predictors of sustained HIV-1 control in the absence of ART, DNA). Table S6. Summary of shotgun metagenomics sequencing yield providing a potential target for future treatment strategies from longitudinally-collected fecal samples. Statistics of reads mapped to and opening up new avenues for a functional HIV cure. integrated gene catalog (IGC) are also shown. Acknowledgements Abbreviations The authors thank all volunteers for participating in this study and the BCN02 ART​: Antiretroviral therapy; ATI: Antiretroviral therapy interruption; BH- study group. adjusted: Benjamini-Hochberg-adjusted; HUMAnN2: HMP unified metabolic BCN02 Study Group. IrsiCaixa AIDS Research Institute-HIVACAT Hospital analysis network v.2; MAP: Monitored antiretroviral therapy pause; MVA: MVA. Universitari Germans Trias i Pujol, Badalona, Spain: Susana Benet, Christian HIVconsv; RMD: Romidepsin; SCFA: Short chain fatty acids. Brander, Samandhy Cedeño, Bonaventura Clotet, Pep Coll, Anuska Llano, Javier Martinez-Picado, Marta Marszalek, Sara Morón-López, Beatriz Mothe, Roger Supplementary Information Paredes, Maria C. Puertas, Miriam Rosás-Umbert, Marta Ruiz-Riol. Fundació Lluita contra la Sida, Infectious Diseases Department, Hospital Universitari The online version contains supplementary material available at https://​doi.​ Germans Trias i Pujol, Badalona, Spain: Roser Escrig, Silvia Gel, Miriam López, org/​10.​1186/​s40168-​022-​01247-6. Cristina Miranda, José Moltó, Jose Muñoz, Nuria Perez-Alvarez, Jordi Puig, Boris Revollo, Jessica Toro. Germans Trias i Pujol Research Institute, Badalona, Spain: Ana María Barriocanal, Cristina Perez-Reche. Clinical Pharmacology Unit, Hos‑ Additional file 1: Supplementary methods. Supplementary results. Fig‑ pital Universitari Germans Trias i Pujol, Badalona, Spain: Magí Farré. Pharma‑ ure S1. BCN02-microbiome study design and sample collection strategy. cokinetic/pharmacodynamic modeling and simultation, Institut de Recerca de Figure S2. Overview of sample disposition for multi-omic analysis. Figure l’Hospital de la Santa Creu i Sant Pau-IIB Sant Pau, Barcelona, Spain: Marta Valle. S3. Taxonomic classification of fecal samples at the order level. Figure Hospital Clinic- HIVACAT, IDIBAPS, University of Barcelona, Barcelona, Spain: S4. Differential abundance in Methanobacteriales between controllers Christian Manzardo, Juan Ambrosioni, Irene Ruiz, Cristina Rovira, Carmen and non-controllers. Figure S5. Linear Discriminant Analysis (LDA) effect Hurtado, Carmen Ligero, Emma Fernández, Sonsoles Sánchez-Palomino, and size (LEfSe) at the species level. Figure S6. Longitudinal feature-volatility Jose M. Miró. Projecte dels NOMS-Hispanosida, BCN Checkpoint, Barcelona, analysis of Bacteroidales and Clostridiales species. Figure S7. Bray-Curtis Spain: Antonio Carrillo, Michael Meulbroek, Ferran Pujol and Jorge Saz. The dissimilarity index between controllers and non-controllers. Figure S8. Jenner Institute, The Nuffield Department of Medicine, University of Oxford, Gut microbiome profiling excluding B07 participant from non-controllers UK: Nicola Borthwick, Alison Crook, Edmund G. Wee and Tomáš Hanke. arm. Figure S9. Differentially abundant metabolic pathways from Bacteroidales and Clostridiales at the study entry. Figure S10. Differential Authors’ contributions metabolic pathways between controllers and non-controllers. Figure S11. R.P, B.M, J.M, B.C, J.M.M, and C.B conceived and designed the study. B.M, R.P, Longitudinal variation of differentially abundant pathways over the trial. C.M, J.M.M, and C.B recruited the study participants and performed their Figure S12. Differential archaeal metabolic pathways between controllers clinical evaluations. M.P and M.C performed fecal DNA extraction, library and non-controllers. Figure S13. Spearman’s correlation between clinical preparation and sequencing, under the supervision of M.N.J, Y.G, and R.P. B.O data, vaccine response and gut microbial variables. Figure S14. Differen‑ and C.D performed PBMC transcriptomics and soluble factors determinations, tially expressed PBMC host genes between controllers and non-controllers under the supervision of M.R.R and C.B. M.C.P performed the viral reservoir at baseline. Figure S15. Protein inflammation markers from controllers size determinations, under the supervision of J.M.P. L.N.R, M.D.L, S.K, and K.B and non-controllers at baseline. Figure S16. Functional enrichment of performed fecal metaproteomics experiments and data analysis, under the transcripts correlated with the ratio Bacteroidales:Clostridiales and viral supervision of A.Bu. A.Bo performed bioinformatics and statistical analyses reservoir. Figure S17. Integrated analysis of microbiome, metaproteome of metagenome, transcriptome, soluble factors, clinical and integration data, and transcriptome data. under the supervision of M.N.J and R.P. L.N.R performed the bioinformatics and Additional file 2: Dataset S1. Longitudinal microbial composition statistical analyses of fecal metaproteome data, under the supervision of A.Bu. (relative abundance) in viremic controllers and non-controllers analyzed F.C.M, M.N.J, A.Bo, B.O, and Y.G contributed to data management. A.Bo and R.P by MetaPhlAn2.0. Dataset S2. Raw bacterial protein counts at baseline wrote the paper, which was reviewed, edited, and approved by all authors. in viremic controllers and non-controllers analysed using Scaffold Q+ software v4.9.0. Dataset S3. Raw gene counts at baseline in viremic Funding controllers and non-controllers estimated using RSEM v1.3.0. Dataset This study was funded by Instituto de Salud Carlos III through the project S4. Soluble factors (normalized protein expression units) from the “PI16/01421” (co-funded by European Regional Development Fund “A way Olink inflammation panel at baseline in viremic controllers and non- to make Europe”). The project was sponsored in part by Grifols and received controllers. Dataset S5. List of transcripts significantly correlated with funding from the European Union’s Horizon 2020 Research and Innovation the Bacteroidales:Clostridiales ratio adjusted for multiple comparisons Programme under Grant Agreement N° 847943 (MISTRAL). The BCN02 clinical (spearman’s rho and corresponding q-value<0.05 are reported). Dataset trial was an investigator-initiated study funded by the ISCIII PI15/01188 grant, S6. Spearman’s correlation (rho=0.5) between viral reservoir (CA HIV-1 the HIVACAT Catalan research program for an HIV vaccine and the Fundació RNA and HIV-1 DNA), bacterial species (Bacteroidales and Clostrdiales), Gloria Soler. Some sub-analyses of the BCN02 trial were partly funded by the transcripts correlated with the ratio Bacteroidales / Clostridiales and dif‑ European Union’s Horizon 2020 research and innovation program under grant ferentially abundant microbial proteins (p<0.025). Dataset S7. Benjamini– agreement 681137-EAVI2020 and by NIH grant P01-AI131568. A.Bu lab was Hochberg multiple hypothesis correction for unadjusted p-values ≤ 0.05 supported by grants from the Canadian Institutes of Health Research (HB3- (metagenomics and metaproteomics datasets). 164066) and the National Institutes for Health Research (R01DK112254). J.M.P lab was supported by grant PID2019-109870RB-I00 from the Spanish Ministry Additional file 3: Table S1. List of differentially expressed genes between of Science and Innovation and in part by Grifols. J.M.M. received a personal controllers and non-controllers at baseline (adjusted p-value <0.1and 80:20 research grant from Institut d’Investigacions Biomèdiques August Pi i log2 FoldChange = 0). Table S2. Detailed information of differentially Sunyer (IDIBAPS), Barcelona, Spain, during 2017–21. expressed genes (DEGs) between controllers and non-controllers at baseline (adjusted p-value <0.1 and log2 FoldChange = 0). Table S3. GO Availability of data and materials terms from enrichment analysis (biological process) based on upregulated Datasets supporting the conclusions of this study are available as additional genes in viremic controllers (n. transcripts =15). Table S4. Enriched GO information (Additional file 2). Metagenome and RNA-seq data have been Borgognone et al. Microbiome (2022) 10:59 Page 14 of 15 deposited in the European Nucleotide Archive (ENA) and are accessible 10. Zimmermann P, Curtis N. The influence of the intestinal microbiome on through ENA accession numbers PRJEB42384 and PRJEB43195. The code and vaccine responses. Vaccine. 2018;36(30):4433–9. databases used for data analysis are available at 10.5281/zenodo.4876340 11. Ciabattini A, Olivieri R, Lazzeri E, Medaglini D. Role of the microbiota in the modulation of vaccine immune responses. Front Microbiol. 2019;10:1305. Declarations 12. Sui Y, Lewis GK, Wang Y, Berckmueller K, Frey B, Dzutsev A, et al. Mucosal vaccine efficacy against intrarectal SHIV is independent of Ethics approval and consent to participate anti-Env antibody response. J Clin Invest. 2019;129:1314–28. This study was approved by the institutional ethical review board of the par‑ 13. Pantaleo G, Janes H, Karuna S, Grant S, Ouedraogo GL, Allen M, et al. ticipating institutions (Reference Nr AC-15-108-R) and by the Spanish Regula‑ Safety and immunogenicity of a multivalent HIV vaccine compris‑ tory Authorities (EudraCT 2015-002300-84). All participants provided written ing envelope protein with either DNA or NYVAC vectors (HVTN informed consent in accordance to the principles expressed in the Declaration 096): a phase 1b, double-blind, placebo-controlled trial. Lancet HIV. of Helsinki and local personal data protection law (LOPD 15/1999). 2019;6:e737–49. 14. Cram JA, Fiore-Gartland AJ, Srinivasan S, Karuna S, Pantaleo G, Tomaras Consent for publication GD, et al. Human gut microbiota is associated with HIV-reactive Where individual patient data is included, all patients provided informed consent. immunoglobulin at baseline and following HIV vaccination. PLoS One. 2019;14(12):e0225622. Competing interests 15. Williams WB, Han Q, Haynes BF. Cross-reactivity of HIV vaccine The authors declare that they have no competing interests. responses and the microbiome. Curr Opin HIV AIDS. 2018;13(1):9–14. 16. Eloe-Fadrosh EA, McArthur MA, Seekatz AM, Drabek EF, Rasko DA, Author details Sztein MB, et al. Impact of oral typhoid vaccination on the human 1 IrsiCaixa AIDS Research Institute, Hospital Universitari Germans Trias i Pujol, gut microbiota and correlations with S. Typhi-specific immunological Barcelona, Catalonia, Spain. 2 CIBERINFEC, Madrid, Spain. 3 University of Vic– responses. PLoS One. 2013;8(4):e62026. Central University of Catalonia (UVic–UCC), Vic, Catalonia, Spain. 4 Universitat 17. Harris VC, Armah G, Fuentes S, Korpela KE, Parashar U, Victor JC, et al. Autonoma de Barcelona (UAB), Barcelona, Catalonia, Spain. 5 Center for Global Significant correlation between the infant gut microbiome and rotavi‑ Health and Diseases, Department of Pathology, Case Western Reserve Univer‑ rus vaccine response in rural Ghana. J Infect Dis. 2017;215:34–41. sity, Cleveland, OH, USA. 6 Department of Obstetrics & Gynecology, University 18. Huda MN, Lewis Z, Kalanetra KM, Rashid M, Ahmad SM, Raqib R, of Manitoba, Winnipeg, Manitoba, Canada. 7 Institut Mar d’Investigacions et al. Stool microbiota and vaccine responses of infants. Pediatrics. mediques (IMIM), CIBERONC, Barcelona, Catalonia, Spain. 8 Infectious Diseases 2014;134(2):e362–72. Service, Hospital Clinic‑Institut d’Investigacions Biomèdiques August Pi i 19. Julg B, Dee L, Ananworanich J, Barouch DH, Bar K, Caskey M, et al. Sunyer (IDIBAPS), University of Barcelona, Barcelona, Catalonia, Spain. 9 Fight Recommendations for analytical antiretroviral treatment interruptions AIDS Foundation, Infectious Diseases Department, Germans Trias i Pujol in HIV research trials—report of a consensus meeting. Lancet HIV. University Hospital, Barcelona, Catalonia, Spain. 10 Department of Infec‑ 2019;6(4):e259–68. tious Diseases Service, Germans Trias i Pujol University Hospital, Barcelona, 20. Mothe B, Rosás-Umbert M, Coll P, Manzardo C, Puertas MC, Morón- Catalonia, Spain. 11 Catalan Institution for Research and Advanced Studies López S, et al. HIVconsv vaccines and romidepsin in early-treated HIV- (ICREA), Barcelona, Catalonia, Spain. 12 Department of Medicine Solna, Center 1-infected individuals: safety, immunogenicity and effect on the viral for Molecular Medicine, Karolinska Institute, Karolinska University Hospital, reservoir (study BCN02). Front Immunol. 2020;11:823. Stockholm, Sweden. 21. Wei DG, Chiang V, Fyne E, Balakrishnan M, Barnes T, Graupe M, et al. His‑ tone deacetylase inhibitor romidepsin induces HIV expression in CD4 T Received: 18 June 2021 Accepted: 16 February 2022 cells from patients on suppressive antiretroviral therapy at concentra‑ tions achieved by clinical dosing. PLoS Pathog. 2014;10(4):e1004071. 22. Mothe B, Manzardo C, Sanchez-Bernabeu A, Coll P, Morón-López S, Puertas MC, et al. Therapeutic vaccination refocuses T-cell responses towards conserved regions of HIV-1 in early treated individuals (BCN 01 References study). EClinicalMedicine. 2019;11:65–80. 1. Volberding PA, Deeks SG. Antiretroviral therapy and management of 23. Létourneau S, Im EJ, Mashishi T, Brereton C, Bridgeman A, Yang H, et al. HIV infection. Lancet. 2010;376(9734):49–62. Design and pre-clinical evaluation of a universal HIV-1 vaccine. PLoS 2. Finzi D, Hermankova M, Pierson T, Carruth LM, Buck C, Chaisson RE, One. 2007;2(10):e984. et al. Identification of a reservoir for HIV-1 in patients on highly active 24. Søgaard OS, Graversen ME, Leth S, Olesen R, Brinkmann CR, Nissen SK, antiretroviral therapy. Science (80-). 1997;278:1295–300. et al. The depsipeptide romidepsin reverses HIV-1 latency in vivo. PLoS 3. Siliciano JD, Kajdas J, Finzi D, Quinn TC, Chadwick K, Margolick JB, et al. Pathog. 2015;11(9):e1005142. Long-term follow-up studies confirm the stability of the latent reser‑ 25. Truong DT, Franzosa EA, Tickle TL, Scholz M, Weingart G, Pasolli E, et al. voir for HIV-1 in resting CD4+ T cells. Nat Med. 2003;9:727–8. MetaPhlAn2 for enhanced metagenomic taxonomic profiling. Nat 4. Chun TW, Davey RT, Engel D, Lane HC, Fauci AS. AIDS: Re-emergence of Methods. 2015;12(10):902–3. HIV after stopping therapy. Nature. 1999;401:874–5. 26. Li J, Wang J, Jia H, Cai X, Zhong H, Feng Q, et al. An integrated catalog 5. Namazi G, Fajnzylber JM, Aga E, Bosch RJ, Acosta EP, Sharaf R, et al. The of reference genes in the human gut microbiome. Nat Biotechnol. control of HIV after antiretroviral medication pause (CHAMP) study: 2014;32:834–41. posttreatment controllers identified from 14 clinical studies. J Infect 27. Franzosa EA, McIver LJ, Rahnavard G, Thompson LR, Schirmer M, Wein‑ Dis. 2018;218:1954–63. gart G, et al. Species-level functional profiling of metagenomes and 6. Turnbull EL, Wong M, Wang S, Wei X, Jones NA, Conrod KE, et al. Kinet‑ metatranscriptomes. Nat Methods. 2018;15:962–8. ics of expansion of epitope-specific T cell responses during primary 28. Klatt NR, Cheu R, Birse K, Zevin AS, Perner M, Noël-Romas L, et al. Vaginal HIV-1 infection. J Immunol. 2009;182:7131–45. bacteria modify HIV tenofovir microbicide efficacy in African women. 7. Liu MKP, Hawkins N, Ritchie AJ, Ganusov VV, Whale V, Brackenridge S, Science (80-). 2017;356:938–45. et al. Vertical T cell immunodominance and epitope entropy determine 29. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: HIV-1 escape. J Clin Invest. 2013;123:380–93. ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21 Available 8. Posteraro B, Pastorino R, Di Giannantonio P, Ianuale C, Amore R, Ricciardi from: http://​www.​ncbi.​nlm.​nih.​gov/​pubmed/​23104​886. Cited 2017 Feb 22. W, et al. The link between genetic variation and variability in vaccine 30. Li B, Dewey CN. RSEM: accurate transcript quantification from RNA- responses: systematic review and meta-analyses. Vaccine. 2014;32:1661–9. Seq data with or without a reference genome. BMC Bioinformatics. 9. Li JZ, Etemad B, Ahmed H, Aga E, Bosch RJ, Mellors JW, et al. The size of 2011;12:323. the expressed HIV reservoir predicts timing of viral rebound after treat‑ 31. Love MI, Huber W, Anders S. Moderated estimation of fold change and ment interruption. AIDS. 2016;30:343–53. dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):1–21. Borgognone et al. Microbiome (2022) 10:59 Page 15 of 15 32. Assarsson E, Lundberg M, Holmquist G, Björkesten J, Thorsen SB, Ekman 56. De Cock KM, Fowler MG, Mercier E, De Vincenzi I, Saba J, Hoff E, et al. D, et al. Homogenous 96-plex PEA immunoassay exhibiting high sensitiv‑ Prevention of mother-to-child HIV transmission in resource-poor ity, specificity, and excellent scalability. PLoS One. 2014;9(4):e95192. countries: translating research into policy and practice. J Am Med Assoc. 33. Rohart F, Gautier B, Singh A, Lê Cao KA. mixOmics: an R package for 2000;283:1175–82. ‘omics feature selection and multiple data integration. PLoS Comput Biol. 57. Li SX, Armstrong AJS, Neff CP, Shaffer M, Lozupone CA, Palmer BE. Com‑ 2017;13(11):e1005752. plexities of gut microbiome dysbiosis in the context of HIV infection and 34. Aratani Y. Myeloperoxidase: its role for host defense, inflammation, and antiretroviral therapy. Clin Pharmacol Ther. 2016;99:600–11. neutrophil function. Arch Biochem Biophys. 2018;640:47–52. 58. Noguera-Julian M, Rocafort M, Guillén Y, Rivera J, Casadellà M, Nowak 35. Holm J, Hansen SI. Characterization of soluble folate receptors (folate P, et al. Gut microbiota linked to sexual preference and HIV infection. binding proteins) in humans. Biological roles and clinical potentials in EBioMedicine. 2016;5:135–46. infection and malignancy. Biochim Biophys Acta Proteins Proteomics. 59. Bailon L, Llano A, Cedeño S, Lopez MB, Alarcón-Soto Y, Coll P, et al. A 2020;1868(10):140466. placebo-controlled ATI trial of HTI vaccines in early treated HIV infection. 36. Iljazovic A, Roy U, Gálvez EJC, Lesker TR, Zhao B, Gronow A, et al. Perturba‑ In: CROI - Virtual Conf Retroviruses Opportunistic Infect; 2021. tion of the gut microbiome by Prevotella spp. enhances host susceptibil‑ 60. Boessen R, Heerspink HJL, De Zeeuw D, Grobbee DE, Groenwold RHH, ity to mucosal inflammation. Mucosal Immunol. 2020. Roes KCB. Improving clinical trial efficiency by biomarker-guided patient 37. Lopetuso LR, Scaldaferri F, Petito V, Gasbarrini A. Commensal clostridia: selection. Trials. 2014;15(1):1. leading players in the maintenance of gut homeostasis. Gut Pathog. 61. Wilson NL, Moneyham LD, Alexandrov AW. A systematic review of probi‑ 2013;5(1):23. otics as a potential intervention to restore gut health in HIV infection. J 38. Guillén Y, Noguera-Julian M, Rivera J, Casadellà M, Zevin AS, Rocafort M, Assoc Nurses AIDS Care. 2013;24:98–111. et al. Low nadir CD4+ T-cell counts predict gut dysbiosis in HIV-1 infec‑ 62. Vujkovic-Cvijin I, Rutishauser RL, Pao M, Hunt PW, Lynch SV, McCune tion. Mucosal Immunol. 2019;12:232–46. JM, et al. Limited engraftment of donor microbiome via one-time 39. Le Chatelier E, Nielsen T, Qin J, Prifti E, Hildebrand F, Falony G, et al. fecal microbial transplantation in treated HIV-infected individuals. Gut Richness of human gut microbiome correlates with metabolic markers. Microbes. 2017;8:440–50. Nature. 2013;500:541–6. 63. Rosel-Pech C, Chávez-Torres M, Bekker-Méndez VC, Pinto-Cardoso S. 40. Medzhitov R, Janeway CJ. Advances in immunology: innate immunity. N Therapeutic avenues for restoring the gut microbiome in HIV infection. Engl J Med. 2000;343:338–44. Curr Opin Pharmacol. 2020;54:188–201. 41. d’Hennezel E, Abubucker S, Murphy LO, Cullen TW. Total lipopolysaccha‑ ride from the human gut microbiome silences toll-like receptor signaling. mSystems. 2017;2(6):e00046–17. Publisher’s Note 42. Fassarella M, Blaak EE, Penders J, Nauta A, Smidt H, Zoetendal EG. Gut Springer Nature remains neutral with regard to jurisdictional claims in pub‑ microbiome stability and resilience: elucidating the response to pertur‑ lished maps and institutional affiliations. bations in order to modulate gut health. Gut. 2020. 43. Zheng D, Liwinski T, Elinav E. Interaction between microbiota and immu‑ nity in health and disease. Cell Res. 2020;30(6):492–506. 44. Avey S, Cheung F, Fermin D, Frelinger J, Gaujoux R, Gottardo R, et al. Mul‑ ticohort analysis reveals baseline transcriptional predictors of influenza vaccination responses. Sci Immunol. 2017;2(14):eaal4656. 45. Kotliarov Y, Sparks R, Martins AJ, Mulè MP, Lu Y, Goswami M, et al. Broad immune activation underlies shared set point signatures for vaccine responsiveness in healthy individuals and disease activity in patients with lupus. Nat Med. 2020;26:618–29. 46. Fourati S, Cristescu R, Loboda A, Talla A, Filali A, Railkar R, et al. Pre-vacci‑ nation inflammation and B-cell signalling predict age-related hypore‑ sponse to hepatitis B vaccination. Nat Commun. 2016;7:10369. 47. de Jong SE, Olin A, Pulendran B. The impact of the microbiome on immu‑ nity to vaccination in humans. Cell Host Microbe. 2020;28(2):169–79. 48. Rocafort M, Noguera-Julian M, Rivera J, Pastor L, Guillén Y, Langhorst J, et al. Evolution of the gut microbiome following acute HIV-1 infection. Microbiome. 2019;7(1):1–7. 49. Parbie PK, Mizutani T, Ishizaka A, Kawana-Tachikawa A, Runtuwene LR, Seki S, et al. Dysbiotic fecal microbiome in HIV-1 infected individuals in Ghana. Front Cell Infect Microbiol. 2021;11:646467. 50. Vesterbacka J, Rivera J, Noyan K, Parera M, Neogi U, Calle M, et al. Richer gut microbiota with distinct metabolic profile in HIV infected elite con‑ trollers. Sci Rep. 2017;7(1):6269. 51. Kim CJ, Kovacs C, Chun TW, Kandel G, Osborne BJW, Huibner S, et al. Antiretroviral therapy in HIV-infected elite controllers: impact on gut Ready to submit your research ? Choose BMC and benefit from: immunology, microbial translocation, and biomarkers of serious non- AIDS conditions. J Acquir Immune Defic Syndr. 2014;67(5):514–8. • fast, convenient online submission 52. Pastor-Ibáñez R, Díez-Fuertes F, Sánchez-Palomino S, Alcamí J, Plana M, Torrents D, et al. Impact of transcriptome and gut microbiome on the • thorough peer review by experienced researchers in your field response of hiv-1 infected individuals to a dendritic cell-based hiv thera‑ • rapid publication on acceptance peutic vaccine. Vaccines. 2021;9(7):694. • support for research data, including large and complex data types 53. Cai Y, Deeks SG, Brinson C, Ramgopal M, Jones N, DeJesus E, et al. HIV rebound in controllers is associated with specific fecal microbiome • gold Open Access which fosters wider collaboration and increased citations profile. In: CROI - Virtual Conf Retroviruses Opportunistic Infect; 2021. • maximum visibility for your research: over 100M website views per year 54. Carvajal-Rodríguez A, de Uña-Alvarez J, Rolán-Alvarez E. A new multitest correction (SGoF) that increases its statistical power when increasing the At BMC, research is always in progress. number of tests. BMC Bioinformatics. 2009;10:209. 55. Gaulke CA, Sharpton TJ. The influence of ethnicity and geography on Learn more biomedcentral.com/submissions human gut microbiome composition. Nat Med. 2018;24(10):1495–6.