Akorli et al. BMC Infectious Diseases (2024) 24:1020 https://doi.org/10.1186/s12879-024-09948-z RESEARCH Open Access © The Author(s) 2024. 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://creativecommons.org/licenses/by/4.0/. BMC Infectious Diseases High abundance of butyrate‑producing bacteria in the naso‑oropharynx of SARS‑CoV‑2‑infected persons in an African population: implications for low disease severity Jewelna Akorli1*, Millicent Opoku1,4, Francis Appiah‑Twum1, Margaret Sena Akpo1, Rahmat Yusif Ismail1, Georgina Yaa Kwartemaa Boamah1, Elizabeth Obeng‑Aboagye2, Dina Adu‑Asamoah3 and Irene Owusu Donkor2  Abstract  Background  The association of the oral microbiome with SARS-CoV-2 infections and disease progression has been documented in European, Asian, and American populations but not in Africa. Methods  We conducted a study in Ghana to evaluate and compare the naso-oropharyngeal microbiome in SARS- CoV-2-infected and uninfected persons before (pre-vaccine) and after vaccine availability (post-vaccine) in the coun‑ try. 16S rRNA V3-V4 variable region was sequenced and analysed from DNA extracted from naso-oropharyngeal swabs. Results  Considering only the infection status, infected and uninfected groups had no difference in their within- group diversity and was evident in the study population pre- and post-vaccine availability. The introduction of vac‑ cines reduced the diversity of the naso-oropharyngeal microbiome particularly among SARS-CoV-2 positive per‑ sons and, vaccinated individuals (both infected and uninfected) had higher microbial diversity compared to their unvaccinated counterparts. SARS-CoV-2-positive and -negative individuals were largely compositionally similar varying by 4–7% but considering vaccination*infection statuses, the genetic distance increased to 12% (P = 0.003) and was mainly influenced by vaccination. Common among the pre- and post-vaccine samples, Atopobium and Fine- goldia were abundant in infected and uninfected individuals, respectively. Bacteria belonging to major butyrate- producing phyla, Bacillota (particularly class Clostridia) and Bacteroidota showed increased abundance more strikingly in infected individuals before vaccines were available. They reduced significantly after vaccines were introduced into the country with Fusobacterium and Lachnoanaerobaculum being the only common bacteria between pre- vaccine infected persons and vaccinated individuals, suggesting that natural infection and vaccination correlate with high abundance of short-chain fatty acids. Conclusion  Our results show, in an African cohort, the abundance of bacteria taxa known for their protective pathophysiological processes, especially during infection, suggesting that this population is protected against severe COVID-19. The immune-related roles of the members of Bacillota and Bacteroidota that were found associated *Correspondence: Jewelna Akorli jakorli@noguchi.ug.edu.gh Full list of author information is available at the end of the article http://creativecommons.org/licenses/by/4.0/ http://crossmark.crossref.org/dialog/?doi=10.1186/s12879-024-09948-z&domain=pdf Page 2 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 with infection and vaccination require further studies, and how these may be linked to ethnicity, diet and age. We also recommend expansion of microbiome–disease association studies across Africa to identify possible bacterial- mediated therapeutics for emerging infections. Keywords  Naso-oropharyngeal microbiome, SARS-CoV-2, COVID-19, Bacillota, Bacteroidota, Butyrate producers Background SARS-CoV-2, which causes COVID-19, emerged at the end of 2019 and spread rapidly through populations. By March 2020, when the WHO declared the disease a pan- demic, approximately 118,000 cases and over 4000 deaths had already been recorded across all continents [1]. At the peak of the COVID-19 pandemic, cases and mortal- ity varied widely across the globe. Mortality became the measure by which the burden of COVID-19 was evalu- ated. Together, Europe and the Americas recorded more than 70% of deaths resulting from COVID-19 [2], and it was initially expected that Africa would be the worst hit by the pandemic. This assumption is because among other factors, health care systems are crucial in deter- mining the general recorded outcomes of infected popu- lations, and Africa is notable for low healthcare delivery [3]. However, the African continent recorded relatively low cases and mortality [2]. Several reasons have been attributed to these statistics, including underreport- ing, limited testing capacity, high tropical temperatures, increased use of herbal medicine, malaria coinfection with COVID-19 and the frequent use of anti-malarial drugs that inhibit SARS-CoV-2 replication [4, 5]. Host factors also contribute significantly to disease outcome resulting from SARS-CoV-2 infection [6], and identifying these factors is important in explaining indi- vidual and population variations in disease epidemiology. The interaction between the host and virus is complex and initiated at the site of entry. For SARS-CoV-2, the naso-oropharyngeal cavity is the major site for viral entry and infection initiation [7]. Here, the binding of the viral receptor to human angiotensin-converting enzyme 2 (ACE2) expressed in epithelial cells is crucial for estab- lishing infection [8, 9]. The virus may lodge here for several weeks, replicating in epithelial cells and elicit- ing a cascade of immune responses that characterize the course of the disease. Low levels of ACE2 expression in the nasal cavity would therefore lead to decreased viral acquisition [10]. Another cellular activation known to be associated with SARS-CoV-2 infection is the expression of transmembrane serine protease-2 (TMPRSS2), which acts as a significant determinant of the entry pathway for the virus; overexpression inhibits viral infection [11]. These findings represent a few examples of our under- standing of the intricacy of SARS-CoV-2 infection which could contribute to potential avenues for therapy. Interactions between bacteria and viruses are common in viral infections. Bacteria may impact viral infectivity and stability, and these interactions could influence how the host responds to viral infections (reviewed in [12]). Oral bacteria could complicate respiratory infections through various suggested mechanisms including pro- moted adhesion of pathogens to mucosal surfaces and alteration of infection site tissues through periodontal- originating cytokines [13]. The human naso-oropharynx is an environment with a residing community of bacte- ria, some of which have been reported to be associated with various clinical statuses of COVID-19. Bacterial taxa linked to poor oral hygiene and some opportunistic microbes have been shown to proliferate in COVID-19 patients [14, 15]. Whether this dysbiosis results from the viral infection and facilitates the progress of the disease severity remains unclear. A decrease in gut and oral bac- terial diversity has also been linked to elevated levels of proinflammatory cytokines which characterises SARS- CoV-2 [14] and other viral infections, such as hepatitis C virus (HCV) [16] and human immunodeficiency virus (HIV) [17]. Understanding the mechanism of the asso- ciation of oral bacteria in COVID-19 infection could make useful contributions to our knowledge of COVID- 19 pathogenesis and improve care through advanced therapeutics. In this study, we focused on the naso-oropharyngeal microbiome of SARS-CoV-2-infected and uninfected individuals in an African population with the aim of identifying signature microbes whose known proper- ties may potentially explain the reduced disease sever- ity recorded. We also assessed potential changes in the naso-oropharyngeal microbiome following vaccination, as COVID-19 vaccines have been shown to cause dysbio- sis of the gut microbiome [18]. Our study contributes to addressing recent concerns about the neglect of microbi- ome research in Africa [19, 20], particularly in the face of emerging infections. Methods Clinical samples Two sets of clinical samples were used in this study; those obtained during the peak of the COVID-19 pan- demic, and the other after vaccination was rolled out in the country. We refer to these samples as ‘pre-vaccine’ and ‘post-vaccine’ sample sets, respectively. For the Page 3 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 pre-vaccine set, we randomly selected 49 and 40 previ- ously confirmed SARS-CoV-2 positives and negatives, respectively, from archived naso-oropharyngeal swabs that were tested prior to March 2021 when the AstraZen- eca vaccines first became available in Ghana through the UN-partnered COVAX initiative [21, 22]. SARS-CoV-2 detection was performed with standard RT‒qPCR using the Veri-Q PCR 316 COVID-19 Detection Kit (MiCo Biomed Corporation, South Korea), with Ct values < 40 reported as positive. We set a selection criterion to retrieve only naso-oropharyngeal swabs that were col- lected into and eluted in sterile water. This was to avoid possible loss of microbial richness that could result from the use of viral transport medium (VTM) which may be incorporated with antibiotics. The post-vaccine samples were swabs collected in June – July 2022 as part of a previous study [23]. These swabs were also placed into 2 mL of sterile water for the same reason already explained and were used for RT‒qPCR viral detection as previously described. Based on the results of the RT-qPCR and the vaccination information collected from participants, samples were grouped into unvaccinated uninfected (UU), unvaccinated infected (UI), vaccinated uninfected (VU) and vaccinated infected (VI). Sample processing for microbiome analyses Total DNA was extracted from 500 µL each of 89 pre- vaccine and 232 post-vaccine samples, respectively, using the ZymoBIOMICS DNA Miniprep kit (Zymo Research, USA). Extractions were performed in batches based on sample availability, especially with the post-vaccine sam- ples which was a running study. To check for potential contamination in downstream analyses, mock (no tem- plate) samples were included in each extraction batch. The mock samples totalled 5 and 9 for the pre- and post- vaccine sets, respectively. Eluted DNA was quantified with the Qubit Fluorometer 3.0 (Invitrogen). Based on the DNA quantities obtained and funding availability, 89 pre-vaccine, 68 post-vaccine and all 14 mock sam- ples were prepared for sequencing. Briefly, the bacterial 16S rRNA V3-V4 region was amplified with barcoded primers: 341F: 5’-CCT​AYG​GGRBGCASCAG- 3’ and 806R: 5’- GGA​CTA​CNNGGG​TAT​CTAAT- 3’. Thermal cycling consisted of initial denaturation at 98℃ for 1 min, followed by 30 cycles of denaturation at 98℃ for 10 s, annealing at 50℃ for 30 s, and elongation at 72℃ for 30 s, then a final extension at 72℃ for 5 min. Amplicons with the required size were selected, pooled by equimolar concentrations, end-repaired, A-tailed and ligated with Illumina adapters. Libraries were purified and sequenced. Eighty (out of 89) pre-vaccine and 65 (out of 68) post- vaccine test samples successfully amplified for 16S rRNA and were processed for sequencing. The pre-vaccine samples consisted of 33 SARS-CoV-2-negative and 47 SARS-CoV-2-positive samples. The post-vaccine samples included 19 unvaccinated uninfected (UU), 7 unvacci- nated infected (UI), 19 vaccinated uninfected (VU) and, 20 vaccinated infected (VI). All 14 mock samples were sequenced, whether they amplified or not. The two sets of test samples (including their mocks) were sequenced separately but with the same sequencing criteria on a NovaSeq 6000 platform to generate 250-paired-end reads at a depth of 50 K tags of raw data per sample. Sequence processing Primers and adapters were trimmed off raw reads and resulting sequences < 60 bp long were removed. Reads that contained more than 10% N’s (ambiguous bases) and quality base score ≤ 5 in over 50% of total read length were also filtered out. These resulted in a total of 15,356,104 filtered paired end reads from the pre- vaccine (minimum = 123,189, maximum = 268,461, median = 180,257.0), and 12,707,463 from the post- vaccine set (minimum = 71,449, maximum = 189,728, median = 179,516.0). The sequences obtained from the two sets of samples were merged and processed using customized pipelines and scripts in Quantitative Insights into Microbial Ecology (QIIME2) package version 2023.7 [24] and R-software [25]. The paired sequences were demultiplexed, derepli- cated and filtered of chimeras using dada2 [26] to obtain amplicon sequence variants (ASVs). The ASVs were tax- onomically assigned against the SILVA 138.1 database [27] using a customized classifier based on the V3-V4 primers used. Unassigned reads and those identified as Eukaryota, Archaea, Chloroplasts and Mitochondria were removed from the ASV table and representative sequences. A midpoint rooted tree was obtained follow- ing the align-to-tree-MAFFT-fasttree pipeline under the q2-phylogeny plugin in QIIME2. The resulting rooted tree and ASV table were exported into R-software for detec- tion and removal of potential contaminants based on ASV prevalence in the mock samples using decontam [28]. A total of 895 ASVs were detected as ‘contaminants’ and excluded from the dataset. Further downstream pro- cessing and analyses were performed using R-custom scripts on a phyloseq [29] object built with the 145 test samples only. We set an arbitrary filtering criterion to retain taxa that were observed more than once in at least 3% of samples. This identified and removed 4885 ASVs as singletons (taxa represented by one sequence). Because the two sample sets were sequenced separately, Condi- tional Quantile Regression (ConQuR) was applied on the ASV table to remove batch effect [30]. Rarefaction was performed on this ‘corrected’ data to allow visualization Page 4 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 of the adequacy of the sequencing depth to represent the entire microbial contents of all test samples. Diversity analyses Total taxa richness, Shannon‒Wiener, and Simpson ⍺-diversity indices were estimated on rarefied data. The indices were compared among infection and vaccination status using a pairwise Wilcoxon Rank Sum test with a Benjamin-Holchberg (BH) P-adjusted correction. Beta (β) diversity analyses were performed in the microeco package [31] and were based on Bray‒Curtis dissimilarity [32] and weighted UniFrac distances [33] to allow estima- tions of variation based on count and taxa phylogenetic distances, respectively. The degree of similarity was esti- mated using Analysis of Similarity (ANOSIM), while group dispersion was statistically tested with betadisp. The amount of variation explained by bacterial composi- tion between test groups was tested with Permutational Multivariate ANOVA (PERMANOVA) accepting as sig- nificant adjusted P-values < 0.05. Differential microbial abundance Differential taxa abundance analysis was performed at the genus level with microeco package [31] in R based on LEfSe [34]. The linear discriminant analysis (LDA) score was set at a threshold of 3 for discriminative features instead of the default of 2 to make the discovery more stringent. Each analysis was bootstrapped 1000 times and P-value for test of significance was maintained at the default 0.05. Results Sequence exploration and statistics After accounting for and removing batch effects, 6,618,391 reads remained for the merged dataset (Table  S1). The average number of reads was 45,644.07; minimum and maximum reads was 1514 and 111,200, respectively. The two datasets had positively skewed dis- tributions but with different read frequencies, indicat- ing that few samples had relatively large number of reads (Fig. 1A). Each dataset was rarefied separately with their respective median read (pre-vaccine = 43,932; post-vac- cine = 50,262) providing evidence that the sequencing depth and data processes applied resulted in adequate representations of the total bacterial content per sample and test group without sample size bias (Fig.  1B). The median number of ASVs was higher in the pre-vaccine than in the post-vaccine population (Fig. 1B). Within‑group microbial diversity increased with vaccination Alpha (⍺) diversity did not differ between infection sta- tus (positive vs negative) pre- or post-vaccine availability (Fig S1). When infected individuals from the two sam- ple sets were compared, however, species richness and ⍺-diversity estimated by Shannon–Wiener index were higher among the population before the availability of the vaccine (P.adj < 0.05) (Fig. 2A). Uninfected individuals in both sample sets only differed in species richness, again being significantly higher pre-vaccine (P.adj = 0.0001) (Fig.  2B). Therefore, individual-to-individual microbial diversity was more varied in the study population prior to the introduction of vaccines than after people received COVID-19 vaccination. We further investigated the post-vaccine dataset to identify whether vaccination influenced within-group diversity. Overall, all vaccinated individuals (infected and infected) showed more variable richness than unvac- cinated individuals (P.adj = 0.006), but other ⍺-diver- sity indices were similar (Fig.  2C), implying that the vaccinated group had many rare species. When infec- tion statuses were considered (vaccination * infection), Fig. 1  Sequence distribution and rarefaction for sample sets analysed following removal of batch effects. Kernel density plot (A) show the reads frequency distribution for pre- and post-vaccine sample sets. Both curves show multimodal distribution and positively skewed distribution. Sub-sampling of resulting sequences produced rarefaction curves (B) that depict adequate saturation of the taxa richness from both sample sets Page 5 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 vaccinated individuals showed more varied richness whether they were positive (VI) or negative (VU) for the virus compared to their unvaccinated counterparts (VU vs UU [P.adj = 0.02]; VI vs UI [P.adj = 0.04) (Fig.  2D). In addition, Shannon–Wiener index was higher in vacci- nated infected (VI) compared to unvaccinated infected individuals (UI) (P.adj = 0.024), suggesting a significant increase in taxa abundance and richness following vac- cination. The impact of vaccination is further confirmed with increased within-group diversity among vaccinated uninfected (VU) compared to unvaccinated infected (UI) individuals (Shannon–Wiener and Simpson indices: P. adj < 0.05) (Fig. 2D). The significant Simpson index indi- cates a change in evenness between microbial diversity associated with being infected without vaccination, and receiving the attenuated virus through vaccines. Phylogenetically distinct taxa explain higher diversity between groups Principal coordinate analysis (PCoA) was used to visual- ize the ordination of dissimilarity between sample groups based on Bray‒Curtis (BC), which considers species com- position and weighted UniFrac (wUF) distances for bac- terial phylogenetic relatedness. Both approaches showed that infected and uninfected groups shared largely simi- lar microbes with few taxa driving little yet significant differences between them. Variation based on bacteria abundance was between 4–7% (PERMANOVA: R2 > 0.03; P.adj = 0.001) (Fig.  3A). Diversity between infected and uninfected individuals in the post-vaccine population was lower than observed pre-vaccine with no difference between the infection groups (PERMANOVA: R2 < 0.015; P-adj. > 0.05), unless vaccination statuses were considered (Fig. 3B). Clustering based on phylogenetic distances always explained a higher variation between test groups than compositional dissimilarity based on counts (Fig.  3). However, there were also significant group dispersions (betadisp: Bray–Curtis [P.adj = 0.005]; weighted UniFrac [P.adj = 0.01]) in the pre-vaccine population suggest- ing that besides distinct taxa, β-diversity was also influ- enced by differences in taxa composition within groups Fig. 2  Comparison of taxon richness, Shannon and Simpson alpha (⍺) diversity indices. SARS-CoV-2-infected (A) and uninfected (B) samples are compared separately between pre-vaccine and post-vaccine sample sets. Among the post-vaccine sample set, comparisons are made first between vaccination statuses (without considering infection status) (C) and while considering the virus infection status (D). Colored dots represent individual samples, and box plots show first and third quartiles of the distribution. The solid horizontal line in the box shows the median index value per group. Pairwise comparison was achieved with the Wilcoxon test with Benjamin-Holchberg (BH) P-adjusted (P.adj) correction Page 6 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 in this sample set. Despite this, the microbial differences between SARS-CoV-2-positive and -negative groups were still larger than individual-to-individual differences within each group (⍺-diversity) (ANOSIM: R = 0.17; P = 0.001). Bacillota and bacteroidota constitute majority of high abundant bacteria The relative abundance of bacterial taxa was observed at the genus level and plotted to show those with > 0.01 total relative abundance. Out of 200 genera identified in the merged dataset, 23 (pre-vaccine = 21; post-vaccine = 16) constituted those with relative abundance > 0.01 and were considered ‘high abundant’ bacteria (Fig.  4). Majority (22 out of 23) of these genera belonged to 5 phyla that together made up > 90% of the bacteria present (Fig S2). Nine genera belonged to Bacillota (formerly Firmicutes) and four were Bacteroidota (formerly Bacteroidetes). Twelve of these genera, including Prevotella, Lachnoan- aerobaculum, Rothia, Actinomyces and Fusobacterium, were common among pre- and post-vaccine populations. Brevundimonas, Corynebacterium, Dolosigranulum and Finegoldia were ‘high abundant’ taxa only among pre-vaccination samples, while Methylobacterium- Methylorubrum and Leifsonia were prevalent among post-vaccine samples (Fig.  4). Distribution of bacteria was generally very patchy, with between 1–5 constitut- ing majority (> 90%) of the microbes in any given sample (Fig. 4). Many butyrate‑producing bacteria are associated with infection and vaccination Differentially abundant bacteria that are associated with test groups were also investigated at the genus Fig. 3  Non-metric dimensional scaling (NMDS) showing sample ordination based on Bray–Curtis dissimilarity and weighted UniFrac phylogenetic distances. The microbial diversity is compared between SARS-CoV-2-positive and -negative individuals before vaccine (A) and post-vaccine (B) availability considering the vaccination status in the latter sample set. UU = unvaccinated uninfected; VU = vaccinated uninfected; UI = unvaccinated infected and VI = vaccinated infected. Samples are coloured dots and ellipses depict 95% confident intervals of the sample clustering per group. The inserted table are PERMANOVA results for vaccination*infection status groups, showing the F-statistic (effect size of variance between compared pair), R2 (quantified variation) and P. adjusted value (significant results are in bold.) Page 7 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 level since species level classification with short reads of the 16SrRNA variable regions are often inconclusive [35]. The range of identification was not limited to ‘high abundant’ (total average relative abundance > 0.01) but, included low-abundance taxa if they were at least 0.01 relative abundance in one of the tested groups. Although many bacteria could be identified as differentially signifi- cant between groups, linear discriminant analyses (LDA) were only plotted for bacteria that satisfied this crite- rion (Table S2) and had an LDA score ≥ 3 instead of the default LDA score of 2 [34]. A total of 16 bacterial genera were differentially abun- dant between SARS-CoV-2-positive and SARS-CoV- 2-negative individuals pre-vaccine availability (Fig.  5A). These included 13 of the ‘high abundant’ taxa previously observed (Fig. 4A) and 3 ‘low abundant’ bacteria. Two of these ‘low abundant’ or rare bacteria were differential dis- criminants of infected individuals (Fig.  5A). About 64% of the differentially abundant bacteria in SARS-CoV-2 positive samples were members of the phylum Bacteroi- dota and Bacillota (particularly class Clostridia), which are known to be important butyrate producers [36]. Two members of class Clostridia; Finegoldia and Peptoniphi- lus, were also more abundant in negative individuals. In the post-vaccine population, however, only Atopo- bium differed between infection status (Fig. 5B), reiterat- ing reduced diversity in this cohort. It was observed to be more abundant in positive individuals as was observed in the pre-vaccine population (Fig.  5A). Accounting for vaccination status, 7 bacteria were found to be differen- tially prevalent among the test groups (Fig. 5C). Interest- ingly, Finegoldia was again associated with unvaccinated uninfected (UU) individuals confirming its significant correlation with non-infection in the study population with regards to SARS-CoV-2. It is worth noting that Fusobacterium, another butyrate producer, was identi- fied as a discriminating microbe, increasing in vaccinated uninfected (VU) persons pre-vaccine and, in positive persons before vaccines were available (Fig. 5A, 5C). Discussion This study is the first to compare the microbial diversity of an African population in relation to infection with SARS-CoV-2 since the pandemic. This study was con- ducted to address the missing information on how the African microbiome correlated with significant variables that characterised the course of the disease. Two sam- ple sets were obtained in a cross-sectional study design, allowing assessment of the microbial community of the study population at two significant milestones of the COVID-19 pandemic. ie. the peak of the pandemic and after vaccination against SARS-CoV-2 was introduced into the study country. Ghana first received the Oxford- AstraZeneca vaccine, which was primarily rolled-out during the vaccination campaign after which others, including Pfizer BioNTech, Johnson & Johnson and Mod- erna also became available by June 2022 when the second sample collection was done. While comparing the naso- oropharyngeal microbiome of infected and uninfected within each sample set, we also compared between the two sample sets to conceptualize microbiome changes in the population given the timelines presented. Based on infection status alone, we describe the study population as homogenous with dysbiosis and individual-individual microbial differences evident only when vaccination sta- tus is accounted for. While Atopobium was significantly Fig. 4  Heatmaps of bacterial genera with average relative abundance > 0.01 in disease groups pre-vaccine availability (A) and vaccination*infection status in the post-vaccine (B) population. Each coloured rectangular block represents the relative abundance of a genus in a sample Page 8 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 associated with SARS-CoV-2 infection, several mem- bers of two major butyrate-producing taxa, Bacillota and Bacteroidota, dominated the naso-oropharyngeal micro- biome in both infected and uninfected groups, whether vaccinated or not. Overall, our results are suggestive of a population with protective immunity that may be linked to high abundance of butyrate-producers. Studies correlating naso-oropharyngeal or oral micro- biome with COVI-19 have been conducted in developed countries on other continents including Asia, Europe and North America [14, 15, 37–39]. Their results generally showed significant differential diversity among infected persons and microbial dysbiosis between infected and uninfected groups. While some reported increased diver- sity [15], others showed a reduction in infected individu- als [14, 37]. Contrary to these results, the current study population reports no differences in ⍺-diversity between infected and uninfected persons. These results are indicative of differential microbial responses to infection within and between geographically distant populations. Human microbiomes differ across continents and, between significantly distant settings within the same geographical location (e.g. country) [40, 41]. It is becom- ing increasingly evident how the extent of this diversity has been largely underestimated [42, 43]. Among sev- eral confounding factors that explain these differences, the ones usually studied in association with microbi- omes are ethnicity, age, diet and disease [42, 44–47]. This current study used samples received or collected at a COVID-19 testing centre in Accra, the capital city of Ghana which has the largest population of any region in the country. Information on ethnicity was not collected because it was not relevant for testing suspected cases of COVID-19, and because the samples were not purposely collected for microbiome analyses. However, all study participants were resident in Accra (data collected from Fig. 5  Linear discriminant analysis (LDA) plots for the identification of significant differential microbes between sample groups. The SARS-CoV-2-infected and uninfected persons are compared among pre-vaccine (A) and post-vaccine (B) populations. Individuals are also compared grouped according to their vaccination*infection status (C) Page 9 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 questionnaire) and only 4 were foreign nationals. Accra is the most cosmopolitan city in Ghana, and the study participants evident from their names (data not shown) originated from various ethnic groups within the coun- try. Their residency in Accra creates the avenue for sev- eral common urbanised lifestyles that can potentially promote homogeneity in their microbiomes. During the pandemic, it became apparent that COVID- 19 severity was linked to the age structure of the popu- lation, particularly that of confirmed cases [48–50]. The association of elderly population with COVID-19 sever- ity has also been linked to differences in microbiome structure between old and young [15]. Africa is consid- ered to have the youngest population in the world, hav- ing a median age of 19 [51]. More than 70% of Ghana’s population is 35 years and below [52]. The age structure of our study population had a median age of 35 (range 18–68) and 37 years as the median age of confirmed cases (Table S1), which is much younger than reported in oral microbiome studies conducted in Asia and Europe [14, 39]. Only one participant in the post-vaccination group was above 60. Although age data was missing for the pre-vaccination sample set, a previous study con- ducted in Ghana pre-vaccination demonstrated that 60% of confirmed cases are reported to be in the 20–39 age bracket [53]. Following COVID-19 vaccination, the SARS-CoV-2 spike protein is detectable in plasma and triggers inflam- mation in epithelial and mucosal sites like the gut [54, 55], and its association with immunity and the gut dys- biosis has been described [55, 56]. Vaccination against COVID-19 has been shown to decrease gut microbial diversity and correlate with stronger immune responses [56]. Similarly, we found our post-vaccination population to have reduced diversity in the naso-oropharynx com- pared to the study population prior, indicative of positive responses and increased immunity resulting from the vaccines. Given the physical connection and chemical communication between the oral-gut axis [57], we can imply that the dysbiosis observed at both sites are linked, but whether the same microbes are regulated at both sites will have to be confirmed with matched analyses of stool and naso-oropharyngeal swabs. Comparable to other reports of gut and oral micro- biome association with COVID-19, we found several opportunistic periodontitis species correlating with infection. Among these Prevotella [14, 38], Atopobium [14], Actinomyces [58], Porphyromonas [15] Lachno- spiraceae, and Leptotrichia [39] significantly increased in infected persons. Although these bacteria were not identified to the species level in the current study, their general lactic acid-producing characteristics and the formation of biofilms [59] support their links with the naso-oropharynx in COVID-19 infections. Particularly, Atopobium was found significantly associated with SARS-CoV-2 positivity in both pre- and post-vaccine groups, and even in vaccinated-infected persons. Based on these results, Atopobium appears to be a striking microbial marker for detecting SARS-CoV-2 infection in this population, but further studies are required to confirm their predictive efficiency. Prevotella is known to be the most abundant genus of the oral microbiome [60] and a notably abundant characteristic microbe in the gut composition of African populations [40, 43] due to high plant and fibre diets [43, 61]. They are also known to be linked to epithelial cytokines and neu- trophil recruitment in acute respiratory and chronic diseases [62–64], which would explain their increased abundance in infected persons, primarily in the pre- vaccine population which was presumably novel to the virus. Prevotella, Alloprevotella and Leptotrichia, also have liposaccharide (LPS)-producing properties which are recognised for their involvement in immunoinhibi- tory pathways [65, 66] but can also be an important host immune stimulant depending on the associated microbial species (reviewed in [67]). About 61% of all differentially abundant bacteria observed belonged to phyla Bacillota (particularly, class Clostridia) and Bacteroidota, two major butyrate- producing taxa [68–71]. While we find some of these butyrate-producing taxa in uninfected persons, the rich- ness of these microbes was further increased in the pres- ence of the viral infection. Although vaccination reduced species diversity, the bacteria that were differentially abundant in those vaccinated comprised of two Bacillota; Solobacterium and Lachnoanaerobaculum. Butyrate is a short-chain fatty acid that is significant in pathophysi- ological processes in humans related to inflammatory diseases. They are involved in reducing mucosal inflam- mation, influencing the fortification of epithelial bar- riers and promoting the relief of oxidative stress [72]. Butyrate-producing microbes form an essential part of the human gut and oral cavity, are acquired in infancy [73] and are maintained in the adult gut through an increase in fibrous diets [74]. In SARS-CoV-2 infection, butyrate has been shown to be effective in regulating the expression of ACE2, proinflammatory cytokines and other genes linked to the progression and disease out- comes of COVID-19 [75] and are predicted to be low in severe COVID-19 [76]. A recent study has directly linked butyrate to reduced cell apoptosis and upregulation of immunity against SARS-CoV-2 infection in experimen- tal mice [77]. Higher abundance of these bacteria in the current study population therefore could be linked to low disease severity and high prevalence of asymptomatic infections [78, 79]. Page 10 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 Conclusion Our study is limited by the small number of partici- pants, particularly in each group. There is also demo- graphic information about the participants we did not have access to such as their disease progression, num- ber of times they have had the infection, and the time lapse between when they were vaccinated and when they were enrolled into the study. These important con- founding factors could have helped explain our results better. However, no similar study has collected all of these in their microbiome-related research, reiterating the importance of standardizing methods for micro- biome studies. Despite these the data presented here has shown results that can be used as the basis for fur- ther investigating the significance of high abundance of butyrate-producers to disease outcomes, particularly in African populations. The use of other high-throughput techniques such shotgun metagenomics can provide comprehensive functional information of the bacteria associated with infection and vaccination. Supplementary Information The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s12879-​024-​09948-z. Supplementary Material 1. Table S1. Metadata of samples used in this study. For each sample, the re-coded sample ID, SARS-CoV-2 status, sex and test Ct values are shown. Supplementary Material 2. Table S2. Relative abundance results of discriminant bacteria from performing LEfSe. Only bacteria with ≥0.01 mean abundance in at least one of the groups compared, and with linear discriminant analysis (LDA) score ≥3 were plotted (Fig 5). Supplementary Material 3. Fig S1. Alpha diversity compared between infection groups without considering vaccination status. Supplementary Material 4. Fig S2. Average relative abundance of ‘high abundant’ bacterial phyla identified in sample sets. These phyla repre‑ sented more than 0.01 average relative abundance and constituted more than 60% of all phyla identified. Acknowledgements We are grateful to the study participants for their involvement in our COVID- 19 projects. We appreciate the support provided by the staff of the COVID-19 Testing Centre, NMIMR, in sample collection and viral testing. Clinical trial number Not applicable. Authors’ contributions JA, MO, and IOD conceptualized the study. JA and MO designed the experi‑ ments. MO, MSA, GYKB, RYI, EO-A, and DA-A collected and analysed postvac‑ cinated samples for viral detection and extracted DNA. JA and FA-T analysed the data and drafted the manuscript. All the authors have read and approved the final manuscript. Funding This study was partly supported by funds from TIBA Out of Africa Fellowship (funded by the National Institute of Health) to JA, National Research Founda‑ tion COVID-19 Africa Rapid Grant Fund (COV19200608529373) to JA and the NMIMR Office for Research Support Fund (EC/P25421/03) to JA. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information file. The demultiplexed sequence reads generated in this study are available in the NCBI SRA database under BioProject PRJNA995054. Declarations Ethics approval and consent to participate Two sets of samples were used in this study. The pre-vaccination samples were retrospective naso-oropharyngeal samples previously collected at or sent to the Noguchi Memorial Institute for Medical Research (NMIMR), University of Ghana, during the peak of the pandemic. Ethical approval (004/20–21) for use of these samples for COVID-19 research was provided by the Noguchi Institu‑ tional Ethical Review Board (Federal Assurance #: 00001824), and consent was waived. Prospective samples for the post-vaccination analyses were drawn from a larger approved project (025/21–22) that collected naso-oropharyn‑ geal swabs and data on the vaccination status of consenting participants. All samples were recoded to ensure anonymity. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Author details 1 Department of Parasitology, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon‑Accra, Ghana. 2 Department of Epidemi‑ ology, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon‑Accra, Ghana. 3 Department of Virology, Noguchi Memorial Institute for Medical Research, University of Ghana, Legon‑Accra, Ghana. 4 Present address Department of Environment and Genetics, School of Agriculture, Bio‑ medicine and Environment, La Trobe University, Bundoora, VIC 3086, Australia. Received: 31 August 2023 Accepted: 17 September 2024 References 1. World Health Organization. WHO Director-General’s opening remarks at the media briefing on COVID-19 - 11 March 2020. 2020. https://​www.​ who.​int/​direc​tor-​gener​al/​speec​hes/​detail/​who-​direc​tor-​gener​al-s-​openi​ ngrem​arks-​at-​the-​media-​brief​ing-​on-​covid-​19---​11-​march-​2020. 2. World Health Organization. Geneva: Available online: https://​covid​19.​ who.​int/ (last cited: [Dec 29, 2022]). 2020. WHO COVID-19 Dashboard. 3. Azevedo MJ. The State of Health System(s) in Africa: Challenges and Opportunities. In: Historical Perspectives on the State of Health and Health Systems in Africa, Volume II. African Histories and Moderni‑ ties. 2017. p. 1–73. Palgrave Macmillan,Cham. https://​doi.​org/​10.​1007/​ 978-3-​319-​32564-4_1. 4. Abu-Hammad O, Alnazzawi A, Borzangy S, Abu-Hammad A, Fayad M, Saadaledin S, et al. Factors Influencing Global Variations in COVID-19 Cases and Fatalities; A Review. Healthcare. 2020Jul 17;8(3):216. 5. De Weerdt J, Van Damme W. Health, Wealth, and the Double Paradox of COVID-19 Mortality in Low-Income Countries [Internet]. SSRN. 2021 Feb [cited 2024 Jul 8]. Available from: De Weerdt, Joachim and Van Damme, Wim, Health, Wealth and the Double Paradox of COVID-19 Mortality in Low-Income Countries (February 8, 2021). Available at SSRN: https://​ssrn.​ com/​abstr​act=​37934​27 or https://​doi.​org/​10.​2139/​ssrn.​37934​27 6. Merad M, Martin JC. Pathological inflammation in patients with COVID- 19: a key role for monocytes and macrophages. Nat Rev Immunol. 2020Jun 17;20(6):355–62. 7. Hou YJ, Okuda K, Edwards CE, Martinez DR, Asakura T, Dinnon KH, et al. SARS-CoV-2 Reverse Genetics Reveals a Variable Infection Gradient in the Respiratory Tract. Cell. 2020Jul;182(2):429–446.e14. 8. Li F, Li W, Farzan M, Harrison SC. Structure of SARS Coronavirus Spike Receptor-Binding Domain Complexed with Receptor. Science (1979). 2005;309(5742):1864–8. https://doi.org/10.1186/s12879-024-09948-z https://doi.org/10.1186/s12879-024-09948-z https://www.who.int/director-general/speeches/detail/who-director-general-s-openingremarks-at-the-media-briefing-on-covid-19---11-march-2020 https://www.who.int/director-general/speeches/detail/who-director-general-s-openingremarks-at-the-media-briefing-on-covid-19---11-march-2020 https://www.who.int/director-general/speeches/detail/who-director-general-s-openingremarks-at-the-media-briefing-on-covid-19---11-march-2020 https://covid19.who.int/ https://covid19.who.int/ https://doi.org/10.1007/978-3-319-32564-4_1 https://doi.org/10.1007/978-3-319-32564-4_1 https://ssrn.com/abstract=3793427 https://ssrn.com/abstract=3793427 https://doi.org/10.2139/ssrn.3793427 Page 11 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 9. Li W, Moore MJ, Vasilieva N, Sui J, Wong SK, Berne MA, et al. Angiotensin- converting enzyme 2 is a functional receptor for the SARS coronavirus. Nature. 2003Nov;426(6965):450–4. 10. Lecarpentier Y, Vallée A. The key role of the level of ACE2 gene expression in SARS-CoV-2 infection. Aging. 2021Jun 15;13(11):14552–6. 11. Koch J, Uckeley ZM, Doldan P, Stanifer M, Boulant S, Lozach PY. TMPRSS2 expression dictates the entry route used by SARS-CoV-2 to infect host cells. EMBO J. 2021;40(16):e107821. 12. Neu U, Mainou BA. Virus interactions with bacteria: Partners in the infec‑ tious dance. PLoS Pathog. 2020Feb 11;16(2): e1008234. 13. Scannapieco FA. Role of Oral Bacteria in Respiratory Infection. J Periodon‑ tol. 1999Jul;70(7):793–802. 14. Soffritti I, D’Accolti M, Fabbri C, Passaro A, Manfredini R, Zuliani G, et al. Oral Microbiome Dysbiosis Is Associated With Symptoms Severity and Local Immune/Inflammatory Response in COVID-19 Patients: A Cross- Sectional Study. Front Microbiol. 2021Jun;23(12): 687513. 15. Gupta A, Bhanushali S, Sanap A, Shekatkar M, Kharat A, Raut C, et al. Oral dysbiosis and its linkage with SARS-CoV-2 infection. Microbiol Res. 2022;261(127055):1–10. Available from: https://​doi.​org/​10.​1016/j.​micres.​ 2022.​127055. Cited 2024 Jun 28. 16. Inoue T, Nakayama J, Moriya K, Kawaratani H, Momoda R, Ito K, et al. Gut Dysbiosis Associated With Hepatitis C Virus Infection. Clin Infect Dis. 2018Aug 31;67(6):869–77. 17. Sun Y, Ma Y, Lin P, Tang YW, Yang L, Shen Y, et al. Fecal bacterial microbi‑ ome diversity in chronic HIV-infected patients in China. Emerg Microbes Infect. 2016Jan 25;5(1):1–7. 18. Ng SC, Peng Y, Zhang L, Mok CK, Zhao S, Li A, et al. Gut microbiota composition is associated with SARS-CoV-2 vaccine immunogenicity and adverse events. Gut. 2022Jun;71(6):1106–16. 19. Allali I, Abotsi RE, Tow LAh, Thabane L, Zar HJ, Mulder NM, et al. Human microbiota research in Africa: a systematic review reveals gaps and priori‑ ties for future research. Microbiome. 2021Dec 15;9(1):241. 20. Makhalanyane TP, Bezuidt OKI, Pierneef RE, Mizrachi E, Zeze A, Fos‑ sou RK, et al. African microbiomes matter. Nat Rev Microbiol. 2023Jun;16(21):479–81. 21. United Nations. UN News Global perspective Human stories. 2021. Ghana and Côte d’Ivoire begin history-making COVAX vaccination drives. Avail‑ able from: https://​www.​news.​un.​org/​en/​story/​2021/​03/​10860​72. Cited 2023 Jul 12. 22. World Health Organization. Ghana finds success in COVID-19 mass vac‑ cination campaigns. 2021. https://​www.​afro.​who.​int/​count​ries/​ghana/​ news/​ghana-​finds-​succe​ss-​covid-​19-​mass-​vacci​nation-​campa​igns. Cited 2023 Jul 12. 23. Opoku M, Obeng-Aboagye E, Boamah GYK, Adu-Asamoah D, Ismail RBY, Akpo MS, Dogbatse EE, Abraham J, Odoom JK, Owusu Donkor I, Akorli J. Assessing the performance of regular surgical nose masks as a sampling method for SARSCoV-2 detection in a cross-sectional study. PLoS One. 2023;18(10):e0293001. 24. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. QIIME allows analysis of high-throughput community sequenc‑ ing data. Nat Methods. 2010;7(5):335–6. Available from: https://​doi.​org/​ 10.​1038/​nmeth.f.​303. Cited 2011 Aug 9. 25. R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2016. Available from: http://​www.r-​proje​ct.​org 26. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods. 2016Jul 23;13(7):581–3. Available from: http://​www.​nature.​ com/​artic​les/​nmeth.​3869. Cited 2020 Feb 11. 27. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41:D590–6. 28. Davis NM, Proctor DM, Holmes SP, Relman DA, Callahan BJ, Simple statisti‑ cal identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome [Internet]. [cited 2019 Sep 5];6:226. Available from: 2018. https://​doi.​org/​10.​1186/​s40168-​018-​0605-2. 29. McMurdie PJ, Holmes S. phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLoS ONE. 2013Apr 22;8(4): e61217. 30. Ling W, Lu J, Zhao N, Lulla A, Plantinga AM, Fu W, et al. Batch effects removal for microbiome data via conditional quantile regression. Nat Commun. 2022Sep 15;13(1):5418. 31. Liu C, Cui Y, Li X, Yao M, microeco: an R package for data mining in micro‑ bial community ecology, FEMS Microbiol Ecol. 2021;97(2):fiaa255. 32. Bray JR, Curtis JT. An ordination of upland forest communities of southern Wisconsin. Ecol Monogr. 1957;27:325–49. 33. Lozupone C, Knight R. UniFrac : a New Phylogenetic Method for Compar‑ ing Microbial Communities. Appl Environ Microbiol. 2005;71(12):8228–35. 34. Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12(6):R60. Available from:  https://​www.​pmc/​artic​les/​PMC32​18848/. Cited 2021 Aug 11. 35. Johnson JS, Spakowicz DJ, Hong BY, Petersen LM, Demkowicz P, Chen L, et al. Evaluation of 16S rRNA gene sequencing for species and strain-level microbiome analysis. Nat Commun. 2019Nov 6;10(1):5029. 36. Chen W, Zhang S, Wu J, Ye T, Wang S, Wang P, Xing D. Butyrate-producing bacteria and the gut-heart axis in atherosclerosis. Clin Chim Acta. 2020;507:236–41. 37. Mostafa HH, Fissel JA, Fanelli B, Bergman Y, Gniazdowski V, Dadlani M, et al. Metagenomic Next-Generation Sequencing of Nasopharyngeal Specimens Collected from Confirmed and Suspect COVID-19 Patients. mBio. 2020 Dec 22;11(6). 38. Rosas-Salazar C, Kimura KS, Shilts MH, Strickland BA, Freeman MH, Wessinger BC, et al. SARS-CoV-2 infection and viral load are associated with the upper respiratory tract microbiome. J Allergy Clin Immunol. 2021Apr;147(4):1226–1233.e2. Available from: https://​linki​nghub.​elsev​ier.​ com/​retri​eve/​pii/​S0091​67492​10017​79. 39. Gao M, Wang H, Luo H, Sun Y, Wang L, Ding S, Ren H, Gang J, Rao B, Liu S, Wang X, Gao X, Li M, Zou Y, Liu C, Yuan C, Sun J, Cui G, Ren Z. Characteri‑ zation of the human oropharyngeal microbiomes in SARS-CoV-2 infec‑ tion and recovery patients. Adv Sci (Weinh). 2021;8(20):e2102785. 40. B Hansen ME, Rubel MA, Bailey AG, Ranciaro A, Thompson SR, Campbell MC, et al. Population structure of human gut bacteria in a diverse cohort from rural Tanzania and Botswana. Genome Biol. 2019;20(16):1. Available from: https://​doi.​org/​10.​1186/​s13059-​018-​1616-9. 41. Gaulke CA, Sharpton TJ. The influence of ethnicity and geography on human gut microbiome composition. Nat Med. 2018Oct 1;24(10):1495–6. 42. Shanahan F, Ghosh TS, O’Toole PW. Human microbiome variance is underestimated. Curr Opin Microbiol. 2023Jun;1(73): 102288. 43. Deschasaux M, Bouter KE, Prodan A, Levin E, Groen AK, Herrema H, et al. Depicting the composition of gut microbiota in a population with varied ethnic origins but shared geography. Nat Med. 2018Oct 27;24(10):1526–31. 44. Rackerby B, Kim HJ, Dallas DC, Park SH. Understanding the effects of dietary components on the gut microbiome and human health. Food Sci Biotechnol. 2020Nov 29;29(11):1463–74. 45. Dwiyanto J, Hussain MH, Reidpath D, Ong KS, Qasim A, Lee SWH, et al. Ethnicity influences the gut microbiota of individuals sharing a geo‑ graphical location: a cross-sectional study from a middle-income country. Sci Rep. 2021Jan 29;11(1):2618. 46. Wu GD, Chen J, Hoffmann C, Bittinger K, Chen YY, Keilbaugh SA, et al. Linking Long-Term Dietary Patterns with Gut Microbial Enterotypes. Sci‑ ence (1979). 2011;334(6052):105–8. 47. Chong CW, Ahmad AF, Lim YAL, Teh CSJ, Yap IKS, Lee SC, et al. Effect of ethnicity and socioeconomic variation to the gut microbiota composi‑ tion among pre-adolescent in Malaysia. Sci Rep. 2015Aug 20;5(1):13338. 48. Dudel C, Riffe T, Acosta E, van Raalte A, Strozza C, Myrskylä M. Monitoring trends and differences in COVID-19 case-fatality rates using decomposi‑ tion methods: Contributions of age structure and age-specific fatality. PLoS ONE. 2020Sep 10;15(9): e0238904. 49. Hulíková Tesárková K, Dzúrová D. The age structure of cases as the key of COVID-19 severity: Longitudinal population-based analysis of European countries during 150 days. Scand J Public Health. 2022Aug 20;50(6):738–47. 50. Mizumoto K, Dahal S, Chowell G. Spatial variability in the risk of death from COVID-19 in Italy. Int J Tuberc Lung Dis. 2020Aug 1;24(8):829–37. 51. Cilliers J, Kwasi S, Yeboua K, Oosthuizen M, Alexander K, Pooe TK, Moyer JD. Impact of COVID-19 in Africa: A Scenario Analysis to 2030 (July 23, 2020). Africa Report, Institute for Security Studies, Pretoria, 2020. p. 1-40. https://doi.org/10.1016/j.micres.2022.127055 https://doi.org/10.1016/j.micres.2022.127055 https://www.news.un.org/en/story/2021/03/1086072 https://www.afro.who.int/countries/ghana/news/ghana-finds-success-covid-19-mass-vaccination-campaigns https://www.afro.who.int/countries/ghana/news/ghana-finds-success-covid-19-mass-vaccination-campaigns https://doi.org/10.1038/nmeth.f.303 https://doi.org/10.1038/nmeth.f.303 http://www.r-project.org http://www.nature.com/articles/nmeth.3869 http://www.nature.com/articles/nmeth.3869 https://doi.org/10.1186/s40168-018-0605-2 https://www.pmc/articles/PMC3218848/ https://linkinghub.elsevier.com/retrieve/pii/S0091674921001779 https://linkinghub.elsevier.com/retrieve/pii/S0091674921001779 https://doi.org/10.1186/s13059-018-1616-9 Page 12 of 12Akorli et al. BMC Infectious Diseases (2024) 24:1020 Available at SSRN: https://​ssrn.​com/​abstr​act=​36608​66 or http://​dx.​doi.​ org/​10.​2139/​ssrn.​36608​66. 52. Atiglo YD. The Conversation Africa. 2023 [cited 2024 Jul 6]. Ghana’s population is young and rapidly urbanising - policies need to match the data. Available from: https://​theco​nvers​ation.​com/​ghanas-​popul​ation-​is-​ young-​and-​rapid​ly-​urban​ising-​polic​ies-​need-​to-​match-​the-​data-​20951​ 0#:​~:​text=​The%​20age%​20com​posit​ion%​20of%​20Gha​na’s,people%​ 20are%​20in%​20urb​an%​20are​as. 53. Odikro MA, Kenu E, Malm KL, Asiedu-Bekoe F, Noora CL. Epidemiology of COVID-19 outbreak in Ghana, 2020. Ghana Med J. 2020;54(4s):5–15. 54. Röltgen K, Nielsen SCA, Silva O, Younes SF, Zaslavsky M, Costales C, et al. Immune imprinting, breadth of variant recognition, and germinal center response in human SARS-CoV-2 infection and vaccination. Cell. 2022Mar;185(6):1025–1040.e14. 55. Boston RH, Guan R, Kalmar L, Beier S, Horner EC, Beristain-Covarrubias N, et al. Stability of gut microbiome after COVID-19 vaccination in healthy and immuno-compromised individuals. Life Sci Alliance. 2024Apr 5;7(4): e202302529. 56. Ray S, Narayanan A, Vesterbacka J, Blennow O, Chen P, Gao Y, 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. 2023Dec 20;9(1):104. 57. Park SY, Hwang BO, Lim M, Ok SH, Lee SK, Chun KS, et al. Oral-Gut Microbiome Axis in Gastrointestinal Disease and Cancer. Cancers (Basel). 2021Apr 28;13(9):2124. 58. Tao W, Zhang G, Wang X, Guo M, Zeng W, Xu Z, et al. Analysis of the intestinal microbiota in COVID-19 patients and its correlation with the inflammatory factor IL-18. Medicine in Microecology. 2020Sep;5: 100023. 59. Swidsinski A, Mendling W, Loening-Baucke V, Ladhoff A, Swidsinski S, Hale LP, et al. Adherent Biofilms in Bacterial Vaginosis. Obstet Gynecol. 2005;106:1013–23. 60. Dewhirst FE, Chen T, Izard J, Paster BJ, Tanner ACR, Yu WH, et al. The Human Oral Microbiome. J Bacteriol. 2010Oct;192(19):5002–17. 61. De Filippo C, Cavalieri D, Di Paola M, Ramazzotti M, Poullet JB, Massart S, et al. Impact of diet in shaping gut microbiota revealed by a compara‑ tive study in children from Europe and rural Africa. Proc Natl Acad Sci. 2010Aug 17;107(33):14691–6. 62. Könönen E, Gursoy UK. Oral Prevotella Species and Their Connection to Events of Clinical Relevance in Gastrointestinal and Respiratory Tracts. Front Microbiol. 2022Jan;6:12. 63. Könönen E, Fteita D, Gursoy UK, Gursoy M. Prevotella species as oral residents and infectious agents with potential impact on systemic condi‑ tions. J Oral Microbiol. 2022;14(1):2079814. 64. de Aquino SG, Abdollahi-Roodsaz S, Koenders MI, van de Loo FAJ, Pruijn GJM, Marijnissen RJ, et al. Periodontal Pathogens Directly Promote Auto‑ immune Experimental Arthritis by Inducing a TLR2- and IL-1–Driven Th17 Response. J Immunol. 2014May 1;192(9):4103–11. 65. 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. 66. Vatanen T, Kostic AD, d’Hennezel E, Siljander H, Franzosa EA, Yassour M, et al. Variation in Microbiome LPS Immunogenicity Contributes to Auto‑ immunity in Humans. Cell. 2016May;165(4):842–53. 67. Mohr AE, Crawford M, Jasbi P, Fessler S, Sweazea KL. Lipopolysaccharide and the gut microbiota: considering structural variation. FEBS Lett. 2022Apr 15;596(7):849–75. 68. Shah HN, Bowden GH, Hardie JM, Williams RAD. Comparison of the Bio‑ chemical Properties of Bacteroides melaninogenicus from Human Dental Plaque and Other Sites. J Appl Bacteriol. 1976Dec;41(3):473–92. 69. Louis P, Flint HJ. Diversity, metabolism and microbial ecology of butyrate- producing bacteria from the human large intestine. FEMS Microbiol Lett. 2009May;294(1):1–8. 70. Rivière A, Selak M, Lantin D, Leroy F, De Vuyst L. Bifidobacteria and Butyrate-Producing Colon Bacteria: Importance and Strategies for Their Stimulation in the Human Gut. Front Microbiol. 2016Jun;28:7. 71. Manson JM, Rauch M, Gilmore MS. The commensal microbiology of the gastrointestinal tract. In: Huffnagle GB, Noverr MC, editors. GI Microbiota and Regulation of the Immune System Advances in Experimental Medi‑ cine and Biology. New York, NY: Springer; 2008. p. 15–28. 72. Hamer HM, Jonkers D, Venema K, Vanhoutvin S, Troost FJ, Brummer RJ. Review article: the role of butyrate on colonic function. Aliment Pharma‑ col Ther. 2007Oct 26;27(2):104–19. 73. Appert O, Garcia AR, Frei R, Roduit C, Constancias F, Neuzil-Bunesova V, et al. Initial butyrate producers during infant gut microbiota development are endospore formers. Environ Microbiol. 2020Sep 20;22(9):3909–21. 74. Wächtershäuser A, Stein J. Rationale for the luminal provision of butyrate in intestinal diseases. Eur J Nutr. 2000Sep 1;39(4):164–71. 75. Paparo L, Maglio MA, Cortese M, Bruno C, Capasso M, Punzo E, et al. A New Butyrate Releaser Exerts a Protective Action against SARS-CoV-2 Infection in Human Intestine. Molecules. 2022Jan 27;27(3):862. 76. Nguyen LH, Okin D, Drew DA, Battista VM, Jesudasen SJ, Kuntz TM, et al. Metagenomic assessment of gut microbial communities and risk of severe COVID-19. Genome Med. 2023Jul 12;15(1):49. 77. Yu H, Yuan L, Yan Z, Zhou M, Ye J, Wu K, et al. Butyrate Protects against SARS-CoV-2-Induced Tissue Damage in Golden Hamsters. Int J Mol Sci. 2023Sep 16;24(18):14191. 78. Owusu Donkor I, Mensah SK, Dwomoh D, Akorli J, Abuaku B, Ashong Y, Opoku M, Andoh NE, Sumboh JG, Ohene SA, Owusu-Asare AA, Quartey J, Dumashie E, Lomotey ES, Odumang DA, Gyamfi GO, Dorcoo C, Afatodzie MS, Osabutey D, Ismail RBY, Quaye I, Bosomprah S, Munster V, Koram KA. Modeling SARS-CoV-2 antibody seroprevalence and its determinants in Ghana: A nationally representative cross-sectional survey. PLOS Glob Public Health. 2023;3(5):e0001851. 79. Quashie PK, Mutungi JK, Dzabeng F, Oduro-Mensah D, Opurum PC, Tapela K, et al. Trends of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibody prevalence in selected regions across Ghana. Wellcome Open Res. 2021Jul;6(6):173. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. https://ssrn.com/abstract=3660866 https://dx.doi.org/10.2139/ssrn.3660866 https://dx.doi.org/10.2139/ssrn.3660866 https://theconversation.com/ghanas-population-is-young-and-rapidly-urbanising-policies-need-to-match-the-data-209510#:~:text=The%20age%20composition%20of%20Ghana’s,people%20are%20in%20urban%20areas https://theconversation.com/ghanas-population-is-young-and-rapidly-urbanising-policies-need-to-match-the-data-209510#:~:text=The%20age%20composition%20of%20Ghana’s,people%20are%20in%20urban%20areas https://theconversation.com/ghanas-population-is-young-and-rapidly-urbanising-policies-need-to-match-the-data-209510#:~:text=The%20age%20composition%20of%20Ghana’s,people%20are%20in%20urban%20areas https://theconversation.com/ghanas-population-is-young-and-rapidly-urbanising-policies-need-to-match-the-data-209510#:~:text=The%20age%20composition%20of%20Ghana’s,people%20are%20in%20urban%20areas High abundance of butyrate-producing bacteria in the naso-oropharynx of SARS-CoV-2-infected persons in an African population: implications for low disease severity Abstract Background Methods Results Conclusion Background Methods Clinical samples Sample processing for microbiome analyses Sequence processing Diversity analyses Differential microbial abundance Results Sequence exploration and statistics Within-group microbial diversity increased with vaccination Phylogenetically distinct taxa explain higher diversity between groups Bacillota and bacteroidota constitute majority of high abundant bacteria Many butyrate-producing bacteria are associated with infection and vaccination Discussion Conclusion Acknowledgements References