Gates Open Res Gates Open Research 2572-4754 F1000 Research Limited London, UK 10.12688/gatesopenres.14041.1 Research Article Articles Characterization of longitudinal nasopharyngeal microbiome patterns in maternally HIV-exposed Zambian infants

[version 1; peer review: 2 approved with reservations]

 Odom Aubrey R. Data Curation Formal Analysis Methodology Project Administration Software Validation Visualization Writing – Original Draft Preparation https://orcid.org/0000-0001-7113-7598 a 1 Gill Christopher J. Conceptualization Funding Acquisition Investigation Supervision Writing – Review & Editing https://orcid.org/0000-0003-3353-0617 2 Pieciak Rachel Project Administration Writing – Review & Editing https://orcid.org/0000-0002-1514-5320 2 Ismail Arshad Resources Writing – Review & Editing https://orcid.org/0000-0003-4672-5915 3 4 Thea Donald Investigation Supervision Writing – Review & Editing 2 MacLeod William B. Data Curation Investigation Methodology Writing – Review & Editing https://orcid.org/0000-0001-8003-8874 2 Johnson W. Evan Conceptualization Formal Analysis Methodology Supervision Writing – Review & Editing 1 5 Lapidot Rotem Conceptualization Data Curation Investigation Methodology Supervision Writing – Review & Editing 6 7 Bioinformatics Program, Boston University, Boston, MA, 02118, USA Department of Global Health, Boston University School of Public Health, Boston, MA, 02118, USA Sequencing Core Facility, National Institute for Communicable Diseases of the National Health Laboratory Service, 2131 Johannesburg, South Africa Department of Biochemistry and Microbiology, University of Venda, Thohoyandou 0950, South Africa Division of Infectious Disease, Center for Data Science, Rutgers New Jersey Medical School, Newark, NJ, 07103, USA Pediatric Infectious Diseases, Boston Medical Center, Boston, MA, 02118, USA Department of Pediatrics, Boston University School of Medicine, Boston, MA, 02118, USA aodom@bu.edu

No competing interests were disclosed.

 11 11 2022 2022 6 143 4 11 2022 Copyright: © 2022 Odom AR et al. 2022 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Previous studies of infants born to HIV-positive mothers have linked HIV exposure to poor outcomes from gastrointestinal and respiratory illnesses, and to overall increased mortality rates. The mechanism behind this is unknown, but it is possible that differences in the nasopharyngeal (NP) microbiome between HIV-unexposed and HIV-exposed infants could play a role in perpetuating some outcomes.

Methods: We conducted a longitudinal analysis of 170 NP swabs of healthy HIV-exposed, uninfected (HEU; n=10) infants and their HIV(+) mothers and HIV-unexposed, uninfected (HUU; n=10) infants and their HIV(-) mothers. These swabs were identified from a sample library collected in Lusaka, Zambia between 2015 and 2016. Using 16S rRNA gene sequencing, we characterized the maturation of the microbiome over the first 14 weeks of life to determine what quantifiable differences exist between HEU and HUU infants, and what patterns are reflected in the mothers' NP microbiomes.

Results: In both HEU and HUU infants, Staphylococcus and Corynebacterium began as primary colonizers of the NP microbiome but were in time replaced by Dolosigranulum, Streptococcus, Moraxella and Haemophilus. When studying differences between infants, the microbe Staphylococcus haemolyticus indicated a distinctive high association with HIV exposure at birth, even when accounting for the interaction between HIV exposure status and time of sampling. When comparing infants to their mothers with paired analyses, HEU infants’ NP microbiome composition was only slightly different from their HIV(+) mothers at birth or 14 weeks, including in their carriage of S. pneumoniae, H. influenzae, and S. haemolyticus.

Conclusions: Our analyses indicate that the HEU infants in our study exhibit subtle differences in the NP microbial composition throughout the sampling interval. Given our results and the sampling limitations of our study, we believe that further research must be conducted in order to confidently understand the relationship between HIV exposure and infants’ NP microbiomes.

 nasopharyngeal microbiome longitudinal cohort study microbial communities HIV exposure children National Institute of Allergy and Infectious Diseases R01AI133080 Gates Foundation OPP1105094 National Institutes of Health R21AI154387 National Institute of General Medical Sciences T32GM100842 This work was supported by the Gates Foundation [OPP1105094] who funded the Southern Africa Mother Infant Pertussis Study: Phase I (SAMIPS-1) and also by the NIAID of the NIH under grant R01AI133080 to C.G. W.E.J and A.R.O.M. were supported in part by the NIH under grant R21AI154387. A.R.O.M. was supported in part by the NIGMS of the NIH under grant T32GM100842. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction

Currently, more than one million infants are born to HIV-infected [HIV(+)] women worldwide every year. 1 Fortunately, with antiretroviral treatment for mothers and prophylaxis for their infants, the vast majority of HIV-exposed infants will not become infected with HIV. 2 However, prevention of mother-to-child transmission (PMTCT) does not eradicate health disparities between HIV-exposed, uninfected (HEU) and HIV-unexposed, uninfected (HUU) infants by eliminating HIV transmission. Data suggests these children are still directly or indirectly affected by their mother’s HIV status. For example, recent meta-analyses published by members of our team reported a 60% increased risk of death 3 and an increased risk of pneumonia and diarrhea 4 among HEU compared with HUU children, thereby supporting the observed phenomenon that infants are vulnerable not only to morbidity, but also to increased mortality. These findings have also been observed in numerous other studies that have shown the linkage between poor outcomes and gastrointestinal and respiratory illnesses. 5– 9 Hypothesized explanations include dysregulation of passive immunity via maternal antibodies, changes in the maturation of infant lymphocytes, exposure to microbes present in the mother’s birth canal at delivery, and/or social factors, which may have an impact on morbidity and mortality rates in the early stages of life.

Alternatively, it remains possible that many of these reported health differences are merely artifacts of various selection biases. Previous studies on HIV exposure phenomenon were cross sectional studies based on convenience sampling and/or lacked precision due to small sample sizes. 3, 4 Further, few such studies were longitudinal, making it difficult to observe changes over time. 3, 4 When studying the microbiome, integral to the early development of the immune system, the lack of longitudinal structure is a crucial limitation given that it evolves dynamically over the first days and weeks of life, as both we and others have previously demonstrated. 10, 11 Thus, it remains possible that much of the HEU phenomenon as currently described could be due to sociological factors, or simply selection bias. To better understand the potential biological basis for this phenomenon, unbiased, systematic data are required to build confidence in its existence.

In the gut, interactions between the microbial community and the host influence the development of the immune system and, consequently, the development of diseases. 12 While much is known about the evolution of the intestinal microbiome, far less is known about the dynamics surrounding the microbiome of the upper respiratory tract, which plays an important role in respiratory health. If the increased rates of respiratory disease observed among HEU children have a biological explanation, we hypothesized that the respiratory microbiomes of HEU infants would differ systematically from HUU infants over the first few months of life, potentially explaining their greater susceptibility to certain respiratory diseases. Previous studies on the impact of HIV exposure have highlighted strong correlations between HIV exposure and increased risk of pneumococcal colonization and disease, 7, 13, 14 whereas others have demonstrated no differences in pathogen carriage between children with HIV infection and control groups. 15– 17 The current analysis seeks to address this important knowledge gap.

Given the complex dynamics of interactions between the host, microbes, and environment beginning at birth, we conducted an exploratory longitudinal comparison of the nasopharyngeal (NP) microbiomes of HEU and HUU infants. We reasoned that this pilot study could provide insight into the presence or absence of immunological factors distinguishing these two groups. We collected 167 samples from 10 HEU and 10 HUU infants and their mothers from Lusaka, Zambia, which were obtained shortly after the infants’ birth and at two-week intervals thereafter through 14 weeks of life. We sought to characterize the NP microbiome of these infants to determine what quantifiable differences exist between HEU and HUU infants.

 Methods

A subset of 170 NP swabs of 20 infants and their mothers were identified from a sample library collected in Lusaka, Zambia between 2015 and 2016. The sample library was part of a nested time-series case comparator study within the prospective longitudinal Southern Africa Mother-Infant Pertussis study (SAMIPS). 18 For the cohort, infants were included if they were otherwise ‘healthy’ when screened at one week of age. Healthy infants were born at term (>37 weeks); not underweight (>2500 grams); had no acute or chronic conditions known at the time of enrollment; were not born via cesarean section; and had no known complications during pregnancy or labor and delivery.

The institutional review boards at Boston Medical Center and Excellence in Research Ethics and Science Converge in Lusaka jointly provided ethical oversight (The ERES Converge, Lusaka. REF# 2015-Jan-002, Date: 01/02/2015; BUMC IRB, Boston. # H-33521, Date: 12/12/2014). All mothers provided written informed consent, with consent forms presented in English and the two dominant vernacular languages spoken in Lusaka: Bemba and Nyanja.The present analysis uses HEU ( n=10) infants and their HIV(+) mothers alongside HUU healthy control ( n=10) infants and their HIV-negative [HIV(-)] mothers collected as part of the SAMIPS study. Characteristics of the study cohort are delineated in Table 1. Mother-infant pairs were recruited during their first scheduled postpartum well-child visit at approximately one week of age. Infants and their mothers were enrolled from the Chawama Primary Health Clinic (PHC) in Chawama compound, a densely populated peri-urban area near central Lusaka. Chawama PHC is the only government-supported clinic in this community and is the primary source of medical care for Chawama residents, allowing for maximal study reach. NP swabs were obtained from infants at enrollment and approximately every two to three weeks thereafter through 14 weeks of age for a total of seven scheduled time points each. Mothers’ samples were gathered at all time points at which the infants were swabbed, but only the first and last time points at weeks zero to two and 12–14 were sequenced for analysis for a total of 40 samples.

Baseline demographic characteristics of the full infant cohort, stratified by HIV exposure status.

IQR refers to the interquartile range, BCG refers to the Bacille Calmette-Guérin vaccine for tuberculosis disease, and OPV refers to oral polio vaccine.

Parameter HIV-Exposed HIV-Unexposed All Subjects
Number of Infants 10 10 20
Place of Birth
University Teaching Hospital % (n) 40.0% (4/10) 30.0% (3/10) 35.0% (7/20)
Chawama Clinic % (n) 60.0% (6/10) 60.0% (6/10) 60.0% (12/20)
Chilenje Clinic % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Home Delivery % (n) 0.0% (0/10) 10.0% (1/10) 5.0% (1/20)
Other % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Male Sex % (n) 50.0% (5/10) 40.0% (4/10) 45.0% (9/20)
Median Age in Days at Enrollment (IQR) 6.0 (6–7) 6.5 (6–7) 6.0 (6–7)
Gestational Age
Median Estimated Gestational Age at Delivery (weeks) (IQR) 40.0 (38–40) 40.0 (39–40) 40.0 (39–40)
Median Birth Weight (IQR) 3050.0 (2800–3200) 3100.0 (3000–3100) 3100.0 (2900–3200)
Twin Birth % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Immunizations at birth
BCG % (n) 30.0% (3/10) 50.0% (5/10) 40.0% (8/20)
OPV % (n) 30.0% (3/10) 20.0% (2/10) 25.0% (5/20)
Number of Mothers 10 10 20
Median Age in Years at Enrollment (IQR) 34.0 (29–36) 34.0 (30–37) 34.0 (30–37)
Married % (n) 100.0% (10/10) 100.0% (10/10) 100.0% (20/20)
Father Lives with Child % (n) . % (./0) . % (./0) . % (./0)
Labor and Delivery Complications
Obstructed Labor % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Birth Asphyxia % (n) 10.0% (1/10) 0.0% (0/10) 5.0% (1/20)
Sepsis % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Haemorrhage % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Other Birth Complication % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Maternal Immunization
Median Number of Tetanus Toxoid Doses (IQR) 5.0 (3–5) 5.0 (3–5) 5.0 (3–5)
Received other vaccines % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Received Influenza % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Received TDaP % (n) 0.0% (0/10) 0.0% (0/10) 0.0% (0/20)
Maternal HIV Status
Mother HIV(+) % (n) 100.0% (10/10) 0.0% (0/10) 50.0% (10/20)
Mother on ART % (n) 100.0% (10/10) . % (./0) 100.0% (10/10)
Mother on ART Prior to Pregnancy % (n) 100.0% (10/10) . % (./0) 100.0% (10/10)
Trimester initiated ART
First Trimester % (n) . % (./0) . % (./0) . % (./0)
Second Trimester % (n) . % (./0) . % (./0) . % (./0)
Third Trimester % (n) . % (./0) . % (./0) . % (./0)
Household composition
Median Household Size (IQR) 5.5 (4–7) 6.0 (5–7) 6.0 (5–7)

HIV(+) mothers enrolled in the SAMIPS study were required to be on antiretroviral therapy (ART) to prevent mother-child transmission. Among the mothers in our immediate cohort, 50% (10/20) were HIV infected, of whom 100% had initiated antiretroviral treatment prior to conception. The study did not collect CD4 counts from study subjects, and assessment of mother’s HIV status relied upon previous testing done at the clinic. Although data on maternal HIV status was available, final HIV status could not be ascertained on the infants themselves, which typically is not possible until the infant is four to six months of age. As all mothers received ART during pregnancy, pooled transmission rates of breastfeeding mothers would be about 3.54% (95% CI: 1.15–5.93%) at the six month mark. 19 Therefore, it can be assumed that infants in this study becoming HIV(+) would be rare, and so all infants born to an HIV(+) mother were classified here as being HIV-exposed with two unknown subpopulations of HEU and HIV(+) infants. All enrolled infants received the pentavalent and pneumococcal vaccines at ages six, 10 and 14 weeks, which offer protection against 10 pneumococcal serotypes and Haemophilus influenzae type B. Additional information about the study structure and sampling methods can be found in Gill et al. (2016). 18

The HEU infants were restricted to subjects who had entered the study in April, May, June, or July and were matched with HUU infants on month of enrollment into study, mother’s age, and mother’s education. Some missingness in the data is present; for ten samples, these visits either never occurred or swabs were not collected. An additional three samples were excluded from analysis because fewer than 10,000 reads aligned to RefSeq reference genomes. This left us with 129 infant samples and 38 mother samples for analysis. In total, 16/20 infants had data for all seven time points, 2/20 infants had only six time points, one infant had three time points, and another infant had only two time points. All infants had swabs for the first and second time points. Only two of 20 mothers lacked both time points. The numbers of infant samples and mother samples collected for each time point are enumerated fully in Table 2.

Counts and percentages of samples available for analysis from both infants and mothers.

These are stratified by time point, infant age in weeks, and HIV status. HEU: HIV-exposed, uninfected; HUU: HIV-unexposed, uninfected.

Infant Mother Overall
( n=129) ( n=38) ( n=167)
Average infant age (weeks)  
0.58 (t=0) 20 (15.5%) 20 (52.6%) 40 (24.0%)
2.88 (t=1) 20 (15.5%) 0 (0%) 20 (12.0%)
6 (t=2) 19 (14.7%) 0 (0%) 19 (11.4%)
8.21 (t=3) 17 (13.2%) 0 (0%) 17 (10.2%)
10.21 (t=4) 17 (13.2%) 0 (0%) 17 (10.2%)
12.22 (t=5) 18 (14.0%) 0 (0%) 18 (10.8%)
14.35 (t=6) 18 (14.0%) 18 (47.4%) 36 (21.6%)
HIV Status  
HUU/HIV(-) 68 (52.7%) 19 (50.0%) 87 (52.1%)
HEU/HIV(+) 61 (47.3%) 19 (50.0%) 80 (47.9%)
Sample processing and storage

NP swabs were obtained from the posterior nasopharynx using a sterile flocked tipped nylon swab (Copan Diagnostics, Merrieta, California). The swabs were then placed in universal transport media, put on ice and transferred to our onsite lab on the same campus, where they were aliquoted and stored at -80°C until DNA extraction. DNA was extracted using the NucliSENS EasyMagG System (bioMérieux, Marcy l’Etoile, France). Extracted DNA was stored at our lab located at the University Teaching Hospital in Lusaka at -80°C. Sample collection, processing and storage were previously described (Gill et al., 2016). 18

16S ribosomal DNA amplification and MiSeq sequencing

For 16S library preparations, two PCR reactions were completed on the template DNA. Initially the DNA was amplified using universal bacterial primers 20 specific to the V3–V4 region of the 16S rRNA gene. 21 Library preparation was performed according to the standard instructions of the 16S Metagenomic Sequencing Library Preparation protocol (Illumina, USA). The 16S primer pairs incorporated the Illumina overhang adaptor (16S forward primer 5’-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3’; 16S reverse primer 5’-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3’)

Each PCR reaction contained DNA template (~12 ng), 5µℓ forward primer (1μM), 5 µℓ reverse primer (1μM), 12.5 µℓ 2 X Kapa HiFi Hotstart ready mix (KAPA Biosystems Woburn, MA), and PCR grade water to a final volume of 25µℓ. PCR amplification was carried out as follows: heated lid 110°C, 95°C for 3 min, 25 cycles of 95°C for 30s, 55°C for 30s, 72°C for 30s, then 72°C for 5 min and held at 4°C. Negative control reactions without any template DNA were carried out simultaneously.

The size of the amplicons was then visualized using the 4200 TapeStation (Agilent Technologies, Germany). Successful PCR products were cleaned using AMPure XP magnetic bead-based purification (Beckman Coulter, IN). The IDT for Illumina Nextera DNA UD Indexes kit (Illumina, San Diego, CA) with unique dual index adapters were used to allow for multiplexing. Each PCR reaction contained purified DNA (5 μℓ), 10 μℓ index primer mix, 25 μℓ 2X Kapa HiFi Hot Start Ready mix and 10 μℓ PCR grade water. PCR reactions were performed on a Bio-Rad C1000 Thermal Cycler (Bio-Rad, Hercules, CA) Cycling conditions consisted of one cycle of 95°C for 3 min, followed by eight cycles of 95°C for 30 s, 55°C for 30 s and 72°C for 30 s, followed by a final extension cycle of 72°C for 5 min. PCR products of negative controls were confirmed negative on Agilent TapeSataion (no band observed).

Prior to library pooling, the indexed libraries were purified with Ampure XP beads and quantified using the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA). Purified amplicons were run on the Agilent TapeStation (Agilent Technologies, Germany) for quality analysis before sequencing. The sample pool (2 nM) was denatured with 0.2N NaOH, then diluted to 4 pM and combined with 10% (v/v) denatured 20 pM PhiX, prepared following Illumina guidelines. Libraries were then sequenced on the Illumina MiSeq sequencing platform (Illumina, USA) at the Sequencing Core Facility, National Institute for Communicable Diseases (NICD) of the National Health Laboratory Service, South Africa, using a 2 x 300 cycle V3 kit, following standard Illumina sequencing protocols. Negative controls were sequenced as well, resulting in extremely low reads that were not further analyzed.

In addition to using negative controls, all samples were processed at random to account for reagent contamination. Lab technicians were blinded to the timing of sample collection and clinical data.

We assessed the quality of the sequencing data using FastQC v0.11.9. 22 Trimmomatic 20 v0.39 was used to trim Illumina adapters and remove low-quality sequences. We performed a sliding window trim, cutting once when the average quality score within a window of six bases falls below 15. We removed both leading and trailing low quality or N bases below quality six. All other parameters used the default settings.

Sequencing data were aligned to bacterial genomes and profiled using the PathoScope 2.0 pipeline. 23 All RefSeq representative bacterial genomes available as of November 2, 2018 were used as a PathoScope reference library. We obtained target read counts delineated by National Center for Biotechnology Information (NCBI) unique identifiers (UIDs) and matched them to the NCBI Taxonomy database to retrieve accurate taxonomic hierarchy information. We then aggregated reads by genera. Data were transformed to relative abundances using counts per million (CPM) and normalized using log CPM for subsequent analyses. In most cases, taxa belonging to genera with average relative abundances of less than 1% were grouped as “Other” in analyses. All code has been made available via Zenodo. 24

Statistical analysis

All analyses were performed using R Statistical Software (v4.2.1; R Core Team 2022).

Microbial abundances across sample groups were visualized using alluvial diagrams and stacked bar plots using the R packages ggplot2 v3.3.5 and alluvial v0.2-0 . The alluvial diagrams illustrate individual genera as stream fields that change position at different time points. The height of a stream field represents the relative abundance of that taxon. At a given time point, stream fields are ranked from the highest to lowest abundance (top to bottom). These were plotted for infants according to HIV exposure status over several time points. Stacked bar plots were used to visualize the relative abundance of microbes at a given taxonomic level in each sample, represented as a single bar, labeled by time point, and plotted within each HIV exposure status group for separate mothers and infant comparisons. These plotting and diagramming techniques allow for an efficient overview of the types of differences inherently present in the data at the group level.

Generalized estimating equations (GEEs) as described in Liang and Zeger (1986) 25 and extended by Agresti (2002) 26 have been widely used for modeling longitudinal data, 27 and more recently for longitudinal microbiome data. 28, 29 For each genus present in the microbial aggregate of samples, we modeled normalized log CPM relative taxon counts, estimating the effects of time point and HIV exposure status and their interaction, while accounting for the underlying structure of clusters formed by individual subjects. We defined the responses Y 1, Y 2,..., Y n as the collection of infant relative abundances for a given taxon in log CPM, with n = 129. We identified the mean model µ ij for the i th infant and j th timepoint. With regression parameters β k representing HIV exposure status and time point, and the AR(1) variance structure V i , we formed the estimating equation:

U ( β ) = ∑ i = 1 n ∂ μ i ∂ β V i − 1 { Y i − μ i ( β ) } .

This then becomes an optimization problem, such that solving U( β) = 0 estimates the parameters β k . We modeled abundances for all 12 genera and nine of the top species. The link function g was chosen to be a Gaussian link. It was assumed that these abundances are correlated within infants for the various sampling time points. As such, we accounted for this with a first-order autoregressive AR(1) working correlation structure with homogenous variances such that the correlation between adjacent time points was assumed to be more similar. Parameter estimates were collected from each model, along with Wald test p-values. Multiple testing was adjusted for using a Bonferroni correction. Models were created in R using geepack v1.3.10. 30

Hotelling’s T 2 tests 31 were used to determine whether the microbiome profiles exhibited notable differences or trends across time and groups. We then used t-tests to identify which genera contributed most to these differences. Groups of mothers and infants or HEU and HUU infants are designated as the two sampling units on which the relative abundances of the p most abundant genera were measured. For paired tests, we chose p = 6 variables to ensure that n < p so that singularity could be avoided and T 2 could be properly computed, where n is the number of measurements in a sampling unit. We generally tested the hypotheses

H 0 : μ y = μ x vs. H A : μ y ≠ μ x

which, in the paired case given μ diff = μ y – μ x , is equivalent to

H 0 : μ d i f f = 0 vs. H A : μ d i f f ≠ 0

to conduct a comparison between groups for the p most abundant genera for testing groups. The population means μ x and μ y represent the two sampling units of a given test, which were generally either μ mothers and μ infants or μ HUU and μ HEU . We treated samples as paired in both cases, such that the mothers are paired with their own infants, or HEU and HUU infants were paired according to the pre-analysis matching carried out previously. When conducting testing on data from HIV positive mothers and HIV negative mothers, we did not pair mothers and instead relied on a standard two-sample test. We assumed that the relevant conditions for testing are met, meaning that the groups are correlated and have a multivariate normal distribution. Normality is met by using microbe abundances in log CPM units.

We conducted three main groups of testing with the T 2 statistic. The first paired test was between mother-infant pairings at the first or last time point ( t ∈ [0, 6]) and for either the mother/infant HIV(+)/HEU or HIV(-)/HUU groups for a total of four tests. These tests offer a clear between-group comparison while accounting for similarity in the mother/infant dyads and calculating genera-specific differences in ruling whether the hypotheses are met. The second testing approach consisted of seven tests. We compared relative abundances of the six most highly abundant genera at each of the seven time points ( t ∈ [0, 1,..., 6]) from paired HEU and HUU infants. The final testing approach compared all HIV(+) and HIV(-) mothers’ samples for all 12 genera across mothers, including the “Other” designation as described previously. All samples were unpaired and therefore utilized the unpaired multivariate T 2 generalization.

Test statistics were calculated to test whether microbiome profiles in paired infants were notably different among the seven time points, and whether HEU mothers and infants had similar profiles at the first and last time points. After calculating the T 2 test statistic, we calculate an equivalent F p,n–p test statistic distributed with p and n – p degrees of freedom and use this distribution to calculate the test’s p-value:

p-value = Pr { F p , n − p + 1 > n − p + 1 n p T 2 }

Paired or two-sample t-tests were used to distinguish which genera are most important to the differences identified with Hotelling’s T 2 tests, and α-levels were adjusted after performing the p tests by using a Bonferroni critical value.

Beta diversity using the Bray-Curtis dissimilarity metric was compared using a non-parametric Wilcoxon rank-sum test between HEU and HUU infants, and between HIV(+) and HIV(-) mothers. Tests were conducted separately for each subset of samples for the first and last time points for a total of four tests. The null hypothesis conjectures that the distributions of both populations are equal, under the general assumption that all observations from both groups are independent of each other.

 Results Sample characteristics

Counts of infant and mother samples available at each time point and for each group of interest can be found in Table 2. Although we included 10 HUU and 10 HEU infants and their mothers in the study, there was some small variation in numbers of samples at the different time points, but they were overall close in evenness. Infants were approximately 7 ± 1 (mean ± SD) days old at time of first swabbing, and were on average 22 ± 2 days, 42 ± 3 days, 58 ± 2 days, 74 ± 2 days, 88 ± 2 days, and 105 ± 2 days respectively at subsequent time points 1–6.

 Longitudinal differences in the nasopharyngeal microbiome composition

Our FastQC analysis indicated that the overall sequencing quality was excellent, with mean Phred quality scores remaining greater than 25 (99.5% accuracy) for at least 175 bp for both forward and reverse reads. Trimmomatic removed less than 3% of reads in any given sample. The analysis covered 129 infant and 38 mother swab samples, with an average of 124,905 ± 299,120 (mean ± SE) reads per infant sample (max = 3,016,276; min = 13,218) and 70,463 ± 48,730 reads per mother sample (max = 208,029; min = 14,339). The number of reads in infants’ samples was significantly higher than that of mothers’ swabs (p = 0.02), which may in part be the result of mothers’ acquired immunity over time and therefore lower overall NP carriage. In our raw data, we uniquely identified 17 phyla, 647 genera, and 758 total species across all samples. NP microbiome composition separated according to HIV exposure status are depicted in Figures 1A and 1B for infants and mothers, respectively. The alluvial plot in Figure 1C graphically depicts the most abundant genera in log CPM present in HEU and HUU infants. As relative abundances (in log CPM) change over time, the position of a flow stream representing a single genus may change position relative to the other genera. At a given time point, the flow streams are stacked according to the abundance relative to other streams.

The maturation over 120 days of the NP microbiomes of A) healthy HIV-exposed, uninfected (HEU; n=10) and HIV-unexposed, uninfected (HUU; n=10) infants and B) HIV(+) (n=10) and HIV(-) mothers (n=10). These stacked bar plots reveal variation in the relative abundance of microbes between groups of either infants or mothers clustered at the genus level. Each bar represents a single time point binned by age and is the average of ~10 samples. Genera with an average relative abundance of <1% across all samples are labeled as “Other.” The alluvial plot in C) depicts the changing presence of genera across all infant samples by HIV exposure status. As relative abundances change over time, the position of a flow stream representing a single genus may change position relative to the other genera.

Due to the heavy presence of taxa with relatively low abundance, we labeled taxa with average relative abundances of less than 1% as “Other” at the genera and species level. Post-grouping, the present taxa were limited to three distinctly identifiable phyla, encompassing 12 genera and 87 species. The most abundant phyla were the Firmicutes (~56.5%), Proteobacteria (~22.4%), and Actinobacteria (~16.1%). The remaining 5% of reads were characterized as “Other.” At the genus level, the most abundant groups were the Dolosigranulum (~23.5%), Staphylococcus (17.2%), Corynebacterium (~16.1%), Streptococcus (~13.5%) and Moraxella (~12.4%) genera.

For both the HEU and HUU infant groups, the respiratory microbiome during the first months of life is dominated by Staphylococci and Corynebacteria. Early on, we observed the emergence of more typical respiratory bacteria such as Moraxella and Streptococcus sp. Additionally, the commensal Dolosigranulum emerges as a dominant member of the microbiome within the first weeks of life. We saw the appearance of Haemophilus somewhat later at around four to six weeks. Overall, these transitions occurred in an orderly and stepwise pattern. These transitional patterns align with those found in a longitudinal East Asian infant cohort. 32

While longitudinal trends were strongly apparent across infant groups, the observed differences between the HEU and HUU infants were subtle. Figure 1A illustrates higher amounts of Streptococcus and suppression of Dolosigranulum among HEU vs. HUU infants. Furthermore, Figure 1C suggests increasing amounts of Streptococcus and Haemophilus over time for HEU infants, whereas Staphylococcus largely declines after time point three in both groups. Together these findings support a time-dependent effect on genus abundance taking place soon after birth.

As this is one of the few studies conducted on the NP microbiome in Zambia, there are no conclusive baseline expectations for a typical microbiome composition. However, we have compiled the results of a published longitudinal study that analyzed NP swabs from a cohort of 234 healthy infants from Washington, D.C. 33 ( Table 3). Swabs in the study were taken at ~two, ~six, and ~12 months. We have included the proportion of genera present across the aggregate of all swabs, and for the ~two-month sampling (no further information on exact timing of swabs was specified). Slight differences appear to be present, which may be in part due to differences in living conditions, sample size, and choice of analysis database used. The Teo et al. study utilized the Greengenes database, which produces far less sensitive results when compared to other databases like RefSeq and Silva. 34 Otherwise, when comparing the Teo et al. cohort at two months to those in our study, we find that the abundance of Staphylococcus is nearly twice as high as in our study for either HEU or HUU infants, although the Corynebacterium, Moraxella, Streptococcus average means are all within a few percentage points for both groups. In our analysis, the most overwhelming difference between the HEU and HUU infants is in Dolosigranulum, for which HUU infants showed a 19.63% higher presence of Dolosigranulum when averaged across time points. Overall, as this study is not a one-to-one comparison to the Teo et al. study, it is difficult to draw exact conclusions, but it appears that the NP microbiome profiles observed here are similar to those seen in healthy infants of a similar age.

Comparison of average genera relative abundance with a NP longitudinal study of a healthy infant cohort from Washington, D.C.

The number of samples, age group by month (m), sequenced 16S region, patient condition and genus mean relative proportions are enumerated. While the studies are not a one-to-one comparison, the relative abundances between studies appear to be similar. NR denotes statistic not reported in original paper.

Study Teo et al. (2015) 33 Teo et al. (2015) 33 This study This study
Number of samples 1,021 ~177 68 61
Age group 2–12 m 2 m only 0–3.5 m 0–3.5 m
16S Region V4 V4 V3–V4 V3–V4
Patient condition Healthy Healthy Healthy HEU
Microbe  
Dolosigranulum 8.80% 14% 32.04% 12.41%
Streptococcus 14% 14% 8.93% 14.54%
Moraxella 31.20% 9% 10.65% 15.65%
Staphylococcus 10.30% 41% 19.71% 14.54%
Corynebacterium 13.50% 22% 17.92% 14.93%
Haemophilus 9.70% NR 4.91% 9.21%
Anaerobacillus NR NR 0.94% 2.02%
Paracoccus NR NR 1.20% 0.57%
Acinetobacter 13% NR 1.10% 0.70%
Pseudomonas NR NR 0.29% 3.55%
Alkahilophilus NR NR 0.15% 0.67%
Other NR NR 2.16% 5.64%
Relationship between nasopharyngeal microbiome composition and HIV exposure

GEEs revealed some differences among genera and species for time point and HIV exposure status when adjusting for subject variation and the interaction effect. Models were created for all 12 genera (including the “Other” grouping) and for the nine species that averaged greater than 1% abundance across all infant samples. Of all taxa tested, four of nine species and six of 12 genera exhibited substantial instability across time points ( Table 4). One species, Staphylococcus haemolyticus, was highly associated with HIV exposure in infants (p <0.01, adj. p = 0.01). Additionally, Streptococcus mitis indicated a strong interaction effect between time and HIV exposure status (p <0.01, adj. p = 0.04). Figure 2 shows the estimated marginal means of the time and HIV-exposure status effects for S. haemolyticus, S. mitis, Haemophilus influenzae, and the Dolosigranulum genus. We chose to include these microbes in the figure as a representative selection of microbes with significant and non-significant effects. In Figure 2a, differences in the abundance of S. haemolyticus for HEU and HUU infants is cleanly pronounced for several time points, reflecting the Wald test result, but this is not the case for the other microbes in Figures 2b, 2c, or 2d. All of these had at least marginally significant p-values when testing the time point effect but lacked strong HIV exposure status effects.

Table of P-values from Wald tests on GEE models.

The effects modeled were HIV status, and time point effects, with their interaction. Results are stratified by species and genus. All p-values less than alpha=0.05 are bolded.

Time point Unadjusted p-values Time point Coefficient Estimate Time point Adjusted p-values HIV Status Unadjusted p-values HIV Status Coefficient Estimate HIV Status Adjusted p-values Interaction Unadjusted p-values Interaction Coefficient Estimate Interaction Adjusted p-values
Genus  
Acinetobacter 0.00 -0.09 0.02 0.78 -0.06 1.00 0.29 0.04 1.00
Alkalihalophilus 0.01 0.18 0.07 0.15 0.67 1.00 0.03 -0.26 0.30
Anaerobacillus 0.19 -0.11 1.00 0.12 0.77 1.00 0.56 -0.06 1.00
Corynebacterium 0.01 -0.18 0.09 0.11 -0.39 1.00 0.18 0.12 1.00
Dolosigranulum 0.01 0.25 0.13 0.35 -0.46 1.00 0.46 0.09 1.00
Haemophilus 0.00 0.43 0.00 0.14 0.69 1.00 0.23 -0.18 1.00
Moraxella 0.00 0.35 0.00 0.21 0.54 1.00 0.07 -0.14 0.89
Other 0.03 -0.09 0.30 0.81 -0.04 1.00 0.25 0.06 1.00
Paracoccus 0.44 -0.04 1.00 0.08 -0.50 0.99 0.09 0.12 1.00
Pseudomonas 0.00 -0.15 0.00 0.19 -0.52 1.00 0.03 0.17 0.40
Staphylococcus 0.00 -0.38 0.00 0.53 -0.13 1.00 0.35 0.04 1.00
Streptococcus 0.00 0.13 0.01 0.43 -0.18 1.00 0.12 0.09 1.00
Species  
Corynebacterium accolens 0.52 -0.27 1.00 0.26 -2.34 1.00 0.18 0.70 1.00
Corynebacterium pseudodiphtheriticum 0.58 0.25 1.00 0.30 -1.91 1.00 0.34 0.51 1.00
Dolosigranulum pigrum 0.01 0.90 0.06 0.26 -1.82 1.00 0.42 0.33 1.00
Haemophilus influenzae 0.00 1.33 0.00 0.11 1.76 0.96 0.30 -0.44 1.00
Staphylococcus epidermidis 0.03 -0.57 0.24 0.97 -0.07 1.00 0.56 -0.21 1.00
Staphylococcus haemolyticus 0.01 -0.45 0.07 0.00 4.04 0.01 0.07 -0.62 0.59
Staphylococcus simiae 0.00 -1.27 0.01 0.11 -4.49 1.00 0.08 0.88 0.69
Streptococcus mitis 0.00 -1.24 0.00 0.06 -2.60 0.53 0.00 0.65 0.04
Streptococcus pneumoniae 0.00 1.81 0.00 0.16 1.85 1.00 0.52 -0.20 1.00
Plots of estimated marginal means of the abundance of a given microbe accounting for estimated GEE effects.

The time of sampling is on the x-axis, and the abundance of the microbe in log CPM is on the y-axis. The estimated change in abundance over time is illustrated for control infants (in red) and HIV-exposed, uninfected (HEU) infants (in blue). The interaction between time and HIV exposure status is accounted for in the differing group slopes. All microbes had at least marginally significant time effects, but only S. haemolyticus had a very strong HIV exposure status effect.

Hotelling’s T 2 test statistics were used to compare relative abundances of the top six most abundant genera between paired HEU and HUU infants at each time point, namely Dolosigranulum, Streptococcus, Moraxella, Staphylococcus, Corynebacterium, and Haemophilus. As previously noted, we chose p = 6 variables to ensure that n < p so that singularity could be avoided and T 2 could be properly computed, where n is the number of measurements in a given sampling unit. Paired infants did not present significantly different microbiome profiles at any of the time points at α = 0.05 ( Table 5). The largest differences seemed to occur at time point 5 (p = 0.07). The fluctuating critical F values indicate that uneven sample distribution across time points contributes to the difficulty in identifying clear differences between populations.

Table of Hotelling’s T <sup>2</sup> test results.

Tests compared log CPM relative genera abundances of paired HIV-exposed, uninfected (HEU) and HIV-unexposed, uninfected (HUU) infants at all seven time points. The first and second degrees of freedom for the test are denoted by df 1 and df 2, respectively.

Time point 0 Time point 1 Time point 2 Time point 3 Time point 4 Time point 5 Time point 6
df 1 6 6 6 6 6 6 6
df 2 4 4 3 1 1 2 2
Critical F value 6.16 6.16 8.94 233.99 233.99 19.33 19.33
F statistic 0.90 1.99 0.47 3.41 22.66 14.56 1.26
p-value 0.57 0.26 0.80 0.39 0.16 0.07 0.50

Overall microbiome composition varied significantly at the times of first and last sampling for HUU and HEU infants. Among all infant samples at time point 0, a Wilcoxon rank-sum test of the Bray-Curtis dissimilarity revealed a non-significant difference in microbiome composition between HUU and HEU infants (p = 0.68). The same test for all infant samples at time point six was contrastingly showed large differences in beta diversity (p <0.01). Within-group, between-subject tests revealed variation at the first and last time points within the HEU group, and at both first and last time points within the HUU group (all p-values < 0.01). However, large variation was not present between time points within the HEU (p = 0.11) or HUU infant samples (p = 0.78). As a result of heavy inter-subject variability, alpha diversity metrics were not deemed useful for this analysis.

 Relationship between infant and mother nasopharyngeal microbiome composition

We also investigated the relationship between the NP microbiome in infants and their mothers. Paired Hotelling’s T 2 tests were again used to test the log CPM of genera as a measure of relative abundance at the first and last time points for the HIV(+)/HEU and HIV(-)/HUU groups ( Table 6). Genera tested were the same as listed for the infant-specific Hotelling’s tests, delineated previously. HIV(+)/HEU mother-infant pairs had marginally significant profile differences at the first time point ( T 6 , 4 2 ; p = 0.10) but were similar enough to not present as having significant profiles at the last time point ( T 6 , 1 2 ; ; p = 0.60). HIV(-)/HUU mother-infant pairs also had marginally significant differences at the first time point ( T 6 , 4 2 ; p = 0.10) but varied notably at the last time point ( T 6 , 3 2 ; p = 0.02). Paired t-tests for the six genera were separately conducted for HIV(-)/HUU mother-infant pairs to distinguish which genera are most important to the identified difference at that time point. The largest difference in abundance occurred for the Haemophilus ( t 8, p = 0.02; adj. p-value = 0.10) and Staphylococcus genera ( t 9; p-value <0.01; adj. p-value <0.01). HUU Infants were noted as having greater Haemophilus carriage than their HIV(-) mothers, whereas mothers had greater Staphylococcus carriage than their infants.

Table of Hotelling’s T <sup>2</sup> test results.

Tests compared log CPM relative genera abundances at the first and last time points for the HIV(+)/HEU and HIV(-)/HUU groups using mother-infant pairs. HEU: HIV-exposed, uninfected; HUU: HIV-unexposed, uninfected.

Time point 0 Time point 1
HEU/HIV(+) HUU/HIV(-) HEU/HIV(+) HUU/HIV(-)
df 1 6 6 6 6
df 2 4 4 1 3
Critical F value 6.16 6.16 233.99 8.94
F Statistic 4.13 4.14 1.20 19.52
p-value 0.10 0.10 0.60 0.02

As part of our inquiry into why HEU infants might be more susceptible to higher morbidity or mortality, we conducted paired t-tests for well-known pathogenic species between mother-infant pairs for Streptococcus pneumoniae, Haemophilus influenzae, and Staphylococcus haemolyticus. Tests compared pairs at the first and last time points, and within either HIV or control subgroups, for a total of four different tests per species ( Table 7). Inequality in pathogen carriage was ascertained by the mean of differences in log CPM. The tests for S. pneumoniae were nominally significant at t = 0 for the HIV(+)/HEU group (p = 0.03, adj. p = 0.4) and t = 6 for the HIV(-)/HUU group (p = 0.01, adj. p = 0.1). In the former, the mothers had more pathogen carriage than the infants, but this was the opposite case for the latter. H. influenzae was more highly abundant in HUU infants than in their HIV(-) mothers at t = 6 (p = 0.01, adj. p = 0.1), and S. haemolyticus was likely to be found in the HUU infants at t = 0 in a higher concentration than in their mothers.

Results table from paired t-tests of log CPM relative abundances of well-known pathogenic species.

Tests compared mother-infant pairs at the first and last time points, and within either HIV or control subgroups, for a total of four different tests per species. Lower and upper 95% confidence interval bounds are denoted by the Lower CI and Upper CI columns. The test degrees of freedom are denoted by df.

Species HIV Status Time point p-value Lower CI Upper CI Mean of differences (log CPM) t df Adjusted p-value
Streptococcus pneumoniae HIV 0 0.03 0.26 3.80 2.03 2.60 9 0.4
6 0.16 -3.47 0.73 -1.37 -1.60 6 1.0
Control 0 0.12 -0.43 3.28 1.42 1.73 9 1.0
6 0.01 -4.86 -0.79 -2.83 -3.20 8 0.1
Haemophilus influenzae HIV 0 0.84 -1.55 1.87 0.16 0.21 9 1.0
6 0.38 -3.87 1.70 -1.08 -0.95 6 1.0
Control 0 0.88 -1.18 1.36 0.09 0.15 9 1.0
6 0.01 -2.69 -0.45 -1.57 -3.23 8 0.1
Staphylococcus haemolyticus HIV 0 0.03 -4.36 -0.27 -2.31 -2.56 9 0.4
6 0.40 -0.94 2.06 0.56 0.91 6 1.0
Control 0 0.01 -3.02 -0.61 -1.81 -3.41 9 0.1
6 0.27 -0.52 1.60 0.54 1.17 8 1.0
Trends in NP microbial community composition between HIV(+) and HIV(-) mothers

We compared HIV(+) and HIV(-) mothers at the time points that we sequenced. We used the Bray-Curtis dissimilarity metric to compare compositional dissimilarity between NP microbiomes. We identified strong dissimilarity between HIV(+) and HIV(-) mothers at time point six (p <0.01), but higher similarity at the first time point (p = 0.70). Additionally, the summed abundances of all twelve genera, including “Other” genera, were used to conduct a two-sample Hotelling’s T 2 test on unpaired mothers. The differences we observed were marginally significant at the first time point ( T 12 , 7 2 ; p = 0.16), but more noticeable at the latter ( T 12 , 5 2 ; p = 0.04). Unpaired t-tests indicated that the “Other” taxa (p = 0.03, adj. p = 0.37), Alkalihalophilus (p = 0.08, adj. p = 1), and Paracoccus (p = 0.10, adj. p=1) were the largest contributors to this difference. These taxa were all more highly abundant in the HIV(-) mothers.

 Discussion

To date, many studies have examined the effect of the microbiome of HIV(+) mothers on their infants and found noticeable differences. Bender et al. (2016) reported that although very few differences were apparent in the microbiomes of mothers with and without HIV infection, maternal HIV infection was associated with changes in the mouth, skin, and gut microbiome of HEU infants from Haiti. 35 Higher abundance of Pseudomonadaceae and Thermaceae, along with decreased bacterial diversity in stools of HEU infants was suggested as one mechanism that accounts for the immunologic derangements and poor growth observed in these children. It has been noted that the microbiome of mother and infant dyads reveals some associations with HIV infection, 36 particularly that infants’ microbiomes reflect the dysbiosis of their mothers, but how this dysbiosis is established in the HEU infant is poorly understood. Significantly higher bacterial diversities have been found in the fecal matter of HIV-exposed infants, compared to HIV-unexposed infants in an African cohort. 37 The relatively small number of studies looking specifically at the microbiome of the nasopharynx in HIV-exposed infants have found few changes associated with HIV infection, 38 even though increased risk of pneumococcal colonization and disease remains apparent for these infants. 7, 13, 14 Although dysbiosis seems to be a reasonable factor in this risk, the few available data contradict this hypothesis, demonstrating no differences in pathogen carriage between children with HIV infection and control groups. 15– 17

Our study explored differences in the NP microbiome among Zambian HEU and HUU infants and their mothers over a three-month period. Our starting point for this analysis was to better understand the observed excess mortality and increased rates of respiratory disease among HEU infants. Assuming that such differences are not merely due to sampling biases, the chief hypotheses explaining them are that 1) HEU infants have subtle immunological deficits; or 2) these are the consequence of confounding due to environmental and sociological factors. These are not mutually exclusive. We had reasoned that shifts in the microbiome between HEU and HUU infants would be harder to explain based on sociology and more likely reflective of the external environment and immunological factors.

Our analyses indicate that the HEU infants in our study exhibited subtle differences in the NP microbial composition throughout the sampling interval. Given our limited number of samples, it is within reason that these differences are a result of the sample variation. Although we cannot exclude true differences between the populations, we are left with uncertain evidence that there is an HIV exposure effect on the NP microbiome in the first 14 weeks after birth.

Initial plots of the microbial relative abundances appeared to show dynamic changes in abundances of certain genera; it is apparent from Figure 1A and 1C that Staphylococcus and Corynebacterium began as primary colonizers of the NP microbiome but were in time replaced by Dolosigranulum, Streptococcus, Moraxella and Haemophilus. The plots appear to show Dolosigranulum may have higher carriage with simultaneous lower carriage of Streptoccocus in HUU infants. From Figure 1B, we observe that HIV(+) mothers may have higher Streptococcus abundance than HIV(-) mothers. The decreased presence of “Other” genera indicates greater bacterial diversity in HIV(-) mothers’ NP microbiomes over HIV(+) mothers.

Some of these observations are supported by the findings from the GEE modeling ( Table 4). For instance, some taxa appeared to be present in larger proportions at certain time points for HEU or HUU infants, but in general, taxa seemed to follow the same general trends in both groups. Regardless of HIV exposure status, it was apparent that as the child grows, an increasing amount of respiratory flora emerges. As time went on, we found that HEU infants diverged in their microflora profiles and diversity from the HUU infants, as was shown by the Wilcoxon test at the last time point. Interestingly, this was not verifiable in the multivariate Hotelling’s T-squared test that collectively tested the top six of the most prevalent genera over time. This indicates that differences may have occurred for individual genera at a given time point yet did not result in holistic trends involving several of the top genera. The largest difference seemed to occur at time point five. It would seem reasonable to suggest that at birth, infants’ microbiome profiles are more similar than different, and that the diversification of these profiles occurs over time; however, the lack of any considerable difference at time point six disputes this argument.

When comparing HEU and HUU infants, the only microbe that appeared to have a distinct presence was that of Staphylococcus haemolyticus. The microbe indicated a distinctive high association with HIV exposure at birth, even when accounting for the interaction between HIV exposure status and time point. In the microbiome, Staphylococcus haemolyticus is not known for any pathogenic characteristics. Yet its relative taxonomic proximity to other pathogenic species of Staphylococcus suggest that a library misidentification, as is possible with any genomic reference library, 34 could result in identifying one or more other Staphylococcus species as S. haemolyticus. There is a possibility that our genomic identification process could be misidentifying what are actually S. aureus reads. S. aureus is one of the most common pathogens colonizing the nasopharynx and the lower airways, 39, 40 and carriage has been found to be significantly higher in RSV than rhinovirus-infected infants. 41 In a case series from India, NP carriage densities of Streptococcus pneumoniae and S. aureus were higher in both mothers and children living in HIV-affected households, regardless of the child's HIV status. 42

From the stacked barplots, we observed that Dolosigranulum seemed to have a higher relative proportion at all time points in HUU infants, but this was not verified as a distinguishable difference in our statistical testing. Yet, it would make sense for this trend to occur, as high Dolosigranulum carriage has been found to be correlated with postive outcomes of RSV, 43 COVID-19, 44 HIV exposure, 45 and Bronchiolitis. 46 It is generally accepted as a marker of a healthy microbiome, positively associated with Corynebacterium and potentially protective against colonization by S. aureus and S. pneumoniae. 47 Yet, it is unclear as to what Dolosigranulum’s exact role is in mediating defense against illness.

When comparing infants to their mothers with paired analyses, HEU infants’ NP microbiome composition was not vastly different from their HIV(+) mothers at birth or 14 weeks, including in their carriage of S. pneumoniae, H. influenzae, and S. haemolyticus. HUU infants were similar to their mothers at birth, but apparently grew apart from their HIV(-) mothers by 14 weeks as infants acquired more Haemophilus (specifically, Haemophilus influenzae) and decreased in Staphylococcus carriage.

As we had sequenced samples from both HIV(+) and HIV(-) women as part of this study, we were interested to see if we could ascertain any general differences from the present samples. We found that the two groups of women had greater dissimilarity at the last time point than the one immediately after birth, with HIV(-) women exhibiting a larger presence of taxa with relatively small abundances (denoted as “Other” in our analyses) revealing more microbial diversity in their NP microbiomes. In the nasopharynx, lower diversity has been associated with individuals with rhinovirus illness 48 and in children with HIV-associated bronchiectasis, 49 suggesting that greater NP diversity is a sign of a healthy microbiome, mirroring a previous finding in the gut. 50

Based on our results, we see some nuance, but must acknowledge that our study has several limitations. First, and most importantly, given this is both a pilot and a longitudinal study, our sample size is small. Second, this is a retrospective analysis of samples that were taken for the purpose of detecting Bordetella pertussis and not aimed for microbiome analysis. Nonetheless, the samples were collected in appropriate transport media and were transferred immediately on ice to the research lab. Our quality control analysis confirmed high quality of the sequencing data. Lastly, as the SAMIPS study was not focused on studying HIV transmission as a main aspect of its structure, several key variables were not measured that would have been informative in studying our results. For example, we do not know the viral loads and CD4 counts of the mothers at any point, and we lack information on whether mothers continued to take ART post-pregnancy. We are also unable to determine whether infants were infected with HIV as proper testing to determine transmission was not conducted as part of the study. On this last point, however, even if infants were HIV(+) rather than HEU, this sampling would have occurred at an early stage in disease and are unlikely to be highly immunologically deficient. Further, the direction of bias would magnify the differences between the two groups rather than diminish them, supporting the differences we saw in our analysis.

These limitations, and especially the small sample sizes, also negatively affect the power of our testing and modeling efforts. This is intended as a case study for learning about what trends may be present in this cohort that merit further research with more samples for comparison. It is also of interest as few microbiome studies are longitudinal, providing us with a new perspective on how the NP microbiome may be characterized across time in the different groups.

 Conclusions

Acknowledging these issues, our findings may suggest that that sociological differences between populations could mask more subtle inter-group nuances. This warrants further research and discussion regarding the HIV exposure phenomenon before readily affirming or denying that HIV exposure affects infants’ NP microbiomes. The potential effect of the HIV(+) mother’s microbiome on an infant may present further changes in pathogen carriage, or community diversity, as was suggested in our analysis. If associations hold, identifying HIV exposure as a predisposing factor to illness and poor health in infants would present opportunities for further research and development to support infants’ living situations, especially in areas of higher HIV transmission.

 Consent

Written informed consent for publication of the patients’ details was obtained from the parents of the patients.

 Data availability Underlying data

Zenodo: Underlying data for ‘Characterization of longitudinal nasopharyngeal microbiome patterns in maternally HIV-exposed Zambian infants’. https://doi.org/10.5281/zenodo.7255313 24

This project contains the following underlying data:

Data file 1: FinalDatOther.rds

Data file 2: FinalDatPICRUSt2.RDS

Data file 3: animalculesFinalHIV.rds

Data file 4: animalcules_data_2021.rds

Data file 5: mappingFaits.csv

Data file 6: mappingFinalHIV.csv

Data file 7: mappingFinalHIV.tsv

Data file 8: preclean_MetaData.txt

Data file 9: samips_immunization.csv

Data files 10–176: *-sam-report.tsv

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).

 Accession numbers

NCBI BioProject: Characterization of longitudinal nasopharyngeal microbiome patterns in maternally HIV-exposed Zambian infants. Accession number PRJNA874826. https://identifiers.org/NCBI/BioProject:PRJNA874826

Software availability

Source code available from: https://github.com/aubreyodom/HIV_Exposed_Infants

Archived source code at time of publication: https://doi.org/10.5281/zenodo.7255313 24

License: CC0 1.0 Public domain dedication

 Sugandhi N Rodrigues J Kim M : HIV-exposed infants: rethinking care for a lifelong condition. AIDS. 2013;27 Suppl 2(0 2):S187–95. 24361628 10.1097/QAD.0000000000000090 4089095 Nielsen-Saines K Watts DH Veloso VG : Three postpartum antiretroviral regimens to prevent intrapartum HIV infection. N Engl J Med. 2012;366(25):2368–79. 22716975 10.1056/NEJMoa1108275 3590113 Brennan AT Bonawitz R Gill CJ : A meta-analysis assessing all-cause mortality in HIV-exposed uninfected compared with HIV-unexposed uninfected infants and children. AIDS. 2016;30(15):2351–60. 27456985 10.1097/QAD.0000000000001211 Brennan AT Bonawitz R Gill CJ : A Meta-analysis Assessing Diarrhea and Pneumonia in HIV-Exposed Uninfected Compared With HIV-Unexposed Uninfected Infants and Children. J Acquir Immune Defic Syndr. 2019;82(1):1–8. 31408450 10.1097/QAI.0000000000002097 8214449 Thea DM St Louis ME Atido U : A prospective study of diarrhea and HIV-1 infection among 429 Zairian infants. N Engl J Med. 1993;329(23):1696–702. 8232458 10.1056/NEJM199312023292304 Shapiro RL Lockman S Kim S : Infant morbidity, mortality, and breast milk immunologic profiles among breast-feeding HIV-infected and HIV-uninfected women in Botswana. J Infect Dis. 2007;196(4):562–9. 17624842 10.1086/519847 von Mollendorf C von Gottberg A Tempia S : Increased risk for and mortality from invasive pneumococcal disease in HIV-exposed but uninfected infants aged <1 year in South Africa, 2009-2013. Clin Infect Dis. 2015;60(9):1346–56. 25645212 10.1093/cid/civ059 Locks LM Manji KP Kupka R : High burden of morbidity and mortality but not growth failure in infants exposed to but uninfected with human immunodeficiency virus in Tanzania. J Pediatr. 2017;180:191–199.e2. 27829511 10.1016/j.jpeds.2016.09.040 5183465 Pierre RB Fulford TA Lewis K : Infectious disease morbidity and growth among young HIV-exposed uninfected children in Jamaica. Rev Panam Salud Publica. 2016;40(6):401–409. 28718488 Lapidot R Faits T Ismail A : Nasopharyngeal Dysbiosis Precedes the Development of Lower Respiratory Tract Infections in Young Infants: a longitudinal infant cohort study. medRxiv. 2021. 10.1101/2021.10.13.21264939 Biesbroek G Bosch AA Wang X : The impact of breastfeeding on nasopharyngeal microbial communities in infants. Am J Respir Crit Care Med. 2014;190(3):298–308. 24921688 10.1164/rccm.201401-0073OC Garcia-Carbonell R Yao SJ Das S : Dysregulation of Intestinal Epithelial Cell RIPK Pathways Promotes Chronic Inflammation in the IBD Gut. Front Immunol. 2019;10:1094. 31164887 10.3389/fimmu.2019.01094 6536010 Farley JJ King JC Nair P : Invasive pneumococcal disease among infected and uninfected children of mothers with human immunodeficiency virus infection. J Pediatr. 1994;124(6):853–8. 8201466 10.1016/s0022-3476(05)83170-1 Kinabo GD van der Ven A Msuya LJ : Dynamics of nasopharyngeal bacterial colonisation in HIV-exposed young infants in Tanzania. Trop Med Int Health. 2013;18(3):286–295. 23320622 10.1111/tmi.12057 Rusen ID Fraser-Roberts L Slaney L : Nasopharyngeal pneumococcal colonization among Kenyan children: antibiotic resistance, strain types and associations with human immunodeficiency virus type 1 infection. Pediatr Infect Dis J. 1997;16(7):656–662. 9239769 10.1097/00006454-199707000-00007 Leibovitz E Dragomir C Sfartz S : Nasopharyngeal carriage of multidrug-resistant Streptococcus pneumoniae in institutionalized HIV-infected and HIV-negative children in northeastern Romania. Int J Infect Dis. 1999;3(4):211–215. 10575151 10.1016/s1201-9712(99)90027-9 Polack FP Flayhart DC Zahurak ML : Colonization by Streptococcus penumoniae in human immunodeficiency virus-infected children. Pediatr Infect Dis J. 2000;19(7):608–612. 10917217 10.1097/00006454-200007000-00005 Gill CJ Mwananyanda L MacLeod W : Incidence of Severe and Nonsevere Pertussis Among HIV-Exposed and -Unexposed Zambian Infants Through 14 Weeks of Age: Results From the Southern Africa Mother Infant Pertussis Study (SAMIPS), a Longitudinal Birth Cohort Study. Clin Infect Dis. 2016;63(suppl 4):S154–S164. 27838668 10.1093/cid/ciw526 5106616 Bispo S Chikhungu L Rollins N : Postnatal HIV transmission in breastfed infants of HIV-infected women on ART: a systematic review and meta-analysis. J Int AIDS Soc. 2017;20(1):21251. 28362072 10.7448/IAS.20.1.21251 5467610 Bolger AM Lohse M Usadel B : Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–2120. 24695404 10.1093/bioinformatics/btu170 4103590 Klindworth A Pruesse E Schweer T : Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41(1):e1. 22933715 10.1093/nar/gks808 3592464 Andrews S : FastQC: a quality control tool for high throughput sequence data.Babraham Bioinformatics, Babraham Institute, Cambridge, United Kingdom; 2010. Hong C Manimaran S Shen Y : PathoScope 2.0: a complete computational framework for strain identification in environmental or clinical sequencing samples. Microbiome. 2014;2:33. 25225611 10.1186/2049-2618-2-33 4164323 Odom-Mabey A : HIV_Exposed_Infants: Initial release (v1.0.0.). Zenodo.[Dataset] [Software] 2022. http://www.doi.org/10.5281/zenodo.7255313 Liang KY Zeger SL : Longitudinal data analysis using generalized linear models. Biometrika. 1986;73(1):13–22. 10.1093/biomet/73.1.13 Agresti A : Categorical Data Analysis.2nd Edition. Edited by: Wiley JSI Hoboken, NJ. 2002. Reference Source Gibbons RD Hedeker D DuToit S : Advances in analysis of longitudinal data. Annu Rev Clin Psychol. 2010;6:79–107. 20192796 10.1146/annurev.clinpsy.032408.153550 2971698 Luo D Liu W Chen T : A Distribution-Free Model for Longitudinal Metagenomic Count Data. Genes (Basel). 2022;13(7):1183. 35885966 10.3390/genes13071183 9316307 Sun H Huang X Huo B : Detecting sparse microbial association signals adaptively from longitudinal microbiome data based on generalized estimating equations. Brief Bioinform. 2022;23(5):bbac149. 35561307 10.1093/bib/bbac149 Højsgaard S Halekoh U Yan J : The R package geepack for generalized estimating equations. Journal of statistical software. 2006;15(2):1–11. 10.18637/jss.v015.i02 Rencher A Christensen W : Methods of Multivariate Analysis, Third Edition ed.Wiley series in probability and statistics. John Wiley & Sons, Inc.;2012. 10.1002/9781118391686 Salter SJ Turner C Watthanaworawit W : A longitudinal study of the infant nasopharyngeal microbiota: the effects of age, illness and antibiotic use in a cohort of South East Asian children. PLoS Negl Trop Dis. 2017;11(10):e0005975. 28968382 10.1371/journal.pntd.0005975 5638608 Teo SM Mok D Pham K : The infant nasopharyngeal microbiome impacts severity of lower respiratory infection and risk of asthma development. Cell Host Microbe. 2015;17(5):704–715. 25865368 10.1016/j.chom.2015.03.008 4433433 Faits T Odom-Mabey AR Castro-Nallar E : Metagenomic profiling pipelines improve taxonomic classification for 16S amplicon sequencing data. bioRxiv. 2022; 2022.07.27.501757. 10.1101/2022.07.27.501757 Bender JM Li F Martelly S : Maternal HIV infection influences the microbiome of HIV-uninfected infants. Sci Transl Med. 2016;8(349):349ra100. 27464748 10.1126/scitranslmed.aaf5103 5301310 González R Maldonado A Martín V : Breast milk and gut microbiota in African mothers and infants from an area of high HIV prevalence. PLoS One. 2013;8(11):e80299. 24303004 10.1371/journal.pone.0080299 3841168 Claassen-Weitz S Gardner-Lubbe S Nicol P : HIV-exposure, early life feeding practices and delivery mode impacts on faecal bacterial profiles in a South African birth cohort. Sci Rep. 2018;8(1):5078. 29567959 10.1038/s41598-018-22244-6 5864830 Kelly MS Plunkett C Yu Y : Non-diphtheriae Corynebacterium species are associated with decreased risk of pneumococcal colonization during infancy. ISME J. 2022;16(3):655–665. 34511605 10.1038/s41396-021-01108-4 8857224 Suárez-Arrabal MC Mella C Lopez SM : Nasopharyngeal bacterial burden and antibiotics: influence on inflammatory markers and disease severity in infants with respiratory syncytial virus bronchiolitis. J Infect. 2015;71(4):458–469. 26149186 10.1016/j.jinf.2015.06.010 Thorburn K Harigopal S Reddy V : High incidence of pulmonary bacterial co-infection in children with severe respiratory syncytial virus (RSV) bronchiolitis. Thorax. 2006;61(7):611–615. 16537670 10.1136/thx.2005.048397 2104657 Rosas-Salazar C Shilts MH Tovchigrechko A : Differences in the nasopharyngeal microbiome during acute respiratory tract infection with human rhinovirus and respiratory syncytial virus in infancy. J Infect Dis. 2016;214(12):1924–1928. 27923952 10.1093/infdis/jiw456 5142087 Khan T Das RS Bhattacharya SD : Increased household exposure to respiratory pathogens in HIV exposed uninfected children due to maternal HIV. J Family Med Prim Care. 2019;8(3):1232–1234. 31041279 10.4103/jfmpc.jfmpc_111_19 6482729 de Steenhuijsen Piters WA Heinonen S Hasrat R : Nasopharyngeal microbiota, host transcriptome, and disease severity in children with respiratory syncytial virus infection. Am J Respir Crit Care Med. 2016;194(9):1104–1115. 27135599 10.1164/rccm.201602-0220OC 5114450 Islam M Albarracin L Melnikov V : Dolosigranulum pigrum modulates immunity against SARS-CoV-2 in respiratory epithelial cells. Pathogens. 2021;10(6):634. 34064210 10.3390/pathogens10060634 8224358 Kelly MS Surette MG Smieja M : The nasopharyngeal microbiota of children with respiratory infections in Botswana. Pediatr Infect Dis J. 2017;36(9):e211–e218. 28399056 10.1097/INF.0000000000001607 5555803 Hasegawa K Mansbach JM Ajami NJ : Association of nasopharyngeal microbiota profiles with bronchiolitis severity in infants hospitalised for bronchiolitis. Eur Respir J. 2016;48(5):1329–1339. 27799386 10.1183/13993003.00152-2016 5459592 Brugger SD Eslami SM Pettigrew MM : Dolosigranulum pigrum cooperation and competition in human nasal microbiota. mSphere. 2020;5(5):e00852–20. 32907957 10.1128/mSphere.00852-20 7485692 Allen EK Koeppel AF Hendley JO : Characterization of the nasopharyngeal microbiota in health and during rhinovirus challenge. Microbiome. 2014;2:22. 25028608 10.1186/2049-2618-2-22 4098959 Bhattacharya SD Niyogi SK Bhattacharyya S : Associations between potential bacterial pathogens in the nasopharynx of HIV infected children. Indian J Pediatr. 2012;79(11):1447–1453. 22570015 10.1007/s12098-012-0762-4 Lozupone CA Stombaugh JI Gordon JI : Diversity, stability and resilience of the human gut microbiota. Nature. 2012;489(7415):220–230. 22972295 10.1038/nature11550 3577372 10.21956/gatesopenres.15340.r36374 Reviewer response for version 1 Vonasek Bryan 1 Referee Michigan State University College of Osteopathic Medicine (Ringgold ID: 43977), East Lansing, Michigan, USA

Competing interests: No competing interests were disclosed.

 12 7 2024 Copyright: © 2024 Vonasek B 2024 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

This is an interesting study that investigates an important question: why do HIV exposed infants have poor health outcomes compared to unexposed infants? The key aspects of this study that are more novel are 1) longitudinal analysis of the microbiome and 2) focus on the NP microbiome. Especially when combined with the paired microbiome data from the mothers and the nice use of controls (HUU infants), this is a good study design that can be scaled up from this exploratory/pilot study. 

As the authors do a good job of describing and acknowledging, it's hard to glean much from these data given the small sample sizes. Especially considering the wide variation within groups that is described, making comparisons between groups then becomes challenging. That said, I think it is important to  Index these findings, but I would first like to see several improvements to allow readers to better understand: 1) the results, 2) some additional key limitations, and 3) coherent interpretations of the results.

General considerations

1. The authors acknowledge the limitation of not actually knowing whether the exposed infants are uninfected. I would be extra-careful with this consideration. Don't define them as HEU, maybe instead just "HIV exposed."

2. From what I can tell, a P-value cutoff to define statistically significant differences was not defined in the Methods. Of course, <0.05 is common and seems to be used in the bolding of some of the P-values in the Results. Especially for a study like this in which there are so many potential confounders that aren't really being addressed and in which the sample size is limited, I would be really careful going too much into differences that aren't statistically significant. Phrases throughout the Results and Discussion like "marginally significant" and "nominally significant" should be de-emphasized or removed. Especially in the Discussion, focus on the significant differences that you did find (including a coherent discussion about these differences), rather than emphasizing too much about differences that weren't significant.

3. The finding of more Staph haemolyticus in the exposed babies at birth is certainly worth highlighting (even more than you already do). But it's confusing as written in the Abstract and in the text: "...even when accounting for the interaction between HIV exposure status and time of sampling." Why "account for" exposure when that is the input variable you're looking at? Why "account for" time of sampling when they are all "at birth." This can be worded more clearly. Also, know that Staph haemolyticus can certainly be pathogenic in humans, especially in immunocompromised individuals (i.e. newborns, HIV exposed or infected)--you can revamp the Discussion with this in mind. In relation to this, also worth commenting on whether the HIV(+) mothers were more likely or not to have S. haemolyticus. 

4. A big factor not addressed is whether the exposed infants and their mothers living with HIV were on Bactrim prophylaxis. If you don't know, this needs to be clearly stated as a limitation. If you do know, please incorporate as a covariate in your analyses. Bactrim would likely have a major impact on the NP microbiome.

5. The manuscript will be much more digestible for the reader describe the various analyses you are doing here in a parallel order in the Methods, Results, and Discussion.

Introduction

1. Overall, this is well written. In fact, the quality of the Introduction compared to the Discussion is so striking to me that I suspect different authors being mainly responsible for these two sections? If I'm correct, then the author that led writing of the Intro should please help polish up the Discussion!

2. "We collected 167 samples...through 14 weeks of life." >> this sentence doesn't belong in the Introduction.

Methods

1. "Characteristics of the study cohort are in Table 1." >> this goes in the Results.

2. Important to give some sense of how the 10 exposed infants were sampled for inclusion in this study from the larger study. Its says it was those entering the study from April to June--all of them from the parent study? Or some sub-set? Random selection?

3. I also think the text related to Table 2 would be more appropriate in the Results.

Results

1. Table 1 -- Remove the rows without any data (Chilenje clinic, father lives with child, all rows are trimester starting HIV)

2. First paragraph: Ages of infants at time of swabbing doesn't match what I see in Table 2. Along these lines, be consistent using either days or weeks.

3. Second paragraph: "...which may in part be the result of..." >> interpretation of data belongs in the Discussion.

4. Fig 1. "120 days" -- This is different than 14 weeks?

5. Third paragraph: Rephrase the first sentence because "heavy presence" and "relatively low" are contradictory. Maybe "diversity" instead of "presence"?

6. "These transitional patterns align with..." >> this belongs in the Discussion.

7. Paragraph starting with "While longitudinal trends..." To me, this should be the first paragraph under the existing heading "Relationship between NP microbiome comp...".

8. The paragraph and data in the table about the Teo study are rather distracting. At the very least, all this needs to be moved to the Discussion. That cohort is very different from the cohort in this study, so I suggest not allocating so much text to that--maybe just a brief comparison as you do with ref 32 (which, again, belongs in the Discussion).

9. Fig 2 -- Define the widths of the lines plotted in the figure description. Double check, and consider rephrasing, your description of the slopes: I'm not sure what "accounted for" means.

10. "As previously noted, we chose p=6..." >> No need to reiterate.

11. "...difficulty in identifying clear differences between populations." >> rephrase (even remove!) as this reads as if you're assuming that there are differences.

12. "Overall microbiome composition varied significantly..." >> Isn't this redundant with what is written at the beginning of the Results? 

13. The paragraph describing beta-diversity needs to be revamped and written more clearly.

14. In the tables and text, just use one of these terms and be consistent throughout: "HEU/HIV(+), etc." vs "HIV(+)/HEU" -- flipping the order around confuses the reader.

15. The sentences about Strep pneumo differences are confusing. Reword to mirror how H. influenzae is described--though in that sentence, write "more abundant" not "more highly abundant." Write your sentence re S. haemolyticus in the same way.

16. Also be consistent with how you write about the time points: "t=6" vs "time point six" vs "last time point" >> just pick one way and stick with it.

17. Table 7 -- Reorder the columns to a more conventional order: Mean, 95% CIs, P-value (left to right). Consider removing 't' and 'df' columns. In the table heading, mention what variables you are adjusting for to get the "adjusted p-value." Why are you now using "HIV" and "Control" to define the groups? I suggest naming the groups the same as for Table 6, unless you have a good reason.

Discussion

Overall, this needs a lot of improvement. 

1. The first paragraph should summarize the keys findings from YOUR data.

2. In the second paragraph, maybe just clearly state microbiome as a third possible hypotheses and acknowledge that all three drivers are likely inter-related and hard to differentiate.

3. Be clear when describing differences in the NP microbiome in the exposed infants 1) longitudinally vs 2) compared to un-exposed infants vs 3) c/t mothers. This is an multiple areas. I can certainly infer your meaning, but it would be easier to read if the comparison is clearly reiterated.

4. The connection between description of Dolosigranulum and the cited studies is not clear. 

5. Rather than describing the diversity of the microbiome for the mothers as having lots of "other" organisms, wouldn't it be more valid to look formally at beta and alpha diversity?

6. You argue pretty clearly that you have good samples. I don't think you need to state "retrospective analysis of samples" as a limitation. 

7. I don't agree with the statements about minimizing the limitation of lack of HIV testing. In this age group, HIV infections can certainly be rapidly progressive. Also, because the underlying causes of HEU having poor health c/t HUU aren't well understood in the first place, one can't assume the effects ("direction of bias") of contaminating the HEU cohort with infected infants.   

8. "This is intended as a case study..." >> Use the term "exploratory" or "pilot" study instead because "case study" has a different, specific meaning.

Conclusions

Focus here on what can be gleaned from your study results. You did not evaluate "sociological differences."

Consent

Because you also report data from the mothers, I hope you also consented them for that (not just the infant's data)?

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

No

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Pediatric infectious diseases

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Odom Aubrey Boston University School of Medicine, Boston, Massachusetts, USA

Competing interests: None to report.

 13 8 2024

Reviewer 2

Comment 2-1: This is an interesting study that investigates an important question: why do HIV exposed infants have poor health outcomes compared to unexposed infants? The key aspects of this study that are more novel are 1) longitudinal analysis of the microbiome and 2) focus on the NP microbiome. Especially when combined with the paired microbiome data from the mothers and the nice use of controls (HUU infants), this is a good study design that can be scaled up from this exploratory/pilot study. 

As the authors do a good job of describing and acknowledging, it's hard to glean much from these data given the small sample sizes. Especially considering the wide variation within groups that is described, making comparisons between groups then becomes challenging. That said, I think it is important to  Index these findings, but I would first like to see several improvements to allow readers to better understand: 1) the results, 2) some additional key limitations, and 3) coherent interpretations of the results.

Response 2-1: We thank the reviewer for providing thorough, thoughtful feedback for this paper.

Comment 2-2: The authors acknowledge the limitation of not actually knowing whether the exposed infants are uninfected. I would be extra-careful with this consideration. Don't define them as HEU, maybe instead just "HIV exposed."

Response 2-2: We thank the reviewer for this feedback. We have changed all instances of “HEU” throughout the manuscript to “HIV-exposed”.

Comment 2-3: From what I can tell, a P-value cutoff to define statistically significant differences was not defined in the Methods. Of course, <0.05 is common and seems to be used in the bolding of some of the P-values in the Results. Especially for a study like this in which there are so many potential confounders that aren't really being addressed and in which the sample size is limited, I would be really careful going too much into differences that aren't statistically significant. Phrases throughout the Results and Discussion like "marginally significant" and "nominally significant" should be de-emphasized or removed. Especially in the Discussion, focus on the significant differences that you did find (including a coherent discussion about these differences), rather than emphasizing too much about differences that weren't significant.

Response 2-3: We have clarified in the methods that α=0.05 (see line 216). Where relevant, a Bonferroni correction is applied to adjust the alpha value (type I error rate) to account for Type I error rate inflation as noted for the GEE and the Hotelling’s T-squared test methods descriptions, and elsewhere in the manuscript (e.g. the table descriptions).

Our rationale for including “marginally significant” results is that Bonferroni corrections are known to be at times overly conservative in deflating Type I error, and thus adjusted p-values that are around the newly calculated alpha value can still be practically interesting. To make this more clear for the reader, we added the following sentence to the methods section:

“Due to the conservative nature of the Bonferroni correction, adjusted p-values between 0.05 and 0.10 were noted as marginally significant as effects that could retain some level of practical significance.”

We removed a few results that were greater than the cutoff of 0.10 and altered the discussion accordingly.

Comment 2-4: The finding of more Staph haemolyticus in the exposed babies at birth is certainly worth highlighting (even more than you already do). But it's confusing as written in the Abstract and in the text: "...even when accounting for the interaction between HIV exposure status and time of sampling." Why "account for" exposure when that is the input variable you're looking at? Why "account for" time of sampling when they are all "at birth." This can be worded more clearly.

Also, know that Staph haemolyticus can certainly be pathogenic in humans, especially in immunocompromised individuals (i.e. newborns, HIV exposed or infected)--you can revamp the Discussion with this in mind. In relation to this, also worth commenting on whether the HIV(+) mothers were more likely or not to have S. haemolyticus.

Response 2-4: We can see how this wording can be confusing to the reader. The inclusion of a variable in a statistical model is by definition to acknowledge or account for the quantifiable effect of that variable on the response. The wording in the abstract is meant to convey that concept. We can understand how this might not be clear to readers lacking a statistical background, and so we rewrote these points:

[Discussion] The microbe indicated a distinctive high association with HIV exposure at birth and across time points, a finding confirmed by separate statistical tests. (lines 452-453)

[Abstract] When evaluating the interaction between HIV exposure status and time of sampling among infants, the microbe Staphylococcus haemolyticus showed a distinctive high association with HIV exposure at birth. (lines 29-31)

Comment 2-5: A big factor not addressed is whether the exposed infants and their mothers living with HIV were on Bactrim prophylaxis. If you don't know, this needs to be clearly stated as a limitation. If you do know, please incorporate as a covariate in your analyses. Bactrim would likely have a major impact on the NP microbiome.

Response 2-5: We thank the reviewer for this observation. Unfortunately, the SAMIPS study did not track Bactrim Pneumocystis jirovecii Pneumonia (PJP) prophylaxis, only acute use of antibiotics in response to infection. This cohort required that infants were healthy as a criterion of limitation, and so antibiotic use in response to infection was not included as a covariate.

We have acknowledged this limitation in the discussion section (lines 494-496).

Comment 2-6: The manuscript will be much more digestible for the reader describe the various analyses you are doing here in a parallel order in the Methods, Results, and Discussion.

Response 2-6: We agree that the current delineation of analyses is difficult to follow, and have implemented the following changes throughout the Methods, Results, and Discussion, with an emphasis on reorganizing text under appropriately titled, standardized subheadings.

Changes to subheadings themselves were made as follows.

In the methods section, we replaced the “Statistical analysis” subheading with ones that referenced the type of analysis conducted. The wording was reflected in the Results and Discussion subsections. These are: “Visualizing longitudinal trends in microbiome composition”, “Modeling effects of HIV status on infant microbiome composition across time”, and “Testing differences in microbiome composition across time and cohort groups”. These subheadings are broad enough to allow all analyses to be grouped together more concisely by method.

In the Results section, we made the following changes to the previous subheadings:

“Longitudinal differences in the nasopharyngeal microbiome composition” was rewritten as “Visualizing longitudinal trends in NP microbiome composition”

“Relationship between nasopharyngeal microbiome composition and HIV exposure” was rewritten as “Effects of HIV status on infant NP microbiome composition across time”

“Relationship between infant and mother nasopharyngeal microbiome composition across time” was rewritten as “Associations within mother and infant duad NP microbiome composition over time”

“Trends in NP microbial community composition between HIV(+) and HIV(-) mothers” was rewritten as “Effect of HIV status on mother NP microbial community composition”

In the Discussion, we added subheaders that were nearly identical to the Results subheaders.

 Introduction

Comment 2-7: Overall, this is well written. In fact, the quality of the Introduction compared to the Discussion is so striking to me that I suspect different authors being mainly responsible for these two sections? If I'm correct, then the author that led writing of the Intro should please help polish up the Discussion!

Response 2-7: We have updated the wording in the Discussion based on suggestions provided by each reviewer, including the order of results discussed, the relocation of two paragraphs discussing previous findings in the literature, and the addition of literature examining S. haemolyticus.

Comment 2-8: "We collected 167 samples...through 14 weeks of life." >> this sentence doesn't belong in the Introduction.

Response 2-8: We have rewritten the relevant sentences of the introduction to remove specific mentions of sample numbers or exact time of analyses to focus on the context of the study (i.e., the locale, the infant groups, and the NP microbiome sampling). This now reads as follows:

“Given the complex dynamics of interactions between the host, microbes, and environment beginning at birth, we conducted an exploratory longitudinal comparison of the nasopharyngeal (NP) microbiomes of HEU and HUU infants and their mothers during the first 14 weeks of life. We reasoned that by characterizing quantifiable differences in NP microbiota distinguishing these two groups, this pilot study could indicate differences encouraging future immunology-focused studies” (lines 81-83).

Methods

Comment 2-9: "Characteristics of the study cohort are in Table 1." >> this goes in the Results.

Response 2-9: We have moved Table 1 and some accompanying text to the Results section (line 267).

Comment 2-10: Important to give some sense of how the 10 exposed infants were sampled for inclusion in this study from the larger study. Its says it was those entering the study from April to June--all of them from the parent study? Or some sub-set? Random selection?

Response 2-10: We added the following clarification to the first paragraph of the Methods section (see lines 89-94):

From 1,981 total mother infant pairs, we selected a subset that had 3 or more study visits, had no siblings under the age of five years, and who enrolled in the study from April to July 2015.  From this subset, we randomly selected 10 mother infant pairs with an HIV positive mother who started ARV treatment before pregnancy.  These 10 mother infant pairs were randomly matched to 10 mother infant pairs with an HIV negative mother by education, month of entry into the study, and maternal age.

Comment 2-11: I also think the text related to Table 2 would be more appropriate in the Results.

Response 2-11: We have moved Table 2 and some accompanying text to the Results section (line 268-273).

Results

Comment 2-12: First paragraph: Ages of infants at time of swabbing doesn't match what I see in Table 2. Along these lines, be consistent using either days or weeks.

Response 2-12: We note the discrepancy and have removed the delineation of days in the first column of Table 2. We updated Table 1 to reflect weeks at enrollment rather than days. We changed the delineation of days at swabbing time points to weeks.

Comment 2-13: Table 1 -- Remove the rows without any data (Chilenje clinic, father lives with child, all rows are trimester starting HIV)

Response 2-13: These rows have been removed and Table 1 has been restructured to better delineate parameter sections/groups.

Comment 2-14: Second paragraph: "...which may in part be the result of..." >> interpretation of data belongs in the Discussion.

Response 2-14: The paragraph on the Teo et al. study has been moved to the discussion section.

Comment 2-15: Fig 1. "120 days" -- This is different than 14 weeks?

Response 2-15: We believe this was written in error. We have changed this caption to read “14 weeks”.

Comment 2-16: Third paragraph: Rephrase the first sentence because "heavy presence" and "relatively low" are contradictory. Maybe "diversity" instead of "presence"?

Response 2-16: We have altered this sentence to read “Due to the high diversity of taxa with relatively low abundance among individual microbes…” (line 288-289).

Comment 2-17: "These transitional patterns align with..." >> this belongs in the Discussion.

Response 2-17: This sentence has been relocated to the discussion section (lines 423-433).

Comment 2-18: Paragraph starting with "While longitudinal trends..." To me, this should be the first paragraph under the existing heading "Relationship between NP microbiome comp...".

Response 2-18: We agree with the reviewer and have moved that paragraph to the heading titled “Relationship between nasopharyngeal microbiome composition and HIV exposure.” (lines 312-316)

Comment 2-19: The paragraph and data in the table about the Teo study are rather distracting. At the very least, all this needs to be moved to the Discussion. That cohort is very different from the cohort in this study, so I suggest not allocating so much text to that--maybe just a brief comparison as you do with ref 32 (which, again, belongs in the Discussion).

Response 2-19: We relocated the Teo et al. study to the discussion and renumbered the tables accordingly (lines 423-433). The text discussing the study has been shortened to remove some excess discussion of differences while still introducing the cohort and thus retaining the table’s relevance.

Comment 2-20: Fig 2 -- Define the widths of the lines plotted in the figure description. Double check, and consider rephrasing, your description of the slopes: I'm not sure what "accounted for" means.

Response 2-20: We agree with the reviewer that the explanation of the figure according to the caption is difficult to parse, and so we have rewritten the caption for clarity in delineating each visual aspect of the plot grid. We removed the “accounted for” phrasing.

“Plots of estimated marginal means of the abundance of a given microbe, given the estimated GEE effects of time point of sampling (x axis) and HIV status (line color). A line illustrates the estimated change in microbe abundance over time, with positive or negative slopes illustrating increased or decreased estimated abundance in log CPM (respectively). HIV-exposed infants (denoted as HEU) are depicted in blue, and HIV-unexposed, uninfected infants (HUU) are depicted in red. Given the separate lines for each status and changing slope across time, these plots depict the interaction effect between time point and HIV status. All microbes had at least marginally significant time effects, but only S. haemolyticus had a very strong HIV exposure status effect.”

Comment 2-21: "As previously noted, we chose p=6..." >> No need to reiterate.

Response 2-21: We have removed this sentence from the Results section.

Comment 2-22: "...difficulty in identifying clear differences between populations." >> rephrase (even remove!) as this reads as if you're assuming that there are differences.

Response 2-22: We agree with the reviewer and acknowledge that our point may have been difficult to grasp especially for audiences without familiarity with the test in question. We have removed the sentence from the paragraph to avoid confusion.

Comment 2-23: "Overall microbiome composition varied significantly..." >> Isn't this redundant with what is written at the beginning of the Results? 

Response 2-23: We have reread the Results and it is unclear what section the reviewer is referring to as redundant. We have retained this section as is.

Comment 2-24: The paragraph describing beta-diversity needs to be revamped and written more clearly.

Response 2-24: We rewrote the beta diversity paragraph describing infant results as follows:

“We examined variations in overall microbiome composition by observing differences in beta diversity between HUU and HIV-exposed infants. We conducted two Wilcoxon rank-sum tests of the Bray-Curtis dissimilarity at t=0 and t=6. We found a non-significant difference in microbiome composition between HUU and HIV-exposed infants at t=0 (p = 0.68), but large differences at t=6 (p <0.01). Performing additional tests within the HIV-exposed and HUU groups themselves, we also identified heavy inter-subject variability at both t=0 and t=6 (all p-values < 0.01). As a result of this inter-subject variability, alpha diversity metrics were not performed for this analysis.” (lines 339-345)

Comment 2-25: In the tables and text, just use one of these terms and be consistent throughout: "HEU/HIV(+), etc." vs "HIV(+)/HEU" -- flipping the order around confuses the reader.

Response 2-25: We have replaced all instances of status groups with HEU/HIV(+) and HUU/HIV(-), and replaced “mother-infant” with “infant-mother” pairs to match the ordering of the descriptors.

Comment 2-26: The sentences about Strep pneumo differences are confusing. Reword to mirror how H. influenzae is described--though in that sentence, write "more abundant" not "more highly abundant." Write your sentence re S. haemolyticus in the same way.

Response 2-26: We rewrote the test results as follows (lines 369-372):

“H. influenzae was more abundant in HUU infants than in their HIV(-) mothers at t = 6 (adj. p = 0.1), and S. haemolyticus was likely to be found in the HUU infants at t = 0 in a higher concentration than in their HIV(-) mothers (adj. p = 0.1). S. pneumoniae was more likely to be found in HUU infants than their HIV(-) mothers at t = 6 (adj. p = 0.1). Each of these findings were marginally significant.”

Comment 2-27: Also be consistent with how you write about the time points: "t=6" vs "time point six" vs "last time point" >> just pick one way and stick with it.

Response 2-27: We thank the reviewer for this feedback. We decided to change all instances of alternate references to time points to variable notation (e.g., t=0 or t=6).

Comment 2-28: Table 7 -- Reorder the columns to a more conventional order: Mean, 95% CIs, P-value (left to right). Consider removing 't' and 'df' columns. In the table heading, mention what variables you are adjusting for to get the "adjusted p-value." Why are you now using "HIV" and "Control" to define the groups? I suggest naming the groups the same as for Table 6, unless you have a good reason.

Response 2-28: We have reordered the columns to reflect the suggested order and removed the t statistic and degrees of freedom columns to reduce table information. We also relabeled “HIV” and “Control” to HEU/HIV(+) and HUU/HIV(-), which reflects the Table 6 nomenclature. We noted in the table caption that a Bonferroni correction was used to obtain the adjusted p-values.

Table 7 was renamed as Table 6.

Discussion

Comment 2-29: The first paragraph should summarize the keys findings from YOUR data.

Response 2-29: We moved the paragraph discussing previous findings in the literature to after the initial paragraph (line 400). We combined the first two paragraphs to now compose the first paragraph of the discussion, which gives an overview of our findings (that there is nuance).

Comment 2-30: In the second paragraph, maybe just clearly state microbiome as a third possible hypotheses and acknowledge that all three drivers are likely inter-related and hard to differentiate.

Response 2-30: We thank the reviewer for this feedback. Given that the hypotheses we proposed were related to microbiota differences as potential upstream or downstream indicators, we have adjusted our wording to read as follows:

“Assuming that such differences are not merely due to sampling biases, the chief hypotheses explaining them are that 1) HEU infants have subtle immunological deficits, or 2) these are the consequence of confounding due to environmental and sociological factors. These are not mutually exclusive, and we acknowledge that all drivers are likely inter-related and difficult to differentiate. We reasoned that they may be in turn associated with differences in the microbiota which could then serve as a quantifiable indicator of differences in respiratory health between HEU and HUU infants.” (lines 392-394)

Comment 2-31: Be clear when describing differences in the NP microbiome in the exposed infants 1) longitudinally vs 2) compared to un-exposed infants vs 3) c/t mothers. This is an multiple areas. I can certainly infer your meaning, but it would be easier to read if the comparison is clearly reiterated.

Response 2-31: We adjusted unclear wording and added headers throughout the discussion to properly delineate the various analysis types (see Response 2-6).

Comment 2-32: The connection between description of Dolosigranulum and the cited studies is not clear. 

Response 2-32: We have rewritten the paragraph discussing Dolosigranulum in the results section (lines 463-469) to better describe Dolosigranulum pigrum in the literature but refrain from making conclusions based on the data given the lack of statistically significant results.

Comment 2-33: Rather than describing the diversity of the microbiome for the mothers as having lots of "other" organisms, wouldn't it be more valid to look formally at beta and alpha diversity?

Response 2-33: It appears that the interpretation of beta diversity results was only implied at the time of writing, and so we have rewritten this paragraph to explain the results more clearly:

“One of our study objectives was to compare the NP microbiomes of the HIV(+) and HIV(-) mothers at t=0 and t=6. From Figure 1B, we observed that HIV(+) mothers may have had higher Streptococcus abundance than HIV(-) mothers overall. A significant inter-group beta diversity test result at t=6 showed strong differences in taxa among HIV statuses, a finding reflected by the stacked bar plots indicating a larger presence of taxa with relatively small abundances in HIV(-) women (denoted as “Other” throughout our analyses). This finding implies more microbial diversity in the NP microbiomes of the HIV(-) women. In the nasopharynx, lower diversity has been associated with individuals with rhinovirus illness50 and in children with HIV-associated bronchiectasis,51 suggesting that greater NP diversity is a sign of a healthy microbiome. This mirrors previous findings in the gut microbiome of healthy individuals.52” (lines 477-486)

Comment 2-34: You argue pretty clearly that you have good samples. I don't think you need to state "retrospective analysis of samples" as a limitation. 

Response 2-34: We agree with the reviewer and have removed this limitation from the Conclusion section.

Comment 2-35: I don't agree with the statements about minimizing the limitation of lack of HIV testing. In this age group, HIV infections can certainly be rapidly progressive. Also, because the underlying causes of HEU having poor health c/t HUU aren't well understood in the first place, one can't assume the effects ("direction of bias") of contaminating the HEU cohort with infected infants.   

Response 2-35: We have removed the two sentences in question that served to minimize the argument against HIV incidence in the cohort.

Comment 2-36: "This is intended as a case study..." >> Use the term "exploratory" or "pilot" study instead because "case study" has a different, specific meaning.

Response 2-36: We have corrected this oversight by using the term “pilot study” (line 504).

Conclusions

Comment 2-37: Focus here on what can be gleaned from your study results. You did not evaluate "sociological differences."

Response 2-37: We appreciate the reviewer’s remarks and have subsequently removed references to sociological differences in the Conclusion. We have altered the first sentence to read as follows:

“Acknowledging these issues, our findings suggest that there are subtle nuances between HIV-exposed and HUU infant populations.” (lines 509-510)

Consent

Comment 2-38: Because you also report data from the mothers, I hope you also consented them for that (not just the infant's data)?

Response 2-38: We thank the reviewer for noticing this oversight. We did obtain consent from the mothers who were also participants in the study. We have added the following to the Consent section of the manuscript (lines 554-556):

“Written informed consent for publication of the infants’ patient details was obtained from the parents of the infants. Consent for inclusion of mothers’ data was also obtained from the participating mothers themselves.”

 10.21956/gatesopenres.15340.r33429 Reviewer response for version 1 Smith Christiana 1 Referee https://orcid.org/0000-0002-4611-6477 University of Colorado, Aurora, CO, USA

Competing interests: No competing interests were disclosed.

 16 6 2023 Copyright: © 2023 Smith C 2023 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

This is a well-written description of a very small study comparing the nasopharyngeal microbiome of Zambian HEU infants (and their mothers) to HUU infants (and their mothers) over the first 14 weeks of life. The hypothesis that the maternal or infant microbiome plays a key role in the increased morbidity/mortality observed in HEU infants has been under-explored in the literature, justifying the indication for the study. However, the differences identified between HEU and HUU infants in this study were rather minimal (the authors use the word “subtle”), which diminishes the overall impact/significance of the publication.

Strengths of the study include that the participants were followed longitudinally over time, and the HEU and HUU cohorts were well-matched and recruited from the same physical location.

The two major limitations of the study, which are described appropriately by the authors, are: 1) the small sample size results in very limited power to identify differences between HEU and HUU infants, leading to “uncertain evidence that there is an HIV exposure effect on the NP microbiome...” and 2) no information is provided on the immune status of the HIV+ mothers (viral load, CD4) or the HIV infection status of the HIV-exposed infants (i.e. HIV NAAT results after birth), which are likely very important contributors to the NP microbiome. For example – a person with poorly controlled HIV and low CD4 count is at risk of opportunistic respiratory infections (TB, PJP, MAC) which would likely influence the composition of their NP microbiome. Was there really no information available to the study authors that could be included to better describe the health of the mothers and infants (any maternal viral loads or CD4 counts, known maternal comorbidities, infant dried blood spot testing at birth, etc.)?

Person-first language should be used throughout the abstract and manuscript. It is probably fine to use the abbreviations as-is (HIV+, HEU, HUU), but phrases such as “HIV-positive mothers” or “HIV-infected women” should be changed to “women/mothers with HIV” or “women/mothers living with HIV.” The term “mother-to-child transmission” is also outdated and should be changed to “vertical transmission” or “perinatal transmission.” For more information on why this matters, please see: Dilmitis et al. (2012 1 ).

Introduction, paragraph 2: the authors reference “the HEU phenomenon” several times without really explaining what “the phenomenon” is. Are they referring to the increased morbidity and mortality observed in this population? Please be more specific.

Introduction, paragraph 4: The authors state that their study might “provide insight into the presence or absence of immunological factors distinguishing [HEU vs. HUU infants].” However, differences in the microbiome do not necessarily indicate differences in “immunological factors.” Recommend revising the text appropriately.

Methods, paragraph 1: it is necessary to state whether all infants breastfed or whether some infants consumed any foods other than breast milk in the first 14 weeks of life, as nutrition sources have a profound ability to impact the microbiome.

There are several examples where the manuscript is organized in a non-standard way:

Tables containing data (such as Table 1 and Table 2) belong in the results section, not the methods section. Recommend moving these tables, and the text describing them, to the beginning of Results.

The text in the methods section describing the reasons that some samples were excluded (beginning with “Some missingness in the data is present…”) also belongs in results, not methods.

There are some “editorial” comments in the results section, which really belong in the discussion instead. For example, page 7, last paragraph: “…which may in part be the result of the mothers’ acquired immunity over time and therefore lower overall NP carriage;” page 8, 3 rd to last paragraph: “These transitional patterns align with those found in a longitudinal East Asian infant cohort.”

 Table 1 is very confusing and poorly organized:

The headers often do not apply to the rows beneath them (male sex and median age should not be categorized under “place of birth;” median birth weight and twin births should not be categorized under “gestational age;” number of mothers, mothers age, married, father lives with child should not be categorized under “immunizations at birth;” etc.

There are many rows for which the result is 0 in every column, or for which there are apparently no data. These rows should be removed from the table completely. For example, why is the Chilenje clinic described in table when only the Chawama clinic is described in the text? Why is “other place of birth” listed in the table if none of the infants were born in an “other place”? Why are “father lives with child” and “trimester initiated ART” included in the table if there are no data for these variables? The L&D complications that did not occur in any infants (obstructed labor, sepsis, hemorrhage, other) can be removed and a sentence could be added to the text that says “none of these complications occurred: X,Y,Z.”

The table shows one infant with birth asphyxia. This infant should be further described in the text. How severe was the birth asphyxia? Did the infant require respiratory support, intubation, oxygen, etc. (since respiratory instrumentation could potentially have impacted this infant’s respiratory microbiome)?

 Table 2:

The text states that two of 20 mothers lacked swabs at both the t=0 and t=6 time points. However, the table shows all 20 mothers had a swab at t=0 and 2 were missing swabs at t=6. Either the text or table should be revised so that they harmonize.

It would be preferred if the table could be stratified by HIV status at each time point, rather than consolidating all of the HEU and HUU results at the bottom. This way it would be easy to see whether either group was over-represented at any one given time point.

 Table 3: please standardize the number of decimal points across columns (some columns have 2 decimal points, others have none).

Table 4: several columns describe “adjusted p-values” but there is no description of what variables were adjusted for. The reader should not have to refer back to the methods section to understand the statistical analysis. This is also a problem in Table 7. Please add a description of what variables the analyses were adjusted for to the Table heading or to a footnote for each table.

Table 6: I believe there is a typo in the table; the two time points being compared are t=0 and t=6 according to the text; however, the table shows t=0 and t=1.

Table 7: It is very difficult to understand this table without reading the text. Specifically, what direction is indicated by the positive and negative numbers in the column titled “mean of differences” (higher abundance in infants or higher abundance in mothers)? After a long period of scrutiny, I was able to determine that negative numbers mean higher abundance in infants, but it should be immediately apparent to the reader. Please explain this in the Table heading or in a footnote.

Also related to Table 7, in the results section, page 12, paragraph 2: The authors describe that the purpose of the analysis in Table 7 is to compare the abundance of 3 particular bacterial species between the mothers and infants within each cohort (HIV+/HEU and HIV-/HUU), and that this comparison will help explain why HEU infants may be more susceptible to morbidity/mortality. It is not clear how comparing the abundance of these bacteria between mothers vs. infants WITHIN each cohort would help explain the HEU infants’ increased susceptibility to morbidity/mortality. It seems that to answer this question, it would be more valuable to compare the abundance of bacteria between HEU vs. HUU infants (or between HIV+ vs. HIV- mothers). Please clarify how you expect this analysis to answer the question proposed, or modify/remove the language describing the reason for this analysis.

Discussion: The authors should add to the list of limitations that their results may not be generalizable to HEU infants around the globe, given that all samples were collected from participants living in a single African country.

Discussion, page 14: There is some redundancy between the last 2 sentences of the first paragraph, and the 6 th paragraph, both of which describe differences in bacterial diversity between HIV+ and HIV- mothers.

There are several typos in the manuscript. For example, page 11, paragraph 3: “The same test for all infant samples at time point six was contrastingly showed….” Also Conclusions: “our findings may suggest that that sociological….”

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

I cannot comment. A qualified statistician is required.

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Pediatric infectious diseases clinician, research focused on the immunology of HIV-exposed and HIV-infected infants and children.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

References : Language, identity and HIV: why do we keep talking about the responsible and responsive use of language? Language matters. Journal of the International AIDS Society .2012;15(4(Suppl 2)) : 10.7448/IAS.15.4.17990 10.7448/IAS.15.4.17990 Odom Aubrey Boston University School of Medicine, Boston, Massachusetts, USA

Competing interests: None to report.

 13 8 2024

Reviewer 1

Comment 1-1: This is a well-written description of a very small study comparing the nasopharyngeal microbiome of Zambian HEU infants (and their mothers) to HUU infants (and their mothers) over the first 14 weeks of life. The hypothesis that the maternal or infant microbiome plays a key role in the increased morbidity/mortality observed in HEU infants has been under-explored in the literature, justifying the indication for the study. However, the differences identified between HEU and HUU infants in this study were rather minimal (the authors use the word “subtle”), which diminishes the overall impact/significance of the publication.

Strengths of the study include that the participants were followed longitudinally over time, and the HEU and HUU cohorts were well-matched and recruited from the same physical location.

Response 1-1: We appreciate the reviewer’s summary and thank the reviewer for providing their feedback for this paper.

Comment 1-2: The two major limitations of the study, which are described appropriately by the authors, are: 1) the small sample size results in very limited power to identify differences between HEU and HUU infants, leading to “uncertain evidence that there is an HIV exposure effect on the NP microbiome...” and 2) no information is provided on the immune status of the HIV+ mothers (viral load, CD4) or the HIV infection status of the HIV-exposed infants (i.e. HIV NAAT results after birth), which are likely very important contributors to the NP microbiome. For example – a person with poorly controlled HIV and low CD4 count is at risk of opportunistic respiratory infections (TB, PJP, MAC) which would likely influence the composition of their NP microbiome. Was there really no information available to the study authors that could be included to better describe the health of the mothers and infants (any maternal viral loads or CD4 counts, known maternal comorbidities, infant dried blood spot testing at birth, etc.)?

Response 1-2: We agree that information on the immune status of the HIV+ mothers is indeed integral to identifying additional factors in the composition of their NP microbiota. We regrettably did not have access to this data as detailed HIV status information was not collected as part of the SAMIPS study, which was focused on Pertussis incidence. It was not optimally designed to look at the impact of HIV exposure as the original study purpose and did not seek to characterize the mothers HIV health status nor collect breastfeeding information. Neither viral load nor CD4 testing were routinely performed in Lusaka at this time and neither of these were deemed essential for the purposes of the original study beyond noting the mothers’ HIV status.

We have clarified this point further in the Methods section on line 118, and in the discussion on lines 492-494 (“several key variables were not measured that would have been informative as potential confounders in our study results”).

Comment 1-3: Person-first language should be used throughout the abstract and manuscript. It is probably fine to use the abbreviations as-is (HIV+, HEU, HUU), but phrases such as “HIV-positive mothers” or “HIV-infected women” should be changed to “women/mothers with HIV” or “women/mothers living with HIV.” The term “mother-to-child transmission” is also outdated and should be changed to “vertical transmission” or “perinatal transmission.” For more information on why this matters, please see: Dilmitis  et al. (2012 1).

References

1. Dilmitis S, Edwards O, Hull B, Margolese S, et al.: Language, identity and HIV: why do we keep talking about the responsible and responsive use of language? Language matters.  Journal of the International AIDS Society. 2012;  15 (4(Suppl 2)).  Publisher Full Text

Response 1-3: We thank the reviewer for their attentiveness to this matter. We have implemented person-first identifying language throughout the manuscript.

Comment 1-4: Introduction, paragraph 2: the authors reference “the HEU phenomenon” several times without really explaining what “the phenomenon” is. Are they referring to the increased morbidity and mortality observed in this population? Please be more specific.

Response 1-4: The reviewer’s inclination is correct; we are referring to the increased morbidity and mortality rates observed in HEU infants. We have altered the manuscript to read more clearly in the introduction, by removing one instance and clarifying in another case the “HEU ‘phenomenon’ of increased morbidity and/or mortality rates” (see lines 51, 64, 66).

In lines 511-513 of the Conclusions, we altered a sentence reading “This warrants further research and discussion regarding the phenomenon” to “The effects we have found here warrant further research and discussion regarding the role HIV exposure plays in infant health before readily affirming or denying that HIV exposure affects infants’ NP microbiomes.”

Comment 1-5: Introduction, paragraph 4: The authors state that their study might “provide insight into the presence or absence of immunological factors distinguishing [HEU vs. HUU infants].” However, differences in the microbiome do not necessarily indicate differences in “immunological factors.” Recommend revising the text appropriately.

Response 1-5: We agree with the reviewer’s reasoning that differences present in the microbiome do not necessarily correlate to immunological differences. We made a jump in reasoning that can only be made with something akin to an analysis of host-microbe interactions, which was not conducted here. We have altered the sentence to read, “We reasoned that by characterizing quantifiable differences in NP microbiota distinguishing these two groups, this pilot study could indicate differences encouraging future immunology-focused studies” (lines 82-83).

Comment 1-6: Methods, paragraph 1: it is necessary to state whether all infants breastfed or whether some infants consumed any foods other than breast milk in the first 14 weeks of life, as nutrition sources have a profound ability to impact the microbiome.

Response 1-6: We agree that information on breastfeeding would be extremely relevant for the purposes of microbiome analysis. After reviewing the case report forms, we found that the study did not assess and collect breast feeding status. For purposes of this study, we can clarify that formula is rarely used and breastfeeding is nearly universal. We have clarified this in the manuscript and added a relevant citation (lines 123-125).

Comment 1-7: There are several examples where the manuscript is organized in a non-standard way:

Tables containing data (such as Table 1 and Table 2) belong in the results section, not the methods section. Recommend moving these tables, and the text describing them, to the beginning of Results.

Response 1-7: We have moved the tables and appropriate text descriptions to the beginning of the Results section (lines 267-273).

Comment 1-8: The text in the methods section describing the reasons that some samples were excluded (beginning with “Some missingness in the data is present…”) also belongs in results, not methods.

Response 1-8: We have moved the description of the excluded samples to the beginning of the Results section (line 268).

Comment 1-9: There are some “editorial” comments in the results section, which really belong in the discussion instead. For example, page 7, last paragraph: “…which may in part be the result of the mothers’ acquired immunity over time and therefore lower overall NP carriage;” page 8, 3 rd to last paragraph: “These transitional patterns align with those found in a longitudinal East Asian infant cohort.”

Response 1-9: We have moved the comments in question to the Discussion section, including the larger paragraph on parallel findings from previous studies (lines 285-287; 421-422; other comments were also moved when relevant).

Comment 1-10: Table 1 is very confusing and poorly organized:

The headers often do not apply to the rows beneath them (male sex and median age should not be categorized under “place of birth;” median birth weight and twin births should not be categorized under “gestational age;” number of mothers, mothers age, married, father lives with child should not be categorized under “immunizations at birth;” etc.

There are many rows for which the result is 0 in every column, or for which there are apparently no data. These rows should be removed from the table completely. For example, why is the Chilenje clinic described in table when only the Chawama clinic is described in the text? Why is “other place of birth” listed in the table if none of the infants were born in an “other place”? Why are “father lives with child” and “trimester initiated ART” included in the table if there are no data for these variables? The L&D complications that did not occur in any infants (obstructed labor, sepsis, hemorrhage, other) can be removed and a sentence could be added to the text that says “none of these complications occurred: X,Y,Z.”

The table shows one infant with birth asphyxia. This infant should be further described in the text. How severe was the birth asphyxia? Did the infant require respiratory support, intubation, oxygen, etc. (since respiratory instrumentation could potentially have impacted this infant’s respiratory microbiome)?

Response 1-10: We thank the reviewer for their attention to this oversight. We have reordered the rows to clearly define parameter sections, and removed all rows for which the parameter did not apply to the infant demographics.

We have added the following sentence in place of the excluded rows: “None of the mothers in this cohort experienced obstructed labor, sepsis, or hemorrhage complications during the labor and delivery period” (lines 97-99).

During the SAMIPS data collection procedures, each mother was asked whether they had any complications with labor and delivery. This was a closed ended question with five possible responses: obstructed labor, birth asphyxia, sepsis, hemorrhage, or other. Follow-up questions about severity or treatment were not asked. Given that the study eligibility criteria limited enrollment to healthy infants with no acute or chronic medical conditions, it is unlikely that the birth asphyxia was severe. However, we cannot rule out the use of respiratory instrumentation that could impact the infant’s respiratory microbiome. We have noted this clarification in our study limitations of the Discussion section (lines 502-503).

Comment 1-11: Table 2:

The text states that two of 20 mothers lacked swabs at both the t=0 and t=6 time points. However, the table shows all 20 mothers had a swab at t=0 and 2 were missing swabs at t=6. Either the text or table should be revised so that they harmonize.

It would be preferred if the table could be stratified by HIV status at each time point, rather than consolidating all of the HEU and HUU results at the bottom. This way it would be easy to see whether either group was over-represented at any one given time point.

Response 1-11: We appreciate the reviewer’s attention to this discrepancy. We verified that the table contains an accurate summary of the data and updated the text to reflect the table (line 273). We have also updated Table 2 to delineate each status/subject group per time point.

Comment 1-12: Table 3: please standardize the number of decimal points across columns (some columns have 2 decimal points, others have none).

Response 1-12: We have adjusted the table to have one decimal point standardized across the columns and right-justified the entries (now Table 7).

Comment 1-13: Table 4: several columns describe “adjusted p-values” but there is no description of what variables were adjusted for. The reader should not have to refer back to the methods section to understand the statistical analysis. This is also a problem in Table 7. Please add a description of what variables the analyses were adjusted for to the Table heading or to a footnote for each table.

Response 1-13: We thank the reviewer for their comment. We acknowledge that there may also be some confusion as to what we mean by adjusted p-values. The concept of adjusting p-values is widely used in ‘omics analysis, and stems from the application of a correction to limit the Type I error (false positive) inflation in multiple hypothesis testing. There are several techniques that can be used for this purpose; we used the Bonferroni correction which is the most conservative. The actual implementation of this correction is to identify a corrected significance level, α, so each p-value is compared to this new value resulting in an adjusted p-value.

We have clarified language throughout the manuscript and caption for Table 7 (now Table 6) by changing to the following language: “A Bonferroni correction was applied to account for multiple hypothesis testing.” (lines 753-754).

We recognize that Table 4 (now Table 3) is especially confusing given the different variables. We have rearranged the columns to list the coefficient estimate followed by the unadjusted and adjusted p-values for each variable included in the model.

Comment 1-14: Table 6: I believe there is a typo in the table; the two time points being compared are t=0 and t=6 according to the text; however, the table shows t=0 and t=1.

Response 1-14: We have adjusted Table 6 accordingly (now Table 5).

Comment 1-15: Table 7: It is very difficult to understand this table without reading the text. Specifically, what direction is indicated by the positive and negative numbers in the column titled “mean of differences” (higher abundance in infants or higher abundance in mothers)? After a long period of scrutiny, I was able to determine that negative numbers mean higher abundance in infants, but it should be immediately apparent to the reader. Please explain this in the Table heading or in a footnote.

Response 1-15: We agree that the mean of differences column is not well explained. We have gone ahead and corrected the table caption and heading to explain the sign directionality for mean of differences values for Table 7 (now Table 6).

Comment 1-16: Also related to Table 7, in the results section, page 12, paragraph 2: The authors describe that the purpose of the analysis in Table 7 is to compare the abundance of 3 particular bacterial species between the mothers and infants within each cohort (HIV+/HEU and HIV-/HUU), and that this comparison will help explain why HEU infants may be more susceptible to morbidity/mortality. It is not clear how comparing the abundance of these bacteria between mothers vs. infants WITHIN each cohort would help explain the HEU infants’ increased susceptibility to morbidity/mortality. It seems that to answer this question, it would be more valuable to compare the abundance of bacteria between HEU vs. HUU infants (or between HIV+ vs. HIV- mothers). Please clarify how you expect this analysis to answer the question proposed or modify/remove the language describing the reason for this analysis.

Response 1-16: We agree that the justification for performing this set of tests is invalid as written, and so we have rewritten the test justification as follows (see lines 363-366):

“Although all infants in our cohort were free from illness, we found that certain pathogenic species were present in modest abundance across samples. To ascertain the possibility of HIV-exposed infants acquiring pathogens from their HIV(+) mothers, we conducted paired t-tests for three well-known pathogenic species: Streptococcus pneumoniae; Haemophilus influenzae; and Staphylococcus haemolyticus.”

Comment 1-17: Discussion: The authors should add to the list of limitations that their results may not be generalizable to HEU infants around the globe, given that all samples were collected from participants living in a single African country.

Response 1-17: We agree with this limitation and have added it to our conclusion accordingly on lines 490-491. (“Second, given that all samples were collected from participants born in Zambia, these results may not be generalizable to HIV-exposed infants in other countries.”)

Comment 1-18: Discussion, page 14: There is some redundancy between the last 2 sentences of the first paragraph, and the 6 th paragraph, both of which describe differences in bacterial diversity between HIV+ and HIV- mothers.

Response 1-18: We have relocated the sentence referring to Figure 1B (“From Figure 1B, we observed that…”) to lines 478-479 and removed the repetitive sentence referring to increased microbial diversity.

Comment 1-19: There are several typos in the manuscript. For example, page 11, paragraph 3: “The same test for all infant samples at time point six was contrastingly showed….” Also Conclusions: “our findings may suggest that that sociological….”

Response 1-19: We have made the appropriate corrections throughout the manuscript and thank the reviewer for their attention to this matter.