NEW YORK (GenomeWeb) – Mothers' HIV status affects the microbiome development of uninfected infants, according to a new study.
More than a million babies are born each year to women who are HIV positive. While most of these infants do not become infected with the virus, they still experience increased morbidity and twice the mortality of infants who were not exposed to HIV.
Researchers from the Children's Hospital Los Angeles examined the microbiomes of mother-and-baby pairs from Haiti in which half of the mothers were HIV-positive. As they reported today in Science Translational Medicine, the microbiomes of HIV-exposed infants differed from those of unexposed infants, even though the mothers all had roughly similar microbiomes, no matter their HIV status.
"In contrast to the mostly consistent microbial communities identified in all of the mothers, the microbiomes of HIV-exposed, uninfected infants were strikingly different from infants born to HIV-negative women in the same community," first author Jeffrey Bender from Children's Hospital Los Angeles said in a statement.
While some of this dysbiosis could be traced to microbial transfer between mother and child, Bender and his colleagues also uncovered changes in the human milk oligosaccharide (HMO) content of infected mothers' breast milk that appeared to have downstream effects on the infants' microbiomes and, possibly, their immune systems.
He and his colleagues enrolled 50 mother-and-infant pairs from Port-au-Prince, Haiti, into their study, and of this group, half the mothers had HIV. All infected mothers were on anti-retroviral therapy, had low plasma HIV RNA levels, and had high CD4 T cell counts. The researchers collected microbial samples from six sites from each pair — areolar skin, breast milk, and vaginal samples from the mothers and mouth, skin, and stool samples from the infants.
The researchers performed 16S ribosomal RNA gene sequencing on the samples to yield some 77 million reads, which they then classified into nearly 700 operation taxonomic units.
Broadly, the researchers found that the microbiomes of infected and uninfected mothers were similar. They only noted increased β-proteobacteria levels among vaginal samples from HIV-infected mothers.
As previous studies had uncovered differences in stool microbiota based on HIV status, Bender and his colleagues said that they were surprised to see such similarities in the microbiomes of infected and uninfected women. They suggested that this discrepancy could be because the mothers in their study were relatively healthy and had high CD4 T cell counts and because they did not examine their stool microbiome.
The microbiomes of uninfected infants born to infected mothers, though, differed from those of infants born to uninfected mothers. The stool microbiomes of exposed infants had lower α-diversity, and that tracked with their mothers CD4 T cell count: the lower the count, the less diverse their infants' microbiomes were. At the same time, the researchers noted a difference in microbial taxonomic composition of infant stool, also based on their mothers' infection status. Exposed infants, they reported, had higher levels of Pseudomonadaceae in their stool while unexposed infants had higher levels of Prevotellaceae in their stool.
A random forest classifier, meanwhile, found that levels of Prevotellaceae, Alcaligenaceae, Desulfovibrionaceae, and Pseudomonadaceae in infants' stool best predicted their mothers' HIV status.
The researchers also compared the microbiomes of their cohort to those from a healthy Bangladeshi infant cohort. They developed a relative maturity index for each sample to find that HIV-exposed babies had less mature microbiomes than their unexposed peers. A single species — Bacteroides fragilis — was more abundant in exposed infants.
As the mother-to-infant transfer of microbes couldn't fully explain the difference between exposed and unexposed infant microbiomes — the mothers had broadly similar microbiomes — Bender and his colleagues also examined the composition of mothers' breast milk. In low-resources settings, they noted, women with HIV are encouraged to breastfeed their infants in conjunction with anti-retroviral therapy, as breastfeeding confers additional benefits.
Using high-performance liquid chromatography, the researchers found that the HMO composition of breast milk varied between HIV-positive and -negative mothers. Bender and his colleagues noted, though, that owing to their small sample size, this finding was not statistically significant, though it tracked with what they'd reported previously observing in a Zambian cohort.
In particular, they found that HIV-positive mothers had higher levels of 3'-sialyllactose, 3-fucosyllactose, and 2'-fucosyllactose, while HIV-negative mothers had higher levels of lacto-N-tetraose and lacto-N-neotetraose. These HMOs also correlated with the presence of certain bacteria in infant stool. For instance, increases in 3'-sialyllactose in the breast milk of HIV-negative women correlated with increases in Enterococcacceae and Fusobacteriaceae in infants' stool.
This, the researchers said, suggests that breast milk from HIV-positive women, in addition to the slight changes seen in their microbiomes, contribute to the differences seen in their infants' microbiomes.
"[T]he relatively immature and dysbiotic microbiome could potentially compromise development of the infant's immune system," Bender added.
He and his colleagues further noted that additional study of the microbiomes of HIV-exposed infants could reveal ways — such as through probiotics or prebiotics — to prevent the increased morbidity and mortality they experience.