Human cytomegalovirus (CMV), a prevalent herpesvirus, is commonly transmitted to infants through breast milk. While CMV transmission via breast milk is generally considered harmless for full-term, healthy infants, it can have severe consequences for preterm or immunocompromised infants, potentially causing sepsis, thrombocytopenia, and long-term neurodevelopmental issues. The transmission rate in full-term infants from seropositive mothers can be as high as 70%. Despite the frequency and clinical importance of CMV reactivation in the mammary gland during lactation, research into its relationship with human milk composition remains limited. One study found increased pro-inflammatory cytokines in milk with maternal CMV reactivation. This study investigates the association between CMV in breast milk and alterations in milk composition, specifically focusing on the impact of these changes on the infant gut microbiome and the infant's growth. Understanding this association is crucial, as it can provide insight into the potential long-term health effects of CMV exposure during infancy, even in seemingly benign cases.

Previous research highlights the global seroprevalence of CMV in women of childbearing age (~85%), emphasizing its ability to infect various cell types, including epithelial, endothelial, and immune cells. Following initial, often asymptomatic, infection, CMV establishes lifelong latency. Postnatal transmission via breast milk is frequently observed, with CMV reactivating in the mammary glands of nearly all seropositive women. While the effects of postnatal CMV transmission on preterm infants are well-documented (including sepsis, thrombocytopenia, and neurodevelopmental issues), the consequences for full-term, healthy infants remain unclear. Existing literature offers conflicting evidence concerning the effect of CMV transmission on infant growth. However, prior research has not comprehensively investigated the connection between CMV in breast milk and the broader composition of the milk itself, including the hundreds of bioactive factors that influence infant development. This study fills this gap in the literature by investigating the impact of CMV on milk composition and its relationship to infant health outcomes.

This study analyzed data from the Mothers and Infants Linked for Healthy Growth (MILK) study, focusing on exclusively breastfeeding mother-infant pairs with full-term infants. Milk samples were collected at one month postpartum. Shotgun DNA sequencing and qPCR were used to determine CMV status in milk samples. RNA sequencing analyzed the milk transcriptome, while metabolomics investigated the milk metabolome. Infant fecal DNA sequencing assessed the infant gut microbiome at 1 and 6 months postpartum. Infant growth was measured via weight-for-length Z-score (WLZ). Statistical analyses included differential gene expression analysis (DESeq2), differential abundance analysis for metabolites, principal component analysis (PCA) for microbiome data, and linear regression and structural equation modeling (SEM) to examine associations between CMV status, milk composition, gut microbiome, and infant growth. Various covariates, including maternal characteristics and infant factors, were included in these analyses to account for potential confounding effects.

The study identified CMV DNA in 35% of the milk samples using shotgun DNA sequencing. This was validated by qPCR, which showed strong agreement with the sequencing results. CMV-positive milk samples exhibited an upregulation of the IDO1 gene and increased levels of kynurenine and kynurenic acid, suggesting an elevated activity in the IDO tryptophan-to-kynurenine metabolic pathway. Infant gut microbiomes of those exposed to CMV+ milk showed decreased abundances of *Bifidobacterium* species and increased *Clostridium tertium*. Infant growth showed a complex interplay with CMV exposure: while infants fed CMV+ milk had higher weight-for-length Z scores at one month compared to those fed CMV- milk, higher CMV viral load (proportion of CMV-mapped reads) was negatively correlated with weight-for-length and positively correlated with length-for-age in the CMV+ group. Structural equation modeling suggested that the positive association between CMV+ milk status and one-month weight-for-length was likely driven by increased milk kynurenine levels rather than a direct effect of the virus itself.

The study's findings challenge the prevailing notion that CMV presence in breast milk is inconsequential for healthy term infants. The observed upregulation of the IDO tryptophan-to-kynurenine pathway in CMV+ milk suggests a possible immune response to CMV reactivation in the mammary gland. The changes in the infant gut microbiome, with lower abundances of beneficial *Bifidobacterium* and increased potentially pathogenic *Clostridium tertium*, warrants further investigation into potential long-term health consequences. The opposing effects of kynurenine and viral load on infant weight-for-length emphasize the complexity of CMV's influence on infant development. The positive correlation between milk kynurenine and infant weight-for-length at one month, independent of CMV status, highlights the importance of milk composition as a factor in early growth. Future studies should explore the long-term impact of these findings and further investigate the mechanisms behind CMV's effects on milk composition and its influence on infant health.

This study demonstrates a significant association between CMV DNA in human milk and alterations in milk composition, infant gut microbiome, and early infant growth. The upregulation of the IDO pathway and changes in the infant gut microbiome highlight potential mechanisms through which CMV exposure might affect infant development. While the observed effects on growth in full-term infants were of moderate magnitude and may not be clinically significant, the findings underscore the need for further research into the long-term implications of CMV exposure through breast milk, especially in vulnerable populations such as preterm infants. Further research could focus on longitudinal studies tracking viral transmission and infant health outcomes, along with exploring the mechanistic underpinnings of CMV's influence on milk composition and its impact on infant development.

A limitation is the unknown serostatus of the infants, as infant blood samples were unavailable. Therefore, the degree of postnatal CMV acquisition in infants fed CMV+ milk is uncertain. Additionally, congenital CMV infection was not tested for. The observational nature of the study prevents drawing causal conclusions, and the associations observed may be influenced by unmeasured confounding factors. The sample size and demographic characteristics of the cohort may also limit the generalizability of the findings to broader populations.