Cellular and transcriptional diversity over the course of human lactation

Sarah K. Nyquist, Patricia Gao, Tessa K. J. Haining, Michael R. Retchin, Yarden Golan, Riley S. Drake, Kellie E. Kolb, Benjamin E. Mead, Nadav Ahituv, Micaela E. Martinez, Alex K. Shalek, Bonnie Berger, Brittany A. Goods

TL;DR

This study leverages single-cell RNA sequencing (scRNA-seq) to provide an unprecedentedly detailed look at the diverse cell types and their changing gene expression within human breast milk over the course of lactation. The findings reveal dynamic shifts in epithelial cell populations and identify distinct macrophage subtypes, offering new insights into the complex biological processes supporting maternal and infant health.

What Problem Does This Paper Address?

Human breast milk (hBM) is a vital source of nutrition and immune factors for infants, yet its cellular composition and the functional roles of these cells remain poorly understood. While previous studies have offered glimpses, they often relied on bulk analysis or limited single-cell datasets, hindering a comprehensive understanding of how hBM's cellular landscape evolves over the extended period of lactation (up to 632 days postpartum). This lack of detailed knowledge limits our ability to fully appreciate the dynamic interplay between maternal health, infant needs, and the biological functions of breast milk. This paper aims to bridge this gap by providing a high-resolution, longitudinal characterization of hBM cells.

Methods

The researchers employed single-cell RNA sequencing (scRNA-seq) to analyze the transcriptomes of individual cells isolated from human breast milk. To optimize data quality, they rigorously tested various sample handling and cell isolation methods, ultimately favoring fresh processing of milk. For the main analysis, they collected longitudinal samples from 15 donors and processed over 48,000 cells.

Key analytical techniques included:

  • Unsupervised clustering to identify distinct cell populations.
  • Marker gene identification to annotate these clusters based on known cell-type specific genes.
  • Generalized additive models (GAMs) to associate cell frequencies with maternal-infant metadata (e.g., time postpartum, daycare attendance, hormonal birth control use).
  • Subclustering to further dissect the heterogeneity within major cell types like epithelial cells and macrophages.
  • Functional enrichment analyses (e.g., Reactome, GO terms) to infer the biological pathways and functions associated with identified cell clusters.
  • Module scoring to assess the M1/M2 polarization status of macrophages.
  • Differential expression analysis to identify genes and pathways that change over the course of lactation.

Datasets

The primary dataset is a newly generated single-cell RNA sequencing dataset of human breast milk cells. This dataset comprises over 48,478 cells from 50 samples collected longitudinally from 15 human donors across various stages of lactation, from early postpartum (3 days) to over 600 days. This dataset is unique in its longitudinal depth and the inclusion of detailed maternal-infant dyad metadata.

Key Insights and Ideas

  • Epithelial Cell Diversity Increases Over Time: Contrary to a static view, the study reveals that the diversity of epithelial cell populations within breast milk actually increases as lactation progresses, suggesting a specialization of function.
  • Macrophages are Key Immune Players with Tolerogenic Roles: Macrophages constitute the majority of immune cells in hBM and exhibit distinct subclusters, many of which are enriched for tolerogenic functions, hinting at a role in immune modulation and protection.
  • Cellular Composition is Influenced by Maternal/Infant Factors: Beyond time, factors like daycare attendance and hormonal birth control use are associated with shifts in specific epithelial cell populations, underscoring the dynamic interaction between maternal lifestyle and milk composition.
  • A "Division of Labor" in Epithelial Cells: Different epithelial cell subclusters (LC1 and secretory lactocytes) show opposing transcriptional responses to hormonal signaling (e.g., prolactin receptor), suggesting a potential specialization where some cells focus more on structural support and others on milk production as lactation advances.

Key Results

  • The majority of cells in hBM are lactocytes, a specialized epithelial cell type.
  • Cell-type frequencies shift significantly over lactation, with epithelial diversity increasing at later stages.
  • Five distinct macrophage subclusters were identified, with most scoring higher for M2 (anti-inflammatory/tolerogenic) gene signatures.
  • LC1 epithelial cells showed increased abundance associated with daycare attendance and hormonal birth control use.
  • LC1 and secretory lactocyte subclusters exhibited opposite transcriptional changes in key hormone receptor genes (e.g., ESR1, INSR, PRLR) over the course of lactation.
  • Genes related to metabolism and milk component biosynthesis generally decreased in expression over time, while genes involved in signaling and response to stimulus increased.

Limitations and Future Work

The authors acknowledge that the relatively small donor cohort size limits the statistical power for associating metadata with cell proportions, particularly after multiple testing correction. They also note that donor-specific effects can be conflated with metadata features and time, necessitating larger, prospectively designed studies.

Future work suggested includes:

  • Investigating the functional implications of macrophage subclusters and their potential role in immune tolerance.
  • Further exploring the "division of labor" hypothesis in epithelial cells and how hormonal regulation drives these functional specializations.
  • Understanding how external factors like pumping, formula supplementation, and daycare impact hBM cellular composition and function.
  • Validating these findings in breast tissue to disentangle changes in exfoliated cells from actual tissue composition shifts.
  • Leveraging these insights for therapeutic interventions to support healthy lactation and milk production.

Summary updated

Abstract

Human breast milk (hBM) is a dynamic fluid that contains millions of cells, but their identities and phenotypic properties are poorly understood. We generated and analyzed single-cell RNA-sequencing (scRNA-seq) data to characterize the transcriptomes of cells from hBM across lactational time from 3 to 632 d postpartum in 15 donors. We found that the majority of cells in hBM are lactocytes, a specialized epithelial subset, and that cell-type frequencies shift over the course of lactation, yielding greater epithelial diversity at later points. Analysis of lactocytes reveals a continuum of cell states characterized by transcriptional changes in hormone-, growth factor-, and milk production-related pathways. Generalized additive models suggest that one subcluster, LC1 epithelial cells, increases as a function of time postpartum, daycare attendance, and the use of hormonal birth control. We identify several subclusters of macrophages in hBM that are enriched for tolerogenic functions, possibly playing a role in protecting the mammary gland during lactation. Our description of the cellular components of breast milk, their association with maternal–infant dyad metadata, and our quantification of alterations at the gene and pathway levels provide a detailed longitudinal picture of hBM cells across lactational time. This work paves the way for future investigations of how a potential division of cellular labor and differential hormone regulation might be leveraged therapeutically to support healthy lactation and potentially aid in milk production.
BibTeX
@inproceedings{nyquist2022cellular,
  author = {Sarah K. Nyquist and Patricia Gao and Tessa K. J. Haining and Michael R. Retchin and Yarden Golan and Riley S. Drake and Kellie E. Kolb and Benjamin E. Mead and Nadav Ahituv and Micaela E. Martinez and Alex K. Shalek and Bonnie Berger and Brittany A. Goods},
  title = {{Cellular and transcriptional diversity over the course of human lactation}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2022},
  doi = {10.1073/pnas.2121720119},
}
Figure