Mitochondrial Metabolism Underlies Developmental Programming of Hematopoietic Stem Cell Function By Prenatal Folate

Metabolism is a potent regulator of hematopoietic stem cell (HSC) function. Here, we discovered that varying prenatal folate status metabolically alters HSCs during development and affects adult HSC self-renewal and engraftment potential into adulthood. Folate status varies greatly in the global population based on nutrition, common genetic polymorphisms, and widespread supplementation. Folate-derived one-carbon metabolism (OCM) regulates cellular methylation, de novo nucleotide biosynthesis and mitochondrial metabolism, processes critical to HSC function and establishment. Despite widespread supplementation, there is limited understanding of how prenatal folate supplementation affects metabolism of the developing hematopoietic compartment. We propose that varying prenatal folate influences HSC mitochondrial metabolism during fetal development and persists into adult offspring.

To test how prenatal folate modulated HSC function, female mice were assigned to one of three experimental diets to model population-wide folate consumption: 0mg/kg (deficient, FD), 2mg/kg (control, FC) and 8mg/kg (supplemented, FS). Both FD and FS offspring revealed changes in BM hematopoiesis as evidenced by expanded lymphoid progenitors, however only FS resulted in more mature lymphoid cells in the BM. To determine effects on cell function, we performed primary and secondary transplantation of LT-HSCs isolated from adult offspring. FD offspring showed impairments in HSC function measured by worse engraftment across all peripheral blood (PB) populations compared to FC. In contrast, FS offspring HSCs demonstrated increased total cellular output of donor cells across all PB lineages except platelets, revealing that prenatal folate affects HSC function in both FD and FS adult offspring in opposing ways. Secondary transplantation revealed that FS HSCs retained enhanced total cellular output and superior engraftment compared to FC, whereas FD HSCs demonstrated complete lack of self-renewal.

To investigate underlying metabolic changes effecting function in offspring HSCs, we performed metabolomic analysis. Metabolomic profiling suggested that FD HSCs were heavily reliant on glycolysis due to impaired mitochondrial metabolism. FD HSCs exhibited signs of mitochondrial dysfunction through, 1) accumulating glucose and lactate suggesting reliance on glycolysis, 2) accumulating succinate and glutamine suggesting complex II impairment and inadequate utilization of mitochondrial metabolism and 3) accumulation of carnitines outside the mitochondria. Surprisingly, the same changes associated with mitochondrial dysfunction in FD HSCs persisted downstream into lymphoid-biased multipotent progenitors. FS offspring HSCs also showed signs of altered metabolism and metabolic distinction as shown by metabolite heatmap and PCA plot, suggesting that effects of prenatal folate on adult offspring HSC metabolism may be folate dose dependent.

To investigate transcriptional alterations associated with metabolic changes in adult hematopoiesis by prenatal folate, we performed single-cell RNA sequencing (Scseq). Scseq analysis revealed most differential gene expression (DE) in HSCs, with less DE in progenitors. Additionally, the most significant differential expression was evident in metabolic and protein translation genes, resulting in translation being highly downregulated in both FD and FS offspring HSCs. Together, our data suggest that prenatal folate status programs hematopoietic stem cell function into adulthood and is driven by alterations to metabolic activity.

We hypothesized that initiation of metabolic changes in HSCs begins during fetal development. Profiling of fetal liver (FL) hematopoietic cells revealed that FD caused expansion of LT-HSCs that was propagated across the hematopoietic hierarchy, including myeloid and lymphoid progenitors and mature myeloid cells. In contrast, FS decreased FL cell output. FD fetal HSPCs exhibited higher mitochondrial membrane potential and lower glycolytic activity, whereas FS fetal HSPCs displayed both higher glycolytic activity and oxygen consumption despite impaired production of downstream hematopoietic cells. Ongoing experiments seek to understand how prenatal folate specifically alters HSC metabolism, and the relationship between altered metabolism and its effect on protein translation in offspring HSCs.

No relevant conflicts of interest to declare.