Evolution-guided yeast complementation reveals functional differences in human PSPH variants.

Deciphering how human genetic variants affect conserved metabolic enzymes is essential for understanding their evolutionary and clinical significance. Here, we combine sequence analyses of temporally stratified human genomes with a quantitative Saccharomyces cerevisiae complementation assay in a strain lacking SER2, the yeast gene required for the final step of L-serine biosynthesis, to examine functional differences among human phosphoserine phosphatase (PSPH) variants. Population-genomic comparisons between ancient hunter-gatherers and present-day humans identified two PSPH exons with elevated differences in nucleotide diversity, guiding the selection of two ancient-genome-prioritized variants (R27S and Q83H) for functional testing. To place their effects in functional context, we expressed each variant individually and compared complementation with the modern PSPH allele and two disease-associated alleles (D32N and A35T) across multiple environmental conditions. Human PSPH enhanced growth of the SER2 deletion mutant and revealed reproducible quantitative differences among alleles. The modern allele generally conferred the strongest complementation, while the ancient genome variants supported measurable but more condition-dependent rescue, and the disease-associated alleles showed the weakest complementation. These differences were broadly consistent across conditions, while specific environmental perturbations revealed context-dependent shifts in effect size. Together, our results establish a scalable framework that links evolutionary genomics with experimental functional assays to identify and evaluate human metabolic enzyme variants with measurable in vivo effects.