Human Amylase Diversity

Representative publications on this topic include:

1. Yilmaz, F., Karageorgiou, C., Kim, K.*, Pajic, P., Scheer, K., Human Genome Structural Variation Consortium, Beck, C. R., Torregrossa, A., Lee, C., & Gokcumen, O. (2024). Reconstruction of the human amylase locus reveals ancient duplications seeding modern-day variation. Science, 386(6724), eadn0609. *denotes co-first author
2. Karageorgiou, C., Gokcumen, O., & Dennis, M. Y. (2024). Deciphering the role of structural variation in human evolution: a functional perspective. Current opinion in genetics & development, 88, 102240.

Media highlights on this topic:

1. How Early Humans Evolved to Eat Starch, written by Carl Zimmer for the NYT.

2. Why do we love carbs? Origins predate agriculture, maybe even our split from Neanderthals, written by Tom Dinki for UB Now.

The amylase gene encodes an enzyme that digests starch; prior studies have shown high variation in the amylase locus. However, its sequence complexity has left much of the structural variation unresolved, hindering our understanding of its evolutionary history. Using long-read sequencing and optical mapping, we identified 30 distinct structural haplotypes at the amylase locus. Recurrent amylase duplications have been discussed within the context of dietary adaptation in mammals.

The AMY1 gene shows wide diversity across human populations, with no clear geographic specificity. We observe the three-copy haplotypes of the AMY1 gene across almost all populations in Africa. This suggests the primary two duplications of the gene occurred before the out-of-Africa migration, approximately 60KYA to 70KYA, offering insight into the early evolutionary history of the AMY1 gene.

The amylase locus exhibits a high mutation rate, far exceeding most other regions of the genome. This elevated rate has likely contributed to the extensive copy number variation seen in AMY1, making it a key locus for studying human SVs.

Strong negative selection limits functional variation among amylase gene copies. Despite the extensive variation in copy number, the coding sequences of all AMY genes remain highly conserved due to strong negative selection operating in the locus.

Our findings suggest that archaic hominins had increased AMY1 copy numbers, potentially dating the duplication as far back as 800KYA. However, several possible scenarios could explain these findings, and more research is needed to understand the origins of these duplications.

Over the past 4,000 years, an increase in AMY1 copy number is observed in European populations. This increase coincides with the spread of agriculture, possibly reflecting changes in diet during this period, but the exact pressures driving this increase remain uncertain.

The oldest anatomically modern human genome analyzed, Ust’-Ishim from Siberia (~45KYA), had six AMY1 copies per diploid cell. Similarly, the oldest modern human from Europe, Peştera Muierii (~34KYA), had eight AMY1 copies. These findings indicate that high AMY1 copy numbers (≥6 copies) had already spread across Eurasia as far back as ~45KYA.

Duplications at the AMY locus are primarily driven by non-allelic homologous recombination (NAHR). This mechanism has been key to generating the structural and copy number variation seen across different human populations today.

We also examined the evolution of the AMY2 gene. Although AMY2 shows less variation in copy number compared to AMY1, our findings suggest that its duplications occur primarily through non-recurrent mechanisms, specifically microhomology-mediated break-induced replication (MMBIR).

Taken together: Our findings suggest that the initial AMY1 duplication event occurred far before the rise of agriculture, possibly even before the human-Neanderthal split around 800KYA. This highlights how structural variation at the amylase locus has deep evolutionary roots. The significant increase in AMY1 copy numbers in European farmers over the last 4,000 years may reflect an adaptive response to changes in diet, particularly the increased consumption of starch. However, the pre-existing variation laid the groundwork for this shift.

By identifying both non-recurrent and recurrent mechanisms behind the structural variation in AMY1 and AMY2 genes, this study sheds light on how gene duplications and other genomic changes have influenced human evolution. Our work provides a robust framework for understanding how lifestyle factors, such as diet, have interacted with the genome over time, leading to the complex structural variation observed today in the amylase locus.

If you are in mechanisms generating SVs, check out our mutational network, where we connect all distinct structural haplotypes, tracing the most parsimonious scenarios and inferring duplications, deletions, and inversions.