Cattle mitogenome variation reveals a post-glacial expansion of haplogroup P and an early incorporation into northeast Asian domestic herds

Modern cattle descend from aurochs (*Bos primigenius*), with *Bos taurus* and *Bos indicus* resulting from independent domestication events in the Near East and Indus Valley, respectively. Mitochondrial DNA (mtDNA) analysis reveals a few major haplogroups in modern cattle, with T and I common in *B. taurus* and *B. indicus*, and P, Q, and R rare and found only in taurine breeds. Haplogroup P was prevalent in extinct European aurochs but is rarely found in modern European cattle. Its presence in a few modern Asian cattle samples, especially at high frequency (45.9%) in Japanese Shorthorn cattle, suggested a possible local domestication or introgression event in Asia. This study aimed to clarify the origin of haplogroup P in Northeast Asian cattle using complete mitogenome sequencing of Japanese Shorthorn cattle to resolve the phylogenetic relationships and estimate the timing of its spread.

Previous research using mtDNA control-region surveys indicated a Near Eastern origin for domestic cattle and the absence of European aurochs domestication. Studies on ancient DNA revealed haplogroup P's dominance in European aurochs. However, its presence in modern Asian cattle was initially attributed to introgression from wild aurochs. Recent findings of a high frequency of haplogroup P in Japanese Shorthorn cattle challenged this assumption, suggesting a potential separate Asian domestication or introgression event. The existing literature highlighted the need for whole mitogenome sequencing for higher resolution phylogenetic analysis and more accurate coalescence time estimation.

The researchers sequenced the entire mitogenome from 14 Japanese Shorthorn cattle samples previously identified as haplogroup P. This revealed ten distinct haplotypes (JS1-JS10), with two haplotypes (JS1 and JS3) being represented multiple times. Sequence alignment with previously published P mitogenomes and the bovine reference sequence identified variant sites, including indels. Phylogenetic trees were constructed using maximum likelihood (ML) methods, incorporating both the new and previously published mitogenomes, along with representative mitogenomes of other haplogroups. The analysis used both complete mitogenomes and 410-bp long control-region sequences to include a larger dataset. Coalescence times for haplogroup P nodes were estimated using the BASEML program in PAML, considering both all codon positions and third codon positions only to account for different substitution rates. Bayesian skyline plots (BSPs) were constructed using BEAST to infer past demographic trends of haplogroup P, utilizing control-region sequences from aurochs and modern cattle. The mutation rate in the control region was assumed uniform between aurochs and cattle, and the mutation rate estimates from aurochs were applied to cattle data.

Phylogenetic analyses revealed that the ten Japanese Shorthorn haplotypes formed a distinct sub-cluster, named P1a, characterized by six transitions. P1a also included a previously reported Korean mitogenome. Two European aurochs mitogenomes branched earlier in the phylogeny, suggesting a European origin for haplogroup P. Coalescence time estimates for haplogroup P, P1, and P1a were 16,180-8770 YBP, 12,530-8770 YBP, and 3720-2900 YBP, respectively. These timeframes aligned with the early post-glacial period. The BSP for haplogroup P showed a significant population size increase in aurochs around the early post-glacial period. The BSP for P1a showed a sharp increase around 650 YBP. This coincides with historical records of cattle importation from Mongolia/Russia to northern Japan in 1454-1456 AD. In contrast, other Japanese cattle breeds trace their ancestry to Korea, suggesting a different origin for Japanese Shorthorn cattle. Analyses of ancient DNA from Chinese aurochs (from the Houtaomuga ruins) did not reveal the presence of haplogroup P, implying that the integration of haplogroup P into Asian cattle herds happened later.

The findings support a model where haplogroup P originated in Europe and expanded into Asia during the post-glacial period. The emergence of the P1a sub-haplogroup in northeastern Asia likely represents an early incorporation of aurochs carrying this sub-haplogroup into local domestic herds. The high frequency of P1a in Japanese Shorthorn cattle reflects a relatively recent introduction from a northern continental source, potentially Mongolia or Russia, consistent with historical records. The absence of haplogroup P in ancient Chinese aurochs and in Yakutian cattle suggests that its integration into Asian cattle populations occurred later and may have been through a separate introduction.

This study provides strong evidence for a post-glacial expansion of aurochs haplogroup P from Europe to Asia, its incorporation into early Northeast Asian domestic herds, and a recent arrival in Japan. Further research using mitogenomes from modern autochthonous taurine breeds and ancient aurochs from relevant regions is needed to fully elucidate the origin and spread of haplogroup P in Asia.

The study relies primarily on mtDNA data, which reflects only the maternal lineage. Further studies incorporating nuclear DNA analysis would provide a more comprehensive understanding of the genetic history of cattle. The limited number of ancient DNA samples from relevant regions restricts the ability to draw definitive conclusions about the exact location and timing of events. The assumption of a uniform mutation rate for the control region between aurochs and cattle might introduce some uncertainty in the demographic estimates.