Endemic areas of human T-lymphotropic virus type 1 (HTLV-1) have been reported in Japan as well as tropical Africa, Central and South America and Melanesia. The existence of two subgroups, i.e., the transcontinental and Japanese subgroups, was reported in Japan. In the present study, we provide data on the ratio of the two subgroups in each endemic area and infection foci and examine the distribution of HTLV-1 in Japan and neighboring areas. A 657 bp fragment of env region of HTLV-1 proviral genome was successfully amplified for 183 HTLV-1 positive DNA samples. The subgroup determination was done by RFLP reactions using endonucleases HpaI and HinfI. The northern part of mainland Kyushu, represented by Hirado and Kumamoto, was monopolized by the Japanese subgroup, while the transcontinental subgroup ranged from 20 to 35% in the Pacific coast areas of Shikoku (Kochi), the Ryukyu Archipelago (Kakeroma and Okinawa) and Taiwan. An interesting finding in the present study is the presence of the transcontinental subgroup in Kochi, suggesting the endemicity of the transcontinental subgroup along the Kuroshio Current.

We conducted phylogenetic analyses and an estimation of coalescence times for East Asian strains of HTLV-1. Phylogenetic analyses showed that the following three lineages exist in Japan: “JPN”, primarily comprising Japanese isolates; “EAS”, comprising Japanese and two Chinese isolates, of which one originated from Chengdu and the other from Fujian; and “GLB1”, comprising isolates from various locations worldwide, including a few Japanese isolates. It was estimated that the JPN and EAS lineages originated as independent lineages approximately 3,900 and 6,000 years ago, respectively. Based on archaeological findings, the “Out of Sunda” hypothesis was recently proposed to clarify the source of the Jomon (early neolithic) cultures of Japan. According to this hypothesis, it is suggested that the arrival of neolithic people in Japan began approximately 10,000 years ago, with a second wave of immigrants arriving between 6,000 and 4,000 years ago, peaking at around 4,000 years ago. Estimated coalescence times of the EAS and JPN lineages place the origins of these lineages within this 6,000–4,000 year period, suggesting that HTLV-1 was introduced to Japan by neolithic immigrants, not Paleo-Mongoloids. Moreover, our data suggest that the other minor lineage, GLB1, may have been introduced to Japan by Africans accompanying European traders several centuries ago, during or after “The Age of Discovery.” Thus, the results of this study greatly increase our understanding of the origins and current distribution of HTLV-1 lineages in Japan and provide further insights into the ethno-epidemiology of HTLV-1.