Tajima and Sonoda reply

In Latin America, the prevalence of HTLV-I is decreasing and at present is not as high (up to 10%; ref. 10) as reported previously. The same phenomenon can be seen in an endemic island in Japan that had a high prevalence of HTLV-I in 1985 (23% in 2,582 adults from 2–50% among 123 villages; ref. 11) and now has a very low prevalence rate in young adults (down to 3%). The rates of transmission of HTLV-I may have easily changed over time depending on transmission routes and infectious conditions. Therefore, it seems reasonable that one of two Andean mummies about 1,500-year-old is positive for HTLV-I.

In investigating the origin of HTLV-I, it is very important to select the appropriate source of materials from natives of South America. Some African HTLV-I strains have spread around the world, and many native people have recently been infected with HTLV-I from other ethnic groups, especially through prostitutes and drug users in large cities. To obviate these confounding factors, we have collected blood samples in very isolated villages in the Andes10. It is important to note that Andean samples from Quechua in Cusco3 would not be suitable to investigate ethnic relationship between Andean HTLV-I and its African lineage, because that city represents an admixture of many ethnic groups whose HTLV-1 may have come from several lineages.

The evolutionary speed of HTLV-I is as slow as 0.4 × 10−7–6.8 × 10−7 base substitutions/year (ref. 12), which is 100 times less than the estimate for HTLV-II (0.3 × 10−4–1 × 10−4; ref. 13). Based on the HTLV-II ‘time clock’, all of the Transcontinental subgroup of HTLV-I might have been spread throughout the ‘New Continent’ during the last 100–400 years. Based on the HTLV-I evolutionary speed, however, aboriginal strains of HTLV-1 among Mongolian populations in South America and Asia, including Japanese and Ainu, produce distinct ethnic clusters separating in the period 10,000–40,000 years BP (before present).

We propose that Asian Mongolians with HTLV-I moved into Andes during the prehistorical period long before the colonial era. Our results with HTLV-I DNA extracted from a 1,500 year-old Andean mummy substantiate this.