Spoke to Sanjay Kumar who has better knoweldge about the M haplogroup and he sent me this report,
TARTU UNIVERSITY FACULTY OF BIOLOGY AND GEOGRAPHY, INSTITUTE OF MOLECULAR AND CELL BIOLOGY, DEPARTMENT OF EVOLUTIONARY BIOLOGY
Mait Metspalu
COMMON MATERNAL LEGACY OF INDIAN CASTE AND TRIBAL POPULATIONS
M.Sc. Thesis
The excerpt from this report in brief which is for our purpose is as follows,
Five Indian populations (Lodha, Bhoksa, Tharu, Kanet and Kurmi) are surveyed here for mtDNA variation. The populations are chosen in order to compare mtDNA variation between geographical regions as well as on social axis. The dispute over indigenous inhabitants of South Asia has largely been an open question while tribals and Austro-Asiatic speakers in particular have most often collected the fame. Here we test this conjecture by comparing mtDNA lineages of Austro-Asiatic Lodha to those of other tribals and caste groups. Gene flow from adjacent geographical areas will be followed and defining new lineage groups will hopefully refine classification of Indian-specific mtDNA lineages.
The samples used in this study were collected from four scheduled tribes (Lodha n=56, Bhoksa n=23, Tharu n=36 and Kanet n=34) and one social community (Kurmi n=55) from West Bengal and Northern regions of India. (Fig. 9). Some of the samples, namely Bhoksa, Tharu, Kanet were sent to us as purified DNA and some (Lodha, Kurmi) as bloodstains. S. Mastana and S.S. Papiha kindly provided all the samples.
The Kanet are a tribal population in the Kinnaur district of Himachal Pradesh and make up most of the districts population. In (Singh 1997) all the inhabitants of the Kinnaur district are referred to as scheduled tribe Kinnaura, Kinnara or Kinnaurese. The two major social groups of the Kinnaura are the Khosia and the Beru. The Khosia 30 are Rajput and are also known as Kanet, Khash or Khasa. They own land and are agriculturists. The Kinnaura speak the Kinnauri dialect, which belongs to the Himalayan group of Tibeto-Burman family of languages. They use different local dialects of the IndoAryan language Himachili for inter-group communication. The Kinnaura religion is an admixture of Buddhism and Hinduism. The traditional occupations of the Kinnaura are agriculture, trade and sheep rearing, which they continue till today. ~70% of the workers are cultivators. The total population of the Kinnaura is ~48000 (1981 census). (Singh 1997) pp. 533-534
The Lodha are a tribal population living mostly in western part of Midnapore district of West Bengal were they are also known as Kheria and Kharia. To a lesser extent they are also present in the Mayurbhanj and Baleswar districts of Orissa. The total population of the Lodha is ~59000 (1981 census). Their mother tongue, lodha, is akin to Savara, an Austro-Asiatic language. They are fluent in Bengali, which they use to communicate with other communities (the Lodhas in Orissa also speak Oriya). Traditionally the Lodhas have provided themselves by forest dwelling, hunting and gathering (grass-rope making in Orissa). Of the 40% of workers among the Lodhas 40% are engaged in forestry, fishing, hunting, etc., and another 40% are agricultural labourers. In Orissa the per cent of agriculturists is higher. Vast majority of the Lodhas are Hinduists. ~17% claim to be Christians. (Singh 1997) pp. 694-697 The Bhoksa are a Himalayan community (scheduled tribe) that inhabits the terai* areas of Bijnor district of Uttar Pradesh and Dehradun, Nainital and Pauri Garhwal districts of Uttaranchal. In Dehradun district they are also referred to as Mehre or Mehra. They speak Hindi and write in Devanagari script. The total population of the Bhoksa is ~32000 (1981 census). The traditional and primary occupations of the Bhoksa are agriculture and animal husbandry. Over 99% of the Bhoksa are Hinduists. (Singh 1997) pp. 146-149 * a belt of marshy land at the foot of the Himalayas mountains: moderate climate, dense to thin forests and medium rainfall, also tarai 31 The Tharu are a well-studied community (scheduled tribe) of Uttar Pradesh who live close to the border of Nepal, and are widely dispersed in the Districts of Baharaich, Gonda, Gorakhpur, Kheri (Lakhimpur) and Nainital district of Uttaranchal. Their total population in India is ~96000 (1981 census). ~99% of the Tharus are rural. Most of the Tharus live in southern Nepal (terai areas) where they number about 720,000. The Tharu trace their origin to Rajput forefathers, who fled from the great battle described in the epic Mahabharata. (For popular article on the Tharus see also: National Geographic Magazine, September 2000). They inhabit the terai areas. Their mother tongue Tharu belongs to the central group of the Indo-Aryan family of languages. They use Hindi for inter-group communication and write in Devanagari script. The Tharu are a landholding community with individual proprietorship of land. They did hunt and gather food in the past, but presently they depend on settled cultivation. Although nearly 100% of the Tharus are Hinduists, they use alcoholic beverages and eat beef. Despite their patrilineal social system, women have property rights greatly exceeding those recognized in Hindu society. (Singh 1997) pp. 1137-1140 Apart from West Bengal
The Kurmi are also concentrated in Bihar and UP where they represented respectively 3.6 and 3.5% of the population in 1931. The Kurmi generally work as cultivators and are looked at as middle caste peasants but they claim to be Kshatriyas.
Treatment of bloodstains Several discs of 3 mm diameter were cut from the bloodstains on Guthrie cards. The discs were then vortexed in 1 ml of deionised water for 30 minutes (modification from (Makowski et al. 1995). Following the aspiration of the water, the discs were incubated in 100ยตl methanol for 15 minutes, after what the methanol was removed. Next, 100ยตl 5mM NaOH/NaCl mix and 20ยตl EDTA (end concentration 0,2 mM) was added. Mineral oil was added to protect the sample from evaporation while heating at 100ยบC for 10 minutes. Then the samples were placed on ice. Method was developed in our department by Jรผri Parik. All the samples were kept at -20ยบC. 32 PCR conditions Various regions of the mtDNA were amplified using the polymerase chain reaction (PCR) (Saiki et al. 1988): Hypervariable Segments I and II (HVS-I HVS-II) in D-loop and different RFLP sites over mitochondrial DNA coding region. PCR was carried out with the thermocycler “Biometra UNO II” usually in total volume of 15-20ยตl.
It is often speculated that the tribal populations (especially the Austro-Asiatic speakers in the east and Dravidian-speaking tribes in the south) of India might be the relics of the first wave of the anatomically modern human immigration to India (Papiha 1996; Cavalli-Sforza et al. 1994). So far, mtDNA studies have revealed no grounds for such speculations. The lineages present in tribals fit well into the framework of the variation seen in non-tribal groups (Kivisild et al. 1999a) (Kivisild et al. manuscript in preparation). It has to be noted, though, that no detailed mtDNA study on Austro-Asiatic speaking tribals has been published yet.
The Lodha are a tribal population living mostly in western part of Midnapore district of West Bengal were they are also known as Kheria and Kharia. To a lesser extent they are also present in the Mayurbhanj and Baleswar districts of Orissa. The total population of the Lodha is ~59000 (1981 census). Their mother tongue, lodha, is akin to Savara, an Austro-Asiatic language. They are fluent in Bengali, which they use to communicate with other communities (the Lodhas in Orissa also speak Oriya). Traditionally the Lodhas have provided themselves by forest dwelling, hunting and gathering (grass-rope making in Orissa). Of the 40% of workers among the Lodhas 40% are engaged in forestry, fishing, hunting, etc., and another 40% are agricultural labourers. In Orissa the per cent of agriculturists is higher. Vast majority of the Lodhas are Hinduists. ~17% claim to be Christians.
Apart from West Bengal the Kurmi are also concentrated in Bihar and UP where they represented respectively 3.6 and 3.5% of the population in 1931. The Kurmi generally work as cultivators and are looked at as middle caste peasants but they claim to be Kshatriyas.
Table 3 presents the frequencies of the mtDNA haplogroups found in the studied five populations (see also supplementary material for full data table). For better characterisation of spatial differences in mtDNA lineages distribution in India, the populations were grouped by their geographical origin: Kanet from Himachal Pradesh, Tharu and Bhoksa from northern districts of Uttar Pradesh and Uttaranchal as a northern group; Lodha and Kurmi from West Bengal as an eastern group. As already established in several studies (Passarino et al. 1996a; Passarino et al. 1996b; Bamshad et al. 1997; Kivisild et al. 1999a; Bamshad et al. 2001), the dominant mtDNA lineage cluster in Indian populations is the Asian-specific M defined by gains of DdeI and AluI restriction sites at np 10394 and 10397, respectively. An average frequency of haplogroup Hg M in the studied populations was 76%, while the eastern group showed considerably higher Hg M frequency than the northern one, 93% and 57%, respectively. All the Lodhas included in this study fall into Hg M. It should be noted, however, that in another study where 32 Lodha mtDNAs were typed for Hgs M and U, the corresponding frequencies were 82% and 18% (Roychoudhury et al. 2000).
In concordance with previous reports, the subclusters of Hg M found in our study were largely Indian-specific (Quintana-Murci et al. 1999; Kivisild et al. 1999b; Bamshad et al. 2001). Eastern Asian M derivates C, D and E, accounted for only 3% each in the Kanet population. Among the Tharus the frequencies for Hgs C and D were 3% and 6%, respectively. It is worthwhile noting that the populations from West Bengal lacked eastern Asian Hg M varieties completely. This is also true for the upper cast people from West Bengal (our unpublished data).
Of the Hg M subclusters, M3 is the most widespread in the populations under study – the Kurmi being the only ones lacking this haplogroup (see Table 3). M4a and M6 discriminate the northern and eastern populations, as M4a is present only in the former and M6 only in the latter. This segregation is not maintained when additional data of many different Indian populations is included – both M4 and M6 have representatives from many populations of different social rank, geographical origin and linguistic background (our unpublished data).
A number of Hg M lineages could not be ascribed to any of the defined Hg M subhaplogroups. The greedy network (Bandelt et al. 2000) based on HVS-I sequence variation of these lineages (M*) (Fig. 11) reveals an extensive diversity, in particular in the northern populations. The Lodhas show the least amount of variation and fall into only a few (8) haplotypes. In general, this analysis does not reveal any strictly population- or region-specific lineage groups. There is only one lineage with considerable length (substitutions at nps 16147, 16189, 16243, 16278, 16362) what is present only among one population, the Kurmis. Given the seemingly starlike topology of the tree (Fig.11), it was possible to calculate the coalescence time for these lineages which yielded 62,000 ± 6500 years BP. Coalescence estimate of 40,000 ± 2000 years BP. for M* lineages was calculated from a much large dataset including 360 individuals (Mountain et al. 1995; Bamshad et al. 1996; Kivisild et al. 1999a) (our unpublished data). The large contrast in these estimates is most likely caused by demographic histories of the Kurmis and Lodhas. Probably because of a bottleneck event and/or by a founder effect, most of them harbour only few haplotypes. This, in turn, disrupts the starlike topology of the tree. Indeed, when Lodhas and Kurmis are excluded from the expansion time calculation, the result becomes close to that observed with the large dataset, being 43,000 ± 7300.
A vast majority of the Hg U lineages in the studied populations belong to two Indianspecific varieties - U2i and U7 (Kivisild et al. 1999a). The frequency of Hg U lineages is higher among the northern populations - 16%, while that of the eastern group is only 5% (8% if data of (Roychoudhury et al. 2000) is included). This is because the Lodhas included into this study lack Hg U. European-specific U4 and U5 are both represented by only one individual from the Tharu and Kanet, respectively.
Based on genetic studies of classical markers (summarised in (Papiha 1996), linguistic data and archaeology, peopling of India is usually discussed bearing in mind just two putative large-scale immigration waves of anatomically modern humans to the subcontinent. Firstly, the demic diffusion of Dravidic speakers coinciding with the arrival of several varieties of wheat and other cereals some 8000 – 9000 years ago from the Fertile Crescent (Diamond 1997; Renfrew 1989). Secondly, a more widely discussed scenario is in a presumed invasion of nomadic Indo-Aryan tribes around 4000 BP either from the west or from the Central Asian steppes in the north. Literature about the latter is huge and still growing, often mixed with clearly political rhetoric. However, both theories leave completely open the question about the “indigenous”, pre-Neolithic inhabitants of India. In some papers the present-day tribal populations (especially the Austro-Asiatic speakers) of India are considered to be descendants of these original inhabitants of South Asia (Papiha 1996; Cavalli-Sforza et al. 1994; Gadgil 1997).
To this date no Austro-Asiatic speaking Indian tribal population has been studied in detail for mtDNA variation. The Lodha, Munda and Santal tribals have been typed for the frequencies of haplogroups M, U, A and D (Roychoudhury et al. 2000). These results showed that in haplogroup frequencies Austro-Asiatic tribals are composed as the rest of Indians - of Hgs M and U.
Here we analysed mtDNA sequence variation in one of the Austro-Asiatic speaking tribals – the Lodhas – in detail. It became evident that the Lodhas have gone through demographic bottleneck and/or represent a population with strongly manifested narrow founder effect and, in result, exhibit only a limited extent of variation in their maternal lineages. Nevertheless, the Hg M lineages present among the Lodhas fit well into the framework of Indian varieties of this super-cluster of human mtDNA. Moreover, all the Hg M and U lineages found in the studied four tribal populations, with the exception of one Kurmi lineage, have representatives in a wide range of different Indian populations, described earlier by us and others (Mountain et al. 1995; Kivisild et al. 1999a; Kivisild et al. 2000) (our unpublished data). The mtDNA data, therefore, suggest a common origin for Indian tribal and caste groups. This may seem 42 to be in conflict with earlier interpretations by i) Das and colleagues for example, who demonstrated by frequency distributions of classical genetic markers that IndoEuropean and Austro-Asiatic speaking tribals showed little genetic affinity (Das et al. 1996) and to ii) overall conclusion of S. S. Papiha in the review of classical genetic studies of Indian populations, that tribal populations are in general well differentiated from the nontribal castes or communities (Papiha 1996). However, these differences are likely due to different approaches used: allele frequency based statistics and genealogical approach.
Our mtDNA-based analyses do not support the idea that tribals or Austro-Asiatic speakers in particular, are genetically different from the cast groups of India in principle. Rather, the differences (even significant) can be attributed to genetic drift (including bottlenecks, founder effects etc.), changing frequencies but not lineage clusters (clades), which the tribal populations share with the rest of Indian populations – and, as a rule, do not share with other Eurasian or African populations. Nevertheless, we admit that additional data on other Austro-Asiatic speakers of India and beyond is needed to draw more detailed picture of their genetic affinities within India and with contiguous areas.
In contrast to that, studied by us eastern Indian tribal populations did not show any admixture with East Asians whatsoever. For the Lodha and Kurmi, their particular demographic histories could be indicative: both populations show only a little variation and, therefore, they might have lost the East Asian lineages. Yet, most likely they never had East Asian lineages at substantial frequencies as, given the number of lineages left among these populations (30 haplotypes out of 111 samples examined), the probability that they have lost all East Asian haplotypes by means of drift is negligible. The argument of lineage loss is also not applicable to explain the lack of East Asian maternal lineages among the upper cast sample of West Bengal (our unpublished data), who show no sign of severe bottleneck in their demographic history. High population density over a long time period, making the region less prone to the effect of immigration could, among other interpretations, serve as an explanation.
The mtDNA lineages arising from the central node of Hg M that have so far not been assigned to any subcluster of Hg M, form the M* lineages. As the Lodhas and Kurmis did not show starlike topology on the greedy network of M* (Fig. 11), they had to be excluded from the coalescence time calculation. Based on the remaining data of Tharus, Bhoksas and Kanets the expansion time for the Indian M* was estimated as 43,000 ± 7300 BP. To further narrow the error margins, a dataset of 362 M* lineages 44 covering different areas and socio-cultural backgrounds of India, was included and in result, expansion time of 40000 ± 2000 years BP was found. Analyses yielding somewhat or significantly earlier Indian Hg M coalescence estimates, ranging from ~47,000 – ~65,000 BP (Kivisild et al. 1999b) and (Mountain et al. 1995) respectively, included sequence information of all Indian M subclusters in the former, and additionally even some African sequences in the latter case, therefore blurring the expansion time estimate of the central M node in India.
The report is very technical to be read and understood. However, a meticulous reading and interpretation of the same says that Lodhas and Kurmis (It should be read as Kudmis as the same has been misspelled as usual in the report; In West Bengal we have Kudmis and not Kurmis) are the idegenous group originated much before other indigenous groups of Jahrkhand (greater Jahrakhnd which encompasses areas of Jharkhnad, Bihar, West Begnal and Orissa.
Sanjay has added the following which is more easy to understand.
เค
เคซ्เคฐीเคा เคฎें anatomically modern human เคชเคจเคชा । เคिเคธเคो mitochondrial DNA Haplogroup L เคเคนा เคเคฏा । L mutate เคนो เคเคฐ L1, L2, L3, L4...... เคฌเคจा । L3 เค
เคซ्เคฐीเคा เคธे เคฌाเคนเคฐ เคจिเคเคฒा । เคฌाเคนเคฐ เค เคเคฐ L3 mutate เคนो เคเคฐ N เคเคฐ M เคฌเคจा । N เคเคฒा เคเคฏा Europe เคी เคเคฐ । M เคฆो เคญाเค เคฌंเค เคเคฏा । เคเค เคนिเคฎाเคฒเคฏ เคे เคเคค्เคคเคฐ เคी เคเคฐ เคธे เคนोเคคे เคนुเค เคชूเคฐ्เคตी เคเคถिเคฏा เคฎें เคเคฒा เคเคฏा เคเคฐ เคฆुเคธเคฐा เค เคเคฏा south Asia เคฎें । South Asia เคตाเคฒा M เคฌिเคเคฐเคคे เคนुเค เคธเคฎुเคฆ्เคฐ เคे เคिเคจाเคฐे เคिเคจाเคฐे south east Asia เคนोเคคे เคนुเค Australia เคคเค เคชเคนुंเค เคเคฏा । เคธเคญी เคเคเคน เคธเคฎเคฏ เคे เคธाเคฅ mt DNA Haplogroup M เคฎें mutation เคे เคธाเคฅ เคชเคฐिเคตเคฐ्เคคเคจ เคนोเคคा เคเคฏा । เคญाเคฐเคค เคฎें เคญी Haplogroup M เคฎें เคธเคฎเคฏ เคे เคธाเคฅ mutation เคนोเคคा เคฐเคนा เคเคฐ เคंเคธाเคจ เค
เคฒเค-เค
เคฒเค เค्เคฐुเคช เคฎें เคฌंเคเคคे เคเค । เค
เคญी เคคเค เคญाเคฐเคค เคฎें macrohaplogroup M mutate เคนो เคเคฐ M 2, M3, M4, M5, M6, M18, M25, M30, M31, M33, M34, M35, M36, M37, M38, M39, M40, M41, M42, M43, M45, M48, M49, M50 เคฎें เคฌंเคे เคนैं । เคเคธเคे เค
เคฒाเคตे เคเคจ เคธเคญी เคे subgroup เคญी เคฌเคจे เคนैं । เคฏเคนां เคคเค เคिเคธी เคाเคคि เคเคจเคाเคคि เคจเคถ्เคฒ เคा เคोเค เค
เคฐ्เคฅ เคจเคนीं เคนै เค्เคฏोंเคि เคो เคเค เคธाเคฅ เคฐเคน เคเค เคเคจ्เคนोंเคจे เคเค เคाเคคि เคเคจเคाเคคि เคจเคถ्เคฒ เคा เคจिเคฐ्เคฎाเคฃ เคเคฐ เคฒिเคฏा । เคเคธी เคฒिเค เคญाเคฐเคค เคฎें เคฐเคนเคจे เคตाเคฒे เคธเคญी เคจเคถ्เคฒ เคे เคฒोเคों เคฎें 70% เคฎें เคตिเคญिเคจ्เคจ M เค्เคฐुเคช เคฎिเคฒเคคा เคนी เคนै । เคเคนां เคคเค เคญाเคฐเคค เคฎें ancient settler เคे เคชเคนเคाเคจ เคी เคฌाเคค เคนै เคคो เคฒोเค เคฎाเคจ เคเคฐ เคเคฒเคคे เคนैं เคी เคเค เคो เคเคฆिเคตाเคธी ST เคนैं Haplogroup M2 เคตाเคฒे เคตे เคนी ancient settlers เคนैं । เคชเคฐ เคเคธ เคคเคฅ्เคฏ เคो เคจเคเคฐंเคฆाเค เคเคฐ เคฆेเคคे เคนैं เคी M2 , Haplogroup M เคธे เคจिเคเคฒा เคนुเค เคนै เคเคฐ เคญाเคฐเคค เคฎें เคเค เคธเคฎुเคน เคนैं เคिเคจเคฎे Haplogroup M เคा mt DNA , 85 เคธे 95% เค
เคฌ เคญी เคฎौเคूเคฆ เคนै । เคคो เคญाเคฐเคค เคा ancient settler เคिเคธे เคเคนเคจा เคाเคนिเค? เคाเคนिเคฐ เคธी เคฌाเคค เคนै เคिเคธ เคाเคคि เคเคจเคाเคคि เคธเคฎुเคน เคे เคธैเคฎ्เคชเคฒ เคฎें mt DNA Haplogroup M เคธเคฐ्เคตाเคงिเค เคชाเคฏा เคाเคคा เคนै । เคฏाเคจि เคिเคจเคฎें เคเค เคญी mt DNA Haplogroup M เคे เคीเคจ เค
เคชเคจी เคชुเคฐाเคจी เค
เคตเคธ्เคฅा เคฎें เคฎौเคूเคฆ เคนै ।