Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
171

171

Ratings: (0)|Views: 7|Likes:

More info:

Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

06/20/2012

pdf

text

original

 
c
Indian Academy of Sciences
RESEARCH NOTE
Genetic study of scheduled caste populations of Tamil Nadu
M. VIJAYA*, S. KANTHIMATHI and A. RAMESH
 Department of Genetics, Dr ALM PGIBMS, University of Madras, Taramani, Chennai 600 113, India
Introduction
The present study aims to describe the genetic structure of the scheduled caste populations in Tamil Nadu state, and toassess their relationships with contemporary people of dif-ferent socio-economic groups of the state. We have stud-ied eight human-specific indels (insertion
 / 
deletion polymor-phisms) in DNA samples from five Tamil Nadu caste groupsusing phylogeneticand principalcomponentanalysis (PCA),and compared the data with earlier data on the three othercaste groups of the state. The results indicate that scheduledcastes are genetically heterogenous, but show closeness withother socio-economic groups of the state.The social structure of India is dominated by the Hinducaste system. Caste refers to a rigid system of ranked socialinequality with significant barriers to mobility or to intimateassociations between di
ff 
erentstrata. ‘Caste’ comes from thePortuguese word
casta
, meaning lineage, breed, race or cat-egory, and is described by one or both of the two ancientIndian concepts of varna and jati (Pillay 1979): ‘Studies inthe history of India with special reference to Tamil Nadu’.The caste (varna) system has four ranked categories:priests (Brahmins), warriors (Kshatriyas), farmers and mer-chants (Vaisyas), and labourers and servants (Sudras). How-ever, the status of the Sudra was actually higher than that of a fifth varna, the Panchama. This fifth varna was added at alater stage to include the so-called untouchables, who wereexcluded from the other four varna (Elder 1996). Based onsocio-economic hierarchy, the varna system can be classi-fied into three castes: upper (Brahmins), middle (Kshatriyas,Vaisyas andSudras)andlower(Panchama). IncontemporaryIndia, the governmentrefers to Panchama or untouchablesasscheduled castes.Scheduled caste members comprised about 139 millionpeople, or more than 16 per cent of the total population of 
*For correspondence. E-mail: vijim@mail.com.
India. About 10.7 million of them are distributed in TamilNadu state, the majority being Hindus (Census of India2001). Major scheduled castes of Tamil Nadu are Pallan,Paraiyan and Chakkiliyar, belonging to the Dravidian lin-guistic group.The present study was undertaken to delineate the ge-netic a
nity of scheduled castes among themselves and withother varnas of the Hindu caste system. We have genotypedthe study groups using seven human specific
Alu
insertions(Stoneking
et al
. 1997), a mitochondrial DNA insert in thehuman nuclear genome (Zischler
et al
. 1995), and a deletionof256bpofa 285-bp
 Alu
elementat the
CD4
locus(Edwardsand Gibbs 1992).
Alu
elements are stable genetic markersand their frequency distribution provides information on thedemographic history and migration patterns of human popu-lations. The knowledgeof both the ancestral state and the di-rection of mutational change facilitate phylogenetic analysis(Batzer
et al
. 1995). Earlier study data of other Tamil Naducaste groups(uppercaste: Iyer and Iyengar,lower caste: Pal-lan) were used for comparative analysis.
Caste groups
 Pallan
The Pallan are an ancient community,engagedextensivelyinwet landfarminganddistributedmainlyinThanjavur,Madu-rai and Ramanathapuram districts (Ramaiah 2004). The Pal-lan rank themselves highest among the scheduled castes, andnumber around 2.2 million (Census of India 2001).
 Paraiyan
The term Paraiyan is derived from the Tamil word paraimeaning drum, as some of them act as drummers at funeralsand village festivals. They are also engaged in cultivation,grass cutting and weaving. Fifteenth century literature indi-cates that theParaiyanswere also engagedin leather process-ing from dead animals, which is traditionally considered tobe defiling and polluting. They are also knownas Sambavan,
Keywords.
caste system; genetic a
nity; scheduled castes; socio-economic groups; Tamil Nadu; principal component analysis.
Journal of Genetics,
Vol. 87, No. 2, August 2008
171
 
 M. Vijaya et al.
Sambavar and Samban (Singh 1998), and their number wasestimated to be around 1.8 million (Census of India 2001).
Chakkiliyar
Chakkiliyars are also referred to as Arundhatiyar, Madari,Madiga and Pagadai, and represent the lowest strata evenamong the schedule castes. No literature is available to in-dicate their historical origin, although it is thought that theymight be immigrants from the neighbouring state of AndhraPradesh, as their mother tongue is Telugu. They are mainlya landless community. Their traditional occupations aresweeping, scavenging and removing dead animals, makingand repairing footwear, as well as working as farm labourers(Singh 1998). The size of the populationwas estimated to beover 0.7 million (Census of India 2001).
 Kallar
Kallars are known to be the oldest immigrants of Neolithicperiod with Mediterranean racial elements. Traditionally,they were described as semiagriculturists and semiwarriors.They are mostly agriculturists, and are mainly distributedin Thanjavur and Madurai districts and to a lesser extent inTiruchirapalli, Ramanathapuram, Tirunelveli and Pudukkot-tai districts (Singh 1998). The size of the population wasestimated to be over 0.8 million (Second Backward ClassesCommission of Tamil Nadu 1989).
 Maravar
They are largely distributed in the districts of Tiruchirapalli,Madurai, Kanniyakumari, Tirunelveli, Virudhunagar, Than- javur, Pudukkottai, Ramanathapuram and Sivaganga dis-tricts. Traditionally,theyareagriculturists(Singh1998). Thepopulation size was estimated around 0.53 million (SecondBackward Classes Commission of Tamil Nadu 1989).
 Agamudaiyar
Agamudaiyar of northern Tamil Nadu use the title ‘Mu-daliartitle whereasthoselivingin southernpartsuse thetitle‘Udayar’. They are distributed in the districts of Thanjavur,Nagapattinam, Tiruchirapalli and Pudukkottai (Singh 1998),and number around 0.82 million (Second Backward ClassesCommission of Tamil Nadu 1989).
Subjects and methods
The populations selected for the present study includesscheduled castes (lower caste: Paraiyan and Chakkiliyar),and middle castes (Kallar, Maravar and Agamudaiyar). Inaddition to these, upper caste (Iyer and Iyengar) and lowercaste (Pallan) groups, for which comparable data were avail-able and were also included for comparative analysis (Basu
et al
. 2003). All these populations belong to Dravidian lin-guistic group.Blood samples were drawn with informed consent, from266 unrelated individuals belonging to two scheduled caste:Paraiyan (
n
=
50) and Chakkiliyar (
n
=
47), and three mid-dle caste: Kallar (
n
=
54), Maravar (
n
=
68) and Agamu-daiyar (
n
=
47). Information about age, sex and caste pro-vided by the respondents were recorded. Institutional humanethical committee clearance was obtained prior to the collec-tion of blood samples. Paraiyansamples were collected fromCuddalore, Dharmapuri and Chennai, Chakkiliyar sampleswere from Salem and Erode districts, Kallar samples fromMadurai and Thanjavur, Maravar samples from Kammuthiand Agamudaiyar samples from Thanjavur district.DNA from the peripheral blood leucocytes was isolatedusing theprotocolof Miller
et al
.(1988). DNA samples werescreenedforeightindelconsistingofseven
 Alu
insertionsites(
 Alu APO
,
Alu CD4
,
Alu PV92
,
Alu FXIIIB
,
Alu ACE 
,
Alu PLAT 
and
 Alu D1
) andone mitochondrialDNA insertionsite(
mtNUC 
) (Viswanathan
et al
. 2004). These DNA markerswere selected by the Department of Biotechnology (DBT),Government of India as part of its Human Genome Diversity(HGD) initiative.The allele frequencies and their standard errors werecomputed by gene counting method. Chi-square tests forthe departure from Hardy–Weinberg expectations were per-formed. Average heterozygosity was calculated using theestimated allele frequencies for each population. DISPAN(genetic distance and phylogenetic analysis) software (Ota1993) was used to compute average heterozygosity and genediversity (Nei 1973). The software also constructs phylo-genetic trees (dendrograms) by using the neighbour-joiningmethod. Population relationships were also analysed viaPCA, using SPSS 11.0 version.
Results
 Allele frequencies
The
Alu
deletion allele (
) at
CD4
locus is human specificandits presenceis consideredto be theancestral state. Chim-panzees, gorillas, orangutans and gibbons are monomorphicfor the (
+
) allele at this locus, thereby indicating that thisdeletion event probably occurred after the divergence of hu-mans from the great apes about 4 to 6 million years ago(Tishko
ff 
et al
. 1996). The frequencies of these allele wereless than 11% in the study populations. The insertion allele(
+
) at
mtNUC 
locus is human specific and reported to haveoccurred before the migration of human populations fromAfrica (Zischler
et al
. 1995). The frequency of this allelevaried from 30.2% in Chakkiliyar to 61.8% in Iyengar. Nosignificant departure from the Hardy–Weinberg equilibriumwas observed (
 P
<
0
.
05).
 Average heterozygosity
The average heterozygosity was obtained for all loci alongwith three caste populations of Tamil Nadu for which thedata are available (Basu
et al
. 2003). The values rangedfrom 37.4% in Maravar to a maximum of 45% in Iyengar.The average heterozygosity values of the study populations172
Journal of Genetics,
Vol. 87, No. 2, August 2008
 
Genetic study of scheduled caste
are higher than the other global populations studied, withthe exception of African populations (Stoneking
et al
. 1997;Novick 
et al
. 1998). Thus, the DNA markers attest that thecaste populations of Tamil Nadu exhibit high levels of ge-nomic diversity.
Table 1.
Gene diversity analysis for individual loci and forall loci considered jointly.Loci
t
s
G
st
mtNUC 
0.4944 0.4727 0.0400
 Alu ACE 
0.4994 0.4795 0.0398
 Alu APO
0.3370 0.3254 0.0344
 Alu FXIIB
0.4997 0.4714 0.0567
 Alu D1
0.3994 0.3712 0.0707
 Alu CD4
0.1366 0.1341 0.0183
 Alu PLAT 
0.4959 0.4744 0.0432
 Alu PV92
0.4464 0.4265 0.0445All loci 0.4136 0.3944 0.0464
 H 
t
–Total genomic diversity among the populations.
 H 
s
–Diversity between individuals within population.
G
st
–Genomic diversity between population.
Genomic diversity between populations
The amount of genetic di
ff 
erentiation among populations,
G
ST
values (a measure of the interpopulation variability) foreach polymorphic locus were determined. The results of thegene diversity analysis separately for each locus and for allloci taken together are presented in table 1. It is seen that ex-cept for the
Alu CD4
locus (0
.
018) and
Alu APO
(0
.
034), thetotal genomic diversity (
 H 
T
) for each locus was quite high.The values of 
T
varied between 0.137 (
 Alu CD4
) to 0
.
499(
 Alu FXIIIB
). However, most of the genomic diversity is at-tributable to diversity between individualswithin the popula-tions (
 H 
S
). When all loci are jointly considered, 4.6% of thetotal genomic diversity is attributable between-populationsvariation.
Genomic a
  ffi
 nities among populations
The neighbour-joiningtree depictingthe populationrelation-ships is presented in figure 1. The eight populations (includ-ing three other caste populations of Tamil Nadu; Basu
et al
.2003)groupedin six clusters: (Kallar and Maravar), Agamu-daiyar, Chakkiliyar, Paraiyan, Pallan and (Iyer and Iyengar).Upper caste and middle caste populations formed separateclusters, while the scheduled caste populations fall on inde-pendent branches.PCA of the allele frequencies at the eight polymorphicIndel loci was also performed (figure 2). The two princi-pal components, explains about 94% of the allele variance,which provide the most information for a two-dimensionaldepiction of population relationships.
Discussion
The present study aimed at revealing the genetic relation-ship of the scheduled caste (lower caste) among themselvesand with other socio-economicgroups (upper and middle) inTamil Nadu.The Phylogenetic analysis illustrated that the clusteringpattern is similar to that of the di
ff 
erent socio-economicgroups, except for the lower caste (figure 1). Upper andmiddle caste populations showed close a
nity among them-selves, whereas, the scheduledcastes did notshow anyclose-ness. These findings demonstrated that the scheduled castepopulations of Tamil Nadu are genetically heterogenous.Tamil Nadu state was colonized by several waves of mi-grants during di
ff 
erent time periods. Based on morphologi-cal traits Malhotra
et al
. (1981) recognized the existence of three morphological types among the populations of TamilNadu. The Proto–Australoid features were identified inParaiyan, Chakkiliyar and Mutracha. People with these fea-tures, accordingtothem,mighthaveenteredtheTamilregionbetween 15,000–20,000 years ago. Paleo–Mediterraneanand Mediterranean features were identified in Kallar andVanniyar. The antiquity of Mediterranean populations has
Figure 1.
Unrooted neighbour-joining tree depicting genomic relationshipsamong 8 caste populations of Tamil Nadu. Number indicate (%) the fractionof 5000 bootstrap replicates that supported a particular grouping.
Journal of Genetics,
Vol. 87, No. 2, August 2008
173

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->