Matrilineal genetic structure within and among populations of the cooperatively breeding common marmoset, Callithrix jacchus

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report
Category:

Legal forms

Published:

Views: 48 | Pages: 8

Extension: PDF | Download: 0

Share
Description
Matrilineal genetic structure within and among populations of the cooperatively breeding common marmoset, Callithrix jacchus
Transcript
   Molecular Ecology (2003) 12  , 1101–1108© 2003 Blackwell Publishing Ltd  BlackwellPublishingLtd.  SHORT COMMUNICATION  Matrilineal genetic structure within and among populations of the cooperatively breeding common marmoset, Callithrix  jacchus  C. G. FAULKES  *  , M. F. ARRUDA  †  and M. A. O. MONTEIRO DA CRUZ  ‡  *  School of Biological Sciences, Queen Mary, University of London, Mile End Road, London E1 4NS, UK, †   Federal University of Rio Grande do Norte, Brazil, ‡   Federal Rural University of Pernambuco, Brazil Abstract Common marmosets are members of the family Callitrichidae, South American primatescharacterized by highly social group living and cooperative breeding. In this study we ana-lysed 1112 base pairs (bp) of the mitochondrial control region in 59 Callithrix jacchus   indi-viduals, sampled mainly from two geographically distinct field sites in N.E. Brazil.Analysis of molecular variation revealed a highly significant genetic structuring of haplo-types between social groups and between populations. Examination of matrilineal geneticstructure within social groups revealed that seven of nine recorded breeding pairs werefrom different maternal lineages, indicating assortative mating and outbreeding. In addi-tion to the breeders, at least six of 10 groups contained adult individuals from differentmatrilines, with five haplotypes present in one social group of nine animals. Groups of mixedlineages raise questions about potential reproductive conflicts of interest, and the extent ofkin-selected altruism in the evolution and maintenance of cooperative breeding in this species.  Keywords  : Callithrix jacchus  , Callitrichidae, cooperative breeding, kin structure, mitochondrialDNA, phylogeography Received 29 May 2002; revision received 6 January 2003; accepted 6 January 2003  Introduction  Marmosets and tamarins of the Family Callitrichidae aresmall arboreal primates endemic to the northern half of SouthAmerica. The four genera of 40 or so species are distributedwidely in a variety of primary and secondary foresttypes. Within the family, cooperative breeding strategies arewidespread and virtually all species are characterized bysmall territorial groups of approximately 4–15 individuals,where reproduction is monopolized by one or a smallnumber of dominant individuals of each sex (high repro-ductive skew; for review see French 1997; Tardif 1997).The social and reproductive system of common marmosets,  Callithrix jacchus  , is typical of the Callitrichidae. Groupscontain 3–15 individuals, and typically a single dominantfemale breeds, normally producing dizygotic twins. However,plural breeding among females can occur (Digby & Ferrari1994; Nievergelt et al  . 2000), and mating systems can bevariable both within and between callitrichid species. Poly-andry, polygyny and monogamy have all been reportedfrom behavioural observations (Ferrari & Digby 1996).Nievergelt et al  . (2000) used microsatellite genotyping toinvestigate three social groups of common marmosets inNisia Floresta, Brazil, and found two breeding females pergroup mating with mainly a single dominant male (poly-gynmonandry). However, the presence of more than one breeding female per group may be atypical (M. F. Arruda  et al  . unpublished; Rothe & Darms 1993). Irrespective of mating system, twin offspring are usually produced incallitrichids, and the main role of nonreproductive helpersin the group is to assist in the care of the breeding female’soffspring. This is principally by sharing the burden of carry-ing the relatively bulky twin infants around their arborealhabitat, although food provisioning may also occur (Tardif   et al  . 1993). Determining the genetic structure and patterns  Correspondence: Chris G. Faulkes. Fax: 020 89830973; E-mail: c.g.faulkes@qmul.ac.uk   1102  C. G. FAULKES, M. F. ARRUDA and M. A. O. MONTEIRO DA CRUZ  © 2003 Blackwell Publishing Ltd,  Molecular Ecology  , 12, 1101–1108  of relatedness within social groups is thus crucial in under-standing the role of kin selection in the alloparental behaviour observed in marmosets, where these nonbreed-ing helpers forego their own reproduction and incur ener-getic costs in altruistic behaviour (Tardif 1997; Sánchez  et al  . 1999; Bales et al  . 2000). To date, only one study hasquantified relatedness and has supported behaviouralobservations suggesting that most group members areclose relatives (Nievergelt et al  . 2000).Little is known of the intraspecific population genetics of any of the Callitrichidae, and understanding the broaderpopulation genetic structure and phylogeography of specieswill further our understanding of their social behaviourand have implications for their conservation. Molecularphylogenetic analysis of the genus Callithrix (  marmosets) by Tagliaro et al  . (1997) provides strong support for a divi-sion into Amazonian and Atlantic forest clades, the latterincluding the common marmoset. The common marmosetand its sister species the black-eared marmoset (  Callithrixpenicillata  ) are perhaps two of the most adaptable of theAtlantic forest callitrichid primates, and also exploit morehostile, seasonal habitats such as the caatinga (semi-aridthorn scrub). Although feeding on the exudate of tree bark(gum) supplements a diet of fruit and insects in all speciesof the genus, the trait of tree gouging and gum feeding ismost highly developed in C. jacchus  and C. penicillata  . Thisincreased specialization undoubtedly enables them to sur-vive in habitats where fruit may be scarce for long periodsof time (Rylands 1984; Ferrari 1993), and may thereforealso influence patterns of dispersal, gene flow and geneticstructuring of populations. We report here the first exten-sive intraspecific genetic study of a callitrichid primateusing mitochondrial control region sequence variation toinvestigate micro- and macrogeographical genetic struc-turing of populations. Specifically we aim to (i) quantifythe matrilineal structure of social groups to unambigu-ously examine kin structure, and (ii) determine therelationships of mitochondrial sequences in the context of population structuring and phylogeographical history.  Materials and methods  Study sites and samples  Wild common marmosets in two areas of N.E. Brazil(Tapacura and Nisia Floresta) which have been the subjectof ongoing long-term behavioural studies were examinedin detail. Nisia Floresta experimental EFLEX-IBAMAforestry station is located 45 km south of Natal, the statecapital of Rio Grande do Norte. Of a total area of 180 ha,80 ha are secondary Atlantic coastal forest, while 40 ha arecomposed of experimental plantations of pine, eucalyptus,coconut and commercial species (Santee & Arruda 1994).Study groups occur in both areas of the forest and two of the four social groups investigated here (Belém and Chui)were observed continuously from 1991 to 1996. TapacuraEcological Field Station lies within an isolated 390 ha blockof secondary Atlantic Coastal Forest, surrounded by sugarcane plantations. The vegetation is tropical semideciduouswith areas of dense undergrowth, but in places the forestis broken up with smallholdings, grassland and somenonindigenous trees and shrubs. Five main social groupswere studied in detail, together with two peripheral animals.Individuals were chosen for analysis from a large sampleset to reflect social structure during a particular timeperiod, as the composition of groups can be dynamic. ForNisia, samples were taken from the period 1995–97, and forTapacura 1995–96. Further samples were obtained from onecolony from the Botanical Gardens in Recifé, approx. 40 kmeast of Tapacura, and two published sequences obtainedfrom animals collected at Extremos, 15 km north of Natal andapproximately 45 km distant from Nisia (Tagliaro et al  . 1997).The locations of these sites in Brazil are shown in Fig. 1.Skin biopsies of approximately 2 mm diameter weretaken from the ear during routine capture of animals.These were placed in alcohol and stored at −  20 °  C. Altern-atively, or in addition to the biopsies, 30–40 hairs wereplucked, sealed into bags and stored at −  20 °  C.   Mitochondrial DNA analysis  DNA was extracted from hair samples using the proceduredescribed for buccal cells with the GFX™ Genomic BloodDNA Purification Kit (Amersham Pharmacia Biotech), andfrom the tissue samples using a standard phenol:chloroformmethodology. Polymerase chain reaction ( PCR) amplifica-tion of the control region of the mitochondrial DNA wasachieved using combinations of five ‘universal’ primers (A,B, C, D and G) and a standard PCR protocol, as describedfor African mole-rats (Faulkes et al  . 1997).Initially, sequencing of double-stranded DNA productswas performed manually using a Sequenase kit (UnitedStates Biochemical), and products were separated using a6% polyacrylamide gel. Sequencing was carried out in bothdirections using combinations of all the above primers toobtain complimentary partially overlapping strands. Forsome samples, and to replicate some manual sequencing,an Amersham MegaBace 1000 automated sequencer wasused to separate fragments produced using the DynamicET Terminator Sequencing kit (Amersham). Sequenceshave been deposited in NCBI with Accession nosAY196755–AY196775.   Analysis of mitochondrial DNA sequences  Sequences were compiled for analysis and alignedmanually using MacClade version 3 (Madison & Madison1992). Maximum parsimony was performed using both the   MATRILINEAL STRUCTURE OF COMMON MARMOSET POPULATIONS  1103  © 2003 Blackwell Publishing Ltd,  Molecular Ecology  , 12, 1101–1108   branch and bound and heuristic search options in paup  *version 4.0b10 (Swofford 2001), with all characters havingan equal weighting and gaps treated as a ‘fifth base’. Priorto maximum likelihood analysis, modeltest  3.06 (Posada& Crandall 1998) was used to establish the evolutionarymodel most appropriate for the data, and these parametersthen used in paup  *. modeltest  was also used to performa likelihood ratio test on trees generated with and with-out a molecular clock enforced. Bayesian analysis of thesequence data was performed using MrBayes (Huelsenbeck& Ronquist 2001), with starting trees generated bothrandomly and by prior neighbour-joining analysis with  paup  *, which was also used to construct a consensus treefrom the resulting data output. The published sequence for  C. penicillata  (Tagliaro et al  . 1997) was used for outgroupcomparison in all phylogenetic analyses.The frequency of haplotypes and proportion of sequencevariation (calculated from pairwise distances) amongsocial groups within regions and among regions wasinvestigated using analysis of molecular variance (  amova  ;Excoffier et al  . 1992). amova  produces estimates of variancecomponents and  F  -statistic analogues Φ   ST  , Φ   SC  and Φ   CT  ,which describe variation at the following levels, respect-ively: within colonies, among colonies within populationsand among populations.  Results   Haplotype diversity  Contiguous mitochondrial control region sequences,1112 base pairs (bp), were obtained from 57 individuals.Inclusion of two published sequences from Tagliaro  et al  . (1997) gave a total of 59 sequences from which 23distinct haplotypes were identified as follows (no.individuals/no. social groups in parentheses): NisiaFloresta, 11 haplotypes (27/4); Extremos, two haplotypes(2/–) (Tagliaro et al  . 1997); Botanical Gardens, Recifé, twohaplotypes (2/1) and at Tapacura, eight haplotypes (28/6).While haplotype diversity was relatively high, consideringparents and offspring were included in the data set (23haplotypes in 59 animals sampled), rates of divergence between them were low, with uncorrected genetic dis-tances ranging from 0.09 to 1.99%. HKY85 correctedvalues ranged from 0.09 to 2.03%. as follows: within Nisia,0.72 ±  0.06%, n  = 55 pairwise comparisons; within Tapa-cura, 0.84 ±  0.08%, n  = 28; between Nisia and Tapacura,1.43 ±  0.02%, n  = 88. Comparing C. jacchus  with the outgroupspecies C. penicillata  , the average HKY85 corrected gene-tic distance was 6.36 ±  0.06% (6.02 ±  0.05% uncorrected;  n  = 23).   Matrilineal structure of social groups  The distribution of haplotypes within and among groupsat Nisia and Tapacura is shown in Table 1.It was not possible to obtain samples from every animalin all groups, but some missing data were inferred fromknown relationships between individuals and the establishedpattern of mitochondrial DNA inheritance: mothers andoffspring should have the same mitochondrial haplotype,assuming no mutation or paternal leakage. In all 13 casesof samples from known mothers (  n  = 8) and offspring andsiblings/twins that were sequenced, haplotypes wereidentical.In seven of the 10 groups the data were sufficientlycomplete to compare the haplotypes of breeding indi-viduals. Within these seven groups, of nine recorded breeding pairs seven were found to be from differentmaternal lineages. Table 1 shows clearly that in addition tothe breeding male, at least six of the 10 groups also containindividuals from different matrilines, up to a maximum of five in the case of Group 4 in Nisia. Although sample sizewas relatively small, our results suggest no apparent sexdifferences in animals of differing haplotypes, indicatingthat dispersal of both sexes occurs in common marmosets. Fig. 1 Sampling map showing the species range for C. jacchus (shaded area on main map), and the relative locations of the foursampling sites in N.E. Brazil (inset).   1104  C. G. FAULKES, M. F. ARRUDA and M. A. O. MONTEIRO DA CRUZ  © 2003 Blackwell Publishing Ltd,  Molecular Ecology  , 12, 1101–1108  Table 1  Distribution of control region haplotypes and social group structure of common marmoset populations at (a) Nisia and (b)TapacuraSocial group (sampling period)AnimalStatusHaplotype(a)BelémBrejeiroBreeding male17(May 1996)BratAdult male17BhaskaraImmigrant male (joined April 1996)17BonitaBreeding female18Bia  9   Juvenile of Bonita18  Babi  8   Juvenile of Bonita18Breno  9   Juvenile of Bonita18Beta  8   Juvenile of Bonita18Beto   9   Juvenile of Bonita18Bethoven  8   Juvenile of Bonita18  ChuiChrisAdult male (ex breeding male)17(September 1995)CazuzaAdult male (breeder with Betty)13Colega 9  Adult female (Catrina†)13Cacá  8  Adult female (Catrina†)13ChiquitaAdult female (Clara†)13  CaluaAdult female (Clara  †  )13BettyImmigrant female17  CézarJuvenile (Betty)17  CarlaJuvenile (Betty)17  EspanhaEnriqueBreeding male22(April 1997)EricAdult male23EduadoJuvenile male (Estela); Expelled by Enrique24ElviraAdult female; Expelled by Goeth22AdrianaBreeding female25  ElaineJuvenile female (Adriana)25ElainaJuvenile female (Adriana)  25  EdaJuvenile female (Adriana)  25Group FourGandhiAdult male (ex breeder)16(September 1995)GreciaBreeding female21GoethBreeding male (August 96)21GiocondaAdult female14GabrielaAdult female23GustavoAdult male15GracaAdult femaleGilda  9   Juvenile female (Grecia)21Gardelia  8   Juvenile female (Grecia)21(b)EsparancinsUrsulaBreeding female10(1995)Tiana  9  Adult female (Ursula)10Sebastian  8  Adult male (Ursula)10  Pamela  9  Adult female (Ursula)10  Xorona  8   Juvenile female (Ursula)10  Zapata  9   Juvenile female (Ursula)10Fatima  8   Juvenile female (Ursula)10Edinilza   9   Juvenile female (Ursula)10Leticia   8   Juvenile female (Ursula)10ClovisJuvenile male (Ursula)10Sao PauloBreeding male?10LiJuvenile male (Ursula)10  Estovadins (Est)QueridaBreeding female7(1995)ZorroBreeding male? 9  Emigrated6MarceloBreeding male? 8  6Xuxa 9   Juvenile female (Querida)7  Irineu   8   Juvenile male (Querida)7  ElainePeripheral female10  IkuJuvenile male (Querida)7   MATRILINEAL STRUCTURE OF COMMON MARMOSET POPULATIONS  1105  © 2003 Blackwell Publishing Ltd,  Molecular Ecology  , 12, 1101–1108  In one group from Tapacura (Esparancins), all groupmembers were from the same matriline.  Population genetic structuring and phylogeographical trends  We used amova  to contrast the molecular variance forcontrol region haplotypes (including the inferred haplo-type data in Table 1), both within and among social groupsand among geographical locations (Nisia Floresta, Extremos,Recifé, Tapacura). Using a null distribution generated by1000 permutations of the data set, there was a highlysignificant genetic structuring both between social groupswithin populations (  Φ   SC  = 0.36; P  < 0.001) and betweenpopulations (  Φ   CT  = 0.48; P  < 0.001). Of the total variance,48% was explicable among the geographical locations,19% among social groups within locations and 33% withinsocial groups. At the geographical level, all the haplo-types identified in the study were restricted in their dis-tribution to their respective regions, with none shared between the four different geographical locations. Thissuggests that dispersal over long distances is not occurring.Following hierarchical likelihood ratio tests of differentmodels of evolution, the one found to best fit the data cor-responded to the HKY85 + G  + I model of sequence evolu-tion (  P  < 0.0001). An additional ratio test using the log-likelihood scores of the trees obtained using this modelwith and without a molecular clock enforced was not sig-nificant (2  δ = 34.36; d.f. = 23; P  = 0.052). The phylogramobtained is shown in Fig. 2.A similar model of evolution to that used for maximumlikelihood was also used in the Bayesian analysis. A totalof 1 000 000 generations were run in MrBayes, with likeli-hood scores stabilizing after 30 000 generations. Trees weresaved every 100 generations to give a total of 10 000, of which the first 300 were rejected (corresponding to thoseobtained before the likelihood scores had stabilized).Of the 1112 characters used, 1019 were constant and 93were variable. Of the latter, 38 were parsimony informa-tive, while 55 characters were uninformative. Nine equallyparsimonious trees were produced that differed only inthe branching order of the Recifé and Extremos cladesand haplotype ‘25’ relative to each other and the mainTapacura and Nisia clades. Support for each internal  Esmirradins (Esm)Ilka MariaBreeding female9(1996)MarcianaBreeding female9KeridaEx breeding female6Professor RobertoBreeding male?10OquéiBreeding male?6Linneau 9  Adult male (Kerida)6  Fernanda  8  Adult female (Kerida)6  NelsonJuvenile male (Ilka/Marciana?)8Vexame 9   Juvenile male (Ilka/Marciana)9Wilma  8   Juvenile female (Ilka/Marciana)9Escomprimidins (Esc)WilliamDominant male10(1995)Tonheta 9   Juvenile male (Ute)5  Valente  8   Juvenile male (Ute)5WaldemarJuvenile male (Wanda)?Wanda  Breeding femaleNo sampleEspivitadins (Esv)ClaudinhaBreeding female10(1996)QuideJuvenile male (Claudinha)10LúcioMauroBreeding male?10YpsiloneBreeding male?4  Isabel JoaquinaJuvenile female (Claudinha)10BárbaraAdult female (Claudinha)10FernandoAdult male (Claudinha?)?HugoJuvenile male (Claudinha?)? Alagadins (Ala)616Peripheral male10(1995)620Peripheral female11†Animal dead or missing. Data from three individuals in italics were sampled ‘outside’ of the temporal scheme and not included in the analysis. Bold type indicates that the sample was not sequenced, and the haplotype inferred from known mother–offspring or sibling relationship. Known twins are joined by parentheses.Social group (sampling period)AnimalStatusHaplotype Table 1 Continued
Recommended
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x