Kromosom Rearrangement

|
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.
 14 views
of 8

Please download to get full document.

View again

Description
aa
Share
Tags
Transcript
  TRENDSin Ecology & Evolution  Vol.16 No.7 July 2001 http://tree.trends.com0169–5347/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0169-5347(01)02187-5 351 Review Loren H.Rieseberg Dept of Biology, IndianaUniversity, Bloomington,IN 47405, USA.e-mail:lriesebe@indiana.edu Most species of plants and animals differ in their KARYOTYPES 1,2 (see Glossary). This observation,combined with evidence that chromosomalrearrangements might reduce the fertility of heterozygous hybrids (Box1), has led someresearchers to argue for a causative role forchromosomal change in SPECIATION 1,3 . For example,White concludes 1 that chromosomal rearrangementshave ‘played the primary role in the majority of speciation events’. The opposing and more widelyheld view is that the accumulation of chromosomaldifferences between populations is largely incidentalto speciation 4–6 .The most widely cited reasons for doubting animportant role for karyotypic change in speciationinclude: (1)the observation that many chromosomalrearrangements have little effect on fertility 5,7,8 ;(2)theoretical difficulties associated with fixingchromosomal rearrangements that are strongly UNDERDOMINANT (i.e. reduce the fitness of heterozygotes) 9,10 ; (3)the supposed ineffectiveness of chromosomal differences as barriers to gene flow 4,11,12 ;and (4)the widespread belief that premating and/orecological barriers arise earlier than chromosomalrearrangements in the speciation process and thusare more likely to cause speciation 6,13,14 .Here, I discuss prominent models of chromosomalspeciation and population genetic issues associatedwith the establishment and spread of chromosomalrearrangements, and consider the validity of thesemodels, particularly with respect to the effects of chromosomal rearrangements on the fitness of plantand animal hybrids and on interspecific gene flow. Iargue that chromosomal rearrangements reduce geneflow more often through their effects onrecombination rates than through their effects onfitness. I conclude by discussing the implications of this non-traditional view of chromosomalrearrangements with respect to: (1)traditionalchromosomal speciation models; (2) SYMPATRIC or PARAPATRIC models of speciation; and (3)the survival of  NEOSPECIES that have come back into contact withtheir progenitor(s). Models of chromosomal speciation There are many overlapping, largely untested,models of chromosomal speciation 3 (Box2). Thesemodels share one fundamental feature: chromosomaldifferences that have accumulated between theneospecies and its progenitor(s) are assumed toimpair the fertility or viability of interspecific hybrids(Box1), thereby reducing gene flow 15 . However,almost all other assumptions vary between models(Table1), including whether geographical isolation isrequired for speciation, the proposed means by whichchromosomal rearrangements arise and become fixedin populations, and the effects of individualrearrangements on the fitness of chromosomallyheterozygous individuals. In addition, some modelssuggest that adaptive differences associated with thechromosomal repatterning might allow theneospecies to colonize new habitats (quantum model)or to invade the habitat of the parental form(stasipatric model).The primary difficulty with most chromosomalspeciation models is that the fixation of stronglyunderdominant chromosomal rearrangementsthrough drift is unlikely, except in small, inbredpopulations 9,10,15,16 . The difficulties associated withfixing underdominant mutations are exacerbated by SYMPATRY . Thus, it is unsurprising that mostchromosomal models assume some sort of geographical isolation (Box2; Table1). Even hybrid or‘recombinational’speciation, which must be initiatedin sympatry, is most probable when the hybridneospecies becomes spatially isolated from itsparental taxa following a hybrid founder event 17 . MEIOTICDRIVE has been promoted as acomplementary mechanism to drift for the fixation of rearrangements (stasipatric model), but it appears tobe infrequent 18,19 . If chromosomal rearrangementsare neutral or only weakly underdominant, theconditions required for their fixation are relaxed, butthey are then less likely to reduce hybrid fitness. Onlyunder special conditions, such as those outlined in themonobrachial centric fusion model 20 , arerearrangements that were initially neutral likely tocause sterility in interpopulation hybrids. However, Several authors have proposed that speciation frequently occurs when apopulation becomes fixed for one or more chromosomal rearrangements thatreduce fitness when they are heterozygous.This hypothesis has littletheoretical support because mutations that cause a large reduction in fitnesscan be fixed through drift only in small,inbred populations.Moreover,theeffects of chromosomal rearrangements on fitness are unpredictable and varysignificantly between plants and animals.I argue that rearrangements reducegene flow more by suppressing recombination and extending the effects oflinked isolation genes than by reducing fitness.This unorthodox perspectivehas significant implications for speciation models and for the outcomes ofcontact between neospecies and their progenitor(s). Chromosomal rearrangements andspeciation Loren H.Rieseberg  there is some evidence that rearrangements that areweakly underdominant individually might bestrongly underdominant in combination 21 – therationale that underpins cascade and chain models of chromosomal speciation (Box 2; Table 1). Therecombinational model is immune to the problem of fixing underdominant rearrangements becausehydridization brings them to a frequency of 50 % anddrift becomes incidental to the process.Unsurprisingly, the two models (monobrachial centricfusion and recombinational) that lack the theoreticalproblems associated with the fixation of underdominant mutations also provide the mostconvincing examples of chromosomal speciation innature 22–24 .Arelated issue concerns the fitness of the novelchromosomal homozygote. If the new rearrangementis associated with a favorable gene complex,conditions for the establishment and spread of therearrangement are slightly relaxed. However, theadvantage with respect to establishment is presentonly when the population is small and selectionagainst the underdominant mutation is weak  10 . As aresult, homozygote advantage is rarely viewed as animportant feature in the establishment of chromosomal rearrangements, but it is necessary forthe migration of a new rearrangement into otherpopulations (stasipatric model) and the colonizationof habitats not occupied by the progenitor species(quantum model). With reference to therecombinational model (Table1), new genecombinations arising from hybridization are assumedto contribute to ecological divergence 17 , but the srcinof these new gene combinations might be incidental tochromosomal repatterning.Afinal consideration is the speed with which newchromosomal rearrangements arise and becomeestablished in populations 9,25 and the strength of theresulting sterility barrier. Most models rely onspontaneous mutation to drive chromosomalevolution but, in many taxa, this process will be slowrelative to the development of ecological isolatingbarriers or the accumulation of genic sterility factors.Also, significant isolation will probably requiremultiple rearrangements 11,12 , further slowing theprocess of chromosomal speciation. Two of the modelslisted in Table1 (saltational and recombinational)suggest means by which multiple chromosomalrearrangements are fixed simultaneously and veryearly in the speciation process. In the saltationalmodel 26 , inbreeding induces chromosomal breakage,but there is little quantitative evidence for enhancedchromosomal mutation rates in inbred populations.The recombinational model 4,27 postulates that rapidkaryotypic evolution occurs through the sorting of chromosomal rearrangements that differentiate the TRENDSin Ecology & Evolution  Vol.16 No.7 July 2001 http://tree.trends.com 352 Review There are several kinds of chromosomalrearrangement. These include the fusion orfission of chromosomes, the duplication ordeletion of a chromosomal segment, theinversion of a segment and thetranslocation of segments between non-homologous chromosomes. In individualsthat are heterozygous for one or more ofthese rearrangements, recombinationbetween chromosomes that differ for therearrangement(s) often generates UNBALANCEDGAMETES (see Glossary) thatmight themselves die or that cause zygotesto die. In an example of a PERICENTRICINVERSION (Fig.I; reproduced, with permission, fromRef.a), recombinant chromosomesresulting from a single crossover contain aduplication and a deficiency (small circlesindicate centromeres). Gametes or zygoteswith a recombinant chromosome might beinviable, leading to the selective recovery ofnon-recombinant chromosomes in viableoffspring and an effective reduction inrecombination. Reference aGriffiths, A.J.F. et al. (1993)  An Introduction toGenetic Analysis (5th edn), W.H. Freeman & Co. Box 1.Chromosomal rearrangements and meiosis TRENDS in Ecology & Evolution  ABCDACBDPairingAACBDDBCrossoverSegregationEnd of meiosis IEnd of meiosis IIAABBCDCADDBBCDCAABCDABCANormal productDuplication A armDeletion D armDBCDDBCADuplication D armDeletion A armInversion product I  parental species and the accumulation of additionalrearrangements induced by hybridization. Both of these mechanisms have contributed to karyotypicchange in a strongly isolated hybrid sunflowerspecies,  Helianthus anomalus 24 .This discussion illustrates the paradoxtraditionally associated with chromosomalspeciation 12 . Only rearrangements that are stronglyunderdominant are considered likely to contribute tospeciation, but these kinds of rearrangements areexceedingly difficult to fix in natural populations. Themonobrachial centric fusion and recombinationalmodels are immune to this particular difficulty, butthey are likely to be rare for other reasons. However,the effects of chromosomal rearrangements on geneflow – the parameter that really matters – are notnecessarily strongly correlated with their effects onfertility. Moreover, in certain organismal groups,chromosomal sterility might evolve with sufficientrelative speed to play an important role in thesurvival of neospecies that have come into contactwith their progenitor(s). TRENDSin Ecology & Evolution  Vol.16 No.7 July 2001 http://tree.trends.com 353 Review Many models of chromosomal speciationhave been published over the past 50years. Some of the most prominentmodels include:ãChain or Cascade models, whichassume that REPRODUCTIVEISOLATION (seeGlossary) arises via the accumulation ofchromosomal rearrangements that areindividually weakly underdominant a .ãThe Chromosomal Transilience model,which suggests that a stronglyunderdominant chromosomalrearrangement might become fixedthrough drift and inbreeding in anisolated population. REINFORCEMENT withrespect to the ancestral KARYOTYPE mightcomplete speciation b .ãThe Monobrachial Fusion model,whichproposes that isolatedsubpopulations becomeindependently fixed for differentcentric fusions, which individuallycause little or no loss of fertility whenheterozygous. However, hybridsbetween the two subpopulationswould be intersterile becausedifferent combinations of chromosomearms had been fused in the twosubpopulations c .ãThe Recombinational model ,whichdescribes a process in whichhybridization between chromosomaldivergent populations leads tochromosomal breakage and to thesorting of preexisting rearrangementsthat differentiate the parental species.A new recombinant genotype couldbecome stabilized if it is sufficientlykaryotypically divergent (and therebyreproductively isolated) from eitherparental species d .ãThe Quantum speciation model,which suggests that chromosomalrearrangements might become fixedvery rapidly in a peripheral founderpopulation through drift andinbreeding, leading to reproductiveisolation. This model is similar to thechromosomal transilience modelexcept that the new gene arrangementsresulting from karyotypic change arethought be adaptive d .ãThe Stasipatric model, which assumesthat a strongly underdominantchromosomal rearrangement arises andbecomes fixed in a population that iswithin the range of the progenitorspecies. Unlike other models, thestasipatric model postulates an importantrole for meiotic drive in the fixation ofchromosomal rearrangements a .ãThe Saltational model, which proposesthat inbreeding in a peripheral founderpopulation could induce chromosomalbreakage. However, as in most othermodels, chromosomal rearrangements(which later serve as isolating barriers)are fixed through drift in small, inbredpopulations e . References aWhite, M.J.D. (1978)  Modes of Speciation , W.H.Freeman & Co.bTempleton, A.R. (1981) Mechanisms of speciation – a population genetic approach.  Annu. Rev. Ecol. Syst. 12, 23–48cBaker, R.J. and Bickham J.W. (1986)Speciation by monobrachial centric fusions. Proc. Natl. Acad. Sci. U. S. A. 83, 8245–8248dGrant, V. (1981) Plant Speciation , ColumbiaUniversity PresseLewis, H. (1966) Speciation in floweringplants. Science 152, 167–172 Box 2.Models of chromosomal speciationTable 1.Differences among chromosomal speciation models ModelGeographicalMutational originsFitness of individualMeans ofChromosomalRefsisolation?rearrangementsestablishmentrepatterningadaptive? Chain/CascadeYesSpontaneousWeak underdominanceDriftNo1rearrangementChromosomalYesSpontaneousStrong underdominanceDriftNo27transiliencerearrangementMonobrachialYesSpontaneousWeak or noDriftNo20centric fusionrearrangementunderdominanceRecombinationalProbablyHybridizationWeak or strongFertility selectionMaybe2underdominanceQuantumYesSpontaneousStrong underdominanceDriftYes2rearrangementStasipatricNoSpontaneousStrong underdominanceDrift/meiotic driveYes1rearrangementSaltationalYesInbreedingStrong underdominanceDriftMaybe26  Chromosomal rearrangements as isolatingmechanisms The impact of chromosomal rearrangements onhybrid fertility or viability is generally assumed to besynonymous with their effectiveness as barriers togene flow. However, this is not necessarily the case.Rather, we need to consider two separate issues: (1)dochromosomal rearrangements affect hybrid fitness;and (2)can chromosomal rearrangements affect geneflow through mechanisms other than reduced hybridfitness? Effects on hybrid fitness  It has long been recognized that different kinds of chromosomal perturbations vary in their effects onfitness. For example, TRANSLOCATIONS , FUSIONS , FISSIONS and INVERSIONS are typically viewed asunderdominant mutations, whereas HETEROCHROMATIN additions and deletions are not. Even supposedlyunderdominant rearrangements are unpredictable intheir fitness effects because of the mechanisms thatalleviate or prevent malsegregation at meiosis, suchas partial or complete suppression of recombination 8 .There are two other relevant complications. First,it can be extremely difficult to distinguish betweenthe effects of chromosomal rearrangements on hybridsterility from those of genes 28 . For example, contraryto cytogenetic predictions, hybrids of chromosomallysimilar species sometimes exhibit abnormal meioticpairing, whereas hybrids of chromosomally divergentspecies sometimes pair normally. Second, the effectsof the same kinds of rearrangements appear to varyacross organismal groups. In plants, for example,most rearrangements appear to have large effects onfertility 29 , whereas in animals, karyotypicheterozygosity seems less likely to have negativefitness consequences 5,8 .This apparent difference between plants andanimals with respect to chromosomal sterility wasfirst recognized by Dobzhansky 7 , who noted that thedoubling of the chromosomal complement in planthybrids typically led to a complete restoration of fertility. In  Drosophila hybrids, however, chromosomaldoubling failed to restore pairing or fertility. Asexplained by Dobzhansky, chromosomal doublingfurnishes an exact homolog for each chromosome inthe hybrid genome, thereby restoring pairing andfertility in chromosomally divergent hybrids.However, chromosomal doubling should have no effecton the action of complementary genes, so genicsterility is preserved. Based on this reasoning, sterilityin  Drosophila was interpreted as being caused bygenes, whereas sterility in the referenced planthybrids was assumed to be chromosomal in srcin.The conclusion that chromosomal rearrangementsare more likely to contribute to the sterility of plantsthan to the sterility of animals is reinforced byadditional evidence that has accumulated since theseclassic experiments 29,30 . This includes a much longerlist of sterile plant hybrids that recover full fertilityupon chromosomal doubling 29 , as well as evidencethat sterility in plants often maps to chromosomalrearrangements 31 . By contrast, sterility in animals(i.e.  Drosophila and platyfish) has been mapped togenes 6,32 . Asurprising observation is that sterile planthybrids with normal pairing sometimes recoverfertility following chromosomal doubling 29 . Sterilityin these hybrids is interpreted as resulting from‘cryptic structural differentiation’. This has led to aparadoxical situation in which the burden of proof with respect to plant sterility is on those whointerpret it as resulting from the action of genes,whereas for sterility in animals, the burden of proof falls onto those who argue for a chromosomal basis.So why does chromosomal sterility appear to bemore important in plants than in animals? Onepossibility relates to the observation that most genesare expressed in the male gametes of plants 33 , but notof animals 34 . As a result, pollen carrying a deletion asa result of chromosomal irregularities will probablyabort, whereas sperm are typically unaffected 35,36 .Female gametes of both plants and animals tend to betolerant of deletions because of contributions of mRNAs and proteins from surrounding maternaltissue 36,37 . Although chromosomal deletions can causethe death of diploid offspring, many deficiencies arerescued by genes from the alternative gamete innewly formed zygotes. This reduces the fitness losscaused by chromosomal rearrangements, or at leastdefers the loss to later generations, making it moredifficult to associate fitness reductions withchromosomal rearrangements.Asecond possible explanation relates to theprevalence of differentiated sex chromosomes inanimals versus plants. Most animals are dioeciousand possess a degenerate Y chromosome, whereasmost plants are hermaphroditic. Even in the smallfraction of plants that are dioecious, Y-degeneration isminimal 38 . The degenerate Y makes most animals,but not plants, subject to HALDANE ’ SRULE , which statesthat if only one hybrid sex is sterile or inviable,it is always the heterogametic sex 39 . The acceptedexplanations for Haldane’s rule are that mostX-linked alleles that cause hybrid problems arepartially recessive 40,41 , and that genes that contributeto male sterility evolve more quickly than do thosethat cause female sterility 42 . The important pointhere is that genic sterility will evolve most rapidly inorganisms with a large X chromosome and a verydegenerate Y, such as  Drosophila . By contrast, genicsterility is likely to evolve slowly in plants or inanimals that lack a degenerate Y or that have only asmall number of genes on the X chromosome. Rapidmale evolution also seems most probable inorganisms with differentiated sex chromosomes andthereby less constrained in their response to SEXUALSELECTION 43 . The bottom line is that, althoughkaryotypic changes might accumulate at similar ratesin plants and animals, they will probably make adisproportionately large contribution to sterility in TRENDSin Ecology & Evolution  Vol.16 No.7 July 2001 http://tree.trends.com 354 Review
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