Documents

Rare Inherited Variation in Autism: Beginning to See the Forest and a Few Trees

Description
1-s2.0-S0896627313000548-main
Categories
Published
of 3
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.
Related Documents
Share
Transcript
  Neuron Previews Rare Inherited Variation in Autism:Beginning to See the Forest and a Few Trees Jason L. Stein, 1 Neelroop N. Parikshak, 1 and Daniel H. Geschwind 1, * 1 Program in Neurogenetics, Department of Neurology and Center for Autism Research and Treatment, Semel Institute, David Geffen Schoolof Medicine, University of California Los Angeles, Los Angeles, CA 90095, USA *Correspondence: dhg@mednet.ucla.eduhttp://dx.doi.org/10.1016/j.neuron.2013.01.010 In this issue of   Neuron , two papers ( Lim et al., 2013; Yu et al., 2013 ) use whole-exome sequencing (WES) to elucidate the contribution of inherited variation to the risk for autism by leveraging the increased penetranceofhomozygousandcompoundheterozygousrarevariantsinautosomesandhemizygousrarevariantsinthe Xchromosome of males. Together, they expand our knowledge about the geneticarchitecture of ASD, verifypreviouslyidentifiedgenes,andidentify novel mutationsthatwillguidethediscoveryofthecriticalbiologicalprocesses disrupted in autism.  Autism is a spectrum of neurodevelop-mental disorders (referred to as autismspectrumdisorder, ASD) affecting around1 in 88 individuals (2012 CDC estimate).Genetic risk factors for ASD playa substantial role and come in manyforms: those transmitted from parents toaffected child or those appearing denovo in the germline, as well as thosefound commonly in the population, orthose only rarely observed ( Figures 1and 2 ). Genetic investigations have re-vealed that hundreds of genomic loci arelikely involved and it is now important toadvance our understanding of ASD’sgenetic architecture while simultaneouslyidentifying specific deleterious variantsto understand the biological pathwaysinvolved ( Berg and Geschwind, 2012 ).Inherited variation from both commonand rare alleles provides the largestgenetic contribution to population-wide ASD risk, explaining   40% (95% confi-dence interval: 8%–84%) of the risk fordeveloping ASD ( Hallmayer et al., 2011 ).Though common variants are estimatedtobealargedrivingfactorforthedisorder(40%–60% variance explained; Klei et al.,2012 ), the effect size of individualcommon variants is small (estimated tobe odds-ratio < 1.2; Anney et al., 2012 ).This observation has led to a search forrare variants, which may exhibit largerindividual effect sizes.To this end, cost-effective high-throughput genome sequencing of ASDpatients analyzed in paradigms wherethe effect of a mutation can be seen overrandom chance plays a critical role ( Iossi-fovetal.,2012;Nealeetal.,2012; O’Roak et al., 2012; Sanders et al., 2012 ). However, this process is complicated bythe fact that each of us inherits over 100nonsense, loss-of-function mutations inour genomes, leading to about 20 com-pletely inactivated genes and severaldozen de novo variants, some of whichmay be functional, but none of which areclearly associated with disease ( MacAr-thur et al., 2012 ). In the latter category,predicted protein structure altering rarevariants have been observed to be morefrequent in ASD cases, and recent worksuggests nontransmitted, de novo muta-tions that delete genic regions, perturbsplicing, or truncate protein productsmay contribute to the development of  ASD in 15%–20% of cases ( Devlin andScherer, 2012 ). Although inherited muta-tions have been identified in rare families,no population-based exome sequencingstudy has demonstrated a significant rolefor inherited deleterious variants in ASDrisk, leaving the contribution of this classof genetic variation unknown. Now, asshown in two papers in this issue of  Neuron  ( Lim et al., 2013; Yu et al., 2013 ), it is clear that ASD risk is increased whentwo rare variants deleteriously affectbothcopiesofaproteincodinggene,con-sistentwitharoleforrecessiveinheritanceof nonsynonymous mutations in ASD.Starting with a population-based ap-proach, Lim et al. (2013) performed WESin933ASDcasesand869controls,allow-ing them to see the complete genomesequence in protein-coding and flank-ing regions. They focused on the mostdamaging type of mutations (nonsenseand essential splice site) transmittedfrom the mother and father to the child.In autosomes, they studied loci wherethe mutations truncate both copies of a protein, and in the X chromosome inmales, they studied mutations that trun-cate the only copy. Only rare alleles of this type (minor allele frequency  % 5%)were studied, based on the hypothesisthat double mutations commonly foundin the population are not pathological.These rare, putative recessively actingmutations were observed twice as manytimes in ASD patients as controls in auto-somes(6% of cases and 3.3%of controlscarried such a mutation). When overlap-ped with a data set of brain gene expres-sion, the overall odds-ratio (OR) for ASDincreased to 2.7. Importantly, this enrich-ment of double mutation variants wasreplicated in an independent data set,thistimeshowingasmaller,butstillsignif-icant effect (7.6% of cases and 5.5% of controls). The authors estimate an overall3% contribution to risk for ASD from thisclass of mutations. Rare hemizygousmutations on the X chromosome, whichwould also be depleted of the normalprotein in males, were also enriched inmale ASD cases compared to controls(4.8% versus 3.1%), revealing involve-ment in 1.7% of male ASD cases. Thisstudy fills an important gap in our knowl-edge of the genetic architecture of ASDbyestimatingthatabout5%ofASDcasesmay be affected by rare inherited loss-of-function homozygous, compound hetero-zygous, or X chromosome mutations in Neuron  77  , January 23, 2013 ª 2013 Elsevier Inc.  209  males. In this way, Lim et al.(2013) significantly advanceour population-level under-standing of risk for ASD.One key issue is that mostcurrent studies on the contri-bution of rare variants togenetic risk are not wellpowered to identify specificgenes. Typically, functionalstudies or repeated instancesof rare mutations in the samegene are needed to ensurethat a variant is indeed func-tional and associated with ASD. Alternatively, other evi-dence, such as gene expres-sion data can be used tofunctionallyprioritizevariants,as many mutations, includingbut not limited to de novodeletions (e.g., copy numbervariations; CNVs), or reces-sive mutations causing intel-lectual disability (ID) or ASDare expected to significantlyreduce RNA levels even inperipheral blood ( Luo et al.,2012 ). We predict that futurestudies combining genomesequencing with gene ex-pression will have increasedpower to detect pathogenicinherited or de novo geneticvariation. Short of this, verylarge sample sizes (larger than currentGWAS samples ranging   10,000 sub- jects) or unique families with multipleaffected children (multiplex families) willbe necessary.In the second study, Yu et al. (2013)took an elegant approach to identifyspecific inherited rare variants by study-ing multiplex families, combining multiplegene-mapping modalities, and validatingvariant pathogenicity experimentally.Specifically, the authors searched con-sanguineous and/or multiplex familiesusing a combination of heterozyogsitymapping and linkage-directed WES anal-ysis. This identified regions in three fami-lies with high linkage scores (likelihood>600:1 of the genetic region containinga variant tracking ASD in that family)where specific nonsynonymous or frame-shift rare variants were also found.This analysis identified three genes,  AMT  ,  PEX7  , and  SYNE1  as affected byhomozygous or compound heterozygousrare variation linked to ASD. To demon-strate the pathogenicity of each of theserare variants, the authors performeddetailed in vitro functional analysis, whichhas not been previously performed inpopulation-based WES studies. Forexample,one familycontained ahomozy-gousI308Fchangein  AMT  ,agenecodingfor a glycine-degradation enzyme inwhichseveremutationshavebeenshownto lead to neonatal nonketotic hypogly-cemia (NKT), a life-threatening neonatalmetabolic syndrome. This mutation wasanalyzed biochemically and its effecton protein solubility was evaluated inbacteria, suggesting that the mutationmay induce a protein-folding defectand result in a functional hypomorph.Indeed, the family contained threeaffected individuals with an atypical,milder manifestation of NKT symptomsin addition to ASD. The authors proposethat, in general, less severemutations in genes involvedin recessive neurodevelop-mental ID syndromes maylead to ASD. A similar inheritance pat-tern and relationship to syn-dromicdiseasewasobservedwith  PEX7  , which is thecausative gene for rhizomelicchondrodysplasia punctata,and  SYNE1 , where null muta-tions have been linked tocerebellar ataxia and arecessive form of arthrogry-posis multiplex congenita.The functional data pre-sented by the authors andabsence of the strong clas-sical phenotypes suggestedthese changes were also hy-pomorphic, and their experi-mental investigations suppor-ted this. Yu and colleagues’approach exemplifies howvarious lines of direct andindirect evidence—genome-widelinkage,exonicvariation,biological validation, andclinical relevance to existingsyndromes—may be used toconvincingly implicate a rarevariant, which itself maynever reach population-wide,genome-wide significance.Yu et al. (2013) further generalized thismutational pattern and its relationship tosyndromic disease by compiling a set of risk genes comprising these 3 and 70other previously implicated ID-relatedgenes.Theythencomparedrareinheritedvariants from WES in 163 consanguin-eousand/ormultiplexfamiliesto831pop-ulation-matched individual exomes tofind additional homozygous, compoundheterozygous, or hemizygous rare muta-tions. They find five families with previ-ously unidentified nonsense or frameshiftvariants and 11 with previously unidenti-fied missense variants in their set of riskgenes. Intriguingly, two families hadnonsense or frameshift mutations in  PAH and one had a likely functional missensemutation in  AMT  , suggesting a potentialmetabolic contribution to their phenotypeand perhaps autistic symptoms. Thishighlights the potential for better geneticdiagnoses and treatment by immediate Rare de novoRare transmitted         0        1        2        3        4        5     O    d    d   s    R   a   t    i   o Large deletion or duplication Contribution of Classes of Rare Variation to ASD Risk Single base pair mutation that truncates protein product Single base pair mutation that changes protein product     D   e   n   o   v   o   S   N   V   n   o   n   s   e   n   s   e Mother Father Child w/ ASD    D   e   n   o   v   o   C   N   V Mother Father Child w/ ASD    T   r   a   n   s   m   i   t   t   e   d   2  -   h   i   t   S   N   V Mother Father Child w/ ASD    D   e   n   o   v   o   S   N   V   m   i   s   s   e   n   s   e Mother Father Child w/ ASDMother Father Child w/ ASD    T   r   a   n   s   m   i   t   t   e   d   C   N   V   T   r   a   n   s   m   i   t   t   e   d   1  -   h   i   t   S   N   V Mother Father Child w/ ASD Figure 1. Risk Imparted by Different Kinds of Rare Genetic Variants The population-level enrichment in ASD versus controls of each form of raremutationstudiedtodate(representativesinglenucleotidevariant,orSNVesti-mates from Sanders et al., 2012 but also see Neale et al., 2012; O’Roak et al., 2012; Iossifov et al., 2012; copy number variation, or CNV estimates from Sanders et al., 2012; two-hit transmitted SNV estimates from Lim et al.,2013 ). An emerging trend in ASD genetics is that the most damaging typesof rare variation are observed more often in ASD than controls. De novo vari-ants are overrepresented in autism more than transmitted variants. Moreover,the most damaging class of variants have the greatest degree of overrepre-sentation. These risk estimates do not include confidence intervals or all of thestudiesbutattempttoconveytherelativeeffectsizes.Belowthebargraphare diagrams illustrating the method of inheritance and type of mutation. Neuron Previews 210  Neuron  77  , January 23, 2013 ª 2013 Elsevier Inc.  intervention in a subset of ASD, aswasrecentlyhighlightedinarecentlydiscovered metabolic form of ASD( Novarino et al., 2012 ). In total, thisstudy’s cohort found novel changeslinked to ASD in the novel genes  AMT  ,  PEX7  ,  SYNE1 ,  VPS13B/ COH1 ,  PAH , and  POMGNT1 , aswell as previously implicated genes NLGN4X   and  MECP2 .These studies provide two uniquevistas on ASD genetics. Given anetiologically and phenotypicallyheterogeneous neurodevelopmentaldisorder that may involve hundredsof genes, the field has attempted togain a foothold on biological path-ways by identifying genes usinghighly penetrant mutations that arelinked to ASD. These studies alsoprovide convincing statistical evi-dence for the role of homozygous orcompound heterozygous loss of functionmutationsinASDrisk.More-over,theresultsemphasizehowdiffi-cult it can be to assign blame to asinglegene,giventherarityofevents,their occurrence in controls, and thenumbers necessary to attain gen-ome-wide significance. Observingtheeventmoretimesinaffectedindi-viduals is still necessary to providedefinitive proof of genetic associa-tionwithdisease.Thisisfurthercom-plicated by the observation that mutationrates vary by several orders of magnitudeacross the genome, which suggests theneed for locus-specific calculations of variant significance in ASD ( Michaelsonet al., 2012 ).Finally,theclearlydefinedroleforCNVsand SNVs that delete or alter either one orboth copies of a gene or its isoforms indi-cates that transcript levels play a signifi-cant role in ASD susceptibility. From thisperspective, we predict that future workwill identify variants in regulatory regionsthat affect transcript levels, such aspromoters and enhancers, and that non-coding regions will be just as functionallyimportant to ASD as currently implicatedprotein-coding regions. This is also sup-ported by the high mutability of DNasehypersensitivesitesandCpG-richregions( Michaelson et al., 2012 ). So far, thecontribution of regulatory elements isunexplored at a population level, asWES does not effectively measure theseregions. Nevertheless, these current ap-proaches provide unprecedented insight(e.g., Figure 2 ) that will aid in explaining ASD’s pathobiology. REFERENCES  Anney, R.,Klei, L.,Pinto,D.,Almeida,J.,Bacchelli,E., Baird, G., Bolshakova, N., Bo ¨ lte, S., Bolton,P.F., Bourgeron, T., et al. (2012). Hum. Mol. Genet.  21 , 4781–4792.Berg, J.M., and Geschwind, D.H. (2012). GenomeBiol.  13 , 247.Devlin, B., and Scherer, S.W. (2012). Curr.Opin. Genet. Dev.  22 , 229–237.Hallmayer,J.,Cleveland,S.,Torres,A.,Phil-lips, J., Cohen, B., Tsrcoe, T., Miller, J., Fe-dele, A., Collins, J., Smith, K., et al. (2011). Arch. Gen. Psychiatry  68 , 1095–1102.Hosmer, D., and Lemeshow, S. (2000). Applied Logic Regression, Second Edition(New York: John Wiley & Sons, Inc.), pp.164–167.Iossifov, I., Ronemus, M., Levy, D., Wang,Z., Hakker, I., Rosenbaum, J., Yamrom, B.,Lee, Y.H., Narzisi, G., Leotta, A., et al.(2012). Neuron  74 , 285–299.Klei,L.,Sanders,S.J.,Murtha,M.T.,Hus,V.,Lowe, J.K., Willsey, A.J., Moreno-De-Luca,D., Yu, T.W., Fombonne, E., Geschwind,D., et al. (2012). Mol. Autism  3 , 9.Lim, E.T., Raychaudhuri, S., Sanders, S.J.,Stevens, C., Sabo, A., MacArthur, D.G.,Neale, B.M., Kirby, A., Ruderfer, D.M.,Fromer, M., et al. (2013). Neuron  77  , thisissue, 235–242.Luo,R., Sanders,S.J.,Tian,Y.,Voineagu, I.,Huang,N.,Chu,S.H.,Klei,L.,Cai,C.,Ou,J.,Lowe, J.K., etal.(2012).Am.J.Hum. Genet. 91 , 38–55.MacArthur, D.G., Balasubramanian, S.,Frankish, A., Huang, N., Morris, J., Walter,K., Jostins, L., Habegger, L., Pickrell, J.K.,Montgomery, S.B., et al.; 1000 GenomesProject Consortium. (2012). Science  335 ,823–828.Michaelson, J.J., Shi, Y., Gujral, M., Zheng,H., Malhotra, D., Jin, X., Jian, M., Liu, G.,Greer, D., Bhandari, A., et al. (2012). Cell 151 , 1431–1442.Neale, B.M., Kou, Y., Liu, L., Ma’ayan, A., Samo-cha, K.E., Sabo, A., Lin, C.F., Stevens, C., Wang,L.S., Makarov, V., et al. (2012). Nature  485 ,242–245.Novarino, G., El-Fishawy, P., Kayserili, H., Meguid,N.A., Scott, E.M., Schroth, J., Silhavy, J.L., Kara,M., Khalil, R.O., Ben-Omran, T., et al. (2012).Science  338 , 394–397.O’Roak, B.J., Vives, L., Girirajan, S., Karakoc, E.,Krumm, N., Coe, B.P., Levy, R., Ko, A., Lee, C.,Smith, J.D., et al. (2012). Nature  485 , 246–250.Sanders, S.J., Murtha, M.T., Gupta, A.R.,Murdoch, J.D., Raubeson, M.J., Willsey, A.J.,Ercan-Sencicek, A.G., DiLullo, N.M., Parikshak,N.N.,Stein,J.L.,etal.(2012).Nature  485 ,237–241.Yu, T.W., Chahrour, M.H., Coulter, M.E., Jiraler-spong, S., Okamura-Ikeda, K., Ataman, B.,Schmitz-Abe, K., Harmin, D.A., Adli, M., Malik, A.N., et al. (2013). Neuron  77  , this issue,259–273. Tagged by Common VariationDe Novo CNVsNonsense de Novo SNVsMissense de Novo SNVs2-Hit LoF Rare TransmittedUnexplored Rare GeneticInfluences/Epistasis/Environment  Global Contribution of Typesof Genetic Variation to ASD Figure 2. The Percentage of Variance Explained by  Various Forms of Genetic Risk Factors for ASD Common variants capture a large percentage of populationrisk for ASD ( Klei et al., 2012 ), whereas current studies showthat rare exonic variants explain a smaller amount of variance.Percent variance explained for rare variants, based on citedreferencesinFigure1,isestimatedviathe squaredcorrelationof outcome with predicted probabilities from logistic regres-sion ( Hosmer and Lemeshow, 2000 ). Note that thoughcommon variants explain a large proportion of the variance,each locus is expected to be of small effect, whereas, thoughrare variants explain a small proportion of the variance, eachlocus is expected to be of larger effect. As-yet-unexploredrare variation (in the 98% of the genome not studied byWES),interactionsamonggenes,andtheinfluenceoftheenvi-ronment are potential culprits to explain the remaining risk. Additionally, because syndromic forms of autism are typicallyexcludedfrommostpopulation-levelstudiesfromwhichthesegenetic estimates have been derived, they are not includedhere. Clinical ascertainment suggests that their prevalence ison the order of 5% within the entire population of autism. Neuron  77  , January 23, 2013 ª 2013 Elsevier Inc.  211 Neuron Previews
Search
Similar documents
View more...
Tags
Related Search
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