Copyright � 2009 by the Genetics Society of AmericaDOI: 10.1534/genetics.109.104695

Fork Stalling and Template Switching As a Mechanism for PolyalanineTract Expansion Affecting the DYC Mutant of HOXD13,

a New Murine Model of Synpolydactyly

Olivier Cocquempot,*,†,1 Veronique Brault,*,† Charles Babinet‡ and Yann Herault*,†,§,2

*Universite d’Orleans, UMR6218, Molecular Immunology and Embryology, 45071 Orleans, France †Centre National de la RechercheScientifique (CNRS), UMR6218, MIE, 3B rue de la Ferollerie, 45071 Orleans Cedex 2, France, ‡Unite de Biologie du Developpement,

URA 1960, CNRS, Institut Pasteur, 25 rue du Docteur Roux 75015 Paris, France and §CNRS, UPS44, TAAM,Institut de Transgenose, 45071 Orleans, France

Manuscript received May 5, 2009Accepted for publication June 15, 2009

ABSTRACT

Polyalanine expansion diseases are proposed to result from unequal crossover of sister chromatids thatincreases the number of repeats. In this report we suggest an alternative mechanism we put forward whilewe investigated a new spontaneous mutant that we named ‘‘Dyc’’ for ‘‘Digit in Y and Carpe’’ phenotype.Phenotypic analysis revealed an abnormal limb patterning similar to that of the human inheritedcongenital disease synpolydactyly (SPD) and to the mouse mutant model Spdh. Both human SPD andmouse Spdh mutations affect the Hoxd13 gene within a 15-residue polyalanine-encoding repeat in the firstexon of the gene, leading to a dominant negative HOXD13. Genetic analysis of the Dyc mutant revealed atrinucleotide expansion in the polyalanine-encoding region of the Hoxd13 gene resulting in a 7-alanineexpansion. However, unlike the Spdh mutation, this expansion cannot result from a simple duplication ofa short segment. Instead, we propose the fork stalling and template switching (FosTeS) described forgeneration of nonrecurrent genomic rearrangements as a possible mechanism for the Dyc polyalanineextension, as well as for other polyalanine expansions described in the literature and that could not beexplained by unequal crossing over.

HIGHLY conserved Hox genes encode transcriptionfactors containing a homeobox and forming

multimeric complexes with other specific DNA-bindingpartners to regulate the transcription of specific genes(Moens and Selleri 2006). They play key roles in thecontrol of positional information during body axisspecification and limb patterning. In mammals, Hoxgenes are organized in four clusters A to D in whichthey are expressed with a precise spatial and temporalpattern that reflects their genomic organization, with59genes expressed late and in more posterior and distalregions (Kmita and Duboule 2003; Deschamps andVan Nes 2005). In the appendices, only the more 59

located genes (9–13) are expressed, with a predomi-nant role of A- and D-cluster genes. Duplications anddeletions analyses (Herault et al. 1998; Kmita et al.2005) have revealed a complex network of interactionsamong Hox genes from the different clusters as well asa global regulation of genes within a cluster.

Despite their central role in vertebrate body pattern-ing, human congenital abnormalities attributable to

mutations in HOX genes are rare and have only mildeffects (for review see Goodman 2002). Mutations inHOX genes have been mostly associated with somehuman congenital malformation syndromes of the devel-oping limbs. Whereas mutations in HOXA13 cause hand-foot-genitalia syndrome (HFGS, OMIM no. 140000),characterized by short thumbs and halluces, and clino-brachydactyly of the second and fifth fingers (Warot

et al. 1997; Goodman et al. 2000), mutations in HOXD13give rise to a wide spectrum of limb malformations withvariable penetrance and expressivity. Missense mutationswithin the homeodomain of HOXD13 result in a loss offunction that leads to brachydactyly type D (BDD, MIM113200) or to brachydactyly type E (BDE, MIM 113300).Expansion (Akarsu et al. 1996; Muragaki et al. 1996;Goodman et al. 1997) and, in few cases, frameshiftdeletions (Goodman et al. 1998; Calabrese et al. 2000)of the polyalanine tract in exon 1 of HOXD13 triggersynpolydactyly (SPD, MIM 186000), a semidominantlimb malformation syndrome characterized by digitduplication and fusion in heterozygotes and a combina-tion of syndactyly, polydactyly, and shortening of thehands and feet in homozygotes (Caronia et al. 2003;Yucel et al. 2005).

A spontaneous mouse mutant of Hoxd13 was isolatedby Johnson and collaborators (1998) that, in its homo-

Supporting information is available online at http://www.genetics.org/cgi/content/full/genetics.109.104695/DC1.

1Present address: Universite de Limoges, 87060 Limoges, France.2Corresponding author: UMR6218, IEM, Uni Orleans, CNRS, UPS44,

TAAM, Institut de Transgenose, 3B rue de la Ferollerie, 45071 OrleansCedex 2 France. E-mail: herault@cnrs-orleans.fr

Genetics 183: 23–30 (September 2009)

zygous state, phenotypically and molecularly modeledhuman SPD. The Spdh (synpolydactyly homolog) muta-tion is a 21-bp in-frame duplication within the poly-alanine stretch of exon 1, expanding the number ofalanines from 15 to 22, and it corresponds to the majortype of expansion found in human SPD. Spdh heterozy-gous mice have only a slight shortening of digits 2 and 5,whereas homozygotes present a severe shortening of allfingers from both forelimbs and hindlimbs, associatedwith a combination of syndactyly and polydactyly due toa substantial delay in the ossification of the limb bonyelements. In addition, homozygous mice lack preputialglands and males are not fertile (Bruneau et al. 2001;Albrecht et al. 2002).

We identified a new spontaneous mutant originatedin the BALB/c line that presented an alteration of thedistal part of the limbs and was hence named Dyc for‘‘Digit in Y shape and Carpe.’’ This mutation appearedspontaneously in the last backcross of a transgenic lineon the BALB/c genetic background, during the finalintercross to generate mice homozygous for the trans-gene. The phenotype was similar to human synpolydac-tyly and to phenotypes observed in mutations affectingHox genes. Phenotypic and genetic characterization ofDyc mice demonstrated that Dyc is a new allele of Hoxd13affecting the polyalanine cluster. Even though theresulting mutant allele contains, like the Spdh mutant,a 7-alanine expansion, sequencing analysis revealed thatthe nucleotide expansion in the Dyc mutant is differentfrom that of the Spdh. Spdh polyalanine expansion is astraightforward reduplication of a short segment of theimperfect trinucleotide repeat, whereas the Dyc expan-sion seems to result from two smaller duplications.Unlike classical long unstable trinucleotide repeatexpansions that have been described to result frompolymerase slippage during DNA replication (seeKovtun et al. 2004 for review), polyalanine expansionsare short, imperfect, and stable when transmitted fromgeneration to generation and hence have been pro-posed to result from an unequal crossing over betweentwo mispaired wild-type alleles (Warren 1997). How-ever, the Dyc expansion, like a few other alanine tractexpansions reported in the literature do not all fit withthe model of unequal crossing over (Goodman et al.1997; De Baere et al. 2003; Robinson et al. 2005;Trochet et al. 2005). Until now, no mechanism otherthan replication slippage, which does not fit with the‘‘nature’’ of the polyalanine repeats, was proposed(Trochet et al. 2007). In the article, we describe thephenotype and mutation of the Dyc allele of Hoxd13,apply the mechanism of fork stalling and templateswitching (FosTeS) (Lee et al. 2007), on the basis ofthe switching forward or backward of the replicationfork along the DNA template using microhomologysequences to generate the Dyc expansion, and pro-pose FosTeS as a new mechanism for polyalanineextensions.

MATERIALS AND METHODS

Mouse lines and complementation test: The Dyc spontane-ous mutation was isolated during the intercross of a mutantline backcrossed at N10 on the BALB/c genetic background.C57BL/6J HoxD,Del3. (Zakany and Duboule 1996) micewere obtained from D. Duboule through the European MouseMutant Archive (EMMA; www.emmanet.org). The lines weremaintained in specific pathogen-free conditions and all theexperiments were carried out in accordance with the Euro-pean and French regulations. Complementation testing wasdone in a B6C hybrid background.

Skeletal preparations: Preparations of skeletons wererealized for adults, newborns, and embryos, according toJohnson et al. (1998). Briefly, the adults and newborns areeviscerated, skinned, and fixed in ethanol. Staining of thebones is obtained with alizarin red and of the cartilage(newborns) with alcian blue. The flesh is clarified in glycerol.The embryos are fixed in Bouin solution, stained with alcianblue, and the flesh is clarified in methyl salicylate.

In situ hybridization: Whole-mount in situ hybridization wasperformed using digoxigenin-labeled probes for Hoxd13,Hoxd12, Hoxd11, and Hoxd10 genes as described previously(Herault et al. 1998) and embryos of stages 12.5 dayspostcoitum (dpc) obtained by time matings from Dyc/1 3Dyc/1. Embryos were collected, fixed in 4% paraformalde-hyde at 4�, dehydrated, and stored in methanol at �20� untiluse. After rehydratation, bleaching with H2O2 and a pre-treatment with proteinase K (10 mg/ml), the embryos wereincubated over night at 70� with the probes (1 mg/ml). Theywere then washed several times, incubated with anti-DIGalkaline phosphatase antibody (Roche Biochemicals) at thedilution of 1:5000 overnight at 4�. After washing of theantibody, staining was performed with BM purple (Roche).

Exon sequencing of Hoxd13 to Hoxd10 genes: To screen formutation in the coding sequence of Hoxd10-13 genes, specificprimers were designed with the DsGene program for ampli-fication of the exons of those genes by PCR ( Johnson et al.1998). Exons were amplified by classical PCR. Then, bothstrands of the PCR fragments were sequenced using the BigDye Terminator reaction mix (Applera, France) on an ABI310sequencer. The sequence was analyzed with the DSGenesoftware (Acceleris, UK).

RESULTS

Dyc mutants show digit malformations and agenesisof preputial glands: Mice with abnormal digits in fore-and hindlimb feet were discovered in the BALB/c lineduring a backcross of a transgene to generate homozy-gous animals. In the progeny of this intercross, 71% ofthe animals (n ¼ 109 produced progeny) had a digitalteration, with 48% having a slight phenotype affectingdigits 2 and 5, and 23% having a stronger phenotypewith shortening of all the fingers. This was indicative ofthe appearance of a semidominant mutation with thegroup of animals bearing the mild phenotype beingheterozygous and the group with the strong phenotypebeing homozygous and suggested that the parents wereheterozygotes. Looking back at the parents, they wereindeed found to have the mild digit phenotype.

In heterozygous adult animals, skeletal analysis(Figure 1a) revealed a shortening of digits 2 and 5.Homozygous animals exhibited a complete alteration of

24 O. Cocquempot et al.

the distal part of the five digits and malformation of allthe bones from the ankle (metacarpal, carpal, meta-tarsal, and tarsal bones). These alterations were asso-ciated with a syndactyly for digits 4 and 5 and theappearance of an extra sixth postaxial digit (polydac-tyly). The hindlimbs had an additional strong deformityof the metatarsal bone of digit 1. This new mutant wascalled Dyc for Digit in Y and Carpe (Y-shaped finger andcarpe). No other alteration of the axial skeleton wasfound (n ¼ 10 wt and 10 Dyc/Dyc; 12 weeks old).Necropsy analyses (n ¼ 3 wt and 4 Dyc/Dyc) did notreveal any other drastic organ malformation and ahematological analysis (n ¼ 3 wt and 4 Dyc/Dyc; 12weeks old) indicated a normal globular numeration.Homozygous Dyc males are not fertile. Analysis of theurogenital apparatus of these males revealed that pre-putial glands were absent. Hence, the Dyc mutationproduces two major alterations: a severe malformationof the autopod and the lack of preputial glands in males.These two phenotypes were similar to those of the Spdhmutant mouse, a model for human synpolydactylycharacterized by an expansion of unstable repeatswithin the 59 end of the Hoxd13 gene (Johnson et al.1998).

Effect of the Dyc mutation on the ossification: Wewent further to check whether the origin of thephenotype of the Dyc mutant during bone and cartilagedevelopment was similar to the Spdh mutant. Alcianblue/alizarin red staining of the skeleton was per-formed at several time points during embryonic andpostnatal development to compare cartilage and bonedevelopment between Dyc/Dyc mutants and wild-typelittermates (supporting information, Figure S1). Alcianblue staining of autopods at E14.5 dpc reveals a de-velopmental delay in individualization of the fingers

and cartilage formation in the Dyc/Dyc autopod. Super-numerary fingers are present. At 15.5 dpc, deformationat the level of the metatarsus and metacarpus are visible.At birth (0.5 days postpartum or dpp), ossificationcenters are detected in all wild-type digits but not inDyc/Dyc mutant forelimbs. Polydactily is accompaniedby syndactily. At 4.5 dpp, while ossification of thephalanx is completed in wild-type autopods, in Dyc/Dycautopods, the ossification center’s location is not clearlydefined and there are defects in segmentation of thefinger elements, making it hard to distinguish meta-carpus and phalanx.

Dyc mutation is an allele of Hoxd genes: Limbphenotypes associated with the Dyc mutation are moresevere than those found in the Hoxd13 mutationobtained by genetic inactivation, but similar to thealterations found in mutants of the posterior Hoxdgenes, controlling limb patterning (Zakany et al.1997; Kmita et al. 2002a,b). Heterozygote HoxDDel3/1

mutant mice (Zakany and Duboule 1996) bearing adeletion that inactivates the expression of Hoxd11,Hoxd12, and Hoxd13 genes simultaneously have ashortening of digits 2 and 5 like Dyc heterozygousanimals (Figure 1, b vs. a). HoxDDel3/Del3 homozygous forthis deficiency showed reduction in the length of all thedigits (ectrodactyly) with a rougher and stiffer aspect ofthe digits and polydactyly/syndactyly in the hindlimbs(Figure 1b). The Dyc mutation in its homozygous formseems hence to be less severe than the Del3 one. Wecarried out a complementation test by crossing Dycanimals with HoxDDel3 mutant mice. HoxDDel3/Dyc trans-heterozygous mice displayed digit alteration similar tothe ones present in HoxDDel3/Del3 homozygotes (Figure1b). Both have a shortening of all fingers with a mal-formation of toe 1 and polydactyly in the hindlimbs.

Figure 1.—Alizarin skeletalpreparations of adult distal fore-limbs and hindlimbs of (a) wild-type (1/1), heterozygous (Dyc/1),and homozygous (Dyc/Dyc) mu-tants and (b) heterozygous Del3,homozygous Del3, and transheter-ozygous Dyc/Del3 mice. Fingersare numbered from 1 to 5, fromthe thumb to the little finger andshortening of fingers is shownby arrows. Dyc/1 and Del3/1 au-topods have shortening of fingers2 and 5. Dyc/Dyc have shorteningand malformation of all fingersand appearance of a supernumer-ary finger, noted 6, in postaxialposition. Syndactily between fin-gers 3 and 4 with fusion at themetacarpus or metatarsus level isalso present. In hindlimbs, thehallux is shortened with the meta-

tarsus having a characteristic form. Del3/Del3 have shortening of all fingers and high deformation of toe 1, syndactily of toes 4 and5 (arrow), and supernumerary toe 6 in the hindlimb. Transheterozygote Dyc/Del3 animals show shortening of all fingers with ahigher deformation of toe 1 (arrow).

FosTeS Mechanism Leading to Polyalanine Expansion Mutation 25

Interestingly synpolydactily is rarely observed in trans-heterozygote animals. These results indicate that the Dycmutation is an allele of one of the genes from the HoxDcomplex.

Posterior genes from the HoxD do not have theirexpression perturbed by the Dyc mutation: We checkedwhether the Dyc mutation affected the transcriptionalcontrol of posterior genes from the HoxD complex inthe limb bud. Only the more 59 located genes (9–13) areexpressed in the limbs. They are expressed in over-lapping domains in more or less anterior-to-posteriorgradients, are involved in the proper patterning ofproximodistal elements, and impact the developmentof the anteroposterior axis of the limbs. Comparing theexpression patterns of Hoxd13, Hoxd12, Hoxd11, andHoxd10 genes in the limb buds of wild-type controls andDyc/Dyc mutant embryos at 12.5 dpc using whole-mountin situ hybridization revealed identical patterns of ex-pression between mutant and control animals in thedistal part of the autopod (Figure S2). Additionalhybridization of Hoxd13 at 11.5 dpc and 13.5 dpc andof Hoxa13 at 13.5 dpc did not reveal any abnormalexpression of these markers whereas physical alterationis already visible (data not shown).

The Dyc mutation affects the Hoxd13 proteinsequence: After having checked that expression pat-terns of the posterior genes from the HoxD complexwere not perturbed, we looked for the presence ofmutations within the coding sequences of those genes.Sequencing analysis of Hoxd10, Hoxd11, and Hoxd12 didnot show any mutation within those genes in Dycmutants. However, PCR amplification of the 59 part ofexon 1 of Hoxd13 indicated an increase in the size ofthe amplified segment compared to BALB/c andC57BL/6J controls (data not shown). Sequencing ofthis fragment revealed a duplicated region responsiblefor the increased size (Figure 2a). This amplificationcorresponds to 21 bp containing mainly trinucleotiderepeats, coding for a series of 15 alanine residues in thewild-type control and for 22 alanine residues in the Dycmutant. Hence, the Dyc mutation corresponds to poly-alanine expansion similar to the Hoxd13Spdh mutationthat creates a gain of function of HOXD13 protein.However, alignment between Spdh and Dyc polyalaninestretches (Figure 2b) indicates that the nature of theexpansion is different between the two allelic mutants.Whereas the Spdh expansion appears to be a straightfor-ward reduplication of the nucleotide repeats corre-

Figure 2.—Analysis of the Dyc mutation. (a) Alignment between BALB/c (sequence of reference) and Dyc/Dyc showingthe amplification in 59 of Hoxd13 exon 1. (b) Alignment between the BALB/c, C57BL/6J, Spdh, and Dyc sequences. The duplicatednucleotides in the two mutants are shown in red and purple and the corresponding fragments in the normal sequence are under-lined in the same color. The arrows show the straightforward tandem duplication in the Spdh mutant and the imperfect ampli-fication in Dyc where the unequal allelic homologous recombination model is at fault. (c) Multiple alignments of the BALB/csequence to the Dyc (yellow) at microhomologies (red letters) that are boxed. The extended sequence in the Dyc mutant is under-lined. Aligned sequences are in blue, green, and purple; nonaligned ones in gray.

26 O. Cocquempot et al.

sponding to alanines 1–7, the Dyc expansion is morecomplex, with a duplication of nucleotide repeatscoding for alanines 1–5 and the two first nucleotidesof alanine 7, whereas a small duplication correspondingto the sequence AGCA shifted in front of the firstduplication instead of being after as in the wild-typesequence.

DISCUSSION

The Dyc mutation is a semidominant mutation caus-ing severe limb malformations that are strikingly similarto those of the Spdh mutant, a 21-bp in-frame duplica-tion within the polyalanine-encoding region of theHoxd13 gene (Johnson et al. 1998). Like Spdh/Spdhmutants, homozygous Dyc mice have delayed distal limbdevelopment and altered chondrogenesis resultingin reduced size of all the fingers associated with poly-dactyly and syndactyly. Heterozygous animals are lessaffected, with only a slight shortening of fingers 2 and 5,indicating that the Hoxd13 wild-type allele is able topartly compensate for the effect of the Dyc allele. Like inthe Spdh mutant, alterations found in Dyc homozygousmice are more severe than those of null mutants ofHoxd13 (Dolle et al. 1993; Davis and Capecchi 1996)but are similar to those found in the loss of function ofthe three most posterior genes of the HoxD complex(Hoxd13, Hoxd12, and Hoxd11) (Zakany and Duboule

1996). However, normal pattern expression of thosegenes during limb development of Dyc/Dyc embryosindicates that inactivation does not occur at the level ofthe transcription as it could have been expected fromthe transcriptional control loops between Hox genesthat have been described in the literature (Gavalas

et al. 2003).Mapping of the Dyc mutation revealed an expansion

of the trinucleotide repeat in the upstream exon ofHoxd13 similar to that found in the Spdh spontaneousmutant that was discovered in a mouse colony withthe B6C3Fe genetic background. Both Dyc and Spdhalleles give an additional 7 alanines at the 15-alaninesstretch, conferring a dominant-negative property to theHOXD13 protein, probably by interfering with theactivity of the wild-type HOXD13 and other HOXproteins. This suggests that the polyalanine tract is im-portant either for interaction with other partners or forformation of a functionally important structure of theprotein and that whatever the genetic background,there is a threshold length of polyalanine extensionabove which the extension becomes pathologic anddigital anomalies appear. Polyalanine tract expansion isalso the common mutation found in the typical form ofhuman SPD (Akarsu et al. 1996; Muragaki et al. 1996;Horsnell et al. 2006), with additional alanine residuesvarying from 7 to 14 (Goodman et al. 1997). Like in themouse, a minimum of 7 additional alanine residues isneeded in human HOXD13 for a phenotype to appear

and the genetic background does not seem to influencethe severity of the phenotype that is influenced by thelength of the alanine expansion (Goodman et al. 1997).But human and mouse differ in their heterozygous andhomozygous states. Whereas digit fusions and duplica-tions are present in human heterozygotes, this pheno-type is only visible in mouse homozygotes indicating adifference in molecular interplay between human andmouse.

Polyalanine tract expansions are found in at least ninedisorders (Amiel et al. 2004; Brown and Brown 2004),mostly in genes that are coding for transcription factorsinvolved in developmental processes (Karlin et al. 2002;Lavoie et al. 2003). Unlike classical long unstable tri-nucleotide repeat expansions found in genes impli-cated in hereditary neurodegenerative diseases, thepolyalanine expansions are imperfect, short, and stablethrough meiosis and mitosis. This suggests a differentmechanism of mutation from strand slippage occurringduring DNA replication, repair, or recombination in theother diseases of unstable repeat expansion (Sinden

et al. 2002; Pearson et al. 2005; Kaplan et al. 2007;Kovtun and McMurray 2008). Indeed, the expandedpolyalanine tract in SPD is quite stable (no increase inmutation size) when transmitted from one generationto another. In addition, polyalanine tracts are short,with the longest nonpathogenic tract being of 20residues (Lavoie et al. 2003), whereas tracts of �35perfect trinucleotide repeats are required for instabilityand expansion (Eichler et al. 1994; Kunst and Warren

1994). Warren (1997) suggested that polyalanineexpansion found in the mouse Spdh allele and humanSPD may have resulted from unequal crossing over inthe HOXD13 gene during meiosis due to misalignmentbetween triplet repeats. This mechanism could alsoexplain expansions found in other polyalanine disor-ders (Robinson et al. 2005; Arai et al. 2007). However,unlike the Spdh mutation, the complex duplicationfound in the Dyc mutant is not a straightforward re-duplication and cannot be explained by unequalcrossing over. Looking at the 7-alanine expansion tract(Figure 2b), the additional six last alanines can corre-spond to a reduplication of alanines 1–6, but there isan additional GCA codon in front of those 6 alanineswhose origin cannot be explained. We checked forcodon polymorphism in the B6C3Fe and BALB/c linesthat could explain this atypical insertion, but could notfind any. In addition, even if this additional GCA codonis ignored, the 6-alanine expansion cannot result fromsimple unequal allelic homologous recombination.Other mutations have been reported in the literature(Goodman et al. 1997; De Baere et al. 2003; Robinson

et al. 2005; Trochet et al. 2007), which do not fit theunequal allelic recombination model for mutationalmechanism described by Warren (1997). Trochet

and collaborators (2007) already refuted the unequalcrossing-over mechanism as the explanation of poly-

FosTeS Mechanism Leading to Polyalanine Expansion Mutation 27

alanine expansions. However, they only suggested thatthe replication slippage mechanism proposed for dis-eases with unstable triplet expansions was also applyingto polyalanine extensions.

We therefore tried to find another mechanism thatcould lead to the Dyc expansion. Among the mecha-nisms that have kept our attention is the FosTeSdescribed by Lee et al. (2007). FosTeS was proposed toexplain some nonrecurrent chromosomal rearrange-ments (duplications, deletions) causing genomic disor-ders and that could not be explained by the usualnonallelic homologous recombination (NAHR) andnonhomologous end joining (NHEJ) (Shaw andLupski 2004; see Gu et al. 2008 for review of the threemechanisms). Performing a breakpoint sequence anal-ysis of nonrecurrent duplications, they uncoveredcomplex arrays of normal, duplicated, and triplicatedsequences at the junctions of the duplications andidentified few base-pair microhomologies that couldact as ‘‘bridges’’ for the DNA replication fork to skipalong the chromosome to create these complex arrays.They proposed that the DNA fork could skip forward orbackward along the chromosome using these shortsequence homologies when encountering a complexgenomic architecture or a DNA lesion. Since the prop-osition of FosTeS as a mechanism to explain the non-recurrent rearrangements in PMD patients, anothergenomic disorder has been analyzed for submicroscopicrearrangements, revealing the potential role of FosTeSin an increasing number of complex pathologic rear-rangements (Carvalho et al. 2009). We checked if thislong-range template switching mechanism could beapplied to the Dyc mutation and more generally to

polyalanine extensions via short-range template skip-ping. We made multiple alignment of the wild-typeHoxd13 polyalanine tract sequence to the Dyc sequenceat microhomologies to try to reconstitute the Dycexpansion (Figure 2c). We could identify two micro-homology regions of 5 bp each that could have made thereplication fork skip two times forward along the wild-type polyalanine sequence to create the Dyc extension.We then listed the polyalanine expansions referenced inthe literature that could not be explained by unequalcrossing over (Table 1) and we tried the same micro-homology alignment method between the expandedalleles and their normal control alleles to determine theprocess of nucleotide amplification. We encounter thesame problem as Goodman et al. (1997) to interpretthe 9-alanine expansion of pedigree P. This one can beinterpreted as either a simple reduplication of alanines1–9 via the mechanism of unequal crossing over or ofFosTeS (Figure 3) only if there is a polymorphism or amutation in the triplet coding for alanine 8 (GCT–GCG). Both the 15-bp (15 alanines) and 18-bp (16alanines) insertions found in PHOX2B can be inter-preted as a skipping ‘‘backward’’ of the replication forkfollowed by a skipping ‘‘forward.’’ The 21-bp insertion(17 alanines) could be interpreted as a single skipping‘‘backward.’’ However, the microhomology found is onlyof a single nucleotide. The 15-alanine extension inFOXL2 can be recovered by three skipping backward ofthe replication fork along the normal polyalanine tract.Finally, the 17 alanine extension in the PABPN1polyalanine can be the result of a first skipping back-ward at a 2-bp homology followed directly by a secondskipping backward of three nucleotides.

TABLE 1

List of the alanine track expansions that cannot be explained by the unequal recombination mechanism

Gene Repeat Name Reference Sequence

HOXD13 15 ala Normal allele GCGGCGGCGGCGGCAGCGGCGGCTGCGGCGGCGGCGGCGGCAGCC

HOXD13 19 ala Pedigree P Goodman et al.(1997)

GCGGCGGCGGCGGCAGCGGCGGCTGCGGCGGCGGCGGCGGCAGCGGCGGCGGCGGCGGCGGCGGCGGCAGCC

PHOX2B 20 ala Normal allele GCAGCAGCAGCGGCGGCGGCCGCGGCAGCGGCGGCGGCGGCAGCGGCAGCGGCGGCAGCT

PHOX2B 15 ala Trochet et al.(2005, 2007)

GCAGCAGCAGCGGCGGCGGCCGCGGCAGCGGCGGCGGCGGCAGCGGCAGCGGCGGCGGCGGCAGCGGCAGCGGCT

PHOX2B 16 ala Trochet et al.(2005, 2007)

GCAGCAGCAGCGGCGGCGGCCGCGGCAGCGGCGGCGGCGGCAGCGGCGGCGGCCGCGGCAGCGGCGGCGGCGGCAGCT

PHOX2B 17 ala Trochet et al.(2005, 2007)

GCAGCAGCAGCGGCGGCGGCGGCAGCAGCAGCGGCGGCGGCCGCGGCAGCGGCGGCGGCGGCAGCGGCAGCGGCGGCAGCT

FOXL2 14 ala Normal allele GCGGCAGCCGCAGCGGCTGCAGCAGCTGCGGCTGCAGCCGCGFOXL2 115 ala 921�935

trip15De Baere et al.

(2003)GCGGCAGCCGCAGCGGCTGCAGCAGCTGCGGCTGCAGCAGCTGC

GGCTGCAGCAGCTGCGGCTGCAGCAGCTGCGGCTGCAGCCGCGPABPN1 10 ala Normal allele GCGGCGGCGGCGGCGGCGGCAGCAGCAGCGPABPN1 17 ala (GCG) 13 Robinson et al.

(2005)GCGGCGGCGGCGGCGGCGGCGGCGGCGGCGGCGGCGGCG

GCGGCAGCAGCAGCG

These expansions were all found in human pedigrees.

28 O. Cocquempot et al.

Hence, FosTeS was tested on seven atypical poly-alanine extensions and allowed to reconstitute the patho-logic expansion for at least five of them. We thereforepropose FosTeS as an alternative mutational mechanismto unequal crossing over in the generation of polyala-nine expansions. Our hypothesis is further supported bythe fact that FosTeS is supposed to be engendered, or atleast facilitated, by a surrounding high-order genomicarchitecture that contains cruciform or other non-Bstructures (Lee et al. 2007). Such alternative DNAstructures are facilitated by specific sequence motifssuch as low-copy repeats (LCR), symmetrical features,and short, direct, or inverted repeats, and the poly-alanine trinucleotide repeat sequence possesses suchsequence motifs and symmetry elements. Looking forspecific features flanking the breakpoints of complexrearrangements occurring in the MECP2 region andproposed to result from FosTeS, researchers foundincreased frequency of the sequences 59-CTG-39/59-CAG-39 (Carvalho et al. 2009). They suggest that thesemotifs might represent a cis-acting sequence that may bea recognition site for proteins involved in priming DNAreplication in eukaryotes. Interestingly, 59-CTG-39/59-CAG-39 are motifs that are also found around thebreakpoints in the polyalanine expansion tracts (seeFigures 2c and 3).

Hence, after having been described as a mechanismfor large genomic rearrangements, our findings that

FosTeS could also be implicated in monogenic traits viavery short stretches of sequence duplications revealsthat a variety of genomic rearrangements ranging fromgross DNA changes of several kilobases or megabases torearrangements involving only few nucleotide expan-sions within single genes could originate from the samemolecular mechanism.

LITERATURE CITED

Akarsu, A. N., I. Stoilov, E. Yilmaz, B. S. Sayli and M. Sarfarazi,1996 Genomic structure of HOXD13 gene: a nine polyalanineduplication causes synpolydactyly in two unrelated families.Hum. Mol. Genet. 5: 945–952.

Albrecht, A. N., G. C. Schwabe, S. Stricker, A. Boddrich,E. E. Wanker et al., 2002 The synpolydactyly homolog (spdh)mutation in the mouse: a defect in patterning and growth of limbcartilage elements. Mech. Dev. 112: 53–67.

Amiel, J., D. Trochet, M. Clement-Ziza, A. Munnich andS. Lyonnet, 2004 Polyalanine expansions in human. Hum.Mol. Genet. 13: R235–R243.

Arai, H., T. Otagiri, A. Sasaki, T. Hashimoto, K. Umetsu et al.,2007 De novo polyalanine expansion of PHOX2B in congenitalcentral hypoventilation syndrome: unequal sister chromatidexchange during paternal gametogenesis. J. Hum. Genet. 52:921–925.

Brown, L. Y., and S. A. Brown, 2004 Alanine tracts: the expandingstory of human illness and trinucleotide repeats. Trends Genet.20: 51–58.

Bruneau, S., K. R. Johnson, M. Yamamoto, A. Kuroiwa andD. Duboule, 2001 The mouse Hoxd13(spdh) mutation, a poly-alanine expansion similar to human type II synpolydactyly (SPD),

Figure 3.—Application of FosTeS to atypical polyalanine extensions. Multiple alignments of the normal human sequences ofHOXD13, PHOX2B, FOXL2, and PABPN1 to the sequences with atypical polyalanine extensions (yellow) at microhomologies (redletters) that are boxed. The extended sequence in the mutants is underlined. Aligned sequences are in blue, green, and purple;nonaligned ones in gray.

FosTeS Mechanism Leading to Polyalanine Expansion Mutation 29

disrupts the function but not the expression of other Hoxdgenes. Dev. Biol. 237: 345–353.

Calabrese, O., S. Bigoni, F. Gualandi, C. Trabanelli, G. Camera

et al., 2000 A new mutation in HOXD13 associated with footpre-postaxial polydactyly. Eur. J. Hum. Genet. 8: 140.

Caronia, G., F. R. Goodman, C. M. E. Mckeown, P. J. Scambler andV. Zappavigna, 2003 An 147L substitution in the HOXD13homeodomain causes a novel human limb malformation byproducing a selective loss of function. Development 130:1701–1712.

Carvalho, C. M. B., F. Zhang, P. Liu, A. Patel, T. Sahoo et al.,2009 Complex rearrangements in patients with duplicationsof MECP2 can occur by Fork Stalling and Template Switching.Hum. Mol. Genet. 18: 2188–2203.

Davis, A. P., and M. R. Capecchi, 1996 A mutational analysis of the59 HoxD genes: dissection of genetic interactions during limbdevelopment in the mouse. Development 122: 1175–1185.

De Baere, E., D. Beysen, C. Oley, B. Lorenz, J. Cocquet et al.,2003 FOXL2 and BPES: mutational hotspots, phenotypic vari-ability, and revision of the genotype-phenotype correlation. Am.J. Hum. Genet. 72: 478–487.

Deschamps, J., and J. Van Nes, 2005 Developmental regulation ofthe Hox genes during axial morphogenesis in the mouse. Devel-opment 132: 2931–2942.

Dolle, P., A. Dierich, M. Lemeur, T. Schimmang, B. Schuhbaur

et al., 1993 Disruption of the Hoxd-13 gene induces localizedheterochrony leading to mice with neotenic limbs. Cell 75:431–441.

Eichler, E. E., J. J. A. Holden, B. W. Popovich, A. L. Reiss, K. Snow

et al., 1994 Length of Uninterrupted Cgg repeats determinesinstability in the Fmr1 gene. Nat. Genet. 8: 88–94.

Gavalas, A., C. Ruhrberg, J. Livet, C. E. Henderson andR. Krumlauf, 2003 Neuronal defects in the hindbrain ofHoxa1, Hoxb1 and Hoxb2 mutants reflect regulatory interac-tions among these Hox genes. Development 130: 5663–5679.

Goodman, F., M. L. Giovannucci-Uzielli, C. Hall, W. Reardon,R. Winter et al., 1998 Deletions in HOXD13 segregate withan identical, novel foot malformation in two unrelated families.Am. J. Hum. Genet. 63: 992–1000.

Goodman, F. R., 2002 Limb malformations and the human HOXgenes. Am. J. Med. Genet. 112: 256–265.

Goodman, F. R., S. Mundlos, Y. Muragaki, D. Donnai, M. L.Giovannucciuzielli et al., 1997 Synpolydactyly phenotypescorrelate with size of expansions in HOXD13 polyalanine tract.Proc. Natl. Acad. Sci. USA 94: 7458–7463.

Goodman, F. R., C. Bacchelli, A. F. Brady, L. A. Brueton,J. P. Fryns et al., 2000 Novel HOXA13 mutations and the phe-notypic spectrum of hand-foot-genital syndrome. Am. J. Hum.Genet. 67: 197–202.

Gu, W., F. Zhang and J. R. Lupski, 2008 Mechanisms for humangenomic rearrangements. Pathogenetics 1: 4.

Herault, Y., M. Rassoulzadegan, F. Cuzin and D. Duboule,1998 Engineering chromosomes in mice through targeted mei-otic recombination (TAMERE). Nat. Genet. 20: 381–384.

Horsnell, K., M. Ali, S. Malik, L. Wilson, C. Hall et al.,2006 Clinical phenotype associated with homozygosity for aHOXD13 7-residue polyalanine tract expansion. Eur. J. Med.Genet. 49: 396–401.

Johnson, K. R., H. O. Sweet, L. R. Donahue, P. Ward-Bailey,R. T. Bronson et al., 1998 A new spontaneous mouse mutationof Hoxd13 with a polyalanine expansion and phenotype similarto human synpolydactyly. Hum. Mol. Genet. 7: 1033–1038.

Kaplan, S., S. Itzkovitz and E. Shapiro, 2007 A universal mecha-nism ties genotype to phenotype in trinucleotide diseases. PloSComp. Biol. 3: 2291–2298.

Karlin, S., L. Brocchieri, A. Bergman, J. Mrazek and A. J. Gentles,2002 Amino acid runs in eukaryotic proteomes and diseaseassociations. Proc. Natl. Acad. Sci. USA 99: 333–338.

Kmita, M., and D. Duboule, 2003 Organizing axes in time andspace; 25 years of colinear tinkering. Science 301: 331–333.

Kmita, M., N. Fraudeau, Y. Herault and D. Duboule, 2002a Serialdeletions and duplications suggest a mechanism for the collin-earity of Hoxd genes in limbs. Nature 420: 145–150.

Kmita, M., B. Tarchini, D. Duboule and Y. Herault,2002b Evolutionary conserved sequences are required for theinsulation of the vertebrate Hoxd complex in neural cells. Devel-opment 129: 5521–5528.

Kmita, M., B. Tarchini, J. Zakany, M. Logan, C. J. Tabin et al.,2005 Early developmental arrest of mammalian limbs lackingHoxA/HoxD gene function. Nature 435: 1113–1116.

Kovtun, I. V., and C. T. McMurray, 2008 Features of trinucleotiderepeat instability in vivo. Cell Res. 18: 198–213.

Kovtun, I. V., C. Spiro and C. T. McMurray, 2004 Triplet repeatsand DNA repair: germ cell and somatic cell instability in trans-genic mice. Methods Mol. Biol. 277: 309–319.

Kunst, C. B., and S. T. Warren, 1994 Cryptic and polar variation ofthe fragile-X repeat could result in predisposing normal alleles.Cell 77: 853–861.

Lavoie, H., F. Debeane, Q. D. Trinh, J. F. Turcotte, L. P. Corbeil-Girard et al., 2003 Polymorphism, shared functions andconvergent evolution of genes with sequences coding for polya-lanine domains. Hum. Mol. Genet. 12: 2967–2979.

Lee, J. A., C. M. B. Carvalho and J. R. Lupski, 2007 A DNA repli-cation mechanism for generating nonrecurrent rearrangementsassociated with genomic disorders. Cell 131: 1235–1247.

Moens, C. B., and L. Selleri, 2006 Hox cofactors in vertebratedevelopment. Dev. Biol. 291: 193–206.

Muragaki, Y., S. Mundlos, J. Upton and B. R. Olsen, 1996 Alteredgrowth and branching patterns in synpolydactyly caused bymutations in HOXD13. Science 272: 548–551.

Pearson, C. E., K. N. Edamura and J. D. Cleary, 2005 Repeatinstability: mechanisms of dynamic mutations. Nat. Rev. Genet.6: 729–742.

Robinson, D. O., S. R. Hammans, S. P. Read and J. Sillibourne,2005 Oculopharyngeal muscular dystrophy (OPMD): analysisof the PABPN1 gene expansion sequence in 86 patients reveals13 different expansion types and further evidence for unequalrecombination as the mutational mechanism. Hum. Genet.116: 267–271.

Shaw, C. J., and J. R. Lupski, 2004 Implications of human genomearchitecture for rearrangement-based disorders: the genomicbasis of disease. Hum. Mol. Genet. 13: R57–R64.

Sinden, R. R., V. N. Potaman, E. A. Oussatcheva, C. E. Pearson,Y. L. Lyubchenko et al., 2002 Triplet repeat DNA structuresand human genetic disease: dynamic mutations from dynamicDNA. J. Biosci. 27: 53–65.

Trochet, D., S. J. Hong, J. K. Lim, J. F. Brunet, A. Munnich et al.,2005 Molecular consequences of PHOX2B missense, frame-shift and alanine expansion mutations leading to autonomicdysfunction. Hum. Mol. Genet. 14: 3697–3708.

Trochet, D., L. De Pontual, B. Keren, A. Munnich, M. Vekemans

et al., 2007 Polyalanline expansions might not result fromunequal crossing-over. Hum. Mutat. 28: 1043–1044.

Warot, X., C. Fromental-Ramain, V. Fraulob, P. Chambon andP. Dolle, 1997 Gene dosage-dependent effects of the Hoxa-13 and Hoxd-13 mutations on morphogenesis of the terminalparts of the digestive and urogenital tracts. Development 124:4781–4791.

Warren, S. T., 1997 Polyalanine expansion in synpolydactyly mightresult from unequal crossing-over of HOXD13. Science 275:408–409.

Yucel, A., I. Kuru, M. E. Bozan, M. Acar and M. Solak,2005 Radiographic evaluation and unusual bone formationsin different genetic patterns in synpolydactyly. Skeletol Radiol.34: 468–476.

Zakany, J., and D. Duboule, 1996 Synpolydactyly in mice with atargeted deficiency in the HoxD complex. Nature 384: 69–71.

Zakany, J., C. Fromentalramain, X. Warot and D. Duboule,1997 Regulation of number and size of digits by posteriorHox genes: a dose-dependent mechanism with potentialevolutionary implications. Proc. Natl. Acad. Sci. USA 94:13695–13700.

Communicating editor: N. Arnheim

30 O. Cocquempot et al.

Supporting Information http://www.genetics.org/cgi/content/full/genetics.109.104695/DC1

Fork Stalling and Template Switching as a Mechanism for Polyalanine Tract Expansion Affecting the DYC Mutant

of HOXD13, a New Murine Model of Synpolydactyly

Olivier Cocquempot, Véronique Brault, Charles Babinet and Yann Herault

Copyright © 2009 by the Genetics Society of America DOI: 10.1534/genetics.109.104695

O. Cocquempot et al. 2 SI

FIGURE S1.—Autopod development in wild-type and Dyc/Dyc mutant mice. In each panel, a wild-type paw (left) is compared to Dyc/Dyc one (right). (a ) At 14.5 dpc, wild-type autopods have well individualized fingers with visible cartilaginous elements at the level of the carpus metacarpus, tarsi and metatarsi, whereas Dyc/Dyc autopod development is delayed: cartilaginous blastema have only formed in the first phalanx and are not visible in the more distal parts of the future digits that are shorter. (b)At 15.5 dpc, of the cartilaginous elements are visible in both wild-type and mutant autopods, but the digits in Dyc/Dyc embryos reveals shortened and supernumerary digits (6+7) with deformation at the level of the metatarsus and metacarpus. (c) Alizarin red staining of the skeleton at 0.5 dpp shows ossification centres in all wild-type digits but not in Dyc/Dyc mutant forelimbs. In the anterior limbs polydactily is accompanied by syndactily of digits 2, 3, 4 and 5, 6. In posterior limbs, there is also polydactily with syndactily of digits 2, 3 and 4, 5. At 4.5 dpp, ossification of the phalanx is completed in wild-type autopods. In Dyc/Dyc autopods, some ossification has occurred, but localisation of the ossifications centres is not clearly defined and there are defects in segmentation of the finger elements, making it hard to distinguish metacarpus and phalanx.

O. Cocquempot et al. 3 SI

FIGURE S2.—Expression of Hox genes in wild-type and homozygous (Dyc/Dyc) forelimbs at 12.5 dpc. Whole-mount

in situ hybridization of Hoxd10, Hoxd11, Hoxd12, Hoxd13 and Hoxa13 probes indicates that there is no difference in expression profiles of theses genes between wild type and homozygote.