Analysis of De Novo HOXA13 Polyalanine Expansions
Diane Roulston,3Christine A. Larsen,4H. Scott Stadler,4and Jeffrey W. Innis1,2*
1Department of Pediatrics and Communicable Diseases, University of Michigan, Ann Arbor, Michigan
2Department of Human Genetics, University of Michigan, Ann Arbor, Michigan
3Department of Pathology, Cytogenetics Laboratory, University of Michigan, Ann Arbor, Michigan
4Department of Molecular and Medical Genetics, Oregon Health Sciences University and Shriners Hospital for Children, Portland, Oregon
Manuscript Received: 3 October 2012; Manuscript Accepted: 12 December 2012
from unequal chromosomal recombination, yet mechanistic
studies are lacking. We identified two de novo cases of hand-
foot-genital syndrome (HFGS) associated with polyalanine
expansions in HOXA13 that afforded rare opportunities to
investigate the mechanism. The first patient with HFGS was
heterozygous for a de novo nine codon polyalanine expansion.
Haplotype investigation showedthat the expansionarose on the
maternally inherited chromosome but not through unequal
crossing over between homologs, leaving unequal sister chro-
matid exchange during mitosis or meiosis or slipped mispairing
patient with HFGS was mosaic for a six codon polyalanine
expansion. Multiple tissue PCR and clonal analysis of paternal
fibroblastsshowedonlyexpansion/WT and WT/WT clones, and
haplotype data showed that two unaffected offspring inherited
the same paternal allele without the expansion, supporting a
father does not support sister chromatid exchange in the origin
of the expansion. Mosaicism for HOXA13 polyalanine expan-
sions may be associated with a normal phenotype, making
examination of parental DNA essential in apparently de novo
HFGS cases to predict accurate recurrence risks. We could not
exchange has been proven for any polyalanine expansion, sug-
(and contractions) is slipped mispairing without repair or that
the true frequencyof unequal sister chromatid exchange involv-
ing these repeats is low. ? 2013 Wiley Periodicals, Inc.
Key words: hand-foot-genital syndrome; HOXA13; polyalanine
Hand-foot-genital syndrome (HFGS; OMIM 140000) is an auto-
somal dominant condition characterized by limb and urogenital
vesicoureteral reflux, incomplete M€ ullerian fusion in females, and
hypospadias in males [Innis, 2006]. The HFGS is caused by muta-
tions in the HOXA13 gene (OMIM 142959) [Mortlock and Innis,
1997; Goodman, 2000]. The HOXA13 gene includes two exons;
18 residues (labeled I through III for simplicity), while exon two
encodes thehomeodomain.Themajorityofreported mutationsin
the HOXA13 gene in patients with HFGS have been in-frame,
meiotically and mitotically stable, short (7–14 residue) expansions
within any one of the three large polyalanine tracts, while point
mutations comprise nearly all of the remaining mutations [Innis,
At least nine distinct diseases are associated with short poly-
alanine expansions in other genes encoded by imperfect triplet
Grant sponsor: University of Michigan, Department of Human Genetics
and the Department of Pediatrics.
Jeffrey W. Innis, 4909 Buhl Building, 1241 E. Catherine Street, Ann Arbor,
MI 48109-5618. E-mail: firstname.lastname@example.org
Article first published online in Wiley Online Library
(wileyonlinelibrary.com): 26 March 2013
How to Cite this Article:
Owens KM, Quinonez SC, Thomas PE,
Keegan CE, Lefebvre N, Roulston D, Larsen
novo HOXA13 polyalanine expansions
supports replication slippage without repair
in their generation.
Am J Med Genet Part A 161A:1019–1027.
? 2013 Wiley Periodicals, Inc.
expansions have been associated with earlier onset and shortened
median survival in amyotrophic lateral sclerosis [Blauw et al.,
2012]. While replication slippage and associated mechanisms
[Chen et al., 2005; Pearson et al., 2005; McMurray, 2010] have
been shown to cause the unstable contractions and expansions
associated with classic trinucleotide repeat diseases, polyalanine
[Warren, 1997; Brown and Brown, 2004]; however, this has not
been proven in HOXA13 polyalanine expansion mutations in
patients with HFGS. Replication slippage or slipped mispairing
mediated by misalignment of short repetitive sequences during
expansion (or contraction) mutations. Unequal sister chromatid
exchange or homologous recombination is not supported by the
observations of (1) polyalanine tract expansions involving at least
two duplications in single alleles of HOXD13 and FOXL2 [Brown
tion in PHOX2B [Trochet et al., 2007], and (3) lack of deletion
patients with polyalanine expansions in PHOX2B [Trochet et al.,
de novo HFGS associated with HOXA13 polyalanine expansions
allowing for experimental testing of these hypotheses.
MATERIAL AND METHODS
This study was approved by the University of Michigan Medical
Institutional Review Board. Genomic DNA was extracted from
HOXA13 exons 1 and 2 were amplified by PCR using custom
oligonucleotide primer pairs designed using Primer3 as previously
described [Innis et al., 2004]. To obtain independent clones, PCR
products were ligated into the p-GEM?T-Easy Vector system
(Promega Corp., Fitchburg, WI), electroporated MegaX DH10b
T1R Electrocomp? cells (Invitrogen/Life Technologies Corp.,
Grand Island, NY), and blue-white colony selection was used.
We used SNPs upstream and downstream of HOXA13 and
inverse PCR to determine SNP phase on each chromosome. BstBI
restriction enzyme sites located outside of the region of interest
by inverse PCR to appose upstream and downstream SNPs. The
using custom primers produced the final haplotypes (Fig. 3).
Luciferase assays were performed in NG108-15 cells comparing
G177R mutant construct was generated by using the QuikChange
site-directed mutagenesis kit (Stratagene, La Jolla, CA), and the
presence of this mutation was confirmed by DNA sequencing.
NG108-15 cells (ATCC#HB-12317) were maintained in DMEM
media (Gibco/Life TechnologiesCorp., Grand Valley, NY) supple-
mented with 10% FBS (Atlanta Biologicals, Norcross, GA), HAT
(Invitrogen), and 1% penicillin/streptomycin. Cells (1?105)
were seeded in 12-well plates and grown for 24hr at 378C with
5% CO2. Transfections were performed using FuGENE6 transfec-
tion reagent (Roche Applied Science, Indianapolis, IN), 0.1mg
pRL-CMV Renilla, and 0.25mg pCAGGS-HOXA13 wild type or
mutant,alongwith 0.5mgofapGL4.23plasmid (Promega) encod-
ing luciferase and containing an EphA7 cis-regulatory element
previously shown to be regulated by HOXA13 [Stadler et al.,
2001]. Empty pGL4.23 and pCAGGs expression vectors were
used as controls (red bars). The Dual-Glo Luciferase Assay system
(Promega) was used to detect luciferase activity 24hr posttrans-
fection in OptiPlate-96F black plates using a Fusion Microplate
Analyzer (Perkin Elmer, Waltham, MA). Six replicates of each
ed a total of three separate times. Results were normalized for
described by the manufacturer (Promega). Because the pGL4.23
vector also contains several HOXA13 binding sites, background
activation of the empty pGL4.23 by HOXA13 was also subtracted
from the final luciferase levels after Renilla normalization.
Patient 1 is a female born to nonconsanguineous parents. No
pregnancy or birth history was available. The patient had a history
of normal developmental milestones. We first evaluated her at 17
years, 10 months of age. At that time, her height was 150.2
centimeters (?3rd centile), her weight was 45.7kg (5–10th centile),
and her head circumference was 54.0cm (?35th centile). The
patient’s medical history was significant for genitourinary abnor-
malities including a small bladder, urinary leakage, and multiple
urinary tract infections necessitating numerous Mitrofanoff and
bladder augmentation procedures. Additional medical history was
significant for back and heel pain, a right inguinal hernia status—
postrepair, a left heminephrectomy, and scoliosis with no thor-
acolumbar vertebral segmentation anomaly. Clinical examination
and radiographic findings showed a short distal phalanx of the
thumbs bilaterally, micronychia of the thumbs bilaterally, limited
flexion of the DIP joint and distal phalanges of the fourth fingers
bilaterally, radial fifth finger clinodactyly bilaterally, short distal
phalanges of all toes bilaterally, short middle phalanges of all toes
bilaterally, fusion of the middle and distal phalanges of all toes
bilaterally, hypoplasia of the cuneiform bilaterally with right-sided
fusion with the metatarsal, and micronychia of the great toes
the patient was diagnosed with HFGS. The patient’s mother and
maternal half-sister were also examined and were unaffected
(Fig. 1B). The patient’s father is deceased but reportedly did not
have thumb, toe, or genitourinary abnormalities. A maternal half-
brother and paternal half-brother were also both reportedly unaf-
fected. Basedonthefamily history andphysicalexams, wehypoth-
esized that the first patient had de novo HFGS.
father who are nonconsanguineous. The patient was born at 39
weeks’ gestation and weighed 4.1kg (>90th centile) and was
Shortly after birth, the patient was diagnosed with bacterial pneu-
monia and spent time in the neonatal intensive care unit. The
him at 35 months of age. At that time, his height was 94.8
1020 AMERICAN JOURNAL OF MEDICAL GENETICS PART A
hishead circumferencewas49.8cm (54thcentile).Physicalexami-
well as partial cutaneous two-three-toe syndactyly bilaterally
(Fig. 5C,D). The patient was also found to have mild glandular
hypospadias with no history of urinary tract infections and occa-
sional toe-walking. A skeletal survey showed the following abnor-
malities: short middle phalanx of all fingers, short proximal and
distal phalanges of the thumbs, radial fifth finger clinodactyly
of all toes bilaterally, short proximal phalanges of the great toes,
absent ossification of the middle phalanges of toes 2 through
5 bilaterally, and short first metatarsals. Based on these findings,
the patient was diagnosed with HFGS.
The patient’s mother, father, brother, sister, maternal
grandfather, and maternal grandmother were also examined
and were unaffected. Based on the family history and physical
exams, we hypothesized that the second patient also had de novo
Amplification of an exon 1 segment including the large 18 residue
type length (arrow, Fig. 2A) and a larger amplicon. Sequencing of
independent clones from several independent amplifications
showed one allele with a nine-alanine expansion in tract III
(c.357_383 dup27, Fig. 2B) and, surprisingly, a second allele
with a point mutation, c.529G>C (p.Gly177Arg) (Genbank
accession no. JN628255; Fig. 2C), showing the two sequence
variations on separate homologs.
To determine if either mutation was inherited, maternal geno-
mic DNA from two independent blood samples was analyzed for
the presence of the nine-alanine expansion and the c.529G>C
point mutation identified in Patient 1, which showed that the
mother carried neither. While the mother could still be a low-level
HOXA13 PCR products, were detected in repeated amplifications
of leukocyte derived DNA. A SNP-based haplotype upstream and
her mother to determine the parental origin of the polyalanine
expansion. Patient 1’s HOXA13 SNP haplotype showed that the
polyalanine expansion must have originated de novo on the
maternal chromosome while the point mutation must have origi-
nated on the paternally inherited chromosome (Fig. 3).
To determine whether the polyalanine expansion occurred as a
result of unequal crossing-over between homologs during meiosis,
we used inverse PCR. Patient 1 was homozygous for the upstream
SNPs rs7812039 and rs13243033, whereas her mother was hetero-
zygous at these loci. Figure 3 also shows the results of inverse PCR
with maternal DNA. The de novo polyalanine expansion did not
arise by unequal crossing-over between maternal chromosome 7
homologs, and must have occurred as a result of either slipped
mispairing or unequal sister chromatid exchange in maternal mito-
sis prior to gamete production or during meiosis II, or by either
mechanism early on postzygotically. Although we did not examine
for mosaicism in Patient 1, a deleted product by PCR analysis was
never detected, suggesting that if the expansion had occurred
postzygotically, unequal sister chromatid exchange was unlikely.
Given that homozygous loss of Hoxa13 function is lethal
[Mortlock et al., 1996; Warot et al., 1997; Shaut et al., 2008],
heterozygous for two disease-causing mutations. However, the
c.529G>C (p.Gly177Arg) point mutation found in Patient 1
has not been previously reported in dbSNP (build 135) or in the
NHLBI Exome Sequencing Project (snp.gs.washington.edu/EVS/)
at position 177 is highly conserved in HOXA13 among vertebrates
[Mortlock et al., 2000; Lehoczky and Innis, 2008] and lies within a
(basic, hydrophilic) is not a conservative substitution for glycine
(small, neutral). Polyphen-2 analysis (genetics.bwh.harvard.edu/
pph2/)showedthissubstitutiontobe‘‘probably damaging’’ witha
score of 0.992. Analysis of the c.529G>C point mutation using
showed a rejected substitution score of 3.56, suggesting evolution-
ary selection against substitution at this nucleotide.
Therefore, to test the effect of this sequence variation on func-
tion, the p.Gly177Arg mutation was introduced into a plasmid
containing the mouse Hoxa13 cDNA through site-directed muta-
genesis and then the p.Gly177Arg HOXA13 transcriptional activa-
promoter in vitro. The EphA7 promoter is regulated in vivo by
Hoxa13 [Stadler et al., 2001]. As seen in Figure 4, the p.Gly177Arg
FIG. 1. Phenotype and pedigree of Patient 1. A: Limb phenotype in
thumbs and great toes with micronychia of the thumb and great
toe bilaterally (arrows). B: Pedigree of the patient’s family. Open
squares or circles mean the individual is unaffected. An asterisk
indicates individuals who were clinically examined.
OWENS ET AL.
mutation did not significantly affect transcriptional activation by
HOXA13 in this in vitro assay (P¼0.631). This result does not
preclude an effect of this amino acid substitution on regulatory
activity at other target loci in different cellular contexts, but it
underscores the importance of functional analysis of sequence
variants. Also, if Patient 1’s father was heterozygous for the
of paternal HFGS features.
The PCR analysis of large repeat tract III followed by cloning and
(c.358_375 dup18, inherited from his father) and a 10-alanine
contraction allele (c.355_384 del30, inherited from his asymptom-
atic mother and her asymptomatic father). Close examination of
Patient 2’s father showed a normal extremity exam (Fig. 5A,B),
suggesting either lack of penetrance or mosaicism. Paternal and
maternal HOXA13 PCR sequences matched Patient 2’s two alleles
(Fig. 5E,F). A SNP-based haplotype analysis upstream and down-
stream of HOXA13 was performed to identify the paternal chro-
mosome carrying the expanded alanine tract (Fig. 6). Both
unaffected siblings carried the same paternal allele as Patient 2
without the expanded alanine tract, suggesting that the father was
mosaic for the expansion. The PCR sequencing on the paternal
grandparents showed that neither were carriers, supporting the
hypothesis that Patient 2’s father sustained a postzygotic polyala-
the father to look for WT, expanded, and contracted polyalanine
tract III alleles. One week later, after minimal expansion, the
fibroblasts were harvested, and a sample was plated into ten 96-
well plates at an average concentration of 0.8 cells per well. Many
wells showed no growth, but genomic DNA from 58 wells was
failed to amplify. Thirty-four clones had both an expansion allele
and a normal allele; 10 clones showed only amplification of the
normal sized allele. Thus, 34/44 clones (77%) showed expansion
by PCR. A separate cloning experiment performed later after
further expansion of the original culture, plated at higher density,
detected no contracted allele among 159 samples. An additional
paternal DNA sample derived from saliva showed only six-alanine
alanine contraction allele product was observed in any genomic
PCR with paternal DNA from any tissue or with any of the clonal
FIG. 2. PolyalanineexpansionandrarevariantinPatient1.A:ThePCRandclonesequencinganalysisofHOXA13polyalaninetractIII.Lane1:Patient1
adjacent to the marker lane was a separate control reaction. B: Sequencing of cloned expansion products in the patient showed a 27bp
(nine-alanine) tandem duplication (c.357_383 dup27). Imperfect GCN triplets encoding alanine residues are depicted as shown; * nine-alanine
triplet codons duplicated in the expansion allele. C: Sequencing downstream of the polyalanine tract III showed a point mutation (c.529G>C,
asterisk indicates the mutated base in a PCR clone from the nonexpansion allele (c.529G>C).
1022 AMERICAN JOURNAL OF MEDICAL GENETICS PART A
FIG. 3. Haplotypes of Family 1. Haplotype analysis of Patient 1 showed that the polyalanine expansion occurred on the maternally inherited allele,
while the p.Gly177Arg mutation originated on the paternally inherited allele. Diamond refers to inferred haplotypes. SNPs and hg19 reference
genomic positions are: rs7812039 (chr7:27,248,891), ?9.2kb centromeric (upstream) of the 50end of HOXA13; rs13243033 (chr7:27248373),
?8.7kb centromeric of HOXA13; rs2189239 (chr7:27237453), ?2.2kb telomeric of the HOXA13 50end; rs757181 (chr7:27236958), ?2.7kb
telomeric of the HOXA13 50end; and rs2391397 (chr7:27236558), ?3.1kb telomeric of the HOXA13 50end.
OWENS ET AL.
Family 2 is the most informative regarding mechanism(s) for
HOXA13 polyalanine expansions. From four different sites
fibroblast clones of skin, only the expanded and normal allele sizes
were observed, supporting somatic and gonadal mosaicism in the
asymptomatic father of Patient 2. In addition, if the mechanism of
there should have been three different paternal cell lines: a six-
alanine expanded/WT, WT/WT and a six-alanine contracted/WT.
either a cell line with a six-alanine contraction/WT genotype was
is not unequal sister chromatid exchange. The former explanation
frequency polymorphism in the normal population, and indeed
survives quite well in Patient 2, his mother, and maternal grandfa-
residues would be deleterious. Thus, slipped mispairing appears
most likely as the mechanism of the postzygotic expansion in the
1997; Brown and Brown, 2004], which is alsotheoretically possible
data from either of the patients described here.
Arai et al.  proposed that unequal sister chromatid ex-
change may have led to de novo polyalanine expansions in the
PHOX2B gene in patients with congenital central hypoventilation
syndrome (CCHS; OMIM 209880); however, no data proving this
mechanism was presented. The authors proposed that this mecha-
well as the presence of normal contraction alleles. Contraction
alleles, although quite rare, have also been observed as presumably
normal variants in the HOXA13 gene [Brown and Brown, 2004;
Innis et al., 2004; Innis, 2006] and other genes with polyalanine
expansion (and the expected reciprocal deleted product) in
either maternal or proband DNA. Failure to observe the deleted
product does not eliminate the possibility of either unequal sister
chromatid exchange or slipped mispairing occurring in the mater-
However, our data in Family 2 strongly support slipped mis-
pairing during replication as the mechanism for this instance of
ly proposed as a mechanism for the instability of repetitive DNA
tracts in yeast mutants for mismatch repair genes [Strand et al.,
1993]. Trochet et al.  also presented a family where unequal
sister chromatid exchange was not a suitable explanation for a
PHOX2B polyalanine expansion in a mosaic (leukocyte DNA)
 also documented somatic mosaicism for PHOX2B expan-
sions, and did not identify shorter products expected for unequal
sister chromatid exchange in mosaic individuals. Additional poly-
alleles of HOXD13 and FOXL2 [Brown and Brown, 2004], also
support slipped mispairing. By replication slippage, both expan-
sions and contractions could occur, depending on whether tem-
plate or nascent strand misalignment was involved [Strand et al.,
1993; Chen et al., 2005; McMurray, 2010]. Finally, Cocquempot
et al.  suggested FosTES (Fork stalling and TEmplate
Switching) [Lee et al., 2007] as the mechanism for polyalanine
expansion in the Hoxd13 gene in the mouse mutant Dyc, whose
structure is quite different from typical HOXA13 or HOXD13
polyalanine expansions. The authors stated that polyalanine tracts
contain microhomology, symmetry, and specific sequence motifs
that commonly lead to FosTES.
In our report, we have proven that slipped mispairing during
1. Whereas paternal mosaicism in Family 2 indicates a postzygotic
origin for the expansion, the timing of expansion in Family 1 is
unclear. In addition, the 10-alanine contraction in Proband 2 was
inherited from the maternal grandfather. Thus, not unlike poly-
alanine expansions, this contraction has been stable in this family
allele would be an expected product of unequal sister chromatid
Wereported onthemolecular basisofHypodactyly inmice,a50
base pair deletion in Hoxa13, and proposed that the deletion arose
from slipped mispairing or unequal crossing over catalyzed by
nearby 10 base pair GC-rich microhomology in the first exon of
expansions as well as the Hypodactyly deletion in HOXA13 exon 1
associated with numerous GC-rich microhomologies supports the
hypothesis that the structure predisposes to expansions, contrac-
tions, anddeletionsof several typesdepending on thenatureof the
Finally, other than for PHOX2B [Parodi et al., 2008; Trochet
et al., 2007], this is the third investigation of mosaic polyalanine
expansion carriers documenting results consistent with slipped
FIG. 4. In vitro transcriptional activation function of p.Gly177Arg
HOXA13 on the EphA7 promoter. Percent luciferase activities
(relative to wt control) are indicated on the y-axis. HOXA13 wt or
G177R (normalized to Renilla and empty vector controls) are
1024AMERICAN JOURNAL OF MEDICAL GENETICS PART A
mispairing. In contrast, despite the possibility of unequal sister
chromatid exchange or homologous recombination generating
polyalanine expansions, there has been no report of a proven
occurrence. It will be important to identify mosaic individuals
for other polyalanine expansion disorders, and possibly for con-
tractions, and investigate particularly by clonal analysis whether
replication slippage is the predominant mechanism or not.
NHLBI Exome Sequencing Project: http://snp.gs.washington.
UCSC Genome Browser: http://genome.ucsc.edu/dbSNP135:
FIG. 5. Phenotype and polyalanine repeat size in Family 2. A,B: Hands and feet of asymptomatic father of Patient 2. C,D: Hands and feet of Patient 2
demonstrating typicalHFGS skeletal features. E:HOXA13 PCRproducts across large repeatIIIin Family2.Lane 1, asymptomatic father(6-alanine
expansion (c.358_375 dup18)/WT alleles); lane 2, asymptomatic mother (10-alanine contraction (c.355_384 del30)/WT); lane 3, Patient 2 (6-
alanine expansion/10-alanine contraction); lanes 4,5, asymptomatic sister and brother (WT/WT); lane 6, normal control; lane 7, water control. F:
Imperfect GCX triplets encoding alanine residues are depicted as shown; * six-alanine triplet codons duplicated in the expansion allele. The
contraction allele sequence of triplet codons inherited from the maternal grandfather is shown.
OWENS ET AL.
The authors thank the patients and their families for their research
participation. We thank Dr. Tom Glover for helpful comments.
This research was funded by the University of Michigan, Depart-
ment of Human Genetics and the Department of Pediatrics.
Dr. Innis is the Morton S. and Henrietta K. Sellner Professor of
FIG. 6. Pedigree and haplotypes of Family 2. Patient 2 with HFGS, filled in square. Dotted line: father of proband is mosaic for HOXA13 polyalanine
in tract III) from his asymptomatic mother. All individuals with an * were examined. Asymptomatic siblings of Patient 2 inherited the same
chromosome haplotype from their father, without the polyalanine expansion observed in Patient 2.
1026 AMERICAN JOURNAL OF MEDICAL GENETICS PART A
REFERENCES Download full-text
Arai H, Otagiri T, Sasaki A, Hashimoto T, Umetsu K, Tokunaga K,
Hayasaka K. 2007. De novo polyalanine expansion of PHOX2B in
congenital central hypoventilation syndrome: Unequal sister chromatid
exchange during paternal gametogenesis. J Hum Genet 52:921–925.
Blauw HM, van Rheenen W, Koppers M, Van Damme P, Waibel S,
Lemmens R, van Vught PWJ, Meyer T, Schulte C, Gasser T, Cuppen
LH. 2012. NIPA1 polyalanine repeat expansions are associated with
amyotrophic lateral sclerosis. Hum Mol Genet 21:2497–2502.
illness and trinucleotide repeats. Trend Genet 20:51–58.
Chen J-M, Chuzhanova N, Stenson PD, Ferec C, Cooper DN. 2005. Meta-
analysis of gross insertions causing human genetic disease: Novel muta-
tional mechanisms and the role of replication slippage. Hum Mutat
Cocquempot O, Brault V, Babinet C, Herault Y. 2009. Fork stalling and
template switching as a mechanism for polyalanine tract expansion
affecting the DYC mutant of Hoxd13, a new murine model of synpoly-
dactyly. Genetics 183:23–30.
Cooper GM, Stone EA, Asimenos G, Comparative NISC Program Se-
quencing, Green ED, Batzoglou S, Sidow A, 2005. Distribution and
intensity of constraint in mammalian genomic sequence. Genome Res
Goodman FR, Fryns J-P, Mortlock DP, Innis JW, Holmes LB, Donnenfeld
HOXA13 mutations and the phenotypic spectrum of hand-foot-genital
syndrome. Am J Hum Gen 67:197–202.
Stephens K, editors. GeneReviews [Internet]. Seattle, WA: University of
Washington Seattle; 2006.
Innis JW, Mortlock D, Chen Z, Ludwig M, Williams ME, Williams TM,
2004. Polyalanine expansion in HOX A13: Three new affected families
and the molecular consequences in a mouse model. Hum Mol Genet
generating nonrecurrent rearrangements associated with genomic dis-
orders. Cell 131:1235–1247.
Lehoczky J, Innis JW. 2008. Expanded HOXA13 polyalanine tracts in a
monotreme. Evol Dev 10:433–438.
human development. Nat Rev Genet 11:786–799.
Mortlock DP, Innis JW. 1997. Mutation of HOXA13 in hand-foot-genital
syndrome. Nat Genet 15:179–180.
(Hd): A deletion in Hoxa13 leads to arrest of digital arch formation. Nat
Mortlock DP, Sateesh P, Innis JW. 2000. Evolution of N-terminal sequen-
ces of the vertebrate HOXA13 protein. Mamm Genome 11:151–158.
Parodi S, Bachetti T, Lantieri F, Di Duca M, Santamaria G, Ottonello G,
Matera I, Ravazzolo R, Ceccherini I. 2008. Parental origin and somatic
mosaicism of PHOX2B mutations in congenital central hypoventilation
syndrome. Hum Mutat 29:206–215.
of dynamic mutations. Nat Rev Genet 6:729–742.
Shaut CAE, Keene DR, Sorensen LK, Li DY, Stadler HS. 2008. HOXA13 is
essential for placental vascular patterning and labyrinth endothelial
specification. PLoS Genet 4:e1000073.
Stadler HS, Higgins KM, Capecchi MR. 2001. Loss of Eph-receptor
expression correlates with lossof celladhesion andchondrogenic capac-
ity in Hoxa13 mutant limbs. Development 128:4177–4188.
simple repetitive DNA in yeast by mutations affecting DNA mismatch
repair. Nature 365:274–276.
Tishkoff DX, Filosi N, Gaida GM, Kolodner RD. 1997. A novel mutation
DNA mismatch repair. Cell 88:253–263.
Trochet D, de Pontual L, Keren B, Munnich A, Vekemans M, Stanislas L,
Amiel J. 2007. Polyalanine expansions might not result from unequal
crossing-over. Hum Mutat 28:1043–1044.
Warot X, Fromental-Ramain C, Fraulob V, Chambon P, Doll? e P. 1997.
Gene dosage-dependent effects of the Hoxa-13 and Hoxd-13 mutations
on morphogenesis of the terminal parts of the digestive and urogenital
tracts. Development 124:4781–4791.
Warren ST. 1997. Polyalanine expansion in synpolydactyly might
result from unequal crossing-over of HOXD13. Science 275:408–
Williams TM, Williams ME, Innis JW. 2005. Range of HOX/TALE super-
class associations and protein domain requirements for HOXA13-MEIS
interaction. Dev Biol 277:457–471.
OWENS ET AL.