brachydactyly - Journal of Medical Genetics

Document Sample
brachydactyly - Journal of Medical Genetics Powered By Docstoc
					                               Downloaded from on January 7, 2013 - Published by

IJMed Genet 1995;32:851-854                                                                                                    851

                               Evaluation of candidate                           genes       for familial
                               Joan M Mastrobattista, Pascal Dolle, Susan H Blanton, Hope Northrup

                               Abstract                                            complexes (HOXA, B, C, D) and encode po-
                               Type Al brachydactyly in humans is a                sitional information along the anterior-pos-
                               recognisable syndrome characterised by              terior body axis and the limb axes.3589 These
                               shortening of the middle phalanx of all             genes have been conserved in evolution from
                               digits with occasional fusion of the middle         invertebrates to humans.39 10 A loss of function
                               and terminal phalanges. The purpose of              mutation of the murine Hoxd-13 gene through
                               this study was to evaluate candidate genes          gene targeting resulted in mice with tetramelic
                               for type Al brachydactyly in two families           skeletal abnormalities." The defects, which
                               with multiple affected members. Several             were restricted to the distal extremities of the
                               classes of genes have been implicated in            limbs (the forefeet and hindfeet), included the
                               the control of distal limb development              reduction or absence of some skeletal elements
                               including homeobox containing genes                 as well as skeletal fusions. Muscles, tendons,
                               (MSXI, MSX2), some members of the                   and skin were apparently not affected. In these
                               homeobox gene family, and genes encod-              respects, the defects found in the murine Hoxd-
                               ing growth factors of the FGF, TGF, and             13 mutation are similar to those found in
                               PDGF families. Homeobox (Hox) genes                 human type Al brachydactyly.
                               are a family of developmental control                  Another group of candidate genes are certain
                               genes activated early in embryogenesis              growth factor genes which are specifically ex-
                               that encode positional information along            pressed during limb development and can have
                               the anterior-posterior body axis and spe-           stimulatory or inhibitory effects on limb growth
                               cify distinct spatial domains within de-            and patterning.'2-'5 Growth factors have also
                               veloping limbs. Growth factor genes can             been shown to influence Hox gene expression
                               regulate the proliferation and differ-              in in vitro experiments.3 Growth factor genes
                               entiation of various embryonic structures           of interest include: fibroblast growth factor
                               including limb buds and have been shown             (FGF)-1,-2,-4,12-15 transforming growth factor-
                               to influence Hox gene expression. Can-              alpha (TGF-ot),'6 platelet derived growth fac-
                               didate genes HOXD, MSX1, MSX2, FGF-                 tor-alpha (PDGF-a),17 and platelet derived
                               1, and FGF-2 were excluded in one family.           growth factor-beta (PDGF-[).'61" Many of
Department of                  The brachydactyly type Al gene or locus             these genes have been cloned and poly-
Gynecology, and                was not found in either of the two families         morphisms described.
Reproductive                   studied.                                               The purpose of this study was to evaluate
Sciences,                                                                          candidate genes for type Al brachydactyly in
University of Texas
Medical School-                (J Med Genet 1995;32:851-854)                       humans. The similarities between the human
Houston, Texas, USA                                                                type Al brachydactyly and the murine Hoxd-
J M Mastrobattista                                                                 13 mutation31' suggest that the human disease
Institut de Genetique          Congenital hand anomalies have an estimated         may result from a mutation in the HOXD
et de Biologie                 prevalence at birth of approximately 5/1000.1       complex. We therefore analysed the HOXD
Moleculaire et                 One type of congenital hand anomaly is bra-         locus as well as the MSX1 and MSX2 genes
Cellulaire, CNRSI              chydactyly which Bell classified into seven         by sequencing gel electrophoresis for linkage
Strasbourg, France             types: Al, A2, A3, B, C, D, and E.2 In the          to type Al brachydactyly in two families con-
P Dolle                        type A brachydactylies, variable abnormalities      taining a total of 11 affected persons (figure).
Department of                  of the middle phalanges are observed. Type Al       We also evaluated growth factor genes FGF-1,
Pediatrics, University         brachydactyly (Farabee type, MIM 112500) is         FGF-2, PDGF-oa, PDGF-3, and TGF-a by
of Virginia Medical            an autosomal dominant condition2 for which          sequencing gel electrophoresis or Southern
School,                        Bell's criteria are: shortening of the middle       blotting.
Virginia, USA                  phalanx of all digits (both hands and feet),
S H Blanton                    shortening of the proximal phalanges of the
Division of Medical
                               first digit, and occasional fusion between the      Patients, materials, and methods
Genetics, Department           middle and terminal phalanges. Variable ex-         Two families diagnosed with type Al bra-
of Pediatrics,                 pressivity is common.                               chydactyly were ascertained through a review
University of Texas               Two classes of homeobox genes have been          of medical records at the Shriner's Hospital
Medical School-
Houston, 6431 Fannin,          implicated in the control of distal limb de-        for Crippled Children-Houston Unit and the
MSB 3.144, Houston,            velopment, namely some of the HOXD and              University of Texas Houston Medical Genetics
Texas 77030, USA               HOXA genes (previously called HOX4 and              clinic. One family was of Scandinavian descent
H Northrup                     HOX1)3-5 as well as the MSX1 and MSX2               (five affected members) and the other of Mex-
Dr Northrup.
                 to:           genes (previously called HOX7 and HOX8,             ican descent (six affected members). The in-
                               although these genes do not structurally belong     stitutional review boards at both institutions
Received 30 March 1995
Revised version accepted for   to the Hox gene family).67 Hox genes are a          approved the protocol and written informed
publication 19 June 1995       family of genes which are clustered in four         consent was obtained from both families.
                          Downloaded from on January 7, 2013 - Published by

852                                                                                                   Mastrobattista, Dolle, Blanton, Northrup

                                                                                            Samples were overlaid with mineral oil and
                                        -0o                                                 processed through 30 temperature cycles con-
                                                                                            sisting of denaturation, annealing, and elong-
                                                                                            ation (temperature cycles varied depending on
                                                                                            the primer pair involved). Aliquots of amplified
                                                                                            DNA were mixed with a stop solution con-
                                                                                            taining formamide and electrophoresed on
                                                                                            standard polyacrylamide sequencing gels. Gels
                                                                                            were processed, dried, and exposed to auto-
      0         0         0           0                                                0
                                                                                            radiography for 24 to 48 hours at room tem-
                                                                                            perature. A standard M-13mp18 sequencing
                                                                                            ladder was used to compare DNA banding
                                                                                                      Primer pairs from within HOXD,2'
                                                                                            MSX1,22 MSX2,23 FGF-1,24 and PDGF-1325
                                                                        I                 I were used in the experiments.
                                            9                             *             0      One candidate gene, TGF-oz,26 contained a
                                                                                            TaqI site detectable on an agarose gel stained
                                                                                            with ethidium bromide after PCR am-
                                                                                            plification. A DNA molecular weight marker
                                                                                            was used as a standard and specific fragments
                                                                                            were identified.
                                                                                               Two candidate genes, FGF-2 and PDGF-oa,
                                                                                            contained restriction fragment length poly-
                                                                                            morphisms (RFLPs) located in introns. FGF-
                                                                                            2 contains a HindIII site polymorphism. A pair
                                                                                            of synthetic oligonucleotides, (TCAAGCTA-
                                                                                            CAACTTCAAGCA) and (AGAAGCCAG-
                      11                                                                    TAATCTTCCATC) were designed to amplify
                                                                                            the second exon and used as a probe in the
                                             * 0                                            experiment.27 PDGF-c contains a Styl poly-
                                                                                            morphism28 which was detected by Southern
                                                                                            blotting and probing with a 1 7 kb PDGF-ct
                                                                                            probe (clone-IB657; ATCC number-86274).19
                                                                                               Linkage analysis was performed using the
                                                                                            MLINK program of the LINKAGE package
                                                                                            (version 5.03).29 Lod scores between the dis-
Family 1 (top) and family 2 (bottom) are depicted. The filled circles beneath the pedigree  ease and each marker locus were generated
symbols denote people enrolled in thestudy.                                                 for each family and then summed over both
                                                                                            families.30 The two families had the potential
                                                                                            of generating a maximum total lod score of
                                   Of the 17 participants in the study, 13 were 3-01 of which family 1 contributed 2- 107 and
                                examined by one of the authors (JM or HN). family 2 contributed 0 903. Markers within the
                                The remaining four were examined by the various genes were studied for inclusion or
                                family's private physician who provided specific exclusion as candidate genes; therefore, a lod
                                information. The thorough examination of score of -2 eliminated the gene as a candidate.
                                each participant's hands and feet included:
                                hand length, middle finger length, evaluation
                                of palmar creasing patterns, and shoe size. The Results
                                proband in each family underwent radiological The families analysed contained 11 persons
                                evaluation oftheir hands and feet. Persons were affected with type Al brachydactyly (figure).
                                regarded as affected with type Al brachydactyly Two point lod scores between type Al bra-
                                if Bell's criteria were fulfilled. Pedigrees of the chydactyly and the candidate genes are pre-
                                families are shown in the figure.                           sented in the table. HOXD was informative in
                                   Genomic DNA was extracted from fresh, both families, and combined lod scores at 0=
                                anticoagulated blood samples from indicated 0 05 were less than - 2 0; however, family
                                members of the affected families (figure)."9 Five 2 was informative without any recombination
                                candidate loci (HOXD, MSX1, MSX2, FGF- yielding a lod score of 0 9. Lod scores in family
                                1, PDGF-P) contained dinucleotide repeat 1 were negative. Therefore HOXD was elim-
                                polymorphisms which were tested in the famil- inated as a candidate locus in family 1 but not
                                ies.                                                        in family 2. MSXl was informative only in
                                   Standard polymerase chain reactions (PCR), family 1 and was eliminated as a candidate
                                as described by Weber and May,20 were modi-                 gene with a lod score of -3 90 at 0=0 01.
                                fied and carried out in 20 p1 volumes which MSX2 was informative only in family 2 and
                                contained 20 ng genomic DNA, 100 ng each was eliminated as a candidate gene with a
                                of oligonucleotide primer, 200,mol/I each of lod score of -2-10 at 0=0 001. FGF-1 was
                                dGTP, dCTP, dTTP, dATP, 0-08 gl ocP32 informative in both families and excluded as a
                                dCTP at 3000 Ci/mmol, 2-0 gl of 10 x buffer candidate gene with a lod score of - 2 10 at a
                                (100 mmol/l Tris-HCl, 15 mmol/l MgC12, 0 of 0 05. FGF-2 was informative only in family
                                500 mmol/l KC1, pH 8-3 at 20°C), 0-25 units 1 and was eliminated as a candidate gene with
                                Taq (Perkin-Elmer Cetus, Norwalk, CT). a lod score of -2-10 at 0=0 001. Other can-
                               Downloaded from on January 7, 2013 - Published by

Evaluation of candidate genes for familial brachydactyly                                                                                                      853

Lod scores and number of informative families for each locus. PDGF-x, PDGF-/3, and            responsible for the mouse brachypodism muta-
TGF-x were uninformative in both families.
                                                                                              tion, whose phenotype is a reduction in the
                  Family Reconmbination fraction(V.)
                                                                                              length and number of the long bones of the
                         0 00     0 001      01      005      0-10     0-15   0-20   0 30     limb.3' In addition, the fibroblast growth factor-
HOXD              1       -x       -9 90    -5-91    -3-16    -2-03 -1-40 -1-00
                                                                                              4 protein (FGF-4) has been shown to stimulate
                                                                                     -0 45
                  2         0 90     0 90     0-89     0-81     0-72   0-62   0 57     0 30   proliferation of cells in distal limb mesenchyme
                  1 &2
                                                              - 1-31
                                                                     -0-78 -0 43
                                                                     -0-65 -0-38
                                                                                              and cause limb outgrowth while bone mor-
HOXMSX2           2       -x       -2-10    -1 11    -0-46    -0-23  -0-12 -0-06     -0-01    phogenetic protein-2 (BMP-2) can counteract
FGF-1             1
                                   -5 40
                                            -3 40
                                                                     -1-05 -0-80
                                                                     -0 04 -0-02
                                                                                              this growth promoting effect.'4 Very recently,
                  1 &2    -x       - 5 57   - 3 55   -2-10    - 1-46 - 1-09 -0-82    -0 44    the FGF-8 gene was also shown to be spe-
FGF-2             1       -x       -2*10    -1-11    -0 44    -0-19 -0-06     0-01     0 07   cifically expressed in developing limbs.32 All of
Lod scores are recorded in all cases where informative by family number. When both families   these genes represent additional candidates to
are informative, combined scores are calculated.                                              investigate. The endeavour to find candidate
                                                                                              genes should help unravel the link between
                                                                                              altered gene regulation and the perturbation of
                               didate genes tested which were uninformative                   a developmental process such as limb pat-
                               in both families were PDGF-oi, PDGF-,B, and                    terning.
                               TGF-oc.                                                        We wish to thank Jacqueline T Hecht, PhD and Jonathan D
                                                                                              Stein, BS for supplying several primers used in this project. We
                                                                                              also wish to thank the families involved in the study for their
                                                                                              participation and the Shriners Hospital for Crippled Children-
                               Discussion                                                     Houston Unit.
                               DNA from members of two families (17
                               people) affected with brachydactyly type Al                       1 Zguricas J, Snijders PJLM, Hovius SER, Heutink P, Oostra
                                                                                                     BA, Lindhout D. Phenotypic analysis of triphalangeal
                               was tested for linkage to eight candidate genes                       thumb and associated hand malformations. _7 Med Genet
                               to determine whether one might be the gene                             1994;31 :462-7.
                                                                                                2 Temtamy S, McKusick V. Brachydactyly as an isolated
                               responsible for the condition in these two famil-                     malformation. In: Bergsma D, Mudge JR, Paul NW,
                               ies. While the number and size of the families                        Greene SC, eds. The genetics of hand malformations. Vol
                                                                                                      14(3). New York: A R Liss, 1978:187-226.
                               were small, we had the capability either to                      3 Redline RW, Neish A, Holmes LB, Collins T. Biology of
                               include linkage with a maximum lod score of                           disease-homeobox genes and congenital malformations.
                                                                                                     Lab Invest 1992;66:659-70.
                               3-01 as calculated by MLINK or exclude link-                     4 Dolle P, Izpisua-Belmonte JC, Boncinelli E, Duboule D.
                               age with a lod score of <-2. With only two                            The Hox-4.8 gene is localized at the 5' extremity of the
                                                                                                     Hox-4 complex and is expressed in most posterior parts
                               families, we were unable to detect genetic                            of the body during development. Mech Dev 199 1;36:3-13.
                              heterogeneity.                                                    5 Duboule D. The vertebrate limb: a model system to study
                                                                                                     the hox/HOM gene network during development and
                                  The homeobox genes and growth factor                               evolution. BioEssays 1992; 14:375-84.
                              genes were among the most promising can-                          6 Taylor-Zimmerman GP, Muneoka K. Embryonic limb bud
                                                                                                     regulation correlates with msx-1 expression. Proceedings of
                               didate genes available for the following reasons.                     the fourth international conference on limb development and
                              Earlier studies on mutations in Drosophila and                         regeneration. Asilomar, CA, July 1992: A846.
                                                                                                7 Coelho CND, Sumoy L, Rogina B, Upholt WB, Kosher
                              Xenopus embryos strongly suggested that mem-                           RA. Roles of the chicken homeobox-containing genes
                              bers of the FGF and TGF families may provide                           ghox-8, ghox-7, and ghox-4.6 in pattern formation during
                                                                                                     limb development. In: Fallon JF, Goetinck PF, Kelley
                              signals required for the early establishment of                        RO, Stocum DL, eds. Limb development and regeneration,
                              the body plan.3 Some growth factors belonging                          Part A. 4th ed. New York: Wiley-Liss, 1993:61-70.
                                                                                                8 Thompson MW, McInnes RR, Willard HF. The human gene
                              to the same families are specifically expressed                        map: gene mapping and linkage analysis. In: Thompson and
                              in the developing vertebrate limb and are able                         Thompson genetics in medicine. 5th ed. Philadelphia: W B
                                                                                                     Saunders, 1991:167-99.
                              to interfere with limb growth and patterning                      9 Watson JD, Gilman M, Witkowski J, Zoller M, eds. Genes
                              under experimental conditions.314 Several                              control development of drosophila. In: Recombinant DNA.
                                                                                                     2nd ed. New York: W H Freeman, 1992:389-407.
                              HOXA, C, and D genes display specific tem-                       10 Acampora D, D'Esposito M, Faiella A, et al. The human
                              poral and spatial expression domains in de-                            HOX gene family. Nucleic Acids Res 1989;17:10385-402.
                                                                                               11 Dolle P, Dierich A, LeMeur M, et al. Disruption of the
                              veloping limbs.45 Furthermore, a Hoxd-13                               Hoxd-13 gene induces localized heterochromy leading to
                              targeted mutation in the mouse showed a                                mice with neotenic limbs. Cell 1993;75:431-41.
                                                                                               12 Anderson R, Landry M, Muneoka K. In vitro maintenance
                              phenotype somewhat analogous to human bra-                             of posterior (ZPA) signalling in mouse limb bud cells. In:
                              chydactyly type Al ." The MSX1 and MSX2                                Fallon JF, Goetinck PF, Kelley RO, Stocum DL, eds.
                                                                                                     Limb development and regeneration. Part A. 4th ed. New
                              homeobox genes are believed to be involved in                         York: Wiley-Liss, 1993:371-80.
                              epithelium/mesenchyme interactions which are                    13 Reginelli AD, Anderson R, Muneoka K. ZPA signaling, fgf-
                                                                                                     2 and hox-d gene expression. Proceedings of the fourth
                              critical for limb growth and patterning.67                             international conference on limb development and regeneration.
                                  The brachydactyly type Al gene was not                            Asilomar, CA, July 1992:A849.
                                                                                               14 Niswander L, Martin GR. Fgf-4 and bmp-2 have opposite
                              found among the candidate genes tested. Over-                         effects on limb growth. Nature 1993;361:68-71.
                              all, five loci (HOXD, MSX1, MSX2, FGF-1,                        15 Baird A, Klagsbrun M. The fibroblast growth factor family:
                                                                                                    an overview. Ann N YAcad Sci 199 1;638:xi-ii.
                              and FGF-2) were excluded in either one or                       16 Adamson ED. Oncogenes in development. Development
                              both families. Candidacy of HOXD remains,                              1987;99:449-7 1.
                                                                                              17 Carrington JL, Potts JD. Detection of messenger RNA for
                              however, equivocal as it was weakly linked in                         a platelet-derived growth factor (PDGF) a receptor-like
                              one family but not the other. Future objectives                       molecule in chicken limb buds using the polymerase chain
                                                                                                    reaction (PCR). In: Fallon JF, Goetinck PF, Kelley RO,
                              are to ascertain additional families and to test                      Stocum DL, eds. Limb development and regeneration. Part
                              these and other candidate genes.                                      B. 4th ed. New York: Wiley-Liss, 1993:667-72.
                                                                                              18 Chen P, Carrington JL, Paralkar VM, Pierce GF, Reddi
                                 Future areas of study include the evaluation                       AH. Chick limb bud mesodermal cell chondrogenesis:
                              of additional Hox genes, such as those be-                            inhibition by isoforms of platelet-derived growth factor
                                                                                                    and reversal by recombinant bone morphogenetic protein.
                              longing to the HOXA and HOXC complexes,                               Exp Cell Res 1922;200:110-17.
                                                                                              19 Martin AO, Northrup H, Ledbetter DH, et al. Prenatal
                              in brachydactyly type Al patients or in other                         detection of 46,XY,rec(5),dup q,inv(5) (p1 3q33) using
                              human congenital limb abnormalities. Recent                           DNA analysis, flow cytometry, and in situ hybridization
                                                                                                    to supplement classical cytogenetic analysis. Am _7 Med
                              investigations have shown that a member of                            Genet 1988;31:643-54.
                              the TGF-, superfamily (the Gdf-5 gene) is                       20 Weber JL, May PE. Abundant class of human DNA poly-
      Downloaded from on January 7, 2013 - Published by

854                                                                                        Mastrobattista, Dolle, Blanton, Northrup
             morphisms which can be typed using the polymerase chain          TaqI polymorphic regions. Am _7 Hum Genet 1993;53:
             reaction. Am Hum Genet 1989;44:388-96.                           168-75.
        21 Rosen DR, Brown RH Jr. Dinucleotide repeat poly-              27 Abraham JA, Whang JL, Tumolo A, et al. Human basic
             morphism in the HOX4E locus. Hum Mol Genet 1993;2:               fibroblast growth factor: nucleotide sequence and genomic
             617.                                                             organization. EMBO_7 1986;5:2523-8.
        22 Padanilam BJ, Stadler HS, Mills KA, et al. Characterization   28 Ferns GAA, Ross R. StyI RFLP of the human platelet
             of the human HOX7 cDNA and identification of poly-               derived growth factor (PDGF) A-chain gene. Nucleic Acids
             morphic markers. Hum Mol Genet 1992;1:407-10.                    Res 1990;18:208.
        23 Jabs EW, Muller U, Li X, et al. A mutation in the ho-         29 Lathrop GM, Lalouel JM, Julier C, Ott J. Strategies for
             meodomain of the human MSX2 gene in a family affected            multilocus linkage analysis in humans. Proc Natl Acad Sci
             with autosomal dominant craniosynostosis. Cell 1993;75:          USA 1984;81:3443-6.
             443-50.                                                     30 Ott J. Analysis of human genetic linkage. Baltimore: Johns
        24 Li X, Jaye M, Crumley G, et al. Dinucleotide repeat poly-          Hopkins University Press, 1985.
             morphism in the human fibroblast growth factor acidic       31 Storm EE, Huynh TV, Copeland NG, Jenkins NA, Kingsley
             (FGFA) gene on chromosome 5. Hum Mol Genet 1992;1:               DM, Lee S. Limb alterations in brachypodism in mice due
             216.                                                             to mutations in a new member of the TGF3-superfamily.
        25 Patterson H, Mitchell PJ, Cooper CS, Stratton MR. Di-              Nature 1994;368:639-43.
             nucleotide repeat polymorphisms at the SIS locus. Nucleic   32 Crossley PH, Martin GR. The mouse Fgf8 gene encodes a
             Acids Res 1990;18:5917.                                          family of polypeptides and is expressed in regions that
        26 Qian JF, Feingold J, Stoll C, May E. Transforming growth           direct outgrowth and patterning in the developing embryo.
             factor-alpha: characterization of the BamHI, RsaI and            Development 1995;121:439-51.
                    Downloaded from on January 7, 2013 - Published by

                                  Evaluation of candidate genes for familial
                                  J M Mastrobattista, P Dollé, S H Blanton, et al.

                                  J Med Genet 1995 32: 851-854
                                  doi: 10.1136/jmg.32.11.851

                                  Updated information and services can be found at:

                                  These include:
         References               Article cited in:

     Email alerting               Receive free email alerts when new articles cite this article. Sign up in
           service                the box at the top right corner of the online article.


To request permissions go to:

To order reprints go to:

To subscribe to BMJ go to:

Shared By: