Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Linkage studies of x-linked cleft palate and ankyloglossia in a British Columbia native kindred Gorski, Sharon M. 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_gorski_sharon.pdf [ 6.81MB ]
JSON: 831-1.0086108.json
JSON-LD: 831-1.0086108-ld.json
RDF/XML (Pretty): 831-1.0086108-rdf.xml
RDF/JSON: 831-1.0086108-rdf.json
Turtle: 831-1.0086108-turtle.txt
N-Triples: 831-1.0086108-rdf-ntriples.txt
Original Record: 831-1.0086108-source.json
Full Text

Full Text

LINKAGE STUDIES OF X-LINKED CLEFT PALATE AND ANKYLOGLOSSIAIN A BRITISH COLUMBIA NATIVE KINDREDbySHARON M. GORSKIB.Sc., Simon Fraser University, 1990A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESGENETICS PROGRAMMEWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIANovember 1992© Sharon M. Gorski, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of Medical Genetics (Genetics Programme)The University of British ColumbiaVancouver, CanadaDate ^11./ 1479g-DE-6 (2/88)ABSTRACTHuman craniofacial malformations are a class of common congenitalanomalies. Their etiology is heterogeneous and often poorly understood. Toelucidate the nature of craniofacial defects at the molecular level, one approach is tostudy the exceptional examples of malformations which segregate in families assingle gene disorders. This study employs such an approach: it is directed towardthe isolation of a locus responsible for cleft palate and ankyloglossia which segregateas a single X-linked trait (CPX) in a British Columbia (B.C.) Native kindred.The original description (Lowry 1970) of the clefting defect in the B.0 kindredincluded submucous cleft palate and bifid or absent uvula. Sixty-three of the members were clinically reevaluated and it was observed that some of theaffected males and carrier females also present with ankyloglossia (tongue-tie).Ankyloglossia previously has been associated with X-linked cleft palate in anIcelandic kindred in which a locus responsible for cleft palate was provisionallyassigned to the Xq21.3-q22 region (Moore et al. 1987; Ivens et al. 1988). This thesisdescribes linkage analyses in the B.C. kindred which were initiated with DNA markersfrom the Xq21-q22 region and were later expanded to include markers from the Xq13region. No recombination was observed between CPX and the DNA markersDXS447 (peak lod score [Zmax = 9.38), DXS72 (Zmax = 7.74; Gorski et al. 1992) andDXS326 (Zmax = 2.27). A new polymorphic DNA marker, X850A/L-7, was generatedand mapped to the Xq21.1-q21.3 region. X850A/L-7 was found to be partiallyinformative in the B.C. family and also nonrecombinant with respect to CPX (Zmax =1.91). Recombination was observed between CPX and PGK1 (Zmax = 7.63 atrecombination fraction [4] = 0.03) and between CPX and DXYS1 (Zmax = 5.59 at 'd =0.04). These results localize B.C. CPX between PGK1 and DXYS1 in the Xq13.3-q21.31 region.iiTABLE OF CONTENTSABSTRACT^ iiTABLE OF CONTENTS^ iiiLIST OF TABLESLIST OF FIGURES^ viACKNOWLEDGEMENTS vii1. INTRODUCTION^ 11.1 General introduction^ 11.2 Development of the palate in humans^ 21.3 Development of the tongue in humans 31.4 Congenital malformations of the palate in humans^ 41.5 Congenital malformations of the tongue in humans 61.6 Positional cloning^ 71.6.1 Regional localization^ Linkage mapping DNA polymorphisms^ 121.6.2 Identification of nearest flanking markers^ 141.7 Genetic analysis of isolated cleft palate in humans 172. MATERIALS AND METHODS^ 212.1 Pedigree and clinical evaluation of family members^ 2122 Cytogenetic analysis 232.3 DNA preparation, Southern transfer, and hybridization 2324 DNA polymorphisms^ 242.5 Linkage analyses 282.6 Cosmid library screen 282.7 Ce# lines^ 292.8 Field inversion gel electrophoresis (Sfil test blot)^ 302.9 Preparative field inversion gel electrophoresis to enrich for theDXS95 region^ 302.10 Human repeat element-mediated PCR^ 313. RESULTS^ 333.1 Regional localization of CPX^ 333.1.1 Cytogenetics 333.1.2 Genotypes^ 333.1.3 Two-point linkage analysis^ 393.1.4 Multipoint linkage analysis 423.2 Refinement of crossover map positions 443.2.1 The DXS95 region: screening for additional polymorphismsand generation of a cloning source^ 443.2.2 Human repeat element-mediated PCR 473.2.3. Mapping human repeat element-mediated PCR products bySouthern blot hybridization: localization to the DXS72 -DXS95 region^ 473.2.4 Cloning and mapping of A1 B/L 1 S PCR products^533.2.5 Physical mapping of DXS326^ 563.2.6 Screening for polymorphisms with new markers from Xq21^574. DISCUSSION^ 604.1 Regional localization of CPX^ 604.2 The Xq13.3-q21.31 region of the human genome^ 644.3 Generation of DNA markers from within the CPX candidate region^664.4 The B.C. CPX phenotype 724.5 Conclusions^ 744.6 Summary 764.7 Proposals for further research^ 77REFERENCES^ 79APPENDIX 1. Cytogenetics reports^ 88ivLIST OF TABLESTable 1. X chromosome markers used in present study^ 25Table 2. B.C. kindred genotypes for informative markers from Xq13-q22^35Table 3. Two-point lod scores between CPX and X chromosome marker loci^40vLIST OF FIGURESFigure 1. Pedigree of B.C. CPX kindred members used for linkage analysis^22Figure 2. Autoradiograph of Southern blot illustrating DXS447 genotypes for aportion of the B.C. CPXfamily^ 37Figure 3. Demonstration of DNA marker haplotypes for a portion of the B.C. family ^38Figure 4. X chromosome map illustrating the B.C. CPX candidate region^41Figure 5. Multipoint linkage analysis of CPX vs. a fixed loci map^43Figure 6. Schematic representation of preparative FIGE gel size fractionation andSouthern blot analysis of Sfil digested Cl2D DNA 46Figure 7. Human repeat element-mediated PCR products^ 48Figure 8. Deletion breakpoints of the four male Xq21 deletion cell lines used inthis study^ 50Figure 9. Mapping human repeat element-mediated PCR products by Southernblot hybridization: localization to the DXS72 - DXS95 region^51Figure 10. Mapping individual Al B/L1 S PCR products by Southern blothybridization^ 54Figure 11. X850A/L-7 Tacit polymorphism^ 59viACKNOWLEDGEMENTSI thank my supervisor, Dr. Paul Goodfellow, for providing guidance,enthusiasm, support and opportunity. I would also like to thank the members of mysupervisory committee, Dr. J. Friedman, Dr. D. Juriloff and Dr. R. McMaster, for theirencouragement, helpful comments and discussions. I am grateful to Patricia Birch forupdating the B.C. family pedigree, and performing blood sampling and clinicaldiagnoses, Dr. Fred Dill and Dr. Dagmar Kalousek for performing the cytogeneticanalyses, Dr. Steve Wood for providing helpful assistance with the LINKAGEprograms, and Dr. Virginia Diewert for sharing her time for interesting discussions.Thank-you to the members of the laboratory, Karen Adams, Angie Brooks-Wilson, Helen McDonald, Diane Miller, and Duane Smailus who all provided experttechnical advice and much moral support during the past two years. Special gratitudeto Karen Adams who performed many of the DNA sample preparations and DNAgenotypings, and to Diane Miller who shared with me, step by step, the experiencesof graduate school. I am grateful to my family and friends for their encouragementand support throughout. Special thanks to my husband, Marco Marra, for sharinginterest in my work, providing patient support, love, and friendship. Thank-you alsofor never failing to make me laugh.I thank the B.C. family members for their participation which has made thisstudy possible. Financial assistance was provided by a Medical Research Council ofCanada studentship.vii1. INTRODUCTION1.1 General introductionCleft palate is a common congenital malformation in humans. It may be afeature of malformation syndromes which are caused by chromosomal, Mendelian orenvironmental factors (reviewed in Melnick et al. 1980; Sperber 1989). Alternatively,cleft palate may occur in isolation where there is evidence for involvement of bothgenetic and environmental factors, but underlying mechanisms are unknown(reviewed in Melnick et al. 1980). The complex etiology of cleft palate makes its studyparticularly difficult. One approach to the investigation of cleft palate is the study ofthe exceptional examples of cleft palate which segregate in families as single genedisorders. Such forms of cleft palate are amenable to analysis by virtue of thechromosomal location of the underlying gene. The process of the isolation of a genebased on the map position of a mutation in that gene is a strategy termed positionalcloning. The isolation and characterization of a locus responsible for a rareMendelian form of cleft palate may provide insight into the pathogenesis of the morecommon multifactorial forms of cleft palate.The main objective of this study was to map the locus responsible for an X-linked form of cleft palate (CPX) associated with ankyloglossia (tongue-tie) whichsegregates in a British Columbia (B.C.) Native family. To better understand the natureof the molecules that may be involved in the etiology of CPX, the normal developmentof the human palate and tongue is reviewed. Some anomalies of palate and tonguedevelopment are subsequently described, followed by a description of the positionalcloning method and its applications to a subset of palatal anomalies.11.2 Development of the palate in humansA complex sequence of events is involved in the development of the humanpalate. Ectomesenchymal tissue arises from the crests of the neural fold by aboutthree weeks postconception (PC). In the cranial region, neural crest cells migrate tothe six pairs of branchial arches and form the major source of skeletal and connectivetissues for the face and anterior neck (Sperber 1989). In the first branchial arch, inassociation with pharyngeal ectoderm, neural crest cells form the bilateral maxillaryprominences. Fusion of the maxillary prominences and the medial nasal and lateralnasal prominences, at about five to six weeks PC, forms the primary palate andestablishes the initial separation between oral and nasal cavities (Diewert 1985;Johnston and Bronsky 1991). The primary palate gives rise to the upper lip, thealveolus, and anterior palatal region around the incisive foramen (Diewert 1985).Extensions arise from the maxillary prominences at about six and a half weeksPC. These extensions become the bilateral palatal shelves which initially growvertically down the sides of the intervening tongue. At approximately eight weeks, theoronasal cavity expands vertically, Meckel's cartilage and the tongue extend forwardbeneath the primary palate, and the palatal shelves elevate rapidly to a horizontalposition above the tongue (Diewert 1983). Concomitant with these events are liftingof the upper face away from the thorax and reflex opening of the mouth (Diewert1985). Fusion of the palatal shelves in the midline and anteriorly with the primarypalate forms the secondary palate, completing the separation of the oral and nasalcavities. At the same time, fusion occurs between the downward growing nasalseptum and the upper surface of the future hard palate region. The regions that willform the soft palate and uvula remain unattached to the nasal septum (Sperber 1989).Palatal shelf midline fusion initiates in the anterior third of the palate andproceeds bidirectionally (Ferguson 1988). Fusion occurs by formation of an epithelialseam in the hard palate region, and merging of the soft palate is subsequent to2closure of the hard palate (Burdi and Faist 1967). The epithelial seam is comprised ofmedial edge epithelial cells of the opposing palatal shelves which adhere to eachother by means of cell surface glycoproteins and desmosomes (Ferguson 1988). Theepithelial cells of this seam either degenerate or transform into mesenchymalconnective tissue cells, resulting in mesenchymal continuity across the hard palate(Ferguson 1988; Fitchett and Hay 1989). Ossification of the anterior two-thirds of thesecondary palate proceeds during the eighth week, giving rise to the hard palate.Myogenic mesenchyme of the first, second, and fourth branchial arches migrates intothe posterior third of the secondary palate, giving rise to the muscle of the soft palate(Sperber 1989).1.3 Development of the tongue in humansFormation of the human tongue initiates in the floor of the pharynx during thefourth week PC. Proliferation of mesenchyme from the first branchial arches results inthe two lateral lingual swellings which subsequently enlarge and fuse with eachother. The lingual swellings also fuse with a median elevation, the tuberculum impar,to form the mucosa of the body (anterior two-thirds) of the tongue (Pansky 1982;Sperber 1989). The mucosa of the posterior third, or root of the tongue, is provided bythe copula (connector) derived from the second, third, and , to a lesser extent, thefourth branchial arches. The anterior body and posterior root of the tongue aredemarcated by the V-shaped sulcus terminalis in the adult (Corliss 1976; Pansky1982).Underneath the fused lingual swellings and tuberculum impar (i.e. anteriorbody of the tongue), there is an accumulation of epithelial cells (Corliss 1976). Theseepithelial cells proliferate into the underlying mesenchyme of the floor of the primitiveoral cavity and anchor the developing tongue to that floor. Beginning in week five,most of these cells degenerate to free the tongue body from the floor of the mouth.3The midline cell mass that remains is the lingual frenulum (Corliss 1976; Sperber1989).The muscles of the tongue, thought to be derived from occipital myotomes,push forward from the floor of the pharynx and underneath the mucosal layer of thetongue. The hypoglossal nerve follows along with this muscle mass, while the nervesupply of the other tongue components is derived from their respective diversebranchial arch origins (Pansky 1982; Sperber 1989).1.4 Congenital malformations of the palate in humansA disruption in any of the processes of normal palate development could leadto palatal clefting defects. For example, failure of the medial prominences to mergeresults in median cleft lip. Failure of the lateral, medial and maxillary prominences tofuse results in unilateral or bilateral cleft lip. Severity varies from a small notch of thelip to a complete cleft of the primary palate (still referred to as "cleft lip") extending intothe floor of the nostril and through the alveolar process. Cleft lip may be associatedwith cleft palate and, in this situation, clefting of the palate occurs secondarily to the lipclefting. Nonsyndromic cleft lip with or without cleft palate has a frequency of 0.1% inEuropean white populations (Farrall and Holder 1992), but is about one half that inblack populations (Corliss 1976) and twice that in the Japanese population (reviewedin Natsume et al. 1989).Clefts of the secondary palate range in severity to include clefts of the uvula,soft palate, and hard palate. Cleft or bifid uvula is a split in the midline of the uvulaand is considered, by some, to be the least severe form of cleft palate (Pansky 1982;Sperber 1989). The prevalence of bifid uvula is approximately 1% to 3% in a mixedracial population (Weatherley-White et al. 1972; Shprintzen et al. 1985; Wharton andMowrer 1992), and as frequent as 10% and 11 % in the British Columbia Native andNavajo Indian populations, respectively (Lowry 1970; Jaffe and De Blanc 1970). Bifid4uvula is often indicative of submucous cleft palate, a condition where the soft palate iscomposed of mucosa with little underlying muscle (Lowry et al. 1973; Shprintzen et al.1985). Other indicators of submucous cleft palate are a notch in the posterior edge ofthe hard palate and muscle separation in the midline of the soft palate (Calnan 1954).The effects of submucous clefts may include nasality, speech difficulties, and minorswallowing problems (Lowry et al. 1973). It is possible that submucous clefts are dueto deficient movement of mesenchyme into the soft palate region (V. Diewert,personal communication). Another suggestion is that some submucous clefts are theresult of cell death in the posterior regions of the maxillary prominence (reported inJohnston and Bronsky 1991).The frequency of submucous cleft palate is approximately 0.08% in a mixedracial population (Weatherley-White et al. 1972). In a survey of 868 B.C. Nativeschool children, Lowry (1979) found the frequency of submucous cleft palate to be0.3%. Many submucous clefts remain undiagnosed even throughout adulthood(McWilliams 1991) and varying criteria for the diagnosis of submucous cleft palate isevident in the literature (reviewed in Peterson-Falzone 1991). It appears that clefts ofthe submucous variety, therefore, are often not included in estimates of isolated cleftpalate frequencies.Clefts of the hard palate, which almost always include soft palate clefts, are themost severe of the secondary palate clefts. The frequency of isolated clefts of thehard palate is approximately 0.07% and varies little between populations (Leck 1984;Lowry et al. 1989). Such clefts of the secondary palate are caused by a failure of thelateral palatal shelves to meet and/or fuse with each other and the nasal septum.Studies in the mouse indicate that the failure is attributable to two major factors: oneis reduced palatal shelf size, and the other is delayed movement of the palatalshelves from a vertical to a horizontal position (Diewert and Pratt 1981). Similar5alterations in development are thought to be responsible for isolated hard palateclefts in humans (Diewert 1986; Johnston and Bronsky 1991).Clefts which include the hard palate region occur twice as frequently in femalesas in males (reviewed in Burdi and Silvey 1969). Observations of human embryoshave indicated that closure of the palatal shelves takes place approximately oneweek later in females than in males, and, therefore, females may be susceptible toany disruptive factors for a longer period of time (Burdi and Silvey 1969). Genderdifference in the incidence of cleft palate is, however, dependent on cleft severity.The data reviewed by Burdi and Silvey (1969) indicated a female to male ratio of only1.2:1 for clefts of the soft palate. Similar observations were made in the Japanesepopulation where the female to male ratio was as high as 3:1 for complete secondarypalate clefts but only 1:1 for bifid uvula (Natsume et al. 1989). In the Danishpopulation, the female to male ratio was 1.1:1 in surgically treated cases of cleftpalate and 0.7:1 in nonoperated cases (Christensen et al. 1992).1.5 Congenital malformations of the tongue in humansAnomalies of tongue development in humans include aglossia, macroglossia,microglossia, cleft tongue, bifid tongue, and ankyloglossia. Aglossia occurs when thetongue fails to develop, and is very rare. Alterations in the normal growth rate of thetongue result in macroglossia, an excessively large tongue, and microglossia, anabnormally small tongue. Incomplete fusion and absence of fusion of the laterallingual swellings lead to cleft tongue and bifid tongue, respectively (Pansky 1982;Sperber 1989).Ankyloglossia is manifested by an abnormally short and often thick lingualfrenulum (Warden 1991). Complete ankyloglossia occurs when there is a total fusionbetween the tongue and the floor of the mouth. Partial ankyloglossia, or tongue-tie,varies in degree from those individuals with a shortened lingual frenulum to those in6which there is a marked fibrosis of both the lingual frenulum and the underlyinggenioglossus muscle (major extrinsic muscle of the tongue). Ankyloglossia limits thenormal range of motion of the tongue and can contribute to speech difficulties (Hortonet al. 1969; Warden 1991).The prevalence of partial ankyloglossia in neonates has been reported as1.7% to 4.4% with no population (black and white) predilection but a significant 3:1(male:female) gender difference (Friend et al. 1990; Harris et al. 1992; Jorgenson etal. 1982). Earlier studies reported partial ankyloglossia frequencies of only 0.04% to0.1% and, in one report, a 1:1 male to female ratio (McEnery and Gaines 1941;reviewed in Warden 1991). These differences may be attributed to the fact that theearlier studies were based on diagnoses in children and adults. At birth, the tongue isoften short with the frenulum extending to the tip. During the early weeks to the firsttwo years of life, the tongue grows longer and the frenulum stretches, often alleviating"neonatal ankyloglossia" (Horton et al. 1969; Sanders 1979; Warden 1991).1.6 Positional cloningThe malformations of the palate and tongue described above are not wellunderstood at the molecular level. One way to identify the molecules involved inthese malformations is to identify families in which palate and/or tongue defectssegregate as single gene disorders. Genes responsible for heritable disorders canbe isolated, without prior knowledge of their function, by the positional cloningmethod. Positional cloning is the process of the isolation of a gene (generally a formof a gene that causes a disease) based on its position in the genome. An outline ofthe steps involved in the positional cloning strategy is as follows: 1) localization of thedisease gene to a chromosomal region, 2) identification of nearest flanking markers todefine a candidate region, 3) physical mapping of the candidate region, 4) isolationand mapping of markers across the candidate region, 5) identification and cloning of7candidate gene sequences, and 6) mutation searching (Wicking and Williamson1991; Collins 1992). This outline is a generalized one, dependent primarily on theresources available for any given mapping project. The first two stages of positionalcloning, detailed below, were employed in this study directed ultimately toward theisolation and characterization of an X-linked cleft palate and ankyloglossia locus(CPX).1.6.1 Regional localizationThe subchromosomal localization of a disease gene may be accomplished bythe identification of cytogenetically detectable chromosomal rearrangements(deletions, duplications, inversions, or translocations) in affected individuals. Forexample, the localization of the Duchenne muscular dystrophy gene (DMD) to theXp21 region was made possible by the detection of X/autosome translocations inaffected females (Verellen et al. 1978; Jacobs et al. 1981; Boyd et al. 1986), and adeletion in an affected male (Francke et al. 1985). The assignment of DMD to Xp21was subsequently confirmed by using polymorphic DNA probes from the Xchromosome short arm in family linkage studies (Goodfellow et al. 1985). In theabsence of any cytogenetically detectable chromosomal abnormalities, regionallocalization of a disease gene by positional cloning depends solely on linkagemapping. Regional localization by linkage mapping may be expedited by testing forlinkage between the disease locus and previously mapped genes with functionsand/or expression patterns that make them reasonable candidates for the disease. Linkage mappingA linkage map is a linear array of markers (genes or anonymous DNAsequences, each with at least two alleles - N.B. it is now convention in humangenetics to use the term alleles to describe alternative forms of a locus, not only of a8gene) where each marker is genetically linked to at least one other marker. Two lociare said to be genetically linked when they exist on the same chromosome and therecombination fraction between them is less than 0.5. The extent of genetic linkage,then, is measured by the recombination fraction, defined as the probability that agamete produced by a parent is a recombinant (Ott 1991). In organisms such asCaenorhabditis elegans and Drosophila melanogaster, it is relatively easy, due toshort generation times and large numbers of progeny, to determine the recombinationfraction between two loci. An unlimited number of planned matings can be set up andthe number of resultant nonparental (i.e. recombinant) progeny is divided by the totalnumber of progeny to obtain the fraction of recombinants. A prerequisite to accuratelinkage mapping in humans is the availability of large multigeneration families inwhich the segregation of loci can be observed. Even then, the number of meioses islimited so mathematical models have been developed to obtain the maximum amountof information from segregation data.The lod score method of linkage analysis, commonly used in human genetics,is based on Morton's (1955) sequential tests procedure. The lod score methoddetermines an odds ratio for observing the segregation pattern of two loci in a familyon the assumption that the two loci are linked at a given recombination fraction (9)value (eg. between 0.0 and 0.5) compared to the assumption that the two loci areunlinked. The likelihood that two loci are linked at a given 9 value [L(9)]is divided bythe likelihood that the two loci are unlinked [L(0.5)] to give a likelihood ratio (L*), orodds for linkage: L*(0) = L(9)/L(0.5). For any given value of 0, the correspondinglikelihood ratio can be calculated and that 0 value associated with the highestlikelihood ratio is deemed the maximum likelihood estimate of the recombinationfraction between the two loci in question. The likelihood ratio is convenientlyexpressed as its logarithm to the base ten, or lod (logarithm of the odds) score(Barnard 1949). Lod (Z) scores at the same 9 values but generated from different9families can thus be simply added since the summation of logarithms is equivalent tothe multiplication of independent likelihood ratios. It is advantageous that likelihoodratios or Z scores from different families can be combined because it is often thesmaller or the less informative two-generation families that are more frequentlyavailable for analysis. Traditionally (Morton 1955), a lod score of 3 (1000:1 odds infavour of linkage) indicates that there is sufficient support for linkage of autosomalloci, a lod score of 2 indicates that there is sufficient support for linkage of loci knownto be X-linked, and a lod score of -2 is considered sufficient for the absence oflinkage. These values were determined by the aim that the false positive probabilitystay below 5 percent (Ott 1991; Risch 1992). The prior probability that two randomlyselected loci are linked (eg. for 8 < 0.3) is approximately 2 percent (Elston and Lange1975). At a lod score of 3, the posterior odds for linkage is the product of the priorodds and the odds from the data, or 1/50 x 1000/1 = 20/1. The posterior probability oflinkage is thus 20/21 or 95.2 percent; the posterior probability of no linkage (i.e. falsepositive probability) is 1/21 or less than 5 percent (Risch 1992). It appears that if twoloci are known to be X-linked, then the prior probability that they will be linked (for 8 <0.3) is at least 20 percent. Accordingly, the posterior odds for linkage would be 20:1at a lod score of just 2 (1/5 x 100/1 = 20/1), and the false positive probability would bebelow 5 percent.The large increase in the number of polymorphic markers on the humangenetic linkage map has made possible the routine use of multipoint linkage analysis.Multipoint linkage analysis is the simultaneous analysis of several linked loci (Ott1991). It accounts for the nonindependence of recombination estimates and theresult is that multipoint analysis is often more efficient than two-point analysis (Lathropet al. 1985; Ott 1991). In human linkage studies, for example, some families orportions of families are often informative for different subsets of marker loci. Multipointanalysis allows the incorporation of information from all available data and can10increase the number of families informative for a given linkage study (Lathrop et al.1985).In the positional cloning approach, multipoint analysis is often used todetermine the support for various positions, x, of a disease locus relative to a map offixed marker loci. A likelihood ratio (R) is calculated as follows: R = L(x, 023, 034,...) /L(., 023, 034 ,...) where x specifies a location of the disease locus in a specified intervalon the fixed map and oo indicates that the disease locus is off the map (Keats et al.1989). Likelihood ratios (R) are often expressed as location scores, S, where S = -2In(R), or as multipoint lod (Z) scores, where multipoint Z logio(R) (Ott 1991).Calculation of lod scores can now be accomplished by a number of computerprograms such as LIPED (Ott 1974) and LINKAGE (Lathrop et al. 1984, 1985). TheLINKAGE package of programs was used for the linkage analyses described in thisthesis. The input into the LINKAGE programs is divided into a "pedfile" containingpedigree and genotype data, and a "datafile" containing loci descriptions, allelefrequencies, penetrance values, recombination rates and gene order. Prior toanalysis, input data is pre-processed by a program called Unknown. Unknown inferspossible genotypes for parents with unspecified genotypes, and reports anyinconsistencies in the data (Linkage Analysis Package User's Guide). The MLINKprogram performs likelihood calculations for specified incremental values of therecombination rate between two loci. Likelihood values (i.e. for 0 = 0.5 and thespecified 0) and lod scores are reported. The !LINK program determines, by iteration,the maximum likelihood estimate of recombination fractions for a designated numberof markers. When only two loci are considered, the maximum lod score is alsodetermined (Lathrop et al. 1985; Linkage Analysis Package User's Guide). Inmultipoint analysis by the LINKMAP program, the location of a test locus is estimatedrelative to a map of known loci whose order and interlocus recombination fractionsare specified and assumed to be fixed. A likelihood ratio is then calculated, for a11number of specified points within each interval on the fixed map, by comparing thelikelihood of each location to the likelihood that the test locus is unlinked to themultilocus map. The resultant likelihood ratios are reported as location scores. Inaddition, the support for one order over another is quantified by odds ratios (Ott,1991).To place loci on a genetic linkage map, it is necessary to transformrecombination fractions (8) to map distances, measured in map units or centimorgans(cM). The genetic map distance (in cM) between two loci is the expected percentageof crossovers occurring in the interval between them. For closely linked loci, 1 cM isequivalent to a recombination fraction of .01 (or 1%). For loci further apart, mapdistance values (cM) are greater than their corresponding recombination values (%)due to the occurrence of double crossovers between two loci which are notobservable as recombinants. The occurrence of double crossovers, however, is alsoproportional to the level of interference between two loci. Interference is thenonindependence of crossovers, or the observation that one crossover "interferes"with the occurrence of another crossover nearby. Some conversions ofrecombination fraction to map distance (mapping functions) incorporate differentlevels of interference along with the probability of double crossover events into theirconversion formulas. Several mapping functions have been developed for differentorganisms (reviewed in Ott 1991). For human data, the Haldane (1919) and Kosambi(1944) mapping functions are most commonly used. DNA polymorphismsFor a DNA locus to be useful in linkage studies, it must be polymorphic. Inhuman genetics, a mating is termed potentially informative for linkage between twoloci when at least one of the two parents is a double heterozygote. The degree ofusefulness of a locus, therefore, depends on the number and frequency of its alleles.12Measures used to express the usefulness, or degree of polymorphism, of a locus arethe heterozygosity value and the PIC (polymorphism information content) value (seesection 3.2.6). Heterozygosity is the probability that a random individual isheterozygous for any two alleles at a given locus (Ott 1991); PIC is the probability thata meiotic event will be informative for a given locus (Botstein et al. 1980).Polymorphic loci may be genes or anonymous (of unknown function) DNA markers.In the case of genes, the variant may be detected as the expression of one particulargenotype or another (i.e. as a phenotype), or it may be detected, like the anonymousDNA markers, at the level of DNA sequence.Variation in DNA sequence is detected in the form of restriction fragment lengthpolymorphisms (RFLPs). RFLPs can be attributed to DNA insertions or deletions.Alternatively, RFLPs can result from the occurrence of a point mutation which eithercreates or destroys the recognition sequence of a restriction endonuclease. Thepresence or absence of the recognition site can be distinguished as a difference inlength of DNA restriction fragments by Southern blot analysis. Single nucleotidemutations are common in the human genome with a disproportionately highoccurrence at cytosine residues in CpG dinucleotides (Bird 1987). The increasedmutation rate is due to an increased methylation of such cytosines (5-methylcytosine)followed by deamination. If not repaired, the result is a cytosine to thymine transition(Bird 1987). Therefore, restriction enzymes that contain CpG in their recognitionsequence (eg. Mspl, Taql) are especially useful for detecting RFLPs. Most marker locidetected as RFLPs have only two alleles (i.e. a restriction site is either present orabsent) and thus are often of limited use in a given linkage study; the maximumprobability that an individual will be heterozygous is 50%.Tandemly repeated sequences frequently result in more informativepolymorphisms because they usually have several alleles corresponding to differentnumbers of repeating units. Minisatellites are short 11 to 60 by core sequences13(Jeffreys et al. 1985) which are present as variable numbers of tandem repeats(VNTRs) (Nakamura et al. 1987). Differences in repeat number can be detected byrestriction enzyme digestion and Southern blot analysis (i.e. as RFLPs) or by thepolymerase chain reaction (PCR). Microsatellites consist of 1 to 6 by motifs iterated10 - 50 times in tandem arrays which occur randomly throughout most eukaryoticgenomes (Hamada et al. 1982; Weber 1990). The type of microsatellite most oftenused for linkage studies in humans is.the CA dinucleotide. The CA dinucleotidemicrosatellite is present approximately 50,000 to 100,000 times in the human genome(Hamada and Kakunaga 1982). Blocks of (dC-dA)n • (dG-dT)n (i.e. CA repeats) oftenvary by multiples of 2 by between individuals and have been demonstrated to behighly polymorphic (Weber and May 1989; Litt and Luty 1989). Microsatellitepolymorphisms are detected by PCR using primers flanking the CA repeats followedby polyacrylamide gel electrophoresis.Anonymous polymorphic loci are assigned names according to guidelines setby members of the DNA Committee of the Human Gene Mapping Workshops(Williamson et al. 1991). Each locus name begins with the letter D (for DNA) followedby the chromosome number (or X, Y), a symbol indicating DNA complexity (eg. S forsingle copy), and a number to individually identify different loci on the samechromosome.1.6.2 Identification of nearest flanking markersOnce a gene has been mapped to a chromosomal region, additionalpolymorphic DNA markers from that region are used in linkage mapping to determinethe nearest markers flanking (i.e. proximal and distal to) the disease gene. This stageidentifies critical recombinant individuals and defines the candidate region.Refinement of the candidate region may be possible by identifying additionalindividuals who are recombinant with respect to markers within the candidate region.14Alternatively, additional polymorphic markers from within the candidate region may beisolated to more precisely localize the crossovers in the recombinant individualsalready identified. The second approach often entails isolating the chromosomalregion of interest and may employ the use of flow sorted human chromosomes,somatic cell hybrids (i.e. human and rodent cells fused in culture) containing singlehuman chromosomes, somatic cell hybrids containing portions of humanchromosomes, or microdissected human metaphase chromosomes. DNA fragmentscan be isolated from these chromosomal sources by constructing recombinant DNAlibraries which are subsequently screened with total human DNA, or by utilizing Aluelement-mediated PCR (Nelson et al. 1989; Brooks-Wilson et al. 1990). Variousapplications of the latter technique have been used to isolate human-specific DNAmarkers from a subchromosomal region of interest (eg. Ledbetter et al. 1990; Bernardet al. 1991).In Alu element-mediated PCR (Brooks-Wilson et al. 1990), a primerhomologous to the consensus sequence of the extreme 3' end of the human Alufamily is used to amplify human DNA found between two adjacent elements inopposite orientations. Alu elements are members of the SINE (short interspersedrepeat) family of mammalian repetitive elements and occur on average every 4 kb orabout 900,000 times in the human genome (Britten et al. 1988). They areapproximately 300 by long and near their middle most contain a single restriction sitefor the enzyme Alul. There is sufficient sequence divergence between human Aluelements and their rodent homologs to allow human specific inter-Alu amplification byPCR from somatic cell human/hamster hybrid DNA.Alu element-mediated PCR has been extended to include the use of primersdirected to the 3' end of L1 elements (Ledbetter et al. 1990; Brooks-Wilson et al.1992). L1 elements are members of the LINE (long interspersed repeat) family ofmammalian repetitive elements, occur about 10,000 times in the human genome,15have an approximate size of 6 kb, contain a Kpnl restriction enzyme site, and displayspecies-specific nucleotide sequence conservation (reviewed in Scott et al. 1987).The rationale behind the inclusion of L1 element homologous primers in what is nowcalled human repeat element-mediated PCR, is that the SINE and LINE families ofrepeat elements are inversely distributed in the human genome (Korenberg andRykowski 1988; Bickmore and Sumner 1989). The LINE family dominates the Gpositive bands of Giemsa stained human chromosomes while the SINE familydominates the G negative bands. Depending on the subchromosomal region ofinterest, primers homologous to human Alu and L1 elements would be expected togenerate different amounts and varying ratios of inter-Alu, inter-L1,and inter-Alu/L1PCR products.Enrichment for a subchromosomal region is possible by restriction enzymedigestion and preparative pulsed field gel electrophoresis of DNA from a somatic cellhybrid containing the chromosome (or chromosome portion) of interest. Pulsed fieldgel electrophoresis (PFGE) utilizes non-uniform, alternately pulsed electrical fields tofacilitate the separation of very large fragments (up to about 6,000 kb) of DNA(Schwartz and Cantor 1984; Q-life systems Inc. 1991). When PFGE is carried out witha low melting temperature agarose (preparative) gel, a DNA fragment of desired sizecan be excised and used as a source for cloning. Typically, the DNA is purified fromthe agarose gel, restriction enzyme digested, and used to construct a recombinantDNA library which is subsequently screened with total human DNA to recover human-specific clones from the starting mixed DNA source (Anand et al. 1988; van de Pol etal. 1990). Alternatively, an isolated and purified DNA fragment can be used as atemplate for Alu-PCR (Burright et al. 1991). Both approaches have demonstratedenrichment for clones from a defined subchromosomal region.Once markers are generated from a subchromosomal region, it is necessary toconfirm their localization in the genome. Physical mapping can be done by in situ16hybridization to chromosomes, or by hybridization to Southern blots or PCRamplification of DNA from somatic cell hybrids containing portions of humanchromosomes or DNA from human cell lines carrying deletions or duplications. On asmaller scale, new markers can be hybridized to Southern blots of human or hybridDNA separated by PFGE to look for the comigration of new markers with any existingmarkers already mapped. Markers that are demonstrated to be from thesubchromosomal region of interest are subsequently screened for polymorphisms(section for use in linkage mapping. If informative in the disease family andrecombinant with respect to the disease gene, a new marker may enable a decreasein the size of the candidate region. Such a new marker would then be identified as anearest flanking marker.Given the current state of the human genetic map, further positional cloningexperiments would likely include cloning the candidate interval for physical mapping,isolation of new markers, and identification of gene sequences (i.e. steps 3 to 5 of thepositional cloning method). The final objective of positional cloning is thedemonstration that an isolated gene is, in fact, the disease gene of interest. Aninterim goal of the Human Genome Project is to construct a genetic linkage map ofpolymorphic markers spaced 2 cM apart (Roberts 1990; Caskey and Rossiter 1992).A map of that resolution would ideally allow the localization of a disease gene to acandidate interval of 2 cM, or within approximately 2 megabases (Mb) of DNA(Renwick 1969). A subsequent plan of the Human Genome Project is to sequencethe human genome; realization of that goal will allow exploitation of the candidategene approach and thus increase the efficiency of the positional cloning process.1.7 Genetic analysis of isolated cleft palate in humansThe first genetic analysis of cleft palate was carried out by Fogh-Anderson(1942) who used family recurrence data to show that isolated nonsyndromic cleft17palate is etiologically distinct from cleft lip and cleft palate. While family recurrencedata suggest the involvement of genetic factors in isolated cleft palate, a simpleMendelian pattern of inheritance is not usually observed. Instead, the multifactorial(i.e. combination of several genetic and environmental factors) threshold theory ispostulated for most cases of isolated cleft palate but neither the genetic nor theenvironmental components are well understood (Fraser 1980a; Nora and Fraser1989). There are some exceptional examples of families, however, in which thesegregation of isolated cleft palate does suggest a major gene determinant (Lowry1970; Rushton 1979; Rollnick and Kaye 1986; Bixler 1987; Hall 1987; Moore et al.1987; Gorski et al. 1992). Such forms of cleft palate can be investigated by thepositional cloning approach, provided that adequate family material is available.Isolation and characterization of a gene involved in one of the rare Mendelian formsof cleft palate may provide insight into the nature of genes and gene productsinvolved in the more frequently occurring, non-Mendelian cleft palate in the generalpopulation.There exist reports of two families exhibiting autosomal dominant inheritance ofisolated cleft palate. In one report, in which the ethnic background of the family is notstated, the palatal defect is described as being a cleft of the soft palate (Jenkins andStady 1980). In a second study, Rollnick and Kaye (1986) describe a German/Dutchfamily with palatal anomalies varying from bifid uvula to a cleft of the posterior half ofthe hard palate. Another autosomal dominant form of cleft palate occurs as part of aclefting syndrome, the Van der Woude syndrome (VWS). In VWS families, an affectedindividual may have only an isolated cleft palate, but his/her family pedigree will alsoinclude paramedian lower-lip pits and/or cleft lip and/or hypodontia (Murray et al.1990). Linkage analyses based on a combined candidate gene and candidatecytogenetic region approach were performed in six three-generation VWS familiesand resulted in the localization of the VWS gene to 1 q32-1 q41 (Murray et al. 1990).18X-linked inheritance of isolated cleft palate has been documented in at least sixfamilies (Lowry 1970; Rushton 1979; Rollnick and Kaye 1986; Bixler 1987; Hall 1987;Moore et al. 1987; Gorski et al. 1992). In one of the first reports of X-linked cleftpalate, Lowry (1970) presented a large Native kindred from British Columbia (B.C.),Canada. The cleft palate in this family was described as being inherited as an X-linked trait and characterized by submucous cleft palate and bifid or absent uvula(Lowry 1970, 1971). This thesis includes a clinical reevaluation of a portion of theB.C. Native family which led to the observation that some of the affected males andcarrier females present with ankyloglossia. X-linked cleft palate has been associatedwith ankyloglossia in at least three other families. The first family is German (Rollnickand Kaye 1986), the second is from Kentucky (Hall 1987), and the third is Icelandic(Moore et al. 1987). The cleft palate in these families is characterized by hard palateclefts (Icelandic family and, possibly, Kentucky family), soft palate clefts, submucousclefts and bifid or absent uvula (Rollnick and Kaye 1986; Hall 1987; Bjornsson et al.1989). Linkage studies in the Icelandic kindred resulted in a provisional assignmentof an X-linked locus responsible for cleft palate (CPX) to the Xq21.3-q22 regionbetween DXYS12 and DXS17 (Moore et al. 1987; Ivens et al. 1988; Mandel et al.1989; Stanier et al. 1991). Linkage studies suggest that the cleft palate locus in theGerman family maps to the same region (reported in Moore et al. 1990).The large size of the B.C. Native family, the availability of a large number ofpolymorphic markers on the X chromosome, and the provisional assignment of theIcelandic CPX, provide an ideal framework for mapping the B.C. cleft palate andankyloglossia locus (B.C. CPX). Directed toward this goal, this thesis has two mainobjectives. The first objective is the regional localization of B.C. CPX on the Xchromosome. This goal was accomplished by performing linkage studies in the with existing DNA polymorphic markers from the Xq21-q22 region (i.e. theIcelandic CPX candidate region). Linkage analyses were later expanded to include19markers from the Xql 3 region. The second main objective of this thesis was to refinethe B.C. CPX candidate region to the maximum resolution possible given theavailable family material. To address this objective, new DNA markers weregenerated from the CPX candidate region to facilitate the precise localization of twocritical crossovers.202. MATERIALS AND METHODS2.1 Pedigree and clinical evaluation of family membersThe B.C. family consists of more than 160 living individuals in 4 generations.Observations described in this thesis and previous records establish classical X-linked inheritance over 5 generations. In addition to the submucous cleft palate andbifid or absent uvula already reported (Lowry 1970), ankyloglossia was observed ina significant number of affected males and carrier females. Individuals wereconsidered to be affected if they exhibited submucous cleft palate and/orankyloglossia. Bifid or absent uvula and/or a notch in the posterior edge of the hardpalate were used as indicators of submucous cleft palate (Lowry et al. 1973;McWilliams 1991). Bifid uvula alone was not considered as adequate for assigningdisease status in light of its high frequency (10%) in the British Columbia Nativepopulation (Lowry 1970).Sixty-three family members were examined clinically (P. Birch, J. Friedman,and P. Goodfellow, personal communication; for details of the family structure, seepedigree in fig.1). Ten of the 16 affected males have both submucous cleft palateand ankyloglossia. Four affected males have submucous cleft palate alone, and oneaffected male has ankyloglossia alone (111-5). We were unable to examinepersonally one of the 15 affected males (11-7), but it was observed and reportedelsewhere that he has a visible incomplete cleft (Lowry 1970). It is unknown whetherhe has ankyloglossia. Females were considered carriers if they are clinicallyaffected, have affected sons, or have an affected father. Six (75%) of the eight carrierfemales are affected. Five of the affected females exhibit ankyloglossia, and onefemale (11-4) has both ankyloglossia and submucous cleft palate. The diseasephenotype of one individual (IV-8) is uncertain at present.2172^-42 34^5^6^7• 1:5 6 o III-o i^ r!I 1 0 i 0 0 ill • 0 i^0-119 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29.16 i ^43^4^5^6 7 8Iv^b IS ci a ill1 2 3 4 5iir6 523 o^ a6 7 8^9 10 11 12 13 14^15 16■•1^2n.)N ■ affected• affectede obligate carrierFigure 1. Pedigree of B.C. CPX kindred members used for linkage analysis. Genotype data were obtained for all living individuals shown.The disease phenotype of individual IV-8 is uncertain at present. A slash (/) through a symbol indicates that the individual is deceased.2.2 Cytogenetic analysisHigh-resolution analysis of X chromosomes was requested for two B.C. familymembers, an affected female (III-7) and her affected son (IV-11). The GTG (Gbanding using Trypsin and Giemsa) technique was performed on prophasechromosome preparations from blood lymphocyte cultures by D. Kalousek(Cytogenetics Laboratory, B.C. Children's Hospital). Interpretation of the cytogeneticanalysis was confirmed by F. Dill.2.3 DNA preparation, Southern transfer, and hybridizationBlood samples (10 to 30 ml) were obtained from 50 family members (fig.1).Genomic DNA was extracted using standard SDS (sodium dodecyl sulphate) /proteinase K digestion and phenol extraction protocols (Sambrook et al. 1989).Aliquots of 5 p.g of genomic DNA were digested with appropriate restriction enzymesfor Southern blot analyses. The DNA was separated by gel electrophoresis on 0.7-1.0% agarose gels and transferred by the method of Southern (1975) toGeneScreenPlus (Du Pont) membrane. Prehybridization, hybridization, and washeswere carried out according to the manufacturer's recommendations. In general,prehybridization was performed for at least four hours at 65 0C in GeneScreenPlushybridization solution (1% SDS, 1 M sodium chloride, 10% dextran sulphate).Hybridization was carried out overnight at 65°C in the same solution with theaddition of denatured salmon sperm DNA (10014/ml hybridization solution).Membranes were washed for 10 to 20 min at RT in 2 X SSC, and 15 to 60 min at650C in 0.2 X SSC and 0.2% SDS (1 X SSC is 150 mM sodium chloride, 15 mM tri-sodium citrate). Autoradiography was at -70 0C for 1-10 days with intensifyingscreens.23Blots were stripped in 0.4N NaOH for 40 min at 43 0C, followed byneutralization in 0.2M Tris, 0.2 X SSC, and 0.2% SDS for 30 min at 43 0C. When notin use, blots were stored in sealed plastic bags at 40G.2.4 DNA polymorphismsThe B.C. family was tested for DNA polymorphisms at twenty-five loci. Table 1lists the twenty-five DNA marker loci, names of probes used, X chromosome regionalassignments, enzymes with which variants are detected, allele sizes andfrequencies, and genotypes for the markers uninformative in the B.C. family. Thelatter is reported for possible future reference. Information in table 1 is from D. Barker(personal communication), P. Bridge (personal communication), Browne et al.(1991), Fain et al. (1991), S. Gorski (sections 3.1.2 and 3.2.5), Lafreniere et al.(1991), Molloy et al. (1992), M-A. O'Reilly (personal communication) and Williamsonet al. (1991).To facilitate DNA manipulations, the inserts for pXG7a (DXS95) and pDP34(DXYS1) were recloned into pBluescriptllKS+ (Stratagene). The pBluescript DNAand clone DNA were digested with the appropriate restriction enzymes. DigestedDNAs in a vector/insert molar ratio of 1:2 were combined in a 10 ill ligation reactionwith 1 X T4 DNA ligase buffer (5 X DNA Ligase Reaction Buffer = 250 mM Tris, 50mM MgCl2, 5 mM ATP, 5 mM DTT, 25% polyethylene glycol-8000; GIBCO BRL) and1 U of T4 DNA ligase (GIBCO BRL). Ligation was carried out overnight at 16 0C. Theligation reaction was diluted 5 X with TE prior to transformation into competent E. colistrain DH5a (prepared by K. Adams). Transformants were selected on ampicillin (50gg/m1) and X-Gal/IPTG containing LB agar plates (Sambrook et al. 1989). Theresultant pBluescript clones were named pXG7a-Bls and pDP34-Bls (table 1).Plasmid DNA was prepared according to a standard minipreparation alkaline lysisprotocol (Sambrook et al. 1989).24Table 1. X chromosome markers used in present study.Locus Probe RegionalAssignmentEnzyme Alleles (kb) Frequencies *Genotype ifuninformativeDXS1 P8 Xq11-q12 Taql 15.0/9.0 .84/.16 1DXS159 cpX289 Xq12 Pstl 5.5/1.6 .67/.33 1DXS106 cpX203 Xq12 BgAI 1.0/5.8 .64/.36 1DXS441 pRX214H1(c) Xq13.2-q13.3 Taql 2.7/2.5 .64/.36 2PGK1 pXPGK-RI0.9 Xq13.3 Psi! 5.2/1.8 .85/.15DXS447 pRX404E2(b) Xq21.1 BgA I 4.4/3.6 .65/.35DXS355 pRX167R3 Xq13-q21.2 7.5/6.5 .83/.17 2DXS349 pRX98H3 Xq13-q21.2 Mspl 14/7.2/7.0/5.2 .10/.67/.13/.10 2 or 3DXS364 pRX272E2 Xq21 Taql 6.5/2.6 .60/.40 1DXS346 pRX86R2(a) Xq21.1 BgA I 2.4/2.1 .38/.62 2DXS72 pX65H7 Xq21.1 Hindi II 7.2/0.7 .45/.55DXS367 pCH1 Xq21.1-q21.2 EcoRI 7.8/5.6 .47/.53 2DXS540 pz11a Xq21.2 EcoRV 7.3/5.3 .53/.47 1pz11a Mspl 7.5/16.5 .69/.31 1pz11c EcoRI 4.5/4.2 .64/.36 1DXS326 pOST38M1(c) Xq21.1-q21.3 Mspl 3.7/2.7,1.0 .62/.38DXS95 pXG7apXG7a-BlsXq21.2-q21.3 Taql 10.5/9.0 .90/.10 1DXS262 pKZ040H2 Xq21.2-q22.1 BgA I 14.5/8.3 .36/.64 1DXYS1 , pDP34pDP34-BlsXq21.31 Taqi 10.6/11.8 .60/.40primers: (CA)n .174/.172/ .54/.42/.04 **DXYS1/4-1A .170DXYS1/4-1 BDXYS12 St25-1 Xq21.31-q21.33 Taql 2.1/1.6 .23/.77DXS3 p19-2 Xq21.3 Taql 3.0,2.0/5.0 .62/.38 1Msp 4.4/12 .75/.25PLP primers: 4 Xq21.33-q22 Ahall .333/.235 + .74/.265 (Acy1) .09825Table 1. contd.Locus Probe RegionalAssignmentEnzyme Alleles (kb)1/2/3/4/5Frequencies Genotype ifuninformativeDXS178 p212/9 Xq21.33-q22 Taql 3.2/1.8 .70/.30DXS94 pXG-12 Xq22 Psi! 6.5/7.2 .52/.48 1DXS101 cX52.5 Xq22 Mspi 7.7/7.5 .35/.65 2DXS17 S21 Xq22 Tacii 2.2/2.0 .65/.35 1DXS456 primers: Xq21-q22 (CA)n .156/.148/ *".20/.20/XG3OBL .146/.154/ .20/.20/.20XG3OBR .144*Genotypes given are those observed in the B.C. CPX kindred. Genotype numberscorrespond to the order of the alleles (kb) given for each locus. Generally, the larger sizedrestriction fragment is designated 1 and the smaller sized restriction fragment is designated 2.**Uninformative but allele size not determined.***Allele population frequencies unknown. Frequencies given are those used for the MLINKprogram.NOTE.-HGM11 reference markers are underlined. Other marker loci are listed primarily inphysical map order (D. Barker, personal communication; S. Gorski, section 3.2.5; Lafreniereet al. 1991; Williamson et al. 1991).26DNA probes for all loci, with the exception of PLP and DXS456 describedbelow, detected RFLPs on Southern blots. Probes were labeled using the BethesdaResearch Laboratories Nick Translation System with [ 32PJdCTP. Alternatively,probes were gel isolated and labeled by the random primer method of Feinberg andVogelstein (1984). Gel isolation was generally carried out in 1 X TBE buffer (0.1 MTris, 0.1 M boric acid, 2 mM EDTA) and 1% agarose succeeded by a secondisolation in 1 X TBE and 0.8% low-melting-point agarose (GIBCO BRL). Any probescontaining repetitive sequences were preassociated (15 ng) with 0.5 mg ofdenatured sheared human placental DNA in 1 ml of GeneScreenPlus hybridizationsolution for 1 hr at 650C. Hybridization was then carried out at 700C.The RFLP at the PLP locus was detected by PCR amplification with 0.4each of primers 4 and 5 (table 1; P. Bridge, personal communication). Reactionconditions were altered to consist of 50-100 ng genomic DNA, 200 I.LM of each dNTP,1.1 U Promega Taq polymerase, and 1 x Promega reaction buffer. Thirty cycles of 2min at 950C, 2 min at 550C and 3 min at 72 0C, followed by 10 min at 72 0C, wereperformed with a Perkin-Elmer Cetus thermal cycler. The PCR product was digestedwith Acyl (Promega) and electrophoresed on a 3% NuSieve (FMC) + 1% agarosegel.The DXS456 locus represents a microsatellite (CA) n polymorphism. Primersand reaction conditions used were as described elsewhere (Luty et al. 1990) with thefollowing modifications: Promega reaction buffer and Taq polymerase (1.1 U) wereused with the addition of 0.25 mM spermidine. Thirty cycles of 2 min at 95 0C, 1 minat 470C and 2 min at 720C, with a final 72°C incubation for 10 min, were performedwith a Perkin-Elmer Cetus thermal cycler.272.5 Linkage analysesLinkage analyses were performed using the MLINK, !LINK, and LINKMAPprograms of the LINKAGE package (v5.10) (Lathrop et al. 1984, 1985). Two-pointlinkage analyses between CPX and marker loci were conducted using MLINK. Lodscore (Z) values were calculated at various recombination fraction (0) values (table3). The maximum two-point Z's (Zmax) and corresponding A's were determinedusing !LINK (table 3). The frequency of the CPX mutant allele was set at 0.0005.The penetrance of CPX was set at 1.0 in males and 0.75 in heterozygous females.Females were scored as affected or unaffected according to clinical diagnosis. Alldeceased individuals were scored as unknown.Multipoint linkage analysis was performed using LINKMAP. Location scores(twice the natural logarithm-of-the-odds ratio) were calculated for all possiblepositions (10 points within each interval) of B.C. CPX relative to a map of six of themost closely linked marker loci deduced from the two-point analyses. The order ofthe marker loci and the distances between them were assumed to be fixed; the locusorder used was PGK1-DXS447-DXS72-DXYS1-DXYS12-DXS3, with respective 8'sof .001, .039, .044, .048 and .027. The O's are approximations based on combineddata from Keats et al. (1989, 1990) and Puck et al. (1991). DXS326 was notincluded in the multipoint analysis because there are no previous reports of its mapdistance relative to the other markers used. Genetic distances were calculated usingHaldane's (1919) mapping function: d = -1/2 In (1-20) where d = genetic distance (inM) and 0 = recombination fraction.2.6 Cosmid library screenThe pWE15 cosmid library is a commercially available library (Stratagene)made with human male lymphocyte DNA in the pWE15 vector. The library wasplated by A. Brooks-Wilson and duplicate filter lifts (with Hybond-N [Amersharn]28membrane) were prepared by D. Smailus. To enable repeated use, the library"plates" were stored at -700C and duplicate filters were stored at 4 0C (A. Brooks-Wilson and D. Smailus). I screened approximately 7.5 X 10 5 cosmids with 200 ng ofgel isolated pXG7a-Bls insert radiolabeled by the random primer method (section2.4). Prehybridization and hybridization were carried out in Denhardt's hybridizationsolution (5 X SSPE [0.9 M NaCI, 50 mM sodium phosphate, 5 mM EDTA], 5 XDenhardt's solution [100 X Denhardt's = 2% bovine serum albumin, 2% Ficoll, 2%polyvinylpyrrolidone], 0.5% SDS) with the addition of 100 p.g/ml denatured salmonsperm DNA during the hybridization period. Duplicate filters were washed for 2 X 10min at RT in 2 X SSC, and 3 X 20 min at 65 0C in 0.2 X SSC and 0.2% SDS.Autoradiography was for 8 days at -70 0C with intensifying screens.2.7 Cell linesCl2D is a fibroblast hybrid cell line with a single human X chromosome in aChinese hamster background (Goss and Harris 1975). Cl2D cells were grown underHAT (sodium hypoxanthine, aminopterin, thymidine) selection in Dulbecco'sModified Eagle Medium (D-MEM) supplemented with 10% fetal calf serum and 1 Xantibiotic-antimycotic (all from GIBCO).The lymphoblast cell line A.S. (used for the Sfil control blot) was kindlyestablished by A. Junker. Cell line A.S. was derived from an affected member (IV-16) of the B.C. CPX kindred. The lymphoblast Xq21 deletion cell lines, XL45, CM,XL62 and NP, were a gift from F. Cremers. The extent of the deletion in each cell line(Cremers et al. 1989) is indicated in fig.8 (section 3.2.3). All lymphoblast cell lineswere expanded in culture in RPM, 1640 medium (GIBCO) supplemented with 10%fetal calf serum and 1 X antibiotic-antimycotic.Preparation of genomic DNA from the lymphoblast cell lines was essentiallythe same as described in section 2.3 except that resuspension and SDS/proteinase29K digestion were initiated from a cell pellet derived from a dense 100 ml culture ofgrowing cells (Sambrook et al. 1989). Genomic DNA from Cl2D (described above),WT49 (human lymphoblast cell line), and CHOK1 (hamster fibroblast cell line;described in Brooks-Wilson et al. 1992) was prepared previously by P. Goodfellow.2.8 Field inversion gel electrophoresis (Sfil test blot)Agarose blocks containing approximately 5 X 10 5 cells (= 3.5^DNA) wereprepared (Herrmann et al. 1987) from the Cl2D and A.S. cell lines. Agarose cellblocks from four male controls were kindly provided by A. Brooks-Wilson. Each blockwas preincubated overnight at 4 0C in 20 U Sfil (BRL) and 1 X M buffer (Maniatis etal. 1982) in a total volume of 200 gl to allow diffusion of the buffer/restriction enzymeinto the block. Digestion was performed for 5 h at 50 0C. DNA samples wereelectrophoresed in 0.8% agarose in 1 X TBE using ROM card 5 (100 - 1100 kbp) ofthe AutoBase electrophoresis system (Q-liFE Systems Inc.). The AutoBase systemuses computer generated electrophoresis parameters optimized for DNA separationin pulsed fields with a one dimensional electrode configuration (i.e. field inversion).Individual memory (ROM) cards store the protocols to facilitate standardization ofexperimental conditions (AutoBase Instruction Manual). Prior to the usual Southernblot procedure (section 2.3), the field inversion agarose gel was soaked in 0.25NHCI for 20 min to facilitate transfer of large DNA fragments (exposure to 0.25N HCIcauses partial depurination of DNA; subsequent exposure of the DNA to strong basecauses hydrolysis of the phosphodiester backbone at the sites of depurination).Hybridization was carried out as described in section Preparative field inversion gel electrophoresis to enrich for the DXS95 regionTwenty-six Cl2D cell blocks were digested with Sfil (section 2.8) andtransferred to a 2 mm wide and 130 mm long slot in a 1% low melting point agarose30gel. An additional three Sfil digested Cl2D blocks and a DNA size marker(Saccharomyces cerevisiae strain YNN295 chromosomes, BioRad) were loaded intoseparate wells. Field inversion gel electrophoresis was performed with ROM card 5of the AutoBase electrophoresis system. After electrophoresis, the lanes containingthe separate Cl2D samples and size marker were cut off and Southern blotted(section 2.3). The remainder of the gel was cut into 2 mm wide horizontal strips(section 3.2.1, fig.6) which were individually stored in 0.5 M EDTA, pH8 at 4 0C.The Southern blot of the 3 Cl2D samples and size marker was probed withpXG7a-Bls insert. The resultant hybridization signal corresponded to several of theagarose gel horizontal strips (fig.6). Segments (3-lane-wide) from each of thesestrips were electrophoresed in 0.8% agarose (ROM card 5), Southern blotted, andprobed with pXG7a-Bls insert. The agarose gel strip with the greatest hybridizationsignal was designated X850 ( X chromosome /IQ kb Sfil fragment).2.10 Human repeat element-mediated PCRHuman repeat element-mediated PCR was performed using the primers andreaction conditions described by Brooks-Wilson et al. (1990, 1992). The primersAl B and Ll S are homologous to the consensus sequence of the extreme 3' end ofthe human Alu and Ll repeat elements, respectively. To facilitate cloning of PCRproducts, the primers have been modified by the addition of restriction enzyme sites.The Al B primer contains a BamHI recognition site and the Ll S primer contains aSall recognition site (Brooks-Wilson et al. 1992). Reaction templates included 100ng of WT49 (total human XX), CHOK1 (total hamster XX) and Cl2D DNAs, and 5111and 15µl aliquots of X850 gel slice (washed 2 X 30 min in TE; TE = 10 mM Tris, 1mM EDTA). For each template, PCR amplification was performed with the Al Bprimer, Ll S primer, and the Al B and Ll S primers together. For the X850 template(15 gl aliquot/high concentration reaction) two rounds of amplification were31performed. Following the first reaction, 1^aliquots (Al B, Ll S and Al B plus Ll S)were used as templates for a second amplification reaction. Reaction products werevisualized on an ethidium bromide (EtBr) stained 1.2% agarose gel.The PCR amplification products from each of the X850 reactions were used asprobes for Southern blot analysis. They were extracted twice with an equal volumeof chloroform/isoamyl alcohol (24:1), precipitated with 1/10 X volume sodium acetate[2.5M] and 2.5 X volume 95% ethanol, and resuspended in 1 X TE. Approximately15 ng of products from each of the three X850 reactions were then radiolabeled bythe random primer method and preassociated as described in section 2.4. Southernblots were also preassociated or "flooded" with denatured salmon sperm DNA (200µg/ml hybridization solution) and denatured sheared human placental DNA (100p.g/ml hybridization solution) during the prehybridization and hybridization periods.Individual Al B/L1 S PCR products were cloned by ligating 25 ng ofBamHI/Sall digested Al B/L1S product pool DNA with 5 to 40 ng of BamHI/SaAdigested pUC18. Ligation, transformation, and selection were carried out asdescribed in section 2.4 except that Library Efficiency DH5aTM competent cells(GIBCO BRL) were used. Positive transformants (white colonies) were tested for thepresence of DNA plasmids with inserts by using small portions of the bacterialcolonies as templates in Alu/L1 PCR. Thirty colonies were tested and approximatelyone-third of each PCR mixture was electrophoresed in a 1.2% agarose gel andtransferred to a Hybond-N (Amersham) nylon membrane. The remaining two-thirdsof select PCR mixtures was gel isolated, radiolabeled (as above), and hybridized(solutions as in section 2.6) to the membrane to determine the number and identitiesof distinct products.323. RESULTS3.1 Regional localization of CPXIn an effort to define the region of the X chromosome in which CPX is located,two approaches were taken. First, X chromosome cytogenetic analysis wasrequested to look for any gross structural abnormalities that might be causallyassociated with CPX. Second, linkage analyses were performed to look forcosegregation of the cleft palate and ankyloglossia phenotype (CPX) withpolymorphic DNA markers already regionally mapped on the X chromosome.3.1.1 CytogeneticsGTG (G banding using Trypsin and Giemsa) analysis was performed onprophase chromosome preparations from blood lymphocyte cultures of an affectedfemale (III-7) and her affected son (IV-11) (D. Kalousek, Cytogenetics Laboratory, B.C.Children's Hospital). High resolution analysis of the X chromosomes, at the 850 bandlevel or greater, revealed no abnormalities (D. Kalousek and F. Dill, personalcommunication). The cytogenetics reports are appended (Appendix #1).3.1.2 GenotypesGenotypes for twenty-five polymorphic markers from the Xq11-q22 region weredetermined for the purpose of linkage analyses. Fifteen markers were uninformative.These markers and their corresponding B.C. kindred genotypes are included in table1. The ten remaining markers were fully or partially informative and their genotypedata for each living B.C. family member is listed in table 2. Individual ID is the familymember's blood sample code number; B.C. pedigree ID number corresponds to thenumbering in the pedigree (fig.1); allele numbers for each locus correspond to those33given (in kb) in table 1 (- = not determined); order of markers is chromosomal order,proximal to distal; the positions of DXS326 and PLP are provisional.All genotypes, except for the markers PLP and DXS456 (described in 2.4),were determined by conventional RFLP analyses. An example of an autoradiographfrom Southern blot analysis with the dimorphic marker DXS447 is shown in figure 2.Deduced haplotypes for a portion of the family, which includes three recombinantindividuals (III-1 0, IV-6 and IV-10), are presented in figure 3.34Table 2. B.C. kindred genotypes for informative markers from Xq13-q22B.C.indiv. pedigreeID^ID PGK1 DXS447 DXS72 DXS326 DXYS1 DXYS12DXS3 PLP DXS178 DXS4561 V-2 2 2 2 1 1 2 2 2 1 12 11-9 22 22 12 12 12 22 12 12 12 123 11-4 2 2 2 1 1 2 2 1 2 14 11-15 12 12 12 12 11 -- 11 11 12 155 11-10 1 1 1 1 1 2 2 1 2 16 11-5 1 1 1 1 2 1 1 1 2 17 11-12 2 2 2 1 1 2 2 1 2 18 V-15 22 12 12 12 12 22 1^1 1^1 22 139 11-17 2 2 2 1 1 2 2 2 1 110 11-2 1 1 1 1 2 1 1 1 2 111 11-14 1 1 1 1 2 1 1 2 1 112 V-3 2 2 2 1 1 2 2 2 1 113 11-16 1 1 1 1 2 1 1 2 1 114 1-2 2 2 1 2 2 2 1 2 1 215 11-11 2 2 2 1 2 2 1 2 116 1-3 12 12 12 11 12 12 12 12 12 1117 11-24 1 1 1 1 2 1 1 2 1 118 11-26 1 1 1 1 2 1 1 2 1 119 11-25 2 2 2 1 1 2 2 1 2 120 V-12 2 2 1 2 2 1 2 1 221 V-11 1 1 1 1 2 1 1 2 1 122 11-8 2 2 2 2 1 2 1 2 1 2 2 2 1 2 1 2 1 2 1 223 11-7 1 2 1 2 1^1 1 2 2 2 1 2 1^1 2 2 1^1 1 224 V-10 1 1 1 1 2 2 1 2 1 225 11-6 2 1 1 2 2 2 2 1 126 V-9 1 1 1 1 1 2 1 127 V-13 12 22 12 12 12 22 1^1 12 1^1 2428 11-1 22 22 12 12 12 22 12 22 1^1 1229 V-4 2 2 2 2 2 1 2 1 230 11-29 1 2 -- 1 2 1 2 -- 1^1 -- 2 2 1^1 1^131 11-28 2 2 1 1 1 2 2 132 V-14 2 2 2 1 1 2 2 1 2 133 11-13 22 22 -- 12 12 22 12 12 12 1234 V-1 22 -- -- 12 -- -- -- 12 1^1 2535 V-5 2 2 2 1 1 2 2 1 136 11-18 1 1 1 1 2 1 1 2 1 137 1-4 1 2 1 2 1 2 1^1 -- 1 2 1 2 1 2 1 2 1^138 1-5 12 12 12 11 12 12 12 12 12 1139 V-6 1 2 1 2 2 2 1 2 1 240 V-7 1 1 1 1 2 1 1 2 1 241 V-8 1 1 1 1 2 1 1 2 1 135Table 2. contd.B.C.indiv. pedigreeID^ID PGK1 DXS447 DXS72 DXS326 DXYS1 DXYS12 DXS3 PLP DXS178 DXS45642 11-3 1 2 1 2 1 1 1 2 2 2 1 2 1 1 2 2 1 1 1 243 11-21 2 2 2 1 1 2 2 1 2 144 11-22 2 2 2 1 1 2 2 1 2 145 11-19 2 2 2 1 1 2 2 1 2 146 11-23 1 1 1 1 2 1 1 2 1 147 11-20 1 1 1 1 2 1 1 2 1 148 11-27 2 2 2 1 1 2 2 1 2 -49 1-7 1 1 1 1 2 1 1 2 1 150 V-16 1 1 1 1NOTE.-Allele numbers correspond to those given in Table 1; - indicates that thegenotype is not determined.Order of markers is chromosomal order; left to right is proximal to distal.3635 36 37 38 39 40 41 42 43 44 45 46 47 48 494.4 kb3.6 kb3.0 kbFigure 2. Autoradiograph of Southern blot illustrating DXS447 genotypes for a portion of theB.C. CPX family. Lane numbers correspond to individual IDs in table 2. The DXS447 probe(pRX404E2) detects a BgiIl constant band of 3.0 kb and alleles of 4.4 kb and 3.6 kb.212222112212122211221222122212212221212111-3IV-6212221212Marker loci:PGK1DXS447DXS72^CPXDXS326DXYS 1DXYS 12DXS3PLPDXS178DXS456Figure 3. Demonstration of DNA marker haplotypes for a portion of the (11-1, 11-3, 111-3, 111-7, III-10, IV-6, and IV-1 0 in fig.1). Phase was inferredso as to minimize the number of crossovers. The recombination eventsdetected in IV-6 and III-10 localize CPX distal to PGK1 and proximal toDXYS1, respectively. The genotypes segregating with CPX are indicated byshading ( 0 ). Genotypes indicated by diagonal lines ( 13 ) are uninformativewith respect toCPX recombination. Other symbols are as in fig.1.383.1.3 Two-point linkage analysisLod score (Z) values for two-point disease-to-marker analysis are summarizedin table 3. The number of informative meioses for each marker is indicated. Norecombination was observed between CPX and DXS447, DXS72, and DXS326(Zmax = 9.38, 7.74, and 2.27, respectively). A single recombination event betweenCPX and the marker PGK1 (Zmax = 7.63 at Ei = 0.03) places CPX distal to PGK1(fig.3). This critical crossover was detected in an unaffected male (IV-6). Sevenrecombination events were observed between CPX and markers distal to DXS326.One crossover was detected between CPX and DXYS1 (Zmax = 5.59 at 13 = 0.04) andplaces CPX proximal to DXYS1 (fig.3). This critical crossover was detected in anaffected male (III-10). A second distal recombination event was observed betweenCPX and DXYS12 (Zmax = 6.46 at 8 = 0.06) (fig.3). This recombination event was alsodetected in an affected male (IV-10), who was uninformative for DXYS1. Four of therecombination events occurred between CPX and both PLP and DXS178, and thefinal event occurred between CPX and DXS456. No multiple crossover events wereobserved in the interval examined. The combined results of the two-point linkageanalyses, then, localize B.C. CPX between PGK1 and DXYS1 in the Xq13.3-q21.31region (fig. 4). The genetic distance between these two loci is approximately 8.8 cM(Keats et al. 1990).390Table 3. Two-point lod scores between CPX and X chromosome marker loci.Z at 0 ofLocus 0.00 .001 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 Zmax 9PGK1 -00 6.56 7.57 7.14 6.54 5.85 5.08 4.24 3.32 2.31 1.20 7.63 0.03DXS447 9.38 9.37 8.66 7.90 7.11 6.28 5.40 4.47 3.48 2.41 1.25 9.38 0.00DXS72 7.74 7.73 7.12 6.48 5.81 5.11 4.39 3.62 2.81 1.94 1.00 7.74 0.00DXS326 2.27 2.27 2.08 1.88 1.67 1.47 1.25 1.03 0.80 0.55 0.29 2.27 0.00DXYS1 -00 4.44 5.56 5.26 4.82 4.29 3.71 3.08 2.40 1.66 0.85 5.59 0.04DXYS12 -00 3.78 6.46 6.29 5.84 5.25 4.58 3.82 2.99 2.08 1.08 6.46 0.06DXS3 -00 4.56 5.71 5.43 4.99 4.47 3.89 3.25 2.54 1.77 0.92 5.73 0.04PLP -co -6.30 0.75 1.70 2.03 2.09 1.98 1.77 1.45 1.05 0.56 2.09 0.19DXS1 78 -00 -6.43 0.64 1.61 1.96 2.04 1.96 1.75 1.44 1.05 0.56 2.04 0.20DXS456 -00 -3.64 -0.42 0.00 0.17 0.23 0.25 0.23 0.19 0.14 0.08 0.25 0.24NOTE.-Linkage data for PGK1, DXS447, DXS72, DXS326, DXYS1, DXYS12, DXS3, PLP, DXS178, and DXS456 werecontributed primarily by 31, 30, 24, 7, 23, 30, 24, 24, 24, and 6 meioses, respectively.Map Position(cM)Marker0.0 PGK1 —DXS447DXS72DXS326DXS958.8 DXYS1 —DXYS1 2Xq22Xq13Xq21— 16.6 DXS3PLPDXS1 78— 26.1 DXS17DXS456B.C. CPX22.322.222.121.321.221.111.411.311.2311.2211.2111.112.21321.121.221.322.232425262728XFigure 4. X chromosome map illustrating the B.C. CPX candidate region(Xq13.3-q21.31). Linkage relationships of the DNA markers informative forlinkage analyses in the B.C. kindred are shown. Two uninformative markers,DXS95 and DXS17, are also indicated. Map positions (in cM) are from HGM10.5(Keats et al. 1990). PGK1 was arbitrarily set at 0.0 cM on the genetic map.413.1.4 Multipoint linkage analysisMultipoint linkage analysis supports the CPX localization suggested by therecombination events described in section 3.1.3. The CPX location score results andfixed loci map are presented in figure 5. The location score, S (S = -21n [likelihoodratio]), is maximum at the position of DXS447 (S = 49.0). This value corresponds to amultipoint Z score (= 1og10 [likelihood ratio] or S / 21n10]) of 10.64. The supportinterval (1 lod difference from the maximum-likelihood estimate of location) for CPX atDXS447 is from 0.01 cM to 7.8 cM and is indicated by the dashed line in figure 5. Theodds against CPX localization are greater than 100:1 (i.e. the value generallyconsidered sufficient for exclusion of an interval) in all intervals tested outside of thesupport interval. These results indicate that CPX can be excluded from regionsproximal to PGK1 and distal to DXYS1.42re a^lc(11^(7) (i)k k X0,.^cZ4002 30500Ei0 2010-10^0^10^20^30Genetic distance (cM)Figure 5. Multipoint linkage analysis of CPX vs. a fixed loci map. The position ofPGK1 was arbitrarily set at 0 cM. The overall maximum location score (= -21n[like-lihood ratio]) occurs at the DXS447 position. The dashed line is drawn 1 lod intervalbelow the maximum-likelihood estimate of CPX location.433.2 Refinement of crossover map positionsTwo-point and multipoint linkage analysis served to assign B.C. CPX to theinterval between PGK1 and DXYS1. Two, and possibly three, recombination eventswithin this region were identified (fig.3). One approach to refining the localization ofB.C. CPX is to precisely localize these crossovers by observing the segregationpattern of CPX with additional markers from within the candidate region. All reportedand several as yet unpublished polymorphic markers were tested for informativenessin the B.C. kindred. In an effort to refine the localization of the disease locus, myefforts were directed to screening for additional polymorphisms and generating newmarkers from the DXS95 region. This region was chosen for the following reasons:i) The proximal crossover (detected in IV-6) had already been localized to the regionbetween PGK1 and DXS447. The distance between these loci is relatively smallaccording to previously published data (Puck et al. 1991).ii) In the current X chromosome map, DXS95 is the closest marker proximal to DXYS1(CPX distal flanking marker) and, if informative, would have the potential to refine thelocation of both distal crossovers.iii) A cloning source that enriches for the DXS95 locus (described below) had alreadybeen prepared in response to an unconfirmed report that DXS95 was the closestproximal flanking marker for a cleft palate and ankyloglossia locus in another kindred.That report was later learned to be erroneous. The cloning source had, however,been prepared and represented a potentially efficient means of generating newmarkers from the DXS95 region.3.2.1 The DXS95 region: screening for additional polymorphisms and generation ofa cloning sourceThe existing DXS95 probe (pXG7a-Bls) was tested for its ability to detectpolymorphic restriction sites in the B.C. family. Southern blot analysis of relevant B.C.44CPX family member DNAs restricted with enzymes (see section 3.2.6 below) failed toreveal additional RFLPs. Probe pXG7a-Bls was then used to screen a total human(male) genomic DNA pWE15 cosmid library in an effort to expand the locus and todevelop a highly informative microsatellite repeat marker system. No DXS95-positivecosmids were obtained. Finally, pXG7a-Bls was used to identify a cloning source(described below) that enriches for DXS95. The cloning source was prepared for thegeneration of region-specific markers which subsequently could be used to screenSouthern blots for RFLPs and ultimately to expand the locus if required.The DXS95 cloning source was prepared as follows. The probe pXG7a-Blswas hybridized to a Southern blot of FIGE size separated Sfil digested DNA from thesomatic cell hybrid Cl2D (human X chromosome only). A single Sfil restrictionfragment of approximately 850 kb was detected, consistent with previously publishedreports (Cremers et al. 1989; Merry et al. 1989). The FIGE blot containing the Sfildigested Cl2D DNA was generated from a portion of a larger preparative gel, theremainder of which had been cut into horizontal strips and stored (section 2.9 andfigure 6). The pXG7a-Bls hybridization signal corresponded to several of theseagarose strips. To increase DXS95 enrichment, segments of each strip wereelectrophoresed again on a field inversion agarose gel. Observation of thephotograph (not shown) of the ethidium bromide stained gel indicated that during thesecond FIGE, the DNA in each agarose strip was further resolved. The agarose gelwas Southern blotted, probed with pXG7a-Bls, and the strip with the greatesthybridization signal was identified. This agarose gel strip was designated X850 (for Xchromosome 850 kb Sfil fragment). The hybridization signal in X850 corresponded toa band size of approximately 850 kb. The gel photograph indicated that DNA in thissize range comprised at least 50% of the DNA in the X850 gel strip while theremainder of the DNA was observed as a smear ranging from 650 kb to 900 kb insize. Despite containing comigrating Sfil fragments, X850 represented the cloning45 Cl2D^Cl2D M• 850 kbFigure 6. Schematic representation of preparative FIGE gel size fractionation (left)and Southern blot analysis of Sfil digested Cl2D (human X only hybrid) DNA (right).The autoradiograph shows the results of the pXG7a-Bls hybridization (to threeseparate lanes containing Cl2D DNA) used to determine which 2 mm widepreparative gel slices contained the DXS95 sequences. M = Bio-Rad DNA sizestandard - Yeast chromosomes, Saccharomyces cerevisiae.46source most enriched for DXS95 and was subsequently used for generating newmarkers by human repeat element-mediated PCR.3.2.2 Human repeat element-mediated PCRAs a means of obtaining human clones from the Cl2D 850 kb Sfil agarose gelfragment (X850) identified above, aliquots of the agarose gel strip were used directlyas templates in human repeat element-mediated PCR. As expected (Brooks-Wilsonet al. 1992), a total human genomic DNA control template (WT49) gave rise to smearson an ethidium bromide stained agarose gel with the Al B primer and the Al B/L1 Sprimers (figure 7). A series of more discrete bands was observed with the LIS primeralone. The Cl2D genomic DNA control template yielded a distinct pattern of discretebands with each of the three primer combinations (fig.7), consistent with a reduction inhuman template complexity and varied distributions of Alu and Ll elements in thehuman genome. The total hamster genomic DNA control template (CHOK1; fig.7) andthe X850 template did not produce any visible PCR products with any of the primers(not shown). When aliquots of each of the three (Al B, Ll S, Al B/L1 S) X850 PCRmixtures were used in second PCR runs, several distinct products and a backgroundsmear were visible in each reaction (fig.7). The products ranged in size fromapproximately 0.2 kb to 1.4 kb. Hamster DNA subjected to a second round ofamplification yielded no visible products.3.2.3 Mapping human repeat element-mediated PCR products by Southern blothybridization: localization to the DXS72 - DXS95 regionThe PCR products present in each second X850 reaction mixture couldoriginate from (i) the human DXS95 850 kb Sfil fragment, (ii) other human Xchromosome 850 kb Sfil fragments, (iii) any hamster chromosome 850 kb Sfilfragments, and/or (iv) human and hamster Sfil fragments of a different size that have47U)u) u) cn =+ Upco^u)^L. -+ 1:1 = c n co^u)^03 + + , _..103 CO .1 up ::-- i up :tZ m m ‹FT ,-<1 <I- .‹..-- 7  CI3 L  L. -up -J^1.-j 6 < a r c°+x.0-) , 6-4- 0^C) Q 6 cS 6 c) , - 1- cS ^8 .7c.-.1-^ u)I-- cq 2 D: 0 E- N 2 D: d I— N op D: ci.2 3: Z5 x c) ^Z5 x 0 c 3: Z3 x c) cIFigure 7. Human repeat element-mediated PCR products. Amplification productswere fractionated on a 1.2% agarose gel and stained with ethidium bromide.Templates and primers are as indicated. The X850 amplification products are thosegenerated after two rounds of PCR amplification. M = HindlIl digested lambda DNAand Haelll digested (DX174 DNA (BRL); WT49 = total human genomic DNA; CHOK1 =total hamster genomic DNA; Cl2D = somatic cell hybrid DNA (human X chromosomeonly); X850 = Cl2D 850 kb Sfil agarose gel strip; n.c. = negative control (no DNA)48comigrated with X850. To help distinguish between these possibilities, pools of theX850 PCR products from each second reaction were hybridized to Southern blotscontaining EcoR1 digested genomic DNA from the following: WT49, a human cell linecontrol, CHOK1, a hamster cell line control, and four male Xq21 deletion human celllines (see section 2.7). The deletion cell line DNAs have in common a deletion of theregion between DXS72 and DXS95 (figure 8). The smallest of the deletions, presentin the XL45 cell line, breaks within the DXS95 locus. PCR products generated fromthe DXS95 region, therefore, are expected to be absent from the DNA of CM, XL62,NP, and may or may not be absent from XL45.The Al B PCR products hybridized to five distinct EcoRI fragments in WT49 andall four male Xq21 deletion cell line DNAs (figure 9A). Faint hybridization to twoEcoRl fragments in hamster DNA was detected. The Ll S PCR products hybridized toat least nine EcoR1 fragments in all the human DNAs (figure 9B). Hybridization tohamster DNA was also significant; seven EcoR1 fragments, most of a different sizethan those observed in human DNA, were detected. The intensity of hybridization ofthe Ll S PCR products was greater in hamster than human DNA. This result suggeststhat the X850 cloning source contains a region of the hamster genome rich in L1-likesequences, or at least enriched for Ll S oligonucleotide sequences. The Al B/L1 SPCR products hybridized to four discrete EcoR1 fragments in both WT49 and the fourXq21 deletion cell lines (figure 9C). A 3 kb band was detected in WT49 DNA but notin any of the four male Xq21 deletion cell line DNAs, suggesting that a portion of theAl B/L1 S PCR products was derived from the DXS72 - DXS95 region. An alternate,but less likely, explanation is that the 3 kb band represents one allele of an EcoRlpolymorphism with the alternate allele represented by one of the larger bandsobserved on the autoradiograph (N.B. WT49 DNA is XX).49Deletion cell lines^Map PositionXL45 CM XL62 NP (cM)Xql 3Marker— 0.0 PGK1 —DXS72 B.C. CPXDXS95— 8.8 DXYS1 —III Xq21DXYS1 216.6 DXS3Xq22 — 26.1 DXS1 7Figure 8. Deletion breakpoints of the four male Xq21 deletion cell lines used inthis study (Cremers et al. 1989). Deleted regions are represented by vertical linesand approximate breakpoints are indicated by horizontal lines. The distalbreakpoint in the XL45 cell line is within the DXS95 locus. The X chromosomemap on the right shows the marker constitution of the deletion cell lines. Geneticdistances (in cM) are from Keats et al. 1990. The B.C. CPX candidate region isindicated.50Figure 9. Mapping human repeat element-mediated PCR products by Southern blothybridization: localization to the DXS72 - DXS95 region. Pools of the X850 PCRproducts generated using the Al B, L1 S, and Al B/L1 S primers (panels A, B, and C,respectively) were used to probe Southern blots containing EcoRI digested genomicDNA from WT49 (human cell line), XL45, CM, XL62, NP (male Xq21 deletion celllines), and CHOK1 (hamster cell line). One fragment detected by the Al B/L1 S PCRproducts appears to be from the DXS72 - DXS95 region as a 3 kb band is detected inWT49 but not in any of the deletion cell lines (panel C). Lane marked M containsHindlll digested lambda DNA and Haelll digested bX174 DNA (BRL) as sizestandards.51CHOK1NPXL62CMXL45WT49COCHOK1NPXL62CMXL45WT49CHOK1NPXL62CMXL45WT49M523.2.4. Cloning and mapping of A1B/L1S PCR productsThe Al B/L1 S products were restriction digested with Sall and BamHI andcloned into pUC18. Seven different clones were identified from among thirty whichwere analyzed by PCR amplification of inserts, gel electrophoresis, and Southern blotanalysis (see section 2.10). Sizes of the amplified inserts ranged from approximately0.15 kb to 0.7 kb, similar to those seen in figure 7. The PCR amplification productsfrom three randomly chosen clones, X850A/L-4 (0.5 kb insert), X850A/L-7 (0.55 kbinsert), and X850A/L-26 (0.4 kb insert), were individually hybridized to the Southernblots containing the Xq21 deletion cell line DNAs. Clone X850A/L-4 detected anEcoR1 fragment of 8 kb in WT49 and all four Xq21 deletion cell lines, but not inCHOK1 (figure 10A). Clone X850A/L-4, thus, is derived from a human X chromosomeSfil fragment located outside of the DXS72-DXS95 region. The second clone,X850A/L-7 also detected an 8 kb EcoR1 fragment, but it was present in WT49 DNAand not present in any of the Xq21 deletion cell line DNAs (figure 10B).Reexamination of the PCR pool preliminary mapping autoradiograph (figure 9C)revealed a band of this size with increased signal intensity in WT49 compared to theXq21 deletion cell lines. The third clone, X850A/L-26 detected the 3 kb EcoR1fragment, present only in WT49, that was observed in the preliminary mapping (figure10C). X850A/L-7 and X850A/L-26, therefore, are both human derived clones whichmap to the DXS72-DXS95 region, and demonstrate the human specific regionalenrichment of the combined preparative FIGE and human repeat element-mediatedPCR methods.To determine if X850A/L-7 and X850A/L-26 originated from an X chromosome850 kb Sfil fragment, each was hybridized against a Southern blot containing Sfildigested genomic DNA from Cl2D and five different male controls (Sfil test blot;hybridization results not shown). For accurate size comparisons, the DXS95 probewas first hybridized to the Sfil test blot so that the resultant autoradiograph could be53Figure 10. Mapping individual Al B/L1 S PCR products by Southern blothybridization. Inserts from three of the Al B/L1 S PCR product clones were hybridizedto the same blots as shown in fig. 9. The resultant autoradiographs are as follows:A. clone X850A/L-4, B. clone X850A/L-7, and C. clone X850A/L-26. Clones X850A/L-7 and X850A/L-26 map to the DXS72 - DXS95 region.54CHOK1NPXL62CMXL45WT49111111•11r91111111110.11111111111rrCHOK1NPXL62CMXL45WT49COCHOK1NPXL62CMXL45WT49M55overlayed with those from X850A/L-7 and X850A/L-26. The DXS95 probe was foundto detect an Sfil fragment of approximately 785 kb in all samples and an additional430 kb fragment in two of the human cell line samples. The size difference (i.e. 785kb instead of 850 kb) can be attributed to differences in agarose, buffer, and DNAconcentration. The altered Sfil restriction pattern in the two individuals is believed torepresent a normal variant. The clone X850A/L-7 hybridized to 320 kb and 650 kbSfil fragments. X850A/L-26 hybridized to a 650 kb fragment and also hybridizedweakly to a 710 kb Sfil fragment. X850A/L-7 was then used to probe the Southernblot used to identify X850. The greatest hybridization signal was detected in twoagarose gel strips containing the majority of DNA in the 650 kb size range, and a fainthybridization signal was detected in the same size range in the X850 gel strip. Itappears that X850A/L-7 was derived from a 650 kb partial Sfil digestion product thatcomigrated in the X850 cloning source. X850A/L-26 is likely derived from a 650 kbSfil fragment or 710 kb partial Sfil digestion fragment that also comigrated in theX850 cloning source. Because X850A/L-7 and X850A/L-26 produced distinctnybridization patterns on the same Sfil test blot, the clones must be derived fromdifferent Sfil fragments.A 625 kb Sfil fragment containing the polymorphic marker DXS540 (table 1) islocated within the DXS72 - DXS95 region (Cremers et al. 1990). To test thepossibility whether either X850A/L-7 or X850A/L-26 could be derived from theDXS540 Sfil fragment, the probe (pZ11c) for DXS540 was hybridized to the Sfil testblot. A single Sfil fragment of 600 kb was detected, indicating that neither X850A/L-7nor X850A/L-26 is located on the same Sfil fragment as DXS540.3.2.5 Physical mapping of DXS326The deletion cell lines used for mapping the Al B/L1 S PCR products alsoproved useful in refining the location of an existing marker, DXS326. The DXS32656locus was partially informative in the B.C. family and nonrecombinant with respect toCPX. DXS326 was known to be located between DXS447 and DXYS1 (D. Barker,personal communication), but had not been mapped relative to the other markers inthis interval. The probe for DXS326 was hybridized to the Southern blot that includesDNA from the male Xq21 deletion cell lines. No bands were detected in any of thefour deletion cell line DNAs, indicating that DXS326 lies between DXS72 andDXS95, or at DXS95. When the DXS326 probe was subsequently hybridized to theSfil test blot, a single band size of 320 kb was observed. Since DXS326 does not lieon the same Sfil fragment as does DXS95, it can be concluded that DXS326 mapsbetween DXS72 and DXS95. This result served to refine the crossover position inindividual IV-10 (fig.3).3.2.6 Screening for polymorphisms with new markers from Xq21To facilitate the genetic mapping of CPX as well as the mapping efforts ofothers working in the Xq21 region, the X850A/L-7 and X850A/L-26 clones werescreened to determine if they detect polymorphisms. To screen for conventionalRFLPs, the clones were hybridized to Southern blots of relevant B.C. family members'DNAs digested with the following restriction enzymes: Bg111, EcoRl, Hindlll, Mspl, Pstl,Pvull, and Taql. Clone X850A/L-26 did not detect any RFLPs. A Taql RFLP wasdetected, however, with X850A/L-7 (K. Adams, personal communication). The RFLPscreening filter indicated that the Taql polymorphism was informative in some of theB.C. family members, including individual III-7. The RFLP, therefore, had the potentialto localize the crossover observed in individual IV-10. Like DXS326, the X850A/L-7clone did not recombine with CPX (fig. 11) and thus localized the crossover distal toX850A/L-7. The relative order of X850A/L-7 and DXS326 is not known.The Taql RFLP detected by X850A/L-7 is dimorphic with allele sizes of 5.0 kb(allele 1) and 4.7 kb (allele 2). DNA samples from unrelated Native individuals were57not available. DNA samples from 12 unrelated Caucasian individuals (8 females, 4males; i.e. 20 different X chromosomes) were digested with Taql, Southern blotted (H.Jenkins, personal communication) and probed with X850A/L-7. Frequencies of 0.30and 0.70 were observed for allele 1 and allele 2, respectively. The heterozygosityvalue (H) for X850A/L-7 is thus 0.42 (H = 1 - I pi2 where pi is the population frequencyof the ith allele). The PIC value (Botstein et al. 1980) for X850A/L-7 is equivalent tothe heterozygosity value since, for X-linked loci, a meiotic event in a heterozygousfemale will always be informative (see section^10^11^125.0 kb4.7 kbFigure 11. X850A/L-7 Taql polymorphism. Segregation of the 5.0 kband 4.7 kb alleles of X850A/L-7 in a portion of the B.C. CPX kindred(III-7, IV-9 to IV-12 in fig.1) is illustrated. No recombination was detectedbetween CPX and X850A/L-7. Symbols are as in fig.1.594. DISCUSSION4.1 Regional localization of CPXThe human X chromosome is estimated to consist of 150 to 200 Mb of DNAcontaining 2500 genes (Keats et al. 1989; Caskey and Rossiter 1992). The gene mapof the X chromosome is relatively dense, with at least 160 genes assigned to it on thebasis of X-linked inheritance (McKusick et al. 1990). DNA marker linkage studieshave been successful in regionally localizing a large number of these X-linked genes(Mandel et al. 1989; Davies et al. 1991). This thesis describes the localization, bylinkage analyses, of an X-linked locus responsible for cleft palate and/orankyloglossia (CPX) in a B.C. Native kindred. Both two-point and multipoint linkageanalyses resulted in localization of B.C. CPX to the Xq13.3-q21.31 region betweenPGK1 and DXYS1. The intervals proximal to PGK1 and distal to DXYS1 wereexcluded as candidate locations for CPX with odds of at least 100:1 by multipointanalyses. Linkage between CPX and markers in the Xq13-q21.3 region is supportedby high lod score values. The designation of PGK1 and DXYS1 as flanking markersis dependent on critical recombination events.The crossovers localizing B.C. CPX proximal to DXYS1 and distal to PGK1were detected in an affected male (I11-10) and an unaffected male (IV-6), respectively.Individual 111-10 is affected, having ankyloglossia and a short bifid uvula typical ofthose seen in other affected members of the B.C. family. Individual IV-6 shows noindications of being affected; there is no notch in the posterior edge of the hard palate,the soft palate appears intact, and the uvula and tongue appear normal. Penetrancein the male members of the B.C. family was assumed to be 1.0 for the linkageanalyses. In no instance was the complete CPX-associated haplotype observed in anunaffected male. Nonpenetrance has been reported in three obligate affected males,one in each of the three previously reported families with X-linked cleft palate and60ankyloglossia (Rollnick and Kaye 1986; Hall 1987; Bjornsson et al. 1989). However,two of these males were unexamined and nonpenetrance was assumed on the basisof past written records or from interviews with family members (Hall 1987; Bjornssonet al. 1989). Experience with the B.C. family indicates that these are unreliablediagnostic methods due to the sometimes subtle nature of the CPX defect. The thirdobligate affected male was examined and no palatal anomaly was detected (Rollnickand Kaye 1986). It was not reported whether he has ankyloglossia. A total of over100 males were clinically examined in the four reported families with cleft palate andankyloglossia (Lowry 1970; Rollnick and Kaye 1986; Hall 1987; Bjornsson et al. 1989;Gorski et al. 1992); there are no other reports of putative nonexpressing obligateaffected males and no reports of unaffected fathers with affected or carrier daughters.Given these data, it appears unlikely that individual IV-6 in the B.C. kindred is anonexpressing carrier of CPX.Other values assumed for the linkage analyses, besides penetrance, were theCPX mutant allele frequency and the DNA marker allele frequencies. The CPXmutant allele frequency was set at 0.0005 for the linkage analyses reported in thisthesis. Lowry (1979) found the frequency of submucous cleft palate in the B.C. Nativepopulation to be 0.003. This value (0.003) represents a possible upper limit for thefrequency of the CPX mutant allele in the B.C. Native population (no ankyloglossiawas noted), and is probably an over-estimate as the affected individuals likely presentwith forms of submucous cleft palate other than CPX. Linkage analyses carried outwith the CPX mutant allele frequency set at 0.003 resulted in lod score values veryclose to those reported in table 3; the greatest difference was a decrease in Zmax of0.02 (data not shown). Similarly, calculations at a range of marker allele frequenciesmade little difference to the lod score results. The marker allele frequencies used forthe analyses in this thesis were determined primarily in Caucasian populations (table2) but could be different in the B.C. Native population (eg. by founder effects and61inbreeding). However, the structure of the B.C. pedigree is such that very fewgenotypes were inferred for the linkage analyses: the sibships are large with manymales; there are few deceased individuals; disease status is known in all cases butone (fig.1).The B.C. CPX localization is inconsistent with that proposed for Icelandic CPX.The initial report of the localization of CPX in the Icelandic kindred included clinicalexaminations of 182 individuals from four generations. Blood was collected from 82individuals, including nine males with cleft palate and ankyloglossia, ten females withankyloglossia alone, one female with cleft palate alone and one male with a high-vaulted palate (Moore et al. 1987). Clefts ranged in severity from hard palate clefts tobifid uvula (Bjornsson et al. 1989). Individuals with any form of cleft and/orankyloglossia were considered affected. Linkage data was contributed primarily byeight meioses and indicated linkage between CPX and DXYS1 with a Zmax of 3.07 at= 0.0 (Moore et al. 1987). Further analyses with additional distal markers identifiedfour recombinant individuals, two each delineated by the probes for DXYS12(proximal flanking marker) and DXS17 (distal flanking marker) (see fig.4 for markerpositions) (Ivens et al. 1988; Stanier et al. 1991). Both recombination events placingIcelandic CPX proximal to DXS17 were detected in affected males and recombinationevents placing CPX distal to DXYS12 were detected in an affected female and acarrier/affected (i.e. not reported) female (Ivens et al. 1988; Stanier et al. 1991). Amultipoint lod score of 4.1 was obtained for Icelandic CPX mapping between DXYS12and DXS17 (Ivens et al. 1988).In the B.C. kindred, two recombination events observed in affected males placeB.C. CPX proximal to DXYS12 (fig.3). One crossover event was between CPX and allinformative markers distal to, and including, DXYS1 (detected in individual III-10discussed above). The second crossover event occurred between CPX and allinformative markers distal to, and including, DXYS12. In the second individual (IV-6210), DXYS1 was uninformative. Individual IV-10 was diagnosed as having asubmucous cleft palate and ankyloglossia. In the B.C. kindred multipoint analysis, theintervals distal to DXYS12 were excluded with odds of greater than 1,000:1.There are several possible explanations for the apparent inconsistency in thelocalization of the CPX gene(s) in the Icelandic and B.C. kindreds. First, CPX in thetwo families may be due to mutations in different genes. Genetic heterogeneityseems unlikely, however, in light of the similar phenotypes (see section 4.4) and theproximity of B.C. CPX and Icelandic CPX on the X chromosome. Second, it ispossible that blood or DNA sample mix-up occurred in either family and was notdetected. A third possibility is that DNA marker genotypes were assigned incorrectly.A fourth explanation for the inconsistent localization of B.C. and Icelandic CPX is thatdisease status was assigned incorrectly in one or more individuals. Both DXYS12recombinant individuals in the B.C. family are affected sons of obligate carrierfemales. The DXYS12 recombinant individuals in the Icelandic kindred are bothdaughters of carrier females. Detailed clinical descriptions and pedigree informationwere not reported for these recombinant Icelandic females. One is known to beaffected, but it is uncertain whether she has ankyloglossia or is the female with cleftpalate only (and what the nature of that cleft is). It is possible that one or more of theaffected DXYS12 recombinant individuals in the B.C. and Icelandic kindreds arephenocopies. For example, an individual in the Icelandic family with a cleft of thehard palate could have the common multifactorial form of cleft palate. Isolatedsubmucous cleft palate and isolated ankyloglossia are also both common birthdefects (Weatherley-White et al. 1972; Lowry 1979; Warden 1991). Since there aretwo DXYS12 recombinants in both the B.C. and Icelandic families, it is likely that theapparent inconsistency in CPX regional assignments is the result of more than one ofthe above proposals.634.2 The Xq13.3-q21.31 region of the human genomeLinkage analyses defined the B.C. CPX candidate region as the intervalbetween the polymorphic loci PGK1 and DXYS1 in the Xq13.3-q21.31 region. It isestimated that PGK1 and DXYS1 are separated by 8.8 Mb of DNA (Keats et al. 1990).The PGK1 marker is from the gene for phosphoglycerate kinase (PGK), an enzymeinvolved in the glycolytic pathway. DXYS1 is an anonymous DNA marker, the XYindicating that it detects homologous sequences on the Y chromosome. The PGK1 -DXYS1 region is also the candidate region, at least in part, for a number of other lociresponsible for human genetic disorders. These loci include the gene(s) for 1) X-linked dominant Charcot-Marie-Tooth (CMT) disease, a progressive motor andsensory neuropathy (lonasescu et al. 1992), 2) a rare X-linked form of dystonia-parkinsonism syndrome (XPD) (Kupke et al. 1992), 3) Aland Island eye disease(AIED) (Alitalo et al. 1991), 4) X-linked severe combined immunodeficiency (SCIDX1)(Puck et al. 1991), 5) Allan-Herndon Syndrome (AHS) type of mental retardation(Schwartz et al. 1990), 6) nonsyndromic deafness (Reardon et al. 1991), and 7)Simpson-Golabi-Behmel (SGB) syndrome, an overgrowth dysplasia syndrome(Hughes-Benzie et al. 1992).The localization of SGB in a Dutch-Canadian family to the Xqcen-X821.3region is of particular interest because submucous cleft palate and ankyloglossia arefound in some SGB pedigrees (Hughes-Benzie et al. 1992). Other features of SGBsyndrome include coarse face, macrosomia, visceromegaly, renal dysplasia, hernias,midline groove of lower lip, grooved tongue, congenital heart defects and skeletalabnormalities. Heterozygous females show partial expression of SGB which, in theDutch-Canadian family, includes ankyloglossia. Also in that family is one SGB malewith a unilateral cleft lip and palate. Because of its complex phenotype, it is possiblethat SGB syndrome is due to a chromosomal rearrangement involving several genes,one of which might be CPX. However, recent linkage analyses including a second64family (British) suggest that the SGB locus may map distal to the B.C. CPX candidateregion (A. Mackenzie, personal communication).The only known gene that maps between PGK1 and DXYS1 and has beencloned (Cremers et al. 1990) is that for choroideremia (CHM) also called tapeto-choroideal dystrophy (TCD). Choroideremia is an eye disorder characterized byprogressive dystrophy of the choroid, retinal pigment epithelium, and retina in affectedmales. The locus for choroideremia was mapped to the Xq13-q22 region by linkageanalyses and the identification, by cytogenetic analyses, of affected individuals withdeletions and translocations in Xq21 (reviewed in Cremers et al. 1989 and Merry etal. 1989). Clinical symptoms associated with the Xq21 deletions can includechoroideremia, mental retardation, and deafness. Female carriers of the deletionsare asymptomatic, with the exception of retinal changes characteristic of thechoroideremia carrier state (Tabor et al. 1983; Hodgson et al. 1987; Nussbaum et al.1987; Cremers et al. 1989; Merry et al. 1989; Wells et al. 1991). The onlydocumented Xq21 deletion individual (N.P.) (Tabor et al. 1983) with cleft palate hasunilateral cleft lip and palate. Cleft palate in association with cleft lip is believed to beetiologically distinct from isolated cleft palate (Fogh-Anderson 1942; Fraser 1980b)and has not been observed in any of the kindreds with cleft palate and ankyloglossia(Lowry 1970; Rollnick and Kaye 1986; Hall 1987; Bjornsson et al. 1989; Gorski et al.1992).There are several possible explanations for the absence of cleft palate inmales with Xq21-region deletions. It is possible that the CPX mutation is not anamorph (loss-of-function mutation). This suggestion is consistent with the apparentdominant nature of the mutation. The cleft palate and/or ankyloglossia phenotypemay be the result of an alteration in the CPX product. Another possibility is that CPXis located outside the region encompassed by the Xq21 deletions. The distalbreakpoint of the largest male viable deletion (individual RvD) is just centromeric to65DXS1 7, and the proximal breakpoint is between DXS72 and PGK1 (Cremers et al.1989). It would follow, then, that CPX is located proximal to DXS72.If CPX is a dominant mutation, the CPX phenotype may be due to an increasein the amount or activity of the CPX gene product (i.e. a hypermorph). An extra copyof CPX, therefore, might give rise to cleft palate and ankyloglossia. There existreports of at least five viable males with duplications which, together, encompass theentire CPX candidate region (Steinbach et al. 1980; Vejerslev et al. 1985; Schwartz etal. 1986; Cremers et al. 1988; Muscatelli et al. 1992) Anomalies associated with theduplications include growth retardation, psychomotor retardation, cryptorchidism, andhypotonia but not cleft palate or ankyloglossia. One male with a duplication includingXq21-q24 has a high arched palate (Schwartz et al. 1986), a feature noted in theIcelandic CPX family (Bjornsson et al. 1989). The absence of CPX, however,suggests that the CPX mutation may not be a hypermorph. It is also possible that theobserved anomalies in Xq21 duplication individuals are not due to gene dosage but,instead, result from position effects or interruption of coding sequences at theduplication breakpoints.4.3 Generation of DNA markers from the CPX candidate regionAt the present time, one of the steps often necessary in the positional cloningapproach is the generation of DNA probes from the candidate region. In this study,efforts were directed to the derivation of polymorphic markers from the DXS95 region.The DXS95 locus is located within the CPX candidate region, approximately one totwo cM proximal to DXYS1 (F. Cremers, personal communication). If a polymorphicmarker identified in the DXS95 region was recombinant with respect to CPX (in III-10or IV-10, fig.3), it would be the nearest distal flanking marker and thus refine thecandidate region. If a DXS95-region polymorphic marker was nonrecombinant withrespect to CPX (in III-10 or IV-10, fig.3), it could further localize two of the previously66identified crossovers and thus more clearly define the genetic limitations of theavailable B.C. family material. A combined preparative FIGE and human repeatelement-mediated PCR strategy was employed with the intention of generating newmarkers from the DXS95 locus. It was known beforehand that the DXS95 probe waslocated on an approximately 850-950 kb Sfil restriction fragment (Merry et al. 1989;Cremers et al. 1989). Anand et al. (1988) used densitometric scans of ethidiumbromide-stained gel photographs to show that the bulk of Sfil DNA fragments from ahuman X chromosome only hybrid is between 100-500 kb in size and a relatively lowamount of DNA is present in the 850 kb size range. This information indicated that ahuman X chromosome only hybrid (Cl2D) 850 kb Sfil agarose gel fragment (X850)isolated previously (by S. Gorski) using FIGE would be enriched for the DXS95 locus.The efficacy of deriving a clone from the DXS95 region is considerably greaterusing a DXS95-containing 850 kb Sfil fragment as a cloning source compared tousing the entire X chromosome as a cloning source. Assuming a total length of thehuman X chromosome of 200 Mb, an 850 kb fragment represents approximately0.43% (1/233) of its total length. The use of a single DXS95-containing 850 kb Sfilfragment rather than the entire X chromosome as a cloning source would thusrepresent an enrichment of 233-fold. However, in the preparative FIGE method,multiple Sfil fragments of 850 kb as well as comigrating Sfil fragments of other sizescould be contained in the cloning source. For this reason, an enrichment of less than233-fold would be expected. Also, it is known that not all regions of the humangenome can be cloned with equal efficiency. In a study by Anand et al. (1988) inwhich an X only hybrid 840 kb Sfil fragment isolated using preparative PFGE wasused as a cloning source, 14% of the human clones isolated were from the Xchromosome 840 kb fragment of interest. This value represents an enrichment of 33-fold in comparison to the frequency of clones expected to be derived from a single Xchromosome 840 kb fragment using the entire X chromosome as a cloning source67(14% [frequency of human clones isolated] x 238 [enrichment expected if 100% ofhuman clones were derived from the X chromosome 840 kb fragment of interest] = 33;assumptions being that the X chromosome is 200 Mb in length and that clones aregenerated randomly along that length). It is reasonable to assume that a similar levelof enrichment for clones from the DXS95 850 kb Sfil fragment is possible.Human repeat element-mediated PCR (Ledbetter et al. 1990; Brooks-Wilson etal. 1992) was chosen as a relatively quick method of obtaining human-specific clonesfrom X850. Human specific amplification of hybrid DNA fractionated by pulsed-fieldgel electrophoresis has been described previously (Burright et al. 1991) but wasmodified to eliminate the time-consuming DNA recovery (electroelute from agarose,dialyze against TE, sequentially extract with phenol, chloroform, and ether, precipitateand resuspend) step. Rather, an aliquot of the X850 gel slice was used directly in oneround of PCR amplification, and then an aliquot of that PCR reaction mixture wasused as a template for a second amplification reaction. It is not known whether thetwo methods result in different distributions or yields of PCR products from a given gelslice.Because the distribution of repeat sequences in the human genome is regiondependent (Bickmore and Sumner 1989), a combination of Alu, L1, and Alu/L1 PCRamplification was used. The cytogenetic location of DXS95 is bands Xq21.2-21.3(Davies et al. 1991); band Xq21.2 is an Alu-rich Giemsa (G) -negative band andXq21.3 is a L1-rich G-positive band.The identification of relevant clones from X850 was determined by Southernblot analysis using DNA from a human cell line (WT49) and four male cell lines withdeletions encompassing DXS95. When an enriched cloning source such as X850 isused as a template in human repeat element-mediated PCR, the pattern and numberof products generated is highly specific - individual products may not be similarlyrepresented in the human repeat element-mediated PCR products derived from total68human DNA (K. Adams, personal communication). For this reason, total EcoRidigested genomic DNA from WT49 and the deletion cell lines was used for theSouthern blot analysis instead of the more standard approach of using repeatelement-mediated PCR products from the mapping resource (Brooks-Wilson et al.1992). The human specificity of the X850 PCR products was tested concomitantlywith map position by including a EcoR1 digested hamster cell line DNA sample nextto the WT49 and Xq21 deletion cell line DNAs on the Southern blot.The size of the DXS72 - DXS95 region is estimated to be approximately 2.6 to3.6 Mb in size (Keats et al. 1989,1990; F. Cremers, personal communication). A 625kb Sfil fragment containing part of the CHM locus and the polymorphic markerDXS540 (table 1) is known to be located within the DXS72 - DXS95 region (Cremerset al. 1990). To test the possibility whether either X850A/L-7 or X850A/L-26 could bederived from that same Sfil fragment, the probe for DXS540 was hybridized to the Sfiltest blot. A single Sfil fragment of 600 kb was detected, indicating that neither of theX850A/L probes are physically linked to DXS540 on the same Sfil fragment. Mymapping data thus suggest that the region deleted in the XL45 cell line (i.eapproximately equivalent to the interval between DXS72 and DXS95) containsdifferent Sfil fragments of 650 kb (partial Sfil digestion fragment containing X850A/L-7), 710 kb (partial Sfil digestion fragment containing X850A/L-26), 600 kb (containingDXS540), 320 kb (containing DXS326; see section 3.2.5), and 785 kb (containingDXS95). The distal breakpoint of the deletion in the XL45 cell line is within theDXS95 locus but it is not known what fraction of the 785 kb fragment is located withinthe deletion region. The summed sizes of all the different Sfil fragments listed aboveis 3065 kb. This value is consistent with the size estimates for the DXS72 - DXS95region.The Alu/L1 PCR clone mapping results suggest that the X850 cloning sourcecontains comigrating 650 kb and, possibly, 710 kb Sfil fragments. PFGE is very69sensitive to DNA concentration and thus DNA overloading may have resulted ininadequate separation of Sfil fragments. This possibility was further substantiatedupon observation of the photograph (not shown) of the ethidium bromide stainedagarose gel which was used in Southern blot analysis to identify which singleagarose gel strip (i.e. X850) would be the optimal DXS95 cloning source (section3.2.1). The photograph indicated that during the second FIGE, the DNA in X850 wasfurther resolved than in the first FIGE (fig. 5). It appeared that at least 50% of the DNAcorresponded to a fragment size of 850 kb but there was a smear of additional DNAfragments ranging from 650 kb to 900 kb in size. When X850A/L-7 was used to probethe Southern blot used to identify X850, the agarose gel strips containing DNApredominantly in the 650 kb size range gave the greatest hybridization signal. Fainthybridization was detected in the 650 kb region of X850. These results suggest thatthe clone X850A/L-7 is derived from a comigrating 650 kb partial Sfil digestionproduct. Similarly, X850A/L-26 is likely derived from a 650 kb Sfil fragment or the 710kb partial Sfil digestion fragment.The Alu/L1 PCR clone mapping results were unexpected in that two of threeclones mapped to the DXS72 - DXS95 region. Furthermore, the two clones mappedto that region despite not being derived from the size selected DXS95 850 kb Sfilfragment. Enrichment for the DXS95 Sfil fragment comparable to that obtained byAnand et al. (1988) for a 840 kb Sfil fragment of interest might still have beenachieved but, unlike Anand's method (preparation of a recombinant library), thederivation of clones from X850 was nonrandom. Rather, clone generation wasdependent on the distribution of repeat sequences contained in X850 DNA,suggesting that relatively Alu-L1 rich 650 kb and, possibly, 710 kb Sfil fragments arecontained therein. It would be inaccurate to make conclusions regarding the possibleenrichment for clones from the DXS95 850 kb Sfil fragment because of the smallsample size of clones that were investigated. However, the apparent failure to70generate Alu/L1 clones from the DXS95 Sfil fragment could be explained by ascarcity of either Alu or L1 elements, or a distribution of Alu and L1 elements notamenable to Alu/L1 PCR. It is possible that either the inter-Alu or inter-L1 PCRproduct pools are more greatly enriched for clones from the DXS95 850 kb Sfilfragment. Despite the apparent failure of clone derivation from DXS95, the Xq21deletion mapping filter allowed relevant clones to be detected. A polymorphic clonederived from anywhere between DXS72 and DXS95 still has the potential to refinetwo of the previously identified distal crossovers (fig. 3).The X850A/L-7 clone was found to detect a Taql RFLP which was partiallyinformative in the B.C. family and nonrecombinant with respect to CPX. This resultserved to localize the crossover observed in individual IV-10 to the region distal toX850A/L-7. The relative order of X850A/L-7 and DXS326, also in the DXS72 -DXS95 region and nonrecombinant with CPX, is not known. The X850A/L-7 result,however, does illustrate how the generation of polymorphic markers from thecandidate region is used to define the limitations of the genetic information obtainablefrom a given individual.Besides the availability of genetic material, another limitation of the positionalcloning approach is illustrated by individual 111-10 and his mother 11-3. The carrierfemale 11-3 is uninformative for both X850A/L-7 and DXS326 and, therefore, thelocation of the crossover detected in 111-10 cannot be further refined. The paucity ofhighly informative markers in a relevant region is thus a second limitation of thepositional cloning approach. It would be useful, at this point, to use X850A/L-7,X850A/L-26, and/or the DXS326 probe(s) to screen a cosmid library in an effort toexpand the loci and to develop a highly informative microsatellite repeat markersystem. Other limitations or factors influencing the success of any positional cloningproject include the amount of penetrance, heterozygote detection (in the case of71recessive alleles), time of onset of clinical manifestations, diagnosis of clinicalmanifestations, and genetic heterogeneity (Nora and Fraser 1989) .4.4 The B.C. CPX phenotypeThe cleft palate and/or ankyloglossia phenotype appeared to be fully penetrantin the male members of the B.C. family. In no instance was the CPX-associatedhaplotype observed in an unaffected male, and in no instance is there an unaffectedmale with an affected or obligate carrier daughter. In carrier females, cleft palateand/or ankyloglossia was 75% penetrant. This value is similar to the 82% penetranceobtained for carrier females in the Icelandic family (Moore et al. 1987).The expressivity of cleft palate and ankyloglossia is highly variable in bothmales and females but, generally, is greater in males. These observations areconsistent with the phenotypic variability described in the three previously reported X-linked cleft palate and ankyloglossia families. Differences in severity of CPXexpression could be due to environmental factors, other genetic factors, and/or, incarrier females, random X inactivation.Bifid or absent uvula was not used as a sufficient indicator of affected status inthe B.C. kindred because of the high frequency (10%) of such anomalies in the B.C.Native population (Lowry 1970). However, bifid or absent uvula in the B.C. kindredwas observed in 63% of affected males and in 38% of carrier females, compared with6% of unaffected males and 0% of noncarrier females. The affected individuals withbifid uvula have relatively short uvulas in comparison to the unaffected individual withbifid uvula. Bifid or absent uvula has also been observed and reported in the threeother families with X-linked cleft palate and ankyloglossia (Rollnick and Kaye 1986;Hall 1987; Bjornsson et al. 1989).As suggested by Hall (1987), ankyloglossia may be a useful diagnostic markerfor X-linked cleft palate. In the B.C. kindred, ankyloglossia was present in 69% of72affected males and in 75% of carrier females. Ankyloglossia is not usually associatedwith multifactorial forms of isolated cleft palate. In two studies describing the clinicalfindings in the oral cavity of neonates, ankyloglossia was found in approximately2.2% of the 2,758 neonates examined (Jorgenson et al. 1982; Friend et al. 1990).There was only one instance of absent uvula reported. However, both study samplesconsisted of neonates from well baby nurseries and it is possible that neonates withsubmucous cleft palate would require special care (Lowry et al. 1973) and thus not beincluded in the study sample.At this stage in the present study, insight into the etiology of CPX remainslimited to observations of the phenotype. Reduction in the amount of soft palatemuscle, separation of soft palate muscle, and/or absent or short bifid uvulas areobserved in some affected CPX individuals. These findings suggest that the amountof mesenchyme which migrates into the posterior third of the secondary palate maybe insufficient and/or that the mesenchyme of the palatal shelves may fail to mergeproperly. Normal mesenchyme migration and merging are thought to include theaction of extracellular matrix (ECM) molecules, growth factors and growth factorreceptors (Ferguson 1987,1988). The locus for Stickler syndrome, a Mendeliansyndrome associated with cleft palate, has been linked to the gene for type II collagen(i.e. an ECM molecule) on chromosome 12 (Francomano et al. 1987). In the Report ofthe Second X Chromosome Workshop (Davies and Craig 1991), is mention of theisolation and mapping to Xq21-q22 of a new gene or pseudogene related to theepidermal growth factor family (TDGF3). If the report is true, the TDGF3 locus couldbe a candidate for CPX. A mesenchymal protein essential for epithelial cellorganization in a variety of epithelial tissues has been identified recently (Hirai et al.1992). A molecule of this type might also be involved in palatal epithelial-mesenchymal interactions.73Cleft palate and ankyloglossia may be caused by a mutation in the same gene,mutations in two independent genes, or one may cause the other (eg. ankyloglossiacauses cleft palate). It has been suggested that one of the events most subject toerror in the development of the human secondary palate is removal of the tongue frombetween the palatal shelves (Johnston and Bronsky 1991). Failure of tongue removalcould prevent or delay palatal shelf elevation and contact (Diewert 1986; Johnstonand Bronsky 1991). It is possible, then, that the ankyloglossia observed in the causes cleft palate by delaying shelf elevation and thus mesenchymal mergingin the posterior soft palate region. Sexual differences in the timing of palatal shelfelevation (one week earlier in males) might account for the increased severity of CPXin males; assuming no sexual differences in the timing of tongue development,females would have a greater time period to alleviate or reduce the severity ofankyloglossia before shelf elevation occurred. Ankyloglossia is not observed in allCPX affected individuals but may have been present earlier in development at thecritical period of shelf elevation. If ankyloglossia is responsible for the cleft palate,then a mutation which prevents degeneration of the cells attaching the tongue to thefloor of the mouth might cause CPX. A mutation which results in absent or reducedamounts of an epidermal growth factor (EGF)-like molecule is a good candidate: EGFin mice has been shown to initiate the breakdown of the fusion of the eyelid in new-born mice (Cohen 1962; D. Juriloff, personal communication). A question thatremains, however, is why ankyloglossia is not more commonly associated with cleftpalate. It is possible that careful examination for microforms of cleft palate has notbeen conducted in most ankyloglossia cases.4.5 ConclusionsThis study was directed toward the identification of the locus responsible forcleft palate and ankyloglossia in a B.C. Native kindred. The main objectives were74twofold. The first objective was to map the locus to a chromosomal region. Thesecond goal was to refine the regional map position to the maximum resolutionpossible given the available B.C. family material.Clinical investigations of the B.C. kindred and update of the B.C. pedigreeconfirmed that the palatal defect is X-linked as originally proposed by B. Lowry(1970). In addition, the clinical reevaluations revealed that some of the affectedmales and carrier females have ankyloglossia. This observation indicated that thephenotype in the B.C. kindred is similar to the X-linked cleft palate and ankyloglossiaphenotype reported in three other families (Rolinick and Kaye 1986; Hall 1987; Mooreet al. 1987). In one of these other families (Icelandic; Moore et al. 1987), the locusresponsible for X-linked cleft palate and ankyloglossia (CPX) was provisionallyassigned to the Xq21.3-q22 region (Moore et al. 1987; Ivens et al. 1988; Mandel et al.1989) and, therefore, provided a starting point for mapping B.C. CPX. Regionallocalization of B.C. CPX and identification of a distal flanking marker wereaccomplished by linkage analysis using polymorphic markers from the Xq21-q22region. Linkage analyses were then extended to include markers from the Xql 3region and resulted in the identification of a proximal flanking marker for B.C. CPX.Results of both two-point and multipoint linkage analysis support the localization ofB.C. CPX to the Xq13.3-q21.31 region between the markers PGK1 and DXYS1. Thedistance between these two B.C. CPX flanking markers is approximately 8.8 cM(Keats et al. 1990). The position of B.C. CPX is further proximal on the X chromosomelong arm than the location proposed for CPX based on recombination eventsdetected in the Icelandic kindred. Further studies are required to determine if theapparent inconsistency is due to genetic heterogeneity or procedural error(s).The second objective of this research was to refine the B.C. CPX candidateregion by localizing more precisely two crossovers identified during the B.C. kindredlinkage analyses. New X chromosome markers were generated and one of these,75X850A/L-7, was found to detect an RFLP and map to the Xq21.1-q21.3 region. Anexisting polymorphic marker, DXS326, was also mapped to the Xq21.1-q21.3 region.The physical mapping of these two polymorphic markers may be useful for otherlinkage studies involving genes localized to the Xq21 region. In the B.C. kindredlinkage studies, both markers were partially informative and nonrecombinant withrespect to CPX. These results did not narrow the CPX candidate region but didfurther refine the map position of one of the crossovers. Refinement of the position ofthe second crossover was limited by marker informativeness. Based on theknowledge that the second crossover occurs within the interval between DXS72 andDXYS1, it can be estimated that the size of the candidate region can potentially bedecreased by a maximum of 4.6 cM. Still, the remaining candidate region would be atleast 4.2 cM; fine mapping of CPX in the B.C. kindred is, therefore, limited by thenumber of meioses available. Further strategies for cloning CPX should thus includean expansion of the genetic resources. Alternately, the Human Genome Project mayallow future studies to exploit the candidate gene approach for identifying CPX. It ispossible that the eventual elucidation of the molecular defect underlying cleft palateand ankyloglossia in the B.C. family will further the understanding of at least one ofthe factors involved in craniofacial development.4.6 Summary1. Sixty-three members of the B.C. Native family including sixteen affected males andeight carrier females were examined clinically. An important observation was made -some of the affected males and carrier females present with ankyloglossia.Penetrance of cleft palate and/or ankyloglossia appears to be complete in males and75% in females. Expressivity of CPX is highly variable in both sexes but, generally, isgreater in males.762. Cytogenetic analyses detected no chromosomal abnormalities associated withCPX .3. Two-point linkage analyses revealed no recombination between CPX and thepolymorphic loci DXS447, DXS72, and DXS326 (Zmax = 9.38, 7.74, and 2.27,respectively). A single recombination event between CPX and PGK1 (Zmax 7.63 at0 = 0.03) places CPX distal to PGK1. A single recombination event between CPX andDXYS1 (Zmax = 5.59 at 0 = 0.04) places CPX proximal to DXYS1.4. Multipoint linkage analysis resulted in a maximum likelihood estimate of locationfor CPX at the position of DXS447 (S = 49.0). The odds against CPX localization aregreater than 100:1 in the regions proximal to PGK1 and distal to DXYS1.5. New markers were generated from a DXS95-enriched cloning source by acombined preparative FIGE and human repeat element-mediated PCR method.Seven PCR products were cloned and three were characterized in more detail.6. Two new markers mapped within the DXS72 - DXS95 region. The markerdesignated X850A/L-7 hybridized to 320 kb and 650 kb Sfil fragments. The cloneX850A/L-26 hybridized to 650 kb and 710 kb Sfil fragments.7. An existing marker, DXS326, was mapped between DXS72 and DXS95. DXS326hybridizes to a 320 kb Sfil fragment. The fine mapping of DXS326 allowedrefinement of the position of the crossover detected in IV-10 to the region distal toDXS326.8. Clone X850A/L-7 was found to detect a Taql RFLP. Clone X850A/L-7 is partiallyinformative in the B.C. family, nonrecombinant with respect to CPX, and thus localizesthe crossover detected in IV-10 distal to X850A/L-7.4.7 Proposals for further research1. Attempt to physically link X850A/L-7 and X850A/L-26 with existing markers by longrange restriction mapping using additional restriction enzymes.772. Clone and map the PCR products from the Al B and LIS X850 product pools.Compare the amount and distribution of repeat element-mediated PCR productsderived from X850, the DXS95 850 kb Sfil fragment, and other Sfil fragments.3. Develop a highly informative marker system for probes X850A/L-7, X850A/L-26and/or DXS326 for the purpose of refining the position of the crossover detected inindividual 111-10.4. Continue to monitor the PGK1 - DXYS1 region in humans for the mapping andcloning of any putative CPX candidate genes. In addition, continue to monitor thePGK1 - DXYS1 homologous region in the mouse for the mapping and cloning of anyputative CPX candidate gene homologues (i.e. comparative mapping).5. Expand the genetic resources.a) In the original report of the B.C. Native family, Lowry (1970) presented a pedigreeindicating that individual 1-2 (of the B.C. CPX pedigree, Fig.1) had a sister who was anobligate CPX carrier. It might be possible to locate descendants of that carrier female.b) A large Manitoba Mennonite family with a similar X-linked cleft palate andankyloglossia phenotype has been identified (C. Greenberg, personalcommunication). This family presents an opportunity to address the question of CPXheterogeneity that was raised by this study.c) Another Native kindred with an apparently similar X-linked cleft palate andankyloglossia phenotype has been identified and DNA linkage studies have beencarried out (C. Johnson, personal communication). It is possible that this Nativekindred may share a common ancestral CPX mutation with the B.C. Native kindred. Ifthat is the case, one strategy for potentially narrowing the CPX candidate regionwould be to identify a common core haplotype for DNA polymorphisms in the region.78REFERENCESAlitalo T, Kruse TA, Forsius H, Eriksson AW, de la Chapelle A (1991) Localization ofthe Aland Island eye disease locus to the pericentromeric region of the Xchromosome by linkage analysis. Am J Hum Genet 48:31-38Anand R, Honeycombe J, Whittaker PA, Elder JK, Southern EM (1988) Clones froman 840-kb fragment containing the 5' region of the DMD locus enriched by pulsedfield gel electrophoresis. Genomics 3:177-186Barnard GA (1949) Statistical inference. J R Statist Soc B11:115-139Bernard LE, Brooks-Wilson AR, Wood S (1991) Isolation of DNA fragments from ahuman chromosomal subregion by Alu PCR differential hybridization. Genomics9:241-246Bickmore WA, Sumner AT (1989) Mammalian chromosome banding - an expressionof genome organization. TIG 5:144-148Bird AP (1987) CpG islands as gene markers in the vertebrate nucleus. TIG 3:342-347Bixler D (1987) Letter to the editor: X-linked cleft palate. Am J Med Genet 28:503-505Bjornsson A, Arnason A, Tippet P (1989) X-linked cleft palate and ankyloglossia in anIcelandic family. Cleft Palate J 26:3-8Botstein D, White RL, Skolnick M, Davis RW (1980) Construction of a genetic linkagemap in man using restriction fragment length polymorphisms. Am J Hum Genet32:314-331Boyd Y, Buckle V, Holt S, Munro E, Hunter D, Craig I (1986) Muscular dystrophy ingirls with X;autosome translocations. J Med Genet 23:484-490Britten RJ, Baron WF, Stout DB, Davidson EH (1988) Sources and evolution ofhuman Alu repeated sequences. Proc Natl Acad Sci USA 85:4770-4774Brooks-Wilson AR, Goodfellow PN, Povey S, Nevanlinna HA, de Jong PJ, GoodfellowPJ (1990) Rapid cloning and characterization of new chromosome 10 DNA markersby Alu element-mediated PCR. Genomics 7:614-620Brooks-Wilson AR, Smailus DE, Weier H-UG, Goodfellow PJ (1992) Human repeatelement-mediated PCR: cloning and mapping of chromosome 10 DNA markers.Genomics 13:409-414Browne DL, Zonana J, Litt M (1991) Dinucleotide repeat polymorphism at theDXYS1X locus. Nucleic Acids Res 19:1721Burdi AR, Faist K (1967) Morphogenesis of the palate in normal human embryos withspecial emphasis on the mechanisms involved. Am J Anat 120:149-15979Burdi AR, Silvey RG (1969) Sexual differences in closure of the human palatalshelves. Cleft Palate J 6:1-7Burright EN, Reyner EL, Gorski JL (1991) Human-specific amplification of radiationhybrid DNA fractionated by pulsed-field gel electrophoresis. Nucleic Acids Res19:401-402Calnan J (1954) Submucous cleft palate. Br J Plast Surg 6:264-272Caskey CT, Rossiter BJF (1992) The Human Genome Project. Purpose and potential.J Pharm Pharmacol 44 (Suppl 1):198-204Christensen K, Holm NV, Olsen J, Kock K, Fogh-Andersen P (1992) Selection bias ingenetic-epidemiological studies of cleft lip and palate. Am J Hum Genet 51:654-659Cohen S (1962) Isolation of a mouse submaxillary gland protein accelerating incisoreruption and eyelid opening in the new-born animal. J of Biol Chem 237:1555-1562Collins FS (1992) Positional cloning: let's not call it reverse anymore. Naturegenetics 1:3-6Corliss CE (1976) Patten's human embryology: elements of clinical development.McGraw-Hill Book company, New YorkCremers FPM, van de Pol TJR, Wieringa B, Hofker MH, Pearson PL, Pfeiffer RA,Mikkelsen M, Tabor A, Ropers HH (1988) Molecular analysis of male-viable deletionsand duplications allows ordering of 52 DNA probes on proximal Xq. Am J Hum Genet43:452-461Cremers FPM, van de Pol DJR, Diergaarde PJ, Wieringa B, Nussbaum RL, SchwartzM, Ropers H-H (1989) Physical fine mapping of the choroideremia locus using Xq21deletions associated with complex syndromes. Genomics 4:41-46Cremers FPM, van de Pal DJR, van Kerkhoff LPM, Wieringa B, Ropers H-H (1990)Cloning of a gene that is rearranged in patients with choroideraemia. Nature347:674-677Davies KE, Mandel J-L, Monaco AP, Nussbaum RL, Willard HF (1991) Report of thecommittee on the genetic constitution of the X chromosome. Cytogenet Cell Genet58:853-966Davies KE, Craig IW (1991) Report of the second X chromosome workshop.Genomics 10:842-846Diewert VM, Pratt RM (1981) Cortisone-induced cleft palate in NJ mice: failure ofpalatal shelf contact. Teratology 24:149-162Diewert VM (1983) A morphometric analysis of craniofacial growth and changes inspatial relations during secondary palatal development in human embryos andfetuses. Am J Anat 167:495-52280Diewert VM (1985) Development of human craniofacial morphology during the lateembryonic and early fetal periods. Am J Orthod 88:64-76Diewert VM (1986) Craniofacial growth during human secondary palate formationand potential relevance of experimental cleft palate observations. J Craniofac GenetDev Biol (Suppl) 2:267-276Elston RC, Lange K (1975) The prior probability of autosomal linkage. Ann HumGenet 38:341-350Fain PR, Luty JA, Guo Z, Nguyen K, Barker DF, Litt M (1991) Localization of the highlypolymorphic microsatellite DXS456 on the genetic linkage map of the human Xchromosome. Genomics 11:1155-1157Farrall M, Holder S (1992) Familial recurrence-pattern analysis of cleft lip with orwithout cleft palate. Am J Hum Genet 50:270-277Feinberg AP, Vogelstein B (1984) A technique for radiolabeling DNA restrictionendonuclease fragments to high specific activity: addendum. Anal Biochem 137:266-267Ferguson MWJ (1988) Palate development. Development 103 (Suppl):41-60Ferguson MWJ (1987) Palate development: mechanisms and malformations. Irish JMed Sci 156:309-315Fitchett JE, Hay ED (1989) Medial edge epithelium transforms to mesenchyme afterembryonic palatal shelves fuse. Dev Biol 131:455-474Fogh-Andersen P (1942) Inheritance of hare lip and cleft palate. Arnold Busck,CopenhagenFrancke U, Ochs HD, de Martinville B, Giacalone J, Lindgren V, Disteche C, PagonRA, Hofker MG, van Ommen G-JB, Pearson PL, Wedgwood RJ (1985) Minor Xp21chromosome deletion in a male associated with expression of Duchenne musculardystrophy, chronic granulomatous disease, retinitis pigmentosa, and McLeodsyndrome. Am J Hum Genet 37:250-267Francomano CA, Liberfarb RM, Tatsuo H, Maumenee IH, Strecter EA, Meyers DA,Pyeritz RE (1987) The Stickler syndrome: evidence for close linkage to the structuralgene for type II collagen. Genomics 1:293-296Fraser FC (1980a) The William Allan memorial award address: evolution of apalatable multifactorial threshold model. Am J Hum Genet 32:796-813Fraser FC (1980b) The genetics of cleft lip and palate: yet another look. In: Pratt RM,Christiansen KL (eds) Current research trends in prenatal craniofacial development,Elsevier, New York, pp357-366Friend GW, Harris EF, Mincer HH, Fong TL, Carruth KR (1990) Oral anomalies in theneonate, by race and gender, in an urban setting. Ped Dent 12:157-16181Goodfellow PN, Davies KE, Ropers HH (1985) Report of the committee on the geneticconstitution of the X and Y chromosomes. Cytogenet Cell Genet 40:296-352Gorski SM, Adams KJ, Birch PH, Friedman JM, Goodfellow PJ (1992) The generesponsible for X-linked cleft palate (CPX) in a British Columbia Native kindred islocalized between PGK1 and DXYS1. Am J Hum Genet 50:1129-1136Goss SJ, Harris H (1975) New method for mapping genes in human chromosomes.Nature 255:680-684Haldane JBS (1919) The combination of linkage values, and the calculation ofdistances between the loci of linked factors. J Genet 8:299-309Hall BD (1987) Letter to the editor: A further X-linked isolated nonsyndromic cleftpalate family with a nonexpressing obligate affected male. Am J Med Genet 26:239-240Hamada H, Kakunaga T (1982) Potential Z-DNa forming sequences are highlydispersed in the human genome. Nature 298:396-398Hamada H, Petrino MG, Kakunaga T (1982) A novel repeated element with Z-DNA-forming potential is widely found in evolutionarily diverse eukaryotic genomes. ProcNatl Acad Sci USA 79:6465-6469Harris EF, Friend GW, Tolley EA (1992) Enhanced prevalence of ankyloglossia withmaternal cocaine use. Cleft Palate-Craniofac J 29:72-76Herrmann BG, Barlow DP, Lehrach H (1987) A large inverted duplication allowshomologous recombination between chromosomes heterozygous for the proximal tcomplex inversion. Cell 48:813-825Hirai Y, Takebe K, Takashina M, Kobayashi S, Takeichi M (1992) Epimorphin: amesenchymal protein essential for epithelial morphogenesis. Cell 69:471-481Hodgson SV, Robertson ME, Fear CN, Goodship J, Malcolm S, Jay B, Bobrow M,Pembrey ME (1987) Prenatal diagnosis of X-linked choroideremia with mentalretardation, associated with a cytologically detectable X-chromosome deletion. HumGenet 75:286-290Horton CE, Crawford HH, Adamson JE, Ashbell TS (1969) Tongue-tie. Cleft Palate J6:8-23Hughes-Benzie R, Hunter AGW, Allanson JE, Mackenzie AE (1992) Simpson-Golabi-Behmel syndrome associated with renal dysplasia and embryonal tumor: localizationof the gene to Xqcen-q21. Am J Med Genet 43:428-435lonasescu VV, Trofatter J, Haines JL, lonasescu R, Searby C (1992) Mapping of thegene for X-linked dominant Charcot-Marie-Tooth neuropathy. Neurology 42:903-90882Ivens A, Moore GE, Chambers J, Arnason A, Jensson 0, Bjornsson A, R. Williamson(1988) X-linked cleft palate: the gene is localized between polymorphic DNA markersDXYS12 and DXS17. Hum Genet 78:356-358Jacobs PA, Hunt PA, Mayer M, Bart RD (1981) Duchenne muscular dystrophy (DMD)in a female with an X/autosome translocation: further evidence that the DMD locus isat Xp21. Am J Hum Genet 33:513-518Jaffe BF, De Blanc GB (1970) Cleft palate, cleft lip, and cleft uvula in Navajo Indians:incidence and otorhinolaryngologic problems. Cleft palate J 7:300-305Jeffreys AJ, Wilson V, Thein SL (1985) Hypervariable 'minisatellite' regions in humanDNA. Nature 314:67-73Jenkins M, Stady C (1980) Dominant inheritance of cleft of the soft palate. Hum Genet53:341-342Johnston MC, Bronsky PT (1991) Animal models for human craniofacialmalformations J Craniofac Genet Dev Biol 11:277-291Jorgenson RJ, Shapiro SD, Salinas CF, Levin LS (1982) Intraoral findings andanomalies in neonates. Pediatrics 69:577-582Keats B, Ott J, Conneally M (1989) Report of the committee on linkage and geneorder. Cytogenet Cell Genet 51:459-502Keats BJB, Sherman SL, Ott J (1990) Report of the committee on linkage and geneorder. Cytogenet Cell Genet 55:387-394Korenberg JR, Rykowski MC (1988) Human genome organization: Alu, Lines, andthe molecular structure of metaphase chromosome bands. Cell 53:391-400Kosambi DD (1944) The estimation of map distances from recombination values. AnnEugen 12:172-175Kupke KG, Graeber MB, Muller U (1992) Dystonia-Parkinsonism syndrome (XDP)locus: flanking markers in Xq12-q21.1. Am J Hum Genet 50:808-815Lafreniére RG, Brown CJ, Powers VE, Carrel L, Davies KE, Barker DF, Willard HF(1991) Physical mapping of 60 DNA markers in the p21.1-q21.3 region of the humanX chromosome. Genomics 11:352-363Lathrop GM, Lalouel JM, Julier C, Ott J (1984) Strategies for multilocus linkageanalysis in humans. Proc Natl Acad Sci USA 81:3443-3446Lathrop GM, Lalouel JM, Julier C, Ott J (1985) Multilocus linkage analysis in humans:detection of linkage and estimation of recombination. Am J Hum Genet 37:482-498Leck I (1984) The geographical distribution of neural tube defects and oral clefts. BritMed Bull 40:390-39583Ledbetter SA, Nelson DL, Warren ST, Ledbetter DH (1990) Rapid isolation of DNAprobes within specific chromosome regions by interspersed repetitive sequencepolymerase chain reaction. Genomics 6:475-481Litt M, Luty JA (1989) A hypervariable microsatellite revealed by in vitro amplificationof a dinucleotide repeat within the cardiac muscle actin gene. Am J Hum Genet44:397-401Lowry (1970) Sex-linked cleft palate in a British Columbia Indian Family. Pediatrics46(1):123-128Lowry RB (1971) X-linked cleft palate. Birth Defects: Original Article Series VII:76-79Lowry RB, Courtemanche AD, MacDonald C (1973) Submucous cleft palate and thegeneral practitioner. CMA Journal 109:995-997Lowry RB (1979) Genetic studies of cleft lip and cleft palate in the North AmericanIndians of British Columbia. Thesis, Doctor of Medicine, Queen's University, BelfastLowry RB, Thunem NY, Uh SH (1989) Birth prevalence of cleft lip and palate in BritishColumbia between 1952 and 1986: stability of rates. Can Med Assoc J 140:1167-70Luty JA, Guo Z, Willard HF, Ledbetter DH, Ledbetter S, Litt M (1990) Five polymorphicmicrosatellite VNTRs on the human X chromosome. Am J Hum Genet 46:776-783Mandel JL, Willard HF, Nussbaum RL, Romeo G, Puck JM, Davies KE (1989) Reportof the committee on the genetic constitution of the X chromosome. Cytogenet CellGenet 51:384-437Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning: a laboratory manual.Cold Spring Harbor Laboratory Press, USAMcEnery E, Gaines F (1941) Tongue-tie in infants and children. J Pediatr 18:252-255McKusick VA, Francomano CA, Antonarakis SE (1990) Mendelian inheritance inman. The Johns Hopkins University Press, BaltimoreMcWilliams BJ (1991) Submucous clefts of the palate: how likely are they to besymptomatic? Cleft Palate-Craniofacial J 28:247-249Melnick M, Bixler D, Shields ED (1980) Etiology of cleft lip and cleft palate. Alan RLiss, Inc, New YorkMerry DE, Lesko JG, Sosnoski DM, Lewis RA, Lubinsky M, Trask B, van den Engh G,Collins FS, Nussbaum RL (1989) Choroideremia and deafness with stapes fixation: acontiguous gene deletion syndrome in Xq21. Am J Hum Genet 45:530-540Molloy CM, van de Pol TJR, Brohet RM, Ropers H-H, Cremers FPM (1992) ThreeRFLPs for pZ11 (DXS540) in the choroideremia gene at Xq21.2. Nucleic Acids Res20:143484Moore GE, Ivens A, Chambers J, Farrall M, Williamson R, Page DC, Bjornsson A,Arnason A, Jensson 0 (1987) Linkage of an X-chromosome cleft palate gene. Nature326:91-92Moore GE, Ivens A, Newton R, Balacs MA, Henderson DJ, Jensson 0 (1990) Xchromosome genes involved in the regulation of facial clefting and spina bifida. CleftPalate J 27:131-135Morton NE (1955) Sequential tests for the detection of linkage. Am J Hum Genet7:277-318Murray JC, Nishimura DY, Buetow KH, Ardinger HH, Spence MA, Sparkes RS, FalkRE, Falk PM, Gardner RJM, Harkness EM, Glinski LP, Pauli RM, Nakamura Y, GreenPP, Schinzel A (1990) Linkage of an autosomal dominant clefting syndrome (Van derWoude) to loci on chromosome 1q. Am J Hum Genet 46:486-491Muscatelli F, Verna JM, Philip N, Moncla A, Mattei MG, Mattei JF, Fontes M (1992)Physical mapping of an Xq-proximal interstitial duplication in a male. Hum Genet88:691-694Nakamura Y, Leppert M, O'Connell P, Wolff R, Holm T, Culver M, Martin C, FujimotoE, Hoff M, Kumlin E, White R (1987) Variable number of tandem repeat (VNTR)markers for human gene mapping. Science 235:1616-1622Natsume N, Miura S, Kawai T (1989) Effect of sexual differences on the developmentof cleft palate in the human. Plast Reconstr Surg 84:854-855Nelson DL, Ledbetter SA, Corbo L, Victoria MF, Ramirez-Solis R, Webster TD,Ledbetter DH, Caskey CT (1989) Alu polymerase chain reaction: a method for rapidisolation of human-specific sequences from complex DNA sources. Proc Natl AcadSci USA 86:6686-6690Nora JJ, Fraser FC (1989) Medical genetics: principles and practice, 3rd ed. Lea andFebiger, PhiladelphiaNussbaum RL, Lesko JG, Alan Lewis R, Ledbetter SA, Ledbetter DH (1987) Isolationof anonymous DNA sequences from within a submicroscopic X chromosomaldeletion in a patient with choroideremia, deafness, and mental retardation. Proc NatlAcad Sci USA 84:6521-6525Ott J (1974) Estimation of the recombination fraction in human pedigrees: Efficientcomputation of the likelihood for human linkage studies. Am J Hum Genet 26:588-597Ott J (1991) Analysis of human genetic linkage, rev ed. The Johns Hopkins UniversityPress, BaltimorePansky B (1982) Review of medical embryology. Macmillan Publishing Co Inc, NewYorkPeterson-Falzone SJ (1991) Commentary. Cleft Palate-Craniofacial J 28:250-25185Puck JM, Bailey LC, Conley ME (1991) Update on linkage of X-linked severecombined immunodeficiency (SCIDX1) to loci in Xq13. Human Gene Mapping 11.Cytogenet Cell Genet 58:2082-2083Reardon W, Middleton-Price HR, Sandkuijl L, Phelps P, Bellman S, Luxon L,Pembrey ME, Malcolm S (1991) A multipedigree linkage study of X-linked deafness:linkage to Xq13-q21 and evidence for genetic heterogeneity. Genomics 11:885-894Renwick JH (1969) Progress in mapping human autosomes. Br Med Bull 25:65-73Risch N (1992) Genetic linkage: interpreting lod scores. Science 255:803-804Roberts L (1990) Whatever happened to the genetic map? Science 247:281-282Rollnick BR, Kaye CI (1986) Mendelian inheritance of isolated nonsyndromic cleftpalate. Am J Med Genet 24:465-473Rushton AR (1979) Sex-linked inheritance of cleft palate. Hum Genet 48:179-181Sanders B (1979) Pediatric oral and maxillofacial surgery. The C.V. Mosby Company,St. LouisSambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual.Cold Spring Harbor Laboratory Press, USASanders, B. (1979) Pediatric oral and maxillofacial surgery. The C.V. MosbyCompany, St. LouisSchwartz CE, Ulmer J, Brown A, Pancoast I, Goodman HO, Stevenson RE (1990)Allan-Herndon Syndrome. II. Linkage to DNA markers in Xq21. Am J Hum Genet47:454-458Schwartz DC, Cantor CR (1984) Separation of yeast chromosome-sized DNAs bypulsed field gradient gel electrophoresis. Cell 37:67-75Schwartz S, Schwartz MF, Panny SR, Peterson CJ, Waters E, Cohen MM (1986)Inherited X-chromosome inverted tandem duplication in a male traced to agrandparental mitotic error. Am J Hum Genet 38:741-750Scott AF, Schmeckpeper BJ, Abdelrazik M, Theisen Comey C, O'Hara B, PrattRossiter J, Cooley T, Heath P, Smith KD, Margolet L (1987) Origin of the human L1elements: proposed progenitor genes deduced from a consensus DNA sequence.Genomics 1:113-125Shprintzen RJ, Schwartz RH, Daniller A, Hoch L (1985) Morphologic significance ofbifid uvula. Pediatrics 75:553-561Sperber GH (1989) Craniofacial Embryology, 4th ed. Butterworths, Cambridge86Stanier PM, Newton R, Ivens A, Arnason A, Bjornsson A, Moore GE (1991) Newpolymorphisms in the region of X-linked cleft palate and ankyloglossia. Human GeneMapping 11. Cytogenet Cell Genet 58:2086Steinbach P, Horstmann W, Scholz W (1980) Tandem duplication dup(X)(q13q22) ina male proband inherited from the mother showing mosaicism of X-inactivation. HumGenet 54:309-313Tabor A, Andersen 0, Lundsteen C, Niebuhr E, Sardemann H (1983) Interstitialdeletion in the "critical region" of the long arm of the X chromosome in a mentallyretarded boy and his normal mother. Hum Genet 64:196-199van de Pol TJR, Cremers FPM, Brohet RM, Wieringa B, Ropers H-H (1990) Derivationof clones from the choroideremia locus by preparative field inversion gelelectrophoresis. Nucleic Acids Res 18:725-731Vejerslev LO, Rix M, Jespersen B (1985) Inherited tandem duplication dup(X) (g131-g212) in a male proband. Clin Genet 27:276-281Verellen C, Markovic V, DeMeyer R, Freund M, Laterre C, Worton R (1978)Expression of an X-linked recessive disease in a female due to non-randominactivation of the X chromosome. Am J Hum Genet 30:97AWarden PJ (1991) Ankyloglossia: a review of the literature. Gen Dent 39:252-253Weatherley-White R, Sakura C, Brenner L, Stewart J, Ott J (1972) Submucous cleftpalate. Plast Reconstr Surg 49:297-304Weber JL, May PE (1989) Abundant class of human DNA polymorphisms which canbe typed using the polymerase chain reaction. Am J Hum Genet 44:388-396Weber JL (1990) Informativeness of human (dC-dA) n •(dG-dT) n polymorphisms.Genomics 7:524-530Wells S, Mould S, Robins D, Robinson D, Jacobs P (1991) Molecular and cytogeneticanalysis of a familial microdeletion of Xq. J Med Genet 28:163-166Wharton P, Mowrer DE (1992) Prevalence of cleft uvula among school children inkindergarten through grade five. Cleft Palate-Craniofac J 29:10-12Wicking C, Williamson B (1991) From linked marker to gene. TIG 7:288-293Williamson R, Bowcock A, Kidd K, Pearson P, Schmidtke J, Ceverha P, ChipperfieldM, et al (1991) Report of the DNA committee and catalogues of cloned and mappedgenes, markers formatted for PCR and DNA polymorphisms. Cytogenet Cell Genet58:1190-183287APPENDIX 1. Cytogenetics reports. High-resolution X chromosome analysis wasperformed for two members of the B.C. CPXfamily: an affected female (III-7 in fig.1; p.89) and her affected son (IV-11 in fig.1; p.90) (F. Dill and D. Kalousek, personalcommunication).88.^Rijn: 22 APR .1991 •• 1149 B.C.'S CHILDREN'S HOSPITALCYTOGENETICS REPORT• •• '.^•^PAGE 1-FOR-.: ..22 APR 1991(U:7502758)9 MAY 1949 F •OUT (17 DEC)DR. JAN M. FRIEDMAN90-C602389g COLL: 17/12/90 LOG: 17/12/90 . 1037PHYSICIAN: JAN M.'..fRIEDHRCOMMENT: cc: Dr.P.GoodfellowMG,^ .CLINICAL HISTORY: Son (IMMINNPMMISMONNUMNEP: 90-C62288X -linked cleft palate.1111971X===== ==112=2:21=2=3=73*=======i 2C.^•--•H19K•resolutioniioto9enetic analysis of the X chroaosOies •is•noraal(^asp band lavell.,;- "=> SIGNED OUT BY: D.E. MCFADDENC TYX123 t100150^.C 10X220 IL1100 PP3160 03174 1P3186 Z41072 3. .^,•-is•SURGICAL PROCEDURE(S): ANOMEA^.TISSUE(S): CHROMOSOME PAIR 23, BLOOD.LYMPNOCYTE,.•AA FINAL CYTOGENETIC DIAGNOSIS A*. CYTOGENETICS LABORATORYBritisheolumbirs.m.^•\ Children's HospitalDEPARTMENT OF PATHOLOGY89 IiarinsAcantirs..Children's l-lospitai• • • DEPARTMENT OF PATHOLOGY4 :-CYTOGENETICS 'LABORATORYI.ti'2 Rti: 22 APR ggi"1149 /.O.C.'S CHILDRENS HOSPITAL :^-PAO114'CYTOGENETICS REPORT^•^114;.22 APR •OMMOOMPAlliiiIMMID_Ai:0034913)31 MAY 1374 ft OUT (1.7 DEC)DR. JAN M. FRIEDMAN90-C402299R COLL: 17/12/90 LOW 17/12/90 1034 .-^PHYSICIAN: SAN. M. FRIEDMAit^•^• •COmENT: cc: .;CH-411,Dr.P.Good allow •• .^. .^•CLINICAL - HISTORY: X-linkeil alert palate. ••••.Mather tionoimi 90-0O229 .9 . .-•SURGICAL PROCEDVRE(S): ANOKA• TIMMS): CHRONOSOMEPAIR 23, eLOOD LYMPHOCYTEAA FINAL CYTOGENETIC DIAGNOSIS AA. 3tamaisessmaga=usa aaaaa a=s0a=massumis^. ' 4•• • .^ „resolutien cytagenatie analysis of the X chrostosaienormal CS) band level),IiHED OUT BY: D.E.. EEDDEd.^.^ .22/4191::-.•*.•••^:• .•^• .••^•^•^• -• •1., -TYX123.M00150 :3^•^•^• . • z^- • "?'„,•10X220 IL1100 PP3160 PP3174 PP3188 ZO1073 3 .•90


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items