of Cystic Fibrosis? Analyses of the CFTR Gene in 67 Patients

Transcription

1 m. J. Hum. enet. 56: , 1995 Is ongenital Bilateral bsence of Vas Deferens a Primary Form of ystic Fibrosis? nalyses of the FR ene in 67 Patients B. Mercier, ' * * Lissens,2 Silber,3 Novelli,4 Bonduelle,2 udrezet,' and. Ferec' 'entre de Biogenetique.D..S., Brest; 'University Hospital VUB, Brussels; 'St Luke's Hospital, St. Louis; and 4Human enetics, atholic University of Rome, Rome Summary ongenital bilateral absence of the vas deferens (BVD) is an important cause of sterility in men. lthough the genetic basis of this condition is still unclear, it has been shown recently that some of these patients carry mutations in their cystic fibrosis transmembrane conductance regulator (FR) genes. o extend this observation, we have analyzed the entire coding sequence of the FR gene in a cohort of 67 men with BVD, who are otherwise healthy. We have identified four novel missense mutations (800, 149R, R258, and E193K). We have shown that 42% of subjects were carriers of one F allele and that 24% are compound heterozygous for FR alleles. hus, we have been unable to identify 76% of these patients as carrying two FR mutations. Furthermore, we have described the segregation of FR haplotypes in the family of one BVD male; in this family are two male siblings, with identical FR loci but displaying different phenotypes, one of them being fertile and the other sterile. he data presented in this family, indicating a discordance between the BVD phenotype and a marked carrier (F508) chromosome, support the involvement of another gene(s), in the etiology of BVD. Introduction ongenital bilateral absence of the vas deferens (BVD) accounts for -6% of male sterility (Holsclaw et al. 1971). he molecular basis of this condition is unknown, but recently it has been shown that a large proportion of these sterile men carry a mutation in at least one of their cystic fibrosis (F) genes (Dumur et al. 1990). he F transmembrane conductance regulator (FR) gene, which was *Both authors contributed equally to this work. Received June 20, 1994; accepted for publication September 29, ddress for correspondence and reprints: Dr. laude Ferec, entre de Biogenetique,.D..S., BP 459, Brest edex, France by he merican Society of Human enetics. ll rights reserved /95/ $02.00 cloned in 1989, has been analyzed extensively in the last four years (Kerem et al. 1989; Riordan et al. 1989; Rommens et al. 1989; sui 1992). he F phenotype, which is characterized by lung disease and/or pancreatic exocrine insufficiency, can also include failure of reproductive function both in males and in females. More than 95% of males are sterile because of altered Wolffian ducts with atrophy of the vas deferens, tail and body of the epididymis, and seminal vesicles (Boat et al. 1989). he relationship between male sterility in BVD and F patients was studied recently at the molecular level by Rigot and coworkers. hey showed that in a cohort of asymptomatic azoospermic healthy males with BVD, there was a high frequency of F508 carriers (Rigot et al. 1991). hey also reported the frequent occurrence of a second mutation in their group of patients, the mutation R117H (ervais et al. 1993). hese observations have prompted the hypothesis that BVD men could represent an incomplete form of F (nguiano et al. 1992). o address the relationship between BVD and F, we have conducted a large collaborative study including >60 BVD patients. We have analyzed the entire coding sequence of the FR gene for each individual, by a clamp denaturing gradient-gel electophoresis (DE) assay and DN sequencing. Where possible, we have performed segregation studies in families with at least one affected sibling. he results we obtained confirmed the implication of FR abnormalities in the pathogenesis of BVD, but our results also suggest that the genetic basis of BVD could be more heterogeneous than previously suggested and that at least a second gene could be involved in the etiology of the disease. Patients, Material, and Methods Patients Sixty-seven unrelated azoospermic patients with congenital absence of the vas deferens were included in this study. heir diagnoses were initially suggested by the clinical observation of impalpable vasa and were subsequently confirmed by seminal analyses. hese patients have no history of respiratory disease, nor have they any signs or 272

2 Mercier et al.: Bilateral bsence of Vas Deferens in F symptoms of F. We have included only patients with bilateral absence of the vas deferens. hose patients with unilateral absence of the vas deferens were excluded, as were patients with renal malformations and other urogenital abnormalities. For each of these patients, a spermogram was performed, and all were azoospermic. Ninety percent of these patients had been included in a microsurgical epididymis sperm aspiration (MES) program, and the diagnosis of BVD was confirmed at the time of the surgical sperm aspiration (Silber et al. 1990). While none of the patients had signs or symptoms of F, a systematic measurement of steatorrhea was not performed for these patients. here was no difference in the phenotype presented by our three groups of patients, defined at the genotype level, particularly between the group of patients with two F mutations and the group in which we were unable to find any mutation. However, we have not performed objective measurements of pancreatic status and pulmonary function. Sweat chloride tests were not routinely performed. Family R For individual Ill, the diagnosis of BVD was made by clinical examination showing impalpable vasa and was confirmed at the time of surgical sperm aspiration, when this patient was enrolled in a MES program for in vitro fertilization using epididymal sperm. He has undergone three trials in this program. One of the assays was successful until the pregnancy was terminated spontaneously at 6 wk. his patient has no history of upper and/or lower respiratory symptoms from childhood. Individual II1 was azoospermic and the FSH values were normal. he sweat chloride test performed recently showed a value of 48 mmol, which is within the normal range. s individual I14 is fertile and his paternity has been confirmed, no sperm count was carried out. DN nalysis DE.-Oligonucleotides for amplification of exons and the amplification reactions for the DE analyses have been described elsewhere (Lerman et al. 1987; Myers et al. 1987). mplification products were run on 6.5% polyacrylamide gels containing a linear denaturing gradient from 20% to 70%, where 100% denaturant equals 7 M urea with 40% formamide (v/v). Electrophoresis was performed at 75 V for 7-9 h, with respect to the melting map of the amplified sequence. PR fragments displaying an altered behavior in the gel, indicating the presence of a mutation or a polymorphism in this particular part of the gene, were then sequenced. he ystic Fibrosis enetic nalysis onsortium rules for coding each nucleotide change, as well as the primers, have been described elsewhere (sui 1992; udrezet et al. 1993). Microsatellite Markers.-Oligonucleotides and amplification conditions for the analysis of the microsatellite 273 markers have been described elsewhere (Morral et al. 1992). mplification for the IVS 17b and IVS 17b repeat markers were carried out simultaneously. Primers were end labeled with 4 polynucleotide kinase in the presence of [32PlyP. mplification was carried out for 25 cycles, using an annealing temperature of 50. single PR was performed for IVS8, with an annealing temperature of 57 for 35 cycles. wo microliters of PR product were mixed with 2 pl of formamide, were loaded on a polyacrylamide gel, and electrophoresed for a period of 4.5 h at 2,000 V. Dried gels were then exposed to X-ray film overnight. Results Survey We studied a group of BVD patients from Europe and the United States (23 patients from France, 7 from Italy, 14 from Belgium, and 23 from the United States). ltogether, we looked at 67 infertile men, none of whom had any of the classic symptoms of F. We scanned the entire coding sequence of the FR gene of these patients, using DE, followed by DN sequencing. his was made possible by the use of 37 primer pairs, which permitted the analysis of all 27 exons and their intron/exon boundaries (udrezet et al. 1993). We identified 28 F508, 5 W1282X, 1 542X, 1 R553X, and 2 N1303K mutations. hese changes that were identified in F patients are believed to be disease causing, this being determined either directly, by a study of their effect on the activity of the protein, or indirectly, by their exclusion from non-f chromosomes. In addition, we identified the following missense mutations: four R668, one 800, one (628R + S1235R, borne on the same chromosome), one (R74W + D1270N, borne on the same chromosome), six R117H, one F1052V, one R117, one S1235R, one 149R, one R258, two R347H, one R1066H, one R75L, and one E193K. ogether with the four new alleles that we identified, these missense mutations correspond to a total of 23 different alleles (see table 1). Moreover, we systematically looked for the transmission of FR mutations in pedigrees of BVD patients in which there were at least two affected males or an affected male with one or several nonsterile brothers. In order to define the phase of these chromosomes precisely, we employed polymorphic dinucleotide repeat markers, situated in introns 8 (IVS 8 ) and 17b (IVS 17b and IVS 17b ) (Morral and Estivill 1992; Morral et al. 1992). hese dinucleotide markers allowed us to construct haplotypes for the individual chromosomes bearing each mutation. mong the families studied, we collected and analyzed data from three large families. In one family, there were two BVD brothers who shared the same genotype. In the second family, the microsatellites were noninformative. he third family is described below.

3 274 able I enotypes of the 67 BVD Patients: Details of F Mutations Identified in Patients with Isolated BVD No. of Patients enotypea 1 F508 + (628R + S1235R) 1 F508 + (R74W + D1270N) 2 F508 + R668 4 F508 + R117H 1 F508 + R258 1 F508 + R75L 1 E193K + N1303K 1 R347H + R1066H 1 R117 + W1282X 1 R553X + R R + R668 1 R117H+R117H 18 F508/unidentified 4 W1282X/unidentified 1 542X/unidentified 1 N1303K/unidentified 1 S1235R/unidentified 1 R347H/unidentified 1 800/unidentified 1 F1052V/unidentified 23 unidentified/unidentified a In parentheses are the two mutations located on the same haplotype. Identification ofnovel Mutations mong the missense mutations, four have not been observed previously: (i) DE analysis of exon 13 showed one DN sample displaying a modified pattern of migration, which was investigated by direct sequencing of amplified DN. nucleotide change, -o. at position 2531, was identified. he result of this change is the substitution of an alanine by a glycine at codon 800. his converts a nonpolar-lateral chain () to a polar-lateral chain (). his amino acid modification occurs in the R domain, which regulates the opening of the chloride channel. his patient carries no mutation on the other allele, according to our analysis (fig. 1). (ii) he second change, situated in exon 4, is -o. at position 577 and corresponds to the substitution of a glycine for an arginine (149R). It has been reported elsewhere that such an amino acid change could be considered as mild, as for 628R in exon 13 (Fanen et al. 1992) and 1061R in exon 17b (Mercier et al. 1993b) (fig. 1). (iii) he third novel mutation we found is R258, and the nucleotide change is - at position 904 in exon 6b. It corresponds to the substitution of an arginine for a glycine (R258) his mutation is situated in a region that encodes part of the transmembrane region of the protein, and the mutation results in a change of polarity at codon 258. (iv) he fourth novel mutation we have observed occurs at position 709 and corresponds to a --. change m. J. Hum. enet. 56: , 1995 that results in a glutamic acid codon instead of a lysine codon (E193K). his change modifies the polarity at this position (fig. 1). he three latter mutations correspond to changes in amino acids that have been conserved throughout evolution, as observed in dogfish, Xenopus, bovine, mouse, and human species (Yorifuji et al. 1991; ucker et al. 1992). his conservation suggests the functional importance of these amino acids. For all four of these new mutations, a segregation analysis was performed in each family, allowing us to show that 149R, R258, and E193K were carried by a particular allele and that these mutations were not de novo mutations. Moreover, we screened 250 F patients and never observed such mutations. We also screened 300 normal chromosomes, so the possibility that these variations are polymorphisms is low. Of the 67 patients analyzed, 16 were compound heterozygotes for two mutations in the coding portion of FR, including those changes we identified in this study (24%). wenty-eight subjects (42%) were carriers of one FR allele. For 23 individuals (34%), we were unable to find any mutations in the coding reading or splice junctions of FR, even with our comprehensive screening procedure (table 1). wo patients were carriers of two missense mutations on the same allele (268R + S1235 R and R74W + D1270N) (table 1). he association of these two missense mutations on the same haplotype is interesting, as both S1235R and R74W have been independently reported elsewhere to be FR alleles (laustres et al. 1993; uppens et al. 1993). We also had previously reported that haplotype R74W + D1270N was present in a healthy 23 le b 12 t 113 t 11,4 t *, a Figure I FR locus for brothers II-1 and II-4. Microsatellite markers (IVS 8, IVS 17b, and IVS 17b ) permitted the construction of four different phases (16/28/13, 16/31/13, 16/43/13, and 23/30/13), and their segregations show that the two brothers, 11-1 and 11-4, are identical at the FR locus. he number "16" corresponds to the number of repeats at the IVS 8 locus, "28" to the number of repeats at the 17b locus, and "13" to the IVS 17b, which in this case is not polymorphic. he F508 is situated in exon 10, between IVS 8 and IVS 17b, as indicated. Both brothers II-1 and II-4 are identical at the FR locus. Individual 11-1 is a BVD patient, and his brother (11-4) is fertile (his paternity was confirmed). he half-shading designates an individual heterozygote for F508.

4 Mercier et al.: Bilateral bsence of Vas Deferens in F - E193K a -> SOO - a Figure 2 utoradiographs showing nucleotide sequence of portions of exons 5, 13, and 4 of FR and demonstrating the mutations E193K, 800, and 149R, respectively. he sequences were generated by asymmetric PR and direct sequencing. Each of the mutations occurs on only one allele of each patient. mother of an affected child (Verlingue et al. 1993), her genotype being (R74W + D1270N) o. If we consider that BVD may be a mild form of F, these observations suggest that a second mutation in a FR allele may result in a reversion or partial reversion of phenotype. onsistent with this possibility are two independent observations obtained first by mutation screening and second from studies in a yeast model system. Dork and coworkers (1991) identified an individual bearing both the F508 and R553Q mutations on the same allele. his F patient presented with a relatively mild phenotype. In the experimental model system involving a yeast SE6 and human FR chimeric protein, a reversion of mating deficiency was obtained by introduction of a second alteration on the mutated allele (eem et al. 1993). In both the mildly affected patient and the model system, the changes are predicted to be relatively close within a single working domain, the first nucleotide binding fold, of FR. It is reasonable to suspect that the first change disrupts structure, while the second change, at least partially, relieves this disruption. It is not clear how the double changes we identified may lead to a reversion, as the changes correspond to alterations that are predicted to be in the amino-terminal region and the second nucleotide binding fold. However, these alterations would both be predicted to be on the cytoplasmic surface of FR. Discordant Segregation ofa ommon F Mutation and BVD In family R (fig. 2), subject Ill was a 45-year-old BVD male. Our analysis (table 1) showed him to be a carrier of the mutation F508 that was inherited from his mother, I- 2. His full brother, 11-4, also had inherited the F508 allele. he three polymorphic markers used to build the phase show complete informativity, the four haplotypes being clearly identified. F508 was carried on the haplotype, 23/30/13, the most common haplotype associated with F508 (Morral et al. 1994). We have established, without doubt, that the two brothers, 11-1 and 11-4, are identical at the FR genotype level (23/30/13 and 16/ 28/13). Subject 114 is fertile, having fathered two children. We performed a paternity test using polymorphic markers situated on different chromosomes (HL DR DQ, polymorphism of intron 40 of factor von Willebrand [aucher et al. 1992], and microsatellite IVS 17b of the FR gene). Paternity could not be excluded for these two children, with a probability of paternity of >99.9%. paternity test was also confirmed for I2 and his two children, II1 and II4. hese results show clearly that these two brothers are carrying two identical FR haplotypes and suggest that another locus could account for the phenotype observed for subject II,. Discussion 275 It has been postulated since 1971 (Holsclaw et al. 1971) that BVD could represent an incomplete form of F, as the majority of men suffering from F are sterile because they lack the vas deferens. s soon as the FR gene was cloned, and in the following years, when the majority of F mutations were identified, this hypothesis was reviewed by Rigot et al. (1991). hey showed that 8/19 azoospermic men were carriers of F508. his observed frequency was significantly higher than the expected frequency of 1/25. his high frequency has since been observed in additional studies (Patrizio et al. 1993). Further, a second mutation, R117H, known to lead to a mild phenotype in F patients, was also found to occur at high frequency in these men (ervais et al. 1993). More recently, the characterization of a large number of mutations in the FR gene has permitted nguiano et al. to confirm these initial results, by describing 3/25 BVD patients to be compound heterozygotes for FR mutations and so to propose that BVD could be a primary form of F (nguiano et al. 1992). he results of FR mutation screenings in earlier studies have shown very similar results: Osborne et al. (1993) found that 2/26 men with BVD were compound heterozygotes (8 were carriers of one mutation, and no mutation was identified in the 16 other patients). Patrizio et al. (1993) reported 59% (26/ 44) of BVD subjects as carrying one FR allele, and ulard et al. (1994) found 50% among 12 BVD patients. Different strategies have been used to scan the FR gene, but Osborne et al., ulard et al., and we have explored the coding sequence of the FR gene extensively, using either DE, SSP, or direct DN sequencing, and have obtained quite similar results. o date, evaluation of techniques allowing the identification of mutations in a cloned gene is still a matter of discussion. he DE tech-

5 276 m. J. Hum. enet. 56: , 1995 nique has probably the highest sensitivity of detection (-99%) compared with the other available techniques (SSP, chemical cleavage, and RNase protection assay) (Fodde and Losekoot 1994). In this large study, in which we analyzed 67 BVD patients from different countries, only 24% could be identified as compound heterozygotes. One can assume that our analysis of the coding sequence of the FR gene was very thorough, as we have been able to identify >98% of FlR mutations in our cohort of F patients (Ferec et al. 1992) and -95% of FR mutations in F patients of various origins (Mercier et al. 1993a). lthough we did not analyze the promotor and intronic regions of the FR, these results support the idea that FR plays a role in the genesis of BVD, but they are inconclusive, as 34% of these patients displayed no mutation at all in the coding sequence of the gene, and 42% were carriers of only one allele. his prompted us to look at segregation of FR alleles in families with at least one BVD male and a large number of siblings. One of our families, family R, (fig. 2) included two brothers who carry the common FS08 mutation (one of whom was obviously fertile, because he fathered two children). His full brother had BVD but had inherited identical chromosomes, as determined by microsatellite markers and haplotype analysis. he paternity of individual I14 to these two children has been confirmed using polymorphic markers. his transmission shows clearly, for the first time, to our knowledge, that at least one other gene is implicated in the pathophysiology of BVD. he precise involvement of FR in the development of organs is poorly understood. In male sex organs, the F pathology is observed as atrophy or absence of the body of the epididymis or of the vas deferens. he origin of this pathology may result either from defective development or from luminal obstruction of the ducts, as has been observed in the F epididymis. It has been observed that 1%- 2% of F males have been able to father children (Feigelson et al. 1969). In summary, after having studied this cohort of BVD patients, we have been able to show that -42% of these subjects are carriers of one FR mutation and that 24% are compound heterozygotes for one severe and one mild mutation. hese genotypes could be considered extremely mild forms of F, the more frequent genotype for this condition being FS08/R117H. It has been shown by Kiesewetter et al. (1993) that R117H occurs on two chromosomal backgrounds, one carrying a S variant with respect to a polypyrimidine tract in intron 8, which alters exon 9 splicing, and the other carrying a 7 variant associated with normal exon 9 splicing. he R117H mutations in this study are associated only with the 7 variant, indicating that the chromosomal background associated with BVD may be specific. hese data show that the BVD phenotype is determined by the association of the R117H with a particular chromosomal background. Our results lead us to propose that BVD is a heterogeneous and complex genetic disorder. Between 20% and 25% of cases may actually represent very mild forms of F, while 34% appear not to be associated with the F gene. he incidence of F mutations on just one of the alleles of the remaining proportion of BVD patients (43%) may implicate a role for FR, as suggested by the family we present. arrier status is insufficient to lead to BVD, and additional factors or genes are necessary for BVD pathophysiology in these cases. cknowledgments We would like to thank our colleagues-dr. David, Nantes; Dr. Parent and Dr. habaud, Brest; Dr. lavel, Reims; Dr. Delalande, Paris; Dr. B. Simon Bouy, Paris; and Prof. Dallapiccola, Rome-for storing samples. We also thank Odile Raguene's and Isabelle Quere for technical assistance and hantal Baclet for secretarial assistance. We thank Mr. R and his family for their kind comprehension and collaboration. his work was supported by the ssociation Francaise de Lutte contre la Mucoviscidose and by the Italian NR (P.F. FM). References nguiano, Oates RD, mos J, Dean M, errard B, Stewart, Maher, et al (1992) ongenital bilateral absence of the vas deferens- primarily genital form of cystic fibrosis. JM 267: udrezet MP, Mercier B, uillermit H, Quere I, Verlingue, Rault, Ferec (1993) Identification of 12 novel mutations in the FR gene. Hum Mol enet 2:51-54 Boat F, Welsh MJ, Beaudet L (1989) ystic fibrosis. In: Scriver R, Beaudet L, Sly WS, Valle D (eds) he metabolic basis of inherited disease. Mcraw Hill, New York, pp laustres M, Laussel M, Desgeorges M, iansily M, ulard JF, Razakatsara, D le J (1993) nalysis of the 27 exons and flanking regions of the cystic fibrosis gene: 40 different mutations account for 91.2% of the mutant alleles in southern France. Hum Mol enet 2: ulard JF, Desgeorges M, osta P, Laussel M, Razakatzara, Navratil H. D le J, et al (1994) nalysis of the whole FR coding regions and splice junctions in azoospermic men with congenital bilateral aplasia of epididymis or vas deferens. Hum enet 93: uppens H, Marynen P, De Boeck, assiman JJ (1993) Detection of 98.5% of the mutations in 200 Belgian cystic fibrosis alleles by reverse dot-blot and sequencing of the complete coding region and exon/intron junctions of the FR gene. enomics 18: Dork, Wulbrand U, Richter, Neumann, Wolfes H, Wulf B, Maa, et al (1991) ystic fibrosis with three mutations in the cystic fibrosis transmembrane conductance regulator gene. Hum enet 87: Dumur V, ervais R, Rigot JM, Lafitte JJ, Manouvrier S, Biserte J, Mazeman E, et al (1990) bnormal distribution of F F508 allele in azoospermic men with congenital aplasia of the epididymis and vas deferens. Lancet 336:512

6 Mercier et al.: Bilateral bsence of Vas Deferens in F 277 Fanen P, hanem N, Vidaud M, Besmond, Martin J, ostes P, Plassa F, et al (1992) Molecular characterization of cystic fibrosis: 16 novel mutations identified by analysis of the whole cystic fibrosis transmembrane conductance regulator (FR) coding regions and splice site junctions. enomics 13: Feigelson J, Pecau Y, Shwachman H (1969) propos d'une paternite chez un malade atteint de mucoviscidose. rch Fr Pediatr 26: Ferec, udrezet MP, Mercier B, uillermit H, Moullier P, Quere I, Verlingue (1992) Detection of over 98% cystic fibrosis mutations in a eltic population. Nature enet 1: Fodde R, Losekoot M (1994) Mutation detection by denaturing gradient gel electrophoresis (DE). Hum Mutat 3:83-94 aucher, Mercier B, Mazurier (1992) Von Willebrand disease family studies: comparison of three methods of analysis of the von Willebrand factor gene polymorphism related to a variable number tandem repeat sequence in intron 40. Br J Haematol 82:73-80 ervais R, Dumur V, Rigot JM, Lafitte JJ, Roussel P, laustres M, D le J (1993) High frequency of the R117H cystic fibrosis mutation in patients with congenital absence of the vas deferens. N Engl J Med 328: Holsclaw DS, Lober B, Jockin H, Schwachman H (1971) enital abnormalities in male patients with cystic fibrosis. J Urol 106: Kerem BS, Rommens JM, Buchanan J, Markiewicz D, ox K, hakravarti, Buchwald M, et al (1989) Identification of the cystic fibrosis gene: genetic analysis. Science 245: Kiesewetter S, Macek M Jr, Davis, urristin SM, hu S, raham, Shrimpton E, et al (1993) mutation in FR produces different phenotypes depending on chromosomal background. Nature enet 5: Lerman LS, Silverstein K (1987) omputational simulation of DN melting and its application to denaturing gradient gel electrophoresis. Methods Enzymol 155: Mercier B, Lissens W, udrezet MP, Bonduelle M, Liebaers I, Ferec (1993a) Detection of more than 94% cystic fibrosis mutations in a sample of Belgian population and identification of four novel mutations. Hum Mutat 2:16-20 Mercier B, Lissens W. Novelli, Kalaydjieva L, de rce M, Kapranov N, anki Klain N, et al (1993b) Identification of eight novel mutations in a collaborative analysis of a part of the second transmembrane domain of the FR gene. enomics 16: Morral N. Bertranpetit J, Estivill X, Nunes V, asals, imenez J, Reis, et al (1994) racing the origin of the major cystic fibrosis mutation (F508) in European populations. Nature enet 7: Morral N, Estivill X (1992) Multiplex PR amplification of three microsatellites within the FR gene. enomics 13: Morral N, irbau E, Zielenski J, Nunes V, asals, sui L, Estivill X (1992) Dinucleotide (/) repeat polymorphism in intron 17b of the cystic fibrosis transmembrane conductance regulator gene. Hum enet 88:356 Myers RM, Maniatis, Lerman LS (1987) Detection and localization of single base changes by denaturing gradient gel electrophoresis. Methods Enzymol 155: Osborne LR, Lynch M, Middleton P, lton EWFW, eddes DM, Pryor JP, Hodson ME, et al (1993) Nasal epithelial ion transport and genetic analysis of infertile men with congenital bilateral absence of the vas deferens. Hum Mol enet 2: Patrizio P. sch RH, Handelin B, Silber SJ (1993) etiology of congenital absence of vas deferens: genetic study of three generations. Hum Reprod 8: Rigot JM, Lafitte JJ, Dumur V, ervais R, Manouvrier S, Biserte J, Mazeman E, et al (1991) ystic fibrosis and congenital absence of the vas deferens. N Engl J Med 325:64-65 Riordan JR, Rommens JM, Kerem BS, lon N, Rozmahel R, rzelczak Z, Zielenski J, et al (1989) Identification of the cystic fibrosis gene: cloning and characterization of complementary DN. Science 245: Rommens JM, lannuzzi M, Kerem BS, Drumm ML, Melmer, Dean M, Rozmahel R, et al (1989) Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245: Silber SJ, Ord, Balmaceda JP, Patrizio P, sch RH (1990) ongenital absence of vas deferens: study on the fertilizing capacity of human epididymal sperm. N Engl J Med 323: eem JL, Berger H, Ostedgaard LS, Rich DP, sui L, Welsh MJ (1993) Identification of revertants for the cystic fibrosis F508 mutation using SE6-FR chimeras in yeast. ell 73: sui L (1992) Mutations and sequence variations detected in the cystic fibrosis transmembrane conductance regulator (FR) gene: a report from the ystic Fibrosis enetic nalysis onsortium. Hum Mutat 1: ucker SJ, annahill D, Higgins F (1992) Identification and developmental expression of the Xenopus laevis cystic fibrosis transmembrane conductance regulator gene. Hum Mol enet 1:77-82 Verlingue, David, udrezet MP, Le Roux M, Mercier B, Moisan JP, Ferec (1993) symptomatic carrier of two FR mutations: consequences for prenatal diagnosis? Prenat Diagn 13: Yorifuji, Lemna WK, Ballard F, Rosenbloom L, Rozmahel R, Plavsic N, sui L, et al (1991) Molecular cloning and sequence analysis of the murine cdn for the cystic fibrosis transmembrane conductance regulator. enomics 10: