Distinct Spectrum Of Cftr Gene Mutations

Hum Genet (1997) 100 : 365–377

© Springer-Verlag 1997

O R I G I N A L I N V E S T I G AT I O N

Thilo Dörk · Bernd Dworniczak · Christa Aulehla-Scholz · Dagmar Wieczorek · Ingolf Böhm · Antonia Mayerova · Hans H. Seydewitz · Eberhard Nieschlag · Dieter Meschede · Jürgen Horst · Hans-Jürgen Pander · Herbert Sperling · Felix Ratjen · Eberhard Passarge · Jörg Schmidtke · Manfred Stuhrmann

Distinct spectrum of CFTR gene mutations in congenital absence of vas deferens Received: 15 April 1997 / Accepted: 29 April 1997

Abstract Congenital absence of the vas deferens (CAVD) is a frequent cause for obstructive azoospermia and accounts for 1%–2% of male infertility. A high incidence of mutations of the cystic fibrosis transmembrane conductance regulator (CFTR) gene has recently been reported in males with CAVD. We have investigated a cohort of 106 German patients with congenital bilateral or unilateral absence of the vas deferens for mutations in the coding region, flanking intron regions and promotor sequences of

T. Dörk (Y) · J. Schmidtke · M. Stuhrmann Institut für Humangenetik, OE 6300, Medizinische Hochschule Hannover, D-30625 Hannover, Germany Tel.: +49-511-5323876; Fax: +49-511-5325865 B. Dworniczak · D. Meschede · J. Horst Institut für Humangenetik, Universität Münster, Münster, Germany C. Aulehla-Scholz · H.-J. Pander Abteilung für Klinische Genetik der Frauenklinik Berg, Stuttgart, Germany D. Wieczorek · E. Passarge Institut für Humangenetik, Universitätsklinikum Essen, Essen, Germany I. Böhm Genetisches Labor Dr Waldenmaier, München, Germany A. Mayerova Institut für Humangenetik der Universitätskinderklinik, Universität Freiburg, Freiburg, Germany H. H. Seydewitz Klinisch-chemisches Labor der Universitätskinderklinik, Universität Freiburg, Freiburg, Germany E. Nieschlag Institut für Reproduktionsmedizin, Universität Münster, Münster, Germany H. Sperling Klinik für Urologie, Universitätsklinikum Essen, Germany F. Ratjen Zentrum für Kinderheilkunde, Universitätsklinikum Essen, Essen, Germany

the CFTR gene. Of the CAVD patients, 75% carried CFTR mutations or disease-associated CFTR variants, such as the “5T” allele, on both chromosomes. The distribution of mutation genotypes clearly differed from that observed in cystic fibrosis. None of the CAVD patients was homozygous for ∆F508 and none was compound heterozygous for ∆F508 and a nonsense or frameshift mutation. Instead, homozygosity was found for a few mild missense or splicing mutations, and the majority of CAVD mutations were missense substitutions. Twenty-one German CAVD patients were compound heterozygous for ∆F508 and R117H, which was the most frequent CAVD genotype in our study group. Haplotype analysis indicated a common origin for R117H in our population, whereas another frequent CAVD mutation, viz. the “5T allele” was a recurrent mutation on different intragenic haplotypes and multiple ethnic backgrounds. We identified a total of 46 different mutations and variants, of which 15 mutations have not previously been reported. Thirteen novel missense mutations and one unique amino-acid insertion may be confined to the CAVD phenotype. A few splice or missense variants, such as F508C or 1716 G→A, are proposed here as possible candidate CAVD mutations with an apparently reduced penetrance. Clinical examination of patients with CFTR mutations on both chromosomes revealed elevated sweat chloride concentrations and discrete symptoms of respiratory disease in a subset of patients. Thus, our collaborative study shows that CAVD without renal malformation is a primary genital form of cystic fibrosis in the vast majority of German patients and links the particular expression of clinical symptoms in CAVD with a distinct subset of CFTR mutation genotypes.

Introduction Cystic fibrosis (CF) is a fatal autosomal recessive exocrinopathy that affects approximately one in 2000 individuals in Caucasian populations (Welsh et al. 1995). The physiological basis of the disease is characterized by abnormal secretion of electrolytes and fluid across the ep-

366

ithelial membranes of most exocrine organs. Clinical hallmarks of CF include chronic infections and pulmonary obstruction, pancreatic insufficiency, neonatal meconium ileus, failure to thrive, cholestasis, diabetes mellitus, elevated sweat electrolytes and male infertility (Welsh et al. 1995). CF, in its classic form, is usually diagnosed by two or more of these symptoms during the first few years of life and is associated with a mean life expectancy of currently about 30–35 years (FitzSimmons 1993). More than 95% of adult CF males are infertile because of obstructive azoospermia. The spectrum of observed Wolffian duct abnormalities in CF includes the absence or atrophy of the epididymal body and tail, seminal vesicles and the ejaculatory ducts and, in many cases, the congenital bilateral absence of the vas deferens (CBAVD; Kaplan et al. 1968; Taussig et al. 1972; Stern et al. 1982; Heaton and Pryor 1990; Wilschanski et al. 1996). CBAVD (McKusick 277180) has long been noted as a frequent cause of male infertility that is inherited in an autosomal recessive fashion (Nelson 1950; Schellen and van Stratten 1980); it occurs in 1%–2% of men presenting with infertility at urological departments and may have an incidence of up to 1:1000 in Caucasian males (Charny and Gillenwater 1965; Dubin and Amelar 1971; Holsclaw et al. 1971; Jequier et al. 1985; Oates and Amos 1994; Mak and Jarvi 1996). Although isolated CBAVD is considered a distinct clinical and genetic entity, it was suggested as long ago as 1971 that some males with CBAVD may have an unusually mild form of CF (Holsclaw et al. 1971). This assumption has been corroborated following the identification of the cystic fibrosis transmembrane conductance regulator (CFTR) gene, which is the causative defective in CF (Rommens et al. 1989; Riordan et al. 1989; Kerem et al. 1989). The observation that many men presenting with CBAVD have mutations in their CFTR genes has led to the proposal that CBAVD may be a primary genital form of CF (Dumur et al. 1990; Anguiano et al. 1992). Subsequent reports have confirmed the high incidence of CFTR gene mutations in otherwise healthy men with CBAVD (Patrizio et al. 1993; Oates and Amos 1994; Augarten et al. 1994; Casals et al. 1995; Mercier et al. 1995). However, the initial work also revealed that many investigated CBAVD patients did not seem to carry mutations on both copies of the gene. Two explanations were found for this apparent lack of CFTR mutations in this large subgroup of patients. First, CBAVD concomitant with congenital renal malformations is thought to represent a different clinical and embryological aetiology that is unrelated to the expression of CFTR (Rubin 1975; Augarten et al. 1994; Oates and Amos 1994; Dumur et al. 1996). Second, a partially penetrant CFTR splice variant termed the “5T” allele had initially not been recognized as a diseasecausing mutation but turned out to be associated with the phenotype of CBAVD in more recent studies (Chillón et al. 1995; Costes et al. 1995; Zielenski et al. 1995). In the present study, we have investigated the proportion and distribution of CFTR gene mutations in a large cohort of 106 German males who presented at various urological centres with congenital absence of the vas def-

erens (CAVD). We describe the results of this collaborative effort with particular emphasis on novel mutations that appear to be specific for CAVD and on the differences of CFTR genotypes between isolated CAVD and classic CF. This approach should improve our understanding of genotype-phenotype relationships for both of these common diseases.

Materials and methods Patients Our study group comprised 106 infertile men who were diagnosed as having obstructive azoospermia at various urology departments. All patients were found to have either bilateral (n = 101) or unilateral (n = 5) absence of the vas deferens by scrotal exploration and the clinical observation of impalpable vasa. Absence of seminal vesicles was confirmed in some cases by semen analysis (low fructose, low volume, low pH). Histological examinations were performed in most cases and revealed reduced spermatogenesis in a subset of patients. Unilateral renal agenesis was found in 9 men by abdominal ultrasonography. The patients were 25–38 years of age and most of them had parents of German descent. We only included patients who were apparently unrelated. In one family, two brothers were ascertained as having CBAVD and the CFTR genotype ∆F508/L375F. Mutation analysis: All patients underwent a genetic evaluation for the most common CFTR gene mutations at one of the participating centres in Hannover, Münster, Freiburg, Stuttgart, Essen or Munich. At this first step of mutation screening, 18 CFTR gene mutations known to be frequent in Germany were tested directly by allele-specific means, such as restriction enzyme analysis or heteroduplex analysis. This initial screening included the mutations ∆F508, G542X, R553X, G551D, N1303K, 1717–1 G→A, 3272–26 A→G, Y1092X, 2143delT, R347P, R347H, R334W, I336K, R117H, R117C, 2789+5 G→A, 3849+10kB C→T and the “5T” allele, the latter two splice variants being tested according to the instructions of Highsmith et al. (1994) and Chillón et al. (1995). We next investigated those patients with only one or no identified mutations and with no known renal malformations for novel and rare mutations of the CFTR gene by single-strand conformation analysis (SSCP) of all 27 exons and flanking intron and promotor sequences. Samples with an abnormal migration pattern in the SSCP analysis were directly sequenced by using the Sequenase 2.0 polymerase chain reaction (PCR) product sequencing kit (Amersham/USB). Primer sequences and SSCP conditions were as described elsewhere and had previously been optimized to detect more than 95% of mutations in 350 German CF families and 100% of mutations in 63 Austrian CF families (Dörk et al. 1994b; M. Stuhrmann et al. in preparation). However, we exchanged one upstream primer for the amplification of exon 23, as we had observed a frequent G/A polymorphism located within the primer binding site of the commonly used 23i-5 primer (Zielenski et al. 1991a), which under stringent conditions could result in an amplification bias from heterozygote samples. Our new primer 23i-5(II) had the sequence 5′-AGAAGTACTGGTGATTCTAC-3′ and could be used together with primer 23i-3 at an annealing temperature of 58° C. In addition to the coding region and flanking intron sequences, a 512bp portion containing the 5′UTR and minimum promotor region of the CFTR gene was amplified by primers 5′-GTCCTCCAG CGTTGCCAACTGG-3′ (forward) and 5′-CAACGCTGGCCTTTTCCAGAGG-3′ (reverse) under standard PCR conditions with annealing at 66°C. These products were also screened for mutations by SSCP analysis but, except for polymorphism 125 G→C, no variations could be detected in that region.

367 Evaluation of CF symptoms: All patients with CAVD were offered genetic counselling and an investigation of other phenotypic features consistent with CF. Several males refused any further clinical evaluation. Those who participated were asked for a history of respiratory disease, gastrointestinal or hepatobiliary manifestations or other disease symptoms. Sweat electrolytes were then measured in some cases by pilocarpine iontophoresis according to established methods (Gibson and Cooke 1959).

Results Mutation screening Direct analysis of the most common CFTR mutations was performed on 106 male German individuals presenting with CAVD: 92 had CBAVD without renal agenesis, 9 had CBAVD with concomitant unilateral renal malformation and 5 had unilateral absence of the vas deferens. Two CF mutations (∆F508 and R117H) and one splicing variant (the “5T” allele) were found to be particularly common in these patients. The 3-bp deletion ∆F508 in exon 10 was found on 57 chromosomes (26%) and was the most frequent mutation, although its incidence was much lower than among German CF patients (allele frequency of 72%). Missense substitution R117H was identified on 24 chromosomes (11%), which is an approximately 30fold increase in allele frequency compared with German CF chromosomes (Dörk et al. 1994b). This frequency of R117H seems to be the highest number reported so far in a patient subpopulation and confirms a particular association of R117H with the CAVD phenotype in Mid-Europe (Gervais et al. 1993). All the R117H alleles were from patients of German or Austrian descent and carried the same diagnostic marker haplotype composed of four intragenic dimorphisms and the 7T allele (Table 1). This haplotype is otherwise rare on normal or CF chromosomes and accounts for less than 0.5% in the German population, suggesting a common origin of the R117H mutation, which occurs at a CpG dinucleotide, in Mid-European populations. The third frequent gene variant in our CAVD cohort, collectively termed the “5T” allele, reflects at least three different alleles that can be distinguished by their different intragenic backgrounds. The “5T” reduction of the polythymidine tract in intron 8 was preceded by 13, 12 or 11 GT dinucleotides, respectively, and was associated with each of the three most common intragenic marker haplotypes (Table 1), indicating recurrent mutational events. In further support of recurrence, the “5T” variant was present in patients of German, Turkish, Lebanese, Vietnamese, Greek or Austrian descent. It is interesting to note that the most common “5T” allele in CAVD, viz. (TG)125T, is located on the most frequent dimorphic marker haplotype usually associated with (TG)117T in intron 8. Similarly, dimorphic marker analysis indicated that the (TG)135T allele resides on the most frequent (TG)127T haplotype, whereas the (TG)115T allele results from a (TG)107T haplotype. This suggests a

recurrent mutational mechanism that involves nucleotide misincorporation rather than deletion/insertion mutagenesis. Screening for the three mutations ∆F508, R117H and “5T” together allowed the identification of some 40% of mutations in our CAVD cohort. Only four other common CF mutations were found in more than one family by this initial screening: 2789+5 G→A on 4 alleles, R347H on 3 alleles, G542X and 3272–26 A→G each on 2 alleles (Table 1). No mutation was found in the nine CBAVD patients with known renal involvement. We then searched for additional novel and rare CFTR gene mutations by SSCP analyses of all exons and flanking intron regions in samples from those individuals in whom CFTR mutations had not been found on both alleles and renal malformation had not been documented. The 46 different mutations and variants that we were finally able to identify by using this approach are listed in Table 1. Fifteen mutations were uncovered that had never been observed previously on normal or CF chromosomes. Interestingly, thirteen of these are new missense substitutions, some of which may be subtle changes specific for the CAVD phenotype, and one mutation is a unique amino-acid insertion within the first transmembrane region. Missense mutations were also the predominant type of lesion on the other CAVD chromosomes where a CFTR mutation could be found, most of these missense mutations occurring within the CFTR transmembrane domains (Fig. 1). In contrast, only 8 alleles with nonsense or frameshift mutations were identified. The incidence of CFTR “null alleles” is therefore about three-fold lower in CAVD males than in German CF patients. New mutations Four novel missense mutations affect charged amino acids in the amino terminal transmembrane domain (TMD) of the CFTR protein. The E56K mutation was found in a German CBAVD patient who carried ∆F508 on the maternal allele and E56K on the paternal allele. The D58N mutation was uncovered in a Lebanese CBAVD patient who had inherited D58N from his father and a “5T” allele from his mother. The R297W mutation was identified on a Q1352H chromosome in a Vietnamese CBAVD patient (see below). The R334L mutation occurs at a position at which another mutation, R334W, has been described in mild to moderate CF and has been found to affect chloride channel conductivity significantly (Gasparini et al. 1991; Sheppard et al. 1993). The German CBAVD patient in our study was compound heterozygous for R334L and for the missense mutation I336K in the same exon. Six other new missense mutations were located in the carboxyterminal transmembrane domain, particularly in the third cytoplasmic loop (R933S, V938G, L973F, D979A). The R933S substitution was found together with the R75Q allele in a heterozygous CBAVD patient who also carried the ∆F508 deletion. The V938G substitution was identified in two unrelated patients, one homozygote with unilateral ab-

368 Table 1A Frequency distribution and haplotypes of CFTR mutations in 106 CAVD patients Mutationa

Nucleotide changesb

Locationc

Frequencyd Haplotypee

Referencef

174delA E56K D58N D110H R117H A120T hL138 L206W M265R R297W 1078delT R334W R334L I336K R347H L375F ∆F508

deletion of A at 174 G→A at 298 G→A at 304 G→A at 460 G→A at 482 G→A at 490 insertion of CTA after 546 T→G at 749 T→G at 926 C→T at 1021 deletion of T at 1078 C→T at 1132 G→T at 1133 T→A at 1139 G→A at 1172 A→C at 1257 deletion of 3 bp between 1652–1655 G→T at 1756 C→T at 1789 G→T at 1836 insertion of A at 2184 G→A at 2789+5 A→T at 2931 T→G at 2945 T→A at 3048 and C→T at 3049 A→C at 3068 G→C at 3123 A→G at 3227 A→G at 3272-26 G→C at 3586 T→A at 3590 deletion of C at 3659 G→A at 3978 C→G at 4041 A→G at 4183 G→C at 4261 T→A at 4295

exon exon exon exon exon exon exon exon exon exon exon exon exon exon exon exon

1 3 3 4 4 4 4 6a 6b 7 7 7 7 7 7 8

1 1 1 2 24 1 1 1 1 1 1 1 1 1 3 1

D3 B3 C2 C2 B6 n.p. A2 B8 A2 C2 C2 B1 D3 A2 D1 B3

This study This study This study Dean et al. (1990) Dean et al. (1990) Chillón et al. (1994) This study Claustres et al. (1993) Schwarz et al. (pers. comm.) This study Claustres et al. (1992) Gasparini et al. (1991) This study Cuppens et al. (1993) Cremonesi et al. (1992) Jézéquel et al. (1996)

exon exon exon exon exon intron exon exon

10 11 11 12 13 14b 15 15

57 2 1 1 1 4 1 3

B1 B1 A4 B3 D3 D3 n.p. D3

Kerem et al. (1989) Kerem et al. (1990) Cutting et al. (1990) This study Dörk et al. (1994b) Highsmith et al. (1997) This study This study

exon exon exon exon intron exon exon exon exon exon exon exon exon

16 16 17a 17a 17a 18 18 19 20 21 22 22 23

1 1 1 1 2 3 1 1 1 1 1 1 1

D3 n.p. A2 B3 D3 C2, A2 B3 C2 D3 B1 A2 C1 n.p.

This study This study Fanen et al. (1992) This study Fanen et al. (1992) Highsmith et al. (pers. comm.) This study Kerem et al. (1990) Vidaud et al. (1991) Osborne et al. (1991) This study Costes et al. (1995) This study

G→A at 356 C→G at 1184 reduction of oligoT tract to 5T at 1342-12 T→G at 1655 G→A at 1716 G→C at 1859 C→T at 2134 T→G at 3837 G→C at 4188

exon exon

3 7

2 1

A2 C4

Zielenski et al. (1991b) Mercier et al. (1993)

intron exon exon exon exon exon exon

8 10 10 12 13 19 22

26 3 3 2 2 1 2

C2, A4, D3, A2 C2 D3 D3 D3 n.p. C2

Chu et al. (1991) Kobayashi et al. (1990) Kerem et al. (1990) Fanen et al. (1992) Fanen et al. (1992) Cuppens et al. (1993) Nukiwa and Seyama (pers. comm.)

G542X R553X L568F 2184insA 2789+5 G→A R933S V938G L973F D979A L997F Y1032C 3272-26 A→G D1152H V1153E 3659delC W1282X N1303K K1351E D1377H L1388Q Variants: R75Qg T351S 5Th F508C 1716 G→A G576Ai R668Ci S1235R Q1352Hh a The

nomenclature of mutations follows Beaudet and Tsui (1993) The symbol “h” is used to designate an amino-acid insertion b Nucleotides are numbered according to the cDNA sequence of Riordan et al. (1989) c Exons and introns are numbered according to Zielenski et al. (1991a) d Allele frequency is given as number of chromosomes e Haplotypes were defined as listed in B below. Minor haplotypes are shown in italics; (n.p.) phase unknown f Personal communications to the Cystic Fibrosis Genetic Analysis Consortium (http://www.genet.sickkids.on.ca/cftr.html): M265R

by M. Schwarz, A. Haworth and G. Malone, D1152H by W. E. Highsmith jr., L. Burch, K. J. Friedman, B. M. Wood, A. Spock, L. M. Silverman and M. R. Knowles, Q1352H by T. Nukiwa and K. Seyama are indicated g Missense substitutions R933S and R75Q occurred together in a ∆F508 heterozygous patient h Q1352H is associated with 5T and R297W, respectively i Missense substiutions G576A and R668C are linked on the same allele in both CBAVD patients

369 Table 1B Dimorphic marker haplotypes. Haplotypes were defined following a previous study of CF chromosomes (Dörk et al. 1994b) by using the dimorphic extragenic RFLPs XV2c and KM.19 and the four intragenic markers (GATT), M470V, T854, TUB18 as previously published (Estivill et al. 1987; Dörk et al. 1992)

Extragenic XV-2c

KM.19

Intragenic

(GATT)n

M470V

T854

TUB18

A B C D

1 2 1 2

1 2 3 4 5 6 7 8

6 7 7 6 7 7 6 8

1 2 1 1 1 1 1 1

1 1 2 2 1 2 1 2

1 1 2 1 1 1 2 2

1 1 2 2

Fig. 1 Distribution of CAVD missense mutations in the transmembrane domains (TMD) of CFTR. Missense mutations were frequent in the sixth transmembrane helix of the amino terminal TMD (top) and throughout the first intracytoplasmic loop of the carboxyterminal TMD (bottom). New amino-acid substitutions are indicated in bold, whereas missense variants of uncertain significance are shown in italics. The membrane topology of the CFTR protein was adapted from the original report and has been confirmed in vitro by glycosylation site mutagenesis (Riordan et al. 1989; Chang et al. 1994)

sence of the vas deferens (CUAVD) and one heterozygote with CBAVD. The homozygous case may classify V938G as the first “pure” CUAVD mutation, whereas the heterozygous CBAVD patient carries a potentially severe mutation, the new frameshift deletion 174delA, on the other chromosome. The amino-acid substitution L973F results from a tandem nucleotide substitution in exon 16; this gives rise to a silent change in codon 972 and to the phenylalanine for leucine substitution at codon 973. Mutation Y1032C is an amino-acid substitution encoded by exon 17a and is predicted to reside within the tenth transmembrane helix of CFTR; it was found in a German CBAVD patient heterozygous for Y1032C and for ∆F508. D979A and V1153E are non-conservative amino-acid changes in exons 16 and 18, respectively, at positions adjacent to other known CBAVD mutations, viz. I980K and D1152H (Bienvenu et al. 1996; W. E. Highsmith et al. personal communication). Three other new missense substitutions, viz. L568F, K1351E and L1388Q, are located within or flanking the CFTR nucleotide-binding domains (NBD). L568F was detected in a CBAVD male of Turkish descent. This mutation targets the conserved “Walker B” ATP-binding motif of NBD1 at a crucial position (Higgins 1992). K1351E is located at a consensus motif of NBD2, which is essential for the signature of ATP-binding cassette transporters and which is thus a frequent site of CF mutations. Our finding of a German CBAVD patient heterozygous for ∆F508 and K1351E adds evidence to previ-

ous indications that ABC signature mutations have less severe consequences in the second than in the first CFTR nucleotide-binding fold (e.g. the CF mutation pair G551D/ G1349D). Missense mutation L1388Q flanks NBD2 at its carboxyterminal side and was found in a CBAVD male heterozygous for ∆F508 and L1388Q. This finding indicates that even substitutions within the final part of the CFTR protein exert subtle effects on CFTR function. Another novel mutation was a 3-bp insertion, termed hL138, within exon 4 of CFTR. This mutation was uncovered in a CBAVD male who is a compound heterozygote for the insertion and for a “5T” allele, from Southern Germany. His father had inherited the hL138 mutation and was of Polish descent; his ancestors had lived in the region of former Eastern Prussia for centuries. The hL138 mutation is predicted to add a fourth leucine residue to the second transmembrane helix at the cytosolic mouth of the channel pore. The molecular and pathological consequences of this unusual mutation, the first amino-acid insertion known so far in CFTR, remain to be defined. CFTR variants Several patients carried no typical CF mutation but a potentially disease-associated CFTR variant on at least one chromosome (Table 1). Splicing variants, such as “5T” or 1716 G→A, and missense variants, such as F508C or the

370 Table 2 Frequency distribution of CFTR variants in different subgroups of individuals. Allele frequencies of six CFTR variants were compared in randomly selected anonymous blood donors visiting the Medical School, Hannover, in the parents of CF patients Variant

125G→C R75Q 5T F508C 1716G→A G576A-R668C

(non-CF chromosomes), in our CAVD cohort and in a large cohort of CF patients. The latter group includes and extends that of a previous study (Dörk et al. 1994b). Complex alleles are indicated

Allele frequency n (% of chromosomes) Random donors

Non-CF

CBAVD

CF

15/178 4/188 9/186 0/188 5/188 0/188

n.d. 3/130 (2.1%) 2/65 (2.9%) n.d. 3/212 (1.5%) n.d.

2/212 (0.9%) 2/212 (0.9%)b 26/212 (12.3%)c 3/212 (1.4%) 3/212 (1.4%) 2/212 (0.9%)f

1/1000 1/1000 3/1000 2/1000 2/1000 3/1000

(8.5%) (2.2%) (4.8%) (2.6%)

a One

(0.1%)a (0.1%) (0.3%) (0.2%)d (0.2%)e (0.3%)f

e One

CF allele with R75X and 125G→C CBAVD allele with R75Q and R933S c One CBAVD allele with 5T and Q1352H d Two CF alleles with F508C and S1251N

CF allele with 1716G→A and L619S and R668C were linked on two CBAVD and three CF alleles, whereas two additional CF alleles carried R668C together with the 3849+10kB C→T mutation (Dörk and Stuhrmann 1995)

double mutant allele G576A and R668C, have previously been reported to be benign but no other mutation could be detected on these alleles in our patients after scanning the whole coding region, except for one case where the R75Q and R933S mutations were found together in a ∆F508 heterozygote. In order to investigate whether these variants are preferentially associated with the CAVD phenotype, we compared their frequencies in males with CAVD with the frequencies in a random panel of healthy German blood donors and in CF families (Table 2). The promotor polymorphism 125 G→C is unlikely to be a disease-associated substitution, as it is present, on two different haplotypes, in some 8.5% of chromosomes from the general population and has been seen only twice in our CAVD cohort. Conversely, the “5T” variant has an incidence of some 5% in the random donors but reaches a significantly increased allele frequency of 12.3% in our CAVD study group. The association of the “5T” allele with this particular disease phenotype is explained by the extensive skipping of exon 9 in epithelial CFTR mRNA transcripts (Chu et al. 1993; Chillón et al. 1995; Teng et al. 1997). The next two most frequent variants were the splice site substitution 1716 G→A and missense substitution R75Q. Splicing variant 1716 G→A was identified in three of our CBAVD males. This variant affects the final nucleotide of exon 10 and results in extensive exon skipping in carrier lymphocyte mRNA (data not shown). However, no specific association with CAVD was found in the present data set and we observed two proven fathers with the compound heterozygous genotypes ∆F508/1716 G→A or G551D/1716 G→A, respectively, in our cohort of CF families. On the other hand, we saw the compound heterozygous genotype ∆F508/1716 G→A twice in our CAVD cohort, which is 20-fold more often than that expected from the allele frequencies of each change in the general population. Interestingly, some CF patients had an unusually mild form of the disease and were compound heterozygous for ∆F508 and the splicing variant 1716 G→A, with no other CFTR mutation being detected, despite exhaustive searches (Cuppens et al. 1993; our un-

published data). Therefore, exclusion of the possibility that splicing variant 1716 G→A has the potential to result in very mild disease is premature and future investigations should examine possible genetic factors that may modify the penetrance of this variant. Similar considerations may apply to the five missense substitutions that have been described as polymorphic or rare variants on normal Caucasian chromosomes. R75Q is a candidate mutation but there is no evidence of a specific association with CAVD. We have identified, in a set of 65 CF families, two proven fathers with the compound heterozygous genotype ∆F508/R75Q and additional mutations have been detected here and elsewhere in patients with R75Q alleles (Zielenski and Tsui 1995; D. Hughes, personal communication). However, these complex alleles do not explain all cases and a final classification of R75Q requires further examination. The less frequent F508C variant was initially reported to be benign (Kobayashi et al. 1990) but is present in three of our CBAVD males who are all compound heterozygous for ∆F508 and F508C with no other detected mutation (one patient has been reported previously by Meschede et al. 1993). The missense substitutions G576A and R668C occurred in cis on the same allele in two of our CBAVD males and in three further patients with very mild CF (data not shown); it is thus difficult to decide whether one or both of these changes are required to predispose towards mild disease (Anguiano et al. 1992; Fanen et al. 1992; Chillón et al. 1995). Two other double mutant alleles were identified in a CBAVD patient of Vietnamese descent. This male was homozygous for a missense mutation, Q1352H, in exon 22 of the CFTR gene, but heterozygous for a “5T” allele and for the novel R297W missense mutation. Missense mutation Q1352H, located in the ABC signature of the NBD2 of CFTR, has so far only been observed in the heterozygous state in 1 out of 18 Japanese patients with diffuse panbronchiolitis (K. Seyama, personal communication). The Q1352H mutation may be insufficient to cause CBAVD but the additional occurrence of one “5T”

b One

f G576A

371 Table 3 CFTR mutation genotypes in 106 males with CAVD Genotype

PolyT

Frequency Ethnic descent

Diagnosis

∆F508/R117H ∆F508/5T ∆F508/F508C ∆F508/R347H ∆F508/1716 G→A ∆F508/3272-26 A→G ∆F508/E56K ∆F508/M265R ∆F508/R334W ∆F508/T351S ∆F508/L375F ∆F508/G576A & R668C ∆F508/R933S ∆F508/L997F ∆F508/Y1032C ∆F508/D1152H ∆F508/K1351E ∆F508/D1377H ∆F508/L1388Q ∆F508/unknown

9/7 9/5 9/7 9/9 9/7 9/7 9/7 9/7 9/9 9/9 9/7 9/7 9/7 9/9 9/7 9/7 9/7 9/7 9/7 9/7

21 9 3 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 4

German, Austrian German, Austrian German German German German German German-Portuguese German German Volga German German German German German German German Portuguese German German

20 CBAVD, 1 CUAVD 8 CBAVD, 1 CUAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD 3 CBAVD, 1 CUAVD

5T/5T 5T/G542X 5T/D58N 5T/hL138 5T/1078delT 5T/R553X 5T/2184insA 5T/D979A 5T/D1152H 5T/3659delC 5T/S1235R 5T/W1282X 5T & Q1352H/ R297W & Q1352H 5T/unknown

5/5 5/9 5/7 5/7 5/7 5/7 5/7 5/7 5/7 5/7 5/7 5/7

2 2 1 1 1 1 1 1 1 1 1 1

German German, Turkish Lebanese German-Polish German German Turkish Vietnamese Turkish German Greek German

CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD CBAVD

5/7 5/7

1 1

Vietnamese German

CBAVD CBAVD

R117H/L206W R117H/2789+5 G→A R117H/unknown

7/9 7/7 7/7

1 1 1

German German German

CBAVD CBAVD CBAVD

2789+5 G→A/2789+5 G→A 2789+5 G→A/L973F

7/7 7/7

1 1

Lebanese German

CBAVD CBAVD

V938G/V938G V938G/174delA

7/7 7/7

1 1

Greek German

CBAVD CBAVD

D110H/D110H R334L/I336K R347H/N1303K L568F/D1152H 3272-26 A→G/V1153E

7/7 7/7 9/9 7/7 7/7

1 1 1 1 1

Turkish German German Turkish German

CBAVD CBAVD CBAVD CBAVD CBAVD

R75Q/unknown A120T/unknown 1716G→A/unknown G576A & R668C/unknown 2752-15 C→G/unknown Unknown/unknown

7/7 9/7 7/7 7/7 7/7

1 1 1 1 1 17

German German German German Iranian German, Turkish

CBAVD CBAVD CBAVD CBAVD CBAVD 7 CBAVD and 1 CUAVD without observed renal agenesis, 9 CBAVD with renal agenesis

372

Fig. 2 Spectrum of CFTR mutation genotypes in CF patients (left) and in patients with congenital absence of the vas deferens (right). Mutation genotypes were classified according to the nature of the CFTR mutation (premature termination mutations, missense substitutions or splice site mutations) to allow better comparison between the two patient populations. Note that the two most common genotype classes in CF (∆F508/∆F508 and ∆F508/termination) were absent in the males with isolated congenital absence of the vas deferens

allele and the R297W mutation on a homozygous Q1352H background may then reduce CFTR function to a diseasecausing level. In summary, whereas all these variants are candidates for disease-associated mutations with a reduced penetrance, further studies are required to establish their physiological significance. A firm association with CAVD has only been established for the “5T” variant.

Table 4 Clinical symptoms in 26 CAVD males. Patients are compiled for whom sweat chloride was elevated and/or CF-like symptoms were recorded. Height and weight parameters were in the normal range, or even indicated overweight in a few individuals. Sweat chloride values are given as the mean of two or three inde-

pendent measurements by pilocarpine iontophoresis (40–60 mM borderline, > 60 mM pathological; n.d. not determined). Lung function tests indicated initial pulmonary deterioration in a few cases (FEV1 forced expiratory volume in 1 s, given as percent predicted)

Subject

Age (years)

Genotype

Height (cm)

Weight (kg)

Sweat C1– Symptoms (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

33 37 31 32 33 31 33 28 32 34 33 28 25 34 33 31 31 31 29 32 37 28 33

∆F508/R117H ∆F508/R117H ∆F508/R117H R117H/unknown ∆F508/E56K ∆F508/M265R ∆F508/R334W ∆F508/R347H ∆F508/F508C ∆F508/Y1032C ∆F508/3272-26 A→G ∆F508/unknown 5T/D58N 5T/hL138 5T/1078delT 5T/G542X 5T/2184insA 5T/D979A 5T/D1152H 5T/W1282X 5T/unknown D110H/D110H R334L/I336K

172 178 181 164 193 192 182 n.d. 192 n.d. 172 185 184 177 187 181 n.d. n.d. n.d. 180 180 175 170

75 83 91 70 100 112 78 n.d. 98 n.d. 82 95 99 80 87 85 n.d. n.d. n.d. 76 74 80 65

46 31 n.d. 33 85 59 n.d. n.d. 32 n.d. 125 n.d. 55 53 56 79 60 55 57 n.d. n.d. n.d n.d.

24 25 26

35 30 29

N1303K/R347H V938G/174delA V938G/V938G

167 n.d. 197

77 n.d. 115

93 42 n.d.

Dyspnoe Nasal polyposis Nasal polyposis Recurrent infections Sinusitis, recurrent bronchitis Recurrent infections, pancreatitis Recurrent infections, pneumonia Recurrent infections Pneumonia Recurrent bronchitis Recurrent infections, maldigestion, FEVI 73% Maldigestion – – Recurrent bronchitis – Borderline pancreatic sufficiency Recurrent infections, FEVI 76% – Recurrent infections, nasal polyposis Nasal polyposis Asthma bronchiale, obstipation Recurrent infections, nasal polyposis, maldigestion, salt depletion episodes – – Asthma bronchiale

373

Spectrum of CFTR genotypes and associated phenotypes Some intriguing aspects of genotype-phenotype relationships for CF and CAVD arise from the analysis of the complete CFTR mutation genotypes in our CAVD cohort, as the genotype distribution and frequencies are markedly different from those in CF (Table 3, Fig.2). Whereas none of the males with isolated CAVD was homozygous for ∆F508, homozygosity was observed for four other mutations: one Lebanese and one Turkish male with CBAVD were homozygous for the mutations D110H and 2789+5 G→A, respectively, both of which had previously been identified as very mild CF mutations (Dean et al. 1990; Highsmith et al. 1997). As mentioned above, one further male presenting with unilateral absence of the vas deferens as the only clinical symptom was found to be homozygous for missense mutation V938G. Finally, two CBAVD patients were homozygous for the “5T” variant, one being compound heterozygous (TG)135T/(TG)125T and the other being homozygous for (TG)125T. Interestingly, none of the CAVD patients was compound heterozygous for ∆F508 and a nonsense or frameshift mutation or for any two other CF mutations that had previously been classified as “severe” with regard to pancreatic function, indicating that the correlation between mutation genotype and CAVD phenotype is remarkably strict. By far the most frequent CAVD genotype in our population was the ∆F508/R117H genotype that was present in every fifth German male presenting with CBAVD. A few other genotypes overlapped with those previously observed in some CF adults with mild or very mild disease, e.g. compound heterozygosity for ∆F508 and mutations R334W, R347H, 3272–26 A→G or D1152H. This partial overlap of clinical spectra prompted us to examine retrospectively whether additional disease symptoms were recorded or observable in males with these CF-like mutation genotypes. Cases with positive findings were compiled if patient history and clinical records were informative (Table 4). Sweat chloride measurements were initiated in a series of patients and yielded borderline or even pathological values in several cases. On closer examination, a subgroup of males presenting with CAVD had episodes of further clinical symptoms that most frequently involved the upper respiratory tract (Table 4).

Discussion CF and CAVD are such different phenotypes that they appear to be distinct clinical entities. CF is a life-threatening disease that is usually diagnosed by pediatricians in children during their first few years, because of gastrointestinal and pulmonary manifestations and failure to thrive. CAVD is diagnosed by urologists in adult males presenting with an infertility problem; some of these patients have additional renal malformations, which is not a symptom of CF. However, despite these differences, a flurry of recent work indicates a genetic commonality for both diseases in the majority of patients. From the results of our

study, the type of mutation and the genotype of the CFTR gene appear to be pivotal to the differences in the clinical and tissue-specific expression of disease in these individuals. CAVD is a genetically heterogeneous disorder, involving CFTR gene mutations and variants in over 80% of our cases; 75% of our patients are compound heterozygous for two CFTR gene mutations or variants. Nine out of the 17 males with no CFTR mutation detected had unilateral renal agenesis or malformations concomitant with Wolffian duct abnormalities. This indicates a developmental insult to the mesonephric duct before weeks 6–7 of gestation and thus represents a distinct clinical aetiology and perhaps the involvement of other genes (Augarten et al. 1994; Oates and Amos 1994). However, the eight other males of our study who had neither detectable CFTR mutations nor an apparent renal maldevelopment did not appear to fall into this category. A similar proportion of 10– 30% of CBAVD males with no detectable CFTR mutation has been noted by others (Chillón et al. 1995; Costes et al. 1995; Zielenski et al. 1995) and one of these studies did not include patients with renal agenesis (Costes et al. 1995). This could indicate a failure to detect all CFTR mutations within the investigated portions of the gene (although it would be unusual to have so many patients unresolved on both alleles) or the presence of subtle forms of kidney malformation. Alternatively, these males may belong to a minor entity of CAVD that is neither caused by CFTR gene mutations nor defined by mesonephric duct maldevelopment. Further clinical and molecular studies are required to address this issue. Recent work on CFTR-knockout mice suggests that CF-associated male infertility is directly related to an impairment of chloride transport, the cardinal function of CFTR (Leung et al. 1996). In this regard, it is interesting that the vast majority of CAVD-associated mutations in our study are missense substitutions, many of which target transmembrane regions known to regulate chloride permeability. Six different amino-acid substitutions are located within the sixth transmembrane domain and two others flank the transmembrane helix 12. These two helices appear to contribute most to the anion conductance of CFTR (McDonough et al. 1994; Cheung and Akabas 1996). A cluster of four novel missense substitutions, including the “ultramild” V938G mutation, target the third intracytoplasmic loop. This portion of CFTR is thought to be involved in channel gating (Seibert et al. 1996). Two further CAVD missense mutations, viz. R117H and R347H, have been thoroughly studied in vitro and both were shown to result in a pH-sensitive decrease of chloride conductance (Sheppard et al. 1993; Tabcharani et al. 1993). We can conclude that this type of mutation, which alters but does not abolish the CFTR channel conductivity, is particularly frequent in men with isolated CAVD. A different class of mutation associated with CAVD may be represented by the frequent “5T” variant, which is thought to be a partially penetrant and “leaky” splicing mutation (Chillón et al. 1995; Costes et al. 1995; Zielenski et al. 1995). Homozygotes for “5T” have been de-

374

scribed who are phenotypically normal, although the “5T” variant induces the skipping of the essential exon 9 in approximately 90% of their CFTR mRNA (Chu et al. 1992, 1993). Our study indicates that homozygosity for “5T” can be sufficient to cause disease and adds further weight to the association of the “5T” variant with CAVD in another large cohort of individuals. It is noteworthy that the “5T” variant recurs on diverse ethnic backgrounds, such as German, Turkish, Lebanese or Vietnamese. This suggests that the CFTR gene is involved in many cases of male infertility in populations where CF is rare and ∆F508 is absent. Indeed, given their total allele frequency of some 5% in most investigated populations, the “5T” variants collectively appear to represent the most common disease-causing mutations of the CFTR gene worldwide. “5T” alleles have also been noted in some CF patients with very mild symptoms, perhaps defining the other end of the clinical spectrum associated with “5T” (Dörk et al. 1994a; Chillón et al. 1995). Severe disease results when additional mild mutations such as R117H occur in cis on a “5T” allele in certain populations (Kiesewetter et al. 1993) and the joint effect of both mutations strongly supports the view of CBAVD as representing a partial form of CF. Partial penetrance could be a feature not only of the “5T” variant, but also of other CFTR variants, which have initially been described as benign polymorphisms. The two linked missense substitutions G576A and R668C, for example, have previously been classified as polymorphisms on normal chromosomes (Fanen et al. 1992), as polymorphisms on CBAVD alleles (Osborne et al. 1993; Culard et al. 1994) or as separate disease-causing mutations in CBAVD (Anguiano et al. 1992; Chillón et al. 1995; Mercier et al. 1995). The penetrance of a CBAVD allele probably depends on the severity of the other allele in trans (Deltas et al. 1996). In addition, the expression of a clinical phenotype can be influenced by modifying genes, as has recently been observed in CF mice (Rozmahel et al. 1996). Therefore, missense variants, such as F508C or G576A, or splicing variants, such as 1716 G→A, deserve closer examination with regard to what extent they can impair CFTR function in an epithelial tissue, such as the vas deferens. The concept of partial penetrance may finally explain some observations of discordant brothers with and without CBAVD and who share the same CFTR genotype, although these cases could be equally well accommodated on the assumption of a different causative gene (Mercier et al. 1995; Rave-Harel et al. 1995). Striking CFTR genotypic differences are observed between our CAVD cohort and German patients with CF. Homozygosity for ∆F508 or compound heterozygosity for ∆F508 and a nonsense or frameshift mutation account for most cases with classic CF but are missing among the males presenting with isolated CAVD. Previous studies have reported few homozygous CBAVD patients for mutations R117H, D1152H and “5T” (Costes et al. 1995; Rave-Harel et al. 1995; Zielenski et al. 1995). Our analysis adds mutations D110H, 2789+5 G→A and V938G to the growing list of CFTR mutations that, in the homozy-

gous state, can result in a restricted and primarily genital expression of disease. Compound heterozygosity of these mutations with a “severe” CF allele, such as ∆F508, either leads to CBAVD or extends this phenotype to a pancreatic sufficient form of CF, which concurs with the proposed overlap between different CFTR-related phenotypes (Estivill 1996; Stern 1997). It will be interesting to determine whether these genotype-phenotype relationships hold true for the other congenital Wolffian duct abnormalities that have now been associated with compound heterozygosity for mild CFTR gene mutations, such as unilateral absence of the vas deferens or idiopathic epididymal obstruction (Mickle et al. 1995; Jarvi et al. 1995), and for possible female equivalents, such as hypofertility with thick cervical mucus (Gervais et al. 1996). The high frequency of mild or very mild CFTR mutations in patients with CAVD suggests that the vas deferens is one of the tissues most susceptible to the effect of changes in CFTR activity. However, the most common mild CF mutation in German CF adults with borderline sweat chloride and pancreatic sufficiency, viz. splicing mutation 3849+10kB C→T (Highsmith et al. 1994), has not been found in any of our 106 CAVD males. Interestingly, this mutation has repeatedly been observed in the few fertile male CF patients (Highsmith et al. 1994; Stern et al. 1995; Dörk and Stuhrmann 1995; Dreyfus et al. 1996). Thus, there seem to be two classes of mild CFTR mutations in males: those that primarily target the male genital tract (e.g. R117H, “5T”, D1152H) and those that may leave the vas deferens patent but exert deleterious effects at a more advanced age and lead to late-onset disease of the pulmonary tract, as is often observed for the 3849+10kb C→T mutation. Because some 75% of the CAVD males are homozygous or compound heterozygous for two CFTR mutations, some of them can be expected to have elevated sweat chloride levels or additional symptoms consistent with a mild form of CF, as seen here. Similar observations have been documented in a few initial studies (Costes et al. 1995; Durieu et al. 1995; Colin et al. 1996; Dumur et al. 1996). Whether these CAVD males should be treated as CF patients is doubtful. Few means exist to prevent CF-type symptoms, such as chronic infections, and it is not known whether these males would benefit from presymptomatic therapy. Instead, learning to cope with “having CF” in addition to infertility may raise anxiety in some patients or anger in others who feel healthy. Nevertheless, knowledge of their predisposition could prove helpful, if complications occur later in life, and may guide important clinical decisions, e.g. regarding the management of pulmonary disease. We believe that CAVD patients with two confirmed CFTR mutations should be counselled with all appropriate caution and should be offered clinical follow-up to ascertain potential complications as soon as possible. With experience gained from future prospective studies, mutation analysis of the CFTR gene may become of prognostic value in estimating the life-time risk for CF-like symptoms in men presenting with CAVD.

375 Acknowledgements We cordially thank all the men with CAVD and our CF families who agreed to participate in this study. We are indebted to our colleagues B. Fasselt, T. Dirksen, B. Albrecht, G. Gillessen-Kaesbach (Essen), J. Denil, S. Morlot (Hannover), L. Bispink (Bad Münder), I. Weber, G. Utermann (Innsbruck), D. Emmerich, B. Schmieders, G. Wolff (Freiburg), H. Schindler (Baden-Baden), W. Weidner (Gießen), M. Claßen (Bremen), M. Pruggmayer (Peine), S. Palm (Köln), W. Ihring (Leinefelde), H.L. Zienau (Darmstadt), R. and D.J. Ziegler (Ottersweier), E. Schröder (Düsseldorf), A. Freksa (Mainz), D. Wöhrle (Neu-Ulm) and H. Theile (Leipzig) for their contribution of patients and support of this study. Special thanks are due to Anja Schridde, Hildegard Frye, Andrea Korte, Andrea Trefilov, Kathrin Rommel, Wolfram Antonin, Alexander Stegh, Britta Skawran, Margit Ebhardt and Anneke Loos for their experimental contributions to haplotype analysis and mutation screening, and to Christoph Lanfer for his help with the manuscript. We gratefully acknowledge the collaboration with members of the Cystic Fibrosis Genetic Analysis Consortium (directed by Prof. Lap-Chee Tsui, Toronto). Part of this work was supported by a grant from the Deutsche Forschungsgemeinschaft.

References Anguiano A, Oates RD, Amos JA, Dean M, Gerrard B, Stewart C, Maher TA, White MB, Milunsky A (1992) Congenital bilateral absence of the vas deferens. A primary genital form of cystic fibrosis. J Am Med Assoc 267:1794–1797 Augarten A, Yahav Y, Kerem BS, Halle D, Laufer J, Szeinberg A, Dor J, Mashiach S, Gazit E, Madgar I (1994) Congenital bilateral absence of vas deferens in the absence of cystic fibrosis. Lancet 344:1473–1474 Beaudet AL, Tsui L-C (1993) A suggested nomenclature for designating mutations. Hum Mutat 2:245–248 Bienvenu T, Hubert D, Setbon E, Dusser D, Kaplan J-C, Beldjord C (1996) A novel missense mutation in exon 16 of the cystic fibrosis transmembrane conductance regulator (CFTR) gene identified in CBAVD patients. Hum Mutat 7:182 Casals T, Bassas L, Ruiz-Romero J, Chillón M, Giménez J, Ramos MD, Tapia G, Narváez H, Nunes V, Estivill X (1995) Extensive analysis of 40 infertile patients with congenital absence of the vas deferens: in 50% of cases only one CFTR allele could be detected. Hum Genet 95:205–211 Chang X-B, Hou Y-X, Jensen TJ, Riordan JR (1994) Mapping of cystic fibrosis transmembrane conductance regulator membrane topology by glycosylation site insertion. J Biol Chem 269:18572–18575 Charny CW, Gillenwater JY (1965) Congenital absence of the vas deferens. J Urol 93:399–401 Cheung M, Akabas MH (1996) Identification of cystic fibrosis transmembrane conductance regulator channel-lining residues in and flanking the M6 membrane-spanning segment. Biophys J 70:2688–2695 Chillón M, Casals T, Giménez J, Nunes V, Estivill X (1994) Analysis of the CFTR gene in the Spanish population: SSCP screening for 60 known mutations and identification of four new mutations (Q30X, A120T, 1812–1 G→A, and 3667del4). Hum Mutat 3:223–230 Chillón M, Casals T, Mercier B, Bassas L, Lissens W, Silber S, Romey M-C, Ruiz-Romero J, Verlingue C, Claustres M, Nunes V, Férec C, Estivill X (1995) Mutations in the cystic fibrosis gene in patients with congenital absence of the vas deferens. N Engl J Med 332:1475–1480 Chu CS, Trapnell BC, Murtagh JJ jr, Moss J, Dalemans W, Jallat S, Mercenier A, Pavirani A, Lecocq J-P, Cutting GR, Guggino WB, Crystal RG (1991) Variable deletion of exon 9 coding sequences in cystic fibrosis transmembrane conductance regulator gene mRNA transcripts in normal bronchial epithelium. EMBO J 10:1355–1363

Chu CS, Trapnell BC, Curristin SM, Cutting GR, Crystal RG (1992) Extensive posttranscriptional deletion of the coding sequences for part of nucleotide binding fold 1 in respiratory epithelial mRNA transcripts of the cystic fibrosis transmembrane conductance regulator gene is not associated with the clinical manifestation of cystic fibrosis. J Clin Invest 90:785–790 Chu CS, Trapnell B, Curristin SM, Cutting GR, Crystal RG (1993) Genetic basis of variable exon 9 skipping in cystic fibrosis transmembrane conductance regulator mRNA. Nat Genet 3: 151–156 Claustres M, Gerrard B, White MB, Desgeorges M, Kjellberg P, Rollin B, Dean M (1992) A rare mutation (1078delT) in exon 7 of the CFTR gene in a Southern French adult with cystic fibrosis. Genomics 13:907–908 Claustres M, Laussel M, Desgeorges M, Giausily M, Culard J-F, Razakatsara G, Demaille J (1993) Analysis 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 Genet 2:1209–1213 Colin AA, Sawyer SM, Mickle JE, Oates RD, Milunsky A, Amos JA (1996) Pulmonary function and clinical observations in men with congenital bilateral absence of the vas deferens. Chest 110:440–445 Costes B, Girodon E, Ghanem N, Flori E, Jardin A, Soufir JC, Goossens M (1995) Frequent occurrence of the CFTR intron 8 (TG)n5T allele in men with congenital bilateral absence of the vas deferens. Eur J Hum Genet 3:285–293 Cremonesi L, Ferrari M, Belloni E, Magnani C, Seia M, Ronchetto P, Rady M, Russo MP, Romeo G, Devoto M (1992) Four new mutations of the CFTR gene (541delC, R347H, R352Q, E585X) detected by DGGE analysis in Italian patients, associated with different clinical phenotypes. Hum Mutat 1:314–319 Culard J-F, Desgeorges M, Costa P, Laussel M, Ratzakatzara G, Navratil H, Demaille J, Claustres M (1994) Analysis of the whole CFTR coding region and splice junctions in azoospermic men with congenital bilateral aplasia of epididymis or vas deferens. Hum Genet 93:467–470 Cuppens H, Marynen P, De Boeck C, Cassiman 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 CFTR gene. Genomics 18:693–697 Cutting GR, Kasch LM, Rosenstein BJ, Zielenski J, Tsui L-C, Antonarakis SE, Kazazian HH jr (1990) A cluster of cystic fibrosis mutations in the first nucleotide-binding fold of the cystic fibrosis conductance regulator protein. Nature 346:366–369 Dean M, White M, Amos J, Gerrard B, Stewart C, Khaw KT, Leppert M (1990) Multiple mutations in highly conserved residues are found in mildly affected cystic fibrosis patients. Cell 61: 863–870 Deltas CC, Boteva K, Georgiou A, Papageorgiou E, Georgiou C (1996) Description of a symptomless cystic fibrosis L346P/ M348K compound heterozygous Cypriot individual. Mol Cell Probes 10:315–318 Dörk T, Stuhrmann M (1995) Severity of disease in cystic fibrosis. Lancet 346:1036–1037 Dörk T, Neumann T, Wulbrand U, Wulf B, Kälin N, Maass G, Krawczak M, Guillermit H, Ferec C, Horn G, Klinger K, Kerem BS, Zielenski J, Tsui L-C, Tümmler B (1992) Intra- and extragenic marker haplotypes of CFTR mutations in cystic fibrosis families. Hum Genet 88:417–425 Dörk T, Fislage R, Neumann T, Wulf B, Tümmler B (1994a) Exon 9 of the CFTR gene: splice site haplotypes and cystic fibrosis mutations. Hum Genet 93:67–73 Dörk T, Mekus F, Schmidt K, Boßhammer J, Fislage R, Heuer T, Dziadek V, Neumann T, Kälin N, Wulbrand U, Wulf B, Hardt H von der, Maass G, Tümmler B (1994b) Detection of more than 50 different CFTR mutations in a large group of German cystic fibrosis patients. Hum Genet 94:533–542

376 Dreyfus DH, Bethel R, Gelfand EW (1996) Cystic fibrosis 3849+10kB C→T mutation associated with severe pulmonary disease and male fertility. Am J Respir Crit Care Med 153: 858–860 Dubin L, Amelar RD (1971) Etiologic factors in 1294 consecutive cases of male infertility. Fertil Steril 22:469 Dumur V, Gervais R, Rigot J-M, Lafitte J-J, Manouvrier S, Biserte J, Mazeman E, Roussel P (1990) Abnormal distribution of CF ∆F508 allele in azoospermic men with congenital aplasia of epididymis and vas deferens. Lancet 336:512 Dumur V, Gervais R, Rigot J-M, Delomel-Vinner E, Decaestecker B, Lafitte J-J, Roussel P (1996) Congenital bilateral absence of vas deferens (CBAVD) and cystic fibrosis transmembrane regulator: correlation between genotype and phenotype. Hum Genet 97:7–10 Durieu I, Bey-Omar F, Rollet J, Calemard L, Boggio D, Lejeune H, Gilly R, Morel Y, Durand DV (1995) Diagnostic criteria for cystic fibrosis in men with congenital absence of the vas deferens. Medicine 74:42–47 Estivill X (1996) Complexity in a monogenic disease. Nat Genet 12:348–350 Estivill X, Scambler PJ, Wainwright BJ, Hawley K, Frederick P, Schwartz M, Baiget M, Kere J, Williamson R, Farrall M (1987) Pattern of polymorphism and linkage disequilibrium for cystic fibrosis. Genomics 1:257–263 Fanen P, Ghanem N, Vidaud M, Besmond C, Martin J, Costes B, Plassa F, Goossens M (1992) Molecular characterization of cystic fibrosis: 16 novel mutations identified by analysis of the whole cystic fibrosis conductance transmembrane regulator (CFTR) coding regions and splice site junctions. Genomics 13: 770–776 FitzSimmons S (1993) The changing epidemiology of cystic fibrosis. J Pediatr 122:1–9 Gasparini P, Nunes V, Savoia A, Dognini M, Morral N, Gaona A, Bonizzato A, Chillón M, Sangiulo F, Novelli G, Dallapiccola B, Pignatti PF, Estivill X (1991) The search for South European cystic fibrosis mutations: identification of two new mutations, four variants and intronic sequences. Genomics 10:193– 200 Gervais R, Dumur V, Rigot J-M (1993) High frequency of the R117H cystic fibrosis mutation in patients with congenital absence of the vas deferens. N Engl J Med 328:446–447 Gervais R, Dumur V, Letombe B, Larde A, Rigot JM, Roussel P, Lafitte JJ (1996) Hypofertility with thick cervical mucus: another mild form of cystic fibrosis? J Am Med Assoc 276:1638 Gibson LE, Cooke RE (1959) A test for concentration of electrolytes in sweat in cystic fibrosis of the pancreas utilizing pilocarpine by iontophoresis. Pediatrics 23:545–549 Heaton ND, Pryor JP (1990) Vasa aplasia and cystic fibrosis. Br J Urol 66:538–540 Higgins CF (1992) ABC transporters: from microorganisms to man. Annu Rev Cell Biol 8:67–113 Highsmith WE jr, Burch LH, Zhou Z, Olsen JC, Boat TF, Spock A, Gorvoy JD, Quittell L, Friedman KJ, Silverman LM, Boucher RC, Knowles MR (1994) A novel mutation in the cystic fibrosis gene in patients with pulmonary disease but normal sweat chloride concentrations. N Engl J Med 331:974–980 Highsmith WE jr, Burch LH, Zhou Z, Olsen JC, Strong TV, Smith T, Friedman KJ, Silverman LM, Boucher RC, Collins FS, Knowles MR (1997) Identification of a splice site mutation (2789+5G→A) associated with small amounts of normal CFTR mRNA and mild cystic fibrosis. Hum Mutat 9 : 332–338 Holsclaw DS, Perlmutter AD, Jockin H, Shwachman H (1971) Genital abnormalities in male patients with cystic fibrosis. J Urol 106:568–574 Jarvi K, Zielenski J, Wilschanski M, Durie P, Buckspan M, Tullis E, Markiewicz D, Tsui L-C (1995) Cystic fibrosis transmembrane conductance regulator and obstructive azoospermia. Lancet 345:1578 Jequier AM, Ansell ID, Bullimore NJ (1985) Congenital absence of the vasa deferentia presenting with infertility. J Androl 6: 15–19

Jézéquel P, Chauvel B, Le Treut A, Le Gall JY, David V, Le Lannou D, Blayau M (1996) Identification of a novel mutation in CFTR gene exon 8 (L375F) in a CUAVD phenotype. Hum Genet 97:548–549 Kaplan E, Shwachman H, Perlmutter AD, Rule A, Khaw KT, Holsclaw DS (1968) Reproductive failure in males with cystic fibrosis. N Engl J Med 279:65–69 Kerem BS, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, Buchwald M, Tsui L-C (1989) Identification of the cystic fibrosis gene: genetic analysis. Science 245:1073– 1080 Kerem BS, Zielenski J, Markiewicz D, Bozon D, Gazit E, Yahaf J, Kennedy D, Riordan JR, Collins FS, Rommens JM, Tsui L-C (1990) Identification of mutations in regions corresponding to the two putative nucleotide (ATP-) binding folds of the cystic fibrosis gene. Proc Natl Acad Sci USA 87:8447–8451 Kiesewetter S, Macek M jr, Davis C, Curristin MS, Chu C-S, Graham C, Shrimpton AE, Cashman SM, Tsui L-C, Mickle J, Amos J, Highsmith WE jr, Shuber A, Witt DR, Crystal RG, Cutting GR (1993) A mutation in CFTR produces different phenotypes depending on the chromosomal background. Nat Genet 5:274–277 Kobayashi K, Knowles M, O’Brien WE, Beaudet AL (1990) Benign missense variations in the cystic fibrosis gene. Am J Hum Genet 47:611–615 Leung AYH, Wong PYD, Yankaskas JR, Boucher RC (1996) cAMP- but not Ca2+-regulated Cl– conductance is lacking in cystic fibrosis mice epididymides and seminal vesicles. Am J Physiol 271 (Cell Physiol 40):C188-C193 Mak V, Jarvi K (1996) The genetics of male infertility. J Urol 156:1245–1257 McDonough S, Davidson N, Lester HA, McCarty NA (1994) Novel pore-lining residues in CFTR that govern permeation and open-channel block. Neuron 13:623–634 Mercier B, Lissens W, Audrezet MP, Bonduelle M, Liebaers J, Ferec C (1993) 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, Verlingue C, Lissens W, Silber SJ, Novelli G, Bonduelle M, Audrezet MP, Ferec C (1995) Is congenital absence of vas deferens a primary form of cystic fibrosis? Analyses of the CFTR gene in 67 patients. Am J Hum Genet 56:272–277 Meschede D, Eigel A, Horst J, Nieschlag E (1993) Compound heterozygosity for the ∆F508 and F508C cystic fibrosis transmembrane conductance regulator (CFTR) mutations in a patient with congenital bilateral aplasia of the vas deferens. Am J Hum Genet 53:292–293 Mickle J, Milunsky A, Amos JA, Oates RD (1995) Congenital unilateral absence of the vas deferens: a heterogeneous disorder with two distinct subpopulations based upon aetiology and mutational status of the cystic fibrosis gene. Hum Reprod 10: 1728–1735 Nelson RE (1950) Congenital absence of the vas deferens: a review of the literature and report of three cases. J Urol 63:176– 182 Oates RD, Amos JA (1994) The genetic basis of congenital bilateral absence of the vas deferens and cystic fibrosis. J Androl 15:1–8 Osborne L, Knight RA, Santis G, Hodson M (1991) A mutation in the second nucleotide binding fold of the cystic fibrosis gene. Am J Hum Genet 48:608–612 Osborne LR, Lynch M, Middleton PG, Alton EWFW, Geddes DM, Pryor JP, Hodson ME, Santis GK (1993) Nasal epithelial ion transport and genetic analysis of infertile men with congenital absence of the vas deferens. Hum Mol Genet 2:1605–1609 Patrizio P, Asch RH, Handelin B, Silber SJ (1993) Aetiology of congenital absence of vas deferens: genetic study of three generations. Hum Reprod 8:215–220

377 Rave-Harel N, Madgar I, Goshen R, Nissim-Raffinia M, Ziadni A, Rahat A, Chiba O, Kalman YM, Brautbar C, Levinson D, Augarten A, Kerem E, Kerem B (1995) CFTR haplotype analysis reveals genetic heterogeneity in the etiology of congenital bilateral aplasia of the vas deferens. Am J Hum Genet 56:1359–1366 Riordan JR, Rommens JM, Kerem BS, Alon N, Rozmahel R, Grzelczak Z, Zielenski J, Lok S, Plavsic N, Chou JL, Drumm ML, Iannuzzi ML, Collins FS, Tsui L-C (1989) Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 245:1066–1073 Rommens JM, Iannuzzi MC, Kerem BS, Drumm ML, Melmer G, Dean M, Rozmahel R, Cole JL, Kennedy D, Hidaka N, Zsiga M, Buchwald M, Riordan JR, Tsui L-C, Collins FS (1989) Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245:1059–1065 Rozmahel R, Wilschanski M, Matin A, Plyte S, Oliver M, Auerbach W, Moore A, Forstner J, Durie P, Nadeau J, Bear C, Tsui L-C (1996) Modulation of disease severity in cystic fibrosis transmembrane conductance regulator deficient mice by a secondary genetic factor. Nat Genet 12:280–287 Rubin SO (1975) Congenital absence of the vas deferens. Scand J Urol Nephrol 9:94–99 Schellen TMCM, Stratten A van (1980) Autosomal recessive hereditary congenital aplasia of the vasa deferentia in four siblings. Fertil Steril 35:401–404 Seibert FS, Linsdell P, Loo TW, Hanrahan JW, Riordan JR, Clarke DM (1996) Cytoplasmic loop three of cystic fibrosis transmembrane conductance regulator contributes to regulation of chloride channel activity. J Biol Chem 271:27493–27499 Sheppard DN, Rich DP, Ostedgaard LS, Gregory RJ, Smith AE, Welsh MJ (1993) Mutations in CFTR associated with mild disease form Cl- channels with altered pore properties. Nature 362:160–164 Stern RC (1997) The diagnosis of cystic fibrosis. N Engl J Med 336:487–491 Stern RC, Boat TF, Doershuk CF (1982) Obstructive azoospermia as a diagnostic criterion for the cystic fibrosis syndrome. Lancet I:1401–1403

Stern RC, Doershuk CF, Drumm ML (1995) 3849+10kB C→T mutation and disease severity in cystic fibrosis. Lancet 346: 274–276 Tabcharani JA, Rommens JM, Hou X-Y, Chang X-B, Tsui L-C, Riordan JR, Hanrahan JW (1993) Multi-ion pore behaviour in the CFTR chloride channel. Nature 366:79–82 Taussig LM, Lobeck CC, Di Saint’Agnese PA, Ackerman DR, Kattwinkel J (1972) Fertility in males with cystic fibrosis. N Engl J Med 287:586–589 Teng H, Jorissen M, Van Poppel H, Legius E, Cassiman J-J, Cuppens H (1997) Increased proportion of exon 9 alternatively spliced CFTR transcripts in vas deferens compared with nasal epithelial cells. Hum Mol Genet 6:85–90 Vidaud M, Fanen P, Martin J, Ghanem N, Nicholas S, Goossens M (1990) Three mutations in the CFTR gene in French cystic fibrosis patients: identification by denaturing gradient gel electrophoresis. Hum Genet 85:446–449 Welsh MJ, Tsui L-C, Boat TF, Beaudet AL (1995) Cystic fibrosis. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds) The metabolic and molecular bases of inherited disease, 7th edn. McGrawHill, New York, pp 3799–3876 Wilschanski M, Corey M, Durie P, Tullis E, Bain J, Asch M, Ginzburg B, Jarvi K, Buckspan M, Hartwick W (1996) Diversity of reproductive tract abnormalities in men with cystic fibrosis. J Am Med Assoc 276:607–608 Zielenski J, Tsui L-C (1995) Cystic fibrosis: genotypic and phenotypic variations. Annu Rev Genet 29:777–807 Zielenski J, Rozmahel R, Bozon D, Kerem BS, Grzelczak Z, Riordan JR, Rommens JM, Tsui L-C (1991a) Genomic DNA sequence of the cystic fibrosis transmembrane conductance regulator (CFTR) gene. Genomics 10:214–228 Zielenski J, Bozon D, Kerem BS, Markiewicz D, Durie P, Rommens JM, Tsui L-C (1991b) Identification of mutations in exons 1 through 8 of the cystic fibrosis transmembrane conductance regulator (CFTR) gene. Genomics 10:229–235 Zielenski J, Patrizio P, Corey M, Handelin B, Markiewicz D, Asch R, Tsui L-C (1995) CFTR gene variant for patients with congenital absence of vas deferens. Am J Hum Genet 57:958–960