© 1993 Oxford University Press Human Molecular Genetics, 1993, Vol. 2, No. 10 1551-1556

Direct sequencing of the complete CFTRgene: the molecular characterisation of99.5% of CF chromosomes in WalesJeremy P.Cheadle*, Mary C.Goodchild1 and Alison L.MeredithInstitute of Medical Genetics and 'Department of Chid Health (CF Unit), University of Wales College of Medicine, Heath Park, Cardiff CF4 4XN, UK

Received July 8, 1993; Revised and Accepted August 12, 1993

We have performed an extensive mutation analysis on 184 CF families In Wales. In our previous study) mutationson 329/369 CF chromosomes were Identified after screening for delta F508 and sixteen other mutations. To identifythe mutations on the remaining 40 uncharacterised CF chromosomes, we have carried out direct DNA sequencingover the complete coding region, Intron splice sites, and part of the promoter region of the CFTR gene. Duringthis study we have designed a set of internal sequencing primers which allow clear sequencing through theaforementioned regions. Sequence analysis revealed 15 further mutations (4 of which are novel), and 10 previouslydescribed polymorphisms. In total, we have Identified 29 mutations, the distribution of which provides furtherInsight into the functional domains of the CFTR protein. We have characterised 99.5% of the CF chromosomes(365/367, one sample degraded). In order to ascertain accurate frequency data for the Welsh population; CF familieswith at least 3 'Welsh' grandparents were strictly regarded as 'Welsh'. Of these 91 families, delta F508 accountsfor 71.6%, 621 + 1 G - T 6.6% and 1898 + 1 G - A 5.5%. The Implications for CF population screening in Walesare discussed.

INTRODUCTION

Cystic Fibrosis (CF) is the most common severe autosomalrecessive disorder of the Caucasian population, with an estimatedincidence of 1/2,500 and a carrier frequency of 1/25. In 1989,the cystic fibrosis transmembrane conductance regulator {CFTR)gene, and the most common mutation (delta F508), wereidentified (1,2, 3). The predicted protein, consisting of 1480amino acids, has two membrane spanning domains, twonucleotide binding domains, and a regulatory domain. Recentexperiments (4), have demonstrated that both the regulatorydomain and the nucleotide binding domains play crucial rolesin the functioning of CFTR as a low-conductance chloridechannel. Molecular studies on naturally occuring mutations allowfurther insight into the functional domains of the protein, andsince the characterisation of the CFTR gene, over 150 additionalmutations have been identified (5).

There are marked variations in the frequency of delta F508and some of the rarer CF mutations in different geographicallocations. For example, in Northern European populations, deltaF508 accounts for approximately 70% of all CF chromosomes,but in Southern Europe this decreases to as low as 30% (6).Furthermore, different ethnic groups also have varyingfrequencies of CF mutations; for instance W1282X, a nonsensemutation in exon 20, accounts for 60% of all Ashkenazi JewishCF chromosomes, but only 2% of the Sepharadic Jewish CFchromosomes (7). Therefore, to offer an effective diagnosticservice in a study population, one needs to know which mutationsare present (and their relative frequencies), to screen for themand to determine accurate risk estimates. Furthermore, one needs

to ascertain what proportion of CF chromosomes can be screenedrapidly, to assess the possibility of CF carrier screening in thatpopulation.

In a previous report (8), we identified the mutations on 329of our total 369 CF chromosomes, after screening for delta F508and sixteen other rarer mutations using the procedures describedby the original investigators. In this study, we have carried outdirect DNA sequencing over the complete coding region, theintron splice sites, and part of the promoter region of the CFTRgene, to search for the mutations on the remaining 40uncharacterised CF chromosomes. We have designed specificinternal sequencing primers for those regions for which wereunable to obtain a clear sequence from the PCR primers, whichhas enabled us to characterise the molecular defect in 99.5% ofour CF chromosomes.

RESULTS

The primers used for sequencing are presented in Table 1. Indesigning the internal sequencing primers, particular care wastaken to avoid homology to the termini of the PCR fragments,as our unpublished data indicate that even as few as eight non-contiguous homologous bases in a 20-mer is sufficient to yieldsequence from the ends of the PCR fragments. Using the PCRprimer PF, one can sequence clearly from -520 to -229 in thepromoter region, (nucleic acid sequence ref. 11). Chou et al.(11), have demonstrated that this region contains a negativeregulatory element (-345 to -277), and probably part of a

• To whom correspondence should be addressed

at Russian A

rchive on Decem

ber 24, 2013http://hm

g.oxfordjournals.org/D

ownloaded from

1552 Human Molecular Genetics, 1993, Vol. 2, No. 10

Table 1. Primers used for direct sequencing

EXON

Prom1

23

4

e6A

6B7

80

1 0

1 11 2

13*

SEQUENCING PRIMER

PF (PCR) I-620 TO -2291 .1t8'f: 6'-TGGAAGCAAATGACATCACAQ-3'ExJbPF (PCR) •3BS': 6--GCAQAQAATQQQATAQAQAQCTQaC-3'3e3'a: 6'-GTACAAATGAGATCCTTACCCC-3'4I5' (PCR) k4I3' (PCR) b5I3- (PCR) k6Ae6': 6--CACATTTCQTQTQQATCQCTCCTTTQC-3-6AB3 ' : 6'-GGGCAAGGACTATCAGaAAACC-3'6AI3' (PCR) b6CI31 (PCR) b718' (PCR) k713' (PCR) k813- (PCR) b986': 6'-GGTGATGACAGCCTCTTCTTCAG-3'T16L o1013' (PCR) b108: 6'-AGAATATACACTTCTGCTTAGGATG-3'1118' (PCR) b12«3': 6'-GCAATCTATGATGGGACAGTCTGTC-3'1316- (PCR) bR13e3': S'-AGGAGACAGGAGCATCTCCTTCTAATG-S'13«6b: 6-TGATTCTTTCGACCAATTTAGTG-3'

•This study, bZielenski el al. (9), cKerem et al. (10).

EXON

13b

14A14B16

16

17A

17B

18

19

20

2122

23

24

Table 2. Novel mutations identified during this study are in italic type

SEQUENCING PRIMER

C1-1M (PCR for 13a) bR13s3'b: 5'-TTQTCTTTCQQTQAATQTTCTQAC-3'R 1 3 B 3 ' « : 6'-CAAGCCAGTTTCTTGAGATAACCT-3'

1 3 B 6 C : S ' -CAQAACATTCACCQAAAQACAAC-3 P

14AI3' (PCR) b14Be6': 6-GTGQGCATGGQAGGAATAQG-3'1516' (PCR) b16e6b: 6'-TTQCTTQCTATQQQATTCTTCAQ-3'16e6': 6'-AQQAATTTQTCATCTTQTATATTA-3'16e3': 6-TACATACCTQQATQAAQTCAAATA-3'17At3' (PCR) k17A823': S-TTQQAQQAAATATQCTCTCAACAT-3'17BI5' (PCR) b17Bs3' (PCR) •18e3'b: S'-ACAQCCCACTQCAATQTACTC-3'1 8 B S C . 8'-TACQTCTTTTQTQCATCTATA-3'1SI3' (PCR) b19fl3b: S'-TTGACAQTCATTTQGCCCCC-3'2OI6' (PCR)k20fl6'c: 8-TGGATCAQGGAAGAGTACTTTG-3'2183': 6'-GTGTTGGTATGAGTTACCCC-3'22I3' (PCR)k22B3 ' . 6'-CTGGATCCAAATGAGCACTGGG-3'

23I6' (PCR)b23I3' (PCR)b24IS' (PCR)b248S': S'-CAAAGTGCGGCAGTACGATTCC-3'

HAPLOTYPE

LOCATION

Exon

Exon

Exon

Exon

Exon

Exon

Exon

Intron

Exon

Exon

1

3

8A

8B

7

11

11

12

13

13

Exon 14A

Intron

Exon

Exon

17A

17B

19

In t ror 1^

Exon 21

MUTATION

M1V

E60X

O220X

977 imA

107S delT

SS49N

8S49R

1B98«1 0-A

2184 dsIA

2184 InsA

W846X1

3272-26 A>Q

L1077P

3069 dtIC

1A4Q«10khC>TO O ̂ W lUHDv* 1

401B IntT

AMINO ACIDALTERATION

TRANSLATION1N1T1 AT 10 NMUTATION

Qlu>8top I t 00

Oln'Stop i t 2 :o

FRAME8HIFT

FRAMtBHIFT

Skr'Akn 11 540

8«r>Arg i t 549

8PLICINOM U TAT 10 N

FHAMESHIFT

FRAME8HIFT

Trp>8t0p kt 146

• PUCIHO.MUTATION

L«o>Pro It 1077

FRAME8HIFT

ACTIWTION OF

e r -L i c i S I T E

FRAMESHIFT

NUCLEOTIDEALTERATION

A'Q i t 133

Q>T It 310

C>T I I 780

1 bp Iniarlion

1 bp dt l f l lon

Q.A i t 1778

T'Q kl 1779

Q-A «l 1888M

1 bp delation

1 bp Insertion

Q>A at 2809

A'Q at 3272-20

T'C kt 3382

1 bp dt l t l lonC>T Ik k « .Zkk

10kb ttom i i s n 1(

1 bp i n s e r t i o n

o

^

2

2

2

2

2

1

1

2

2

-

1

2

1

-

2

|

4

1

2

1

1

-

2

1

1

1

1

2

1

1

2

1

s1

1

2

1

1

2

2

1

1

1

2

2

1

1

2

1

32-4

1

1

2

2

2

2

2

2

2

2

2

-

2

1

2

1p

p 1

PI

P I

P I

> •

P I

P I

P I

P 1

P I

>a

PI

PI

P I

PI

ALTERNATIVE MEANSOF DETECTION

NlMlll (-) Btgl (')

Mtelll <-)

DAftc 8'l«BAi S'-TQCTQaOAAQAAOCAATOO-a'r * u t I ' latAi 8-TACTOTCTTAAOTTTTCAATCA-S1

10 7INi fi'-aCACAQATAAAAACACCACOAA-S'ASP 1076Mi S'-aCACAOATAAAAACACCACTAO-a'

C- 718' IC- ( n n l T<sp- l t ' c

Ddal (-)

-

4 O _ l898*1Ni S'-OAACATACCT TTCAA-3'ASO i8t8*1Mi f'-TTQAAAOATATaTTC-S'

—

—

NIMIII (-)

-

-

I I I I K I 8-TTOTATaaTTTaaTTOACTAOa-J'ASP JSItMi S*-TTOTATQOTTTaaTTOACTCGT-3

C> 1011' IC- IsonS Tcnp>Sl'c

Hohl (*)* * fJ * 9 ' \ f

• B n <010H: •-TATTTATTTTTTCT00AAC-3'

REF

. b

- 0

1 2

1 3

1 4

1 0

1 8

- d

- «

1 6

17

- I

10

-o

12

•CheadleJ.P., BeUoni.E., Ferran.M., Millar-Jones.L. and Meredidi,A.L., manuscript in preparation. bMalone,G., Schwarz.M. and Super,M.,personal comm. to the CF genetic analysis consortium. cShackleton,S., Harris.A., Schwartz,M. and Holraberg,L., personal comm. to theCF genetic analysis consortium. dChevalier.F., Bozon.D., Dork.T., BosshammerJ., Merkus,F. and Tummler.B., personal comm. to theCF genetic analysis consortium. eKalin,N., Dork,T. and Tummler.B-, personal comm. to the CF genetic analysis consortium. fBozon.D.et al, personal comm. to the CF genetic analysis consortium. «Highsmith,W.E. Jr, Burch.L., Boat.T.F., Boucher.R.C, Silverman,L.M.and Knowles.M.R. personal comm. to the CF genetic analysis consortium.

positive regulatory element (-277 to -242). The PCR primerPR yields very poor sequence, and to date, our attempts to designa good internal sequencing primer to cover the bulk of the positiveregulatory element (—226 to -89) have been hampered by thehigh GC content.

In this sequencing study, we have identified fifteen mutations(Table 2 and Figure 1), four of which are novel (described in

detail elsewhere, refs 12, 16, and Cheadle et al. manuscript inpreparation), and ten previously described polymorphisms (datanot shown). We have characterised the molecular defect on 36/39CF chromosomes (one sample degraded). In the three CFchromosomes in which we excluded the site of the mutation fromthe complete coding regions, the intron splice sites, and part ofthe promoter region (see above) of the CFTR gene, we screened

at Russian A

rchive on Decem

ber 24, 2013http://hm

g.oxfordjournals.org/D

ownloaded from

Human Molecular Genetics, 1993, Vol. 2, No. 10 1553

S549Nheterozygote

G A T C

1078 delTheterozygote

G A T C

G T. T G

G G. G T

T GG T

' T T. T T -. C

TTG

: S

S549Rheterozygote

G A T C

3659 deICheterozygote

C T A G

" , T > G -»• Wt__

T A

• C T, T A• A C

C C .AA

3272-26A>Gheterozygote

G A T C

Figure 1. Direct sequencing of some of the previously published mutations. S549N (G—A a: 1778) and S549R (T—G at 1779) were sequenced from 1115', 1078delT from 7i5\ 3659 deIC from 19e3'b, and 3272-26A-G from 17Bi5'.

Membrane spanning

n n nn nnNBF1 R-domaln Membrane spanning

ii in n nn nnNBF2

1 2 3 4 S 8» Ob 7 8 8 10 1 1 1 2 13 14«14b IS 1 0 1 7 . 17b 18 19 20 21 22 23 24

M1VE60X

O220X

| R117H

G85E

621*1G>T

977msA

AF508I

AI6071898*1G>A

G642X 2184delA1078delT I

I 8549N 2184insA

I I1154insTC S549R

G651DI

R553XI

R560T

I1717-1G>A

W846X1

3272-26A>G

L1077P

R12B3U

3669delC4 016 in aT

IN1303K

3849*10kbC>T

Figure 2. Mutations identified in the 'Welsh' group are in bold type, and those which we identified are in italics.

for the mutation 3849+10 kb C —T, and identified one positive.For the mutations which did not alter restriction enzyme sites,and which were either relatively frequent in our population, orwhich would require screening in extended family members,allele specific primers or allele specific oligonucleotides weredesigned (Table 2).

The spectrum and distribution of the mutations in the CFTRgene

Including the data from our previous study, we have identified29 mutations in 183 CF families from Wales: 7 frameshiftmutations, 5 nonsense mutations, 9 missense mutations, 2 singleamino acid deletions, 5 splice mutations, and 1 translationinitiation mutation. These mutations are distributed over all ofthe domains of the CFTR protein (Figure 2): 8 are located in thefirst nucleotide binding fold (NBF), 3 in the second NBF, 2 in

the regulatory (R) domain, 3 in transmembrane spanningsegments, and the other 13 are in either the regions adjoiningthe aforementioned domains (7), non-coding regions (5), or thetranslation initiation codon (1). For our two remaininguncharacterised CF chromosomes, we did not identify anychanges in the promoter region screened.

Frequencies of CF mutations in the Welsh populationThe frequencies of the mutations in the 183 CF families in Walesare presented in Table 3. In our 'Welsh' group, 17 mutationsaccount for 99.5% (182/183) of the CF chromosomes. We havepreviously reported the unusually high frequency of 621 + 1G—T(6.6%) in the Welsh population (8), however, we were surprisedto find in this sequencing study another mutation, that of1898+1G^ A, which is also at a relatively high frequency(5.5%). To our knowledge, no other centre within the United

at Russian A

rchive on Decem

ber 24, 2013http://hm

g.oxfordjournals.org/D

ownloaded from

1554 Human Molecular Genetics, 1993, Vol. 2, No. 10

Table 3. Frequencies of mutations in 183 CF families m

MUTATION

delta FSOB621«1<3>T

18aS«1Q>AQ661OQ642XQ85E

R663X1078delTR1283M

3669delCR117H

delta I607N1303K

1717-1Q>AR660T

M1VE80X

Q220XS77lnaA

1154inaTC8S49N8649R

2184dtlA2184ineAW846X1

3272-26A>QL1077P

3849>10kbC>T4018ln«T

Total

Total

TOTAL

Welsh

131121045

24311212

1

1

11

182

183

71.8%6.6%5.5%2.2%2.7%

1.1%2.2%1.6%0.5%0.6%1.1%0.6%1.1%

0.6%

0.6%

0.5%0.5%

99.6%

0 6%

Other

1 14677442

11

1

11

111

111111

167

168

72.2%3.2%4.4%4.4%2.6%2.6%1.3%

0.6%0.6%

0.8%

0.8%0.8%

0.6%0.6%0.6%

0.6%0.6%0.8%0.8%0.6%0.8%

99.4%

0 6%

Undefined

2221

1

28

26

84.6%7.7%3.8%

3.8%

100%

Wales

TOTAL

2671918119644322222111111111111111

3 6 6

2

3 6 7

72.8%5.2%4.9%3.0%2.6%1 4%1.1%1.1%0.8%0.6%0.6%0.6%0.6%0.6%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%0.3%

09 6%

0 6%

Table 5. PCR cycling conditions

'Welsh' families were defined as those having three or more 'Welsh' grandparents.

Table 4. PCR primers designed in this study

Region

Promoter

Exon 1

Eun 2

Eun 14B

Eun 17A

Eun 17B

Prlnar

PP. B'-AOQCACATTTTCCTTCCCTTTTCAA-S1

PR. B'-CACCACCCCTTCCTTTTQCTCTTT-S'

ElilP, B-TCOOCTTTTAACCTOOaCAOTQA-3-Exim, B-0TO00CACOT0TCTTTCCQAAOC-3p

I i2bPP. B'-CATAATTTTCCATATOCCAO-S'ExZbPH. 8--CACCATACTTaOCTCCTATT-Si

I4BI1S'. «'-CACAAAaCAAAa0AA0ATQAAAT-3I4BI23'. 6'-OOQAOAAAraAAACAAAOTOOA-3'

17A»5'. e'-AOAAATAAATCACTQACACACTT-a'17A«3'> S'-AACAXTAAAOAATCTCAAATAaC-3'

17BIB'. S'-TTCAAAaAAraOCACCAQTQT-3' »17Ba3'. 8-OAATOCAOCATTTTATTOTTOA-3'

Produot alze

8 2 0

3 0 6

3 3 3

2 8 8

3 0 2

4 8 3

bp

bp

bp

bp

bp

bp

Exon 1 and 2 prlmara were daalgnad by C.WIoking (St.Mary'a)• Zlalanakl at al. (•)

Exon 1 and 2 primers were designed by C.Wicldng (St Mary's). @ Zielenskiet al (9).

Kingdom has reported similar frequencies for these twomutations. Furthermore, in a Celtic population from Brittany (18),only one 621 + 1G—T and no 1898+ 1G—A mutations wereidentified; this is particularly interesting since the Welshpopulation is believed to have similar origins to other 'Celtic'populations. The two most frequent mutations in the Bretonpopulation (excluding delta F508), are 1078 delT (4.9%) andG551D (4.1%) (ref. 18); both of these mutations account for2.2% of our 'Welsh' CF chromosomes.

In our total population, 29 mutations (of which 15 are foundonly once) account for 99.5% (365/367) of the CF chromosomes.The two CF chromosomes in which we have excluded the codingregions, intron splice sites, and part of the promoter of the CFTRgene for mutations, belong to two individuals, both of whom

Annealing

6 4*C

82 °C

61 °C

6 0°C

69 °C

68 C

67 °C

6«°C

6 6*C

64°C

62 °C

No. of

3 0

3 0

3 2

3 6

3 0

3 0

3 0

3 3

3 0

3 0

3 0

3 0

3 0

1

5,21,22

17B

Promoter

16,16

19

13a

6A

2,4,7,8,12,13b, 14A, 18,23,24

6B,9,10,11

14B.17A

3

2 0

Exon 13 was amplified as part of two fragments (a and b) as described by Zielenskiet al. (9).

convincingly met the criteria for a diagnosis of CF, and wereheterozygous for delta F508. Since one of these patients ispancreatic sufficient, whereas the other is insufficient, we suggestthat the two uncharacterised chromosomes harbour different CFmutations.

DISCUSSION

To date, three ethnically discrete populations, the Ashkenazi Jews(19), the Bretons (18), and the Hutterites (20) have characterised97% (5 mutations), 98.1% (19 mutations), and 100% (2mutations), of their CF chromosomes, respectively. Most otherpopulations have struggled to determine the mutations on 90%of their chromosomes (17, 21, 22). In our previous study weidentified the molecular defects on 329/369 CF chromosomes,by screening for delta F508 and many of the rarer mutationsprimarily through the use of allele specific primers, allele specificoligonucleotides and restriction analysis. The aim of this studywas to identify the mutations on our 40 remaining uncharacterisedCF chromosomes. By direct sequence analysis, we have identified15 further mutations accounting for 36/39 chromosomes.Furthermore, one of the three remaining CF chromosomes inwhich we had excluded the site of the mutation from the codingregion, splice sites, and part of the promoter region of the CFTRgene, was positive for the mutation 3849+10 kb C —T, whichis located deep within one of the introns. In total therefore, wehave identified the mutations on 365/367 CF chromosomes,99.5% (29 mutations).

It must be emphasized however, that the direct sequencingapproach to screening used here is far more labour intensive thanthe use of denaturing gradient gel electrophoresis, single strandconformational polymorphism, chemical cleavage, or hydrolinkanalysis. Conversely, all of the latter techniques need to be usedin conjunction with sequencing to characterise any aberrant bands(as they may be polymorphisms), so direct sequencing as the solemutation detection method may be justified, provided that thesample size is not too great. Furthermore, since these 'rapid'mutation detection techniques do not detect 100% of mutations,some mutations may be missed early on in the analyses and hencetime would be wasted screening the rest of the gene. Thesequencing primers that we describe here, allow clear sequencethroughout the coding region, splice sites and part of the promoter

at Russian A

rchive on Decem

ber 24, 2013http://hm

g.oxfordjournals.org/D

ownloaded from

Human Molecular Genetics, 1993, Vol. 2, No. 10 1555

region of the CFTR gene. These primers have been designed(often to be used in conjunction with the PCR primers of Zielenskiet al. ref. 9), to sequence through the regions which cannot beclearly obtained using the PCR primers, either because they lietoo far away from the region of interest or because of secondarystructure complications. These primers should make the directsequencing approach less time consuming and therefore morefeasible.

The mutations identified in the 183 CF families in Wales arespread over 14 exons and 5 introns of the CFTR gene. Themutations are fairly evenly distributed over this area, apart fromsix mutations which are clustered across 19 codons in exon 11,a region which has previously been proposed to be a hot spotfor mutations (23). In terms of establishing the importantfunctional domains of the CFTR protein, missense mutations areprobably the most informative type of mutation. Of the 9 missensemutations, 6 are in the nucleotide binding domains (4 in the first,2 in the second), 1 is in a transmembrane spanning segment, 1(Rl 17H) immediately flanks a transmembrane spanning segment,and 1 lies well within a region adjoining two transmembranespanning segments. Interestingly no missense mutations wereidentified in the R-domain, however we found the non-conservative substitution R668C (in exon 13) on a normalchromosome, implicating that it is benign, as reported by Fanenet al. (17). This may suggest that alterations in the R-domainare not critical for CFTR function; this is supported by theobservation of degenerate phosphorylation within this region (24).

Over the past two years, we have found that the ARMSstandard test kit (25), used in our pilot population screeningprogram, is a rapid and reliable (> 98% first time success rate,our unpublished data) means of screening for mutations. FurtherARMS tests are in development and in principle it should bepossible to use a three tube ARMS test to assay rapidly for 13mutations (delta F508, 621 + 1G-T, G551D, G542X, R1283M,W1282X, R553X, N13O3K, delta 1507, 1898+1G-A, R117H,R560T, and 1717-1G—A) (S.Little, personal comm.). Thiswould identify 92.6% of our total CF chromosomes, therebyreducing the risk of having a CF child to 1/1350 for a couplewhere one partner is a carrier and the other is negative for themutations tested. Some scientists (26) advocate that until adetection rate of 96% can be attained (where upon the risk ofhaving a CF child to the aforementioned couple is reduced tothe general population risk), population screening should bepostponed.

The two remaining uncharacterised mutations may well liewithin the large introns, the positive regulatory element of thepromoter, or the 3' untranslated region. Futhermore we cannotexclude the possibility that they may be large (whole exon orgreater) deletions, as these could have been missed by thesequencing approach. However, we conclude that the vastmajority of CF in Northern European populations is caused bymolecular defects within the CFTR gene.

MATERIALS AND METHODSPatients

In a study of 184 CF families, 369 CF chromosomes (one shared between firstcousins) were analysed. During the analysis one sample, a compound heterozygotefor delta F508 degraded, and since a re-sample could not be obtained, it wasexcluded. Although over 98% of the families currently reside in Wales, onlyfamilies with three or more 'Welsh' grandparents (91 in total) were strictly regardedas 'Welsh', in order to ascertain accurate frequency data for the Welsh population.Of the remaining 92 families, 79 had two or fewer 'Welsh' grandparents; the

proportion of grandparents in these families were 46% English, 42% Welsh, 7%Irish, 2.5% Scottish and 2.5% Southern European. In the other 13 families, thenationalities of the grandparents were unknown.

The diagnosis of CF was on the basis of clinical findings and abnormal sweatelectrolyte concentrations. The classification of pancreatic sufficiency (PS)depended on normal values for 3 day faecal fat excretion; only some pancreaticinsufficient (PI) patients had such tests but all these patients required pancreaticenzyme supplementation by clinical criteria. Where it was not possible to determinepancreatic status, the patient was classified as P?

PCR amplification and direct DNA sequencing

Initially, we amplified all of the exons by the polymerase chain reaction usingthe primers as described by Zielenski a al. (9). However, exons 1,2, 14B, 17Aand 17B did not amplify as part of a specific fragment under our standardconditions, so the primers were re-designed. Primers allowing the amplificationof part of the promoter regkm (-25 to -549), proposed to contain both a positiveand negative regulatory element (11), were also designed (Table 4).

The 50 /d reaction mixes contained 100-300 ng of genomic DNA, 0.2 mMof each dNTP, 5 pi of Cetus buffer (100 mM Tris-HCl pH 8.3; 500 mM KC1;15 mM MgCl2; 0.01 % gelatin), 50 pmol of the promoter primers, 25 pmol ofexon 1, 2, 3, 4, 6A, 7, 8, 9, 10, 11, 12, 13a, 13b, 14A, 14B, 16, 17A, 18,19, 20, 23 and 24 primers, 19 pmol of exon 6B primers, 12.5 pmol of exon15, 21 and 22 primers, and 6 pmol of exon 5 and 17B primers, and 1 unit ofCetus Amplitaq DNA polymerase. All PCRs had an initial denaturation at 94°C10 minutes, followed by 30-35 cycles of 52-64°C 1 minute (Table 5), 72°C1 minute, 93°C 30 seconds, and a final step of 52-64°C 1 minute and 72°C10 minutes. For the amplification of the promoter region, the reaction mixes weresupplemented with 10% DMSO and 2 minutes were allowed for each of the cyclingsteps. The amplified products were purified with Geneclean (Stratech).

Direct DNA sequencing with the primers shown in Table 1, was carried outas previously described (12).

3849+10 kb C - T screen

ThU amt mnifrl rait nrmrrting tn ri> JTKtnirrincw rrf HJghxmith.W F. ,Jr, Burch,L.,Boat,T.F., Boucher.R.C, Silverman.L.M. and Knowles.M.R. personal comm.to the CF genetic analysis consortium.

Primer design

Primers were designed with the aid of the oligo primer analysis software(MedProbe). For internal sequencing primers, special care was taken to avoidany homology between the primer (particularly at its 3' terminus) and the terminalends of the PCR fragments. Suitable oligonucleotides were then selected on thebasis of their internal stability distribution. Primers that are stable (have a highfree energy) at their 5'-termini but somewhat unstable on their 3'-ends performbest, as this structure effectively eliminates false priming.

ACKNOWLEDGEMENTS

We would like to thank Drs L.Al-Jader and S.Bell for their clinical advice, andN.Robertson (Cellmark Diagnostics) and R.Snell for helpful discussion duringthe course of this work. We also wish to express our gratitude to Prof. P.Harper,Drs D.Shaw and A.Clarke, for their comments on this manuscript. This workwas supported by the CF Trust (UK), the WSDHSR and the Welsh Office.

ABBREVIATIONS

Prom, Promoter; REF., Reference; PAGE, Polyacrylamide gel electrophorcsis;C, Common primer, IC, Internal control; ASO, Allele specific oligonucleotide;ASP, Allele specific primer; ARMS, Amplification refractory mutation system.

REFERENCES

1. RommensJ.M., Iannuzzi.M.C, Kerem.B., Drumm.M.L., Melmer.G.,Dean,M., Rozmahel,R., ColeJ.L., Kennedy.D., Hidaka.N., Zsiga.M.,Buchwald.M., RiordanJ.R., Tsui.L. and Collins.F.S. (1989) Science 245,1059-1065.

2. Riordan,J.R., Rommens.J.M., Kerem.B., Alon,N., Rozmahel.R.,Grzelczak.Z., ZielenskiJ., Lok,S., Plavsic.N., ChouJ., Drumm.M.L.,Iannuzzi,M.C, Collins.F.S. and Tsui.L. (1989) Science 245, 1066-1073.

3. Kerem.B., RommensJ.M., BuchananJ.A., Markiewicz.D., Cox.T.K.,Chakravarti.A., BuchwaM.M. and Tsui.L. (1989) Science 245, 1073-1080.

4. Anderson,M.P., Berger.H.A., Rich,D.P., Gregory,R.J., Smith.A.E. andWelsh.M.J. (1991) Cell 67, 775-784.

at Russian A

rchive on Decem

ber 24, 2013http://hm

g.oxfordjournals.org/D

ownloaded from

1556 Human Molecular Genetics, 1993, Vol. 2, No. 10

5. Tsui.L. (1992) Hum. Mm. 1, 197-203.6. European Working Group on CF Genetics (1990) Hum. Genet. 85, 391 -445.7. Shoshani,T., Augarten.A., Gazit.E., Bashan.N., Yahav.Y., Rivlin.Y.,

Tal.A., Seret,H., Yaar.L. , Kercm.E. and Kerem.B. (1992) Am. J. Hum.Genet. 50, 222-228.

8. CheadleJ., Myring.J., Al-Jader,L. and Meredith,L. (1992) J. Med. Genet.29, 642-646.

9. ZielenskiJ., Rozmahd.R., Bozon,D., Kerem.B., Grzelczak.Z., RiordanJ.R.,Rommens.J. and Tsui.L. (1991) Genomics 10, 214-228.

10. Kerem.B., Zielenski.J., Markiewicz.D., Bozon.D., Gazit,E., YahavJ.,Kennedy.D., RiordanJ.R., CoUins.F.S., RommensJ.M. and Tsui.L. (1990)Proc. Natl. Acad. Sci. USA 87, 8447-8451.

11. Chou,J., Rozmahel.R. and Tsui.L. (1991)7. Bid. Oiem. 266, 24471-24476.12. Cheadle.J.P., A]-Jader,L.N. and Meredith.A.L. (1993) Hum. Mol. Genet.

2, 317-319.13. Claustres.M., Gerrard.B., Whke.M.B., Desgeorges.M., Kjellberg.P.,

Rollin,B. and Dean.M. (1992) Genomics 13, 907-908.14. Cutung.G.R., Kasch.L.M., Rosenstein.B.J., ZielenskiJ., Tsui.L.,

Antonarakis.S.E. and Kazazian.H.H.Jr (1990) Nature 346, 366-368.15. Strong, T.V., Smh, L.S., Nasr, N., Wood, D.L., Cote, J.L., Ianuzzi, M.C.,

Stem, R.C. and Collins, F.S. (1992) Hum.Mut., 1, 380-387.16. CheadleJ.P., Al-Jader.L.N. and Meredith.A.L. (1993) Hum. Mol. Genet.

2, 1067-1068.17. Fanen.P., Ghanera.N., Vidaud.M., Besmond.C, Martin^., Costes.B.,

Plassa.F. and Goossens.M. (1992) Genomics 13, 770-776.18. Ferec.C, Audrezet.M.P., Mercier.B., Guillermit.H., MouJlier,P., QuereJ.

and Verlingue.C. (1992) Nature Genet. 1, 188-191.19. Abeliovich.D., Lavon.I.P., Lerer,I., Cohen.T., Springer.C, Avital.A. and

Cutting.G. (1992) Am. J. Hum. Genet. 51, 951-956.20. Zielenski.J., Fujiwara.T.M., Markiewicz.D., Paradis.A.J., Anacleto.A.I.,

Richaids.B., Schwartz,R.H., Klinger,K.W., Tsui.L. and Morgan,K. (1993)Am. J. Hum. Genet. 52, 609-615.

21. Cutting.G.R., Curristin,S.M., Nash,E., Rosenstein.B.J., Lerer,I.,Abeliovich.D., Hill.A. and Graham,C. (1992) Am. J. Hum. Genet. 50,1185-1194.

22. Kristidis.P., Bozon.D., Corey.M., Markiewicz.D., RommensJ., Tsui.L.and Durie.P. (1992) Am. J. Hum. Genet. 50, 1178-1184.

23. Tsui.L. (1992) Trends Genet. 8, 392-398.24. Cheng.S.H., Rich.D.P., Marshall,.!., Gregory.R.J., Welsh.M.J. and

Smith.A.E. (1991) Cell 66, 1027-1036.25. Ferric,R.M., Schwarz.M.J., Robertson.N.H., Vaudin.S., Super.M.,

Malone.G. and Little.S. (1992) Am. J. Hum. Genet. 51, 251-262.26. Ten Kate.L.P. (1989) Nature 342, 131.

at Russian A

rchive on Decem

ber 24, 2013http://hm

g.oxfordjournals.org/D

ownloaded from