Survey of CF Mutations Part 2

[Guest guest]

Outline   Results

Patients Patients and probands referrals to our laboratory for CF mutation

analyses were based on1) clinical symptoms typical for/indicative of CF,

(the clinical symptoms that prompted analysis for CF gene status

comprised: prior positive tests for immune reactive trypsinogen (IRT),

positive sweat test (> 60 mmol chloride/L), lung disease

(bronchiectasies), and others,2) indication for in vitro fertilisation

(IVF – group 2a: men, 2b: their spouses), and3) gene status determination

because of anticipated parenthood and partners or relatives affected by

CF.The first group (symptomatic) was comprised of 83 male (mean age: 29

yrs, range: 19–48 yrs) and 88 female patients (mean age: 31 yrs, range:

19–61 yrs) with 114 children/adolescents (60 male, 54 female) from < 1 to

18 yrs, the mean age of this group was 13.5 yrs. The clinical symptoms

for the individuals with mutations (where given) are shown in table 2), a

complete list is available upon request.The group 2a) (IVF) referred to

us from IVF clinics included 20 men (mean age: 34 yrs, range 24–47 yrs),

group 2b (IVF) twelve of their spouses with a mean age of 30 yrs (range:

22–36 yrs) and a presumable risk of being a CF carrier equal to the

population risk. Clinical symptoms in men included CBAVD, atrophy of the

testes, aplasy of the ductus deferens, aplasy of the epidymes, and

azoospermia. The women in group 2b (all without clinical symptoms) were

referred to CF analysis for gene status determination in the context of

in vitro fertilisation.Group 3 (mean age: 29 yrs, range: 9–69 yrs)

consisted of 26 men and 28 women. These persons wanted their gene status

to be determined, because either their children or close relatives were

afflicted by CF. The group included 17 couples and 12 individuals that

wanted to be tested in the context of planned parenthood.Amplicor cystic

fibrosis testThe Roche Amplicor CF mutation test strip was evaluated by

analysing in parallel 100 samples with a panel of techniques established

in our laboratory (Inno LiPa reverse dot blot – first generation, DGGE,

sequencing). For 27 persons in this cohort this included sequencing of

one to five exons. Besides various polymorphisms, only one splice site

and one frameshift mutation were found. Thus, 97% of all mutations in

this cohort that we could find routinely following our established

protocols could be analysed with the Roche Amplicor Cystic Fibrosis

test.The test is based on three major processes: Polymerase chain

reaction target amplification, hybridisation of the amplified products to

specific oligonucleotide probes, and detection of the probe-bound

amplified product by colour formation.DNA is extracted from minute

amounts of patients blood by lysis of the red blood cells and incubation

in Extraction Buffer at 100°C. This method is quick (45 min) and results

in very good quality of DNA. Subsequently, the DNA is PCR amplified with

8 pairs of biotinylated primers that simultaneously amplify eight

different regions of the CFTR gene (Table 1 – amplified regions plus the

mutations they encompass). In addition to the analysis of 16 common

mutations, the assay yields information about the poly-T tract in intron

8 and polymorphisms at locus F508. Single base exchanges at this locus

could lead to a lesser extent of hybridisation to the gene probes that

distinguish between F508 and ?F508. By comparison to the hybridisation

pattern at the polymorphic probes (see Fig. 1), misinterpretation can be

excluded.After the PCR, the amplicons are alkali denatured to form single

strands which then will hybridise to bound probes. After stringent

washes, a horseradish peroxidase-streptavidin complex binds to the

biotin-labelled amplicons captured by the membrane bound probes via the

streptavidin moiety. This conjugate is reacted after further washing

steps with hydrogen-peroxide and TMB to form a colour complex.DNA

studiesA total of 71 patients had CFTR mutations or the 5T allele or both

(table 2 – final results for all methods used), 19 patients were found to

have two mutations, 49 to have one mutation (not counting the 5T

allele).Among 114 children < 18 yrs in group 1), we found 9 patients to

be homozygote for ?F508, two compound heterozygote for ?F508/G542X, one

compound heterozygote for ?F508/3849+10kbC-T, five heterozygote for

?F508, one G551D/WT, one R117H/WT, one homozygote for 621+1G-T, and one

girl with 5T/7T alleles in intron 8 (total of 18% with

mutations).Twenty-two percent of the adults in group 1) had CFTR

mutations, namely two ?F508/?F508, thirteen ?F508/WT, one compound for

R117H/WT and 1466delAATT (frameshift mutation in exon 9), one

R117H/G542X, one G551D/WT, one 3849+10kb C-T/WT, one compound

heterozygote for ?F508/1248+1 A ? G (splice mutation in intron 7), and

two individuals with ?F508/3849+10kb C-T. Table 2) gives the details for

these individuals. Specifics for persons with no identified mutation

albeit clinical indications for CF are not shown because of space

limitations. They comprise 133 individuals, 54 of these were analysed for

16 mutations, 79 for 29 mutations of which 13 were examined further with

DGGE and sequencing of suspicious fragments (a detailed documentation is

available upon request). Virtually none of these developed CF as

confirmed by our referring clinicians [22].Group 2) consisted of 12 women

(all WT/WT and 7T/7T or 7T/9T) and 20 men. Nine of these men (45%) had

normal alleles (and a benign thymidine polymorphism in intron 8) at the

loci analysed, seven men had a genotype of 5T/9T with ?F508/WT, one

showed 7T/9T with ?F508/WT, one had 9T/9T with ?F508/WT, and one had

7T/9T with G542X/WT. As such, of the 20 male patients referred to us

because of infertility, 45% had the ?F508 mutation and 77% of those were

additionally affected in intron 8 by the 5T mutation (table 2). The women

in this group (all without clinical symptoms and normal genotype)

supposedly have the normal population risk of being a CF carrier.

Consequently, they are arranged in the separate category 2b.Of 54

individuals in group 3, tested because their partners or relatives had

CFTR mutations, 37% had mutated alleles: seventeen persons had the

genotype ?F508/WT, one had G551D/WT, one had 3849+10kb C-T, and one

person had G542X/WT (table 2).In total, the ?F508 mutation represented

83% (57 of 68) of all exon mutations in this cohort (table 2). The ?F508

mutation has a strong association with the 9T allele on the same

chromosome. This can be inferred because all ten ?F508 homozygotes had

9T/9T alleles, and 37 of 38 ?F508 heterozygotes had at least one 9T

allele. On the other hand, only 25 out of 188 WT/WT genotypes had a

7T/9T, four had a 9T/9T polymorphism, but 157 WT/WT genotypes were

associated with a 7T/7T polymorphism. This association of ?F508 and 9T

has been observed before [13]. Eight 5T alleles of a total of fifteen

were found in individuals other then the men with fertility problems

referred from IVF clinics.Of all 257 persons analysed for CF mutations,

105 were examined only for 16 mutations (Roche assay), 152 for 29

mutations (Roche + Innogenetics assay), 31 individuals were tested

further by DGGE of all exons and sequencing of suspicious fragments. Of

the 71 CFTR mutations, 69 were found merely by the assay with 16

mutations (n = 257), none was found additionally with the second reverse

line probe (n = 152), and 2 mutations were found by DGGE and sequencing

(n = 31). Evaluation of our results by contacting the referring

clinicians attested good performance of our assays: Virtually all

patients with CF were found, whereas probands without mutations (by our

tests) did not develop CF later on [22].In comparison to CF

mutation-frequencies in some European countries, the CF alleles we found

(>2%) with our tests show the following distribution in our cohort:

?F508: 83% (Romania:27%, Switzerland: 43%, Denmark: 87,2% [19]); G542X:

4,4% (France: 3,1%, Italy: 4,8%, Spain: 7,7%); G551D: 4,4% (UK: 3,1%,

Czechia: 4,0%, Ireland: 6,9%), 3849+10kbC-T: 2,9% (Germany: 1,2%, Poland:

2,6%, Latvia: 12,5%), R117H: 2,9% (Greece: 1,2%, Ireland: 2,0%, Norway:

3,0%)Because we participate in the European Quality assessment trial for

Cystic Fibrosis, we could evaluate the quality of the Roche Amplicor CF

test in this regard as well. In the case of the G551D/WT R553X/WT

genotype, the Roche test needs careful interpretation (and ought to be

improved) because the hybridisation at the one locus destabilised the

hybridisation at the other locus which is only 4 bp apart. Accordingly,

the line assay showed a G551D/G551D and R553X/R553X (two homozygote

mutations) genotype instead of the correct G551D/WT and R553X/WT

(compound heterozygote) genotype. The phenotypic classification for such

patient samples (full CF), however, was accurate, all other samples were

typed correctly. This limitation is not unique to the Roche CF assay and

is a characteristic of all mutation-specific assays in that sequence

variations near the interrogated mutation affect test accuracy.

Outline   Discussion

Since the discovery of the cystic fibrosis transmembrane conductance

regulator (CFTR) gene [14-16] the number of identified CF mutations has

increased enormously. The mutations can be classified according to their

influence of CFTR-mediated chloride secretion [17]. Mutations of category

I, II, and III lead to complete loss of gene function (associated with

pancreatic insufficiency), whereas class IV and V mutations confer

altered conductance properties or reduced synthesis of the CFTR protein.

Such reduced synthesis caused by varying degrees of exon 9 splicing have

been found associated with three length variants within the splice

acceptor site in intron 8 [18].Numerous methods for analysing CFTR

mutations are available, the number of mutations that should be tested

for routine diagnostic purposes still being a question of debate. It may

be best to have a two-tiered approach for mutation detection, with an

economical first step and further more elaborate techniques for

identifying rare mutations [19]. Because we perform the " economical first

step " in our laboratory routinely, up to 29 of the most common CFTR

mutations were analysed with line probe assays.Persons to be tested for

CF mutations are referred to our laboratory because of 1) symptoms

typical for/indicative of CF, 2) fertility problems, and 3) CF in

relatives/partners. This classification was kept for ease of

presentation. Because of financial and personnel restrictions, it is not

possible in our laboratory to analyse all specimen by DGGE [20] and

sequencing. Thus, reverse line probe assays are used as a first step

analysis tool. From DNA extraction to the results the turnaround time is

approximately 6 hrs. Both assays used are not automated and thus require

various manual steps. But the ease with which 30 mutations and, very

importantly, the thymidine polymorphism/mutation in intron 8 can be

tested, make these tests adept as a diagnosis tool for CF in the clinical

laboratory. Certainly, these tests were developed for the US and Europe

and cannot be adapted for other ethnic groups. In summary, the frequency

of CF chromosomes in the cohort examined with our tests was 26%, with the

?F508 allele affecting 83% of the CF chromosomes. This is considerably

higher than the proportion of ?F508 alleles of 55% found in a previous

study in southern Austria [21]. Five other mutations in CF chromosomes

were found with these tests, four of which with a frequency above 2%. The

only other mutations detected by sequencing, were a splice site mutation

in intron 7 (1248+1 A-G) and a frame shift mutation in exon 9

(1466delAATT).Austrian law requires genetic counselling for persons

seeking genetic diagnosis. In this context, careful explanation of

positive and negative test results must be given. Whereas the meaning of

a positive test result would yield clear conclusions, a negative result

does not exclude possible mutations at loci not examined by the test

used. As more than 800 putative mutations in the CFTR gene are known,

this has to be explained thoroughly to the patients seeking advice.The

phenotype/genotype correlation in our cohort was rather moderate: In the

group of individuals referred to us because of symptoms indicative for

CF, mutations were found in only 22% of the samples. However, our test

system is not inappropriate, as we have found practically all patients

with CF, whereas probands without mutations did not develop clinical

signs of CF later on [22]. Thus, the increasingly pressing need for fast

clinical reports was met by our tests. Even slight indications for

putative disease must be analysed for their possible significance,

nowadays, explaining the high number of samples with no mutations

found.Somewhat higher results (37%) were found for individuals referred

to us for genotype determination because their partners/relatives had

known CF mutations. Thirty samples with a WT/WT genotype by the reverse

line blot assays were analysed subsequently by DGGE and sequencing that

resulted in only two further determinations of the mutant genotype. In

both groups, the assays did not necessarily depict unknown mutations, but

the panel of the commonest CF mutations was determined with

certainty.Very informative, though, were the results for the patients

referred from in vitro fertility clinics. For more than half of the

afflicted males, a conclusive analysis could be offered by the tests, and

the exclusion of the most common mutations in their female partners was

of utmost importance for their family planning. In this context, the

determination of the 5T mutation in intron 8 was very significant. With

the wealth of information now available, the association of the 5T

mutation with fertility problems is evident. The 5T polymorphism is

considered now to be a mutation with incomplete penetrance [12]. The

diminished amount of CFTR gene product generated by the splicing defect

in intron 8 no longer sustains the physiological functions required.

    Conclusions

Our results show that current techniques allow a routine first-line

analyses of the CFTR gene status. It is a substantial improvement for

routine diagnostics to have available a test system for 30 mutations plus

the polypyrimidine length variants in intron 8. Certainly. further

developments will advance the diagnostic possibilities for CF

considerably. In this regard, it will be interesting to review our

samples with the new generation line probe assays that are being

developed, allowing a survey of up to 60 mutations [23].

Becki

YOUR FAVORITE LilGooberGirl

YOUNGLUNG EMAIL SUPPORT LIST

www.topica.com/lists/younglung

Pediatric Interstitial Lung Disease Society

http://groups.yahoo.com/group/InterstitialLung_Kids/