Main

The Quebec population numbers more than seven million, of which roughly six million have descended from French settlers. Among the 25,000 settlers who came from various provinces of France between the beginning of the 17th century and the British conquest of 1763, only about 8,500, including 1,600 women, settled permanently.13 It was demonstrated that the 2,600 settlers established in “Nouvelle-France” before 1680 contributed about two thirds of the gene pool of the current Francophone population.1,4 A mosaic of founder effects was observed in rural regions recently opened to colonization and, while inter-regional migrations have increased with time, regional genetic variation persists throughout the Province of Quebec (PQ). The Saguenay–Lac-Saint-Jean (SLSJ) region is a well-documented example of this phenomenon.5 It is a geographically isolated region located in the northeastern part of Quebec. Several studies have demonstrated an increased incidence of autosomal dominant and recessive disorders in this region.6,7 In SLSJ, it was documented that the incidence of cystic fibrosis (CF) reached 1 live birth per 902 between 1975 and 1988, which corresponds to a carrier rate of 1/15.8,9 This is three times higher than the average incidence of 1/2500 live births in other Caucasian populations (or a carrier rate of 1/25).

CF is a multisystemic disorder caused by mutations in the CF transmembrane conductance regulator (CFTR) gene.10 It affects both the respiratory and the digestive systems, and can also induce infertility in men.11 A variable genetic correlation was observed between CFTR mutations and pulmonary function,12 pancreatic insufficiency,13 and congenital bilateral absence of the vas deferens.14 This subject is reviewed at length in a recent review by Dorfman and Zielenski.15 Since it was cloned in 1989, over 1,500 CFTR mutations have been documented.16,17 Some of them, such as ΔF508, are commonly distributed, whereas others are found in specific populations or ethnic groups, such as the M1101K in Hutterites.18,19 The protein encoded by the CFTR gene is expressed in the apical membrane of exocrine epithelial cells, and is a cyclic adenosine monophosphate (cAMP)-induced chloride channel that can also regulate other ion channels.10,20

For CF neonatal screening, many programs throughout the world have adopted a two-tier combination of trypsinogen and DNA analysis with either ΔF508 allele alone or a panel of CF-causing mutations.21 In the same way, multimutation platforms can be built to offer carrier screening in different populations. To do so, there is a need to determine which mutations should be part of a CFTR mutation screening panel to reach a high sensitivity (>90%) in a particular population.22 In this report, after having assessed and compared the distribution of CFTR mutations in the SLSJ region with that of three other regions of PQ, we propose a mutation panel for carrier screening purposes in the SLSJ population.

MATERIALS AND METHODS

Subjects

The Canadian Consortium for Cystic Fibrosis Genetic Studies has genotyped samples from 6 of the 10 different CF clinics of the PQ, totaling 482 patients with CF (Canadian Consortium for Cystic Fibrosis Genetic Studies, unpublished data). It represents 45% of the patients with CF living in the PQ. All patients have an established diagnosis of CF. The phenotypic description of these patients, including sex ratio, mean age, percent predicted value of forced expiratory volume in one second (FEV1), and body mass index (BMI), is shown in Table 1. The project was approved by the ethics committee of the Hospital for Sick Children and informed consent was obtained from all subjects. In this study, we grouped the samples of the six clinics into four populations (see Fig. 1). The first population is from the SLSJ region, and is mainly composed of Francophones (Centre de santé et de services sociaux de Chicoutimi; n = 85). The second population, also mainly composed of Francophones, is from Sherbrooke (Centre hospitalier universitaire de Sherbrooke; n = 42), a city located southeast of Montreal and close to the United States border. Montreal is the largest city of the PQ and the most ethnically diversified. It is composed of individuals (Francophones and Anglophones) who have been living there for many generations, individuals (mostly Francophone) coming from all other regions of the PQ, and immigrants (multiethnics). Samples from the four CF clinics in Montreal were grouped into two populations: the third and the fourth populations of this study; one is mostly composed of Francophones (Centre hospitalier de l'Université de Montréal and Centre hospitalier universitaire Sainte-Justine; n = 292) and the other is mostly composed of Anglophones and immigrants (Montreal Children's Hospital and Montreal Chest Institute; n = 63).

Table 1 Phenotypic description of the subjects
Fig 1
figure 1

Map of the Province of Quebec. The different regions of the Province of Quebec. The numbers one, two, three, and four represent the Saguenay–Lac-Saint-Jean region (Chicoutimi CF clinic), the Sherbrooke CF clinic, the Montreal Francophone CF clinics, and the Montreal Anglophone and multiethnic CF clinics, respectively.

CFTR mutation screening

The CFTR genotyping was performed using two methods: multiplexed heteroduplex analysis23 and high-resolution melt analysis with SYTO9 (Invitrogen, Burlington, Canada) and Corbett Rotor-Gene 6000HRM (Corbett Life Science, Sydney, Australia) instrument24 followed by resequencing of the identified fragments. Large deletion detection was performed using established conditions.25,26 Some samples with incomplete CFTR genotype (n = 128) were screened for large deletions by Quest Diagnostics (Madison, NJ).27,28

Statistical analysis

A χ2 test was used to make comparisons between the allele distributions of the different populations, a P-value <0.05 was considered significant.

RESULTS

Data from the Canadian Consortium for Cystic Fibrosis Genetic Studies make it possible to describe the distribution of CFTR mutations in the SLSJ region and to compare it with that of three other regions in the PQ (see Table 2 for allelic frequencies). Three mutations are prevalent in the SLSJ population (ΔF508, 621 + 1G>T, and A455E); according to data provided by the genetic counseling services and the Chicoutimi CF clinic, three other mutations are present in at least three different families (711 + 1G>T, 3199del6, and Y1092X). Figure 2 illustrates the differences in the distribution of these six mutations in the SLSJ region compared with the three other regions. Although the ΔF508 and 621 + 1G>T mutations are more frequent in all populations studied, their distribution in the SLSJ region is different. The mutation ΔF508 is less represented in the SLSJ (Fig. 2,A) population than in the other French populations studied (P = 0.011) (Fig. 2,B andC). Moreover, the 621 + 1G>T is three to four times more frequent in the SLSJ population than in the two other Francophone population described here (P < 10−12) and more than 25 times more frequent than in the Anglophone and multiethnic population of Montreal (P < 10−7) (Fig. 2,D). Similarly, the A455E mutation frequency is two to three times higher in the SLSJ population compared with the other Francophone population studied (P = 0.004) and eight times higher than in the Anglophone and multiethnic population of Montreal (P = 0.013). Moreover, there is only one unknown CF allele in the SLSJ population compared with 25 only in the Francophone populations (P = 0.027) and also 25 in the Anglophone and multiethnic population (P < 10−8). Altogether, the six mutations represent 95.89% of the CFTR allele of CF patients in the SLSJ population, whereas the proportions are 86.85, 85.27, and 61.18% for the Sherbrooke CF clinic, the Montreal Francophone CF clinics, and the Montreal Anglophone and multiethnic CF clinics, respectively (P = 0.010, P < 10−3, and P < 10−13, respectively) (Figs. 2 and3).

Table 2 Cystic fibrosis mutations present in the four populations studied
Fig 2
figure 2

Distribution of the CFTR alleles in the Saguenay–Lac-Saint-Jean population in comparison with three other populations of the Province of Quebec. The percentage of six cystic fibrosis transmembrane conductance regulator (CFTR) alleles in four populations; subjects from (A) the Chicoutimi CF clinic in the Saguenay–Lac-Saint-Jean (SLSJ) region, (B) the Sherbrooke CF clinic, (C) the two CF clinics representing the Francophone population of Montreal, and (D) the two CF clinics representing the Anglophone and multiethnic population of Montreal. The three most common alleles in the SLSJ population are the ΔF508, 621 + 1G>T and A455E mutations. The frequency of the ΔF508 mutation is lower in the SLSJ population than in the other Francophones population (P = 0.011) but the frequency of the 621 + 1G>T and A455E mutation is greater in this region than in any other region described here (P < 10−12 and P = 0.004 for the Francophone populations, and P < 10−7 and P = 0.013 for the Anglophone and multiethnic population, respectively). Moreover, the percentage of unknown alleles is only 0.59% in the SLSJ region. It is lower than any other regions described in this study (P = 0.027 in Francophone and P < 10−8 in Anglophone and multiethnic populations).

Fig 3
figure 3

Percentage of CFTR allele covered with three or six mutations in the Saguenay–Lac-Saint-Jean population in comparison with three others populations of the Province of Quebec. With only three mutations, the rate of covered cystic fibrosis transmembrane conductance regulator (CFTR) allele obtained in the Saguenay–Lac-Saint-Jean (SLSJ) population (Chicoutimi CF clinic) is significantly higher than in the other regions, reaching 94.12% compared with 82.89, 80.36, and 59.48% for the Sherbrooke population, the Francophone population of Montreal, and the Anglophone and multiethnic population of Montreal, respectively (P = 0.005, P < 10−4, and P < 10−12). Moreover, with six mutations the rate is 95.89% for the SLSJ region, also significantly higher than 86.85, 85.27, and 61.18% for the other ones (P = 0.010, P < 10−3, and P < 10−13, respectively).

DISCUSSION

Bobadilla et al.21 proposed a 50-mutation platform for CF neonatal diagnosis, considering the most prevalent mutations and those that are present in different ethnic subgroups in the US population. Previously, the American College of Medical Genetics (ACMG) and the American College of Obstetricians and Gynecologists (ACOG), in conjunction with the National Human Genome Research Institute, have also proposed a panel for CF screening in the US population comprising 23 mutations.29,30 Those examples illustrate the great challenge encountered when building a multimutation screening program for CF disease in multiple populations, i.e., reaching a high sensitivity (>90%) while avoiding exclusion of any minority populations.22 In the SLSJ population, a founder effect occurred during the settlement of the region, thus diminishing the genetic diversity of its population; this effect is well illustrated for the CF population.31 Moreover, the SLSJ region is characterized by a low immigration rate; according to the 2001 census of the Quebec population made by the “Instituts de la statistique du Québec,” immigrants represent only 0.73% of the population (2,040/278,279 individuals).32 Finally, because there are few rare CF mutations in the SLSJ region, researchers have identified almost all the patients with CFTR mutations.

Table 1 gives phenotypic data about the subjects of the four regions studied. The mean age and body mass index (kg/m2) are similar for the four groups of children and adults who live in those regions. The only difference observed is for the FEV1 (% pred), which is lower for the children and adults from the Anglophone and multiethnic population of Montreal than for subjects of Francophone populations studied. As observed in the data from the Canadian Consortium for Cystic Fibrosis Genetic Studies (Figs. 2 and3), the distribution of CFTR mutations in SLSJ region is different from that in other regions in the PQ. In the study of Rozen et al.,33 the authors also observed that the frequency for ΔF508 mutation was lower in SLSJ region (58.0%) than in the other regions of the PQ (71%) (P = 0.047), and that subjects from the SLSJ region also have a higher 621 + 1G>T (23.2%) frequency than those of the remaining regions of the PQ (12.84%) (P = 10−5).21,33

Four of our six most frequent mutations (ΔF508, 621 + 1G>T, A455E, and 711 + 1G>T) are present in the ACMG-ACOG panel of 23 mutations, representing a detection rate of 94.71% in the SLSJ population. However, according to our results, a multimutation panel for carrier screening in the SLSJ region could include only the three principal mutations (ΔF508, 621 + 1G>T, and A455E), covering a total of 94.12% of the CFTR alleles present in the SLSJ region (Fig. 3). Another possibility is to include the three additional mutations that are at least present in three different families (711 + 1G>T, 3199del6, and Y1092X), reaching a detection rate of the CFTR alleles of 95.89% (Fig. 3). These two possibilities are greater than the 90% sensitivity threshold accepted for screening tests and would be at a low cost.22 The implementation of a CF carrier screening program could diminish the incidence of CF in the SLSJ region and, more importantly, allow future parents to make informed decisions about family planning.