Ann. Hum. Genet. (1991),55, 283-290 Printed in Great Britain
Population genetics of vitamin D-dependent rickets in northeastern Quebec M. DE BRAEKELEER AND J. LAROCHELLE Human Genetics, Department of Human Sciences, Universite'du Qudbec a Chicoutimi, Chicoutimi, Qudbec, Canada and Hdpital de Chicoutimi, Chicoutimi, Que'bec
Vitamin D-dependent rickets (VDD1) is an autosomal recessive disorder that was recognized in Saguenay-Lac-St-Jean (SLSJ)in 1970. The great majority of the VDDl cases reported in the French Canadian population of Quebec originated from SLSJ, Charlevoix, and the Haute CGte Nord, all regions located in northeastern Quebec. The prevalence a t birth in SLSJ was estimated a t 112916 live borns, and the carrier rate was estimated a t 1/27 inhabitants in the SLSJ region. The mean coefficient of inbreeding was not elevated in the VDDl group of SLSJ compared with three matched control groups. The mean coefficient of kinship was 2.5 times higher in the VDDl group than in the control groups. In the SLSJ region, the places of origin of the VDDl children and their children did not show a clustered non-uniform distribution. Endogamy was not found to be higher in the VDDl group than in control groups. The genealogical reconstruction showed all the obligate carriers of the VDDl gene, but one, to be related to a small set of founders who settled in New France in the 17th century. All these results, as well as a strong linkage disequilibrium between RFLPs located on the long arm of chromosome 12 and the VDDl locus, support the hypothesis of a founder effect for VDDl. They also suggest that a unique mutation accounts for most, if not all, of the cases known in northeastern Quebec. INTRODUCTION
Saguenay-Lac-St-Jean (SLSJ) is a geographically isolated region located 200 km northeast of Quebec City (Fig. 1). The region under study has been described in full detail elsewhere (De Braekeleer, 1991a ; De Braekeleer & Larochelle, 1990). Vitamin D-dependent rickets (VDD1) was first reported in Saguenay-Lac-St-Jean in 1970 (Scriver, 1970; Arnaud et al. 1970). It is an autosomal recessive disorder, caused by a defective 1-a-hydroxylation of 25(OH)D (Scriver, 1970; Arnaud et al. 1970; Fraser et al. 1973). Therefore, serum levels of 1,25(OH)ZD are very low but usually detectable (Scriver et al. 1978; Delvin et al. 1981). The disorder is known to be common in the SLSJ region. The gene frequency was estimated to be 0.02 ; this high frequency was attributed to a founder effect rather than to a high rate of inbreeding (Bouchard et al. 1984, 1985). In the last years, research has focused on mapping the VDDl locus. Recently, Labuda et al. (1990) reported a linkage between chromosome 12 markers and the VDD1 locus. The most Correspondence to: Dr Marc De Braekeleer, Department of H u m a n Sciences, Universitd du Qu6bec a Chicoutimi, ,555 Boulevard de l'liniversitd Chicoutimi, Qudbec G7H 2B1, Canada.
M. DE BRAEKELEER AND J. LAROCHELLE
probable order of loci was also determined to be cen-COL2Al-VDD1-(D12S14,D12S17)D12S6-DlZS8qter. A strong linkage disequilibrium was found between a three-marker haplotype (DlZSl4,D12S17,and D12S6)and VDDl;indeed, 11 of 16 VDDl chromosomes (69%) carried the same haplotype (Labuda et al. 1990). The present study was aimed a t analysing the geographical distribution of VDDl in the northeastern region of the province of Quebec and, more particularly, in SLSJ. Prevalence a t birth, carrier rate, inbreeding and kinship, endogamy and distances between places of birth of the obligate carriers were also studied in the SLSJ region. Genealogies were also reconstructed to search for a possible founder effect.
MATERIALS AND METHODS
The places and years of birth of the children with VDDl were extracted from the patient files kept a t the Clinique de PBdiatrie in Chicoutimi and the Chicoutimi Hospital. Seventy-two patients had been diagnosed at the end of 1989. The diagnosis was established clinically ; it was confirmed by measurement of blood 1,25(OH)2D in 26 patients. The places of residence of the parents (obligate carriers) a t the time of birth of the VDDl children in SLSJ were obtained from the families or extracted from the SLSJ population register, developed and maintained a t SOREP (Interuniversity Centre for Research on Populations established a t the Universite du Qukbec it Chicoutimi). The places of residence of the parents (grandparents of the VDDl children) at the time of birth of the obligate carriers were also extracted from the population register. The prevalence a t birth was calculated by dividing the total number of newborns later diagnosed as having VDDl by the total number of liveborns during the same period. The carrier rate was estimated using the Hardy-Weinberg equilibrium law. Three control groups were created using the SLSJ population register developed and maintained a t SOREP, as previously described in detail elsewhere (De Braekeleer & Larochelle, 1990). They were used in the study of consanguinity, kinship and the geographical distribution of the places of residence, as previously described (De Braekeleer & Larochelle, 1990). The statistical analyses of the spatial distributions used the Monte Carlo simulations, as previously described (De Braekeleer & Larochelle, 1990). Genealogies of the VDDl probands whose parents were married in SLSJ were automatically reconstructed using MEDIC^, a program developed at SOREP (Simard & Kouladjian, 1986). The coefficients of inbreeding and kinship were determined using the programs CONB and APPB also developed at SOREP. The genealogies of all the probands from the other two regions of northeastern Quebec were reconstructed using various sources such as computerized population registers, marriage repositories and genealogical dictionaries to an average depth of 14 generations, allowing recognition of the founders of the French Canadian population in Europe in the 17th century (Ue Braekeleer et al. 1991 ; De Braekeleer, 1991 b ) . They were then recorded in a computerized genealogical database (BELGE, developed a t UQAC) and analysed using various programs, notably PED BELGE (also developed a t UQAC) which, based on an algorithm determining the closest relationship between individuals a t each generation, identifies the most likely founders in a set of families with a given disorder (De Braekeleer, 1991b). A founder was defined as an
Population genetics of VDDl
Tern! scam ingue 't
"i Outaouais '-.**
--- - -
Interprovincial l i m i t s
Fig. 1. Spatial distribution of the places of residence of the parents a t the time of birth of the 72 patients with vitamin D-dependent rickets.
individual born outside Quebec (usually in France) who contributed to the genealogy. The methodology of genealogical reconstruction and analysis has been described in full detail elsewhere (De Braekeleer et al. 1991 ; De Braekeleer, 1991 b ) .
Fifty-four VDDl families living in or originating from the northeastern region of Quebec have now been identified. The geographical distribution of the places of residence of the parents a t the time of birth of the 72 VDDl patients distributed in these 54 families is shown in Fig. 1. Fifty patients were born in the SLSJ region, 9 in Charlevoix, and 8 on the Haute C6te Nord. The places of residence of the parents a t the time of birth of the 50 patients from SLSJ were distributed in 14 of the 66 municipalities within the region. Since, in some families, there was more than one affected individual, the Monte Carlo simulations were performed on the places of residence of the parents a t the time of birth of the 35 probands. They showed four municipalities to have a higher number of places of residence than expected ( P < 0 0 5 ) . The prevalence, a t birth, of VDDl in the SLSJ between 1970 and 1987 was calculated to be 30/87471 live births or 1/2916. The carrier rate was estimated to be 1 in 27 inhabitants by using the Hardy-Weinberg law. The geographical distribution of the places of residence in SLSJ of the parents of the obligate carriers (that is the grandparents of the probands) is shown in Fig. 2. They were distributed in 31 of the 66 municipalities of the SLS,J region. The Monte Carlo simulations showed 5 municipalities (identified by an asterisk in Fig. 2) to have a higher number of places of residence than expected ( P < 0.05). Sixteen places of residence of the parents a t the time of birth of the obligate carriers were located in Charlevoix and ten on the Haute C6te Nord.
M. DE BRAEKELEER AND J. LAROCHELLE
Fig. 2. Spatial distribution of the places of residence in Saguenay-Lac-St-Jean of the parents of the obligate carriers of the VDDl gene. The asterisk indicates a municipality with a higher number of places of residence than expected ( P < 0 0 5 ) .
Table 1. Distribution of the places of origin of the parents married i n SLSJ of the VDDl patients and those of the control individuals Group VDDI Controls
Both inside SLSJ
One inside and one outside SLSJ
Both outside SLSJ
X = 1.41; P = 0.50 From the same From contiguous municipality municipalities of SLSJ within SLSJ 14
x = 094;P = 0 6 2
From non-contiguous municipalities of SLSJ 15 50
Forty-three couples having had a t least one affected child got married in SLSJ; they were matched to three control groups. Parents of VDDl children and those of control children were equally likely to have been born in the SLSJ region or outside (Table 1). Among parents born in SLSJ, parents of affected children and parents of control children were equally likely to have been born in the same municipality, in contiguous municipalities or in non-contiguous municipalities (Table 1).
Population genetics of VDDl
Table 2. Mean coeficients of inbreeding and kinship in VDDl and control groups in SLSJ Mean coefficient of inbreeding Group VDDl Control 1 Control 2 Control 3
( x 10-4)
5.68 13.63 7'72 I 6.4I
Mean coefficient of kinship ( x 10-4) 716 2.52
of the d V l i t h cent.urq France
Fig. 3. Distribution of the places of origin in France of the founders of the VDDl gene.
The mean coefficients of inbreeding and kinship of the VDDl and control groups are given in Table 2 . The mean coefficient of inbreeding in VDDl was not found to be higher than the average value of the mean coefficients of inbreeding of the three control groups. There was only one marriage between second-degree cousins in the VDDl group. The mean kinship coefficient was found to be 2.5 times higher in the VDDl group than the average value of the mean kinship coefficients of the three control groups pooled together. This increase was due to the fact that 2 probands were related as aunt and nephew, 2 as first-degree cousins, 5 as first-degree cousins once removed, and 9 as second-degree cousins. Genealogical reconstruction was attempted in all 108 VDD 1 obligate carriers. However, the analysis was only performed on 101 genealogies because the genealogical data were too fragmentary in 5 genealogies and there had been adoptions in the remaining 2. No founder
M. DE BRAEKELEER AND J. LAROCHELLE
common to all 101 obligate carriers was found. Indeed, one carrier was of Acadian origin and her parents originated from New Brunswick and Nova Scotia, two Canadian provinces situated east of Quebec. A set of 17 founders was found to be common to all other 100 VDDl carriers. All these founders originated from France, including 1 from Aunis, 3 from Normandie, and 10 from Perche (Fig. 3), the remaining 3 founders being of unknown origin. The analysis of these 100 VDDl genealogies showed only one common ancestor to be born in Quebec in 1624, whereas all the other common ancestors were indeed considered as founders, because they were born in France. However, it should be remembered that there are many pitfalls in genealogical reconstruction (adoption, non-paternity, missing and false links).
Vitamin D-dependent rickets is an autosomal recessive disorder that has been rarely reported (Stoop et al. 1967; Dommergues et al. 1978; Karpouzas et al. 1979; Bravo et al. 1986). The disorder was reported to be unusually frequent in SLSJ (Bouchard et al. 1984, 1985). Indeed, based on cases born between 1970 and 1987, the gene frequency was estimated to be 0.037. The mean inbreeding coefficient for the VDDl families from the SLSJ region was low. In particular, close inbreeding was only found in 1 of the 43 families ascertained in this region. No close inbreeding was found in the 11 families originating from Charlevoix and the Haute CGte Nord, also suggesting that the gene frequency in these two regions must be high. The mean kinship coefficient was elevated in the VDDl group from SLSJ, but the most represented founder couple was found in only 5 of the 43 genealogies. Therefore, the VDDl gene must have been introduced by quite a large number of carriers. Although it is difficult to determine the number of carriers immigrating to the SLSJ region, the genealogical reconstruction showed that the most likely carriers came from Charlevoix. Since the migration from Charlevoix to SLSJ was of the familial type, a large number of children carriers of the VDDl gene may have entered the SLSJ region with their parents (Roy et al. 1988; Gauvreau et al. 1991). The presence of several VDDl families on the Haute Cbte Nord is the result of the immigration of families mainly coming from Charlevoix and SLSJ in the 19th century. All 101 genealogies, but one, traced back to a set of founders who settled in Nouvelle France in the 17th century. The only VDDl carrier who does not share these founders is of Acadian extraction. Indeed, several VDDl Acadian families are known; they mainly reside in the province of New Brunswick (Dr Glorieux, personal communication). Since only 1 common ancestor to the 100 remaining VDDl carriers did not originate in Europe, it is likely that the VDDl gene was brought to New France in the 17th century. Indeed, the disorder has been reported, although rarely, in western Europe (Stoop et al. 1967; Dommergues et al. 1978; Karpouzas et al. 1979; Bravo et al. 1986). The present-day French Canadian population of Quebec is mainly issued from some 8500 migrants who settled in the St Lawrence valley in the 17th and 18th centuries (Charbonneau et al. 1987; Charbonneau & Robert, 1987). The great majority of these migrants came from the Ile de France (Paris) and the western provinces of France (mainly Normandie, Poitou, Aunis and Saintonge) (Charbonneau & Robert, 1987). Although the province of Perche, a small region located between Normandie and Ile de France, only contributed 217 of the 8500 founders to the French Canadian population, it contributed 10 founders common to all 100 VDDl carriers. The
Population genetics of VDDl Monte Carlo simulations showed that the contribution of Perche to the VDDl carriers was higher than expected from its relative weight of founders to the French Canadian population ( P < 0.001). Therefore, it would appear that the province of Perche might have been the most likely centre of diffusion of the VDDl gene for the French Canadian population in the 17th century. However, the contribution of Perche is significantly higher than expected (P< 0-05) in all 10 hereditary disorders from eastern Quebec we have studied so far (De Braekeleer, 1 9 9 1 ~ )Ongoing . studies seem to show that, a t least in some disorders, the contribution of Perche may have acted as a confounding factor. Migrants from Perche in the 17th century were among the first to settle in Nouvelle France; their descendance in 1730 was the highest, compared with settlers from other provinces of France (Charbonneau et al. 1987 ; De Braekeleer 1 9 9 1 ~ ) Furthermore, . migrants from Perche used to arrive in large families whereas the migration from other regions of France was mainly the fact of isolated individuals (Charbonneau et al. 1987; De Braekeleer 1 9 9 1 ~ ) . RFLPs (Dl2814,D12Sl7, D12S6) linked to the VDDl locus have now been identified on the long arm of chromosome 12 (Labuda et al. 1990). A linkage disequilibrium was also noted between these RFLPs and the VDDI locus. Indeed, 11 of the 16 (69%) VDDI chromosomes so far characterized bore the same three-marker haplotype ; the linkage disequilibrium was even stronger among the VDDI chromosomes from the SLSJ and Charlevoix regions (11/13 85 YO) (Labuda et al. 1990). The genealogical reconstruction and the strong linkage disequilibrium suggest that the high incidence of VDDl in northeastern Quebec is likely to be the result of a founder effect (Bouchard et al. 1984; De Braekeleer, 1991a). It also suggests that a unique mutation accounts for most, if not all, of the VDDl cases known in these regions. ~
The authors thank Drs Charles R. Scriver and Francis H. Glorieux for their advice, assistance and encouragement. This study was supported in part by grants from the Fondation de I’UniversitB du Quebec a Chicoutimi, the Fonds de Recherche en Santd du QuBbec (FRSQ) and the Fonds pour 1’Aide a la Recherche et aux Chercheurs (FCAR). REFERENCES
ARNAUD,C., MAIJER, R., READE,T., SCRIVER,C. R. & WHELAN, D. T. (1970). Vitamin D dependency: An inherited postnatal syndrome with secondary hyperparathyroidism. Pediatrics 46,87 1-879. BOUCHARD, G., LABERGE, C., SCRIBER, C. R., GLORIEUX, F. H., DECLOS,M., BERGERON, L., LAROCHELLE, J. & MORTEZAI, S. (1984). Etude dkmographique et gdndalogique de deux maladies hdrdditaires au Saguenay. Cahiers Que‘becois de De‘mographie 13, 118-137. BOUCHARD, G., LABERGE, C. & SCRIBER,C. R. (1985). La tyrosindmie hBrdditaire et le rachitisme vitaminodependant au Saguenay. Union M i d . Canada 114, 633-636. BRAVO,H., ALMEIDA, S., TATO,I. G., BUSTILLO,J . M . & TOJO, R. (1986). Early pseudo-deficiency or Prader’s hypocalcemia familial type I rickets. A n . Esp. Pediatr. 25, 121-124. CHARBONNEAU, H. & ROBERT,N. (1987). The French origins of the Canadian population, 1608-1759. In Historical Atlas of Canada. Vol. 1 : From the beginning to 2800 (ed. R. C. Harris), plate 45. Toronto: University of Toronto Press. CHARBONNEAU, H., DESJARDINS, B., GUILLEMETTE, A., LANDRY, Y., LEGARE, J. & NAULT, F. (1987). Pu’aissance d’une population. Les Franqais dtablis au Canada au XVII‘ sikcle. Paris : Institut National d’Etudes DBmographiques. DE BRAEKELEER, M. (1991 a ) . Hereditary disorders in Saguenay-Lac-St-Jean (Quebec, Canada). Hum. Hered. (in press). M. (1991b). BELGE : U n ensemble d’analyse gBn6tique des gBn6alogies. Chicoutimi, SOREP, DE BRAEKELEER, Universite du Quebec a Chicoutimi. DE BRAEKELEER, M. (1991c). Le Perche : une province d’importance pour la population canadienne-franpaise. Annales de 1’ACFAS (in press). M. & LAROCHELLE, J. (1990). Genetic epidemiology of hereditary tyrosinemia in Quebec and DE BRAEKELEER, in Saguenay-Lac-St-Jean. Am. J. Hum. Genet. 47, 302-307.
M. DE BRAEKELEER AND J. LAROCHELLE
DE BRAEKELEER, M., DIONNE,C., GAQNE,C., JULIEN, P., BRUN,D., VEN MURTHY,M. R. & LUPIEN,P. J. (1991). Founder effect in familial hyperchylomicronemia among French Canadians of Quebec. Hum. Hered. (in press). E. E., GLORIEUX, F. H., MARIE, P. J. & PETTIFOR, J . M. (1981). Vitamin D dependency: replacement DELVIN, therapy with calcitriol. J. Pediatr. 99, 2G34. DOMMERGUES, J. P., GARABEDIAN, M., GUERIS,J., LEDEUNFF, M. J., CREIQNOU, L., COURTECUISSE,V. & BALSAN, S. (1978). Effets des principaux deriv6s de la vitamine D : chez trois enfants d’une fratrie atteints de rachitisme ‘pseudo-carentiel ’. Arch. Fr. Pkdiatr. 35, 105Ck1057. FRASER, D., KOOH,S. W., KIND,H. P., HOLICK, M. F., TANAKA, Y. & DELUCA,H. F . (1973). Pathogenesis of hereditary vitamin D dependent rickets. An inborn error of vitamin D metabolism involving defective conversion of 25-hydroxyvitamin D to l-alpha,25-dihydroxyvitaminD. N. Engl. J. Med. 289, 817-822. GAUVREAU, D., GUERIN,M. & HAMEL,M. (1991). De Charlevoix au Saguenay: mesure et caractkristiques du mouvement migratoire avant 1911. In Histoire d’un gknome :population et gknktique dans l’est du Qukbec (ed. G. Bouchard and M. De Braekeleer), pp. 145-150. Sillery: Presses de 1’Universitk du Qukbec. KARPOUZAS, J., PAPATHANASIOU-KLONTZA, D., XIPOLITA-ZACHARIADU, A., BENETOS, S. & MATSANIOTIS, N. (1979). Pseudo-vitamin D deficiency rickets: report of a case. Helw. Paediatr. Acta 34, 461464. LABUDA, M., MORGAN,K. & GLORIEUX, F. H. (1990). Mapping autosomal recessive vitamin D dependency type 1 to chromosome 12q14 by linkage analysis. A m . J . Hum. Genet. 47, 28-36. ROY, R., BOUCHARD, G. & DECLOS,M. (1988). La premiere generation de Saguenayens: Provenance, apparentement, enracinement. Cahiers Qukbecois de Debwgraphie 17, 113-134. SCRIVER, C. R. (1970). Vitamin D dependency. Pediatrics 45, 361-363. H. F. & HAMSTRA, A. J . (1978). Serum 1,25-dihydroxyvitamin D levels SCRIVER, C. R., READE,.T.M., DELUCA, in normal subjects and in patients with hereditary rickets or bone disease. N. Engl. J. Med. 299, 976979. K. (1986). Description du programme MEDIC4 de construction automatique des SIMARD, A. & KOULADJIAN, g6nkalogies. Chicoutimi, SOREP, Universitk du Qu6bec a Chicoutimi. M. J . C. & TIDDENS, H. A. W. M. (1967). Pseudo vitamin D deficiency rickets. Acta STOOP,J. W., SCHRAAGEN, Paediat. S c a d . 56, 607-613.