- Email: [email protected]
Refinement of the Locus for Autosomal Dominant Hereditary Gingival Fibromatosis (GINGF) to a 3.8-cM Region on 2p21
Genomics 68, 247–252 (2000) doi:10.1006/geno.2000.6285, available online at http://www.idealibrary.com on
Refinement of the Locus for Autosomal Dominant Hereditary Gingival Fibromatosis (GINGF) to a 3.8-cM Region on 2p21 Shangxi Xiao,* ,† ,‡ Xiurong Wang,§ Bingyin Qu, ¶ Minghua Yang,㛳 Guiyu Liu,‡ Lei Bu,* Ying Wang,** Liqin Zhu,* Hao Lei,* Landian Hu,* Xuejun Zhang,‡ Jing Liu,† Guoping Zhao,* and Xiangyin Kong* ,1 *Shanghai Research Center of Biotechnology, Chinese Academy of Science, Shanghai 200233, People’s Republic of China; †University of Science and Technology of China, Hefei, People’s Republic of China; ‡The First Affiliated Hospital of Anhui Medical University, Hefei, People’s Republic of China; §Central Hospital of Xiangfan City, Xiangfan, People’s Republic of China; 㛳Stomatological Hospital of Nanjing City, Nanjing, People’s Republic of China; ¶Stomatological Hospital of Guiyang City, Guiyang, People’s Republic of China; and **The Forth Military Medical University, Xian, People’s Republic of China Received March 24, 2000; accepted June 12, 2000
Hereditary gingival fibromatosis (HGF, MIM 135300; approved gene symbol GINGF) is an oral disease characterized by enlargement of gingiva. Recently, a locus for autosomal dominant HGF has been mapped to an 11-cM region on chromosome 2p21. In the current investigation, we genotyped four Chinese HGF families using polymorphic microsatellite markers on 2p21. The HOMOG test provided evidence for genetic homogeneity, with evidence for linkage in four families (heterogeneity versus homogeneity test HOMOG, 2 ⴝ 0.00). A cumulative maximum two-point lod score of 5.04 was produced with marker D2S390 at a recombination frequency of ⴝ 0 in the four linked families. Haplotype analysis localized the hereditary gingival fibromatosis locus within the region defined by D2S352 and D2S2163. This region overlaps by 3.8 cM with the previously reported HGF region. Singlestrand conformation polymorphism and sequence analysis of the coding region of cytochrome P450 1B1 (CYP1B1) excluded it as a likely candidate gene. © 2000 Academic Press
INTRODUCTION
Gingival fibromatosis is a benign disorder manifested by enlargement of gingiva. The overgrowth of gingiva associates with increased proliferation of fibroblasts and the increased production of collagen and other extracellular matrix molecules (Noyan et al., 1994; Tipton et al., 1997). Cytokines, such as autocrine transforming growth factor , play a role in stimulating the production of collagen and extracellular matrix molecules by fibroblasts (Tipton and Dabbous, 1998; Hong et al., 1999). Gingival fibromatosis consists of a group of diseases with genetic heterogeneity (Witkop, 1 To whom correspondence should be addressed. Telephone: 86-2164700892. Fax: 86-21-64700244. E-mail: [email protected].
1971; Shashi et al., 1999). It is expressed either as an isolated condition or as a feature of some rare or unrelated syndromes. The reported isolated forms are transmitted mainly as an autosomal dominant trait. An autosomal recessive form has also been reported (Goldblatt and Singer, 1992; Singer et al., 1993). In addition, gingival fibromatosis often appears as one trait in Zimmermann–Laband, Murray–Puretic– Drescher, Rutherfurd, Cowden, and Cross syndromes. Recently, an autosomal dominant gingival fibromatosis locus, GGF1, was mapped to a 37-cM interval on 2p21–p22 in a Brazilian family (Hart et al., 1998). More recently, this region has been further narrowed to an 11-cM interval bounded by D2S1788 and D2S2298 (Shashi et al., 1999). Approximately 14 known genes and more than 100 Unigene clusters have been mapped to this region. Since calcium channel blockers induce the development of gingival fibromatosis in some people, calmodulin 2 (CALM2) and sodium calcium exchanger (NCX1) were suggested as two candidate genes (Hart et al., 1998). Additionally, CYP1B1, an enzyme that oxidizes a variety of structurally unrelated compounds, localizes to this region (Shashi et al., 1999). A boy with a 2p13–p21 duplication also showed gingival overgrowth (Fryns, 1996). It is thus speculated that the duplicated region contains a gene (GGF2) responsible for the gingival overgrowth trait (Fryns, 1996). This region was shown to be proximal to the GGF1 locus region, indicating the presence of another gene involving this disease (Shashi et al., 1999). In this study, we obtained four hereditary gingival fibromatosis families, in which gingival fibromatosis was inherited as an autosomal dominant trait, from four provinces. To map the loci of hereditary gingival fibromatosis in the Chinese population and to refine the previously mapped region, we typed these four families using polymorphic markers on 2p21. In these
247
0888-7543/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
248
XIAO ET AL.
FIG. 1. Haplotype analysis of families I, II, III, and IV for 2p21 markers. Affected and unaffected individuals are represented by black and open symbols, respectively. Disease-gene bearing chromosomes are indicated with boxes, and filled black areas represent the region of the disease gene.
four families, the gingival fibromatosis locus was located to an approximately 8.7-cM region on 2p21. It overlaps by 3.8 cM with the GGF1 locus region. 2 MATERIALS AND METHODS Clinical data and sample collection. Four hereditary gingival fibromatosis families were identified through probands from four
2 The HGMW-approved symbol for the gene described in this paper is GINGF.
provinces in China. All family members received careful oral examination by experienced clinical dentists. In these families, affected individuals had no history of exposure to gingival fibromatosis inducible drugs and no evidence of progressive deafness, hypertrichosis, or distinctive faces. Peripheral blood samples were obtained from available family members. Genomic DNA was extracted using a Qiagen kit following the manufacturer’s recommended protocol. Genotyping. Highly informative polymorphic microsatellite markers were selected from 2p21. Primer sequences were obtained from the Genome Database (http://gdbwww.gdb.org/). PCR amplifications were carried out following the Licor Co. manual using a PTC-225 DNA Engine Tetrad (MJ Research. Inc.) and a touchdown
REFINEMENT OF GINGF LOCUS TO A 3.8-cM REGION ON 2p21
249
FIG. 1—Continued
program modified by the annealing temperature of each primer set. PCR was carried out in a 10-l volume containing 20 ng genomic DNA template, a 2.0 mM concentration of each dNTP, 1.0 pmol of each M13-tailed primer, 1.0 pmol of primer, 1 pmol of labeled M13 primer, 1.5 mM MgCl 2, 10 mM Tris–HCl, 1.0 units of Taq DNA polymerase (Perkin–Elmer Corp.). The sample was initially dena-
tured at 95°C for 8 min, followed by four cycles of denaturation at 95°C for 45 s, annealing at 68°C for 5 min, a four-cycle drop, 2°C/ cycle to 60°C, and elongation at 72°C for 1 min and then by a second set of four or two cycles of denaturation at 95°C for 45 s, annealing at 58°C for 2 min, a 4- or 2-cycle drop of 2°C/cycle to 50 or 54°C, and elongation at 72°C for 1 min, and then 20 to 25 cycles of denaturation
250
XIAO ET AL.
TABLE 1 Two-Point Linkage Analysis between the HGF Locus and the Chromosome 2p21 Markers Lod score at ⫽
Maximum
Marker
Location (cM)
n
het
0.0
0.01
0.05
0.1
0.2
0.3
0.4
Lod
D2S2221 D2S2144 D2S390 D2S352 D2S1788 D2S2230 D2S1346 D2S1356 D2S119 D2S1352 D2S441
44.09 46.37 48.50 50.65 55.51 56.15 59.36 64.29 65.61 73.61 92.61
9 8 9 5 11 4 5 8 7 4 8
0.7494 0.6431 0.8013 0.6752 0.8388 0.6653 0.6529 0.8099 0.6200 0.5909 0.7851
⫺inf. ⫺inf. 5.04 0.18 4.94 3.17 2.98 ⫺2.63 0.48 ⫺12.92 ⫺12.37
2.56 0.66 4.99 0.22 4.90 3.13 2.98 3.02 2.00 ⫺7.10 ⫺5.64
3.46 1.91 4.73 0.33 4.65 2.96 2.91 3.46 2.56 ⫺3.89 ⫺2.93
3.42 2.19 4.31 0.39 4.21 2.68 2.72 3.36 2.58 ⫺2.47 ⫺1.79
2.66 1.95 3.29 0.37 3.15 2.00 2.14 2.72 2.13 ⫺1.11 ⫺0.73
1.60 1.31 2.11 0.28 1.95 1.23 1.41 1.80 1.41 ⫺0.41 ⫺0.22
0.56 0.57 0.89 0.15 0.75 0.44 0.64 0.76 0.58 ⫺0.07 0.00
3.46 2.19 5.05 0.39 4.94 3.17 2.98 3.46 2.58 ⫺0.07 0.00
0.05 0.10 0.00 0.10 0.0 0.0 0.0 0.05 0.1 0.4 0.4
Note. Genetic coordinates in centimorgans according to the Marshfield genetic map are in the column “Location.” n gives the total number of different alleles encountered in the pedigree set; het gives the estimated heterozygosity value. Marker loci showing no recombination with HGF are in boldface type and represent the cosegregating segment. at 95°C for 30 s, annealing at 50 or 54°C for 30 s, and elongation at 72°C for 30 s. Finally the sample was elongated for 15 min at 72°C. PCR products were electrophoresed on a Licor 4200L DNA sequencer (Licor Corp.) on 6% denaturing polyacrylamide gels with fluorescent dye-labeled DNA markers (Licor Corp.). Data were collected and analyzed with Base Image 4.1 and Gene Image 3.12 software (Licor Corp.) and were checked by an experienced operator. Linkage ready pedigree files were prepared using Gene Image software. Linkage and haplotype analysis. The hereditary gingival fibromatosis locus was modeled as an autosomal dominant, two-allele system with 100% penetrance. The affected allele frequency was set to 0.0001. Marker allele frequencies were obtained from our four families’ genotyping data. Two-point linkage analysis was performed using the MLINK and ILINK programs with LINKAGE version 5.10 (Lathrop et al., 1984). Multipoint analyses were conducted by FASTLINK Version 4.1P (Cottingham et al., 1993; Schaffer et al., 1994). Genetic heterogeneity was tested by using the HOMOG program (Ott, 1991). Marker allele frequency calculation and haplotype construction were performed using the program SimWalk2 version 2.31 with the aid of Cyrillic Version 2.0.2 software (Sobel and Lange, 1996). Radiation hybrid mapping. Oligonucleotide primers for possible genes within the HGF critical region were synthesized based on the sequence of each gene. These primers were used to score the highresolution Stanford TNG radiation hybrid panel. Statistical analysis of the RH data was carried out on the mapping server at the Stanford Genome Center (http://shgc-www.stanford.edu/RH/rhserverformnew. html). Marker order was determined with Rhminbrk Version 3.0. Mutation analysis of CYP1B1. A set of oligonucleotide primers was designed covering the whole coding region of CYP1B1 (GenBank Accession No. U56438) using the primer 3.0 program (http://wwwgenome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi). These primers were used to perform PCR-single-strand conformation polymorphism (SSCP) assays for the CYP1B1 gene. The PCR products were run on 10% polyacrylamide gels and silver stained according to standard methods.
RESULTS
Linkage and Haplotype Analysis In our investigation, we typed all available individuals of the four HGF families with polymorphic mark-
ers on 2p21 (Fig. 1). HOMOG program analysis (A test) supported homogeneity of these four families (␣ ⫽ 1.00, 2 ⫽ 0.00). Two-point linkage analysis of these four families yielded a summed maximum lod score of 5.04 with D2S390 at ⫽ 0 (Table 1). The recombination events in individual 19 of family I placed the HGF locus distal to D2S119 (Fig. 1). The crossover in individual 8 of family III defines the HGF locus distal to D2S1346 (Fig. 1). To localize the HGF locus more precisely in family III, we typed this family with two additional polymorphic markers, namely D2S2186 and D2S2163, respectively. The paternal origin haplotype limited the HGF locus distal to D2S2163. A crossover in individual 22 of family I determined the distal boundary to D2S352. Our data thus suggest that the HGF gene lies in an 8.7-cM interval flanked by D2S352 and D2S2163. We conducted three-point analyses with adjacent marker loci and the disease locus. Multipoint analysis with chromosome 2p21 markers produced a maximum lod score of 6.61 at D2S2230. Scanning for Mutations in the CYP1B1 Gene by SSCP Analysis The entire coding region of CYP1B1 was screened for mutations with eight pairs of primers (see Table 2). Genomic DNA from four affected individuals and that from four unaffected individuals from family I was used as templates for PCR amplification. To assess the SSCP system, a positive control and a negative control were included. The results showed no mutations in the coding region in affected individuals, although two polymorphisms were detected with primer set CYP1B1-2 and CYP1B1-7. Although they cosegregated with the HGF trait in family I, they also appeared in unaffected individuals (data not shown).
REFINEMENT OF GINGF LOCUS TO A 3.8-cM REGION ON 2p21
TABLE 2 Primer Sequences Flanking the CYP1B1 Coding Region for Mutation Screening Using Genomic DNA Primer CYP1B1-1 CYP1B1-1 CYP1B1-2 CYP1B1-2 CYP1B1-3 CYP1B1-3 CYP1B1-4 CYP1B1-4 CYP1B1-5 CYP1B1-5 CYP1B1-6 CYP1B1-6 CYP1B1-7 CYP1B1-7 CYP1B1-8 CYP1B1-8
forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse
Sequence 5⬘-CACGCCTTCTCCTTTCTGTC-3⬘ 5⬘-ATCAGTGGCCACGCAAAC-3⬘ 5⬘-GCTGGCCACTGTGCATGT-3⬘ 5⬘-GGACACCACACGGAAGGA-3⬘ 5⬘-CCTCCTTCCGTGTGGTGT-3⬘ 5⬘-CGAAACACACGGCACTCAT-3⬘ 5⬘-GTCATGAGTGCCGTGTGTTT-3⬘ 5⬘-TGTCCAGGATGAAGTTGCTG-3⬘ 5⬘-AACCGCAACTTCAGCAACTT-3⬘ 5⬘-GCCTCCCAGAGGCTTTACC-3⬘ 5⬘-GCTCACTTGCTTTTCTCTCTCC-3⬘ 5⬘-CAGTGGTGGCATGAGGAATA-3⬘ 5⬘-GCCTGTCACTATTCCTCATGC-3⬘ 5⬘-TCATCACTCTGCTGGTCAGG-3⬘ 5⬘-AGGACCTGACCAGCAGAGTG-3⬘ 5⬘-CAGCTTGCCTCTTGCTTCTT-3⬘
Size (bp) 205 295 270 214 269 210 200 244
Radiation Hybrid Mapping In our study, we positioned the HGF locus distal to D2S2163, thus forming an overlapping region in the interval between D2S1788 and D2S2163 with the GGF1 locus region. This region is about 2.8 Mb, containing five known genes, namely, fasciculation and elongation protein zeta 2 (FEZ2), interferoninducible double-stranded RNA-dependent protein kinase (PRKR), protein kinase C nu (PRKCN), breakpoint cluster region protein 1 (BCRP1), CCAAT-box-binding transcription factor (CBF2), and 11 Unigene clusters. Three previously suggested candidate genes, CYP1B, NCX1, and Cdc42 effector protein 3 (CEP3), localize to the centromeric boundary of the HGF locus (http://bioinformatics.weizmann.ac.il/udb/). Splicing factor arginine/serine-rich 7 (SFRS7) was previously mapped to the interval of D2S1788 and CYP1B1. All genes within the critical region may be implicated in HGF disease. To map these genes more precisely, primers were designed to score high-resolution RH Panel TNG. These results show that the candidate locus order is as follows: CALM2–FEZ2–D2S1788 –CBF2–SFRS7–CYP1B1– CEP3–D2S2163–NCX1. DISCUSSION
Gingival fibromatosis is a heterogeneous disease. So far, two gingival fibromatosis loci have been mapped, GGF1 on 2p21 and GGF2 on 2p13 (Hart et al., 1998; Shashi et al., 1999; Fryns, 1996). In this study, we mapped a dominant hereditary gingival fibromatosis locus to a region defined by D2S352 and D2S2163 in Chinese families. This region overlaps with the recently reported GGF1 region (Hart et al., 1998). The overlapping region is about 3.8 cM from D2S1788 to D2S2163 and represents a physical re-
251
gion of 2.8 Mb. This indicates the existence of a common gingival fibromatosis locus in this region between the Brazilian family and the Chinese families. Among families II, III, IV, and I, no common haplotype was found in affected individuals, suggesting the different mutation origins. Since gingival fibromatosis can potentially be induced by calcium channel blockers, two genes within the previously mapped HGF locus region, CALM2 and NCX1, were suggested to be HGF candidate genes. However, high-resolution radiation mapping showed that CALM2 and NCX1 were outside of the HGF critical region. Thus, they are unlikely to be the genes responsible for gingival fibromatosis. Radiation hybrid panel analysis placed CBF2, SFRS7, CYP1B1, and CEP3 in the critical region. Based on the sequences of contigs NT-002229, NT-002723, and NT-002803 (http://www.ncbi.nlm.nih.gov/), FEZ2, PRKR, and PRKCN may also localize to the critical region. Because CYP1B1 plays an important role in the metabolism of a wide range of drugs, it was selected as the first gene to use in performing mutation detection in affected individuals. Mutation analysis of the coding region of CYP1B1 in family I revealed no obvious mutations although two polymorphisms were revealed. The polymorphic alleles cosegregated with the HGF phenotype. However, they also appear in normal control individuals. These results, however, do not exclude CYP1B1 mutations in other families or in noncoding regions. In gingival fibromatosis, the gingiva fibroblast is activated by some autocrines (Tipton and Dabbous, 1998; Hong et al., 1999), so all fibroblast expressed genes within the critical region may be included as candidates. We have not yet examined the expression profiles of other genes within the critical region. Based on the Cancer Genome Anatomy Project Database, at least PRKR is expressed in gingiva tissue (http://www.ncbi.nlm.nih.gov/ncicgap/). The fast progress of human genome sequencing will certainly aid in the cloning of gingival fibromatosis genes. In part of the population, some pharmacological agents such as phenytoin and cyclosporin can also induce gingival fibromatosis. The patients may carry one or more susceptibility genes, which have some defect in metabolizing or interacting with gingival fibromatosis inducible drugs. So far, however, no such susceptibility loci have been identified. In the Brazilian family, recombination suppression was observed between the polymorphic markers D2S1788 and D2S441 (Hart et al., 1998). In contrast, in the present study, our results did not support recombination suppression near the gingival fibromatosis locus. However, before the mutated genes are identified in these families of different racial origins, we cannot exclude the possibility that the controversial results are caused by allelic mutations.
252
XIAO ET AL.
ACKNOWLEDGMENTS We thank all members of gingival fibromatosis families for providing their blood samples. We are grateful to Dr. Fiona Francis for critically reading the manuscript. We gratefully acknowledge Dr. Gengxi Hu for the normal individual DNA samples. This work was supported by Life Science Special Fund of CAS supported by the Ministry of Finance (Grant KJ95T-06 to X.K.).
REFERENCES Cottingham, R. W., Jr., Idury, R. M., and Schaffer, A. A. (1993). Faster sequential genetic linkage computations. Am. J. Hum. Genet. 53: 252–263. Fryns, J. P. (1996). Gingival fibromatosis and partial duplication of the short arm of chromosome 2 (dup(2)(p13 3 p21)). Ann. Genet. 39: 54 –55. Goldblatt, J., and Singer, S. L. (1992). Autosomal recessive gingival fibromatosis with distinctive facies. Clin. Genet. 42: 306 –308. Hart, T. C., Pallos, D., Bowden, D. W., Bolyard, J., Pettenati, M. J., and Cortelli, J. R. (1998). Genetic linkage of hereditary gingival fibromatosis to chromosome 2p21. Am. J. Hum. Genet. 62: 876 – 883. Hong, H. H., Uzel, M. I., Duan, C., Sheff, M. C., and Trackman, P. C. (1999). Regulation of lysyl oxidase, collagen, and connective tissue growth factor by TGF-1 and detection in human gingiva. Lab. Invest. 79: 1655–1667. Lathrop, G. M., Lalouel, J. M., Julier, C., and Ott, J. (1984). Strat-
egies for multilocus linkage analysis in humans. Proc. Natl. Acad. Sci. USA 81(11): 3443–3446. Noyan, U., Yilmaz, S., Arda, O., and Kuru, B. (1994). The ultrastructural examination of gingival fibromatosis. J. Marmara. Univ. Dent. Fac. 2: 409 – 413. Ott, J. (1991). “Analysis of Human Genetic Linkage,” 2nd ed., Johns Hopkins Univ. Press, Baltimore, MD. Schaffer, A. A., Gupta, S. K., Shriram, K., and Cottingham, R. W., Jr. (1994). Avoiding recomputation in linkage analysis. Hum. Hered. 44: 225–237. Shashi, V., Pallos, D., Pettenati, M. J., Cortelli, J. R., Fryns, J. P., von Kap-Herr, C., and Hart, T. C. (1999). Genetic heterogeneity of gingival fibromatosis on chromosome 2p. J. Med. Genet. 36: 683– 686. Singer, S. L., Goldblatt, J., Hallam, L. A., and Winters, J. C. (1993). Hereditary gingival fibromatosis with a recessive mode of inheritance. Case reports. Aust. Dent. J. 38: 427– 432. Sobel, E., and Lange, K. (1996). Descent graphs in pedigree analysis: Applications to haplotyping, location scores, and marker-sharing statistics. Am. J. Hum. Genet. 58: 1323–1337. Tipton, D. A., and Dabbous, M. K. (1998). Autocrine transforming growth factor beta stimulation of extracellular matrix production by fibroblasts from fibrotic human gingiva. J. Periodontol. 69: 609 – 619. Tipton, D. A., Howell, K. J., and Dabbous, M. K. (1997). Increased proliferation, collagen, and fibronectin production by hereditary gingival fibromatosis fibroblasts. J. Periodontol. 68: 524 –530. Witkop, C. J., Jr. (1971). Heterogeneity in inherited dental traits, gingival fibromatosis and amelogenesis imperfecta. South. Med. J. 64(Supp. 11): 16 –25.
Refinement of the Locus for Autosomal Dominant Hereditary Gingival Fibromatosis (GINGF) to a 3.8-cM Region on 2p21 Shangxi Xiao,* ,† ,‡ Xiurong Wang,§ Bingyin Qu, ¶ Minghua Yang,㛳 Guiyu Liu,‡ Lei Bu,* Ying Wang,** Liqin Zhu,* Hao Lei,* Landian Hu,* Xuejun Zhang,‡ Jing Liu,† Guoping Zhao,* and Xiangyin Kong* ,1 *Shanghai Research Center of Biotechnology, Chinese Academy of Science, Shanghai 200233, People’s Republic of China; †University of Science and Technology of China, Hefei, People’s Republic of China; ‡The First Affiliated Hospital of Anhui Medical University, Hefei, People’s Republic of China; §Central Hospital of Xiangfan City, Xiangfan, People’s Republic of China; 㛳Stomatological Hospital of Nanjing City, Nanjing, People’s Republic of China; ¶Stomatological Hospital of Guiyang City, Guiyang, People’s Republic of China; and **The Forth Military Medical University, Xian, People’s Republic of China Received March 24, 2000; accepted June 12, 2000
Hereditary gingival fibromatosis (HGF, MIM 135300; approved gene symbol GINGF) is an oral disease characterized by enlargement of gingiva. Recently, a locus for autosomal dominant HGF has been mapped to an 11-cM region on chromosome 2p21. In the current investigation, we genotyped four Chinese HGF families using polymorphic microsatellite markers on 2p21. The HOMOG test provided evidence for genetic homogeneity, with evidence for linkage in four families (heterogeneity versus homogeneity test HOMOG, 2 ⴝ 0.00). A cumulative maximum two-point lod score of 5.04 was produced with marker D2S390 at a recombination frequency of ⴝ 0 in the four linked families. Haplotype analysis localized the hereditary gingival fibromatosis locus within the region defined by D2S352 and D2S2163. This region overlaps by 3.8 cM with the previously reported HGF region. Singlestrand conformation polymorphism and sequence analysis of the coding region of cytochrome P450 1B1 (CYP1B1) excluded it as a likely candidate gene. © 2000 Academic Press
INTRODUCTION
Gingival fibromatosis is a benign disorder manifested by enlargement of gingiva. The overgrowth of gingiva associates with increased proliferation of fibroblasts and the increased production of collagen and other extracellular matrix molecules (Noyan et al., 1994; Tipton et al., 1997). Cytokines, such as autocrine transforming growth factor , play a role in stimulating the production of collagen and extracellular matrix molecules by fibroblasts (Tipton and Dabbous, 1998; Hong et al., 1999). Gingival fibromatosis consists of a group of diseases with genetic heterogeneity (Witkop, 1 To whom correspondence should be addressed. Telephone: 86-2164700892. Fax: 86-21-64700244. E-mail: [email protected].
1971; Shashi et al., 1999). It is expressed either as an isolated condition or as a feature of some rare or unrelated syndromes. The reported isolated forms are transmitted mainly as an autosomal dominant trait. An autosomal recessive form has also been reported (Goldblatt and Singer, 1992; Singer et al., 1993). In addition, gingival fibromatosis often appears as one trait in Zimmermann–Laband, Murray–Puretic– Drescher, Rutherfurd, Cowden, and Cross syndromes. Recently, an autosomal dominant gingival fibromatosis locus, GGF1, was mapped to a 37-cM interval on 2p21–p22 in a Brazilian family (Hart et al., 1998). More recently, this region has been further narrowed to an 11-cM interval bounded by D2S1788 and D2S2298 (Shashi et al., 1999). Approximately 14 known genes and more than 100 Unigene clusters have been mapped to this region. Since calcium channel blockers induce the development of gingival fibromatosis in some people, calmodulin 2 (CALM2) and sodium calcium exchanger (NCX1) were suggested as two candidate genes (Hart et al., 1998). Additionally, CYP1B1, an enzyme that oxidizes a variety of structurally unrelated compounds, localizes to this region (Shashi et al., 1999). A boy with a 2p13–p21 duplication also showed gingival overgrowth (Fryns, 1996). It is thus speculated that the duplicated region contains a gene (GGF2) responsible for the gingival overgrowth trait (Fryns, 1996). This region was shown to be proximal to the GGF1 locus region, indicating the presence of another gene involving this disease (Shashi et al., 1999). In this study, we obtained four hereditary gingival fibromatosis families, in which gingival fibromatosis was inherited as an autosomal dominant trait, from four provinces. To map the loci of hereditary gingival fibromatosis in the Chinese population and to refine the previously mapped region, we typed these four families using polymorphic markers on 2p21. In these
247
0888-7543/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
248
XIAO ET AL.
FIG. 1. Haplotype analysis of families I, II, III, and IV for 2p21 markers. Affected and unaffected individuals are represented by black and open symbols, respectively. Disease-gene bearing chromosomes are indicated with boxes, and filled black areas represent the region of the disease gene.
four families, the gingival fibromatosis locus was located to an approximately 8.7-cM region on 2p21. It overlaps by 3.8 cM with the GGF1 locus region. 2 MATERIALS AND METHODS Clinical data and sample collection. Four hereditary gingival fibromatosis families were identified through probands from four
2 The HGMW-approved symbol for the gene described in this paper is GINGF.
provinces in China. All family members received careful oral examination by experienced clinical dentists. In these families, affected individuals had no history of exposure to gingival fibromatosis inducible drugs and no evidence of progressive deafness, hypertrichosis, or distinctive faces. Peripheral blood samples were obtained from available family members. Genomic DNA was extracted using a Qiagen kit following the manufacturer’s recommended protocol. Genotyping. Highly informative polymorphic microsatellite markers were selected from 2p21. Primer sequences were obtained from the Genome Database (http://gdbwww.gdb.org/). PCR amplifications were carried out following the Licor Co. manual using a PTC-225 DNA Engine Tetrad (MJ Research. Inc.) and a touchdown
REFINEMENT OF GINGF LOCUS TO A 3.8-cM REGION ON 2p21
249
FIG. 1—Continued
program modified by the annealing temperature of each primer set. PCR was carried out in a 10-l volume containing 20 ng genomic DNA template, a 2.0 mM concentration of each dNTP, 1.0 pmol of each M13-tailed primer, 1.0 pmol of primer, 1 pmol of labeled M13 primer, 1.5 mM MgCl 2, 10 mM Tris–HCl, 1.0 units of Taq DNA polymerase (Perkin–Elmer Corp.). The sample was initially dena-
tured at 95°C for 8 min, followed by four cycles of denaturation at 95°C for 45 s, annealing at 68°C for 5 min, a four-cycle drop, 2°C/ cycle to 60°C, and elongation at 72°C for 1 min and then by a second set of four or two cycles of denaturation at 95°C for 45 s, annealing at 58°C for 2 min, a 4- or 2-cycle drop of 2°C/cycle to 50 or 54°C, and elongation at 72°C for 1 min, and then 20 to 25 cycles of denaturation
250
XIAO ET AL.
TABLE 1 Two-Point Linkage Analysis between the HGF Locus and the Chromosome 2p21 Markers Lod score at ⫽
Maximum
Marker
Location (cM)
n
het
0.0
0.01
0.05
0.1
0.2
0.3
0.4
Lod
D2S2221 D2S2144 D2S390 D2S352 D2S1788 D2S2230 D2S1346 D2S1356 D2S119 D2S1352 D2S441
44.09 46.37 48.50 50.65 55.51 56.15 59.36 64.29 65.61 73.61 92.61
9 8 9 5 11 4 5 8 7 4 8
0.7494 0.6431 0.8013 0.6752 0.8388 0.6653 0.6529 0.8099 0.6200 0.5909 0.7851
⫺inf. ⫺inf. 5.04 0.18 4.94 3.17 2.98 ⫺2.63 0.48 ⫺12.92 ⫺12.37
2.56 0.66 4.99 0.22 4.90 3.13 2.98 3.02 2.00 ⫺7.10 ⫺5.64
3.46 1.91 4.73 0.33 4.65 2.96 2.91 3.46 2.56 ⫺3.89 ⫺2.93
3.42 2.19 4.31 0.39 4.21 2.68 2.72 3.36 2.58 ⫺2.47 ⫺1.79
2.66 1.95 3.29 0.37 3.15 2.00 2.14 2.72 2.13 ⫺1.11 ⫺0.73
1.60 1.31 2.11 0.28 1.95 1.23 1.41 1.80 1.41 ⫺0.41 ⫺0.22
0.56 0.57 0.89 0.15 0.75 0.44 0.64 0.76 0.58 ⫺0.07 0.00
3.46 2.19 5.05 0.39 4.94 3.17 2.98 3.46 2.58 ⫺0.07 0.00
0.05 0.10 0.00 0.10 0.0 0.0 0.0 0.05 0.1 0.4 0.4
Note. Genetic coordinates in centimorgans according to the Marshfield genetic map are in the column “Location.” n gives the total number of different alleles encountered in the pedigree set; het gives the estimated heterozygosity value. Marker loci showing no recombination with HGF are in boldface type and represent the cosegregating segment. at 95°C for 30 s, annealing at 50 or 54°C for 30 s, and elongation at 72°C for 30 s. Finally the sample was elongated for 15 min at 72°C. PCR products were electrophoresed on a Licor 4200L DNA sequencer (Licor Corp.) on 6% denaturing polyacrylamide gels with fluorescent dye-labeled DNA markers (Licor Corp.). Data were collected and analyzed with Base Image 4.1 and Gene Image 3.12 software (Licor Corp.) and were checked by an experienced operator. Linkage ready pedigree files were prepared using Gene Image software. Linkage and haplotype analysis. The hereditary gingival fibromatosis locus was modeled as an autosomal dominant, two-allele system with 100% penetrance. The affected allele frequency was set to 0.0001. Marker allele frequencies were obtained from our four families’ genotyping data. Two-point linkage analysis was performed using the MLINK and ILINK programs with LINKAGE version 5.10 (Lathrop et al., 1984). Multipoint analyses were conducted by FASTLINK Version 4.1P (Cottingham et al., 1993; Schaffer et al., 1994). Genetic heterogeneity was tested by using the HOMOG program (Ott, 1991). Marker allele frequency calculation and haplotype construction were performed using the program SimWalk2 version 2.31 with the aid of Cyrillic Version 2.0.2 software (Sobel and Lange, 1996). Radiation hybrid mapping. Oligonucleotide primers for possible genes within the HGF critical region were synthesized based on the sequence of each gene. These primers were used to score the highresolution Stanford TNG radiation hybrid panel. Statistical analysis of the RH data was carried out on the mapping server at the Stanford Genome Center (http://shgc-www.stanford.edu/RH/rhserverformnew. html). Marker order was determined with Rhminbrk Version 3.0. Mutation analysis of CYP1B1. A set of oligonucleotide primers was designed covering the whole coding region of CYP1B1 (GenBank Accession No. U56438) using the primer 3.0 program (http://wwwgenome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi). These primers were used to perform PCR-single-strand conformation polymorphism (SSCP) assays for the CYP1B1 gene. The PCR products were run on 10% polyacrylamide gels and silver stained according to standard methods.
RESULTS
Linkage and Haplotype Analysis In our investigation, we typed all available individuals of the four HGF families with polymorphic mark-
ers on 2p21 (Fig. 1). HOMOG program analysis (A test) supported homogeneity of these four families (␣ ⫽ 1.00, 2 ⫽ 0.00). Two-point linkage analysis of these four families yielded a summed maximum lod score of 5.04 with D2S390 at ⫽ 0 (Table 1). The recombination events in individual 19 of family I placed the HGF locus distal to D2S119 (Fig. 1). The crossover in individual 8 of family III defines the HGF locus distal to D2S1346 (Fig. 1). To localize the HGF locus more precisely in family III, we typed this family with two additional polymorphic markers, namely D2S2186 and D2S2163, respectively. The paternal origin haplotype limited the HGF locus distal to D2S2163. A crossover in individual 22 of family I determined the distal boundary to D2S352. Our data thus suggest that the HGF gene lies in an 8.7-cM interval flanked by D2S352 and D2S2163. We conducted three-point analyses with adjacent marker loci and the disease locus. Multipoint analysis with chromosome 2p21 markers produced a maximum lod score of 6.61 at D2S2230. Scanning for Mutations in the CYP1B1 Gene by SSCP Analysis The entire coding region of CYP1B1 was screened for mutations with eight pairs of primers (see Table 2). Genomic DNA from four affected individuals and that from four unaffected individuals from family I was used as templates for PCR amplification. To assess the SSCP system, a positive control and a negative control were included. The results showed no mutations in the coding region in affected individuals, although two polymorphisms were detected with primer set CYP1B1-2 and CYP1B1-7. Although they cosegregated with the HGF trait in family I, they also appeared in unaffected individuals (data not shown).
REFINEMENT OF GINGF LOCUS TO A 3.8-cM REGION ON 2p21
TABLE 2 Primer Sequences Flanking the CYP1B1 Coding Region for Mutation Screening Using Genomic DNA Primer CYP1B1-1 CYP1B1-1 CYP1B1-2 CYP1B1-2 CYP1B1-3 CYP1B1-3 CYP1B1-4 CYP1B1-4 CYP1B1-5 CYP1B1-5 CYP1B1-6 CYP1B1-6 CYP1B1-7 CYP1B1-7 CYP1B1-8 CYP1B1-8
forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse
Sequence 5⬘-CACGCCTTCTCCTTTCTGTC-3⬘ 5⬘-ATCAGTGGCCACGCAAAC-3⬘ 5⬘-GCTGGCCACTGTGCATGT-3⬘ 5⬘-GGACACCACACGGAAGGA-3⬘ 5⬘-CCTCCTTCCGTGTGGTGT-3⬘ 5⬘-CGAAACACACGGCACTCAT-3⬘ 5⬘-GTCATGAGTGCCGTGTGTTT-3⬘ 5⬘-TGTCCAGGATGAAGTTGCTG-3⬘ 5⬘-AACCGCAACTTCAGCAACTT-3⬘ 5⬘-GCCTCCCAGAGGCTTTACC-3⬘ 5⬘-GCTCACTTGCTTTTCTCTCTCC-3⬘ 5⬘-CAGTGGTGGCATGAGGAATA-3⬘ 5⬘-GCCTGTCACTATTCCTCATGC-3⬘ 5⬘-TCATCACTCTGCTGGTCAGG-3⬘ 5⬘-AGGACCTGACCAGCAGAGTG-3⬘ 5⬘-CAGCTTGCCTCTTGCTTCTT-3⬘
Size (bp) 205 295 270 214 269 210 200 244
Radiation Hybrid Mapping In our study, we positioned the HGF locus distal to D2S2163, thus forming an overlapping region in the interval between D2S1788 and D2S2163 with the GGF1 locus region. This region is about 2.8 Mb, containing five known genes, namely, fasciculation and elongation protein zeta 2 (FEZ2), interferoninducible double-stranded RNA-dependent protein kinase (PRKR), protein kinase C nu (PRKCN), breakpoint cluster region protein 1 (BCRP1), CCAAT-box-binding transcription factor (CBF2), and 11 Unigene clusters. Three previously suggested candidate genes, CYP1B, NCX1, and Cdc42 effector protein 3 (CEP3), localize to the centromeric boundary of the HGF locus (http://bioinformatics.weizmann.ac.il/udb/). Splicing factor arginine/serine-rich 7 (SFRS7) was previously mapped to the interval of D2S1788 and CYP1B1. All genes within the critical region may be implicated in HGF disease. To map these genes more precisely, primers were designed to score high-resolution RH Panel TNG. These results show that the candidate locus order is as follows: CALM2–FEZ2–D2S1788 –CBF2–SFRS7–CYP1B1– CEP3–D2S2163–NCX1. DISCUSSION
Gingival fibromatosis is a heterogeneous disease. So far, two gingival fibromatosis loci have been mapped, GGF1 on 2p21 and GGF2 on 2p13 (Hart et al., 1998; Shashi et al., 1999; Fryns, 1996). In this study, we mapped a dominant hereditary gingival fibromatosis locus to a region defined by D2S352 and D2S2163 in Chinese families. This region overlaps with the recently reported GGF1 region (Hart et al., 1998). The overlapping region is about 3.8 cM from D2S1788 to D2S2163 and represents a physical re-
251
gion of 2.8 Mb. This indicates the existence of a common gingival fibromatosis locus in this region between the Brazilian family and the Chinese families. Among families II, III, IV, and I, no common haplotype was found in affected individuals, suggesting the different mutation origins. Since gingival fibromatosis can potentially be induced by calcium channel blockers, two genes within the previously mapped HGF locus region, CALM2 and NCX1, were suggested to be HGF candidate genes. However, high-resolution radiation mapping showed that CALM2 and NCX1 were outside of the HGF critical region. Thus, they are unlikely to be the genes responsible for gingival fibromatosis. Radiation hybrid panel analysis placed CBF2, SFRS7, CYP1B1, and CEP3 in the critical region. Based on the sequences of contigs NT-002229, NT-002723, and NT-002803 (http://www.ncbi.nlm.nih.gov/), FEZ2, PRKR, and PRKCN may also localize to the critical region. Because CYP1B1 plays an important role in the metabolism of a wide range of drugs, it was selected as the first gene to use in performing mutation detection in affected individuals. Mutation analysis of the coding region of CYP1B1 in family I revealed no obvious mutations although two polymorphisms were revealed. The polymorphic alleles cosegregated with the HGF phenotype. However, they also appear in normal control individuals. These results, however, do not exclude CYP1B1 mutations in other families or in noncoding regions. In gingival fibromatosis, the gingiva fibroblast is activated by some autocrines (Tipton and Dabbous, 1998; Hong et al., 1999), so all fibroblast expressed genes within the critical region may be included as candidates. We have not yet examined the expression profiles of other genes within the critical region. Based on the Cancer Genome Anatomy Project Database, at least PRKR is expressed in gingiva tissue (http://www.ncbi.nlm.nih.gov/ncicgap/). The fast progress of human genome sequencing will certainly aid in the cloning of gingival fibromatosis genes. In part of the population, some pharmacological agents such as phenytoin and cyclosporin can also induce gingival fibromatosis. The patients may carry one or more susceptibility genes, which have some defect in metabolizing or interacting with gingival fibromatosis inducible drugs. So far, however, no such susceptibility loci have been identified. In the Brazilian family, recombination suppression was observed between the polymorphic markers D2S1788 and D2S441 (Hart et al., 1998). In contrast, in the present study, our results did not support recombination suppression near the gingival fibromatosis locus. However, before the mutated genes are identified in these families of different racial origins, we cannot exclude the possibility that the controversial results are caused by allelic mutations.
252
XIAO ET AL.
ACKNOWLEDGMENTS We thank all members of gingival fibromatosis families for providing their blood samples. We are grateful to Dr. Fiona Francis for critically reading the manuscript. We gratefully acknowledge Dr. Gengxi Hu for the normal individual DNA samples. This work was supported by Life Science Special Fund of CAS supported by the Ministry of Finance (Grant KJ95T-06 to X.K.).
REFERENCES Cottingham, R. W., Jr., Idury, R. M., and Schaffer, A. A. (1993). Faster sequential genetic linkage computations. Am. J. Hum. Genet. 53: 252–263. Fryns, J. P. (1996). Gingival fibromatosis and partial duplication of the short arm of chromosome 2 (dup(2)(p13 3 p21)). Ann. Genet. 39: 54 –55. Goldblatt, J., and Singer, S. L. (1992). Autosomal recessive gingival fibromatosis with distinctive facies. Clin. Genet. 42: 306 –308. Hart, T. C., Pallos, D., Bowden, D. W., Bolyard, J., Pettenati, M. J., and Cortelli, J. R. (1998). Genetic linkage of hereditary gingival fibromatosis to chromosome 2p21. Am. J. Hum. Genet. 62: 876 – 883. Hong, H. H., Uzel, M. I., Duan, C., Sheff, M. C., and Trackman, P. C. (1999). Regulation of lysyl oxidase, collagen, and connective tissue growth factor by TGF-1 and detection in human gingiva. Lab. Invest. 79: 1655–1667. Lathrop, G. M., Lalouel, J. M., Julier, C., and Ott, J. (1984). Strat-
egies for multilocus linkage analysis in humans. Proc. Natl. Acad. Sci. USA 81(11): 3443–3446. Noyan, U., Yilmaz, S., Arda, O., and Kuru, B. (1994). The ultrastructural examination of gingival fibromatosis. J. Marmara. Univ. Dent. Fac. 2: 409 – 413. Ott, J. (1991). “Analysis of Human Genetic Linkage,” 2nd ed., Johns Hopkins Univ. Press, Baltimore, MD. Schaffer, A. A., Gupta, S. K., Shriram, K., and Cottingham, R. W., Jr. (1994). Avoiding recomputation in linkage analysis. Hum. Hered. 44: 225–237. Shashi, V., Pallos, D., Pettenati, M. J., Cortelli, J. R., Fryns, J. P., von Kap-Herr, C., and Hart, T. C. (1999). Genetic heterogeneity of gingival fibromatosis on chromosome 2p. J. Med. Genet. 36: 683– 686. Singer, S. L., Goldblatt, J., Hallam, L. A., and Winters, J. C. (1993). Hereditary gingival fibromatosis with a recessive mode of inheritance. Case reports. Aust. Dent. J. 38: 427– 432. Sobel, E., and Lange, K. (1996). Descent graphs in pedigree analysis: Applications to haplotyping, location scores, and marker-sharing statistics. Am. J. Hum. Genet. 58: 1323–1337. Tipton, D. A., and Dabbous, M. K. (1998). Autocrine transforming growth factor beta stimulation of extracellular matrix production by fibroblasts from fibrotic human gingiva. J. Periodontol. 69: 609 – 619. Tipton, D. A., Howell, K. J., and Dabbous, M. K. (1997). Increased proliferation, collagen, and fibronectin production by hereditary gingival fibromatosis fibroblasts. J. Periodontol. 68: 524 –530. Witkop, C. J., Jr. (1971). Heterogeneity in inherited dental traits, gingival fibromatosis and amelogenesis imperfecta. South. Med. J. 64(Supp. 11): 16 –25.