Screening of the Eight BBS Genes in Tunisian Families: No Evidence of Triallelism

Abstract

purpose. To study Bardet-Biedl syndrome (BBS) in the Tunisian population and determine the presence of triallelism in the eight identified BBS genes.

methods. DNA samples were collected from 19 consanguineous Tunisian families with BBS. Genome-wide scans were performed with microsatellite markers in 12 families, and two-point linkage analyses were performed. Direct sequencing was used to screen patients with BBS for mutations in all eight identified BBS genes.

results. Mutations in the BBS genes were identified in nine families. In addition, a large consanguineous family (57004) showed linkage to the BBS7 locus, although no mutation was identified. Five novel mutations were present in the nine families: one in BBS2 (c.565C>T, p.ArgR189Stop), one in BBS5 (c.123delA, p.Gly42GlufsX11), one in BBS7 (g.47247455_47267458del20004insATA, p.Met284LysfsX7), and two in BBS8 (c.459+1G>A, p.Pro101LeufsX12 and c.355_356insGGTGGAAGGCCAGGCA, p.Thr124ArgfsX43).

conclusions. All families in which mutations were identified show changes in both copies of the mutant gene, and inheritance patterns in all families are consistent with autosomal recessive inheritance excluding any evidence of triallelism in the BBS genes in Tunisia.

Bardet-Biedl syndrome (BBS; OMIM 209900; Online Mendelian Inheritance in Man; http://www.ncbi.nlm.nih.gov/Omim/ provided in the public domain by the National Center for Biotechnology Information [NCBI], Bethesda, MD) is a group of clinically and genetically heterogeneous autosomal recessive disorders classically defined by the association of retinal degeneration, obesity, postaxial polydactyly, hypogonadism, renal disorders, and highly variable mental retardation. Occasionally, the clinical features could include diabetes mellitus, congenital cardiomyopathy, genital anomalies, and behavioral problems. ^{1 2 3 4} To date, eight distinct BBS genes have been identified at the following loci: 11q13, BBS1 ⁵ ; 16q21, BBS2 ⁶ ; 3p13-p12, BBS3 ^{7 8 9} ; 15q22.3-q23, BBS4 ¹⁰ ; 2q31, BBS5 ^{11 12} ; 20p12, BBS6 ^{13 14} ; 4q27, BBS7 ¹⁵ ; and 14q32.1, BBS8. ¹⁶  

Different types of mutations have been reported for the BBS genes including missense, nonsense, deletions, and insertions. Mutations in the BBS1 gene are the most frequently encountered among the white population with p.Met390Arg observed in 78% to 80% of the affected families. ^{17 18 19} Of interest, in some of the studied BBS families, none of the known BBS gene mutations were identified, suggesting that at least one BBS locus remains unidentified. 

Triallelic inheritance is defined as a rare inheritance pattern in which three mutant alleles distributed between two genes are necessary to produce an affected phenotype, whereas two mutations in a single gene result in an unaffected phenotype. BBS resulting from mutations in BBS1, BBS2, BBS3, BBS4, BBS5, BBS6, BBS7, and BBS8 has been reported to display triallelic inheritance, ^{9 12 15 16 17 20 21 22} but the contribution of each allele is difficult to ascertain. Moreover, it has been suggested that there might be a connection between triallelism and the phenotypic severity of the disease. ²⁰ However, another study reported that there is no evidence supporting the complex inheritance mode in this syndrome. ¹⁸ In addition, double heterozygous mice for BBS2 and BBS4 genes are normal. ²³ The BBS proteins (types 1 through 8) have been localized at the basal bodies and/or centrosomes of mammalian cells and have been shown to allow signaling, either directly or indirectly, by maintaining a microtubule-based network that connects the different cellular structures. 

Herein, we report the results of linkage and mutation analysis of all eight identified BBS genes in 19 Tunisian BBS families. The geographic position of Tunisia in North Africa has ensured an eventful population history. Phoenicians, Romans, Vandals, Byzantines, Arabs, Ottomans, and French have controlled or colonized the region at different historical times. Although affected by the admixture of these different groups, the native Tunisian population derives mainly from the Arab-Berber ethnic group known for the high frequency of consanguineous marriages. The syndrome in all affected individuals is consistent with an autosomal recessive mode of inheritance. Linkage analysis revealed that only 9 of the 19 studied families shared the known BBS loci, and sequencing of the eight known BBS genes identified five novel mutations. These findings confirm the genetic heterogeneity of BBS in Tunisia, thus complicating the DNA diagnosis of this syndrome in the Tunisian population. Furthermore, it suggests that genes associated with BBS in a large affected group of the Tunisian population are yet to be identified. 

Material and Methods

Families and Clinical Ascertainment

Informed consent was obtained from each individual family member. This study was approved by the ethics review board of Charles Nicolle Hospital, Tunis, and the National Eye Institute (NEI) Institutional Review Board, and the protocol was consistent with the Declaration of Helsinki. BBS was diagnosed on the basis of the established criterion that four primary features or three primary plus two secondary features are necessary to make the diagnosis. ^{19 24} DNA was isolated from blood by standard procedures. ²⁵ Samples from 19 Tunisian families including 34 patients were collected at the outpatient clinic of the genetic department at Charles Nicolle Hospital in Tunis. Patients were referred by physicians throughout Tunisia. 

Molecular Analysis

Genotyping and Linkage Analysis.

In families 57001, 57003, 57004, 57005, 57006, 57007, 57008, 57009, 57010, 57011, 57012, and 57013, genome-wide scans were performed with 382 fluorescently labeled microsatellite markers (Linkage Mapping Set MD-10; Applied Biosystems, Inc. [ABI], Foster City, CA). Multiplexed PCR was performed as described. ²⁶ Two-point linkage analyses were performed with the FASTLINK implementation of the MLINK program of the LINKAGE package. ^{27 28} BBS was assumed to be a fully penetrant autosomal recessive trait. Equal allele frequencies were assumed for the genome-wide scan. 

Haplotype Analysis.

Microsatellite markers and single-nucleotide polymosphisms (SNPs) were used to determine whether founder mutations were present in the Tunisian patients with BBS in the case that an identical mutation was identified in different families. Haplotypes were constructed using five markers—D16S3039, D16S3110, D16S408, D16S3071, and D16S3057—for the BBS2 mutation and four microsatellite markers—D14S73, D14S1058, D14S1066, and D14S1044—and one SNP identified in intron 2 of the BBS8 gene for the BBS8 mutation. 

Sequence Analysis.

Primer sequences for all eight BBS genes were designed to amplify each exon, including the intron–exon boundaries, and are available from the authors on request. DNA was sequenced using dye termination chemistry (BigDye Terminator cycle sequencing ver. 3.1 and the model 3100 sequencer; ABI) according to the manufacturer’s recommendations. The sequence information was imported into a computer program (SeqMan 5.01; DNAStar Inc., Madison, WI), and sequences of one affected member of each family and one unaffected family member when available and consensus sequences were aligned to identify variations. 

Amplification of the Junctional Fragment in the BBS7 Deletion.

PCR amplification of the junctional fragment in the BBS7 gene of the affected individual of family 57014 was performed in 50-μL reactions with primers BBS7-Ex8fw 5′-ATGGGGGAAGACAGAGGGAGGAG-3′ and BBS7-Ex16rv 5′-AGGAGTTGACCAGATGCAGTTAGATA-3′ with polymerase (LATaq; Takara Bio Inc., Shiga, Japan). PCR cycling consisted of an initial denaturation step of 95°C for 10 minutes; 32 cycles consisting of a 94°C denaturation step for 40 seconds, a 55°C annealing step for 30 seconds, and a 72°C elongation step for 5 minutes; and a final elongation step at 72°C for 5 minutes. The same primers were used for sequencing with the addition of primers P654, 5′-TCAAGAGATCAAGACCA-3′, and P655, 5′-CTGCTCCTTGCAGAACAGGGC-3′. 

Molecular Modeling.

The BBS6 gene wild-type protein (BBS6) model was built by homology modeling based on crystal coordinates for archaeal chaperonin from Thermoplasma acidophilium (Brookhaven protein database [PDB] ²⁹ file 1a6d) as the structural template. Protein recognition (x-fold) was performed using one- and three dimensional sequence profiles coupled with secondary structure and solvation potential information (3D-PSSM; http://www.sbg.bio.ic.ac.uk/∼3dpssm). The primary sequences of BBS6 and archaeal chaperonin were aligned by the method of Needleman and Wunsch, ³⁰ incorporated in the program Look, version 3.5.2 ³¹ for three-dimensional structure prediction. Finally, the monomeric BBS6 was built using the automatic segment-matching method in Look, ³² followed by 500 cycles of energy minimization. The program was used to predict the conformation of BBS6 incorporating the genetic point mutations R517C and G532V and to refine the structure by self-consistent ensemble optimization, which applies the statistical mechanical mean-force approximation iteratively to achieve the global energy minimum structure. 

Results

Mutation Analysis

Ten of the 19 collected families had more than one affected individual. A total of 34 patients are included in the study, in which 5 patients with BBS belonged to the same family 57004 (Fig. 1) ; 3 others came from two different families 57001(Fig. 2a)and 57013 (Fig. 2f) ; and 2 patients each were from families 57002 (Fig. 2b) ; 57003 (Fig. 2c) ; 57007, 57009, 57010 (Fig. 2d) ; and 57011 and 57012(Fig. 2e) . Families 57005, 57006, 57008, 57014, 57015 (Fig. 2g) ; 57016 (Fig. 2h) , 57017 (Fig. 2i) , 57018, and 57019 had only one affected individual. Sequencing of all eight BBS genes for the 34 patients, including the entire coding region and the splice site junctions, showed seven different mutations in nine families, five of them are novel (Table 1) . In addition we identified 37 polymorphic changes (Table 2) . None of the novel pathogenic changes detected in the Tunisian BBS families were found in 60 tested control individuals belonging to the same ethnic population. The unaffected siblings, when available, were genotyped for the respective mutations identified in their families, and none of them was found to carry two mutations. 

BBS1.

Sequence analysis of the 16 exons of the BBS1 gene revealed a homozygous C>T change at position 448 in exon 5, c.448C>T (Fig. 3a)creating a stop codon, p.Arg146Stop, in both affected individuals in family 57007, possibly resulting in nonsense-mediated decay. This change has been reported. ¹⁷ The most frequent mutation in the white population, p.Met390Arg, was not found in any of the Tunisian families. 

BBS2.

Two unrelated families, 57008 and 57018, showed an identical homozygous sequence change, c.565C>T (Fig. 3b) , creating a stop codon in exon 5, pArg189Stop. The resultant BBS2 protein is thus predicted to lack the terminal 532 amino acids and is possibly subject to nonsense-mediated decay. Haplotype analysis using five polymorphic microsatellite markers located within 1 Mb of the BBS2 gene have identified a common haplotype (255, 241, 125, c.565C>T, and 91, and 192) for (D16S3039, D16S3110, D16S408, BBS2, D16S3071, and D16S3057) respectively in both families 57008 and 57018 (Table 3) . 

BBS5.

A previously unreported homozygous mutation in exon 2 of BBS5 is seen in family 57011. A deletion of one base c.123delA (Fig. 3c)creates a frameshift predicted to result in a truncated protein lacking the final 300 amino acids, p.Gly42GlufsX11. 

BBS6-MKKS.

An affected member of family 57005 showed a homozygous nonsense mutation in exon 6, c.1436C>G (Fig. 3d) , predicted to result in a p.Ser479Stop with 92 amino acids missing from the C-terminal end of the protein. In addition, this individual was homozygous for two missense changes in exon 6: c.1595G>T (p.Gly532Val) and c.1549C>T (p.Arg517Cys). These two changes were also seen in heterozygous fashion in four unrelated affected individuals from families 57008, 57015, 57016, and 57018 and in five control individuals (8%), but no control subjects were homozygous for these two changes. 

In an effort to understand the mutations detected in the BBS genes, we used a computer-generated model of BBS6 to predict the effect of the changes p.Arg517Cys and p.Gly532Val. Neither of these changes was detected in homozygous form in 120 control chromosomes. Screening BBS6 using the Web-based 3D-PSSM method for thermosome protein x-fold recognition showed two significant hits with E-values of 1.35e-5 and 2.7e-3, both with the archaeal chaperonin (PDB files 1q3q_A, 1a6e_A, or 1a6d_A). The alignment of BBS6 and archaeal chaperonin sequences exhibit a sequence identity (similarity) of 20% (34%) for 500 common residues. Thus, the atomic coordinates of thermosome (subunit A) were used as a structural template for building the model of the monomeric BBS6 structure. The structural modeling of BBS6, with and without the two identified mutations is shown in Figure 4 . In the wild-type BBS6 structure, amino acid residues 517 and 532 are located on the protein surface. In the p.Arg517Cys mutant, the charged amino acid arginine is replaced by hydrophobic cysteine at the protein surface. In the p.Gly532Val mutant, small glycine is replaced by the hydrophobic residue valine, exposed on the protein surface. These two changes in BBS6 occur on the C-terminal part of the molecule. The nature of the changes in residue surface exposures due to mutations p.Arg517Cys and p.Gly532Val suggest that these two changes do not destabilize the BBS6 structure, but that their effect could primarily be related to protein–protein interactions. Recently, it has been reported that the amino terminal part of the BBS6 apical domain is involved in association with the centrosome, ³³ and so the associations affected by these mutations are presumably related to other proteins. 

BBS7.

In family 57014, the affected individual carried a large homozygous deletion of 20,007 bp which includes exon 9 through exon 15 (g.47247455_47267458del20004insATA (Fig. 3e) . This deletion causes a frameshift at methionine 284 and is predicted to yield a premature termination, p.Met284LysfsX7. 

BBS8.

Three Tunisian families showed two unreported mutations in BBS8, including a homozygous insertion of 16-bp GGTGGAAGGCCAGGCA from nucleotides 355 to 356 bp at exon 4 in all affected individuals of family 57009 (Fig. 3g) . After an initial G, the next 15 bases correspond to a duplication of the sequence from nucleotide 356-370 (c.355_356insGGTGGAAGGCCAGGCA). This sequence is predicted to result in a frameshift and an early stop codon after 42 amino acids (p.Thr124ArgfsX43). In family 57006, the affected individual harbored a homozygous substitution in the invariant AG sequence of the 5′ (donor) splice site of exon 4, c.459+1G>A (Fig. 3f) . Therefore, exon 4 is predicted to be skipped, causing a frameshift and truncation of the protein p.Pro101LeufsX12, possibly resulting in nonsense-mediated decay. This same mutation was also seen in a homozygous fashion in an unrelated affected individual from family 57019. Haplotype analysis using four microsatellite markers within 1 Mb on either side of the BBS8 gene and a rare novel SNP identified in intron 2 of the BBS8 gene show a common haplotype (248, c.459+1G>A, C, 290, and 254) for (D14S1058, BBS8, SNP c.194+48T>C, D14S1066, and D14S1044), respectively, in both families (Table 3) . 

No deleterious mutations were observed in the BBS3 and BBS4 genes. 

Genome-wide Scan

A genome-wide scan was performed to confirm the significance of some of the sequence changes and possibly to identify new BBS loci. The scan was limited to families 57001, 57003, 57004, 57005, 57006, 57007, 57008, 57009, 5010, 57011, 57012, and 57013, which were considered large enough for genome-wide scans, even though some were not large enough to provide significant results when analyzed alone. Linkage analysis for the large consanguineous family 57004, which consists of 14 individuals of whom five are affected (Fig. 1) , yielded the highest two-point lod score with D4S402 (5.52 at θ = 0; Table 4 ). D4S1575 also showed significantly positive lod scores without recombination (4.72 at θ = 0). The high lod score obtained with these markers, which flank the BBS7 locus (D4S402 to 2.5Mb-BBS7 to 12Mb-D4S1575) suggests linkage of family 57004 to the BBS7 locus. Surprisingly, the coding sequence and splice sites of both reported isoforms of BBS7 gene did not show any changes. Linkage analysis of families with no identified mutation in the eight BBS genes (Table 5)revealed that in family 57001 two-point lod scores of 2 or greater were observed with markers D1S207 and D12S326 and for family 57003, linkage analysis yielded a lod score of 1.5 or greater with markers D6S1610, D7S515, D7S486, D7S530, D8S550, and D14S63. Only markers on 7q show clusters with lod score ≥ 1.5. Families 57010, 57012, and 57013 yielded only lod scores < 1 (data not shown). 

Discussion

Herein, we report the results of screening the coding regions and exon–intron boundaries of all eight identified BBS genes in 19 consanguineous Tunisian families in which individuals are affected by BBS. Perhaps as a result of Tunisia’s complicated history, these results show that even in this small population with high levels of consanguinity, ³⁴ BBS is genetically very heterogeneous. In a group of 19 families, mutations were identified in six of the eight known BBS genes with no deleterious changes identified in the remaining 10 families (Figs. 1 2) . In agreement with previous studies in which mutations were identified in 50% of the BBS families, ^{3 5 10 17 18 19 35} in our study, mutations are identified in 9 of 19 families corresponding to a ratio of 47%. However, BBS1 mutations are detected in only one family (5%). This ratio is lower than the 20% to 25% seen in affected white patients with BBS, ^{5 17 20 36} but in concordance with the frequency reported in families from Saudi Arabia which is estimated at 8%. ¹⁷  

An identical mutation in exon 5 of BBS2 gene c.565C>T (p.Arg189Stop) was identified in two families (10%). Even though these two families were unrelated and were from two different cities in Tunisia, they shared the same haplotype extending over 2 Mb, suggesting the presence of a founder effect for this mutation. The BBS gene implicated most frequently in our study was BBS8, with three families carrying mutations including two unrelated families carrying the same 5′(donor) splice site change in exon 4 c.459+1G>A. Haplotype analysis using four informative microsatellite markers and a rare SNP within intron 2 of the BBS8 gene shows that the mutation p.Pro101LeufsX112 is associated with a common founder haplotype in these two families. We also report the first large deletion in the BBS7 gene, with a homozygous loss of exons 9 through 15. 

In triallelic inheritance, also called recessive inheritance with a modifier of penetrance, ³⁷ mutations in both alleles of a single causative gene result in a mild or undetectable phenotype. For full expression of the disease state, a mutation of a single allele at a second gene is necessary. Findings consistent with this inheritance pattern have been reported for BBS in several studies, but were not found in some others. In addition, several knockout mouse models generated for the BBS1, BBS2, BBS4, and MKKS genes, ^{23 38 39 40} appear to require only homozygous absence of a single BBS gene to display the phenotype. No evidence of triallelic inheritance was detected in this study, as none of the unaffected siblings in our families carried two mutations in any of the known BBS genes. This result is consistent with previous studies of BBS in Arab populations, specifically in families originating from Saudi Arabia. One possible explanation may be that the major gene implicated in this phenomenon is BBS1 with the p.Met390Arg mutation in white populations, ^{17 18} and the mutation was not observed in any of our patients. In fact, mutations in the BBS1 gene were observed in only a single family, and the mutations reported in the present study are severe frameshift or truncation mutations rather than missense mutations that might leave some residual activity. 

Family 57004 was a large consanguineous family that was linked to a region of chromosome 4 including the BBS7 locus with a maximum lod score of 5.52 at θ = 0 with D4S402. However, no mutation was detected in affected members of this family. There are several possible explanations for this besides the trivial one of simple sequencing error. These include first that there might have been mutations present in regulatory sequences or in introns (no changes were observed in splice site junctions) that were not detected in our sequencing protocol. A second possibility is that there is a large duplication or inversion not detected by sequencing. Third, there may be missing exons in the known BBS7 gene in which mutations would not be detected by our current sequencing protocol. Finally, the disease in this family could be caused by a new gene physically close to the BBS7 gene. 

There is some suggestion of new BBS loci in the remaining families, although the data are not strong enough for a definitive statement. Two-point linkage in family 57001 showed 2 chromosomal regions with a lod score > 2.5, and in family 57003 a genome-wide scan revealed four potential loci for BBS, each with a lod score ≥ 1.5. Finally, the absence of mutations in 10 families and the haplotype analysis in 3 families (57001, 57002, and 57003; data not shown), excluding association to any of the known loci, confirms the presence of at least one additional BBS locus. 

In conclusion, besides seven different pathogenic changes, five of which are novel, we have identified 37 polymorphic changes (Table 2) . We were unable to find evidence of triallelic inheritance in this population. Finally, our study shows similarity in the mutational spectrum of BBS between the Tunisians and Middle Eastern populations as previously suggested by several genetic analyses, especially Y-chromosome studies. ^{41 42}  

Appendix 1

Databases

GenBank: BBS1 NM_024649, BBS2 NM_031885, BBS3 NM_032146, BBS4 NM_033028, BBS5 NM_152384, BBS6-MKKS NM_018848 and NM_170784, BBS7 NM_018190 and NM_176824, BBS8 NM_144596, NM_198309 and NM_198310. 

Online Mendelian Inheritance in Man (OMIM), http://www.ncbi.nlm.nih.gov/Omim/ 

Human Mutation Database, http://www.hgmd.cf.ac.uk/ 

Protein Fold Recognition, http://www.sbg.bio.ic.ac.uk/∼3dpssm/ 

Brookhaven Protein Database http://www.ch.cam.ac.uk/ 

 Supported in part by the Intramural Research Program of the National Institutes of Health, the National Eye Institute, and the Scientific and Research Minister of Tunisia.

 Submitted for publication October 11, 2005; revised March 17, 2006; accepted June 20, 2006.

 Disclosure: N. Smaoui, None; M. Chaabouni, None; Y. V. Sergeev, None; H. Kallel, None; S. Li, None; N. Mahfoudh, None; F. Maazoul, None; H. Kammoun, None; N. Gandoura, None; A. Bouaziz, None; E. Nouiri, None; R. M’Rad, None; H. Chaabouni, None; J.F. Hejtmancik, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Pedigree and haplotype analysis of family 57004 with the BBS7 locus. Markers D4S402 and D4S1575 are homozygous in all affected individuals. Filled bars: affected haplotypes with homozygous markers.

Figure 2.

 

Pedigrees of families with no identified mutation in eight known BBS genes; +, DNA sample available (family 57004 is shown in Fig. 1 ).

Table 1.

 

Mutations in BBS Genes in Tunisian Families

Family Identifcation	Gene	Location	Nucleotide Change	Effect on Protein Amino Acid Change	GenBank RefSeq	References
57007	BBS1	Exon5	c.448C>T	p.Arg146Stop	NM_024649	Beals PL, et al. 17
57008	BBS2	Exon5	c.565C>T	p.Arg189Stop	NM_031885	This study
57018	BBS2	Exon5	c.565C>T	p.Arg189Stop	NM_031885	This study
57011	BBS5	Exon2	c.123delA	p.Gly42GlufsX11	NM_152384	This study
57005	BBS6	Exon6	c.1436 C>G	p.Ser479Stop	NM_018848	Hichri H, et al. 19
57014	BBS7	Deletion Ex9-Ex15	g.47247455_47267458del20004insATA	p.Met284LysfsX7	NM_018190	This study
57006	BBS8	Splice site exon4	c.459+1G>A	p.Pro101LeufsX12	NM_144596	This study
57019	BBS8	Splice site exon4	c.459+1G>A	p.Pro101LeufsX12	NM_144596	This study
57009	BBS8	Exon4	c.355_356insGGTGGAAGGCCAGGCA	p.Thr124ArgfsX43	NM_144596	This study

Table 2.

 

Single Nucleotide Polymorphisms Identified in 19 Tunisian BBS Families

Gene	mRNA	Family	Localization-Exon	Nucleotide Change	Amino Acid Change	Polarity Change	dB SNP Reference
BBS1	NM_024649	57002, 57003, 57004, 57019, 57012, 57013	4	c.378G>A	p.Leu126Leu	No	rs2298806
		57012, 57013	9	c.736-8G>C			rs10896125
		57002, 57013	11	c.1110G>A	p.Pro370Pro	No	This study
		57019	12	c.1146G>C	p.Gly382Gly	No	This study
		57003, 57004, 57012	14	c.1413C>T	p.Leu471Leu	No	rs3816492
		57010, 57012	17	c.1779+7A>G			rs8432
BBS2	NM_031885	57009, 57014, 57016, 57013	3	c.367A>G	p.Ile123Val	No	rs3177663
BBS3	NM_032146	57012	7	c.327C>G	p.Leu109Leu	No	This study
		57017	6	c.195T>C	p.Phe65Phe	No	This study
BBS4	NM_033028	57009	6	c.405+17C>T			This study
		57002, 57003, 57005, 57006, 57007, 57008, 57015, 57016, 57012, 57013	13	c.1061T>C	p.Ile354Thr	Yes	rs2277598
BBS5	NM_152384	57007, 57019, 57012	1	c.1-40C>G			rs1879466
		57001, 57011	2	c.142+160C>A			rs2353187
		57001, 57015	7	c.618+139T>C			This study
		57010	8	c.619-9T>C			This study
		57001, 57003, 57004, 57005, 57007, 57008, 57009, 57016, 57017, 57018, 57010, 57011, 57012	9	c.682-52A>T			rs12692909
		57005, 57013	11	c.901-49C>G			This study
BBS6	NM_018848	57005, 57008, 57015, 57018	3	c.117C>T	p.Pro39Pro	No	rs17852626
		57005, 57008, 57015	3	c.534C>T	p.Ile178Ile	No	rs17852625
		57015, 57016, 57018, 57019, 57010, 57011, 57012	3	c.985+33G>C			rs221667
		57015, 57016, 57018	3	c.985+16T>G			rs6133917
		57015, 57016	4	c.986-29A>T			rs764266
		57005, 57008	6	c.1595 G>T	p.Gly532Val	No	rs1545
		57005, 57008, 57015, 57016, 57018	6	c.1549 C>T	p.Arg517Cys	No	rs1547
BBS7	NM_018190	57002, 57006, 57007, 57009, 57014, 57019	4	c.166-45C>T			This study
		57006, 57007, 57015, 57018, 57011, 57012	5	c.528+173C>G			rs2661555
		57001, 57008, 57019, 57010, 57013	9	c.934+31delAAGA
		57003, 57006, 57009, 57011	9	c.934+32A>G			rs6824258
		57013	10	c.1037+150C>T			This study
		57006	11	c.1041T>C	p.Asn347Asn	No	This study
		57010, 57013	14	c.1511+25C>A			This study
		57001, 57002, 57003, 57004, 57005, 57007, 57008, 57015, 57016, 57017, 57018, 57010, 57011, 57012	17	c.1866+16A>G			rs1507994
		57006, 57007, 57015, 57016, 57017, 57018, 57011, 57012	18	c.1891-12A>C			rs2706793
BBS8	NM_144596	57006, 57019	2	c.194+48T>C
		57011, 57012, 57013	6	c.594+67G>A			rs12890993
		57012	11	c.880-170T>C			This study
		57011, 57012	15	c.1402-5T>G			This study

Figure 3.

 

Electropherograms from affected individuals and controls. All mutations are homozygous and underlined (a) BBS1 mutation in family 57007, c.448C>T, p.Arg146Stop; (b) BBS2 mutation in families 57008 and 57018 c.565C>T, p.Arg189Stop; (c) BBS5 mutation in family 57011 (deletion of 1 base is indicated by an arrow, c123delA), p.Gly42GlufsX11; (d) mutation in BBS6 in family 57005, c.1436C>G p.Ser479Stop; (e) mutation in family 57014, deletion of exon 9 through exon 15, sequence of the junctional fragment between introns 8 and 15 and the insertion of ATA, g.47247455_47267458del20004insATA, p.Met284LysfsX7; (f) splice mutation in 5′ (donor) splice site of exon 4 c.459+1G>A, p.Pro101LeufsX12, in BBS8 gene in both families 57006 and 57019; (g) mutation c.355_356insGGTGGAAGGCCAGGCA, p.Thr124ArgfsX43, in the BBS8 gene in family 57009, the dashed line corresponds to the inserted sequence of the 16 bp, the continued trait corresponds to the duplicated sequence from nucleotide 356 to 370.

Table 3.

 

Markers and Haplotypes Analysis in BBS2 Mutation for Families 57008 and 57018 (common haplotype is shown in blue color) and BBS8 Mutation for Families 57006 and 57019 (common haplotype is shown in yellow color)

 

Figure 4.

 

Structural model of BBS6 built using homology modeling and effect of genetic mutations. The spatial organization of secondary structure elements of BBS6 built by homology modeling is shown in (A). Red cylinders and yellow arrows show α-helices and β-strands, respectively. The deleted polypeptide chain fragment 479 to 511, which is lost in the structure after the mutation p.Ser479Stop (S479X) is shown in red. Positions of amino acid residues C517 and V532 corresponding to point mutations p.Arg517Cys (R517C) and p.Gly532Val (G532V) in BBS6 structure are represented in green in (B) and (C), respectively.

Table 4.

 

Lod Score BBS7 Locus in Family 57004

Marker	Physical-Map Location (kB) NCBI	Genetic-Map Location (cM)	Lod Score at θ =									θ Maximum	Z max
						0	0.01	0.05	0.1	0.2
D4S406	112076	115.9	−1.43	2.5	2.94	2.84	2.22	0.06	2.95
D4S402	120506	122.6	5.52	5.41	4.96	4.38	3.19	0	5.52
D4S1575	135148	134.5	4.72	4.62	4.25	3.76	2.76	0	4.72
D4S424	142555	143.8	−Infinity	−2.52	−0.13	0.56	0.74	0.2	0.74

Table 5.

 

Two Point Lod Score Analysis for Families 57001 and 57003

Family	Marker	Cytogenetic-Map Location	Lod Score at θ =									θ Maximum	Z max
						0	0.01	0.05	0.1	0.2
57001	D1S207	1p31.1	2.53	2.48	2.25	1.9	1.35	0	2.53
57001	D12S326	12q21.2	2.41	2.36	2.16	1.9	1.35	0	2.41
57003	D6S1610	6p21.2	1.51	1.47	1.32	1.12	0.76	0	1.51
57003	D7S515	7q22.1	1.49	1.46	1.33	1.16	0.83	0	1.49
57003	D7S486	7q31.2	1.96	1.91	1.71	1.46	0.98	0	1.96
57003	D7S530	7q32.1	1.58	1.54	1.38	1.18	0.79	0	1.58
57003	D8S550	8p23.1	1.96	1.91	1.73	1.5	1.04	0	1.96
57003	D14S63	14q23.2	1.9	1.9	1.7	1.4	1	0	1.9

The authors thank the families for their participation.