The Pseudohypoparathyroidism Type 1b Locus ... - Wiley Online Library

Download PDF
6 downloads 54475 Views 894KB Size Report
Comment
located on chromosome 20q13.3, the region containing GNAS1, demonstrated linkage to PHP 1b. Fine- mapping and multipoint linkage analysis of this region ...
JOURNAL OF BONE AND MINERAL RESEARCH Volume 18, Number 3, 2003 © 2003 American Society for Bone and Mineral Research

The Pseudohypoparathyroidism Type 1b Locus Is Linked to a Region Including GNAS1 at 20q13.3 SUZANNE M JAN DE BEUR,1,3 JEFFERY R O’CONNELL,4 RITA PEILA,5 JUSTIN CHO,1,3 ZHICHAO DENG,2,3 STEPHEN KAM,2,3 and MICHAEL A LEVINE2,3 ABSTRACT Pseudohypoparathyroidism (PHP) is characterized by biochemical hypoparathyroidism with elevated parathyroid hormone levels owing to reduced target tissue responsiveness to parathyroid hormone. Patients with PHP 1a have somatic defects termed Albright’s hereditary osteodystrophy (AHO) and exhibit resistance to additional hormones because of heterozygous mutations in the GNAS1 gene that lead to a generalized deficiency of the ␣ subunit of Gs, the heterotrimeric G protein that couples receptors to adenylyl cyclase. By contrast, patients with PHP 1b lack AHO and have selective parathyroid hormone (PTH) resistance, presumably because of an imprinting defect that impairs expression of Gs␣ in the proximal renal tubule. Although an epigenetic defect in GNAS1 has been identified in subjects with PHP1b, the genetic defect is unknown. To define the genetic defect in PHP 1b, we performed a genome-wide linkage analysis in five multi-generational PHP 1b families. Of the 408 polymorphic microsatellite markers examined, markers located on chromosome 20q13.3, the region containing GNAS1, demonstrated linkage to PHP 1b. Finemapping and multipoint linkage analysis of this region demonstrated linkage to a 5.7-cM region between 907rep2 and the telomere. Haplotype analysis established that affected individuals shared a 5-cM region including part of the GNAS1 gene to the telomere. Our data confirm that PHP1b is linked to a region that includes GNAS1, and further refine the locus, although the primary genetic mutation(s) that causes defective imprinting of GNAS1 remains undefined. (J Bone Miner Res 2003;18:424 – 433) Key words:

pseudohypoparathyroidism, linkage analysis, GNAS1, genetic imprinting, G protein INTRODUCTION

SEUDOHYPOPARATHYROIDISM (PHP) is a heterogeneous group of disorders characterized by biochemical hypoparathyroidism caused by failure of proximal renal tubule cells to respond to parathyroid hormone (PTH). Early studies showed that patients with PHP fail to generate either a phosphaturic or nephrogenous cyclic adenosine monophosphate (cAMP) response to exogenously administered PTH,(1) thereby implicating a defect in the PTH receptorheterotrimeric GTP-binding protein (Gs)-adenylyl cyclase signal transduction complex as the basis for PTH resistance.

P

The authors have no conflict of interest.

In the PHP1a variant of PHP, accessible tissues have demonstrated a 50% reduction in expression and activity of the ␣ chain of Gs (Gs␣)(2,3) owing to heterozygous mutations in the GNAS1 gene located at 20 q13.3.(4) Patients with PHP 1a are unresponsive to PTH and to several additional hormones whose receptors are coupled to Gs(5–7) and manifest a distinct constellation of developmental and somatic defects, collectively termed Albright hereditary osteodystrophy (AHO).(8) Many PHP 1a kindreds also contain individuals who have the GNAS1 defect and AHO but also have normal hormone responsiveness, a condition termed pseudopseudohypoparathyroidism (PPHP). Subjects with PPHP inherit a defective GNAS1 allele paternally, whereas subjects with PHP1a inherit the defective allele maternally.(9,10)

1

Departments of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA. Department of Pediatrics, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA. 3 Ilyssa Center for Molecular and Cellular Endocrinology, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA. 4 Department of Genetics, University of Pittsburgh, Pittsburgh, Pennsylvania, USA. 5 Epidemiology, Demography and Biometry Program, National Institute of Aging, National Institutes of Health, Bethesda, Maryland, USA. 2

424

PSEUDOHYPOPARATHYROIDISM 1b LINKS TO 20q13.3

These observations are consistent with current models of imprinting of the GNAS1 locus in both humans(11,12) and mice.(13,14) Indeed, studies of mice that are heterozygous for disruption of the Gnas gene show different phenotypes that depend on the parental origin of the defective allele.(13,15) A second variant, PHP 1b, is clinically and biochemically distinct from PHP 1a. Patients with PHP 1b lack features of AHO, have normal expression of Gs␣ protein in accessible tissues, and manifest hormone resistance only to PTH.(16) Selective resistance to PTH in PHP 1b patients led to the hypothesis that the gene encoding type 1 PTH receptor (PTHR) might be defective in these patients. However, analysis of the coding regions,(17) promoter,(18,19) and cDNA(20) of the PTH1R gene has failed to reveal mutations and linkage analysis excluded both the type 1 and type 2 PTHRs as possible candidate disease genes.(21) Recent studies have demonstrated linkage of PHP 1b to an approximately 9-cM region on chromosome 20q13.3 that includes at least a portion of GNAS1.(22,23) These studies demonstrated that the pattern of inheritance was consistent with imprinting at this locus.(22) Subsequent studies have identified an epigenetic defect in the GNAS1 gene in subjects with PHP 1b that results from loss of maternal imprinting within the differentially methylated region of exon 1A of GNAS1.(24) The genetic mutation that results in this epigenetic defect is unknown, however. To elucidate the genetic defect in PHP 1b, we undertook a genome-wide linkage analysis in several well-characterized, multigenerational PHP 1b kindreds. We confirm that PHP 1b is linked to chromosome 20q13.3, which includes the GNAS1 locus, and have refined the candidate region to a 5-cM region that is telomeric to GNAS1.

MATERIALS AND METHODS Study subjects Informed consent was obtained from each subject for participation in these studies. This study was approved by the Joint Committee on Clinical Investigation of The Johns Hopkins University School of Medicine. Criteria used to establish the diagnosis of PHP 1b included (1) evidence of PTH resistance (e.g., hypocalcemia or normocalcemia with elevated serum levels of intact PTH, and when tested, a defective nephrogenous cAMP response to infusion of PTH); (2) absence of features of AHO; (3) normal responsiveness to thyroid-stimulating hormone (TSH) and other hormones; (4) no evidence of vitamin D deficiency or hypomagnesemia; and (5) normal expression(25) or activity of Gs␣(16) and normal sequence of the 13 coding exons and the exon/intron boundaries of the GNAS1 gene. Three multigenerational kindreds (families M, K, and R) with PHP 1b were analyzed in the genome-wide linkage analysis. An additional two kindreds (families P and B) were added for the expanded linkage analysis of regions with LOD scores that were suggestive of linkage. These five families are unrelated and share no common ancestors. When combined, the study population consisted of 20 affected individuals, 30 unaffected individuals, and 3 obligate gene carriers.

425

GNAS1 mutational analysis DNA was extracted from peripheral blood leukocytes by standard methods.(26) Exons 1–13, the flanking intron sequences, and the exon 1 promoter region of the human GNAS1 gene were amplified by the polymerase chain reaction (PCR) as described previously.(27,28) Amplified DNA fragments were analyzed first by polyacrylamide gel electrophoresis to assess the size of the fragments, and then by denaturing gradient gel electrophoresis to detect mutations.(27) Amplified DNA fragments that migrated anomalously were isolated from gels and sequenced directly using the USB Radiolabeled Terminator Cycle Sequencing Kit (USB Corp., Cleveland, OH, USA).

Genotype analysis A genome-wide scan with an average intermarker distance of 10 cM was performed with 408 microsatellite markers derived from the ABI and Weber 6.0 mapping sets (Research Genetics, Huntsville, AL, USA). The X chromosome was not included in our analysis. Additional markers for expanded analysis of chromosome 20q were selected from previously published markers (Marshfield: http:// research.marshfieldclinic.org/genetics),(22) and some genotyping for novel markers (907rep2, 907rep4, 806M20CA, 543J19-TTA)(29,30) on chromosome 20q was kindly provided by Drs M Baptese and H Juppner. Each marker was amplified in a 10-␮l reaction volume containing 0.1 ␮M of each primer, 0.1 mM dNTPs, 10 mM Tris (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.6 U of TaqDNA polymerase (Perkin Elmer, Foster City, CA, USA), and 60 ng DNA. One of each primer pair was labeled with a fluorescent dye (6FAM,TET, HEX). PCR was performed using a GeneAmp 9600 (Perkin Elmer) or a PTC-225 DNA Engine Tetrad (MJ Research, Cambridge, MA, USA) with an initial denaturation of 4 minutes at 94°C followed by a touchdown PCR program(31) consisting of a series of two cycle steps, with the first two cycle step denaturation at 94°C for 20 s, annealing at 62°C for 20 s, and extension at 72°C for 20 s. The annealing temperature in each subsequent two-cycle step is reduced by 1°C until 52°C, at which point 10 cycles are performed with a final 10-minute extension at 72°C. Labeled products were pooled by multiplexing markers of different size ranges and dyes when possible and electrophoresed on a model 373 DNA Sequencer (PE Applied Biosystems, Foster City, CA, USA) on 6% denaturing polyacrylamide gels. Data were collected and analyzed with Genescan software (PE Applied Biosystems) that calculates fragment length in reference to an internal lane standard (Genescan-500 labeled with TAMRA) and quantifies the amount of fluorescence in each fragment. The data were then imported into Genotyper (PE Applied Biosystems) to identify alleles. Genotyping of microsatellite markers was performed by the Methods Development Laboratory and DNA Analysis Facility of Johns Hopkins Genetic Resources Core Facility and by Sequana Therapeutics, Inc. Polymorphisms within the GNAS1 gene were detected by directly sequencing PCR products or subjecting PCR products to denaturing gradient gel electrophoresis.(27) The

426

JAN DE BEUR ET AL. TABLE 1. PRIMERS

FOR

GNAS1 INTRAGENIC POLYMORPHISMS

Primer

Sequence

Polymorphism

Annealing temperature

Size (bp)

Reference

Exon A forward Exon A reverse Promoter forward

5⬘-ACACTCAGTCGCGTCGGCA-3⬘ 5⬘-GAGCAAACGAAAAAGGGAC-3⬘ 5⬘-CCGGGCGCGCGCTCCCTCCCCTTCCGCCCACCCCAG-3⬘ 5⬘-CTCGGCCGGGCCCCCGCCGCCGCCCCGAGCCAAAAG-3⬘ 5⬘-TCTTGTAGCGCCCTCCCA-3⬘ 5⬘-(GC)40TGCCCATGTGCAGGGCTGTCACTCATGTT-3⬘A8 5⬘-CTCTGAGCCCTCTTTCCAAACTAC-3⬘ 5⬘-(GC)40GGTTATTCCAGAGGGACTGGGGTGAA-3⬘

GGCGCC insert at ⫺2495

60°C

490

24

GGCGGC insert at ⫺486

70°C

409

(ATTគ ) to (ATCគ ) at I131

60°C

308

G to A at ⫹17,583

57°C

295

Promoter reverse Exon 5 forward Exon 5 reverse Intron 7 forward Intron 7 reverse

primer sequences, polymorphism locations, annealing temperatures, and expected fragment sizes are listed in Table 1. The genotype data were verified using SIMWALK2 to check for genotyping errors. Questionable genotypes were selected for confirmation by direct sequencing.

Southern blot analysis DNA samples (5–10 ␮g) were digested overnight with an excess of either EcoRI, PstI, or NcoI according to manufacturer’s directions (Life Technologies, Gaithersburg, MD, USA), size fractionated by electrophoresis through agarose gels, and transferred to Nytran Plus nylon membranes (Schleicher & Schuell, Keene, NH, USA). DNA blots were hybridized overnight to the 1.5-kb NcoI-SalI cDNA fragment that contains the complete coding and 3⬘ untranslated regions of GNAS1.(32)

Linkage analyses Two-point linkage analysis was performed using VITESSE,(33) assuming an 80% penetrant dominant model with disease allele frequency 0.0001. Allele frequencies of each marker were estimated from the subjects in this study. To further refine possible regions of linkage indicated from the two-point results, multipoint linkage analysis was performed using SIMWALK2.(34) SIMWALK2 uses a Markov chain Monte Carlo method to generate approximate LOD scores. SIMWALK2 was used because current software programs are unable to compute exact LOD scores with data sets that contain the large number of markers and subjects of families K and R. To test the expected accuracy of SIMWALK2 on families K and R, we compared the approximate results of SIMWALK2 and exact results of GENEHUNTER(35) on the remaining three families, which were small enough to be analyzed by exact methods. Results derived from SIMWALK2 were consistent with those of GENEHUNTER. Family K, however, is the only family that has offspring (KIII-1, KIV-1) of an obligate male carrier (KII-1). Earlier linkage studies(22) and recent studies(11,12) have suggested imprinting of chromosome 20q in the PHP 1b region; therefore, we analyzed the data under the assumption of imprinting for family K. Because currently available software is incapable of simulating an im-

27

printing penetrance model for kindreds as large as family K, we scored individual KII-1 and all his offspring as unknown phenotypic status and scored individuals (KII-3 and KII-5) as obligate carriers. Haplotypes were statistically reconstructed using SIMWALK2, which finds a phased genotype configuration that maximizes the likelihood of data without regard to affection status.

RESULTS Characterization of PHP 1b families Biochemical characterization. Detailed clinical characteristics, biochemical data, and Gs␣ levels for family M have been published previously(2,16,21) (Fig. 1). In some cases, the diagnosis of PHP 1b was confirmed by infusion of PTH (MII-1, MII-2, MII-1, MIII-3, MIII-5), which produced negligible increases in urinary excretion of nephrogenous cAMP. MII-6 had normal serum calcium levels and a mildly elevated intact PTH level. She was classified as unaffected based on a normal PTH infusion test Affected members of family K had biochemical profiles characteristic of PTH resistance (Fig. 2). Two subjects (KII-3, KII-5) were obligate carriers and had normal serum levels of calcium, phosphorous, and PTH, as well as normal nephrogenous cAMP responses to infusion of PTH.(1–34) One affected child (KIV-2) had normal serum calcium levels and an elevated intact PTH level. This subject was classified as affected with later confirmation of the diagnosis by showing absence of methylation of exon 1A (data not shown). Family R displayed an unusual pattern of inheritance in that there were obligate gene carriers and affected individuals in the same sibship.(36) One subject (RII-2) was an obligate gene carrier with two affected children (Fig. 3). She had normal serum calcium, phosphorous, and PTH levels. Her sister (RII-4) had biochemical and clinical features consistent with PHP1b and had an affected child (RIII-3). RII-1 had normal serum calcium but an elevated intact PTH. Unfortunately, we were unable to perform a PTH infusion test to clarify the affected status. Molecular analysis of exon 1A imprinting revealed a normal heterozygous pattern of methylation, thus supporting her classification as unaffected (data not shown). All the affected members of Family P

PSEUDOHYPOPARATHYROIDISM 1b LINKS TO 20q13.3

427

FIG. 1. Family M pedigree and haplotypes. Circles represent females and squares represent males. Symbols for unaffected individuals are shaded gray and symbols for affected individuals are represented by solid figures. Individuals that were not analyzed in this study are represented by open symbols. The serum calcium and parathyroid hormone levels are expressed as either N (normal), 1 (elevated), 2 (reduced), or ND (not done). The pedigree is above the table with the corresponding haplotypes under the column labeled with each individual’s number. Disease haplotypes are shaded in gray. The columns labeled in italics are haplotypes that have been deduced based on the genotype of the offspring.

displayed clinical and biochemical characteristics consistent with PHP 1b. Two individuals had the diagnosis of PHP 1b confirmed by infusion of PTH (PI.1, PII.1; Fig. 4). Affected individuals in family B demonstrated biochemical evidence of PTH resistance (Fig. 5). One of the affected children, BII-1, had normal serum calcium levels and an elevated intact PTH level. Her affected status was confirmed by showing lack of methylation on the maternal allele of exon 1A (data not shown). Gs␣ protein and GNAS1 gene analyses. Erythrocyte Gs␣ protein levels were normal in affected members of families K (KIII.6, KIII.9, KIV.2) and P (PI.1, PII.1), and Gs␣ activity was normal in affected members of family M (MII.1, MII.2, MIII.3, MIII.5). To examine the structural GNAS1 gene for mutations, DNA from an affected member of each family (MII.1, KIII.6, RIII.1, PI.1, BI.1) was used to amplify exons 1–13 plus the intron/exon borders of GNAS1, and PCR products were analyzed by denaturing gradient gel electrophoresis.(27,28) In each case, no mutations were identified. Moreover, Southern blot analysis of GNAS1 excluded large deletions, insertions, or rearrangements in the gene in an affected individual from each kindred (MII.1, KIII.9, RIII.1, PI.1, BI.1).

Pedigree analysis In each of the five kindreds, a dominant mode of inheritance was observed with affected members present in each successive generation (Figs. 1–5). As previously described(22) for other PHP 1b kindreds, we observed a pattern

of inheritance consistent with imprinting of the PHP 1b locus (Fig. 2). Similar to PHP 1a, where hormone resistance derives from maternal transmission of a defective GNAS1 allele, PHP 1b occurs only in those family members who have maternally inherited the disease allele.

Genome-wide scan for linkage To determine the area of the genome linked to the PHP 1b trait, we performed a genome-wide scan in three kindreds (families K, M, and R) using 408 polymorphic microsatellite markers with an average intermarker distance of 10 cM. For each of the 408 markers, two-point LOD scores were calculated, assuming an 80% penetrant dominant model and disease allele frequencies of 0.0001. Initially, nine microsatellite markers had LOD scores that exceeded 1.0. Seven of the nine were flanked by markers with negative LOD scores and had nonparametric p values that were not statistically significant. An area on chromosome 10q contained two markers (D10S212 and D10S2171) with LOD scores that exceeded 2.0. Fine mapping with 13 additional markers and the addition of two kindreds (families P and B) excluded linkage to this region. Thus, secondary analysis with additional markers and/or additional families (P and B) excluded all markers with LOD scores ⬎1.0 except D20S100 on chromosome 20q13. The maximum combined LOD score for this marker in the initial three kindreds was 3.20 (␪ ⫽ 0), with individual contributions of 1.19, 1.73, and 0.28 for families K, M, and R, respectively.

428

JAN DE BEUR ET AL.

FIG. 2. Family K pedigree and haplotypes. Circles represent females and squares represent males. Symbols for unaffected individuals are shaded gray and symbols for affected individuals are represented by solid figures. Individuals that were not analyzed in this study are represented by open symbols and obligate gene carriers without clinical PTH resistance are represented by cross hatched symbols. The serum calcium and parathyroid hormone levels are expressed as either N (normal), 1 (elevated), 2 (reduced), or ND (not done). The pedigree is above the table with the corresponding haplotypes under the column labeled with each individual’s number. Disease haplotypes are shaded in gray. The columns labeled in italics are haplotypes that have been deduced based on the genotype of the offspring.

Fine mapping of chromosome 20q13 and linkage analysis of the PHP 1b region Two-point linkage analysis: To generate a more detailed linkage map of 20q13, we used additional markers and PHP 1b families. Specifically, we performed genotype analysis with 22 additional markers from published genetic maps (http://research.marshfieldclinic.org)(29,30) as well as known(24,27) and novel polymorphic markers (Table 2) in GNAS1. Moreover, we performed analysis in two kindreds (families P and B) with an additional seven affected individuals. Marker D20S902 comprised the centromeric boundary and D20S93 the telomeric boundary of the approximately 20-cM region (Table 2). Four polymorphic markers within the GNAS1 gene were included in the analysis. The order of the markers was confirmed by the Marshfield genetic map and map orders published in the literature.(22,23)

The maximum LOD score obtained for each family was 2.13 for family M (D20S158, ␪ ⫽ 0), 1.51 for family M (AFMa202yb, ␪ ⫽ 0), 1.20 for family R (543J19-TTA, ␪ ⫽ 0), 0.82 for family P (D20S171, ␪ ⫽ 0), and 0.52 for family B (D20S93, ␪ ⫽ 0). The combined maximum LOD score was 5.73 at D20S93 (␪ ⫽ 0.0). Variability in LOD scores is explained by differences in the number of subjects in each family as well as the relative heterozygosity for marker genotypes. There were seven markers (D20S100, D20S158, D20S25, exon 1A, D20S171, AFMa202yb, and D20S93) across the 20-cM region that had LOD scores that exceeded 3 (Table 2). To narrow this broader area of linkage to the minimal region shared by all affected individuals, we performed multipoint linkage analysis incorporating 22 markers on 20q13. A graphic representation of the LOD score obtained from the multipoint analysis is shown in Fig. 6. Based on

PSEUDOHYPOPARATHYROIDISM 1b LINKS TO 20q13.3

429

FIG. 3. Family R pedigree and haplotypes. Circles represent females and squares represent males. Symbols for unaffected individuals are shaded gray and symbols for affected individuals are represented by solid figures. Individuals that were not analyzed in this study are represented by open symbols and obligate gene carriers without clinical PTH resistance are represented by cross hatched symbols. The serum calcium and parathyroid hormone levels are expressed as either N (normal), 1 (elevated), 2 (reduced), or ND (not done). The pedigree is above the table with the corresponding haplotypes under the column labeled with each individual’s number. Disease haplotypes are shaded in gray. The columns labeled in italics are haplotypes that have been deduced based on the genotype of the offspring.

the multipoint linkage analysis, we identified a 5.7-cM region with LOD scores that exceeded 3.0. The boundaries of the region are 907rep-2 at the centromere and D20S93 at the telomere. Because of poor heterozygosity of some of the markers, especially the markers within and immediately flanking GNAS1, and limited informativeness in some of our kindreds, an area of relatively low LOD scores is flanked by a centromeric (907rep2-exon 1A, maximum LOD ⫽ 5.34) and telomeric (exon 5-D20S93, maximum LOD ⫽ 6.55) region of significant LOD scores. Although not significant, the LOD scores are positive in the region between GNAS1 exon 1A and intron 3, and linkage to that region cannot be excluded. Haplotype analysis. Figures 1–5 display the five PHP 1b pedigrees and corresponding haplotypes for chromosome 20q13. The affected haplotypes are shaded gray. Within a family, all affected individuals share the affected haplotype. The absence of the PHP 1b phenotype in subjects who have the affected haplotype may be explained by imprinting (KII-1, KIII-1, KVI-1), variable penetrance (RII-2, RII-6), or conversion (RIV-2), as discussed below.

The disease haplotype is inherited maternally in families M, R, P, and B. In contrast, the disease allele is inherited paternally in family K. The presence of unaffected obligate gene carriers (KII-3, KII-5) in the second generation is consistent with paternal imprinting of the PHP 1b gene locus. The centromeric limit of the PHP 1b locus is defined by a recombinant in family K (KII-5, KIII-6, KIII-9) between D20S451 and 907rep2 (Fig. 2). There are no recombinants that define the telomeric limit of linkage. KIV-2 has a double recombination that narrows the region shared by all the affected individuals to 5 cM between GNAS1 intron 7 and the end of the chromosome.

DISCUSSION Our genome-wide linkage analysis of five multigenerational PHP 1b kindreds confirms linkage to chromosome 20q13.3,(22) and inclusion of an additional 22 markers on chromosome 20q13.3 has allowed us to refine the PHP 1b locus to a 5-cM region extending from GNAS1 intron 7 to the end of the chromosome. Although this region includes

430

FIG. 4. Family P pedigree and haplotypes. Circles represent females and squares represent males. Symbols for unaffected individuals are shaded gray and symbols for affected individuals are represented by solid figures. Individuals that were not analyzed in this study are represented by open symbols. The serum calcium and parathyroid hormone levels are expressed as either N (normal), 1 (elevated), 2 (reduced), or ND (not done). The pedigree is above the table with the corresponding haplotypes under the column labeled with each individual’s number. Disease haplotypes are shaded in gray.

part of the GNAS1 gene, the site of the molecular defect in PHP 1a, extensive phenotypic and molecular characterization of the kindreds we studied documented that affected subjects had PHP 1b as determined by isolated PTH resistance, absence of somatic features of AHO and normal levels of Gs␣ protein in erythrocyte membranes. Moreover, nucleotide sequences of the 13 coding exons of Gs␣ and the respective exon/intron boundaries were normal when analyzed by PCR-based techniques, and large deletions, insertions, or rearrangements of the gene were excluded by restriction endonuclease analysis. These studies exclude defects in the structural gene for Gs␣ as a cause for PHP 1b in these subjects. The variable phenotype observed in our PHP 1b kindreds and other PHP 1b kindreds(22,23) is similar to the inheritance pattern of hormone resistance exhibited by PHP 1a families in which maternal transmission of Gs␣ deficiency results in PHP 1a (AHO and multiple hormone resistance) and paternal transmission of Gs␣ deficiency results in PPHP (AHO without hormone resistance). In the families we studied, PHP 1b developed only in those subjects who inherited the disease haplotype maternally (e.g., MIII-1, MIII-3, MIII-5,

JAN DE BEUR ET AL.

FIG. 5. Family B pedigree and haplotypes. Circles represent females and squares represent males. Symbols for unaffected individuals are shaded gray and symbols for affected individuals are represented by solid figures. Individuals that were not analyzed in this study are represented by open symbols. The serum calcium and parathyroid hormone levels are expressed as either N (normal), 1 (elevated), 2 (reduced), or ND (not done). The pedigrees for each family are above each table with the corresponding haplotypes under the column labeled with each individual’s number. Disease haplotypes are shaded in gray.

KIII-4, KIII-6, KIII-9, KIV-2, RII-4, RIII-1, RIII-2, RIII-3, PII-1, PIII-1, PIII-2, BII-1, BII-2, BII-3). As predicted from a model of paternal imprinting, subjects (e.g., KII-3, KII-5, KIII-1, KIV-1) who inherit the disease haplotype paternally fail to manifest PHP 1b. Family R is an example of the variable penetrance exhibited in this disorder. An affected individual (RII-4), an obligate gene carrier (RII-2), and a phenotypically unaffected individual (RII-6) are present in the same sibship. Based on our current understanding of the imprinting mechanism at this locus, because all three of these subjects (RII-2, RII-4, RII-6) have maternally inherited the affected haplotype, they are expected to exhibit PTH resistance. As we were not able to perform PTH infusions in these individuals, it is possible that they manifest subclinical PTH resistance with normocalcemia and normal PTH levels.

PSEUDOHYPOPARATHYROIDISM 1b LINKS TO 20q13.3

431

TABLE 2. COMBINED TWO-POINT LOD SCORES

FOR

MARKERS

ON

20q

Markers

Distance

Family M

Family K

Family R

Family P

Family B

Combined

␪

D20S902 D20S120 D20S52 D20S100 D20S102 D20S158 D20S149 D20S25 D20S451 907rep2 907rep4 806M20-CA Exon 1A Promoter Intron 3 Exon 5 Intron 7 543J19-TTA D20S171 AFMa202yb D20S173 D20S93

5.2 cM 1.3 cM 0.0 cM 2.2 cM 0.6 cM 0.0 cM 3.0 cM 0.0 cM 5.6 cM to D20S171 40.0 kb 250.0 kb 35.0 kb 2.0 kb 12.5 kb ⬍1 kb 5.5 kb 50.0 kb

0.88 1.17 ⫺0.74 1.73 0.98 2.13 0.88 1.26 0.06 0.36 0.41 0.00 1.27 1.27 ⫺0.16 1.27 1.67 0.00 2.02 1.27 0.55 2.01

0.66 0.97 0.66 1.19 0.16 1.30 0.93 1.27 0.99 0.00 ⫺0.11 1.06 0.92 0.00 ⫺0.05 ⫺0.15 0.03 1.11 1.08 1.51 0.34 1.45

⫺0.12 0.39 1.07 0.28 ⫺0.28 0.23 0.28 0.28 0.37 ⫺0.13 ⫺0.08 0.12 0.88 0.00 ⫺0.03 ⫺0.02 ⫺0.01 1.20 0.70 1.00 0.54 1.14

0.00 0.21 0.28 0.60 0.09 0.56 0.26 0.25 0.26 0.00 0.00 0.00 0.60 0.00 0.29 0.60 0.00 0.00 0.82 0.00 0.60 0.60

0.43 0.21 0.26 0.00 0.00 0.26 0.40 0.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ⫺0.33 0.00 0.00 0.48 0.00 0.51 0.52

1.86 2.96 1.54 3.80 0.96 4.48 2.75 3.31 1.67 0.22 0.22 1.18 3.67 1.27 0.06 1.37 1.70 2.31 5.10 3.78 2.55 5.73

0.05 0.10 0.05 0.00 0.20 0.05 0.10 0.05 0.05 0.10 0.20 0.00 0.00 0.00 0.30 0.00 0.00 0.00 0.00 0.00 0.00 0.00

2.0 cM 0.5 cM

Significant LOD scores (⬎3.0) are shown in bold.

FIG. 6. Multipoint LOD scores. The curve represents the combined multipoint LOD scores for all five families. The marker order is below the graph; the relative distance in cM is on the x axis and the LOD score is on the y axis. A significant LOD score exceeds 3.0.

Spontaneous normocalcemia (i.e., without treatment) is well described in PHP 1a(37) and suggests that measurement of calcium and/or PTH levels may not be sensitive enough to detect mild PTH resistance. Variable penetrance has been well documented in subjects with PHP1a,(4) and one case report describes conversion from PTH sensitivity (PPHP) to PTH resistance (PHP 1a) with age.(38) An apparent example of this phenomenon for PHP 1b is observed in family R, as RIV-2 inherited the maternal disease-related haplotype; however, at 4 years of age, no PTH resistance is evident (Fig. 3). The GNAS1 locus exhibits a complex pattern of reciprocal genomic imprinting within the 47-kb region of the GNAS1 locus; there are at least four alternate first exons that splice onto exons 2–13 to produce three distinct proteins.(12)

The most downstream promoter (exon 1) produces transcripts encoding Gs␣ and is biallelically expressed from both the maternal and paternal alleles in most tissues. Recent studies indicate that the maternal allele is preferentially expressed in the somatotrophic cells of the pituitary,(39) and at least in mice, in proximal renal tubular cells.(13,40) In contrast, promoters for alternative exons encoding NESP55 and XL␣s, located upstream of exon 1, are expressed from the maternal and paternal alleles, respectively.(12) These proteins do not seem to function in receptor-mediated signal transduction and are most highly expressed in neuroendocrine tissues.(41– 43) The fourth alternative first exon, termed exon 1A, lacks an initiator codon, and transcripts containing this exon are unlikely to be translated.(44,45) These transcripts are also expressed exclusively from the paternal allele.(14,24)

432

JAN DE BEUR ET AL.

FIG. 7. Comparison of PHP 1b minimal interval. The markers are arranged from centromeric (left) to telomeric (right) along chromosome 20q13.3. The approximate 9-cM interval delineated by the checked bar represents the minimal interval of linkage to PHP 1b based on haplotype analysis and a single recombinant reported by Juppner et al.(22,23) The hatched bar represents a 5 cM minimal area of linkage in our study based on a single recombination in individual KIV-2. The interval denoted by ** represents the area of linkage reported by Juppner et al.(22,23) if information derived solely from a single recombinant is eliminated. The interval denoted by * represents the minimal interval of linkage in this study supported by two or more recombinant individuals. Note that there is no overlap between the two regions (hatched and checkered bars) when information supported by only single recombinants is included. However, when intervals supported by more than one recombinant are used, there is overlap from marker 907rep2 to D20S171 (5.6 cM) in both studies.

Recently, Lui et al.(24) described a defect in imprinting of the GNAS1 maternal allele in patients with PHP 1b. The genetic mutation(s) responsible for this epigenetic defect remains unknown, but based on the data presented here, it is likely to be located within the region that spans GNAS1 intron 7 to the telomere. This region, defined by a recombination in family K (KIV-2) does not overlap with the region reported by Bastepe et al.(23) (Fig. 7). Previous studies(22,23) have linked PHP 1b to a 9-cM region that lies between DS20s149 and 806M20-CA, with the telomeric limit defined by a recombinant in a single individual. This region is centromeric to the region defined by our analysis and does not overlap (Fig. 7). Our multipoint linkage analysis reveals significant linkage across the whole region from D20S451 (D20S86) to D20S93 (5.7 cM) except a small area between exon 1A and GNAS1 intron 3. If the data from our study and previously reported studies(22,23) are taken together and information derived solely from a single recombinant individual is eliminated, the minimal region shared by all affected individuals is a 5.6-cM region between 907rep2 and D20S171 (Fig. 7). The discrepancies that result from narrowing the linked region with information based on recombination in a single individual may be artifactual or may reflect unrecognized errors in gene mapping or marker order. It is equally likely that a variety of mutations, in different regions of 20q13, may account for PHP 1b, because the variable loss of imprinting in GNAS1(22,23) suggests the existence of epigenetic and genetic heterogeneity. Our current results confirm and refine the PHP 1b locus to a region on 20q13 that included GNAS1. The genetic mutation that disrupts normal maternal imprinting of GNAS1 and accounts for renal-specific PTH resistance remains to be elucidated. Future studies that further refine the PHP 1b locus will no doubt facilitate identification of important regions that influence imprinting and expression of GNAS1.

ACKNOWLEDGMENTS We thank the PHP 1b patients and families that participated in these studies. We gratefully acknowledge Dr Murat

Bastepe for assistance with genotyping of some our families. We are grateful for expert technical assistance from Michael Gostomski, the staff of Methods Development Laboratory and DNA Analysis Facility of Johns Hopkins Genetic Resources Core Facility, and the staff of Sequana Therapeutics, Inc. We thank Dr Alejandro Schaeffer for contributing to linkage analyses. This research is supported by National Institutes of Health Grants R01DK46720 (MAL), M01RR00052 (GCRC), AG16992 (JRO), and T32DK07751(MAL), The Pearl M Stettler Award for Women Physicians (SMJ), and The Clinician Scientist Award from The Johns Hopkins University School of Medicine (SMJ).

REFERENCES 1. Chase LR, Melson GL, Aurbach GD 1969 Pseudohypoparathyroidism: Defective excretion of 3⬘, 5⬘-cyclic-AMP in response to parathyroid hormone. J Clin Invest 48:1832–1844. 2. Levine MA, Eil C, Downs RW Jr, Spiegel AM 1983 Deficient guanine nucleotide regulatory unit activity in cultured fibroblast membranes from patients with pseudohypoparathyroidism type I. A cause of impaired synthesis of 3⬘, 5⬘-cyclic AMP by intact and broken cells. J Clin Invest 72:316 –324. 3. Levine MA, Ahn TG, Klupt SF, Kaufman KD, Smallwood PM, Bourne HR, Sullivan KA, Van Dop C 1988 Genetic deficiency of the alpha subunit of the guanine nucleotide-binding protein Gs as the molecular basis for Albright hereditary osteodystrophy. Proc Natl Acad Sci USA 85:617– 621. 4. Jan de Beur SM, Levine MA 2001 Pseudohypoparathyroidism: Clinical, Biochemical, and Molecular Features. In: Bilezikian JP, Marcus R, Levine MA (eds.) The Parathyroids, 2nd ed. Academic Press, New York, NY, USA, pp. 807– 825. 5. Wolfsdorf JI, Rosenfield RL, Fang VS, Kobayashi R, Razdan AK, Kim MH 1978 Partial gonadotrophin-resistance in pseudohypoparathyroidism. Acta Endocrinol 88:321–328. 6. Marx SJ, Hershman JM, Aurbach GD 1971 Thyroid dysfunction in pseudohypoparathyroidism. J Clin Endocrinol Metab 33:822– 828. 7. Werder EA, Illig R, Bernasconi S, Kind H, Prader A 1975 Excessive thyrotropin-releasing hormone in pseudohypoparathyroidism. Pediatr Res 9:12–16. 8. Albright F, Burnett CH, Smith PH 1942 Pseudohypoparathyroidism: An example of “Seabright-Bantam syndrome.” Endocrinology 30:922–932. 9. Davies SJ, Hughes HE 1993 Imprinting in Albright’s hereditary osteodystrophy. J Med Genet 30:101–103. 10. Wilson LC, Oude Luttikhuis ME, Clayton PT, Fraser WD, Trembath RC 1994 Parental origin of Gs alpha gene mutations in Albright’s hereditary osteodystrophy. J Med Genet 31:835– 839.

PSEUDOHYPOPARATHYROIDISM 1b LINKS TO 20q13.3 11. Hayward BE, Kamiya M, Strain L, Moran V, Campbell R, Hayashizaki Y, Bonthron DT 1998 The human GNAS1 gene is imprinted and encodes distinct paternally and biallelically expressed G proteins. Proc Natl Acad Sci USA 95:10038 –10043. 12. Hayward BE, Moran V, Strain L, Bonthron DT 1998 Bidirectional imprinting of a single gene: GNAS1 encodes maternally, paternally, and biallelically derived proteins. Proc Natl Acad Sci USA 95:15475–15480. 13. Yu S, Yu D, Lee E, Eckhaus M, Lee R, Corria Z, Accili D, Westphal H, Weinstein LS 1998 Variable and tissue-specific hormone resistance in heterotrimeric Gs protein ␣-subunit (Gs␣) knockout mice is due to tissue-specific imprinting of the Gs␣ gene. Proc Natl Acad Sci USA 95:8715– 8720. 14. Liu J, Yu S, Litman D, Chen W, Weinstein LS 2000 Identification of a methylation imprint mark within the mouse Gnas locus. Mol Cell Biol 20:5808 –5817. 15. Yu S, Gavrilova O, Chen H, Lee R, Liu J, Pacak K, Parlow AF, Quon MJ, Reitman ML, Weinstein LS 2000 Paternal versus maternal transmission of a stimulatory G-protein alpha subunit knockout produces opposite effects on energy metabolism. J Clin Invest 105:615– 623. 16. Levine MA, Downs RW Jr, Moses AM, Breslau NA, Marx SJ, Lasker RD, Rizzoli RE, Aurbach GD, Spiegel AM 1983 Resistance to multiple hormones in patients with pseudohypoparathyroidism. Association with deficient activity of guanine nucleotide regulatory protein. Am J Med 74:545–556. 17. Schipani E, Weinstein LS, Bergwitz C, Iida-Klein A, Kong XF, Stuhrmann M, Kruse K, Whyte MP, Murray T, Schmidtke J 1995 Pseudohypoparathyroidism type Ib is not caused by mutations in the coding exons of the human parathyroid hormone (PTH)/PTH- related peptide receptor gene. J Clin Endocrinol Metab 80:1611–1621. 18. Bettoun JD, Minagawa M, Kwan MY, Lee HS, Yasuda T, Hendy GN, Goltzman D, White JH 1997 Cloning and characterization of the promoter regions of the human parathyroid hormone (PTH)/ PTH-related peptide receptor gene: Analysis of deoxyribonucleic acid from normal subjects and patients with pseudohypoparathyroidism type 1b. J Clin Endocrinol Metab 82:1031–1040. 19. Minagawa M, Watanabe T, Kohno Y, Mochizuki H, Hendy GN, Goltzman D, White JH, Yasuda T 2001 Analysis of P3 promoter of human PTH/PTHrp receptor gene in pseuodohypoparathyroidism type 1b. J Clin Endocrinol Metab 86:1394 –1397. 20. Fukumoto S, Suzawa M, Takeuchi Y, Kodama Y, Nakayama K, Ogata E, Matsumoto T, Fujita T 1996 Absence of mutations in parathyroid hormone (PTH)/PTH-related protein receptor complementary deoxyribonucleic acid in patients with pseudohypoparathyroidism type Ib. J Clin Endocrinol Metab 81:2554 –2558. 21. Jan de Beur SM, Ding CL, LaBuda MC, Usdin TB, Levine MA 2000 Pseudohypoparathyroidism 1b: Exclusion of parathyroid hormone and its receptors as candidate disease genes. J Clin Endocrinol Metab 85:2239 –2246. 22. Juppner H, Schipani E, Bastepe M, Cole DE, Lawson ML, Mannstadt M, Hendy G, Plotkin H, Koshiyama H, Koh T, Crawford JD, Olsen BR, Vikkula M 1998 The gene responsible for pseudohypoparathyroidism type 1b is paternally imprinted and maps in four unrelated kindreds to chromosome 20q13.3. Proc Natl Acad Sci USA 95:11798 –11803. 23. Bastepe M, Pincus JE, Sugimoto T, Tojo K, Kanatani M, Azuma Y, Kruse K, Rosenbloom AL, Koshiyama H, Juppner H 2001 Positional dissociation between the genetic mutation responsible for pseudohypoparathyroidism 1b and the associated methylation defect at exon A/B: Evidence for a long-range regulatory element within the imprinted GNAS1 locus. Hum Mol Genet 10:1231–1241. 24. Liu J, Litman D, Rosenberg MJ, Yu S, Biesecker LG, Weinstein LS 2000 A GNAS1 imprinting defect in pseudohypoparathyroidism type IB. J Clin Invest 106:1167–1174. 25. Patten JL, Levine MA 1990 Immunochemical analysis of the alpha-subunit of the stimulatory G-protein of adenylyl cyclase in patients with Albright’s hereditary osteodystrophy. J Clin Endocrinol Metab 71:1208 –1214. 26. Levine MA, Ahn TG, Klupt SF, Kaufman KD, Smallwood PM, Bourne HR, Sullivan KA, Van Dop C 1988 Genetic deficiency of the alpha subunit of the guanine nucleotide-binding protein Gs as the molecular basis for Albright hereditary osteodystrophy. Proc Natl Acad Sci USA 85:617– 621. 27. Miric A, Vechio JD, Levine MA 1993 Heterogeneous mutations in the gene encoding the alpha subunit of the stimulatory G protein of adenylyl cyclase in Albright hereditary osteodystrophy. J Clin Endocrinol Metab 76:1560 –1568.

433

28. Jan de Beur SM, Deng ZC, Ding CL, Levine MA 1998 Amplification of GC rich exon 1 of GNAS1 and identification of three novel missense mutations. Proceedings of the 80th Annual Meeting of the Endocrine Society. New Orleans, LA, USA, p. 62. 29. Bastepe M, Pincus JE, Juppner H 1999 Two frequent tetranucleotide repeat polymorphisms between VAPB and STX16 on chromosome 20q13. Mol Cell Probes 13:449 – 451. 30. Bastepe M, Juppner H 2000 Identification and characterization of two new, highly polymorphic loci adjacent to GNAS1 on chromosome 20q13.3. Mol Cell Probes 14:261–264. 31. Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS 1991 Touchdown. PCR to circumvent spurious priming during gene amplification. Nucleic Acids Res 19:4008. 32. Patten JL, Johns DR, Valle D, Eil C, Gruppuso PA, Steele G, Smallwood PM, Levine MA 1990 Mutation in the gene encoding the stimulatory G protein of adenylate cyclase in Albright’s hereditary osteodystrophy. N Engl J Med 322:1412–1419. 33. O’Connell JR, Weeks DE 1995 The VITESSE algorithm for rapid exact multilocus linkage analysis via genotype set-recoding and fuzzy inheritance. Nat Genet 11:402– 408. 34. Sobel E, Lange K 1996 Descent graphs in pedigree analysis: Applications to haplotyping, location scores, and marker-sharing statistics. Am J Hum Genet 58:1323–1337. 35. Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES 1996 Parametric and nonparametric linkage analysis: A unified multipoint approach. Am J Hum Genet 58:1347–1363. 36. Allen DB, Friedman AL, Greer FR, Chesney RW 1988 Hypomagnesemia masking the appearance of elevated parathyroid hormone concentrations in familial pseudohypoparathyroidism. Am J Med Genet 31:153–158. 37. Drezner MK, Haussner MR 1979 Normocalcemic pseudohypoparathyroidism. Association with normal vitamin D3 metabolism. Am J Med 66:503–508. 38. Barr DG, Stirling HF, Darling JA 1994 Evolution of pseudohypoparathyroidism: An informative family study. Arch Dis Child 70:337–338. 39. Hayward BE, Barlier A, Korbonits M, Grossman AB, Jacquet P, Enjalbert A, Bonthron DT 2001 Imprinting of the G(s)alpha gene GNAS1 in the pathogenesis of acromegaly. J Clin Invest 107:R31–R36. 40. Schwindinger WF, Lawler AM, Gearhart JD, Levine MA 1998 A murine model of Albright hereditary osteodystrophy. Proceedings of the 80th Annual Meeting of the Endocrine Society, New Orleans, LA, USA, p. 480. 41. Pasolli HA, Klemke M, Kehlenbach RH, Wang Y, Huttner WB 2000 Characterization of the extra-large G protein alpha-subunit XLalphas. I. Tissue distribution and subcellular localization. J Biol Chem 275:33622–36. 42. Klemke M, Pasolli HA, Kehlenbach RH, Offermanns S, Schultz G, Huttner WB 2000 Characterization of the extra-large G protein alpha -subunit XLalpha s. II. Signal transducation properties. J Biol Chem 275:33633–33640. 43. Weiss U, Ischia R, Eder S, Lovisetti-Scamihorn P, Bauer R, Fischer-Colbrie R 2000 Neuroendocrine secretory protein 55 (NESP55): Alternative splicing onto transcripts of the GNAS gene and posttranslational processing of a maternally expressed protein. Neuroendocrinology 71:177–186. 44. Swaroop A, Agarwal N, Gruen JR, Bick D, Weissman SM 1991 Differential expression of novel Gs␣ signal transduction protein cDNA species. Nucleic Acids Res 17:4725– 4729. 45. Ishikawa Y, Bianchi C, Nadal-Ginard B, Homcy CJ 1990 Alternative promoter and 5⬘ exon generate a novel Gs alpha mRNA. J Biol Chem 265:8458 – 8462. arsid16517762

Address reprint requests to: Suzanne Jan de Beur, MD Johns Hopkins University School of Medicine Division of Endocrinology and Metabolism 4940 Easten Avenue, B114 Baltimore, MD 21224, USA E-mail: [email protected] Received in original form April 4, 2002; in revised form July 9, 2002; accepted September 4, 2002.