Do broad GNAS methylation defects modify the Albright's hereditary osteodystrophy phenotypes in pseudohypoparathyroidism type 1?

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: MON 199-237-Disorders of Parathyroid Hormone & Calcium Homeostasis
Monday, June 17, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board MON-204
Toshikatsu Mitsui*1, Akiko Yuno2, Takeshi Usui3, Akie Nakamura4, Toshihiro Tajima5, Keisuke Nagasaki6, Satoshi Narumi1, Tomohiro Ishii7 and Tomonobu Hasegawa8
1Keio University School of Medicine, Tokyo, Japan, 2Kin-i-kyo Chuo Hospital, Sapporo, Japan, 3National Hospital Organization Kyoto Medical Center, Kyoto, Japan, 4Hokkaido University School of Medicine, Sapporo, Japan, 5Hokkaido Univ Sch of Med, Hokkaido, Japan, 6Niigata University Graduate School of Medicine and Dental Sciences, Niigata, Japan, 7Keio Univ, Tokyo, Japan, 8Keio Univ Sch of Medicine, Tokyo, Japan
Background: Pseudohypoparathyroidism (PHP) type 1 is characterized by PTH resistance and classified as type 1a or 1b, by the presence or absence of Albright’s hereditary osteodystropy (AHO).  AHO phenotypes include short stature, obesity, round face, brachydactyly, and subcutaneous calcification.

   It has been well accepted that GNAS mutations cause PHP type 1a, while methylation defects at the differentially methylated regions (DMRs) of GNAS lead to PHP type 1b.  Recent studies have shown two types of the GNAS methylation defect: the localized type and the broad type.  The localized type is characterized by the localized hypomethylation at exon A/B DMR with STX16 deletion, while the broad type by methylation defects at multiple GNAS DMRs (exon A/B, XL, AS, NESP) with negative genetic analyses.  Interestingly, the broad type has been observed in both PHP type 1a and 1b.  Patients with broad type show different methylation patterns of the four GNASDMRs.  However, it is not clear whether these patterns of methylation defects contribute to the variation of AHO phenotypes. 

   In this study, we integrated a cohort of PHP type 1 patients to evaluate the correlation between epigenotype (methylation status) and AHO phenotypes.

Materials and Methods: The study subjects were 16 PHP type 1 patients from 16 families.  None had gene mutations in GNAS nor PRKAR1A, and  submicroscopic deletions in GNAS locus including STX16.  We further divided 16 patients, based on the number of AHO phenotypes, into following three groups to evaluate the correlation between epigenotype and AHO phenotypes.

- “AHO” group (two or more phenotypes) (n=3)

- “Equivocal AHO” group (one phenotype) (n=4)

- “No AHO” group (n=9).

Epigenotype was determined as follows.  First, we analyzed uniparental disomy by microsatellite markers and SNP array CGH.  Second, we analyzed methylation status of four GNASDMRs by Methylation-Specific MLPA and MassARRAY.

Results: Fifteen out of 16 patients had following two patterns of broad GNASmethylation defect.

- Hypo-all: Hypomethylations at all four GNASDMRs (n=6)

- Hypo/Hyper: Hypomethylations at three DMRs and hypermethylation at NESP DMR (n=9)

In “AHO”, “Equivocal AHO”, and “No AHO” group, we identified 1 Hypo-all and 1 Hypo/Hyper, 1 Hypo-all and 3 Hypo/Hyper, 4 Hypo-all and 5 Hypo/Hyper, respectively.

Two patterns of broad methylation defects, Hypo-all and Hypo/Hyper, did not contribute to the variation of AHO phenotypes.  Additionally, methylation defects were not associated with specific AHO phenotypes.

Conclusions: There was no correlation between epigenotype patterns and AHO phenotypes. This indicates that broad GNAS methylation defects are not phenotypic modifying factor of AHO.

Nothing to Disclose: TM, AY, TU, AN, TT, KN, SN, TI, TH

*Please take note of The Endocrine Society's News Embargo Policy at