Gigantism and acromegaly due to Xq26 microduplications and GPR101 mutation

Abstract

Background: Increased secretion of growth hormone leads to gigantism in children and acromegaly in adults; the genetic causes of gigantism and acromegaly are poorly understood.

Methods: We performed clinical and genetic studies of samples obtained from 43 patients with gigantism and then sequenced an implicated gene in samples from 248 patients with acromegaly.

Results: We observed microduplication on chromosome Xq26.3 in samples from 13 patients with gigantism; of these samples, 4 were obtained from members of two unrelated kindreds, and 9 were from patients with sporadic cases. All the patients had disease onset during early childhood. Of the patients with gigantism who did not carry an Xq26.3 microduplication, none presented before the age of 5 years. Genomic characterization of the Xq26.3 region suggests that the microduplications are generated during chromosome replication and that they contain four protein-coding genes. Only one of these genes, GPR101, which encodes a G-protein-coupled receptor, was overexpressed in patients' pituitary lesions. We identified a recurrent GPR101 mutation (p.E308D) in 11 of 248 patients with acromegaly, with the mutation found mostly in tumors. When the mutation was transfected into rat GH3 cells, it led to increased release of growth hormone and proliferation of growth hormone-producing cells.

Conclusions: We describe a pediatric disorder (which we have termed X-linked acrogigantism [X-LAG]) that is caused by an Xq26.3 genomic duplication and is characterized by early-onset gigantism resulting from an excess of growth hormone. Duplication of GPR101 probably causes X-LAG. We also found a recurrent mutation in GPR101 in some adults with acromegaly. (Funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development and others.).

Figure 1. Familial and Sporadic Cases of Gigantism and Male and Female Growth Patterns Due to the Xq26.3 Microduplications

Panel A shows Patient F1C, who has familial gigantism, at the age of 3 years. His growth chart up to 24 months of age shows the rapid acceleration in weight, although the acceleration in height did not begin until after his second birthday (Fig. S4 in the Supplementary Appendix). Panel B shows an unaffected mother and her daughter (Patient S6), who has sporadic gigantism and whose height was 120 cm at the age of 3 years. A growth chart for Patient S4 (Panel C), another girl with sporadic gigantism, illustrates the typical early increase in height and weight seen in patients with Xq26.3 microduplications, starting at the age of 6 months in this child.

Figure 2. Summary of the Genomic Gains on Chromosome Xq26.3

Shown are 10 different Xq26.3 microduplications, as seen on array comparative genomic hybridization, that were found in 12 patients with familial or sporadic gigantism (with the inheritance pattern indicated at right). Duplicated genomic segments (red) and nonduplicated segments (white) are shown. The genomic coordinates are provided at base-pair resolution on the x axis. The two smallest regions of overlap (SRO), SRO1 and SRO2, are identified, showing the genomic contents in the corresponding regions. The symbols next to the gene names represent the structure of the genes, with vertical lines representing exons and horizontal lines (with or without arrows) representing introns. Adapted from the UCSC Genes track in the UCSC Genome Browser.

Figure 3. Imaging and Histopathological Findings in Patients with Xq26.3 Microduplications

Panels A through D show progressive changes from normal pituitary tissue (Panel A) to adenoma (Panel D), as indicated by reticulin staining of the pituitary gland in Patient F1C. In Panel A, normal pituitary gives way to the expanded hyperplastic acini (Panel B), and in Panel C, areas of transformation are evident (circled) with enlarged, hyperplastic, confluent acini that are caused by breakdown of reticulin fibers and that lead to adenoma (Panel D) with disruption of the reticulin fiber network. Increased GPR101 expression was observed in five tested patients with Xq26 microduplications, whereas there is little if any expression in normal pituitary tissue or growth hormone–producing tumors without Xq26.3 microduplications or GPR101 defects (see also Fig. S7 in the Supplementary Appendix); an example is shown here (Panels E through G) from Patient S3. When the staining of growth hormone (Panel E) and the staining of GPR101 (Panel F) are merged, GPR101 seems to be expressed in some of the growth hormone–secreting cells (Panel G, arrows) but not in all such cells. Nuclei (blue) were stained with DAPI. Panel H shows a sagittal view of a macroadenoma on magnetic resonance imaging of Patient S5 with the Xq26.3 microduplication.

Figure 4. Expression of GPR101 in Pituitary Tissue from Children with Xq26.3 Microduplications

The expression of GPR101 in pituitary tissue from children carrying Xq26.3 microduplications was increased by a factor as high as 1000, as compared with the expression in unaffected pituitary tissue (in five samples [NP1 through NP5] obtained on autopsy) and in pituitary tumors from two patients with sporadic acromegaly (GH1 and GH2) who tested negative for the microduplication (Panel A). These findings, which were obtained on quantitative reverse-transcriptase–polymerase-chain-reaction (qRT-PCR) assay and normalized by a housekeeping gene, contrast with those for two other genes, ARHGEF6 (Panel B) and RBMX (Panel C), in the duplicated stretch of DNA; neither of these two genes showed up-regulated expression. Also shown are cell proliferation (Panel D), growth hormone secretion (Panel E), and activation of DNA sequences called cyclic AMP response elements (CRE) (Panel F) in rat GH3 cells transfected with mutant (p.E308D and p.A397K) and nonmutant GPR101 constructs. Values for cells transfected with empty (control) vector were set at 1. Also shown are values for untreated cells (vehicle) and forskolin (which increases CRE activation). Data are expressed as the mean results of three to five independent experiments, each of which was performed in triplicate. The T bars indicate standard deviations. One asterisk denotes P<0.05, two asterisks P<0.01, and three asterisks P<0.001.

Figure 5. Effect of the p.E308D Mutation in GPR101 in 11 Patients with Sporadic Acromegaly

Panel A shows the sequence for GPR101 in growth hormone–producing pituitary tumors obtained from patients with sporadic acromegaly, as compared with normal tissue. Panel B shows results for a patient with a somatic mutation, which was determined by the presence of the mutation in the GPR101 sequence of DNA in the tumor sample but not in the sequence in peripheral-blood mononuclear cells. None of the 13 families with familial isolated pituitary adenomas carried the p.E308D mutation in GPR101. Panel C shows a structural model of GPR101 bearing the p.E308D mutation. Residue A397 is located at the cytosolic end of transmembrane (TM) 6 of GPR101. The mutated D308 residue and the nonmutated A397 residue are shown in space-filling representation and colored according to elements, with carbon atoms in gray, oxygen atoms in red, and nitrogen atoms in blue. The backbone of the receptor and the G protein heterotrimer is schematically represented as a ribbon, with the receptor shown with a spectrum of colors that ranges from red at the N-terminal to purple at the C-terminal; the α, β, and γ subunits of the G protein are in gray, blue, and pink, respectively. The cytosolic ends of TM 5 and TM 6 and intracellular loop (IL) 3, which connects them, are indicated by labels. The blue arrows show directions of the β-sheet domains of the β subunit of the G protein.