9
The Growth Hormone Receptor (GHR) c.899dupC Mutation Functions as a Dominant Negative: Insights into the Pathophysiology of Intracellular GHR Defects Michael A. Derr, Javier Aisenberg, Peng Fang, Yardena Tenenbaum-Rakover, Ron G. Rosenfeld, and Vivian Hwa Department of Pediatrics (M.A.D., P.F., R.G.R., V.H.), Oregon Health & Science University, Portland, Oregon 97239; Pediatric Endocrinology (J.A.), Hackensack University Medical Center, Hackensack, New Jersey 07601; and Ha’Emek Medical Center (Y.T.-R.), Afula and the Technion Faculty of Medicine, Haifa 18101, Israel Context: GH insensitivity (GHI) is a condition characterized by pronounced IGF-I deficiency and severe short stature. We previously identified a novel compound heterozygous GH receptor (GHR) mutation, GHR:p.R229H/c.899dupC, in a patient presenting with GHI. The heterozygous p.R229H (prepeptide) variant was previously associated with GHI despite a lack of adequate functional studies. The novel heterozygous GHR:c.899dupC variant affects the critical JAK2-binding Box 1 region of the GHR intracellular domain; the duplication predicted a frameshift and early protein termination. Objective: The individual and synergistic effect(s) of the p.R229H and c.899dupC mutations on GHR function(s) were evaluated in reconstitution studies. Results: The recombinant human GHR (hGHR):p.R229H variant was readily expressed, and unex- pectedly, GH-induced signal transducer and activator of transcription 5b (STAT5b) phosphorylation was comparable to that induced by wild-type hGHR. The truncated, immunodetected hGHR: c.899dupC variant, in contrast, was unresponsive to GH. To mimic a compound heterozygous state, the two variants were coexpressed, and strikingly, the presence of the hGHR:c.899dupC effectively abolished the GH-induced STAT5b activities that were observed with hGHR:p.R229H alone. Fur- thermore, hGHR:c.899dupC dose-dependently reduced the GH-induced STAT5b activities associ- ated with hGHR:p.R229H. This dominant negative effect was also observed when hGHR:c.899dupC was coexpressed with wild-type hGHR. Conclusion: The p.R229H variant, contrary to an earlier report, appeared to function like wild-type GHR and, therefore, is unlikely to cause GHI. The c.899dupC variant is a novel dominant negative mutation that disrupted normal GHR signaling and is the cause for the GHI phenotype of the reported patient. (J Clin Endocrinol Metab 96: E1896 –E1904, 2011) G H insensitivity (GHI) is characterized by postnatal growth retardation, normal to elevated serum levels of GH, and low serum concentrations of IGF-I. The clin- ical presentations of patients with GHI have been linked to an array of mutations or deletions in genes involved in the GH-IGF-I axis (1, 2), including the GH receptor (GHR) (3), signal transducer and activator of transcription 5B (STAT5B) (4, 5), the acid-labile subunit (IGFALS) (6), and the IGF1 genes (7). Approximately 70 different mo- lecular defects have been identified in the GHR gene in over 300 patients presenting with GHI, highlighting the importance of the GHR in human postnatal growth (2, 3). The human GHR (hGHR) is a glycosylated homodi- meric transmembrane protein. In addition to the signal ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2011 by The Endocrine Society doi: 10.1210/jc.2011-1597 Received May 26, 2011. Accepted August 17, 2011. First Published Online September 7, 2011 Abbreviations: GHBP, GH binding protein; GHI, GH insensitivity; GHR, GH receptor; GHRE, GH response element; hGH, human GH; hGHR, human GHR; JAK2, Janus kinase 2; STAT5B, signal transducer and activator of transcription 5B; WT, wild-type. JCEM ONLINE Advances in Genetics—Endocrine Research E1896 jcem.endojournals.org J Clin Endocrinol Metab, November 2011, 96(11):E1896 –E1904 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Derr MA JCEM 2011

Embed Size (px)

Citation preview

Page 1: Derr MA JCEM 2011

The Growth Hormone Receptor (GHR) c.899dupCMutation Functions as a Dominant Negative: Insightsinto the Pathophysiology of Intracellular GHR Defects

Michael A. Derr, Javier Aisenberg, Peng Fang, Yardena Tenenbaum-Rakover,Ron G. Rosenfeld, and Vivian Hwa

Department of Pediatrics (M.A.D., P.F., R.G.R., V.H.), Oregon Health & Science University, Portland,Oregon 97239; Pediatric Endocrinology (J.A.), Hackensack University Medical Center, Hackensack, NewJersey 07601; and Ha’Emek Medical Center (Y.T.-R.), Afula and the Technion Faculty of Medicine, Haifa18101, Israel

Context: GH insensitivity (GHI) is a condition characterized by pronounced IGF-I deficiency andsevere short stature. We previously identified a novel compound heterozygous GH receptor (GHR)mutation, GHR:p.R229H/c.899dupC, in a patient presenting with GHI. The heterozygous p.R229H(prepeptide) variant was previously associated with GHI despite a lack of adequate functionalstudies. The novel heterozygous GHR:c.899dupC variant affects the critical JAK2-binding Box 1region of the GHR intracellular domain; the duplication predicted a frameshift and early proteintermination.

Objective: The individual and synergistic effect(s) of the p.R229H and c.899dupC mutations on GHRfunction(s) were evaluated in reconstitution studies.

Results: The recombinant human GHR (hGHR):p.R229H variant was readily expressed, and unex-pectedly, GH-induced signal transducer and activator of transcription 5b (STAT5b) phosphorylationwas comparable to that induced by wild-type hGHR. The truncated, immunodetected hGHR:c.899dupC variant, in contrast, was unresponsive to GH. To mimic a compound heterozygous state,the two variants were coexpressed, and strikingly, the presence of the hGHR:c.899dupC effectivelyabolished the GH-induced STAT5b activities that were observed with hGHR:p.R229H alone. Fur-thermore, hGHR:c.899dupC dose-dependently reduced the GH-induced STAT5b activities associ-ated with hGHR:p.R229H. This dominant negative effect was also observed when hGHR:c.899dupCwas coexpressed with wild-type hGHR.

Conclusion: The p.R229H variant, contrary to an earlier report, appeared to function like wild-typeGHR and, therefore, is unlikely to cause GHI. The c.899dupC variant is a novel dominant negativemutation that disrupted normal GHR signaling and is the cause for the GHI phenotype of thereported patient. (J Clin Endocrinol Metab 96: E1896–E1904, 2011)

GH insensitivity (GHI) is characterized by postnatalgrowth retardation, normal to elevated serum levels

of GH, and low serum concentrations of IGF-I. The clin-ical presentations of patients with GHI have been linked toan array of mutations or deletions in genes involved in theGH-IGF-I axis (1, 2), including the GH receptor (GHR)(3), signal transducer and activator of transcription 5B

(STAT5B) (4, 5), the acid-labile subunit (IGFALS) (6),and the IGF1 genes (7). Approximately 70 different mo-lecular defects have been identified in the GHR gene inover 300 patients presenting with GHI, highlighting theimportance of the GHR in human postnatal growth (2, 3).

The human GHR (hGHR) is a glycosylated homodi-meric transmembrane protein. In addition to the signal

ISSN Print 0021-972X ISSN Online 1945-7197Printed in U.S.A.Copyright © 2011 by The Endocrine Societydoi: 10.1210/jc.2011-1597 Received May 26, 2011. Accepted August 17, 2011.First Published Online September 7, 2011

Abbreviations: GHBP, GH binding protein; GHI, GH insensitivity; GHR, GH receptor; GHRE,GH response element; hGH, human GH; hGHR, human GHR; JAK2, Janus kinase 2;STAT5B, signal transducer and activator of transcription 5B; WT, wild-type.

J C E M O N L I N E

A d v a n c e s i n G e n e t i c s — E n d o c r i n e R e s e a r c h

E1896 jcem.endojournals.org J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 2: Derr MA JCEM 2011

peptide (residues 1–18), the hGHR can be structurallydivided into three domains: an extracellular domain, en-coded by exons 2–7 (residues 19–264); a transmembranedomain, encoded by exon 8 (residues 265–288); and anintracellular domain, encoded by exons 9 and 10 (residues289–638). The extracellular domain of the GHR consistsof two functional subdomains (8). Subdomain 1, residues19–141, is involved in GH binding. Subdomain 2, residues146–264, is involved in receptor dimerization and GH-induced receptor rotation (9). In addition, the extracellu-lar region can be proteolytically cleaved to circulate inserum as the GH binding protein (GHBP). Abnormallylow serum levels of GHBP in patients with GHI can be,therefore, a strong indicator of a mutation in the GHRgene.

The hGHR intracellular domain features a proline-richBox 1 motif required for Janus kinase 2 (JAK2) associa-tion. Upon GH binding to GHR, JAK2 becomes activatedand phosphorylates GHR intracellular tyrosines Y534,Y566, and Y627 (10), which then act as docking sites forsignaling molecules, such as cytosolic STAT5b. JAK2phosphorylates the docked STAT5b on a single tyrosine(Tyr699), and the phosphorylated STAT5b subsequentlydissociates from GHR and forms homodimers. Homodi-meric phosho-STAT5b translocates to the nucleus, whereit binds DNA and regulates gene transcription, includingthe IGF1 gene. Genetic defects affecting the GHR intra-cellular domain are rare, with only seven proven muta-tions out of the 70 GHR mutations identified to date (2, 3).Interestingly, all seven mutations involve frameshifts thatresult in early protein terminations.

We previously reported compound heterozygous GHRmutations in a patient with GHI (11). The child presentedwith a height of �4.07 SD score (SDS), normal basal serumGH levels (8 ng/ml), and abnormally low serum concen-trations of IGF-I (16 ng/ml; normal range, 54–178 ng/ml).GHBP was well within the normal range (1379 pmol/liter;normal range, 267-1638 pmol/liter), indicating an atypi-cal GHI phenotype. The compound heterozygous GHRmutations subsequently identified consisted of a previ-ously reported missense mutation, p.R229H (mature pep-tide, R211H) (12) located in subdomain 2 of the extra-cellular domain, and a novel heterozygous duplication ofnucleotide 899 (c.899dupC) within the intracellular Box 1motif, which would predict a frameshift and early proteintermination (12).Wenowpresent functional evidence thatc.899dupC is a dominant negative mutation and is, byitself, the likely etiology for the GHI and IGF deficiencyphenotype observed in the patient. Combining these find-ings with our previously published studies of the criticaland redundant GHR intracellular tyrosines involved inSTAT5b phosphorylation (10) permits development of a

hypothesis on the basis of dominant negative intracellularGHR mutations.

Materials and Methods

AntibodiesAntibodies used were as follows: anti-phospho-Stat5(Tyr694),

anti-phospho-p44/p42-MAPK(Thr202/Tyr204)(197G2), anti-phos-pho-Stat3(Tyr705)(D3A7), anti-Stat3, and anti-phospho-Stat1, allfrom Cell Signaling Technology (Beverly, MA); anti-Stat5b(G2),anti-Stat1-p84/p91(c-136), anti-GHR(S-16), anti-GHR(H300), allpurchased from Santa Cruz Biotechnology (Santa Cruz, CA); anti-ERK1/2, from Upstate Biotechnology (Lake Placid, NY); and an-timouse IgG and antirabbit IgG, from Amersham-Pharmacia Bio-tech (Uppsala, Sweden).

hGHR cDNA variantsFull-length hGHR in pcDNA1/AMP expression plasmid (des-

ignated hGHR), kindly provided by Dr. Richard Ross (Univer-sity of Sheffield, Sheffield, UK) (13), was employed to generatethe GHR variants by QuickChange II site-directed mutagenesisfollowing the recommended protocol (Agilent Technologies,Stratagene, La Jolla, CA). Primers used to generate the variantswere: GHR-R229H-forward, 5�-GGATAAGGAATATGAAG-TGCATGTGAGATCCAAACAACG-3�; GHR-R229H-reverse,5�-CGTTGTTTGGATCTCACATGCACTTCATATTCCTTA-TCC-3�; GHR-899dupC-forward, 5�-GCTGATTCTGC-CCCCCAGTTCCAGTTCCAAAG-3�; and GHR-899dupC-reverse, 5�-CTTTGGAACTGGAACTGGGGGGCAGAAT-CAGC-3�. All resulting hGHR variants were confirmed by DNAsequencing.

Cell culture and cell transfectionHEK293 cells were maintained in DMEM (Cellgrow; Medi-

atech, Herndon, VA) supplemented with 10% fetal bovine serum(Invitrogen Life Technologies, Inc., Grand Island, NY) at 37 C in5% CO2. For reconstitution studies, HEK293 cells were tran-siently transfected with vector, pcDNA3.1, or the relevanthGHR variants, as indicated, using TransIT-LT1 (Mirus, Mad-ison, WI) (14). After 24-h transfection, the cells were starved inDMEM supplemented with 0.1% BSA for 16 h before treatmentwith recombinant human GH (hGH) (Lilly, Fegersheim, France)as indicated. All transfection experiments were performed atleast three independent times.

Western immunoblot analysisFor Western blot analysis, cells were treated with indicated

concentrations of recombinant hGH for 20 min. Cell lysateswere solubilized in Triton X-100 lysis buffer as previously de-scribed (15). Equal quantities of protein (protein assay; Bio-RadLaboratories, Hercules, CA), 35 or 50 �g, were size fractionatedon reducing 7 or 10% sodium dodecyl sulfate-polyacrylamidegels and electroblotted onto nitrocellulose membranes. Mem-branes were blocked with 5% nonfat dry milk in Tris-bufferedsaline/0.1% Tween 20. Western blots were processed with theappropriate primary and secondary antibodies, following themanufacturers’ protocols, and visualized by enhanced chemilu-minescence (PerkinElmer Life Sciences, Inc., Boston, MA).Membranes were stripped using a buffer consisting of 62.5 mM

J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904 jcem.endojournals.org E1897

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 3: Derr MA JCEM 2011

Tris (pH 6.8), 2% sodium dodecyl sulfate, and 0.8% 2-mercap-toethanol for 15 min at 80 C, before reprobing with other pri-mary antibodies.

Luciferase reporter assaysThe luciferase reporter construct carrying 8 � GH response

element (GHRE) from the rat Spi2.1 gene in pGL2 (pGHRE-LUC) was a generous gift from Dr. Peter Rotwein (OregonHealth & Science University, Portland, OR). HEK293 cells weretransfected for 24 h with a total input DNA of 2 �g: 1 �gpGHRE-LUC plus vector or specified hGHR constructs as in-dicated. Transfected HEK293 cells were treated with recombi-nant hGH for 18 h, and cell lysates were subsequently analyzedfor reporter activity (luciferase assay system; Promega Corp.,Madison, WI), following the manufacturers’ protocol. Lu-ciferase activities were measured with a luminometer (Bio TekInstruments Inc., Winooski, VT) and normalized to total proteinconcentration. Each experiment was performed at least threeindependent times in duplicate.

Results

The hGHR:R229H variant is readily expressed anddoes not affect GHR-mediated signaling

The hGHR:R229H variant was previously reported as aheterozygous mutation associated with GHI (12). The priorstudy alluded to regenerating only the extracellular portionof hGHR carrying the R229H variant, and the expression ofthis construct was reported to be extremely poor, leading tothe hypothesis that heterozygous hGHR:p.R229H func-tioned as a dominant negative or that the second, wild-type(WT) allele was poorly expressed (12). To determinewhether this hypothesis could explain the phenotype ob-served in our patient, R229H was regenerated, employingfull-lengthhGHR cDNA. In reconstitution studies, the effectof homozygous GHR:p.R229H on GH-induced signalingwas evaluated. As shown in Fig. 1A, recombinant full-lengthhGHR:p.R229H was expressed at levels similar to the ma-ture fully glycosylated WT hGHR (Fig. 1A, arrow), and thefull glycosylation, furthermore, was indicative of correcttranslocation to the cell surface (16, 17). To evaluate thefunctionality of the expressed hGHR:p.R229H variant,analysisofGH-inducedsignal transductionwasundertaken.GH-induced phosphorylation of STAT5b was essentiallyidentical between cells transfected with hGHR:p.R229H orwith WT hGHR (Fig. 1A). Of the other GH-GHR-mediatedpathways (STAT-1, STAT-3, and ERK1/2), there were also noobservable differences between WT and variant GHR. Corre-latingwiththesignalingprofiles,GH-induced,STAT5b-depen-dent, gene transcriptional activities (assessed by luciferasereporter assays) were indistinguishable between cells ex-pressing hGHR:p.R229H or WT hGHR (Fig. 1B).

The hGHR:c.899DupC variant cannot activateGH-induced signaling pathways

The novel heterozygous hGHR:c.899dupC mutation,in exon 9, is of particular interest because mutations inexons 9 and 10, encoding the intracellular domain of theGHR, are rare. Exon 9 encodes for 23 amino acid residues,nine of which correspond to amino acids 297–305 of theBox 1 motif. Box 1 is a proline-rich region known to becritical for JAK2 association. The duplication of nucleo-

FIG. 1. Analysis of GH-induced STAT5b activation in HEK293 cellsexpressing the hGHR:p.R229H variant. The mutation was regenerated inthe hGHR cDNA for reconstitution studies. A, Western immunoblotanalysis for GH-induced tyrosyl-phosphorylated STAT5b, STAT3, STAT1,and ERK1/2. Cell lysates analyzed were from HEK293 cells transfectedwith pcDNA3.1, WT hGHR, or variant hGHR:p.R229H, untreated ortreated with recombinant hGH as indicated. Antibodies employed forthe sequential immunoblot analysis are indicated on the left. Thephosphorylated form of STAT5b and the fully glycosylated hGHR areindicated by arrows. B, Analysis of GH-induced, pSTAT5b-dependent,GHRE(Spi2.1)-luciferase reporter activities. HEK293 cells werecotransfected with pGHRE-LUC and the indicated constructs. GHtreatment as indicated.

E1898 Derr et al. GHR c.899dupC Dominant Negative Mutation J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 4: Derr MA JCEM 2011

tide 899 predicts a frame-shift that retains proline at po-sition 300 of the GHR prepeptide, the addition of six newresidues, and the introduction of a stop codon (11). Todetermine whether such a truncated GHR variant could beexpressed, the homozygous mutation was regenerated.When reconstituted in HEK293 cells, the mature, fullyglycosylated form was readily expressed (Fig. 2A) and cor-rectly localized. Impact on GH-induced signaling indi-cated that, as might be predicted from loss of the Box 1motif, GH-induced STAT5b phosphorylation was com-pletely abrogated (Fig. 2A). GH-induced phosphorylationof STAT1 and STAT3, furthermore, was also abrogated(Fig. 2A). Results from the Luciferase reporter assays cor-related with the immunoblot analysis because GH-in-duced, STAT5b-dependent transcriptional activities in

cells expressing the hGHR:c.899dupC variant were com-parable to cells expressing vector (Fig. 2B).

Coexpression of hGHR:c.899dupC with eitherhGHR:p.R229H or WT hGHR severely compromisedGH-induced STAT5b phosphorylation

As shown above, the homozygous expression of eachvariant revealed that hGHR:p.R229H was functionallyindistinguishable from WT hGHR, whereas hGHR:c.899dupC was incapable of mediating GH-inducedsignaling. To mimic the heterozygous state found in thepatient, the hGHR:p.R229H and hGHR:c.899dupCvariants were coexpressed in a 1:1 ratio, and GH-in-duced STAT5b phosphorylation was analyzed. Cells co-expressing both variants exhibited markedly reduced lev-els of STAT5b phosphorylation (Fig. 3, lanes 14 and 21),compared with cells expressing hGHR:p.R229H, andonly slightly higher than that detected in cells trans-fected with vector. Interestingly, coexpression of hGHR:c.899dupC with WT hGHR, (Fig. 3, lanes 12 and 19), sim-ilarly showed severe inhibition of GH-induced STAT5bphosphorylation. GH-induced STAT5b phosphorylationin cells coexpressing hGHR:p.R229H and WT hGHR(Fig. 3, lanes 13 and 20) was the same as cells expressingeither WT hGHR or hGHR:p.R229H alone. Altogether,these results demonstrate that the hGHR:c.899dupC vari-ant inhibited STAT5b signaling in a dominant negativemanner when coexpressed with either hGHR:p.R229H orWT hGHR.

The hGHR:c.899DupC variant dose-dependentlyreduces GH-induced STAT5b activities

To further characterize the dominant negative proper-ties of the hGHR:c.899dupC variant, GH-inducedSTAT5b activation was evaluated in cells cotransfectedwith hGHR:p.R229H (1 �g) and increasing increments ofhGHR:c.899dupC (0.25–1 �g). As the dose of transfectedhGHR:c.899dupC increased,GH-inducedSTAT5bphosphor-ylation decreased (Fig. 4A). At a 1:1 ratio (Fig. 4A, lane 16),STAT5b phosphorylation was, again, only slightly higher thanin cells transfected with vector only. Consistent with the immu-noblot analysis, STAT5b-dependent transcriptional activitieswere dose-dependently reduced with increasing amounts oftransfected hGHR:c.899dupC (Fig. 4B).

Discussion

Of the more than 70 GHR mutations that have been iden-tified to date, 34 are missense mutations (2). The over-whelming majority of the GHR missense mutations, 31 of34, have been found in the extracellular domain, with 17located in subdomain 2, a region rich in �-sheet structures.

FIG. 2. Analysis of GH-induced STAT5b activation in HEK293 cellsexpressing the hGHR:c.899dupC variant. The mutation wasregenerated in the hGHR cDNA for reconstitution studies in HEK293cells. A, Immunoblot analysis for GH-induced tyrosyl-phosphorylatedSTAT5b, STAT3, and STAT1. B, Analysis of GH-induced, pSTAT5b-dependent, GHRE(Spi2.1)-luciferase reporter activities.

J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904 jcem.endojournals.org E1899

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 5: Derr MA JCEM 2011

The previously reported association of a heterozygousGHR:p.R229H variant with short stature and apparentGHI, therefore, seemed plausible because R229 is locatedin one of the six �-sheets of subdomain 2, and a substitu-tion of a histidine could disrupt the �-sheet structure. Insupport of this hypothesis, in silico functional analysis(http://genetics.bwh.harvard.edu/pph2/) predicted thatR229H had a high score of 0.984 (of a possible 1.0) to befunctionally damaging. Our reconstitution studies inHEK293 cells were, therefore, particularly significant be-cause they indicated, for the first time, that not only wasthe recombinant full-length hGHR:p.R229H variant nor-mally expressed, but it functioned like WT hGHR. Indeed,when the hGHR:p.R229H variant was expressed in ho-mozygous form, there was no demonstrable loss of GH-induced STAT5b phosphorylation or Luciferase activa-tion, even at physiological doses of GH. These unexpectedresults provide strong evidence that the heterozygoushGHR:R229H variant has little functional significanceand is unlikely to contribute significantly to the short stat-ure observed in our patient or in the previously reportedcase (12). This conclusion was further supported by theobservation that the mother and a sibling of our patientwere also heterozygous carriers of hGHR:p.R229H, butwere of normal stature (Fig. 5A), and, in an unrelatedfamily, R229H heterozygosity was identified not only intwo short-statured children, but also in the normal stat-ured mother (Fig. 5B; and Tenenbaum-Rakover, Y., un-published observations). Interestingly, the R229H variantis reported to have a heterozygous frequency of 0.009 inthe National Center for Biotechnology Information SingleNucleotide Polymorphism database (rs6177). Most im-portantly, our studies underscore the importance of func-tional analysis to evaluate the significance of any GHR

variant associated with a clinical phenotype, before cau-sality is inferred.

Identification of the compound heterozygous hGHR:p.R229H/c.899dupC mutations in a patient with un-equivocal GHI suggested, at first, that the cumulativesignaling defects of each variant might be causally re-lated to the GHI and short stature. Our studies of thehGHR:p.R229H variant, however, indicated that sig-naling was not apparently defective. In contrast, thehGHR:c.899dupC variant displayed no detectable GH-induced activities, consistent with its lack of Box 1. Theseresults do not, in and of themselves, explain the GHI ob-served in our patient because the presence of one func-tional copy of GHR, typically, is generally sufficient toallow normal stature. The compound heterozygous stateof our patient, with a functional GHR variant and a del-eterious GHR variant, suggested that hGHR:c.899dupCmay act as a dominant negative GHR. Consistent with thishypothesis, we demonstrated that coexpression of hGHR:c.899dupC with either hGHR:p.R229H or WT hGHRresulted in a significant reduction of GH-induced STAT5bphosphorylation and transcriptional activity, providingstrong evidence for a dominant negative effect. Althoughthe dominant negative hypothesis predicts that homo- andheterodimers involving hGHR:c.899dupC will be non-functional, in a compound heterozygous hGHR:R229H/c.899dupC state, a proportion of functional homodimerichGHR:R229H would still be present. The predicted pres-ence of some functional hGHR:p.R229H homodimers isconsistent with the phenotype of moderate GHI observedin our patient.

Only two dominant negative heterozygous GHR mu-tations have been described to date, both, interestingly,located in the intracellular domain: 1) a c.876–1G�C

FIG. 3. Dominant negative effect of hGHR:c.899dupC on GH-induced STAT5b phosphorylation. Immunoblot analysis was carried out on lysatesfrom HEK293 cells cotransfected with equal amounts of constructs as indicated. Cells were untreated or treated with 10 ng/ml or 100 ng/mlrecombinant hGH for 20 min. Immunoblots were probed with the antibodies indicated.

E1900 Derr et al. GHR c.899dupC Dominant Negative Mutation J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 6: Derr MA JCEM 2011

transversion affecting the 3� splice acceptor site of intron8 (18); and 2) a c.945�1G�A transition affecting the 5�splice donor site of intron 9 (19) (Fig. 6A). In both cases,exon 9 is spliced out, resulting in the same severely trun-cated GHR variant (hGHR:1–292) that retained onlyseven intracellular amino acid residues, compared with the

12 intracellular amino acids retained bythe hGHR:c.899dupC variant (Fig. 6, Aand B). The hGHR:1–292 and hGHR:c.899dupC variants are structurally andfunctionallysimilar, inthattheBox1mo-tif is either completely, or partially, de-leted, thereby abrogating STAT5b (18,19), aswell as STAT1and -3 signal trans-duction (Fig. 2A).

A potential mechanism for the dom-inant negative effect of the hGHR:1–292 variant came from functional stud-ies in which the expressed truncatedvariant demonstrated prolonged pres-ence on the cell surface (20). This ob-servation was attributed to loss of pu-tative intracellular residues required forreceptor internalization, and as a con-sequence, accumulation of the hGHR:1–292 variant on the cell surface wasproposed to exert a dominant negativeeffect. The membrane accumulation ofthe truncated GHR was thought to con-tribute to an elevation of serum con-centrations of GHBP to levels at least2-fold higher in affected patients thanin normal controls (19). It is, therefore,of note that our patient carrying thec.899dupC variant had serum GHBPconcentrations well within the normalrange (11) and that expression of the

c.899dupC variant in our reconstitution studies did notappear to be higher than that of WT hGHR. Hence, ourresults are not entirely consistent with impaired receptorinternalization as the mechanism for the dominant nega-tive effects of the c.899dupC variant, although furtherstudies are necessary to confirm this hypothesis.

Carriers of the GHR:1–292 variant, similarto our case, present with atypical GHI features,including moderate short stature (Fig. 6B) (18,19). The average reported height SDS of pa-tients with the hGHR:1–292 variant is �3.14and ranges from �2.0 to �3.6. Our patientcarrying the c.899dupC mutation, with aheight SDS of �4.07, is thus the most severelyaffectedcarrierofadominantnegativeGHRmu-tation. For the splice site mutations, c.876–1G�C and c.945�1G�A, variability in splicingefficiency may result in a small proportion ofnormally spliced GHR mRNA from the mutantalleles, which could contribute toward a less severeGHI phenotype. Indeed, a previously described

FIG. 4. Dose-dependent reduction in GH-induced STAT5b activation. Cell lysates analyzedwere from HEK293 cells cotransfected with equal amounts of hGHR:p.R229H and increasingamounts of hGHR:c.899dupC. A, Immunoblot analysis for GH-induced tyrosyl-phosphorylatedSTAT5b. B, Analysis of GH-induced, pSTAT5b-dependent, GHRE(Spi2.1)-luciferase reporteractivities. HEK293 cells were cotransfected with pGHRE-LUC and the indicated constructs.

FIG. 5. Pedigrees of two unrelated family members carrying GHR:p.R229Hgenotype. Height SDS are as indicated. Male, WT GHR gene, �; female, WT GHRgene, E. ?, GHR status unknown. Heterozygous p.R229H is indicated by half-filledsquares or circles. A, The p.R229H/c.899dupC proband (arrow) has been previouslyreported (11). B, The etiology for short stature in this family has yet to be established(Tenenbaum-Rakover, Y., unpublished observation).

J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904 jcem.endojournals.org E1901

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 7: Derr MA JCEM 2011

GHR splicing mutation, hGHR:c.618�792A�G (21),is associated with a range of GHI phenotypes, from mild tosevere (22). The hGHR:c.618�792A�G mutation is lo-cated in intron 6 and leads to recognition of a pseudoexonand inclusion of 108 basepairs between exons 6 and 7, caus-ing impaired trafficking of the mutant protein (22). A pro-portion of normally spliced GHR mRNA was detected infibroblast cells from homozygous carriers of hGHR:c.618�792A�G, demonstrating inefficiency in splicing

events that was proposed to account forthe range of GHI phenotypes (22, 23).

Four other intracellular GHR frame-shift mutations have been identified andassociated with GHI (Fig. 6, A and B)(24–27). All of the affected patients withthese mutations were either homozy-gotes or compound heterozygotes carry-ing a proven second detrimental GHRmutation. The c.899_911del mutation,similar to our c.899dupC variant, lacksan intactBox1, is severely truncated, andmight therefore be expected to exert adominant negative effect. However,heterozygouscarriersofthisGHR intracel-lularmutation(24)didnotshowconsistentshort stature (Fig. 6B). Of the remainingmutations, GHR mRNA transcriptscarrying c.981delC were not detectedin the normal statured heterozygouscarriers (25), heterozygous carriers ofc.1323_1344del22 were not described(26), and the one heterozygous carrier ofthe c.1733_1734delG variant was ofnormal stature (27).The lackofpotentialdominant negative effects of these muta-tions isnotentirelyclearbut ismost likelydue to poor peptide expression or reten-tion of the Box 1 motif.

The high proportion of dominantnegative mutations within the intracel-lular domain (three of seven convincingintracellular mutations) is intriguingbecause none of the 60 identified extra-cellular mutations associated with GHIresults in dominant negative effects.This striking lack of dominant negativeGHR mutations in the GHR extracel-lular domain may be explained by theasymmetrical binding of GH to the ex-tracellular domain of the GHR dimer(28). Because each GHR monomer car-ries all of the necessary GH binding andactivation sites, it can be inferred that,

in the asymmetrical binding of GH, GHR dimers contain-ing one partner with an altered extracellular domain stillhave sufficient sites to bind GH in a manner sufficient forinitiating intracellular signal transduction.

Finally, it is of note that neither homozygous nonsensenor homozygous missense mutations in the GHR intra-cellular domain have been identified to date. Only fourheterozygous missense variants have been identified:

FIG. 6. A, Schematic representation of intracellular GHR variants associated with GHI. Theaffected codons are indicated by filled circles. New amino acid residues as a consequence ofthe frameshift mutations are indicated in bold italics. B, Summary of patients and familymembers carrying intracellular GHR variants. *, hGHR prepeptide numbering; **, NA, Notavailable.

E1902 Derr et al. GHR c.899dupC Dominant Negative Mutation J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 8: Derr MA JCEM 2011

Y332C (10), C440F (29), A496T (30), and P579T (31).Evaluation of Y332F (10) and C440F (32) in reconstitu-tion systems and inconsistent clinical association ofC440F and P579T with short stature indicate that thesevariants, contrary to the original hypotheses, are unlikelyto be responsible for short stature. It remains unclearwhether an A496T exchange has biological significance.As has been previously shown, there is redundancy in theintracellular tyrosines necessary for STAT5b activation,which would seem to permit greater “forgiveness” for mis-sense mutations affecting the intracellular, as opposed tothe extracellular, domain of the GHR (10).

In summary, we provide functional evidence thatc.899dupC is a dominant negative GHR mutation respon-sible for the atypical GHI observed in our previously de-scribed severe short stature subject (11). Only two otherdominant negative GHR mutations have been identifiedto date, and like c.899dupC, they are located in the intra-cellular domain of GHR. All three mutations involved lossof the Box 1 motif and very early protein termination,suggesting that a stably expressed heterozygous GHRvariant is clinically significant only with the loss of allcritical regions within the intracellular domain (i.e. Box 1,intracellular tyrosines, internalization motifs).

Acknowledgments

Address all correspondence and requests for reprints to: MichaelDerr, Department of Pediatrics, CDRCP, Oregon Health & Sci-ence University, 3181 SW Sam Jackson Park Road, Portland,Oregon 97239-3098. E-mail: [email protected].

This work was funded by a March of Dimes grant (to R.G.R.).Disclosure Summary: M.A.D., J.A., P.F., Y.T.-R., and V.H.

have nothing to declare.

References

1. Rosenfeld RG, Hwa V 2009 The growth hormone cascade and itsrole in mammalian growth. Horm Res 71:36–40

2. David A, Hwa V, Metherell LA, Netchine I, Camacho-Hubner C,Clark AJ, Rosenfeld RG, Savage MO 2011 Evidence for a contin-uum of genetic, phenotypic, and biochemical abnormalities in chil-dren with growth hormone insensitivity. Endocr Rev 32:472–497

3. Savage MO, Attie KM, David A, Metherell LA, Clark AJ, Camacho-Hubner C 2006 Endocrine assessment, molecular characterizationand treatment of growth hormone insensitivity disorders. Nat ClinPract Endocrinol Metab 2:395–407

4. Rosenfeld RG, Belgorosky A, Camacho-Hubner C, Savage MO, WitJM, Hwa V 2007 Defects in growth hormone receptor signaling.Trends Endocrinol Metab 18:134–141

5. Hwa V, Nadeau K, Wit JM, Rosenfeld RG 2011 STAT5b deficiency:lessons from STAT5b gene mutations. Best Pract Res Clin Endocri-nol Metab 25:61–75

6. Domene HM, Hwa V, Argente J, Wit JM, Wit JM, Camacho-Hub-ner C, Jasper HG, Pozo J, van Duyvenvoorde HA, Yakar S, Fo-

fanova-Gambetti OV, Rosenfeld RG 2009 Human acid-labile sub-unit deficiency: clinical, endocrine and metabolic consequences.Horm Res 72:129–141

7. Woods KA, Camacho-Hubner C, Savage MO, Clark AJ 1996 In-trauterine growth retardation and postnatal growth failure associ-ated with deletion of the insulin-like growth factor I gene. N EnglJ Med 335:1363–1367

8. de Vos AM, Ultsch M, Kossiakoff AA 1992 Human growth hor-mone and extracellular domain of its receptor: crystal structure ofthe complex. Science 255:306–312

9. Behncken SN, Waters MJ 1999 Molecular recognition events in-volved in the activation of the growth hormone receptor by growthhormone. J Mol Recognit 12:355–362

10. Derr MA, Fang P, Sinha SK, Ten S, Hwa V, Rosenfeld RG 2011 Anovel Y332C missense mutation in the intracellular domain of thehuman growth hormone receptor does not alter STAT5b signaling:redundancy of GHR intracellular tyrosines involved in STAT5b sig-naling. Horm Res Paediatr 75:187–199

11. Aisenberg J, Auyeung V, Pedro HF, Sugalski R, Chartoff A, Rothen-berg R, Derr MA, Hwa V, Rosenfeld RG 2010 Atypical GH insen-sitivity syndrome and severe insulin-like growth factor-I deficiencyresulting from compound heterozygous mutations of the GH recep-tor, including a novel frameshift mutation affecting the intracellulardomain. Horm Res Paediatr 74:406–411

12. Goddard AD, Covello R, Luoh SM, Clackson T, Attie KM, Ge-sundheit N, Rundle AC, Wells JA, Carlsson LM 1995 Mutations ofthe growth hormone receptor in children with idiopathic short stat-ure. The Growth Hormone Insensitivity Study Group. N Engl J Med333:1093–1098

13. Ross RJ, Esposito N, Shen XY, Von Laue S, Chew SL, Dobson PR,Postel-Vinay MC, Finidori J 1997 A short isoform of the humangrowth hormone receptor functions as a dominant negative inhib-itor of the full-length receptor and generates large amounts of bind-ing protein. Mol Endocrinol 11:265–273

14. Fang P, Riedl S, Amselem S, Pratt KL, Little BM, Haeusler G,Hwa V, Frisch H, Rosenfeld RG 2007 Primary growth hormone(GH) insensitivity and IGF deficiency caused by novel compoundheterozygous mutations of the GH receptor gene: genetics andfunctional studies of simple and compound heterozygous states.J Clin Endocrinol Metab 92:2223–2231

15. Hwa V, Little B, Kofoed EM, Rosenfeld RG 2004 Transcriptionalregulation of insulin-like growth factor-I (IGF-I) by interferon-�(IFN-g) requires STAT-5b. J Biol Chem 279:2728–2736

16. Harding PA, Wang XZ, Kelder B, Souza S, Okada S, Kopchick JJ1994 In vitro mutagenesis of growth hormone receptor Asn-linkedglycosylation sites. Mol Cell Endocrinol 106:171–180

17. van den Eijnden MJ, Lahaye LL, Strous GJ 2006 Disulfide bondsdetermine growth hormone receptor folding, dimerisation and li-gand binding. J Cell Sci 119:3078–3086

18. Ayling RM, Ross R, Towner P, Von Laue S, Finidori J, Moutous-samy S, Buchanan CR, Clayton PE, Norman MR 1997 A dominant-negative mutation of the growth hormone receptor causes familialshort stature. Nat Genet 16:13–14

19. Iida K, Takahashi Y, Kaji H, Nose O, Okimura Y, Abe H, ChiharaK 1998 Growth hormone (GH) insensitivity syndrome with highserum GH-binding protein levels caused by a heterozygous splice sitemutation of the GH receptor gene producing a lack of intracellulardomain. J Clin Endocrinol Metab 83:531–537

20. Iida K, Takahashi Y, Kaji H, Takahashi MO, Okimura Y, Nose O,Abe H, Chihara K 1999 Functional characterization of truncatedgrowth hormone (GH) receptor-(1–277) causing partial GH insen-sitivity syndrome with high GH-binding protein. J Clin EndocrinolMetab 84:1011–1016

21. Metherell LA, Akker SA, Munroe PB, Rose SJ, Caulfield M, SavageMO, Chew SL, Clark AJ 2001 Pseudoexon activation as a novelmechanism for disease resulting in atypical growth-hormone insen-sitivity. Am J Hum Genet 69:641–646

22. David A, Camacho-Hubner C, Bhangoo A, Rose SJ, Miraki-Moud

J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904 jcem.endojournals.org E1903

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.

Page 9: Derr MA JCEM 2011

F, Akker SA, Butler GE, Ten S, Clayton PE, Clark AJ, Savage MO,Metherell LA 2007 An intronic growth hormone receptor mutationcausing activation of a pseudoexon is associated with a broad spec-trum of growth hormone insensitivity phenotypes. J Clin EndocrinolMetab 92:655–659

23. Maamra M, Milward A, Esfahani HZ, Abbott LP, Metherell LA,Savage MO, Clark AJ, Ross RJ 2006 A 36 residues insertion in thedimerization domain of the growth hormone receptor results in de-fective trafficking rather than impaired signaling. J Endocrinol 188:251–261

24. Gastier JM, Berg MA, Vesterhus P, Reiter EO, Francke U 2000Diverse deletions in the growth hormone receptor gene cause growthhormone insensitivity syndrome. Hum Mutat 16:323–333

25. Kaji H, Nose O, Tajiri H, Takahashi Y, Iida K, Takahashi T,Okimura Y, Abe H, Chihara K 1997 Novel compound heterozygousmutations of growth hormone (GH) receptor gene in a patient withGH insensitivity syndrome. J Clin Endocrinol Metab 82:3705–3709

26. Milward A, Metherell L, Maamra M, Barahona MJ, Wilkinson IR,Camacho-Hubner C, Savage MO, Bidlingmaier M, BidlingmaierCM, Clark AJ, Ross RJ, Webb SM 2004 Growth hormone (GH)insensitivity syndrome due to a GH receptor truncated after Box 1,resulting in isolated failure of STAT 5 signal transduction. J ClinEndocrinol Metab 89:1259–1266

27. Tiulpakov A, Rubtsov P, Dedov I, Peterkova V, Bezlepkina O,Chrousos GP, Hochberg Z 2005 A novel C-terminal growth hor-mone receptor (GHR) mutation results in impaired GHR-STAT5but normal STAT-3 signaling. J Clin Endocrinol Metab 90:542–547

28. Brown RJ, Adams JJ, Pelekanos RA, Wan Y, McKinstry WJ,Palethorpe K, Seeber RM, Monks TA, Eidne KA, Parker MW, Wa-ters MJ 2005 Model for growth hormone receptor activation basedon subunit rotation within a receptor dimer. Nat Struct Mol Biol12:814–821

29. Kou K, Lajara R, Rotwein P 1993 Amino acid substitutions in theintracellular part of the growth hormone receptor in a patient withthe Laron syndrome. J Clin Endocrinol Metab 76:54–59

30. Goddard AD, Dowd P, Chernausek S, Geffner M, Gertner J, HintzR, Hopwood N, Kaplan S, Plotnick L, Rogol A, Rosenfield R,Saenger P, Mauras N, Hershkopf R, Angulo M, Attie K 1997 Partialgrowth-hormone insensitivity: the role of growth-hormone receptormutations in idiopathic short stature. J Pediatr 131:S51–S55

31. Chujo S, Kaji H, Takahashi Y, Okimura Y, Abe H, Chihara K 1996No correlation of growth hormone receptor gene mutation P561Twith body height. Eur J Endocrinol 134:560–562

32. Iida K, Takahashi Y, Kaji H, Onodera N, Takahashi MO, OkimuraY, Abe H, Chihara K 1999 The C422F mutation of the growthhormone receptor gene is not responsible for short stature. J ClinEndocrinol Metab 84:4214–4219

Share Your Career News! Endocrine News would like to consider your news

for its Smart Moves section. [email protected].

E1904 Derr et al. GHR c.899dupC Dominant Negative Mutation J Clin Endocrinol Metab, November 2011, 96(11):E1896–E1904

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 December 2016. at 23:19 For personal use only. No other uses without permission. . All rights reserved.