| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Studies |
Developmental Endocrinology Branch, National Institute of Child Health and Human Development (K.Y., G.B.C.), and Cell Regulation Section, Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases (L.D.K.), National Institutes of Health, Bethesda, Maryland 20892; Department of Surgery, Johns Hopkins University (M.S.), Baltimore, Maryland 21205; and Department of Pediatrics, Asahikawa Medical College (K.Y., A.O.), Asahikawa 078, Japan
Address all correspondence and requests for reprints to: Koichi Yano, Department of Pediatrics, Asahikawa Medical College, Nishikagura 45-311, Asahikawa 078, Japan.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Of interest, however, among the cases of testicular resistance to LH caused by inactivating mutations of the LH/CGR gene reported recently (12, 13, 14), several have mutations in the region of the sixth or seventh transmembrane domains (12, 13). In one study, genomic DNAs from patients with male pseudohermaphroditism and Leydig cell hypoplasia were found to have a homozygous missense mutation, Ala593 to Pro, in the sixth transmembrane region of the LH/CGR (12). In another report (13), a boy with micropenis had a homozygous mutation of Ser616 to Tyr in the seventh transmembrane domain of the LH/CGR. In expression studies, these mutations demonstrated no increase in cAMP production with human CG (hCG) stimulation.
Using the Garnier program (15), we analyzed the secondary structure of receptors with activating mutations in the sixth transmembrane domain, where most mutations were discovered. The results suggested that Phe576 was a critical bridging residue important for receptor activity and conformation in the sixth transmembrane domain and third intracellular loop. All mutants but one induced changes in the computer-predicted secondary structure in the region between residues 571 and 583 (6), which changed the predicted structure at Phe576 from a helical array to an extended array. We therefore hypothesized that Phe576 might play an important role in the LH/CGR with respect to cAMP signaling and receptor conformation, although no mutations at Phe576 have been found in patients with male-limited precocious puberty. We hoped simultaneously to gain insight into the relationship between Gs and Gq coupling and receptor conformation, because some but not all mutations associated with increased Gs coupling and male-limited precocious puberty also exhibited altered hormone-induced Gq coupling and inositol phosphate formation. We also hoped that studies of Phe576 might provide clues about the role of receptor conformation in loss of Gs coupling in male pseudo-hermaphroditism.
To test this prediction and evaluate this question, we made four amino acid substitutions at residue Phe576 and studied cAMP production, inositol phosphate (IP) production, and hCG binding using cells transfected with each mutant LH/CGR complementary DNA (cDNA).
| Materials and Methods |
|---|
|
|
|---|
Four amino acid substitutions of the Phe576 in LH/CGR cDNA were made by recombinant PCR (16), i.e. Phe to Gly (F576G), Phe to Glu (F576E), Phe to Ile (F576I), and Phe to Tyr (F576Y). In brief, two PCR products that overlap the mutated sequence were created, both of which contained each mutation introduced as part of the PCR primers. The second PCR step used the overlapped PCR products as template; primers contained human LH/CGR DNA sequence 5' or 3' to the mutated region plus a BstXI (5'-end) and a HpaI site(3'-end). The second-step PCR products containing each mutation were digested with BstXI and HpaI and ligated to wild-type human LH/CGR cDNA (2) in pSG5, which had been similarly digested to remove the normal hLH/CGR sequence in this interval. Amplified DNAs were purified by CsCl gradient centrifugation (17) and sequenced (18) to ensure that each DNA had the predicted mutation but no other changes.
Transfection and assays
Wild-type LH/CGR cDNA was the same preparation described previously (2). COS-7 cells were transfected with 25 µg plasmid mutant or wild-type cDNA by electroporation (5, 6, 11, 19, 20); cells transfected with pSG5 alone were negative controls. The same batch of transfected cells was plated in DMEM with 10% FCS in 6-well plates (5 x 105 cells/well) for binding assays or in 24-well plates (1 x 105 cells/well) for cAMP and IP assays (5, 6, 11, 19, 20). The medium was inositol free in the latter assays and supplemented with 2 mCi/L myo-[2-N-3H]inositol (SA, 22.9 Ci/mmol; DuPont-New England Nuclear, Billerica, MA). All assays were initiated simultaneously, 48 h after transfection and after washing the cells with assay buffer (HBSS containing 0.5% BSA and 20 mmol/L HEPES, pH 7.4) (5, 6, 11, 19, 20).
Purified hCG (CR-127; 14,900 IU/mg) used in cAMP, IP, or binding assays was kindly provided by the National Hormone and Pituitary Program (NIDDK, Center for Population Research, of the National Institute of Child Health and Human Development, and the Agricultural Research Service of the United States Department of Agriculture). For binding studies, it was iodinated (SA, 44 µCi/µg) using lactoperoxidase (5, 6, 11, 19, 20). [125I]hCG binding was measured after 8 h at 22 C in incubations containing 1 mL assay buffer, 4 x 104 cpm [125I]hCG, and 010-7 mol/L unlabeled hCG. Specific binding was calculated by subtracting values obtained in cells transfected with the pSG5 control plasmid and incubated with the same concentrations of radiolabeled and unlabeled hCG. Nonspecific binding was less than 10% and was the same as [125I]hCG binding in the presence of 10-7 mol/L unlabeled hCG (5, 6, 11, 19, 20).
Total cAMP and IP levels were measured (5, 6, 11, 19, 20) in the same wells. Incubations were performed for 2 h at 37 C in 0.2 mL assay buffer containing 10 mmol/L LiCl and 0.5 mmol/L isobutylmethylxanthine alone or with 10-11 to 10-7 mol/L hCG. After adding 1.0 mL 5% perchloric acid, samples were centrifuged to remove protein debris, neutralized with KOH, and centrifuged to remove insoluble salts. Total cAMP was measured in aliquots of the supernatant with a RIA kit (New England Nuclear-DuPont); IP formation was determined in aliquots of the same supernatant using Dowex AG1-X8 columns (5, 6, 11, 19, 20). Data were expressed as the fold increase above the negative control value, i.e. cells transfected with pSG5 and exposed to buffer alone.
The cAMP and IP assays were performed in triplicate; binding assays were performed in duplicate. Values in each well were corrected either for cell protein, measured using a Bio-Rad protein assay kit (Bio-Rad Labs., Richmond, CA) and a BSA standard, or for total tritiated inositol incorporated (5, 6, 11, 19, 20). The program LIGAND (21) was used to calculate dissociation constant (Kd) values (5, 6, 11, 19, 20).
Statistic analyses
Values are presented as the mean ± SE from multiple experiments. Statistical differences were calculated using the two-tailed Students t test.
| Results |
|---|
|
|
|---|
In the absence of hCG, cells transfected with the F576I or the
F576G mutants exhibited 3.2 ± 0.4-fold or 2.7 ± 0.2-fold
higher basal cAMP levels, respectively, than cells transfected with
wild-type receptor or with vector alone (Fig. 1
). hCG
could still induce a concentration-dependent increase in cAMP levels
similar to the levels in cells with wild-type receptor. Maximal
(10-7 mol/L) hCG-stimulated cAMP levels in cells
transfected with the F576I mutant, F576G mutant, and the wild-type
hLH/CGR were 10.0 ± 0.6-fold, 10.8 ± 0.7-fold, and
10.0 ± 0.7-fold higher than the control value, respectively.
These values did not differ from each other statistically.
|
The F576E mutant did not alter basal cAMP levels and did not exhibit a statistically significant increase in cAMP levels when exposed to hCG. Basal cAMP levels and the levels at 10-7 mol/L hCG were 1.5 ± 0.2-fold and 1.9 ± 0.5-fold higher than the control value, respectively.
Basal and hCG-induced IP Levels in Cells Transfected with Mutant LH/CGRs
There were no significant differences between basal IP levels of
cells expressing any of the mutant receptors and basal IP levels of
cells expressing the wild-type LH/CGR (Fig. 2
).
|
hCG Binding in Cells Transfected with Mutant LH/CGRs
Binding of hCG to the F576I or the F576G mutants was similar to
wild-type receptor. COS-7 cells transfected with the F576I, the F576G
mutants, and wild-type receptor cDNAs specifically bound 9.1%, 8.8%,
and 10.5% of added [125I]hCG, respectively (Fig. 3A
). Kd values were 10.2 x
10-10 M, 9.2 x 10-10
M, and 9.2 x 10-10 M,
respectively. Cells transfected with the F576Y mutant bound 4.9% of
added [125I]hCG; the Kd value was 5.1 x
10-10 M. These cells exhibited a lower
capacity as well as a higher affinity, as shown in Scatchard analysis
(Fig. 3B
). Cells transfected with F576E exhibited negligible hCG
binding.
|
| Discussion |
|---|
|
|
|---|
In the absence of hCG, cells transfected with the F576I or the F576G
mutants exhibited 3-fold higher basal cAMP levels than cells
transfected with wild-type receptor or with vector alone (Fig. 1
). hCG
could still induce a concentration-dependent increase in cAMP levels
similar to the levels in cells with wild-type receptor. Cells
transfected with the F576Y mutant exhibited 2-fold higher basal cAMP
levels; however, these cells exhibited a lower hCG-induced increase in
cAMP levels. The F576E mutant did not alter basal cAMP values and did
not respond to hCG.
In contrast to the increased basal cAMP levels, there were no
significant differences between basal IP levels of cells expressing any
of the mutant receptors and cells expressing the wild-type LH/CGR (Fig. 2
). Cells transfected with F576I exhibited a similar hCG-induced
increase in IP levels as cells transfected with wild-type receptor; the
maximal stimulated levels by hCG for F576G were slightly lower than the
levels for wild-type receptor. Cells transfected with F576E and F576Y
showed minimal but significant hCG-induced increases in IP levels at
high hCG concentration.
Binding of hCG to F576I or F576G was similar to wild-type receptor.
COS-7 cells transfected with each of the three cDNAs specifically bound
910% of added [125 I]hCG (Fig. 3
). Cells transfected
with F576Y bound 5% of added [125I]hCG and exhibited
higher affinity and lower capacity. Cells transfected with F576E
exhibited negligible hCG binding.
We analyzed protein secondary structure of these mutants using a
computer program (15) that predicted secondary structure (helical,
extended, turn, or coil conformations). The F576G, F576I, and F576Y
mutants are all predicted to change the helical conformation similarly
to an extended conformation in the region of the sixth transmembrane
domain (Fig. 4
). In contrast, the F576E mutant changes
the secondary structure to a totally helical conformation in the
region.
|
These data suggested that the common denominator of the F576I, F576G, and F576Y mutants is altered activation of adenylyl cyclase, i.e. the common change to an extended conformation in this region is linked to increased Gs coupling. The totally different impact on receptor conformation of the F576E mutant and its association with decreased rather than increased Gs coupling strengthens the possibility that the extended conformation induced by the other mutations in this region is important for increased Gs coupling.
Two additional points emerge from these studies. First, the F576Y
mutant has effects that lead to higher binding affinity, lower capacity
of hCG binding, and decreased Gq coupling and phospholipase C
activation, in addition to its effects on Gs coupling, but these
changes cannot be linked to the common conformational shift, because
they are different from F576I and F576G. The impact of the tyrosine, as
opposed to the phenylalanine, glycine, and isoleucine side chains, all
of which are hydrophobic but devoid of the hydroxyl group, must be
considered to result in these changes, in addition to altered
conformation. We already reported similar changes in activating mutant
LH/CGRs that were found in patients with male-limited precocious
puberty (6, 11). The A572V mutant in the sixth transmembrane domain in
the LH/CGR exhibited increased basal cAMP and IP levels with decreased
hCG-induced IP levels (6). The M398T mutant in the second transmembrane
domain in the LH/CGR exhibited increased basal cAMP levels, no increase
in basal IP levels, and decreased hCG-induced IP levels (11). In
addition, these two mutants exhibited higher affinity and lower
capacity of hCG binding. Similar changes in affinity and capacity have
been noted also in mutations altering G protein coupling of the TSH
receptor and
1B adrenergic receptor (19, 22, 23, 24, 25). These
changes in affinity and capacity have been postulated to reflect
distant effects on the extracellular domain of the conformational
changes induced by the mutations in the third cytoplasmic loop-sixth
transmembrane region that cause the constitutively increased G protein
coupling.
Mutations in the carboxyl end of the third intracellular loop, which is
adjacent to the sixth transmembrane region of the
1B-adrenergic receptor, constitutively activate the
receptor, resulting in G protein coupling in the absence of agonist
(24). All 19 amino acid substitutions at Ala293 confer constitutive
activity. This set of mutated receptors exhibits a graded range of
elevated biological activities. The fact that all possible mutations at
this particular site result in increased activity suggests that this
region may function to constrain the G protein coupling of the
receptor, a constraint that is normally relieved by agonist occupancy.
Our results of the LH/CGR study was different from the results of the
1B-adrenergic receptor study. LH/CGR mutations at
Phe576 in the sixth transmembrane domain exhibited three
different functional patterns depending on their predicted change in
the secondary structure. Two patterns were constitutively activating in
cAMP signaling with or without changes in IP signaling or hCG binding.
The other pattern was no cAMP response and a minimal IP response to hCG
stimulation with negligible hCG binding. There must be different
mechanisms of constitutive activation even in similar regions of the
different G protein-coupled receptors.
Second, the fact that there is minimal but significant Gq coupling by
the F576E mutant, despite negligible hCG binding and Gs coupling,
suggests that at least some receptor has been incorporated into the
bilayer, and that it is not a null mutation that results in a failure
of the receptor to enter the bilayer; nevertheless, this cannot be
excluded at this time. Recently, inactivating mutations of the LH/CGR
have been reported in patients with male pseudohermaphroditism
(12, 13, 14). Of these, one, A593P, is in the C-terminal portion of the
sixth transmembrane domain (12). Interestingly, this mutation does not
alter conformation of the Phe576 region (Fig. 5
, dark bar 1), but rather has a
profound impact on the conformation of the C-terminal portion of the
sixth transmembrane domain (Fig. 5
, dark bar 2).
This mutation results in a receptor with a reduced maximal hormone
binding but normal affinity. This result is consistent with the present
studies, because there is no change in the Phe576 region to
an extended conformation that would increase Gs coupling. Nevertheless,
this observation raises the possibility that mutations in the
C-terminal portion of the sixth transmembrane domain that similarly
break up the helical conformation in the region between residues
588594 will decrease Gs coupling. Of interest, a mutation in the
seventh transmembrane domain that results in decreased receptor
activity and micropenis, S616Y (13, 14), has no effect on the
conformation of the sixth transmembrane domain, suggesting diverse
mechanisms through which mutations produce loss of activity.
|
Received January 24, 1997.
Revised April 23, 1997.
Accepted May 1, 1997.
| References |
|---|
|
|
|---|
1-adrenergic receptor determine the selectivity of coupling to
phosphatidylinositol hydrolysis. J Biol Chem. 267:16331639.
1B-adrenergic receptor by all amino acid substitutions
at a single site. J Biol Chem. 267:14301433.
1B-adrenergic receptor
enhances mitogenesis and tumorigenicity. Proc Natl Acad Sci USA. 88:1135411358.This article has been cited by other articles:
![]() |
M. Nisolle, S. Gillerot, F. Casanas-Roux, J. Squifflet, M. Berliere, and J. Donnez Immunohistochemical study of the proliferation index, oestrogen receptors and progesterone receptors A and B in leiomyomata and normal myometrium during the menstrual cycle and under gonadotrophin-releasing hormone agonist therapy Hum. Reprod., November 1, 1999; 14(11): 2844 - 2850. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |