| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
q Mutations
Department of Neurosurgery and Laboratory of Molecular Neurosurgery and Biotechnology (N.M.O., C.-O.E., G.T.T.), Division of Pediatric Endocrinology and Department of Pediatrics (M.R.B., J.S.P.), Division of Endocrinology (L.S.B.), Department of Medicine, Emory University School of Medicine, Atlanta, Georgia 30322
Address all correspondence and requests for reprints to: Dr. Nelson M. Oyesiku, Section of Neurosurgery, The Emory Clinic, Inc., 1365 Clifton Road, N. E., Building B, Suite B2200, Atlanta, Georgia 30322. E-mail: noyesik{at}emory.edu
| Abstract |
|---|
|
|
|---|
-subunits of the G
protein, Gs, gsp ocogene have been discovered in 40% of
GH-secreting pituitary adenomas. Therefore, we hypothesized that a
novel G protein class, G
q, involved in pituitary signal
transduction, might be involved in pituitary tumorigenesis. Recombinant
mutations of G
q result in constitutive activation of phospholipase C
and have transforming activity. Therefore, we screened tumor samples
from 37 pituitary adenomas for the presence of activating mutations of
the G
q gene. Importantly, our sample contains 8 FSH and LH adenomas.
In the pituitary gland, FSH and LH are linked to the GnRH-G
q
signaling cascade, making these tumors a logical choice for screening
for G
q mutations. Complementary DNA (cDNA) was synthesized by RT-PCR
with G
q specific primers to exclude pseudogene transcripts.
Fragments of G
q cDNA-encompassing residues (Arg183,
Gln209) were screened by single-strand conformation
polymorphism and then sequenced in both directions. No mutations were
detected. We conclude that mutations in these regions of the G
q cDNA
occur infrequently, if at all, in human pituitary adenomas. Alternative
mechanisms underlying pituitary tumorigenesis should be explored. | Introduction |
|---|
|
|
|---|
, ß
-subunits (1). The
-subunit is
believed to confer receptor and effector specificity on the
heterotrimer. When the G protein is activated by interaction with
receptor, the
-subunit exchanges bound guanosine diphosphate for
guanosine triphosphate (GTP). The intrinsic GTPase activity of the
-subunit restores it to the basal state in which guanosine
diphosphatephos is bound. At least three
-subunits (Gs,
Gi, and more recently, Gq) have been involved
in pituitary cell signaling. Gs is involved in the
GHRH-adenyl cyclase (AC)-cAMP pathway. Gi is involved in
the SRIH-AC-cAMP pathway. A novel class, Gq (G
q and
G
11) has been identified as the G protein that mediates the TRH and
GnRH signal cascades in pituitary cells (2, 3). These pathways are
mediated by phospholipase C (PLC) and the inositol phospholipids. G
q
and G
11 differ at only four residues over the C-terminal 144 amino
acids, which contains the structural elements required for specific
interactions with effector and receptor.
The most specific molecular abnormality in pituitary tumors identified
to date is a mutation of the G protein, Gs(4). Indeed, 2
single-point mutations in the G
s-subunit (arginine-201
and glutamine-227) give rise to the gsp oncogene. This
mutant G
s results in constitutive activity of AC and
high levels of cAMP and occurs in about 40% of somatotroph adenomas
and the McCune-Albright syndrome (4, 5, 6, 7, 8, 9, 10, 11, 12). In a similar vein, G
q may
be a candidate oncogene. Recombinant mutations G
q (R183C and Q209L)
result in constitutive activity of PLC and increased inositol
phospholipid turnover in transfected cells. Furthermore, G
q Q209L
mutations induce malignant transformation when expressed in NIH 3T3
cells (13, 14). However, there are no reports of G
q mutations in
pituitary tumors. Therefore, we screened samples from 37 pituitary
tumors for the presence of activating somatic mutations within the
GTPase catalytic domain of the G
q genes, using single-strand
conformation polymorphism (SSCP) of RT-PCR product, and direct DNA
sequencing. Importantly, our sample contains 8 FSH and LH adenomas. In
the pituitary gland, FSH and LH are linked to the GnRH-G
q signaling
cascade, making these tumors a logical choice for screening for G
q
mutations.
| Subjects and Methods |
|---|
|
|
|---|
Clinical and pathological data from the 37 patients are shown in
Table 1
. There were 22 males and 15
females ranging from ages 2690 yr. The tumors consisted of 30
clinically nonfunctional tumors (including 8 FSH and LH tumors), 3
GH-secreting adenomas (acromegaly), 2 PRL-secreting adenomas, 1
ACTH-secreting adenoma (Cushings disease), and 1 case of pituitary
apoplexy. Pituitary tumor specimens were obtained during surgery
performed at Emory University Hospital under an approved informed
consent protocol. All tumors were macroadenomas. Specimens were
immediately frozen in liquid nitrogen and stored at -80 C until
analyzed.
|
Total RNA was isolated using an RNAgents Total RNA Isolation System (Promega, Madison, WI). In general, 5060 mg of frozen pituitary tumor tissue was homogenized in 600720 µL of the denaturing solution; 6072 µL of 2 mol/L sodium acetate (pH 4.0) was added and mixed thoroughly. Two phenol:chloroform:isoamyl alcohol extractions were performed; the aqueous phase was collected. The RNA was isopropanol precipitated at -20 C overnight. After centrifugation at 12,000 x g for 15' at 4 C, the pellet was dissolved in 100 µL of 0.5% SDS, 10 mmol/L EDTA, and then ethanol precipitated. After centrifugation, the pellet was washed with 75% ice-cold ethanol, dried briefly in a speed vac, and resuspended in 3040 µL diethyl pyrocarbonate-treated water. The purity and quantity of RNA was analyzed spectrophotometrically.
RT-PCR and SSCP analysis
Primers were designed using Primer Select software (Lasergene,
Madison, WI) and the human G
q complementary DNA (cDNA)
sequence (15). To avoid amplifying the G
q pseudogene,
which is highly homologous to human G
q cDNA, we designed
a pair of primers (20 oligonucleotides) that flanked the GTPase
catalytic domain from codon A168 to D243, which
had 3' mismatches with the pseudogene sequence. The sense-strand primer
also was designed such that it contained a contiguous eight-base
mismatch with the peudogene sequence:
The sense primer was 5'AGCTGACCCTGCCTACCTGC 3', and the antisense
primer was 5' GTCTGACTCCACGAGAACTT 3'. This set of primers amplified
the 229-bp G
q cDNA flanking the catalytic domain. The
oligonucleotide primers were synthesized by Emory University
Microchemical Facility, Winship Cancer Center.
For the RT reaction, we used the first-strand cDNA synthesis kit (Boehringer Mannheim, Indianapolis, IN); total RNA (2.5 µg) was reverse transcribed in a 10-µL reaction in 10 mmol/L Tris, 50 mmol/L KCl (pH 8.3), 5 mmol/L MgCl2, 1 mmol/L deoxynucleotide mix, 1.0 µmol/L antisense primer, 15 U RNase inhibitor, and 12.5 U of AMV RT. The total RNA was incubated at 65 C for 15 min and cooled in ice for 5 min, the enzymes were added, and the mixture was incubated at room temperature for 10 min and then at 42 C for 60 min.
The PCR was performed in a 50-µL reaction with the final
concentration of all components as follows: 10 mmol/L Tris, 50 mmol/L
KCl (pH 8.3), 2.5 mmol/L MgCl2, 0.2 mmol/L deoxynucleotide
mix, 0.2 µmol/L of both primers, 2.5 U of Taq polymerase
(Boehringer Mannheim), and 4 µCi of
-32PdCTP (3000
Ci/mmol). PCR was then performed at the following conditions: 94 C for
2 min followed by 94 C, 30 sec; 60 C, 30 sec; 72 C, 30 sec for 30
cycles.
The PCR product was diluted 1:10 in loading dye and then denatured, chilled, and loaded. Electrophoresis of 3 µL of the denatured sample was performed through a 0.5 XMDE gel (J. T. Baker Inc., Phillipsburg, NJ) at 5 watts at room temperature for 18 h. The gel was dried and exposed to film with intensifying screen at -80 C for 48 h.
RT-PCR and sequencing of G
q c DNA
The same set of primers described above was used to amplify and
sequence, in both directions, the 229-bp G
q cDNA flanking the
catalytic domain from codon A168 to D243. Total
RNA (5 µg) from each sample was subjected to RT as described above.
The PCR was performed, as previously described, except for 33 cycles
and without
32P dCTP (cytosine triphosphate).
Before sequencing, RT-PCR product was purified by eluting from agarose
gel using the Wizard PCR preps DNA purification system (Promega). After
precipitation and resuspension in 25 µL of water, the DNA (5 µL)
was used for double-strand cycle sequencing (GIBCO/BRL, Gaithersburg,
MD). The primers were 5' end labeled with
33P-ATP (2000
Ci/mmol) and T4 polynucleotide kinase. The direct cycle
sequencing reactions were set up as recommended by the manufacturer,
except for the addition of 5 pmols of each labeled primer, 2.5 U of
Taq Polymerase, and 10% dimethyl sulfoxide (Sigma). The
sequencing reaction was performed at the following cycles: 94 C for 2'
followed by 94 C, 30"; 60 C, 30"; 72 C, 60" for 20 cycles and
then 94 C, 30"; 72 C, 60" for 15 cycles. The sequencing reactions
were denatured and resolved on a 6% denaturing polyacrylamide
sequencing gel. The dried gel was exposed to film with intensifying
screen at -80 C for 12 days.
| Results |
|---|
|
|
|---|
q
mutations
We used two strategies to eliminate the possibility of amplifying
the human G
q pseudogene that is highly homologous to human G
q
cDNA. First, the oligonucleotide primers (20 bases) were designed to
have their 3'-ends mismatched with the pseudogene sequence; and second,
we designed the sense-strand primer such that it makes an eight-base
mismatch with the pseudogene sequence. This pair of primers selectively
amplified the 229-bp G
q cDNA flanking the GTPase catalytic domain
from A168 to D243 and encompassing the target
codons of R183 and Q209 (Fig. 1
). To further test the specificity of
the primers, we sequenced the RT-PCR products from both directions and
found that the sequence of the product was identical to the sequence of
the published G
q cDNA (15) with a nucleotide C at position 562, A at
610, G at 613, G at 652, and A at 752. The corresponding pseudogene
nucleotides are, respectively: G, G, A, T, and G. Sequencing from
antisense primer also revealed the existence of eight bases: 5'
CCTGCCTA 3'. These bases are present in the G
q cDNA but absent in
pseudogene sequence. Therefore, we conclude that this set of primers
specifically amplifies the G
q cDNA product of 229 bp and reliably
excludes the pseudogene product.
|
Thirteen pituitary tumor samples were examined by SSCP analysis of
RT-PCR products. We found no alteration in mobility shifts in the SSCP
analysis from any of the tumor samples. To insure that we had not
missed any mutations by SSCP analysis, the G
q cDNA product from all
tumor samples was analyzed by direct sequencing.
RT-PCR and sequencing of G
q cDNA flanking the GTPase domain
Because SSCP analysis is not 100% reliable in detecting point mutations, we did not use this as a screening or a definitive method for detecting mutations, rather we performed direct cycle sequencing of RT-PCR products from all tumor samples.
G
q cDNAs flanking the GTPase domain were successfully amplified from
all 37 pituitary tumors and 2 normal pituitaries, yielding RT-PCR
products of the expected size (229 bp), as shown in the ethidium
bromide agarose gel in Fig. 2
. RT-PCR
products were eluted from the agarose gel, precipitated, and directly
sequenced in both directions with 33P 5'-end-labeled
primers. Representative results for the G
q cDNA at
Arg183 and Gln209 are shown in Fig. 3
. As in the published human G
q cDNA
from temporal cortex sequence cDNA (15), our sequencing clearly
revealed that Arg183 in the human G
q cDNA from 2 normal
pituitaries or from 37 pituitary tumors is coded by CGA, and
Gln209 by CAA. None of the presumed activating mutations at
codons Arg183 and Gln209 were detected.
Furthermore, no alteration in the codons surrounding residues within
the GTPase catalytic domain between codons A168 and
D243 was detected.
|
|
| Discussion |
|---|
|
|
|---|
-subunit (G
q) might be involved in the pathogenesis of
pituitary tumors. The search for G
q mutations at these loci was
stimulated by prior recombinant experiments, in which G
q mutations
were genetically engineered at analogous residues to the G
s
oncogene, which has been found in a subset of pituitary adenomas (4, 11). These recombinant mutants include: 1) R183C, an arginine
183-to-cysteine conversion that activates phosphoinositide-specific PLC
(14, 16); and 2) Q209L, a glutamine 209-to-leucine mutant that also
persistently activates phosphoinositide-specific PLC. Furthermore, when
this mutant is expressed in NIH 3T3 cells, the cells are transformed,
as assessed by focus formation or the soft agar colony formation assay.
However, to date, none of these recombinant mutants (or indeed, any
G
q mutants) have been found in pituitary tumors. A previous study
had exclusively analyzed 8 TSH tumors, which are also linked to G
q
in pituitary cell signaling (17). No mutations were found. Our study
analyzed 37 pituitary tumors of various types, and most importantly,
the samples in this study include 8 FSH and LH tumors. GnRH is the
releasing hormone for FSH and LH and is linked to G
q in pituitary
cell signaling (3). FSH and LH tumors, thus, are the only
hormone-secreting tumors, other than TSH tumors, most likely to harbor
G
q mutations. Furthermore, FSH and LH tumors are more prevalent than
TSH tumors, because they probably account for the majority of
clinically nonfunctioning pituitary adenomas (18). Despite these
tantalizing prospects, our results from direct DNA sequencing of 37
tumor samples, and those of Dong et al. (17), in 8 TSH
adenomas argue strongly against the possibility that G
q is a
significant factor in pituitary tumorigenesis, because there were no
mutations in the sequences encoding Arg183 or
Gln209 or the surrounding bases in the conserved
GTP-binding and hydrolysis domains.
The use of RT-PCR to screen for dominant oncogenic mutations is based
on the assumption that tumor cells express the mutant allele at levels
comparable with those of the normal allele. The quality of the cycle
sequencing data, generated in both directions for regions of interest
in G
q, makes it unlikely that a heterozygous mutant allele would
have gone undetected, even if small proportions of normal tissue had
unintentionally been included during surgical resection of the tumor
samples. SSCP of double-stranded DNA fragments through a nondenaturing
MDE gel also is regarded as a highly reliable method for detection of
heterozygous mutations. SSCP has proven successful in detecting
single-base substitutions in other G protein-coupled receptor genes
(17). The sensitivity of the procedure used here for screening the
G
q coding sequences was validated by direct sequencing.
In contrast to somatotroph adenomas, very little is known about the
pathogenesis of clinically nonfunctional, TSH-, FSH-, or LH-secreting
pituitary tumors. The failure to detect known activating mutations of
the G
q leaves open the search for alternative mechanisms underlying
adenoma formation. Although we have excluded those recombinant
mutations known to impair GTPase activity in G
q as a common cause of
pituitary tumors, we have not excluded the possibility that activating
mutations exist elsewhere in their coding sequence, or that changes in
the expression levels of these
-subunits are associated with
pituitary tumor formation. Another possibility is that dysregulation of
components downstream from G
q contribute to tumorigenesis.
Inappropriate activation of other G proteins that might be coupled to
pituitary cell signaling pathways, including those that control
membrane Ca2+ conductance and MAP kinase activity, remain
to be explored. Mutations that lead to abnormal activity of different
stimulatory hormone or growth factor pathways also need to be
considered. It is also possible that pituitary tumor development is
related to the loss of one or more inhibitory inputs.
In conclusion, the recombinant mutations known to impair GTPase
activity in the G
q genes occur infrequently, if at all, in human
pituitary tumors. Alternative mechanisms, underlying pituitary
tumorigenesis , need to be explored.
| Acknowledgments |
|---|
Received June 9, 1997.
Revised August 21, 1997.
Accepted August 26, 1997.
| References |
|---|
|
|
|---|
-subunit adenomas: clinical findings and endocrinology. In:
Landolt A, Vance M, Reilly P, eds. Pituitary adenomas. New York:
Churchill Livingstone; 127138.
This article has been cited by other articles:
![]() |
C.-O. Evans, A. N. Young, M. R. Brown, D. J. Brat, John. S. Parks, A. S. Neish, and N. M. Oyesiku Novel Patterns of Gene Expression in Pituitary Adenomas Identified by Complementary Deoxyribonucleic Acid Microarrays and Quantitative Reverse Transcription-Polymerase Chain Reaction J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 3097 - 3107. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. L. M. Dahia and A. B. Grossman The Molecular Pathogenesis of Corticotroph Tumors Endocr. Rev., April 1, 1999; 20(2): 136 - 155. [Abstract] [Full Text] |
||||
![]() |
S. L. Asa and S. Ezzat The Cytogenesis and Pathogenesis of Pituitary Adenomas Endocr. Rev., December 1, 1998; 19(6): 798 - 827. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |