| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Eppley Institute for Research in Cancer and Allied Diseases (A.K.M.-L., E.B., C.F.B., R.W.R.), the Departments of Biochemistry and Molecular Biology (A.K.M.-L., R.W.R.), Pharmacology (E.B., R.W.R.), and Obstetrics and Gynecology (E.B., J.R.. V.M.), University of Nebraska Medical Center, Omaha, Nebraska 68198
Address all correspondence and requests for reprints to: Dr. Elliott Bedows, Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, 986805 Nebraska Medical Center, Omaha, Nebraska 68198-6805. E-mail: ebedows{at}unmc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit in Chinese
hamster ovary cells and 293T cells. When we examined ßV79M hCG
biosynthesis, we detected atypical ßV79M hCG folding intermediates,
including a ßV79M conformational variant that resulted in a
ß-subunit with impaired ability to assemble with the -subunit. The
inefficient assembly of ßV79M hCG appeared to be independent of
ß-subunit glycosylation or of the cell type studied, but, rather, was
due to the inability of the ßV79M subunit to fold correctly. The
majority of the V79M ß-subunit synthesized was secreted as
unassembled free ß. Although the amount of ß hCG heterodimer
formed and secreted by ßV79M-producing cells was less than that by
wild-type ß-producing cells, the hCG that was secreted as ß V79M
heterodimer exhibited biological activity indistinguishable from that
of wild-type hCG. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit and
a unique ß-subunit that defines biological specificity (1). hCG is
maximally secreted from the trophoblastic cells of the placenta during
the first trimester to stimulate the corpus luteum to produce
progesterone and maintain pregnancy. Of the six homologous hCGß genes
located on chromosome 19q13.3, hCGß gene 5 is usually the most highly
expressed (2, 3). Clinical studies have shown that single bp mutations in the ß-subunit genes of LH, TSH, and FSH can cause clinical disorders. Weiss et al. (4) showed that a mutation in the LH-ß DNA sequence resulting in a Q54R change in the ß-subunit amino acid sequence prevents the molecule from binding to its receptors and causes infertility and lack of spontaneous puberty. A mutation in TSH-ß (E12STOP) was shown to cause familial hypothyroidism, characterized by mental and growth retardation (5). A single base deletion at TSH cysteine codon 105 that caused a frameshift accounted for hypothyroidism in two related families (6). Moreover, a 2-bp deletion in codon 61 of FSH-ß, producing a frameshift mutation, was shown to be responsible for primary amenorrhea and infertility (7). However, to date, naturally occurring genetic disorders of hCGß have not been reported.
Using the formation of disulfide (S-S) bonds as an index
of folding, the hCGß folding pathway has been elucidated. This
pathway, which has been shown to be the same whether hCGß folding
occurs when the hormone is eutopically expressed in trophoblast-derived
JAR cells (8) or in transfected CHO cells (9) or if the subunit is
folded in vitro (10), is depicted in Eq I below (where
the numbers above the arrows indicate the S-S bonds that
are formed in each folding intermediate):
 |
pß1-early is the earliest intracellular folding
intermediate detectable, and is converted to pß1-late with a
t1/2 of conversion of 2–3 min as the earliest disulfide
linkages, 34–88 and 38–57, are formed. Next, the 9–90 and 23–72
disulfide bridges form converting pß1-late into assembly competent
pß2-uncombined. The formation of the 93–100 disulfide bond coincides
with the assembly of the ß-subunit with . The last bond to form
during the kinetic folding pathway of hCGß is 26–110, which appears
to function as a "seat belt" for the -subunit (12). It should be
noted that the crystal structure of secreted hCG (12, 13) reveals S-S
bonds formed between Cys residues 38–90 and 9–57 rather than between
Cys residues 38–57 and 9–90 and implies that a S-S bond rearrangement
occurs during the folding or processing of hCGß. Previous studies
have demonstrated that mutations in certain cysteine residues that form
the disulfide bonds of hCGß cause the protein to misfold, not
assemble with , or to be secreted abnormally (14, 15, 16). When
asparagine residues located at positions 13 and 30, which form
consensus sequences for N-linked glycosylation, were changed
to glutamine residues by site-directed mutagenesis, the resultant
hCGß proteins exhibited unusually slow folding rates (17). Therefore,
mutations in the glycoprotein ß-subunit cause alterations that affect
biological function, suggesting that naturally occurring variants of
CG-ß may also adversely affect the biological function of the
hormone.
As the correct folding and assembly of hCG are crucial for its
activity, defects of hCG may be at least in part responsible for some
cases of spontaneous abortions or infertility. In our study, the
sequence of ß gene 5 of 334 random individuals, 27 infertile women,
and 14 of their husbands was examined. We report here that although
most of the polymorphisms detected occurred within intron regions or
were silent, one polymorphism was discovered within the coding region
of exon 3 of the hCGß gene 5 in 4.2% of the individuals studied.
This resulted in the expression of a ß-subunit containing the amino
acid substitution V79M. Characterization of the protein-folding
phenotype of the V79M genetic variant revealed that this ß-subunit
generated atypical folding intermediates, including a conformational
variant of the assembly-competent hCG pß2. The V79M variant of pß2
was impaired in its ability to assemble with the glycoprotein hormone
-subunit to produce ß heterodimer, resulting in more than 85%
of the ßV79M secreted as unassembled ß, as opposed to 50% or less
in cells expressing the wild-type ß-subunit.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PCR
hCGß gene 5 was selectively amplified from the white blood
cell DNA of random subjects, infertile patients, and plasmid containing
gene 5 using primers 1158 sense and 3'-antisense (Fig. 1
). Plasmids containing genes 5, 3, and 8
were gifts from Dr. Irving Boime (Washington University, St. Louis,
MO). Plasmids were used to design PCR primers specific to gene 5 and as
templates in the optimization of the gene 5-specific PCR reaction.
Primers were designed with a 3'-terminal nucleotide mismatch to all
other hCGß genes to selectively amplify gene 5 (26). All
oligonucleotide primers were synthesized by the Eppley Institute
Molecular Biology Core Laboratory on an PE Applied Biosystems (Foster City, CA) model 394 DNA/RNA synthesizer.
Reactions were carried out in a total of 50 µl containing 1.25 U
Taq DNA polymerase (Life Technologies, Inc.,
Gaithersburg, MD) in Life Technologies PCR buffer solution
consisting of 20 mM Tris-HCl (pH 8.4); 50 mM
KCl; 0.125 mM each of deoxy (d)-ATP, dCTP, dGTP, and dTTP;
1.25 mM MgCl2; 0.5 µg 1158 sense primer; 0.5
µg 3'-antisense primer; approximately 0.8 µg genomic DNA; and 0.5 U
Perfect Match DNA polymerase enhancer (Stratagene, La
Jolla, CA). Reactions were placed in a Perkin Elmer 2400
thermal cycler under conditions of 94 C for 4 min followed by 35 cycles
of 94 C for 45 sec, 62 C for 45 sec, and 72 C for 45 sec and a final
extension of 7 min at 72 C. As a negative control, DNA was left out of
one PCR reaction for every set of primers used. Fifteen microliters of
each product were electrophoresed on a 1% (wt/vol) agarose gel in
Tris-glacial acetic acid-EDTA buffer. The gel contained 0.2 µg/ml
ethidium bromide to visualize the 990-bp hCGß DNA fragments under UV
light. A cotton-plugged Pasteur pipette was used to remove two agarose
plugs from each amplicon band. Each plug served as an enriched template
for the second PCR reactions to separately amplify the two coding exons
and their flanking intron regions. The second PCR mixture contained
sense primer hCG7 or exon 3 sense, antisense primer hCG2 or
3'antisense, and 2.5 µCi [-32P]dCTP (>3000
Ci/mmol), with the rest of the components identical to those in the
first PCR mixture. Primer location and sequences are illustrated in
Fig. 1. Thermal cycling conditions were the same as those used for the
first PCR reaction. Three microliters of each of the
[-32P]dCTP-labeled PCR reactions were then digested
with a restriction enzyme for 1 h at 37 C. The amplified exon 2
was digested with 5 U HincII, yielding DNA fragments of 233
and 99 bp. The amplified exon 3 was digested with 5 U PvuII,
yielding DNA fragments of 222 and 224 bp. Seven microliters of
formamide stop solution containing 95% formamide, 20 mM
EDTA, 0.05% bromophenol blue, and 0.05% xylene cyanol were added to
the digested material, which was subsequently placed in a boiling water
bath for 4 min and then immediately placed on ice. One normal,
nondenatured sample of each exon was always included as a control.
|
DNA sequencing
DNA showing unusual band migration on the SSCP gels was
reamplified with oligonucleotide primers 1158 sense and 3'-antisense
and sequenced using either hCG9 or hCG10 (Fig. 1
) as the sequencing
primer. Sequencing analyses were performed on a 373 ABI stretch
sequencer (PE Applied Biosystems, Foster City, CA) by the
Eppley Institute Molecular Biology Core Laboratory. Ten random
specimens exhibiting wild-type SSCP patterns for exon 2 and five
specimens for exon 3 were sequenced. Thirty-two random specimens
exhibiting unusual SSCP patterns were sequenced. Each atypical
SSCP pattern was specific for one, and sometimes two, DNA base changes.
DNA from all of the infertile patients was amplified with
oligonucleotide primers 1158 sense and 3'-antisense and sequenced.
Site-directed mutagenesis
Wild-type hCGß inserted into the pGS vector, obtained from Dr.
Tyler White, Scios, Inc. (Mountain View, CA), was mutated
in a two-step PCR reaction. An antisense primer, V79M, was designed to
incorporate the G to A nucleotide change to produce a methionine
residue in place of the wild-type valine (Fig. 1). The 5'- and
3'-primers used in the PCR reactions were responsible for incorporating
restriction enzyme site for cloning into the final products (Fig. 1).
PCR reactions were carried out in a total of 100 µl containing 2.5 U
Pfu DNA polymerase (Stratagene, La Jolla, CA)
in 1 x native Pfu PCR buffer solution consisting of
200 mM Tris-HCl (pH 8.75); 100 mM
(NH4)2SO4; 20 mM
MgSO4; 1% Triton X-100; 1000 µg/ml BSA; 100
mM KCl; 0.125 mM each of dATP, dCTP, dGTP, and
dTTP; 0.5 µg HindIII sense primer (Fig. 1); 0.5 µg V79M
antisense primer; and 15 ng wild-type hCG-ß plasmid. Reactions were
placed in a Perkin-Elmer 2400 thermal cycler (Norwalk, CT)
under conditions of 94 C for 4 min followed by 35 cycles of 94 C for 30
sec, 60 C for 30 sec, and 74 C for 30 sec and a final extension of 7
min at 74 C. Negative controls were performed as well. The products
were purified using the QiaQuick PCR spin column purification kit
(QIAGEN, Valencia, CA) and were used as the sense primer
for the second PCR reaction. The reaction to amplify the entire coding
sequence of hCGß was carried out as described above with the V79M
mega primer as the sense primer and XbaI antisense
(Fig. 1) as the 3'-primer. The product was purified using a
QiaQuick PCR spin column and was then ligated into empty pGS plasmid
using the HindIII and XbaI restriction enzyme
sites and T4 DNA ligase (Life Technologies, Inc.).
Competent DH5 cells (Life Technologies) were
transformed with the plasmid containing the mutated hCGß insert.
Positive clones were grown overnight in 1 ml YT media (8 g tryptone,
5 g yeast extract, and 5 g NaCl), purified by allcaline lysis
and checked for the presence of single, full-length inserts by
restriction digestion. A large scale culture was made of one clone.
Plasmid was subsequently purified using the QIAGEN Maxi
Plasmid Kit, and the plasmid insert was sequenced to check for possible
polymerase errors.
Transfection into Chinese hamster ovary (CHO) cells and human 293T
cells
CHO cells stably expressing hCG were grown on 100-mm culture
dishes to approximately 30% confluence in F-12 medium containing 5%
FBS and 0.2 mg/ml G-418. The G-418 was needed to maintain the hCG
gene in its G-418-resistant vector in the cells. One to 10 µg plasmid
containing either the wild-type ß or V79M ß were diluted to 450
µl in autoclaved water. Fifty microliters of 2.5 M
CaCl2 were added to the plasmid, and this mixture was added
dropwise to 500 µl 2 x HEPES-buffered saline while vortexing.
The resulting precipitate was placed on ice for 10 min, then added
dropwise to the cells with fresh medium and incubated at 37 C in 5%
CO2. The medium with the DNA precipitate was removed from
the CHO cells 24 h later. Cells were fed glutamine free
Ultraculture (BioWhittaker, Wakersville, MO) containing 0.2 mg/ml
G-418, and 30 µM methionine sulfoximine was added to the
cells to select for those that had incorporated the pGs methionine
sulfoximine-resistant ß plasmid. Fresh medium was placed on the cells
every 4 days until colonies appeared. These colonies were isolated and
examined by PCR for the ß insert. Positive colonies were grown to use
for metabolic labeling experiments.
293T cells were grown in 60-mm culture dishes to approximately 70%
confluence in DMEM containing 5% FBS. One to 10 µg plasmid
containing either the wild-type ß, V79M ß, or the wild-type
gene were transfected as above. The 293T cells were used for metabolic
labeling 48 h later.
Metabolic labeling of cells with radioactive substrates and
immunoprecipitation of hCGß
CHO cells or human 293T cells grown to 90–95% confluence in
100-mm plastic dishes were pulse labeled for the times indicated with
L-[35S]cysteine (1100Ci/mmol; DuPont NEN, Boston, MA) at a concentration of 200–300 µCi/ml in
serum-free medium lacking cysteine (9). All pulse incubations were
carried out as previously described (9), and the cells were incubated
for chase times of 0 min to 24 h. Cells were harvested with cold
PBS containing detergents, protease inhibitors, and 50 mM
iodoacetic acid (pH 8.0) to trap the free sulfhydryl groups of the ß
folding intermediates as previously described (9). Cell lysates were
centrifuged, and immunoreactive forms were precipitated with a goat
polyclonal antiserum that recognizes all known folding intermediates of
wild-type hCGß, as previously described (9). Immune complexes were
precipitated with protein A-Sepharose (Sigma Chemical Co.,
St. Louis, MO) and prepared for SDS-PAGE or reverse phase HPLC as
described below.
SDS-PAGE
Radiolabeled hCG forms that adsorbed to protein A-Sepharose
beads were eluted with 2-fold concentrated SDS gel sample buffer (125
mM Tris-HCl, pH 6.8, containing 2% SDS, 20% glycerol, and
40 µg/ml bromophenol blue). Samples run under reducing conditions
were boiled for 4 min in sample buffer containing 2%
ß-mercaptoethanol, whereas samples run under nonreducing conditions
were boiled for 4 min in sample buffer lacking ß-mercaptoethanol. The
samples, including the protein A-Sepharose beads, were applied to
polyacrylamide gradient slab gels (5–20%) that were run by the method
of Laemmli (18). Gels were dried in vacuo on filter paper
and exposed to a phosphor screen (Molecular Dynamics, Inc., Sunnyvale, CA) overnight. The phosphor screen was then
scanned on a PhosphorImager (Molecular Dynamics, Inc.) and
analyzed by ImageQuant software version 4.1 (Molecular Dynamics, Inc.). Quantitation of gel images were obtained by integrating
the volume of each band to give absolute intensities of the bands.
Reverse phase HPLC analysis
The hCGß folding intermediates pß1 and pß2 were purified
by a two-step process (immunoprecipitation followed by C4
reverse phase HPLC) as described by Huth et al. (8).
Briefly, pß1 and pß2 were immunoprecipitated from cell lysates with
polyclonal antisera, and immunocomplexes were precipitated with protein
A-Sepharose beads. To dissociate precipitated immunocomplexes, pellets
were treated with 6 M guanidine-HCl (pH 3, sequenal grade;
Pierce Chemical Co., Rockford, IL) for 16 h at room
temperature with 100 mg myoglobin (Sigma Chemical Co.) as
carrier. After low speed centrifugation to remove protein A-Sepharose
beads, the guanidine eluates were injected onto a Vydac 300-Å
C4 reverse phase column (The Separations Group, Hesperia,
CA) equilibrated with 0.1% trifluoroacetic acid (TFA) and eluted using
an acetonitrile gradient as previously described (8, 9, 10).
Tryptic digestions
Nonreduced hCGß forms were digested for 16 h in silanized
polypropylene tubes containing 100–200 µg myoglobin, 0.03% trypsin
(Sigma Chemical Co.), 5 mM CaCl2,
and 50–150 mM Tris-HCl, pH 8 (8, 9, 10). The digestion was
continued with the addition of two aliquots each of 50 µg trypsin
(0.06%, final concentration).
HPLC purification of tryptic peptides
hCGß tryptic digests were injected onto a Brownlee
C18 300-Å reverse phase column (8, 9, 10) equilibrated with
0.1% TFA. The column was eluted isocratically for 3 min with 0.1% TFA
followed by a 0.32%/min acetonitrile gradient in 0.1% TFA for 100
min. The column was washed with 80% acetonitrile-0.1% TFA for 5 min
and then reequilibrated in 0.1% TFA. The flow rate was 1.0 ml/min;
1-min fractions were collected in silanized polypropylene tubes. Tubes
into which disulfide-linked peptides eluted contained 4.5 µg
myoglobin. Samples were concentrated by Speed-Vac centrifugation
(Savant Instrument Co., Farmingdale, NY) and stored at -70 C. A
portion of each sample was added to 5 ml ECOLUME (ICN Biomedicals, Inc., Costa Mesa, CA) scintillation fluid and
counted in a Beckman Coulter, Inc. model LS 3801 liquid
scintillation spectrometer (Palo Alto, CA).
Ion exchange HPLC
A reverse phase HPLC peak that eluted at 40 min contains two
peptides in the wild-type ß folding intermediates. These peptides
were resolved by ion exchange chromatography using a Waters
diethylaminoethyl-anion exchange column (Waters Corp.,
Milford, MA) (9). The peptides were eluted during a 30-min 0- to 300
mM NaCl gradient at a constant pH of 7 in 20 mM
phosphate buffer. One-minute fractions were collected and analyzed by
scintillation counting to mark the elution positions of
[35S]cysteine-containing peptides.
Assay for hCG biological activity
The biological activity of hCG was assayed as described by Ho
et al. (19). Briefly, immunoreactivity of the hCG
heterodimer secreted from either CHO cells expressing the common
glycoprotein hormone -subunit and either the wild-type ß-subunit
or the ßV79M subunit was determined by a two sandwich-type
immunoenzymometric assay using the B-109 monoclonal antibody as coating
antibody and rabbit anti-hCG CR121 antiserum as secondary antibody
(20). Biological activity measurements of hCG samples were determined
by a luminescence LH/CG bioassay employing a 293T cell line stably
transfected with human LH/CG receptor complementary DNA and a
luciferase reporter gene driven by a cAMP-dependent promoter (21). CR
127 (a gift from Dr. John O’Connor) served as the standard in both
reactions. Ratios of biological activity to immunoreactivity were
calculated for each sample. Each sample was run at least three times. A
biological activity/immunoreativity ratio of unity indicates that the
hCG secreted by either cell type had biological activity equivalent to
that of CR 127.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) were created to target gene 5 only, and have 3'
nucleotide mismatches to the other ß genes. To show the selectivity
of these primers, PCR was performed on control samples consisting of
genomic DNA. The products were sequenced to confirm that gene 5 was the
only ß gene amplified (data not shown). hCGß genes 3, 8, 7, 1, and
2 differ from gene 5 at specific nucleotides in the 5'-end of the PCR
product, and none of these differences were detected in the sequencing
data, indicating that only gene 5 had been amplified. Amplification
reactions were also performed with primers 1158 sense and 3'-antisense,
using plasmids containing ß gene 5, 3, or 8 as DNA templates to show
that hCGß gene 5 was the only ß gene capable of being amplified
with these primers. A PCR product was detected only when the DNA
template was the plasmid construct containing the ß gene 5 insert
(data not shown). Thus, we were able to determine that gene 5 was
amplified without interference from other ß genes.
To examine the DNA from human blood specimens, white blood cell genomic
DNA was subjected to PCR with primers 1158 sense and 3'-antisense. A
portion of each amplification reaction was electrophoresed on an 1%
agarose gel (Fig. 2). The 990-bp
amplification products were isolated from the agarose gel, and the
gel-DNA fragment served as a hCGß gene 5-enriched template for the
subsequent PCR reactions. Exons 2 and 3, which are the coding exons for
the mature protein, were then amplified in individual PCR reactions.
These products were analyzed by SSCP gels. This method allows single
stranded DNA to migrate on the gel according to its size and
three-dimensional conformation. Previous reports (22, 23) have shown
that even a 1-bp change in a nucleotide sequence of 100–300 bp in
length can alter the conformation from the wild-type structure and give
a unique band migration. Examples of the polymorphisms detected in exon
3 are depicted as an autoradiogram in Fig. 3. Although these data indicate that
polymorphisms exist in hCGß gene 5, DNA sequencing of subjects with
altered SSCP banding patterns was required to identify the exact
location of the nucleotide substitutions.
|
|
), a
G to A substitution located 37 bp upstream from exon 2 (-37; Table 1),
and a C to G substitution located 21 bp downstream from exon 2 (+21;
Table 1). In the 334 DNA samples examined for polymorphisms in exon 3,
two substitutions were identified, a silent C to T substitution in
codon 82 (Tyr) and a G to A base change in codon 79 (Val) that changed
this amino acid residue to Met (V79M). The latter was the only
polymorphism detected that changed the primary amino acid sequence of
hCGß. This base change was always found as a heterozygote and was
always detected in conjunction with the silent tyrosine polymorphism in
codon 82. The ßV79M heterozygote was found in 14 of 334, or 4.2%, of
the random subjects analyzed.
|
To determine the genotype frequencies for the polymorphisms identified,
the Hardy-Weinberg principle was employed. This equation states that
the genotypic frequencies of a gene with two different alleles are a
binomial function (p2 + 2pq + q2 = 1) of
the allelic frequencies (24). The allelic frequency is calculated by
dividing the number of each allele by the total number of alleles,
thereby generating the number of persons expected with each genotype.
This can then be compared with the actual number of observed persons
for each genotype. The polymorphisms, frequencies for each allele, and
numbers of expected vs. observed genotypes are shown in
Table 1. The gene frequencies for the infertile patients were virtually
identical to those for the random population (data not shown); however,
the -37, +21, and V79M polymorphisms were not detected in the
infertile patients, most likely due to their low frequencies. As the
only change of significance found that altered the primary amino acid
sequence of the hCGß gene 5 product was the V79M variation, we
studied the biological impact of this alteration further.
Characterization of folding of the V79M variant
We examined the V79M ß-subunit to determine whether the genetic
alteration affected hCG folding and bioactivity. It is known that the
assembly of the ß-subunit with is required for hCG to elicit
biological activity and that folding of the hCGß subunit is the
rate-limiting step for the assembly of the hCG heterodimer (8, 9, 25).
Therefore, we asked whether the ßV79M subunit folded, assembled, and
was secreted like the wild-type ß-subunit. The illustration in Fig. 4 depicts the location of amino acid
valine 79 in the crystal structure of the mature hCG protein (12, 13).
As the V79M amino acid change occurs in a region that could influence
the hydrophobic interactions required during the earliest nucleation
events of protein folding, we wanted to examine the effect of the V79M
genetic variation on the early steps of hCGß protein folding.
|
-subunit and cotransfected with
either wild-type hCGß or ßV79M were pulse labeled for 5 min with
[35S]Cys and chased for the times indicated in Fig. 5. Figure 5A shows that, as previously
seen (9), conversion from the earliest detectable wild-type ß-subunit
folding intermediate, pß1, to the second major folding intermediate,
pß2, occurred in CHO cells expressing wild-type ß subunit with a
t1/2 of 5 min. Moreover, detection of the -subunit after
immunoprecipitation with polyclonal antiserum to the ß-subunit
indicates that heterodimer assembly occurred (Fig. 5A, chase times 30
and 180 min secreted) (8, 9, 10, 14). However, the pß1 folding
intermediate of ßV79M was not detected by nonreducing SDS-PAGE after
0 min of chase (Fig. 5B), although the appearance of the more slowly
migrating pß2-like folding intermediate was seen within 5 min. That
these are folding intermediates and not mature ß is confirmed by the
observation that they reduce to the typical early ß folding forms
seen for wild-type ß (Fig. 5, C and D). Previous evidence (16)
suggests that the three ß isoforms in Fig. 5, C and D, are due to
carbohydrate differences.
|
, A and B after 15–90
min of chase. There was an additional diffuse band in the ßV79M
immunoprecipitates migrating at a Mr from 35,000–40,000
and a band at a Mr between 29,000–30,000 that migrated
slightly slower than the pß2 band from wild-type ß
(Mr = 27,000–28,000). The ßV79M forms collapsed to
bands identical to those seen in cells expressing wild-type ß on
reducing SDS-PAGE (Fig. 5, C and D), demonstrating that differences in
migration on nonreducing SDS-PAGE between ßV79M and wild-type ß are
lost when disulfide bonds are reduced.
The mature ß-subunit is secreted from CHO cells expressing wild-type
ß (Fig. 5A, lane S). Mature ß-subunit is also secreted from CHO
cells expressing ßV79M subunit (Fig. 5B, lane S) and can be discerned
at a Mr of 35,000 from ß folding intermediates by
reducing SDS-PAGE after 3–24 h of chase (Fig. 5D). Therefore, the
ßV79M subunit is capable of forming a secretion-competent form.
The fact that ßV79M pß2-like intermediates were formed with
kinetics similar to those of wild-type pß2 subunit, even though we
were unable to detect the earliest folding intermediate (pß1) under
nonreducing SDS-PAGE conditions, suggests that V79M pß1 has a
conformation different from that of wild-type pß1 and may not be
recognized by our polyclonal hCG antibody. To test this hypothesis, we
performed the following experiment. We repeated the pulse-chase
labeling experiment described above and precipitated with polyclonal
hCG antiserum but instead of analyzing the immunocomplexes generated
after protein A-Sepharose precipitation by nonreducing SDS-PAGE (Fig. 5, A and B), we analyzed aliquots of these reaction products by
reversed phase HPLC (Fig. 6). That
pß1-like folding intermediates were present at early chase times in
the cells expressing ßV79M is shown by the fact that a pß1 form was
detected when the immunoconjugates were eluted from the Sepharose beads
with guanidine hydrochloride and separated by HPLC (Fig. 6). The HPLC
elution profile shown in Fig. 6 for the ß folding intermediates from
the V79M cells is identical to that seen for the ß forms detected in
CHO cells expressing the unmutated ß gene (8, 9, 10). This suggests that
the pß forms produced in the cells expressing ßV79M are less stable
when boiled in SDS (before SDS-PAGE) than pß1 from wild-type cells or
that they may be heterogeneous in nature and cannot be resolved by
SDS-PAGE under nonreducing conditions. The data shown in Fig. 6 also
demonstrate that the t1/2 of conversion of V79M pß1 to
pß2 was 15 min (as calculated by the counts per min present in each
peak over time), 3 times slower than the 5-min t1/2 of
conversion of pß1 to pß2 in CHO cells expressing the wild-type
ß-subunit (Fig. 5) (9).
|
) (8, 9, 10). The
tryptic digestion peak that eluted at 40 min was disproportionally
larger in the ßV79M variant for both pß1 and pß2 (Fig. 7). This
peak contains peptides 69–74 and 105–114 in wild-type ß
intermediates, which indicates that bonds 23–72 and 26–110,
respectively, are not formed. As these peptides coelute at 40 min (8, 9), it was necessary to collect this peak and examine it further by
anion exchange chromatography, which separates the peptides. When anion
exchange was performed on the 40-min pß2 V79M tryptic digest peak,
only peptide 69–74 was detected (data not shown). These data indicate
that the cysteine at amino acid residue 110 formed a disulfide linkage.
In the wild-type folding pathway, disulfide bond 26–110 is the last
bond to form, and it acts as a seat belt for the -subunit that
stabilizes the ß heterodimer (12). Perhaps the premature formation
of a disulfide linkage with cysteine 110 precludes the ability of
ßV79M to assemble with the -subunit.
|
coprecipitated with the hCGß-specific
polyclonal antibody than in cells expressing wild-type CG (Fig. 5, A
and B). Therefore, we were interested in examining whether the hCG
heterodimer assembly or secretion was affected by the change in amino
acid composition in ßV79M. To do this, CHO cells expressing both the
-subunit and the V79M mutant ß-subunit were pulse labeled for 5
min and chased for periods from 0 min to 24 h, and the secreted
material was isolated from chase medium by immunoprecipitation with
polyclonal ß antiserum (Fig. 8).
|
, after the 3-, 8-, and 24-h chases, most of the
secreted ßV79M was unassembled (free) ß. However, as the ratio of
- to ß-subunit expressed in the transfected CHO cells used in this
experiment was 1:15, and as we had previously shown that efficient
ß assembly is dependent on an /ß ratio of 1:2 to 1:1.2 in CHO
cells (9), we used a CHO cell line transfected with ßV79M and the
wild-type -subunit at an /ß ratio of 1:2.5. Cells were pulsed
for 5 min and chased for up to 24 h (Fig. 9). The percentage of ß assembled into
ß heterodimer was calculated for both V79M and wild-type ß. The
percentage of wild-type ß assembled with reached 55%, whereas
<10% of V79M ß was assembled with and secreted. In addition,
293T cells were transiently transfected with ßV79M and at /ß
ratios of 1:4 and 1:6. In these cells, <10% of ß V79M was assembled
into ß heterodimer after 4 h, whereas 45–55% of wild-type
ß was assembled with the -subunit (data not shown).
|
-subunit and
either the wild-type ß-subunit or the ßV79M subunit was determined
as described in Materials and Methods. As Table 2 shows, the hCG secreted from either cell type
displayed comparable biological activity even though the total amount
of ßV79M CG secreted was lower. This indicates that the ßV79M
subunit that was able to combine with the -subunit had full
biological activity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We performed tryptic digests of the V79M pß1 and pß2 intermediates
isolated from C4 reverse phase HPLC to compare the
disulfide bond formation with wild-type bond formation. The V79M pß1
and pß2 tryptic maps differed from wild-type ß maps in that the 40
min tryptic peak was larger for the V79M variant. In wild-type pß1
and pß2 this peak contains two peptides, 69–74 and 105–114;
however, only peptide 69–74 was identified in the V79M pß2. These
data indicate that either native disulfide bond 26–110 forms
prematurely in the variant or that cysteine 110 forms a nonnative
disulfide bond. The 26–110 bond forms after wild-type pß2 has
combined with the -subunit and appears to function as a seat belt to
stabilize the hCG heterodimer (10, 12). Premature formation of a
disulfide bond with cysteine 110 in ßV79M may hinder its ability to
assemble with the -subunit.
There is precedence that a single amino acid substitution can
contribute to misfolding of proteins and also have clinical correlates.
For example, a mutation of valine 18 to alanine in the N-terminal
domain of the amyloid-ß peptide induced a significant increase in
-helical structure in an Alzheimer’s patient (27). Valine 79 is
located in the loop 3 region (H3-ß) of the hCGß subunit (28).
H3-ß is hypothesized to contribute to the secondary structure by
causing a hydrophobic collapse during the initial stages of folding.
Some of the key hydrophobic amino acids that are believed to play a
role in the structure of H3-ß by the formation of hydrogen bonds,
such as I67, L69, V80, and Y82, surround V79 (28). Perhaps the
substitution of valine to methionine disrupts the hydrogen bonding and,
as a result, allows the protein to form unusual conformers. This
methionine residue may destabilize the initial steps in the hydrophobic
collapse of the protein, and in pß1, the C-terminus or other portions
of the molecule may not be able to adopt a stable conformation, instead
yielding many heterogeneous forms. Consistent with this was the
observation that we did not see a distinct pß1 band, as we do with
wild-type hCGß, under nonreducing gel conditions, but did using HPLC.
As the wild-type ß protein folds to a more compact structure,
i.e. pß2, with more disulfide bonds to stabilize the
conformation, the subunit becomes less heterogeneous and can be
detected as a distinct band by nonreducing SDS-PAGE, whereas the
pß2-like intermediate formed by ßV79M remains more
heterogeneous.
A number of mutations in hCGß have been produced by site-directed
mutagenesis and studied in cell culture systems. That these studies
accurately reflect events that occur within the trophoblast cells of
the placenta is consistent with our observations that the folding
pathway of hCGß is indistinguishable in trophoblast-derived JAR
(choriocarcinoma) cells (8), primary syncytiotrophoblast cells (Bedows,
E., unpublished), ß gene-transfected CHO cells (9), and, as described
in this report, ß gene-transfected 293T cells. Many mutant forms of
hCGß are blocked at various steps in the folding pathway and fold
more slowly than wild-type ß or are defective in their assembly with
the -subunit (11, 14, 15). If similar mutations were found in the
human population, it is likely that a nonfunctional form of hCG would
be produced, and that placentas producing such altered forms could not
support pregnancy. However, Bedows et al. have demonstrated
in hCGß mutants, such as C23A-C72A (14) and P73G (Bedows, E.,
unpublished), that even though the ß-subunits fold at a slower rate,
the ß dimer that forms has biological activity.
The ßV79M variant was detected in 4.2% (14 of 334 subjects examined) of the population studied and was not detected in its homozygous state. This is possibly because of the low frequency of the variant, as only 1 in more than 600 people would be expected to be homozygous for a genetic variant of this frequency. Therefore, the homozygous trait may not have been detected in our limited patient population. Alternatively, as the ß gene 5 product is usually the major source of hCGß produced during pregnancy (2, 3), ßV79M gene 5 homozygosity may be an embryo-lethal event. In this situation, infertility may occur in individuals whose source of hCGß arises almost exclusively from ß gene 5, whereas pregnancy might be rescued in individuals who produce greater amounts of other ß gene products, e.g. those of ß gene 3 or 8 (3).
As we identified only a small number of polymorphisms in gene 5, and only one polymorphism was identified in the coding region, changes in sequence may be selected against. One base change was identified in an infertile patient, by direct DNA sequence analysis, that was not seen in the random individuals. An A to T transversion was found in intron 2, 25 bp upstream from exon 3. This base change could be critical if this sequence is directly involved with the spliceosome machinery that produces the mature messenger RNA. Alternatively, this intron base change may be a spontaneous, silent polymorphism that occurred randomly in this particular individual. Moreover, it cannot be ruled out that the SSCP analysis of random population specimens may have missed this base change because it is not as sensitive as direct DNA sequencing. It has been documented that SSCP, even under optimized conditions, may only detect 85–95% of polymorphisms or mutations; however, it is a relatively quick and reproducible method to screen a large number of samples (22, 23).
Although it is not known why there are six homologous hCGß genes, this redundancy may help ensure that early pregnancy can be maintained in the event that expression of one of the genes is lost. As many as 10–20% of all couples suffer from infertility, and some of these are cases of frequent, unexplained spontaneous abortions (4), which may in part result from low hCG activity. However, to demonstrate the clinical importance of the ßV79M genetic variant detected in our study, a larger sampling of individuals will be necessary. These studies are currently underway in our laboratory.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Transgenomic, Inc., 5600 South 42nd Street, Omaha,
Nebraska 68107. Recipient of an Emley Fellowship.
3 Current address: Corporate Office of Science and Technology,
Johnson & Johnson, 410 George Street, New Brunswick, New
Jersey 08901.
Received November 23, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß dimer
assembly-competent form. J Biol Chem 265:312–317-helical to ß strand transition in the amino-terminal fragment
of the amyloid ß peptide modulates amyloid formation. J Biol
Chem 270:3063–3067This article has been cited by other articles:
 |
|
 |
|
  Y. Chen, M. G. Kibriya, F. Jasmine, R. M. Santella, R. T. Senie, and H. Ahsan Do Placental Genes Affect Maternal Breast Cancer? Association between Offspring's CGB5 and CSH1 Gene Variants and Maternal Breast Cancer Risk Cancer Res., December 1, 2008; 68(23): 9729 - 9734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Rull, L. Nagirnaja, V.-M. Ulander, P. Kelgo, T. Margus, M. Kaare, K. Aittomaki, and M. Laan Chorionic Gonadotropin {beta}-Gene Variants Are Associated with Recurrent Miscarriage in Two European Populations J. Clin. Endocrinol. Metab., December 1, 2008; 93(12): 4697 - 4706. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P N Themmen An update of the pathophysiology of human gonadotrophin subunit and receptor gene mutations and polymorphisms Reproduction, September 1, 2005; 130(3): 263 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Lamminen, P. Jokinen, M. Jiang, P. Pakarinen, H. Simonsen, and I. Huhtaniemi Human FSH{beta} subunit gene is highly conserved Mol. Hum. Reprod., August 1, 2005; 11(8): 601 - 605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Jiang, M.-L. Savontaus, H. Simonsen, C. Williamson, R. Mullenbach, J. Gromoll, N. Terwort, M. Alevizaki, and I. Huhtaniemi Absence of the genetic variant Val79Met in human chorionic gonadotropin-beta gene 5 in five European populations Mol. Hum. Reprod., October 1, 2004; 10(10): 763 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. G. Crichton, E. Bedows, A. K. Miller-Lindholm, D. M. Baldwin, D. L. Armstrong, L. H. Graham, J. J. Ford, J. O. Gjorret, P. Hyttel, C. E. Pope, et al. Efficacy of Porcine Gonadotropins for Repeated Stimulation of Ovarian Activity for Oocyte Retrieval and In Vitro Embryo Production and Cryopreservation in Siberian Tigers (Panthera tigris altaica) Biol Reprod, January 1, 2003; 68(1): 105 - 113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. H. Burns and M. M. Matzuk Minireview: Genetic Models for the Study of Gonadotropin Actions Endocrinology, August 1, 2002; 143(8): 2823 - 2835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P. N. Themmen and I. T. Huhtaniemi Mutations of Gonadotropins and Gonadotropin Receptors: Elucidating the Physiology and Pathophysiology of Pituitary-Gonadal Function Endocr. Rev., October 1, 2000; 21(5): 551 - 583. [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 |
