| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Center for Research on Reproduction and Women’s Health, Divisions of Human Reproduction and Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: Jerome F. Strauss, III, Center for Research on Reproduction and Women’s Health, 778 Clinical Research Building, 415 Curie Boulevard, Philadelphia, Pennsylvania 19104. E-mail: jstrauss{at}obgyn.upenn.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This study was approved by the Institutional Review Board of the University of Pennsylvania. Follicular fluid was aspirated from individual mature follicles at the time of ultrasound-guided transvaginal ovum retrieval from patients undergoing controlled ovarian hyperstimulation for in vitro fertilization at the Hospital of the University of Pennsylvania. The stimulation regimen involved the administration of human menopausal gonadotropins (a combination of Pergonal and Metrodin) and down-regulation with GnRH analogue, beginning in the luteal phase (4). Seven of the nine patients had severe tubal factor as a cause of their infertility. The remaining two had minimal endometriosis. The average patient age was 35, with a range of 31–39 yr. The average number of follicles recruited in each patient was 12, with a range of 6–20. To prevent specimen contamination, separate syringes were used for each follicle aspiration and follicle flush. Care was taken to assure that each individual follicle’s contents were kept separate. Aspirates containing blood were not analyzed. Granulosa cells were isolated from follicular fluid by a modification of the technique described by Beckmann et. al. (5). After removal of the oocyte, each follicle aspirate was centrifuged, and cell pellets were then resuspended in phosphate-buffered saline (PBS) and layered onto Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden) and centrifuged at 3000 x g for 10 min. Granulosa cells at the Ficoll-PBS interface were collected. A cytologic smear was prepared and examined microscopically to assess the percentage of contaminating cells (leukocytes) using the Papanicolaou stain and immunocytochemical staining for the steroidogenic enzyme, 3ß-hydroxysteroid dehydrogenase (3ß-HSD), which is expressed by the granulosa cells but not by leukocytes (6). The 3ß-HSD polyclonal antibody was generously provided by Dr. Fernand Labrie (University of Laval, Canada) and immunostaining was carried out using the Vectastain Elite ABC Kit (Vector Labs, Burlingame, CA), which is based on the avidin-biotin-peroxidase method. Monkey kidney COS-1 cells were used as negative controls. The percentage of contaminating cells (cells not staining for 3ß-HSD and cells having the morphologic features of leukocytes) was assessed by a cytopathologist.
An additional purification step was performed on an aliquot of the follicular aspirates from two patients. This step involved the removal of contaminating leukocytes by anti-CD45 magnetic immunobeads (7). Fifty percent of the follicle aspirate was suspended in PBS with 30% FCS and was incubated with 25 µg anti-CD45 magnetic immunobeads (Immunotech, Inc., Westbrook, ME) for 10 min. This suspension was then placed in a magnetic support (Immunotech, Inc.) for 10 min, which clustered the immunobeads and their bound leukocytes to one side of the tubes, thus allowing the aspiration of the remaining granulosa cell-enriched suspension. The cells were then collected by centrifugation, and an aliquot was used to prepare a smear for cytologic evaluation. The other portion of each of these follicle aspirates was processed by the routine procedure described above. The two sets of granulosa cell preparations were matched, and both the cytologic assessment and the PCR results after HpaII digestion were compared.
The granulosa cell preparations, as assessed by the cytologic methods described above, contained a relatively low proportion of contaminating cells, with a range of 2–4%. The additional step of anti-CD45 magnetic immunobead purification did not significantly change the proportion of contaminating cells (1–3%). The contaminating cells consisted of ovarian fibroblasts and leukocytes.
Clonality assay
The assay for clonality is summarized in schematic form in Fig. 1
.
|
Each granulosa cell pellet was resuspended in 200 µL proteinase K lysis buffer [400 µg/mL proteinase K, 10 mmol/L Tris-HCl (pH 8.0), 1 mmol/L ethylenediamine tetra-acetate, and 0.5% SDS] and was digested at 54 C with continuous agitation overnight. Following the methods of Mashal et. al. (8), the lysate was extracted once with phenol, once with phenol-chloroform (1:1), and once with chloroform. DNA was precipitated in a mixture of 0.1 vol 3 mol/L sodium acetate (pH 5.2) and 2.2 vol cold 100% ethanol and then pelleted by centrifugation. After washing with 70% ethanol, the DNA was resuspended in 30 µL 10 mmol/L Tris-HCl, pH 8.0. The DNA concentration was determined by spectrophotometry, and working DNA solutions of 50 ng/µL were made for each specimen.
Aliquots of DNA (400 ng) were incubated in 10 µL of digestion buffer, with or without 5 units of HpaII restriction enzyme (Promega, Madison, WI), for 3 h at 37 C. The DNA was then reextracted from both groups (with and without restriction enzyme) in the same manner described above. The precipitate was resuspended in 32 µL 10 mmol/L Tris-HCl, pH 8.0. To serve as controls, DNA from a known ovarian cancer (monoclonal) and from the same patient’s normal peripheral leukocytes (polyclonal) were subjected to an identical protocol.
PCR
The HpaII digested and undigested granulosa cell DNA
and control DNAs were amplified by PCR using primers chosen to amplify
a polymorphic short tandem repeat (STR) located within the androgen
receptor gene. The amplified product also included the two flanking
HpaII recognition sites (9). The primer pair used was:
5'-TGCGCGAAGTGATCCAGAACC-3' and 5'-TTGGGGAGAA CCATCCTCACC-3'. This
amplification yields a DNA product ranging in size from 173–233 bp.
Approximately 100 ng DNA (8 µL) was added to a 20-µL reaction
containing 1 µmol/L of each primer, 10% dimethyl sulfoxide, 1.5
mmol/L Mg Cl2, 60 mmol/L Tris-HCl (pH 10.0), 15 mmol/L
(NH4)2, one unit Taq polymerase, 4
µCi 
[32P]-dCTP, 0.125 mmol/L dCTP, and 1.25 mmol/L
of each of the following nucleotides: thymidine 5'-triphosphate, dATP,
and 7-deaza-2'-dGTP. The latter reagent has been shown to be superior
to dGTP to facilitate the symmetric amplification of pairs of alleles
containing G-C rich areas, such as the androgen receptor’s STR region
(10). PCR amplification, using the Gene Amp PCR System 9600 (Perkin
Elmer, Foster City, CA), were done under the following conditions: 95 C
for 2 min followed by 35 cycles of denaturing at 94 C for 30 sec,
annealing at 60 C for 30 sec, and extension at 72 C for 30 sec, and
then ending with 72 C for 7 min.
After PCR amplification, 1 vol gel loading buffer (95% formamide, 20 mmol/L ethylenediamine tetra-acetate, 0.05% bromophenol blue, and 0.05% xylene cyanol) was added. An aliquot (4 µL) was loaded on a 0.5-mm thin 6% denaturing polyacrylamide gel and electrophoresed at 72 watts for 2.5 hr. Gels were dried and exposed to x-ray film. Each allele is usually represented by a set of tightly clustered, evenly separated bands (stutter bands), which are frequently observed in PCR performed on DNA containing oligonucleotide repeats. Densitometric analysis of the autoradiographs was carried out using the NIH Image 1.54 (Bethesda, MD) analysis software to allow quantitation of the reduction in band intensity after HpaII digestion. The allelic intensity ratio was expressed as the ratio of the more intense allele divided by the less intense allele. The entire assay (consisting of HpaII digestion, PCR amplification, and gel electrophoresis) was carried out on two separate occasions for all specimens, to confirm results.
The sensitivity and specificity of the clonality assay was determined by mixing two control DNA samples (known monoclonal and polyclonal samples from the same individual) in various ratios, followed by digestion and amplification, as described above.
Interpretation
We used the criteria established by Mashal et al. (8)
in assigning granulosa cell populations as monoclonal or polyclonal.
Results for a given follicle were interpreted as monoclonal if the
HpaII-treatment caused the loss of one allele or caused the
ratio of one allele to the other to become more than 4.0. A polyclonal
pattern was defined as a ratio 
3.0 (7).
The proportion of monoclonal follicles was tabulated, and the standard error of proportion (SEP) was calculated by the formula: SEP = (pq/n)0.5, where P = the proportion found to be monoclonal, q = the proportion found to be polyclonal, and n = the number of informative follicles.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Analysis of DNA extracted from paired cancer and normal tissue
confirmed the validity of the clonality assay. As expected, normal
tissue displayed two alleles, even when predigested with
HpaII, whereas the same patient’s ovarian cancer tissue
exhibited a loss of one allele when pretreated with HpaII
(monoclonal pattern) (Fig. 2) (11).
|
7.5% nongranulosa
cells should not yield a spurious result. However, an admixture
containing 
10% polyclonal DNA consistently gave a
polyclonal appearance (allelic intensity ratio 3). Because the
polyclonal DNA consists of a 50:50 ratio of each inactivated allele, a
mixture containing 5% of the opposite inactivated allele should not be
misconstrued by our assay as being monoclonal. Clonality of granulosa cells
All of the 9 patients studied were heterozygous for the androgen receptor alleles, and all had at least 1 polyclonal follicle, thus proving that none of these patients had an inborn skewing of X-inactivation in their ovarian tissue that would have prohibited accurate interpretation of the assays. A total of 91 follicular aspirates were collected; 72 of which contained sufficient recoverable granulosa cells and thus underwent DNA extraction. DNA samples from 65 of the 72 specimens were amplified successfully by PCR. All amplification reactions gave results that were unambiguous; there were no follicles which yielded allelic intensity ratios between 3.0 and 4.0.
Seventeen samples (26%) were judged to be monoclonal, whereas 48 were
polyclonal (Fig. 3). The SEP
was 5.4%. The 95% confidence interval for the incidence of
monoclonality is 15–37%. Eight of the 9 patients had at least 1
follicle that was monoclonal. There was no correlation between the
chronological order of aspiration and the likelihood of the follicle
being monoclonal.
|
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The X-linked human androgen receptor gene (HUMARA) is highly
polymorphic, because of a variable-length CAG trinucleotide repeat in
exon 1. Adjacent to the hypervariable region, are two sites recognized
by the methylation-sensitive HpaII restriction endonuclease
(Fig. 1
). The inactive X chromosome is heavily methylated and thus is
protected from cleavage by HpaII. PCR primer pairs were
made, which flanked the hypervariable region and the HpaII
sites. Predigestion with HpaII cuts the active X allele,
destroying this as a template for PCR (13). Thus, only the allele from
the inactive X is amplified by PCR. A tissue is considered to be
monoclonal (i.e. all cells having the same X chromosome
inactivated) if only one of the two alleles is amplified by PCR after
HpaII digestion. If a cell population is derived from more
than a few progenitor cells (polyclonal), its pattern of X-inactivation
should be evenly distributed. However, if only a few precursor cells
are responsible for making up the cohort of cells in a sample
(oligoclonal), one would expect to see an unequal distribution of the
inactive X allele. If all of the progenitor cells happened, by chance,
to have the same X allele inactivated, then the clonality assay would
give a result indistinguishable from a truly monoclonal cell
population.
We found that 26 ± 5.4% (proportion ± SEP) of mature human ovarian follicles are made up of a cohort of granulosa cells, all containing the same inactivated X chromosome. This result is unexpected, because numerous studies involving a confined sampling of nonneoplastic tissue have consistently demonstrated a polyclonal origin (14, 15, 16). There are two possible explanations for this rather unique finding: 1) the granulosa cell population of one out of every four follicles is derived from a single progenitor cell, and thus, 25% of follicles have a true monoclonal granulosa cell population, whereas 75% of the follicles have a polyclonal population; or 2) the granulosa cell cohort of a given follicle is derived from a small number of precursor cells and thus is oligoclonal.
In considering the first possible explanation, one is left to ask why 25% of the mature follicles would be created differently from the rest. The more likely explanation is the second one.
Using the following probability equation, we estimated that the number of precursor cells that potentially contribute to the granulosa cell population is three: (1/2)n x 2 = Pm, where Pm = the proportion of follicles with a monoclonal pattern and n = the number of progenitor cells. When n = 3, then Pm is 0.25, which is very close to our observed incidence of 0.26 ± 0.05. If n = 4 or n = 2, then Pm is 0.125 and 0.5, respectively. Both of these Pm values are outside the 95% confidence interval of our results (0.15–0.37); and thus, we can exclude all integers other than 3 from being the true number of progenitor cells (at the time of X inactivation in the embryo) that ultimately give rise to each individual follicle’s granulosa cell cohort. Certain human tissues contain compartments that are derived from a patch of contiguous cells derived from a single precursor, such as the individual intestinal crypts (17). It is quite possible that patches of pre-granulosa cells, derived from a small number of stem cells, form the initial layer of granulosa cells in the primordial follicle, accounting for an oligoclonal pattern.
The oligoclonal interpretation is consistent with the results of a study on the clonality of murine granulosa cells (3, 18), where the population of precursor cells ultimately giving rise to the granulosa cell complement of a given follicle was determined to be oligoclonal (from only five ancestral cells). One experiment was based on a quantitative assay of the two variants of the X-linked enzyme, phosphoglycerate kinase-1, in heterozygous mice, with the enzyme activities being analyzed by binomial equations. The authors also analyzed the subcompartments of the follicle, the mural and cumulus cells, the results of which indicated that the cumulus and mural granulosa cells had a common origin. An additional study (3), using in situ hybridization techniques with chimeric mouse ovaries, confirmed this finding. Boland, Grosden, and colleagues observed that follicles are constructed by the radial proliferation of granulosa cell clones, which form long, thin columns across the follicle wall. The granulosa cells used in our study are most likely antral cells, given that the mural cells would be less likely to be aspirated from the follicle, because they are bound to the basement membrane. From the previously cited studies, it is likely that this selection of a subpopulation of cells for analysis would not alter our results or conclusions.
If the granulosa cells in a mature follicle are derived from only a few, instead of many different progenitor cells, then a particular follicle might potentially be disadvantaged if one of the progenitor cells contained a mutation that was passed on to all of its daughters. An example of this situation is McCune-Albright syndrome, in which affected individuals are believed to be somatic mosaics for an activating Gs protein mutation that constitutively activates the adenylate cyclase of these mutant cells (19). These individuals undergo precocious puberty as a consequence of the autonomous proliferation and production of estradiol by the mutant granulosa cells.
What factors could confound our results? The purity of the starting material is crucial. One could assume that all granulosa cell populations are truly monoclonal and that 75% of our specimens simply contained either cells from more than one follicle or a significant number of contaminating cells. However, we took great care to avoid cross-contamination between follicles. The first aspirate from each patient or each ovary was no more likely to give monoclonal results than subsequent aspirates. The cytologic assessment refutes the notion that there was significant contamination by other cell types. Our analysis indicated that contamination was uniformly less than 5% of the cells, a level of contamination that would not have led to a spurious assignment of a truly monoclonal follicular cell population as polyclonal.
One potential limitation of our study is that only mature ovarian follicles were assessed. It is possible that follicles that never reach maturity differ in the initial number of granulosa cell precursors from those follicles that do reach maturity. Another potential limitation of our study is that the source for all of the follicles was infertile women. However, the cause of the infertility in all cases was structural. Two of the nine women whose follicles were analyzed conceived during their in vitro fertilization procedure, making it unlikely that our findings are skewed by major differences between the competence of follicles from infertile women, as compared with normal fertile females.
In summary, our study suggests that the cohort of granulosa cells in a human preovulatory follicle is derived from a small number (three) of precursor cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 31, 1997.
Revised May 19, 1997.
Accepted May 29, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Hickey, A. Chandy, and R. J. Norman The Androgen Receptor CAG Repeat Polymorphism and X-Chromosome Inactivation in Australian Caucasian Women with Infertility Related to Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., January 1, 2002; 87(1): 161 - 165. [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 |
