| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, University of Maryland School of Medicine, Baltimore, Maryland 21201
Address all correspondence and requests for reprints to: Dr. Eli Y. Adashi, Departments of Obstetrics/Gynecology and Physiology, University of Maryland School of Medicine, 655 West Baltimore Street, BRB 11–010, Baltimore, Maryland, 21201. E-mail: eadashi{at}umabnet.ab.umd.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has been hypothesized that ovulation may constitute a cyclic
inflammatory-like process (5) and that gonadotropin-inducible
intraovarian interleukin-1 (IL-1; 6), an established mediator of
inflammation (7), may play a central role in this regard (8, 9, 10, 11). The
observation (12, 13, 14) that IL-1 is capable of stimulating ovarian PG
biosynthesis (PGE2>PGF2
), an established
mediator of the ovulatory cascade (15), lends additional support to the
notion that ovulation is an inflammatory-like process and that IL-1 may
participate in this process.
It was the objective of this communication to determine whether the ability of IL-1ß to stimulate PG biosynthesis involved and/or required the intervention of PLA2. Our findings establish IL-1ß as an up-regulatory agent for sPLA2 and cPLA2 gene expression, document the presence of a possible phosphatidylethanolamine (PE)-dependent PLA2 activity in cultured whole ovarian dispersates from immature rats, reveal the up-regulatory, receptor-mediated, and time-dependent action of IL-1ß in this regard, and suggest the existence of endogenous PLA2-stimulating/IL-1-like activity. These observations support the proposition that PLA2 is a key component in the IL-1-stimulated biosynthesis of ovarian prostaglandins.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-1-stearoyl-2-arachidonyl
[arachidonyl-5,6,8,9, 11,12,14,15-3H(N), PCSA, 88
Ci/mmo]; PGF2 [5,6,8,9,11,12,14,15-3H(N),
200 Ci/mmol]; PGE2
[5,6,8,9,11,12,14,15-3H(N), 154 Ci/mmol]; AA
[5,6,8,9,11,12,14,15-3H(N), 100 Ci/mmol] and
[-32P] UTP (800 Ci/mmol) were purchased from DuPont,
New England Nuclear (Boston, MA). The corresponding unlabeled
compounds, RNase A, indomethacin, phosphatidylinositol (PI) and
phosphatidylethanolamine (PE) were purchased from Sigma Chemical Co.
(St. Louis, MO). McCoy’s 5a (serum-free) medium,
penicillin-streptomycin solution, BSA, DNase, Moloney Murine Leukemia
Virus reverse transcriptase, and trypan blue stain were obtained from
Life Technologies, Inc. (Grand Island, NY). Collagenase (Clostridium
Histolyticum; CLS type I; 144 U/mg) was from Worthington Biochemical
Corp. (Freehold, NJ). Hexane, methanol, and 2-propanol were from J. T.
Baker, Inc. (Phillipsburg, NJ). Chloroform was from Fisher Scientific
(Fair Lawn, NJ). All other chemicals were purchased from Sigma Chemical
Co. (St. Louis, MO). RNase T1 was from Pharmacia (Piscataway, NJ).
Ecl136II was from New England Biolabs, Inc. (Beverly, MA). T7 RNA
polymerase, pGEM7Zf+ and other molecular grade reagents were from
Promega (Madison, WI). Recombinant human IL-1ß (2 x 107 U/mg) was generously provided by Drs. Errol B. de Souza and C. E. Newton of DuPont Merck Pharmaceutical Co. (Wilmington, DE). A recombinantly expressed preparation of the naturally occurring human IL-1 receptor antagonist (IL-1RA) was generously provided by Dr. Daniel E. Tracey (The Upjohn Co., Kalamazoo, MI).
Animals and cell culture
Immature (25 to 28 day old) Sprague-Dawley female rats,
purchased from Zivic-Miller Laboratories (Zelienople, PA), were killed
by CO2 asphyxiation.
Whole ovarian dispersates were prepared by collagenase digestion as previously described (16) and were cultured in serum-free McCoy’s 5a medium supplemented with L-glutamine (2 mM), penicillin (100 U/ml), and streptomycin sulfate (100 µg/ml). Cells were cultured in either 12 x 75 mm tubes (1 ml media) or in dishes (1–3 ml media) as noted and were maintained for up to 48 h at 37 C under a water-saturated atmosphere of 95% air and 5% CO2.
Arachidonic acid release assay
Whole ovarian dispersates (5 x 105 viable
cells/dish) were initially pulse-labeled with [3H]AA
(0.2–0.4 µCi/ml) for 24 h to allow for the incorporation of the
labeled free fatty acid into esterified lipids. At the end of the
incubation period, unincorporated label was removed by washing the
cells three times with fresh McCoy’s 5a medium. The cells were then
reincubated for the duration indicated in the absence or presence of
specified treatments. At the conclusion of this treatment period,
conditioned media were collected, acidified, and extracted (twice) with
3 ml ethyl acetate. The latter solvent was then evaporated in a
Speed-Vac centrifuge (Savant Instruments, Inc., Farmingdale, NY), and
the residue resuspended in 250 µl hexane/isopropanol/acetic acid,
95/5/0.025 (vol/vol/vol). The resultant products were then fractionated
by HPLC using a Waters HPLC system (Milford, MA) on a normal phase diol
column (LiChrosorb Diol, 10 µm; EM Reagents, Gibbstown, NJ) with a
concave gradient of hexane/isopropanol (95:5 to 60:40) at 2 ml/min. The
column was calibrated with authentic [3H]AA,
[3H]PGE2 and
[3H]PGF2. Radiolabeled AA and major
eicosanoid products (PGE2 and PGF2)
were detected and quantified by on-line scintillation counting with a
Radiomatic Flow Detector (Packard Instrument Co., Meriden, CT).
TLC of [3H]AA-labeled cellular phospholipids
[3H]AA-labeled cells were sonicated on ice (x 2)
for 5s (Vibra Cell, Sonics and Materials Inc., Danbury, CT).
Thereafter, total cellular phospholipids were extracted with 3 ml
chloroform/methanol, 66:33 (vol/vol). The bottom chloroform layer
(containing phospholipids) was dried, and the residue resuspended in 10
µl of the same solvent. The latter phospholipid mixture was then
separated by one- or two-dimensional TLC (Eastman Kodak, Rochester, NY)
using a previously-described (17) solvent system, consisting of
chloroform/methanol/acetic acid/water (90/8/1/0.8; vol/vol/vol/vol) in
the first dimension, and a chloroform/methanol/7 M ammonium
hydroxide (15/6/1; vol/vol/vol/) solvent system in the second
dimension. The radioactivity in 1 cm-long segments along the TLC strips
was determined using a scintillation counter. Labeled prostanoids and
PCSA served as authentic controls. Unlabeled PE and PI also served as
authentic controls and were visualized with iodine vapor.
Cell-free PLA2 enzyme assay
Total cellular PLA2 activity was determined by
measuring the release of [3H]AA from the sn-2
position of [3H]PCSA substrate. Whole ovarian dispersates
(5 x 105 viable cells/tube) were initially cultured
as described for 48 h in the absence or presence of the specified
treatments. Media were then removed, the cells washed once with 100
mM HEPES, pH 8 buffer, and suspended in 1 ml of the same.
Cells were then sonicated on ice (x 2) for 5s as described above. A
sonicate volume representing 1.5 x 105 cells was then
incubated for 1 h at 37 C in a total assay volume of 1 ml. Assay
buffer also contained 5% methanol (to dissolve substrate) and 2
mM CaCl2. Total PCSA substrate concentration
was 5 µM inclusive of 25 x 103 cpm of
labeled substrate. Preliminary experiments disclosed that neither
deoxycholic acid nor presonication of the substrate increased the
reaction efficiency, and thus were not used subsequently. The enzymatic
reaction was terminated with 7% formic acid (final pH 3.5), and the
resultant products were extracted and detected by HPLC as described
above. The ratio of labeled product to labeled substrate was multiplied
by the initial substrate concentration (5 µM) to
calculate the rate of enzymatic conversion to product
(pmol/105 cells·h). Values were also corrected for
substrate availability in the reaction mixture, and for product
recovery.
Nucleic acid probes
A plasmid containing a 750-bp SmaI/EcoRI
insert of the complementary DNA (cDNA) encoding rat type II
sPLA2 (18) was kindly provided by Dr. J. Ishizaki (Shionogi
Research Laboratories, Osaka, Japan). For RNase protection assays, a
BamHI fragment was excised and subcloned into pGEM7Zf+. This
latter construct was then linearized with Ecl136II and transcribed with
T7 RNA polymerase in the presence of [-32P] UTP to
yield a 535-nucleotide (nt) antisense riboprobe which, upon
hybridization, was projected to generate a 452-nt protected
fragment.
A full-length cDNA of the rat 85-kDa cPLA2 (19) was kindly provided by Dr. Yuji Owada (Tohoku University, Sendai, Japan) in a transcribable vector (pBluescript II SK+). Because the transcribed region contained a poly(A) tail and several A-rich regions, the cDNA was modified to ensure the generation of a high specific activity riboprobe. Specifically, use was made of a BamHI to remove some of the 3' end of the cDNA. After self ligation, the product was linearized with PvuII and transcribed with T7 RNA polymerase to yield a 328nt antisense riboprobe which upon hybridization was projected to generate a 253-nt protected fragment corresponding to the translated region of the cDNA. The rat large ribosomal protein 19 (RPL19) normalizing probe was generated as previously described (20).
RNA extraction
Total RNA from cultured cells (1.5 x 106
cells/dish) was extracted with RNAzol B (Tel Test, Friendswood, TX)
according to the manufacturer’s protocol.
RNAse protection assay
Linearized DNA templates were transcribed with T7 RNA polymerase
to specific activities of 800 Ci/mmol [-32P]UTP
(cPLA2 and sPLA2) or 160 Ci/mmol
[-32P]UTP (RPL19). The riboprobes were gel-purified in
an effort to eliminate transcribed products shorter than the full
length probes. The assay was performed as previously described (21).
Gels were exposed to XAR film (Kodak, Rochester, NY) for varying
lengths of time with intensifying screens. To generate quantitative
data, gels were also exposed to a phosphor screen (Molecular Dynamics,
Sunnyvale, CA). The resultant digitized data were analyzed with Image
Quant Software (Molecular Dynamics, Sunnyvale, CA). The hormonally
independent RPL19 messenger RNA (mRNA) signal was used to normalize the
sPLA2 and cPLA2 mRNA data for possible
variation in RNA loading. Specifically, the net signal of the
sPLA2 or cPLA2 protected band (respective
background subtracted)/net RPL19 signal ratio was calculated for each
sample.
Statistical analysis
Except as noted, each experiment was replicated a minimum of
three times. Data points are presented as mean ± SEM.
Statistical significance (Fisher’s protected least significant
difference) was determined by ANOVA and Student’s t test.
Statistical values were calculated using Statview 512+ for MacIntosh
(Brain Power, Inc., Calabasas, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The first radioactive peak (Rf = 0.2;
top panel) comigrated with authentic PCSA (bottom
panel). Authentic standards for PE and PI also comigrated in the
same position (not shown). The second radioactive peak (Rf = 0.9;
top panel) did not comigrate with any of the other authentic
standards tested (AA, PGE2 or PGF2). This
unknown compound may reflect a metabolite of AA (personal
communication, Frank Hertelendy, University of St. Louis, St. Louis,
MO). Importantly, pulse-labeled cells contained little, if any, free
AA. The recovery ratio of labeled cellular phospholipids from AA
(i.e. counts detected at Rf = 0.2/total counts added)
was determined at the end of the incorporation period. The calculated
incorporation rate of [3H]AA into phospholipids was only
20%, significantly lower than the above mentioned 85%, and did not
increase when BSA (0.5 mg/ml) was added to the labeling medium.
|
), the incorporation of [3H]AA into
phospholipids (Rf = 0.2) tended to plateau following 20 h of
culture. These findings suggest the adequacy of 24 h of cellular
labeling with [3H]AA.
|
), treatment with IL-1ß produced progressive,
time-dependent increments in the accumulation of radiolabeled products
as compared with untreated control, reaching statistical significance
at the 48 h time point (P < 0.05). Further
analysis, using HPLC separation of products (Fig. 4),
demonstrated that the IL-1ß-induced increase in total radioactivity
in conditioned media reflected time-dependent increments in
[3H]AA, and its downstream metabolites,
[3H]labeled PGE2 and PGF2.
Replicate experiments confirmed that most of the [3H]AA
released from cells was further metabolized to PGE2 and
PGF2. These observations suggest that IL-1ß increases
rat ovarian PLA2 activity, as well as prostaglandin
endoperoxide synthase activity. However, because all medium
radioactivity is ultimately derived from the processing of
[3H]AA-labeled cellular phospholipids, the radioactivity
released as prostanoid metabolites may be operationally referred to as
AA released and as such may be used as a reflection of ovarian
PLA2 activity.
|
|
), treatment with IL-1ß produced a mean 3-fold
increase in [3H]AA release (P = 0.14),
which was completely blocked by the concurrent addition of IL-1RA
(P = 0.12). Moreover, treatment with IL-1RA by itself
produced a small, albeit significant (P < 0.05)
decrease in basal [3H]AA release. These findings suggest
that the ability of IL-1ß to enhance [3H]AA release by
cultured whole ovarian dispersates is IL-1 receptor-mediated. These
findings also point to the existence of endogenous
PLA2-stimulating/IL-1-like activity.
|
), treatment with IL-1ß resulted in a significant
(P < 0.05) decrease (55%) in the cellular content of
[3H]labeled phospholipids (Rf = 0.2, cf. Fig. 1
).
Treatment with IL-1ß did not eliminate the second peak (Rf =
0.9, data not shown). These data confirm the finding (Figs. 3 and 4)
that IL-1ß stimulates AA release from ovarian cells.
|
), cellular products were separated by TLC as shown in
(Fig. 7). [3H]AA was incorporated into two
compounds that comigrated with authentic PC and PE. Although PC was the
major labeled phospholipid noted, treatment with IL-1ß depleted
mainly the pool which comigrated with PE (1.8-fold decrease;
P < 0.05). The PC pool remained unaffected.
|
), treatment with IL-1ß
produced a 2.6 fold increase (P < 0.05) in total
(cytosolic and secretory) cellular ovarian PLA2 activity as
compared with untreated controls. Concurrent provision of IL-1RA,
produced a significant (P < 0.01) decrease in basal
PLA2 activity and all but abolished the IL-1ß effect
(P < 0.05). These findings also (cf. Fig. 5) suggest
the existence of endogenous PLA2-stimulating/IL-1-like
activity because control levels are further reduced by the addition of
IL-1RA.
|
.
Increasing concentrations of IL-1ß produced only a modest (up to
1.8-fold), albeit significant (P < 0.05), increase in
sPLA2 transcripts. A narrow response range was noted. Of
note was the robust expression of sPLA2 in untreated,
control cells. We have shown that endogenously produced IL-1ß induces
this basal level of expression in control cells (Ben-Shlomo, I., S.
Kol, and E. Adashi, submitted for publication). Taken together, these
observations suggest that sPLA2 gene expression in cultured
ovarian cells is exquisitely sensitive to IL-1ß as reflected by the
ability of endogenously produced IL-1ß to drive sPLA2
expression to near maximal levels.
|
.
Increasing concentrations of IL-1ß produced significant
(P < 0.05) increases (up to 5-fold) in
cPLA2 transcripts. Of note was the dramatic effect of
increasing the IL-1ß concentration from 0.5 to 5 ng/ml. A further
increase in the IL-1ß concentration resulted in only a marginal
effect on cPLA2 gene expression. Taken together, these
observations suggest that cPLA2 gene expression is induced
over a relatively narrow range of IL-1ß doses.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 4), loss of cellular phospholipids (Figs. 6 and 7) and
increased cellular PLA2 activity (Fig. 8) that is
receptor-mediated (Figs. 5 and 8). The IL-1ß-stimulated
PLA2 activity is most likely dependent on PE as substrate
since this compound, rather than PC, was depleted by IL-1ß (Fig. 7).
However, given that the TLC system used may not resolve all
phospholipids, this identity remains to be established. We also noted
that PC was the major labeled phospholipid (Fig. 7). The differential
incorporation of AA into phospholipids (PC>PE) has also been reported
for mast cells (23). Preferential depletion of the PE pool upon
stimulation was also observed. Thus, it would appear that, while PC may
constitute the long-term, more stable pool of arachidonate substrate,
PE serves as the more available source for the biosynthesis of
inflammatory mediators.
Because IL-1ß up-regulates the gene expression of sPLA2
and cPLA2 (Figs. 9 and 10), both species may contribute to
the observed increase in PLA2 activity. The ability of
IL-1ß to stimulate PLA2 activity, not hitherto reported
in ovarian cells, was previously described for a number of extraovarian
cell types (e.g. fibroblasts and mesangial cells; 24–32).
The present observations (Fig. 4) also concur with our previous data
(12), suggesting that IL-1ß induces prostaglandin endoperoxide
synthase activity, an issue beyond the scope of this communication.
The role of PLA2 in the inflammatory processes is well established (33). Increased PLA2 activity has been noted in a host of pathological processes (e.g. septic shock, connective tissue inflammatory diseases, premature labor, hypertension, and pancreatitis). However, this family of enzymes also plays many other physiological roles, including the facilitation of fat digestion, lipoprotein metabolism, lung surfactant metabolism, cell membrane homeostasis, insulin release, and sperm maturation and penetration (2, 3). Our data suggest that PLA2 may play a role in ovarian physiology in general and in ovulation in particular (as an intermediary of IL-1ß). In this connection, in situ hybridization and RNase protection studies identified sPLA2 and cPLA2 transcripts in periovulatory ovarian granulosa cells (Ben-Shlomo, I., S. Kol, and E. Adashi, submitted for publication).
Although the role of PLA2 in ovulation and/or luteinization has not been fully elucidated, an increase in ovarian PLA2 activity in response to the LH surge (34) was recently reported. Because preovulatory ovarian IL-1ß gene expression is probably gonadotropin dependent (6, 35), we speculate that the LH-induced stimulation of PLA2 activity is IL-1ß mediated. This robust enzyme may participate in the rapid cellular and tissue remodeling that characterizes the ovary during the periovulatory period. Moreover, by promoting PG biosynthesis, PLA2 may act in concert with other factors [e.g. plasminogen activator, collagenase (36, 37)] to promote follicular rupture and the release of the cumulus/oocyte complex (38, 39). Reported alterations in PLA2 activity during luteal regression in pseudopregnant and pregnant rats (40) support the notion that PLA2 may also be involved in corpus luteum formation and regression.
Apart from documenting the ability of exogenously provided IL-1ß to
stimulate ovarian PLA2 activity, our present findings also
suggest the existence of endogenous
PLA2-stimulating/IL-1-like activity. Indeed, the ability of
IL-1RA to inhibit basal PLA2 activity (Figs. 5 and 8)
underscores the possibility that cultured whole ovarian dispersates
release IL-1-like bioactivity. Although the presence of IL-1ß (or
IL-1) in media conditioned by whole ovarian dispersates is yet to be
demonstrated, prior studies have clearly established whole ovarian
dispersates as a site of IL-1ß gene expression (6). Preliminary
observations (Kol, S., M. Ando, and E. Adashi, unpublished data)
further suggest that the ovary is also the site of modest IL-1 gene
expression. Taken together, these observations support the concept that
both exogenous and endogenous IL-1 exert an up-regulatory effect on
ovarian PLA2 activity.
At the level of gene expression, our observations suggest that
sPLA2 transcripts are substantially stimulated by what are
probably low levels of endogenously elaborated IL-1ß. This conclusion
is supported by the finding that a further increase in IL-1ß
concentrations (by adding exogenous IL-1ß) results in only a modest
further increase in sPLA2 expression (Fig. 9). Presumably
then, sPLA2 is highly sensitive to stimulation by even low
concentrations of IL-1ß. In contrast, the overall increase in
cPLA2 transcripts following treatment with IL-1ß was far
more substantial compared with sPLA2 (5 vs.
1.8-fold). It seems likely that cPLA2 is less sensitive to
stimulation by IL-1ß because higher, exogenously provided doses of
IL-1ß are required to effect a meaningful increase (Fig. 10). Perhaps
the bulk of steady-state cellular PLA2 activity in
untreated cells is attributable to sPLA2.
In summary, our current observations document the presence of possible PE-dependent PLA2 activity in cultured whole ovarian dispersates from immature rats, reveal the up-regulatory, receptor-mediated, and time-dependent action of IL-1ß in this regard, support the existence of endogenous PLA2-stimulating/IL-1-like activity, and establish IL-1ß as an up-regulatory agent for sPLA2 and cPLA2 gene expression. As such, these observations support the proposition that PLA2 is a key component in the IL-1ß-stimulated biosynthesis of ovarian prostaglandins.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Recipient of a 1994 Merck Senior Fellow Award of The Endocrine
Society and an American Fellowship Award. Current address: Department
of Obstetrics and Gynecology, Rambam Medical Center, Haifa, Israel.
3 Recipient of a Lalor Foundation Fellowship, an International
Fellowship Award from the Israeli Medical Association, a Fullbright
Fellowship, and a Harlea Charitable Trust Award. Current address:
Department of Obstetrics and Gynecology, Haemek Medical Center, Afula,
Israel.
Received July 1, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
stimulates prostaglandin
biosynthesis in serum-activated mesangial cells by induction of a
non-pancreatic (type II) phospholipase A2. J Biol Chem 266:14119–14127This article has been cited by other articles:
 |
|
 |
|
  D. M. Duffy, C. L. Seachord, and B. L. Dozier An Ovulatory Gonadotropin Stimulus Increases Cytosolic Phospholipase A2 Expression and Activity in Granulosa Cells of Primate Periovulatory Follicles J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5858 - 5865. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Saito, M. Ando, D. Sussman, H. Negishi, G. King, and E. Y. Adashi Interleukin 1 Upregulates Ovarian Prostaglandin Endoperoxide Synthase-2 Expression: Evidence for Prostaglandin-Dependent/Ceramide-Independent Transcriptional Stimulation and for Message Stabilization Biol Reprod, December 1, 2001; 65(6): 1759 - 1765. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. W. Chung, M. Ando, and E. Y. Adashi Periovulatory and Interleukin (IL)-1--Dependent Reglation of IL-6 in the Immature Rat Ovary: A Specific IL-1 Receptor-Mediated Eicosanoid-Dependent Effect Reproductive Sciences, September 1, 2000; 7(5): 301 - 308. [Abstract] [PDF]
|
|
|
|
 |
|
 J. V. BONVENTRE The 85-kD Cytosolic Phospholipase A2 Knockout Mouse: A NewTool for Physiology and Cell Biology J. Am. Soc. Nephrol., February 1, 1999; 10(2): 404 - 412. [Full Text]
|
|
|
|
|
|
M. D. Rosenthal, E. D. Albrecht, and G. J. Pepe Developmental Maturation of Baboon Placental Trophoblast: Expression of Messenger Ribonucleic Acid and Protein Levels of Cytosolic and Secretory Phospholipases A2 J. Clin. Endocrinol. Metab., August 1, 1998; 83(8): 2861 - 2867. [Abstract] [Full Text]
|
|
|
|
 |
|
 M. Ando, S. Kol, E. Kokia, K. Ruutiainen-Altman, J. Sirois, R. M. Rohan, D. W. Payne, and E. Y. Adashi Rat Ovarian Prostaglandin Endoperoxide Synthase-1 and -2: Periovulatory Expression of Granulosa Cell-Based Interleukin-1-Dependent Enzymes Endocrinology, May 1, 1998; 139(5): 2501 - 2508. [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 |
