| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Graduate School of Biological Sciences, Nara Institute of Science and Technology, Ikoma, Nara 630–01, Japan
Address all correspondence and requests for reprints to: Dr. Tatsuo Takeya, Graduate School of Biological Sciences, Nara Institute of Science and Technology, Ikoma, Nara 630–01, Japan. E-mail: ttakeya{at}bs.aist-nara ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At present, 12 different genes encoding the respective members of the connexin family have been identified and reported in rodents (4, 5). They consist of both highly conserved and divergent regions; the amino-terminal domains and four membrane-spanning regions are highly conserved, whereas the length and the primary structure of the two intracellular domains, named cytoplasmic and carboxyl-terminal domains, respectively, vary greatly in individual connexins. The molecular diversity seen in each connexin may reflect the functional and regulatory complexities of GJC described above, although details of the structure-function relationships in those two domains are mostly unknown.
It is known that some connexins are expressed rather ubiquitously, whereas other connexins are expressed in limited tissues or cell types, and occasionally, multiple connexins are expressed in a single cell (4). The complexity of such expression profiles makes it difficult to determine the physiological roles of connexins. To establish the role(s) of an individual connexin and how connexins share roles with each other, one feasible approach might be to take a tissue or an organ consisting of several cell types as a source and examine the connexin genes expressed there. We recently identified the expression of multiple connexin genes (Cx-43, -32, -30.3, and -26) in the porcine ovarian follicle (14). These genes showed respective expression profiles in terms of follicular compartments as well as in the estrous cycle. In the present study, we identified a new member of the connexin family expressed in the ovarian follicles and characterized the gene structure as well as its expression profile.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1–7 mm in diameter),
granulosa cells, and cumulus-oocyte complexes were prepared as
previously described (14). Female Wistar rats (4–10 weeks old) were
hypophysectomized and shipped a week later by Japan SLC (Shizuoka,
Japan). The rats were injected ip with 10 IU PMSG (Sigma Chemical Co.,
St. Louis, MO), followed by an ip injection of 10 IU hCG (U.S.
Biochemical Corp., Cleveland, OH) 48 h after the first injection,
when the additional effect of hCG was examined. In control experiments,
PBS was injected instead of the respective hormones. The rats were
killed 8 h after the injection of hCG, and the respective ovaries
were removed from each rat and kept frozen until sectioning.
RNA extraction, RT, and PCR cloning
Total RNA was extracted from tissues by the acid
guanidinium-phenol chloroform procedure (16), and possibly contaminated
genomic DNA was removed by digestion with deoxyribonuclease (DNase)
free of ribonuclease (RNase; RQ1 DNase; Promega, Madison, WI). Ten
micrograms of total RNA were reverse transcribed into complementary DNA
(cDNA) by Superscript II (Life Technologies, Grand Island, NY) with
oligo-(deoxythymidine)12–18 primer for RT-PCR according to
the supplier’s instructions. The conditions for PCR cloning with a
pair of degenerated oligonucleotides were described previously (Fig. 1A
) (14). The products were finally
subcloned in pT7Blue-T cloning vector (Novagen, Madison, WI) and
subjected to DNA sequence analysis using an ABI3700 or ABI310 system
(Perkin Elmer, Foster City, CA). To extend the cDNA clones thus
obtained toward both the 5'- and 3'-ends, we prepared two kinds of
gene-specific primers (GSP-1 and -2 in Fig. 1B) for a rapid
amplification of cDNA ends (RACE) reaction (Marathon cDNA Amplification
Kit, Clontech Laboratory, Palo Alto, CA).
|
) was subcloned into RNA transcription
vector pBluescript II SK+, and a complementary RNA (cRNA)
probe on the antisense was prepared by in vitro
transcription on the linearized constructs using SP6 RNA polymerase in
the presence of digoxigenin-UTP. After the RNA polymerase reactions,
DNA templates were hydrolyzed with RQ1 DNase. cRNA probes for other
connexin genes were prepared based on the respective constructs, as
described in the previous study (14).
|
Cx-60 messenger RNA (mRNA) levels in tissues; Northern blotting and
RT-PCR
Northern blotting with total RNA samples (20 µg) from the
respective tissues was carried out as described previously (14, 17).
The probes used for filter hybridizations were derived from the
respective cDNA clones obtained in this and a previous study (14), and
they were labeled with [32P]deoxy-CTP.
The expression of Cx-60 was examined by RT-PCR with total RNA samples (10 µg) from the respective tissues under the conditions described in the previous report, in which a linear amplification was achieved (14); briefly, 35 cycles of 94 C were performed for 30 sec, 60 C for 1 min, and 72 C for 2 min, followed by a final extension for 5 min at 72 C. For quantitative analysis, a pair of primers specific to Cx-60 (GCACTTTATAGACTCAGGGCCTTTG and AGCAGACATCCTTTCAGAGGGAC) was prepared to produce a 176-bp fragment, and the products were separated by gel electrophoresis using 3.5% NuSieve GTG agarose (FMC, Rockland, ME); the intensities of the products were measured by an ATTO Densitometric Analyzer (ATTO, Tokyo, Japan). A portion of the nucleotide sequences of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene (18) was chosen to construct primers for PCR: 1) AGGTGAAGGCGGAGTCAAC and TACTCCTTGGAGGCCATGTG, and 2) AACAGCCTCAAGATCATCAGC and ACT-CATGGCATGGACTGTGG, which are expected to produce 1-kbp and 123-bp fragments, respectively.
Screening of porcine genomic library
A porcine genomic library (Clontech, PL1010j) was screened using
standard techniques (16). A Cx-60 cDNA fragment used for in
situ hybridization was labeled by
[
-32P]deoxy-CTP, using random hexamers and Klenow
fragment (Takara, Kyoto, Japan), and used as the Cx-60 probe. Purified
genomic clones were directly used for either PCR amplification or DNA
sequencing analysis (Dye Terminator Cycle Sequencing Kit, Perkin
Elmer). For the PCR reactions, following two sets of primers were
prepared: 1) CAGACTGGGAAATCCTTCATAG (from -140 to -119) and
GCTGTGAGCTGTGGTGTAGG (from +1835 to +1855), and 2) TCACAGCTTAGAGCCCAG
(from +333 to +351) and GGCGTAGAACATCAATGGTAAG (from +1423 to +1445),
respectively (the number 1 indicates the first nucleotide of the coding
region).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) (1) , revealed a clone containing a sequence that was not
identical to those of any known connexin genes, yet retained the
characteristic sequences of the family (Fig. 2 and Table 1).
|
-32P]deoxy-CTP as a probe, but
neither library produced positive clones, suggesting the low expression
level of the corresponding mRNA. Therefore, we took another approach by
extending the cDNA clone to both the 5'- and 3'-ends separately by the
RACE system; the nucleotide sequences GCACTTATAGACTCAGGGCCTTTG (GSP-1)
and AGCAGACATCCTTTCAGAGGGAC (GSP-2) within the amplified fragment
(Figs. 1B and 2) were used as gene-specific primers for 5'- and 3'-end
amplification, respectively. Products by both reactions were found to
contain the common region flanked by GSP-1 and -2 as expected, and a
single uninterrupted open reading frame (ORF) was deduced from the
combined nucleotide sequences (Fig. 2), indicating that this approach
was successful.
Screening and characterization of genomic clone
To confirm the nucleotide sequence deduced from the cDNA clones
obtained by the RACE system, an analysis of genomic clones was
performed. We screened the porcine genomic library (Clontech, Palo
Alto, CA) and isolated three independent clones. First, we carried out
PCR amplification with primers corresponding to the sequences either
within or outside the ORF using the genomic clones (Fig. 3). The amplified products gave their
respective sizes as expected from the cDNA analysis, implying that the
deduced ORF could exist on the porcine genome, and also that this new
gene was an intron-less gene, as are the other connexin genes (2, 4).
This idea and the deduced nucleotide sequence from the cDNA clones were
eventually examined by sequencing the corresponding regions on the
genomic clones.
|
); hence,
we named this gene Cx-60. The presence of the four putative
membrane-spanning regions unique to the connexin family divide the
connexin proteins into the nine regions shown schematically in Fig. 1.
The primary sequences of Cx-60 in the respective regions are compared
with those of other rat connexin genes (19, 20, 21, 22, 23, 24) and are summarized in
Table 1. As the Cy and C regions are completely diverged among the
family, the corresponding region is not included. In total, 53 residues
common to all of the rat connexin genes can also be found in Cx-60
(Fig. 2); among them, cysteine residues in EC1 and EC2 (at 54, 61, 65,
189, 194, and 200) that show the conserved sequences
CX6CX3C and CX4CX5C,
respectively, were noted and have been suggested to be involved in
intramolecular disulfide bond formation to link two extracellular loops
(4).
The connexin family genes can be classified into two groups based on
their similarities in primary structures and thus are prefixed with
either or ß; to date, seven genes in group and five genes in
group ß have been identified and characterized in the rodent system
(1, 2, 5). This classification coincides fairly well with the molecular
masses; the ß group includes the connexin proteins with molecular
masses less than 33 kDa, suggesting that newly identified Cx-60 could
be classified as a member of group . Actually, Cx-60 showed more
similarities with connexins larger than Cx-37 in the TM domains, in
particular TM2 and TM3 (Table 1). In addition, group-specific residues,
either or ß, can be pointed out in the primary structures of the
connexin family through the species; they are at Met2
(ß), Gly13 (ß), Val38 (ß),
Val79 (), Leu82 (ß), Tyr92
()/Val92 (ß), Ala97 (ß),
TrpTrp154–155 (ß), Gln/Arg191 (), and
SerLeu224–225 (). In this respect, Cx-60 shows the
characteristic features of group at the corresponding residues.
Expression of Cx-60: in situ hybridization in the porcine ovary
As spatial- and stage-specific expressions of four connexin genes
(Cx-26, -30.3, -32, and -43) have been revealed in porcine ovarian
follicles (14), the expression profile of the newly isolated Cx-60 gene
was studied first in porcine ovarian follicles and compared with those
of other genes.
First, the specificity of the Cx-60 probe was examined against the four
connexin genes, showing no cross-reactivity with the corresponding
regions of other genes (Fig. 4A). By
using this probe, expression of the Cx-60 gene and the size of its
transcript were examined with total RNA from follicles and whole
ovaries by Northern blotting, but no significant signals were obtained.
As the same filter revealed the expression profile of the Cx-43 gene as
that shown in the previous study (14), the low copy numbers of Cx-60
transcript in the respective tissues were evident, and the expression
of Cx-60 mRNA was able to be confirmed only by RT-PCR (Fig. 4B) and
in situ hybridization (described below).
|
, a, b, and d),
where the expression of Cx-26 or -30.3 was detected in the previous
experiments (14). RNase-treatment of the sections before hybridization
failed to give any signals (Fig. 4C, c), indicating that the
hybridization reactions were actually between the probe and the
intrinsic Cx-60 transcripts.
We then examined Cx-60 expression in cumulus cells. As no follicle
sections with cumulus-oocyte complexes (COC) were recovered or detected
in those preparations due to the fixation procedures with antral
follicles, we recovered COC separately from the total contents of
antral follicles and probed them with the Cx-60 probe. Although the
Cx-60 probe did not give positive signals in the granulosa cell
compartment, in contrast to the thecal compartment, as described above,
the same probe gave positive signal in the cumulus cells in five of
seven samples to an extent similar to that in the thecal cells (Fig. 4C, e). In the previous study in which we showed the expression of
Cx-43 and Cx-30.3 in cumulus cells, 10–20% of the COC remained
unstained with the probes as well, whereas neither the Cx-26 nor the
Cx-32 probe gave signals greater than the background level (14). Taken
together, these results indicate that Cx-60 was expressed in cumulus
cells.
Effect of gonadotropin administration on Cx-60 mRNA level in the
rat ovary
To further investigate the above findings obtained using porcine
ovaries and to examine the effect of gonadotropin (PMSG or/and hCG)
stimulation on the expression of Cx-60 mRNA, we carried out the
following experiments using 5- and 11-week-old hypophysectomized rats;
the following results were obtained in both age groups. First, the
expression profiles of five connexin genes, Cx-60, Cx-43, Cx-32,
Cx-30.3, and Cx-26, were essentially identical between porcine and rat
ovaries; those of Cx-60, Cx-43, and Cx-30.3 are shown in Fig. 5. Cx-60 mRNA was detected in the
surrounding thecal cell compartment, where Cx-26 mRNA was also detected
(as in the porcine system), although the compartment seemed to consist
of fewer cell layers than that in the porcine ovary (Fig. 5a). This
presumably made Cx-43 mRNA almost undetectable in the rat thecal cell
compartment unlike in the porcine ovary (Fig. 5c). We detected Cx-60
mRNA expression in cumulus cells in the porcine ovary using recovered
COC for the reason described above (14). However, unlike in the porcine
system, it was very difficult to recover COC in a large quantity from
the rat ovary. Therefore, we have little information on Cx-60 mRNA in
rat cumulus cells.
|
), suggesting that Cx-60 is expressed
constitutively in the rat ovarian follicle under the conditions
observed. This profile was quite different from that of Cx-43 (Fig. 6),
whose expression level was negligible without gonadotropin stimulation,
whereas the level was significantly enhanced by PMSG stimulation and
was down-regulated by additional hCG administration (25, 26, 27).
|
. Despite its overall limited
expression, Cx-60 mRNA levels in restricted tissues such as the colon,
thymus, and spleen (in addition to the ovaries) were noticeable; in
contrast, Cx-60 mRNA levels in the lens and liver were extremely
low.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that Cx-60 was expressed at very low levels in various organs and tissues. However, the possibility that the expression of Cx-60 is greater in other organs and tissues or at specific stages in development cannot be excluded. Although a low expression of Cx-60 mRNA does not necessarily indicate its less physiological significance, the low Cx-60 mRNA content made it difficult to quantitate its expression levels without combining appropriate techniques such as RT-PCR. An analysis at the protein level making use of an anti-Cx-60 antibody might be useful for quantitating the expression level of Cx-60 mRNA.
Despite this difficulty, Cx-60 was observed to have a unique expression profile by in situ hybridization in porcine ovarian follicles compared with those of the previously identified four connexin genes; namely, Cx-60 mRNA was identified in both thecal cells and cumulus cells, and Cx-60 seemed to be expressed constitutively upon gonadotropin stimulation. We previously reported that Cx-26, Cx-30.3, and Cx-32 are ubiquitously expressed in the theca interna, and that Cx-30.3 and Cx-43 are expressed in cumulus cells. The newly identified expression of Cx-60 in these two compartments thus confirms that multiple genes, including Cx-60, are expressed in each compartment, raising essential questions concerning the function of gap junctions formed by the respective connexins and their behaviors, e.g. whether all gap junctions are functional or, rather, cell-cell coupling is via one specific connexin, and whether these connexins can form heterologous channels. To examine these questions, experiments using cell systems expressing the respective connexin genes are required. Such experiments are underway in our laboratory.
In the porcine ovary, Cx-60 and Cx-26 mRNAs were found in the same region of the thecal cells as that previously described, but Cx-60 was different from Cx-26, in that Cx-60 mRNA was also detected in the cumulus cells in antral follicles. As, unlike Cx-30.3 and Cx-43, Cx-60 mRNA was not detected in the granulosa cells, this observation might seem odd. However, it has become evident in recent years that granulosa cells do not constitute a homogeneous tissue but, rather, show interesting regional specialization; there are distinctive differences between the peripheral positions of the membrane granulosa, periantral granulosa cells, and the cumulus region (13, 28, 29). These regions show not only morphological differences, but also distinctive characteristics in several other respects. For example, 1) the mitotic indexes in cumulus cells are 3 times higher than those in the mural granulosa cells (30, 31); 2) the affinity of hCG to rat cumulus and mural granulosa cells is the same, but the number of hCG-binding sites is very different between the two cell types (223 sites/cumulus cells vs. 2000 sites/mural cells) (32, 33); and 3) when the enzyme activity of P450 side-chain cleavage (P450scc) was examined after the administration of PMSG to immature rats, P450scc appeared in normal granulosa cells by 48 h, whereas P450scc was detected in the cumulus cells for the first time at the early onset of the endogenous LH surge (34, 35).
Taken together, these findings suggest that Cx-60 plays some specific role(s) in the cumulus cells in preovulatory and/or ovulating follicles. One specific function for the connexin gene(s) expressed in cumulus cells seems to be the formation of intercellular channels that ensure a communication pathway for the delivery of crucial signals to the oocyte (13, 36). As was found in Cx-37-/- mice (37), Cx-37 is expressed in the oocyte and is essential for oocyte maturation, implying that a functional gap junction between the oocyte and the cumulus cells is crucial for this process. As Cx-43 seems not to be involved in this channel formation (36, 37), Cx-60 as well as Cx-30.3 may be involved in forming heterotypic channels with Cx-37. An investigation of the expression of Cx-60 in murine cumulus cells and/or the expression of Cx-37 in porcine ovary and a construction of animals lacking Cx-60 are needed to understand the physiological role of Cx-60 in oogenesis and ovulation. Regarding the expression of Cx-60 in porcine cumulus cells, as described above, a limited number of COC could not be stained with the Cx-60 probe; it is not clear at this time, however, whether the presence of unstained COC has some physiological significance.
The expression profiles of the connexin genes were further examined in the rat ovary and compared with those in the porcine ovary. We reported previously that the porcine theca interna could be classified into two zones by probing with Cx-43. Although Cx-43 mRNA was detected in both the thecal and granulosa cell compartments, about a quarter of the region close to the lamina basalis (inner zone) appeared to be devoid of Cx-43 mRNA and protein as well (14). Cx-26, Cx-30.3, and Cx-32 mRNAs were detected in this zone. A major difference in the rat ovary was that Cx-43 mRNA was not clearly detected in the thecal cell compartment, unlike that in the porcine ovary. However, Cx-30.3 mRNA was equally expressed in the thecal and granulosa compartments, and Cx-43 is known to be incompatible with both Cx-26 and Cx-32 (Ref. 4 and references therein), suggesting that Cx-30.3 plays a role in connecting the thecal compartment with cumulus cells through the granulosa compartment. The compatibility of Cx-60 with Cx-26 and Cx-32 awaits further study.
The effects of gonadotropin administration on the expression of Cx-43 and Cx-60 were examined using hypophysectomized 5-week-old rats. Cx-43 mRNA was undetectable in the sections prepared from unstimulated ovaries, whereas the expression was induced after FSH administration, followed by a decrease in expression upon LH exposure. In this regard, the Cx-43 mRNA level was examined quantitatively in the porcine estrous cycle in the previous study (14) using RNA from the corpus luteum and corpus albicans in addition to small and large follicles. In that study the expression of Cx-43 mRNA was detected in the small follicles, and the level was constant until luteinization proceeded, then declined to 20% of the follicular level, indicating that the results obtained in these two studies are consistent. The down-regulation of Cx-43 expression by LH has been reported previously in follicle culture systems (26, 27). Cx-60 mRNA, in contrast, was essentially detected in the thecal layer, as was observed in the porcine ovary in all sections to a similar extent, suggesting that Cx-60 is expressed constitutively. However, the results obtained with Cx-60 mRNA in this context are qualitative, and further analyses may be needed to quantitate the change in expression levels and define the cell types in the rat thecal layer. In particular, as the thecal layer is known to be a complex tissue consisting of several different cell types, it might be important to specify the cell type in which Cx-60 is predominantly expressed. Finally, the sensitivity to gonadotropin administration seems to be different for each gene, because Cx-30.3 gave an expression profile similar to that of Cx-43 after PMSG treatment, whereas hCG stimulation did not reduce the Cx-30.3 mRNA level (data not shown).
The wide distribution and conservation of connexins in different cells and organisms as well as their modulation at both the transcription and protein levels indicate their fundamental importance for cell functions. Some of the physiological roles of the respective connexins in specific tissues have been revealed by the different approaches described above: the involvement of Cx-32 in an inherited human disease, CMT disease, and the swelling and blockage of the right ventricular outflow tract of the heart caused by a null mutation in Cx-43. Cx-37 was recently found to be present in the gap junction between oocytes and granulosa cells in mice, and mice lacking Cx-37 were reported to fail to complete meiotic processes and to ovulate as mentioned above (37). In the present study, Cx-60 was revealed to be unique in that its mRNA was detected predominantly in the thymus and spleen in addition to the ovary, where Cx-60 was originally isolated. As no other connexin genes have been reported to be expressed in the thymus or spleen (4), it is intriguing to examine the specific role of Cx-60 in these tissues; in particular, whether Cx-60 is directly involved in immune competence or the maturation of lymphocytes, and if it is required for channel formation between stromal cells and lymphocytes. With regard to these questions, very limited information on molecule selectivity or the regulatory mechanism of the Cx-60 gap junction is available at present; hence, the physiological significance of the rather restricted expression of Cx-60 awaits further clarification.
|
Footnotes |
|---|
2 Present address: Lawrence Berkeley Laboratory, University of
California, Berkeley, California 94720.
3 Present address: Mochida Pharmaceutical Co. Ltd., Kamiya, Kita,
Tokyo 115, Japan.
Received May 22, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. M. Ratchford, C. R. Esguerra, and K. H. Moley Decreased Oocyte-Granulosa Cell Gap Junction Communication and Connexin Expression in a Type 1 Diabetic Mouse Model Mol. Endocrinol., December 1, 2008; 22(12): 2643 - 2654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. SAEZ, V. M. BERTHOUD, M. C. BRANES, A. D. MARTINEZ, and E. C. BEYER Plasma Membrane Channels Formed by Connexins: Their Regulation and Functions Physiol Rev, October 1, 2003; 83(4): 1359 - 1400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. I. Gittens, A. A. Mhawi, D. Lidington, Y. Ouellette, and G. M. Kidder Functional analysis of gap junctions in ovarian granulosa cells: distinct role for connexin43 in early stages of folliculogenesis Am J Physiol Cell Physiol, April 1, 2003; 284(4): C880 - C887. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. E. I. Li, K. Waldo, K. L. Linask, T. Chen, A. Wessels, M. S. Parmacek, M. L. Kirby, and C. W. Lo An essential role for connexin43 gap junctions in mouse coronary artery development Development, March 6, 2003; 129(8): 2031 - 2042. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Granot, E. Bechor, A. Barash, and N. Dekel Connexin43 in Rat Oocytes: Developmental Modulation of Its Phosphorylation Biol Reprod, March 1, 2002; 66(3): 568 - 573. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Shimada, T. Maeda, and T. Terada Dynamic Changes of Connexin-43, Gap Junctional Protein, in Outer Layers of Cumulus Cells Are Regulated by PKC and PI 3-Kinase During Meiotic Resumption in Porcine Oocytes Biol Reprod, April 1, 2001; 64(4): 1255 - 1263. [Abstract] [Full Text]
|
|
|
|
|
|
B. Sommersberg, A. Bulling, U. Salzer, U. Fröhlich, R. E. Garfield, A. Amsterdam, and A. Mayerhofer Gap Junction Communication and Connexin 43 Gene Expression in a Rat Granulosa Cell Line: Regulation by Follicle-Stimulating Hormone Biol Reprod, December 1, 2000; 63(6): 1661 - 1668. [Abstract] [Full Text]
|
|
|
|
 |
|
 D. Manthey, F. Bukauskas, C. G. Lee, C. A. Kozak, and K. Willecke Molecular Cloning and Functional Expression of the Mouse Gap Junction Gene Connexin-57 in Human HeLa Cells J. Biol. Chem., May 21, 1999; 274(21): 14716 - 14723. [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 |
-HindIII size marker.
