Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Tissue engineering wikipedia , lookup
Gap junction wikipedia , lookup
Signal transduction wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
Cell culture wikipedia , lookup
Cell encapsulation wikipedia , lookup
Cellular differentiation wikipedia , lookup
0013-7227/96/503.00/O Endocrinology Copyright 0 1996 by The Endocrine Vol. 137, No. 11 Printed in U.S.A. Society Differential Expression of Four Connexin Genes, Cx-26, Cx-30.3, Cx-32, and Cx-43, in the Porcine Ovarian Follicle KOJI ITAHANA”, YOKO Graduate School of Biological 630-01, Japan MORIKAZU, AND TATSUO Sciences, Nara Institute TAKIZYA of Science and Technology, ABSTRACT Connexin genes expressed in porcine ovaries were isolated by reverse transcription-PCR, and the expression of four connexin genes (Cx-43, -32, -30.3, and -26) was detected by in situ hybridization in internal and surrounding compartments of large antral follicles. Cx-43 and Cx-30.3 were expressed in the theta interna as well as in the granulosa cell compartment. However, two zones were observed in the theta interna by probing with Cx-43; about a quarter of the region close to the lamina basalis appeared devoid of Cx-43 messenger RNA (mRNA) and protein as well. Cx-30.3, in contrast, seemed to be expressed ubiquitously in these two compartments. Cx-26 and Cx-32 G Al’ JUNCTIONS are specialized plasma membrane associations containing intercellular channels that coordinate tissue function by allowing cell to cell passage of ions and small molecules (l-3). Gap junction channels (GJC) are formed by the association of two hemichannels (connexons), one provided by each of the communicating cells. Each connexon is composed of 6 identical subunits, or connexins, that are essential to form a central pore and are products of members of a growing multigene family; at present, 12 different genes in mammals have been identified that code for different members of the connexin family (1, 2). Some connexins have been cloned from multiple species, and their sequences have turned out to be fairly conserved throughout the species; e.g. connexin-43 (Cx-43) (4) appears to be the most ubiquitous connexin, sharing 97% amino acid identity among mammals (2). In addition to the conserved primary structures in the transmembrane domains, the molecular diversity of different connexins in the cytoplasmic domains could be an indicator of the potential structural and regulatory complexities of GJC. Indeed, connexins are expressed in a variety of cells, and each cell type has its own characteristic pattern of expression of the connexin gene (2, 3). In ovarian follicles, intercellular communications between oocytes and somatic cell components through GJC are involved in folliculogenesis and oogenesis (5). First, gap junctions permit the transfer of metabolites such as nucleotides, amino acids, and sugars from the granulosa cells to the oocyte as needed for growth and development. Second, the Received April 29, 1996. 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: [email protected]. *On leave of absence from the Institute for Chemical Research, Kyoto University (Kyoto, Japan). 5036 Ikoma, Nara were expressed only in the thecal cells. Both Cx-43 and -30.3 also gave positive signals in the cumulus cells to the same extent as obtained in the granulosa cells with the respective probes, whereas the expression of Cx-32 and -26 was undetectable. Cx-43 mRNA drastically decreased in the functioning stage of the corpus luteum, whereas Cx-30.3 mRNA increased significantly in the early stage of luteinization in the estrus cycles. These results suggest that Cx-43 is not the only gap junction protein to be expressed in the ovarian follicle; other proteins, such as Cx-32, -30.3, and -26 seem to be expressed, and their are regulated differently. (Endocrinology 137: expressions 5036~5044,1996) communication through GJC may be very important in the regulation of meiosis. Mammalian oocytes are known to become arrested in prophase I of meiosis during the fetal and neonatal periods until the LH surge resumes the meiotic arrest and eventually causes germinal vesicle breakdown at puberty. During this process, the disassembly of gap junctions correlates temporally with the initiation of germinal vesicle breakdown (6-B), and maturation was initiated when the cumulus-oocyte complex was experimentally separated from the follicle well (5,9,10). Although the exact role of LH during the process is not yet clear, one well established role of LH is its involvement, in concert with another gonadotropin, FSH, in follicular steroidogenesis. According to a proposed mechanism described in the two cell-two gonadotropin hypothesis (ll), an intercellular communication between two types of cells, granulosa and thecal cells, may be essential. Taking into consideration the finding that several types of cells are involved in both folliculogenesis and oogenesis, it is reasonable to suggest that multiple connexin genes participate in these processes, and it is important to extend the analysis to the surrounding thecal cell compartment as well. To date, Cx-43 has been the only connexin gene reported to be expressed in granulosa cells (12, 13); no information is available on thecal cells. In this study, we extended the search for connexin genes by examining their expression in porcine ovarian follicles and the surrounding thecal region. We detected transcripts of Cx-32, -30.3, and -26 in addition to Cx-43 by reverse transcription-PCR (RT-PCR) and confirmed their expressions by in situ hybridization. Cx-43 and the other three genes were differentially expressed in the respective compartments. To further analyze the expression in estrous cycles, messenger RNA (mRNA) levels of these connexin genes in follicular and luteal phases of the porcine ovaries were analyzed by Northern blot analysis and RT-PCR. CONNEXIN Animals, Materials tissues, and cells mRNA IN THE OVARIAN and Methods Pig ovaries were obtained from 4- to 12-month-old pigs at a local slaughterhouse (14). Follicular contents were sucked out by syringes from small or large follicles (-1-7 mm in diameter) and used directly as a source of follicular cells containing granulosa cells and oocytes. Cumulus-oocyte complexes could be isolated under the microscope. Corpora lutea were classified into three stages in addition to corpus albicans: early, functional, and regressing stages based on the morphological characteristics of the surface (14). RNA extraction oligonucleotides and RT-PCR cloning with degenerate Total RNA was extracted by acid guanidinium-phenol chloroform procedure from follicles directly or from corpora lutea (15), and possible contaminated genomic DNA was removed by digestion with deoxyribonuclease free of ribonuclease (RQl DNase, Promega, Madison, WI). Ten micrograms of total RNA were reverse transcribed into complementary DNA (cDNA) by Superscript II (BRL) with oligo-(deoxythymidine),,_,, primer for RT-I’CR according to the supplier’s instructions. Two pairs of degenerate oligonucleotides (primers Fl and F2 and primers Rl and R2; F stands for forward primers and R for reverse primers) corresponding to the conserved regions among the connexin family were synthesized (Fig. 1, A and B) (3). Primer Fl corresponds to the amino-terminal region, and primer F2 includes a conserved amino acid sequence, VCYD, within the first extracellular domain. Primer R2 includes a conserved amino acid sequence, VDCF, in the second extracellular domain, and Rl corresponds to the carboxy-end portion within the same domain. To minimize probe degeneracy, not all of the possible sequence variations revealed in the alignment were represented in each probe. RT-PCR reactions (30 ~1 in volume) contained 0.2 pg RT products, 0.2 FM of primer F or R, and 200 PM deoxy-NTP with Tuq Extender Buffer (Stratagene, La Jolla, CA). The addition of AmpliTaq (2 U; I’erkin-Elmer, A Cytoplasmic Membrane Extracellular 3 t t 3 F2 R2 t Rl FOLLICLE Norwalk, CT) and Tuq Extender PCR Additive (2 U; Stratagene) was performed after the reaction reached 94 C to minimize undesired priming during the initial cycle. Thirty cycles at 94 C were performed for 30 set, 60 C for 2 min, and 72 C for 3 min. This was followed by a final extension for 15 min at 72 C. Reaction products were separated and visualized on agarose gel electrophoresis. The products were finally subcloned in pT7Blue-T cloning vector (Novagen) and subjected to DNA sequence analysis by AB13700 (Perkin-Elmer). Tissue sections and hematoxylin primerF1 : ATGGGTGACTGGAG(C/T)(G/T)(C/T)C(C/T)T(G/A)G primer F2 : GGCTG(C/T)(AG)A(G/C)AA(C/T)GTCTGCTA(T/C)G primer Rl : ACCACCA(G/A)CAT(G/A)AAGA(T/C)(G/A)ATGAAG primer RZ : GCC(T/G)GGA(G/A)A(C/T)(G/A)AA(G/A)CAGTCCAC FIG. 1. Primers and reactions of RT-PCR. A, Domains of the connexins are defined according to the method of Bennett et al. (3) and are indicated: amino-terminus (N), cytoplasmic loop (Cy), carboxy-terminus (C), membrane spanning(Ml-M4), and extracellular loop (El and E2). The locations of four primers constructed for RT-PCR are shown side by side (Fl and F2; Rl and R2). B, Nucleotide sequences of each primer. leosin staining Ovarian sections were prepared for in situ hybridization essentially as described previously (15). In brief, the removed ovaries were covered in OCT embedding compound (Miles, Elkhart, IN) and snap-frozen in liquid nitrogen. Cryosections (8 pm thick) were prepared using a Cryostat HM500-OM (Microm, Germany), collected on silane-coated slides, and stored at -80 C until hybridization. For fixation, slides were dipped into freshly prepared 4% paraformaldehyde in PBS for 60 min at room temperature and washed in PBS (10 min), followed by digestion with protenase K (1 pg/ml) for 10 min at 37 C. The slides were washed in 0.1 M triethanolamine for 1 min, rinsed in distilled water, and dipped into 0.1 M triethanolamine containing 0.25% acetic anhydride for 1 min. Sections were washed finally in PBS and sequentially dehydrated for 1 min each at 70%, 80%, 90%, and 100% ethanol (followed by air-drying). Hematoxylin-eosin staining was performed essentially according to the procedure described in Ref. 15. In situ hybridization and probes DNA regions containing unique sequences of the respective connexin genes (Fig. 2) were subcloned into the RNA transcription vector pBluescript II SK+ or pT7Blue T-vector. Complementary RNA probes on the antisense were prepared by in vitro transcription of the linearized constructs using SP6 or T7 RNA polymerase in the presence of digoxigeninUTP. After RNA polymerase reactions, DNA templates were hydrolyzed with RQl deoxyribonuclease. Procedures for the in situ hybridization were essentially based on those of Ausubel et al. (15). Each probe was diluted in hybridization buffer containing 50% formamide, 0.6 M NaCl, 10 mM Tris (pH 7.4), 1 mM EDTA (pH 8.0), 0.25% SDS, 10% dextran sulfate, 1 X Denhardt’s solution, and 200 ~g/ml Esckevickia coli transfer RNA. Probe solution was heated at 85 C for 3 min, chilled on ice, and added to each slide in a volume of 30 ~1. The slides were incubated overnight at 50 C in a covered container humidified with 50% formamide and drained; they were then immersed in 5 X SSC (standard saline citrate), followed by incubation in 2 X SSC containing 50% formamide for 30 min at 50°C. The slides were washed with TEN [0.5 M NaCl, 10 mM Tris (pH 8.0), and 1 mM EDTA] for 10 min at 37 C, then treated with 1 pg/ml ribonuclease A in TEN for 30 min at 37 C and finally with 2 X SSC for 20 min at 50 C. The slides were then washed with 0.2 x SSC at 50 C twice. Sections were stained by the use of a digoxigenin detection kit (Boehringer Mannheim, Mannheim, Germany) according to the supplier’s instructions, mRNA levels: Northern B 5037 blotting and RT-PCR Total RNA samples (20 pg) from follicular cells or the respective stages of corpora lutea (14) were denatured with formaldehyde, electrophoresed in 1.2% agarose gels, and transferred onto nylon membranes (Immobilon, Millipore Corp., Bedford, MA) as described previously (10, 14). Hybridization was performed under conditions of high stringency; filters were washed three times with 2 X SSC-0.1% SDS for 15 min each time, followed by washing in 0.2 X SSC-0.1% SDS at 37 C. A probe used for filter hybridizations was derived from the Cx-43 cDNA clone used for in situ hybridization described above, but the probe was labeled with [32P]deoxy-CTP instead of digoxigenin. The components of RT-PCR reactions were identical to those of reactions with the degenerate oligonucleotides described above unless otherwise indicated, whereas reaction conditions were slightly modified; 35 cycles at 94 C were performed for 30 set, 60 C for 1 min, and 72 C for 2 min, followed by a final extension for 5 min at 72 C. Reaction products were separated and visualized on acrylamide gel electrophoresis. For quantitative analysis, the products were stained with CYBR Green I (FMC Bioproducts), and the intensities were measured by a CONNEXIN mRNA IN THE OVARIAN FOLLICLE 5038 POT cx43 Hum Cx43 Rat cx43 II I,-- ---------- KL LDKVQAYSTA GGKVWLSVLF IFRILLLGTA Primer Fl -- ---------------------------- 1 --- ------- ---------- Endo. Vol137.No VESAWGDEQS AFRCNTQQPG CENVCYDKSF PISHVRFWVL QIIFVSVPTL ---------------------------__________ -----_---- ---------- __________ --_------- 11 LYLAHVFYVM _---__--__ ---------- ---------- ----__---- FKSVFEVAFL ---I------ LIQWYIYGFS _--------- LSAVYTCKRD PCPHQVDCFL ----------------_-_ ---v------ ----_----- ---------- I RKEEKLNKKE EELKVAQTDG VNVEMHLKQI EIKKFKYGIE ---D-----------------_-----------------_------ ---_------ ---E------ ---------- EHGKVKMRGG LLRTYIISIL __________ ----___-__ ---------- ----_----- ---------- SRPTEKT ---_-----_--- POT Cx32 Hum Cx32 Rat Cx32 0 PrimerF2 HF FPISHVRLWS LQLILVSTPA __ __---------_---____ Q----_----__________ LLVAMHVAHQ QHIEKKMLRL __________ ----_-----_--______ ---------- EGHADPLHLE EVKRHKVHIS GTLWWTYVIS WFRLLFEAA ---G-----__________ ______------------v ---G-v---_-_-____-_ ------------------V FMYVFYLLYP _------------------ GYAMVRLVKC EAYPCPNT _-__-------F-------------DV------ EF FPLSHVRLWA LQLILVTCPS __ --.J--------____---- Par cx30.3 Mus Cx30.3 Por Cx26 Hum Cx26 Rat Cx26 LLWMHVAYR QERERKHRLK HGPDARSLYD NPNKKRGGLW WTYLLSLIFK ---------E-----------N-PA--~ -LS------_--------- rl-- ---------primerF2 HY FPISHIRLWA _- -_-------- LQLIFVSTPA ----M-----------s-- LLVAMHVA~R RHEKKRKFIK -----__----------M__________ _______--- GEMKSEFKDI EEIKTQKVRI --I-N-------_-_____ --I--------_-_-__-- AAVDAGFLYI v---S----- EGALWWTYTG SIFFRILFEA --s------T -----VI----s------s -----VI--- FHRLYRNYDM --(-I-Kj-,--- AFMYVFYVMY V------I------_---- SGFSMQRLVK CNAWPCPNT N--F ______ --------PrimerR p---------------0 FIG. 2. Predicted amino acid sequences of cDNA clones amplified by RT-PCR on porcine ovarian follicle mRNA. Amplified products with the primers shown in Fig. 1B were subcloned into pT’IBlue-T vector and subjected to DNA sequence analysis. A search for amino acid sequences thus determined was made in the SWISS-PROT database. Each amplified region of porcine counterpart is aligned with the corresponding region(s) of the mammalian gene(s): human Cx-43 (17), rat Cx-43 (4), human Cx-32 (18), rat Cx-32 (19), mouse Cx-30.3 (20), human Cx-26 (21), and rat Cx-26 (22). Amino acid sequences identical to those of the porcine genes are indicated by dashes. Open. boxes indicate the primer sites. Fluor Imager (Molecular Dynamics, Sunnyvale, CA). A portion of nucleotide sequences of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene that were conserved among the mammals (16) was chosen to construct primers for PCR; the reaction was expected to produce a 122-bp fragment, which was used for preparation of 32P-labeled GAPDH probe as well. Immunocytochemistry Frozen sections were rehydrated into PBS to remove OCT compound and incubated in PBS containing 10% skim milk. Incubation with antiCx-43 monoclonal antibody (Chemicon International) was performed overnight at 4 C, followed by washing three times with PBS. The bound antibodies were revealed with antimouse Ig-fluorescein (5-bromo-2’deoxyuridine labeling and detection kit I, Boehringer Mannheim) according to the manufacturer’s instructions. R.esldts Domain structure of the connexin family of primers for PCR and construction A search was made for connexin genes expressed in the porcine ovarian follicles by applying RT-PCR. To design primers for that purpose, domain structures identified in the connexin family were considered. The general archi- tecture of the connexin genes can be outlined, as shown in Fig. 1, based on the sequence similarities between previously characterized connexin genes and the evidence obtained by biochemical analyses (3); the N- and C-terminals are cytoplasmic (designated N and C, respectively), and there are four membrane-spanning regions (Ml-M4), with a cytoplasmic loop (Cy) and two extracellular loops (El and E2). Of these domains, Cy and C are relatively unique regions for an individual gene, and regions other than these two are well conserved in the family. However, there exist some portions within the N, E2, and M4 regions where identities are shared only by genes such as Cx-32, -31, and -26, making it possible to divide the family into two groups (1,3,11): group I (or /3) consists of genes such as Cx-32, -31, and -26, whereas group II (or) consists of genes such as Cx-43, -46, and -38. Two degenerate oligonucleotide primers, Fl and F2, corresponding to the sense strands of N and El, respectively, and two other degenerate primers, Rl and R2, corresponding to the antisense strand sequences of the 5’- and 3’-portions of E2, respectively, were prepared (Fig. 1B). Fl and Rl are CONNEXIN mRNA IN THE OVARIAN FOLLICLE specific to group II, whereas F2 and R2 are common to all connexin genes. cDNA cloning and DNA sequence analyses RT-PCR amplification with combinations of the abovedescribed primers at a 60 C annealing temperature produced distinct respective bands after agarose gel electrophoresis (data not shown). The products were recovered from gel, subcloned in pT7Blue-T vector, and subjected to DNA sequence analyses. Products with combinations of primers other than with F2 and R2 were found to encode single species of the connexin gene Cx-43, as far as sequenced; a predicted amino acid sequence from a deduced DNA sequence was identical to that of rat Cx-43 except for two residues (Fig. 2) (4). In contrast, products with F2 and R2 were found to contain a mixture of porcine counterparts of mammalian Cx-32, Cx-30.3, and Cx-26 in addition to Cx-43 (Fig. 2) (17-22). All cloned cDNA contained the Cy region that is unique to each connexin gene in addition to conserved M2 and M3 regions (Fig. lA), as expected. The degree of sequence identity in each gene was fairly high, i.e. at least 90%, among the mammals. In situ hybridization As Cx-43 is the only type of gap junction protein reported to be expressed in the granulosa cell compartment (12, 13), we wanted to confirm and examine further the expression of connexin genes other than Cx-43 in the corresponding and neighboring compartments. To do so, a digoxigenin-labeled probe specific to an individual gene was prepared and hybridized with thin sections of porcine ovary (Fig. 3, lanes A and B). To begin with, cross-reactivities between each probe were examined, and their specificities were confirmed (data not shown). Cx-43 was expressed in the granulosa cell compartment (the follicle granulosa cells within the basement membrane) and in the region surrounding the follicles as well (Fig. 3). These two regions were clearly separated by cell layers where no signal was detected; based on the profile of hematoxylin-eosin staining, both stained and unstained compartments were assumed to be included in the theta interna, indicating that the theta interna could be divided into two zones and that about a quarter of internal thecal cell layers adjacent to the basal membrane was devoid of the Cx-43 mRNA. This expression profile was further confirmed at the protein level by immunoblotting using an anti-Cx-43 antibody as well as in a different follicle (Fig. 4, A-C). The expression profile of Cx-30.3 was different from that of Cx-43 in two respects: 1) Cx-30.3 mRNA was detected in both the granulosa cell and the surrounding thecal compartments without interruption as observed in the case of Cx-43; and 2) the signal with Cx-30.3 was more intense in thecal cells than in granulosa cells. The difference in intensities between Cx-43 and Cx-30.3 does not reflect the difference in their mRNA expression levels, because the specific activities and/ or lengths of these two probes were not identical. Cx-26 and Cx-32, in contrast, were expressed exclusively in the thecal cell compartment (Fig. 3). Therefore, Cx-43 seems not to be the only gap 5039 junction protein expressed in the ovarian follicle; other genes, such as Cx-26, -30.3, and -32, appear to be expressed as well. Finally, ribonuclease treatment of the sections before hybridization failed to give any signals with any probe, indicating that hybridization reactions were actually between probes and intrinsic unique transcripts. Furthermore, the expression of all Cx genes examined were restricted to within the follicle and its surrounding thecal regions and not in the regions of surrounding stroma. Expression of connexin genes in cumulus-oocyte complexes Besides GJC that connect single species of cells, intercellular couplings between granulosa cells and cumulus cells or between cumulus cells and oocytes are believed to play important roles in the regulation of both oocyte growth and meiosis (5). We, therefore, examined connexin gene expression in the cumulus cells (Fig. 3, lane C). Due to the fixation procedures or for other reasons, as porcine sections used in the above experiments were derived from either fully grown antral follicles or out-centered portions of follicles, no cumulus-oocyte complexes (COC) were recovered or detected in those preparations. We thereby recovered COC from the total follicle contents, spun them down onto slides glasses, and probed them with respective digoxigenin-labeled DNAs. Four or five COC were examined with each probe, and the typical profiles obtained are shown in Fig. 3 (lane C). Both the Cx-43 and Cx-30.3 probes gave intense signals in the cumulus cells, whereas signals were undetectable with the Cx-32 and Cx-26 probes. It is difficult to conclude that there was a total absence of Cx-32 and Cx-26 transcripts in COC, but it was clear that the intensities given by Cx-43 or Cx-30.3 were at least comparable in the granulosa cell compartment and the cumulus cells, whereas the signal was significantly faint in cumulus cells in the case of Cx-32 and Cx-26 (Fig. 3). The same reasoning may be possible with the oocytes themselves, where no significant signals were obtained with any probes; staining was seen in the central region with the Cx-30.3 probe derived from the surrounding cumulus cells, not from the oocyte. Connexin mRNA levels in estrus cycle The expression profiles of connexin genes in luteinizing ovaries have not yet been reported. As Cx-43 and Cx-30.3 were found in the present study to be expressed in granulosa cells, we prepared total RNA from three stages of corpus luteum (early stage, functional stage, and regressing stage) and corpus albicans in addition to RNA from both small (cl mm in diameter) and large (>5 mm in diameter) follicles and examined their mRNA levels by Northern blot analysis and RT-PCR (Fig. 5). The expression of Cx-43 mRNA was clearly detected by Northern blot analysis; the expression level was relatively high in follicle cells and during the regression stage, whereas the expression was significantly down-regulated in the functional stage of the corpus luteum (Fig. 5A). The same expression profile of Cx-43 mRNA was confirmed by RT-PCR (data not shown). The expression of Cx-30.3 mRNA, on the CONNEXIN mRNA IN THE OVARIAN FOLLICLE Endo . 1996 Vol 137 . No 11 FIG. 3. Expression of Cx-43, -32, -30.3, and -26 mRNAs in the porcine large follicles. Sections (8 pm) adjacent to each other were hybridized with the respective digoxigenin-labeled probes. Lane A shows a representative follicle (magnification, X20); lane B shows the same sample with higher magnification (X 100); lane C shows a representative profile of cumulus-oocyte complex with the respective probes (bar indicates 25 km). The sample on the bottom ro’owindicates a follicle stained with hematoxylin-eosin; an area surrounded by an open &XC is the region shown in lane B. other hand, was difficult to detect quantitatively by Northern blot; hence, we examined its expression by RT-PCR on the same RNA preparations as those studied for Cx-43 expres- sion (Fig. 5B). The condition used for PCR produced amplification in a linear fashion; a 5-fold increase in the template caused an equivalent increase in the product, and a serial CONNEXIN mRNA IN THE OVARIAN FOLLICLE 5041 ne antral follicles. A, A section (8 pm) containing a follicle different from that used in Fig. 3 was stained with anti-Cx-43 antibody as a primary antibody and visualized by fluorescein isothiocyanate-labeled antimouse IgG as a second antibody. B, An adjacent section was visualized with the same second antibody, but without the treatment with anti-Cx-43 antibody. C, A section adjacent to that used in A was hybridized with the digoxigenin-labeled Cx-43 probe. D, A hematoxylin-eosin-stained section. Magnification, x200. decrease in the template resulted in significant reductions of the products to the level where any quantification was difficult (Fig. 58, lanes 3 and 8-10). The peak of Cx-30.3 mRNA expression was seen in large follicles and the early stage of the corpus luteum, and the authenticities of the bands were confirmed by Southern blot with the 32P-labeled Cx-30.3 probe (data not shown). Discussion In this study, we identified the expression of four connexin genes (Cx-43, Cx-32, Cx-30.3, and Cx-26) in the porcine ovary by applying RT-PCR and characterized their expression profiles by in situ hybridization, Northern blot analysis, and RT-PCR. The results revealed that these four types of con- CONNEXIN mRNA IN THE OVARIAN FOLLICLE 5042 Endo. 1996 Vol 137. No 11 123456 Ml2345678910 cx-30.3 Ml2345678910 GAPDH GAPDH \ 100 7 80 z! $ 2 60 40 2 20 0 123456 123456 FIG. 5. Expression profiles of Cx-43 and Cx-30.3 mRNAs in the estrus cycle. A, Total RNA (20 pg) isolated from each stage in the estrus cycle of porcine ovary were run on agarose gel, transferred onto nylon membrane (Immobilon N, Millipore), and probed with the 32P-labeled Cx-43 cDNA. Data are representative of two experiments. The positions of the 28s and 18s ribosomal RNAs are indicated by arrows (top panel). The same filter was probed with the 32P-labeled GAPDH DNA (middle panel). The expression level at each stage was normalized with the densities given by GAPDH mRNA, and their relative estimates against the lane 1 are shown as a linear chart (bottom panel). Each point represents the mean level for two experiments. Lane 1, Small follicles; lane 2, large follicles; lane 3, corpus luteum (early stage); lane 4, corpus luteum (functioning stage); lane 5, corpus luteum (regressing stage); lane 6, corpus albicans. B, The expression of Cx-30.3 was examined by RT-PCR using the same RNA preparations as those described above, and the product profiles (131 bp) obtained by acrylamide gel electrophoresis are shown (top panel). The RT products were examined by PCR with GAPDH primers producing 122-bp fragments (middle panel). Data are representative of two experiments. The expression levels were normalized as described in A, and the level of lane 3 is indicated as 100%; lane numbers correspond to those of the top and middle panels (bottom panel). Each point represents the mean level for two experiments. Lane M, HaeIII-digested pUC18 DNA, lanes 1-6, the same as in A, lane 7, no template; lane 8, same RNA but 5 times as much as in lane 3; lane 9, l/&h the RNA of lane 3; lane 10, 1/25th RNA of lane 3. nexin genes were actually expressed within the follicles and the surrounding thecal cells, and their expression patterns were differentially regulated both temporally and spatially, as schematically summarized in Fig. 6. The signals thus obtained with the respective probes were not simply due to cross-hybridization with other genes, because the specificity of each probe was examined against those of other probes, and no cross-reactivities were detected, as described above. However, in general, the possibility of the existence of a gene(s) that is still unidentified but possesses structural commonalities with the connexin family cannot be excluded. The connexin family genes could be further classified into two groups based on their primary structures: group I (p), including Cx32, Cx-30.3, and Cx-26, and group II (a), Cx-43. Although it is not yet clear whether this classification by structural features reflects any physiological significance, it is clear that both types of connexin genes are expressed in ovarian follicles. Cx-32 and Cx-26 seem to be expressed predominantly in the thecal cell compartment, distinguishing these two genes from Cx-30.3 within group II. The differential expressions of these genes between granulosa and thecal cells may reflect their differences in physiological roles; one possible role is in the biogenesis of steroid hormones. Granulosa and thecal cells have been speculated to have different roles in steroidogenic capacities. One factor to be considered is differences in the cells’ spatial localizations; the granulosa cells of the follicles are avascular, and the granulosa cells are not accessible to low density lipoprotein, which is the principal source of cholesterol used for steroidogenesis. In contrast, thecal intema cells can become uptake precursors for steroidogenesis and respond to LH. These possibilities have been examined by using isolated granulosa cells and thecal cells, leading to the two cell-two gonadotropin theory (ll), which proposed that in response to LH, the theta interna cells convert C,, steroids to Cl9 steroids, and that FSH stimulates granulosa cells to aromatize the preformed C,, steroids to estrogen. To support such a mechanism and make it possible, not only GJC, which might be specific for each cell type to create intercellular synchronization but also another type of GJC that can bridge different CONNEXIN mRNA IN THE Cx32 cx43 TFu- 0 ‘I BM GC \ cx30.3 CUM(-- Cx26 -u-0 FIG. 6. Species and localizations porcine antral follicles expression of connexin ing. TE, Theta externa, granulosa cells; CUM, of connexin genes expressed in the are schematically shown. Regions where the genes were identified are illustrated by shadTI, theca interna; BM, basal membrane; GC, cumulus cells; 00, oocytes. OVARIAN 5043 FOLLICLE cruitment of dominant follicles to be matured or ovulated (27). Observations by Medui et al. (23) and in our study also strongly suggest that Cx-43 and LHR are coexpressed in the rest of the compartments. It has been established, however, that an LH surge causes down-regulation of Cx-43 mRNA expression (28,29) (Itahana, K., and T. Takeya, unpublished observation). The coexpression of LHR and Cx-43 in antral or preovulatory follicles thus suggests that the LH level during the follicle phase of the estrus cycle is not sufficient to repress Cx-43 mRNA expression directly or induce such inhibitory activity, whereas the increased LH level at the surge could be sufficient to trigger the down-regulatory mechanism. In the present study, we newly identified the expression of Cx-30.3 in cumulus granulosa cells. The expression of Cx-43 in this region has been reported (13,30,31). Valdimarsson et al. (30) further demonstrated the expression of Cx-32 in COC; they identified the products at both the protein and mRNA levels. In contrast, the expression of Cx-32 in cumulus cells seemed much fainter than that in thecal cells and was almost undetectable in this study. This does not necessarily mean that Cx-32 is not expressed in cumulus cells, because Valdimarsson et al. (30) used both RT-PCR and an anti-Cx-32 antibody to detect the product, and thereby the sensitivity of their methods may differ from ours. References types of cells might exist. Our present findings on the identities and the expression profiles of connexin genes are relevant to this question. Although Cx-43 and Cx-30.3 are coexpressed in both granulosa and theta interna cells, they showed distinct differences, in that Cx-43 was not expressed specifically in the theta interna region close to the lamina basalis. The immunostaining of large antral follicles by using anti-LH receptor (LHR) antibody previously revealed the presence of two clear-cut zones in the theta interna (23); about one third of the inner layers of theta interna cells adjacent to the basal lamina were not stained, whereas the rest of the internal thecal cells were significantly stained. The functional significance of the existence of the zone devoid of LHR is not yet clear, but the existence of thecal cell layers where key enzymes for steroidogenesis, such as 3-hydroxysteroid dehydrogenase, P450-17 anhydrase, and aromatase could not be detected in human follicles, has also been reported (24-26). Based on the species and stage specificities of the sources of the materials and on the present hematoxylin-eosin staining profile, this LHR-devoid zone seems to correspond to the portion of theta interna detected in this study as a region devoid of Cx-43 mRNA. The existence of the zone lacking Cx-43 between the theta and granulosa cell compartments in the large antral follicles implies the unique role of Cx-43 in folliculogenesis and/or oogenesis. This zone may function as a blockade against the follicle compartment by interrupting the transfer of signal molecules produced in the theta cells by LH stimulation, and we suspect that this mechanism may be crucial for the re- 1. Kumar 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. NM, Gilula NB 1996 The zap iunction communication channel. Cell 84381-388 Beyer EC 1993 Gap junctions. Int Rev Cyt 137C:l-37 Bennett MVL, Barrio LC, Bargiello TA, Spray DC, Hertzberg E, Saez JC 1991 Gap junctions: new tools, new answers, new questions. Ne&on 6:3Og320 Beyer EC, Paul DL, Goodenough DA 1987 Connexin 43: a protein from rat heart homologous to a gap junction protein from liver. J Cell Biol105:2621-2629 Eppig JJ 1991 Intercommunication between mammalian oocytes and companion somatic cells. BioEssays 13:569-574 Gilula NB, Epstein ML, Beers W 1978 Cell-to-cell communication and ovul&ion; a study of the cumulus-oocyte complex. J Cell Biol 78:58-75 Larsen WJ, Wert SE, Brunner GD 1986 A dramatic loss of cumulus cell gap junction is correlated with germinal vesicle breakdown in rat oocytes. Dev Biol 113:517-521 Wert SE, Larsen WJ 1990 Preendocytotic alterations in cumulus cell gap junctions precede meiotic resumption in the rat cumulus-oocyte complex. Tissue Cell 22:827-851 Tsfriri A, Charming Cl’ 1975 An inhibitory influence of granulosa cells and follicular fluid upon oocyte meiosis in vim. Endocrinology 96:922-927 Racowsky C, Baldwin KV 1989 In vitro and in vim studies reveal that hamster oocyte m&tic arrest is maintained only transiently by follicular fluid, but persistently by membrana/cumulus granulosa cell contact. Dev Biol 134297-306 Gore-Langton RE, Armstrong D 1994 Follicular steroidogenesis and its control. In: Knobil E, Neil1 JD (eds) The Physiology of Reproduction. ed 2. Raven Press, New York, vol 1:571-627 Risek B, Guthrie S, Kumar N, Gilula NB 1990 Modulation of gap junction transcript and protein expression during pregnancy in the rat. J Cell Biol 110:269-282 Wiesen JF, Midgley Jr R 1993 Changes in expression of connexin 43 gap junction messenger ribonucleic acid and protein during ovarian follicular growth. Endocrinology 133:741-746 Tanaka T, Andoh N, Takeya T, Sato E 1992 Differential screening of ovarian cDNA libraries detected the expression of the porcine collagenase inhibitor gene in junctional corpora lutea. Mol Cell Endocrinol83:65-71 Ausubel FM, Brent R, Kingston RE, Moore DD, Smith JA, Seidman JG, Shuhl K 1988 Current Protocols in Molecular Biology, suppl 7. Wiley Interscience, New York Tokunaga K, Nakamura Y, Sakata K, Fujimori K, Ohkubo M, Sawada K, Sakiyama S 1987 Enhanced expression of a glyceraldehyde-3-phosphate dehydrogenase gene in human lung cancers. Cancer Res 47:5616-5619 Fishman GI, Spray DC, Leinwand LA 1990 Molecular characterization and VI I 5044 18. 19. 20. 21. 22. 23. 24. CONNEXIN mRNA IN THE OVARIAN functional expression of the human cardiac gap junction channel. J Cell Biol 111:589-598 Kumar NM, Gilula NB 1986 Cloning and characterization of human and rat liver cDNAs coding for a gap junction protein. J Cell Biol 103:767-776 Paul DL 1986 Molecular cloning of cDNA for rat liver gap junction protein. J Cell Biol 103:123-134 Hennemann H, Dahl E, White JB, Schwarz H-J, Lalley PA, Chang S, Nicholson BJ, Willecke K 1992 Two gap junction genes, connexin 31.1 and 30.3, are closely linked on mouse chromosome 4 and preferentially expressed in skin. J Biol Chem 26717225-17233 Lee SW, Tomasetto C, Paul D, Keyomarsi K, Sager R 1992 Transcriptional downregulation of gap junction proteins blocks junctional communication in human mammary tumor cell lines. J Cell Biol 188:1213-1221 Zhang J-T, Nicholson BJ 1989 Sequence and tissue distribution of a second protein of hepatic gap junctions, Cx2.6, as deduced from its cDNA. J Cell Biol 109:3391-3401 Meduri G, Vuhai-Luuthi MT, Jolivet A, Milgrom E 1992 New functional zonation in the ovary as shown by immunohistochemistry of luteinizing hormone receptor. Endocrinology 131:366-373 Inkster SE, Brodie A 1991 Expression of aromatase cytochrome P-450 in premenopausal and postmenopausal human ovaries: an immunocytochemical FOLLICLE Endo. 1996 Vol 137 . No 11 study. J Clin Endocrionol Metab 73:717-726 25. Sasano H, Mori T, Sasano N, Negora H, Mason JI 1990 Immunolocalization of 3-B-hydroxysteroid dehydrogenase in human ovary. J Reprod Fertil 89:743-751 26. Zlotkin T, Fargash Y, Orly I 1986 Cell-specific expression of immunoreactive cholesterol side-chain cleavage cytochrome P-450 during follicular development in the rat ovary. Endocrinology 119:2809-2820 27. Fortune JE 1994 Ovarian follicular growth and development in mammals, Biol Reprod 501225-232 28. Schreiber JR, Beckmann MW, Polacek D, Davies PF 1993 Changes in gap junction connexin-43 messenger ribonucleic acid levels associated with rat ovarian follicular development as demonstrated by in situ hybridiazation. Am J Obstet Gynecol 168:1094-1104 29. Granot I, Dekel N 1994 Phosphorylation and expression of connexin-43 ovarian gap junction protein are regulated by luteinizing hormone. J Biol Chem 269:30502-30509 30. Valdimarsson G, De Sousa PA, Kidder GM 1993 Coexpression of gap junction proteins in the cumulus-oocyte complex. Mol Reprod Dev 367-15 31. Sutovsky P, Flechon JE, Flechon B, Motlik J, Peynot N, Chesne P, Heyman Y 1993 Dynamic changes of gap junctions and cytoskeleton during in vitro culture of cattle oocyte cumulus complexes. Biol Reprod 49:1277-1287