| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
RESEARCH |
Department of Animal Science, 1 Department of Computer Science, 2 Department of Molecular Biology, University of Missouri, Columbia, Missouri 65211, USA and 3 Monsanto Company, St Louis, Missouri 63198, USA
Correspondence should be addressed to M C Lucy; Email: lucym{at}missouri.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Luteinization of follicular cells and ovulation of the follicle are complex processes that require the coordinated expression of a large number of genes (Sarit et al. 2005). Some genes involved in the processes are well characterized, but there are other genes whose role in the process is unknown. Gene-expression analysis using microarrays is a method of identifying genes involved in biological processes (Eisen & Brown 1999, Koopman et al. 2002). We developed a microarray that contained expressed sequence tags (EST) found in porcine ovary (Jiang et al. 2004). The objective of this study was to use the microarray to systematically evaluate genes expressed in porcine ovary and to examine their respective roles during the luteinization of porcine preovulatory follicles.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Preparation of cDNA microarrays
Eleven cDNA libraries were used to generate EST from porcine ovary (Jiang et al. 2004). The cDNA libraries were created using tissue from fetal, neonatal, and prepubertal porcine ovaries, pubertal ovaries on different days of the estrous cycle (day 0 (follicle), 5 (ovary), and 12 (follicle and corpus luteum)), and follicles isolated from weaned sows (2, 4, 6, and 8 mm diameter). The cDNA sequences from the 11 libraries were clustered into a single nonredundant UniGene set using the tlcluster program (Scheetz et al. 2003). The UniGene set contained all of the singleton EST as well as a single representative member for each cluster (clusters were groups of independent clones encoding the same EST; the representative member was the clone with the longest sequence read). The UniGene set contained a total of 8009 genes. The cDNA corresponding to each EST were PCR-amplified from bacterial stocks that contained the representative plasmid. The PCR contained 0.25 mM dNTP and 200 ng M13 forward and reverse primers and 0.85 U Biolase DNA polymerase (Bioline, Randolph, MA, USA). The PCR conditions were 3 min at 95 °C initial denaturation followed by 30 cycles of 30 s at 94 °C denaturation, 30 s at 55 °C annealing, and 3 min at 72 °C extension. The PCR was purified using MultiScreen-PCR Plates (Millipore, Billerica, MA, USA) according to manufacturer’s instructions. The purified PCRs were dried in a vacuum centrifugation apparatus and resuspended in 10 µl 3x SSC.
Microarray slides were printed in the same facility and under the same conditions that were described in a previous paper from our group (Whitworth et al. 2005). Gold Seal glass microscope slides (Fisher Scientific, Hampton, NH, USA) were coated with 0.02% poly-L-lysine (Sigma) in 0.05x PBS and aged for 15 days at room temperature under desiccant (Eisen & Brown 1999). The cDNA was printed onto the slides by using a pick and place robot. Printed slides were rehydrated and cross-linked at 120 mJ/cm2 for 20 s (Spectrolinker, Spectronics Corp., Westbury, NY, USA). Slides were blocked with 0.018% succinic anhydride and 0.043 M sodium borate in 1-methyl-2-pyrrolidinone and stored in a desiccator until hybridization.
Total RNA purification, cDNA labeling, and hybridization of arrays
Eleven individual follicles were used for RNA isolation. Total cellular RNA was extracted using the TRIZOL reagent (Invitrogen) according to the manufacturer’s instructions. The integrity and purity of the extracted RNA were verified by measuring the ratio of absorbencies at 260 and 280 nm and by visual inspection of an RNA aliquot on a Tris–borate/EDTA (0.09 M Tris–borate, 0.02 M EDTA) ethidium bromide-stained agarose gel.
At least two replicates for each RNA sample were subjected to the microarray analysis (29 microarrays contributed to the dataset). A control RNA sample was generated by combining equal amounts of total RNA from 2, 4, 6, and 8 mm porcine follicles as well as corpus luteum and corpus hemmoragicum. The test samples (preovulatory estrogenic and -luteinized) were labeled with Cy5 and the control sample was labeled with Cy3. Indirect labeling of cDNA with amino allyl dUTP was done according to the procedure of Hegde et al.(2000) with the following modifications. Equal concentrations of 3-amino allyl dUTP and dTTP were used in the RT of 15 µg total RNA. The purified cDNA was incubated with approximately 0.02 mg Cy5 (test sample) or Cy3 (control) for 1 h at room temperature before the purification with PCR purification columns (Qiagen). The reactions were dried in a vacuum centrifugation apparatus and resuspended in 26 µl of hybridization buffer containing 48% formamide, 4.8x SSC, 0.1% SDS and 20 µg poly A. The cDNA was denatured by incubating at 95 °C for 3 min and allowed to cool to room temperature before being transferred onto a microarray slide. Microarray slides were hybridized overnight at 42 °C and washed (4 min at 42 °C in 1x SSC, 0.2% SDS; 4 min at room temperature in 0.1x SSC, 0.2% SDS; 4 min at room temperature in 0.1x SSC and rinsed with water). The microarrays were then dried by centrifugation before scanning was done with a GenePix 4000B Scanner (Axon Instruments, Inc., Union City, CA, USA).
The intensity of each spot was determined using GenePix Pro (Version 4.0) software (Axon Instruments, Inc.). Raw data were imported into GeneSpring microarray analysis software (Silicon Genetics, Redwood City, CA, USA). Genes with intensities less than 160 in the control channel (Cy3) in at least half of the arrays were filtered out (n = 2665). Welch t-test and Benjamini and Hochberg multiple testing correction with 1% false discovery rate were performed to determine the genes whose expression changed after luteinization in Lowess normalized and log transformed data. Fold changes in gene expression after luteinization were also determined using GeneSpring software.
Annotation of cDNA clones
A list of cDNA clones and corresponding GenBank Accession numbers are available on the project website (http://genome.rnet.missouri.edu/Swine). Each clone was previously submitted to GenBank and therefore has been clustered by the National Center for Biotechnology Information (NCBI, Bethesda, MD, USA) into the porcine UniGene database. The UniGene cluster and cluster annotation was used for preliminary assignment of name and function for each cDNA clone. The cDNA clones whose expression was found to be differentially expressed were further annotated through a manual annotation procedure. The sequence of individual clones was aligned (Blast analysis) against the RefSeq_rna database within NCBI (Wheeler et al. 2006). The highest homology alignments were examined for alignment score, the sense of the alignment (plus/plus vs plus/minus) and whether the alignment corresponded to the 3' tail of the RefSeq mRNA. The cDNA clones that we used were 3' cDNA (Jiang et al. 2004). Thus, clones with a high alignment score (generally >200) that were plus/minus and whose alignment was to the 3' tail of the RefSeq mRNA were annotated with the official gene name as found within RefSeq. The porcine RefSeq was preferentially used but, in the absence of porcine data, then other species were used (primarily human RefSeq). Clones whose alignments had low score (generally <200) or whose alignments appeared fragmented relative to the RefSeq alignment were annotated as unknown. Clones that were described as ‘weakly similar to’ or ‘moderately similar to’ a known gene was also considered unknown within the context of this work.
Real-time RT-PCR
Real-time RT-PCR was used to validate the microarray results. The test RNA and the control samples used in the microarrays were reverse transcribed using Superscript II reverse transcriptase (Invitrogen). The reactions were carried out using 5 µg total RNA, 3 µg random hexamer, and 1.25 µM oligo dT primers. Cytochrome P450 17A1 (CYP17A1), LH/choriogonadotropin receptor (LHCGR), 3ß-hydroxysteroid/
-5--4 isomerase (3B-HSD), plasminogen activator inhibitor-1 (PLANH1) and ribosomal protein L19 (RPL19) were amplified in an ABI Prism 7700 machine (Applied Biosystems, Foster City, CA, USA) using the QuantiTect Sybr Green PCR Kit (Qiagen) and specific PCR primers (Table 1
). The threshold cycle (CT) for the test samples was subtracted from the CT for the control sample to obtain the change (
) in CT (CT). The amount of each mRNA (relative to the control sample) was calculated by using the equation, relative expression = amplification efficiency CT. The amplification efficiencies (derived from an analysis of CT in a serially diluted samples; data not shown) were 2.2, 2.4, 2.2, 2.0, and 2.3 for CYP17A1, 3B-HSD, LHCGR, PLANH1, and RPL19 respectively.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
). Follicular fluid collected from follicles before estrus had nearly ten-fold greater follicular fluid estradiol concentrations when compared with follicular fluid from follicles collected on the second day of estrus (P<0.01). The relative concentration of steroids in follicular fluid was reversed for progesterone. Follicles collected before estrus had lower follicular fluid progesterone concentrations compared with follicles collected on day 2 of estrus (P = 0.10).
|
and 4. Relative to preovulatory estrogenic, microarray analyses detected 107 and 43 genes whose expression was decreased (Table 3) or increased (Table 4) respectively, in preovulatory luteinized follicles at a minimum threshold of P<0.01. For genes whose expression decreased (i.e. were high in preovulatory estrogenic and then low in luteinized), the predominant functional classifications were cytoskeletal proteins and regulators of cytoskeleton; nuclear proteins (chromatin components) and nucleic acid-binding proteins; metabolic enzymes, mitochondrial transporters, and proteins involved in the oxidative stress response; ligands, receptors, and receptor pathways (predominantly the cAMP response system); and cell proliferation/differentiation. For genes whose expression increased (i.e. were low in preovulatory estrogenic and high in luteinized), the predominant functional classifications were proteins that are involved in cell adhesion and migration, cell growth inhibition, or angiogenesis. There were functional classifications that were shared across both expression profiles (decreased or increased) but contained different genes (i.e. CYP17A1 and 3B-HSD) that were in the same functional class (steroidogenic enzymes), but expression was decreased and increased (respectively) in the analyses). Functional classifications shared across both expression profiles (Tables 3 and 4) contained genes whose proteins were involved in lipid transfer/steroidogenesis, proteolysis (proteases and protease inhibitors), and metabolism.
|
|
). Forty-one of the EST had a decrease in expression from preovulatory estrogenic to preovulatory luteinized (negative-fold change).
|
). The relative expression of RPL19 (a selected house-keeping gene (Lee et al. 2002) that was not differentially expressed in the microarrays) was similar for preovulatory estrogenic (1.2 ± 0.1-fold) and preovulatory luteinized (1.0 ± 0.1-fold).
|
). There were three genes (RPL13A, H3F3B, and TPM3) whose EST did not appear to be preferentially represented in the 6 and 8 mm follicle libraries from Jiang et al.(2004). Other than CO946724
[GenBank]
(CAPNS1; one in p2mm and four in p6mm), the genes that had increased expression in the preovulatory luteinized follicles had fewer than five members in the follicular libraries (p2mm, p4mm, p6mm, and p8mm) of Jiang et al.(2004).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
–5). The regulated genes fell into a variety of functional categories. Of interest was the fact that the functional categories were largely distinct for the different follicle classes (preovulatory estrogenic and -luteinized) owing to the unique transition that occurs during luteinization. Genes that were downregulated during luteinization (i.e. were high in preovulatory estrogenic and low in preovulatory luteinized) fell largely into the categories of cytoskeletal proteins or regulators of the cytoskeleton; nuclear proteins or nucleic acid-binding proteins; metabolic enzymes, mitochondrial proteins, and proteins involved in oxidative stress; ligands, receptors, and receptor pathways (predominantely cAMP); and cell proliferation/differentiation proteins. Genes whose expression increased after luteinization fell largely into the categories of cell adhesion/migration, cell growth inhibition, and angiogenesis. There were 58 EST that were differentially expressed, but gene function could not be categorized because it was unknown (Table 5).
The LH surge in pigs occurs near the onset of estrus (Esbenshade et al. 1982). Ovaries were collected shortly before estrus (preovulatory estrogenic) and on the second day of estrus (preovulatory luteinized; follicles presumably exposed to the LH surge). Collection of preovulatory luteinized follicles on the second day of estrus has been done in the past (Fricke et al. 1996) and this model was used in the present study. The status of individual follicles (estrogenic or luteinized) was verified by the analysis of estradiol and progesterone concentrations in the follicular fluid. As expected, the follicular fluid collected from follicles before estrus (preovulatory estrogenic) had high estradiol concentrations. Follicular fluid collected from follicles after the onset of estrus had relatively low estradiol concentrations and high progesterone concentrations (preovulatory luteinized). The follicular concentrations of estradiol and progesterone presented herein were similar to those reported in our previous publication on gene expression in preovulatory porcine follicles (Liu et al. 2000). Distinct regulation of genes involved in steroidogenesis were also observed and were predictable given the shift from estrogen synthesis to progesterone synthesis that occurred in the isolated follicles (Table 2). StAR protein, CYP17A1 (17
-hydroxylase/17,20 lyase) and P450 oxidoreductase (POR; P450 oxidoreductase, a flavo-protein that donates electrons to P450 enzymes) were elevated in preovulatory estrogenic follicles. The expression of NR5A2 (liver receptor homolog-1), a transcription factor that activates transcription of steroidogenic genes, including aromatase (Sirianni et al. 2002, Mendelson et al. 2005) was also greater in preovulatory estrogenic follicles. 3-ß-Hydroxysteroid dehydrogenase (3ß-HSD) increased in preovulatory luteinized follicles; a change in gene expression that is known to occur during luteinization (Conley et al. 1995, Driancourt et al. 1998). There was also a predictable change in INHBB, AKR1C3, and PTGES (inhibin, ß-B, prostaglandin F synthase, and prostaglandin E synthase; high in preovulatory follicles; Jiang et al. 2004). Previous studies of porcine ovary had found greater aromatase mRNA in follicles approaching estrus (Liu et al. 2000), but the arrays used in this study did not contain the gonadal isoform of aromatase (Corbin et al. 2004). Thus, the failure to detect a change in aromatase may reflect our error in not including the appropriate aromatase variant for ovary. Finally, dimeric dihydrodiol dehydrogenase (SUS2DD or AKR1C1) was differentially expressed in preovulatory estrogenic follicles. Dihydrodiol dehydrogenase metabolizes progesterone to the inactive 20-hydroxyprogesterone (Jez et al. 1997) demonstrating that progesterone may be metabolized to an inactive state in the preovulatory estrogenic follicle.
The phenotype of the preovulatory estrogenic follicle, as revealed by these analyses (Table 3), is one of a tissue whose cells are actively synthesizing cytoskeletal components, regulators of the cytoskeleton, and nuclear proteins (chromatin proteins, proteins associated with chromatin, and nucleic acid-binding proteins). Cytoskeletal and nuclear proteins are structural components of the cell and the high level of gene expression may reflect the proliferative nature of the cells in the preovulatory estrogenic follicle relative to the preovulatory luteinized follicle (Fricke et al. 1996). The high level of cytoskeletal protein gene expression may also be a consequence of gonadotropin stimulation because gonadotropins were shown to increase cytoskeletal gene expression in vitro (Grieshaber et al. 2003). Markers of proliferative cells whose expression was high in preovulatory estrogenic follicles included SIPA1, cyclin D2 (CCND2), and components of the Wnt/ß-catenin signaling pathway. The importance of cyclin D2 expression to granulosa cell function was unequivocally demonstrated by the CCND2 knockout mouse that had a defect in granulosa cell proliferation (Sicinski et al. 1996). The Wnt/ß-catenin signaling pathway controls additional aspects of granulosa cell proliferation. Misregulation of the Wnt/ß-catenin pathway leads to ovarian tumorigenesis (Boerboom et al. 2005) and knockout of pathway components leads to infertility in mice through a defect in corpus luteum development (Hsieh et al. 2005). Two different genes (YAP1 and WBP1) whose gene products bind Yes (a Src family nonreceptor tyrosine kinase proto-oncogene; Summy & Gallick 2003) were differentially expressed (high in preovulatory estrogenic follicles).
The second most striking feature of gene expression within the preovulatory estrogenic follicle was the highly metabolic nature of the cells and the evidence for an ongoing oxidative stress response. Three molecular transporters of the mitochondrion were differentially expressed including two adenine nucleotide transporters (SLC25A4 and SLC25A6). Citrate synthase (CS), a mitochondrial enzyme of the tricarboxylic acid cycle, was differentially expressed as well. Oxidative stress in preovulatory follicles is probably created by the highly metabolic environment within follicular cells and the generation of reduced cofactors for steroidogenesis (i.e. NADPH; Miller 2005). Alleviating oxidative stress in follicular cells is important for cell survival because the accumulation of reactive oxygen species causes apoptosis (Tsai-Turton & Luderer 2006). Relative to the preovulatory luteinized follicle, the preovulatory estrogenic follicle expressed a high level of superoxide dismutase 1 (SOD1) and peroxiredoxin 3 (PRDX3); two enzymes that combat oxidative stress (Chang et al. 2004). The SOD1 gene is required for ovarian function as SOD1 knockout mice are subfertile, but this could be related to low blood gonadotropin concentrations in the knockout model (Matzuk et al. 1998). Reduced glutathione is an antioxidant that relieves oxidative stress. It is synthesized within the follicle to act locally and is also transported into and stored within the oocyte (Luberda 2005). Two enzymes that lead to the synthesis of glutathione, glutamate-cysteine ligase catalytic subunit, (GCLC and esterase D, ESD), were differentially expressed (high in preovulatory estrogenic follicle). GCLC is perhaps the more noteworthy of the two because it is the rate-limiting enzyme in glutathione synthesis. We also detected high expression of microsomal glutathione S-transferase 1 (MGST1); an enzyme with the capacity to reduce oxidative stress, but is also involved in prostaglandin and leukotriene synthesis (Hayes et al. 2005).
The third feature of the gene-expression profile within the preovulatory estrogenic follicle (relative to preovulatory luteinized) was the expression of genes within the G-protein-coupled cAMP second messenger system. LH (as opposed to FSH) is the primary gonadotropin driving the terminal differentiation of preovulatory follicles in the pig (Guthrie 2005). Our previous studies demonstrated that preovulatory estrogenic follicles (6 mm) had high LH receptor (LHCGR) expression (Liu et al. 2000). A similar observation was made in this study (Table 3). In addition to elevated LHCGR, there was increased expression for several genes with the cAMP pathway in the preovulatory estrogenic follicle. One gene was structural (AKAP7; a scaffold protein that binds the protein kinase A (PKA) regulatory subunit; Malbon et al. 2004). Two of the identified genes were regulators of cAMP signaling with an inhibitory tone. For example, the protein product of RGS3 is a GTPase activator that is inhibitory toward G-protein signaling (Ishii & Kurachi 2003). The PRKAR2B gene encodes one of the regulatory subunits of the PKA complex. The PRKAR2B gene is an early response gene for gonadotropin-treated ovarian cells (Chu et al. 2002). The expression of PRKAR2B and RSG3 in preovulatory estrogenic follicles may imply a mechanism through which the preovulatory follicle restrains the cAMP signal and prevents premature luteinization.
The LH surge initiates a physiological program that leads to ovulation and the formation of the corpus luteum. Microarray analyses detected some of the underlying genome-level changes that support the transition within the follicle. Cytoskeletal, nuclear, metabolic, and proliferative genes had high expression in preovulatory estrogenic follicles and reduced expression in preovulatory luteinized follicles (see above). The genes that had high expression in the preovulatory luteinized follicles fell into different functional categories when compared with preovulatory estrogenic follicles. The functional categories of genes that were highly expressed in preovulatory luteinized follicles were reminiscent of the cellular processes that have been classically ascribed to follicular cells after the LH surge (Smith et al. 1994, Murphy et al. 2001).
The predominant functional category for increased gene expression in preovulatory luteinized follicles was cell adhesion and migration (Table 4). The category included cell surface antigens (CD9, CD24, LGALS3, and PTX3) as well as regulatory molecules that control cell migration (CAPNS1, FNBP1L, EFNA5, and NCK1). Of the cell surface molecules, CD24 and pentraxin 3 (PTX3) are particularly relevant to ovarian biology. CD24 is characteristic of migratory cells; and is a known marker for tumorigenesis in a variety of tissues (including ovary; Kristiansen et al. 2004). PTX3 is a secreted protein that binds extracellular matrix and is released in response to growth differentiation factor-9 (GDF-9). PTX3 expression is specifically increased after the LH surge (Varani et al. 2002). The PTX3 knockout mouse is subfertile because of defective ovulation and defective formation of the cumulus–oocyte complex (Varani et al. 2002).
Cells within the porcine follicle initiate migration and the remodeling of the follicle wall before ovulation (Murphy et al. 2001). The migratory nature of cells within the preovulatory luteinized follicle was clearly suggested by the increase in expression for calpain (CAPNS1; Franco & Huttenlocher 2005), ephrin A5 (EFNA5; Surawska et al. 2004), forming binding protein (FNBP1L; Watanabe & Higashida 2004), and NCK1 (Miyamoto et al. 2004); genes whose proteins have well-characterized roles in cell migration. The ATP1A1 gene (Na+/K+ transporter) is included in this category as well because of the recent work implicating the function of this gene in controlling cell movement (Rajasekaran et al. 2005).
The second largest class of differentially expressed genes in the luteinized preovulatory follicle was for cell growth inhibition (antiproliferation). The luteinized follicular tissue was presumably a mixture of granulosa (believed to be nondividing), theca (dividing), and endothelial (actively dividing) cells. Given the diversity of cell types and mitotic states, it was somewhat surprising that cell growth inhibitory molecules were detected in these studies. Nonetheless, several genes were identified and their existence suggests an active process through which some cells within the preovulatory follicle become nonproliferative. Perhaps, the most notable members of this class were protein phosphatase 2Aß-subunit (PPP2CB), a Ser/Thr phosphatase that inhibits cell growth by inactivating the MAPK signal (Theodosiou & Ashworth 2002), and PDCD4, a tumor suppressing gene that acts by suppresses carbonic anhydrase (Goke et al. 2004). The remaining members of the class have known tumor suppressor activities as well but do so through different mechanisms. Tristetraprolin (ZFP36L2) destabilizes growth factor mRNA and increases mRNA turnover (Stoecklin et al. 2003) and microseminal protein ß, MSMB or PSP94; one of the most highly expressed genes in porcine corpus luteum from our previous work; Jiang et al. 2004) is an inhibitor of vascular endothelial growth factor (VEGF) signaling (Lamy et al. 2005). Janus kinase 1 can have proliferative or antiproliferative properties depending on the nature the activating ligand (Verma et al. 2003).
The final functional class that was differentially expressed in preovulatory luteinized follicles was comprised of genes whose proteins play a role in angiogenesis. The process through which angiogenesis is initiated within the luteinized follicles has been described (Murphy et al. 2001) and is of interest because of the rapid development of a vascular bed in the growing corpus luteum (Driancourt et al. 1998.). Two of the enzymes in this group (MAOB and LAP3) are vasodilatory; monoamine oxidase metabolizes vasoactive amines to inactive forms and is found in high amounts within the preovulatory follicle and corpus luteum (Yoshimoto et al. 1986) and LAP3 (also known as cysteine amino peptidase or oxitocinase) metabolizes oxytocin, vasopressin, and angiotensin III (Tsujimoto & Hattori 2005) and is a marker for malignant gynecological tumors (Ino et al. 2004). The importance of VEGF in the angiogenic cascade that occurs in preovulatory luteinized follicles is well documented (Tamanini & De Ambrogi 2004) and DSCR1, a differentially expressed gene shown here, is an early response gene for VEGF (Iizuka et al. 2004). PLANH1, the primary inhibitor of plasminogen activator that has been implicated in tumor growth and cellular remodeling (Stefansson et al. 2003) and annexin a2 (ANXA2), a plasminogen–plasminogen activator co-receptor with angiogenic properties (Kim & Hajjar 2002), were also found to be differentially expressed. The upregulation of both PLANH1 and ANXA2 is curious given their contrasting functions within the plasminogen–plasminogen activator system that has functional roles in both ovulation and angiogenesis (Murdoch & Gottsch 2003). Their coexistence may simply reflect the delicate enzymatic balance that exists during cellular remodeling. The activin A receptor is included in this class because of the role of the transforming growth factor-ß superfamily in angiogenesis (Lebrin et al. 2005).
Analyses of gene expression by microarray typically yield a long list of genes with diverse functions that can present a confusing picture of cellular physiology. We have attempted to simplify the process by applying stringent statistical criteria to our analyses (diminishing the number of spurious results) and also by carefully evaluating each printed clone against a curated database (RefSeq of the NCBI). This process has hopefully created a robust list of genes whose expression changes during luteinization of the porcine follicle. Many of the discovered genes have recognized roles in the process, whereas others could be viewed as novel. We grouped genes into functional categories. These grouping could be criticized because genes can assume many biological roles. The categories that we developed were based on a literature search for each gene in this study. Two additional points should be raised. The first is that many of the genes that we identified here were also found to be highly expressed in preovulatory follicle libraries from our original publication (Jiang et al. 2004; Table 6). The two studies used entirely different approaches and samples and achieved a degree of commonality in their results. An obvious limitation was that we sequenced a limited number clones in the original study and therefore could not estimate the true frequency of EST for all clusters in the original libraries. The second point is that a related study in cattle yielded an expression profile that overlapped with ours. Ndiaye et al.(2005) examined gene expression within the bovine dominant follicle before and after human chorionic gonadotropin (hCG) treatment. They discovered 23 known genes that were decreased by hCG in their differential display analyses. Five of the genes that they described (CYP19A1, LRP8, CJA1, LHCGR, and JAK3) were identical to ours and two additional genes (INHBA and SERPINE2) were very closely related. Thus, approximately one-third of the known genes discovered by Ndiaye et al.(2005) were identical or highly similar to ours. The results of a related study in mice (McRae et al. 2005) in which serial analyses of gene expression (SAGE) was used to study changes in granulosa gene expression were less consistent with ours. The mouse study employed pregnant mare serum gonadotropin followed by a luteinizing dose of hCG. Twelve of the reported genes (STAR, TPM1, ACTA2, FADS2, INHBB, LHCGR, GJA1, SYND1, CTSL2, 3B-HSD, ANXA2, and LRP8) can be found in our lists, but only three of the genes (FADS2, INHBB, and GJA1) undergo the same directional change within the two studies. The significance of this observation depends on the technical (SAGE vs microarray) and biological (pig vs mouse; whole follicle vs granulosa cells) basis for the inconsistency. In all likelihood, none of the studies presents a complete list of genes that are differentially expressed during luteinization. The microarray technology that we used and the approaches used by others (differential display and SAGE) have their limitations. As for microarray, the cDNA are printed on glass and many genes are screened out because of low-intensity signals (about one-third in this study). The genes described herein had the qualities of relatively high signal intensity, low between animal and follicle variance, and high differential expression. Technology that employs greater precision and sensitivity will need to be developed to improve the gene-expression profile and increase our knowledge of luteinization.
Fifty-eight of the discovered genes were classified as EST or unique sequences (Table 5). Some of the genes were highly differentially expressed. For example, clones CO946569
[GenBank]
and CO946735
[GenBank]
had a fold change from preovulatory estrogenic to -luteinized of –18.0 and –8.6 respectively. These clones belong to contigs whose members were highly represented in the 6 and 8 mm follicle libraries from our original publication (Table 6; Jiang et al. 2004). The clones were clustered in the Porcine UniGene cluster of GenBank as Ssc.14541 (CYP19A3, aromatase) and Ssc.31012 (transcribed locus). The annotation associated with the first clone (aromatase) is a good example of the problems encountered with gene annotation in public databases. This clone and a large number of clones within Ssc.14541 have relatively low homology to porcine aromatase. We have attempted to avoid gene annotation problems in this study by individually examining each clone for homology to genes in the RefSeq database. The list of genes with unknown function that we present (Table 5) may contain genes whose functions fit into the categories described in this paper. Their protein products need to be identified and characterized.
In summary, genes with differential expression before and after luteinization were discovered by using a microarray approach. Some of the differentially expressed genes had functions that were already known and their regulation was consistent with the published literature. Most of the discovered genes, however, have never been previously associated with luteinization. Cells within preovulatory estrogenic follicles had a gene-expression profile that was distinctly different from cells within preovulatory luteinized follicles. Some of the differentially expressed genes in this study may be novel candidate genes for proteins that function in cell proliferation or cell growth inhibition, oxidative stress, cell migration, and early stages of angiogenesis.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Berisha B & Schams D 2005 Ovarian function in ruminants. Domestic Animal Endocrinology 29 305–317.[CrossRef][Web of Science][Medline]
Boerboom D, Paquet M, Hsieh M, Liu J, Jamin SP, Behringer RR et al. 2005 Misregulated Wnt/beta-catenin signaling leads to ovarian granulosa cell tumor development. Cancer Research 65 9206–9215.
Chang TS, Cho CS, Park S, Yu S, Kang SW & Rhee SG 2004 Peroxiredoxin III, a mitochondrion-specific peroxidase, regulates apoptotic signaling by mitochondria. Journal of Biological Chemistry 279 41975–41984.
Chu S, Rushdi S, Zumpe ET, Mamers P, Healy DL, Jobling T, Burger HG & Fuller PJ 2002 FSH-regulated gene expression profiles in ovarian tumours and normal ovaries. Molecular Human Reproduction 8 426–433.
Conley AJ, Kaminski MA, Dubowsky SA, Jablonka-Shariff A, Redmer DA & Reynolds LP 1995 Immunohistochemical localization of 3 beta-hydroxysteroid dehydrogenase and P450 17 alpha-hydroxylase during follicular and luteal development in pigs sheep, and cows. Biology of Reproduction 52 1081–1094.[Abstract]
Corbin CJ, Mapes SM, Marcos J, Shackleton CH, Morrow D, Safe S, Wise T, Ford JJ & Conley AJ 2004 Paralogues of porcine aromatase cytochrome P450: a novel hydroxylase activity is associated with the survival of a duplicated gene. Endocrinology 145 2157–2164.
Driancourt MA, Quesnel H, Meduri G, Prunier A & Hermier D 1998 Luteinization and proteolysis in ovarian follicles of Meishan and Large White gilts during the preovulatory period. Journal of Reproduction and Fertility 114 287–297.
Eisen MB & Brown PO 1999 DNA arrays for analysis of gene expression. Methods in Enzymology 303 179–205.[Web of Science][Medline]
Esbenshade KL, Paterson AM, Cantley TC & Day BN 1982 Changes in plasma hormone concentrations associated with the onset of puberty in the gilt. Journal of Animal Science 54 320–324.
Evans AC 2003 Characteristics of ovarian follicle development in domestic animals. Reproduction in Domestic Animals 38 240–246.[CrossRef][Web of Science][Medline]
Fortune JE, Rivera GM & Yang MY 2004 Follicular development: the role of the follicular microenvironment in selection of the dominant follicle. Animal Reproduction Science 82–83 109–126.
Franco SJ & Huttenlocher A 2005 Regulating cell migration: calpains make the cut. Journal of Cell Science 118 3829–3838.
Fricke PM, Ford JJ, Reynolds LP & Redmer DA 1996 Growth and cellular proliferation of antral follicles throughout the follicular phase of the estrous cycle in Meishan gilts. Biology of Reproduction 54 879–887.[Abstract]
Goke R, Gregel C, Goke A, Arnold R, Schmidt H & Lankat-Buttgereit B 2004 Programmed cell death protein 4 (PDCD4) acts as a tumor suppressor in neuroendocrine tumor cells. Annals of the New York Academy of Sciences 1014 220–221.[CrossRef][Web of Science][Medline]
Grieshaber NA, Ko C, Grieshaber SS, Ji I & Ji TH 2003 Follicle-stimulating hormone-responsive cytoskeletal genes in rat granulosa cells: class I beta-tubulin, tropomyosin-4, and kinesin heavy chain. Endocrinology 144 29–39.
Guthrie HD 2005 The follicular phase in pigs: follicle populations, circulating hormones, follicle factors and oocytes. Journal of Animal Science 83 E79–E89.
Hayes JD, Flanagan JU & Jowsey IR 2005 Glutathione transferases. Annual Review of Pharmacology and Toxicology 45 51–88.[CrossRef][Web of Science][Medline]
Hegde P, Qi R, Abernathy R, Gay C, Dharap S, Gaspard R, Earle-Hughes J, Snesrud E, Lee NH & Quackenbush J 2000 A concise guide to cDNA microarray analysis. Biotechniques 29 548–562.[Web of Science][Medline]
Hsieh M, Boerboom D, Shimada M, Lo Y, Parlow AF, Luhmann UF, Berger W & Richards JS 2005 Mice null for Frizzled4 (Fzd4–/–) are infertile and exhibit impaired corpora lutea formation and function. Biology of Reproduction 73 1135–1146.
Hunter MG, Robinson RS, Mann GE & Webb R 2004 Endocrine and paracrine control of follicular development and ovulation rate in farm species. Animal Reproduction Science 82–83 461–477.
Iizuka M, Abe M, Shiiba K, Sasaki I & Sato Y 2004 Down syndrome candidate region 1, a downstream target of VEGF, participates in endothelial cell migration and angiogenesis. Journal of Vascular Research 41 334–344.[CrossRef][Web of Science][Medline]
Ino K, Shibata K, Kajiyama H, Kikkawa F & Mizutani S 2004 Regulatory role of membrane-bound peptidases in the progression of gynecologic malignancies. Biological Chemistry 385 683–690.[CrossRef][Web of Science][Medline]
Ishii M & Kurachi Y 2003 Physiological actions of regulators of G-protein signaling (RGS) proteins. Life Science 74 163–171.[CrossRef][Web of Science][Medline]
Jez JM, Flynn TG & Penning TM 1997 A new nomenclature for the aldo-keto reductase superfamily. Biochemical Pharmacology 54 639–647.[CrossRef][Web of Science][Medline]
Jiang H, Whitworth KM, Bivens N, Ries J, Woods R, Forrester LJ, Springer GK, Mathialagan N, Agca C, Prather RS & Lucy MC 2004 Large-scale generation and analysis of expressed sequence tags from porcine ovary. Biology of Reproduction 71 1991–2002.
Kim J & Hajjar KA 2002 Annexin II: a plasminogen-plasminogen activator co-receptor. Frontiers in Bioscience 7 d341–d348.[Web of Science][Medline]
Koopman P, Bullejos M, Loffler K & Bowles J 2002 Expression-based strategies for discovery of genes involved in testis and ovary development. Novartis Foundation Symposium 244 240–249.[Web of Science][Medline]
Kristiansen G, Sammar M & Altevogt P 2004 Tumour biological aspects of CD24, a mucin-like adhesion molecule. Journal of Molecular Histology 35 255–262.[CrossRef][Web of Science][Medline]
Lamy S, Ruiz MT, Wisniewski J, Garde S, Rabbani SA, Panchal C, Wu JJ & Annabi B 2005 A prostate secretory protein94-derived synthetic peptide PCK3145 inhibits VEGF signalling in endothelial cells: implication in tumor angiogenesis. International Journal of Cancer 118 2350–2358.[CrossRef][Web of Science]
Lebrin F, Deckers M, Bertolino P & Ten Dijke P 2005 TGF-beta receptor function in the endothelium. Cardiovascular Research 65 599–608.
Lee PD, Sladek R, Greenwood CMT & Hudson TJ 2002 Control genes and variability: absence of ubiquitous reference transcripts in diverse mammalian expression studies. Genome Research 12 292–297.
Liu J, Koenigsfeld AT, Cantley TC, Boyd CK, Kobayashi Y & Lucy MC 2000 Growth and the initiation of steroidogenesis in porcine follicles are associated with unique patterns of gene expression for individual components of the ovarian insulin-like growth factor system. Biology of Reproduction 63 942–952.
Luberda Z 2005 The role of glutathione in mammalian gametes. Reproductive Biology 5 5–17.[Medline]
Lucy MC 2000 Regulation of ovarian follicular growth by somatotropin and insulin-like growth factors in cattle. Journal of Dairy Science 83 1635–1647.[Abstract]
Malbon CC, Tao J, Shumay E & Wang HY 2004 AKAP (A-kinase anchoring protein) domains: beads of structure-function on the necklace of G-protein signalling. Biochemical Society Transactions 32 861–864.[CrossRef][Web of Science][Medline]
Matzuk MM, Dionne L, Guo Q, Kumar TR & Lebovitz RM 1998 Ovarian function in superoxide dismutase 1 and 2 knockout mice. Endocrinology 139 4008–4011.
McRae RS, Johnston HM, Mihm M & O’Shaughnessy PJ 2005 Changes in mouse granulosa cell gene expression during early luteinization. Endocrinology 146 309–317.
Mendelson CR, Jiang B, Shelton JM, Richardson JA & Hinshelwood MM 2005 Transcriptional regulation of aromatase in placenta and ovary. Journal of Steroid Biochemistry and Molecular Biology 95 25–33.[CrossRef][Web of Science][Medline]
Miller WL 2005 Minireview: regulation of steroidogenesis by electron transfer. Endocrinology 146 2544–2550.
Miyamoto Y, Yamauchi J, Mizuno N & Itoh H 2004 The adaptor protein Nck1 mediates endothelin A receptor-regulated cell migration through the Cdc42-dependent c-Jun N-terminal kinase pathway. Journal of Biological Chemistry 279 34336–34342.
Murdoch WJ & Gottsch ML 2003 Proteolytic mechanisms in the ovulatory folliculo-luteal transformation. Connective Tissue Research 44 50–57.[Web of Science][Medline]
Murphy BD, Gevry N, Ruiz-Cortes T, Cote F, Downey BR & Sirois J 2001 Formation and early development of the corpus luteum in pigs. Reproduction Supplement 58 47–63.
Ndiaye K, Fayad T, Silversides DW, Sirois J & Lussier JG 2005 Identification of downregulated messenger RNAs in bovine granulosa cells of dominant follicles following stimulation with human chorionic gonadotropin. Biology of Reproduction 73 324–333.
Rajasekaran SA, Barwe SP & Rajasekaran AK 2005 Multiple functions of Na, K-ATPase in epithelial cells. Seminars in Nephrology 25 328–334.[CrossRef][Web of Science][Medline]
Richards JS, Russell DL, Ochsner S & Espey LL 2002a Ovulation: new dimensions and new regulators of the inflammatory-like response. Annual Reviews of Physiology 64 69–92.[CrossRef][Web of Science][Medline]
Richards JS, Russell DL, Ochsner S, Hsieh M, Doyle KH, Falender AE Lo YK & Sharma SC 2002b Novel signaling pathways that control ovarian follicular development, ovulation, and luteinization. Recent Progress in Hormone Research 57 195–220.
Sarit F, Ada D & Abraham A 2005 Ovarian transcriptomes as a tool for a global approach of genes modulated by gonadotropic hormones in human ovarian granulosa cells. Endocrine 26 259–265.[CrossRef][Web of Science][Medline]
Scheetz TE, Trivedi N, Roberts CA, Kucaba T, Berger B, Robinson NL Birkett CL, Gavin AJ, O’Leary B, Braun TA et al. 2003 ESTprep: preprocessing cDNA sequence reads. Bioinformatics 19 1318–1324.
Sicinski P, Donaher JL, Gene Y, Parker SB, Gardner H, Park MY, Robker RL, Richard JS, McGinnis LK, Biggers JD et al. 1996 Cyclin D2 is an FSH-responsive gene involved in gonadal cell proliferation and oncogenesis. Nature 384 470–474.[CrossRef][Medline]
Sirianni R, Seely JB, Attia G, Stocco DM, Carr BR, Pezzi V & Rainey WE 2002 Liver receptor homologue-1 is expressed in human steroidogenic tissues and activates transcription of genes encoding steroidogenic enzymes. Journal of Endocrinology 174 R13–R17.[Abstract]
Smith MF, McIntush EW & Smith GW 1994 Mechanisms associated with corpus luteum development. Journal of Animal Science 72 1857–1872.[Abstract]
Smith MF, Ricke WA, Bakke LJ, Dow MP & Smith GW 2002 Ovarian tissue remodeling: role of matrix metalloproteinases and their inhibitors. Molecular and Cellular Endocrinology 191 45–56.[CrossRef][Web of Science][Medline]
Stefansson S, McMahon GA, Petitclerc E & Lawrence DA 2003 Plasminogen activator inhibitor-1 in tumor growth, angiogenesis and vascular remodeling. Current Pharmaceutical Design 9 1545–1564.[CrossRef][Web of Science][Medline]
Stoecklin G, Gross B, Ming XF & Moroni C 2003 A novel mechanism of tumor suppression by destabilizing AU-rich growth factor mRNA. Oncogene 22 3554–3561.[CrossRef][Web of Science][Medline]
Summy JM & Gallick GE 2003 Src family kinases in tumor progression and metastasis. Cancer Metastasis Reviews 22 337–358.[CrossRef][Web of Science][Medline]
Surawska H, Ma PC & Salgia R 2004 The role of ephrins and Eph receptors in cancer. Cytokine & Growth Factor Reviews 15 419–433.[CrossRef][Web of Science][Medline]
Tamanini C & De Ambrogi M 2004 Angiogenesis in developing follicle and corpus luteum. Reproduction in Domestic Animals 39 206–216.[CrossRef][Web of Science][Medline]
Theodosiou A & Ashworth A 2002 MAP kinase phosphatases. Genome Biology 3 reviews 3009.1–3009.10.
Tsai-Turton M & Luderer U 2006 Opposing effects of glutathione depletion and FSH on reactive oxygen species and apoptosis in cultured preovulatory rat follicles. Endocrinology 147 1224–1236.
Tsujimoto M & Hattori A 2005 The oxytocinase subfamily of M1 aminopeptidases. Biochimica et Biophysica Acta 1751 9–18.[Medline]
Varani S, Elvin JA, Yan C, DeMayo J, DeMayo FJ, Horton HF, Byrne MC & Matzuk MM 2002 Knockout of pentraxin 3, a downstream target of growth differentiation factor-9, causes female subfertility. Molecular Endocrinology 16 1154–1167.
Verma A, Kambhampati S, Parmar S & Platanias LC 2003 Jak family of kinases in cancer. Cancer Metastasis Reviews 22 423–434.[CrossRef][Web of Science][Medline]
Watanabe N & Higashida C 2004 Formins: processive cappers of growing actin filaments. Experimental Cell Research 301 16–22.[CrossRef][Web of Science][Medline]
Webb R, Nicholas B, Gong JG, Campbell BK, Gutierrez CG, Garverick HA & Armstrong DG 2003 Mechanisms regulating follicular development and selection of the dominant follicle. Reproduction Supplement 61 71–90.
Wheeler DL, Barrett T, Benson DA, Bryant SH, Canese K, Chetvernin V, Church DM, DiCuccio M, Edgar R, Federhen S et al. 2006 Database resources of the National Center for Biotechnology Information. Nucleic Acids Research 34 D173–D180.
Whitworth KM, Agca C, Kim JG, Patel RV, Springer GK, Bivens NJ, Forrester LJ, Mathialagan N, Green JA & Prather RS 2005 Transcriptional profiling of pig embryogenesis by using a 15-K member unigene set specific for pig reproductive tissues and embryos. Biology of Reproduction 72 1437–1451.
Yoshimoto Y, Sakumoto T, Arai R, Miyake A, Kimura H, Aono T, Tanizawa O & Maeda T 1986 Monoamine oxidase in rat ovary during the estrous cycle. A histochemical study by a new coupled peroxidatic oxidation method. Endocrinology 119 1800–1804.
This article has been cited by other articles:
 |
|
 |
|
  C. Agca, M. C Lucy, and Y. Agca Gene expression profile of rat ovarian tissue following xenotransplantation into immune-deficient mice Reproduction, June 1, 2009; 137(6): 957 - 967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A Bonnet, K A Le Cao, M SanCristobal, F Benne, C Robert-Granie, G Law-So, S Fabre, P Besse, E De Billy, H Quesnel, et al. In vivo gene expression in granulosa cells during pig terminal follicular development Reproduction, August 1, 2008; 136(2): 211 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A Bonnet, R Dalbies-Tran, and M A Sirard Opportunities and challenges in applying genomics to the study of oogenesis and folliculogenesis in farm animals Reproduction, February 1, 2008; 135(2): 119 - 128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Davoli and S. Braglia Molecular approaches in pig breeding to improve meat quality Brief Funct Genomic Proteomic, January 21, 2008; (2008) elm036v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
