| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
* Department of Animal Science, University of Nebraska, Lincoln, Nebraska 68583
United States Department of Agriculture, Agricultural Research Service, Roman L. Hruska U.S. Meat Animal Research Center, Clay Center, Nebraska 68933
2 Corresponding author: Department of Animal Science, University of Nebraska, Lincoln, NE 68583.
E-mail: dpomp{at}unl.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
1.6 million data points with a mixed-model approach identified 88 and 74 unique probes, representing 71 and 59 unique genes, which are differentially expressed between lines in the ovary and ovarian follicles of different size classes, respectively. These findings indicate that long-term selection for components of litter size has caused significant changes in physiological control of the dynamics of follicular maturation. Genes involved with steroid synthesis, tissue remodeling, and apoptosis, in addition to several genes not previously associated with ovarian physiology or with unknown function, were found to be differentially expressed between lines. This study reveals many potential avenues of investigation for seeking new insights into ovarian physiology and the quantitative genetic control of reproduction.
Understanding physiological alterations underlying genetic change will reveal new information regarding biology of female mammalian reproduction and potentially lead to new approaches for trait improvement. Furthermore, evaluation of the physiological consequences of long-term genetic selection will provide unique insights into genetic architecture of complex quantitative traits such as reproduction. The UNL selection lines provide a unique genetic resource for carrying out such studies. Selection for components of litter size has produced an 50% difference in ovulation rate between the lines (index ovulates on average 6.7 more ova than control; JOHNSON et al. 1999).
A highly dynamic process involving recruitment, development, maturation, and atresia of antral follicles determines ovulation rate. The biological mechanisms regulating the dynamics of follicular development are known to be involved with hormonal feedback mechanisms between the hypothalamus, the anterior lobe of the pituitary gland, and the ovaries, which have also been shown to produce molecules with paracrine and autocrine functions (FOXCROFT and HUNTER 1985; FOXCROFT et al. 1989). Gonadotropins and local factors including steroids, growth factors, and other regulatory peptides are known to be involved in maturation of follicles. In spite of this accumulated knowledge, many of the specific mechanisms involved in this process remain to be elucidated in greater detail.
Several genetic approaches have been taken to attempt to identify the underlying factors responsible for the ovulatory advantage observed in the UNL index selection line (POMP et al. 2001). A study to detect quantitative trait loci (QTL) affecting ovulation rate using an F2 population derived from a cross between generations 10 of the index and control lines (CASSADY et al. 2001) did not yield insights with biological significance. Only one significant QTL for ovulation rate was detected, indicating that selection response was likely manifested by changes in many genes, each with potentially small effects. Furthermore, and similar to other such studies (ROHRER et al. 1999; WILKIE et al. 1999), the broad resolution of QTL location renders identification of the underlying genes an extremely difficult task. A study using a candidate gene approach also failed to reveal specific loci contributing to selection response (LINVILLE et al. 2001).
YEN (1999) proposed that the ovulatory advantage of the index line is a result of changes in dynamics of follicle maturation during the follicular phase of the estrous cycle. This initiated ddPCR-based identification of several genes within ovarian follicles and the anterior pituitary that are putatively expressed differentially between the selection lines (BERTANI et al. 2004; GLADNEY et al. 2004). Evaluation of expression differences of thousands of genes in parallel with cDNA microarrays presents a powerful new opportunity to merge genetic and physiological analyses and to uncover new biological connections between genes and biochemical pathways. In this article we report results of a study employing cDNA microarrays to identify genes whose ovarian expression has been changed as a result of long-term genetic selection for components of reproduction.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
analog (Lutalyse, Upjohn) on day 13 of the estrous cycle (day 0 is the first day of estrus) to induce luteal regression. Given that selection for ovulation rate and embryonic survival in the index line has potentially changed dynamics of follicle maturation during the follicular phase of the estrous cycle (YEN 1999), PGF2 was used to initiate and synchronize the follicular phase. Additional details regarding the PGF2 treatment have been published previously (BERTANI et al. 2004; GLADNEY et al. 2004).
Ovaries were harvested daily at 24-hr intervals by ovariectomy from 2–6 days after PGF2 injection (days 2, 3, 4, 5, and 6), weighed, and numbers of corpora albicantia (CA) counted to determine ovulation rate in the previous cycle. Tissue collection at intervals of <24 hr would have required greater precision in determination of the expected time of ovulation in gilts treated with PGF2. One randomly selected ovary from each animal was snap-frozen in LN2 and stored at –80°. From the remaining ovary, all follicles with diameter 
2 mm were dissected, measured, individually snap-frozen in LN2, and stored at –80°. Total RNA was isolated from ovaries and follicles with Trizol (Gibco Life Sciences, Gaithersburg, MD). Follicles were pooled within each animal during the extraction process on the basis of diameter classifications (YEN 1999): small follicle (SF, 2–2.9 mm), medium follicle (MF1, 3–4.9 mm; MF2, 5–6.9 mm), and large follicle (LF, 7 mm). RNA integrity was checked in ethidium bromide-stained agarose gels and quantity was evaluated by fluorometry.
cDNA clones and preparation of microarray probes:
A total of 3636 unique cDNA clones derived from a normalized ovarian follicle cDNA library (CAETANO et al. 2003) were used for making the microarray. In addition, 816 of these clones were duplicated within the array. cDNA-derived probes for porcine estrogen receptor, follistatin, steroidogenic acute regulatory protein, follicle-stimulating hormone receptor, inhibin ß(b)-subunit, and a set of 106 ovarian follicle cDNAs isolated in a differential display PCR study (GLADNEY et al. 2004) were also used. Eight commercially available clones (Incyte Genomics) containing intergenic regions from the yeast (Saccharomyces cerevisiae) genome were used in duplicate as hybridization and data analysis controls. Considering all the cDNA clones and positive and negative controls, a total of 4608 probes were prepared for printing. Inserts were PCR amplified with primers located on the cloning vector in 100-µl reactions. A small aliquot of each PCR reaction was evaluated in agarose gels for quality control. Inserts that failed to amplify (10%, data not shown) were not replaced. PCR products were precipitated with a solution of ethanol/NaAcetate at –20°, centrifuged, rinsed with 70% ethanol, dried at room temperature, and resuspended in 30 µl of printing buffer (3x SSC and 0.1% N-lauroylsarcosine).
Microarray printing:
Microarrays were printed with a GMS417 Arrayer (Genetic Microsystems) on poly-L-lysine-coated slides prepared locally with standard protocols at the University of Nebraska Medical Center Core Facility. Each slide contained a total of 9216 features. The 4608 probes were printed over four different fields (i.e., subarrays) and an exact replicate of these was printed adjacently on each slide in fields five through eight. Printed slides were exposed to UV light for cross-linking, blocked in a succinic anhydride/sodium borate solution, and excess DNA probe material was washed off in a water bath at 95°. Blocked slides were dehydrated in a bath of 100% ethanol, centrifuged, and stored in a dark, cool, and dry location until use.
RNA pooling and generation of cDNAs:
Ovarian RNA from two or three animals within selection line and treatment day was pooled. Similarly, pools of RNA were made from follicles of the same size class within selection line and treatment day. Pooling was used to enable a greater sampling of the populations and to optimally manage limited array and RNA resources. Polyadenylated RNAs complementary to probes on the microarray derived from yeast intergenic sequences (Yeast RNA Controls, Incyte Genomics) were added into each pool of RNA as positive controls. The Submicro Expression Array Detection Kit (Genisphere) was used for generating cDNAs from each RNA pool according to manufacturer's protocols.
Microarray hybridizations and scanning:
Equimolar amounts of cDNA samples from index and control line ovary pools, each collected on the same treatment day and labeled with Cy3 or Cy5, were cohybridized on the same microarray slide. Each hybridization was replicated three times, and three additional hybridizations were performed with a reversal of the Cy dyes. Thus, a total of six hybridizations were performed with each pair of pooled samples. Hybridizations were conducted overnight at 45° in humidified chambers, according to instructions provided by the Genisphere kit manufacturer. Following the appropriate posthybridization washes in SSC buffer, microarray slides were centrifuged for drying and scanned in a ScanArray5000 (GSI Lumonics) scanner with the following parameters: Cy5 [laser power, 85%; photo multiplier tube (PMT), 85%], Cy3 (laser power, 95%; PMT, 85%). This process was repeated with samples collected in each of the 5 treatment days. Similar procedures were used to compare RNA pools derived from ovarian follicles of the same size class, at each treatment day. A total of six hybridizations were performed with each of nine pairs of samples (day 2, SF, MF1; day 3, MF1, MF2; day 4, MF1, MF2; day 5, MF2, LF; and day 6, LF).
Cohybridization of cDNAs from pooled ovarian RNA labeled with Cy3 and Cy5:
Equal amounts of RNA from each of the 10 ovarian RNA pools (two lines x 5 treatment days) were mixed into a single pool. This RNA was subsequently split into two subsamples that were processed as two different treatment groups. A total of six hybridizations were performed with these samples, as in the other experiments, providing data for quality control testing of methods and analyses.
Microarray analysis:
Microarray images were analyzed with the Imagene4.2 computer package (Biodiscovery). Data generated from microarray spots with problems (i.e., dust, scratches, etc.) were excluded from the final data set. Statistical analysis was performed with a mixed-model ANOVA approach (WOLFINGER et al. 2001) using PROC MIXED from SAS (1996). An initial model was fitted to normalize systematic effects common to all probes in a particular microarray. Residuals from this analysis were used in a second model to estimate line effects on the relative RNA expression level for each individual probe on the array. Model 1 was
 | (1) |
is the stochastic error. L and D were fitted as fixed effects and A, AF, and AL were fitted as random effects.
 | (2) |
is the stochastic error. To minimize the rate of false positives, a Bonferroni correction of 0.1/3636 = 2.75E-5 was used to establish a threshold value of significance for the estimated line effects (L) to achieve an experimentwise false-positive rate of 0.1, considering a total of 3636 unique probes were used. The same approaches were used to analyze data obtained from follicle RNA samples.
Tertiary analysis of significant expression differences:
Tertiary analysis of the data was performed to identify associations between genes whose expression patterns had never been previously correlated. Data collected from probes for which statistically significant differences between lines in level of ovarian RNA expression were detected on at least one treatment day were used for clustering analysis with different methodologies provided by the software Cluster (EISEN et al. 1998). Multiple probes for the same gene were excluded from this analysis. Treeview (EISEN et al. 1998) was used to generate graphs of the data.
Confirmation of microarray results:
The same 10 pools of ovarian RNA evaluated in the microarray hybridizations were used for confirmation analysis with Northern hybridization. Expression levels were evaluated for the following probes (genes): UNL-P-FN-ay-a-05-0-UNL (cytochrome P450 side chain cleavage), UNL-P-FN-bq-f-06-0-UNL (calpain light subunit I), and UNL-P-FN-ct-e-01-0-UNL (no match). A probe for porcine glyceraldehyde 3-phosphate dehydrogenase was used to standardize gel-loading differences.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
|
RNA expression differences in follicular tissue:
Statistically significant ovarian-follicle expression differences between lines were observed for a total of 83 probes in one or more of the day/size treatment groups. Removal of duplicated probes resulted in a total of 74 unique differentially expressed probes, of which 32 were also differentially expressed in the ovary comparison (Table S2, published as supplemental data at http://www.genetics.org/supplemental/). These follicle-based probes represent a total of 59 unique genes, of which 51 have known function, 4 have no known matches in GenBank, and 4 are homologous to sequences of unknown function. The mean fold induction for unique probes with significant expression differences was 1.98 ± 0.05 with a range from 1.29 to 3.62.
Hybridization and data analysis controls:
No probes were differentially expressed after cohybridizing cDNAs derived from the same pool of ovarian RNA. Furthermore, none of the yeast positive controls were differentially expressed in any of the sample groups tested. Significant expression differences were not observed between replicated probes (Figure 1, A and B). Figure 1 also illustrates results obtained with probes that contained different regions of the same genes. We observed examples where no significant differences were found between probes (Figure 1C) as well as cases where significant differences arose according to treatment day (Figure 1D). The latter result suggests the presence of potential differences in RNA splicing between lines. Analysis of control probes provides strong validation for the experimental design and methods used in this study. Consistency observed for probes replicated within the array indicates that estimated differences in RNA expression between the lines are repeatable.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Several genes not previously associated with the processes of follicle selection, maturation, or ovulation were found to be differentially expressed in ovary and follicle pools of the UNL index and control selection lines. Moreover, 11 probes with blastn matches to gene sequences of unknown function, and 7 additional probes with no database matches, were also identified as being differentially expressed. These findings provide possibilities for identifying novel genes and mechanisms involved in controlling the processes of ovulatory follicle selection and maturation. Figure 2 shows a graphical representation of the hierarchical clustering analysis (EISEN et al. 1998) performed with the ovarian data and illustrates the utility of this study to identify associations between genes whose expression patterns had never been previously correlated. Figure 3 shows an example of a representative cluster node from the hierarchical clustering analysis containing probes with a similar pattern of expression that represent genes known to be directly involved with estrogen synthesis, genes never associated with this pathway, and an unknown gene. Future tertiary analysis, incorporating alternate clustering methodologies, may enable discovery of new relationships between genes that combine to orchestrate the process of ovarian follicle maturation and thus determine the phenotype of ovulation rate.
|
|
Selection for ovulation rate and embryo survival in the index line altered expression levels of genes associated with transport of cholesterol into the ovarian follicles and steroid production, processes known to change as follicles mature to ovulatory status (FOXCROFT and HUNTER 1985; LEO et al. 2001; ESPEY and RICHARDS 2002). Differences in gene expression were observed in follicles of similar size between index and control lines on all 5 days investigated, indicating that size was not definitive of physiological status. Collagen type I receptor (CD36 antigen-like 1/CD36L1/SBR1/HDL receptor) was overexpressed in index line ovaries at days 2, 4, and 6 (and in day 6 LF) while being underexpressed at day 3. Thus, the role of high-density lipoproteins (HDL) in steroidogenesis may be greater than previously assumed in that low-density lipoproteins (LDL) are regarded as the primary transporters of cholesterol into cells of species other than rodents (GRUMMER and CARROLL 1988). However, the role of LDL in steroidogenesis is not diminished by the current findings as greater expression of ApoER2 (LRP8/endocytosis of LDL) is seen in index line females in ovaries on days 4, 5, and 6 and in day 4 MF1 and day 6 LF.
Conversion of cholesterol to progesterone requires cytochrome P450 side chain cleavage enzyme (CP450SCC), steroidogenic acute regulatory protein (STAR), and 3-ß-hydroxysteroid dehydrogenase (3ßHSD). STAR was overexpressed in index ovaries at day 2 while being underexpressed at day 3 and in day 4 MF2. CP450SSC, cytochrome C oxidase (CCO), and 3ßHSD were all significantly overexpressed in the day 2 SF from index line gilts (Figure 4). Day 3 MF2, resulting from growth of the day 2 MF1, also overexpress CP450SCC in the index line. However, index ovaries at day 3 expressed CP450SCC, CCO, and 3ßHSD at lower levels than in control animals. These findings, in combination with data from YEN (1999), indicate that follicular maturation is delayed as a consequence of selection for ovulation rate and embryo survival in the index line.
|
-hydroxylase (CYP17), the enzyme responsible for androgen synthesis, was overexpressed in index gilts in day 3 MF2, day 4 MF1, and day 5 MF1. Cytochrome P450 aromatase (CYP19), the rate-limiting enzyme for conversion of androgen to estrogen in porcine follicles (CORBIN et al. 2003), was overexpressed in index line ovaries and MF1 on day 4. Although the concentration of plasma estradiol-17ß was found not to differ between the selection lines during the entire estrous cycle (MARISCAL et al. 1998), the observed differences in RNA concentration between lines for genes involved in the synthesis of estrogen support its autocrine and paracrine functions. Interestingly, no differential expression of hormone receptor genes was detected. It is possible that these transcripts are found in excess and thus do not represent good candidates for ovulation rate differences between selection lines. Tissue remodeling occurs as ovarian follicles undergo maturation or atresia. We identified 24 cases of differential gene expression in ovaries and/or follicles, in genes potentially associated with cellular proliferation and tissue remodeling, as a result of long-term index selection. Plasminogen activator inhibitor 1 (PAI1) is involved in the regulation of plasmin, whose activity in follicular fluid increases prior to rupture of ovulatory follicles (POLITIS et al. 1990). Furthermore, estrogen was found to inhibit PAI1 protein synthesis in oviductal cells (KOUBA et al. 2000). We observed that PAI1 was underexpressed in index ovaries at days 2 and 3 (Figure 4), but the negative relationship of PAI1 gene expression with estrogen synthesis (i.e., CYP19) was not strong. Differences between lines were noted only in ovaries and not in follicles, indicating that primary regulation of PAI1 may not be aligned with estrogen synthesis.
Some genes associated with tissue remodeling likely relate to macrophages that infiltrate the thecal layer as follicles progress to ovulatory status (WUTTKE et al. 1997) and to the immune system as ovulation is an inflammatory-like process (RICHARDS et al. 2002). We had previously found expression of calpain light chain (CLSI) to be downregulated in index follicles during the follicular phase (GLADNEY et al. 2004). Similar results were found in the current study in day 3 and 5 ovaries and day 6 LF. However, CLSI was overexpressed in index gilts in day 4 ovaries. Calpains are involved in apoptosis mechanisms (LU et al. 2002) and mutations of the calpain 10 gene (CAPN10) are associated with polycystic ovarian syndrome in humans (GONZALEZ et al. 2002). Current findings indicate that selection for high ovulation rate has altered RNA levels of CLSI and other genes associated with remodeling of ovarian tissue during the follicular phase as a correlated response. Further investigations are necessary to elucidate the biological mechanisms involved.
Results obtained for days 2, 3, 5, and 6 indicate that animals sampled in this study were representative of the index and control selection lines and that PGF2 treatment resulted in previously observed patterns of follicle development (YEN 1999). Gilts of the index line had more total ovarian follicles on all of the 5 days preceding ovulation than the control gilts did. However, index line gilts had a significantly lower than expected mean number of CA on day 4. In addition, these animals had a higher number of MF1, and consequently a lower number of MF2 follicles, than were anticipated for this treatment day. Therefore, results obtained for this treatment day may be potentially nonrepresentative of what would be expected for the true RNA expression differences between lines. One possible explanation is sampling error. However, follicles from these day 4 index gilts overexpressed genes associated with steroidogenesis and support the overall conclusion of a delay in follicular maturation in the index line. Potentially, these two day 4 index gilts possessed a desirable combination of alleles necessary to recruit more follicles into the MF1 pool, but lacked the alleles required for selection of greater numbers of follicles into the MF2 pool and subsequent ovulation. Collectively, gene expression in all index line gilts supports an altered pattern of steroidogenesis as a consequence of selection for increased ovulation rate.
The threshold level used to determine statistical significance for the estimated RNA expression differences between lines was highly conservative to minimize type I errors and to provide the primary findings from the study. However, this also results in a higher percentage of type II errors, which were likely further inflated by the fact that a percentage of probes from the ovarian follicle library (CAETANO et al. 2003) were not generated because of amplification failure. Further mining of the data set with relaxed significance threshold levels is likely to continue to uncover novel results regarding the genetic and physiological control of female reproduction in pigs. Methods such as the use of a false discovery rate (STOREY and TIBSHIRANI 2003) would enable flexibility and a more liberal criterion while still effectively avoiding many false positives.
Recent publications have highlighted the importance of technical and biological replications in the experimental design of expression profiling studies with cDNA microarrays (KERR and CHURCHILL 2001; DOBBIN et al. 2003; KERR 2003). The experimental design used in this study was chosen to maximize the use of both microarray and biological resources available. Pooling of RNA samples within experimental group was necessary to decrease the total number of microarray slides necessary while still allowing for the accurate estimates of the mean mRNA expression-level differences between genetic lines for the assayed genes/probes. As previously outlined (KERR 2003) this approach has been successfully used to produce significant findings (JIN et al. 2001) and is an acceptable alternative for initial groundwork as described in the present study. Ideally a larger number of animals per experimental group would have been used in a design that would allow for the estimation of the variance associated with the mean mRNA expression level for each of the genes/probes studied in each swine line, and such a design should be encouraged in future microarray analysis in livestock species.
Because genetic differences in ovulation rate and litter size between the index and control lines far exceed what might have occurred due to random genetic drift (JOHNSON et al. 1999), we consider gene expression changes in the index line to be primarily the result of the selection process. An observed difference in mRNA levels for an individual gene is potentially the direct result of selection, with the differential expression being a manifestation of allelic variation within the gene itself (i.e., a QTL). Conversely, differences in mRNA expression levels may result from the trans-interaction of one or more QTL with the particular gene under study. Identification of polymorphisms within differentially expressed genes, and subsequent testing for associations with reproductive phenotypes in segregating populations, will assist in classifying results into these categories. We are currently identifying SNPs within the genes found to be differentially expressed in this study and testing them for associations within the index and control selection lines as well as in other commercially relevant maternal lines of pigs. More powerfully, an integrated approach whereby gene expression measurements are measured in gene mapping populations and considered as phenotypes in a QTL analysis (SCHADT et al. 2003; POMP et al. 2004) could directly identify those genes within which polymorphisms regulate gene expression in cis and/or in trans.
It would be of interest to determine the ontogeny of gene expression changes caused by long-term selection. The average change detected for significantly different follicle/ovarian expression phenotypes in this study was approximately twofold, which would represent just below a 10% rate of change per generation when averaged across the 11 generations of selection that had taken place, on the basis of the index of ovulation rate and embryo survival, prior to the time of tissue collection (the subsequent 7 generations of selection were based on litter size, and ovulation rate likely did not change significantly during this period). However, in the absence of comparing the lines for each generation across an extended time period, such estimates are prone to many potential sources of error. For example, selection could have fixed particular alleles, and thus potentially the impact on gene expression changes resulting from the underlying locus, prior to the time of tissue collection, causing a downward bias in the estimate of gene expression change per generation of selection. Alternatively, genetic heterogeneity could exist within the selection lines, and the lack of significant biological replication in the present experimental design could have led to sampling error. Most importantly, gene expression phenotypes are likely quantitative traits in their own right, regulated by multiple loci in a complex and interactive manner (POMP et al. 2004). Thus, the measurement of expression at any given time point in the course of long-term selection should be viewed in terms of the particular culmination and interaction of many events taking place at, and not the summation of individual events leading up to, the specific time point of measurement.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
BERTANI, G. R., C. D. GLADNEY, R. K. JOHNSON and D. POMP, 2004 Evaluation of gene expression in pigs selected for enhanced reproduction using differential display PCR. II. Anterior pituitary. J. Anim. Sci. 82: 32–40.
BOLET, G., L. OLLIVIER and P. DANDO, 1989 Selection sur la prolificite chez le porc. I. Resultats d'une experience de selection sur onze generations. Genet. Sel. Evol. 21: 93–106.
CAETANO, A. R., R. K. JOHNSON and D. POMP, 2003 Generation and sequence characterization of a normalized cDNA library from swine ovarian follicles. Mamm. Genome 14: 65–70.[CrossRef][Medline]
CASSADY, J. P., R. K. JOHNSON, D. POMP, G. A. ROHRER, L. D. VAN VLECK et al., 2001 Identification of quantitative trait loci affecting reproduction in pigs. J. Anim. Sci. 79: 623–633.
CORBIN, C. J., F. M. MORAN, J. D. VIDAL, J. J. FORD, T. WISE et al., 2003 Biochemical assessment of limits to estrogen synthesis in porcine follicles. Biol. Reprod. 69: 390–397.
DOBBIN, K., J. H. SHIH and R. SIMON, 2003 Questions and answers on design of dual-label microarrays for identifying differentially expressed genes. J. Natl. Cancer Inst. 95: 1362–1369.
EISEN, M. B., P. T. SPELLMAN, P. O. BROWN and D. BOTSTEIN, 1998 Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA 95: 14863–14868.
ESPEY, L. L., and J. S. RICHARDS, 2002 Temporal and spatial patterns of ovarian gene transcription following an ovulatory dose of gonadotropin in the rat. Biol. Reprod. 67: 1662–1670.
FOXCROFT, G. R., and M. G. HUNTER, 1985 Basic physiology of follicular maturation in the pig. J. Reprod. Fertil. Suppl. 33: 1–19.[Medline]
FOXCROFT, G. R., M. G. HUNTER and S. A. GRANT, 1989 The physiology of follicular maturation in the pig. Acta Physiol. Pol. 40: 53–63.[Medline]
GLADNEY, C. D., G. R. BERTANI, R. K. JOHNSON and D. POMP, 2004 Evaluation of gene expression in pigs selected for enhanced reproduction using differential display PCR and human microarrays. I. Ovarian follicles. J. Anim. Sci. 82: 17–31.
GONZALEZ, A., E. ABRIL, A. ROCA, M. J. ARAGON, M. J. FIGUEROA et al., 2002 CAPN10 alleles are associated with polycystic ovary syndrome. J. Clin. Endocrinol. Metab. 87: 3971–3976.
GRUMMER, R. R., and D. J. CARROLL, 1988 A review of lipoprotein cholesterol metabolism: importance to ovarian function. J. Anim. Sci. 66: 3160–3173.
JANSEN, M., C. H. DE MOOR, J. S. SUSSENBACH and J. L. VAN DEN BRANDE, 1995 Translational control of gene expression. Pediatr. Res. 37: 681–686.[Medline]
JIN, W., R. M. RILEY, R. D. WOLFINGER, K. P. WHITE, G. PASSADOR-GURGEL et al., 2001 The contributions of sex, genotype and age to transcriptional variance in Drosophila melanogaster. Nat. Genet. 29: 389–395.[CrossRef][Medline]
JOHNSON, R. K., D. R. ZIMMERMAN and R. J. KITTOK, 1984 Selection for components of reproduction in swine. Livestock Prod. Sci. 11: 541–558.[CrossRef]
JOHNSON, R. K., M. K. NIELSEN and D. S. CASEY, 1999 Responses in ovulation rate, embryonal survival, and litter traits in swine to 14 generations of selection to increase litter size. J. Anim. Sci. 77: 541–557.
KERR, M. K., 2003 Experimental design to make the most of microarray studies. Methods Mol. Biol. 224: 137–147.[Medline]
KERR, M. K., and G. A. CHURCHILL, 2001 Experimental design for gene expression microarrays. Biostatistics 2: 183–201.[Abstract]
KOUBA, A. J., B. R. BURKHARDT, I. M. ALVAREZ, M. M. GOODENOW and W. C. BUHI, 2000 Oviductal plasminogen activator inhibitor-1 (PAI-1): mRNA, protein, and hormonal regulation during the estrous cycle and early pregnancy in the pig. Mol. Reprod. Dev. 56: 378–386.[CrossRef][Medline]
LAMBERSON, W. R., R. K. JOHNSON, D. R. ZIMMERMAN and T. E. LONG, 1991 Direct responses to selection for increased litter size, decreased age at puberty, or random selection following selection for ovulation rate in swine. J. Anim. Sci. 69: 3129–3143.[Abstract]
LEO, C. P., M. D. PISARSKA and A. HSUEH, 2001 DNA array analysis of changes in preovulatory gene expression in the rat ovary. Biol. Reprod. 65: 269–276.
LINVILLE, R. C., D. POMP, R. K. JOHNSON and M. F. ROTHSCHILD, 2001 Candidate gene analysis for loci affecting litter size and ovulation rate in swine. J. Anim. Sci. 79: 60–67.
LU, T., Y. XU, M. T. MERICLE and R. L. MELLGREN, 2002 Participation of the conventional calpains in apoptosis. Biochim. Biophys. Acta 1590: 16–26.[Medline]
MARISCAL, D. V., E. G. BERGFELD, A. S. CUPP, F. N. KOJIMA, K. E. FIKE et al., 1998 Concentrations of gonadotropins, estradiol and progesterone in sows selected on an index of ovulation rate and embryo survival. Anim. Reprod. Sci. 54: 31–43.[CrossRef][Medline]
OLLIVIER, L., and G. BOLET, 1982 Selection for prolificacy in the pig. Pig News Info. 3: 383–384.
POLITIS, I., A. SRIKANDAKUMAR, J. D. TURNER, B. K. TSANG, L. AINSWORTH et al., 1990 Changes in and partial identification of the plasminogen activator and plasminogen activator inhibitor systems during ovarian follicular maturation in the pig. Biol. Reprod. 43: 636–642.[Abstract]
POMP, D., A. R. CAETANO, G. R. BERTANI, C. D. GLADNEY and R. K. JOHNSON, 2001 Applying functional genomics research to the study of pig reproduction, pp. 277–292 in Control of Pig Reproduction VI, edited by R. D. GEISERT, H. NIEMANN and C. DOBERSKA. Society for Reproduction and Fertility, Cambridge, UK.
POMP, D., M. F. ALLAN and S. R. WESOLOWSKI, 2004 Quantitative genomics: exploring the genetic architecture of complex trait predisposition. J. Anim. Sci. 82(Suppl.): E300–E312.
RICHARDS, J. S., D. L. RUSSELL, S. O. OCHSNER and L. L. ESPEY, 2002 Ovulation: new dimensions and new regulators of the inflammatory-like response. Annu. Rev. Physiol. 64: 69–92.[CrossRef][Medline]
ROHRER, G. A., J. J. FORD, T. H. WISE, J. L. VALLET and R. K. CHRISTENSON, 1999 Identification of quantitative trait loci affecting female reproductive traits in a multigeneration Meishan-White composite swine population. J. Anim. Sci. 77: 1385–1391.
SAS, 1996 SAS User's Guide (Release 6.09). SAS Institute, Cary, NC.
SCHADT, E. E., S. A. MONKS, T. A. DRAKE, A. J. LUSIS, N. CHE et al., 2003 Genetics of gene expression surveyed in maize, mouse and man. Nature 422: 397–402.[CrossRef][Medline]
STATON, J. M., A. M. THOMSON and P. J LEEDMAN, 2000 Hormonal regulation of mRNA stability and RNA-protein interactions in the pituitary. Mol. Endocrinol. 25: 17–34.
STOREY, J. D., and R. TIBSHIRANI, 2003 Statistical significance for genomewide studies. Proc. Natl. Acad. Sci. USA 100: 9440–9445.
WILKIE, P. J., A. PASZEK, C. W. BEATTIE, L. L. ALEXANDER, M. B. WHEELER et al., 1999 A genomic scan of porcine reproductive traits reveals possible quantitative trait loci (QTLs) for number of corpora lutea. Mamm. Genome 10: 573–578.[CrossRef][Medline]
WILSON, K. F., and R. A. CERIONE, 2000 Signal transduction and post-transcriptional gene expression. Biol. Chem. 381: 357–365.[CrossRef][Medline]
WOLFINGER, R. D., G. GIBSON, E. D. WOLFINGER, L. BENNETT, H. HAMADEH et al., 2001 Assessing gene significance from cDNA microarray expression data via mixed models. J. Comput. Biol. 8: 625–637.[CrossRef][Medline]
WUTTKE, W., L. PITZEL, I. KNOKE, K. THEILING and H. JARRY, 1997 Immune-endocrine interactions affecting luteal function in pigs. J. Reprod. Fertil. Suppl. 52: 19–29.[Medline]
YEN, H. W., 1999 Follicular development, maturation and atresia during the estrous cycle in gilts expressing high and low ovulation rates. Ph.D. Dissertation, University of Nebraska, Lincoln, NE.
This article has been cited by other articles:
 |
|
 |
|
  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]
|
|
|
|
 |
|
 B. A. Freking, K. A. Leymaster, J. L. Vallet, and R. K. Christenson Number of fetuses and conceptus growth throughout gestation in lines of pigs selected for ovulation rate or uterine capacity J Anim Sci, September 1, 2007; 85(9): 2093 - 2103. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
