Our results demonstrate labor-associated changes in gene expression in three functionally important areas of the human uterus. The primary objective of this study was to identify novel-to-labor genes important for the process of parturition. The second main objective was to identify groups of genes with similar expression profiles in order to recognize those with common regulatory mechanisms. Rather than providing a list of genes, this results provides a map of gene interactions in labor. We postulated that the onset of labor is likely to be caused by a reduction in inhibitory and an increase in stimulatory processes, and our data support this theory; for example, we demonstrated that in the lower segment in labor expression of genes for the stimulatory tumor necrosis factor receptor is increased, whereas that of the relaxatory potassium channel is reduced.
The study was specifically designed to investigate gene expression in human labor because the mechanisms of labor vary between species. Previous gene array studies have documented changes in expression in a rodent model [
]. Such animal models are useful since variability is reduced because of the animals' similar genotypes and exposure to a controlled environment. Expression data from such studies can be compared and contrasted with those from human tissue, thus providing an insight into the similarities and differences between species. However, we consider that data from human studies are the most important for understanding human physiology.
Previous human gene array data [
], has marked differences in methodology from our study. Bethin and colleagues [
] determined the expression profile in human extracts obtained either preterm, prior to labor or preterm, and at term following the onset of labor. In contrast, we designed our study to specifically determine labor-associated alterations and to exclude the marked changes in expression at the end of pregnancy. A further difference in our study was that we analyzed human uterine samples from all three functionally distinct areas of the uterus in the same women. It is the cooperation of these components of the uterus (contraction of the fundus, relaxation of the lower segment, and dilation of the cervix) that result in the process of labor. Our study also differed in that the method of analysis and sample size enabled the individual variation between women to be taken into account—i.e., to preserve these characteristics samples were not pooled. This individual analysis enabled the expression of each gene to be identified in each sample. The genes were then grouped into clusters based on their similarity of expression across individual samples. This similarity of genes' expressions in different samples dramatically increases the power of the cluster analysis and is possible only because the individual sample characteristics are maintained. However, one limitation of an individual analysis is that individual variation in expression in human tissue is likely to be high, not only because there are marked genetic and environmental effects but also because the time to the onset of spontaneous labor in non-labor samples is not known.
Aguan and colleagues utilized a different experimental design and methodology to investigate gene expression in the lower uterine segment before and after the onset of labor [
]. The type of array, number of investigated genes (588), and normalization procedures make valid comparison with our study difficult. The studies differ also in how the fold change in the gene expression calculation was done. However, there are several consistent changes in gene expression. For example, we demonstrated a 91% decrease in G protein–coupled receptor 161 in lower-segment samples, which is consistent with the 84% reduction reported by Aguan et al. We also demonstrated consistent changes in guanine nucleotide binding protein alpha expression.
Chan and colleagues [
] studied uterine samples in labor using a subtractive hybridization technique. Although this study used a different technique from ours, and the number of genes upregulated in labor was small, their findings have shown a consistent with our results, significant increase in the expression of interleukin-8.
Gene array data provide a wealth of information, which presents unique analytical challenges. We determined expression in six samples taken before and seven after the onset of labor at term. In order to compare the differences in samples taken before and after labor, the
-test for the difference in expression was performed and the
value was calculated. The genes were ranked according to this value, not to determine significance, which would be inappropriate for this number of comparisons, but to determine genes that demonstrated the greatest and most consistent change in labor. We did not correct for multiple comparisons since the expression of the different genes is not independently regulated. This method of analysis is likely to provide more consistent data than techniques using fewer samples, duplicate arrays on the same samples, or identification of an arbitrary change in expression [
]. By this method, any difference in expression during labor of those genes with the smallest
values is unlikely to have arisen by chance due to observer and instrument variability. Hence, the genes with the most consistent change in expression during labor are most likely to have an important function, although current methodology does not allow primary changes in expression to be distinguished from those secondary to increased contractility.
Although we cannot exclude false positives and negatives, the lower the
p-value the smaller the probability of a false positive results. However, as the number of genes selected increases, so does the chance of inclusion of a false positive result, while the chance of a false negative one decreases. False negative results may also occur due to wide inter-patient variability. There are many potential causes of inter-patient variability. Particularly important is that it cannot be determined in non-laboring patients when parturition would otherwise commence—that is, how close to the onset of labor a non-laboring patient is.
It is likely that many changes in gene expression precede the clinical signs of labor: for example, the steroid hormones estrogen and progesterone are fundamentally important for the maintenance of pregnancy and the onset of labor [
]. In some species, such as the sheep, pregnancy is maintained by progesterone and labor is caused by a dramatic fall in progesterone. The decrease in progesterone concentration increases the estrogen/progesterone ratio leading to contractions [
]. A fall in plasma progesterone has not been demonstrated in women, although administration of antiprogestins can induce labor [
]. This suggests that the mechanism may be slightly different in women.
Our demonstration of a fall in the expression of a modulator of estrogen receptor activity provides a mechanism whereby the functional estrogen/progesterone ratio could be increased without a change in plasma concentration of either. REA is a protein that competitively and selectively binds to the nuclear receptor reducing its function [
]. Although identified in breast cancer and placental cells, this modulator has not been described in the human myometrium. REA and RXR (which also inhibits estrogen activity) were both clustered into one group based on their decreased expression pattern in labor. The expression of prothymosin alpha (an antagonist of REA) is unchanged in labor and further supports this hypothesis. Jointly, they demonstrate existence of a pathway that may represent a novel mechanism of uterine control [
During labor there are concomitant physiological changes in the fundus, lower uterine segment, and cervix. The fundus generates coordinated forceful uterine contractions while the contractile lower segment elongates over the presenting part. The cervix undergoes softening in late pregnancy with a dramatic shortening and dilation during labor. Our data demonstrate related marked spatial differences in gene expression, consistent with previous publications using alternative techniques for quantification [
]. Some of these differences in gene expression may, however, be due to cell type. We have previously demonstrated that more than 98% of cells in our lower segment biopsies are myometrial [
], and fundal samples were taken from the peritoneal (outer) surface to prevent decidual contamination. It is therefore unlikely that changes in gene expression in the fundus and lower segment were derived from non-myometrial cells. In contrast, the cellular composition of the cervix is more heterogeneous, and expression within the different cell types cannot be discerned. Nevertheless, we considered that maintenance of the physiological cellular environment was more important than a homogenous cell population.
Oxytocin and prostaglandins are known to have a fundamental role in human parturition [
]. Our gene array data are consistent with existing evidence on these oxytocics. Oxytocin is produced by the choriodecidua during human labor [
] and acts on myometrial oxytocin receptors to cause contraction. Since oxytocin is not produced in the myometrium but in other gestational tissues, it is reassuring that there was no increase in myometrial expression of oxytocin in our study. The increase in myometrial oxytocin receptor formation precedes the onset of labor, and uterine expression increases from mid-pregnancy to term rather than at the onset of labor [
]. Consistent with these data, we did not demonstrate an increase in oxytocin expression in labor. In contrast, we have previously demonstrated that expression of myometrial secretory phospholipase A2 is increased in samples taken after the onset of labor [
]. This enzyme catalyzes mobilization of arachidonic acid from membrane phospholipids for the synthesis of prostaglandins. Our gene array results confirm an increase in secretory phospholipase A2 expression in myometrial samples taken after the onset of labor and are consistent with a regulatory role for prostaglandins
There is no generally accepted statistical method to analyze differences in gene expressions between groups, due to correlation of expressions of individual genes. To validate our findings, we confirmed expression of genes with different technique and demonstrated functional relationship of co-expressed genes. We validated a proportion of our microarray findings by RT-PCR, and we were reassured that the results using both techniques were consistent among all tested genes. We also analyzed the patterns of expression by two techniques: K-means and hierarchical clustering. Although these techniques may not be completely independent, the method of gene clustering is different and hence the combination provides additional confidence for the identification of networks of co-regulated genes. Prior studies have shown that co-expressed genes have been demonstrated to be functionally related and to participate in common biological processes defined by the Gene Ontology database. These relationships are identified across species and functional categories [
]. The identification within each cluster of genes with similar functions pertinent to labor strengthens our hypothesis that these genes are co-regulated. It is likely that expression of a particular gene can regulate expression of a second, which may itself influence a third. In this way, a single controlling mechanism may induce a multitude of phenotypic alterations leading to a change in function. Furthermore, transcription regulating factors (such as
CAAT enhancer binding protein, CEBP, which is increased after labor in the lower segment) may promote transcription for numerous contraction-associated genes. Indeed, CEBP binding to the oxytocin receptor promoter has recently been demonstrated [
]. We anticipate that elucidating the networks of genes associated with labor will enable a more holistic understanding of the process, leading to more rational methods for manipulating uterine contraction.
In summary, we have demonstrated consistent changes in gene expression in the human lower segment, fundus, and cervix in association with labor. A number of novel-to-labor genes have been identified in addition to networks of co-expressed genes. There are marked tissue and spatial differences in gene expression in the uterus during parturition.