To our knowledge, this is the first report concerning the expression and localization of CRH, CRH-BP, CRH-R1 and CRH-R2 in human cervical tissue.
With respect to CRH, the mRNA levels in the cervix and in most samples from the corpus were below the level of detection, in agreement with earlier studies, in which much lower levels of CRH were found in the myometrium than in the placenta . Of course, we cannot totally exclude the presence of undetectably low levels of CRH mRNA in the cervix and this tissue did stain positively for the CRH protein in connection with our immunohistochemical analysis. However, it appears likely that this cervical CRH originates from the placenta, which is the main source of this protein in reproductive tissues.
In present investigation the levels of CRH-BP, CRH-R1 and CRH-R2 mRNA were found to be lower in the pregnant cervix and corpus than in the corresponding tissues from women who were not pregnant. Again, this finding is in line with earlier reports that CRH-R1 is down-regulated in the pregnant in comparison to the non-pregnant myometrium [28, 29]. However, there was a discrepancy between these two previous studies: Stevens and coworkers observed up-regulation of CRH-R1 at the time of labor , whereas, as was also the case in our study, Rodriguez-Linares et al found no such significant up-regulation . This discrepancy might be due to the use of different methodologies: Stevens et al employed semi-quantitative RT-PCR, Rodriguez-Linares and coworkers competitive RT-PCR and we used real-time RT-PCR, which allows more exact quantitation with high fidelity and identity of the product. The down-regulation of CRH-BP, CRH-R1 and CRH-R2 in the pregnant uterus could be related to the elevated levels of CRH in maternal plasma or, possibly, less CRH effect is needed in pregnancy than in a non-pregnant state. However, our immunohistochemical analysis revealed no clear reduction in the levels of the CRH-BP, CRH-R1 and CRH-R2 proteins in connection with pregnancy, which indicates that there may be no change in physiological function.
There has been much speculation concerning the role of CRH in connection to preterm labor. Here, we observed no differences between preterm and term laboring groups, but the level of CRH-R1 mRNA was significantly higher in preterm not in labor compared to term not in labor cervix. However, eight of our preterm patients received corticosteroid treatment and earlier studies have shown that administration of exogenous corticosteroid enhances the level of maternal plasma CRH, as well as the intensity of immunohistochemical staining for CRH in the placenta and fetal membranes [30, 31]. Thus, in order to ascertain whether this treatment influenced our findings, the data was re-analyzed following elimination of the women being treated. The only change was that there were no longer any significant differences between all respective preterm and term groups (data not shown), which indicates that the difference in cervical CRH-R1 mRNA levels between women undergoing preterm and term caesarean section prior to labor was due to corticosteroid treatment. It would have been preferable from a scientific point of view to conduct a similar study on a large number of patients not receiving corticosteroid treatment, but current obstetric practice in Sweden and other Western countries makes it extremely difficult to obtain samples from patients in preterm labor who have not received exogenous corticosteroids.
Since there appeared to be no essential differences between preterm and term groups, we decided to analyze the study groups irrespective of gestational age. This analysis revealed down-regulation of the levels of both CRH-R2 and CRH-BP mRNA in the cervix in connection with labor. This result, in combination with our previous findings that changes in cytokine and 15-hydroxyprostaglandin dehydrogenase levels occur in association with labor irrespective of the length of gestation [12, 32], indicate that the process of preterm cervical ripening is similar to the one that occurs at term. This decrease in cervical CRH-BP, both at the mRNA and protein levels at the time of labor is associated with a similar decrease in the maternal plasma , which means that more CRH is bioavailable in cervix and can act on the receptors there.
The down-regulation of CRH-R2 in the cervix and myometrium both at mRNA and protein levels at the time of labor is in some respects to the findings by Jirecek and coworkers that expression of the CRH-R2 protein in the myometrium is markedly lower in association with labor at term . However, these other investigators detected no differences in the levels of this protein between patients in preterm labor and not in labor, whereas we report here the same down-regulation tendency at mRNA levels irrespective of gestational age. On the other hand, we performed immunohistochemical analysis only in the term groups and have no data concerning levels of CRH-R2 protein in preterm groups. Similar down-regulation of this receptor has been described in animal studies  and this phenomenon may be important for progression of normal labor. Since urocortins are the main ligands for CRH-R2 and play a role in human reproduction , it would be important to look for changes in the levels of these hormones in the human cervix during pregnancy and labor.
The detection of substantial expression of the CRH and its receptor proteins, as well as receptor mRNA in the cervix suggests that the cervix may be a target for CRH action. The down-regulation of CRH-R2 and CRH-BP, together with the most intense immunohistochemical staining for CRH-R1 in the cervix at the time of labor indicate that CRH may be involved in cervical ripening. No investigations concerning a possible effect of this hormone on the cervix have been published, but there are several mechanisms by which CRH could be involved in the cervical ripening: by up-regulating expression of NOS , enhancing the production of PGE2 and PGF2α  or stimulating secretion of MMP-9 . However, there are controversial findings concerning the role of CRH in the release of cytokines in different tissues [39, 40].
Earlier studies from our group have identified fetal fibronectine , MMP-8  and Syndican-1 and -3 (unpublished data) in the cervical epithelium. Here, for the first time, we demonstrate the presence of CRH in the cervical epithelium with the highest levels at term not in labor. All these findings support the proposal that the cervical epithelium plays a significant role in the signaling processes involved in cervical ripening.