Intermedin in rat uterus: changes in gene expression and peptide levels across the estrous cycle and its effects on uterine contraction

Background The present study demonstrates the expression of intermedin (IMD) and its receptor components in the uterus of the female rat during the estrous cycle and its effect on uterine contraction. Methods The gene expression level of intermedin and its receptor components and the peptide level of intermedin were studied by real-time RT-PCR and enzyme immunoassay (EIA) respectively. The separation of precursor and mature IMD was studied by gel filtration chromatography and EIA. The localization of IMD in the uterus was investigated by immunohistochemistry. The effect of IMD on in vitro uterine contraction was studied by organ bath technique. Results Uterine mRNAs of Imd and its receptor components and IMD levels displayed cyclic changes across the estrous cycle. Imd mRNA level was the highest at proestrus while the IMD level was the highest at diestrus. IMD was found in the luminal and glandular epithelia and IMD treatment significantly reduced the amplitude and frequency of uterine contraction but not the basal tone. Both calcitonin gene-related peptide (CGRP) receptor antagonist hCGRP8-37 and adrenomedullin (ADM) receptor antagonist hADM22-52 partially abolished the inhibitory effect of IMD on uterine contraction while the specific IMD receptor antagonist hIMD17-47 completely blocked the actions. The enzyme inhibitors of NO (L-NAME) and PI3K (Wortmannin) pathways diminished the IMD effects on uterine contraction while the cAMP/PKA blocker, KT5720, had no effect, indicating an involvement of NO and PI3K/Akt but not PKA. Conclusions IMD and the gene expression of its receptor components are differentially regulated in the uterus during the estrous cycle and IMD inhibits uterine contraction by decreasing the amplitude and frequency.

Both uterine ADM expression [23,24] and the response of uterine contraction to ADM [25,26] have been investigated. Uterine endometrium has been shown to have more ADM than the myometrium in rats [23], and in humans [27]. Rat uterine endometrial ADM expression peaks at proestrus and estrus and is regulated by estrogen [23,24].
Our laboratory has previously investigated the expression and functions of ADM in the reproductive system [28][29][30][31][32][33][34][35][36][37][38]. In the female rat, the gene expression and peptide levels of ADM change during the estrous cycle [33]. ADM inhibits ovarian steroidogenesis [33], stimulates ciliary beating but inhibits muscle contraction in the oviduct [35], and plays important roles in pregnancy [34,36]. As ADM is known to inhibit uterine contraction [25,26], we have investigated the changes of IMD gene expression and peptide levels in the uterus during the estrous cycle and its effect on spontaneous uterine contraction to understand the possible roles of IMD in the uterus.

Animals
Female Sprague-Dawley (SD) rats (12-13 wk) were obtained from the Laboratory Animal Unit, LKS Faculty of Medicine, the University of Hong Kong. The rats were housed at a constant temperature and humidity, under a 12-h light-dark cycle (dark period 07:00h to 19:00h), with water and rat chow ad libitum. Vaginal smears were obtained daily from 36 rats, and 28 rats that showed a regular 4-day estrous cycle were included in this study. Staging of the cycle into estrus and diestrus was based on the cytology of the vaginal smear. All rats were killed between 10:00 a.m. and 12:00 noon and the uteri were collected and snap-frozen in liquid nitrogen for storage at −80°C until further analysis. Immature rats injected with gonadotropin were used for the contraction study (please see a later section). All procedures had been approved by the Committee on the Use of Live Animals for Teaching and Research, the University of Hong Kong.

RT-PCR of Imd, Crlr, and Ramps
Total RNA of the uterus was obtained by homogenization in TRIZOL reagent (Life Technologies, Carlsbad, CA, USA) using a polytron (Kinematica, Switzerland) [34] and subjected to RT-PCR. RNA samples (5 μg) were reverse transcribed into complementary DNA (cDNA) with the  Imd mRNA and IMD levels in the uterus across estrous cycle. n=6, 6, 7, and 9 for the four stages respectively. The mRNA levels were normalized to Actb mRNAs, and each value represents the mean ± SEM. * P<0.01.
SuperScript II reverse transcriptase (Life Technologies, Carlsbad, CA). The real time RT-PCR technique has been previously described [35]. Polymerase chain reactions (PCR) were conducted by an iCycler iQ real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA, USA) using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercule, CA). Three house-keeping genes (ribosomal protein L19, β-actin, and 18S ribosomal RNA) were tested and β-actin (Actb) was used as an internal standard based on its uniform expression across the groups. Standard curves for each primer pair were prepared by the serial dilution of cDNA to determine the PCR efficiency. The PCR efficiencies for Imd, Crlr, Ramp1, Ramp2, Ramp3 and Actb  Table 1). Melt curve analysis for each primer showed only one peak for each product. The identities of all the PCR products were confirmed by gene sequencing (Tech Dragon Limited, Hong Kong).

Measurement of IMD in the uterus
Each tissue sample (0.03 g) was homogenized in 3 ml 2N acetic acid (1 ml/0.01 g tissue, BDH Laboratory Supplies, Poole, England) and then boiled for 10 min. A 50-μl aliquot was taken for the protein assay and the remaining homogenate was centrifuged at 18600 X g for 20 min at 4°C (Sorvall SM 24; Thermo Fisher Scientific, Inc., Waltham, MA). The supernatants were all lyophilized and stored at −20°C until assay. The lyophilized tissue samples were reconstituted in 1X IMD assay buffer. IMD level was measured with an IMD (1-50) (human) EIA kit (Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA). The minimum detectable concentration was 0.26 ng/ml and the range was 0-100 ng/ml. The intra-assay and inter-assay coefficients of variation were <10% and <15% respectively. The amount of protein in each sample was measured with a protein assay reagent (BioRad, Hercules, CA, USA) spectrophotometrically at 595 nm (LKB Ultraspec II; Biochem, Berlin, Germany). The immunoreactive IMD was expressed as pg/mg protein.

Gel filtration chromatography of the uterus
The tissues were extracted with a polytron in 1N acetic acid (BHD Laboratory Supplies, Poole, England) on ice (see above). A 50-μl aliquot of the homogenate was stored at −20°C until protein assay. The lyophilized tissue samples were reconstituted in Milli-Q water and centrifuged at 13000 rpm for 20 min at 4°C. Glacial acetic acid (96%) (Sigma, St. Louis, MO, USA) was added to the supernatant to a final concentration of 1 N acetic acid. The samples (in 500 μl of 1 N acetic acid) were then loaded on a Bio-gel P30 (Bio-Rad, Hercules, CA, USA) column (0.9 X 60 cm) and the column was eluted with 1N acetic acid at a flow rate of 1 ml/10 min for a total of 400 min. One-millilitre fractions were lyophilized and measured for IMD immunoreactivities as stated before. The level of immunoreactive IMD was expressed in terms of pg/ml of fraction/mg protein.

Immunohistochemistry
To localize IMD in the uterus, a Vectastain ABC kit (Vector Laboratories, Burlingame, CA, USA) was used for the avidin-biotin histochemical staining procedure. The uterine tissues were fixed in neutral buffered formalin overnight. Paraffin-embedded sections of 5 μM   [25,26,35]. The entire uterus [25] was then tied, via silk threads, to a tissue holder in a 10-ml organ bath containing Kreb's solution aerated with a mixture of oxygen and carbon dioxide (95:5%) at a constant temperature of 37°C. The tissue holder was attached to a force transducer coupled to a graph recorder. As a pilot study using 1, 10 and 100 nM IMD indicated that the uterine preparation only responded to 100 nM IMD, the response to 100 nM IMD (human IMD 1-53 ) was studied after 45-min equilibration.  For the study on receptor antagonism, the uteri were preincubated with 1 μM hADM 22-52 (ADM receptor antagonist), hCGRP  (CGRP receptor antagonist), or hIMD 17-47 (IMD receptor antagonist) or the vehicle for 1h, before the addition of 100 nM IMD. For the signaling pathways, the uteri were preincubated with KT5720 (1 μM, protein kinase A (PKA) inhibitor), N-nitro-L-arginine methyl ester (L-NAME) (100 μM, nitric oxide (NO) synthase inhibitor), or Wortmannin (1 μM, serinethreonine kinase/phosphoinositide 3-kinase (Akt/PI3K) inhibitor) (all from Sigma Chemicals, St. Louise, MO, USA.) before IMD (100 nM) was added.

Statistical analysis
All the data were expressed as mean ± standard error of the mean (SEM), and statistical significance was assessed by one-way analysis of variance (ANOVA) followed by Student-Newman-Keuls (SNK) test for post hoc comparisons, with P<0.05 taken as significant.

IMD immunoreactivity and mRNA level of Imd
The levels of IMD and Imd mRNA in the uterus of cycling rats were estimated by IMD EIA and real-time RT-PCR respectively, and the results are shown in Figure 1. The Imd mRNA level at the proestrus stage was taken as 1 and it was higher than those at estrus and diestrus (P<0.01), with no difference between estrus and diestrus. The peptide level of IMD was higher at diestrus than those at proestrus and estrus (P<0.01).

mRNA expression of Crlr and Ramps in cycling rats
The Crlr, Ramp1, Ramp2, and Ramp3 mRNA levels estimated by real-time RT-PCR are shown in Figure 2. The Crlr mRNA level was significantly lower at estrus than that at proestrus (P<0.05) while the Ramp2 level was lower at metestrus than all the other stages (P<0.05). No changes were observed for Ramp1 and Ramp3 mRNA levels.

Gel filtration chromatographic analysis of the uterus
At both the estrous (n=5) and diestrous stages (n=5), two immunoreactive IMD peaks were observed on the gel filtration chromatograms at fractions 15 and 25. At estrus, there was a higher peak of IMD than the IMD precursor while at diestrus, a predominant peak of IMD precursor and a smaller peak of IMD were seen ( Figure 3). The IMD peak was much higher at estrus than at diestrus. Since the authentic IMD 1-53 , IMD 1-47 and IMD  were eluted at fractions 22, 25 and 27 respectively, the low molecular peak in the sample corresponded to IMD  .

Immunohistochemical study of IMD
Positive immunostaining was observed in the rat uterus at both estrus and diestrus in the luminal and glandular epithelial cells with similar intensities at the two stages ( Figure 4).

Effects of IMD and its receptor antagonists and inhibitors on uterine contraction
Representative tracings of the rat uteri treated with IMD, receptor antagonists and signaling pathways inhibitors are shown in Figure 5. IMD inhibited uterine contraction in both frequency and amplitude ( Figure 5A). Treatment of rat uteri with IMD significantly lowered the amplitude and frequency of contraction by 33.67 ±0.78% and 20.34 ± 1.33% respectively (P<0.05, Figure 6A and B) but had no effects on the basal tone (results not shown). Both hADM 22-52 and hCGRP 8-37 partially blocked the action of IMD on the amplitude ( Figure 5B and C) and the amplitude decreased only by 16.42 ± 2.01% and 20.27 ± 1.26% respectively (vs 33.67 ± 0.78% for IMD alone, P<0.05, Figure 6A). Only hADM  but not the hCGRP 8-37 partially blocked the IMD action on frequency and the decrease was only 12.42 ± 1.01% (vs 20.34 ± 1.33% for IMD alone, P<0.05, Figure 6B). hIMD  completely blocked the relaxation effect of IMD on both amplitude and frequency ( Figures 5D, 6A and B). The use of KT5720 did not alter the responses of the contraction amplitude and frequency to IMD ( Figure 5E) while the effects of IMD were partially inhibited by both L-NAME and Wortmannin ( Figure 5F and G). In the presence of L-NAME and Wortmannin IMD decreased the amplitude by 19.13 ± 1.27 and 23.28 ± 1.80% (vs 33.67 ± 0.78% for IMD alone, P<0.05, Figure 7A) and the frequency by 15.85 ± 2.2 and 16.92 ± 5.2% (vs 20.34 ± 1.33% for IMD alone, P<0.05, Figure 7B). uterine contraction for a better understanding of the possible roles of IMD in the female reproductive tract. IMD decreased uterine contraction, in line with the findings for ADM [25,26].Our study also confirmed that Crlr, Ramp1, Ramp2, and Ramp3 are expressed in the uterus [26], suggesting that both ADM and CGRP receptors are present in this organ. Uterine IMD pep tide level was higher at diestrus than at estrus with no change in gene expression. However, according to the gel filtration chromatogram, much of the IMD immunoreactivity at diestrus was due to the precursor and the amount of IMD 1-47 was actually lower than at estrus. The relative abundance of the precursor IMD at diestrus may be explained by the decreased processing of the precursor protein or the increased secretion of the active IMD. As the Imd mRNA level at diestrus was not higher than at estrus, an increase in synthesis with a concomitant increase in release appears to be less likely. Rather, the smaller IMD peak at diestrus may be due to the decrease in its formation from the precursor. Our results suggest there is a difference in the proteolytic processing of the IMD precursor molecule. Regardless of the cause, an increase in the IMD 1-47 at estrus would mean a greater IMD action compared with diestrus. This may be related to uterine contraction (the present study) or angiogenesis as intermedin is an angiogenic growth factor [40]. The significance of an increase of Imd mRNA at proestrus remains unclear.

Discussion
We were able to demonstrate for the first time that IMD inhibited spontaneous uterine contraction in rat by reducing the contraction amplitude and frequency by 33.67 and 20.34% respectively. The magnitude of inhibition was similar to that reported for ADM on galanininduced contraction [25], where the entire uterus was used, and was less than that reported for ADM on basal contraction [26], where uterine strips were used. This inhibitory effect of IMD was reversed by both ADM and CGRP receptor antagonists, as was previously reported for ADM [26]. However, there is one minor difference between our study and the study of Yanagita et al. [26] in the way to achieve synchronization of the physiological state in the female rat. We injected the immature female SD rats with PMSG while Yanagita et al. injected the mature female SD rats with estradiol, but both methods synchronized the rats to the estrous stage. This inhibitory effect was reported at estrus when the IMD 1-47 level was higher and may synergize with the inhibitory effect of ADM.
We have not studied the IMD effects in induced uterine contraction. In galanin-induced contraction in the rat, ADM inhibited contraction via CGRP receptor only [25]. CGRP inhibited galanin-induced [25] and substance P-induced contraction in the rat [41], and KCl-induced contraction in the human [42] and these effects were mediated by the CGRP receptor. In some of the CGRP actions on the uterus, the NO pathway was involved [43], but not in others [44].
In this study, both hADM  and hCGRP  partially blocked the inhibitory effect of IMD on rat uterine contraction while hIMD 17-47 exhibited complete inhibition. These results suggest that IMD modulate uterine contraction mostly by specific IMD receptor and partially by CGRP and ADM receptors. In addition, our study has shown that both NO and PI3K pathways are involved in IMD mediated uterine contraction. The use of L-NAME (NO blocker) and Wortmannin (PI3K blocker) significantly reduced the decreases in amplitude and frequency induced by IMD but KT5720 (PKA blocker) did not alter the IMD action. The L-NAME inhibition may be mediated by PKG, which is activated by cGMP in the NO pathway leading to the dephosphorylation of myosin light chain to relax the uterine smooth muscle [43]. Another NO-mediated IMD effect on contraction has been reported in the rat papillary muscle [45]. The PI3K pathway for inhibiting smooth muscle contraction has also been reported in gastrointestinal smooth muscles [46] and an involvement of PI3K pathway for an IMD effect can be found in endoplasmic reticulum stress [21]. The cAMP-PKA system is not involved as the use of PKA blocker had no effect on IMD actions although the elevation of cAMP production was one of the characteristic features in the early study of IMD [2].

Conclusions
In conclusion, IMD and the gene expression of its receptor components are differentially expressed in the uterus across the estrous cycle. IMD inhibits uterine contraction by decreasing the amplitude and frequency. This inhibitory effect at estrus may synergize with the inhibitory effect of ADM when the IMD 1-47 level was higher.