Attenuation of Calcitonin Gene Expression in Pregnant Rat Uterus Leads to a Block in Embryonic Implantation^*

Abstract

The peptide hormone calcitonin plays a key role in calcium homeostasis in many tissues, such as bone and kidney. Our previous studies revealed that the expression of calcitonin is dramatically induced in the glandular epithelium of rat uterus between days 3–5 of pregnancy before the onset of blastocyst implantation on day 5. Calcitonin expression is switched off once implantation has progressed to day 6. The coincidence in timing suggested that calcitonin may function as a regulatory signal in the uterus during the early events leading to implantation. Here we report that the implantation stage-specific expression of calcitonin can be specifically attenuated by administering antisense oligodeoxynucleotides (ODNs) directed against exon IV of calcitonin messenger RNA into the uterine horns on day 2 of gestation. The loss of calcitonin messenger RNA and protein expression upon antisense ODN treatment is accompanied by a severe impairment in implantation of embryos. Based on the observations that 1) treatment with two different antisense ODNs possessing different base compositions produced similar phenotypes; and 2) treatment with the complementary sense ODNs did not affect either calcitonin expression or implantation, we conclude that the effects of antisense ODNs on calcitonin expression and implantation are specific and functionally linked. Our study strengthens the hypothesis that a transient expression of calcitonin in the preimplantation phase uterus is critical for blastocyst implantation.

IMPLANTATION, the initiation of the complex interaction between the embryo and the uterus, is a crucial event in mammalian embryonic growth and development (1–3). The initial adherence of the blastocyst to the uterine surface epithelium is followed by intimate interaction of the blastocyst trophectoderm with epithelial cells, which leads to the progressive phases of implantation. It is generally believed that the endometrium and the fertilized ovum must both undergo concomitant developmental changes for implantation to succeed. To investigate the molecular basis of implantation, we sought to identify the genes whose expression in the endometrium temporally coincides with the onset of implantation. By employing a gene expression technique designed by Wang and Brown (4), we isolated a number of putative implantation stage-specific genes. One of these genes was identified by DNA sequence analysis as that of calcitonin (5).

Calcitonin, a 32-amino acid peptide hormone, has long been known to be synthesized and secreted primarily by the parafollicular C cells of the thyroid gland (6–9). Its most well characterized physiological role is to regulate calcium levels in bone and kidney cells (6–9). Our previous studies revealed that calcitonin synthesis is markedly induced in the pregnant rat uterus during implantation (5). The level of uterine calcitonin messenger RNA (mRNA) in cycling rats is low, i.e. less than 1% of the calcitonin mRNA present in the thyroid gland, the prinicipal site of calcitonin synthesis. The expression of calcitonin mRNA and protein rises dramatically after day 2 (postfertilization) of gestation and reaches a peak on day 4, the day before implantation. At this stage, the level of calcitonin within the uterus reaches about 10–20% of that synthesized by the thyroid gland. After day 5, the day implantation occurs, the expression of the gene starts to decline, and by day 6, when implantation is completed, the calcitonin level falls to below detection limits. The burst of calcitonin expression at the time of implantation is localized in the glandular epithelial cells of the endometrium by immunohistochemical analysis (5). The timing and location of its synthesis in the glandular epithelium of rat endometrium prompted us to speculate that calcitonin may regulate blastocyst implantation in an autocrine or paracrine manner.

A timely interplay of the maternal steroid hormones, estrogen and progesterone, is believed to orchestrate the pronounced morphological and biochemical alterations in the endometrium that prepare it to be receptive to the developing blastocyst (1–3). Our previous studies demonstrated that the expression of calcitonin in the endometrium is induced by progesterone (5). Although progesterone profoundly influences uterine functions during pregnancy, calcitonin is one of the few genes that have been identified as regulated by progesterone in the pregnant uterus (10–13). Uterine expression of calcitonin could, therefore, be one of the links between steroid hormone action in the target cell nuclei and embryo-endometrial interactions during blastocyst implan-tation.

To gain insights into the functional role of calcitonin during implantation, we have employed an antisense oligodeoxynucleotide (ODN) methodology. The sequence-specific inhibition of gene expression by antisense ODNs relies on the ability of an ODN to bind specifically and efficiently to a complementary mRNA sequence (14, 15). Hybridization with antisense ODNs results in suppression of target mRNA levels by triggering degradation of the RNA strand of the RNA-DNA duplex. In addition, the ODNs are thought to prevent translation of target mRNAs (14, 15). Our approach involves suppression of calcitonin gene expression in the preimplantation phase uterus by using antisense ODNs targeted against calcitonin mRNA. Although there is mounting evidence from many laboratories that antisense oligomers targeted to cellular or viral mRNA sequences can produce specific biological effects in cultured cells or in intact tissue (14–27), we report here, for the first time, that antisense ODNs can be successfully employed to regulate gene expression in the uterus. Our study shows that the administration of antisense calcitonin ODNs into the preimplantation phase uterus effectively reduced calcitonin mRNA and protein synthesis without affecting the expression of nontarget genes. Interestingly, the loss of calcitonin expression upon antisense ODN treatment was accompanied by a severe impairment in embryonic implantation. Based on these results, we propose that calcitonin plays a crucial role in the uterus during blastocyst implantation.

Materials and Methods

Synthesis and sequence of ODNs

The ODNs containing phosphorothioate linkages in all positions and C5-propynyl modifications at the uridine and cytidine residues were synthesized at Rockefeller University’s Protein/DNA Technology Center and purified by reverse phase HPLC. The sequences of the first set of ODNs (19 nucleotides in length) were: sense ODN-1, 5′-TCTGAGTACCTGCATGCTG-3′; and antisense ODN-1, 5′-CAGCATGCAGGTACTCAGA-3′. The sequences of the second set of ODNs (22 nucleotides in length) were: sense ODN-2, 5′-CCAAGGACTTGGAGACAA-ACCA-3′; and antisense ODN-2, 5′-TGGTTTGTCTCCAAGTCCTTGG-3′. The antisense ODN-1 and -2 were complementary to bases 2615–2633 and 2720–2741, respectively, within exon IV of the rat calcitonin gene.

Animals

All experiments involving animals were conducted according to NIH Guidelines for the Care and Use of Experimental Animals. Virgin female rats (Sprague-Dawley, from Charles River, Wilmington, MA; 60–75 days of age) in proestrus were mated with adult males. The different stages of the cycle in the nonpregnant rats were ascertained by examining vaginal smears. The presence of a vaginal plug after mating was designated day 1 of pregnancy. In certain experiments, animals were ovariectomized and 14 days later were injected sc with estradiol (2μ g/kg BW). This was followed by an injection of progesterone (40 mg/kg BW) 24 h later.

Treatment of animals with ODNs

Sprague-Dawley rats were deeply anesthetized, and an incision was made in the lower abdomen. The ODNs were mixed with DOTAP (Boehringer Mannheim, Indianapolis, IN) and 20% F127 pluronic gel (Sigma Chemical Co., St. Louis, MO). The solution was maintained in liquid form at 4 C before injection. One hundred to 150 μl of this ODN solution were taken in prechilled syringes and injected into each uterine horn.

The injections were made into the luminal space of each horn. Each ODN sample was administered in two steps at two discrete locations along the length of the horn. We imagined each horn to be divided into three equal parts: the cervical one third, the central one third, and the ovarian one third. The needle was first inserted in the middle of the central one third portion of the horn with the tip pointing toward the ovarian side. Approximately 50% of the sample was delivered into the lumen while pulling the syringe out. The needle was then reinserted within the cervical one third of the horn, and the remaining sample was injected with the tip of the needle pointing toward the ovarian side. The solution turned into a transclucent gel as soon as it came in contact with the tissue. The incision was then closed, and the animals were returned to their cages. We noted that the gel slowly disappeared over several hours. We detected no gelled material in the injected horns 4–6 h after injection.

Detection of ³³P-labeled ODNs in the uterus

The antisense ODN-1 was labeled at the 5′-end using[γ -³³P]ATP (SA, 1500 Ci/mmol) and polynucleotide kinase. The labeled ODNs were administered into the uterus following the procedure described above. The animals were killed after 2 h, and the uteri were collected. Frozen uteri were sectioned at 8 μm. The sections were obtained from both sides of the site of the first injection within the middle one third portion of the uterine horn, mounted on slides, fixed in 5% formaldehyde in PBS, and air-dried. The sections were then coated with Kodak NTB2 emulsion (Eastman Kodak, Rochester, NY) and stored in the dark at 4 C for 1 week. The slides were developed in Kodak Dektol 19 at 20 C for 2 min, washed in distilled water, and fixed with Kodak fixer for 5 min. The sections were then counterstained with hematoxylin, mounted with coverslips, and viewed under darkfield and brightfield illumination using a Nikon microscope (Nikon Corp., Melville, NY).

Northern blot analysis

Uteri were isolated from ODN-treated animals and processed for polyadenylated [poly(A)⁺] RNA isolation employing a fast track mRNA isolation kit (Invitrogen, San Diego, CA). For Northern analysis, 5–8 μg poly(A)⁺ mRNA were separated by formaldehyde agarose gel electrophoresis and transferred to Duralon membrane (Stratagene, La Jolla, CA). After transfer, the membranes were baked at 80 C for 2 h. Blots were prehybridized in 50 mm NaPO₄ (pH 6.5), 5 × SSC (standard saline citrate), 5 × Denhardt’s solution (0.02% each of Ficoll, polyvinylpyrrolidone, and BSA), 50% formamide, 0.1% SDS, and 100μ g/ml salmon sperm DNA for 4 h at 42 C. Hybridization was carried out overnight in the same buffer containing 10⁶ cpm/ml ³²P-labeled calcitonin complementary DNA (cDNA) fragment. The filters were washed twice for 15 min each time in 1× SSC-0.1% SDS at room temperature, then twice for 20 min each time in 0.2 × SSC-0.1% SDS at 55 C, and the filters were exposed to x-ray films for 24–72 h. The intensities of signals on the autoradiogram were estimated by densitometric scanning. To correct for RNA loading, the filters were stripped of the radioactive probe by washing for 5 min in 0.1% SDS at 95 C and then reprobed with ³²P-labeled glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or ferritin light chain (FLC) probes.

Immunohistochemistry and image analysis

A polyclonal antibody against rat calcitonin (obtained from Peninsula Laboratories, Belmont, CA) was diluted 1:1000 for immunohistochemistry. A rabbit antibody against rat clusterin (obtained from Dr. C. Y. Cheng, The Population Council, New York, NY) was used at a dilution of 1:2000 (28). Frozen uteri were sectioned at 7μ m, mounted on slides, and then fixed in 5% formaldehyde in PBS. Sections were washed in PBS for 20 min and then incubated in a blocking solution containing 10% normal goat serum for 10 min before incubation in primary antibody overnight at 4 C. Immunostaining was performed using a streptavidin-biotin kit for rabbit primary antibody (Zymed, South San Francisco, CA). Sections were counterstained with hematoxylin, mounted, and examined under brightfield. Red deposits indicate the sites of immunostaining. Control experiments included incubation of sections with 1) normal rabbit serum or preimmune serum for clusterin, and 2) primary antibody that had been preabsorbed with rat calcitonin (Peninsula Laboratories) or clusterin (C. Y. Cheng Laboratory, The Population Council).

A quantitative analysis of the immunohistochemical data was performed by image analysis. The intensity of calcitonin-specific staining was determined using a Nikon Optiphot-2 microscope equipped with a Dage MTI video camera (CCD 72, Dage, Michigan City, IN). The video images of calcitonin protein signal were then digitized using a frame grabber (Quick Capture, Data Translation, Marlboro, MA) and displayed on an IPC (Sun Microsystems, Inc., Mountain View, CA). The stained cytoplasmic areas of the glandular epithelial cells were traced. The integrated pixel intensity was determined for the traced areas using image analysis software (Image-Pro, Media Cybernetics, Silver Spring, MD). The intensities were normalized by dividing the integrated pixel intensity by the cytoplasmic area (which equaled the total number of pixels within the traced boundary). The background intensities were determined for each group by tracing an unlabeled area adjacent to the labeled cells. The background was subtracted from the values obtained for the labeled cells, and the adjusted values are referred to as the relative signal intensities. There were 30 observations for each group.

Statistical analysis

Statistical evaluation of the data representing the effects of ODN treatments on the number of implanted embryos in the uterus was performed using Student’s t test. P < 0.05 was considered statistically significant.

Results

Detection of radiolabeled antisense ODNs in the glandular epithelium after injection into the uterine lumen

To improve the efficacy of ODN administration in whole animal models, we made several changes in the published procedures (16, 17–19). We employed phosphorothioated ODNs containing C-5 propyne modifications at the uridine and cytidine residues. It is known that ODNs carrying these modifications exhibit improved binding affinity for target mRNAs and increased stability to cellular nucleases (15, 20). They also retain the ability to activate ribonuclease H-mediated degradation of the target RNAs (15, 20). We also used a combination of cationic liposomes N-[1-(2, 3-dioleoxyleoxy)propyl]-N,N,N-trimethylammonium methylsulfate (DOTAP) and F127 pluronic gel, a mild cell-permeabilizing agent, as a medium of ODN delivery to the cells. We employed two different antisense ODNs, As-ODN-1 and As-ODN-2, of different base compositions. As control ODNs, we used the complementary sense ODNs, S-ODN-1 and S-ODN-2.

Our previous studies showed that calcitonin mRNA and protein are synthesized predominantly in the glandular epithelial cells of the uterus. To be an effective blocker of calcitonin gene expression, an antisense ODN has to enter the glandular cells synthesizing this hormone. We, therefore, needed to ascertain that the antisense calcitonin ODNs injected into the uterine lumen were actually reaching these glandular sites within the uterus. In the experiment described in Fig. 1, an antisense ODN, As-ODN-1, was labeled at the 5′-end with ³³P and injected into the uterine lumen of pregnant rats (n = 4) on day 2 of gestation. Two hours after injection, the animals were killed, and the uteri were isolated and sectioned to monitor the presence of radioactivity. As revealed by light microscope autoradiography (Fig. 1, right panels), substantial amounts of ³³P-labeled ODNs accumulated in the glandular epithelial cells. Modest amounts of radioactivity were also observed in the luminal epithelium and stroma. These results showed that the ODNs injected into the uterine lumen were indeed reaching the glandular sites of calcitonin synthesis.

Administration of antisense ODNs specifically suppresses the level of calcitonin mRNA in rat uteri

Our previous studies showed that treatment of ovariectomized rats with progesterone leads to a dramatic induction in calcitonin gene expression within 24 h (5). This model system allowed us to assess the specificity and determine the dosage of antisense ODN administration in uteri of intact rats without having to use pregnant animals for standardization purposes. We, therefore, examined the effects of the ODNs on the expression of the mRNAs encoding calcitonin and a number of unrelated genes, such as FLC and GAPDH.

Two weeks after ovariectomy, the animals were primed with estrogen on day 15 and then injected with progesterone on day 16 to induce calcitonin gene expression. Three hours after progesterone treatment, the animals (n = 8) were subjected to a surgical operation, in which the lower abdominal cavity was opened. Both uterine horns of the same animal were injected with sense ODN, whereas both horns of a second animal were injected with antisense ODN. Twenty-four hours after this operation, the animals were killed, uteri were collected, and mRNAs were isolated for Northern blot analysis. The blot was hybridized with a ³²P-labeled calcitonin cDNA (exon IV) probe as well as with FLC and GAPDH cDNA probes.

The results of these experiments are shown in Fig. 2. The uterine horns that were treated with either 10 or 25 μg As-ODN-1 (upper panel, lanes 2 and 4) consistently exhibited drastically reduced (>90%) calcitonin mRNA compared with the horns that were injected with the same doses of S-ODN-1 (upper panel, lanes 1 and 3). Whereas treatment with 25 μg As-ODN-1 produced a slight decrease (∼20%) in the intensity of the FLC signal (lower panel, lane 2) compared with that in the sense ODN-treated (lower panel, lane 1) animals, treatment with 10 μg antisense or sense ODNs displayed FLC signals of equal intensities (lower panel, lanes 3 and 4). The effects of the ODNs on GAPDH expression were similar to those observed on FLC expression (data not shown). Marked inhibition of both FLC and GAPDH mRNA expression was noted when 40 μg or more of either sense or antisense ODN were administered in the tissue (data not shown). These results demonstrated that treatment with an optimal amount (10 μg) of antisense ODNs could bring about specific and substantial reduction in calcitonin gene expression in the uterus without significantly affecting the expression of nontarget genes.

Antisense ODNs suppress the level of calcitonin mRNA and protein in the uteri of pregnant animals

We next examined the effects of the antisense ODNs targeted to calcitonin mRNAs on calcitonin gene expression in pregnant rats. The transient nature of calcitonin gene expression in the pregnant uterus allows blockage of calcitonin expression by selecting an appropriate time window of ODN administration. We reasoned that this time window should be immediately preceding that of calcitonin mRNA or protein induction, which is between days 3–5 of gestation. Although the precise half-life of calcitonin mRNA or protein in the uterus has not been determined, the modified ODNs are known to have half lives of 24–48 h in certain tissues (15). Therefore, the ODNs injected on the afternoon of day 2 of pregnancy, when calcitonin is barely detectable, are expected to survive in the tissue during the subsequent 3–5 days of gestation to effectively suppress the surge of calcitonin mRNA. We thought that ODN treatment beyond days 2–3 would block calcitonin expression less efficiently because large quantities of calcitonin mRNA or protein would have already accumulated in the tissue before ODN injection. We avoided ODN injection beyond day 3 because of the concern that the injection of the ODNs into the uterine lumen immediately before the arrival of the embryos in the uterus on day 4 might potentially perturb the endometrium and hinder the process of embryonic implantation.

To perform ODN treatments, animals (n = 6/ODN pair, 3 for S-ODN-1 and 3 for As-ODN-1) on day 2 (afternoon) of pregnancy were subjected to a surgical operation as described in Materials and Methods. Both uterine horns were injected with 10 μg of one of the antisense ODNs, As-ODN-1 or As-ODN-2, targeted against exon IV of the calcitonin gene or the corresponding sense ODNs. Forty-eight or 72 h after the operation, the animals were killed, uteri were collected, and mRNAs were isolated for Northern blot analysis. The blot was hybridized with a ³²P-labeled calcitonin cDNA (exon IV) probe and control probes including FLC or GAPDH cDNAs. The results of these experiments are shown in Fig. 3. As shown in the top panel of the figure, the uterine horns that were treated with antisense ODNs (right lane) exhibited drastically reduced calcitonin mRNAs on day 4 compared with horns that were injected with the same doses of sense ODNs (left lane). Hybridization of the blot with FLC or GAPDH probes indicated no reduction in the intensities of these signals in the antisense ODN-treated uterus compared with those in the sense ODN-treated tissue (Fig. 2, middle and bottom panels). Due to a relative overloading of RNA sample in the right lane (As), the FLC and GAPDH signals in this lane appear to be slightly more intense than those in the left lane (S). Upon quantitation and normalization of the mRNA signals, we estimated that more than 90% of the calcitonin expression on day 4 was suppressed by antisense ODN treatment. A similar decline in calcitonin mRNA levels on day 5 was observed upon antisense ODN treatment (data not shown). These results demonstrated that antisense ODN intervention on the afternoon of day 2 of gestation effected a specific and drastic suppression of calcitonin mRNA expression in the pregnant rat uterus immediately before (day 4) and at the time (day 5) of implantation.

We also monitored the level of calcitonin protein after antisense or sense ODN treatment by performing immunocytochemical staining of sections of uteri isolated from ODN-treated pregnant animals (n = 6) using an antibody against calcitonin. The results of these experiments are shown in Fig. 4. The uteri of animals that were injected with either sense ODN, S-ODN-1 or S-ODN-2, on day 2 of pregnancy and killed on day 4 exhibited an intense immunostaining (upper panel a) that was mainly localized in the glandular epithelial cells. This staining was calcitonin specific, as control sections incubated with calcitonin antibody that was preabsorbed with excess rat calcitonin showed no immunoreactivity (Fig. 4c). In contrast, the sections of uteri that were injected with the corresponding antisense ODN, As-ODN-1 or As-ODN-2, showed strikingly reduced calcitonin-specific staining (upper panel b). Quantitation of protein staining by image analysis revealed that greater than 80% of calcitonin immunoreactivity was lost upon antisense ODN treatment (Fig. 4, middle panel).

To ascertain that the ODN treatment did not result in a general suppression of protein synthesis in the glandular epithelial cells, we also monitored the expression of a nontarget protein, clusterin, in these cells. Glands of untreated animals showed specific immunostaining with an anticlusterin antibody (Fig. 4, lower panel b). Our results show that treatment with either antisense ODN (Fig. 4, lower panel c) or sense ODN (data not shown) did not affect clusterin protein expression in the glands. These results indicate that antisense ODNs targeted against calcitonin mRNA can specifically inhibit the synthesis of calcitonin protein in the glandular epithelial cells of the uterus without affecting the expression of nontarget proteins.

Administration of antisense oligonucleotides into rat uteri in the early preimplantation phase reduces the number of the implanted embryos

We next investigated whether the suppression of calcitonin mRNA and protein expression upon antisense ODN treatment influenced embryonic implantation. As described above, we administered either the antisense or the corresponding sense calcitonin ODNs into both uterine horns of rats on the afternoon of day 2 of pregnancy. After surgery, the animals were returned to their cages and killed on day 9 of gestation. The uteri were opened, and the number of implanted embryos in each uterine horn was counted.

On day 9 of pregnancy, control saline-injected rats contained six to eight implanted embryos, on the average, in each uterine horn. When the animals were injected with either S-ODN-1 (n = 8) or S-ODN-2 (n = 10) on day 2 of gestation, no significant change in the number of implanted embryos was observed compared with that in the control animals (Fig. 5a, A). This indicated that the dose of ODNs used in this study is not generally toxic to the embryos. Moreover, the surgical perturbation and the injection of vehicle without ODNs on day 2 did not have any harmful effect on embryo implantation.

When rats were injected with either As-ODN-1 (n = 8) or As-ODN-2 (n = 10) on day 2 of gestation, we noted a marked reduction in the number of implanted embryos in 8 of 10 animals. An example of the antisense effect is shown in Fig. 5a, B. In this As-ODN-1-treated uterus, there was no visible embryo in one horn, whereas the other horn showed only three implanted embryos. The embryos also appeared to be abnormal in that they were larger in the As-ODN-treated animals compared with those in the S-treated animals. Isolated uteri from the sense and antisense ODN-treated animals are compared in Fig. 5b, A and B. As shown in Table 1, typically, a 50–80% reduction in the number of implanted embryos was evident in each uterine horn in several independent sets of experiments performed with two different sets of antisense ODNs as described in Materials and Methods. The inhibition of implantation by the antisense ODNs was not complete, presumably because a low level of calcitonin was still expressed in the glands. Only minimal effects (<15%) on implantation were observed in 1 of 8 animals treated with As-ODN-1 and 2 of 10 animals treated with As-ODN-2. The reasons for this weaker response in 10–20% of treated animals are not clear. A statistical test was applied to the data obtained from several experiments to assess the significance of the difference between the number of implanted embryos obtained after treatment with either antisense or sense ODNs. Such analysis produced a P < 0.05%, indicating that the difference in the number of implantation sites in sense- and antisense-treated animals is indeed significant. Taken together, these results indicate that treatment of the uterine horns with antisense calcitonin ODNs, but not with the complementary sense ODNs, can severely impair embryo implantation.

Discussion

Our previous studies revealed that calcitonin, a peptide hormone that regulates calcium homeostasis, is synthesized by the uterus during pregnancy in an implantation stage-specific manner. One way of investigating the biological role of calcitonin during implantation is by analyzing a calcitonin-deficient mutant mouse model system; however, such a system is currently not available. There is also lack of a calcitonin antagonist that functions effectively in vivo. We have, therefore, taken an alternative approach of blocking calcitonin gene expression by using antisense ODNs targeted against calcitonin mRNA. In this study we demonstrate that administration of antisense oligodeoxynucleotides (ODNs), targeted specifically against calcitonin mRNAs, into the lumen of the preimplantation phase uterus results in a dramatic reduction in the number of implanted embryos. Similar treatment with the corresponding sense ODNs exhibits no effect on implantation. The antisense ODN intervention also markedly suppresses the steady state level of the calcitonin mRNA and protein in the uterus, without affecting the expression of unrelated genes. These results collectively suggest that the blockade of embryonic implantation upon administration of the antisense ODNs into the uterus could be a direct phenotypic consequence of the suppression of calcitonin gene expression in the implantation phase of gestation.

A major concern regarding the antisense methodology is the possibility of nonspecific inhibition of gene expression caused by ODN treatment. This issue has been tackled in the present study by the use of multiple sense and antisense ODNs. We have shown that treatment of pregnant uteri with two different antisense ODNs, As-ODN-1 and As-ODN-2, possessing different base compositions produced the same phenotype, but the corresponding sense ODNs of equal length had no effect. Moreover, the antisense, but not the sense, ODNs inhibited target mRNA expression, whereas neither inhibited nontarget mRNAs. Another concern is any general toxic effect the ODNs or the vehicle may exert on the embryo. In our studies, we have observed that administration of two different sense ODNs (10 μg) or the vehicle on day 2 of pregnancy did not have any significant effect on implantation or subsequent embryo development until day 9 of gestation. It is, therefore, highly unlikely that the administration of optimal levels of ODNs or vehicle will have any general deleterious effect on the embryo.

The mechanism of action of calcitonin in the uterus during implantation is unclear. Implantation is the culmination of a sequence of discrete functional events, such as successful attachment of the blastocysts to the appropriate sites on the luminal epithelium and proper embryonic development. If calcitonin is involved in the preparation of the endometrium for implantation, one would expect that the attenuation of calcitonin expression after treatment with antisense ODNs may prevent acquisition of the receptive state of the endometrium, leading to a failure of implantation. If, on the other hand, calcitonin has a role in early embryogenesis, attenuation of its expression during development by antisense ODN may hinder the proper growth and differentiation of the embryos, and these abnormal embryos might fail to implant.

During implantation in the rat, the initial interaction of the trophoblastic membrane with the endometrial epithelium is followed by the intrusion of the trophoblastic processes between the epithelial cells (1–3). The epithelial cells form tight junctions near their apical regions. These tight junctions need to be relaxed or dissociated before successful intrusion by trophoblastic processes. It has been reported previously that addition of calcium ions to a culture of polarized epithelial cells induces the formation of tight junctions, and removal of these ions leads to relaxation of tight junctions (29). Previous studies also demonstrated that treatment of isolated osteoclasts or cells stably transfected with cloned calcitonin receptor cDNA with calcitonin induces influx of extracellular calcium into these cells (30–32). It is, therefore, conceivable that calcitonin secreted by the glandular epithelium may act on the luminal epithelial cells to regulate embryo attachment. Lowering of extracellular calcium levels in the uterine fluid in the immediate surroundings of the implantation bed may induce junction disassembly, redistribution of junctional proteins, and opening of the tight junction barrier in epithelial cells to promote trophoblast invasion (33, 34). Consistent with this scenario, our recent studies detected the presence of a significant amount of calcitonin in the luminal secretions of pregnant rats only between days 3–5 of gestation (Bove, K., and I. C. Bagchi, unpublished observation).

One can also imagine that calcitonin secreted by the uterine epithelium may act on trophoblast cells to induce calcium uptake by these cells. This would also result in a lowering of the calcium ion concentration in the uterine fluid between the epithelial and the trophoblast cells, and this event, in turn, would trigger relaxation of the tight junctions between epithelial cells to facilitate implantation. Further studies leading to the localization of cells that are targets of uterine calcitonin will help us to decipher mechanism of action of this peptide hormone during implantation.

Calcitonin may also play a role in implantation by acting as a factor controlling early embryonic development. Calcium is known to regulate cell aggregation during early embryogenesis, and calcitonin, due to its hypocalcemic properties, may influence cell adhesion (35). In the rat after fertilization, the embryo enters the uterine lumen on day 4 of pregnancy and may potentially be influenced by maternal calcitonin for at least 24 h before implantation on day 5. The block in uterine calcitonin expression after antisense ODN injection might hinder the proper growth and differentiation of the embryos, and these abnormal embryos might fail to implant. A role of calcitonin in early differentiation of the vertebrate embryo has indeed been suggested by Burgess (36). Studies by this investigator have shown that treatment of Xenopus embryo during gastrulation with calcitonin leads to faulty induction of the neural plate and results in defects in the development of oral-facial and central nervous system architecture. One would also have to consider the possibility that antisense ODNs may directly act on the embryos to inhibit their development, perhaps by blocking early embryonic calcitonin synthesis. Although the target(s) of calcitonin action in the developing embryo remains to be identified, our current findings suggest that calcitonin, in addition to its regulatory role in calcium metabolism, may have a broader spectrum of activity at early stages of development. Such a role might be similar to the proposed functions of the cytokines, colony-stimulating factor-1, and leukemia inhibitory factor, in regulating trophoblast proliferation and differentiation (37, 38).

Acknowledgments

We gratefully acknowledge Dr. Ying-Qing Ding for help in developing the ODN treatment methodology and surgical procedures. We thank Dr. C. Y. Cheng for the generous gift of the clusterin antibody. We also thank Evan Reed for the artwork, and Jean Schweis for carefully reading the manuscript.

Author notes