Abstract
Ovine GH (oGH) is synthesized in placental tissue during maximal placental growth and development. Our objectives were to localize oGH mRNA in the placenta, and study the impact of exogenous GH on twin pregnancies during the normal window (35–55 days of gestational age; dGA) of placental expression. In situ hybridization localized oGH mRNA in uterine luminal epithelium but not in tissues of fetal origin. While maternal GH and IGF-I concentrations were increased (P<0.001) approximately tenfold, uterine, uterine fluid, placental, and fetal weights were unaffected by treatment at either 55 or 135 dGA. Fetal length, liver weight, and liver weight per kg of body weight were unaffected by maternal GH treatment. However, in the cotyledon, IGF-binding protein (BP)-1 and IGFBP-4 mRNA concentrations were increased (P<0.05), while IGFBP-2 mRNA was decreased (P<0.05). The concentration of mRNA for IGFBP-3 was unaffected by treatment. Within the caruncle, IGFBP-1 mRNA was decreased (P<0.05), while IGFBP-3 and IGFBP-4 mRNA were increased (P<0.05), and IGFBP-2 mRNA was unchanged due to GH treatment. While our data indicate that elevated maternal GH and IGF-I concentrations during early and mid-gestation do not enhance placental and fetal growth in twin pregnancies, localization of GH mRNA in uterine luminal epithelium could explain GHs transitory expression from 35 to 55 dGA, since by the end of this period the majority of the uterine luminal epithelium has fused with chorionic binucleate cells forming the placental syncytium.
Introduction
Sheep have two growth hormone (GH) genes (ovine GH-1 (oGH-1) and oGH-2), and oGH-2 has multiple alleles (oGH-2N and oGH-2Z; Valinsky et al. 1990, Gootwine et al. 1993). Expression of the oGH-2 gene has been detected in sheep placenta (Lacroix et al. 1996, 1999) and it was inferred that both the N- and Z-alleles of the gene are expressed (Ofir & Gootwine 1997). oGH mRNA and protein have been detected in the placenta during early to mid-gestation (Lacroix et al. 1996, 1999), with the predominant window of oGH expression occurring between 35 and 55 days of gestational age (dGA). Immunolocalization of oGH within maternal syncytial and stromal cells, as well as the fetal trophoblast, has been reported for 40–45 dGA placentomes (Lacroix et al. 1996). This group also reported in situ hybridization of oGH mRNA in the syncytium and trophoblast layers of the placenta at 40 dGA (Lacroix et al. 1999). The oGH receptor (oGHR) mRNA was initially identified in the placenta at 51, 95, and 120 dGA (Klempt et al. 1993) and was localized to what was identified as uterine epithelium and uterine glands at 120 dGA. Lacroix et al. (1999) identified oGHR mRNA in the uterine endometrium from 8 to 120 dGA and identified the primary site of oGHR mRNA being the trophectoderm at 40 dGA. Collectively, these data suggest that a functional interaction of oGH and its receptor may exist during early to mid-gestation within the sheep uteroplacental unit.
Administration of recombinant porcine GH (rpGH) to gilts during early to mid-gestation increases fetal and placental weight (Sterle et al. 1995), and increases in fetal weight and length have also been reported for underfed gilts given rpGH during mid-gestation (Gatford et al. 2000). However, studies in sheep to elucidate the impact of GH administration during pregnancy have provided variable results. Administration of GH during conceptus elongation and early placentation has no effect on placental development (Spencer et al. 1999). Only endometrial gland proliferation was stimulated and this effect was seen with concurrent interferon-τ administration (Spencer et al. 1999). Wallace et al. (2004) reported an increase in uteroplacental weight due to GH infusion from 35–80 dGA; however, this is due likely to increased fetal fluid weight resulting from polyhydraminios, as total and average placentome weights and fetal weights were not statistically different as a result of GH treatment. Furthermore, the increase in uteroplacental weight (Wallace et al. 2004) was only seen in adolescent ewes maintained on a high dietary intake, a feeding regimen used in adolescent ewes to generate fetal growth restriction (Wallace et al. 1997, Anthony et al. 2003). More recently, treatment of ewes from 35–65 dGA with GH failed to alter fetal or placental development (Wallace et al. 2006), as measured some 65 days later, and the GH was only administered to the high intake adolescent ewes. Interestingly, in over-nourished adolescent ewes, serum GH is decreased while serum insulin-like growth factor-I (IGF-I) is increased (Wallace et al. 1997), indicating a disconnect of the normal endocrine relationship between GH and IGF-I.
Administration of GH during mid-pregnancy (70–83 dGA) in normal sheep pregnancies results in increased maternal serum non-esterified fatty acids, IGF-I and glucose, but no change in fetal growth (Jenkinson et al. 1999). Fetal weight was also increased 10% by GH infusion at 98–111 dGA (Jenkinson et al. 1999); however, several reports saw no effect on fetal or placental weight with similar temporal administration of GH (Stelwagen et al. 1994, Currie et al. 1996, Harding et al. 1997). Interestingly, all of the studies discussed above did not administer exogenous GH concurrent with placental GH expression (∼35–55 dGA) within the placenta (Lacroix et al. 1996, 1999). It remains to be seen what the effects of GH administration, concurrent with the normal placental expression window, will have in sheep. Therefore, it was our objective to further define the location of oGH expression in the placenta and to determine the effects of exogenous recombinant bovine GH (rbGH) administration during the normal window of placental oGH expression in twin sheep pregnancies. We chose to study twin pregnancies, as multiple fetuses place increased demand on the maternal system and placenta to provide adequate nutrition to the fetus.
Materials and Methods
cDNA generation
In order to verify the origin of oGH expression within the placenta of sheep, placentomes were collected from 45 dGA pregnancies following killing according to Colorado State Univeristy IACUC approved protocol 02-304A. Intact placentomes were excised, trimmed to 5 mm cross-sections, and fixed overnight in 4% paraformaldehyde, before being embedded in paraffin. cDNA for oGH, vimentin (marker of fibroblast cell lineage), cytokeratin 18 (marker of epithelial cell lineage), and CD 18 (marker of leukocytes) were generated by RT PCR. The cDNA used for in situ hybridization of ovine placental lactogen (oPL) has previously been reported (Kappes et al. 1992). For the reverse transcriptase reaction, the SuperScript First Strand Synthesis kit (Invitrogen) was used following the manufacturer's protocol. The PCR products were generated by 40 cycles of 95 °C 1 min to denature, 1 min to anneal (°C in Table 1), and 72 °C 1 min for extension. The PCR amplicons were inserted into PCR-Script Amp SK (Stratagene, Cedar Creek, TX, USA) for transformation, sequence validation, and subsequent propagation. Plasmid DNA was harvested by alkaline lysis, and purification by centrifugation through a CsCl gradient (Liang et al. 1999, Limesand & Anthony 2001). The plasmid preparations were sequenced by Macromolecular Resources (Colorado State University) to verify sequence authenticity.
PCR primers, annealing temperatures, and product sizes for cDNA used in real-time PCR analysis
| Forward primer | Reverse primer | Anneal (°C) | Product (bp) | |
|---|---|---|---|---|
| cDNA | ||||
| IGF-I | tcgcatctctcttctatctggccctgt | acagtacatctccagcctcctcaga | 62 | 238 |
| IGF-II | gaccgcggcttctacttcag | aagaacttgcccacggggtat | 62 | 202 |
| IGFBP-1 | tgatgaccgagtccagtgag | gtccagcgaagtctcacac | 62 | 248 |
| IGFBP-2 | caatggcgaggagcactctg | tggggatgtgtagggaatag | 55 | 330 |
| IGFBP-3 | ctcagagcacagacaccca | ggcatatttgagctccac | 54 | 335 |
| IGFBP-4 | tgtgcgtgtgtgttgatg | gagggagccaagatgagt | 62 | 229 |
| RPS15 | atcattctgcccgagatggtg | tgctttacgggcttgtaggtg | 60 | 134 |
In situ hybridization
To generate template for in situ hybridization, oGH (406 bp; EcoRI), oPL (533 bp; BamHI and NdeI), oVimentin (369 bp; EcoRI and NotI), oCytokeratin 18 (1307 bp; XhoI), and oCD 18 (679 bp; EcoRI and SacI) were digested from their respective plasmids and gel purified on 1% agarose gels. For a negative control, ∼400 bp of non-coding sequence from PCR-Script Amp were digested out of intact plasmid with NspI and SacII. Radiolabeling of cDNA for in situ hybridization was performed using a 50–100 ng template with the DECAprime II random prime labeling kit (Ambion, Austin, TX, USA) and radiolabeled cytosine tri-phosphate (33Pα-dCTP) per the manufacturer's instructions. Only probes with specific activities exceeding 5×108 c.p.m./μg DNA were used for in situ analysis. In situ hybridization was performed as described previously (Kappes et al. 1992).
Exogenous GH administration
Twenty-three whiteface crossbred ewes were naturally mated to a single sire, generating twin pregnancies. All procedures were approved under Colorado State University IACUC protocol 02-304A. At 35 dGA, the ewes were assigned to treatment groups for s.c. injections of GH (GH, n=10) or buffer control (CON, n=13). The GH group received 0.2 mg/kg body weight (BW) of rbGH (Monsanto, St Louis, MO, USA) per day reconstituted in 100 mM NH4HCO3 buffer. The rbGH was administered by twice daily injections of 3 ml 12 h apart. The CON group received two injections of 3 ml of 100 mM NH4HCO3 buffer. The injection regimen lasted for 20 days (35–55 dGA) to coincide with the normal expression window of oGH within the placenta (Lacroix et al. 1996, 1999). Prior to, and every 5 days during rbGH administration, blood samples were collected by jugular venipuncture to compare serum hormone concentrations of CON and GH groups.
Between 54–56 dGA, 7 CON, and 5 GH ewes were killed (90 mg/kg sodium pentobarbitol) and uterine, placental, and fetal tissues were collected for analysis. Ewe weight, gravid uterine weight, placental weight, and fetal parameters (sex, weight, crown rump length (CRL), and liver weight) were all recorded. The seven CON pregnancies yielded nine male and five female fetuses, whereas the five GH pregnancies yielded five male and five female fetuses. Using CRL and fetal weight, the ponderal index (PI) for each fetus was calculated (PI=weight in grams/length in cm3). Cotyledon and caruncle tissue from each placenta were manually separated and snap frozen in liquid nitrogen for RNA analysis later. The remaining ewes (CON, n=6; GH, n=5) were gestated until 135 dGA to evaluate the effects of early gestation rbGH administration near term. Ewes were killed between 133 and 138 dGA, and the same data and tissue samples were collected as the 55 dGA groups. The 6 CON pregnancies yielded 2 male and 10 female fetuses, whereas the 5 GH pregnancies yielded 4 male and 6 female fetuses. Additionally, empty uterine weight was measured for use in calculating fluid weight within the uterus. To calculate fluid weight, fetal, placental, and empty uterine weights were subtracted from gravid uterine weight.
Serum hormone analyses
A validated GH RIA was used to determine serum GH levels (Hoefler & Hallford 1987) for duplicate samples. Using NIDDK anti-oGH and NIDDK oGH, the intra-assay coefficient of variation was 7.2% and recovery was 97% for 15 ng. A two-site IRMA (Diagnostic Systems Laboratories, Webster, TX, USA) was used to assay serum IGF-I as directed by the manufacturer. Intra- and interassay coefficients of variation were 3.43 and 3.39% for the 50 ng/ml control respectively, and 2.29 and 2.27% for the 200 ng/ml control.
Northern hybridization analysis
Radiolabeling of cDNA for northern hybridization was performed using 50–100 ng template in the DECAprime II random prime labeling kit (Ambion) and using 32Pα-dCTP per the manufacturer's instructions. Only cDNAs with specific activities exceeding 5×108 c.p.m./μg DNA were used for analysis. Caruncle total cellular RNA (tcRNA) was hybridized with oGH cDNA, while cotyledon tissue was hybridized with oPL cDNA as described previously (Kappes et al. 1992). The membranes were exposed to a Phosphor Screen (Molecular Dynamics, Sunnyvale, CA, USA) overnight for oPL and 48 h for oGH. The screens were scanned using a Storm Imager (Molecular Dynamics) to visualize hybridization. The membranes were then stripped with repeated washes of 200 ml boiling stripping buffer (0.05×SSC, 10 mM EDTA and 0.1% SDS), blocked with pre-hybridization buffer, and hybridized with ribosomal protein S15 (RPS15) cDNA (2×106 c.p.m./ml) to serve as a normalization control. For northern hybridization analysis, the expression of RPS15 (Rozance et al. 2007) was used for normalization. The concentration of RPS15 was used for normalization due to changes by treatment in other housekeeping genes tested (actin and glyceraldehydes-3-phosphate dehydrogenase). Following hybridization, the membranes were exposed to a Phosphor screen overnight and scanned with the Storm Imager. Densitometric analysis of the Storm images for all northern hybridizations was completed with ImageQuant 5.0 (Molecular Dynamics).
Quantitative reverse transcriptase PCR
tcRNA was isolated as described previously (Warren et al. 1990, Kappes et al. 1992). An aliquot of the stock tcRNA samples was diluted to a concentration of 1 μg/μl±0.2 μg/μl for use as a working solution. The concentration of the working solutions was verified using the Ribogreen (Molecular Probes, Eugene, OR, USA) RNA quantification system. Measurements of fluorescence were taken using a Mithras LB 940 (Bertold Technologies, Oak Ridge, TN, USA). The samples were excited at 485 nm and emission was measured at 535 nm. The counting time was 0.1 s and the lamp energy was set at 10 000. The unknowns were compared against an RNA standard curve.
For the reverse transcriptase reaction, 2 μg tcRNA was used in the SuperScript First Strand Synthesis kit (Invitrogen). All samples of tcRNA for both caruncle and cotyledon were incubated at the same time with the same kit to avoid sample variation due to technician and/or kit variation. The yield of cDNA from the RT reactions was quantified with the Oligreen (Molecular Probes) ssDNA quantification kit. Samples were excited and measured with the Mithras LB 940 with the same settings as for the Ribogreen analysis. Following quantification, all samples were set to a cDNA concentration of 12.5 ng/μl to be used in quantitative PCR (qPCR). The qPCR mix was 12.5 μl iQ SYBR Green Supermix (Bio-Rad), 200 nM primer (forward and reverse), 2 μl cDNA (25 ng), and balanced to 25 μl total volume with sterile water with a neutral pH. For all qPCRs, the first denaturing step was 95 °C for 5 min and 45 s thereafter, the annealing time was 1 min with gene-specific temperature (Table 1), and 1 min of extension at 72 °C. The reactions were run for 40 cycles, with a fluorescence measurement after each cycle, followed by a melting curve analysis initiated at 55 °C and measuring fluorescence every 0.5 °C to 95 °C. For the qPCR analysis, cDNA for ovine IGF-I, IGF-II, and IGF-binding proteins (BP) 1, 2, 3, and 4 (de Vrijer et al. 2006) and RPS15 (Rozance et al. 2007) were used as standard templates. The standard curve template concentrations spanned 100 to 1×10−5 pg/μl at tenfold increments. For each standard, 1 μl template was supplied to the reaction mix with the same primer concentration. Table 1 lists the primer sequences, annealing temperatures, and product sizes generated by PCR. Primers were designed to span at least one intron of the mRNA sequence for each gene of interest and produce 130–350 bp of product. For each gene of interest, the qPCRs were optimized as described previously (Erickson-Hagen et al. 2005) to obtain apmplification efficiencies of 90–100%, and were validated by electrophoresis of the PCR products to verify correct product size prior to analysis. The standard reactions were run in duplicate and the unknown reactions were run in triplicate. The means are presented as gene of interest cycle threshold divided by RPS15 cycle threshold (Ct/Ct).
Statistical analysis
For the maternal and fetal data collected at necropsy, two-way ANOVA (treatment and fetal sex as dependent variables) was conducted on the data obtained at 55 (7 CON and 5 GH twin preganancies) and 135 (6 CON and 5 GH twin pregnancies) dGA separately. Since there were no significant sex effects or treatment–sex interactions, fetal sex was removed from the statistical model, making student's t-test appropriate for comparing treatment effects (CON versus GH), at each gestational age (55 or 135 dGA). Serum hormone data means (CON and GH) for each sample day also were separated by student's t-test, as were normalized means for northern and qPCR analyses for the CON and GH.
Results
Localization of oGH mRNA
In situ hybridization analysis of placentome sections revealed a distinct population of cells at the maternal–fetal interface that hybridized with the oGH cDNA (Fig. 1A and B). oGH hybridization did not appear to reside within the fetal trophoblast layer, which contain chorionic binucleate cells (BNCs) as identified by oPL hybridization (Fig. 1C and D). To determine what cell-type expresses oGH, oCytokeratin 18, and oVimentin radiolabeled cDNAs were hybridized to day 45 placentome sections as well. Figure 1E–H represent light and dark phase images of oCytokeratin 18 and oVimentin hybridization respectively. Ovine cytokeratin 18 hybridized to the maternal epithelium (Fig. 1E and F), whereas vimentin hybridization was apparent in the maternal villous stroma and to a lesser extent in the fetal villous stroma (Fig. 1G and H). The pattern of hybridization for oGH and oCytokeratin 18 was similarly located at the maternal–fetal interface, and was not consistenly present throughout the boundary of the maternal tissue (Fig. 1A and E). The oGH-positive cells were in random clusters at outer boundaries of maternal tissue, not in the maternal stroma or fetal trophoblast, and appear to be the remnants of the maternal uterine epithelium. To verify that these cells were not of leukocyte origin, sections were hybridized with our oCD18 cDNA. The CD18-positive cells were deep in the endometrium and myometrium, far from the oGH-positive cells within the maternal epithelium (data not shown).
(A and B) In situ hybridization with oGH cDNA. (A) A light field image of day 45 placentome. (B) The dark field of the same location. Positive hybridization is detected in UE. (C and D) In situ hybridization with oPL cDNA. (C) A light field image of day 45 placentome. (D) The dark field of the same location. Positive hybridization detected in FT binucleate cells. (E and F) In situ hybridization with cytokeratin 18 cDNA. (E) A light field image of day 45 placentome. (F) The dark field of the same location. Positive hybridization is detected in UE. (G and H) In situ hybridization with vimentin cDNA. (G) A light field image of day 45 placentome. (H) The dark field of the same location. Positive vimentin hybridization is detected in MVS. FT, fetal trophoblast; FVS, fetal villous stroma; MVS, maternal villous stroma; UE, uterine epithelium; final magnification=400×.
Citation: Journal of Endocrinology 198, 1; 10.1677/JOE-07-0481
Effect of rbGH administration on maternal and fetal characteristics
Exogenous rbGH was administered twice daily to pregnant ewes from days 35 to 55 of gestation, resulting in greater (P<0.001) concentrations of serum GH by day 5 of treatment (Fig. 2), which remained approximately tenfold higher than the controls throughout the treatment period. The exogenous rbGH was biologically active in sheep as the treated ewes exhibited increases (P<0.001) in serum IGF-I concentrations, as evidenced in Fig. 2.
Serum GH and IGF-I concentrations. Concentrations for control (CON) and GH-treated (GH) ewes throughout the gestation treatment period (days 35–55) are presented as mean±s.e.m. with significant differences marked with an asterisk.
Citation: Journal of Endocrinology 198, 1; 10.1677/JOE-07-0481
The physical data, for the ewes, fetuses and placenta collected at necropsy are summarized in Tables 2 and 3. For the 55 dGA groups, no differences were seen in uterine weight, placental weight, fetal weight, CRL, PI, liver weight, or liver weight expressed per kg of BW between GH and CON pregnancies. Similar results were obtained in the near term (135 dGA) ewes. No significant differences were seen between uterine, placental, calculated fluid, and fetal weights. Additionally, no significant differences were found in CRL, PI, liver weight, or liver weight/kg BW between treatments in the 135 dGA groups. There was no effect of fetal sex on the fetal parameters measured at days 55 and 135.
Day 55 maternal and fetal physical data. Data for control (CON) and GH treated (GH) ewes are presented as mean±s.e.m.
| CON | GH | P value | |
|---|---|---|---|
| Treatment | |||
| Pregnancies (n) | 7 | 5 | |
| Days post coitus | 55.14±0.26 | 55±0.32 | 0.734 |
| Ewe weight (kg) | 85.29±2.31 | 82.82±5.38 | 0.610 |
| Gravid uterine weight (g) | 1773.57±73.72 | 1934±59.32 | 0.336 |
| Placental weight (g) | 693.26±59.58 | 709.06±90.75 | 0.882 |
| Fetal weight (g) | 32.23±1.18 | 32.03±1.28 | 0.910 |
| Crown rump length (cm) | 8.26±0.16 | 8.48±0.15 | 0.346 |
| Ponderal index (g/cm3) | 0.057±0.002 | 0.053±0.001 | 0.075 |
| Liver weight (g) | 2.64±0.15 | 2.48±0.11 | 0.426 |
| Liver weight/kg BW | 81.62±2.09 | 77.56±1.98 | 0.190 |
Day 135 maternal and fetal physical data. Data for control (CON) and GH treated (GH) ewes are presented as mean±s.e.m.
| CON | GH | P value | |
|---|---|---|---|
| Treatment | |||
| Pregnancies (n) | 6 | 5 | |
| Days post coitus | 136.00±0.97 | 136.20±0.80 | 0.880 |
| Ewe weight (kg) | 91.29±4.78 | 97.27±5.36 | 0.425 |
| Gravid uterine weight (kg) | 13.80±0.81 | 14.06±0.66 | 0.816 |
| Placental weight (kg) | 1.45±0.09 | 1.50±0.11 | 0.748 |
| Fluid weight (kg) | 2.26±0.39 | 2.21±0.21 | 0.915 |
| Fetal weight (kg) | 4.45±0.17 | 4.54±0.13 | 0.655 |
| Crown rump length (cm) | 50.36±0.79 | 51.70±0.82 | 0.332 |
| Ponderal index (g/cm3) | 0.035±0.001 | 0.033±0.001 | 0.325 |
| Liver weight (g) | 117.43±6.36 | 125.20±8.39 | 0.462 |
| Liver weight/kg BW | 26.33±0.77 | 27.40±1.31 | 0.473 |
Effect of rbGH administration on placental gene expression
To assess placental tissue response to exogenous GH, cotyledon and caruncle tissues were collected from each 55 dGA pregnancy, and northern hybridization analysis was used to detect changes in cotyledon and caruncle concentrations of oPL and oGH mRNA (Fig. 3). Robust signal was obtained for oGH in maternal caruncles, and for oPL in the fetal cotyledons, further supporting that oGH mRNA resides in the maternal placental component. However, there was no effect of GH treatment on cotyledon oPL mRNA concentrations or maternal caruncle oGH mRNA concentrations (Fig. 3).
Relative expression of oPL in the fetal cotyledon and oGH in the maternal caruncle at 55 dGA. Densitometric values for the samples were normalized by densitometric values for RPS 15. Ten micrograms tcRNA were loaded for northern hybridization.
Citation: Journal of Endocrinology 198, 1; 10.1677/JOE-07-0481
Cotyledon mRNA concentration for IGF-I, IGF-II, or IGFBP-3 was not impacted by GH treatment (Fig. 4). However, the concentration of mRNA encoding IGFBP-1 and -4 was increased (P<0.05), while IGFBP-2 mRNA concentration was decreased (P<0.05) in the GH-treated fetal placenta. Figure 5 presents the comparison of relative expression of the transcripts within the caruncle. No significant differences were detected between treatments for caruncle IGF-I, IGF-II, and IGFBP-2 mRNA concentrations. The concentration of caruncular IGFBP-1 mRNA was significantly (P<0.05) decreased, while the concentrations of IGFBP-3 and -4 mRNA were both significantly (P<0.05) increased by GH treatment.
Relative expression of IGF-I, IGF-I, IGFBP-1, -2, -3, and -4 mRNA in cotyledon tissue. All transcripts were corrected (Ct/Ct) by RPS 15 mRNA. Data are expressed as mean±s.e.m. and significant differences (P<0.05) are indicated by an asterisk.
Citation: Journal of Endocrinology 198, 1; 10.1677/JOE-07-0481
Relative expression of IGF-I, IGF-II, IGFBP-1, -2, -3, and -4 mRNA in caruncle tissue. All transcripts were corrected (Ct/Ct) with RPS 15 mRNA. Data are expressed as mean±s.e.m. and significant differences (P<0.05) are indicated by an asterisk.
Citation: Journal of Endocrinology 198, 1; 10.1677/JOE-07-0481
Discussion
GH has been previously detected in the ovine placenta from days 35 to 55 of gestation; however, the cellular source of GH described (Lacroix et al. 1996, 1999) did not coincide with our preliminary studies (TRH Regnault & RV Anthony unpublished observations). Therefore, the first objective of this research was to definitively localize the expression of oGH in day 45 placentome tissue. The results of our in situ hybridization contrast with those reported by Lacroix et al. who reportedly detected oGH in maternal syncytium and stromal cells, and fetal trophoblast. The use of in situ hybridization and visualization at high magnification reported here specifically detected hybridization of oGH at the maternal–fetal interface (Fig. 1A and B). These cells were not chorionic BNC derived, as marked by oPL (Fig. 1C and D). The cells that hybridized to the oGH cDNA also showed positive hybridization for cytokeratin 18 (Fig. 1E and F) a marker of epithelial cells and not vimentin, a marker of fibroblast-derived cells. Additionally, the oGH-positive cells were negative for CD 18 hybridization precluding them from being leukocyte in origin (data not shown).
The results of the in situ hybridization analysis indicate that oGH is expressed in maternal uterine epithelium and is not detected in maternal villous stroma nor fetal trophoblast or villous stroma. Our detection of oGH in day 45 placental tissue coincides with the reported maximal expression of oGH (Lacroix et al. 1999). The pattern of oGH expression beginning at day 35, peaking at day 45, and decreasing to day 55 reported by Lacroix et al. is further supported by a much less robust detection of oGH in day 55 placentomes (TRH Regnault & RV Anthony unpublished observations). While we do not understand the initiation of oGH in the placenta at ∼35 dGA, the mechanism by which oGH expression is terminated may be the loss of the cells producing oGH. The loss of maternal epithelium within the caruncle due to syncytium formation is well documented, and by 55–60 dGA the epithelium is not detectable (Davies & Wimsatt 1966, Stegeman 1974). Therefore, the cessation of oGH expression is likely not a change in transcriptional regulation, but the loss of oGH producing cells.
Maternal GH infusion late in gestation (days 98–111) has been shown to increase fetal growth in singleton pregnancies (Jenkinson et al. 1999); however, these results are not consistent within the same breed receiving the same oGH treatment (Currie et al. 1996). Wallace et al. initiated GH administration to adolescent ewes at the appropriate time (day 35); however, the GH treatment continued to day 80 (Wallace et al. 2004), well past the reported cessation of oGH expression in the placenta (Lacroix et al. 1996, 1999). Although Wallace et al. reported an increase in placental weight in the over-nourished adolescent ewes with singletons; these ewes are predisposed to placental and fetal growth restriction due to the high plane of nutrition. The over-nourished adolescent ewes exhibit an interesting change in GH/IGF-I endocrinology. These ewes have decreased maternal serum GH while at the same time have increased serum IGF-I (Wallace et al. 1997). This indicates a disconnection of the normal positive endocrinology of GH and IGF-I. Therefore, caution should be used when interpreting the results of maternal oGH infusion in this model. Maternal infusion of GH maternally may have resulted in increases in placental weight; however, this may represent a resetting of the normal GH/IGF-I system thus providing a repartitioning of nutrients toward the pregnancy to normalize placental and fetal growth.
In our research, maternal GH treatment increased (P<0.001) maternal serum concentrations of GH and IGF-I tenfold higher than CON ewes, by the fifth day of treatment, and maintained the increase throughout treatment. The peak GH concentration was similar to that reported by Jenkinson et al. (1999) while the peak IGF-I concentration was similar to Stelwagen et al. (1994) and Jenkinson et al. (1999). Although the ewes responded to GH treatment with increased maternal serum IGF-I, exogenous GH treatment had no effect on uterine, placental, and fetal weight or fetal organ weights measured at either day 55 or 135 of gestation. While the model systems are different, our results none the less agree with the recent reports of GH administration during early gestation (Wallace et al. 2006). The study by Wallace et al. (2006) did not administer GH to the moderate intake control fed adolescent ewes, and the GH-treated high intake adolescent ewes carrying singletons, confounding the comparison between their study and ours. However, taken together, these results support the conclusion that elevated maternal serum GH and IGF-I concentrations during the normal expression window of oGH have little impact on placental and fetal growth. We specifically chose twin pregnancies to evaluate, as the multiple fetuses put increased demand on the maternal system and placenta to provide sufficient nutrition to the fetus. The only report of GH administration to ewes carrying twins was that of Jenkinson et al. (1999), in which ewes were treated from 98 to 111 dGA. A significant increase in fetal weight was reported (Jenkinson et al. 1999), but GH treatment had similar effects on the growth of both singles and twins. The response seen by Jenkinson et al. may well be due to partitioning of maternal nutrients toward the pregnancy late in gestation versus an alteration of placental development.
In an effort to gain insight into the potential effects of GH on gene expression within the placenta, we used northern hybridization and qPCR to quantify GH, oPL, IGF-I, IGF-II, and IGFBP-1, -2, -3, and -4 mRNA concentrations in day 55 caruncle and cotyledon tissue. Maternal GH treatment had no effect on GH expression in the caruncle tissue, nor did GH impact oPL expression in cotyledon tissue. In placental growth, IGF-II plays a crucial role as evidenced by growth restriction in knockout and targeted disruption studies (DeChiara et al. 1990, Constancia et al. 2002). Had the exogenous GH significantly changed the IGF-II levels in the placenta, it would be expected that a change in placental growth would be evident. The role of IGF-I may be as a less potent stimulator of proliferation that has stimulatory effects on trophoblast differentiation and migration (Maruo et al. 1995, Lacey et al. 2002). With no gross changes in placental weight and fetal weight evident at necropsy, the lack of effect on IGF mRNA concentrations is not surprising.
By contrast, within the cotyledon, maternal GH treatment increased expression of IGFBP-1 and -4, while IGFBP-2 expression was decreased and IGFBP-3 was unchanged. Within the caruncle exogenous GH increased expression of IGFBP-3 and -4, while decreasing the expression of IGFBP-1. In the caruncle, IGFBP-2 was unaffected by maternal GH treatment. Previously, it was believed that IGFBP-1 was not expressed beyond 21 dGA (Wathes et al. 1998, de Vrijer et al. 2006); however, here we could detect changes in IGFBP-1 by treatment at 55 dGA. We do not discount the previous results, as our detection of IGFBP-1 is most likely due to the increased sensitivity of qPCR versus hybridization technologies in detecting transcripts. The decrease in caruncle expression of IGFBP-1 can be attributed to two possible mechanisms. First, IGFBP-1 is inversely regulated by insulin (Suikkari et al. 1989), and it has been shown previously that GH administration during gestation leads to hyperinsulinemia (Wallace et al. 2004). Given that we achieved significant stimulation of IGF-I similar to that reported by Wallace et al. one could reason that we generated a similar insulin response that reduced caruncle expression of IGFBP-1. Second, it has been shown that IGF-II can regulate IGFBP-1 expression at the maternal interface to control trophoblast invasion (Hamilton et al. 1998, Irwin et al. 2001). Although we did not measure maternal serum IGF-II concentrations, an increase of serum IGF-II due to GH treatment could also reduce the IGFBP-1 expression in the caruncle. The increase in caruncular IGFBP-3 agrees with reports that both GH and IGF-I are stimulatory to IGFBP-3 expression as reviewed by Clemmons (1997). The increased expression of IGFBP-4 is not surprising, as it has been reported to be a negative regulator of IGF-I action in other tissues (Clemmons 1997). Within the caruncle, IGFBP-4 may be upregulated to mitigate the effects of increased serum IGF-I.
Within the cotyledon, it is more difficult to interpret the changes in IGFBP-1, -2, and -4. Given the low amount of blood in a day 55 fetus, it is not feasible to obtain serum analysis of GH, IGF-I, IGF-II, and insulin in fetal circulation. These hormones, as discussed above, regulate the expression of IGFBPs and thus determination of their concentration would be necessary to give accurate interpretation of the transcriptional changes within the placenta. This is the first evidence of the transcriptional changes caused by GH administration during the normal placental oGH expression window within the placenta. While we did not alter the local expression of oGH in the placenta, this may be indicative of alternative regulation of oGH in the placenta versus the pituitary. Our evidence suggests that increased serum GH does not cause the normal negative feedback on GH expression in the placenta that is seen in the pituitary. Cloning and analysis of the complete promoter sequence for oGH-1 and oGH-2 may lead to discovery of differential regulation of the promoters. Heterologous studies with the human GH genes revealed that hGH-V lacked a proximal pituitary-specific protein (Pit-1) site that is one of two present in the hGH-N promoter (Nickel et al. 1991). Furthermore, we could not detect expression of Pit-1 in maternal- or fetal-derived placental tissues (data not presented). This indicates a tissue-specific transcriptional regulation for human GH in the placenta when compared with the pituitary. With the lack of placental response to maternal GH treatment, further research to elucidate the role of placental oGH will likely necessitate a knockdown of oGH expression to gain insight into oGH function within the placenta.
Declaration of Interest
The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
Funding
This research was supported in part by a grant from Colorado State University, College of Veterinary Medicine and Biomedical Sciences, College Research Council, and National Research Initiative Competitive Grant no. 9902267 from the USDA Cooperative State Research, Education, and Extension Service.
Acknowledgements
The authors would like to acknowledge Drs Jann Rhodes, Kimberly Jeckel, and Abraham Schaeffer, and Ms Meridith Davidsen for assistance with this research. We would also like to thank Dr Dennis Hallford at New Mexico State University for the analysis of serum GH concentrations.
References
Anthony RV, Scheaffer AN, Wright CD & Regnault TRH 2003 Ruminant models of prenatal growth restriction. Reproduction 61 183–194.
Clemmons DR 1997 Insulin-like growth factor binding proteins and their role in controlling IGF actions. Cytokine and Growth Factor Reviews 8 45–62.
Constancia M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, Stewart F, Kelsey G, Fowden A & Sibley C et al. 2002 Placental-specific IGF-2 is a major modulator of placental and fetal growth. Nature 417 945–948.
Currie MJ, Bassett NS, Breier BH, Klempt M, Min SH, Mackenzie DD, McCutcheon SN & Gluckman PD 1996 Differential effects of maternal ovine placental lactogen and growth hormone (GH) administration on GH receptor, insulin-like growth factor (IGF)-1 and IGF binding protein-3 gene expression in the pregnant and fetal sheep. Growth Regulation 6 123–129.
Davies J & Wimsatt WA 1966 Observation on the fine structure of the sheep placenta. Acta Anatomica 65 182–223.
DeChiara TM, Efstratiadis A & Robertson EJ 1990 A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 345 78–80.
Erickson-Hagen AS, Orbus RJ, Wilkening RB, Regnault TRH & Anthony RV 2005 Placental expression of angiopoietin-1, angiopoietin-2 and Tie-2 during placental development in an ovine model of placental insufficiency-fetal growth restriction. Pediatric Research 58 1228–1232.
Gatford KL, Owens JA, Campbell RG, Boyce JM, Grant PA, De Blasio MJ & Owens PC 2000 Treatment of underfed pigs with GH throughout the second quarter of pregnancy increases fetal growth. Journal of Endocrinology 166 227–234.
Gootwine E, Sise JA, Penty JM & Montgomery GW 1993 The duplicated gene copy of the ovine growth hormone gene contains a PvuII polymorphism in the second intron. Animal Genetics 24 319–321.
Hamilton GS, Lysiak JJ, Han VK & Lala PK 1998 Autocrine-paracrine regulation of human trophoblast invasiveness by insulin-like growth factor (IGF)-II and IGF-binding protein (IGFBP)-1. Experimental Cell Research 244 147–156.
Harding JE, Evans PC & Gluckman PD 1997 Maternal growth hormone treatment increases placental diffusion capacity but not fetal or placental growth in sheep. Endocrinology 138 5352–5358.
Hoefler WC & Hallford DM 1987 Influence of suckling status and type of birth on serum hormone profiles and return to estrus in early-postpartum spring-lambing ewes. Theriogenology 27 887–895.
Irwin JC, Suen LF, Faessen GH, Popovici RM & Giudice LC 2001 Insulin-like growth factor (IGF)-II inhibition of endometrial stromal cell tissue inhibitor of metalloproteinase-3 and IGF-binding protein-1 suggests paracrine interactions at the decidua: trophoblast interface during human implantation. Journal of Clinical Endocrinology and Metabolism 86 2060–2064.
Jenkinson CM, Min SH, Mackenzie DD, McCutcheon SN, Breier BH & Gluckman PD 1999 Placental development and fetal growth in growth hormone-treated ewes. Growth Hormone and IGF Research 9 11–17.
Kappes SM, Warren WC, Pratt SL, Liang R & Anthony RV 1992 Quantification and cellular localization of ovine placental lactogen messenger ribonucleic acid expression during mid- and late gestation. Endocrinology 131 2829–2838.
Klempt M, Bingham B, Breier BH, Baumbach WR & Gluckman PD 1993 Tissue distribution and ontogeny of growth hormone receptor messenger ribonucleic acid and ligand binding to hepatic tissue in the midgestation sheep fetus. Endocrinology 132 1071–1077.
Lacey H, Haigh T, Westwood M & Aplin JD 2002 Tissue distribution and ontogeny of growth hormone receptor messenger ribonucleic acid and ligand binding to hepatic tissue in the midgestation sheep fetus. BMC Developmental Biology 2 5.
Lacroix MC, Devinoy E, Servely JL, Puissant C & Kann G 1996 Expression of the growth hormone gene in ovine placenta: detection and cellular localization of the protein. Endocrinology 137 4886–4892.
Lacroix MC, Devinoy E, Cassy S, Servely JL, Vidaud M & Kann G 1999 Expression of growth hormone and its receptor in the placental and feto–maternal environment during early pregnancy in sheep. Endocrinology 140 5587–5597.
Liang R, Limesand SW & Anthony RV 1999 Structure and transcriptional regulation of the ovine placental lactogen gene. European Journal of Biochemistry 265 883–895.
Limesand SW & Anthony RV 2001 Novel activator protein-2α splice-variants function as transactivators of the ovine placental lactogen gene. European Journal of Biochemistry 268 2390–2401.
Maruo T, Murata K, Matsuo H, Samoto T & Mochizuki M 1995 Insulin-like growth factor-I as a local regulator of proliferation and differentiated function of the human trophoblast in early pregnancy. Early Pregnancy 1 54–61.
Nickel BE, Nachtigal MW, Bock ME & Cattini PA 1991 Differential binding of rat pituitary-specific nuclear factors to the 5′-flanking region of pituitary and placental members of the human growth hormone gene family. Molecular and Cellular Biochemistry 106 181–187.
Ofir R & Gootwine E 1997 Ovine growth hormone gene duplication – structural and evolutionary implications. Mammalian Genome 8 770–772.
Rozance PJ, Limesand SW, Zerbe GO & Hay WW 2007 Chronic fetal hypoglycemia inhibits the later steps of stimulus-secretion coupling in pancreatic beta-cells. American Journal of Physiology 292 E1256–E1264.
Spencer TE, Gray A, Johnson GA, Taylor KM, Gertler A, Gootwine E, Ott TL & Bazer FW 1999 Effects of recombinant ovine interferon tau, placental lactogen, and growth hormone on the ovine uterus. Biology of Reproduction 61 1409–1418.
Stegeman JHJ 1974 Placental development in the sheep and its relation to fetal development. Bijdragen Tot de Dierkunde 44 3–72.
Stelwagen K, Grieve DG, Walton JS, Ball JL & McBride BW 1994 Effect of bovine somatotropin administration during the last trimester of gestation on maternal growth, and fetal and placental development in primigravid ewes. Animal Production 58 87–94.
Sterle JA, Cantley TC, Lamberson WR, Lucy MC, Gerrard DE, Matteri RL & Day BN 1995 Effects of recombinant porcine somatotropin on placental size, fetal growth, and IGF-1 and IGF-2 concentrations in pigs. Journal of Animal Science 73 2980–2985.
Suikkari AM, Koivisto VA, Koistinen R, Seppala M & Yki-Jarvinen H 1989 Dose–response characteristics for suppression of low molecular weight plasma insulin-like growth factor-binding protein by insulin. Journal of Clinical Endocrinology and Metabolism 68 135–140.
Valinsky A, Shani M & Gootwine E 1990 Restriction fragment length polymorphism in sheep at the growth hormone locus is the result of variation in gene number. Animal Biotechnology 1 135–144.
de Vrijer B, Davidsen ML, Wilkening RB, Anthony RV & Regnault TRH 2006 Altered placental and fetal expression of IGFs and IGF-binding proteins associated with intrauterine growth restriction in fetal sheep during early and mid-pregnancy. Pediatric Research 60 507–512.
Wallace JM, Da Silva P, Aitken RP & Cruickshank MA 1997 Maternal endocrine status in relation to pregnancy outcome in rapidly growing adolescent sheep. Journal of Endocrinology 155 359–368.
Wallace JM, Milne JS & Aitken RP 2004 Maternal growth hormone treatment from day 35 to 80 of gestation alters nutrient partitioning in favor of uteroplacental growth in the overnourished adolescent sheep. Biology of Reproduction 70 1277–1285.
Wallace JM, Matsuzaki M, Milne J & Aitken R 2006 Late but not early maternal growth hormone treatment increases fetal adiposity in overnourished adolescent sheep. Biology of Reproduction 75 231–239.
Warren WC, Liang R, Krivi GG, Siegel NR & Anthony RV 1990 Purification and structural characterization of ovine placental lactogen. Journal of Endocrinology 126 141–149.
Wathes DC, Reynolds TS, Robinson RS & Stevenson KR 1998 Role of insulin-like growth factor system in uterine function and placental development in ruminants. Journal of Dairy Science 81 1778–1789.
