SEARCH

SEARCH BY CITATION

Keywords:

  • Fetus;
  • Pth/pthrp;
  • Bone mineralization;
  • Ion Transport/placenta;
  • Knockout;
  • Animal Models/rodent;
  • Growth and Development

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Parathyroid hormone (PTH) plays an essential role in regulating calcium and bone homeostasis in the adult, but whether PTH is required at all for regulating fetal-placental mineral homeostasis and skeletal development is uncertain. We hypothesized that despite its low circulating levels during fetal life, PTH plays a critical role in regulating these processes. To address this, we examined two different genetic models of PTH deficiency. Pth null mice have enlarged parathyroids that are incapable of making PTH, whereas Gcm2 null mice lack parathyroids but have PTH that arises from the thymus. Pth nulls served as a model of complete absence of PTH, whereas Gcm2 nulls were a model of severe hypoparathyroidism. We determined that PTH contributes importantly to fetal mineral homeostasis because in its absence a fetal hypoparathyroid phenotype results with hypocalcemia, hypomagnesemia, hyperphosphatemia, low amniotic fluid mineral content, and reduced skeletal mineral content. We also determined that PTH is expressed in the placenta, regulates the placental expression of genes involved in calcium and other solute transfer, and may directly stimulate placental calcium transfer. Although parathyroid hormone–related protein (PTHrP) acts in concert with PTH to regulate fetal mineral homeostasis and placental calcium transfer, unlike PTH, it does not upregulate in response to fetal hypocalcemia. © 2010 American Society for Bone and Mineral Research


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Parathyroid hormone (PTH) plays an essential role in regulating calcium and bone homeostasis in the adult. Absence or peripheral resistance to its actions causes hypocalcemia, hyperphosphatemia, reduced bone turnover, and calcifications of soft tissues and basal ganglia.1

In contrast, whether PTH is required at all for regulating fetal-placental mineral homeostasis and skeletal development is uncertain. The fetal biochemical milieu differs from the adult and is characterized by elevations of ionized calcium, magnesium, and phosphate above the maternal values and active transfer of calcium, magnesium, and phosphate across the placenta.2 The skeleton rapidly mineralizes during the third trimester in humans and the last 4 to 5 days of gestation in rodents.2 Circulating levels of PTH and 1,25-dihydroxyvitamin D (calcitriol) are low compared with adult values, whereas parathyroid hormone–related protein (PTHrP) is increased in the fetal circulation during late gestation.3 These and other observations prompted the hypothesis that PTHrP might assume the actions of PTH during fetal life, and our previous studies of fetal mineral homeostasis in Pthrp null fetuses examined this possibility.

We found that Pthrp null fetuses had hypocalcemia (equal to maternal blood level), hypomagnesemia, hyperphosphatemia, reduced fetal-placental calcium transfer, normal amniotic fluid calcium content, and normal skeletal mineral content.4, 5 Serum PTH was increased threefold in Pthrp null fetuses compared with wild-type (WT) fetuses and may have prevented more severe hypocalcemia.5 However, the biochemical abnormalities in Pthrp null fetuses indicate either that PTH could not fully compensate for the absence of PTHrP or that it was restrained by the actions of the parathyroid calcium-sensing receptor (CaSR) to maintain the lower, adult-normal value of serum calcium.6 These observations reaffirmed the hypothesis that PTHrP is an important regulator of fetal mineral homeostasis, but further study of the role of fetal PTH was warranted.

We next examined Hoxa3 null fetuses as a model for aparathyroidism; these mice also lack the thymus and are completely devoid of PTH.7 We found more profound hypocalcemia than in Pthrp null fetuses such that the blood calcium level was reduced well below the ambient maternal calcium concentration.5, 8Hoxa3 nulls also had hypomagnesemia, hyperphosphatemia, a normal rate of placental calcium transfer, low amniotic fluid mineral content, and reduced skeletal calcium and magnesium content. We also created and studied Hoxa3/Pthrp double mutants that lacked both PTH and PTHrP. These double-mutant fetuses had much lower blood calcium and skeletal mineral content than either of the single mutants; their phenotype was similar to what we observed in mice lacking the PTH/PTHrP receptor.4, 5

Thus our previous investigations showed that absence of parathyroids caused a more substantial reduction in serum calcium and skeletal mineral content than absence of PTHrP despite the fact that PTH normally circulates at low levels in the fetus. Removing both parathyroids and PTHrP (or the PTH/PTHrP receptor) caused even more severe abnormalities, suggesting that PTH can partly compensate for the absence of PTHrP or that both PTH and PTHrP normally contribute to the regulation of fetal blood calcium and skeletal mineralization. However, Hoxa3 null mice have other abnormalities in tissues derived from the third pharyngeal arch that contribute to their mortality after birth and which may affect their biochemical and skeletal phenotype. Thus the specific role of PTH in fetal calcium homeostasis required further investigation.

In this study, we hypothesized that despite its low circulating levels, PTH plays a critical role in regulating fetal calcium homeostasis and skeletal mineralization. To address this—and to avoid being led astray by possibly confounding features of one mutant model—we examined two different genetic models of PTH deficiency. Pth null mice have enlarged parathyroids that are incapable of making PTH; the adults display hypocalcemia and hyperphosphatemia.9Gcm2 null mice lack parathyroids but have PTH that arises from the thymus; the adults also have hypocalcemia and hyperphosphatemia.10Pth nulls served as a model of complete absence of PTH, whereas Gcm2 nulls were a model of severe hypoparathyroidism. In both colonies, the pups are born in the expected Mendelian ratios, but survival of null mice is reduced after birth. We have observed that this is due, in part, to the mother, who selectively tosses these otherwise healthy-appearing pups from the nest (even if the investigator puts them back in); there also may be postnatal hypocalcemia-related deaths. In both colonies, the null mice that survive are fertile and grossly indistinguishable from their WT and heterozygous littermates.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Animal husbandry

The creation of Pth null and Gcm2 null mice has been described previously.9, 10 The original strains were back-crossed into Black Swiss (Taconic, Germantown, NY, USA) for three generations prior to beginning any studies and at least annually thereafter such that the mice are about 10 generations into Black Swiss. The colonies were maintained by breeding heterozygous-deleted mice together. Genotyping was done by PCR on DNA extracted from tail clips of weaned pups. Mice were mated overnight; the presence of a vaginal mucus plug on the morning after mating marked embryonic day (ED) 0.5. Normal gestation in these mice is 19 days. All mice were given a standard chow (1% calcium, 0.75% phosphorus) diet and water ad libitum. All studies were performed with the prior approval of the Institutional Animal Care Committee of Memorial University of Newfoundland.

Chemical and hormone assays

Whole blood, plasma, serum, and amniotic fluid were collected using methods described previously.8 Ionized calcium was measured on whole blood using a Chiron Diagnostics 634 Ca2+/pH Analyzer (Chiron Diagnostics, East Walpole, MA). Total calcium, phosphate, and magnesium were measured using colorimetric assays in the Gcm2 colony (Sigma-Aldrich, Oakville, Ontario, Canada); discontinuation of these kits necessitated that different colorimetric assays had to be used in the Pth colony (Diagnostic Chemicals Limited, Charlottetown, Prince Edward Island, Canada). PTH was measured with a rodent PTH 1-34 Elisa kit that has a detection limit of 1.6 pg/mL (Immutopics, San Clemente, CA, USA). Plasma PTHrP was measured using a sensitive RIA with an antibody directed to an amino-terminal epitope on samples that had been collected in a cocktail of aprotinin and EDTA.11 WT fetuses typically have values of 6 to 8 pmol/L, whereas Pthrp null fetuses had values of 3.8 ± 0.3 pmol/L; this likely represents the detection limit of the assay in fetal mouse plasma8 (and unpublished data).

Placental calcium transfer

This procedure has been described in detail elsewhere.4 Briefly, pregnant dams on ED 18.5 were given an intracardiac injection of 50 µCi 45Ca and 50 µCi of 51Cr-EDTA. Five minutes later, the dams were sacrificed, and each fetus was removed from its placenta. The ratio of 45Ca to 51Cr radioactivity within each fetus was measured using a gamma counter and a liquid scintillation counter, respectively. The mean 45Ca/51Cr activity ratio of the heterozygous fetuses in each litter was set at 100% in order that the results from different litters could be pooled for analysis.

Fetal PTH treatment

On ED 18.5, we exposed the uteri of Pth heterozygous dams and gave half the fetuses an intraabdominal injection of 1 nmol rat PTH 1-84 (Bachem, Torrance, CA) in 2 µL saline, whereas the remaining fetuses received 2 µL saline. Sutures were placed over the gestational sacs of the PTH-injected fetuses so that they could be identified later. The mother's incision was closed, and she was permitted to awaken and move about normally. Eighty-five minutes after the fetal injections, the placental calcium transfer procedure described earlier was carried out (ending 90 minutes from the time of PTH or saline injection). In separate experiments (no radioactivity) to determine the effect of PTH treatment on placental gene expression, at 90 minutes after the fetal injections, the placentas were removed, snap frozen with liquid nitrogen, and stored at −70°C for subsequent RNA extraction, microarray, and real-time quantitative RT-PCR.

Fetal ash and skeletal mineral assay

As described previously,5 intact fetuses (ED 18.5) were reduced to ash in a furnace (500°C × 24 hours). A Perkin Elmer 2380 Atomic Absorption Flame Spectrophotometer (Norwalk, CT) determined the calcium and magnesium content of the ash.

Alizarin red S and alcian blue preparations

As described in detail previously,5 fresh fetuses were fixed in 95% EtOH followed by clearing with acetone, stained with alcian blue 8GS and alizarin red S, and then immersed in 1% aqueous KOH until the fetal skeleton was clearly visible through the surrounding tissue. They were transferred into 100% glycerine for permanent storage.

Histology

Undecalcified fetal tibiae were fixed in paraformaldehyde, dehydrated in graded alcohol series, and embedded in paraffin. Then 5 µm sections were deparaffinized, rehydrated, and transferred to distilled water. For von Kossa staining, the sections were transferred to 1% aqueous silver nitrate solution and exposed for 45 minutes under a strong light. They then were washed thrice in distilled water, placed in 2.5% sodium thiosulfate (5 minutes), and washed thrice again in distilled water. Finally, they were counterstained with methyl green, dehydrated in 1-butanol and xylene, and mounted.

RNA extraction

RNA was extracted from snap-frozen placentas and anterior neck sections using an RNeasy Midi Kit (Qiagen, Valencia, CA). The placental samples were representative of the entire placenta and included the three trophoblast cell types and intraplacental yolk sac; the maternal contribution is limited to endothelial cells that line maternal blood vessels. Quality of the RNA samples used for microarray analysis was assessed with the Agilent 2100 BioAnalyzer (Agilent Technologies, Palo Alto, CA), whereas the quality of other RNA samples was assessed using ultraviolet (UV) spectrophotometry and inspection of ribosomal RNA integrity on the electrophoresced gel.

Microarray

Placental RNA from PTH- or saline-treated fetuses was analyzed at the Centre for Applied Genomics, Microarray Facility, Hospital for Sick Children (Toronto, Ontario, Canada). The Mouse Gene ST 1.0 Array (Affymetrix, Santa Clara, CA) was completed on 12 samples (one sample per chip), representing three samples for each of the four groups: WT and Pth null, each treated with either saline or PTH (1-84).

Primary data analysis at the Statistical Analysis Core Facility of the Centre for Applied Genomics used the March 2008 gene annotation information from Affymetrix. Probesets without gene names/gene assignments were removed, leaving 22,158 probesets. Raw data were normalized using the robust multiarray average (RMA) algorithm,12 and differentially expressed genes then were identified using the local-pooled-error test (LPE).13 False discovery rate (FDR)14 was set at 0.05 such that genes with adjusted p values of less than .05 were considered to be statistically significant. The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE16983 (www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE16983).

Real-time quantitative RT-PCR

We used TaqMan Gene Expression Assays, which are predesigned primers and probes for optimal amplification, to determine expression of S100g (CaBP-9K), VDR, PTHrP, PTH, and TRPV6. In addition to the TaqMan Gene Expression Assays, TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA) was used when PTHrP expression was assessed in both neck and placental RNA samples. Details of conditions and cycle times have been reported previously.15 To assess S100g (CaBP-9K), VDR, PTH, and TRPV6 expression in placental samples, we used the TaqMan RNA-to-CT 1-Step Kit (Applied Biosystems) in addition to the TaqMan Gene Expression Assays. This eliminated the need to carry out a separate cDNA synthesis step prior to real-time quantitative RT-PCR. With use of the TaqMan RNA-to-CT 1-Step Kit, the thermal cycler protocol then consisted of a 15 minute cycle at 48°C, a 10 minute cycle at 95°C, followed by 40 cycles of 15 seconds at 95 °C and 1 minute at 60°C. We performed all real-time quantitative RT-PCR using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems), as described previously.16 All samples were analyzed in triplicate. Relative expression ratios were representative of the threshold cycle (the PCR cycle at which an increase in reporter fluorescence is above a baseline signal) normalized to GAPDH and compared with WT animals.

Statistical analysis

Data were analyzed using SYSTAT 5.2.1 for Macintosh (SYSTAT, Inc., Evanston, IL, USA). ANOVA was used for the initial analysis; a post hoc test was used to determine which pairs of means differed significantly from each other. Real-time PCR results were analyzed by the 2−ΔΔCT method, where the target and reference are amplified in separate wells.17 Two-tailed probabilities are reported, and all data are presented as mean ± SE.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Serum chemistries

As expected, serum PTH was undetectable in all Pth null fetuses (Fig. 1A). In Gcm2 null fetuses, the PTH level ranged from low to undetectable, indicating that some fetuses had circulating PTH likely arising from the thymus (see Fig. 1B). Both Pth null and Gcm2 null fetuses shared identical phenotypes of hypocalcemia, hypomagnesemia, and hyperphosphatemia (Table 1); this pattern is very similar to the phenotype of Pthrp null fetuses in the same genetic background. The ionized calcium was reduced to the maternal level, equal to the level of Pthrp null fetuses but well above the level observed in aparathyroid Hoxa3 null fetuses within the same genetic background (see Fig. 1C, D). Amniotic fluid calcium and magnesium were low in both Pth null and Gcm2 null fetuses (see Table 1), which is consistent with reduced renal filtered load from the lower serum levels of these minerals.

thumbnail image

Figure 1. Lack of PTH causes hypocalcemia but no compensatory increase in PTHrP in either Pth null or Gcm2 null fetuses. Serum PTH was undetectable (<1.6 pg/mL) in all Pth nulls (A) and low but detectable (5.0 pg/mL) in Gcm2 nulls (B). Both null genotypes had hypocalcemia (equal to the maternal calcium concentration) (C, D) but no alteration in plasma PTHrP (E, F) and no upregulation of PTHrP mRNA (as detected by real-time quantitative RT-PCR) in the neck region that contains the parathyroids (G, H). Fetuses from the Pth colony are shown in black; those from the Gcm2 colony are shown in gray. Numbers in parentheses indicate the numbers of pups studied. Dashed lines indicate the mean maternal concentrations for each parameter, which were collected and measured simultaneously with the fetal samples.

Download figure to PowerPoint

Table 1. Biochemical Weight and Skeletal Mineral Measurements from Pth and Gcm2 Null Fetuses Versus Respective Siblings and Mothers
 Pth ColonyGcm2 Colony
WT±NullMaternalWT±NullMaternal
  • Note that the kits used to measure total Ca, Mg, and PO4 in the Pth colony were not the same as the kits used for the Gcm2 colony because the manufacturer discontinued the original kits; thus the results for total Ca, Mg, and PO4 are not directly comparable between the two colonies.

  • *

    p < .001;

  • **

    p < .01;

  • ***

    p < .08 versus respective heterozygous siblings.

Ca2+ (mM)1.75 ± 0.051.74 ± 0.031.31 ± 0.04*1.21 ± 0.061.77 ± 0.021.82 ± 0.021.34 ± 0.03*1.31 ± 0.03
Mg (mM)1.17 ± 0.131.32 ± 0.071.11 ± 0.10***0.80 ± 0.090.90 ± 0.010.92 ± 0.010.87 ± 0.01**0.88 ± 0.02
PO4 (mM)2.97 ± 0.173.03 ± 0.113.54 ± 0.11**2.45 ± 0.203.23 ± 0.263.32 ± 0.164.57 ± 0.15*3.18 ± 0.17
Amniotic total Ca (Mm)2.19 ± 0.142.14 ± 0.101.50 ± 0.15**2.35 ± 0.172.23 ± 0.101.75 ± 0.13**
Amniotic Mg (mM)1.44 ± 0.091.35 ± 0.070.88 ± 0.12**1.15 ± 0.071.20 ± 0.040.90 ± 0.07**
Weight (g)1.03 ± 0.031.02 ± 0.021.03 ± 0.031.03 ± 0.031.05 ± 0.021.03 ± 0.02
Ash (mg)19.1 ± 0.519.7 ± 0.418.1 ± 0.5*20.1 ± 0.521.1 ± 0.318.0 ± 0.4*
Calcium (mg/g ash)69.5 ± 2.569.4 ± 1.759.6 ± 2.3*58.9 ± 2.264.9 ± 1.247.8 ± 1.6*
Magnesium (mg/g ash)24.3 ± 1.022.4 ± 0.719.8 ± 0.9*21.0 ± 0.720.5 ± 0.418.5 ± 0.5*

Response of PTHrP

We had observed previously that PTH upregulates in the absence of PTHrP (i.e., in Pthrp null fetuses); we now examined whether PTHrP upregulates in response to hypocalcemia and absence of PTH. Circulating plasma PTHrP levels were no different from respective WT and heterozygous littermates (see Fig. 1E, F). The lack of a difference is likely real because this assay previously distinguished Pthrp null fetuses from WT fetuses and also found very high (63.5 pmol/L) PTHrP levels in PTH/PTHrP receptor (Pthr1) null fetuses.8 However, in order to rule out a local increase in PTHrP within fetal parathyroids, we extracted RNA from anterior neck sections that included the parathyroids and found no elevation in PTHrP mRNA by quantitative real-time RT-PCR (see Fig. 1G, H).

Skeletal phenotype

Both Pth null and Gcm2 null fetuses showed a grossly normal skeletal phenotype, as shown by alizarin red S– and alcian blue–stained intact specimens in Fig. 2(A, B). In an earlier report, the tibial diaphysis was significantly shortened in Pth nulls from an inbred C57BL/6 background,9 but we observed that long bone and tibial diaphyseal lengths of both Pth null and Gcm2 null fetuses (in an outbred Black Swiss background) were normal at gross and microscopic levels (see Fig. 2A–D). Body weights of the Pth null and Gcm2 null fetuses were no different from those of their respective WT siblings (see Table 1). Histologic sections demonstrated normal endochondral bone development with no alteration in the length or morphology of the cartilaginous zones and the growth plate (see Fig. 2C, D). Von Kossa staining suggested a modest reduction in skeletal mineral content that was confirmed by ash weight and mineral content measurements in both phenotypes (see Fig. 2C, D and Table 1). These reductions in skeletal calcium and magnesium content were about half of what we had observed previously in Hoxa3 null fetuses.5

thumbnail image

Figure 2. Gross and microscopic skeletal morphology of ED 18.5 WT, Pth null, and Gcm2 null fetuses. Skeletal preparations stained with alizarin red S (for mineral) and alcian blue (for cartilage) show that both Pth null fetuses (A) and Gcm2 null fetuses (B) had normal axial and appendicular skeletons, including lengths of long bones and mineralization pattern. Panels C and D display von Kossa–stained tibial sections that were counterstained with methyl green. Pth null and Gcm2 null tibiae showed apparently normal endochondral development with no alteration in the lengths or cellular morphology of the cartilaginous or bony compartments, although less mineral (detected by von Kossa) appeared to be present in both null genotypes.

Download figure to PowerPoint

Placental calcium transfer

Fetal hypocalcemia might be caused by a reduction in maternal-fetal calcium flux; conversely, fetal hypocalcemia might induce a compensatory increase in maternal-fetal calcium transport. However, in Pth null fetuses, no alteration in placental 45Ca transfer was noted (Fig. 3A). In Gcm2 null fetuses, a modest but statistically significant increase in 45Ca transfer was detected (see Fig. 3B). To test whether PTHrP might explain this increase in placental calcium transfer, we assayed placental PTHrP mRNA by quantitative real-time RT-PCR and found no increase in PTHrP in either Pth null or Gcm2 null fetuses (see Fig. 3C, D). As noted earlier, circulating PTHrP levels were not increased in Gcm2 nulls either.

thumbnail image

Figure 3. Placental calcium transfer and placental PTHrP mRNA. After administration of 45Ca and 51Cr-EDTA to the mother, Pth null fetuses showed no alteration in the relative transfer of 45Ca compared with WT and Pth+/− littermates (A). Conversely, Gcm2 null fetuses showed a significant upregulation in 45Ca accumulation (B). This increase in placental calcium transfer was not due to a compensatory increase in PTHrP. Placental PTHrP mRNA was unaltered in Pth null (C) and Gcm2 null (D) fetuses, as assessed by real-time quantitative RT-PCR. Figure 1(E, F) showed that plasma PTHrP also was unaltered in either genotype. Fetuses from the Pth colony are shown in black; those from the Gcm2 colony are shown in gray. Numbers in parentheses indicate the numbers of pups studied.

Download figure to PowerPoint

Role of PTH in placental calcium transfer

Whether PTH contributes to the regulation of placental calcium transfer has been unclear, whereas midmolecular fragments of PTHrP have been shown to stimulate this process in fetal lambs and mice.4, 18, 19 Our previous studies in Pthrp null fetuses had shown no effect of PTH treatment, but those fetuses already had threefold upregulation of endogenous PTH and therefore might have been unable to respond to administration of exogenous PTH.4, 5 We used Pth null fetuses as a model to test the ability of PTH treatment to increase the rate of maternal-fetal calcium transfer. We treated all fetuses in utero with injections of either saline or a dose of PTH 1-84 that was equimolar to the dose of PTHrP 1-86 that had proved effective in Pthrp null fetuses. We then assayed placental calcium transfer 90 minutes after treatment. A statistically significant increase in placental calcium transfer occurred in Pth null fetuses (Fig. 4). This increase in 45Ca accumulation was not secondary to altered systemic calcium homeostasis because PTH treatment did not alter the fetal ionized calcium (1.29 ± 0.09 mM in PTH-treated versus 1.31 ± 0.10 mM in saline-treated Pth null fetuses, p = NS).

thumbnail image

Figure 4. PTH 1-84 stimulates placental calcium transfer in Pth null fetuses. Treatment in utero with 1 nmol PTH 1-84 versus saline resulted in a significant increase in the transfer of 45Ca to Pth null fetuses. Saline-injected fetuses are shown in black bars; PTH 1-84–injected fetuses are shown in hatched bars. Numbers in parentheses indicate the numbers of pups studied.

Download figure to PowerPoint

To identify possible mechanisms through which PTH might be acting on the placenta, we extracted RNA from placentas of PTH- or saline-injected WT and Pth null fetuses and performed a genome-wide microarray. At baseline, Pth null placentas had a 60% reduction in TRPV6 mRNA, a 40% reduction in calbindin D-9K mRNA, and a 20% reduction in VDR mRNA versus WT placentas; the expression of other genes involved in cation and solute transport also was significantly reduced (Table 2). The expression of TRPV6, calbindin D-9K, and VDR mRNAs was assessed independently by quantitative real-time RT-PCR, and each was reduced by 60% to 80% in the Pth null versus WT placentas (Fig. 5A–C). In response to PTH administration, Pth null placentas had a 1.8-fold increase in VDR mRNA, and the expression of several solute carriers also increased (Table 3). Quantitative real-time RT-PCR confirmed that the VDR mRNA had increased 1.5-fold over baseline in response to PTH 1-84 treatment (see Fig. 5D).

Table 2. Selected List of Genes Showing Statistically Significant Differential Regulation at Baseline (i.e., Saline Injection) Between Pth Null Versus WT Placentas
GeneGene nameFold changeAdjusted p value
Fabp1Fatty acid binding protein 12.11.013
Hbb-yHemoglobin Y, beta-like embryonic chain1.88.001
Krt1Keratin 11.54.049
Atp7bATPase, Cu2+ transporting, beta-polypeptide0.96.014
Osbpl6Oxysterol binding protein-like 60.84.002
AfpAlpha-fetoprotein0.84<.001
Lrp2 (megalin)Low-density lipoprotein receptor–related protein 20.83<<.001
Dab2Disabled homologue 2 (Drosophila)0.82<<.001
8430408G22RikRIKEN cDNA 8430408G22 gene0.82.044
Slc27a2Solute carrier family 27 (fatty acid transporter)0.81<<.001
Ahsgα-2-HS-glycoprotein0.81<.001
VdrVitamin D receptor0.81<<.001
Apoc1Apolipoprotein C-I0.80.019
HephHephaestin0.80<<.001
Abcc2ATP-binding cassette0.79<.001
Slc13a3Solute carrier family 13 (sodium-dependent dicarboxylate transporter)0.76.016
Slc7a9Solute carrier family 7 (cationic amino acid transporter0.76<<.001
Apoa2Apolipoprotein A-II0.76<<.001
CubnCubilin (intrinsic factor-cobalamin receptor)0.75<<.001
2010003K11RikRIKEN cDNA 2010003K11 gene0.75.002
ApomApolipoprotein M0.75<<.001
TrfTransferrin0.75<<.001
5033414D02RikRIKEN cDNA 5033414D02 gene0.74<.001
Slc22a2Solute carrier family 22 (organic cation transporter)0.73.001
Apoc2Apolipoprotein C-II0.72<<.001
ApobApolipoprotein B0.72<<.001
Rbp2Retinol-binding protein 20.71<<.001
Ambpα1-Microglobulin0.70<.001
Slc6a19Solute carrier family 6 (neurotransmitter transporter)0.69<.001
Slc5a1Solute carrier family 5 (sodium/glucose cotransporter)0.69<<.001
Apoa1Apolipoprotein A-I0.68<<.001
Gc (DBP)Group specific component (vitamin D–binding protein)0.67<<.001
1300017J02RikRIKEN cDNA 1300017J02 gene0.66<<.001
TtrTransthyretin0.66<<.001
SfpdSurfactant-associated protein D0.65.003
Slc3a1Solute carrier family 30.64<<.001
Apoa4Apolipoprotein A-IV0.64<<.001
AlbAlbumin0.63<<.001
Mcoln3Mucolipin 30.62.041
S100g (CaBP-D9K)S100 calcium–binding protein G (calbindin D9k)0.58<<.001
Aqp8Aquaporin 80.46<<.001
Trpv6Transient receptor potential cation channel, subfamily V, member 60.40<<.001
thumbnail image

Figure 5. Placental expression of TRPV6, calbindin D-9k, and VDR. Placentas were harvested from PTH 1-84– and saline-treated fetuses and analyzed by genome-wide microarray. Saline-treated Pth null placentas showed significant downregulation versus WT placentas in the expression of mRNAs for TRPV6 (A), calbindin D-9k (B), and VDR (C). Conversely, following PTH 1-84 treatment, Pth null placentas showed significant upregulation of VDR mRNA versus saline-treated Pth null placentas (D). Numbers in parentheses indicate the number of placentas studied.

Download figure to PowerPoint

Table 3. Selected List of Genes Showing Differential Regulation at 90 Minutes Between PTH 1-84–Injected and Saline-Injected Pth Null Placentas
GeneGene nameFold changeAdjusted p value
2010109I03RikRIKEN cDNA 2010109I03 gene3.68<<.001
Slc39a8Solute carrier family 39 (metal ion transporter)2.20<<.001
SfpdSurfactant-associated protein D1.90.067
VdrVitamin D receptor1.79<<.001
AlbAlbumin (Alb)1.78<.001
2210415F13RikRIKEN cDNA 2210415F13 gene1.55<.001
Slc23a3Solute carrier family 23 (nucleobase transporters)1.43.083
Gc (DBP)Group-specific component (vitamin D–binding protein)1.40<<.001
Apoa4Apolipoprotein A-IV1.26<<.001
1300017J02RikRIKEN cDNA 1300017J02 gene1.25.012
Rbp2Retinol-binding protein 21.20<.001
CubnCubilin (intrinsic factor-cobalamin receptor)0.89.001

Comparison of Pth and Gcm2 null placentas and detection of placental PTH

We examined the expression TRPV6, calbindin D-9K, and VDR mRNAs in Gcm2 null placentas and found that TRPV6 and calbindin D-9K were reduced by 80% and 55%, respectively, compared with their WT siblings (p < .001), whereas VDR mRNA was unchanged from the WT value (p = NS; data not shown). The real-time RT-PCR then was repeated using RNA from Gcm2 and Pth null placentas compared side by side. The relative expressions of TRPV6, calbindin D-9K, and VDR all were significantly higher in Gcm2 null placentas than in Pth nulls (Fig. 6A–C).

thumbnail image

Figure 6. Placental expression of TRPV6, calbindin D-9k, VDR, and PTH in Gcm2 versus Pth null placentas. Compared with their respective WT counterparts, Gcm2 nulls had reduced expression of TRPV6 and calbindin D-9k but normal expression of VDR (not shown). When compared simultaneously with Pth null placentas, Gcm2 null placentas had significantly higher expression of TRPV6 (A), calbindin D-9k (B), and VDR (C). WT and Gcm2 null placentas expressed PTH (not shown), and direct comparison of Gcm2 null with Pth null placentas revealed a sevenfold higher expression of PTH mRNA in Gcm2 null placentas (D). Numbers in parentheses indicate the numbers of placentas studied.

Download figure to PowerPoint

Our findings prompted the consideration that PTH might be expressed in the placenta; if so, it should be absent in Pth nulls but present in Gcm2 nulls. By real-time quantitative RT-PCR, PTH mRNA was detected in WT placentas obtained from both colonies; furthermore, PTH expression was 1.2-fold higher in Gcm2 null placentas than in WT placentas (p = NS). Direct comparison of RNA from Gcm2 and Pth null placentas studied side by side by real-time PCR showed PTH expression to be almost sevenfold higher in Gcm2 null placentas than in Pth null placentas (p < .001) (see Fig. 6D). Expression in Pth nulls did not appear until the thirty-first cycle and may indicate a false-positive detection or the presence of maternal sources of PTH mRNA.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

The PTH/PTHrP receptor is well known to mediate many of the N-terminal actions of PTH and PTHrP. An unresolved paradox is how these two ligands can both be present in the fetal circulation but still carry out distinct roles. We undertook the current studies to further clarify the role of PTH, expecting to see a phenotype identical to aparathyroid Hoxa3 null fetuses—that is, marked hypocalcemia (well below the maternal calcium level), hypomagnesemia, hyperphosphatemia, low amniotic fluid mineral content, and a 25% reduction in skeletal calcium content. Instead, we found a milder hypoparathyroid phenotype with modestly reduced blood calcium concentration (equal to the maternal level), hyperphosphatemia, low amniotic fluid mineral content, and a 10% reduction in skeletal calcium content.

The modestly reduced skeletal mineral content of Pth null and Gcm2 null fetuses confirms that PTH is required to achieve normal mineralization of the skeleton prior to term. The role of PTH may be to directly drive skeletal mineralization by stimulating osteoblast function or to maintain the normally high fetal blood calcium concentration and thereby facilitate mineralization of newly formed bone. Absence of PTH did not alter endochondral bone development or limb lengths in either genetic model, which suggests that the role of PTH to facilitate mineralization may be through its role to maintain the serum calcium concentration and not through any effect on chondrocyte or osteoblast physiology. The original report of Pth null fetuses had indicated significant shortening of the tibial diaphysis, whereas we observed no such shortening in Pth or Gcm2 null fetuses9; the difference between that report and this one may be the respective genetic backgrounds of the mice (C57BL/6 versus Black Swiss). The fact that two distinct models of PTH deficiency shared the same skeletal phenotype suggests that it was caused directly by PTH deficiency and not confounded by unknown factors.

In our prior studies, the phenotype of lower blood calcium and reduced skeletal mineral content was progressively more severe in aparathyroid Hoxa3 fetuses and PTH/PTHrP receptor (Pthr1) null fetuses.5 In this study, two different models of PTH deficiency in animals of the same genetic background had modestly reduced blood calcium and skeletal mineral content. The more modest phenotype of the PTH-deficiency models may indicate that parathyroid tissue participates in regulating fetal mineral homeostasis through the release of other as yet unidentified factors. Further comparative study of other parathyroid-deleted mutants (Pax1 and others) and double mutants of Pth and Hoxa3 may reveal why genetic deletion of parathyroids in Hoxa3 nulls or absence of the PTH/PTHrP receptor in Pthr1 nulls caused more severe hypocalcemia and skeletal undermineralization than in Pth null and Gcm2 null fetuses.

In both Pth null and Gcm2 null fetuses there was no increase in PTHrP mRNA expression in the neck or placenta, no increase in plasma PTHrP, and no reduction in placental calcium transfer. Similarly, we reported earlier that despite more profound hypocalcemia, Hoxa3 null fetuses had no upregulation of placental PTHrP mRNA, no increase in plasma PTHrP, and no alteration in placental calcium transfer. These findings indicate that PTHrP must be regulated differently from PTH during fetal life. PTH is regulated by the CaSR on parathyroids, increasing in response to fetal hypocalcemia (such as in Pthrp null fetuses5 and maternal hypocalcemia20) and increasing in response to inactivating mutations of the CaSR.6 In contrast, fetal PTHrP does not respond to any of these stimuli; the only situation in which we have found PTHrP to be increased is in Pthr1 null fetuses, which lack the PTH/PTHrP receptor.8 PTHrP may be produced autonomously by the placenta or regulated by other factors, such as the sensing of the calcium content exchanged across the trophoblasts and intraplacental yolk sac.3, 21

We observed that Pth null placentas had reduced mRNAs for TRPV6, calbindin D-9K, VDR, vitamin D–binding protein, and other solute or cation transporters. The independent effect of some of these alterations can be predicted from our previous studies. First, in collaboration with the Hediger Laboratory, we found that ablation of TRPV6 in mice significantly lowered placental calcium transfer.22 Second, we noted reduced expression of calbindin D-9K in the intraplacental yolk sac in Pthrp nulls, which have reduced placental calcium transfer.4, 21 Third, ablation of VDR (Vdr null fetuses) was associated with normocalcemia but an increased rate of placental calcium transfer and increased placental expression of TRPV6.16 The studies in Vdr null fetuses may indicate that calcitriol and the VDR act as a brake on the rate of placental calcium transfer; in the absence of VDR, TRPV6 expression and placental calcium transfer increase. Taken together, these previous studies predict that some of the altered gene expression in Pth null placentas should increase net placental calcium transfer, whereas other alterations should decrease it. The decreased expression of TRPV6, calbindin D-9K, and VDR (and other factors) may have offset each other to lead to no net change in placental calcium transfer, exactly as we observed in Pth null fetuses.

Although placental calcium transfer was not reduced in the absence of PTH in either Pth null or Gcm2 null fetuses, the blood calcium, amniotic fluid calcium, and skeletal mineral content all were reduced. This suggests that the rate of backflux or reverse flow of calcium from fetus to mother must have been increased in order to account for where the mineral went. The short 5 minute interval between administration of the isotopes and removal of each fetus from its placenta in this technique means that largely forward flow from mother to fetus is measured, whereas backflux is not. Within this time frame, the transferred isotopes become diluted in the total blood volume of each fetus; consequently, only a small amount of isotope can be expected to return via the umbilical artery and across the placenta to the mother within 5 minutes. The effect of backflux became apparent in our previous studies of Pthrp knockout mice, where we observed that the relative difference in placental calcium transfer among WT, Pthrp+/−, and Pthrp null fetuses increased from 5 to 15 to 30 minutes, likely indicating the progressive effect of backflux on the observed net fetal accumulation of isotopes. Consequently, we have since used the 5 minute time point in the placental calcium transfer experiment because this is unlikely to be confounded by backflux of isotope.

The placental calcium-transfer methodology used in our studies has the advantage of studying intact fetuses with very minimal intervention: an intracardiac injection of isotopes given to a pregnant mouse that is anesthetized for less than 30 seconds. Typically, fetuses from 6 to 10 pregnant mice are required to compare the baseline rate of placental calcium transfer among WT, heterozygous, and null fetuses. The number of mice required more than doubles in order to compare the effect of a single active treatment versus saline at one time point, and measurements can be done only at a single time point after injection (it is a terminal experiment). Consequently, it was not feasible to study a time course of PTH injections or determine a full dose-response curve. Thus we chose the dose of PTH and time point based on our earlier studies in which an equimolar dose of PTHrP 1-86 or 67-86 increased placental calcium transfer in Pthrp null fetuses.4 We demonstrated an increase in placental calcium transfer with PTH treatment of Pth null fetuses, which suggests that PTH may contribute to the normal regulation of placental calcium transfer. Our finding is consistent with earlier data that found that PTH 1-34 treatment could increase calcium transport in vesicles created from human syncytiotrophoblast basement membranes23 and with the intense expression of the PTH/PTHrP receptor in the intraplacental yolk sac.21

PTH 1-84 increased placental calcium transfer within 90 minutes of administration. At this early time point, it is likely that PTH acted directly through its receptor to open channels within the calcium-transporting cells of the placenta. Nevertheless, even at 90 minutes, we observed changes in the expression of mRNAs for VDR, vitamin D–binding protein, and several solute transporters. These changes in mRNA expression are unlikely to cause rapid enough changes in protein expression to account for the increased placental calcium transfer observed by 90 minutes, but the observed changes in these mRNAs indicate that PTH can regulate the expression of calciotropic genes and other solute transporters within the placenta. Moreover, the reduced placental expression of TRPV6, calbindin D-9K, VDR, and other solute transporters at baseline in Pth null fetuses is another indication that PTH may play a role in regulating placental function.

Although Pth null and Gcm2 null fetuses were biochemically indistinguishable, the latter did have a low level of circulating PTH and a small increase in the apparent rate of placental calcium transfer. TRPV6, calbindin D-9K, and VDR each had significantly higher expression in Gcm2 null compared with Pth null placentas, and this may in part explain why the rate of placental calcium transfer also differed between the two null genotypes. Moreover, PTH expression was detected in the placentas of WT and Gcm2 null mice. Since the PTH/PTHrP receptor is intensely expressed in the placenta,21 it is possible that PTH acts in a paracrine fashion to regulate placental genes and calcium transport. It is also conceivable that the small amount of plasma PTH in Gcm2 nulls is derived from both thymic and placental sources.

In conclusion, we have determined that PTH contributes importantly to fetal calcium homeostasis because in its absence a fetal hypoparathyroid phenotype results with hypocalcemia, hypomagnesemia, hyperphosphatemia, low amniotic fluid mineral content, and reduced skeletal mineral content. PTH regulates the placental expression of genes involved calcium and other solute transfer and may contribute to the regulation of placental calcium transfer. PTH may contribute to placental gene expression and function both through both endocrine/systemic (parathyroid-derived) and paracrine (placental-derived) pathways. To our knowledge, no fetal or cord blood calcium measurements have been reported from human fetuses that lack parathyroids, such as owing to DiGeorge syndrome. Our results predict that DiGeorge syndrome will cause hypocalcemia in utero and impaired skeletal mineralization.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

We acknowledge additional technical support from Neva J Fudge, Beth J Kirby, and Pat Ho. Dr T John Martin's provision of plasma PTHrP measurements is greatly appreciated. This work was supported by operating grants to CSK from the Canadian Institutes of Health Research. Presented in part at the American Society of Bone and Mineral Research annual meetings and the joint annual meetings of the Canadian Diabetes Association/Canadian Society of Endocrinology and Metabolism. CSS (née Noseworthy) received a Young Investigator Award and a Travel Award from ASBMR in 2004 and 2007, respectively, and a Travel Award from CDA/CSEM in 2007. CSK received a Canadian Institutes of Health Research grant.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  • 1
    Goltzman D, Cole DE. Hypoparathyroidism. In: FavusMJ, ed. Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism 6th ed. Washington: ASBMR Press; 2006; 216219.
  • 2
    Kovacs CS. Skeletal physiology: fetus and neonate. In: FavusMJ, ed. Primer on the Metabolic Bones Diseases and Disorders of Mineral Metabolism 6th ed. Washington: ASBMR Press; 2006; 5055.
  • 3
    Kovacs CS. Fetal mineral homeostasis. In: GlorieuxFH, PettiforJM, JüppnerH, eds. Pediatric Bone: Biology and Diseases. San Diego: Academic Press; 2003; 271302.
  • 4
    Kovacs CS, Lanske B, Hunzelman JL, Guo J, Karaplis AC, Kronenberg HM. Parathyroid hormone-related peptide (PTHrP) regulates fetal-placental calcium transport through a receptor distinct from the PTH/PTHrP receptor. Proc Natl Acad Sci USA, 1996; 93: 1523315238.
  • 5
    Kovacs CS, Chafe LL, Fudge NJ, Friel JK, Manley NR. PTH regulates fetal blood calcium and skeletal mineralization independently of PTHrP. Endocrinology. 2001; 142: 49834993.
  • 6
    Kovacs CS, Ho-Pao CL, Hunzelman JL, et al. Regulation of murine fetal-placental calcium metabolism by the calcium-sensing receptor. J Clin Invest. 1998; 101: 28122820.
  • 7
    Chisaka O, Capecchi MR. Regionally restricted developmental defects resulting from targeted disruption of the mouse homeobox gene hox-1.5. Nature. 1991; 350: 473479.
  • 8
    Kovacs CS, Manley NR, Moseley JM, Martin TJ, Kronenberg HM. Fetal parathyroids are not required to maintain placental calcium transport. J Clin Invest. 2001; 107: 10071015.
  • 9
    Miao D, He B, Karaplis AC, Goltzman D. Parathyroid hormone is essential for normal fetal bone formation. J Clin Invest. 2002; 109: 11731182.
  • 10
    Günther T, Chen ZF, Kim J, et al. Genetic ablation of parathyroid glands reveals another source of parathyroid hormone. Nature. 2000; 406: 199203.
  • 11
    Grill V, Ho P, Body JJ, et al. Parathyroid hormone-related protein: elevated levels in both humoral hypercalcemia of malignancy and hypercalcemia complicating metastatic breast cancer. J Clin Endocrinol Metab. 1991; 73: 13091315.
  • 12
    Irizarry RA, Hobbs B, Collin F, et al. Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP 2003 Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4(2): 24964.
  • 13
    Jain N, Thatte J, Braciale T, Ley K, O'Connell M, Lee JK. Local-pooled-error test for identifying differentially expressed genes with a small number of replicated microarrays. Bioinformatics. 2003; 19: 19451951 .
  • 14
    Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B (Methodological). 1995; 57: 289300.
  • 15
    Woodrow JP, Sharpe CJ, Fudge NJ, Hoff AO, Gagel RF, Kovacs CS. Calcitonin plays a critical role in regulating skeletal mineral metabolism during lactation. Endocrinology. 2006; 147: 40104021.
  • 16
    Kovacs CS, Woodland ML, Fudge NJ, Friel JK. The vitamin D receptor is not required for fetal mineral homeostasis or for the regulation of placental calcium transfer. Am J Physiol Endocrinol Metab. 2005; 289: E133144.
  • 17
    Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2[−ΔΔC(T)] method. Methods. 2001; 25: 402408.
  • 18
    Care AD, Abbas SK, Pickard DW, et al. Stimulation of ovine placental transport of calcium and magnesium by mid-molecule fragments of human parathyroid hormone-related protein. Exp Physiol. 1990; 75: 605608.
  • 19
    Wu TL, Vasavada RC, Yang K, et al. Structural and physiologic characterization of the mid-region secretory species of parathyroid hormone-related protein. J Biol Chem. 1996; 271: 2437124381.
  • 20
    Kovacs CS, Kronenberg HM. Maternal-fetal calcium and bone metabolism during pregnancy, puerperium and lactation. Endocr Rev. 1997; 18: 832872.
  • 21
    Kovacs CS, Chafe LL, Woodland ML, McDonald KR, Fudge NJ, Wookey PJ. Calcitropic gene expression suggests a role for intraplacental yolk sac in maternal-fetal calcium exchange. Am J Physiol Endocrinol Metab. 2002; 282: E721732.
  • 22
    Suzuki Y, Kovacs CS, Takanaga H, Peng JB, Landowski CP, Hediger MA. Calcium TRPV6 is involved in murine maternal-fetal calcium transport. J Bone Miner Res. 2008; 23: 12491256.
  • 23
    Farrugia W, de Gooyer T, Rice GE, Moseley JM, Wlodek ME. Parathyroid hormone(1–34) and parathyroid hormone-related protein(1–34) stimulate calcium release from human syncytiotrophoblast basal membranes via a common receptor. J Endocrinol. 2000; 166: 689695.