Skeletal and Reproductive Abnormalities in Pth-Null Mice
(4773 total words in this text) -
read: 173 times
We have examined the role of parathyroid hormone (PTH) in the postnatal state in a mouse model of PTH-deficiency generated by targeting the Pth gene in ES cells.
Mice homozygous for the ablated allele, when maintained on a normal calcium intake, developed hypocalcemia, hyperphosphatemia, and low circulating 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] levels consistent with primary hypoparathyroidism.
Fertility in mutant females was diminished due to abnormal ovarian function manifested in part by impaired angiogenesis in the developing corpus luteum. Even in the presence of ovarian dysfunction, bone turnover was reduced and trabecular and cortical bone volume were increased in PTH-deficient mice.
When placed on a low calcium diet, fertility in female mice was completely abolished. Moreover, renal 25-hydroxyvitamin D 1 alpha-hydroxylase (Cyp27b1) expression increased despite the absence of PTH, leading to a rise in circulating 1,25(OH)2D3 levels, marked osteoclastogenesis, and profound bone resorption.
These studies demonstrate the dependence of the reproductive and skeletal phenotype in animals with genetically depleted PTH on the external environment as well as on internal hormonal and ionic circulatory factors.
They point to the importance of calcium balance in reproduction and show that while PTH action is the first defense against hypocalcemia, 1,25(OH)2D3 can be mobilized, even in the absence of PTH, to guard against extreme calcium deficiency.
Introduction
Parathyroid hormone (PTH), the major peptide hormone regulator of calcium homeostasis, is produced almost exclusively by the parathyroid glands and is secreted in response to a decrease in extracellular calcium concentration. PTH enters the circulation and interacts with the type 1 PTH receptor (PTHR1) in target tissues, primarily bone and kidney.
This G protein-coupled cell surface receptor recognizes the 1-34 region of PTH, a sequence in which all the classic biological actions of the hormone (stimulation of bone resorption, increase in renal calcium reabsorption, phosphaturia, bicarbonaturia, 1α-hydroxylation of 25-hydroxyvitamin D, and cAMP production) reside.
PTHR1 possesses the unusual property of binding PTH as well as the paracrine factor PTH-related peptide (PTHrP) with nearly equal affinity. PTHrP was initially identified as the factor responsible for humoral hypercalcemia in patients with malignancy.
The capacity of PTHR1 to bind both PTH and PTHrP is based on sequence similarity in the N-terminal portion of these two ligands. Yet, PTHrP is distinct from PTH in many structural features and certain biological effects, particularly in fetal development and physiology.
Targeted disruption of either Pthrp or Pthr1 in mice and defective PTHrP/PTHR1 signaling in man, lead to a form of lethal skeletal dysplasia characterized by decreased proliferation and accelerated differentiation of growth plate chondrocytes.
PTH also interacts with the type 2 PTH receptor (PTHR2), although the natural ligand for this receptor is likely the neuropeptide tuberoinfundibular peptide of 39 residues TIP39, rather than PTH itself.
Characterization of a third PTH receptor with specificity for the carboxyl-terminal region of PTH has also been reported in rat parathyroid cells, osteoblasts, and osteocytes that presumably exerts an antiresorptive effect on bone by impairing osteoclast differentiation.
It would seem, therefore, that several distinct properties could be attributed to PTH, likely mediated by a variety of receptors. Whether these non-classic biological effects of PTH have potential physiological relevance remains to be determined.
To better understand the physiologic actions of PTH on skeletal homeostasis and other biological systems, we have generated mice homozygous for a null Pth allele and examined the consequences associated with PTH deficiency in the post-natal state and the influence dietary calcium has in its absence.
Results
As a first step in generating PTH-null mice, we isolated mouse recombinant genomic DNA encompassing Pth and characterized the genomic organization and nucleotide sequence of the murine gene.
The strategy for ablating Pth focused on replacing the XhoI/ XbaI genomic DNA fragment that encompasses part of exon 3 encoding the mature PTH peptide with the neomycin resistance (neor) selection cassette gene.
Successfully targeted ES clones were used to generate mice heterozygous for the mutation, which were then intercrossed to generate progeny homozygous for the disrupted Pth allele.
Pth-null mice were obtained at nearly the predicted Mendelian frequency and although lethality due to severe hypocalcemia was anticipated, the mice were viable.
Yet, they did exhibit serum biochemical changes characteristic of primary hypoparathyroidism including moderate to severe hypocalcemia, hyperphosphatemia, and decreased serum 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] levels with complete absence of circulating PTH.
Histological examination of the post-natal parathyroids showed massive, diffuse enlargement of the glands consistent with the continuous stimulation by the prevailing hypocalcemia.
PTH immunoreactivity was absent in the homozygous Pth-null glands thereby confirming the successful targeted disruption of the Pth gene.
In contrast, expression of the calcium-sensing receptor (CaSR) protein was not markedly altered.
To delineate possible mechanism(s) that maintain calcemia to levels that sustain survival in the Pth-/- animals, we measured circulating serum levels of PTHrP. These were shown to be equivalent in wild-type and mutant litter mates (1.22 ± 0.06 and 1.12 ± 0.01 pmol/L, respectively).
Moreover, thymectomy failed to reduce the survival of Pth-null mice compared to their normal litter mates (results not shown), indicating that the thymus does not serve as the tissue source of additional potential calcium-regulating factors other than PTH.
Despite abnormalities in mineral homeostasis, mice grew normally and were fertile, although females were much less so than wild-type littermates (approximately 50-60% success rate).
Intrigued by the decreased fertility of the Pth-null females, gross anatomic examination of the uterii and ovaries of 2-month-old mice was undertaken. The reproductive organs of these animals were similar in size to those from wild-type littermates.
Histological assessment disclosed only very modest impairment in the development of the endometrium (data not shown) while in the ovaries, interstitial tissue was diffusely increased with rare corpora lutea.
Associated with these alterations in ovarian morphology was the observation that serum levels of progesterone were significantly decreased in Pth-/- female mice relative to wild-type littermates at the diestrus stage of the estrous cycle (8.2 ± 0.3 and 13.7 ± 3.7 nmol/ L, respectively), while serum estradiol levels were comparable at the estrus stage (464.3 ± 26.6 and 437.1 ± 62.3 pmol/L, respectively).
Male reproductive organs on the other hand were grossly and histologically normal in the mutant animals (data not shown).
Angiogenesis is very active during formation of the corpus luteum.
We therefore examined the Pth-/- ovaries for expression of factors required for the development of this extensive neovascularization.
Immunoreactivity for vascular endothelial growth factor (VEGF) and fibroblast growth factor-2 (FGF2) was diminished in the mutant tissue, implicating these angiogenic proteins as regulators of luteal angiogenesis whose expression is altered by the prevailing hypoparathyroid state.
Because PTH has major effects on bone remodeling, we next studied the skeletons of Pth-/- animals fed a normal calcium diet at 2, 4, 6 and 9 months of age.
Histologically, cartilage development at the growth plate was normal at all times examined, with proper zone organization and adequate mineralization, suggesting that, in the postnatal state, PTH does not play a major role in chondrocyte biology.
However, major bony alterations were observed in the mutant animals.
Trabecular and cortical bone volume was consistently increased (1.8-fold) in these mice compared to sex-matched littermates.
To define the cause of this increase in bone volume, dynamic histomorphometric analysis of bone was undertaken following double administration of calcein.
Endosteal and trabecular Mean Apposition Rate (MAR), a parameter of bone formation, was profoundly decreased in the mutant mice compared with normal littermates and osteoblasts lining bone surfaces were also diminished.
As well, the number of osteoclasts was reduced (50%) in the Pth-null animals, indicating that loss of PTH is associated with a generalized state of low bone turnover.
We next examined whether alterations in the calcium content of the diet consumed could alter the phenotype of the Pth-/- mice.
For these studies, homozygous and heterozygous 2-month-old PTH deficient mice and wild-type litter mates were fed either regular or low calcium diets for 8 weeks at which time, samples of serum and bones were analyzed.
Normal and heterozygous littermates were indistinguishable in all parameters examined (results not shown).
In contrast, Pth-/- mice had persistent hypocalcemia and hyperphosphatemia that remained unaltered despite the dietary modification.
Circulating levels of PTHrP also remained unchanged.
In sharp contrast, serum 1,25(OH)2D3 levels rose markedly in both groups of mice following institution of a low-calcium diet (4.9-fold in wild-type vs 6.0-fold in mutant mice), indicating that mechanisms had been set in motion in the Pth-null animals to overcome the absence of the stimulatory effect of PTH on vitamin D synthesis.
This increase was reflected, in part, by a rise in renal proximal tubule 25-hydroxyvitamin D 1 alpha-hydroxylase (Cyp27b1) transcript levels and protein immunoreactivity, the former being more pronounced in wild-type animals likely due to the prevailing secondary hyperparathyroidism.
Interestingly, under these dietary conditions, female Pth-/- mice lost completely their capacity to conceive, while the reproductive capability of male animals remained unaffected.
The source of calcium mobilized for maintenance of circulating calcium levels on the calcium-deficient diet became apparent when bones from these animals were examined.
Trabecular bone volume in the wild-type lumbar vertebrae was reduced to 54% of the pretreatment content, while in the mutant specimens the corresponding reduction was to 29% of the basal level.
Cortical bone thickness was unaffected in the wild-type mice but significantly decreased in the Pth-null animals (97% vs 56% of pretreatment thickness, respectively).
Because of the changes in bone volume that arose as a consequence of low dietary calcium, we then examined TRAP staining and quantified osteoclast number and size in these animals.
In Pth-/- mice fed the normal calcium diet, these parameters were decreased compared to the wild-type littermates. However, when animals were moved to a low calcium diet for 3 days, TRAP staining and the number and size of osteoclasts increased, considerably more so in the Pth-/- than in wild-type mice.
These observations imply that a low calcium diet can potently increase 1,25(OH)2D3 levels in vivo, even in the absence of PTH, and enhance bone resorption for maintenance of calcium homeostasis, but in the process can impact negatively on bone mass.
Discussion
Our findings in mice with targeted disruption of the Pth gene, demonstrate the profound manner in which developmental changes not only in the internal milieu but also in the external environment can modify the phenotype of animals with a single genetic alteration.
Thus, our previous studies demonstrated that PTH-negative mice, in the protected intra-uterine environment of the fetus, show abnormalities at the chondro-osseous junction of the growth plate and in formation of the primary spongiosa and trabecular bone which point to an anabolic role for PTH at this stage of development.
Our current studies show that this anabolic effect of the hormone is transformed into a catabolic function in the post-natal state and this effect is modulated by both external factors as well as the ambient level of regulatory hormones and ions.
Furthermore, reproductive capacity is also influenced by these considerations.
Thus, in the Pth-/- mice, reproductive organs appear to form normally in utero but in the post-natal state, defective ovarian function was noted with poor development of corpora lutea.
The primary function of the corpus luteum is the secretion of progesterone, which is required for maintenance of normal pregnancy.
The corpus luteum develops from residual follicular granulosal and thecal cells after ovulation, as endothelial cells invade the ovulation site to form an extensive network of neovasculature that supports its rapid growth, which can exceed that of most rapidly growing tumors.
Hence, mediators of angiogenesis play a pivotal role in this process. VEGF, which is highly expressed during formation of the corpus luteum, and FGF2 have been implicated in luteal cell proliferation or turnover during early pregnancy, and may thereby contribute to the maintenance of luteal function, critical for the successful establishment of pregnancy.
We show that VEGF and FGF2 expression was profoundly decreased in the ovaries of the Pth knockout mice, and these alterations negatively influenced neovasculature formation.
Although it is unclear at present whether this diminished expression is a consequence of the lack of PTH per se, the prevailing hypocalcemia or of the decreased circulating 1,25(OH)2D3 levels, the available experimental evidence would support hypocalcemia.
For example, the reproductive dysfunction of the VDR-null mutant mice was corrected following initiation of a high calcium diet.
Moreover, here we demonstrate that the rise of endogenous 1,25(OH)2D3 levels by low calcium dietary intake is insufficient to normalize the defective development of the corpus luteum, which in fact worsens, as fertility is completely lost.
Lastly, the alterations in reproductive function observed in the Pth-/- mice have been described in humans with hypoparathyroidism and can be avoided by the early introduction of vitamin D and calcium replacement in this setting.
This would suggest that PTH deficiency per se is not responsible for the impaired development of the corpus luteum. The skeletal findings in the Pth-null mice are also of interest.
First, they indicate that in the post-natal state PTH does not play a major role in cartilage development or in the development of the primary spongiosa, roles that may be subserved by locally produced PTHrP.
Second, in the presence of a normal calcium diet, the absence of PTH in the post-natal state was notable for markedly decreased bone turnover with resorption being particularly compromised.
This leads to the marked increases in bone volume noted by our histomorphometric measurements. Observations in patients with hypoparathyroidism tend to support these findings.
Increased bone mineral density (10-32%) has been reported in patients with chronic hypoparathyroidism following surgery for either thyroid cancer or hyperparathyroidism. In addition, this condition provides protection against age-related bone loss in postmenopausal women, perhaps due to attenuation of the high turnover bone loss associated with menopause.
Although supplementation with vitamin D and calcium may have contributed to the increased bone mass in these patients, it would appear that high bone mineral density is a feature of hypoparathyroidism per se, as it has also been observed in untreated individuals with the disorder.
Consequently, in the post-natal state, where the maintenance of normal circulating calcium concentration in the organism is to a great extent dependent on more direct access to calcium in the external environment, the function of PTH appears to have evolved in order to primarily defend against decreases in the ambient calcium.
This appears to involve stimulation of Cyp27b1 expression to raise 1,25(OH)2D3 levels, and induction of a catabolic action on bone to maintain normocalcemia.
Our findings in the post-natal Pth-/- mice and clinical observations in hypoparathyroid patients raise the possibility, therefore, that regulation of PTH secretion can provide a novel therapeutic avenue for the treatment of metabolic bone disease.
Preliminary studies in animals tend to add credence to this hypothesis, as daily transient decreases in PTH levels following administration of the calcimimetic NPS R-568, a calcium-sensing receptor agonist, were shown to have an anabolic effect on uremic bones, and to slow the rate of bone loss following ovariectomy.
Finally, our studies of the Pth-null mice exposed to limiting amounts of calcium in the external environment point to additional mechanisms that can be mobilized to retain circulating calcium concentrations.
Thus, in the presence of a low calcium diet, even in the absence of PTH, Cyp27b1 expression in the kidney was increased, circulating 1,25(OH)2D3 concentrations were augmented, bone resorption was enhanced, and the increased bone volume noted in the hypoparathyroid mice on a normal calcium diet was converted to an osteopenic state.
Most likely, limiting amounts of dietary calcium resulted in transient further reduction of the hypocalcemia observed in the hypoparathyroid animals on normal calcium intake.
This initiated Cyp27b1 stimulation, increases in 1,25(OH)2D3, augmented osteoclastogenesis in bone and mobilization of calcium stores from bone.
Hence, a new steady state was reached in which severe hypocalcemia was “re-set” to the moderate levels but at the expense of extreme osteopenia.
This is consistent with previous reports suggesting that extracellular calcium concentrations can, independently of PTH, regulate Cyp27b1 activity in vivo and in vitro.
However, our studies suggest that in the presence of a normal calcium intake, the ensuing moderate hypocalcemia is less effective in enhancing Cyp27b1 expression than in the presence of a reduced calcium intake where more extreme hypocalcemia may transiently exist.
Consequently, the first line of defense in stimulating Cyp27b1 transcription and maintaining a normal circulating calcium concentration appears to be augmentation of PTH levels whereas 1,25(OH)2D3 may be directly mobilized, even in the absence of PTH, as hypocalcemia becomes more extreme.
An additional possibility is that intestinal epithelial cells directly play a pivotal role in defending against a further fall in calcium when dietary calcium is reduced.
It is possible, that enterocytes have the capacity to sense the decreasing levels of dietary calcium intake and, in turn, release a putative signal, perhaps a circulating agent that acts at the level of the kidney to increase Cyp27b1 expression.
A concomitant effect of such a factor on the skeleton to directly promote bone resorption cannot be excluded.
Materials and Methods
Generation of Pth Knockout Mice
To clone the murine Pth gene, a radiolabeled cDNA encoding human PTH was used as probe to screen a 129/Sv mouse genomic DNA library.
Following isolation and characterization of the Pth gene, the targeting vector was generated by introducing a 2.4 kb BamHI/XhoI fragment corresponding to the 5’ region of homology in the XhoI site of the pPNT vector, while a 3.7 kb BamHI fragment corresponding to the 3’ region of homology was ligated in the unique BamHI site of the vector.
The strategy for ablating Pth focused on deleting PTH-encoding sequences from the mouse genome.
In this construct, part of exon 3 encoding the entire sequence of the mature PTH form was replaced by the neomycin resistance (neor) selection gene cassette.
The vector was linearized with NotI and electroporated into D3 ES cells maintained on mitotically-inactivated mouse primary embryonic fibroblasts resistant to G418.
Following G418 (300 µg/ml) and 1-[2-deoxy,2-fluoro-β-D-arabinofuranosyl]-5-iodouracil (FIAU; 0.2 µM) selection, resistant colonies were isolated and the fidelity of the targeting event was verified by Southern blot analysis of genomic DNA.
Appropriately targeted clones were then injected into 3.5-day pc C57B/L6 blastocysts and extensively chimeric male mice were mated to C57B/L6 females.
Following germline transmission of the mutation, mice were bred to generate animals homozygous for the targeted Pth allele.
Animals used in the present studies were obtained following at least six backcrossings into the C57B/L6 background.
Animal Experimentation
All animal experiments were reviewed and approved by the institutional animal care committee. The mice were housed in a 12-h light/12-h dark cycle.
They were maintained in cages with wooden shavings and consumed water and either a normal calcium diet (0.95% calcium, 0.67% phosphorus and 4.5 IU/g vitamin D3; PMI Feeds, Inc., St.
Louis, MO) or a low calcium diet (0.001-0.005% calcium, 0.4% phosphorus and 2.4 IU/g vitamin D3) ad libitum for the indicated time periods.
Fertility in mice was defined as the number of successful pregnancies following visualization of a vaginal plug. The estrous cycle was staged by examining vaginal smears.
Serum Biochemistry
Serum concentrations of calcium and inorganic phosphorus were determined by routine methods using Sigma Diagnostics reagents (Sigma Diagnostics).
Serum intact PTH was measured with an ELISA assay (Immutopics, Inc., San Clemente, CA) while serum PTHrP and 1,25(OH)2D3 determinations were performed using commercially available RIA kits (Nichols Institute Diagnostics, San Clemente, CA and Immunodiagnostic Systems, UK, respectively).
Serum progesterone and estradiol levels were measured with an immunoassay from ADVIA Centaur Immunoassay System (Bayer Diagnostics, Tarrytown, NY).
Histology and Histochemistry
Ovaries, uterii, thyroparathyroidal tissue, femurs, tibiae, and vertebrae were removed from 6-week-old mice (Pth +/+ and Pth–/– taken from the same litter) and fixed in PLP fixative (2% paraformaldehyde containing 0.075 M lysine and 0.01 M sodium periodate solution) overnight at 5oC prior to processing. Occasionally, bones were decalcified in ethylene-diamine tetra-acetic acid (EDTA) glycerol solution for 5-7 days at 5oC.
Tissue samples were dehydrated and embedded in paraffin after which 5 µm sections were cut on a rotary microtome.
The sections were stained with haematoxylin and eosin (H&E), for tartrate-resistant acid phosphatase (TRAP) or immunostained, as described below. Undecalcified bones were embedded in LR White acrylic resin (London Resin Company Ltd, U.K.).
1µm sections were cut on an ultramicrotome and stained for mineral with the von Kossa staining procedure using toluidine blue as counterstain. Calcein labeling was performed by intra-peritoneal injection with 10 µg calcein/g bodyweight (C-0875, Sigma Chemical Co., St. Louis, MO) at 10 and 3 days before sacrifice.
Bones were harvested and embedded in LR White acrylic resin.
Serial sections were cut and the freshly cut surface of each section was imaged using fluorescence microscopy.
The double calcein interlabel width in cortical and trabecular bone was measured using Northern Eclipse v6.0 (Empix Imaging Inc., Mississauga, ON) image software and the mineral apposition rate (MAR=interlabel width/labeling period) was calculated. For immunohistochemistry, paraffin sections of thyroparathyroidal tissue were stained for PTH and calcium-sensing receptor (CaSR) immunoreactivity by the avidin-biotin-peroxidase complex (ABC) technique using goat serum against PTH 1-34 and mouse anti-CaSR monoclonal antibody, as described.
Ovaries were stained for vascular endothelial growth factor (VEGF) and fibroblast growth factor-2 (FGF2) immunoreactivity using goat antiserum against VEGF and rabbit antiserum against FGF2 (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Kidney sections were immunostained for Cyp27b1 using purified rabbit antiserum.
Computer-Assisted Image Analysis
Computer-assisted image analysis was performed, as previously described.
For determining the area of the mineralized and unmineralized matrix, and the number and size of osteoclasts in stained bone sections, images of primary spongiosa and cortical bone were digitally recorded using a rectangular template and three different fields.
In the primary spongiosa, each image was photographed from the edge of the metaphyseal border of the growth plate (i.e., at the level of the zone of vascular invasion).
In cortical bone, images were taken from the diaphyseal bone close to the metaphysis. All digital images were captured with a Sony digital camera at a magnification of x200, producing a field area of 0.4 mm2.
The positive and negative areas staining in trabecular and cortical bone were measured by digital image analysis using Northern Eclipse v6.0 image software.
Northern Blot Analysis
A cDNA fragment corresponding to nucleotides 421-1474 of mouse 25-hydroxyvitamin D3 1α-hydroxylase (Cyp27b1; Accession Number AB006034) was prepared by RT-PCR of mouse kidney RNA, subcloned, and sequenced.
DNA probes for Cyp27b1 and Glyceraldehyde-3-phosphate dehydrogenase (Gapd) were prepared by Random Primed DNA La beling Kit (Roche) and [α-32P]dCTP (800 Ci/mmol; NEN) Total RNA was isolated from kidney with Tripure Isolation Reagent (Roche), and 20 µg aliquots were fractionated by electrophoresis on a 1% formaldehyde agarose gel, transferred to nitrocellulose membranes and hybridized to the radiolabeled cDNA fragments (48% formamide, 10% dextran sulfate, 5xSSC, 1xDenhardt’s and 100 µg/ml salmon sperm DNA) at 42°C overnight.
The membranes were washed and autoradiograms were prepared using Kodak BioMax film at –80°C with intensifying screens.
Quantification of signal intensity on autoradiograms was performed by Molecular Dynamics Personal Densitometer using ImageQuant software.
Statistical Analysis
Data from biochemical and image analyses are presented as means ± SEM. Statistical comparisons were made using Student’s t test, with P< 0.05 being considered significant.
