Vascular and bone biology have many features and regulatory mechanisms in common1, 2; both systems are subject to tissue remodeling in response to various physiologic and pathologic stimuli, and both are known to be exquisitely sensitive to alterations in the mechanical environment. As a result, osteoporosis and vascular disease often coincide in the elderly,3 with the most common associations being elevated blood pressure, calcification of the vascular wall, and reduced bone mineral density, although the paradox of arterial calcification in osteoporotic patients remains to be elucidated.4
The renin-angiotensin system (RAS) is a central regulator of blood pressure as well as fluid and electrolyte balance.5 In the classic endocrine cascade, renin produced by the juxtaglomerular apparatus of the kidney cleaves liver-produced angiotensinogen to angiotensin I, which is then cleaved by angiotensin-converting enzyme (ACE) to generate the biologically active angiotensin II. In addition, mounting evidence supports the existence of a local RAS loop, although the physiologic function of autocrine/paracrine RAS is not fully understood.6 Locally produced or circulating angiotensin II binds to two G protein-coupled receptors, AT1 and AT2, whereas the vasopressive and aldosterone-secreting actions of angiotensin II are mainly mediated through AT1.7, 8
We previously showed that bone cells express components of RAS, including ACE, AT1, and AT2.9 Furthermore, we9 and others10 have elucidated the link between RAS and bone metabolism by showing that transgenic activation of RAS in the mouse9 or chronic infusion of angiotensin II in the rat10 leads to an osteopenic phenotype, mainly as a result of excessive bone resorption. In both studies, angiotensin II has been shown to act on osteoblasts and stimulate the production of the osteoclastogenic cytokine receptor activator of NF-κB ligand (RANKL), thereby increasing the differentiation of osteoclasts indirectly.9, 10 Thus, these studies provide an account of the cellular and molecular mechanisms underlying the comorbidity of hypertension and osteoporosis with aging. However, whether RAS plays any physiologic role in bone metabolism has remained undetermined. The objective of the present study was to elucidate the physiologic function of RAS in the regulation of bone metabolism by using a mouse model lacking the major angiotensin II receptor, AT1a.
Materials and Methods
We purchased mouse RANKL and macrophage colony-stimulating factor (M-CSF) from R&D systems (Minneapolis, MN, USA), and 1α,25-Dihydroxyvitamin D3 [1α,25(OH)2D3] from Nacalai Tesque Inc. (Kyoto, Japan).
We intercrossed AT1a heterozygous knockout mice to produce homozygous knockout mice, as previously described.11 The mice were of C57BL/6 genetic background. We identified different genotypes by PCR analysis of genomic DNA extracted from the tail, using the following primers: 5′-GGAAACAGCTTGGTGGTG-3′ and 5′-CTGAATTTCATAAGCCTTCTT-3′ for AT1a; 5′-CAAGACCGACCTGTCCGGTG-3′ and 5′-CGACGAGATCCTCGCCGTCG-3′ for Neo, respectively.
Mice were raised under standard laboratory conditions of 24 ± 2°C and 50% to 60% humidity, and were allowed free access to tap water and commercial standard rodent chow (CE-2) containing 1.20% calcium, 1.08% phosphate, and 240 IU/100 g vitamin D3 (Clea Japan Inc., Tokyo, Japan). We measured blood pressure by a noninvasive tail-cuff method, as described previously,12 using a Model BP-98A (Softron, Tokyo, Japan). Urine was collected for 24 hours, and blood samples were centrifuged to obtain plasma and serum.
All experiments were performed in accordance with NCGG's ethical guidelines for animal care, and the experimental protocols were approved by the animal care committee of NCGG.
Blood and urine biochemistry
Serum osteocalcin, carboxy-terminal collagen crosslinks (CTX), calcium, arg8-vasopressin, and leptin were determined using a mouse osteocalcin enzyme immunoassay (EIA) kit (Biomedical Technologies Inc., Stoughton, MA, USA), a Rat Laps EIA kit (Immunodiagnostic Systems Inc., Fountain Hills, AZ, USA), a calcium E-HA test Wako kit (Wako Pure Chemical Industries Ltd., Osaka, Japan), an arg8-Vasopressin EIA kit (Enzo Life Sciences Inc., Plymouth Meeting, PA, USA), and a mouse leptin ELISA kit (Morinaga Institute of Biological Science Inc., Yokohama, Japan), respectively. Plasma intact parathyroid hormone (PTH) and serotonin (5-hydroxytryptamine [5-HT]) were determined by a Mouse Intact PTH ELISA kit (Immutopics Inc., San Clemente, CA, USA) and high-performance liquid chromatography (HPLC) (SRL Inc., Tokyo, Japan), respectively.
Bone and histological analyses
Micro-computed tomography (micro-CT) scanning was performed on the proximal tibia using a µCT-40 device (SCANCO Medical, Brüttisellen, Switzerland) at a resolution of 12 µm, and the microstructural parameters were calculated three-dimensionally,13 according to recently published guidelines.14 The scanning conditions include X-ray energy of 45 kVp, X-ray current of 177 µA, voxel size of 12.0 µm, and field of view (FOV) of 12.3 mm.
Tibiae were fixed in 70% ethanol, and bone histomorphometry was performed on undecalcified sections, with calcein and tetracycline double-labeling (administered s.c. at a 2-day interval), and histomorphometric parameters were measured at the Ito Bone Science Institute (Niigata, Japan).
Cell culture and ex vivo osteoclastogenesis assay
Bone marrow cells were isolated from the tibiae and femurs of 8-week-old wild-type (WT) and knockout (KO) mice, and cultured in α modified essential medium (α-MEM) supplemented with 10% fetal bovine serum (FBS), 100 µg/mL streptomycin, and 100 units/mL penicillin and M-CSF for 3 days, and adherent bone marrow macrophages (BMMs) were used as osteoclast precursors, as previously described.15 BMMs were treated with M-CSF (50 ng/mL) and various concentrations of RANKL for 3 days, fixed in 4% paraformaldehyde, and stained for tartrate-resistant acid phosphatase (TRAP). Multinucleate (≧3 nuclei), TRAP-positive cells were counted as osteoclasts.
Osteoblasts were isolated from newborn mouse calvaria as described previously,16 and co-cultures with bone marrow cells were performed in the presence of 10−8 M 1α,25(OH)2D3, according to an established technique.17 The number of TRAP-positive cells with more than three nuclei was counted under light microscopy.
RNA extraction and RT-PCR
Total RNA was isolated from primary osteoblasts and bone with TRIzol Reagent (Invitrogen, San Diego, CA, USA), and was reverse-transcribed using a high-capacity cDNA reverse transcription kit (Applied Biosystems). Gene expression was assessed by RT-PCR at 94°C for 30 seconds, 55°C for 1 minute, and 72°C for 1 minute with the following primer sets: 5′-GCATCATCTTTGTGGTGGG-3′ and 5′-ATCAGCACATCCAGGAATG-3′ for AT1a (32 cycles); 5′-GCATCATCTTTGTGGTGGG-3′ and 5′-ATGAGCACATCCAGAAAAC-3′ for AT1b (32 cycles); 5′- AGTGCAAACTGGCATGGG-3′ and 5′-AAAACGCCTGGAATCTGA-3′ for AT2 (32 cycles); and 5′-CATCGTGGGCCGCTCTAGGCACCA-3′ and 5′-CGGTTGGCCTTAGGGTTCAGGGGG-3′ for β-actin (26 cycles).
For quantitative RT-PCR, samples were analyzed using PowerSYBR Green PCR master mix and an ABI7300 real-time PCR system (Applied Biosystems). The sequences of the primers used for the real-time PCR were as follows: RANKL, 5′-TGAAGACACACTACCTGACTCCTG-3′ and 5′-CCCACAATGTGTTGCAGTTC-3′; osteoprotegerin (OPG), 5′-ATGAACAAGTGGCTGTGCTG-3′ and 5′-CAGTTTCTGGGTCATAATGCAA-3′; stromal cell-derived factor (SDF)-1α, 5′-TGGAAGGGAGGAGAGTGATG-3′ and 5′-GCCCAAGTGAGAGGAAAGC-3′; Runx2, 5′-TGCCCAGGCGTATTTCAG-3′ and 5′-TGCCTGGCTCTTCTTACTGAG-3′; SOST, 5′-CTTCAGGAATGATGCCACAGAGGT-3′ and 5′-ATCTTTGGCGTCATAGGGATGGTG-3′; MEPE, 5′-GTGCTGCCCTCCTCAGAAATAT-3′ and 5′- GTTCGGCCCCAGTCACTAGA-3′; DMP1, 5′-GGCTGTCCTGTGCTCTCCCAG-3′ and 5′-GGTCACTATTTGCCTGTCCCTC-3′; tryptophan hydroxylase (Tph)1, 5′-TGAAGTCGGAGGACTCATAAAAG-3′ and 5′-CGGGACTCGATGTGTAACAG-3′; Tph2, 5′-GAGCTTGATGCCGACCAT-3′ and 5′-TGGCCACATCCACAAAAT-3′; and β-actin, 5′-AAGGCCAACCGTGAAAAGAT-3′ and 5′-GTGGTACGACCAGAGGCATAC-3′. The target mRNA amount was corrected by that of β-actin mRNA.
Data are expressed as the mean ± SD. Statistical analysis was performed using unpaired Student's t test or ANOVA followed by Student's t test. Values were considered statistically significant at p < 0.05.
High bone mass in AT1a knockout mice
As a first step to explore the physiologic function of AT1a in bone metabolism, the tibial metaphysis of AT1a-deficient male mice was scanned by micro CT, and compared with that of the wild-type and heterozygous littermates. Representative micro CT images of three genotypes are shown in Figure 1A. Detailed three-dimensional (3D) morphometric analysis revealed that AT1a homozygous knockout mice had a significantly higher trabecular bone volume fraction (BV/TV) than age- and sex-matched wild-type mice (Fig. 1B). Microstructure analysis revealed that the trabecular morphology in the AT1a knockout mice was characterized by an increased number and connectivity and decreased separation (Fig. 1B). The heterozygous mice exhibited an intermediate phenotype in terms of both bone mass and trabecular microstructure. The high bone mass phenotype of the AT1a-deficient mice was also found in the female mice (Fig. 1C, D). There was no significant difference in the cortical bone area or volume between AT1a knockout and wild-type mice, as determined by micro CT scanning of tibial diaphysis (data not shown).
We examined the effects of aging and estrogen deficiency on the high bone mass phenotype of AT1a-deficient mice. As shown in Figure 2A, 25-month-old male AT1a knockout mice maintained higher bone mass than age- and sex-matched wild-type mice. Also, as shown in Figure 2B, AT1a-deficient female mice were protected against ovariectomy-induced bone loss.
Elevated bone turnover in AT1a knockout mice
Biochemical analysis of blood samples indicated that the serum concentrations of osteocalcin, a product of osteoblasts, and CTX, a degradation product of type I collagen, were both significantly higher in AT1a-deficient mice than in age- and sex-matched wild-type mice (Fig. 3), which indicates a state of high bone turnover. Serum and urinary calcium and plasma PTH levels did not differ among the three genotypes (Fig. 3), which makes it unlikely that the high bone mass and high bone turnover found in AT1a knockout mice are secondary to alterations in systemic calcium metabolism.
The mechanism underlying the high bone mass of the AT1a knockout mice was further explored at the tissue and cell levels by histomorphometric analysis of the proximal tibia, which revealed that the number of osteoclast (N.Oc/BS), the bone surface area covered by osteoclasts (Oc.S/BS), and the eroded surface (ES/BS) were all significantly elevated in the homozygous knockout mice (Fig. 4), indicating that the osteoclast number and bone resorption activity are increased in the absence of AT1a.
The osteoid surface (OS/BS), osteoblast surface (Ob.S/BS), and bone formation rate (BFR/BS) were significantly increased in the AT1a-deficient mice (Fig. 4), indicating that bone formation was also stimulated along with accelerated bone resorption. Taken together, the histomorphometric analysis results are consistent with the blood biochemical analysis, and suggest that AT1a knockout mice exhibit a high bone mass phenotype coupled with heightened bone turnover.
Ex vivo analysis of AT1a-deficient osteoblasts and osteoclasts
In view of the findings that the trabecular bone of AT1a knockout mice contained increased numbers of osteoclasts and osteoblasts, it was implied that osteoclastogenesis and osteoblastogenesis are both accelerated in the absence of AT1a. AT1a is expressed in bone in vivo and in primary osteoblasts (Fig. 5A), and also in osteoclast precursors, although weakly.9 In addition, elimination of AT1a did not affect the expression of AT1b or AT2 (Fig. 5A). In order to test whether the increased bone cell activities are a result of cell-intrinsic alterations in differentiation potential, AT1a-deficient precursor cells were isolated from the knockout mice, and their potential to differentiate into osteoclasts or osteoblasts was assessed in ex vivo assays.
First, BMMs were isolated as osteoclast precursor cells from both AT1a homozygous knockout and wild-type mice, and their differentiation potential in response to RANKL and M-CSF was assessed in ex vivo cultures. As shown in Figure 5B, the BMMs derived from the AT1a knockout mice tended to generate more TRAP-positive osteoclasts at a higher concentration of RANKL, compared with the BMMs from the wild-type mice, but without a statistically significant difference. Thus, it was concluded that there was no major difference in the potential of hematopoietic precursor cells to differentiate into osteoclasts.
In order to examine the effect of AT1a deficiency on osteoblast maturation, primary osteoblasts were isolated from the calvaria of newly born wild-type and AT1a knockout mice, and their differentiation was assessed by alkaline phosphatase and alizarine red staining. No apparent difference was observed (Fig. 5C), implying that the increase in osteoblasts and bone-forming function observed in the homozygous knockout mice was not the result of cell-autonomous alterations in AT1a-deficient osteoblasts. Osteoblastogenesis assays using bone marrow adherent cells from AT1a knockout and wild-type mice did not reveal any difference in the number of fibroblastic colony-forming units (CFU-F) or osteoblastic colony-forming units (CFU-OB) (data not shown).
We also examined the formation of osteoclasts in a co-culture of calvaria-derived primary osteoblasts and BMMs in the presence of 1α,25(OH)2D3. Again, there was no difference in the number of TRAP-positive osteoclasts formed, whether either or both of the AT1a-deficient BMMs and osteoblasts were used (Fig. 5D). Thus, ex vivo assays did not disclose any functional difference in AT1a-deficient hematopoietic “seed cells” or in osteoblastic “soil cells” that constitute the microenvironment which supports osteoclastogenesis.
Gene expression in AT1a-deficient bone
In order to further explore the mechanism by which bone formation as well as resorption is accelerated in AT1a-deficient mice in vivo, the expression of molecular markers of osteoclasts, osteoblasts, and osteocytes was assessed by quantitative RT-PCR using RNAs extracted from the femur and tibia of AT1a-deficient mice and wild-type mice. As shown in Figure 6, we found a significant elevation of the RANKL/OPG ratio and increased expression of SDF1α in the AT1a knockout bone compared with wild-type bone (Fig. 6), which may account for the increased number and activity of osteoclasts in the AT1a knockout mice.
We also noted a significant reduction in the expression of the SOST gene that codes for an osteocyte-specific secretory protein, sclerostin (Fig. 6). Because sclerostin is known to counter the bone-anabolic action of Wnt, its decreased expression may be one of the reasons for the stimulated bone formation in AT1a-deficient mice.
Metabolic phenotype of the AT1a-deficient mice
As expected, both male and female AT1a-deficient mice exhibited hypotension relative to control mice (Fig. 7A and data not shown). In addition, AT1a-deficient mice exhibited polyuria and polydipsia with elevated serum arginine vasopressin concentrations under the current breeding conditions (Fig. 7B). However, judging from the blood hematocrit levels, it is unlikely that they suffered from overt dehydration (data not shown).
AT1a-deficient mice exhibited a modestly lean phenotype with substantially decreased adipose tissue, especially with aging (Fig. 8A). In fact, the serum leptin concentrations were reduced in AT1a knockout mice (Fig. 8B). There was no difference in the plasma serotonin concentrations or the expression of tryptophan hydroxylase (Tph)1 (in duodenum) and Tph2 (in brainstem) between the AT1a knockout and wild-type mice (Fig. 8B, C).
Excessive RAS activation is implicated in the age-related development of cardiovascular, metabolic, and kidney disease, and the blockade of RAS with ACE inhibitors or angiotensin receptor blockers (ARBs) has been widely used not only in the management of hypertension, but also for the improvement of cardiovascular and renal outcomes.18 Recently, we9 and others10 have shown that excessive activation of RAS causes osteoporosis, mainly through an elevation of osteoclastic bone resorption, by using a transgenic mouse model overproducing human renin and angiotensinogen9 or an infusion of angiotensin II in ovariectomized rats,10 respectively. The present study further supports RAS as an important physiologic regulator of bone remodeling by showing that AT1a-deficient mice exhibit a high bone mass phenotype with elevated bone turnover.
In the absence of AT1a, both bone resorption and formation were elevated, resulting in a net increase in bone mass (Fig. 9). According to our present as well as previous analysis,9 primary osteoblasts express both AT1 and AT2, whereas only weak expression of AT1 is present in the osteoclast lineage. AT1a-deficient osteoblasts, however, exhibited no difference from wild-type cells in the differentiation potential in ex vivo cultures, which makes it unlikely that AT1a functions in osteoblasts in a cell-autonomous manner. Increased bone formation in AT1a knockout mice was associated with a decreased expression of SOST in bone. SOST encodes sclerostin, a secretory product of osteocytes that counters Wnt signaling, thereby negatively regulating bone formation.19 Taken together with the findings that no evident alterations in osteocyte morphology or number were observed in AT1a-deficient bone (data not shown), it is suggested that SOST expression in osteocytes is functionally reduced in AT1a deficiency, which is responsible at least in part for the stimulation of bone formation (Fig. 9). Thus, it is surmised that in addition to PTH receptor signaling20 and mechanical stimulation,21 angiotensin II signaling through AT1a is an important regulator of SOST expression in osteocytes.
As powerful regulatory circuits of bone formation, a central mechanism which involves leptin, brainstem serotonin synthesis, and the sympathetic nervous system,22 as well as Wnt signaling,23, 24 has attracted considerable attention in recent years. It is conceivable that the decreased SOST expression in osteocytes stimulates bone formation in the AT1a-deficient mice through a relative increase in local Wnt signaling in osteoblasts. It has also been shown that low-density lipoprotein receptor-related protein (LRP) signaling in enterochromaffin cells of the duodenum is linked to bone formation through circulating serotonin25; in AT1a knockout mice, however, we did not find any alterations in the plasma serotonin concentrations or the expression of Tph1 in the duodenum.
RAS is an important regulator of the release of leptin from adipocytes. According to a study on the local versus systemic effects of RAS activation, angiotensin II is suggested to act directly on adipocytes to stimulate leptin production and release, whereas a systemic infusion of angiotensin II resulted in a progressive decline in the circulating leptin concentrations through sympathetic activation.26 Intracerebroventricular as well as systemic administration of angiotensin II has also been shown to reduce body weight, along with increased energy expenditure and water drinking in rats.27, 28 In the present study a lean phenotype was observed in AT1a knockout mice, with decreased adipose tissue and leptin levels as well as polyuria and polydipsia (Fig. 9). Increased energy expenditure with elevated urinary excretion of catecholamines has been documented in AT1a knockout mice.29 Intriguingly, a recent report points to the brain RAS in energy and fluid balance; transgenic mice with RAS activation exclusively in the brain exhibit an altered metabolic phenotype, including polyuria, polydipsia, increased energy expenditure and thermogenesis, elevated sympathetic nervous activity and metabolic rate, and decreased circulating angiotensin II levels.30 Because systemic AT1a deficiency seems to mirror brain RAS hyperactivity in terms of the metabolic phenotype, and because circulating angiotensin II concentrations are elevated in AT1a knockout mice,31 there exists the possibility that activation of AT1b or AT2 is responsible for these metabolic disorders. It would follow, then, that AT1a knockout mice exhibit increased bone formation despite increased sympathetic tone, which would otherwise be expected to suppress bone formation according to the current widely accepted model.22 In contrast, if the reduced leptin levels in AT1a knockout mice are simply interpreted as decreased leptin signaling in the brainstem and therefore lowered sympathetic tone via increased serotonin in the hypothalamus, it is likely to contribute to the stimulation of bone formation. However, this possibility is not supported by the result that there was no significant difference in the expression of Tph2, an enzyme for serotonin synthesis in the brain, between AT1a-null and wild-type mice.
Increased osteoclastic bone resorption is another feature of AT1a-deficient mice disclosed in the present study. Because osteoclastogenesis from bone marrow macrophages under the stimulation of RANKL was not affected by the absence or presence of AT1a, a cell autonomous alteration within the hematopoietic lineage is not likely to account for the increased osteoclasts in vivo. Instead, an elevated RANKL/OPG ratio, mainly as a result of decreased OPG expression, was observed along with increased expression of SDF1α, a chemokine known to regulate chemotaxis of osteoclast precursors.32 Thus, it is suggested that these alterations in the microenvironment contributed to the increased differentiation and recruitment of osteoclasts (Fig. 9).
In conclusion, the elevated bone turnover and bone mass found in AT1a-deficient mice establishes AT1a as a negative regulator of bone remodeling; signaling through AT1a decreases osteoblast as well as osteoclast activities through non-cell autonomous mechanisms. During the course of conducting this study, it was reported that pharmacological blockade of the AT2 receptor, as well as genetic inactivation of the AT2 gene, results in higher bone mass with increased bone formation.33 Intriguingly, AT1a-deficient mice have been reported to live longer than control mice through an attenuation of oxidative damage and upregulation of prosurvival genes such as Nampt and Sirt3.34 In view of the present results that AT1a-deficient mice maintain higher bone mass with aging and are also protected against bone loss caused by estrogen deficiency, it is conceivable that prevention of RAS activation or blockade of AT1a signaling may be beneficial for multiple age-related disorders including osteoporosis.
All authors state that they have no conflicts of interest.
This study was supported in part by a grant-in-aid for Longevity Science from the Ministry of Health, Labor and Welfare of Japan (to KI), by a grant for the Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation (NIBIO) of Japan (06-31 to KI), and by a grant from the Naito Foundation (to KI). Pacific Edit reviewed the manuscript before submission. We thank Drs. Sunao Takeshita and Ken Watanabe of our Department (NCGG) for valuable suggestions and reagents.
Authors' roles: Study design: KK and KI. Study conduct: KK, MI, TF, and RF. Data collection: KK and MI. Data analysis: KK and MI. Data interpretation: KK, MI, JI, AF, and KI. Drafting manuscript: KK and KI. Revising manuscript content: MI, JI, and AF. Approving final version of manuscript: KK, MI, TF, RF, JI, AF, and KI. KI takes responsibility for the integrity of the data analysis.