Y F Ma, M Stimpel 1, H Liang, S Pun and WSSJee

Transcription

1 467 Impact of antihypertensive therapy on the skeleton: effects of moexipril and hydrochlorothiazide on osteopenia in spontaneously hypertensive ovariectomized rats Y F Ma, M Stimpel 1, H Liang, S Pun and WSSJee Radiobiology Division, University of Utah School of Medicine, Salt Lake City, Utah 84112, USA and 1 Department of Cardiovascular Clinical Research, Schwarz Pharma AG, Monheim, Germany (Requests for offprints should be addressed to WSSJee, Radiobiology Division, University of Utah, Building 586, Salt Lake City, Utah 84112, USA) Abstract Skeletal effects of moexipril, an angiotensin-converting enzyme (ACE) inhibitor, and hydrochlorothiazide (HCTZ), a thiazide diuretic, were studied in ovariectomized (OVX) spontaneously hypertensive rats (SHR). Moexipril (10 mg/kg per day), HCTZ (10 mg/kg per day), alone or in combination, as well as 17α-estradiol (30 µg/kg per day) were given to OVX SHR immediately after surgery and studied for short- and long-term effects (14 and 56 days respectively). All drugs were given orally. Histomorphometric data on the secondary spongiosa of proximal tibial metaphyses (cancellous bone) and tibiofibular junctions of tibial shafts (cortical bone) were analyzed. Ovariectomy induced cancellous bone loss in SHR by inducing negative bone balance. Estrogen prevented ovariectomy-induced cancellous bone loss in the SHR by reducing bone turnover and partially suppressing the coupling of bone formation to resorption on the endocortical surface. HCTZ reduced blood pressure after 1 week of treatment, yet this effect was no lower than that seen in controls after 3 weeks of treatment. Two weeks of HCTZ transiently prevented ovariectomy-induced increases in bone turnover rate and eroded surface. This delayed ovariectomy induced trabecular bone loss in the proximal tibial metaphysis, but had no effect on the tibial shaft. Like HCTZ, moexipril also reduced blood pressure after the first week of treatment but it had no apparent effect on either the proximal tibial metaphysis or the tibial shaft. A combination of moexipril and HCTZ exhibited a much more potent hypotensive effect and had the same effects on bone mass and dynamic end-points as HCTZ alone. Our data indicate that (1) HCTZ treatment has some transient beneficial effects on both antihypertension and osteoprotection in hypertensive osteopenic rats, (2) the combination of moexipril with HCTZ improved the antihypertensive effect but did not potentiate or hamper the osteoprotective effect of HCTZ, and (3) the skeletal effect of estrogen is not impacted by the hypertensive state. These findings are relevant for the use of ACE inhibitor and thiazide diuretics, alone or in combination, in antihypertensive therapy in postmenopausal women. Introduction Both osteoporosis and hypertension are major causes of morbidity and mortality in elderly people. It is generally accepted that angiotensin-converting enzyme (ACE) inhibitors and diuretics have a synergistic antihypertensive action (Ambrosioni et al. 1987, Dickstein et al. 1994). Thus, co-treatment with these two classes of drugs is a common strategy for the management of hypertension (Scalbert et al. 1992). Because approximately 50% of the human hypertensive population are postmenopausal females (Safar et al. 1994), any concomitant drug therapy in postmenopausal women must not adversely affect the osteoprotection of hormone replacement therapy, nor should it increase bone loss when given alone. We have reported previously that the ACE inhibitor moexipril has no effect on the proximal tibial metaphysis (PTM) of either sham- or ovariectomized (OVX) rats and that it does not hamper the osteoprotective effects of 17β-estradiol in intact rats (Stimpel et al. 1995). The study was, however, conducted in normotensive rats. It is well known that, like hypertensive humans, spontaneously hypertensive rats (SHR) have abnormal calcium metabolism, such as low serum calcium, vitamin D 3, magnesium and high serum parathyroid hormone (PTH) levels (Lucas et al. 1986, Merke et al. 1989, Wise et al. 1996). In addition, a relatively low cortical and cancellous bone along with less osteoblast and more osteoclast in SHR, when compared with their normotensive genetic control Wistar Kyoto rats, have also been reported (Wang et al Journal of Endocrinology Ltd Printed in Great Britain /97/ $08.00/0

2 468 YFMAand others Moexipril and HCTZ in hypertensive OVX rats 1993). Furthermore, most patients treated with an ACE inhibitor are hypertensive and thiazide diuretics are the only antihypertensive drugs which have been shown to increase bone mineral density clinically. Thus, it is necessary to test the skeletal effects of these regimens on a hypertensive animal model. The current study was designed to evaluate the skeletal effects of moexipril and hydrochlorothiazide (HCTZ), alone or in combination, and of 17α-estradiol in OVX SHR. The study examined the histomorphometry of OVX SHR after the operation, at 14 days (short timeeffect) and at 56 days (long time-effect). The findings from this study generated new information on (1) the cortical and cancellous bone tissue and cellular profiles of 23- to 31-week-old female SHR, (2) the skeletal response to ovariectomy in SHR, (3) the skeletal responses of OVX SHR to 17α-estradiol, a mimic of hormone replacement therapy in the hypertensive state, and (4) the skeletal responses of OVX SHR to moexipril and HCTZ, alone and in combination; a mimic of antihypertensive therapy in the estrogen-deficient state. The study might be of clinical relevance when antihypertensive therapy is initiated, particularly in hypertensive women suffering from postmenopausal osteoporosis. Materials and Methods Ninety-six 12-week-old virgin SHR weighing about 220 g were obtained from Taconic Farms (Germantown, NY, USA). They were acclimated to local vivarium conditions and allowed free access to water and diet (Teklad Rodent Chow No. 8604; Harlan Taklad, Madison, WI, USA). At 23 weeks of age, six rats were killed as basal controls. The remaining rats were divided into sham-ovx or OVX groups. Bilateral ovariectomy was performed by the dorsal approach ( Jee et al. 1990). Moexipril (10 mg/kg per day) and HCTZ (10 mg/kg per day), alone or in combination, or 17α-estradiol (30 µg/kg per day) alone were given to OVX rats immediately after operation for 14 or 56 days. All drugs were given orally. Powdered moexipril hydrochloride (Schwarz Pharma, Monheim, Germany) and HCTZ (Schwarz Pharma) were dissolved in deionized water to a concentration of 10 mg/ml. 17α-Estradiol (Sigma Chemical Co., St Louis, MO, USA) was dissolved in a mixture of 95% corn oil and 5% benzyl alcohol to a concentration of 30 µg/ml. Because our previous study (Stimpel et al. 1995) showed that 56 days of deionized water administration has no skeletal effect, and many studies have also shown that a mixture of 95% corn oil and 5% benzyl alcohol has no skeletal effect (Wronski et al. 1988), we used a common vehicle for all control groups, a mixture of 95% corn oil and 5% benzyl alcohol. Rats were weighed every 4 days, and the volume of solution was adjusted accordingly. All rats were doublefluorescent labeled with demeclocycline (15 mg/kg; Sigma) on days 14 and 13 and with calcein (5 mg/kg; Sigma) on days 4 and 3 before they were killed. Heart rate and blood pressures (systolic, diastolic and mean) were measured once a week on the tail of unanesthetized rats using an indirect blood pressure measuring system (Stoelting, Wood Dale, IL, USA). To reduce variation, each measurement was repeated three times and the average number was used for statistical analyses. When the animals were killed the tibiae were excised, freed from soft tissues and cut into three parts. All specimens were stained with Villanueva bone stain for 4 days, dehydrated in graded alcohol and acetone, and embedded in methyl methacrylate. Using a low-speed saw, undecalcified sections were first cut to 230 µm, then ground to 20 µm (PTM) or 30 µm (tibial shaft) and coverslipped for histomorphometric measurements as previously described ( Jee et al. 1990). All the measurements were carried out blind. A digitizing image analysis system was used for taking the measurements. The secondary spongiosa of the proximal tibia between 1 and 4 mm proximal to the growth plate metaphyseal junction was measured. Measured parameters included total tissue area, trabecular area and perimeters, trabecular eroded, osteoid, single-labeled and double-labeled perimeters, and interlabeling width. These measurements were used to calculate the percent trabecular area, number and thickness, percent eroded, osteoid and labeled perimeters, mineral apposition rate, and bone formation rates (BFR) (tissue volume (TV), BFR/TV; bone volume (BV), BFR/BV; and bone surface referent (BS), BFR/BS) as described previously ( Jee et al. 1983, Parfitt et al. 1987). In the tibial shaft, cross-sections at the tibio-fibular junction were used for histomorphometry. Total tissue area, marrow area, marrow trabecular area, minimal cortical width, eroded, osteoid, single- and double-labeled perimeters, and interlabeling width were measured separately. These parameters were used to calculate the total bone area, marrow space area, percent cortical bone area, percent marrow area, percent marrow space area, mineral apposition rate, surface referent bone formation rate at both periosteal and endocortical surfaces, and percent eroded and osteoid perimeters at the endocortical surface ( Jee et al. 1983, Parfitt et al. 1987). In addition, the eroded surface was classified into four different categories: (1) bare eroded surface, (2) osteoid covered eroded surface, (3) label covered eroded surface, and (4) total eroded surface (the sum of the above). These eroded categories allowed us to evaluate the spatial changes of coupling bone formation to resorption during the specific period (Eriksen et al. 1990). Differences among group means were evaluated using an ANOVA and the Fisher PLSD test. Differences between the results of basal control and other treatment groups were compared using the two-tailed Student s

3 Moexipril and HCTZ in hypertensive OVX rats YFMAand others 469 t-test. P<0 05 was considered significant. Data are presented as mean S.E.M. Results Body weight changes Ovariectomy increased body weight by 5 22% which is similar to the results reported by others (Wronski et al. 1987). The 17α-estrogen treatment delayed this body weight increase in OVX SHR. However, there was no difference in body weight between the estrogen- and vehicle-treated rats after 3 weeks of treatment. Moexipril, HCTZ and their combination did not alter the body weight of the OVX SHR (data not show). Heart rate and blood pressure changes Neither ovariectomy nor estrogen treatment induced any significant change in heart rate or blood pressure. Moexipril lowered the blood pressure by 12% during the first week, but did not alter the heart rate of OVX SHR. These results were maintained throughout the study. HCTZ lowered blood pressure by 10% during the first week and heart rate by 15% during the second week. However, both heart rate and blood pressure in HCTZ-treated rats were no different from those in controls by the third week. The changes in heart rate in rats treated with combined moexipril and HCTZ were the same as those in HCTZ-treated SHR, while the changes in blood pressure were greater than those of moexipril-treated SHR (data not show). Cancellous bone changes in the PTM secondary spongiosa Effects of aging Figures 1a and e and 2g and h depict the age-related decreases in longitudinal growth rate in PTM and the ratio of labeled to eroded perimeter in secondary spongiosa. An increase in activation frequency which led to a loss of 18% trabecular bone in secondary spongiosa was also found in SHR between 5 5 and 7 5 months of age. Effects of ovariectomy Figure 1a and c shows a 27% trabecular bone loss accompanied by a reduction in trabecular number and thickness at 2 weeks after ovariectomy. At 8 weeks after ovariectomy, further trabecular bone loss ( 51% of that of aging controls) and further reduction in trabecular number occurred. Figures 1e,2d h and 3 reveal that ovariectomy stimulated the longitudinal growth rate, activation frequency, bone formation rate and eroded surface, with bone resorption exceeding bone formation in OVX SHR which resulted in a significant decrease in the ratio of labeled to eroded perimeter when compared with the sham-operated rats. Effects of estrogen Figure 1a and c show that the 17α-estradiol treatment prevented ovariectomy-induced trabecular bone loss and maintained the trabecular number in OVX SHR. Dynamically, 2 weeks of 17α-estradiol treatment prevented ovariectomy-induced increases in longitudinal growth rate, percent osteoid perimeter, activation frequency and osteoid-covered eroded surface. Rats treated with 17α-estradiol for 8 weeks had lower bone resorption, formation indices, bone turnover (BFR/BV) and activation frequency than those of vehicle-treated OVX rats (Figs 1e, 2 and 3). Effects of moexipril As shown in Figs 1 and 3, moexipril treatment for 8 weeks prevented an ovariectomyinduced increase in the longitudinal growth rate and had no effect on bone mass, architecture or dynamic profiles in OVX SHR. Effects of HCTZ Figures 1a, 2c,dand h and 3c show that 2 weeks of HCTZ treatment prevented the trabecular bone loss in the PTM which had been associated with decreases in bone turnover (BFR/BV), mineral apposition rate, osteoid covered eroded surface, and activation frequency. However, 8 weeks after the treatment, the trabecular bone mass and dynamic end-points in the HCTZ-treated SHR were no different from the vehicletreated OVX SHR. Effects of combination of moexipril and HCTZ Figures 1 3 summarize the changes in bone mass and dynamic parameters after the combination treatment in OVX SHR. None of the perimeters were different from those of HCTZ-treated animals. Cortical bone changes in the tibial shaft Effects of aging There was an 8% reduction in cortical area ( vs mm 2 ) in the tibial shafts of 7 5-month-old SHR when compared with those of 5 5- month-old rats. This was mainly due to the decrease in periosteal bone formation rate (undetectable in 7 5- month-old vs µm/day in 5 5-month-old rats). Effects of ovariectomy At 8 weeks after ovariectomy, there was no alteration in bone mass, but there was stimulation of periosteal ( vs %) and endocortical ( vs %) labeled surfaces as compared with the sham-ovx controls. Effects of estrogen The treatment prevented ovariectomy-induced stimulation of the endocortical eroded surface ( vs %). Effects of moexipril There were no significant differences in the selected static and dynamic indices between the moexipril- and vehicle-treated OVX SHR. Effects of HCTZ No significant differences in the selected static and dynamic indices were observed between the HCTZ- and the vehicle-treated OVX SHR.

4 470 YFMAand others Moexipril and HCTZ in hypertensive OVX rats Figure 1 Static changes in PTM in sham-ovx (aging), OVX SHR, and OVX SHR treated with moexipril (M), HCTZ, alone or in combination, and 17α-estradiol (E2). Note that ovariectomy resulted in obvious bone loss in the PTM by lost trabecular number and thickness (b and c); 17α-estradiol prevented ovariectomy-induced cancellous bone loss (a); HCTZ also prevented ovariectomy-induced cancellous bone loss at 2 weeks (wks) but not at 8 weeks (a); moexipril alone had no effect on PTM; the combination of moexipril and HCTZ produced the same skeletal effects as HCTZ alone (a d). a P<0 05, aging and OVX controls vs basal controls; b P<0 05, OVX controls vs aging controls; c P<0 05, treatment groups vs OVX age-matched controls; d P<0 05, 8-week groups vs the same treatment 2-week groups. Effects of combination of moexipril and HCTZ There were also no significant differences in the selected static and dynamic indices measured in the rats given the combination treatment and the vehicle-treated OVX SHR. Discussion In our current study of SHR, we found that: (1) there was an age-related bone loss in both cancellous (18%) and cortical (8%) bone sites of the tibia between 23 and 31 weeks of age; (2) ovariectomy induced cancellous bone loss of PTM but not cortical bone of the tibial shaft; (3) 17α-estradiol prevented the ovariectomy-induced cancellous bone loss; (4) HCTZ lowered the blood pressure after 1 week of treatment, but blood pressure stabilized to control levels by week 3; (5) HCTZ delayed cancellous bone loss in OVX SHR by transiently inhibiting bone turnover, but it had no effect on cortical bone; (6) moexipril also lowered blood pressure after 1 week of treatment and throughout the duration of the study; (7) moexipril alone had no skeletal effect on the PTM or tibial

5 Moexipril and HCTZ in hypertensive OVX rats YFMAand others 471 Figure 2 Dynamic changes in PTM in sham-ovx (aging), OVX SHR, and OVX SHR treated with moexipril (M), HCTZ, alone or in combination, and 17α-estradiol (E2). Note that ovariectomy increased bone formation and activation frequency, but decreased labeled to eroded ratio; 17α-estradiol corrected ovariectomy changes and HCTZ had almost the same effect as 17α-estradiol at 2 weeks (wks) and was no different from OVX controls at 8 weeks; moexipril alone showed no difference from OVX controls at both 2 and 8 weeks; the combination of moexipril and HCTZ produced almost the same effect as HCTZ alone. a P<0 05, aging and OVX controls vs basal controls; b P<0 05, OVX controls vs aging controls; c P<0 05, treatment groups vs OVX age-matched controls; d P<0 05, 8-week groups vs the same treatment 2-week groups.

6 472 YFMAand others Moexipril and HCTZ in hypertensive OVX rats Figure 3 Eroded surface changes in PTM in sham-ovx (aging), OVX SHR, and OVX SHR treated with moexipril (M), HCTZ, alone or in combination, and 17α-estradiol (E2). Note that ovariectomy stimulated all kinds of eroded surfaces (a d); 17α-estradiol prevented ovariectomy-induced resorption changes (c and d) and HCTZ had the same effect as estradiol at 2 weeks (wks) but lost their effect at 8 weeks (c and d); moexipril had no effect on eroded surfaces in OVX rats; the combination of moexipril and HCTZ demonstrated the same changes as HCTZ alone (a d). a P<0 05, aging and OVX controls vs basal controls; b P<0 05, OVX controls vs aging controls; c P<0 05, treatment groups vs OVX age-matched controls; d P<0 05, 8-week groups vs the same treatment 2-week groups. shaft of OVX SHR; and (8) the combination of moexipril and HCTZ decreased blood pressure more than either agent alone, but produced the same skeletal effect as HCTZ alone in SHR. The age-related bone loss in the SHR was associated with a significant decrease in longitudinal bone growth, and an increase in age-related remodeling-dependent bone loss which was evidenced by a non-significant decrease in bone balance (labeled to eroded surface ratio), increases in the turnover (activation frequency) and the resorption index (eroded surface) in cancellous bone. There was also a decrease in periosteal bone formation rate in cortical bone. Hypertensive impacts on circulation, metabolism, and kidneys may contribute to age-related bone loss over the shorter life-span of SHR. The average life-span of SHR is 18 months, while that of the parent Wistar Kyoto rat is at least 24 months (Yamori et al. 1974). Our study shows that 17α-estradiol had the same osteoprotective effect in OVX SHR as in normotensive OVX Sprague Dawley rats by inhibiting bone turnover (Wronski et al. 1988). Furthermore, estrogen produces uncoupling of bone formation from resorption, as shown by the early appearance of a decrease in eroded surfaces covered by osteoid. However, the skeletal effect of estrogen was not diminished by the hypertensive state. Numerous studies have shown that thiazide increases bone mineral content and bone mineral density, and decreases the incidence of fractures in patients (Adland- Davenport et al. 1985, Cauley et al. 1993). Long-term thiazide therapy has resulted in higher bone mineral density and lower hip fracture rates which are attributed to the increase in serum calcium levels and a decrease in parathyroid activity which, in turn, leads to a decrease in bone resorption (Hall & Schaueblin 1994). Steiniche et al. (1989) further reported that 6 months of HCTZ treatment of patients resulted in a reduction in the extent of eroded surface and bone formation rate as well as a decrease in the osteoid thickness. However, the treatment had no effect on the trabecular bone volume. Only one animal study exists. Jayo et al. (1994) report that estrogen, alone and in combination with thiazide, is effective in preventing the bone loss induced by ovariectomy in the monkey. Both treatments were equally effective. The effect of thiazide alone was not reported in the study. Thus the effects of thiazide on the rat skeleton have not yet been studied. In the current study, HCTZ (10 mg/kg per day) treatment delayed cancellous bone loss in SHR at 2 weeks after ovariectomy. The osteoprotective effect of HCTZ was due mainly to the decrease in bone turnover (BFR/BV), activation frequency and eroded surface. Such

7 Moexipril and HCTZ in hypertensive OVX rats YFMAand others 473 decreases in the bone resorption index and related biomarkers have also been reported after thiazide treatment (Steiniche et al. 1989, Peh et al. 1993, Hall & Schaueblin 1994). However, there are conflicting reports concerning the effects of thiazide on bone formation. Both decreased (Steiniche et al. 1989) and increased (Song & Wergedal 1993) bone formation have been reported in a few limited in vitro (Song & Wergedal 1993) and human iliac crest biopsy (Steiniche et al. 1989) studies. Our in vivo histomorphometric data do not support the in vitro findings that HCTZ stimulates osteoblast formation. There are some factors that might contribute to the depression of bone resorption by HCTZ. These appear to be the increase in urinary sodium and the decrease in urinary calcium excretions and lower serum PTH levels (Peh et al. 1993). As we know, calcium plays an important role in antihypertension by both regulating the smooth muscle relaxation and contraction of the blood vessels, and in suppressing bone resorption by decreasing PTH secretion. Conversely, there seems to be no ready explanation as to why the osteoprotective effect disappeared after 8 weeks of treatment. The antihypertensive effect also vanished at the same time. We do not yet know whether or not the skeletal effects of HCTZ share the same pharmacodynamics as antihypertensive treatments. Serum and urine electrolyte analyses may supply some answers to the loss of blood pressure and skeletal effects noted in our study. Moreover, since only one dose level was employed in the current study, a dose response study in rats is also needed. The rationale for investigating the effect of ACE inhibitor on bone turnover in hypertensive state is based on the following. (1) Capillaries are observed immediately adjacent to osteoclasts at resorptive sites on bone surface and intense vascularity is a common feature in states of excessive bone resorption. (2) Osteoclasts are target cells for substances released by the endothelium of the bone microvasculature, such as endothelin (Alam et al. 1992), nitric oxide (Maclntyre et al. 1991) and prostaglandins ( Jee et al. 1990), suggesting that osteoclasts might also be locally regulated by endothelial products. Since angiotensin II is generated by endothelial cells as well, it suggests that the renin-angiotensin system may be involved in the regulation of osteoclast bone resorption. The speculation is supported by an in vitro experiment showing that the ACE inhibitor moexipril blocked the bone resorption effect of angiotensin II (Hatton et al. 1997). In the current study, moexipril had no effect on the cancellous or cortical bone mass in hypertensive OVX SHR as was reported previously in intact and OVX Sprague Dawley rats (Stimpel et al. 1995). Neither did moexipril treatment hamper the osteoprotective effects of 17β-estradiol (Stimpel et al. 1995). Furthermore, we found that moexipril did not alter the skeletal effects of HCTZ on either the PTM or the tibial shaft in hypertensive, estrogen-deficient rats. The findings indicate that the renin-angiotensin system and its pharmacological inhibition may not be important in regulating bone turnover under in vivo conditions of hypertension and estrogen deficiency. The current study is the first report that this combination treatment of ACE inhibitors and thiazide diuretics produces both desirable antihypertensive and osteoprotective effects in experimental hypertensive animals. Our data indicate that the HCTZ treatment had some transient beneficial effects on both antihypertension and osteoprotection in the hypertensive osteopenic rat skeleton. The combination of moexipril and HCTZ improved the antihypertensive effect but did not potentiate or hamper the osteoprotective effects of HCTZ. The skeletal effect of estrogen was not impacted by the hypertensive state and these findings are clinically relevant for antihypertensive therapy in postmenopausal women suffering from osteoporosis. References Alam ASMT, Gallagher A, Shankar V, Ghatei MA, Datta HK & Huang CLH 1992 Endothelin inhibits osteoclastic bone resorption by a direct effect on cell motility: implications for the vascular control of bone resorption. Endocrinology Adland-Davenport P, McKenzie MW, Notelovitz M, McKenzie LC & Pendergast JF 1985 Thiazide diuretics and bone mineral content in postmenopausal women. American Journal of Obstetrics and Gynecology Ambrosioni E, Borghi C& CostaFV1987 Captopril and hydrochlorothiazide: rationale for their combination. British Journal of Clinical Pharmacology 23 43S 50S. Cauley JA, Cummings SR, Seeley DG, Black D, Browner W, Kuller LH & Nevitt MC 1993 Effects of thiazide diuretic therapy on bone mass, fractures, and falls. The Study of Osteoporotic Fractures Research Group. Annals of Internal Medicine Dickstein K, Aarsland T, Ferrari P, Todd M & Stimpel M 1994 Comparison of the efficacy of three dose levels of moexipril versus placebo as add-on therapy to hydrochlorothiazide in patients with moderate hypertension. Journal of Cardiovascular Pharmacology Eriksen EF, Hodgson SF, Eastell R, Cedel SL, O Fallon WM & Riggs BL 1990 Cancellous bone remodeling in type I (postmenopausal) osteoporosis: quantitative assessment of rates of formation, resorption, and bone loss at tissue and cellular levels. Journal of Bone and Mineral Research Hall TJ & Schaueblin M 1994 Hydrochlorothiazide inhibits osteoclastic bone resorption in vitro. Calcified Tissue International Hatton R, Stimpel M & Chambers TJ 1997 Angiotensin II is generated from angiotensin I by bone cells and stimulates osteoclastic bone resorption in vitro. Journal of Endocrinology Jayo MJ, Register TC, Calson CS, Rankin SE & Sulistiawati E 1994 Effects of estrogen and thiazide on bone mineral content and serum chemistries of ovariectomized cynomolgus monkeys. Journal of Bone and Mineral Research 9 (Suppl) B138. Jee WSS, Inoue MJ, Jee KW & Haba T 1983 Histomorphometric assay of the growing long bone. In Handbook of Bone Morphology, pp Ed H Takahashi. Niigata City, Japan: Nishimura Co. Ltd. Jee WSS, Mori S, Li XJ & Chan S 1990 Prostaglandin E 2 enhances cortical bone mass and activates intracortical bone remodeling in intact and ovariectomized female rats. Bone

8 474 YFMAand others Moexipril and HCTZ in hypertensive OVX rats Lucas PA, Brown RC, Drueke T, Lacour B, Metz JA & McCarron DA 1986 Abnormal vitamin D metabolism, intestinal calcium transport, and bone calcium status in the spontaneously hypertensive rat compared with its genetic control. Journal of Clinical Investigation Maclntyre I, Zaidi M, Alam ASMT, Datta HK, Moonga BS & Lidbury P 1991 Osteoclastic inhibition: an action of nitric oxide not mediated by cyclic GMP. Proceeding of the National Academy of Sciences of the USA Merke J, Lucas PA, Szabo A, Cournot WG, Mall G, Bouillon R, Drueke T, Mann J & Ritz E 1989 Hyperparathyroidism and abnormal calcitriol metabolism in the spontaneously hypertensive rat. Hypertension Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PG, Ott SM & Recker RR 1987 Bone histomorphometry: Standardization of nomenclature, symbols, and units. Journal of Bone and Mineral Research Peh CA, Horowitz M, Wishart JM, Need AG, Morri HA & Nordin BE 1993 The effect of chlorothiazide on bone-related biochemical variables in normal post-menopausal women. Journal of the American Geriatrics Society Safar M, Stimpel M & Zanchetti A (Eds) 1994 Hypertension in Postmenopausal Women. Berlin: Springer-Verlag. Scalbert E, Abdon D, Devissaguet M & Juggi JS 1992 Interaction between an angiotensin converting enzyme inhibitor, perindopril, and a thiazide diuretic in the spontaneously hypertensive rat. Canadian Journal of Cardiology Song X & Wergedal JE 1993 Hydrochlorothiazide stimulates proliferation of human osteoblasts in vitro. Journal of Bone and Mineral Research 8 S362. Steiniche T, Mosekilde L, Christensen MS & Melsen F 1989 Histomorphometric analysis of bone in idiopathic hypercalciuria before and after treatment with thiazide. APMIS Stimpel M, Jee WSS, Ma YF, Yamamoto N & Chen YY 1995 Pharmacological inhibition of angiotensin converting enzyme alone or in combination with 17β-estradiol in the treatment of ovariectomy (postmenopausal)-induced osteoporosis in rats. Journal of Hypertension Wang TM, Hsu JF, Jee WSS & Mattews JL 1993 Evidence for reduced cancellous bone mass in the spontaneously hypertensive rat. Bone and Mineral Wise KJ, Bergman EA, Sherrard DJ & Massey LK 1996 Interactions between dietary calcium and caffeine consumption on calcium metabolism in hypertensive humans. American Journal of Hypertension Wronski TJ, Schlenck PA, Cintron M & Walsh CC 1987 Effect of body weight on osteopenia in ovariectomized rats. Calcified Tissue International Wronski TJ, Yen C-F & Scott KS 1988 Estrogen and diphosphonate treatment provide long-term protection against osteopenia in ovariectomized rats. Journal of Bone and Mineral Research Yamori Y, Tomimoto K, Oshima A, Hazama F & Okamoto K 1974 Development course of hypertension in the SHR-substrains susceptible to hypertensive cardiovascular lesions. Japanese Heart Journal Received 19 November 1996 Accepted 7 April 1997