... Institute of Medical and Veterinary Science, Box 14, Rundle Mall Post Office, Adelaide, South ... Six-month-old, female Sprague Dawley rats (n -- 126) were ob-.
Calcif Tissue Int (1996) 59:121-127
Tissue hltentafioJud 9 1996 Springer-Verlag New York l n c
Increased Bone Resorption Precedes Increased Bone Formation in the Ovariectomized Rat N. A. Sims, 1'3 H. A. Morris, t'3 R. J. Moore, z T. C. Durbridge 2 ~Division of Clinical Biochemistry, Institute of Medical and Veterinary Science, Box 14, Rundle Mall Post Office, Adelaide, South Australia 5000 -'Division of Tissue Pathology, Institute of Medical and Veterinary Science, Box 14, Rundle Mall Post Office, Adelaide, South Australia 5000 ~Department of Physiology,University of Adelaide, South Australia Received: 22 September 1995 / Accepted: 15 January 1996
Abstract. This study describes an increase in biochemical and histomorphometric markers of bone resorption prior to increased bone formation and trabecular bone loss in the ovariectomized rat. Six-month-old, female Sprague Dawley rats were either sham operated or ovariectomized (Ovx) and killed at 0, 6, 9, 15, 18, 21, and 42 days postoperation when femora were collected and trabecular bone volume (BV/TV) was determined from von Kossa silver-stained sections using the Quantimet 520 image analysis system in the distal region. A number of these sections were also examined unstained for fluorochrome labels, and stained for acid phosphatase to detect osteoclast-like cells (ACP surface). At 18 days postoperation, lumbar vertebrae were examined. Blood and urine specimens were analyzed for bone-related biochemical variables. A C P surface was significantly greater in Ovx rats compared with sham at 6 days postoperation (mean ACP surface (%TS) _+SEM: sham 36.4 + 1.9; Ovx 40.3 + 1.2, P < 0.05) as was urinary hydroxyproline excretion. Serum osteocalcin and alkaline phosphatase activity were not elevated in Ovx rats compared with Sham until 9 days postoperation. Mineral apposition rate (MAR) was increased at 12 days after ovariectomy (mean MAR (ixm/day) + SEM: sham 0.85 + 0.06; Ovx 1.23 _+ 0.06, P < 0.05). Trabecular bone volume (BV/TV) at a specific site in the metaphyseal-diaphyseal core area was significantly lower at 15 days postoperation (mean (%) _+ SEM: Sham 7.40 _+ 1.23, Ovx 4.25 0 0.65, P < 0.05). There was no difference in lumbar vertebral BV/TV between the two groups at 18 days postoperation, however, ACP surface was elevated in the Ovx rats (P < 0.05). A systemic increase in bone resorption at 6 days postovariectomy precedes increased formation whereas the length of time required for the dissolution of trabeculae postoperation is determined locally.
over at the level of the basic multicellular unit of bone remodeling (BMU) with a greater level of bone resorption relative to bone formation [1, 2], or increased BMU activation frequency by itself, giving rise to increased trabecular perforation, thus providing a physical basis for inadequate bone formation [4]. Alternatively, a primary effect of estrogen deficiency on the calcium economy has been suggested to produce a secondary effect on bone mineral status [5]. The mature ovariectomized rat is a suitable model for studying bone loss due to estrogen deficiency [6], displaying both increased bone turnover [6-9] and rapid bone loss [10, 11], effects that can be blocked by estradiol treatment [8, 11]. Wronski et al. [7] reported increased histomorphometric markers of bone resorption and formation coincident with trabecular bone loss in the proximal tibial metaphysis at 15 days postovariectomy, and biochemical markers of bone turnover have been found to be significantly elevated at 21 days after ovariectomy [9]. More recently, Dempster et al. [12] demonstrated cancellous bone loss in the proximal tibial metaphysis by day 10 after ovariectomy. This loss was preceded, at day 5, by increased bone resorption, with increased bone formation not detected until day 10 when bone loss has already begun. Here we report data of both local histomorphometric variables at the distal femur and vertebrae, as well as systemic biochemical markers of bone turnover prior to and following this time period of bone loss. We confirm that increased bone resorption precedes the increase in formation, but find the delay in bone formation may be as short as 3 days. Though the increase in resorption is systemic, the time required for the abrupt disappearance of trabeculae is a local phenomenon. Short-term alterations in serum and urinary parameters of calcium and phosphate metabolism and their relationship to the bone loss after Ovx are also described.
Key words: Ovariectomy - - Histomorphometry - - Biochemical markers - - Bone turnover - - Estrogen deficiency - - Calcium.
Materials and Methods
Estrogen deficiency at the menopause causes increased bone turnover with bone loss throughout the skeleton [1-3]. The exact sequence of events remains unclear and a number of theories have been proposed including imbalanced turnCorrespondence to: H. A. Morris at the Division of Clincial Bio-
chemistry
Six-month-old, female Sprague Dawley rats (n -- 126) were obtained from Gilles Plains Animal Resource Center (Gilles Plains, South Australia). Between 0900 and 1200 hours rats were supplied with tap water ad libitum and 15 g/day commercial rat chow containing 0.76% calcium, 0.46% available phosphorus, and 2000 IU/kg vitamin D 3 (Milling Industries, South Australia). The amount of food supplied was slightly below the average daily food consumption of 16.3 g/rat/day determined over a 2-week period to prevent weight gain in Ovx rats [6]. All animal procedures were approved by the IMVS Animal Ethics Committee.
122 For histomorphometric measurements, six animals were killed as a baseline group. The remaining animals were randomly selected for sham operation, in which ovaries were exposed but not removed, or ovariectomy, both performed under halothane anesthesia. At 6, 9, 15, 18, 21, and 42 days postoperation, 10 Sham and 10 Ovx rats were killed by cervical dislocation under halothane anesthesia, and femora were removed. In animals killed at day 18, lumbar vertebrae were also removed. Bone samples were fixed in 10% buffered formol-saline for 24 hours and stored in 70% ethanol until further processing. Before killing, calcein (Sigma Chemical Co, St Louis, USA) and demeclocycline (Lederle, Sydney, Australia), both at 10 mg/kg body weight, were administered i.p. at 9 and 3 days, respectively, to all rats except those killed at 6 and 42 days postop. Animals killed at each time point were from the same breeding group, allowing a cross-sectional analyses of the histomorphometric data. For biochemical measurements, food was withheld from animals 36 hours before blood collection and 12 hours before 24-hour urine collections were commenced. Urine was collected at 3, 9, 15, 21, and 42 days from animals killed at 42 days, and 6 days postoperation from those killed at day 15. Blood was collected from the tail vein under halothane anesthesia, followed by 2 ml saline injection i.p. to replace lost fluid. At day 42, blood was collected by cardiac puncture under halothane anesthesia. Serum and urine samples were stored at -70~ until analyzed.
Biochemistry Serum levels of total calcium, phosphate, albumin, creatinine, and alkaline phosphatase activity (ALP) were analyzed by standard methods on Technicon SMAC II (Tarrytown, New York, USA) using recommended reagents. Serum osteocalcin was measured by radioimmunoassay (RIA) as described previously [9]. Urinary hydroxyproline was determined by the method of Bergman and Loxley [13]. Urinary phosphate was determined in acidified urine specimens (pH < 1) by colorimetry and molybdate [14] on Roche Cobas Bio (Roche, Switzerland) centrifugal analyzer. Urinary creatinine was measured on Olympus Reply (Integrated Sciences, Sydney, Australia) by a picric acid reaction. Maximal renal tubular phosphate reabsorption (TraP) was calculated by the method of Marshall [15]. Urinary bydroxyproline (OHPrE) and phosphate excretions were calculated by multiplying the ratio to urinary creatinine by serum creatinine levels. Serum 17[3-estradiol was measured in Sham and Ovx rats at day 42. Levels were determined in samples extracted with 60% cyclohexane/40% ethyl acetate and reconstituted in human charcoal-stripped serum using a commercial RIA kit (Spectria, Orion Diagnostica, Turku, Finland).
N. A. Sims et al.: Yime Course of Bone Loss After Ovariectomy
Fig. 1. The distal third of a rat femur from an Ovx rat stained by a modified yon Kossa method. Morphometric parameters were measured in the metaphyseal-diaphyseal core. ACP surface, mineral apposition rate, and extent of double fluorochrome-labeled surface were measured in the anterior metaphysis.
postovariectomy with relatively high numbers of bone cells present [17]. In contrast, the metaphyseal-diaphyseal core area was unsuitable for such measurements because the extensive loss of bone resulted in highly variable quantities of trabecular bone surface. Extent of double fluorochrome-labeled surface as a percentage of trabecular bone surface (dLS) was determined on unstained sections by point counting using a Zeiss fluorescence microscope with an ocular mounted Weibel II graticule at x 100 magnification. Interlabel distance was measured at multiple sites along the labeled surface using an ocular-mounted micrometer at x400 magnification. Mineral apposition rate (MAR) was calculated by dividing the interlabel distance by the time interval between labels (6 days). The extent of osteoclast-like, acid phosphatase-positive cells located on bone surface (ACP surface) were determined as percentage of total surface (%TS) in sections stained for acid phosphatase using an enzyme histochemical method [19] by point counting at x200 magnification. Two-way analysis of variance and linear regressions were performed using Minitab 8.0 statistics package. Significant differences were identified by Tukey's Post-Hoc test. Serum 1713estradiol levels were analyzed by the Kruskal-Wallis test, as levels were not normally distributed. Biochemical and histomorphometric data from randomly paired rats in each operation were analyzed by univariate regression. P < 0.05 was considered significant.
Results
Histomorphometry The distal third of each femur and the lumbar vertebral bodies were bisected in the sagittal plane using a low speed saw (Buehler Ltd, Evanston, USA), with a diamond-tipped cutting blade. Onehalf was dehydrated at room temperature in graded ethanol and embedded undecalcified in 50:50 glycol:methyl methacrylate; 5-1,-thick longitudinal sections were cut using a Jfing K motorized microtome (Reichert, Heidelberg, Germany). The percentage of marrow cavity occupied by mineralized bone and its surface length were measured in sections stained with a modified yon Kossa silver technique [16]. Measurements were made in the metaphyseal-diaphyseal core (Fig. 1), an area of bone highly sensitive to ovariectomy [17]. An Olympus BH-2 microscope (Olympus, Tokyo, Japan) interfaced with a Quantimet 520 Image Analysis System (Cambridge Instruments, Cambridge, UK) was used for all measurements. Trabecular bone volume (BV/TV), trabecular number (Tb.N), and trabecular thickness (Tb.Th) were calculated from the percentage area and surface area obtained [18]. Dynamic markers of bone turnover were measured in the secondary spongiosa of the anterior metaphyseal area (Fig. 1). This region maintains a considerable amount of bone for up to 9 weeks
Serum estradiol levels in Ovx rats were significantly lower than Sham levels at 42 days postoperation. Median values and ranges were as follows: sham 27.5, (8-230) pmol/liter; ovx 6.5, (1-9) pmol/liter. The value range in Sham rats was consistent with variation over the 4 - 5 day estrous cycle of the rat [20]. The change in OHPrE from preoperative levels for Sham and Ovx rats was calculated for each rat, and analyzed for the effect of ovariectomy (Fig. 2). OHPrE increased significantly in Ovx rats, and was significantly greater compared to Sham rats from 6 days postoperation onwards. ACP surface was significantly greater in Ovx compared with sham rats; mean values (%TS) _+ SEM: Sham 30.7 _+ 1.3; Ovx 39.0 _+ 1.3 (P < 0.001). The ratios of ACP surface of Ovx to Sham rats (as a percentage) at each time point are presented in Figure 3. The Ovx increase of ACP surface was first detectable at 6 days postoperation (P < 0.05). There was a significant positive correlation between femoral ACP surface and OHPrE in Ovx rats, described by the equation:
N. A. Sims et al.: Time Course of Bone Loss Alter
Ovariectomy
0.7
LJ_ 0.6
(.9 ._1
~-o E
/ ~J]i
0.5 0.4
"-
y,
"'"
i
0.2
123
slightly, though not significantly, depressed in Sham rats at this time point. MAR in the femoral anterior metaphysis was significantly increased in both Sham and Ovx rats at 6 days postop compared with preop (Fig. 5). In Sham rats, MAR returned to preop levels at 12 days postop, whereas in Ovx rats, this marker continued to rise, remaining greater than levels in Sham rats for the duration of the experiment, dLS in this area was increased in Ovx rats compared with Sham levels from day 12 postop (Fig. 5). ALP and dLS were significantly correlated in Ovx rats with the relationship described by the regression equation dLS = 19.3 + 0.098 • ALP; R = 0.125; P = 0.047.
0.0 -0.1
0
6
12
18 24 30
36 42
These variables were not significantly correlated in Sham rats. Serum osteocalcin and MAR also were significantly correlated in Ovx rats, but not in Sham rats. In Ovx rats, the regression equation was
Days Post-Operation Fig. 2. The change in urine hydroxyproline excretion (OHPrE) was significantly greater in Ovx (El) compared with the Sham rats (11) from 6 days postoperation. Values are mean 4- SEM. *P < 0.05 vs Sham by two-way analysis of variance and Tukey post hoc test.
~.200 ~"
*
180
9
9
t I/
~120 .~ m
~
........."
I j"
"". i ....
100
80
i
-
~ F
12 18 24 30 36 42 Days Post-Operation Fig. 3. The ratio of Ovx to Sham ACP surface (expressed as a percentage) increased until 21 days postop. *Time points when the Ovx level was significantly greater than the Sham level (P < 0.05). 0
6
ACP surface = 32.6 + 10.8 • OHPrE; R = 0.115; P = 0.025. This relationship was not significant in Sham rats. In Ovx rats, serum osteocalcin was increased above Sham levels at 9 days postop, reaching a plateau at 15 days postop, and maintained for the duration of the experiment (Fig. 4). There was a transient increase in serum osteocalcin in Sham rats at 3 and 6 days postop compared with preop levels which normalized at 9 days postop and remained stable for the duration of the experiment. ALP was significandy greater in Ovx rats at 9 days postop compared with Sham controls, remaining elevated for the duration of the experiment (Fig. 4). ALP was significantly reduced compared with preop levels in Ovx rats at 6 days postop and was
MAR = 0.72 + 0.015 • osteocalcin; R = 0.125; P = 0.047. BV/TV in the diaphyseal-metaphyseal core was significantly lower in Ovx rats compared with Sham controls at day 15, with no further loss of trabeculae detected in this area (Fig. 6). The loss of bone volume was accompanied by a concomitant decrease in Tb.N (Table 1). Tb.Th did not differ significantly between Sham and Ovx rats, although there was a slight but significant increase with time in both operation groups (Table 1). BV/TV and Tb.N were nonsignificantly increased at 6 days postop in both Sham and Ovx rats after operation, consistent with the rise in MAR and osteocalcin. At 18 days postop, vertebral trabecular bone volume, thickness, and number were not significantly altered by ovariectomy (Table 2). Vertebral ACP surface was greater in Ovx compared with Sham but these levels were significantly lower than levels measured in the distal femur at day 18 postop. There was no significant change in total serum calcium related to ovariectomy (Table 3). Serum albumin was reduced in both Sham and Ovx rats at 3 days postop. However, serum albumin in Sham rats returned to preop levels by day 21, whereas levels in Ovx rats remained significantly reduced at all times postoperation (Table 4). Serum phosphate and TmP differed significantly between Sham and Ovx rats at time of operation (mean values ___SEM; phosphate (mmol/liter): Sham 1.50 _+ 0.04; Ovx 1.44 _ 0.02 (P < 0.05); Trap (retool/liter GF): Sham 1.78 + 0.06; Ovx 1.68 + 0.03 (P < 0.05)). The change in both variables from the time of operation was analyzed to determine the effect of ovariectomy. Serum phosphate was increased in Ovx rats compared with Sham controls from day 6 postop for the duration of the experiment, as was the tubular maximum of phosphate reabsorption (Table 4). There was no significant effect of time or operation on phosphate excretion. Serum phosphate and OHPrE were significantly related in Ovx rats but not in Sham. The regression equation in Ovx rats was: P O 4 ---
1.61 + 0.435 • OHPrE, R = 0.180,
P < 0.005.
Discussion
This study confirms and extends the report of Dempster et al. [12] by describing the time course of increased bisto-
124
281
N. A. Sims et al.: Time Course or" Bone Loss After Ovariectomy
t
t
110
26
- -..
,,,'
100 -1
A
~:~24
,'
~" 90
r-.
'o 22
a. 80
o o
i'
70
u) 20
,;
6o i.
O 18 16
'
I
0
'
6
I
'
I
12
'
18
I
'
I
24
'
30
I
'--]
36
50
}
40
' I ' I 0 6 12
42
I ' 1 ' I ' I ' I 18 24 30 36 42
Days Post-Operation
Days Post,Operation
Fig. 4. Biochemical markers of bone formation were increased by 9 days after ovariectomy. Serum osteocalcin and alkaline phosphatase activity (ALP) increased in Ovx ([-1) compared with Sham controls (B), significant from 9 days postop until the end of the experiment. Values are mean + SEM. *P < 0.05 vs Sham by two-way analysis of variance and Tukey post hoc test.
1.3 1.2
"~ 1.1
7
"""
''
30
'o
E ::L 1.0 m
