Differential effects of alendronate treatment on bone from growing osteogenesis imperfecta and wild-type mouse

Bone 36 (2005) 150 – 158 www.elsevier.com/locate/bone

Differential effects of alendronate treatment on bone from growing osteogenesis imperfecta and wild-type mouse Barbara M. Misof a, Paul Roschger a,*, Todd Baldinib, Cathleen L. Raggioc, Vivien Zraickb, Leon Rootc, Adele L. Boskeyb, Klaus Klaushofer a, Peter Fratzld, Nancy P. Camachob a

Ludwig Boltzmann Institute of Osteology at the Hanusch Hospital of WGKK and AUVA Trauma Centre Meidling, 4th Med. Department Hanusch Hospital, Kundratstr. 37, A-1120 Vienna, Austria b Research Division, The Hospital for Special Surgery, New York, NY 10021, USA c Department of Orthopaedics, The Hospital for Special Surgery, New York, NY 10021, USA d Max Planck Institute of Colloids and Interfaces, Department of Biomaterials, 14424 Potsdam, Germany Received 1 June 2004; revised 8 October 2004; accepted 12 October 2004 Available online 23 November 2004

Abstract Bisphosphonates have been reported to decrease the number of fractures in children with osteogenesis imperfecta (OI). The current study sought to further explore bisphosphonate-associated bone changes in OI by investigating the effects of alendronate (ALN) treatment on bone mechanical and material properties in osteogenesis imperfecta (oim/oim) and wild-type (+/+) mice treated with 26–73 Ag kg 1 day 1 of ALN for 8 weeks via subcutaneously implanted pumps. Femoral three-point bend tests to evaluate cortical bone were combined with geometric and material density analysis. Cortical and trabecular architecture of metaphyseal bone were histomorphometrically evaluated and material density assessed by quantitative backscattered electron imaging (qBEI). For the cortical oim/oim bone, which revealed principally inferior biomechanical properties compared to +/+ bone, ALN neither improved cortical strength or any other mechanical property, nor affected cortical width (Ct.Wi.) or material density. In contrast, for the +/+ mice, bone strength was enhanced (+22%, P b 0.05) though coupled with increased brittleness (+28%, P b 0.05). This mechanical improvement was associated with an increase in Ct.Wi. (+17.3%, P = 0.02) and a reduction in heterogeneity of cortical mineralization (CaWidth, 4%, P = 0.04). In the metaphysis, ALN raised cancellous bone volume (BV/ TV) significantly in oim/oim as well as in +/+ mice (+97%, P = 0.008 and +200%, P b 0.0001, respectively). This occurred without any change in either material density or trabecular thickness (Tb.Th.) in the oim/oim mice, while in the +/+ mice, material density increased slightly but significantly (+3%, P = 0.004), and Tb.Th. increased by 77% ( P b 0.0001). Taken together, these results illustrate the differential effects of ALN on oim/oim vs. +/+ bone, as well as on specific skeletal sites, i.e., cortical vs. trabecular bone. ALN augmented the mechanical, geometrical, and material properties of +/+ cortical and trabecular bone, while the only observable improvement to the oim/oim bone was increased cancellous bone volume. This suggests that in this mouse model of OI, the previously demonstrated bisphosphonateassociated reduction in fractures is primarily attributable to increased metaphyseal bone mass. D 2004 Elsevier Inc. All rights reserved. Keywords: Osteogenesis imperfecta; Bisphosphonates; Bone mineralization density distribution; Quantitative backscattered electron imaging; Biomechanics

Introduction Antiresorptive agents play a prominent role in the prevention and treatment of osteoporosis and other metabolic

* Corresponding author. Fax: +43 1 60150 2651. E-mail address: [email protected] (P. Roschger). 8756-3282/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2004.10.006

bone disorders [1–5]. In the disease osteogenesis imperfecta (OI), current advances in the use of bisphosphonates have shown great promise [6]. OI is a heritable disorder of the connective tissues caused predominantly by mutations in the genes that encode for type I procollagen [7–9]. Patients with OI exhibit reduced bone density, altered bone histomorphometry [10], abnormal bone mineralization [11–13], multiple fractures, and in more severe cases, skeletal deformities.

B.M. Misof et al. / Bone 36 (2005) 150–158

Historically, treatment to decrease fracture rate in children with OI was limited to bracing of the limbs and stabilization of the bone structure by insertion of intramedullary rods [14]. Studies by Glorieux et al. [15] and Plotkin et al. [16], however, reported that treatment with the bisphosphonate pamidronate improved clinical outcome in children with OI. In both studies, bone mineral density increased and fracture rate decreased after 1–5 years of therapy. In addition, the histomorphometrical analysis of paired transiliac bone biopsies from these patients before and after treatment revealed a 65% and 88% increase of BV/TV and Ct.Wi., respectively [15]. Although these initial results of bisphosphonate therapy on OI are quite encouraging, several important issues must be considered regarding the effect of bisphosphonates on bone quality. In general terms, bone quality can be defined as the material and compositional properties that mediate the normal functioning of a bone. These properties include, but are not limited to, the overall macro-architecture of the bones (such as geometrical parameters and bone mass), the microarchitecture (trabecular structures), the material strength of the tissue, and the composition of the mineral and organic matrix components at the ultrastructural level. These parameters and the concerns about bone quality are not easily studied in humans. Some of these questions have been investigated in our laboratory in a naturally occurring animal model of OI, the oim/oim mouse [17]. Mice homozygous for the oim mutation are deficient in proa2(I) collagen, and only homotrimeric (a1(I)3) collagen is deposited in the extracellular matrix of oim/oim mice [18]. The oim/oim mice exhibit a moderate-to-severe OI phenotype characterized by osteopenia, skeletal fracture and deformities, a phenotype similar to that observed in a clinical correlate [19]. Although this specific mutation is not typical of OI, the oim/oim mouse provides a model in which fractures occur naturally, and therefore the potential for therapies in OI to reduce fractures can be readily monitored on a fixed genetic background. Bone from the oim/oim animals has also been characterized by its extreme brittleness, which arises from both collagen and mineral components. Alterations in collagen molecules cause deviations in collagen structure from the normal [20], including decreased cross-linking [21], highly reduced mechanical quality [22] and decreased quantity of collagen [23] together with altered nanostructure [24] and abnormally high calcium concentration [25,26]. Increased material density has been shown to be directly correlated with increased brittleness [27–29] and decreased collagen content [30,31]. Furthermore, increased ash weight (material density) [31] has been linked to brittle bone in other mouse models. We recently reported that treatment of growing oim/oim mice with the third generation bisphosphonate alendronate (ALN) over an 8-week period, from 6 to 14 weeks of age, reduced fracture incidence by almost threefold. This was accompanied by a 50% increase in radiographic femoral

151

metaphyseal density compared to nontreated mice [17], whereas mineral crystallinity as measured by Fourier transform infrared spectroscopy was not affected. In addition, cortical density did not improve in the treated oim/oim mice, but did increase in the treated wild-type +/+ mice. In another recent study designed to evaluate treatment of infant mice, ALN treatment from 2 to 14 weeks of age improved bone density in the oim/oim femur and spine, but also resulted in significantly shorter femur lengths in the +/+ mice [32]. Thus, the effects for the oim/ oim and +/+ mice were not equivalent. Further, it has been demonstrated that high doses of bisphosphonates in animal studies lead to accumulation of microdamage in the bone, potentially leading to a more brittle material [33–35]. These results are particularly germane given the rising interest in the quality of bone formed during the course of bisphosphonate therapy [36,37]. In particular, the abnormally high material density in oim/oim mice [25,26] and human OI [11] bone could be a critical determinant in how the bone responds to treatment with bisphosphonates, which have been shown to increase material density due to their antiresorptive properties [38,39]. In the current study, we investigated whether alendronate treatment would have equivalent effects on bone quality in growing oim/oim and +/+ mice. Femoral threepoint bend biomechanical tests were combined with geometric analysis to derive cortical bone material properties, and morphometric analysis for the quantification of the microarchitecture, and quantitative backscattered electron imaging measurements (qBEI) for the evaluation of material density (degree of mineralization) were performed.

Materials and methods Animals All animal procedures were performed under an IACUC-approved protocol. Breeder mice heterozygous for the oim mutation, oim/+, were initially obtained from the Jackson Laboratory (Bar Harbor, ME). Since the survival rate of the oim/oim mice is low (~50%, due to frailty at birth), additional oim/oim–oim/+ pairs were bred to obtain an increased number of oim/oim mice. Mice were kept at ~248C (758F) and received Purina rodent chow 5001 and tap water ad libitum. Pups were weaned at 4 weeks of age, and 1-cm-long tail snips were obtained for genotype determination. Genomic DNA was extracted from mouse tail snip samples using a DNA Isolation kit (Invitrogen, San Diego, CA) and amplified by the PCR technique using specific primers [40]. PCR products were then analyzed on an ABI Prism 377 automated DNA sequencer (Perkin Elmer, Foster City, CA). Details of the genotyping procedure were described previously [41].

152

B.M. Misof et al. / Bone 36 (2005) 150–158

Alendronate treatment Treatment began at 6 weeks of age and continued for 8 weeks, a rapid growth period in mice considered to be analogous to adolescence in children, just prior to closing of the long bone growth plates. Genotype-identified oim/oim and +/+ mice (n = 12–18) received either alendronate (ALN) (generously provided by Merck and Co., West Point, PA) at a dosage of 73 Ag alendronate/kg/day for 4 weeks followed by 26 Ag kg 1 day 1 for 4 weeks or saline (as control) via subcutaneously implanted Alzet pumps (Alza Corp., Palo Alta, CA). Determination of an appropriate dosage was carried out in an earlier study, and is in the range of the bpharmaceutically effective doseQ for mice recommended by Merck [17]. No difference in activity level was noted between the ALN and saline-treated mice. Mice were sacrificed by methoxyflurane inhalation followed by terminal exsanguination via cardiac puncture. Males and females in approximately equal number were included for all groups. Mechanical and geometrical properties Individual femora were carefully dissected at sacrifice, cleaned of soft tissue, and faxitron radiographs (Hewlett Packard, Rockville, MD) obtained in the anterior–posterior (AP) and medial–lateral (ML) planes. Femoral length and endosteal (d e) and periosteal (d p) diameters in the AP and ML planes were measured from digitized radiographs (resolution of 1524  1012 pixels) using Sigma Scan software (Jandel Scientific, San Rafael, CA) and the moment of inertia was evaluated (assuming an elliptical femoral cross section) as described previously [32,42]. Femora were kept frozen at 208C until testing, and thus underwent only one freeze–thaw cycle. Femurs that showed no fractures or obvious deformities on the radiographs were selected for mechanical testing. Three-point bend tests, performed on the intact femur to assess diaphyseal cortical bone properties, were conducted at room temperature using a closed-loop servo-hydraulic test machine (MTS Systems Corporation, Eden Prairie, MN) with Instron electronic controls (Instron Corporation, Canton, MA). The femurs were tested such that the anterior side was loaded in tension and the posterior in compression. The outside loading points were placed 8 mm apart. The tests were performed at a rate of 0.5 N/s, and load and displacement data were acquired at 20 Hz with a PC equipped with a 12 Bit A/D data acquisition board. Specimens were kept wet throughout the test. Structural mechanical properties that are dependent on both geometry and intrinsic material (tissue) properties of the bone were measured: Ultimate load was defined as the maximum load the bone sustained during testing, the stiffness, a measure of the rigidity of the whole bone, was determined by taking the slope of the elastic region of the load-displacement curve, and the work to failure, a measure of the energy required to fracture the bone, was calculated as the area under the entire curve. Stress–strain curves,

reflective of biomechanically derived intrinsic material properties, were calculated by normalization to the geometrical property moment of inertia as described in a previous work [32]. Young’s Modulus, the slope of the elastic region of the curve and a measure of the intrinsic stiffness of the material, and ultimate stress, the maximum stress before failure, were determined from the stress–strain curves. Toughness, a measure of the amount of energy needed to fracture the bone material, was calculated as the area under the stress–strain curve. Materials that sustain very little post-yield strain before fracture are termed bbrittleQ, and thus the bbrittlenessQ of the cortical bone, a material property, was calculated by dividing the yield strain (point of maximum elastic deformation) by the ultimate strain [43]. Morphometrical analysis and quantitative backscattered electron imaging (qBEI) Undecalcified distal pieces from femora previously fractured in 3-point bend tests were dehydrated in a graded series of ethanol and embedded in polymethylmethacrylate for backscattered electron imaging in the digital scanning electron microscope (SEM). Cross- and longitudinal sections were cut, polished and carbon-coated for the SEM study (N = 10 per group, +/+ and oim/oim, ALN-treated and control). Backscattered electron images (nominal magnification 20, working distance 15 mm) of cross sections and of longitudinal sections of the femora, which display the mineralized tissue exclusively, were analyzed for structural morphometric parameters (nomenclature according to Parfitt et al. [44]). For cortical bone, cortical width (Ct.Wi) and cortical area (Ct.Ar.) were measured on cross-sectional images. For trabecular bone, volume per tissue volume (BV/ TV = 100  B.Ar./T.Ar.) and trabecular thickness [Tb.Th. = (2 B.Ar.)/(1.2 B.Pm.)] at the distance of 500–1500 Am from the growth plate were determined in longitudinal sections. These measurements were performed by a custom-made automated routine using NIH Image 1.63, W. Rasband, National Institutes of Health, USA. Bone mineralization density distributions (BMDD) were measured on cortical bone in the cross sections and on metaphyseal trabecular bone in the longitudinal sections (Fig. 1) using quantitative backscattered electron imaging (qBEI) as described in detail in previous works [45,46]. Briefly, the accelerating voltage of the electron beam was adjusted to 20 kV, the probe current to 110 pA and the working distance to 15 mm in the SEM (DSM 962, Zeiss, Oberkochen, Germany). Backscattered electrons were collected by a 4-quadrant semiconductor backscattered electron (BE)-detector. Cortical and cancellous bone areas were imaged at 200 nominal magnification, corresponding to a pixel resolution of 1 Am. The digital 512  512 pixel BE images were generated by one single frame, using a scan speed of 100 s per frame. Five images within each cortical and trabecular area were taken at random per specimen. The gray levels of these digital images were determined by the

B.M. Misof et al. / Bone 36 (2005) 150–158

153

Fig. 1. Backscattered electron images of longitudinal views of femora from nontreated and alendronate-treated oim/oim (a,b) and +/+ mouse (c,d). Bone mineralization density distributions (BMDD) were measured on cortical bone in the cross sections (not shown) and on metaphyseal trabecular bone in the longitudinal sections. Morphometrical parameters were measured at the distance of 500–1500 Am from the growth plate.

intensity of the BE signal that was proportional to the weight concentration of the mineral in the bone. From the digital images, gray-level histograms were deduced displaying the percentage of bone area occupied by pixels of a certain gray level. The calibration of the images and the transformation of the gray-level histograms to BMDD was performed as previously described [45]. Two different parameters obtained from the BMDD were used for the characterization of the mineralization density (also termed material density) in this study: CaPeak, which represents the most frequently observed calcium concentration within the cortical or cancellous bone area in each sample (the peak position of the histogram) and CaWidth, which denotes the peak width (full width at half maximum) of the BMDD (indicating the heterogeneity of the Ca concentration within the specimen), were measured. A smaller CaWidth corresponds to a greater homogeneity of mineralization. Statistical analysis Statistical analyses for the geometrical and biomechanical data were performed on SigmaStat software (SPSS Inc., Chicago, IL). Univariate statistics were collected on all variables (including means, medians, ranges and standard deviations). Comparisons were made between ALN-treated and nontreated animals, and between oim/oim and +/+ animals. Although the oim gene is not sex-linked, all measurements were analyzed separately for males and females. Significant differences were not found based on gender, and thus the data for males and females were grouped together. Biomechanical as well as histomorphometric and BMDD data were analyzed by two-factor analysis of variance (ANOVA) to test for the simultaneous effects of genotype

and ALN treatment on the outcome variables. Three null hypotheses were examined: (1) there is no effect of genotype, (2) there is no effect of ALN treatment, and (3) the potential effect of ALN treatment is independent of genotype (the interaction term). If significant differences were found for any factors, multiple comparisons were performed using the Student–Newman–Keuls procedure to test for differences with alendronate treatment within either oim/oim or +/+ mice [47]. Values were considered significantly different at P b 0.05.

Results Cortical bone biomechanical data At 14 weeks of age, after the 8-week experimental treatment period, all of the structural mechanical properties (dependent on bone geometry and material properties), and the biomechanically derived intrinsic material properties of oim/oim cortical bone, were inferior compared to the +/+ cortical bone, with the exception of Young’s modulus (Table 1). This indicates that not only was the oim/oim cortical bone structure weaker as a whole, but the intrinsic material that comprises the oim/oim bone, independent of size and shape, is less competent than that of the +/+ bone, and more brittle. There were no significant changes in any of the oim/oim mechanical properties with ALN treatment. However, ALN treatment did increase the bone ultimate load, stiffness, ultimate stress, and Young’s modulus of the +/+ bone, but was coupled with a significant increase in brittleness. The disparity in the effect of ALN on +/+ vs. oim/oim bone was evidenced by the significance of the ANOVA interaction factor for Young’s modulus and ultimate stress.

154

B.M. Misof et al. / Bone 36 (2005) 150–158

Table 1 Biomechanical properties of femora from 14-week-old mice N Moment of inertia (mm4) Ultimate load (N)a Stiffness (N/mm)a Work to failure (N/mm)a Ultimate stress (MPa)b,c Young’s modulus (GPa)b,c Toughness (N m/m3)b Brittleness (%)b

+/+

ALN +/+

oim/oim

ALN oim/oim

14 0.181 F 0.049 12.33 F 1.53 26.70 F 5.14 5.58 F 2.72 89.72 F 13.94 1.690 F 0.436 4.83 F 1.371 51.11 F 11.89

12 0.161 14.70 34.72 4.04 109.2 2.351 4.95 65.30

15 0.0907 6.13 17.83 1.21 77.12 2.437 1.49 91.31

18 0.109 6.50 19.92 1.19 69.21 2.341 1.35 90.16

F F F F F F F F

0.038 2.25* 9.52* 2.49 17.29* 0.695* 1.417 19.06*

F 0.030** F 1.27** F 3.19** F 0.43** F 16.92** F 0.658** F 0.564 ** F 14.85 **

F F F F F F F F

0.041 1.48 7.64 0.34 16.88 0.588 0.431 15.48

Data are means F SD. a Parameters are dependent on bone geometry and material properties. b Intrinsic material parameters. c Two-way ANOVA interaction factor is significant. * P b 0.05 vs. nontreated group of same genotype. ** P b 0.05 oim/oim versus +/+.

Cortical bone morphometrical analysis

Trabecular bone morphometrical analysis

Significant differences in the mean values of Ct.Wi ( 18.8%, P b 0.009) and Ct.Ar ( 26.2%, P b 0.001) at the midshaft region between oim/oim and +/+ were detected prior to treatment. ALN treatment did not influence Ct.Wi. and Ct.Ar. for the oim/oim bone, but did significantly increase Ct.Wi. (+17.3%, P = 0.02) and Ct.Ar. (+16.7%, P = 0.03) for the +/+ bone (Table 2).

No significant differences in the mean values of BV/TV and Tb.Th. of metaphyseal spongiosa between oim/oim and +/+ mice could be detected prior to treatment. Even though there were some markedly low BV/TV values (b7%) in the oim/oim animals, the large variation in this parameter resulted in a mean BV/TV similar to that of the +/+. In both genotypes, ALN treatment had a significant effect on BV/TV (Fig. 1 and Table 2). BV/TV was increased by 97% ( P = 0.008) and more than 200% ( P b 0.0001) in oim/oim and +/+ mice, respectively. Interestingly, there was a substantial difference in the response to ALN for Tb.Th., which remained unchanged for the oim/oim mice, but was significantly increased in the +/+ mice by 77% ( P b 0.0001).

Cortical bone mineralization density distribution (BMDD) The BMDD measurements as performed by quantitative backscattered electron imaging (qBEI) revealed greater CaPeak values (+2.7%, P = 0.004) for the untreated oim/oim cortical bone compared to the untreated +/+ bone, indicative of a greater material density in the oim/oim mice. The CaWidth ( 20.4%, P b 0.001) was significantly less for oim/oim compared to +/+ cortical bone, indicative of a greater homogeneity of mineralization in oim/oim cortical bone. ALN treatment did not effect CaPeak for both oim/oim and wild types (Table 3, Fig. 2) but CaWidth ( 4.1%, P = 0.04) was decreased for the ALN treated +/+ cortical bone.

Trabecular bone mineralization density distribution (BMDD) Similar to cortical bone, in untreated animals, the oim/oim trabecular bone material showed a greater material density (CaPeak, +3.6%, P = 0.003) compared to +/+ bone, while there was no difference in CaWidth between the two genotypes. ALN did not have any effect on the BMDD

Table 2 Morphometrical analysis of cortical and trabecular bone of femora from 14-week-old mice +/+ a

N Ct.Wi [mm] Ct.Ar. [mm2] N BV/TVb [%] Tb.Th.b [Am]

8 0.202 0.92 10 14.8 33.9

F 0.014 F 0.11 F 4.8 F 1.7

ALN +/+

oim/oim

8 0.237 F 1.07 F 10 45.7 F 60.0 F

9 0.164 F 0.68 F 10 11.1 F 31.7 F

0.034* 0.19* 12.6 * 4.8*

ALN oim/oim 0.029 0.12 7.2 2.6

8 0.167 F 0.67 F 10 21.8 F 29.0 F

Data are means F SD. There was a significant difference in the cortical parameters between untreated oim/oim and +/+ ( P b 0.01), but not in the trabecular parameters. a Only samples with cross sectional areas of cortical bone of the midshaft region were included. b Two-way ANOVA interaction factor is significant. * P b 0.05 vs. nontreated group of same genotype.

0.31 11 7.8* 3.1

B.M. Misof et al. / Bone 36 (2005) 150–158

155

Fig. 2. Typical bone mineralization density distributions (BMDD) of all groups of mice. (a) Trabecular bone BMDDs from ALN and nontreated +/+. There is a significant shift of the peak towards higher degree of mineralization and a decrease in peak width (CaWidth) with ALN treatment. (b) Trabecular bone BMDDs from ALN and nontreated oim/oim mice. There is no significant effect of ALN. (c) Cortical bone BMDDs from ALN and nontreated +/+ mice. There is a tendency of a shift towards higher Ca concentration and a significant reduction in CaWidth. (d) Cortical bone BMDDs from ALN and nontreated oim/oim mice. No significant effect of ALN can be detected.

oim/oim mice, without any accompanying improvement in radiographically determined cortical bone density [17]. In the current study, we have evaluated several mechanical, material and morphometrical parameters to better understand this beneficial effect of bisphosphonates in growing oim/oim mouse bone, and potentially also in bone of children with OI. The untreated oim/oim mice had inferior cortical bone mechanical properties and a higher degree of mineralization than the +/+ mice, in agreement with previous studies [7,25,26,32]. The one exception to this was Young’s modulus, which was greater in the oim/oim mice compared to the +/+ mice. Since Young’s modulus is essentially a ratio

parameters in oim/oim animals but ALN increased CaPeak (+3.4%, P = 0.004) and decreased CaWidth ( 10.9%, P = 0.04) in +/+ animals, corresponding to increased material density and increased homogeneity of mineralization in the +/+ due to treatment (Fig. 2, Table 3).

Discussion Recently, we have shown that after an 8-week period of alendronate (ALN) treatment, fracture incidence of treated oim/oim mice was approximately one-third that of untreated

Table 3 Mineralization density properties of femora from 14-week-old mice +/+ N CaPeak (wt.%) (cortical bone cross section)a CaWidth (wt.% Ca) (cortical bone cross section)b,c CaPeak (wt.%) (trabecular bone, longitudinal section) CaWidth (wt.% Ca) (trabecular bone, longitudinal section)c

10 25.43 3.18 23.52 3.57

F F F F

0.46 0.14 0.52 0.29

ALN +/+

oim/oim

10 25.52 3.05 24.32 3.18

10 26.12 2.53 24.38 3.76

F F F F

0.53 0.17* 0.50* 0.26*

F F F F

ALN oim/oim 0.44** 0.08** 0.66** 0.53

Data are means F SD. a All cortical CaPeak values were significantly greater compared to trabecular CaPeak values within the same genotype/treatment group. b All cortical CaWidth values were significantly smaller compared to trabecular CaWidth values within the same genotype/treatment group. c Two-way ANOVA interaction factor is significant. * P b 0.05 vs. nontreated group of same genotype. ** P b 0.05 oim/oim versus +/+.

10 26.16 2.61 24.60 4.08

F F F F

0.51 0.14 0.65 0.50

156

B.M. Misof et al. / Bone 36 (2005) 150–158

of the stress/strain, it is apparent that the higher oim/oim value arises from the greatly reduced strain compared to the +/+ bone. Surprisingly, ALN treatment of oim/oim mice did not result in a change in cortical parameters such as Ct.Wi. or density, unlike the bisphosphonate pamidronate in clinical studies of OI children, where cortical bone changes were observed [15]. This is likely due to a combination of factors, which include the slower growth rate of the oim/oim cortical bone [17], and differences between bone cellular activity and remodeling pathways in oim/oim mouse cortical bone compared to human OI cortical bone. Notwithstanding these differences, it should be noted that in the current study, all mechanical, geometrical and material parameters of cortical bone were only assessed at mid-diaphysis. It is quite possible that improvements in oim/oim cortical bone geometry could be found in other regions, perhaps closer to the metaphysis, adjacent to the region of greatest modeling activity in these growing animals. This possibility is supported by the observation that oim/oim long bone (tibial) bowing tended to be reduced with ALN treatment [17], a finding that reflects an improvement in the overall bone geometry that would conceivably contribute to increased resistance to fracture. Interestingly, ALN treatment also did not affect the cortical mineralization pattern in the oim/oim animals. Both CaPeak (the typical calcium concentration in the sample) and CaWidth (the homogeneity of mineralization) remained unchanged. This is consistent with the biomechanical measurements that revealed no change in any structural mechanical or intrinsic material property, such as bone strength, stiffness, toughness, Young’s modulus, and no increase in the moment of inertia of the femora. On a positive note, oim/oim bone material brittleness also was not increased by the ALN treatment. In contrast to oim/oim mice, ALN improved some mechanical and material properties in the bone of the +/+ mice. The higher stiffness and ultimate load values obtained are consistent with the increase in cortical width and area measured in ALN-treated +/+ mice. Apparently, the observed increase in brittleness was compensated by an increase in ultimate stress and Young’s modulus, resulting in unchanged toughness and mechanical competence for the treated +/+ animals. Interestingly, we found an increase in material mineralization homogeneity without any change of the typical calcium concentration in cortical bone from the wild types. The increase in homogeneity of mineralization reflects a more static tissue consisting of fewer regions of newly formed bone matrix. This could be one factor that contributes to the significant increase in brittleness in the +/ + bone. Other hierarchical structures may also play an important role in this context, but remain to be investigated. In trabecular bone, we found a significant increase in metaphyseal bone volume (BV/TV) for the oim/oim animals after ALN treatment, but again no change in bone material (bone mineralization) properties. However, the increase in BV/TV in this study parallels the increased metaphyseal

density determined radiographically in our previous work on ALN-treated oim/oim mice [17], and is in agreement with results from studies of bisphosphonate-treated children and adults with OI [48–51]. This points to the increased quantity of mineralized tissue as the critical factor underlying the reduction in fracture incidence in the oim/oim mice, perhaps by rendering the bone more resistant to external stresses in the metaphyseal region. We did not observe any change in trabecular thickness, a finding also reported by others studying bisphosphonate-treated children and adults with OI [48,50]. Similarly, our previously reported finding of persistent calcified cartilage through much of the metaphyseal bone in these animals [17] is consistent with other studies of bone from bisphosphonatetreated children with OI. Interestingly, in the +/+ animals, ALN affected both trabecular bone architecture and the trabecular bone mineralization pattern. The trabecular architecture of the +/+ femora metaphyseal bone was improved by an increase in bone volume and trabecular thickness. This is also in agreement with previous studies reporting increased metaphyseal density for these mice [17,32]. Similar effects of increased cancellous bone volume and trabecular thickness have also been observed for other antiresorptive treatments: Estrogen was shown to antagonize cancellous bone loss in ovarectomized growing rats [52], and human recombinant osteoprotegerin caused a progressive increase in bone volume in growing rats [53]. In contrast to the oim/oim mice, we did find a change in trabecular bone mineralization for the +/+ after ALN treatment. CaWidth was significantly decreased and CaPeak significantly increased in trabecular bone from the +/+. Both findings are consistent with alendronate’s antiresorptive activity, and are similar to those reported in studies on the effects of ALN on bone mineralization in osteoporotic women [38,39] and in a pre-clinical study on minipigs [55]. In all the aforementioned studies, the mineralization properties of the ALN-treated tissues correspond to a more static tissue with suppressed bone turnover. Taken together, the results of the current study show a striking disparity between the effects of ALN on +/+ and oim/ oim bone. For the oim/oim mice, the effect of ALN treatment was limited to augment the volume of trabecular bone, while material characteristics such as Capeak, brittleness, and the crystal maturity (as shown in an earlier study [54]), and trabecular thickness, was unchanged. It is probable that the abnormally high calcium concentration (mineral density) reported for some cases of human OI [11] and for oim/oim mouse bone [26,25] is a critical factor towards understanding the lack of the entire spectrum of effects of ALN treatment. Conceivably, the highly elevated material density in oim/oim mouse bone reflects a state whereby the collagen–mineral composite has reached its saturation of collagen mineralization, and thus cannot accommodate any further mineral deposition. It should be emphasized that the previously reported increases in material density with bisphosphonate

B.M. Misof et al. / Bone 36 (2005) 150–158

treatment were for cases of lowered [39,56] and normal material density in diseased or control bones [38,55], and not in cases where material density was elevated prior to treatment, as in the current study. In conclusion, the current study demonstrates that in this OI mouse model, ALN treatment does not cause changes at the bone material level. However, the significantly increased metaphyseal cancellous bone volume, possibly in conjunction with overall improvements in metaphyseal geometry, sufficed to reduce fracture incidence. Future studies should address whether the lack of material changes with bisphosphonate treatment are also valid for human OI bone. In particular, for the treatment of children with OI, it could be important to elucidate whether the beneficial effects of bisphosphonates might be tempered by an eventual additional deterioration of the bone material due to suppressed bone remodeling. At this stage, however, it appears that the benefits of bisphosphonate therapy in OI, and in other pediatric bone disorders [57,58], outweigh any potential disadvantages associated with this therapy.

Acknowledgments This study was supported by NIH DE11803 (NPC), The Children’s Brittle Bone Foundation 007FY9698 (NPC), and utilized the facilities of the Core Center for Skeletal Integrity NIH AR46121. Additionally, this work was financially supported in part by the AUVA (Research funds of the Austrian workers compensation board) and by the WGKK (Viennice sickness insurance fonds). The authors thank Dr. Timothy Wright, Dr. Nadja Fratzl-Zelman, and Dr. Klaus Misof for their critical review of the manuscript. We are also grateful for the technical assistance of Elizabeth Miller, Michael Hossack, and Mona Jain (HSS), and Gerda Dinst, June Thorvig, and Phaedra Messmer (Ludwig Boltzmann Institute of Osteology, Vienna, Austria). The authors also thank Merck and Co., West Point, PA, for supplying the alendronate used in this study.

References [1] Crandall C. Risedronate: a clinical review. Arch Intern Med 2001;161:353 – 60. [2] Hochberg M. Preventing fractures in postmenopausal women with osteoporosis. A review of recent controlled trials of antiresorptive agents. Drugs Aging 2000;17:317 – 30. [3] McClung MR. Bisphosphonates in osteoporosis: recent clinical experience. Expert Opin Pharmacother 2000;1:225 – 38. [4] Black DM, Thompson DE, Bauer DC, Ensrud K, Musliner T, Hochberg MC, et al. Fracture risk reduction with alendronate in women with osteoporosis: the Fracture Intervention Trial. FIT Research Group. J Clin Endocrinol Metab 2000;85:4118 – 24. [5] Siris E. Alendronate in the treatment of osteoporosis: a review of the clinical trials. J Women’s Health Gend-Based Med 2000;9:599 – 606. [6] Camacho NP, Raggio CL, Blank RD. Osteogenesis Imperfecta. Curr Opin Orthop 2000;11:341 – 6.

157

[7] Rowe DW, Shapiro JR. Osteogenesis Imperfecta. In: Avioli LV, Krane SM, editors. Metabolic bone disease and clinically related disorders. San Diego7 Academic Press; 1998. p. 651 – 83. [8] Kuivaniemi H, Tromp G, Prockop DJ. Mutations in fibrillar collagens (types I, II, III, and XI), fibril-associated collagen (type IX), and network-forming collagen (type X) cause a spectrum of diseases of bone, cartilage, and blood vessels. Hum Mutat 1997;9:300 – 15. [9] Pihlajaniemi T, Dickson LA, Pope FM, Korhonen VR, Nicholls A, Prockop DJ, et al. Osteogenesis imperfecta: cloning of a pro-alpha 2(I) collagen gene with a frameshift mutation. J Biol Chem 1984;259: 12941 – 12944 [JID-2985121R]. [10] Rauch F, Travers R, Parfitt AM, Glorieux FH. Static and dynamic bone histomorphometry in children with osteogenesis imperfecta. Bone 2000;26:581 – 9. [11] Boyde A, Travers R, Glorieux FH, Jones SJ. The mineralization density of iliac crest bone from children with osteogenesis imperfecta. Calcif Tissue Int 1999;64:185 – 90. [12] Jones SJ, Glorieux FH, Travers R, Boyde A. The microscopic structure of bone in normal children and patients with osteogenesis imperfecta: a survey using backscattered electron imaging. Calcif Tissue Int 1999;64:8 – 17. [13] Traub W, Arad T, Vetter U, Weiner S. Ultrastructural studies of bones from patients with osteogenesis imperfecta. Matrix Biol 1994;14: 337 – 45. [14] Root L. The treatment of osteogenesis imperfecta. Orthop Clin North Am 1984;15:775 – 90. [15] Glorieux FH, Bishop NJ, Plotkin H, Chabot G, Lanoue G, Travers R. Cyclic administration of pamidronate in children with severe osteogenesis imperfecta. N Engl J Med 1998;339:947 – 52. [16] Plotkin H, Rauch F, Bishop NJ, Montpetit K, Ruck-Gibis J, Travers R, et al. Pamidronate treatment of severe osteogenesis imperfecta in children under 3 years of age. J Clin Endocrinol Metab 2000;85: 1846 – 50. [17] Camacho NP, Raggio CL, Doty SB, Root L, Ilg WA, Toledano TR, et al. A controlled study of the effects of alendronate in a growing mouse model of osteogenesis imperfecta. Calcif Tissue Int 2001;69:94 – 101. [18] Chipman SD, Sweet HO, McBride Jr DJ, Davisson MT, Marks Jr SC, Shuldiner AR, et al. Defective pro alpha 2(I) collagen synthesis in a recessive mutation in mice: a model of human osteogenesis imperfecta. Proc Natl Acad Sci U S A 1993;90:1701 – 5. [19] Nicholls AC, Osse G, Schloon HG, Lenard HG, Deak S, Myers JC, et al. The clinical features of homozygous alpha 2(I) collagen deficient osteogenesis imperfecta. J Med Genet 1984;21:257 – 62. [20] McBride Jr DJ, Choe V, Shapiro JR, Brodsky B. Altered collagen structure in mouse tail tendon lacking the alpha 2(I)chain. J Mol Biol 1997;270:275 – 84. [21] Miles CA, Sims TJ, Camacho NP, Bailey AJ. The role of the alpha2 chain in the stabilization of the collagen type I heterotrimer: a study of the type I homotrimer in oim mouse tissues. J Mol Biol 2002;321: 797 – 805. [22] Misof K, Landis WJ, Klaushofer K, Fratzl P. Collagen from the osteogenesis imperfecta mouse model (oim) shows reduced resistance against tensile stress. J Clin Invest 1997;100:40 – 5. [23] Camacho NP, Hou L, Toledano TR, Ilg WA, Brayton CF, Raggio CL, et al. The material basis for reduced mechanical properties in oim mice bones. J Bone Miner Res 1999;14:264 – 72. [24] Fratzl P, Paris O, Klaushofer K, Landis WJ. Bone mineralization in an osteogenesis imperfecta mouse model studied by small angle X-ray scattering. J Clin Invest 1996;97:396 – 402. [25] Camacho NP, Landis WJ, Boskey AL. Mineral changes in a mouse model of osteogenesis imperfecta detected by Fourier transform infrared microscopy. Connect Tissue Res 1996;35:259 – 65. [26] Grabner B, Landis WJ, Roschger P, Rinnerthaler S, Peterlik H, Klaushofer K, et al. Age and genotype dependence of bone material properties in the osteogenesis imperfecta murine model (oim). Bone 2001;29:453 – 7.

158

B.M. Misof et al. / Bone 36 (2005) 150–158

[27] Zioupos P, Currey JD, Casinos A. Exploring the effects of hypermineralisation in bone tissue by using an extreme biological example. Connect Tissue Res 2000;41:229 – 48. [28] Currey JD. The design of mineralised hard tissues for their mechanical functions. J Exp Biol 1999;202:3285 – 94. [29] Currey JD. Effects of differences in mineralization on the mechanical properties of bone. Philos Trans R Soc Lond, B Biol Sci 1984;304: 509 – 518. [30] Jepsen KJ, Schaffler MB, Kuhn JL, Goulet RW, Bonadio J, Goldstein SA. Type I collagen mutation alters the strength and fatigue behavior of Mov13 cortical tissue. J Biomech 1997;30:1141 – 7. [31] Jepsen KJ, Pennington DE, Lee YL, Warman M, Nadeau J. Bone brittleness varies with genetic background in A/J and C57BL/6J inbred mice. J Bone Miner Res 2001;16:1854 – 62. [32] McCarthy EA, Raggio CL, Hossack MD, Miller EA, Jain S, Boskey AL, et al. Alendronate treatment for infants with osteogenesis imperfecta: demonstration of efficacy in a mouse model. Pediatr Res 2002;52:660 – 70. [33] Hirano T, Turner CH, Forwood MR, Johnston CC, Burr DB. Does suppression of bone turnover impair mechanical properties by allowing microdamage accumulation? Bone 2000;27:13 – 20. [34] Mashiba T, Hirano T, Turner CH, Forwood MR, Johnston CC, Burr DB. Suppressed bone turnover by bisphosphonates increases microdamage accumulation and reduces some biomechanical properties in dog rib. J Bone Miner Res 2000;15:613 – 20. [35] Mashiba T, Turner CH, Hirano T, Forwood MR, Johnston CC, Burr DB. Effects of suppressed bone turnover by bisphosphonates on microdamage accumulation and biomechanical properties in clinically relevant skeletal sites in beagles. Bone 2001;28:524 – 31. [36] Delmas PD. How does antiresorptive therapy decrease the risk of fracture in women with osteoporosis? Bone 2000;27:1 – 3. [37] Weinstein RS. True strength. J Bone Miner Res 2000;15:621 – 5. [38] Boivin GY, Chavassieux PM, Santora AC, Yates J, Meunier PJ. Alendronate increases bone strength by increasing the mean degree of mineralization of bone tissue in osteoporotic women. Bone 2000;27: 687 – 94. [39] Roschger P, Rinnerthaler S, Yates J, Rodan GA, Fratzl P, Klaushofer K. Alendronate increases the degree and uniformity of mineralization in cancellous bone and decreases the porosity in cortical bone of osteoporotic women. Bone 2001;29:185 – 91. [40] McBride Jr DJ, Shapiro JR. Confirmation of a G nucleotide deletion in the Cola-2 gene of mice with the osteogenesis imperfecta mutation. Genomics 1994;20:135 – 7. [41] Camacho NP, Dow D, Toledano TR, Buckmeyer JK, Gertner JM, Brayton CF, et al. Identification of the oim mutation by dye terminator chemistry combined with automated direct DNA sequencing. J Orthop Res 1998;16:38 – 42. [42] Roark RJ, Young WC. Formulas for stress and strain. New York7 McGraw-Hill, Inc; 1975. p. 292. [43] Turner CH, Burr DB. Basic biomechanical measurements of bone: a tutorial. Bone 1993;14:595 – 608 [JID-8504048]. [44] Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, et al. Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR histo-

[45]

[46]

[47] [48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

morphometry nomenclature committee. J Bone Miner Res 1987; 2:595 – 610. Roschger P, Plenk H, Klaushofer K, Eschberger J. A new scanning electron microscopy approach to the quantification of bone mineral distribution: backscattered electron image grey-levels correlated to calcium K alpha-line intensities. Scanning Microsc 1995;9:75 – 86. Roschger P, Fratzl P, Eschberger J, Klaushofer K. Validation of quantitative backscattered electron imaging for the measurement of mineral density distribution in human bone biopsies. Bone 1998;23: 319 – 26. Zar JH. Biostatistical analysis. New Jersey7 Prentice Hall; 1999. p. 260 – 1. Rauch F, Travers R, Plotkin H, Glorieux FH. The effects of intravenous pamidronate on the bone tissue of children and adolescents with osteogenesis imperfecta. J Clin Invest 2002;110: 1293 – 9. Rauch F, Plotkin H, Zeitlin L, Glorieux FH. Bone mass, size and density in children and adolescents with osteogenesis imperfecta: effect of intravenous pamidronate therapy. J Bone Miner Res 2003;18:610 – 4. Shapiro JR, McCarthy EF, Rissuter K, Ernest K, Gelman R, Fedarko N, et al. The effect of intravenous pamidronate on bone mineral density, bone histomorphometry, and parameters of bone turnover in adults with type IA osteogenesis imperfecta. Calcif Tissue Int 2003;72:103 – 12. Adami S, Gattl D, Colapietro F, Francassi E, Braga V, Rossini M, et al. Intravenous neridronate in adults with osteogenesis imperfecta. J Bone Miner Res 2003;18:126 – 30. Schmidt IU, Wakley GK, Turner RT. Effects of estrogen and progesterone on tibia histomorphometry in growing rats. Calcif Tissue Int 2000;67:47 – 52. Capparelli C, Morony S, Warmington K, Adamu S, Lacey D, Dunstan CR, et al. Sustained antiresorptive effects after a single treatment with human recombinant osteoprotegerin (OPG): a pharmacodynamic and pharmacokinetic analysis in rats. J Bone Miner Res 2003;18:852 – 8. Camacho NP, Carroll P, Raggio CL. Fourier transform infrared imaging spectroscopy (FT-IRIS) of mineralization in bisphosphonatetreated oim/oim mice. Calcif Tissue Int 2003;72:604 – 9. Roschger P, Fratzl P, Klaushofer K, Rodan G. Mineralization of cancellous bone after alendronate and sodium fluoride treatment: a quantitative backscattered electron imaging study on minipig ribs. Bone 1997;20:393 – 7. Haas M, Leko-Mohr Z, Roschger P, Kletzmayr J, Schwarz C, Domenig C, et al. Osteoprotegerin and parathyroid hormone as markers of high-turnover osteodystrophy and decreased bone mineralization in hemodalysis patients. Am J Kidney Dis 2002; 39:580 – 6. Cundy T, Wheadon L, King A. Treatment of idiopathic hyperphosphatasia with intensive bisphosphonate therapy. J Bone Miner Res 2004;19:703 – 11. Srivastava T, Alon US. The role of bisphosphonates in diseases of childhood. Eur J Pediatr 2003;162:735 – 51.