Sclerostin Antibody Improves Skeletal Parameters in a Brtl/+ Mouse Model of Osteogenesis Imperfecta

Abstract

Osteogenesis imperfecta (OI) is a genetic bone dysplasia characterized by osteopenia and easy susceptibility to fracture. Symptoms are most prominent during childhood. Although anti-resorptive bisphosphonates have been widely used to treat pediatric OI, controlled trials showed improved vertebral parameters but equivocal effects on long-bone fracture rates. New treatments for OI are needed to increase bone mass throughout the skeleton. Sclerostin antibody (Scl-Ab) therapy is potently anabolic in the skeleton by stimulating osteoblasts via the canonical wnt signaling pathway, and may be beneficial for treating OI. In this study, Scl-Ab therapy was investigated in mice heterozygous for a typical OI-causing Gly->Cys substitution in col1a1. Two weeks of Scl-Ab successfully stimulated osteoblast bone formation in Brtl/+ and WT mice, leading to improved bone mass and reduced long-bone fragility. Image-guided nanoindentation revealed no alteration in local tissue mineralization dynamics with Scl-Ab. These results contrast with previous findings of antiresorptive efficacy in OI both in mechanism and potency of effects on fragility. In conclusion, short-term Scl-Ab was successfully anabolic in osteoblasts harboring a typical OI-causing collagen mutation and represents a potential new therapy to improve bone mass and reduce fractures in pediatric OI.

Introduction

Osteogenesis imperfecta (OI), or “brittle bone disease”, is a heritable disorder caused predominantly by mutations in type I collagen or in proteins that promote the folding or post-translational modification of collagen. Patients with OI have increased bone fragility and are susceptible to fracture from minimal force, skeletal deformities and growth deficiency. OI fragility symptoms are generally most prominent in children; the disorder ranges in severity from mild forms with slightly elevated fracture risk to perinatal lethality ⁽¹⁾.

Current treatment options for OI focus on anti-resorptive bisphosphonates (BP) which have been shown effective at increasing vertebral areal bone mineral density and height in clinical trials. However, BP effects in long bones are less evident and most pediatric OI trials observe little or no functional benefit ^(2–6). Moreover, while BP therapy has been generally well tolerated in pediatric OI patients, there are concerns about long term retention of BPs in the skeleton and use in a growing population ⁽⁷⁾.

Sclerostin antibody (Scl-Ab) is a novel anabolic bone therapeutic presently in clinical trials for treatment of osteoporosis ⁽⁸⁾. Sclerostin is secreted primarily by osteocytes and negatively regulates bone formation by binding to the LRP4/5/6 complex, inhibiting anabolic canonical wnt signaling in osteoblasts ^(9–11). As a result, treatment with a neutralizing Scl-Ab reduces sclerostin inhibition of canonical wnt signaling and is highly anabolic in pre-clinical models and in a phase 1 clinical trial of men and postmenopausal women ^(8,12,13). Whether Scl-Ab therapy is capable of stimulating osteoblast activity in cells harboring a typical OI-causing mutation has yet to be demonstrated.

The Brtl/+ mouse is a knock-in model for moderately severe Type IV OI with a G349C point mutation on col1a1 that recreates an identical defect found in an OI patient ⁽¹⁴⁾. The Brtl/+ mouse recapitulates multiple features of the observed clinical phenotype, including smaller size, reduced BMD, more severe phenotype at young ages, increased bone brittleness and increased bone turnover ^(15,16), making it an appropriate model for testing the anabolic efficacy of Scl-Ab in OI. The purpose of this study was to determine whether a short-term intervention with Scl-Ab in the Brtl/+ model of OI would be effective at stimulating an anabolic skeletal response and improving long bone strength.

Materials and Methods

Animals

Wildtype (WT) and Brtl/+ mice are maintained on a mixed background of Sv129/CD-1/C57BL/6S, and all Brtl/+ animals were the product of breeding male Brtl/+ with female WT. 8 week old male WT and Brtl/+ mice were randomly assigned to Scl-Ab (Scl-AbIV, Amgen, Thousand Oaks, CA) treatment or vehicle injection (PBS) with n=7/group. Sclerostin antibody was injected subcutaneously at 25mg/kg, two times per week, for two weeks. Calcein (30mg/kg) was injected at the start of experiment, after 1 week, and 1 day prior to sacrifice by intraperitoneal injection to facilitate dynamic histomorphometry and nanoindentation placement. Body weights were recorded with each injection. Blood samples were collected at sacrifice by intracardiac puncture, serum separated by centrifuge, and stored at −80°C until analyzed by ELISA.

Left femurs were collected for microCT and mechanical testing, and right femurs for dynamic histomorphometry and nanoindentation tests. Both were stored at −20°C in lactated ringers solution (LRS) soaked gauze until testing or further specimen preparation. All protocols and procedures involving animals were approved by the University of Michigan’s Committee on Use and Care of Animals.

Serum Markers

To measure osteoblast activity, serum osteocalcin (OCN) was quantified with a commercially available ELISA kit (BT-470, BTI, Stoughton, MA). To quantify osteoclast number, serum TRACP5b was measured with a commercially available solid phase immunofixed enzyme activity assay (MouseTRAP, IDS, Fountain Hills, AZ). Both serum tests were performed in duplicate.

Micro Computed Tomography (µCT)

Left femora were scanned in water using cone beam computed tomography (eXplore Locus SP, GE Healthcare Pre-Clinical Imaging, London, ON, Canada). Scan parameters included a 0.5 degree increment angle, 4 frames averaged, an 80kVp and 80µA x-ray source with a 0.508mm Al filter to reduce beam hardening artifacts, and a beam flattener around the specimen holder ⁽¹⁷⁾. All images were reconstructed and calibrated at an 18µm isotropic voxel size to a manufacturer supplied phantom of air, water and hydroxyapatite. Regions of interest (ROI) were located for both cortical and trabecular parameters. A diaphyseal cortical ROI spanning 15% of total femur length was located midway between the distal growth plate and third trochanter. Cortical bone was isolated with a fixed threshold of 2000 Hounsfield Units for all experimental groups. Parameters including cortical thickness, cross sectional area, tissue mineral density (TMD), anterior-posterior bending moment of inertia, endosteal perimeter, and periosteal perimeter were quantified with commercially available software (MicroView v2.2 Advanced Bone Analysis Application, GE Healthcare Pre-Clinical Imaging, London, ON, Canada). A trabecular ROI 10% of total femur length was located immediately proximal to the distal femoral growth plate and defined along the inner cortical surface with a splining algorithm. Due to the large differences in morphology induced by Scl-Ab treatment, a fixed threshold could not be utilized without bias. Trabecular metaphyseal bone was isolated with a more conservative autothresholding algorithm for each specimen based on the bimodal distribution between marrow and bone ⁽¹⁸⁾. Parameters including bone volume fraction (BV/TV), trabecular thickness (TbTh), and trabecular number (TbN) were quantified using standard stereology algorithms (MicroView v2.2). A 3D sphere fitting algorithim was also used to confirm the stereology data for TbTh ⁽¹⁹⁾.

Mechanical Testing (Whole Bone) – Four-Point Bending

Following µCT scanning, left femora were loaded to failure in four-point bending using a servohydraulic testing machine (MTS 858 MiniBionix, Eden Prairie, MN). All specimens were kept hydrated in LRS-soaked gauze until mechanical testing. In the same mid-diaphyseal region analyzed by µCT, the mid-diaphysis was loaded in four point bending with the posterior surface oriented under tension. The distance between the wide, upper supports was 6.26mm, and span between the narrow, lower supports 2.085mm. The vertical displacement rate of the four-point bending apparatus in the anterior-posterior direction was 0.5mm/sec. Force was recorded by a 50lb load cell (Sensotec) and vertical displacement by an external linear variable differential transducer (LVDT, Lucas Schavitts, Hampton, VA), both at 2000Hz. A custom MATLAB script was used to calculate stiffness, yield load, yield displacement, ultimate load, failure displacement, post-yield displacement, and energy to failure. Combining anterior-posterior bending moment of inertia data from µCT with mechanical stiffness from four point bending, the estimated elastic modulus was calculated using standard beam theory as previously described ⁽¹⁵⁾.

Mechanical Testing (Tissue Level) – Fluorescent Guided Nanoindentation

Right femora were dehydrated, encased in epoxy (Kold Mount, Vernon-Benshoff, Albany, NY), and cut transversely at the mid-diaphysis with a low-speed saw (IsoMet, Beuhler, Lake Bluff, IL). The distal section of tissue was polished using progressive grades of silicon carbide abrasive paper (1200, 2400, and 4000 grit) under water irrigation for two minutes at each grade. To further decrease surface roughness the encased specimens were polished on a felt pad for 5 minutes with a ¼ µm diamond suspension (Struers Inc., Cleveland, OH). Specimens were then ultrasonically cleansed in a water bath for 10 minutes to remove surface debris, and glued to magnetic specimen plates for nanoindentation testing. The final root mean square (RMS) roughness of the specimens’ surface was 10.0 ± 3.7 nm, as assessed with Scanning Probe Microscopy of a 5×5 µm² region of the posterior mid-cortex.

A custom 950 TI TriboIndenter (Hysitron, Minneapolis, MN) instrumented with a fluorescent light-source and FITC filter allowed for simultaneous visualization of calcein labeling in specimens and accurate positioning of indents to locations matched for tissue age and treatment status with 0.5µm spatial resolution. Four regions of interest were mechanically tested in the posterior aspect of the femoral cross section: the mid-cortex (defined as midway between the first calcein label on the periosteal and endosteal surfaces, if any), along the first calcein label on 15 day old bone, along the second calcein label on 8 day old bone, and along the third and outer calcein label along 1 day old bone. Indentation consisted of loading a diamond Berkovich indenter tip into samples at 300µN/s, holding at a maximal load of 3,000 µN for 10 s, and unloading at 300µN/s. The indentation modulus E was calculated from the load-displacement curves using the standard Oliver-Pharr method ⁽²⁰⁾. Eight indents, 10 µm apart, were made along each calcein label and the mid cortex, with values averaged for each site and mouse.

Dynamic Histomorphometry

Using the same specimens tested by nanoindentation, dynamic histomorphometry was performed at the mid-diaphysis on the first and third labels according to standard nomenclature ⁽²¹⁾. Briefly, fluorescent images were acquired using a Zeiss Axiovert 200M inverted microscope equipped with Apotome imaging system to minimize out-of-plane light, negating the need for several micron thin sections. Fluorescent images were taken with a 10× objective of calcein (excitation 485/20nm, emission: 540/25nm) labels in bone. The 10× images of the cortex were merged into a single image (Photoshop, Adobe), and these merged images were analyzed using commercially available software (Bioquant Osteo v7.20.10, Nashville, TN). Bone surface (BS), mineral apposition rate (MAR), mineralizing surface to bone surface (MS/BS), and bone formation rate (BFR) were quantified.

Statistics

Based on previous studies examining the phenotype of Brtl/+ at eight weeks of age⁽¹⁵⁾ and on the therapeutic effects of bisphosphonate in Brtl/+⁽¹⁶⁾, a power analysis suggests that samples sizes of seven per genotype per group would sufficiently detect a therapeutic response equivalent to that of alendronate after 12 weeks of therapy. A multivariate ANOVA with LSD post-hoc was used to make all comparisons with the exception of nanoindentation data. For nanoindentation data along calcein labels, a repeated measures ANOVA was used with genotype and treatment status as between-subjects factors and tissue age as the within-subjects factor with simple contrasts to mid-cortical reference values. In all cases, p<0.05 was considered significant. All data is presented as mean±S.D.

Results

Body Weight and Bone Length Remain Unchanged with Short-Term Sclerostin Antibody Treatment

Consistent with the reduced body size with OI, body weight was significantly lower in male Brtl/+ (25.4±1.8g) compared to male WT mice (30.4±2.8g) at 8 weeks of age. Twice weekly dosing of these mice with 25 mg/kg Scl-Ab for 2 weeks did not significantly change body weight gains (data not shown). Scl-Ab treatment also had no effect on femoral length in WT mice (15.9±0.6mm Veh vs. 16.0±0.5mm Scl-Ab) or the shorter Brtl/+ mice (15.2±0.4 Veh mm vs. 15.5±0.4mm Scl-Ab) after 2 weeks.

Serum Markers Show Brtl/+ Anabolic Response to Scl-Ab

Untreated Brtl/+ animals showed marginally elevated serum osteocalcin (p=0.060) and serum TRACP5b (p=0.083) relative to WT (Fig 1), consistent with reported increases in Brtl/+ bone turnover ⁽¹⁶⁾. Scl-Ab therapy significantly increased serum osteocalcin in both WT and Brtl/+ (Fig 1), demonstrating a rapid systemic anabolic response to therapy. Scl-Ab yielded trends toward reduced levels of serum TRACP5 in both WT (p=0.10) and Brtl/+ (p=0.082), consistent with published reports observing reduced serum markers of bone resorption in rats and humans ^(8,12).

Figure 1.

Serum OCN (A) is elevated and serum TRACP5b (B) is reduced with Scl-Ab therapy in both WT and Brtl/+, suggesting a decoupled effect on bone formation and bone resorption. In Vehicle treated animals, Brtl/+ trends toward higher OCN (A) and TRACP5b (B) levels suggesting increased bone turnover in Brtl/+. * p<0.05 Scl-Ab vs. Veh, + p<0.05 Brtl/+ Scl-Ab vs WT Veh.

Scl-Ab Differentially Increases Cortical Bone Formation in Brtl/+ and WT

At the femoral mid-shaft, Scl-Ab treatment increased periosteal BFR/BS in Brtl/+ and WT animals by 76% and 108% respectively (Fig. 2). In Brtl/+, these increases were caused by a 50% increase in MS/BS, with no significant improvement in MAR. Conversely, WT improvements in BFR/BS resulted from a significant 76% increase in MAR with no significant change in MS/BS.

Figure 2.

Dynamic histomorphometry at the periosteal femoral mid-diaphysis reveals anabolic effect of Scl-Ab in WT and Brtl/+. Femoral cortical mid-diaphyseal periosteal BFR (A) increases with Scl-Ab are a result of increased periosteal MAR (B) in WT and periosteal MS/BS (C) in Brtl/+. Brtl/+ Veh n=6, all other groups n=7. * p<0.05 Scl-Ab vs. Veh, + p<0.05 Brtl/+ Scl-Ab vs WT Veh.

Scl-Ab Increases Trabecular Bone Apposition on Existing Bone Surfaces

Brtl/+ animals have lower BV/TV relative to WT caused by reduced trabecular number, but not thickness (Fig. 3). Scl-Ab significantly increased distal femoral metaphyseal BV/TV in WT animals with a trend in Brtl/+ (p=0.077). These improvements resulted from significant increases in trabecular thickness in both WT and Brtl/+, with no change in trabecular number in either genotype.

Figure 3.

Micro computed tomography of the distal femur metaphysis reveals anabolic changes with Scl-Ab therapy. BV/TV (A), and Tb.Th (B) were improved by Scl-Ab with no effect on Tb.N (C). D–G: Representative microCT images of distal metaphyseal trabecular bone reflect these changes. For each experimental group, the animal with the median BV/TV value is shown. A conservative specimen specific threshold is applied as described in the methods. ^*p<0.05 Scl-Ab vs. Veh; ^#p<0.05 WT Veh vs. Brtl/+ Veh, + p<0.05 Brtl/+ Scl-Ab vs WT Veh.

Cortical Shape and Size, Not Mineralization, are Increased with Sclerostin Antibody

At the femoral mid-diaphysis, Brtl/+ animals have a reduced cortical thickness and area relative to WT as measured by microCT (Table 1). Scl-Ab treatment significantly increased both cortical thickness and area in Brtl/+ and WT, primarily through increased femoral cortical periosteal perimeter. Moreover, Scl-Ab treatment of Brtl/+ restored cortical thickness and area to levels not significantly different from WT Veh. Endocortical perimeter showed no significant differences with Scl-Ab treatment in either WT or Brtl/+. In combination, these individual factors contributed to increasing trends in the anterior-posterior bending moment of inertia with Scl-Ab treatment in both WT and Brtl/+, reflecting a structurally stronger bone. Mean cortical tissue mineralization (TMD) calculated by microCT was unaffected by Scl-Ab in either WT or Brtl/+.

Table 1. Cortical MicroCT and Mechanical Properties.

Femoral mid-diaphyseal cortical microCT and mechanical four point bending data.

	WT Veh		WT Scl-Ab		Brtl Veh		Brtl Scl-Ab
Cortical Micro CT
Thickness (mm)	0.21	± .01	0.24	± .01*	0.17	± .02#	0.20	± .02*
Cross Sectional Area (mm^2)	1.01	± .10	1.16	± .10*	0.76	± .10#	0.91	± .10*
Endosteal Perimeter (mm)	4.51	± .36	4.43	± .22	4.09	± .13#	4.21	± .18 +
Periosteal Perimeter (mm)	5.80	± .42	5.95	± .32	5.15	± .16#	5.48	± .21*, p=0.06
Bending Moment of Inertia (mm^4)	0.21	± .05	0.23	± .04	0.12	± .03#	0.16	± .03p=.08, p=0.06
Tissue Mineral Density (mg/cm^3)	1052	± 43	1051	± 50	1081	± 39	1092	± 41 n.s., p=0.1
Mechanical Four Point Bending
Ultimate Load (N)	27.1	± 4.5	38.7	± 10.2*	18.2	± 5.0#	26.8	± 7.8*
Stiffness (N/mm)	237	± 28	297	± 66*	192	± 35p=0.063	256	± 33*
Energy to Failure (J)	3.1	± 1.6	10.6	± 7.6*	1.0	± .6	2.0	± 1.1
Post Yield Displacement (mm)	0.058	± .049	0.247	± .143*	0.026	± .021	0.041	± .037
Estimated Elastic Modulus (GPa)	4.6	± 1.2	4.8	± .5	6.0	± 1.2#	6.0	± .7 +

^*p<0.05 Scl-Ab vs. Veh;

^#p<0.05 WT Veh vs. Brtl/+ Veh,

⁺p<0.05 Brtl/+ Scl-Ab vs. WT Veh.

Sclerostin Antibody Improves Whole Bone Mechanical Properties

Brtl/+ femora have inherently lower ultimate load and stiffness (p=0.063) compared to WT (Table 1). Two weeks of Scl-Ab treatment significantly increased ultimate load and stiffness in both WT and Brtl/+ mice. Importantly, Scl-Ab treatment of Brtl/+ improved functional outcomes including ultimate load and stiffness to levels not significantly different from WT Veh, suggesting that the phenotypic deficit was rescued by therapy. The estimated tissue elastic modulus, as calculated by standard beam theory, was not affected by Scl-Ab treatment in WT or Brtl/+, corroborating similar findings in TMD.

A hallmark feature of OI bone is increased tissue brittleness. Post-yield displacement (PYD) during four point bending is a measure of ductility, an inverse indicator of material brittleness. Phenotypically, although Brtl/+ PYD was not significantly different than WT, five out of seven WT samples had a PYD greater than the largest Brtl/+ PYD value, consistent with previous reports increased brittleness at a similar age ⁽¹⁵⁾. Unexpectedly, post yield displacement was increased 324% with Scl-Ab in WT (p<0.05), and 58% with Scl-Ab in Brtl/+, although failing to achieve statistical significance due to high variability. These data suggest Scl-Ab reduced bone brittleness, improving bone ductility in both Brtl/+ and WT.

Nanoindentation Reveals Similar Patterns of Mineralization in Scl-Ab Treated Animals

Nanoindentation measures local micromechanical properties associated with local tissue mineralization by measuring the tissue elastic modulus. Phenotypically, in the overall statistical model, genotype was a significant predictor suggesting a greater tissue elastic modulus in Brtl/+ compared to WT. However, comparisons between genotypes at specific tissue ages were not significant. Nanoindentation revealed that Scl-Ab treatment did not change the tissue elastic modulus of bone in either WT or Brtl/+. This indicates that despite the sudden change in osteoblastic anabolic function with Scl-Ab treatment, no differences in tissue elastic modulus were seen in bone grown under the influence of Scl-Ab (Fig. 4). The tissue age at which newly formed bone achieved mature tissue elastic modulus was rapid, as only bone laid down 1 day prior to euthanasia had a significantly reduced elastic modulus relative to mid-cortical values consistent with the rapid establishment of mechanical properties which has been reported in rats ^(22,23). No difference was found between 2 week old bone and 1 week old bone relative to mid-cortical values.

Figure 4.

Nanoindentation was performed along calcein labels to match tissue age between treated and untreated animals. White Light, FITC, and merged images showing placement of indents (A). Nanoindentation revealed no difference in tissue elastic modulus across treatment groups in WT (B) and Brtl/+ (C). *notes significance relative to mid-cortical reference and denotes p<0.05.

Discussion

In this study, mice with a genetic knock-in that recreates a typical OI-causing mutation in type I collagen showed a positive response following short-term Scl-Ab therapy. Two weeks of Scl-Ab significantly improved bone mass at both cortical and trabecular sites, and significantly reduced long bone fragility. In the context of existing anabolic and anti-resorptive treatment options, Scl-Ab is novel and has high potential as a therapeutic for OI patients. Current anabolic treatment options have not shown uniform success when applied to OI. The anabolic bone agent teraparatide is currently approved as an osteoporosis therapy, but has been associated with osteosarcoma after long term treatment in growing rats ⁽²⁴⁾, and is therefore contraindicated for use in children with open growth plates. Growth hormone has also been implemented as an anabolic therapy for OI. While effective in some patients at increasing BMD as well as improving bone histomorphometry, the benefits are selective for patients who respond to rGH with increased linear growth. Other OI patients treated with rGH experience no positive effect, and may exacerbate their existing low bone mass phenotype by increasing osteoclast surface ⁽²⁵⁾.

As a result of limited anabolic therapeutics, anti-resorptive bisphosphonates have been widely used in OI. Several controlled pediatric OI clinical trials have demonstrated that bisphosphonate treatment is effective at increasing vertebral BMD ^(2–6). Functionally, intravenous bisphosphonates are effective at increasing vertebral height, an indirect indicator of bone strength that suggests increased resistance to compression ^(3,4). However, in long bones a functional benefit is less clear, even in controlled trials up to 125 patients ^(1–6).

While bisphosphonate therapy increases areal BMD in pediatric OI populations, recent evidence suggests potential for several undesirable effects intrinsic to the BP mechanism of action. First, BP binds strongly to the skeleton and has been detected in urinary markers 8 years after cessation of treatment, resulting in long term suppression of bone turnover ⁽²⁶⁾. Long-term BP therapy has been associated with the accumulation of microdamage in several animal models ^(27–31). Brtl/+ mice have an increased propensity to form and accumulate microdamage ⁽³²⁾ and thus BP treatment may potentially exacerbate this effect. During growth, antiresorptive therapy results in retention of calcified cartilage near the growth plates. This retention of primary spongiosa manifests radiographically as sclerotic metaphyseal banding coinciding with each BP treatment cycle, and these changes can bias measured BMD gains ⁽³³⁾ without concomitant improvements in strength. Previous studies of BP treatment in Brtl/+ found increased trabecular number, but not thickness ⁽³³⁾. Notably, in the present study with Scl-Ab, gains in trabecular bone mass resulted from increased trabecular thickness, not number.

In this study, we found that unlike WT osteoblasts, Brtl/+ osteoblasts did not increase their mineral apposition rate with Scl-Ab treatment. This lack of an MAR increase in Brtl/+ may be suggestive of defective osteoblasts unable to keep pace with Scl-Ab-induced bone formation demands. In support of this, Brtl/+ mice demonstrate a mild delay in secretion of collagen in cell culture, and an enlarged ER has been observed in Brtl/+ fibroblasts, suggesting ER-stress correlated with collagen production, and possibly degradation of mutant collagen ⁽³⁴⁾. As a consequence, Brtl/+ may respond to short-term Scl-Ab by increasing mineralizing surface, and thus osteoblast recruitment, rather than individual cellular activity. This increase in mineralizing surface still leads to a significant increase in overall bone formation rate, resulting in the improved bone size and strength observed. Whether this result also exists in the various OI mouse models with different mutations will be important to contrast with the data presented here. Alternatively, animal age and treatment duration of the Brtl/+ mouse model used in this study may also modulate any Scl-Ab induced MAR effect. Preliminary data from Brtl/+ mice treated for 5 weeks with Scl-Ab suggest an similar increase in MAR in both WT and Brtl/+ animals ⁽³⁵⁾. While a significant increase in periosteal perimeter was found in Brtl/+ when treated with sclerostin antibody, no significant differences were observed on the endosteal surface. This contrasts with previous findings of female rats treated with sclerostin antibody following ovariectomy, where a dominant treatment effect was observed on the endosteal bone surface ⁽¹²⁾. This may suggest a differential response of periosteal vs. endosteal surface to sclerostin antibody dependent on animal age or hormonal status, reflecting either altered cell response to drug or differential bioavailability at these sites.

Consistent with previous studies of linear growth with Scl-Ab or in male SOST knockout mice ^(36,37), we found no differences in either femur length or body mass, suggesting that longitudinal growth modulation is less sensitive than periosteal growth after two weeks of therapy. Unexpectedly, we found that Scl-Ab significantly reduced bone brittleness (increased post-yield displacement) in WT animals as measured by whole bone mechanical four-point bending. Although reductions in bone brittleness were not statistically significant in Brtl/+ animals in this study, our finding of reduced brittleness in WT has broader implications about bone quality changes associated with Scl-Ab treatment. While the differing bone geometries induced by Scl-Ab treatment raises the potential for experimental bias, the large magnitude of our observed difference in WT (324% increase in post yield displacement) remains suggestive. In contrast to the Scl-Ab data presented here, previous bisphosphonate treatment in Brtl/+ found neither significant changes nor increasing trends in PYD. Moreover, BP-treated WT showed marginal reductions in PYD reflecting no improvements in bone brittleness with bisphosphonates despite large gains in femoral bone mass ⁽³³⁾. Interestingly, in the contralateral femurs of the same animals which demonstrated reductions in post yield displacement, we did not observe any difference in the elastic modulus of bone formed under the influence of Scl-Ab by fluorescence guided nanoindentation. These findings represent the first tissue level mechanical test of bone formed during Scl-Ab treatment through coupling of fluorescent imaging to nanoindentation. Further studies are required to determine post-yield behavior of bone at the tissue level, unlike those reported in this study.

The primary goals of this study were to demonstrate the potential for Scl-Ab efficacy at inducing an osteoblast response in Brtl/+ at an age that has been well characterized (8 weeks) ^(15,16). While clinical efficacy will likely require an earlier onset of therapy and longer duration, these findings of significantly improved cortical bone mass, stiffness, and breaking loads with such a short duration of therapy are encouraging, particularly given the mixed response of bisphosphonate efficacy at cortical sites ^(2–6). Thus, these findings support further exploration of Scl-Ab at earlier ages and for longer duration in treatment of OI.

In this study, Scl-Ab rapidly increased bone formation resulting in increased cortical and cancellous bone volume and cortical bone strength without altering bone mineralization kinetics. While Brtl/+ and WT both responded favorably to Scl-Ab, it is unknown if there are differences in sclerostin expression in OI patients or OI mouse models throughout growth which could modulate the therapeutic effect. More broadly, as sclerostin mediates the Wnt signaling pathway, any differences existing in Brtl/+, including the receptors LRP4/5/6, competing ligands including DKK1, or regulation of b-catenin signaling may determine the skeletal Scl-Ab response.

Furthermore, it should be recognized that some Scl-Ab studies have found a decrease in bone resorption outcomes ^(8,12). Given that an upregulation in bone resorption is phenotypic of OI along with the trends of reduced serum TRACP5b data with Scl-Ab from this study, the osteoclast response will be explored in greater detail in future studies with longer term therapy. Nevertheless, the trabecular microCT data in this study shows a large increase in trabecular thickness with no change in trabecular number and is suggestive of a dominantly osteoblast mediated effect.

In summary, we have demonstrated that two weeks of Scl-Ab therapy was capable of increasing bone formation rate in cells harboring a classical OI-causing Gly->Cys substitution in col1a1. These gains lead to improved bone mass and whole bone mechanical properties without altering the timing of tissue mineralization. While bisphosphonate therapy is in widespread use for pediatric OI patients, there is pressing need for more effective anabolic therapeutics to improve patient outcomes. Scl-Ab is a unique and promising anabolic therapy which may be particularly useful for the treatment of pediatric OI. The present data suggest Scl-Ab may be beneficial for the treatment of OI patients by stimulating the osteoblasts and reducing fracture risk.