An Antibody to Notch2 Reverses the Osteopenic Phenotype of Hajdu-Cheney Mutant Male Mice

Abstract

Notch receptors play a central role in skeletal development and bone remodeling. Hajdu-Cheney syndrome (HCS), a disease characterized by osteoporosis and fractures, is associated with gain-of-NOTCH2 function mutations. To study HCS, we created a mouse model harboring a point 6955C>T mutation in the Notch2 locus upstream of the proline, glutamic acid, serine, and threonine domain, leading to a Q2319X change at the amino acid level. Notch2^Q2319X heterozygous mutants exhibited cancellous and cortical bone osteopenia. Microcomputed tomography demonstrated that the cancellous and cortical osteopenic phenotype was reversed by the administration of antibodies generated against the negative regulatory region (NRR) of Notch2, previously shown to neutralize Notch2 activity. Bone histomorphometry revealed that anti-Notch2 NRR antibodies decreased the osteoclast number and eroded surface in cancellous bone of Notch2^Q2319X mice. An increase in osteoclasts on the endocortical surface of Notch2^Q2319X mice was not observed in the presence of anti-Notch2 NRR antibodies. The anti-Notch2 NRR antibody decreased the induction of Notch target genes and Tnfsf11 messenger RNA levels in bone extracts and osteoblasts from Notch2^Q2319X mice. In vitro experiments demonstrated increased osteoclastogenesis in Notch2^Q2319X mutants in response to macrophage colony-stimulating factor and receptor activator of nuclear factor–κB ligand, and these effects were suppressed by the anti-Notch2 NRR. In conclusion, Notch2^Q2319X mice exhibit cancellous and cortical bone osteopenia that can be corrected by the administration of anti-Notch2 NRR antibodies.

A mouse model of Hajdu Cheney syndrome harboring a Notch2 gain-of-function mutation exhibits osteopenia that can be reversed by administering antibodies directed to the Notch2-NRR region to prevent Notch2 activation.

Notch receptors are single-pass transmembrane receptors that play a role in skeletal development and homeostasis and in osteoblast and osteoclast differentiation (1–4). Activation of Notch follows its interactions with ligands of the Jagged and Delta-like families, resulting in the proteolytic cleavage of Notch and the release of the Notch intracellular domain (NICD) (3). Translocation of the NICD into the nucleus induces the transcription of target genes, such as those encoding Hairy/Enhancer of Split (Hes)1, 5, 7 and Hes-related with YRPW motif (Hey)1, 2, and L (5–8). Activation of Notch1 in cells of the osteoblast lineage inhibits their maturation and causes osteopenia, whereas activation of Notch1 in osteocytes induces osteoprotegerin and suppresses bone resorption, causing osteopetrosis (9–11). Additional studies have demonstrated that Notch1 inhibits bone resorption; however, this is not the case for Notch2, which by inducing nuclear factor of activated T-cells cytoplasmic 1 (Nfatc1) enhances osteoclastogenesis (12, 13).

Hajdu-Cheney syndrome (HCS) is a dominant inherited disease characterized by craniofacial developmental abnormalities, acroosteolysis, platybasia, and generalized osteoporosis with fractures (14, 15). HCS is associated with point mutations in or short deletions of exon 34 of NOTCH2 that lead to the creation of a stop codon upstream of the proline (P), glutamic acid (E), serine (S), and threonine (T) (PEST) domain (16–20). Because the PEST domain is necessary for the ubiquitination and degradation of NOTCH2, the mutations result in the translation of a stable protein product and presumably gain-of-NOTCH2 function. Despite the pronounced skeletal abnormalities and neurologic complications reported in HCS, little is known about the mechanisms underlying the bone loss and ways to correct them, and results from the analysis of iliac crest bone biopsies have been inconclusive (21–24).

To gain understanding of the HCS skeletal phenotype and the mechanisms involved, we used homologous recombination to create a mouse model of the disease. A Notch2 mutation (6955C>T) in the mouse, corresponding to a 6949C>T mutation in NOTCH2 found in a subject affected by the disease, was introduced by homologous recombination into the murine genome (17, 19, 25). The 6955C>T Notch2 mutation creates a stop codon in exon 34 leading to the translation of a truncated Notch2 protein of 2318 amino acids (Notch2^Q2319X) (25). Following the validation and sequencing of the Notch2^Q2319X mouse model of HCS, it was demonstrated that mutant mice exhibit cancellous and cortical bone osteopenia due to enhanced bone resorption and osteoclast differentiation.

The purpose of the present work was to determine whether the phenotype of the HCS mouse model could be reversed by pharmacological intervention. To this end, Notch2^Q2319X mice were treated with a specific and well-characterized antibody directed to the negative regulatory region (NRR) of Notch2 (26). The NRR consists of three LIN-12 Notch repeats and a heterodimerization domain and contains the initial cleavage sites (S1 and S2) of Notch, which are required for protein maturation and signal activation (3, 4). The epitope for the anti-Notch2 NRR bridges the LIN-12 Notch repeat and heterodimerization domain so that the antibody locks the receptor in its quiescent state, preventing Notch activation (26, 27). To determine the effect of the anti-Notch2 NRR antibody, Notch2^Q2319X and control mice were treated and characterized by bone microarchitectural and histomorphometric analyses. The effects of the antibody in osteoblast and osteoclast cultures also were tested.

Materials and Methods

Hajdu-Cheney mutant mice

To create a mouse model of HCS, a 6955C>T substitution was introduced into the mouse Notch2 locus by homologous recombination, as previously reported (25). Following the removal of the neomycin selection cassette, the Notch2 mutation was confirmed by sequencing of genomic DNA (GENEWIZ, South Plainfield, NJ) from F1 pups, and mice were backcrossed into a C57BL/6J background for eight or more generations. Genotyping of Notch2^Q2319X mice was conducted in tail DNA extracts by polymerase chain reaction (PCR) using forward primer Nch2Lox gtF 5′–CCCTTCTCTCTGTGCGGTAG-3′ and reverse primer Nch2Lox gtR 5′–CTCAGAGCCAAAGCCTCACTG-3′.

In this study, 1-month-old mutant mice heterozygous for the Notch2^Q2319X mutation and control mice were obtained by crossing heterozygous mutants with wild-type mice. One-month-old male Notch2^Q2319X mutant and control sex-matched littermate mice were treated with anti-Notch2 NRR or anti-ragweed antibody (Genentech, South San Francisco, CA) suspended in phosphate-buffered saline and administered intraperitoneally at a dose of 10 mg/kg twice a week for 4 consecutive weeks and were euthanized at 2 months of age (Table 1). The dose of antibody was based on available data demonstrating the effects of anti-Notch2 NRR antibody in vivo at doses between 2 and 30 mg/kg, information on the lack of antibody-dependent gastrointestinal toxicity when used at doses of 5 mg/kg, and its predicted half-life based on the known half-life of immunoglobulin G (26, 28–31). Studies were approved by the Institutional Animal Care and Use Committee of UConn Health.

Table 1.

Antibodies Used in This Study

Peptide/Protein Target	Antigen Sequence	Name of Antibody	Manufacturer and Individual Providing the Antibody	Species Raised in; Monoclonal or Polyclonal	Dilution Used	RRID
Notch2 NRR	PATCQSQYCADKARDGICDEACNSHACQWDGGDCSLTMEDPWANCTSTLRCWEYINNQCDEQCNTAECLFDNFECQRNSKTCKYDKYCADHFKDNHCDQGCNSEECGWDGLDCASDQPENLAEGTLIIVVLLPPEQLLQDSRSFLRALGTLLHTNLRIKQDSQGALMVYPYFGEKSAAMKKQKMTRRSLPEEQEQEQEVIGSKIFLEIDNRQCVQDSDQCFKNTDAAAALLASHAIQGTLSYPLVSVFSELESPRNAQ	Anti-Notch2 NRR	Genentech, Christian W. Siebel	Synthetic antibody generated by phage display	In vivo 10 mg/kg; in vitro 10 μg/mL	Not listed

Abbreviation: RRID, Research Resource Identifier.

Microcomputed tomography

Bone microarchitecture of femurs from experimental and control mice was determined using a microcomputed tomography instrument (µCT 40; Scanco Medical AG, Bassersdorf, Switzerland), which was calibrated periodically using a phantom provided by the manufacturer (32, 33). Femurs were scanned in 70% ethanol at high resolution, with an energy level of 55 kVp, an intensity of 145 µA, and an integration time of 200 ms. A total of 100 slices at midshaft and 160 slices at the distal metaphysis were acquired at an isotropic voxel size of 216 µm³ and a slice thickness of 6 µm and were chosen for analysis. Trabecular bone volume fraction and microarchitecture were evaluated starting approximately 1.0 mm proximal from the femoral condyles. Contours were manually drawn a few voxels away from the endocortical boundary every 10 slices to define the region of interest for analysis. The remaining slice contours were iterated automatically. Trabecular regions were assessed for total volume, bone volume, bone volume fraction (bone volume/total volume), trabecular thickness, trabecular number, trabecular separation, connectivity density, and structure model index (SMI) using a Gaussian filter (σ = 0.8) and user-defined thresholds (32, 33). For analysis of femoral cortical bone, contours were iterated across 100 slices along the cortical shell of the femoral midshaft, excluding the marrow cavity. Analyses of bone volume/total volume, porosity, cortical thickness, total cross-sectional and cortical bone area, periosteal perimeter, endosteal perimeter, and material density were performed using a Gaussian filter (σ = 0.8, support = 1) and user-defined thresholds.

Bone histomorphometric analysis

Static and dynamic cancellous bone histomorphometry was carried out on experimental and control mice after they were injected with calcein, 20 mg/kg, and demeclocycline, 50 mg/kg, both intraperitoneally, at an interval of 5 days; they were euthanized 2 days after the demeclocycline injections. Five-micron longitudinal sections of undecalcified femurs embedded in methyl methacrylate were cut on a microtome (Microm; Richard-Allan Scientific, Kalamazoo, MI) and were stained with 0.1% toluidine blue. Static parameters of bone formation and resorption were measured in a defined area between 360 and 2160 µm from the growth plate using an OsteoMeasure morphometry system (OsteoMetrics, Atlanta, GA). For dynamic histomorphometry, mineralizing surface per bone surface and mineral apposition rate were measured on unstained sections under UV light using a triple diamidino-2-phenylindole/fluorescein/Texas red set long-pass filter, and bone formation rate was calculated.

For cortical histomorphometry, femurs were embedded in methyl methacrylate and were cut through the mid-diaphysis with an EXAKT Precision Saw. Slides were ground to a thickness of approximately 15 μm using an EXAKT 400 CS Micro Grinding System and were surface polished (Alizee Pathology, Baltimore, MD). Slides were stained with hematoxylin and eosin and were analyzed at a magnification of ×400 using OsteoMeasure XP software. Stained sections were used to draw the cortical bone, marrow space, osteoid, and cell surfaces as well as to count osteocytes within the cortex and osteoblasts and osteoclasts along the endocortical surface. The terminology and units used for cancellous and cortical bone are those recommended by the Histomorphometry Nomenclature Committee of the American Society for Bone and Mineral Research (34, 35).

Osteoblast-enriched cell cultures

The parietal bones of 3- to 5-day-old control and Notch2^Q2319X mutant mice were exposed to Liberase TL 1.2 U/mL (Sigma-Aldrich, St. Louis, MO) for 20 minutes at 37°C, and cells were extracted in five consecutive reactions (36). Cells from the last three digestions were pooled and seeded at a density of 10 × 10⁴ cells/cm², as described (37). Osteoblast-enriched cells were cultured in Dulbecco’s modified Eagle medium supplemented with nonessential amino acids (both from Life Technologies, Grand Island, NY) and 10% heat-inactivated fetal bovine serum (FBS; Atlanta Biologicals, Norcross, GA) in a humidified 5% CO₂ incubator at 37°C in the presence of anti-Notch2 NRR or control antibody at 10 μg/mL, a dose known to inhibit Notch2 signaling (26). Confluent osteoblast-enriched cells were exposed to Dulbecco’s modified Eagle medium supplemented with 10% heat-inactivated FBS, 100 µg/mL of ascorbic acid, and 5 mM of β-glycerophosphate (both from Sigma-Aldrich).

Bone marrow–derived macrophage cultures, osteoclast formation, and bone resorption

To obtain bone marrow–derived machophages (BMMs), bone marrow cells were isolated from long bones by flushing the marrow with a 26-gauge needle. Red blood cells were lysed in lysis buffer containing 150 mM of NH₄Cl, 10 mM of KHCO₃, and 0.1 mM of EDTA (pH 7.4). The cell suspension was centrifuged, and the pellet was suspended in α-minimum essential medium (Life Technologies) containing 10% heat-inactivated FBS (Thermo Fisher Scientific, Waltham, MA) and recombinant human macrophage colony-stimulating factor (M-CSF) at 30 ng/mL. M-CSF complementary DNA (cDNA) and expression vector were obtained from D. Fremont (St. Louis, MO), and M-CSF was purified as previously reported (38). Cells were seeded at a density of 3 × 10⁵ cells/cm² on uncoated Petri dishes and were cultured for 3 to 4 days.

For osteoclast formation, cells were collected following treatment with 0.05% trypsin/EDTA for 5 minutes and were seeded at a density of 4.7 × 10⁴ cells/cm² on tissue culture plates in the presence of M-CSF at 30 ng/mL and murine receptor activator of nuclear factor (Nf)-κB ligand (Rankl) at 10 ng/mL until the formation of multinucleated tartrate-resistant acid phosphatase (TRAP)–positive cells. Rankl cDNA and expression vector were obtained from M. Glogauer (Toronto, ON, Canada), and glutathione S-transferase–tagged Rankl was expressed and purified as described (39). Cultures were carried out in the presence of anti-Notch2 NRR at 10 μg/mL or anti-ragweed antibody at 20 μg/mL; TRAP enzyme histochemistry was conducted using a commercial kit (Sigma Aldrich) in accordance with the manufacturer’s instructions. TRAP-positive cells containing more than three nuclei were considered osteoclastlike cells.

For in vitro bone resorption, BMMs were collected and seeded at a density of 4.7 × 10⁴ cells/cm² on bovine cortical bone slices and were cultured in α-minimum essential medium with 10% FBS, M-CSF at 30 ng/mL, and Rankl at 10 ng/mL in the absence or presence of anti-Notch2 NRR or anti-ragweed antibody at 10 μg/mL (26). To visualize resorption pits, bone slices were sonicated to remove osteoclasts and were stained with 1% toluidine blue in 1% sodium borate. To evaluate the ability of osteoclasts to resorb bone, the resorption area/total bone area was measured on images acquired with an Olympus DP72 camera using cellSens Dimension software v1.6 (Olympus Corporation, Center Valley, PA). Resorption area/total bone area was corrected for the total number of TRAP-positive multinucleated cells (40).

Quantitative reverse transcription

Total RNA was extracted from cultured cells, tibiae, and femurs following the removal of the bone marrow by centrifugation, and messenger RNA levels were determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) (41, 42). For this purpose, equal amounts of RNA were reverse-transcribed using the iScript RT-PCR kit (Bio-Rad, Hercules, CA) according to manufacturer’s instructions and were amplified in the presence of specific primers (Table 2) (all primers were from Integrated DNA Technologies (Coralville, IA) and iQ SYBR Green Supermix (Bio-Rad) at 60°C for 35 cycles. Transcript copy number was estimated by comparison with a serial dilution of cDNA for acid phosphatase 5, tartrate-resistant (Acp5; from Thermo Fisher Scientific), Hes1 (from American Type Culture Collection (Manassas, VA), Hey1 and Hey2 (both from T. Iso), HeyL (from D. Srivastava, Dallas, TX), nuclear factor of activated T cells 1 (Nfatc1; from A. Rao, La Jolla, CA), or tumor necrosis factor superfamily member 11 (Tnfsf11, encoding for Rankl; from Source BioScience, Nottingham, UK) (43–47).

Table 2.

Primers Used for qRT-PCR Determinations

Gene	Strand	Sequence 5′–3′	GenBank Accession Number
Acp5	Forward	GACAAGAGGTTCCAGGAGAC	NM_001102404;
			NM_001102405;
	Reverse	TTCCAGCCAGCACATACC	NM_007388
Hes1	Forward	ACCAAAGACGGCCTCTGAGCACAGAAAGT	NM_008235
	Reverse	ATTCTTGCCCTTCGCCTCTT
Hey1	Forward	ATCTCAACAACTACGCATCCCAGC	NM_010423
	Reverse	GTGTGGGTGATGTCCGAAGG
Hey2	Forward	AGCGAGAACAATTACCCTGGGCAC	NM_013904
	Reverse	GGTAGTTGTCGGTGAATTGGACCT
HeyL	Forward	CAGTAGCCTTTCTGAATTGCGAC	NM_013905
	Reverse	AGCTTGGAGGAGCCCTGTTTC
Nfatc1	Forward	GCGCAAGTACAGTCTCAATGGCC	NM_198429;
	Reverse	GGATGGTGTGGGTGAGTGGT	NM_001164110;
			NM_001164111;
			NM_001164112;
			NM_00116641091;
			NM_016791
Notch2	Forward	CATCGTGACTTTCCA	NM_010928
	Reverse	GGATCTGGTACATAGAG
Rpl38	Forward	AGAACAAGGATAATGTGAAGTTCAAGGTTC	NM_001048057;
	Reverse	CTGCTTCAGCTTCTCTGCCTTT	NM_001048058;
			NM_023372
Tnfsf11	Forward	TATAGAATCCTGAGACTCCATGAAAAC	NM_011613
	Reverse	CCCTGAAAGGCTTGTTTCATCC

GenBank accession numbers identify transcript recognized by primer pairs.

To measure levels of the Notch2^6955C>T mutant transcript, total RNA was reverse transcribed with Moloney murine leukemia virus reverse transcriptase in accordance with manufacturer’s instructions (Life Technologies) in the presence of reverse primers for Notch2 and reverse primers for ribosomal protein L38 (Rpl38) (Table 2). Notch2 cDNA was amplified by PCR in the presence of specific primers (Table 2), a tetrachloro fluorescein–labeled DNA probe of sequence 5′-CATTGCCTAGGCAGC-3′ covalently bound to a 3′-minor groove binder quencher (Life Technologies), and SsoAdvanced Universal Probes Supermix (Bio-Rad) at 60°C for 45 cycles (48). Notch2^6955C>T transcript copy number was estimated by comparison with a serial dilution of a synthetic DNA fragment (Integrated DNA Technologies) containing ∼200 bp surrounding the 6955C>T mutation in the Notch2 locus and was cloned into pcDNA3.1 (Life Technologies) by isothermal single-reaction assembly using commercially available reagents (New England Biolabs, Ipswich, MA) (49).

Amplification reactions were conducted in a CFX96 qRT-PCR detection system (Bio-Rad), and fluorescence was monitored during every PCR cycle at the annealing step. Data are expressed as copy number corrected for Rpl38 copy number, estimated by comparison with a serial dilution of Rpl38 cDNA (from American Type Culture Collection) (50).

Statistics

Data are expressed as mean ± standard error of the mean (SEM). Statistical differences were determined by analysis of variance with the Holm-Sidak or Tukey post hoc analysis for pairwise or multiple comparisons.

Results

General characteristics, femoral microarchitecture, and histomorphometry of Notch2^Q2319X mutant mice

In the present studies, heterozygous Notch2^Q2319X mutant mice were compared with wild-type mice because the homozygosity of Notch2^Q2319X mutant mice results in perinatal lethality and heterozygous Notch2^Q2319X mice exhibit an osteopenic phenotype (25). Heterozygous Notch2^Q2319X mutant male mice were obtained following heterozygous crossings with wild-type mice, all in a C57BL/6J genetic background, and were compared with sex-matched littermate controls. At 2 months of age, Notch2^Q2319X heterozygous mice had a weight that was comparable to that of littermate controls, although the femoral length was slightly shorter (Fig. 1). Administration of anti-Notch2 NRR or control antibody did not result in any apparent unwanted effects or obvious gastrointestinal toxicity; the mice appeared healthy, and their weight was not affected by anti-Notch2 NRR antibody.

Figure 1.

Weight and femoral length of male Hajdu-Cheney Notch2^Q2319X mutant mice (black bars) and littermate wild-type controls (white bars) treated with anti-Notch2 NRR (N2 NRR) or control antibodies (Control) for 4 weeks. Values are means ± SEM; n = 3 to 8. *Significantly different between Notch2^Q2319X mutant and control mice, P < 0.05.

Confirming previous observations, μCT of the distal femur revealed that 2-month-old Notch2^Q2319X mutant male mice had ∼50% decrease in trabecular bone volume associated with reduced connectivity and higher SMI (25). Administration of anti-Notch2 NRR antibody caused a modest, nonstatistically significant increase in trabecular bone volume of wild-type mice, and the osteopenia of Notch2^Q2319X mutant mice was not observed in mice treated with the antibody (Fig. 2). As a result, bone volume/total volume, connectivity, and SMI of Notch2^Q2319X mice treated with anti-Notch2 NRR antibody were not different from values obtained in wild-type mice treated with control antibody (Fig. 2). There were pronounced changes in the cortical bone structure of Notch2^Q2319X mutants; cortical bone was thin and porous, and bone area and cortical thickness were reduced (Fig. 3). The cortical osteopenia was not observed in Notch2^Q2319X mutant mice treated with anti-Notch2 NRR antibody, so the cortical bone area and cortical thickness in Notch2^Q2319X treated mice were not different from that of wild-type mice treated with control antibodies.

Figure 2.

Cancellous bone microarchitecture assessed by μCT of the distal femur from 2-month-old Notch2^Q2319X mutant male mice (black bars) and sex-matched littermate controls (white bars) treated with N2 NRR (n = 5 to 8) or control antibody (Ctrl; n = 4 to 8), both at 10 mg/kg for 4 weeks prior to being killed. Parameters shown are bone volume/tissue volume (BV/TV); trabecular separation (Tb.Sp), number (Tb.N), and thickness (Tb.Th); connectivity density (Conn.D); SMI; and density of material expressed as milligrams of hydroxyapatite (HA) per cubic centimeter⁻³. Values are means ± SEM. *Significantly different between Notch2^Q2319X and control, P < 0.05; **P < 0.053. #Significantly different between N2 NRR and control antibody, P < 0.05. Representative images show cancellous bone osteopenia in Notch2^Q2319X mutant mice and its reversal by N2 NRR antibodies.

Figure 3.

Cortical bone microarchitecture assessed by μCT of the femoral midshaft from 2-month-old Notch2^Q2319X mutant male mice (black bars) and sex-matched littermate controls (white bars) treated with N2 NRR (n = 5 to 8) or antiragweed control antibody (Ctrl; n = 4 to 8), both at 10 mg/kg for 4 weeks prior to euthanizing. Parameters shown are bone volume/tissue volume (BV/TV); cortical porosity (Ct.Po) and thickness (Ct.Th); total area (TA) and bone area (BA); periosteal perimeter (Ps.Pm) and endocortical perimeter (Ec.Pm); and density of material expressed as milligrams of hydroxyapatite (HA) per centimeter⁻³. Values are means ± SEM. *Significantly different between Notch2^Q2319X and control, P < 0.05. #Significantly different between N2 NRR and control antibody, P < 0.05. Representative images show cortical bone osteopenia in Notch2^Q2319X mutant mice and its reversal by N2 NRR antibodies.

Cancellous bone histomorphometric analysis of femurs from Notch2^Q2319X mutant mice confirmed the findings obtained by μCT and demonstrated that they had decreased bone volume/tissue volume secondary to a decrease in trabecular number (Fig. 4). There was a nonstatistically significant 10% increase in osteoclast number and eroded surface in Notch2^Q2319X mutant mice and no change in osteoblast number or bone formation rate. Treatment of Notch2^Q2319X mutants with anti-Notch2 NRR antibodies reversed the trabecular structural phenotype; thus, in these mice, the bone volume/total volume was not different from that of control mice treated with anti-Notch2 NRR antibody. In addition, osteoclast number and eroded surface/bone surface were significantly reduced by anti-Notch2 NRR antibody compared with Notch2^Q2319X mutants treated with control antibody. These results suggest that a Notch2-NRR antibody inhibitory effect on bone resorption was responsible for the normalization of the phenotype.

Figure 4.

Cancellous bone histomorphometry of the distal femur from 2-month-old Notch2^Q2319X mutant male mice (black bars) and sex-matched littermate controls (white bars) treated with N2 NRR (n = 10 to 11) or control antibody (Ctrl; n = 9 to 11), both at 10 mg/kg for 4 weeks prior to euthanizing. Values are means ± SEM. Static parameters shown are bone volume/tissue volume (BV/TV); trabecular number and thickness (Tb.N and Tb.Th); number of osteoblasts/bone perimeter (N.Ob/B.Pm); osteoid surface/bone surface (OS/BS); number of osteoclasts/bone perimeter (N.Oc/B.Pm); and eroded surface/bone surface (ES/BS). Parameters for dynamic histomorphometry are mineral apposition rate (MAR); mineralizing surface/bone surface (MS/BS); and bone formation rate (BFR). Values for these parameters, n = 4 to 7. *Significantly different between Notch2^Q2319X and control, P < 0.05. #Significantly different between N2 NRR and control antibody, P < 0.05. Representative static cancellous bone histological sections stained with toluidine blue show decreased number of trabeculae in Notch2^Q2319X mice and its reversal by treatment with N2 NRR antibodies. Bars in right corner indicate 200 μm.

Cortical bone histomorphometry revealed a decrease in osteoblasts and an increase in osteoclasts on the endocortical surface of Notch2^Q2319X mutant mice but no difference in osteocyte number between control and Notch2^Q2319X mutants (Fig. 5). There was an increase in endocortical eroded surface in Notch2^Q2319X mutants, but it was not statistically significant (P < 0.084). There were no differences in endocortical cellular parameters between Notch2^Q2319X and control mice treated with anti-Notch2 NRR antibody, suggesting a partial rescue of the phenotype by the anti-Notch2 NRR antibody. Cells in the periosteal perimeter could not be identified with sufficient confidence and are not reported.

Figure 5.

Cortical histomorphometry of the mid-diaphysis from 2-month-old Notch2^Q2319X mutant male mice and control male littermates treated with N2 NRR (n = 5) or control antibody (Ctrl; n = 4 to 5), both at 10 mg/kg for 4 weeks prior to euthanizing. Parameters for cortical bone are bone volume/tissue volume (BV/TV); BA; number of osteocytes/bone area (N.Ot/B.Ar); number of osteoblasts/endocortical bone perimeter (N.Ob/Ec.B.Pm) and number of osteoclasts/endocortical bone perimeter (N.Oc/Ec.B.Pm); osteoid surface/bone surface (OS/BS); osteoclast surface/bone surface (Oc.S/BS); and ES/BS. Values are means ± SEM. *Significantly different between Notch2^Q2319X and control, P < 0.05. Representative cross-sectional cortical bone stained with hematoxylin and eosin is shown. Arrows point to osteoclasts on the endocortical surface.

Gene expression in Notch2^Q2319X mutant mice

qRT-PCR of RNA extracted from femurs and tibiae from mutant and control mice revealed an induction (values: means ± SEM normalized to control after correction for Rpl38; n = 3 to 6) of 2.0 ± 0.2–, 1.8 ± 0.3–, and 3.2 ± 0.5–fold in Hey1, Hey2, and HeyL, respectively, in bones from mutant mice compared with control mice treated with control antibody (P < 0.05 for Hey1 and HeyL). Expression in control mice was 1.0 ± 0.2, 1.0 ± 0.3, and 1.0 ± 0.1 for Hey1, Hey2, and HeyL, respectively. The induction of these Notch target genes confirms activation of Notch signaling in skeletal tissue. Administration of anti-Notch2 NRR antibodies resulted in a decrease in Hey1, Hey2, and HeyL expression to 0.7 ± 0.1, 0.8 ± 0.1, and 0.5 ± 0.1, respectively, levels that did not differ from those found in control bones (P > 0.05) but were lower than those in mutant mice treated with control antibody (all P < 0.05). These results demonstrate that the antibody effectively prevented Notch activation. In accordance with previous observations and the resorptive phenotype observed, we found a 2.1 ± 0.4–fold increased expression of Tnfsf11, encoding for Rankl, in tibiae and femurs from Notch2^Q2319X mutant mice compared with expression in control mice (P < 0.052) (25). Following the administration of the anti-Notch2 NRR antibody, Tnfsf11 transcript levels were reduced to 1.1 ± 0.3 (P > 0.05 vs control), possibly contributing to the reversal of the skeletal phenotype.

Osteoblast-enriched cell cultures

Notch2^6955C>T mutant transcripts were detected only in cells from Notch2^Q2319X mutant mice. The expression of the Notch target genes Hey1, Hey2, and HeyL was increased in Notch2^Q2319X osteoblasts, confirming activation of Notch signaling (Fig. 6). Addition of anti-Notch2 NRR antibody to the culture opposed these effects, confirming that the antibody acts by preventing Notch activation. In accordance with the enhanced bone resorption observed in Notch2^Q2319X mice and confirming prior observations, expression of Tnfsf11 was increased in osteoblasts from mutant mice, and the increase was prevented by the addition of the anti-Notch2 NRR antibody to the culture (Fig. 6) (25).

Figure 6.

Calvarial osteoblast-enriched cells from Notch2^Q2319X mutant (black bars) and wild-type (white bars) littermate controls were isolated and cultured in the presence of N2 NRR or control antibody (Control) at 10 μg/mL. Total RNA was extracted, and gene expression was determined by qRT-PCR in the presence of specific primers and probes. Data are expressed as Notch2^6955T>C (Notch2^Q2319X) mutant, Hey1, Hey2, HeyL, and Tnfsf11 (Rankl) copy number corrected for Rpl38. Values are means ± SEM; n = 4. *Significantly different between Notch2^Q2319X mutant and wild-type control cells, P < 0.05. #Significantly different between N2 NRR and control antibodies.

Bone marrow macrophage cultures and osteoclast formation

Confirming previous work, the Notch2^Q2319X mutation resulted in enhanced osteoclastogenesis in BMMs cultured in the presence of M-CSF and Rankl (25). There was a significant increase in the number of TRAP-positive multinucleated cells in BMMs from Notch2^Q2319X mutants compared with controls (Fig. 7). In accordance with the reversal of the phenotype in vivo, the addition of the anti-Notch2 NRR antibody to the cultures decreased osteoclastogenesis so that the number of osteoclasts was comparable to that observed in control cultures. These results confirm previous work demonstrating that Notch activation is required for the effects of Notch2 on osteoclastogenesis and explain the reversal of the skeletal phenotype by the administration of anti-Notch2 NRR antibodies (25).

Figure 7.

Bone marrow mononuclear cells harvested from long bones of fore and hind limbs of Notch2^Q2319X mutants (black bars) and wild-type littermate controls (white bars) were cultured for 72 hours in the presence of M-CSF at 30 ng/mL and then M-CSF and Rankl at 10 ng/mL in the presence of N2 NRR at 10 μg/mL or control antibody at 20 μg/mL for osteoclast formation or 10 μg/mL for pit assay. (a) Cultures were assessed for TRAP by enzyme histochemistry or for resorption area following toluidine blue staining. Representative images of TRAP-stained multinucleated cells and of resorption pits on toluidine blue–stained bone chips are shown in the upper panel. Data are expressed as number of TRAP-positive multinucleated cells, resorption pit area/total area, and resorption pit area/total area corrected for number of TRAP-positive cells. Values are means ± SEM; n = 3 to 4. (b) Total RNA was extracted, and gene expression was measured by qRT-PCR in the presence of specific primers and probes. Data are expressed as Notch2^6955T>C (Notch2^Q2319X) mutant, Hes1, Nfatc1, and Acp5 copy number corrected for Rpl38 copy number and normalized to expression in control cultures exposed to the anti-ragweed antibody. Values are means ± SEM; n = 6 to 14. *Significantly different between Notch2^Q2319X mutant and control cells. #Significantly different between N2 NRR and control antibodies.

To determine the effect of the Notch2^Q2319X mutation on bone resorption, BMMs from control and Notch2^Q2319X mutants were cultured on bovine bone slices for 5 days in the presence of M-CSF and Rankl with and without anti-Notch2 NRR antibody. Pit resorption area/total bone area was significantly increased in the context of the Notch2^Q2319X mutation, and treatment of Notch2^Q2319X cultures with anti-Notch2 NRR antibodies prevented the effect. However, normalization of pit resorption area/total bone area by the number of TRAP-positive multinucleated cells resulted in no significant differences between Notch2^Q2319X and control cultures, indicating that the enhanced resorption was secondary to the increase in osteoclast number. Notch2^6955C>T mutant transcripts were detected only in cells from Notch2^Q2319X mice; and although Hes1 expression was similar in control and Notch2^Q2319X mutant cultures, anti-Notch2 NRR antibodies reduced Hes1 messenger RNA in mutant cultures. Confirming the mechanism of Notch2 action in osteoclast precursors, the expression of Nfatc1 and Acp5 was increased in Notch2^Q2319X cells, an effect that was reversed by the anti-Notch2 NRR antibody (13).

Discussion

Our findings confirm that a Notch2 gain-of-function mutation, replicating the one found in HCS in humans, causes cancellous and cortical bone osteopenia. Although the phenotype was manifested in mice of both sexes, at a younger age it seemed to be more pronounced in male mice (25). Consequently, in the current study, we treated 1-month-old male mice with an anti-Notch2 NRR antibody for a 4-week period in an attempt to reverse the osteopenic phenotype. Although the anti-Notch2 NRR was effective in opposing the phenotype, one needs to be cautious and not extrapolate the results to female mice because only male mice were tested. Another limitation of the work presented is that the same animal could not be analyzed before and after the treatment with anti-Notch2 NRR because the analysis would have required the euthanizing of mice before the initiation of the treatment. The anti-Notch2 antibody used binds to the NRR and prevents Notch2 activation and, as a result, neutralizes Notch2 function (26). The antibody is specific to Notch2 so that the results obtained should not be attributed to the neutralization of other Notch receptors (26).

The phenotype of the Notch2^Q2319X mutant mouse recapitulates aspects of HCS, but Notch2^Q2319X mutant mice do not exhibit detectable acral osteolysis or the obvious neurologic manifestations reported in humans affected by the disease. This is possibly because our work was limited to the study of relatively young heterozygous mice, and additional phenotypic manifestations may require greater or protracted exposure to increased levels of Notch2 and may appear later in life. Recently, we found that an occasional newborn homozygous mutant manifests craniofacial abnormalities resembling the human syndrome, including micrognathia (E. Canalis, unpublished observations). Unfortunately, homozygous Notch2^Q2319X mutants do not survive, and it was not possible to conduct further studies on the characterization of their phenotype or its reversal by anti-Notch2 NRR antibodies.

In contrast to the inhibitory actions of Notch1 on osteoblast and osteoclast differentiation, we confirmed that Notch2 activation in vitro induces osteoclastogenesis and that this effect is reversed by anti-Notch2 NRR antibodies. The effect of the anti-Notch2 NRR antibody on control osteoclastogenesis was modest, and the previously reported effect of γ-secretase inhibitors is transient, whereas their inhibitory effects in mice harboring the Notch2^Q2319X mutation are clear. The difference is most likely due to the modest basal expression of Notch2 in bone marrow cells of the myeloid lineage, and the results are in accordance with the absence of a skeletal phenotype in mice harboring a Notch2 inactivation in LysM-expressing cells (12, 51). The anti-Notch2 NRR antibody had a modest (nonsignificant) effect on gene expression in control wild-type cultures and, as a consequence, had a minimal effect on osteoclastogenesis under basal conditions. The phenotype of the Notch2^Q2319X mutation is in accordance with a gain-of-Notch2 function and the known effects of Notch2 on osteoclast differentiation. By interacting with NF-κB on Nfatc1 regulatory elements in cells of the osteoclast lineage, Notch2 enhances osteoclastogenesis (13). It is important to note that Notch2^Q2319X mutant mice did not exhibit an increase in osteoblast number or a bone-forming response to the increase in bone resorption, suggesting a possible negative regulation of osteoblastogenesis or osteoblastic function by the Notch2 mutation. This could be due to the direct effect of Notch2 on cells of the osteoblast lineage or secondary to changes on an alternative signaling pathway. Recent work confirmed that inactivation of Notch2 in cells of the osteoblastic lineage results in increased bone mass and osteogenic potential, indicating a role in osteoblast differentiation and function (51). The authors did not report a phenotype following the inactivation of Notch2 in cells of the osteoclast lineage. This could be because the basal expression of Notch2 in osteoclast precursors is modest or because the mating scheme reported used one LysM-Cre allele to delete Notch2^loxP/loxP alleles, possibly not achieving sufficient Cre-mediated recombination and gene inactivation (52).

In a previous report, we demonstrated that Notch2^Q2319X mutants exhibit cancellous osteopenia secondary to increased osteoclast number and bone resorption (25). In the current study, the cellular effect was modest (∼10%) in cancellous bone, although the effect was significant on the endocortical surface. We do not have an explanation for the different results observed in cancellous bone in the current study. However, Notch2^Q2319X mutant mice exhibited increased osteoclast number and eroded surface at 1 month of age and not at 3 months (25). This suggests that the mice characterized in the current study had increased osteoclasts and bone resorption at 1 month of age when anti-Notch2 NRR treatment was initiated, explaining the reversal of the structural skeletal phenotype.

Notch2^Q2319X mutants expressed increased levels of Rankl, possibly contributing to the phenotype observed. Rankl expression in osteoblasts was dependent on Notch activation because it was not observed following treatment with anti-Notch2 NRR antibodies. The increased number of osteoclasts and bone resorption appear to be secondary to various mechanisms, including an increased expression of Rankl and in the capacity to differentiate into mature osteoclasts. In vitro experiments demonstrated that anti-Notch2 NRR antibody decreased the induction of osteoclast maturation by Notch2^Q2319X mutant cells. This is congruent with previous work demonstrating that Notch2 activation is required for its effects on osteoclastogenesis (25).

The approach to downregulate Notch signaling has been diverse and includes the use of biochemical inhibitors of Notch activation, antibodies to Notch receptors or their ligands, and small permeable molecules that prevent the formation of an NICD/Rbpjk/Maml ternary complex (53). γ-Secretase inhibitors were developed to reduce amyloid-β protein aggregates in Alzheimer’s disease and are frequently used to block the cleavage of the Notch receptor induced by presenilins (54). However, γ-secretase inhibitors are not specific, and about 90 substrates of the γ-secretase complex are known (55). Moreover, γ-secretase inhibitors prevent the indiscriminate activation of all Notch receptors. Thapsigargin is an inhibitor of the sarco/endoplasmic reticulum Ca²⁺-ATPase that precludes the maturation and folding of Notch and its exit from the endoplasmic reticulum and has been used to prevent the effects of Notch (56, 57). Synthetic small-cell permeable molecules that prevent the assembly of an active Notch transcriptional complex also can be used for the inhibition of all Notch isoforms, but their long-term efficacy is unknown (58).

To target specific Notch receptors, antibodies to the NRR of Notch1, Notch2, and Notch3 have been developed (26, 59). The targeting of this region prevents the initial cleavage and activation of specific Notch isoforms, making their use ideal for the neutralization of individual Notch receptors. Because of this, anti-Notch2 NRR antibodies were selected to reverse the phenotype of the HCS mutant mouse model. However, one needs to be cautious with the extrapolation of results from these studies to the human disease because information on the inhibition of Notch signaling in humans is quite limited. Moreover, Notch neutralization is not without unwanted events, and it may cause vascular tumors and gastrointestinal toxicity (60, 61). An increase in bone resorption appears to explain the osteopenia in a mouse model of HCS, and it is possible that the pronounced bone loss experienced by subjects with HCS is due to enhanced bone resorption. If this is the case, antiresorptive therapy could prove beneficial to patients with HCS presenting with osteoporosis (62). However clinical data on its effectiveness are sparse.

In conclusion, Notch2^Q2319X mutant mice replicating the mutation found in subjects with HCS exhibit marked cancellous and cortical bone osteopenia, which can be prevented or reversed by the administration of anti-Notch2 NRR antibodies.