Mice lacking beta3 integrins are osteosclerotic because of dysfunctional osteoclasts

Abstract

Osteoclasts express the alphavbeta3 integrin, an adhesion receptor that has been implicated in bone resorption and that is therefore a potential therapeutic target. To assess the role of this heterodimer in skeletal development in vivo, we engineered mice in which the gene for the beta3 integrin subunit was deleted. Bone marrow macrophages derived from these mutants differentiate in vitro into numerous osteoclasts, thus establishing that alphavbeta3 is not necessary for osteoclast recruitment. Furthermore, the closely related integrin, alphavbeta5, does not substitute for alphavbeta3 during cytokine stimulation or authentic osteoclastogenesis. beta3 knockout mice, but not their heterozygous littermates, develop histologically and radiographically evident osteosclerosis with age. Despite their increased bone mass, beta3-null mice contain 3.5-fold more osteoclasts than do heterozygotes. These mutant osteoclasts are, however, dysfunctional, as evidenced by their reduced ability to resorb whale dentin in vitro and the significant hypocalcemia seen in the knockout mice. The resorptive defect in beta3-deficient osteoclasts may reflect absence of matrix-derived intracellular signals, since their cytoskeleton is distinctly abnormal and they fail to spread in vitro, to form actin rings ex vivo, or to form normal ruffled membranes in vivo. Thus, although it is not required for osteoclastogenesis, the integrin alphavbeta3 is essential for normal osteoclast function.

Figure 1

β3^–/– BMMs differentiate into osteoclasts that fail to spread. Seven-day osteoclastogenic cultures, containing BMMs of 2-month-old female littermates and ST2 stromal cells, were stained for TRAP activity (red reaction product). β3^+/+ and β3^+/– BMMs form numerous well spread, TRAP-expressing giant cells. TRAP-expressing cells in cultures with β3^–/– BMMS, in contrast, fail to spread (×100). High magnification of the same cultures (bottom row) demonstrates that the TRAP-expressing giant cells, in each circumstance, contain multiple nuclei (arrow) (×250).

Figure 2

β3^–/– BMMs fail to express αvβ3 and the integrin is not induced by IL-4. Pure populations of β3^+/– or β3^–/– BMMs were treated for 48 hours with vehicle (C) or murine IL-4 in an amount (5 ng/mL) known to optimally enhance β3 expression (10). The cells were then surface labeled with NHS-biotin and lysed, and then the lysate was immunoprecipitated with hamster antimurine β3 mAb and the immunoprecipitate subjected to SDS-PAGE. Similar to wild-type BMMs (10), β3^+/– cells express minimal αvβ3 and the integrin is upregulated by IL-4. In contrast, β3^–/– BMMs fail to express αvβ3 in the presence or absence of IL-4.

Figure 3

The β5 integrin subunit does not compensate for absence of β3 in GM-CSF–treated β3^–/– BMMs. Pure populations of β3^+/– or β3^–/– BMMs were treated for 48 or 72 hours with vehicle (–) or murine GM-CSF (+) in an amount (10 ng/mL) known to optimally suppress β5 mRNA expression by wild-type BMMs (17). Total RNA was isolated and subjected to Northern analysis using a murine β5 cDNA. The cytokine inhibits β5 mRNA expression equally in β3^+/– and β3^–/– BMMs.

Figure 4

Compensatory upregulation of β5 integrin does not occur in β3^–/– osteoclasts. Wild-type (WT) and β3^–/– BMMs were grown in the presence of M-CSF alone or in combination with OPGL for 4 days. Ten micrograms of total RNA was run in each lane for Northern analysis. The blot was probed first with a combination of β5 and GAPDH probes, then stripped and probed with β3. In WT cultures, OPGL upregulates expression of β3 and downregulates expression of β5. In β3^–/– cultures, no β3 mRNA is seen, and the downregulation of β5 is equivalent to that seen in WT cultures.

Figure 5

β3^–/– mice develop radiographic osteosclerosis. Tails of 3 6-month-old β3^–/– mice and their heterozygous or wild-type sex-matched littermates were radiographed simultaneously to avoid technical artifacts. The mutant, in all circumstances, shows enhanced bone density.

Figure 6

β3^–/– mice develop histologic osteosclerosis. Distal femora of 2 sets of 6-month-old, sex-matched, β3^+/+, β3^+/–, and β3^–/– littermates. Note the marked increase in cortical and trabecular bone mass in β3^–/– mice as compared with wild-type and heterozygotes, which are indistinguishable. (Hematoxylin and eosin stained; ×40).

Figure 7

β3^–/– mice generate increased numbers of osteoclasts. Histologic sections of distal femoral metaphysic of 2-month-old β3^+/– and β3^–/– littermates were stained for TRAP activity (red reaction product) to identify osteoclasts that are increased in the mutant (×250).

Figure 8

Ruffled membrane formation in osteoclasts of β3^–/– mice is abnormal. Electron micrographs of osteoclasts resident in metaphyseal bone of β3^+/– and β3^–/– mice. The ruffled membranes of heterozygous osteoclasts are indistinguishable from wild-type, whereas those of each β3^–/– osteoclast consists of thickened and blunted villous structures. Scale bar: l.0 μm.

Figure 9

Actin organization in β3^–/– osteoclasts is abnormal. β3^+/– and β3^–/– marrow was blotted onto slides. The cells were stained for TRAP activity, permeabilized, and then incubated with rhodamine-phalloidin. TRAP-expressing multinucleated cells were identified and f-actin was visualized by confocal microscopy. β3^+/– osteoclasts develop a characteristic actin ring, whereas F-actin in β3 osteoclasts is diffusely distributed throughout the cytoplasm.

Figure 10

The resorptive capacity of β3^–/– osteoclasts is defective. Osteoclasts were generated by culturing β3^+/– or β3^–/– BMMs with ST2 cells on dentin. After 7 days, the cells were removed, and resorption lacuna visualized by scanning electron microscopy. Whereas β3^+/– osteoclasts produce numerous, well defined resorption lacunae, those formed by β3^–/– osteoclasts have indistinct borders and are shallow.