Figure 1.
CD68 is expressed by macrophages and osteoclasts.
(A) CD68 and β-actin expression in mouse bone marrow suspension cells treated with 44 ng/mL M-CSF +/− 100 ng/mL RANKL for indicated days (d). M-CSF increases expression of CD68 in a time dependant manner, and RANKL ligand induces an accelerated gel migration rate. Image is representative of 3 independent experiments. Quantification of relative band density is aggregate of 3 independent experiments; data shown is mean + standard deviation. (B) CD68 and β-actin expression in RAW264.7 cells with or without 100 ng/mL RANKL treatment for indicated times. RAW264.7 cells have constitutive M-CSF-stimulated signaling and continuous expression of CD68. RANKL induces similar changes in gel migration as those seen in primary macrophages. Image is representative of 3 independent experiments. Quantification of relative band density is aggregate of 3 independent experiments; data shown is mean + standard deviation. (C) Flow cytometry histograms of BMMs with 3-day treatments of 44 ng/mL M-CSF +/– 100 ng/mL RANKL with (Total) and without (Surface) permeablization with .5% saponin. CD68 can be detected on the surface of BMMs, and neither surface nor total detectible levels of CD68 are altered by addition of RANKL. Image is representative of 2 independent experiments.
Figure 2.
CD68 can be found in a dome-like pattern in osteoclasts cultured on bovine cortical bone slices.
(A) BMMs were seeded onto bone slices with 44 ng/mL M-CSF and 100 ng/mL RANKL and differentiated into osteoclasts over 4 days. Cells were fixed with 4% paraformaldehyde/PBS and stained with Alexa-488-conjugated phalloidin (actin, green), Hoescht (nuclei, blue), and either anti-CD68 (CD68, red) or rat non-immune IgG2a (IgG2a Merge) antibody followed by Alexa-647-conjugated anti-Rat IgG. Scale bars for CD68 staining are 40μm; scale bar for IgG2a staining is 20μm. Images are representative of 3 independent experiments. (B) Enlarged merge image from A. * and + indicate corresponding XZ and YZ cross sectional images, respectively. Scale bar is 40μm. Image is representative of 3 independent experiments. (C) 3-D reconstruction of osteoclast cross sectioned in B with actin in green, nuclei in blue, and CD68 in red.
Figure 3.
BMMs from CD68−/− mice lack expression of CD68.
(A) Vector diagram with a neomycin phosphotransferase expression cassette (PGK-NEO) flanked by sequences with homology to targeted genomic sequence. A thymidine kinase expression cassette (PGK-TK) lies outside the homology region of vector. (B) Structure of targeted wild type allele. A southern blot probe can hybridize to a genomic sequence outside of the homology region. P1 and P2 are primers that specifically amplify the sequence of CD68 targetted for replacement. (C) Recombined allele with exons 1 and 2 of CD68 gene replaced with PGK-NEO. Properly targeted recombined alleles do not contain the thymidine kinase expression cassette. P3 and P4 are primers that specifically amplify a region of PGK-NEO. (D) Tail tip extracts from each of three resultant genotypes were subjected to genotyping PCR using P1, P2, P3, and P4. Each genotype produced a unique pattern of PCR products. (E) Lysates from BMMs from each genotype cultured with 220 ng/mL M-CSF were immunoblotted with antibodies against CD68 and α-Tubulin. While expression of CD68 was seen in +/+ and +/− BMMs, no CD68 could be detected in lysates from −/− BMMs.
Table 1.
Mouse Birth Ratios.
Figure 4.
CD68−/− mice have increased trabecular bone and decreased trabecular tissue mineral density.
(A) Representative trabecular µCT images from 6-month-old female mice of each genotype (5 mice per group). (B) µCT analysis of distal femoral trabecular bone revealed that CD68−/− mice have increased bone volume (BV/TV), increased trabecular number (Tb. N), decreased trabecular spacing (Tb. Sp.) and decreased trabecular tissue mineral density (Trabecular TMD). There was no significant difference in trabecular thickness (Tb. Th.). (C) Representative cortical µCT images from each genotype. (D) Quantification of measured parameters. There was no significant difference between genotypes in either cortical thickness (Ct. Th.) or cortical tissue mineral density (Cortical TMD). *, p<.05; **, p<.01. Data presented are means + standard deviation.
Figure 5.
Histological and histomorphometric analysis of CD68+/+, +/−, and −/− mice.
(A) Representative formalin-fixed, paraffin-embedded histological sections from each genotyped stained for TRAP activity and counterstained with hematoxylin at 40X magnification. (B) 400X magnification of area defined in (A). (C) Quantification of number of osteoclasts per bone surface (N. OC/BS). There was no significant difference in N. OC/BS between genotypes. (D) Representative 70% ethanol-fixed, plastic-embedded histological sections from each genotype stained with Goldner’s trichrome at 40X magnification. (E) 400X magnification of area defined in (D). (F) Histomorphometric analysis of trichrome-stained sections. There was no significant difference between genotypes in numbers of osteoblasts (N. OB/BS) per bone surface. There was a significant increase in bone volume per total volume (BV/TV) in CD68−/− mice compared to CD68+/+ animals and trabecular number (N. Tb.) in CD68−/− mice compared to CD68+/+ and CD68+/− animals. (G) Quantification of mineral apposition rate (MAR) via analysis of calcein double labeling. The MAR was significantly higher in CD68−/− mice compared to both CD68+/+ and CD68+/− animals, and MAR was significantly higher in CD68+/− mice compared to CD68+/+ animals. *, p<.05; **, p<.01. Data presented are means + standard deviation.
Figure 6.
CD68 is not expressed by osteoblasts.
CD68 and β-actin expression in mouse BMMs(Mφ) cultured with 220 ng/mL M-CSF and cultured mouse calvarial osteoblasts (OB) was determined by Western immunoblotting. While expression of CD68 is high in BMMs, expression of CD68 could not be detected in osteoblasts.
Figure 7.
CD68−/− osteoclasts do not efficiently resorb bone.
BMMs from CD68+/+, +/−, and −/− mice were seeded onto bovine cortical bone slices and differentiated into osteoclasts over 4 days. Differentiated osteoclasts were allowed to resorb the slices for an additional 3 days. (A) Representative images of resorbed bone slices generated using laser scanning confocal microscopy. A pit from each image is marked with an arrow. Scale bars are 70μm. (B) Quantification of resorbed area. Data represented as means + standard deviation. *, p<.001. 3 visual fields each from 3 separately resorbed bone slices were assessed. Data presented are means + standard deviation. Images and data are representative of 2 independent experiments.
Figure 8.
CD68−/− have abnormal morphology.
(A) Prior to TRAP staining, osteoclasts of all three genotypes were of relatively similar size. CD68−/− osteoclasts demonstrated intracellular vacuole-like structures that were not present to such an extent in CD68-expressing cells. Scale bars are 100μm. Images are representative of 3 independent experiments. (B) During the fixation process (fixative: 25 mL citrate solution [18 mM citric acid, 9 mM sodium citrate, 12 mM sodium chloride, pH 3.6], 68 mL acetone, 8 mL 37% formaldehyde), many CD68−/− osteoclasts were reduced in size and partially detached from the culture substrate resulting in a smaller size following TRAP staining. Scale bars are 200μm. Images are representative of 3 independent experiments.