Caveolae are 50- to 100-nm plasmalemmal vesicles formed by oligomerized caveolin, a 22-kDa phosphoprotein. These organelles have been implicated in critical signal transduction and molecular transport processes. Here, we show for the first time that osteoblasts express caveolin and have abundant caveolae. Membrane fractionation techniques indicate that osteoblast caveolin is found in detergent-resistant membranes that have the buoyant density characteristic of caveolae, whereas immunoblotting and reverse-transcription polymerase chain reaction (RT-PCR) show that osteoblasts express both caveolin-1 and −2 isoforms. Electron microscopy (EM) and immunofluorescence reveal the hallmarks of caveolae in osteoblasts: abundant 50- to 100-nm noncoated cell surface invaginations (caveolae) and abundant punctate clusters of immunostained caveolin.
OSTEOBLASTS ARE cells responsible for the production of bone matrix and are critical for bone development and for maintaining bone density.(1) These cells, which develop from mesenchymal stem cells, influence multiple aspects of bone by (i) secreting type I collagen and noncollagenous proteins of the bone matrix, (ii) regulating the mineralization of the bone matrix, and (iii) controlling the differentiation and function of osteoclasts. As such, osteoblasts play critical roles in normal skeletal development and homeostasis, as well as in various pathological conditions of the skeleton, including osteoporosis, osteopetrosis, bone cancer, Paget's disease, and hyperthyroidism, to name a few. Thus, a more complete understanding of the biochemistry/cell biology of osteoblasts is required to diagnose, treat, and prevent human disease.
In all eukaryotic cell membranes, sphingolipids and cholesterol appear to form detergent-resistant membrane microdomains (DRMs) by self-aggregation during transport from the trans-Golgi network to the cell surface. DRMs can be thought of as discrete lipid rafts in the general lipid bilayer milieu of the plasma membrane. Various proteins including those with glycosylphosphatidylinositol (GPI) anchors and caveolin (the structural component of caveolae) also are sorted to these “microdomains,” although ultimately not necessarily to the same ones.(2,3) At least two varieties of DRM exist on many cell surfaces and are physically separable, morphologically distinguishable, and chemically distinct with different protein markers.(3) One variety of cell surface DRM contains GPI-anchored proteins, src family protein kinases, G proteins, and other components. (4–8) These DRMs appear by electron microscopy (EM) as “flat” areas of the cell membrane; we refer to these as G domains. A second variety of DRMs is called caveolae and contains caveolin, certain growth factor receptors, src family protein kinases, vesicle transport molecules, and G proteins (as well as other components).(3, 9, 10) By EM, caveolae appear as striated 50- to 100-nm membrane “invaginations” coated with oligomerizedcaveolin. (11–13) Although many cell types have both caveolae and G domains, some cell varieties do not express caveolin (i.e., hemopoietic cells) and may only have G domains.(4–8, 14, 15)
Caveolae are distinctive DRMs present in adipocytes, myocytes, endothelial cells, and in some other cell types. Caveolae have been implicated in molecular transport processes and in signal transduction.(9, 10 16–25) In accord with their role in cell signaling, a number of receptors and signal transduction molecules have been localized to caveolae: endothelin receptors,(26) thrombin receptors,(27) platelet-derived growth factor (PDGF) receptors,(9, 17, 18) a calcium pump,(10,28) an inositol 1,4,5-trisphosphate receptor,(10,29) guanosine triphosphate (GTP)-binding proteins (heterotrimeric, large, and small),(10, 23, 30–32) src family nonreceptor tyrosine kinases, phosphatidylinositol 3 kinase (PI3-kinase), phospholipase Cγ (PLCγ), protein kinase Cα (PKCα), and protein kinase Cβ (PKCβ).(9) Central to the concept of G domains and caveolae is the hypothesis that these structures are important in: (1) organizing signal transduction elements in proximity to receptors and (2) creating local environments in which signal propagation, amplification, and even possibly cross-talk between signal transduction elements can occur. Thus, it is hypothesized that caveolae and other DRMs are centers of signal amplification and are central in creating signal cascades that ultimately direct gene transcription and cell cycle regulation. The present study establishes the foundation for testing this hypothesis in the osteoblast cell lineage.
Caveolin, a 22-kDa protein, is a major structural component of caveolae(12) and appears necessary for caveolae biogenesis.(15) Three isoforms of caveolin (caveolin-1, −2, and −3) have been discovered. Endothelial cells, adipocytes, and most other cells examined express caveolin-1 and −2, which form hetero-oligomers that are the central structural unit of caveolae.(19,33) In contrast, myocytes express only caveolin-3, which form homo-oligomeric caveolae.(34,35) Caveolin has been shown to associate with a variety of signal transduction molecules(21, 30–32, 35–38) including heterotrimeric G proteins,(30,31) the small G protein Ras,(31,32) and src family protein kinases.(31) It has been suggested that caveolin is a negative regulator of signaling molecule activity and may act as a “scaffold” to which disparate signaling elements (including the ones indicated above) can attach in their inactive state.(31) Thus, by both binding and regulating the activities of a variety of signaling elements, caveolin may play a central role in controlling certain signal transduction cascades.
In this report we show that both human and murine osteoblasts express caveolin and have caveolae. Here, we reveal that osteoblasts express both caveolin-1 and −2; that osteoblast caveolin is expressed in Triton X-100-resistant membranes with the buoyant density characteristic of caveolae; that caveolin is abundantly expressed with a punctate distribution in the membranes of osteoblasts; and that osteoblasts have abundant 50- to 100-nm flask-shaped plasmalemmal vesicles.
MATERIALS AND METHODS
Human fetal osteoblasts(39) (hFOBs; generously provided by T. Spelsberg, Mayo Clinic, Rochester MN, U.S.A.) were grown in Dulbecco's modified Eagle medium (DMEM) supplemented with 15% fetal bovine serum (FBS), L-glutamine, penicillin/streptomycin, and gentamicin. MC3T3-E1 (murine osteoblastic cells; referred to as MC3T3 cells(40) were grown in α-modified Eagle medium (α-MEM), 10% FBS, L-glutamine, and penicillin/streptomycin. Both cell types were cultured in 5% CO2 at 37°C and were verified to express the osteoblast-specific protein osteocalcin by radioimmunoassay of conditioned media.
Reverse-transcription polymerase chain reaction
Total RNA from hFOB and MC3T3 cells was extracted using RNeasy kit (Qaigen, Valencia, CA, U.S.A.) according to the manufacturer's instructions. Reverse-transcription (RT) reactions containing 0.8 μg total RNA were carried out using rTth DNA polymerase (Perkin Elmer, Norwalk, CT, U.S.A.) and 50 pmol of Oligo d(T)16 at 42°C for 5 minutes, ramped to 65°C over 5 minutes, and continued for 30 minutes at 65°C. Complementary DNAs (cDNAs) made as described previously and a cDNA library made from primary human osteoblasts (see the following; generously provided by S. Ashkar, Children's Hospital, Boston, MA, U.S.A.) were amplified by the addition of 0.45 μM specific caveolin-1, −2, and −3 degenerative oligonucleotide primers to the reaction tubes (see the following) with the following profile: 1 cycle of 94°C for 2 minutes; set cycles of 95°C for 1 minute; 43°C for 1 minute; and 72°C for 1 minute followed by a single cycle of 60°C for 7 minutes. The following caveolin-specific oligonucleotide primers were used in polymerase chain reaction (PCR) reactions with hFOB and human mandible osteoblast cDNA: 1F (AAGGAGATC/TGACCTGGTG/CAAC) corresponding to caveolin-1 amino acid residues 47-53 and caveolin-3 amino acid residues 20-26; 2F (TTC/TGAA/GGAT/CGTGATC/TGCAGA) and 1R (TCTGCA/GATCACG/ATCC/TTCA/GAA) corresponding to human caveolin-1 amino acid residues 68-74, caveolin-2 amino acid residues 40-46, and caveolin-3 amino acid residues 41-47; 2R (CCAGATGTGG/CAA/GAC/AAGCA/TGA) corresponding to human caveolin-1 amino acid residues 122-128, caveolin-2 amino acid residues 94-100, and caveolin-3 amino acid residues 95-101, used in the following combinations (1F/1R, 2F/2R, and 1F/2R). The following caveolin-specific oligonucleotide primers were used in PCR reactions with MC3T3 osteoblast cDNA: 1F (TTTGAAGATGTGATTGCGGAA) corresponding to murine caveolin-1 amino acid residues 67-74 and 1R (CCAGATGTGCAGGAAGGAGAG) corresponding to murine caveolin-1 amino acid residues 121-128; 2F (TCGAGTCACGACCGGGATCCT) corresponding to murine caveolin-2 amino acid residues 35-42 and 2R (CCAGATGTGCAGACAGCTGAG) corresponding to murine caveolin-2 amino acid residues 107-114.
RNA from primary human osteoblasts
Human osteoblasts were isolated from bone removed during the extraction of a third molar. Bone fragments were cleansed of connective tissue, washed three times in α-MEM containing 10% FBS, and cultured for 2 weeks in α-MEM and 10% FBS in 5% CO2 at 37°C with media changes every 2 days. Cells were allowed to seed out of the explants for 2 weeks until reaching 60-70% confluence. These cells expressed osteoblast markers (alkaline phosphatase, osteopontin, and bone sialoprotein) and differentiated into mineralizing osteoblasts in the presence 10 μg/ml ascorbic acid and 100 mM β-glycerol phosphate. Total RNA was extracted from 5 × 106 cells using Tri-Reagent (Molecular Research Center, Cincinnati, OH, U.S.A.) according to the manufacturer's instructions and pBluescript cDNA library was constructed from the RNA using random primers and RT.
The following monoclonal (mAb) and polyclonal (pAb) antibodies were used in these studies: anticaveolin mAb clone 2234, anticaveolin pAb, (Transduction Labs, San Diego, CA, U.S.A.), anticaveolin-1 (Zymed, San Francisco, CA, U.S.A.), anticaveolin-2 (Transduction Labs), anti-β-actin mAb (Sigma, St. Louis, MO, U.S.A.), and anti-ϵ-coatomer protein (COP) pAb (gift of Dr. Monty Kreiger, Whitehead Institute, Cambridge, MA, U.S.A.).
Successive detergent extraction method
Solubility analysis of membrane constituents was performed essentially as described.(6,7) In brief, cells were resuspended in buffer A (25 mM 2-[N-morpholino]ethanesulfonic acid [MES] and 150 mM NaCl, pH 6.5). To this, an equal volume of the same buffer with 2% Triton X-100, 2 mM Na3VO4, and 2 mM phenylmethylsulfonyl fluoride (PMSF) was added (1% final Triton X-100; buffer A complete), and the cells were incubated on ice for 30 minutes. Insoluble fractions were pelleted in a microcentrifuge (14,000g) for 20 minutes at 4°C. The supernatant was removed (“S” fraction; soluble) and the insoluble pellet was resuspended in buffer B (1% Triton X-100, 10 mM Tris, pH 7.6, 500 mM NaCl, 2 mM Na3VO4, 60 mM β-octylglucoside [Sigma], and 1 mM PMSF) for 30 minutes on ice. Debris was pelleted in a microcentrifuge (14,000g) for 20 minutes at 4°C, and the supernatant was collected. This fraction is referred to as DRM.
Cell fractions were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using either 10% polyacrylamide gels or 5-15% gradient polyacrylamide gels, followed by electrotransfer to nitrocellulose (NC) membranes. NC membranes were stained with Ponceau S to verify uniform transfer and equal protein loading of gel lanes (where appropriate). The NC was incubated in blocking buffer (5% nonfat dry milk [wt/vol] in Tris-buffered saline [TBS], pH 7.6, and 0.05% Tween 20 [vol/vol]) and then incubated with various monoclonal and polyclonal antisera in blocking buffer. Incubations with primary antibodies were for 2 h except for the anticaveolin-2 antibody, which was overnight. After washing (two times in TBS-0.05% Tween 20), the blots were incubated with the appropriate reporter antibody (donkey anti-mouse or sheep anti-rabbit immunoglobulin G [IgG]) conjugated to horseradish peroxidase (Amersham Corp. Piscataway, NJ, U.S.A.), washed (three times in TBS-0.05% Tween 20), developed by treatment with enhanced chemiluminescence reagents (Pierce, Rockford, IL, U.S.A.), and visualized by exposure to X-ray film. In many experiments, signals on immunoblots were quantified on a phosphoimager (Molecular Dynamics, Sunnyvale, CA, U.S.A.) using ImageQuant software (Molecular Dynamics). In some experiments the NC was stripped (100 mM β-mercaptoethanol, 2% SDS [wt/vol], 62.5 mM Tris, pH 6.7, 60°C for 20 minutes), blocked, and immunoblotted with another antibody as described previously.
Cells were resuspended in buffer A complete and homogenized at 4°C with a Teflon homogenizer. The homogenate was centrifuged at 100,000g using a 5-40% sucrose gradient. Successive fractions were collected from the top of the gradient and subjected to immunoblot analysis as described previously.
MC3T3 and hFOB cells were grown on coverslips, fixed in methanol at −20°C, incubated in phosphate-buffered saline (PBS) with 3% goat serum, and incubated with anticaveolin pAb or with normal rabbit serum. After washing three times in PBS, incubating with anti-rabbit Texas red conjugated antibody (Pierce), washing three times in PBS, and washing three times in deionized, distilled H2O, coverslips were viewed on a fluorescence microscope and digital images were processed using either ONCOR or MetaMorph software.
MC3T3 and hFOB cells were fixed (2% paraformaldehyde, 2.5% glutaraldehyde, and 0.1 M NaCacodylate, pH 7.4), treated with osmium collidine, washed in 0.05 M NaMaleate, pH 5.2, incubated with 2% uranyl acetate in 0.05 M NaMaleate, pH 4.2-4.4, washed, dehydrated in ethanol, infiltrated with epon, and processed for transmission EM as described.(41)
Protein concentrations in membrane fractions were quantified by MicroBCA assay (Pierce) using a bovine serum albumin standard.
There are four critical criteria for establishing the presence of caveolae: (i) show the presence of caveolin; (ii) show that the caveolin is present in detergent-resistant buoyant membrane fractions; (iii) show that the caveolin is highly expressed with a punctate distribution; and (iv) reveal the presence of small, noncoated, flasklike membrane invaginations (caveolae) in surface membranes by EM. By all four criteria we show here that osteoblasts have caveolae.
Osteoblasts express caveolin
Caveolin, a 22-kDa structural component of caveolae,(12,15) was found to be highly enriched in the DRMs of the osteoblasts obtained by successive detergent extraction (Figs. 1A and 2A). In both hFOB and MC3T3 cells, >90% of the caveolin resisted Triton solubilization but could be solubilized by octylglucoside (thus located in DRM), whereas <10% of the caveolin was Triton soluble (hereafter this fraction will be referred to as “S”). Interestingly, despite being subject to SDS-PAGE, caveolin oligomers were apparent in overexposed autoradiographs (Fig. 1B). Multimers of caveolin that migrated to approximately 40-45 kDa, 60-66 kDa, and 120 kDa (with other fainter species at approximately 90 kDa) were present in the DRMs of both MC3T3 and hFOB cells. These results indicate that the DRMs of osteoblasts contain caveolin and caveolin oligomers, thus prompting us to further investigate the distribution of caveolin in these cells.
Cells were lysed in Triton X-100, homogenized, and subjected to sucrose gradient ultracentrifugation. Successive fractions collected from the top of the sucrose gradient were then analyzed by SDS-PAGE and immunoblotting. Most of the caveolin (>85%) was found in distinct bands, which fractionated at 15-20% sucrose, a characteristic property of caveolae(3, 9, 10, 22, 23, 30–32, 41) (fractions 5-7; Fig. 1C). These membrane fractions were all but completely free of any cytoskeletal (β-actin) and Golgi (ϵ-COP) markers. A small amount of the total cellular caveolin (<5%) was found in the soluble fraction and some caveolin (≈10%) was present in the insoluble cell pellet, perhaps suggesting a cytoskeletal association.
As described in the Introduction, three isoforms of caveolin (caveolin-1, −2, and −3) have been reported to date. To identify the forms of caveolin expressed by our osteoblasts, we performed immunoblot analysis of S and DRM fractions using caveolin isoform-specific mAbs. Caveolin-1 and −2 were both found in the DRMs of hFOB and MC3T3 cells (Fig. 2A). Caveolin-1 was found abundantly, and caveolin-2, although apparently less abundant, also was found in the osteoblast DRMs. In contrast, we were unable to detect any expression of caveolin-3 in these cells (data not shown).
To confirm our identification of the forms of caveolin expressed by these cells, RNA extracted from hFOB cells was subjected to RT-PCR using degenerate oligonucleotides corresponding to DNA sequences that have the least variation among the three caveolin isoforms (Fig. 2B). Products generated by RT-PCR were then subjected to sequence analysis to confirm their identity. We determined that osteoblasts express both caveolin-1 and −2, whereas no caveolin-3 gene transcripts were detected. We also tested a cDNA library made from primary human osteoblasts (Fig. 2C). The putative caveolin-1 and −2 amplimers were excised and sequenced, proving 100% identity with the known human cDNA sequences. Consistent with our other RT-PCR results, we were able to generate PCR products encoding caveolin-1 and −2 but not −3 from this cDNA library. Consistent with our immunoblotting, we also obtained RT-PCR results similar to those described previously using mouse caveolin specific oligonucleotides and RNA extracted from MC3T3 cells (Fig. 2D).
To explore the distribution of caveolin in osteoblasts further, the two osteoblast cell lines hFOB (Fig. 3) and MC3T3 (Fig. 4) were fixed and stained with either normal rabbit serum (Figs. 3C and 4C) or with an anticaveolin pAb, followed by anti-rabbit Texas-red conjugated antibody (Figs. 3A, 3B, 4A, and 4B). Normal rabbit serum gave a faint, diffuse staining pattern, whereas the anticaveolin antibody reproducibly produced a bright, punctate pattern of staining. Although in some cells the distribution of caveolin was unclear, in other cells caveolin staining appeared to be localized to the outer membrane and had some degree of polarization. Overall, >90% of all the cells examined stained brightly with the anticaveolin antibody (Figs. 4 and 5 and data not shown). Little nuclear staining with the anticaveolin antibody was noted. Thus, these osteoblasts express cell surface caveolin, although like other cultured cell varieties, they are likely to have a cytoplasmic pool of nonsurface caveolin. (42–44)
Osteoblasts have caveolae
In combination, our immunoblot, RT-PCR, and immunofluorescence analysis revealed that osteoblasts express caveolin with the biochemical properties described in other cell types that have morphological caveolae. Those experiments suggested that osteoblasts also might have morphological caveolae. To investigate this possibility, hFOB and MC3T3 cells were subjected to transmission EM analysis. Both hFOB (Fig. 5) and MC3T3 cells (Fig. 6) had abundant 50- to 100-nm noncoated, flasklike plasmalemmal invaginations. Analysis of the number of caveolae in osteoblasts (Table 1) indicated that there are approximately 13 caveolae per 10 μm of plasma membrane in EM thin sections of MC3T3 cells, whereas there are approximately 7.5 caveolae per 10 μm of plasma membrane in hFOB cells. In both cell lines the size of the caveolae (Table 1) was consistent with the 50- to 100-nm size of caveolae determined from other cell varieties.(11,13) In many sections, larger and distinctively different coated invaginations (e.g., clathrin-coated pits) were observed nearby the more abundant noncoated invaginations (caveolae; Figs. 5A–5C, 6A, and 6B and Table 1). The noncoated vesicles were not observed in internal cell membranes (e.g., nuclear membrane), although other structures such as nuclear pores could be observed (data not shown). Some of the flask-shaped invaginations were found to have a striated appearance (Fig. 5B), which is thought to be caused by oligomerized caveolin.(12)
Table Table 1.. Transmission EM Analysis of Osteoblasts
Osteoblasts are essential regulators of both bone mineralization and osteoclast function and must receive and interpret a wide variety of extracellular stimuli. These signals include those originating from endocrine hormones, cell contact with bone matrix and neighboring cell types, soluble signals from osteoclasts, and, likely, biomechanical stress. It is essential to understand the organization of signaling components (cytoplasmic messengers, receptors) in the osteoblast cell membrane. Toward this end, we report here that human and murine osteoblasts have caveolae and caveolin-enriched DRMs (CEDRMs).
Caveolae and caveolin
DRMs are insoluble in cold nonionic detergents with low critical micelle concentrations [CMCs; e.g., Triton X-100, nonidet P-40 (NP-40), and (octylphenoxy)polyethoxyethanol (IGEPAL)]. (45–47) The insolubility of these aggregates is caused directly and specifically by the biochemical nature of the lipids involved. (45–47) Although insoluble in cold nonionic detergents with low CMCs, these DRM aggregates are soluble in octylglucoside, a nonionic detergent with high CMCs,(6, 7, 48–50) thus establishing an operationally simple yet distinct biochemical characteristic of a DRM. Because this property stems directly from the lipid foundation of DRM, it is employed as a critical criterion to establish the presence and composition of DRM. A related property of DRM (including caveolae) that stems directly from its lipid foundation is a buoyant density equivalent to 15-20% sucrose. This second fundamental property of DRM also was employed as a diagnostic tool in the analysis of DRMs in osteoblasts. Caveolin (the major structural protein component of caveolae) localizes to DRM and oligomerizes to form caveolae. The detection of caveolin in DRM is an essential aspect of caveolae analysis and is indicative of the presence of caveolae in a cell. Last, caveolae, which were defined as morphological structures in the 1950s(11,13) and literally mean “little caves,” can only truly be defined by the types of EM ultrastructural investigations used here. Although caveolin-enriched membranes that are isolated based on properties such as buoyant density or selective detergent extraction are derived from caveolae, they are more properly referred to as DRMs or CEDRMs. In the absence of EM ultrastructural investigations, caveolae cannot be identified or inferred regardless of the amount of caveolin a cell may express or the quantity of DRMs a cell may possess.
Osteoblasts express caveolin
Our determination that osteoblasts express caveolin is based on three independent methods used in the analyses of human and murine osteoblasts: immunoblotting, RT-PCR, and immunofluorescence. First, caveolin expression was determined by immunoblotting osteoblast cell fractions for the presence of caveolin ( Figs. 1A and 1C). In both human and murine cells, caveolin expression was found to be enriched highly in fractions that were resistant to solubilization by Triton X-100 but were solubilized by octylglucoside. This result indicates that the vast majority of caveolin expressed in both hFOB and MC3T3 cells is contained in membranes that have the same solubility profile of membranes containing caveolin in other cells. Thus, the expectation that caveolin should localize principally to DRMs is supported by our new data. Second, we found that the CEDRMs float in 15-20% sucrose, a characteristic property of caveolae and other similar membrane domains. It also was apparent in our immunoblots that oligomerized, higher molecular mass forms of caveolin were present (Fig. 1B). Although visualization of these caveolin multimers required overexposed autoradiographs due to their relative paucity after SDS-PAGE, their presence directly shows that caveolin multimers, which are thought to be the central structural unit of caveolae, are present in osteoblasts.
Our immunoblot analysis permitted us to identify the expression of the caveolin-1 and −2 isoforms in the DRMs of osteoblasts (Fig. 2A). Because caveolin-1 is expressed typically in concert with caveolin-2,(51,52) this result is not surprising. Although both isoforms were present, it appeared that caveolin-1 was more abundant. Although this result is possibly a reflection of inefficient antibody recognition in our caveolin-2 immunoblots, it also may indicate higher levels of caveolin-1 versus caveolin-2 expression. Because caveolae can be formed by caveolin-1 without caveolin-2,(15) the suggested paucity of caveolin-2 might not affect the number or function of caveolae in these cells. The question of caveolin-1 versus caveolin-2 levels in osteoblasts certainly will require additional investigation before any conclusion can be drawn. The identification of both isoforms was confirmed by RT-PCR analysis of osteoblast cDNAs ( Figs. 2B and 2D). This analysis revealed that both hFOB (Fig. 2B) and MC3T3 cells (Fig. 2D) express caveolin-1 as well as caveolin-2 messenger RNAs (mRNAs). We also found expression of both caveolin-1 and caveolin-2 by PCR analysis of a primary human osteoblast cDNA library (Fig. 2C). Thus, the expression of caveolin in osteoblast cell lines is not an artifact of tissue culture.
Last, our immunofluorescence analysis confirmed that both the MC3T3 and the hFOB cells express caveolin and additionally showed that there are many punctate foci of caveolin, many of which are distributed on the cell surface ( Figs. 3 and 4). This punctate expression of caveolin is expected in cells with caveolae.(12,31) In addition, our immunofluorescence analysis provides evidence that caveolin is expressed by >90% of individual cells in both the human and the murine osteoblast cell lines ( Figs. 3 and 4 and data not shown). This supports the conclusion that caveolin expression is a general feature of osteoblasts and is not restricted to only a few cells.
Caveolae and osteoblasts
The expression of caveolin-1 and −2 in detergent-insoluble membranes and the punctate distribution of caveolin are consistent with the presence of caveolae in osteoblasts. To identify caveolae in osteoblasts, both the human hFOB cells and the murine MC3T3 cells were subjected to EM ultrastructural analysis. This analysis revealed abundant 50- to 100-nm cell-surface invaginations in the surface of osteoblasts with the characteristic size (Table 1), shape, and appearance of caveolae found in other cells ( Figs. 5 and 6). Our analysis also showed that these osteoblasts have more caveolae (7.5-13/10 μm) than previously reported for human fibroblasts (1.7/10 μm(43) but less than previously reported for rat lung microvascular endothelial cells(9) (17-18/10 μm; Table 1). Given that caveolae were defined historically by their unique morphology,(11,13) our EM observations provide essential proof for the presence of caveolae in osteoblasts. Based on the results described previously, we conclude that osteoblasts have caveolae that are similar in size, shape, and composition to the caveolae described in other cell varieties. Interestingly, as part of our continuing analysis of caveolae in osteoblasts, we have noted that changes in growth characteristics of osteoblast cell lines may alter the number of caveolae (e.g., faster growing cells have less caveolae; Keith R. Solomon, unpublished observation, 2000). The analysis of osteoblast growth and differentiation with regard to caveolin expression and caveolae assembly remains part of our ongoing investigations.
Osteoblasts are critical for skeletal maintenance. Their regulatory function is influenced by multiple extracellular stimuli including soluble growth factors, extracellular matrix, and contact with other cells; thus, for the proper maintenance of the skeleton, these various signals must be properly orchestrated both spatially and temporally. Caveolae, in general, are membrane organelles that influence the distribution and function of receptors and inner-leaflet signaling molecules and, as revealed in our accompanying article,(53) function in this capacity within osteoblasts. Continuing the evaluation of caveolae function in osteoblasts certainly will be a critical element in a more complete understanding of osteoblast responses to external stimuli.
The authors thank J.R. Kasser for his support, D. Brown (Beth-Israel Deaconess Medical Center, Boston, MA, U.S.A.) for technical assistance with the EM studies, T. Spelsberg (Mayo Clinic, Rochester, MN, U.S.A.) for providing the immortalized hFOB cell line, and J.E. Schnitzer for coining the term “G domains.” This work was supported by grants to P.V.H. from the National Institutes of Health (R01-AR44046) and to K.R.S. from the National Institutes of Health (National Research Service Award-HL09984).