US 5,624,801 AGrant
Methods of Identifying Human Osteoclast-Specific and Related Genes
Issue Date:1997-04-29
•3 Claims
•1 Drawing Sheets
Abstract
The present invention relates to a method of identifying DNA encoding an osteoclast-specific or -related gene product. The method comprises hybridizing DNA with a stromal cell+, osteoclast+probe, or with cDNA or mRNA from an osteoclastoma, and with a stromal cell+, osteoclast-probe, or with cDNA or mRNA from stromal cells; and identifying DNA which hybridizes to the stromal cell+, osteoclast+probe, or to cDNA or mRNA from osteoclastoma, but not to the stromal cell+, osteoclast-probe, or to cDNA or mRNA from stromal cells.
Metadata
Assignee
- Forsyth Dental Infirmary for Children
Inventors
- Philip Stashenko
- Yi-Ping Li
- Anne L. Wucherpfennig
Application Information
Application Number:US 4573048
Filing Date:1995-06-01
Priority Date:1993-04-06
Art Unit:187
Classifications
IPC:
C07H 2102C07H 2104C12Q 168
Field of Search:
4355369356;320.1;252.323.177;78
Patent Drawings (1 sheets)
Description
Background of the Invention
Excessive bone resorption by osteoclasts contributes to the pathology of many human diseases including arthritis, osteoporosis, periodontitis, and hypercalcemia of malignancy. During resorption, osteoclasts remove both the mineral and organic components of bone (Blair, H. C., et al., J. Cell Biol. 102:1164 (1986)). The mineral phase is solubilized by acidification of the sub-osteoclastic lacuna, thus allowing dissolution of hydroxyapatite (Vaes, G., Clin. Orthop. Relat. 231:239 (1988)). However, the mechanism(s) by which type I collagen, the major structural protein of bone, is degraded remains controversial. In addition, the regulation of osteoclastic activity is only partly understood. The lack of information concerning osteoclast function is due in part to the fact that these cells are extremely difficult to isolate as pure populations in large numbers. Furthermore, there are no osteoclastic cell lines available. An approach to studying osteoclast function that permits the identification of heretofore unknown osteoclast-specific or -related genes and gene products would allow identification of genes and gene products that are involved in the resorption of bone and in the regulation of osteoclastic activity. Therefore, identification of osteclast-specific or -related genes or gene products would prove useful in developing therapeutic strategies for the treatment of disorders involving aberrant bone resorption.
Summary of the Invention
The present invention relates to isolated DNA sequences encoding all or a portion of osteoclast-specific or -related gene products. The present invention further relates to DNA constructs capable of replicating DNA encoding osteoclast-specific or -related gene products. In another embodiment, the invention relates to a DNA construct capable of directing expression of all or a portion of the osteoclast-specific or -related gene product in a host cell.
Also encompassed by the present invention are prokaryotic or eukaryotic cells transformed or transfected with a DNA construct encoding all or a portion of an osteoclast-specific or -related gene product. According to a particular embodiment, these cells are capable of replicating the DNA construct comprising the DNA encoding the osteoclast-specific or -related gene product, and, optionally, are capable of expressing the osteoclast-specific or -related gene product. Also claimed are antibodies raised against osteoclast-specific or -related gene products, or portions of these gene products.
The present invention further embraces a method of identifying osteoclast-specific or -related DNA sequences and DNA sequences identified in this manner. In one embodiment, cDNA encoding osteoclast is identified as follows: First, human giant cell tumor of the bone was used to 1) construct a cDNA library; 2) produce .sup.32 P-labelled cDNA to use as a stromal cell.sup.+, osteoclast.sup.30 probe, and 3) produce (by culturing) a stromal cell population lacking osteoclasts. The presence of osteoclasts in the giant cell tumor was confirmed by histological staining for the osteoclast marker, type 5 tartrate-resistant acid phosphatase (TRAP) and with the use of monoclonal antibody reagents.
The stromal cell population lacking osteoclasts was produced by dissociating cells of a giant cell tumor, then growing and passaging the cells in tissue culture until the cell population was homogeneous and appeared fibroblastic. The cultured stromal cell population did not contain osteoclasts. The cultured stromal cells were then used to produce a stromal cell.sup.+, osteoclast.sup.-32 P-labelled cDNA probe.
The cDNA library produced from the giant cell tumor of the bone was then screened in duplicate for hybridization to the cDNA probes: one screen was performed with the giant cell tumor cDNA probe (stromal cell.sup.+, osteoclast.sup.+), while a duplicate screen was performed using the cultured stromal cell cDNA probe (stromal cell.sup.+ osteoclast.sup.-) Hybridization to a stromal.sup.+, osteoclast.sup.- probe, accompanied by failure to hybridize to a stromal.sup.+, osteoclast.sup.31 probe indicated that a clone contained nucleic acid sequences specifically expressed by osteoclasts.
In another embodiment, genomic DNA encoding osteoclast -specific or -related gene products is identified through known hybridization techniques or amplification techniques. In one embodiment, the present invention relates to a method of identifying DNA encoding an osteoclast-specific or -related protein, or gene product, by screening a cDNA library or a genomic DNA library with a DNA probe comprising one or more sequences selected from the group consisting of the DNA sequences set out in Table I (SEQ ID NOs: 1-32). Finally, the present invention relates to an osteoclast-specific or -related protein encoded by a nucleotide sequence comprising a DNA sequence selected from the group consisting of the sequences set out in Table I, or their complementary strands.
Brief Description of the Figure
The Figure shows the cDNA sequence (SEQ ID NO: 33) of human gelatinase B, and highlights those portions of the sequence represented by the osteoclast-specific or -related cDNA clones of the present invention.
Detailed Description of the Invention
As described herein, Applicant has identified osteoclast-specific or osteoclast-related nucleic acid sequences. These sequences were identified as follows: Human giant cell tumor of the bone was used to 1) construct a cDNA library; 2) produce .sup.32 P-labelled cDNA to use as a stromal cell.sup.+, osteoclast.sup.+ probe, and 3) produce (by culturing) a stromal cell population lacking osteoclasts. The presence of osteclasts in the giant cell tumor was confirmed by histological staining for the osteoclast marker, type 5 acid phosphatase (TRAP). In addition, monoclonal antibody reagents were used to characterize the multinucleated cells in the giant cell tumor, which cells were found to have a phenotype distinct from macrophages and consistent with osteoclasts.
The stromal cell population lacking osteoclasts was produced by dissociating cells of a giant cell tumor, then growing the cells in tissue culture for at least five passages. After five passages the cultured cell population was homogeneous and appeared fibroblastic. The cultured population contained no multinucleated cells at this point, tested negative for type 5 acid phosphatase, and tested variably alkaline phosphatase positive. That is, the cultured stromal cell population did not contain osteoclasts. The cultured stromal cells were then used to produce a stromal cell.sup.+, osteoclast.sup.-32 P-labelled cDNA probe.
The cDNA library produced from the giant cell tumor of the bone was then screened in duplicate for hybridization to the cDNA probes: one screen was performed with the giant cell tumor cDNA probe (stromal cell.sup.+, osteoclast.sup.+), while a duplicate screen was performed using the cultured stromal cell cDNA probe (stromal cell.sup.+, osteoclast.sup.-). Clones that hybridized to the giant cell tumor cDNA probe (stromal.sup.+, osteoclast.sup.+), but not to the stromal cell cDNA probe (stromal.sup.+, osteoclast.sup.-), were assumed to contain nucleic acid sequences specifically expressed by osteoclasts.
As a result of the differential screen described herein, DNA specifically expressed in osteoclast cells characterized as described herein was identified. This DNA, and equivalent DNA sequences, is referred to herein as osteoclast-specific or osteoclast-related DNA. Osteoclast-specific or -related DNA of the present invention can be obtained from sources in which it occurs in nature, can be produced recombinantly or synthesized chemically; it can be cDNA, genomic DNA, recombinantly-produced DNA or chemically-produced DNA. An equivalent DNA sequence is one which hybridizes, under standard hybridization conditions, to an osteoclast-specific or -related DNA identified as described herein or to a complement thereof.
Differential screening of a human osteoclastoma cDNA library was performed to identify genes specifically expressed in osteoclasts. Of 12,000 clones screened, 195 clones were identified which are either uniquely expressed in osteoclasts, or are osteoclast-related. These clones were further identified as osteoclast-specific, as evidenced by failure to hybridize to mRNA derived from a variety of unrelated human cell types, including epithelium, fibroblasts, lymphocytes, myelomonocytic cells, osteoblasts, and neuroblastoma cells. Of these, 32 clones contain novel cDNA sequences which were not found in the GenBank database.
A large number of cDNA clones obtained by this procedure were found to represent 92 kDa type IV collagenase (gelatinase B; E.C. 3.4.24.35) as well as tartrate resistant acid phosphatase. In situ hybridization localized mRNAIfor gelatinase B to multinucleated giant cells in human osteoclastomas. Gelatinase B immunoreactivity was demonstrated in giant cells from 8/8 osteoclastomas, osteoclasts in normal bone, and in osteoclasts of Paget's disease by use of a polyclonal antisera raised against a synthetic gelatinase B peptide. In contrast, no immunoreactivity for 72 kDa type IV collagenase (gelatinase A; E.C. 3.4.24.24), which is the product of a separate gene, was detected in osteoclastomas or normal osteoclasts.
The present invention has utility for the production and identification of nucleic acid probes useful for identifying osteoclast-specific or -related DNA. Osteoclast-specific or -related DNA of the present invention can be used to express osteoclast-specific or -related gene products useful in the therapeutic treatment of disorders involving aberrant bone resorption. The osteoclast-specific or -related sequences are also useful for generating peptides which can then be used to produce antibodies useful for identifying osteoclast-specific or -related gene products, or for altering the activity of osteoclast-specific or -related gene products. Such antibodies are referred to as osteoclast-specific antibodies. Osteoclast-specific antibodies are also useful for identifying osteoclasts. Finally, osteoclast -specific or -related DNA sequences of the present invention are useful in gene therapy. For example, they can be used to alter the expression in osteoclasts of an aberrant osteoclast -specific or -related gene product or to correct aberrant expression of an osteoclast-specific or -related gene product. The sequences described herein can further be used to cause osteoclast-specific or -related gene expression in cells in which such expression does not ordinarily occur, i.e., in cells which are not osteoclasts.
Example 1
Osteclast cDNA Library Construction
Messenger RNA (mRNA) obtained from a human osteoclastoma (`giant cell tumor of bone`), was used to construct an osteoclastoma cDNA library. Osteoclastomas are actively bone resorptive tumors, but are usually non-metastatic. In cryostat sections, osteoclastomas consist of .about.30% multinucleated cells positive for tartrate resistant acid phosphatase (TRAP), a widely utilized phenotypic marker specific in vivo for osteoclasts (Minkin, Calcif. Tissue Int. 34:285-290 (1982)). The remaining cells are uncharacterized `stromal` cells, a mixture of cell types with fibroblastic/mesenchymal morphology. Although it has not yet been definitively shown, it is generally held that the osteoclasts in these tumors are non-transformed, and are activated to resorb bone in vivo by substance(s) produced by the stromal cell element.
Monoclonal antibody reagents were used to partially characterize the surface phenotype of the multinucleated cells in the giant cell tumors of long bone. In frozen sections, all multinucleated cells expressed CD68, which has previously been reported to define an antigen specific for both osteoclasts and macrophages (Horton, M. A. and M. H. Helfrich, In Biology and Physiology of the Osteoclast, B. R. Rifkin and C. V. Gay, editors, CRC Press, Inc. Boca Raton, Fla., 33-54 (1992)). In contrast, no staining of giant cells was observed for CD11b or CD14 surface antigens, which are present on monocyte/macrophages and granulocytes (Arnaout, M. A. et al. J. Cell. Physiol. 137:305 (1988); Haziot, A. et al. J. Immunol. 141:547 (1988)). Cytocentrifuge preparations of human peripheral blood monocytes were positive for CD68, CD11b, and CD14. These results demonstrate that the multinucleated giant cells of osteoclastomas have a phenotype which is distinct from that of macrophages, and which is consistent with that of osteoclasts.
Osteoclastoma tissue was snap frozen in liquid nitrogen and used to prepare poly A.sup.+ mRNA according to standard methods. cDNA cloning into a pcDNAII vector was carried out using a commercially-available kit (Librarian, InVitrogen). Approximately 2.6.times.10.sup.6 clones were obtained, >95% of which contained inserts of an average length 0.6 kB.
Example 2
Stromal Cell mRNA Preparation
A portion of each osteoclastoma was snap frozen in liquid nitrogen for mRNA preparation. The remainder of the tumor was dissociated using brief trypsinization and mechanical disaggregation, and placed into tissue culture. These cells were expanded in Dulbecco's MEM (high glucose, Sigma) supplemented with 10% newborn calf serum (MA Bioproducts), gentamycin (0.5 mg/ml), 1-glutamine (2 mM) and non-essential amino acids (0.1 mM) (Gibco). The stromal cell population was passaged at least five times, after which it showed a homogenous, fibroblastic looking cell population that contained no multinucleated cells. The stromal cells were mononuclear, tested negative for acid phosphatase, and tested variably alkaline phosphatase positive. These findings indicate that propagated stromal cells (i.e., stromal cells that are passaged in culture) are non-osteoclastic and non-activated.
Example 3
Identification of DNA Encoding Osteoclastoma-Specific or -Related Gene Products by Differential Screening of an osteoclastoma cDNA Library
A total of 12,000 clones drawn from the osteoclastoma cDNA library were screened by differential hybridization, using mixed .sup.32 p labelled cDNA probes derived from (1) giant cell tumor mRNA (stromal cell.sup.+, OC.sup.+) and (2) mRNA from stromal cells (stromal cell.sup.+, OC.sup.-) cultivated from the same tumor. The probes were labelled with .sup.32 [P]dCTP by random priming to an activity of .about.10.sup.9 CPM/.mu.g. Of these 12,000 clones, 195 gave a positive hybridization signal with giant cell (i.e., osteoclast and stromal cell) mRNA, but not with stromal cell mRNA. Additionally, these clones failed to hybridize to cDNA produced from mRNA derived from a variety of unrelated human cell types including epithelial cells, fibroblasts, lymphocytes, myelomonocytic cells, osteoblasts, and neuroblastoma cells. The failure of these clones to hybridize to cDNA produced from mRNA derived from other cell types supports the conclusion that these clones are either uniquely expressed in osteoclasts, or are osteoclast-related.
The osteoclast (OC) cDNA library was screened for differential hybridization to OC cDNA (stromal cell.sup.+, OC.sup.+) and stromal cell cDNA (stromal cell.sup.+, OC.sup.31 ) as follows:
NYTRAN filters (Schleicher & Schuell) were placed on agar plates containing growth medium and ampicillin. Individual bacterial colonies from the OC library were randomly picked and transferred, in triplicate, onto filters with preruled grids and then onto a master agar plate. Up to 200 colonies were inoculated onto a single 90-mm filter/plate using these techniques. The plates were inverted and incubated at 37.degree. C. until the bacterial inoculates had grown (on the filter) to a diameter of 0.5-1.0 mm.
The colonies were then lysed, and the DNA bound to the filters by first placing the filters on top of two pieces of Whatman 3 MM paper saturated with 0.5N NaOH for 5 minutes. The filters were neutralized by placing on two pieces of Whatman 3 MM paper saturated with 1M Tris-HCL, pH 8.0 for 3-5 minutes. Neutralization was followed by incubation on another set of Whatman 3 MM papers saturated with 1M Tris-HCL, pH 8.0/1.5 M NaCl for 3-5 minutes. The filters were then washed briefly in 2.times. SSC.
DNA was immobilized on the filters by baking the filters at 80.degree. C. for 30 minutes. Filters were best used immediately, but they could be stored for up to one week in a vacuum jar at room temperature.
Filters were prehybridized in 5-8 ml of hybridization solution per filter, for 2-4 hours in a heat sealable bag. An additional 2 ml of solution was added for each additional filter added to the hybridization bag. The hybridization buffer consisted of 5.times. SSC, 5.times. Denhardt's solution, 1% SDS and 100 .mu.g/ml denatured heterologous DNA.
Prior to hybridization, labeled probe was denatured by heating in 1.times. SSC for 5 minutes at 100.degree. C., then immediately chilled on ice. Denatured probe was added to the filters in hybridization solution, and the filters hybridized with continuous agitation for 12-20 hours at 65.degree. C.
After hybridization, the filters were washed in 2.times. SSC/0.2% SDS at 50.degree.-60.degree. C. for 30 minutes, followed by washing in 0.2.times. SSC/0.2% SDS at 60.degree. C. for 60 minutes.
The filters were then air dried and autoradiographed using an intensifying screen at -70.degree. C. overnight.
Example 4
DNA SeqUencing of. Selected Clones
Clones reactive with the mixed tumor probe, but unreactive with the stromal cell probe, are expected to contain either osteoclast-related, or in vivo `activated` stromal-cell-related gene products. One hundred and forty-four cDNA clones that hybridized to tumor cell cDNA, but not to stromal cell cDNA, were sequenced by the dideoxy chain termination method of Sanger et al. (Sanger F., et al. Proc. Natl. Acad. Sci. USA 74:5463 (1977)) using sequenase (US Biochemical). The DNASIS (Hitatchi) program was used to carry out sequence analysis and a homology search in the GenBank/EMBL database.
Fourteen of the 195 tumor.sup.+ stromal.sup.- clones were identified as containing inserts with a sequence identical to the osteoclast marker, type 5 tartrate-resistant acid phosphatase (TRAP) (GenBank accession number J04430 M19534). The high representation of TRAP positive clones also indicates the effectiveness of the screening procedure in enriching for clones which contain osteoclast-specific or related cDNA sequences.
Interestingly, an even larger proportion of the tumor.sup.+ stromal.sup.- clones (77/195; 39.5%) were identified as human gelatinase B (macrophage-derived gelatinase) (Wilhelm, S. M. J. Biol. Chem. 264:17213 (1989)), again indicating high expression of this enzyme by osteoclasts. Twenty-five of the gelatinase B clones were identified by dideoxy sequence analysis; all 25 showed 100% sequence homology to the published gelatinase B sequence (Genbank accession number J05070). The portions of the gelatinase B cDNA sequence covered by these clones is shown in the Figure (SEQ ID NO: 33). An additional 52 gelatinase B clones were identified by reactivity with a .sup.32 P-labelled probe for gelatinase B.
Thirteen of the sequenced clones yielded no readable sequence. A DNASIS search of GenBank/EMBL databases revealed that, of the remaining 91 clones, 32 clones contain novel sequences which have not yet been reported in the databases or in the literature. These partial sequences are presented in Table I. Note that three of these sequences were repeats, indicating fairly frequent representation of mRNA related to this sequence. The repeat sequences are indicated by .sup.a,b, superscripts (Clones 198B, 223B and 32C of Table I).
Sequence analysis of the OC.sup.+ stromal cell.sup.- cloned DNA sequences revealed, in addition to the novel sequences, a number of previously-described genes. The known genes identified (including type 5 acid phosphatase, gelatinase B, cystatin C (13 clones), Alu repeat sequences (11 clones), creatnine kinase (6 clones) and others) are summarized in Table II. In situ hybridization (described below) directly demonstrated that gelatinase B mRNA is expressed in multinucleated osteoclasts and not in stromal cells. Although gelatinase B is a well-characterized protease, its expression at high levels in osteoclasts has not been previously described. The expression in osteoclasts of cystatin C, a cysteine protease inhibitor, is also unexpected. This finding has not yet been confirmed by in situ hybridization. Taken together, these results demonstrate that most of these identified genes are osteoclast-expressed, thereby confirming the effectiveness of the differential screening strategy for identifying DNA encoding osteoclast-specific or -related gene products. Therefore, novel genes identified by this method have a high probability of being OC-specific or-related.
In addition, a minority of the genes identified by this screen are probably not expressed by OCs (Table II). For example, type III collagen (6 clones), collagen type I (1 clone), dermatansulfate (1 clone), and type VI collagen (1 clone) are more likely to originate from the stromal cells or from osteoblastic cells which are present in the tumor. These cDNA sequences survive the differential screening process either because the cells which produce them in the tumor in vivo die out during the stromal cell propagation phase, or because they stop producing their product in vitro. These clones do not constitute more than 5-10% of the all sequences selected by differential hybridization.
Example 5
In situ Hybridization of OC-Expressed Genes
In situ hybridization was performed using probes derived from novel cloned sequences in order to determine whether the novel putative OC-specific or -related genes are differentially expressed in osteoclasts (and not expressed in the stromal cells) of human giant cell tumors. Initially, in situ hybridization was performed using antisense (positive)and sense (negative control) cRNA probes against human type IV collagenase/gelatinase B labelled with .sup.35 S-UTP.
A thin section of human giant cell tumor reacted with the antisense probe resulted in intense labelling of all OCs, as indicated by the deposition of silver grains over these cells, but failed to label the stromal cell elements. In contrast, only minimal background labelling was observed with the sense (negative control) probe. This result confirmed that gelatinase B is expressed in human OCs.
In situ hybridization was then carried out using cRNA probes derived from 11/32 novel genes, labelled with digoxigenin UTP according to known methods.
The results of this analysis are summarized in Table III. Clones 28B, 118B, 140B, 198B, and 212B all gave positive reactions with OCs in frozen sections of a giant cell tumor, as did the positive control gelatinase B. These novel clones therefore are expressed in OCs and fulfill all criteria for OC-relatedness. 198B is repeated three times, indicating relatively high expression. Clones 4B, 37B, 88C and 98B produced positive reactions with the tumor tissue; however the signal was not well-localized to OCs. These clones are therefore not likely to be useful and are eliminated from further consideration. Clones 86B and 87B failed to give a positive reaction with any cell type, possibly indicating very low level expression. This group of clones could still be useful but may be difficult to study further. The results of this analysis show that 5/11 novel genes are expressed in OCs, indicating that .about.50% of novel sequences likely to be OC-related.
To generate probes for the in situ hybridizations, cDNA derived from novel cloned osteoclast-specific or -related cDNA was subcloned into a BlueScript II SK(-) vector. The orientation of cloned inserts was determined by restriction analysis of subclones. The T7 and T3 promoters in the BlueScriptII vector was used to generate .sup.35 S-labelled (.sup.35 S-UTP, 850 Ci/mmol, Amersham, Arlington Heights, Ill.), or UTP digoxygenin labelled cRNA probes.
In situ hybridization was carried out on 7 micron cryostat sections of a human osteoclastoma as described previously (Chang, L.-C. et al. Cancer Res. 49:6700 (1989)). Briefly, tissue was fixed in 4% paraformaldehyde and embedded in OCT (Miles Inc., Kankakee, Ill. ). The sections were rehydrated, postfixed in 4% paraformaldehyde, washed, and pretreated with 10 mM DTT, 10 mM iodoacetamide, 10 mM N-ethylmaleimide and 0.1 triethanolamine-HCL. Prehybridization was done with 50% deionized formamide, 10 mM Tris-HCl, pH 7.0, 1.times. Denhardt's, 500 mg/ml tRNA, 80 mg/ml salmon sperm DNA, 0.3M NaCl 1 mM EDTA, and 100 mM DTT at 45.degree. C. for 2 hours. Fresh hybridization solution containing 10% dextran sulfate and 1.5 ng/ml .sup.35 S-labelled or digoxygenin labelled RNA probe was applied after heat denaturation. Sections were coverslipped and then incubated in a moistened chamber at 45.degree.-50.degree. C. overnight. Hybridized sections were washed four times with 50% formamide, 2.times. SSC, containing 10 mM DTT and 0.5% Triton X-100 at 45.degree. C. Sections were treated with RNase A and RNase T1 to digest single-stranded RNA, washed four times in 2.times. SSC/10 mM DTT.
In order to detect .sup.35 S-labelling by autoradiography, slides were dehydrated, dried, and coated with Kodak NTB-2 emulsion. The duplicate slides were split, and each set was placed in a black box with desiccant, sealed, and incubated at 4.degree. C. for 2 days. The slides were developed (4 minutes) and fixed (5 minutes) using Kodak developer D19 and Kodak fixer. Hematoxylin and eosin were used as counter-stains.
In order to detect digoxygenin-labelled probes, a Nucleic Acid Detection Kit (Boehringer-Mannheim, Cat. #1175041) was used. Slides were washed in Buffer 1 consisting of 100 mM Tris/150 mM NaCl, pH7.5, for 1 minute. 100 .mu.l Buffer 2 was added (made by adding 2 mg/ml blocking reagent as provided by the manufacturer) in Buffer 1 to each slide. The slides were placed on a shaker and gently swirled at 20.degree. C.
Antibody solutions were diluted 1:100 with Buffer 2 (as provided by the manufacturer). 100 .mu.l of diluted antibody solution was applied to the slides and the slides were then incubated in a chamber for 1 hour at room temperature. The slides were monitored to avoid drying. After incubation with antibody solution, slides were washed in Buffer 1 for 10 minutes, then washed in Buffer 3 containing 2 mM levamisole for 2 minutes.
After washing, 100 .mu.l color solution was added to the slides. Color solution consisted of nitroblue/tetrazolium salt (NBT) (1:225 dilution) 4.5 .mu.l, 5-bromo-4-chloro-3-indolyl phosphate (1:285 dilution) 3.5 .mu.l levamisole 0.2 mg in Buffer 3 (as provided by the manufacturer) in a total volume of 1 ml. Color solution was prepared immediately before use.
After adding the color solution, the slides were placed in a dark, humidified chamber at 20.degree. C. for 2-5 hours and monitored for color development. The color reaction was stopped by rinsing slides in TE Buffer.
The slides were stained for 60 seconds in 0.25% methyl green, washed with tap water, then mounted with water-based Permount (Fisher).
Example 6
Immunohistochemistry
Immunohistochemical staining was performed on frozen and paraffin embedded tissues as well as on cytospin preparations (see Table IV). The following antibodies were used: polyclonal rabbit anti-human gelatinase antibodies; Ab110 for gelatinase B; monoclonal mouse anti-human CD68 antibody (clone KP1) (DAKO, Denmark); (anti-CD11b) and Mo2 (anti-CD14) derived from ATCC cell lines HB CRL 8026 and TIB 228/HB44. The anti-human gelatinase B antibody Ab11O was raised against a synthetic peptide with the amino acid sequence EALMYPMYRFTEGPPLHK (SEQ ID NO: 34), which is specific for human gelatinase B (Corcoran, M. L. et al. J. Biol. Chem. 267:515 (1992)).
Detection of the immunohistochemical staining was achieved by using a goat anti-rabbit glucose oxidase kit (Vector Laboratories, Burlingame Calif.) according to the manufacturer's directions. Briefly, the sections were rehydrated and pretested with either acetone or 0.1% trypsin. Normal goat serum was used to block nonspecific binding. Incubation with the primary antibody for 2 hours or overnight (Ab110: 1/500 dilution) was followed by either a glucose oxidase labeled secondary anti-rabbit serum, or, in the case of the mouse monoclonal antibodies, were reacted with purified rabbit anti-mouse Ig before incubation with the secondary antibody.
Paraffin embedded and frozen sections from osteoclastomas (GCT) were reacted with a rabbit antiserum against gelatinase B (antibody 110) (Corcoran, M. L. et al. J. Biol. Chem. 267:515 (1992)), followed by color development with glucose oxidase linked reagents. The osteoclasts of a giant cell tumor were uniformly strongly positive for gelatinase B, whereas the stromal cells were unreactive. Control sections reacted with rabbit preimmune serum were negative. Identical findings were obtained for all 8 long bone giant cell tumors tested (Table IV). The osteoclasts present in three out of four central giant cell granulomas (GCG) of the mandible were also positive for gelatinase B expression. These neoplasms are similar but not identical to the long bone giant cell tumors, apart from their location in the jaws (Shafer, W. G. et al., Textbook of Oral Pathology, W. B. Saunders Company, Philadelphia, pp. 144-149 (1983)). In contrast, the multinucleated cells from a peripheral giant cell tumor, which is a generally non-resorptive tumor of oral soft tissue, were unreactive with antibody 110 (Shafer, W. G. et al., Textbook of Oral Pathology, W. B. Saunders Company, Philadelphia, pp. 144-149 (1983)).
Antibody 110 was also utilized to assess the presence of gelatinase B in normal bone (n=3) and in Paget's disease, in which there is elevated bone remodeling and increased osteoclastic activity. Strong staining for gelatinase B was observed in osteoclasts both in normal bone (mandible of a 2 year old), and in Paget's disease. Staining was again absent in controls incubated with preimmune serum. Osteoblasts did not stain in any of the tissue sections, indicating that gelatinase B expression is limited to osteoclasts in bone. Finally, peripheral blood monocytes were also reactive with antibody 110 (Table IV).
Equivalents
Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments described herein. Such equivalents are intended to be encompassed by the following claims.
Claims
We claim:
1. A method of identifying DNA encoding an osteoclast-specific or -related gene product, comprising the steps of: a) contacting a sample containing DNA with a stromal cell+, osteoclast+probe under conditions appropriate for hybridization of the probe to the DNA; b) contacting said sample with a stromal cell+, osteoclast-probe under conditions appropriate for hybridization of the stromal cell+, osteoclast-probe to the DNA; and c) identifying DNA which hybridizes to the stromal cell+, osteoclast+probe, but does not hybridize to the stromal cell+, osteoclast-probe, thereby identifying DNA encoding an osteoclast-specific or -related gene product.
2. A method of claim 1 wherein the probe is a nucleic acid probe.
3. A method of identifying DNA encoding an osteoclast-specific or -related gene product, comprising the steps of: a) contacting a sample containing DNA with cDNA or mRNA from an osteoclastoma under conditions appropriate for hybridization of the DNA with the cDNA or mRNA from an osteoclastoma; b) contacting said sample with cDNA or mRNA from stromal cells cultured from an osteoclastoma under conditions appropriate for hybridization of the DNA with the cDNA or mRNA from said stromal cells; and c) identifying DNA which hybridizes with cDNA or mRNA from an osteoclastoma but does not hybridize with cDNA or mRNA from the stromal cells cultured from an osteoclastoma, thereby identifying DNA encoding an osteoclast-specific or -related gene product.
Non-Patent Literature (11)
- Blair, Harry C., et al., "Extracellular-matrix degradation at acid pH Avian Osteoclast acid collagenase isolation and characterization," Biochemical Journal 290(3):873-884 (15 Mar. 1993).
- Tezuka, Ken-Ichi, et al., "Identification of osteopontin in isolated rabbit osteoclasts," Biochemical and Biophysical Research Communications 186(2):911-917 (31 Jul. 1992).
- Tezuka, Ken-Ichi, et al., "Molecular cloning of a possible cysteine proteinase predominantly expressed in osteoclasts'," Journal of Biological Chemistry, 269(2): 1106-1109, (14 Jan. 1994).
- Horton, M.A. et al., "Monoclonal Antibodies ot Osteoclastomas (Giant Cell Bone Tumors): Definition ot Osteoclast-specific cellular Antigens," Cancer Research, 45:5663-5669 (Nov. 1985).
- Davies, J. et al., "The Osteoclast Functional Antigen, Implicated in the Regulation of Bone Resoption, Is Biochemically Related to the Vitronectin Receptor," The J. of Cell Biology, 109:1817-1826 (Oct. 1989).
- Hayman, A.R. et al., "Purification and Characterization of a Tartrate-resistant Acid Phosphatase form Human Osteoclastomas," Biochem. J., 261:601-609 (1989).
- Sandberg, M. et al., "Localization of the Expression of Types I, III, and IV Collagen, TFG-.beta.1 and c-fos Genes in Developing Human Calvarial Bones," Developmental Biology, 130:324-334 (1988).
- Sandberg, M. et al., "Enhanced Expression of the TGF-.beta.and c-fos mRNAs in the growth Plates of Developing Human Long Bones," Development, 102:461-470 (1988).
- Ek-Rylander, B. et al., "Cloning, Sequence, and Developmental Expression of a Type 5, Tartrate-Resistant, Acid Phosphatase of Rat Bone," The J. Biological Chem., 266:24684-24689 (Dec. 25, 1991).
- Ek-Rylander et al., J. Biol. Chem. 266(36): 24684-24689, 1991.
- Tezuka et al., Biochem. Biophys. Reseach Comm. 186(2) 911-917, 1992.