CA2064878A1 - Bone-specific protein - Google Patents

Abstract

Proteins, having sequence (I) where R1 is Ala ou Thr, R2 is Asn or His, R3 is Thr or Ser, R4 is Ala or Thr, R5 is Ile or Val, R6 is Val or Ile and R7 is Phe or Ile, is disclosed.
These proteins are present in osteoblasts and osteocytes and may be used as markers to follow bone and/or cartilage metabolism.

Description

~ W091/027~4 PCr/US90tO~745 . . .
87`~
1 ' . . ':
., "
, .~.`, ,. ~ .. `~.' BON~-SPECIFIC PROTEIN

~ De~

,jj!; ~ Technical Field ~ ~
The present~invention relates to protein~ ;
chemistry. More~particularly, it relates~a protein that ~.
~ occurs specifically in~bone, antibodies to such proteins, Q~ and immunoassays~for~detectLng~suc~h proteins~

15 ~ ackq~ound~Art O~hers~hàve fractionated~bone~in an attempt to identify proteins which can stimulate t~he formation of new bone when~placed~ln~contact wlth~living systems. (Urist, ,.~;
M.~R., Clin~Orthop~(196~) 56:37;~Science (1965) 150:893; `~;

2~0 ~Redd~ A~ H.~,~ ét~aI.;,~Proc Natl Acad Sci (USA);(1972) `.
69`~ 601.;) ;A "bonè~morphogenic protein~ (BMPj was ~ :
~extracted from demineralized~bone~using urea or guanidine ~-hydro~chlorlde a~nd~reprec;ip1tated~;~according to ~the~
dlscl sures~ln~U~ Patents~ os.~4;, 9~4~,753~a~nd 4,455,256 2`-5~to~UrIst~ Urist subseque~ntly~reported (Urist, M. R., Clin Orthop~Rel~Res`~(`1982)~ 162:219)~that ion exchange pùrif1catLon~ of~this crude proteln mlxture yielded an activity which was~unadsorbed to carboxymethyl cellulose resin (CMC) at~pH`4.8. Urist's reports in Science (1983) ..
220:680-685, Proc Natl Acad Sciènce (USA) (1984) 81:371 375~, and~U.S. Pat. No. 4,789,732 describe`BMPs having mGlecular weights of 17,500 and~l8,500 daltons. Urist~s patent publicat~ion,~EPA Publication No. 0212474,;describes BMP fragments of 4,000~to 7~,000 daltons obtained by limited proteolysis of BMP.

~ WO91/027~ PCT/U~9OtO4745 ~ ~Q~ 2-u.S. Patent No. 4,608,199 describes a bone- ' i derived protein of 30,000-32,000 daltons. The protein is described as b~ing water soluble and having no affinity ; for concanavalin A tConAj.
WO 88/00205 reports four proteins, designated ,~ BMP-1, BMP-2 Class I, BMP-2 Class II and BMP-3, that are :~! alleged to have osteogenic activity by themselves or in combination with other factors. Sequences are provided for each of these proteins which show no homology to the 10 se~uence (see below) of the protein of the present ~ -invention.
Commonly owned U.S. 4,434,094 reported the partial purification of a bone -generation-stimulating, ~ :~
bone-derived protein by extraction with chaotropic agents, 1 15 ~ractionation on anion and cation exchange columns, and , ; recover~ of the activity from a fraction adsorbed to CMC ~`~
at pH 4.8. This~new protein fraction was termed "osteogenic fac~tor" (OF) and was characterized as having a molecular weight below about 30,000 daltons.
~;~ ;20 Commonly owned~U.S. Patent No. 4,774,332 describes two proteins that were purified to homogeneity using a purificat1on procedure that is similar in part to that disclosed ~in U~.~S. 4~,434,094. Those two proteins eluted from~CMC~at~about a 150-200 mM NaCl gradient.
25 ~Thes~e two proteins~were originally called cartilage-inducing factor (CIF)~A and CIF B. CIF A was subsequently found to be ident~ic~a~l to a previously identified protein now called transforming growth factor betal (TGF-betal)~
CIF B has been found to be a novel form of TGF-beta and is now known as TGF-beta2. These proteins and homologous proteins exhibiting similar activity are collectively referred to as TGF-beta.
Commonly owned V.S. Patent No. 4,627,982 concerns a partially~purified bone-inducing factor present in the CMC-bound fraction of U.S. 4,434,094 that elutes in the portion of the NaC1 gradient below that in which the : : ~ :
: ~ :
:
' ~ . , ~YO91/02744 P~TJUS90/04745 }.~
~ _3 2~ 7~

major portions of TGF-betal and TGF-beta2 elute (i.e., below about 150 mM NaCl). The present invention relates to the identification of an ingredient of that fraction.

Disclosure of the Invention ~ .
One aspect of the invention is a substantially ~-pure polypeptide that is found in bone and has the ~ -following sequence~

(H2N)-Ala-Lys-Tyr-Asn-Lys-Ile-Lys-Ser-Arg-Gly-Ile-Lys-Ala-Asn-Rl-Phe-Lys-Lys-Leu-R2-Asn-Leu-R3-Phe-Leu-Tyr-Leu-Asp-His-Asn-Ala-Leu-Glu-Ser-Val-Pro-Leu-Asn-Leu-Pro-Glu-Ser-Leu-Arg-Val-Ile-His-Leu-Gln-Phe-Asn-Asn-Ile-R -Ser-Ile-Thr-Asp-Asp-Thr~
Phe-Cys-Lys-AIa-Asn-~sp-Thr-Ser-Tyr-Ile-Arg-Asp-Arg-Ile-Glu-Glu-Ile-Arg-Leu-Glu-Gly-Asn-Pro-R5-R5-Leu-Gly-Lys-His-Pro-Asn-Ser-Phe-Ile-Cys-Leu-Lys-Arg-Leu-Pro-~;~ Ile-Gly-Ser-Tyr-R -(COOH), where Rl is Ala or Thr, R2 is Asn or His, R3 is Thr or Ser, R4 is ~la or~Thr, R5 is Ile or Val, R6 is Val or Ile and R7 is Phe or Ile,- and substantially pure polypeptides tha~t are and substantiaLly homologous thereto.
Deglycosylated analogs of the above-described polypeptides are another aspect of the invention. `:~
Further aspects of the invent.ion are recombinant materials (i.e., recombinant DNA, recombinant vectors, and recombinant cells or microorganisms) and processes for producing ths polypeptides of the invention, antibodies specific tc he polypeptides, and immunoassays for the polypeptides.

, - :
'' ' ~ .

WO91/~2744 PCT/US90/0~745 .~ 4~" 4-; Brief Description of the Drawinqs In the drawings:
Figure l is a flow chart of the process that was used to isolate the bovine species of the bone-specific protein of the invention from deminexalized bovine bone.
Figure 2 is a graph of the optical densities (absorbances at 280 nm) of the gel filtration fractions of ;` the gel filtration fractions of the example (~IC).
Figure 3 is a graph of the optical densities (absorbances at 280 nm) of eluate fractions from the preparative ion exchange chromatography of the example (UD).
Figure 4 is a graph of the opti.cal densities (absorbances at 280 nm) of eluate fractions from the lS cross-linked ConA chromatography step of the example (11E);
Figure 5 is a graph of the optical densities (absorbances at 280 nm) of eluate fractions from the heparin-sepharose chromatography step of the example (~IF);
Figure 6 is a graph of the optical densities (absorbances at 230 nm) of the gradient fractions from the Cl8-RP-HPLC chromatography step of the example (~IG);
Figure 7 is a ~able showing results of amino acid sequencing of the bovine isblate of the invention and locations of the sequenced fragments in the overall sequence.
- Figure~8 is a photograph of an autoradiograph of SDS-PAGE analyses of the purified bovine protein that are described in the example ~1lH) (lanes A and C show glycosylated protein; lanes B and D show enzymatically deglycosylated protein)~
Figure 9 is a schematic diagram illustrating the structure of the bovine gene that encodes the mature bovine species of the protein of the invention. Various restriction sites are indicated.

: ~ ~
WO91/027q4 PCT/US90/~4745 . ~ '.' .
5 - 2 ~ 7,~ ~ ~

Figure 10 is a restriction map of the region of :
the human gene that encodes the human species of the ,~
protein. : ~
Figure 11, in parts A-E, shows the DNA sequence ~ ~ , -` 5 and deduced amino acid sequence of the mature human spe-cies of the protein of the invention~ in comparison with the sequences of~the bovine gene and protein. Part A ,~
shows~the prelimin~ary unconfirmed sequence of a putative amino terminal"~precursor exon of the human gene for the prepropolypeptide. The extent of this exon~has not~been determined. Bases~ of uncertain identity in the DNA
sequence are represented by numbers. The~signal sequence indicated in the figure is putative and was identifie~ by the von Heijne algorithm. Part ~ shows~the sequence of 15~ another precursor~exon,~ designated ~ that is downstream '`~
o the amino~terminal exon and upstream of the exon that '~
encodes the amino~terminal of the mature protein. Parts , .
C-E show the sequences of exons 1-3 which encode the mature protein.~ Asterisks designate points of identity ,, 20~ be~tween~the bovine~and human DNA sequences. Amino acid numbering is~;re;l~ative to the mature sequence with the ,~
first amino acid thereof designated~ . Consensus 3' and ~-5~ exon s~plice~sites~are indicated by YYYYYYY~YYYNYAG and ~:
JAGGTRAGT, respectively.~
25~ Flgure~ 2~ s a~schematlc diagram of the mamm`alian~expres~sion~vector phOIFl8 which contains the , human gene that~encodes the human species of the invention :, -~ prot~ein.

Modes of CarrYinq Out the Invention Isolatlon~ef Proteln~from Bons~
I,t is believed that the protein of the present ~' invention has been highly conserved among mammalian 35 species--i.e., cbrresponding proteins from different :~
mammalian species (herein called ~species analogs") will ~ WO9l/027~ PCT/US90/0474S

2~$ ~ - -6- ~

have substantially homologous amino acid sequences that vary from the bovine or human proteins described hercin, if at all, in one or more amino acid residue additions, deletions or substitutions and/or substantially similar glycosylation patterns. The amino acid sequences of substantially homologous" proteins will usually be at least 50~ identical, more usually at least 80~ identical, and preferably at least 90~ identical to the bovine/human amino acid sequence described herein. Such proteins may be derived from bone or other tissues of diverse mammalian origin or synthesized using recombinant DNA procedures.
The term is intended to include muteins or analogs of the native protein that are altered in manners known in the art, such as by substitution of cysteines with neutral ~uncharged) amino acids to avoid improper disulfide bonding, by substitution or elimination of residues in the asparagine-linked gLycosylation sites of the proteins to alter glycosylation patterns, by substitution of methionines to make the molecules less susceptible to oxidationi by conservative substitution of other residues, by chemical modification of one or more residues, by substitution with nonnatural amino acids or by elimination or alteration of~side-chain sugars. The source of protein prepared by purification from native sources is advantageously porcine or ~ovine long bone because of its ready availability.~
The process for isolating the protein frcm bone is as follows. The bone is first cleaned using mechanical or abrasive techniques, fragmented, and further washed with, for example, dilute aqueous acid preferably at low temperature. The bone is then demineralized by removal of the calcium phosphates in their various forms, usually by extraction wlth stronger acid These techniques are understood in the art, and are disclosed, for example, in U.S. 4,434,094. The resulting preparation, a .
;

~ -. . . . . . . . . . . ` .

:,~ . !`, , W091~02744 PCT/US90/04745 ~g 2~$~ 7~

7- `
:, -demineralized bone, is the starting material for the `
preparation of the protein from native sources.
;~ The lnitial extraction is designed to remove the ~
nonfibrous (e.g., noncollagenous) proteins from the ~ ~ `
demin~aralized bone. This can be done with the use of chaotropic agents such as guanidine hydrochloride ~at least about 4 molar), urea (8 molar) plus salt, or sodium ~` dodecylsulfate~(at least about 1~ by volume) or such other ~`
chaotropic agents as are known in the art (Termine et al., J Biol Chem (1980) 255:9760-0772; and Sajera and Hascall, J Biol Chem (1969) 244:77-87 and 2384-2396). The extrac-tion is preferably carried out at reduced temperatures to ;
reduce the likelihood of digestion or denaturation of the : -extracted protein. A protease inhibitor may be added to the extractant, if desired. The pH of ~he medium depends upon the extractant selected. The process of extraction generally takes on the order of about 4 hr to 1 day.
After extraction, the extractant may be removed ` ~ by suitable means such as dialysis against water, preceded ;
by concentration by ultrafiltration if desired. Salts can also be removed bX controlled electrophoresis, or by molecular sieving, or by any other means known in the art.
It is also preferred ~to maintain a low temperature during ~;
this~ process so as~to minimize denaturation of the ~ `
2~5 ~proteins. Alte~rnatlvely, the extractant chaotropic agent ~`;
need not be removed, but rather the solution need only be concentrated, for example, by ultrafiltration.
The extract, dissolved or redissolved in :
chaotropic agent, is subjected to gel filtration to obtain fractions of molecular weight in the range of about 20,000 to 36,000 daltons. Gel sizing is done using standard techniques, preferably on a Sephacryl S-200 column at room (10C-25C) temperature.
The sized~fractivn is then subjected to ion exchange chromatography using CMC at approximately pH 4.5-5.2 preferably about 4.8, in the presence of a nonionic .

:

WO9l/02744 PCT/~S90/04745 ~!3~ 8- ~-chaotropic agent such as 6M urea. Other cation exchangers may be used, including those derived from polyacrylamide and cross-linked dextran; however cellulosic cation exchangers are preferred. Of course, as in any ion exchange procedure, the solution must be freed of competing ions before application to the column. The protein is adsorbed on the column and is eluted in an increasing salt concentration gradient in the range of about 10 mM to about 150 mM. This fraction is designated CMB-l for convenience.
CMB-l is lyophilized and the dry CMB-1 is dis-solved in aqueous sodium deoxycholate (DOC~, pH 8Ø This solution is affinity chromatographed on an equilihrated column of ConA cross-linked to resin. The ConA-bound material is eluted~from the resin with aqueous DOC
containing a displacement carbohydrate. This fraction is designated CAB-l for convenience.
CAB-l is reequilibrated for heparin-sepharose chromatography by desalting on a GH-25 column equlllbrated on heparin-sepharose buffer, 6M urea, 0.lM NaCl, 50 mM
Tris-HCl pH 7.2. The desalted fraction is loaded onto a heparin-sepharose column. After washing, bound material is eluted from the column using the same buffer at a 0.5M
NaCl salt~concentration. The resulting eluate is ;;25 des~ignated ~HSB-l for convenience.
HSB-I is dlluted and adjusted to pH 2 and loaded onto a C18-RP-HPLC column. Bound proteins were gradient eluted from the column using a solvent consisting of 90%
acetonitrile in 0.1~ aqueous TPA (solvent B). The protein of the invention slutes at approximately 47-50~ of solvent B (42-45% acetonitril~) by volume.
Proteins eluted by ths C18 chromatography were iodinated by the chloramine-T method. Analysis of the fraction by SDS-PAGE and autoradiography shows a major broad band at 20,000 to 28,000 daltons comprising the , ..
:

.... ~.. , .. , . . ~ .,, .. ; . ... ,.~ . ... . ........... . .. . .... ... . .. . .

` WO~1~02744 PCT/US90~04~45 -9- ~5'~
.. . .:
-- protein. The smearing' of the protein is beli0ved to mainly be the result of heterogeneity in the glycosylation of the molecule or perhaps variable post-translational modification or proteolytic degradation. After enzymatic S or chemical deglycosylation, SDS-PAGE analysis of the protein gives a single band of approximately 10,000 `
` daltons. Reduction of the deglycosylated protein with dithiothreitol does not affect its migration.
Initial amino acid sequence analysis of the :
glycosylated protein so isolated from bovine bone yielded the following internal sequence in the N-terminal portion of the protein~
.. ..
".: ~ .'' . ".
-Lys-Tyr-Asn-Lys-Ile-Lys-Ser-Arg-Gly-Ile-Lys- -15 Ala-Asn-Thr-Phe-Lys-Lys-Leu-His-Asn-Leu-Ser-Phe-X-Tyr-Thr- ;
; Asp-His-Asn-Ala-Leu-Glu- ;~

The initial amino acid (Lys) in the above sequence is nearest to the N-terminal. Initially, the nature of the signal obtained for the residue designated X did not permit this residue to be identified. Re~eated sequencing of the entire peptide and sequencing of oligopeptides generated~from endoproteinase Lys-C (an enzyme that ~`
cl~ea~es proteins at Lys residues) and endoproteinase Glu-C
(a~n enzyme that~cleaves proteins a~ Glu residues) digests have revealed that the above sequence is preceded by an Ala residue~which is the N-terminus, that the residue designated X is Leu, that the second Thr residue (the 26th residue in the above sequence) was incorrect and that this residue is actually a Leu rssidue, and that the isolate consists of a protein oE approximately 106 amino acids.
Figure 7 provides a summary of these sequence analyses.
The symbol CHOI designates a carbohydrate substituent. ;
The symbol COOH~ rèpresents a carboxyl group and designates the carboxy terminus. The first column (on the :, ~
:,,'' -' ~.'. .

WO9~/0~744 PCT/US90/04745 left) provides the sequence analysis of the N-terminal fragment described above. The second, fourth, and sixth columns give the sequences of three major Lys-C fragments of the isolate. The third and fifth columns glve the sequences of two Glu-C fragments.
Subsequent isolation of the gene for this , protein confirmed the sequence shown in Figure 7 with the i sole exception that the deduced sequence lacked the Asp residue at the carboxy terminal. Accordingly, the sequence for the native bovine protein is as follows:

(H2N)-Ala-Lys-Tyr-Asn-Lys-Ile-Lys-Ser-Arg-Gly-Ile-Lys-Ala-Asn-Thr-Phe-Lys-Lys-Leu-His-Asn-Leu-Ser-Phe-Leu-Tyr-Leu-Asp-His-Asn-Ala-Leu-Glu-Ser-Val-Pro-Leu-Asn-Leu-Pro-Glu-Ser-Leu-Arg-Val-Ile-His-Leu-Gln-Phe-Asn-Asn-Ile-Thr-Ser-Ile-Thr-Asp-Asp-Thr-Phe-Cys-Lys-Ala-Asn-Asp-Thr-Ser-Tyr-Ile-Arg-Asp-Arg-Ile-Glu-GIu-Ile-Arg-Leu-Glu-Gly-Asn-Pro-Val-Ile-Leu-Gly-Lys-His-Pro-Asn-Ser-Phe-Ile-Cys-Leu-Lys-Arg-Leu-Pro-Ile-Gly-Ser-Tyr-Ile-(COOH), The sequence of the corresponding human protein was determined by obtaining the human gene uslng DNA
probes based on the~bovlne DNA sequence, sequenclng the human gene~and deducing the amino acid sequence of the protein encoded thereby. The sequence of the human protein was found to be as follows.

(H2N)-Ala-Lys-Tyr-Asn-Lys-Ile-Lys-Ser-Arg-Gly-Ile-Lys-Ala-Asn-Ala-Phe-Lys-Lys-Leu-Asn- -Asn-Leu-Th~-Phe-Leu-Tyr-Leu-Asp-His-Asn-Ala-Leu-Glu-Ser-Val-Pro-Leu-Asn-Leu-Pro-Glu-Ser-Leu-Arg-Val-Ile-His-Leu-Gln-Phe- -,:
~:
:; ' . ' '"'':.

, WO9l/02744 PCT/US90/04745 ~ . ' ~ .
1 1 2 ~

Asn-Asn-Ile-Ala-Ser-Ile-Thr-Asp-Asp-Thr- ~' Phe-Cys-Lys-Ala-Asn-Asp-Thr-Ser-Tyr-Ile- ~ ,,',' ' Arg-Asp-Arg-Ile-Glu-Glu-Ile-Arg-Leu-Glu~
Gly-Asn-Pro-Ile-Val-Leu-Gly-Lys-His-Pro- ,' , Asn-Ser-Phe-Ile-Cys-Leu-Lys-ALg-Leu-Pro- '' :~
Ile-Gly-Ser-Tyr-Phe-(COOH), ~ - ' . . .
.: . . .
A comparison of the human sequence with the bovine sequence shows that there are,seven differences at positions lS, 20, 23, 54, 84, 85 and 105 of the sequence. ' ~
Accordingly, at least the residues~ at those positions may ' ;' be interchanged.~ It is po~-.ible, of course, that ''-' sequences of other mammalian~or,avian speci~es may exhibit other differences~
15~ ~ In the course of obtaining the genes for the mature bovine and~human proteins it was discovered that ~' the genes each encode~a precursor segment. Portions of `, the~ precursor segments for the bovine and human proteins ;are shown in F~igure l1, parts A and'B. ~Accordingly~, it is 20~ ~believed that the protein occu.rs~as a prepropolypeptide ~ ~ ' ' and is~processed~ lnto the mature protein defined by the ~ ; ' sequences indicated above. Polypeptides comprising the , '`'~
mature~protein sequence and including a portion or all of ~, ' the;~precursor segments~a~re intended to be within the scope -25~ o;f~,the invention~
The~invention provides the protein ~in ~' ,subst'antially pure'~`form in~whi'ch it is essentially free of other molecules with~which it is associate~ in nature. In .
this regard, the term "substantially purs" intends a 30 composition containing less than about 30~ by weight ~ ~
contaminating protein, preferably less than about 10% ' ;, contaminating protein, and most preferably less than about , .' 5% by weig~ht~contàminating protein. The term -~, substantially pure" is used relative to proteins with ~ ' 3S which the protein is associated in nature and is not intended to exclude c~ompositions in which the protein is WO91/02744 PCT~US90/04745 admixsd with nonproteinaceous carriers or vehicles, or proteinaceous carriers or vehicles, provided other .~ protein(s) with which it is associated naturally are absent. The invention also provides the protein in novel S partially glycosylated or totally deglycosylated forms (both of which are referred to herein as deglycosylated").
Based on the above amino acid sequences, ~ oligonucleotide probes which contain the codons for a por--~ 10 tion or all of the determined amino acid sequences are prepared and used to screen DNA libraries for substantially homologous genes that encode related

3~ proteins. The homologous qenes may be from other species of mammals or animals (e.g., avians) or may represent i~ 15 other members of a family of related genes. The basic 31 strategies for preparing oligonucleotide probes and DNA
libraries, as well as their screening by nucleic acid hybridization, are well known to those of ordinary skill in the art. See, e.q., DNA CLONING: VOLUME I (D.M. Glover ed. 1985); NUCLEIC ACID HYBRIDIZATION (B.D. Hames and S.J.
Higgins eds. 1985); OLIGONUCLEOTIDE SYNTHESIS (M.J. Gate ed. 1984); T. Maniatis, E.F. Frisch & J. Sambrook, MOLECULAR CLONING: A LABORATORY MANUAL (1982).
First, a DNA library is prepared. Since the initially identified protein was bovine, it was logical to probe a bovine library first, find full length clones and use the full length~bovine clones to probe libraries of other mammalian species to identify the protein gene (and thus the amino acid sequences) of other species. The library can consist of a genomic DN~ library. Bovine and human genomic l.ibraries are lcnown in the art. See, e.q., Lawn et al., Cell (1978) 15:1157-1174. DNA libraries can ; also be constructed of cDNA prepared from a poly-A RNA
(mRNA) fraction by~reverse transcription. See, ~, U.S. :
Patent Nos. 4,446,235; 4,440,859; 4,433,140;

:: :
:. .:

:

` : :
W~91/0~744 P~TtUS90/0474

4,431,740; 4,370,417; 4,363,877. The mRNA is isolated from an appropriate cell line or tissue that expresses the factor. Libraries from cells involved in bone formation (e.g., osteoblasts) or from osteotumors (e.g., osteosarcoma lines) are likely sources to probe for the nucleic acids that encode the protein. cDNA (or genomic DNA) is cloned into a vector suitable for construction of a library. A preferred vector is a bacteriophage vector, such as phage lambda. The construction of an appropriate library is within the skill of the art.
Once the library is constructed, oligonucleo-tides to probe the library are prepared and used to isolate the desired genes. The oligonucleotides are synthesized by any appropriate method. The particular nucleotide sequences selected are chosen so as to correspond to the codons encoding the known amino acid sequences of the protein. Since the genetic code is redundant, it will often be necessary to synthesize seYeral oligonucleotides to cover all, or a reasonable number, of the possible nucleotide sequences which encode a particular region of a protein. Thus, it is generally preferred in selecting a region upon which to base the probes, that the region not contain amino acids whose codons are highly degenerate. It may not be necessary, ~ however, to prepare probes containing codons that are rare -;~ in the mammal from which the library was prepared. In certain circumstances, one of skill in the art may find it desirable to prepare probes that are fairly long, and/or ;`
encompass regions of the amino acid sequence which would 30 have a high degree of deeeneracy in corresponding nucleLc acid sequences, particularly if this lengthy and/or -;~ degenerate region is highly characteristic of the protein.
Probes covering the complete gene, or a substantial part of the genome, may also be appropriate, depending upon the 35expected degree of homology. Such would be the case, for example, if a cDNA of a bovine protein was used to screen . ., WOsl/02744 PCT/USsO/0474s :
a human gene library for the corresponding human protein gene. It may also be desirable to use two probes (or sets of probes)~ each to different regions of the gene, in a single hybridization experiment. Automated oligonucleo-tide synthesis has made the preparation of large familiesof probes relatively straightforward. While the exact length of the probe employed is not critical, generally it is recognized in the art that probes from about 14 to about 20 base pairs are usually effective. Longer probes of about 25 to about 60 base pairs are also used.
The selected oligonucleotide probes are labeled with a marker, such as a radionucleotide or biotin using -~
standard procedures. The labeled set of probes is then used in the screenlng s~ep, which consists of allowing the single-stranded probe to hybridize to isolated ssDNA from ~;~ the library, according to standard techniques. Either stringent or permissive hybridization conditions could be appropriate, depending upon several factors, such as the length of the probe and whether the probe is derived from the same species as the library, or an evolutionarily close or distant species. The selection of the appropri-ate conditions is within the skill of the art. See aener-ally, NUCLEIC ACID HYBRIDIZATION, supra. The basic requirement is that hybridization conditions be of suf-ficient stringency~so that selective hybridization occurs;
i.e., hybridization ~s due to a sufficient degree of nucleic~acld homology (e.g., at least about 7S~), as op-posed to nonspecific binding. Once a clone from the screened library has been identified by positive `
hybridization, it can be confirmed by restriction enzyme analysis and DNA sequencing that the partLcular library insert contains a gene for the protein.
Alternatively, a DNA coding sequence for a protein can be prepared synthetically from overlapping oligonucleotides whose sequence contains codons for the amino acid sequence of the protein. Such oligonucleotides ~ :' -W~'~1/02744 PCTtUS90/04745 ~ .~ -15~ ~ 7~

are prepared by standard methods and assembled into a complete coding sequence. See, e.~ dge, Nature (1981) 292:756; Nambair et al., Science ~1984) 223:1299; Jay et ~;
al., J Biol Chem (1984) 259:6311.
; 5 Accordingly recombinant polynucleotides that ` encode the polypeptidès may be preparéd and isolated by one or more of the above described techniques. The term ~;
recombinant polynucleotide'i as used herein denotes a ~; ; polynucleotide of genomic, cDNA, semisynthetic or ~` 10 synthetic origin which, by virtue of its origin or manipulation ~ I ? is not associated with aLl or a portion of the nucleic acid with which it is associated in~nature .~ or in the form of a library, ~2~ is linked to a polynucleotide to which it is not linked in nature or in a libra~y, or ~3) is not found in nature or in a library.
The DNA~sequence coding for the protein can be cloned in any suitable vector, identified, isolated, and thereby maintained~in a composition substantially free of vectors that do~not contain the codLng sequence of the protein (e.g., other~library clones). Numerous cloning vectors are known,to those of skill in the art, and the ~-selection of an~appropriate cloning vector is a matter of choice. Examplas of recombinant DNA vectors for-cloning ;and~the~host c~e~lls which they transform include 25~bac~teriophage~1ambda~ ~(E. coli), pBR322 (E. coli), pACYCl77 ( ~coli~ pKT230 (gram-negative bacteria)l ;~
pGV1~106 (gram-negàtive bacteria), pLAFRl ~gram-negative bacteria), pME290~(non-E. coli gram-negative bacteria), pHV14 (E. coli and~Bacillus subtills)t pBD9 (Bacillus), pIJ61 (strsPtomyces?/ pUC6 (streptomycQs)~ actinophage C31 (strsptomyces)/ YIp5 ~yeast), YCpl9 (yeastJ,~ and bovine ~ ;
papilloma virus (mammalian cells). See ~enerally, DNA ~
CLONING: VOLUMES I h II, supra; MOLEC~LAR CLONING: A ;~-LABORATORY MANUAL,~su~ra.
In one embodiment of the present invention, the coding sequence for gene encoding the protein i.s placed : : .

WO91~02744 PCT/US90/04745 ~ t~ 16-under the control of a promoter, ribosome binding site (for bacterial and eucaryotic expression), and optionally an operator ,tcollectively referred to herein as "control~
sequences), so that the DNA sequence encoding the protein (referred to herein as the "c;oding" seguence) is transcribed into RNA and the RNA translated into protein in the host cell transformed by the vector. The coding sequence may or may not contain a signal peptide or leader sequence. The determination of the point at which the precursor protein begins and the signal peptide ends'is easily determined from the N-terminal amino acid sequence of the precursor protein. The protein can also be expressed in the form of a fusion protein, wherein a heterologous amino acid se~uence is expressed at the N-terminal. See, e.q., U.S. Patents Nos. 4,431,739;
4,425,437.
The recombinant vector is constructed so that the protein coding sequence is located in the vector with the appropriate control sequences, the positioning and 20 ~orientation of the protein coding sequence with respect to the control sequences being~such that the coding sequence is transcribed under the control of the control sequences (i.e., by RNA polymerase which attaches to the DNA
molecule at the~control sequences)~. The control sequences 25 ~may be~ligated to the coding sequence prior to insertion -into a vector, such~as the cloning vectors described above. Alternatively, the coding sequence can be cloned directly into an expression vector which already contains ;
the control sequence and an appropriate restriction site ~;
downstream from control sequences. For expression of the protein coding sequence in procaryotes and yeast, the control sequences will be heterologous to the coding - sequence. If the selected host cell is a mammalian cell, the control sequences can be heterologous or homologous to 35 the protein coding sequence, and the coding sequence can ;
be genomic DNA, cDNA or synthetic DNA. Either genomic or - .:

',:

WO9l/02744 PC~/US90~04745 L~ ~ 7 ~ ~

cDNA coding sequence may be expressed in yeast. If secretory expression in eukaryotic cells is necessary or ; desirable, sequences such as the yeast alpha factor signal .~ sequence or other sequences that direct secretion are ! 5 included in the control sequence. If glycosylation similar to the native molecule is desired, the gene may be expressed in yeast or mammalian cells (COS, CHO, or CV-l cells) using vectors and procedures known in the art.
., A number of procaryotic expxession vectors are known in i~ 10 the art. See, e.q., U.S. Patent Nos. 4,440,859; ~
~` 4,436,815; 4,431,740;~ 4,431~,739; 4,428,941; 4,425,437; ;
4,418,149; 4,411,994; ~,366,246; ~,342,832. See also British Patent Specifications GB 2,121,054; GB 2,008,123;
GB 2,007,675; and European Patent Specification 103,395.
Yeast expression vectors are known in the art. See, e.q., U.S. Patent Nos. 4,446,235; 4,443,539; 4,430,428. See also European Patent Specifications 103,409; 100,561; and ~` 96,491.
~ Recombinant protein can be produced by growing ~ ~ 20 host cells transformed by the expression plasmid described above under conditions whereby the protein is produced.
The protein is then isolated from the host cells and purified. If the expression system secretes protein into growth media, the protein can be purified directly from cell-free media. If the recombinant protein is not secre~ed, it is isolated from cell lysates. The selection of the appropriate growth conditions and recovery methods are within the skill of the art or are apparent from the ; recovery methods used to isolate the native proteins. The recombinant protein may be recoversd by affinity chromatography using the antibodies produced in accordance with the invention. Recombinant protein may be unglycosylated or have a different glycosylation pattern than the native molecule depending upon the host that is used to produce it. The proteins are useful for making ~

WO91/V27~ PCT/USgO/04745 antibodies that recognize sequential epitopes of the protein, and are useful as bone markers.
Either native, deglycosylated, or synthetic (recombinant) protein can be used to produce antibodies, both polyclonal and monoclonal. The term "antibody" is ~ -intended to inclùde whole Ig of any isotype or species as ; well as antigen binding fragments, chimeric constructs and single chain antibodies. If polyclonal antibodies are desired, purified protein is used to immunize a selected 10 mammal (e.g., mouse, rabbit, goat, horse, etc.) and serum from the immunized animal later collected and treated ~ ~
according to known procedures. Compositions containing ~ ;
polycIonal antibodies to a variety of an~igens in addition ` -to the protein can~be made substantially free of~anti-;15~bodies which do not bind specifically to the protein ~`
bodies by passing~the composition through a column to which protein has been bound. After washing, polyclonal antibodies to the protein are eluted from the column.
Monoclonal anti-protein antibodies can also be readily 20 produced by one skilLed in the art. The general methodol-ogy for making~monoclonal antibodies by hybridomas is well

5~ known. I = ortal, ant~ibody-producing cell lines can also ; be created~by techniques other than fusion, such as direct ;i transformation of B lymphocytes with oncogenic DNA, or ~transfectlon~with~Epstein-Barr virus. See, e.q., Schreier,~M., et ~al.~ HYBRIDOMA TECHNIQUES (1980~
Hammerling et; al~ MONOCLONAL ANTIBODIES AND T~CELL
HYBRIDOMAS~(1981)~; Kennett et al., MONOCLONAL ANTIBODIES
(1980). ~ ~
:, .
By employing the bone-specific protein (native~
deglycosylated or synthetic) as an antigen in the im-munization of the source of the B-cells immortalized for the production ~of~monoclonal antibodies, a panel of monoclonal~antibodies recognizing epitopes at different sites on the~protein molecule can be obtained. Antibodies which xecognize an epitope in the binding region of the ~ -: : :,.

WO91/02744 PCT/U~90/~474~
-19~ 7',C~ ~

protein can be readily identified in competition assays between antibodies and protein. Antibodies which recognize a site on the protein are useful, for example, in the purification of the protein from cell lysates or 5 fermentation media, in charact~rization of the protein and -in identifying immunologically related proteins. Such immunologically related proteins (i.e., that exhibit common epitopes with the protein) are another aspect of the invention. In general, as is known in the art, the 10 anti-protein antibody is fixed (immobilized) to a solid :
support, such as a column or latex beads, contacted with a solution containing the.protein, and separated from the solution. The protein, bound to the immobilized antibodies, is then eluted. Antibodies to the protein may be used to identify osteoblasts and osteocytes by conventional immunoassay procedures. Such identification may be used to follow bone and/or cartilege turnover.

Examples The following is intended to further illustrate processes for preparing the proteins of the invention and their use in preparing antibodies. These examples are not intended to limit the invention in any manner.
~
A. Preparation of Demineralized Bone ~ :
Bovine metatarsal bone was obtained fresh from the slaughterhouse and transported on ice. Bones were cleaned of all periosteum and marrow with high pressure water, crushed into fraqments using a liquid-nitrogen-cooled grinder and pulverized into powder using a liquid-nitrogen-cooled mill. The pulverized bone was washed four ~
times for 20 minutes in 4C deionized water (8 liters/kg). -The bone was then washed overnight with the same volume of deionized water at 4C. The bone powder was demineralized for 5 hr in 0.5 N HCl (21 liter/kg) at 4C. The acid was ,,.:: . , , ~, - ; : . - .

: ' . : . ' ' . .: .

r, 0~

~;~ decanted, and the demlneralized bone powder was washed several times with 4C deionized water until the wash -,, reached a pH>3. The excess water was removed on a suction filter.

B Extractlon of Noncollaqenous Proteins -~s Demineralized bone powder was extracted with 4M
guanidine-HCl, 10 mM EDTA pH 6.8 (2 liters/kg bone powder) for 16 hr at 4C. The suspension was suction-filtered to 10 recover the guanidine-HC1-soluble fraction and concentrated at least 5-fold by ultrafiltration using a ~ -lO,OOO dalton cùt-off membrane (SlOY10 Amicon spiral cartridge). ~
.:: .
C. Gel ~iltration~
The extract ~rom llB, redissolved in 4M
guanidine-HCl, was ~fractionated on a Sephacryl S-200 column equilibrated in 4M guanidine-HCl, 0.02% sodium azide, 10 mM EDTA, pH 5.8. Fractions were assayed by ~` 20 their absorbance at 280 nm and the fractions were combined as s~hown in Figure 2. The fraction indicated by <~ in Figure 2 constitutes~a low molecular weight (LMW, 10,000- ~ ;
30,000 daItons)lprotein fraction possessing the greatest activity.~ This fracti~on was pooled and dialyzed against 6 changes of~l~O~volumes of deioni ed water and lyophilized.
Al~l~operations~except lyophilization and dialysis (4C) were conducted at~room temperature.

D. Ion Exchanqe Chromatoqraphy The pooled fraction from llC was dissolved in 6M
urea, lO mM NaCl, 1 mM N~M, 50 mM sodlum acetate, pH 4.8 and centrifuged at 10,000 rpm for 5 min~ The supernatant was fractionated on a CM52~(a commercially available CMC) column equilibra~ted~ Ln the same buffer. Bound proteins ~i~
were eluted from the;column using a 10 mM to 400 mM NaCl gradient in the same~buffer, and a total volume of 350 ml :

WO91~02744 PCT/US90/04745 .'.~3~
-21- 2 ~ (J 7J ,3 a~ a flow rate of 27 ml/hr. Proteins eluted with 10-150 mM NaCl (the <----> of Figure 3) were collected and 1 dialyzed against 6 changes of llO volumes of deionized water for 4 days and lyophilized. All of the foregoing ` 5 operations were conducted at room temperature except dialysis (4C). ,"

~ E. Con~ Chromatography J The fraction obtained in step D above was 1~ enriched by affinity chromatography using concanavalin A
(ConA)-SepharQse 4B (pharmacia). In order to minimize leaching of ConA from the column during chromatography, the resin was cross-linked with glutaraldehyde essentially as described by K.P. CampbelL, D.H. MacLennan~ J Biol Chem -~
(1981) 256:4626. Briefly, resin was pelleted (500 x g, ; 5 min) and washed twice with 4 volumes of Z50 mM NaHCO
pH 8.8. The resin was then equilibrated in the same buffer for 6-8 hrs at 4C. After pelleting, the resin was `
cross-linked by the addition of 4 volumes of 250 mM ~ :
NaHC03, pH 8.8,~250 mM methyl-alpha-D-mannopyranoside (alpha-MM), 0.03% glutaraldehyde with gentle mixing for 1 hr at room temperature. The reaction was quenched by washing the resin twice in lM Tris-HCl, pH 7.8. The resin was stored in the same buffer containing 0.01% Thimersol at 4C until use.
Samples for ConA chromatography were solubilized in 1% deoxycholate~at pH 8Ø Any small amount of precipitate was removed by centrifugation 12,000 x g, 5 minutes.
Prior to chromatography, cross-linked resin was first equilibrated with ~5 column ~olumes of 50 mM Tris, ~' pH 8.0 followed by ~5 column volumes of l~ sodium deoxycholate. Samples were loaded and nonbound fractions collected by washlng with l~ DOC. Elution was monitored 35 by OD280. Bound material was eluted with 0.5M alpha-MM in 1% DOC as shown in Figure 4.

:

WO~1/V2744 PCT/US90/04745 ,: ~ . "
~ 22-J
F. Chromatoqraphy on Heparin-Sepharose . The bound fraction eluted from the ConA column ; was reequilibrated by chromatography on a GH-25 column ~-''f 5 (Pharmacia) equilibrated in 6M urea, O.lM NaCl, 50 mM
-, Tris-HCl pH 7.2 heparin-sepharose buffer. Approximately ;. 80 mg (l mg/ml) were loaded on a 25 ml large heparin sepharose column (Pharmacia). The column was washed of -~
all unbound material. Then bound proteins were eluted with the same equilibrating buffer but containing 0.5M
NaCl as shown in Figure 5. About 5-8 mg of heparin- -` sepharose bound proteins were recovered.
,: . . .
G. Chromatoqra~y on C18-RP-HPLC
The pH of~the heparin bound fraction was lowered below 5 by adding TFA. Final purification of the heparin- :
bound fraction was achieved using reversed phase HPLC.
The columns used were a Vydac TP-RP18 4.6 mm x 25 cm and 1.0 x 25 cm. Solvent A was 0.1% aqueous trifluoroacetic acid (TFA) ~and B 90~ acetonitrile in A. ~ound proteins were eluted from the column with a 32-62~ B solvent gradi-; ent at a rate of 1~/min. The protein composition eluted between~47-50% solvent B as shown in Figure 6. 140-200 ug protein were recovered. Amino aci~d composition and amino acid sequen~ces of the~protein were determined using standard procedures and are described above and shown in Figure~7.

H. Deqlycosylation `
Glycopeptidase F cleaves N-linked oligo- `~
saccharides at the innermost ~-acetylglucosamine residue.
High mannose, hybrid and complex oligosaccharides are ;~ susceptible to the enzyme. Protein was iodinated by the -`
chloramine-T method. Labeled protein was digested for 12 lS hours with 6.7 units/ml glycopeptidase F (Boehringer Mannheim) in O.lM Tris-HCl, pH 7.4, 10 mM EDTA, at 37C.

'' ~' ~.:

.-' WO91/Ot744 P~T/USgO/0474 .~ ~ .'.71 . ~ .
-23- ~ 6~t~ ~:J

~ Both the glycosylated and deglycosylated forms ;~ were analyzed by sodium dodecyl sulfate/15% polyacrylamide slab gels prepared according to standard methods. ~Figure ~;~
. 8 is a photograph of the autoradiograph.
` 5 I. Isolation of Bovine Protein Gene -The protein is designated "OIF" in the drawings.
The following four 20-mer oligonucleotide probes were synthesized using a Bioséarch 8600 DNA synthesizer.
The sequences of these probes were derived from the amino acid sequence of~the bovine protein that was isolated from bone.

GCNAARTAYA~YAARATQAA
TTYCGNTTRTGNAARTTYTT
CTRCTRTGNAARACRTTYCG
CTYCCNTTRGGNCANTAWGA -where A is adenine, C is cy~osine, G is guanine, T is thymine, N is A, C, G or T, Q is Aj C, or T, R is A or G, is A, G or T and Y is C or T.
These~probes were used to analyze a lambda bacteriophage Illibrary' containing DNA fragments from ; bovine~ er~. The;l~ambda phage ~ector, EM~L3 (Frischauf, 25 ~A~.M~ et al., J~Mol Biol (1983) 170:827) was purchased (Stratagene, 1190 North Torrey Pines Road, La Jolla, CA
92037) ~and used as described. Bovine liver was collected .~ ~ at a slaughterhouse and quickly frozen in liquid nitrogen.
The frozen tissue was pulverized and lysed with sarkosyl NL-97A and proteinase ~. Cellular DNA was purified by CsCl density gradient centrifugation, treated with Sau3A, and fractionated by sucrose gradient centrifugation ~fter phenol-chloroform extraction.
~`~ The DNA was concentrated to 1 mg/ml and 1 ug was mixed with 1 ug EMBL3 DNA. The mixture was treated with DNA ligase as described by the supplier and packaged via ~ ~ .

~ .

~ WO91/02744 PcT/uS9Ofo474s ~Q~

the lgigapack kit (Stratagene) to make a library stock.
Approximately 10,000 viable phage were plated on each of 60 plates (150 mm) (see Molecular Cloning: A Laboratory Manual, Maniatis, Fritsch and Sambrook, Cold Spring Harbor ; 5 Laboratory, Cold Spring Harbor, New York (1983)).
Phage plaques were transferred to nitrocellulose filtexs (4 replicates per plate). Absorbed DNA was de- -natured by treatment with 0.5M NaOH, 1.5M NaCl. Filters ~.
were neutralized in 0.SM Tris, pH 8.0, l.SM NaCl, washed with 2XSSC, air dried, and baked for 2 hr at 80C under vacuum.
, .
Phage containing the gene were identified by hybridizing P-labeled oligonucleotides with filters containing DNA transferred;from phage plaques. Plaques - 15~ hybridizing in duplicate to~at least two of the four oligonucleotides were purified and further characterized.
From three experiments representing a total of 6 x 10 plaques (2-4 bovine genome equivalents) only two plaques were shown on final analysis to conta~in s~equences compat- ~
20~ ible with the protein~sequence.~ These two~phage (bOIF21 ,!~,::.,.. ;
and bOIF39) were shown by restriction site mapping and Southern analysis to contain identical sequences in the bOIF region although the extent of bovlne DNA in the two was different~
Seq~encing of these clones revealed the gene structure shown ;in Figure 9~and the DMA sequence of Figure As ~shown,~ the~mature~bovine~sequence is encoded by three exons,~designated 1, 2, and 3 in Figure 9. Exon 1 encodes the first 17 amino acids, Exon 2 amino acids I8 through 49, and Exon 3 the remaining residues. The pre-; cursor segment of ths prepropolypeptids is sncoded by a portion of Exon 1 and separate exons upstream from Exon 1. .
To confirm the splicing of the exons, a bovine cancell~ous bone mRNA~library wàs prepared using the ; ~ 35 poIymerase chain reaction (PCR) with primers selected from -within Exons 1 and 3. A clone containing the bOIF cDNA
'~
: ~ ;

, : . , f~
- 2 5 - ` 2 ~ ~ q ~
.
sequence was isolated and the nucleotide sequence determined. The resulting sequence confirmed the exon ~`
splicing contemplated by Figures 9 and 11.

5 J. Isolation of Human Protein Gene .
,i Hùman fetal liver DNA was isolated, treated with Sau3A as was described for bovine DNA (above). Phage plaques (2.5 x 10 ) form the EMBL3 human liver DNA library (generated as described above~ and a similar number of plaques from a human liver DNA library produced in lambda phage Charon 4a (Lawn, R.M., et al., Cell (1978) 15:1157) were probed with a radioactively-labeled ( EcoRI) DNA frag-ment (Fragment l, Figure 9) containing the first exon of the bovine gene. All positive appearing plaques (13 from the EMBL3 library and 6 from the Charon 4a) were isolated and reprobed with the radioactive exon 1 DNA fragment as well as a second DNA fragment containing exon 3 (Fragment 2, Figure 9) of the bovine gene. Only one phage from the EMBL3 library (phage 41) hybridized to both DNA fragments and a second phage (phage 28) from the gene library ; hybridized with only the exon 1 fragment upon rescreening.
:~ All other phage did not hybridize with the radioactive ~ probes upon rescreening, indicating that their original ; identification was~;either an artifact of hybridization or that the~human protein cognate DNA was lost upon replication during~isolation of the phage.
Phage~41 and 28 were subjected to restriction site mapping and Southern blot analysis to locate the sequences for the human probes and the DNA sequence of these regions was determined.
Sequencing of these clones showed that the human gene structure parallele~ that shown in Figure 9, that is the human gene for the mature protein comprLses three exons of identical length to the bovine exons. A restric-tion map of the human gene region is shown in Figure 10.It was further determined that the human protein, as the ~: ; ~ -:
``:
....... _ ,., . , . . ~ . .~ . - -' WO91/02744 PCT/U~90/04745 ~ , ., ~ ".
2 6-- ~

bovine, occurs as a prepropolypeptide with the precu~rsor ~' .~' segment being encoded by a portion of Exon 1 and upstream ~`, exons. The human genomic DNA sequence is shown in Figure ". 11. ', ~ 5 '~ , K. Construction of Mammalian Expression vector _ ntaininq Human Protein Gene t ~ ~: A HindIII fragment of the human protein gene ' ,;
"' (site 15422 to~site 25814 in Figure 10) was cloned into a ~ .
', ~ 10 HindITI site of~plasmid pSC614 (see Figure 12) in ' .
transcriptional alignment with the SV40 promoter or in the ', ~
~;~ opposite orientation to yield plasmids phOIF17 and phOIF18 . ,-respectively. Pigure 12 is a schematic diagram of plasmid ,~
phOIF18. ~ ,' ~; L. Transfection of COS-7 Cells with phOIP18 COS-7 celIs were transfected with phOIP18 ~see ,.
Felgner, P.L., et al.~, Proc Natl Acad Sci (USA) (1'987) ,' 84:7413). After transfection the cells were allowed to :", 20 grow in medium with or without'serum (5~). In these ,l~.
experiments only the cells grown in serum-conta'ining media ', synthesized the protein. ~' M. Production and~Testinq of Antibodies to the Protein ;;
- : :
' M.1. ~Production of Polyclonal Antibodies Polyc~lonal antibodies to ('1) a synthetic 30-mer ,~
polypeptide having a sequence corresponding to ~he amino' acids 1-~30 of Flgure~7 except for a Leu ----> Asn substitution at position 25 and (2) the native protein 30 purified from bone as described above were prepared and ,, characterized as follows. ~ ' Antiserum to the 1-30-mer was raised in a rabbit ~ , by injecting the rabbit with 500 ug of the polypeptide in ' ~.
complete Freund's adjuvant (CFA~, followed by boosts of 35 500 ug of~the polypeptide in incomplete Freund's'adjuvant , ' (ICFA) at approximately three week intervals. The :' ..:

'~;-:

. . ... . . , ..... . . .... , . .... .,, , , .. ~, .. ... ..

W~91/02744 PCT~US90/04745 antiserum was obtained after the fourth boost and had a titer as measured by ELISA of >1:10,~00. Rabbit antiserum to the native protein was raised similarly using an initial injection of 50 ug protein in CFA followed by boosts of 50 ug protein in ICFA. This antiserum had a titer of >1:10,000 by ELISA.
The antiserum to the 1-30-mer was tested in Western blots on the purified native protein, deglycosylated native protein, and on crude native protein (Con-A bound material), all fixed post-blotting with 0.2 glutaraldehyde. The antiserum detected the purified native` protein at ~1 ug and also recognized the deglycosylated protein and the crude protein. The antiserum to the native protein recognized the native 15 protein at >100 ng in Western blots. ~ -M.2. Product1on of Monoclonal Antibodies Murine~monoclonal antibodies to the purified native protein were prepared as follows. From two fusions 25 positive wells were identified by immunoprecipitation.
A group of female Balb/c mice was injected ~ intraperitoneally (IP) with 10-20 ug of purLied native -; ~ protein in CFA. The~animals were boosted with l0-2b ug of protein in ICFA~ ~Following the third boost, the mice were ~25~ bled and serum antibody titers against the~ protein checked by ELISA. Two animals were found to have titers of 40,000. ~They were given a final intravenous (IV) injection of 20 ug protein four days prior to the ~usion.
; Fusion to the SP2/0 myeloma ~GM3659 B, NIGMS
Human Genetic Mutant Cell Repository, Camden, NJ) was performed essentially according to the protocol of Oi and Herzenberg, "Immunoglobulin-producing Hybrid Cell Lines"
in Selected Methods in Cellular Immunoloqy, Mishell and Shiigi, eds., W.H. Freeman and Co., San Francisco, pp.
357-362, (1980). Spleen cells from the animals were mixed with SP2/0 at a ratio of 5:1. 50% polyethylene glycol .

.
.
....... .. . . . ... . ...... ......

~,~ . ;, . i,: ' . ~ i ' , ` . . , , J !, WO~1/02744 PC~/USgO/04745 ~
~ 28-: . .
1500 (Boehringer-Mannheim Biochemicals, Indianapolis, IN~
was used as the fusagen. Cells were plated at 106 cells/
well along with resident peritoneal cells at 4 x 103 cells/well in DMEM with high glucose (4.5 g/l) sup~
plemented with 20% FCS (Hyclone Laboratories, Logan, UT), 2 mM L-glutamine,~2 mM sodium pyruvate, nonessential ~mino acids, penicillin and streptomycin. In this procedure, ~-aminopterin was replàced by azaserine (sigma) according to the procedure by Larrick et al., Proc Natl Acad Sci (USA) (1983) 80:6376, and added along with thymidine and ;~
hypoxanthine on day 1 after the fusion. :
From two fusions 25 positive wells were identi-fied by immunoprecipitation of 125I-labeled proteln.
A11 25~were also positive in an ELISA against the protein. In addition, several other wells were positive by ELISA but negative by immunoprecipitation.
The supernatant from one uncloned well (3B2.17, previously designated FO13-3B2) was particularly positive and was used in a Western blot. In this testing synthetic peptides corresponding to amino acid segments 1-30, 62-95 , and 76-105 of the protein sequence were made and 1-2 ug of each was appLied to separate lanes in the gel. Blots were probed with 50-lOO~ug/ml of purified antibody. This anti-body recognized >300 ng protein as well as deglycosylated ~ -protein. The antibody~also picked up the protein in a crude fraction (total Con-A bound) and was found to recognize the C-terminal peptide (76-105) but not the N-terminal peptide ~1-30). Another clone, designated 2C11.6, was found to recognize the intsrnal 62-95 segment.
Clones 3B2.17 and 2C11.6 wers subcloned by limiting dilu-tion and were found to be stable and to be IgG isotype.
These clones have been deposited in the American Type Culture Collection (ATCC) on 5 April 1989 under ths provi-sions of ths Budapest Treaty. Their ATCC dssignations ;~
are, respectively, HB10099 (3B2.17) and HB10098 (2C11.6).
: "

.

WO91/027~ PCT/US90/04745 -29- 2~ 7, ~::
;, . . .
M.3. Immunostalninq With Antibodies Rat fetuses (19 days old) and 3 day old rats 1 were used for tissue sections. Tissue was fixed in 10~
formalin and 5u sections were prepared in paraffin. The sections were treated with xylene (deparaffinized).and washed with Tris-buffered saline~(TBS), pH 7.6, three times. The washed sections were then contacted with a mixture of TBS, 0.05% Tween, 0.5% bovine serum albumin (BSA), 10~ normal mouse serum (NMS), for 1 hr at 25C or overnight at 4C.~ Monoclonal antibody 3B2 (see M.2) at 5 ug/ml in TBS/Tween/BSA/NMS was added and ~he sections were incubated for 1 hr~at room temperature. The sections were then washed thr~ee times in TBS/-Tween and contacted with biotinyl~ated goat anti-mouse antibody for 10 min at room temperature. The séctions were then washed again three ; times with TBS/Tween and contacted with streptavidin-horseradish peroxidase for 10 min at room temperature.
Thereafter, the sections were washed a final three times with TBS/Tween ~and contacted with substrate. After substrate treatment, the sections were washed in water, counterstained with~Mayer's hematoxylin, washed again in water, dehydrated~in lQQ~% ethanol, and mounted.
The most intenst staining occurred in hypertrophic calcifying cartilage in the growth plate.
25 There :was also a~clear staining pattern~in osteoblasts and osteocytes~ with the~`darkest in more mature cells. No detectable staining was observed in soft tissue. A faint pattern was seen in bone itself.
Modifications of the above-described modes of carryiing out the invention that are obvious to those of skill in the arts relsvant to the invention are intended to be within the scope of the following claims.

:: : : :

Claims (14)

Claims

1. A substantially pure polypeptide having the following amino acid sequence:

, where R1 is Ala or Thr, R2 is Asn or His, R3 is Thr or Ser, R4 is Ala or Thr, R5 is Ile or Val, R6 is Val or Ile and R7 is Phe or Ile, and substantially pure polypeptides that are and substantially homologous thereto or are immunologically related thereto.

2. The polypeptide of claim 1 wherein R1 is Ala, R2 is Asn, R3 is Thr, R4 is Ala, R5 is Ile, R6 is Val, and R7 is Phe.

3. The polypeptide of claim 1 wherein R1 is Thr, R2 is His, R3 is Ser, R4 is Thr, R5 is Val, R6 is Ile, and R7 is Ile.

4. A substantially pure prepropolypeptide comprising (a) the polypeptide of claim 1 and (b) at least a portion of a precursor segment having the sequence shown in Figure 11A or Figure 11B, or a sequence substantially homologous thereto.

5. A substantially pure prepropolypeptide comprising (a) the polypeptide of claim 2 and (b) at least a portion of a precursor segment having the human sequence shown in Figure 11, Parts A and B, or a sequence substantially homologous thereto.

6. A substantially pure prepropolypeptide comprising (a) the polypeptide of claim 3 and (b) at least a portion of a precursor segment having the bovine sequence shown in Figure 11, Part B, or a sequence substantially homologous thereto.

7. A polypeptide having the following amino acid sequence:

, where R1 is Ala or Thr, R2 is Asn or His, R3 is Thr or Ser, R4 is Ala or Thr, R5 is Ile or Val, R6 is Val or Ile and R7 is Phe or Ile, wherein said polypeptide is deglycosylated relative to a native polypeptide having said sequence and deglycosylated polypeptides that are substantially homologous thereto or are immunologically related thereto.

8. Antibody that binds to a polypeptide of claim 1, 2, 3, or 7.

9. A recombinant polynucleotide encoding a polypeptide of claim 1, 2 or 3.

10. A recombinant polynucleotide encoding a prepropolypeptide of claim 4, 5, or 6.

11. A recombinant vector containing a recombinant polynucleotide of claim 9 and capable of directing the expression of the polypeptide encoded thereby.

12. A recombinant vector containing a recombinant polynucleotide of claim 10 and capable of directing the expression of the prepropolypeptide encoded thereby.

13. A recombinant host cell or microorganism containing the recombinant vector of claim 11 and capable of permitting expression of said polypeptide.

14. A recombinant host cell or microorganism containing the recombinant vector of claim 12 and capable of permitting expression of said polypeptide.