Skip to main content

Full text of "USPTO Patents Application 09804625"

See other formats


J) 



Europaisches Patentamt 

® ^JjJ European Patent Office @ Publication number: 0 336 760 

Office europeen des brevets A2 



@ EUROPEAN PATENT APPLICATION 



® Application number: 89303417.3 © Int. CI. 4 : C 07 K 7/00 

_ A 61 K 37/02 

@ Date of filing: 06.0439 





Priority: 06.04.88 US 178133 08.11.88 US 268835 


Nathan, Ranga 




23.01.89 US 299408 


6104 Robertson Avenue 
Newark California 94560 (US) 


® 


Date of publication of application: 
11.10.89 Bulletin 89/41 


Rosen, David M. 




3655 Shadyhoilow Court 


® 




San Jose California 35148 (US) 


Designated Contracting States: 




AT BE CH DE ES FR GB GR IT U LU NL SE 


Dasch, James R. 
3181 Morris Drive 


® 


Applicant: COLLAGEN CORPORATION 


Palo Alto California 94303 (US) 


2500 Faber Place 


Seyedln, Saeld M. 
20761 Lowena Court 
Saratoga California 95070 (US) 




Palo Alto, California 94303 (US) 


@ 


Inventor: Bentz, Hanne 






36125 Toulouse Street 
Newark California 94560 (US) 


@ Representative: Goldin, Douglas Michael et a) 
J.A. KEMP & CO. 14, South Square Gray's Inn 
London WC1R5EU (GB) 



The microorganism(s) has (have) been deposited with ATCOHB under numbers 10099 and 10098. 
@ Bone-Inducing protein. 

© An osteogentcally active protein, having the sequence 

(H2N)-Ala-Lys-Tyr-Asn-l_ys-lle-Lys-Ser-Arg-Gly-lle-Lys-Ala- 

Asn-Thr-Phe-Lys-Lys-Leu-His-Asn-Leu-Ser-Phe-Leu-Tyr-Leu- 

Asp-His-Asn-Ala-Leu-Glu-Ser-Vai-Pro-Leu-Asn-Leu-Pro-Glu- 

Ser-Leu-Arg-Vai-lle-His-Leu-Gln-Phe-Asn-Asn-lle-Thr^Ser-lle- 

Thr-As p-Asp-Thr-Phe-Cys-Lys-AIa-Asn-Asp-Thr-Ser-Tyr-lle- 

Arg-Asp-Arg-lle-Glu-Giu-lle-Arg-Leu-Glu-Gly-Asn-Pro-Val-lle- 

Leu-Gly-Lys-His-Pro-Asn-Ser-Phe-lle-Cys-Leu-Lys-Arg-Leu- 

Pro-lle-Gry-Ser-Tyr-lle-Asp-(COOH), 

and substantially pure polypeptides that are substantially 

equivalent and substantially homologous thereto is disclosed. 

Pharmaceutical compositions containing these polypeptides 

and methods to use them are also disclosed. 



Bundesdruckerei Berlin 



1 



EP 0 336760 A2 



2 



Description 

BONE-INDUCING 



Technical Field 

The present invention relates to protein chemistry s 
and osteoplasty. More particularly, it relates to a 
protein that induces bone growth, implant and 
pharmaceutical compositions containing that pro- 
tein, methods for promoting bone growth using such 
compositions, methods for stimulating bone marrow w 
progenitor cells to divide and differentiate into bone 
marrow cells, and methods for treating diseases 
associated with dysfunction/malfunction of bone 
generation and/or bone resorption, such as osteo- 
porosis. 15 

Background Art 

It has been established that bone contains 
materials which can stimulate the formation of new 
bone when placed in contact with living systems. 20 
(Urist, M. R., Clin Orthop (1968) 56:37; Science 
(1965) 150:893; Reddi, A. H. f et al„ Proc Natl Acad 
Sci (USA) (1972) 69:1601.) Attempts have been 
made to purify whatever factors are responsible for 
this activity. A "bone morphogenic protein" (BMP) 25 
was extracted from demineralized bone using urea 
or guanidine hydrochloride and reprecipitated ac- 
cording to the disclosures in U.S. Patents 
Nos. 4,294,753 and 4,455,256 to Urist. Urist subse- 
quently reported (Urist, M. R., Clin Orthop Rel Res 30 
(1982) 162:219) that ion exchange purification of this 
crude protein mixture yielded an activity which was 
unadsorbed to carboxymethyl cellulose resin (CMC) 
at pH 4.8. Urist's reports in Science (1983) 
220:680-685 and Proc Natl Acad Science (USA) 35 
(1984) 81 :371-375 describe BMPs having molecular 
weights of 17,500 and 18,500 daltons. Urist's patent 
publication, EPA Publication No. 0212474, describes 
BMP fragments of 4,000 to 7,000 daltons obtained by 
limited proteolysis of BMP. 40 

U.S. Patent No. 4,608,199 describes a bone- 
derived protein of 30,000-32,000 daltons. The protein 
is described as being water soluble and having no 
affinity for concanavalin A (ConA). 

WO 88/00205 reports four proteins, designated 45 
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 sequence (see 50 
below) of the osteogenic protein of the present 
invention. 

U.S. 4,434,094 to Seyedin and Thomas reported 
the partial purification of a bone generation-stimulat- 
ing, bone-derived protein by extraction with 55 
chaotropic agents, fractionation on anion and cation 
exchange columns, and recovery of the activity from 
a fraction adsorbed to CMC at pH 4.8. This new 
protein fraction was termed "osteogenic factor" 
(OF) and was characterized as having a molecular 60 
weight below about 30,000 daltons. 

Commonly owned U.S. Patent No. 4,774,332 
describes two proteins that were purified to homo- 



PROTEIN 

geneity using a purification 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 
NaCI gradient. These two proteins were originally 
called cartilage-inducing factor (CIF) A and CIF B. 
CIF A was subsequently found to be identical to a 
previously identified protein called transforming 
growth factor beta (TGF-beta). CIF B has been 
found to be a novel form of TGF-beta and is now 
known as TGF-beta2. 

Commonly owned U.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 NaCI gradient below 
that in which the major portions of CIF A and CIF B 
elute (i.e., below about 150 mM NaCI). The present 
invention relates to the identification of the active 
ingredient of that fraction. In this regard, at the time 
that patent was filed it was not known whether the 
bone-inducing activity was attributable to a single 
protein or a plurality of proteins acting in concert. 
Identification of the protein(s) responsible for bone- 
inducing activity was complicated by the large 
number of proteins in the fraction (estimated to be 
several hundred), the lack of a conclusive in vitro 
assay for bone-inducing activity, and extensive 
difficulty in isolating the active protein from other 
proteins in the fraction. Indeed, it has taken 
applicants approximately three years of effort-in 
which a variety of protein fractionation procedures 
were attempted-to obtain the bone-inducing pro- 
tein from the CMC-bound fraction at a level of purity 
at which it could be sequenced and identified as 
being responsible for the activity observed in the 
crude fraction. 

As discussed in detail below, the bone-inducing 
activity of the fraction has been found to be 
attributable to a glycoprotein component having a 
variable (apparently due to variation in glycosylation) 
molecular weight (in the glycosylated state) of 
approximately 20,000-28,000 daltons as determined 
by sodium dodecyl sulfate polyacrylamide gel elec- 
trophoresis (SDS-PAGE) analysis. Amino acid se- 
quencing of the active component indicates it is 
composed of a protein that has a sequence different 
from any previously reported sequence. 

Disclosure of the Invention 

The invention relates to a substantially pure 
polypeptide that induces bone formation in vivo in 
mammals. 

Accordingly, one aspect of the invention is a 
substantially pure osteogenically active polypeptide 
haying an internal sequence selected from the group 
consisting of 

a) -Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg-Glu-lle- 
Lys-Ala-Asn-Thr-Phe-Lys-Lys-Leu-His-Asn- 
Leu-Ser-Phe-Leu-Tyr-Leu-Asp-His-Asn-Aia- 
Leu-Glu-; 

b) -Leu-His-Asn-Leu-Ser-Phe-Leu-Tyr-Leu- 
Asp-His-Asn-Ala-Leu-Glu-Ser-Val-Pro-Leu- 



2 



3 



EP 0336760 A2 



4 



Asn-Leu-Pro-Glu-; 

c) -Ser-Leu-Arg-Val-lle-His-Leu-Gln-Phe- 
Asn-Asn-lle-Thr-Ser-lle-Thr-Asr^Asp-Thr-Phe- 
Cys-Lys-Ala-; 

d) -Ala-Asn-Asp-Thr-Ser-Tyr-lle-Arg-Asp- 
Arg-lle-Glu-Glu-lle-Arg-Leu-Glu-Gly-Asn-Pro- 
Val-lle-; 

e) -Gly-Asn-Pro-Val-lle-Leu-Gly-Lys-His-Pro- 
Asn-Ser-Phe-ile-Cys-Leu-Lys-Arg-Leu-Pro-tle- 
Gly-Ser-Tyr-; and/or a carboxy terminal se- 
quence as follows 

-Arg-Leu-Pro-lle-Gly-Ser-Tyr-lle-Asp-(COOH), 
and substantially pure polypeptides that are sub- 
stantially equivalent and substantially homologous 
thereto. 

Another aspect of the invention is a substantially 
pure osteogenically active polypeptide having the 
following amino acid sequence: 
(H2N)-A!a-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg-Gly-lle- 
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- 
lle-His-Leu-Gln-Phe-Asn-Asn-lle-Thr-Ser-lle-Thr-As 
p-Asp-Thr-Phe-Cys-Lys-Ala-Asn-Asp-Thr-Ser-Tyr- 
lle-Arg-Asp-Arg-lle-Glu-Glu-lle-Arg-Leu-Glu-Gly- 
Asn-Pro-Val-lle-Leu-Gly-Lys-His-Pro-Asn-Ser-Phe- 
lle-Cys-Leu-Lys-Arg-Leu-Pro-lle-Gly-Ser-Tyr-lle- 
Asp-(COOH), 

and substantially pure polypeptides that are sub- 
stantially equivalent and substantially homologous 
thereto. 

Another aspect of the invention is a composition 
for inducing bone formation or bone marrow cell 
production in vivo comprising (a) an effective 
amount of one or more of the above-described 
osteogenically active polypeptides and (b) an effec- 
tive amount of TGF-beta combined with (c) a 
pharmaceutically acceptable carrier. 

Another aspect of the invention is a method of 
inducing bone formation in vivo at a desired site 
comprising implanting the above described compo- 
sition in a mammal at said site. 

Yet another aspect of the invention is a method of 
inducing bone marrow cell production in a living 
mammal comprising administering systemically to 
the mammal an effective amount of one or more of 
the above described polypeptides. 

Another aspect of the invention is a method of 
treating an individual for a condition characterized by 
insufficient bone formation and/or undesired bone 
resorption comprising administering systemically to 
the individual an effective amount of one or more of 
the above-described polypeptides. 

Still another aspect of the invention is an improve- 
ment in a method for treating a living mammal for 
cancer of the hematopoietic system wherein the 
mammal is subjected to irradiation to kill neoplastic 
hematopoietic cells the improvement comprising 
administering systemically to the mammal prior to 
irradiation a sufficient amount of TGF-beta to 
suppress hematopoietic stem cell division and 
administering systemically to the mammal after 
irradiation a sufficient amount of one or more of the 
above described polypeptides to stimulate hemato- 
poietic stem cell division. 



Another aspect is a method for isolating a 
substantially pure osteogenically active protein com- 
position from demineralized bone comprising : 

(a) extracting demineralized bone with a 
5 chaotropic (dissociative) extractant that solu- 

bilizes nonfibrous proteins; 

(b) subjecting the extract from step (a) to gel 
filtration to recover a fraction containing pro- 
teins of molecular weight of approximately 

10 20,000-36,000 daltons; 

(c) adsorbing the fraction from step (b) onto 
a carboxymethyl cellulose cation exchanger at 
approximately pH 4.5-5.5 under denaturing 
conditions; 

15 (d) eluting a fraction from the cation ex- 

changer with a sodium chloride gradient of 
about 10 mM to about 150 mM; 

(e) adsorbing the eluate of step (d) onto a 
cross-linked ConA column; 

20 (f) eluting bound protein from the column of 

step (e); 

(g) adsorbing the eluate of step (f) onto a 
heparin-sepharose column ; 

(h) eluting bound protein from the column of 
25 step (g); and 

(I) chromatographing the eluate of step (h) on 
an RP-HPLC column using a trifluoroacetic 
acid- acetonitrile system and recovering the 
substantially pure osteogenically active protein 
30 composition as the fraction eluting from the 

column at approximately 42-45% acetonitrile. 
Further aspects of the invention are recombinant 
materials (i.e., recombinant DNA, recombinant vec- 
tors, and recombinant cells or microorganisms) and 
35 processes for producing the osteogenic proteins of 
the invention. 

Brief Description of the Drawings 
In the drawings: 
40 Figure 1 is a flow chart of the process that 

was used to isolate osteogenic protein from 
demineralized bovine bone. 

Figure 2 is a graph of the optical densities 
(absorbances at 280 nm) of the gel filtration 
45 fractions of the gel filtration fractions of the 

example (©C). 

Figure 3 is a graph of the optical densities 
(absorbances at 280 nm) of eluate fractions 
from the preparative ion exchange chromato- 
50 graphy of the example ( © D). 

Figure 4 is a graph of the optical densities 
(absorbances at 280 nm) of eluate fractions 
from the cross-linked ConA chromatography 
step of the example (<£> E) ; 
55 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 (©F); (absorbances at 230 
nm) of the gradient fractions from the 
60 Figure 6 Is a graph of the optical densities 

(absorbances at 230 nm) of the gradient 
fractions from the C18-RP-HPLC chromato- 
graphy step of the example (©G) ; 
Figure 7 is a table showing results of amino 
65 acid sequencing of the osteogenically active 



3 



5 



EP 0336 760 A2 



6 



isolate of the invention and locations of the 
sequenced fragments in the overall sequence. 

Figure 8 is a photograph of an autoradio- 
graph of SDS-PAGE analyses of the purified 
osteogenic protein that are described in the 
example (©H) (lanes A and C show glycosyl- 
ated protein; lanes B and D show enzymaticalty 
deglycosylated protein). 

Figure 9 is a bar graph showing the results of 
the assays described in the example (©1.3 and 
1.4). 

Figure 10 is a bar graph of the results of the 
assays described in © J.1 . of the examples. 

Figure 1 1 is a graph of the results of the tests 
described in ©J.3. of the, examples. 

Modes of Carrying Out the Invention 

Isolation of Bone-Inducing Protein from Bone 

In view of the showing that bone inductive 
proteins from human, monkey, bovine and rat are 
nonspecies specific in their ability to produce 
endochondral bone in xenogeneic implants (Sam- 
path, T.K., et ai, Proc Natl Acad Sci (USA) (1983) 
80:6591) it is believed that the bovine protein 
described herein has been highly conserved among 
mammalian species--i.e., corresponding bone-in- 
ducing proteins from different mammalian species 
(herein called "species analogs") will have substan- 
tially homologous amino acid sequences that vary 
from the bovine protein, if at all, in one or more amino 
acid residue additions, deletions or substitutions 
and/or substantially similar glycosylation patterns 
that do not affect the nonspecies-specific ability of 
the molecule to induce bone formation. In this 
regard, the terms "substantially equivalent" and 
"substantially homologous" are intended to mean 
proteins, regardless of species or method of 
preparation, that have the same amino acid se- 
quence as the bovine osteogenic protein described 
in the examples and proteins of similar but different 
amino acid sequence, which difference(s) does not 
affect nonspecies-specific endochondral bone- in- 
ducing activity adversely. The amino acid sequences 
of such "substantially homologous" proteins will 
usually be at least 50% homologous, more usually at 
least 80% homologous, and preferably at least 90% 
homologous to the bovine sequence described 
herein. Accordingly, such proteins may be derived 
from bone of diverse mammalian origin or syn- 
thesized 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 that are 
not essential for activity with neutral (uncharged) 
amino acids to avoid improper disulfide bonding, by 
substitution or elimination of residues in the as- 
paragine-linked glycosylation sites of the proteins to 
alter glycosylation patterns, by substitution of meth- 
ionines that are not necessary for activity to make 
the molecules less susceptible to oxidation, by 
chemical modification of one or more residues, or by 
elimination or alteration of side-chain sugars. The 
source of protein prepared by purification from 
native sources is advantageously porcine or bovine 



long bone because of its ready availability. 

The process for isolating the osteogenic protein 
from bone is as follows. The bone is first cleaned 
using mechanical or abrasive techniques, frag- 
5 mented, 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 with stronger acid. These techniques are 

10 understood in the art, and are disclosed, for 
example, in U.S. 4,434,094. The resulting prepara- 
tion, a demineralized bone, is the starting material 
for the preparation of the claimed osteogenic protein 
from native sources. 

15 The initial extraction is designed to remove the 
nonfibrous (e.g., noncollagenous) proteins from the 
demineralized bone. This can be done with the use 
of chaotropic agents such as guanidine hydro- 
chloride (at least about 4 molar), urea (8 molar) plus 

20 salt, or sodium dodecylsulfate (at least about 1°/o 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 arid Hascall, J Biol Chem 
(1969) 244:77-87 and 2384-2396). The extraction is 

25 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 the 
medium depends upon the extractant selected. The 

30 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 

35 desired. Salts can also be removed by controlled 
electrophoresis, or by molecular sieving, or by any 
other means known in the art. (t is also preferred to 
maintain a low temperature during this process so as 
to minimize denaturation of the proteins. Alterna- 
te? tively, 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 

45 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 Sephac- 
ryl S-200 column at room (10°C-25°C) temperature. 
The sized fraction is then subjected to ion 

50 exchange chromatography using CMC at approxi- 
mately pH 4.5-5.2 preferably about 4.8, in the 
presence of a nonionic chaotropic agent such as 6 M 
urea. Other cation exchangers may be used, 
including those derived from polyacrylamide and 

55 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 factor is adsorbed on the column and is eiuted in 

60 an increasing salt concentration gradient in the 
range of about 10 mM to about 150 mM. This fraction 
is designated "CMB-1" for convenience. 

CMB-1 is lyophilized and the dry CMB-1 is 
dissolved in aqueous sodium deoxycholate (OOC), 

65 pH 8.0. This solution is affinity chromatographed on 



4 



7 



EP 0 336 760 A2 



8 



an equilibrated 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-1" 
for convenience. 

CAB-1 is reequilibrated for heparin-sepharose 
chromatography by desalting on a GH-25 column 
equilibrated on heparin-sepharose buffer, 6 M urea, 
0.1 M NaCl, 50 mM Tris-HCI pH 7.2. The desalted 
fraction is loaded into a heparin-sepharose column. 
After washing, bound material is eluted from the 
column using the same buffer at a 0.5 M NaCl salt 
concentration. The resulting eluate is designated 
"HSB-1" for convenience. 

HSB-1 is diluted 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 o/o aqueous TFA 
(Solvent B). The osteogenic protein of the invention 
elutes at approximately 47-500/o of solvent B 
(42-450/0 acetonitrile) by volume. 

Proteins eluted by the 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 osteogenic protein. The 
"smearing" of the protein is believed to mainly be the 
result of heterogeneity in the glycosylation of the 
molecule or perhaps variable post-translational 
modification or proteolytic degradation. After enzy- 
matic or chemical deglycosylation, SDS-PAGE ana- 
lysis of the protein gives a single band of approxi- 
mately 9600 daltons. Reduction of the deglycosyl- 
ated protein with dithiothreitol does not affect its 
migration. 

Initial amino acid sequence analysis of the glyco- 
sylated protein yielded the following internal se- 
quence in the N-terminal portion of the protein: 
-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg-Gly-lle-Lys-Ala- 
Asn-Thr-Phe-Lys-Lys-Leu-His-Asn-Leu-Ser-Phe- 
X-Tyr-Thr-Asp-His-Asn-AIa-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. Repeated 
sequencing and sequencing of endoproteinase 
Lys-C (an enzyme that cleaves proteins at Lys 
residues) and endoproteinase Glu-C (an enzyme 
that cleaves proteins at 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 residue, and 
that the isolate consists of a protein of approxi- 
mately 106 amino acids. Figure 7 provides a 
summary of these sequence analyses. The symbol 
"CHO" designates a carbohydrate substituent The 
symbol "COOH" represents a carboxy! group and 
designates the carboxy terminus. The first column 
(on the 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 give the sequences of two Glu-C 



fragments. 

The positioning of the fragments shown in Fig. 7 
are based on the apparent overlap In the various 
fragments. Based on this positioning, the most likely 

5 sequence for the protein is: 

(H2N)-Ala-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg-X3ly-lle- 
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- 

10 lle-His-Leu-Gln-Phe-Asn-Asn-lle-Thr-Ser-lle-Thr-As 
p-Asp-Thr-Phe-Cys-Lys-Ala-Asn-Asp-Thr-Ser-Tyr- 
lie-Arg-Asp-Arg-lle-Glu-Glu-lle-Arg-Leu-Glu-GIy- 
Asn-Pro-Val-ile-Leu-Gly-Lys-His-Pro-Asn-Ser-Phe- 
lle-Cys-Leu-Lys-Arg-Leu-Pro-lle-Gly-Ser-Tyr-lie- 

15 Asp-(COOH), 

It will be appreciated that this sequence is not 
conclusive due to the nature of the analysis 
technique. Accordingly, the sequences given above 
may not be entirely accurate. The sequence may be 

20 confirmed by identifying the gene for the protein, 
sequencing the gene and deducing the amino acid 
sequence therefrom. 

Amino acid composition analyses of. the isolated 
deglycosylated protein were carried out with and 

25 without performic acid oxidation (performic acid 
oxidation permits detection of cysteic acid 
residues). The results of these analyses are indi- 
cated below. 

30 Amino Acid Without With Performic 
Performic Oxidation 
Oxidation Integer 
Integer Res/Mol 
Res/Mol 



45 



Asx 1 


11 


12 


Thr 


4 


4 


Ser 


6 


7 


Glx 2 


8 


7 


Gly 


5 


8 


Ala 


4 


4 


Val 


4 


3 


lie 


6 


6 


Leu 


9 


9 


Tyr 


3 


ND 


Phe 


3 


3 


Lys 


7 


6 


His 


3 


3 


Arg 


5 


5 


Pro 


5 


4 


Met 


0 


0 


Cys 


ND 


2 


TOTAL 


*83 


83(86 3 ) 



1 1ncludes both Asp and Asn 
includes both Glu and Gin 

ND <= not determined 

60 3|ncludes three Tyr residues 

'based on SDS-PAGE determined molecular 
weight of approximately 9600 

65 To determine the amino acid composition of the 



5 



9 EP 0 336760 A2 10 



protein, the protein was hydrolyzed for 24 hr at 
110°C in 6 N HCI containing 0.10/o phenol. To 
determine the presence of Cys residues, performic 
acid oxidation was done prior to hydrolysis. Amino 
acids were detected on a Beckman 6300 analyzer 
using ninhydrin detection. It should be understood 
that these compositions are only an approximation 
due to the limitations of the analytical technique. 
Further it is clear that the molecular weight deter- 
mined by SDS-PAGE does not conform with the 
molecular weight projected from the amino acid 
sequence analyses and is substantially lower than 
that projection. The invention provides the oste- 
ogenic protein in substantially pure form in which it 
is essentially free of other molecules with which it is 
associated in nature. In this regard, the term 
"substantially pure" intends a composition contain- 
ing less than about 30Vo by weight contaminating 
protein, preferably less than about 10% contaminat- 
ing protein, and most preferably less than about 5<Vo 
by weight contaminating protein. The term "substan- 
tially pure" is used relative to proteins with which the 
osteogenic protein is associated in nature and is not 
intended to exclude compositions in which the 
osteogenic protein is admixed with nonproteina- 
ceous pharmaceutical carriers or vehicles or protei- 
naceous pharmaceutical carriers or vehicles. The 
invention also provides the osteogenic protein in 
novel partially glycosylated or totally deglycosylated 
form (both of which are referred to herein as 
"deglycosylated"). The term "osteogenic" intends 
the ability to induce new bone formation either alone 
or in combination with a co-factor. Assays for 
osteogenic activity are described in the examples, 
infra . 

Further characterization of the osteogenic protein 
of this invention may be carried out using proce- 
dures known in the art. Its isoelectric focusing 
pattern, isoelectric point, susceptibility to degrada- 
tion by proteases or other chemicals such as acids 
or bases, and affinity to other materials such as 
other lectins, and the like may be so determined. 

Based on the above amino acid sequence, 
oligonucleotide probes which contain the codons 
for a portion or all of the determined amino acid 
sequence are prepared and used to screen DNA 
libraries for genes encoding the osteogenic protein 
and substantially homologous genes that encode 
related proteins having osteogenic activity. The 
homologous genes may be from other species of 
mammals or animals (e.g., avians) or may represent 
other members of a family of related genes. The 
basic 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, ejj., DNA 
CLONING: VOLUME I (D.M. Glover ed. 1985); 
NUCLEIC ACID HYBRIDIZATION (B.D. Hames and 
S.J. Higgins eds. 1985); OLIGONUCLEOTIDE SYN- 
THESIS (MJ. Gate ed. 1984) ;T. Maniatis, E.F. Frisch 
& J. Sambrook, MOLECULAR CLONING: A LABOR- 
ATORY MANUAL (1982). 

First, a DNA library is prepared. Since the 
identified protein is bovine, it is 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 osteogenic 
protein gene (and thus the amino acid sequences) 
of other species. The library can consist of a 
5 genomic DNA library. Bovine and human genomic 
libraries are known in the art. See , e.g ., Lawn et al M 
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, e.g. , 

10 U.S. Patent Nos. 4,446,235; 4,440,859; 4,433,140; 
4,431,740; 4.370,417; 4,363.877. The mRNA is iso- 
lated from an appropriate cell line or tissue that 
expresses the factor. Libraries from cells involved in 
bone formation (e.g., osteoblasts) or from osteotu- 

15 mors (e.g., osteosarcoma lines) are likely sources to 
probe for the osteogenic protein nucleic acids. 
Representative examples of such cell lines are the 
human amniotic line WISH (ATCC CCL25), the 
human osteosarcoma lines TE-85 (ATCC CRL1547) 

20 and MG63 (ATCC CRL1427), the human prostate 
carcinoma line PC-3 (ATCC CRL1435), the rat 
osteosarcoma line UMR-106 (ATCC CRL1661) and 
the mouse osteosarcoma line DUNN (available from 
Dr. vital Ghanta, University of Alabama Medical 

25 School). 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. 

30 Once the library is constructed, oligonucleotides 
to probe the library are prepared and used to isolate 
the desired osteogenic protein genes. The oligonu- 
cleotides are synthesized by any appropriate 
method. The particular nucleotide sequences se- 

35 lected are chosen so as to correspond to the 
codons encoding the known amino acid sequences 
of the osteogenic protein. Since the genetic code is 
redundant, it will often be necessary to synthesize 
several oligonucleotides to cover all, or a reasonable 

40 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 

45 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, 

50 and/or encompass regions of the amino acid 
sequence which would have a high degree of 
redundancy in corresponding nucleic acid sequen- 
ces, particularly if this lengthy and/or redundant 
region is highly characteristic of the protein. Probes 

55 covering the complete gene, or a substantial part of 
the genome, may also be appropriate, depending 
upon the expected degree of homology. Such would 
be the ^ase, for example, if a cDNA of a bovine 
osteogenic protein was used to screen a human 

60 gene library for the corresponding human oste- 
ogenic protein. 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 oligonucleotide synthesis has made the 

€5 preparation of large families of probes relatively 



6 



11 



EP 0336 760 A2 



12 



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 screening step, which 
consists of allowing the single-stranded probe to 
hybridize to isolated ssDNA from the library, accord- 
ing to standard techniques. Either stringent or 
permissive hybridization conditions could be appro- 
priate, 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 
evolutionary close or distant species. The selection 
of the appropriate conditions is within the skill of the 
art. See generally , NUCLEIC ACID HYBRIDIZATION, 
supra . The basic requirement is that hybridization 
conditions be of sufficient stringency so that 
selective hybridization occurs; i.e., hybridization is 
due to a sufficient degree of nucleic acid homology 
(e.g., at least about 75%), as opposed to nonspe- 
cific 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 particular library insert 
contains a gene for the osteogenic protein. 

Alternatively, a DNA coding sequence for an 
osteogenic protein can be prepared synthetically 
from overlapping oligonucleotides whose sequence 
contains codons for the amino acid sequence of the 
protein. Such oligonucleotides are prepared by 
standard methods and assembled into a complete 
coding sequence. See , e.g. , Edge, Nature (1981) 
292:756; Nambair et al., Science (1984) 223:1299; 
Jay et al., J Biol Chem (1984) 259:6311. 

Accordingly recombinant polynucleotides that 
encode the osteogenic polypeptides may be pre- 
pared and isolated by one or more of the above 
described techniques. The term "recombinant poly- 
nucleotide" as used herein denotes a polynucleotide 
of genomic, cDNA, semisynthetic or synthetic origin 
which, by virtue of its origin or manipulation (1) 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 finked to a polynucleotide to which it is 
not linked in nature or (3) is not found in nature. 

Recombinant DNA molecules containing the cod- 
ing sequence for the osteogenic protein can be 
cloned in any suitable vector and thereby maintained 
in a composition substantially free of vectors that do 
not contain the coding sequence of the osteogenic 
protein (e.g., other library clones). Numerous clon- 
ing vectors are known to those of skill in the art, and 
the selection of an appropriate cloning vector is a 
matter of choice. Examples of recombinant DNA 
vectors for cloning and the host ceils which they 
transform include bacteriophage lambda ( E. coli ), 
PBR322 ( E. coli ), pACYC177 ( E. coli ), pKT230 
(gram-negative bacteria), pGV1106 (gram-negative 
bacteria), pLAFRI (gram-negative bacteria), 
pME290 (non -E. coli gram-negative bacteria), pHV14 
( E. coli and Bacillus subtilis ), pBD9 ( Bacillus ), plJ61 



( Streptomyces ), pUC6 ( Streptomyces ), actinophage 
C31 ( Streptomyces ), Ylp5 (yeast), YCp19 (yeast), 
and bovine papilloma virus (mammalian cells). See 
generally , DNA CLONING: VOLUMES I & II, supra ; 
5 MOLECULAR CLONING: A LABORATORY MA- 
NUAL, supra . 

In one embodiment of the present invention, the 
coding sequence for an osteogenic protein gene is 
placed under the control of a promoter, ribosome 

10 binding site (for bacterial expression) and, option- 
ally, an operator (collectively referred to herein as 
"control" sequences), so that the DNA sequence 
encoding the osteogenic protein (referred to herein 
as the "coding" sequence) is transcribed into RNA 

15 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 mature protein begins and the 
signal peptide ends is easily determined from the 

20 N-terminal amino acid sequence of the mature 
protein. The osteogenic protein can also be ex- 
pressed in the form of a fusion protein, wherein a 
heterologous amino acid sequence is expressed at 
the N-terminal. See e.g. , U.S. Patents 

25 Nos. 4,431,739; 4,425,437. 

The recombinant vector is constructed so that the 
osteogenic protein coding sequence is located in 
the vector with the appropriate control sequences, 
the positioning and orientation of the osteogenic 

30 protein coding sequence with respect to the control 
sequences being such that the coding sequence is 
transcribed under the control of the control sequen- 
ces (i.e., by RNA polymerase which attaches to the 
DNA molecule at the control sequences). The 

35 control sequences 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 con- 

40 trol sequence and an appropriate restriction site 
downstream from control sequences. For express- 
ion of the osteogenic protein coding sequence in 
procaryotes and yeast, the control sequences will be 
heterologous to the coding sequence. If the se- 

45 lected host cell is a mammalian cell, the control 
sequences can be heterologous or homologous to 
the osteogenic protein coding sequence, and the 
coding sequence can be genomic DNA, cDNA or 
synthetic DNA. Either genomic or cDNA coding 

50 sequence may be expressed in yeast. If glycosyl- 
ation similar to the native molecule is desired, the 
gene may be expressed in yeast or mammalian cells 
(COS, CHO, or CV-1 cells) using vectors and 
procedures known in the art. In this regard, initial 

55 tests at selected concentrations indicate tentatively 
that the totally deglycosylated protein is not active. 
For this reason, expression in eukaryotes that are 
capable of effecting glycosyiation may be essential 
to make an active protein by recombinant proce- 

60 dures. 

A number of procaryotic expression vectors are 
known in the art. See , e.g. , 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; 

65 4,366,246 ; 4,342,832. See also British Patent Specif i- 



13 



EP 0 336 760 A2 



14 



cations 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.g. , 
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 osteogenic protein can be pro- 
duced by growing host cells transformed by the 
expression vector described above under condi- 
tions whereby the osteogenic protein is produced. 
The osteogenic protein is then isolated from the 
host cells and purified. If the expression system 
secretes osteogenic protein into growth media, the 
protein can be purified directly from cell-free media. 
If the recombinant protein is not secreted, it is 
isolated from cell lysates. The selection of the 
appropriate growth conditions and recovery meth- 
ods are within the skill of the art or are apparent from 
the recovery methods used to isolate the native 
osteogenic proteins. The recombinant protein may 
be recovered by affinity chromatography using the 
antibodies produced in accordance with the inven- 
tion. Recombinant osteogenic protein may be 
unglycosylated or have a different glycosylation 
pattern than the native molecule depending upon 
the host that is used to produce it. As indicated 
above, the deglycosylated protein may not be active. 
It would be useful, however, for making antibodies 
that recognize sequential epitopes of the protein. 

Either native, deglycosylated, or synthetic (rec- 
ombinant) osteogenic protein can be used to 
produce antibodies, both polyclonal and monoclo- 
nal: The term "antibody" is intended to include whole 
Ig of any isotype or species as well as antigen 
binding fragments and chimeric constructs. If poly- 
clonal antibodies are desired, purified osteogenic 
protein is used to immunize a selected mammal 
(e.g., mouse, rabbit, goat, horse, etc.) and serum 
from the immunized animal later collected and 
treated according to known procedures. Composi- 
tions containing polyclonal antibodies to a variety of 
antigens in addition to the osteogenic protein can be 
made substantially free of antibodies which are not 
anti-osteogenic protein antibodies by passing the 
composition through a column to which osteogenic 
protein has been bound. After washing, polyclonal 
antibodies to the osteogenic protein are eluted from 
the column. Monoclonal anti-osteogenic protein 
antibodies can also be readily produced by one 
skilled in the art. The general methodology for 
making monoclonal antibodies by hybridomas is well 
known. Immortal, antibody-producing cell lines can 
also be created by techniques other than fusion, 
such as direct transformation of B lymphocytes with 
oncogenic DNA, or transfection with Epstein-Barr 
virus. See, ejj., M. Schreier et a!., HYBRIDOMA 
TECHNIQUES (1980); Hammerling et al., MONO- 
CLONAL ANTIBODIES AND t-CELL HYBRIDOMAS 
(1981); Kennett et al., MONOCLONAL ANTIBODIES 
(1980). 

By employing osteogenic protein (native, deglyco- 
sylated 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 osteogenic protein molecule 
can be obtained. Antibodies which recognize an 
epitope in the binding region of the protein can be 
readily identified in competition assays between 

5 antibodies and protein. Antibodies which recognize 
a site on the osteogenic protein are useful, for 
example, in the purification of the protein from cell 
lysates or fermentation media, in characterization of 
the protein and in identifying immunologically related 

10 proteins. Such immunologically related proteins (i.e., 
that exhibit common epitopes with the osteogenic 
protein) are another aspect of the invention. In 
general, as is known in the art, the antiosteogenic 
protein antibody is fixed (immobilized) to a solid 

15 support, such as a column or latex beads, contacted 
with a solution containing the osteogenic protein, 
and separated from the solution. The osteogenic 
protein, bound to the immobilized antibodies, is then 
eluted. 

20 

Osteogenic Compositions 

The osteogenic protein of the invention may be 
used to induce de novo bone formation in circum- 
stances where bone is not normally formed. The 

25 protein may thus be used prophylactically to reduce 
the likelihood of fracture, improve fixation of artificial 
joints, repair congenital or trauma-induced bone 
defects, or in cosmetic plastic surgery. The protein 
may also be used to enhance bone formation in 

30 instances where bone is normally formed, such as in 
fracture repair, replacement of surgically removed 
bone, or repair of bone damaged by periodontal 
disease or in other tooth or alveolar ridge repair 
processes. In such uses, the protein wilt be 

35 administered locally, such as by Implantation, at the 
desired site of bone formation. 

The protein may also be administered systemi- 
cally, such as intravenously, to treat indications 
associated with insufficient bone formation and/or 

40 undesirable levels of bone resorption such as 
localized, regionalized or generalized osteoporosis 
or to stimulate bone marrow progenitor cells in the 
treatment of malfunctions or dysfunctions of the 
hematopoietic system such as chronic and acute 

45 mylocytic leukemia and other cancers of the hema- 
topoietic system or in post-irradiation treatment to 
stimulate bone marrow stem cells to divide and 
differentiate. In this regard TGF-beta may be used 
pre-irradiation to suppress marrow stem cell repro- 

50 duction and differentiation and the invention protein 
may be used post-irradiation to stimulate such cells. 

Initial tests of the osteogenic protein composition 
indicate that it is necessary, in the concentrations 
and formulations tested, to coadminister a protein 

55 having TGF-beta activity to achieve bone induction 
at nonbony sites. It may be that TGF-beta induces 
proliferation of bone forming cells and the oste- 
ogenic protein of the invention induces differentia- 
tion of such cells. In this regard, TGF-beta (TGF- 

60 betal, TGF-beta2, other members of the TGF-beta 
family, or mixtures thereof) may enhance the 
process of bone induction through ancillary acti- 
vities such as antiinflammatory activity, chemotactic 
activity, and the like. Other molecules that exhibit 

65 such activities may also be useful as co-factors for 



8 



15 



EP 0 336 760 A2 



16 



bone-induction. The protein may, of course, be 
active at other concentrations or in other formula- 
tions. Further it may not be necessary to coadminis- 
ter TGF-beta at bony sites since the amount of 
endogenous TGF-beta present at the site of action 
or in systemic applications may be sufficient. 

The osteogenic protein of the invention will 
normally be formulated in osteogenically effective 
amounts with pharmaceuticaliy acceptable solid or 
fluid carriers, for local injection or implantation at the 
desired site of activity or systemic administration by 
conventional parenteral routes. Preferably the for- 
mulations for local administration include a matrix 
material that is capable of presenting the protein at 
the desired site of activity as well as providing a 
structure for developing bone and cartilage. Poten- 
tial matrices may be biodegradable or nonbiode- 
gradable and be chemically or biologically defined. 
Examples of such materials are calcium sulfate, 
hydroxyapatite, tricalciumphosphate, polyorthoes- 
ters, polylactic-polyglycolic acid copolymers, col- 
lagen, bioglass, and the like. Formulations for 
systemic administration will typically involve liquid 
vehicles that are commonly used for parenteral 
administration of proteinaceous therapeutics. 

The osteogenic protein of the invention may be 
conjugated with other molecules to increase its 
water-solubility, increase its half-live, or enhance its 
ability to bind to bone. For instance, it may be 
conjugated to polyethylene glycol to increase its 
water solubility or to bone-binding molecules such 
as bisphosphonates (e.g. 1-hydroxyethylidene- 
1,1-bisphosphonic acid, dichloromethylene bis- 
phosphonic acid, and 3-amino-1-hydroxypropy- 
lidene-1,1-bisphosphonic acid) and fluorochromes 
(e.g. tetracyclines, calcein blue, xylenol orange, 
calcein green, and alizarin complexone red) to target 
the protein to bony sites. Various agents for 
conjugating molecules to proteins are well known in 
the art and include aldehydes, carbodiimides, and 
other Afunctional moieties. 

The amount of osteogenic protein administered 
may vary depending upon the carrier used, the 
patient (age, sex, medical history, species) and the 
site and condition being treated. For local implanta- 
tion, the weight ratio of osteogenic protein to carrier 
in the formulation will typically be in the range of 
about 1:5,000 to 1:50,000. The weight ratio of 
osteogenic protein to TGF-beta in the composition 
will usually be in the range of 10:1 to 1:10. The 
implant may be placed at a predetermined site in the 
patient by conventional surgical techniques, such as 
implantation or injection. 

For systemic administration the amount of oste- 
ogenic protein will usually range between 30 g/kg 
body weight and 1 mg/kg body weight. TGF-beta 
may be added to the systemic formulations, if 
necessary, in the above proportions. In addition it 
may be desirable to combine the osteogenic protein 
with other therapeutics, such as, for instance in the 
case of osteoporosis, fluoride, calcitonin, vitamin D 
metabolites, and parathyroid hormone. Because the 
protein is nonspecies specific in its activity it may be 
used to treat mammals in general including sport, 
pet, and farm animals and humans. 



Examples 

The following is intended to illustrate the process 
for purification of native osteogenic protein as 
5 applied to a particular sample and the osteogenic 
activity of the isolated protein. It is not intended to 
limit the invention. 

A. Preparation of Demineralized Bone 

10 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 fragments using a 
liquid-nitrogen-cooled grinder and pulverized into 

15 powder using a liquid-nitrogen-cooled mill. The 
pulverized bone was washed four times for 20 
minutes in 4°C deionized water (8 liters/kg). The 
bone was then washed overnight with the same 
volume of deionized water at 4°C. The bone powder 

20 was demineralized for 5 hr in 0.5 N HCI (21 liter/kg) 
at 4°C. The acid was decanted, and the demin- 
eralized bone powder was washed several times 
with 4°C deionized water until the wash reached a 
pH > 3. The excess water was removed on a suction 

25 filter. 

B. Extraction of Noncollagenous Proteins 
Demineralized bone powder was extracted with 4 

M guanidine-HCI, 10 mM EDTA pH 6.8 (2 liters/kg 
30 bone powder) for 16 hr at 4°C. The suspension was 
suction-filtered to recover the guanidine-HCI-so- 
luble fraction and concentrated at least 5-fold by 
ultrafiltration using a 10,000 dalton cut-off mem- 
brane (S10Y10 Amicon spiral cartridge). 

35 

C. Gel Filtration 

The extract from ©B, redissolved in 4 M gua- 
nidine-HCI, was fractionated on a Sephacryl S-200 
column equilibrated in 4 M guanidine-HCI, 0.02% 

40 sodium azide, 10 mM EDTA, pH 6.8. Fractions were 
assayed by their absorbance at 280 nm and the 
fractions were combined as shown in Figure 2. The 
fraction Indicated by < — > in Figure 2 constitutes 
a low molecular weight (LMW, 10,000-30,000 dal- 

45 tons) protein fraction possessing the greatest 
activity. This fraction was pooled and dialyzed 
against 6 changes of 180 volumes of deionized water 
and lyophilized. All operations except lyophilization 
and dialysis (4°C) were conducted at room tempera- 

50 ture. 

D. Ion Exchange Chromatography 

The pooled fraction from -©C was dissolved in 6 M 
urea, 10 mM NaCI, 1 mM NEM, 50 mM sodium 

55 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 equilibrated in 
the same buffer. Bound proteins were eluted from 
the column using a 10 mM to 400 mM NaCI gradient 

60 in the same buffer, and a total volume of 350 ml at a 
flow rate of 27 ml/hr. Proteins eluted with 10-150 mM 
NaCI (the < — > of Figure 3) were collected and 
dialyzed against 6 changes of 110 volumes of 
deionized water for 4 days and lyophilized. All of the 

€5 foregoing operations were conducted at room 



9 



17 



EP 0 336 760 A2 



18 



temperature except dialysis (4°C). 

E. ConA Chromatography 

The fraction obtained in step D above was 
enriched in osteogenic activity by affinity chromato- 
graphy using concanavalin A (ConA)-Sepharose 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 
250 mM NaHC03, pH 8.8. The resin was then 
equilibrated in the same buffer for 6-8 hrs at 4°C. 
After pelleting, the resin was cross-linked by the 
addition of 4 volumes of 250 mM NaHCOs, pH 8.8, 
250 mM methyl-alpha-D-mannopyranoside (alpha- 
MM), 0.030/0 glutaraldehyde with gentle mixing for 1 
hr at room temperature. The reaction was quenched 
by washing the resin twice in 1 M Tris-HCI, pH 7.8. 
The resin was stored in the same buffer containing 
0.010/0 Thimersol at 4°C until use. 

Samples for ConA chromatography were solu- 
bilized in 1% deoxycholate at pH 8.0. Any small 
amount of precipitate was removed by centrifuga- 
tion 12,000 x g, 5 minutes. 

Prior to chromatography, cross-linked resin was 
first equilibrated with > 5 column volumes of 50 mM 
Tris, pH 8.0 followed by > 5 column volumes of 1% 
sodium deoxycholate. Samples were loaded and 
nonbound fractions collected by washing with Wo 
DOC. Elution was monitored by OD280 Bound 
material was eluted with 0.5 M alpha-MM in 1% DOC 
as shown in Figure 4. 

F. Chromatography on Heparin-Sepharose 

The bound fraction eluted from the ConA column 
was reequilibrated by chromatography on a GH-25 
column (Pharmacia) equilibrated in 6 M urea, 0.1 M 
NaCI, 50 mM Tris-HCI pH 7.2 heparin-sepharose 
buffer. Approximately 80 mg (1 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 0.5 M NaCI 
as shown in Figure 5. About 5-8 mg of heparin-se- 
pharose bound proteins were recovered. 

G. Chromatography 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 0.10/0 aqueous trifluoroacetic acid (TFA) and B 
900/0 acetonitrile in A. Bound proteins were eluted 
from the column with a 32-620/o B solvent gradient at 
a rate of l0A>/min. The osteogenic protein composi- 
tion eluted between 47-500/o solvent B as shown in 
Figure 6. 140-200 ug protein were recovered. Amino 
acid composition and amino acid sequences of the 
protein were determined using standard procedures 
and are described above and shown in Figure 7. 



H. Deglycosylation 

Glycopeptidase F cleaves N-linked oligosac- 
charides at the innermost N-acetylglucosamine 
residue. High mannose, hybrid and complex olfgo- 

5 saccharides are susceptible to the enzyme. Oste- 
ogenic protein was iodinated by the chloramine-T 
method. Labeled protein was digested for 12-15 
hours with 6.7 units/ml glycopeptidase F 
(Boehringer Mannheim) in 0.1 M Tris-HCI, pH 7.4, 10 

10 mM EDTA, at 37°C. 

Both the glycosylated and deglycosylated forms 
were analyzed by sodium dodecyl sulfate/1 50/0 
polyacrylamide slab gels prepared according to 
standard methods. Figure 8 Is a photograph of the 

15 autoradiography 

I. In vivo Bioassay of Materials 

The osteoinductive activity of the protein compo- 
sition described above was evaluated in vivo as 
20 follows. 

1.1. Formulation of Osteogenic Protein Composition 
The protein was used as a 24 ug/ml solution in 

TFA, approximately 450/0 acetonitrile. TGF-beta2 
25 was used as a 30 ug/ml solution in 10/0 TFA, 
approximately 450/0 acetonitrile. 

One ml of osteogenic protein solution and 1.4 ml 
of the TGF-beta solution were stirred with 9 ml of 
Vltrogen® collagen-in-solution (Collagen Corpora- 
te tion, Palo Alto, CA) at 4°C for 5 min. 243 mg of 
porous hydroxylapatite/tricalciumphosphate ce- 
ramic in particulate form (Zimmer Corp., Warsaw, IN) 
was added and the mixture was incubated at 4°C 
for 5 min and then lyophilized. This mixture provided 
35 sufficient material for 6 implants. A like formulation 
was made lacking TGF-beta. 

1.2. Implantation 

The ability of the formulations of 1.1 to induce 
40 endochondral bone formation in rats was deter- 
mined as follows. Portions of the lyophilized formula- 
tions were [hydrated with approximately one volume 
of water, allowed to soak for 5 min and molded into 
approximately 5 x 5 mm bodies for implantation. The 
45 implants contained -4 ug/implant of osteogenic 
protein and 0 or -7 ug/implant of TGF-beta. The 
implants were surgically placed in 34-40 day old male 
Sprague-Dawley rats on either side of the ventral 
thoracic region. Explants were removed at 14 days 
50 and evaluated biochemically for bone formation. 

1.3. Assay for Alkaline Phosphatase 

The level of alkaline phosphatase (AP) activity in 
the explants is a measure of osteogenic activity. To 

55 determine alkaline phosphatase (AP), the explants 
were cut in small pieces and homogenized in 10 
volumes (1/10) of ice-cold 1.5 M NaCI, 3 mM 
NaHC03, pH 7.5 The homogenized samples were 
then centrif uged at 12,000 rpm for 50 min at 4° C, and 

60 an aliquot of the supernatant was diluted 1:10 in 
cold, distilled water. The method of Huggins, et al., J 
EXP Med (1961) 114:761, was used to assess 
alkaline phosphatase using polystyrene plates. 



10 



19 



EP 0 336 760 A2 



20 



I.4. Results 

Figure 9 is a bar graph showing the results of the 
AP assay. As indicated the formulation containing no 
TGF-beta (a) was essentially non-active in the assay; 
whereas the formulation with TGF-beta (b) showed 
substantial activity. Histological examination of the 
explants confirmed the assay results. 

Similar tests were carried out using osteogenic 
protein compositions that had been reduced and 
alkylated or trypsinized. These tests indicate that the 
protein is deactivated by such treatments. 

J. In vitro Assay of Materials 

J.1. Effect on Rat Osteosarcoma Cell Line (ROS 
17/2.8) 

Cells were seeded in multiwell plates at 2500 
cells/cm 2 . After overnight attachment in Hams F12 
medium, 10% fetal bovine serum, the cells were 
incubated with varying amounts of the osteogenic 
protein composition for 72 hours in serum-free 
Hams F12 medium. The control was serum-free 
Hams F12 medium containing no osteogenic pro- 
tein. After incubation the ceils were rinsed twice with 
PBS and lysed with 0.1°/o Triton. Aliquots were taken 
and tested for AP activity using p-nitrophenylphos- 
phate (PNP) as a substrate. Figure 10 shows the 
results of those assays. As shown, the osteogenic 
protein produced a significant increase in AP activity 
which was dose-dependent at concentrations of 
3-30 ng/ml. 

J.2. Effect on Adult Baboon Explant Bone Cells 

Cancellous bone was obtained from adult baboon 
long bone by biopsy. The bone was minced and 
cellular outgrowth followed for up to three weeks. 
Cells were trypsined and used immediately or 
reptated. Replated cells were used within the first 
three passages. Tests for AP activity were carried 
out as in J.1. above. The results of those tests were 
reported in the table below. 

Osteogenic Protein AP specific activity 
Concentration (ng/ml) (nmPNP/ug 

protein/min) 

100 6.0 x 10" 3 

50 9.1 x 10-3 

25 4.8 x 10-3 

12.5 3.5 x 10-3 

Control 3.0 x lO" 3 

Similar tests were carried out on the human 
osteoblastic cell line MG-63 and on rat primary 
osteoblastic cells. However, no increase in AP 
activity was observed. 

J.3. Effect on Multinucleated Ceil (MNC) Formation 

Multinucleated cells are "osteoclast-! ike" cells. 
Osteoclasts are involved in bone resorption. Inhibi- 
tion of the formation of osteoclasts is believed to 
limit bone resorption. 

Bone marrow mononuclear cells from normal 
human donors were isolated using Hypaque-Ficoll 
(Histopaque-1077) gradient centrifugation. Mononu- 
clear cells were cultured in alpha-MEM containing 



20% horse serum at 10 6 cells/ml in 24-well plates 
(5x1 0 5 cells/well). All cultures were maintained in a 
humidified atmosphere of 4<Vo C02-air at 37° C. 
Cultures were fed weekly by removing half of the 

5 medium and replacing an equal volume of fresh 
medium. Osteogenic protein was added to the 
cultures in the presence of 1 ,25-dihydroxyvitamin D 
(D3), 10* fi M. After designated periods of culture 
(usually 3 weeks) cells were fixed with 5% glutaral- 

10 dehyde and stained with Wright's stain. Cells 
containing 3 or more nuclei were counted as MNCs 
using an inverted phase microscope. Figure 11 is a 
graph showing the results of these tests. As shown, 
the osteogenic protein caused a dose-dependent 

15 inhibition of MNC formation at the concentrations 
tested. 

K. Production and Testing of Antibodies to the 
Osteogenic Protein 

20 

K.1. Production of Polyclonal Antibodies 

Polyclonal antibodies to (1) a synthetic 30 mer 
polypeptide having a sequence corresponding to 
the amino acids 1-30 of Figure 7 except for a Leu 

25 — > Asn substitution at position 25 and (2) the 
native protein 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 

30 in complete Freund's adjuvant (CFA), followed by 
boosts of 500 ug of the polypeptide in incomplete 
Freund's adjuvant (ICFA) at approximately three 
week intervals. The antiserum was obtained after the 
fourth boost and had a titer as measured by ELISA of 

35 > 1:10,000. 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. 

40 The antiserum to the 1-30 mer was tested in 
Western blots on the purified native osteogenic 
protein, deglycosylated native osteogenic protein, 
and on crude native osteogenic protein (Con-A 
bound material), all fixed post-blotting with 0.20/0 

45 glutaraldehyde. The antiserum detected the purified 
native osteogenic protein at £1 ug and also 
recognized the deglycosylated protein and the 
crude protein. The antiserum to the native oste- 
ogenic protein recognized the native protein at 

50 £100 ng in Western blots. 

K.2. Production of Monoclonal Antibodies 

Murine monoclonal antibodies to the purified 
native osteogenic protein were prepared as follows. 

55 From two fusions 25 positive wells were identified. A 
group of female Balb/c mice was injected intraperi- 
toneally (IP) with 10-20 ug of purified native 
osteogenic protein in CFA. The animals were 
boosted with 10-20 ug of protein in ICFA. Following 

60 the third boost, the mice were bled and serum 
antibody titers against the protein checked by 
ELISA. Two animals were found to have titers of 
£ 1 :40,000. They were given a final intravenous (IV) 
injection of 20 ug protein four days prior to the 

65 fusion. 



11 



21 



EP 0336760 A2 



22 



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 Immunology , 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 
1500 (Boehringer-Mannheim Biochemicals, Indiana- 
polis, IN) was used as the fusagen. Cells were plated 
at 10 6 cells/well along with resident peritoneal cells 
at 4 x 10 3 cells/well in DMEM with high glucose (4.5 
g/l) supplemented with 2fJO/o FCS (Hyclone Labora- 
tories, Logan, UT). 2 mM L-glutamine, 2 mM sodium 
pyruvate, nonessential amino acids, penicillin and 
streptomycin. In this procedure, aminopterin was 
replaced 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 identified. 

All 25 were positive for immunopreclpitation of 
125 l-labeled osteogenic protein and 23 of the 25 
were positive in an ELISA against the protein. The 
supernatant from one uncloned well (3B2.17, pre- 
viously designated F013-3B2) was particularly posi- 
tive 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 osteogenic 
protein sequence were made and 1-2 ug of each was 
applied to separate lanes in the gel. Blots were 
probed with 50-100 ug/ml of purified antibody. This 
antibody 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 internal 62-95 segment. Clones 
3B2.17 and 2C11.6 were subcloned by limiting 
dilution and were found to be stable and to be IgG 
isotype. These clones are being deposited in the 
American Type Culture Collection (ATCC) under the 
provisions of* the Budapest Treaty. 

The antibodies produced by the two clones were 
testing for their ability to neutralize or block the 
activity of the osteogenic protein. ROS or BMS-2 
cells were cultured with 30 ng/ml of the osteogenic 
protein in the presence or absence of antibody at 
varying concentrations. Both of these cell lines 
normally exhibit substantial increases in AP when 
cultured in media containing osteogenic protein at 
30 ng/ml. The antibody from 3B2.17 was found to 
neutralize this effect of the osteogenic protein at 
concentrations > 10 ug/ml. 

Modifications of the above-described modes of 
carrying out the invention that are obvious to those 
of skill in the arts relevant to the invention are 
intended to be within the scope of the following 
claims. 



Claims 

5 1. A substantially pure polypeptide having 

osteogenic activity and an internal sequence in 
the N-terminal portion as follows 
-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg-Gly-lle-Lys- 
Ala-Asn-Thr-Phe-Lys-Lys-Leu-His-Asn-Leu- 
10 Ser-Phe-Leu-Tyr-Leu-Asp-His-Asn-Ala-Leu- 
Glu- 

and substantially pure polypeptides that are 
substantially equivalent and substantially ho- 
mologous thereto. 
15 2. The polypeptide of claim 1 wherein the 

amino acid immediately preceding the initial Lys 
of the internal sequence is Ala and defines the 
amino terminal of the polypeptide. 

3. A polypeptide having an internal sequence 
20 in the N-terminal portion of the polypeptide as 

follows: 

-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg-Gly-lle-Lys- 

Ala-Asn-Thr-Phe-Lys-Lys-Leu-His-Asn-Leu- 

Ser-Phe-Leu-Tyr-Leu-Asp-His-Asn-Ala-Leu- 

25 Glu- 

wherein said polypeptide is deglycosylated 
relative to a native osteogenically active poly- 
peptide having said sequence and deglycosyl- 
ated polypeptides that are substantially equival- 

30 ent and substantially homologous thereto. 

4. A substantially pure osteogenically active 
polypeptide having the following amino acid 
sequence: 

(H2N)-Ala-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg- 
35 Gly-lle-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-lle-His-Leu-Gln-Phe-Asn- 
Asn-lle-Thr-Ser-lle-Thr-As p-Asp-Thr-Phe-Cys- 
40 Lys-Ala-Asn-Asp-Thr-Ser-Tyr-lle-Arg-Asp-Arg- 
lle-Glu-Glu-lle-Arg-Leu-Glu-Gly-Asn-Pro-Val- 
lle-Leu-Gly-Lys-His-Pro-Asn-Ser-Phe-lle-Cys- 
Leu-Lys-Arg-Leu-Pro-IIe-Gly-Ser-Tyr-lle-Asp- 
(COOH), 

45 and substantially pure polypeptides that are 

substantially equivalent and substantially ho- 
mologous thereto. 

5. A polypeptide having the following amino 
acid sequence: 

50 (H2N)-Ala-Lys-Tyr-Asn-Lys-lle-Lys-Ser-Arg- 
Gly-lle-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-lle-His-Leu-Gln-Phe-Asn- 

55 Asn-lle-Thr-Ser-lle-Thr-As p-Asp-Thr-Phe-Cys- 

Lys-Ala-Asn-Asp-Thr-Ser-Tyr-lle-Arg-Asp-Arg- 
lle-Glu-Glu-lle-Arg-Leu-Glu-Gly-Asn-Pro-Val- 
lle-Leu-Gly-Lys-His-Pro-Asn-Ser-Phe-lle-Cys- 
Leu-Lys-Arg-Leu-Pro-lle-<aly-Ser-Tyr-lle-Asp- 

60 (COOH), 

wherein said polypeptide is deglycosylated 
relative to a native osteogenically active poly- 
peptide having said sequence and deglycosyl- 
ated polypeptides that are substantially equival- 

€5 ent and substantially homologous thereto. 



12 



23 



EP 0336760 A2 



24 



6. A composition for inducing bone formation 
and/or inhibiting bone resorption comprising an 
effective amount of an osteogenically active 
polypeptide of claim 1 , 2 or 4. 

7. The composition of claim 6 which further 5 
includes an effective amount of TGF-beta. 

8. A method of inducing bone formation in 
vivo at a predetermined site in a living mammal 
comprising placing the composition of claim 6 

or 7 at said site. 10 

9. A method of inducing bone marrow cell 
production In a living mammal comprising 
administering an effective amount of the com- 
position of claim 6 or 7 to the mammal 
systemically. 15 

10. A method of treating osteoporosis in a 
living mammal comprising administering an 
effective amount of the composition of claim 6 
or 7 to the mammal systemically. 

11 . In the method of treating a living mammal 20 
for a cancer of the hematopoietic system 
comprising irradiating the mammal to kill neo- 
plastic hematopoietic cells, the improvement 
comprising administering a sufficient amount of 



the composition of claim 6 or 7 to the mammal 
systemically after said Irradiation to stimulate 
hematopoietic stem cell division. 

12. The method of claim 1 1 wherein a sufficient 
amount of TGF-beta is administered systemi- 
cally to the mammal prior to said irradiation to 
suppress hematopoietic stem cell division. 

13. Antibody that binds to a polypeptide of 
claim 1,2, 3, 4 or 5. 

14. A recombinant polynucleotide encoding a 
polypeptide of claim 1 , 2, 3, 4 or 5. 

15. A recombinant vector containing a recom- 
binant polynucleotide of claim 14 and capable of 
directing the expression of the polypeptide 
encoded thereby. 

16. A recombinant host cell or microorganism 
containing the recombinant vector of claim 15 
and capable of permitting expression of said 
polypeptide. 

17. A process for producing a polypeptide of 
claim 1, 2, 3, 4 or 5 comprising culturing the 
recombinant host cell or microorganism of 
claim 16. 



30 



35 



40 



45 



50 



55 



60 



65 



13 



EP 0336760 A2 



DEMINERAUZED BONE POWDER ( 0.5 N HCL ) 

1 

4 M GUANIDINE HYDROCHLORIDE EXTRACT 

I 

SEPHACRYL S - 200 ( 20 - 36KD MW ) 

I 

CM - CELLULOSE ( 10 * 150 mM NaCL ) 

I 

CONCAN AVAU N • A ( 0.5 M Mathyl <* 6 - Mannopyrandsidd ) 

I 

HEPARIN - SEPHAROSE ( 0.1 - 0.5 M NaCL ) 

I 

C 18 . REVERSED PHASE - HPLC (42-45% Acatonltrile ) 



FIGURE 1 



EP 0336 760 A2 



mMNiQ ( ■ ) 




( ) UJU09Z *t?urqiosQY v - «k 



in 




EP 0336760 A2 




v 



EP 0336 760 A2 




FIGURE 7 



EP 0 336 760 A2 



t 




FIGURE 8 



✓ 



EP 0336 760 A2 




FIGURE 9 



EP 0 336760 A2 




This Page is Inserted by IFW Indexing and Scanning 
Operations and is not part of the Official Record 

BEST AVAILABLE IMAGES 

Defective images within this document are accurate representations of the original 
documents submitted by the applicant. 

Defects in the images include but are not limited to the items checked: 

□ BLACK BORDERS 

□ IMAGE CUT OFF AT TOP, BOTTOM OR SIDES 
□(FADED TEXT OR DRAWING 
[^BLURRED OR ILLEGIBLE TEXT OR DRAWING 

□ SKEWED/SLANTED IMAGES 

□ COLOR OR BLACK AND WHITE PHOTOGRAPHS 

□ GRAY SCALE DOCUMENTS 

□ LINES OR MARKS ON ORIGINAL DOCUMENT 

□ REFERENCE(S) OR EXHIBIT(S) SUBMITTED ARE POOR QUALITY 

□ OTHER: 

IMAGES ARE BEST AVAILABLE COPY. 
As rescanning these documents will not correct the image 
problems checked, please do not report these problems to 
the IFW Image Problem Mailbox.