Skip to main content

Full text of "USPTO Patents Application 09804625"

See other formats


PCX 



WORLD INTELLECTUAL PROPERTY ORGANIZATION 
International Bureau 




INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) 



(51) InternatioDa] Patent Qassification ^ : 

C12N 15/12, 5/10, C12P 21/08 
C07K 13/00, A61K 37/02 



Al 



(11) international Publication Number: 
(43) Inteniational Publication Date : 



WO 93/09228 

13 May 1993(13.05.93) 



(21) International Application Number: PCT/US92/09326 

(22) International Filing Date: 30 October 1992 (30.10.92) 



(30) Priority data: 
786,063 



31 October 1991 (31.10.91) US 



(71) AppDcant: WHITEHEAD INSTITUTE FOR BIOMEDI- 

CAL RESEARCH [US/US]; Nine Cambridge Center, 
Cambridge, MA 02142 (US). 

(72) Inventors: LIN, Herbert, Y. ; 550 Memorial Drive, Apt. 

12D1, Cambridge, MA 02139 (US). WANG, Xiao-Fan ; 
872 Massachusetts Avenue, No 401, Cambridge, MA 
02139 (US). WEINBERG, Robert, A ; 25 Copley Street, 
Brookline, MA 02146 (US). LODISH, Harvey, F. ; 195 
Fisher Avenue, Brookline, MA 02146 (US). 



(74) Agents: GRANAHAN, Patricia et al.; Hamilton, Brook, 
Smith & Reynolds, Two Militia Drive, Lexington. MA 
02173 (US). 



(81) Designated States: AU. CA, JP, European patent (AT, BE, 
CH. DE, DK, ES, FR, GB, GR, IE, IT,-LU, MC, NL, 
SE). 



Published 

H^th international search report. 



(54) Title: TGF.p TYPE RECEPTOR cDNAS AND USES THEREFOR 



(57) Abstract 



DNA encoding TGF-P type III receptor of mammalian origin, DNA encoding TGF-P type 11 receptor of mammalian ori- 
gin, TGF-P t^t III receptor, TGF-P type II receptor and uses therefor. 



FOR THE PURPOSES OF INFORMATION ONLY 

Codes used lo identify States party to the PCT on the front pages of pamphlets publishing international 
applications under the PCT. 



AT 


Aiotria 


FR 


AU 


Australia 


GA 


BB 


Barbados 


GB 


BE 


Bcleium 


GN 


BF 


Burkina Foso 


GR 


6C 


Bulgaria 


HU 


BJ 


Benin 


IE 


BR 


Bro/il 


rr 


CA 


Canada 


JP 


CF 


Cji:nlral Afrkan Republic 


KP 


CO 




KR 


CH 


Swiuirftand 


CI 


C'otc d'lvoin: 


KZ 


CM 


(*unicroun 


LI 


cs 


(.Vuchu&liivakia 


LK 


cz 


Orcli KcpubliL- 


I.U 


0£ 


Gcrmany 


MC 


DK 


Denmark 


MG 


ES 


Spain 


Ml. 


Fl 


Fintaml 


MN 



France 
Gabon 

Unucd Kingdom 

Guinea 

Greece 

Hungary 

Ireland 

Italy 

Japan 

Democratic HeopIe*s Republic 
or Korea 

Republic of Korea 

Ka/iikhstan 

Uechlenslcin 

Sri tunka 

Ijixcmbourg 

Monaco 

Madagascar 

Man 

Mongolia 



MR 


Mauritania 


MW 


Malawi 


NU 


Nelherlamb 


NO 


Norway 


HZ 


New Zealand 


PL 


Pokind 


FT 


Portugal 


RO 


Romania 


RU 


Ru^tan Federation 


SD 


Sudan 


SE 


Sweden 


SK 


Slovak Republic 


SN 


Senegal 


SU 


Soviet Union 


TD 


Chad 


TC 


Togo 


UA 


Ukraine 


US 


United Slates oT America 


VN 


Vict Nam 



N 



wo 93/09228 



PCT/US92/09326 



-1- 

TGF-jg TYPE RECEPTOR cDNAs AND USES THEREFOR 

Description 

Background 

Transforming growth factor-beta (TGF-^) is a member 

5 of a family of structurally related cytokines that elicit 
a variety of responses, including growth, differentia- 
tion, and morphogenesis, in many different cell types. 
(Roberts, A.B. and M.B. Sporn, In: Peptide Growth 
Factors and Their Receptors , Springer-Verlag, Heidelberg, 

10 pp. 421-472 (1990); Massague, J., Annu. Rev. Cell. Biol. 
6:597-641 (1990)) In vertebrates at least five different 
forms of TGF-^, termed TGF-^1 to TGF-^5, have been 
identified; they all share a high degree (60%-80%) of 
amino-acid sequence identity. While TGF-^1 was initially 

15 characterized by its ability to induce anchorage- 
independent growth of normal rat kidney cells, its 
effects on most cell types are anti-mitogenic. (Altschul, 
S.F. et al., J. Mol. Biol. 215:403-410 (1990); Andres, 
J.L. et al., J. Cell. Biol. 109 :3137-3145 (1989)). It is 

20 strongly growth-inhibitory for many types of cells, 

including both normal and transformed epithelial, endo- 
thelial, fibroblast, neuronal, lymphoid, and hemato- 
poietic cells. In addition, TGF-^ plays a central role 
in regulating the formation of extracellular matrix and 

25 cell-matrix adhesion processes. 

In spite of its widespread effects on cell phenotype 
and physiology, little is known about the biochemical 
mechanisms that enable TGF-^ family members to elicit 
these varied responses. Three distinct high-affinity 



wo 93/09228 



PCT/US92/09326 



-2- 



cell-sxirface TGF-^ -binding proteins, termed type 1, II 
and III, have been identified by incubating cells with 
radiolabelled TGF-^1, cross-linking bound TGF-pl to cell 
surface molecules, and analyzing the labelled complexes 
by polyacrylamide gel electrophoresis. (Hassague, J. and 

5 B. Like, J. Biol. Chem. 260 :2636-2645 (1985); Cheifetz, 
S. et aljj^ J. Biol. Chem. 261:9972-9978 (1986).) The 
binding constants are about 5-50pM for the type I and II 
receptor and 30-300 pM for the type III receptor. (Boyd, 
F.T. and J. Massague, J. Biol. Chem. 264:2272-2278 

10 (1989)) 

The type I and II receptors, of estimated 53 and 
70-100 kilodaltons mass respectively, are M-glycosylated 
transmembrane proteins that are similar in many respects. 
Each of these receptors has a distinct affinity for each 

15 member of the TGF-^ family of ligands. (Boyd, F.T. and 
J. Massague, J. Biol. Chem> 264 :2272-2278 (1989)) In 
contrast, the type III receptor shows comparable affin- 
ities for all TGF-p isotypes; the type III receptor is 
the most abundant cell-surface receptor for TGF-^ in many 

20 cell lines (upwards of 200,000 per cell), and is an 
integral membrane proteoglycan. It is heavily modified 
by glycosaminoglyccui (GAG) groups, and migrates hetero- 
geneously upon gel electrophoresis as proteins of 280 to 
330 kilodaltons. When deglycosylated with heparitinase 

25 and chondrontinase, the protein core migrates as a 
100-110 kilodalton protein. The TGF-^ binding site 
resides in this protein core, as non-glycosylated forms 
of this receptor that are produced in cell mutants 
defective in GAG synthesis are capable of ligand binding 

30 with affinities comparable to those of the natural 

receptor. (Cheifetz, S. and J. Massague, J. Biol. Chem. , 
264:12025-12028 (1989) A variant form of type III 



wo 93/09228 



PCr/US92/09326 



-3- 



receptor is secreted by some types of cells as a soluble 
molecule that apparently lacks a membrane anchor. This 
soluble species is found in low amounts in serum and in 
extracellular matrix. 

The type III receptor, also called betaglycan, has a 
5 biological function distinct from that of the type I and 

II receptors. Some mutant mink lung epithelial cell 
(HvlLu) selected for loss of TGF-^ responsiveness no 
longer express type I receptors; others, similarly 
selected, lose expression of both the type I and II 

10 receptors. However, all these variants continue to 
express the type III receptor. (Boyd, F.T. and J. 
Massague, J. Biol. Chem. 264 ; 2272-2278 (1989); Laiho, M. 
et al^, J. Biol. Chem. 265:18518-18524 (1990)) This has 
led to the proposal that types I and II receptors are 

15 signal -transducing molecules while the type III receptor, 
may subserve some other function, such as in concen- 
trating ligand before presentation to the bona fide 
signal-transducing receptors. The secreted form of type 

III receptor, on the other hand, may act as a reservoir 
20 or clearance system for bioactive TGF-^. 

Additional information about each of these TGF-^ 
receptor types would enhance our understanding of their 
roles and make it possible, if desired, to alter their 
functions. 

25 Summary of the Invention 

The present invention relates to isolation, sequen- 
cing and characterization of DNA encoding the TGF-^ type 
III receptor of mammalian origin and DNA encoding the 
type II receptor of mammalian origin. It also 
30 relates to the encoded TGF-^ type III and type II 

receptors, as well as to the soluble form of each; uses 



wo 93/09228 



PCrAJS92/09326 



-4- 



of the receptor-encoding genes and of the receptors them- 
selves; antibodies specific for TGF-^ type III receptor 
and antibodies specific for TGF-^ type II receptor. In 
particular, it relates to DNA encoding the TGF-^ type III 
receptor of rat and human origin, DNA encoding the TGF-^ 

5 type II receptor of human origin and homologues of each. 
The TGF-^ receptor-encoding DNA of the present 
invention can be used to identify equivalent TGF-^ 
receptor type III and type II genes from other sources, 
using, for example, known hybridization-based methods or 

10 the polymerase chain reaction. The type III receptor 
gene, the type II receptor gene or their respective 
encoded products can be used to alter the effects of 
TGF-p (e.g., by altering receptivity of cells to TGF-^ or 
interfering with binding of TGF-^ to its receptor) , such 

15 as its effects on cell proliferation or growth, cell 
adhesion and cell phenotype. For example, the TGF-^ 
receptor type III gene, the TGF-^ receptor type II gene, 
or a truncated gene which encodes less than the entire 
receptor (e.g., soluble TGF-^ type III receptor, soluble 

20 TGF-^ type II receptor or the TGF-^ type III or type II 
binding site) can be administered to an individual in 
whom TGF-^ effects are to be altered. Alternatively, the 
TGF-^ type III receptor, the TGF-^ type II receptor, a 
soluble form thereof (i.e., a fonn lacking the membrane 

25 anchor) or an active binding site of the TGF-^ type III 
or the type II receptor can be administered to an indivi- 
dual to alter the effects of TGF-^. 

Because of the many roles TGF-^ has in the body, 
availability of the TGF-p receptors described herein 

30 makes it possible to further assess TGF-^ function 
utilizing in vivo as well as in vitro methods and to 
alter (enhance or diminish) its effects. 



wo 93/09228 



PCr/US92/09326 



-5- 



Brief Description of the Drawings 

Figure 1 is the DNA sequence (SEQ ID NO. 1) and the 
translated amino acid sequence (SEQ ID NO, 2) of type III 
TGF-^l receptor cDNA clone R3-0FF (full insert size 6 
kb) , in which the open reading frame with flanking 

5 sequences of the clone are shown. The transmembrane 
domain is indicated by a single underline. Peptide 
sequences from purified type III receptor, mentioned in 
text, that correspond to the derived sequence, are in 
italics and underlined. Potential N-1 inked glycosylation 

10 sites are indicated by #, and extracellular cysteines by 
A consensus protein kinase C phosphorylation site is 
indicated by $ . The last non- vector encoded amino acid 
of Clone R3-0F (2.9 kb) is indicated by §. Consensus 
proteoglycan attachment site is indicated by +++. Other 

15 potential glycosaminoglycan attachment sites are 

indicated by +. The upstream in-frame stop codon (-42 to 
-44) is indicated by a wavy line. Signal peptide 
cleavage site predicted by vonHeijne's algorithm (von 
Heijne, G., Nucl. Acid. Res. 14; 4683-4690 (1986) is 

20 indicated by an arrow. 

Figure 2 is the nucleotide sequence of the full- 
length type II TGF-^ receptor cDNA clone 3FF isolated 
from a human HepG2 cell cDNA library (full insert size 5 
kb) (SEQ ID NO. 3). The cDNA has an open reading frame 

25 encoding a 572 amino acid residue protein. 

Figure 3 is the amino acid sequence of the full- 
length type II TGF-^ receptor (SEQ ID NO. 4). 

Detailed Description of the Invention 

The subject invention is based on the isolation and 
30 sequencing of DNA of vertebrate, particularly mammalian, 
origin which encodes TGF-^ type III receptor and DNA of 
mammalian origin which encodes TGF-p type II receptor, 



wo 93/09228 



PCr/US92/09326 



-6- 



expression of the encoded products and characterization 
of the expressed products. As described, a full-length 
cDNA which encodes TGF-^ receptor type III has been 
isolated from a cDNA library constructed from a rat 

5 vascular smooth muscle cell line and a full-length cDHA 
which encodes TGF-^ type II receptor has been isolated 
from a human cDNA library. The human homologue of the 
type III gene has also been cloned. A deposit of human 
TGF-^ type III CDNA in the plasmid pBSK has been made 

10 under the terms of the Budapest Treaty at the American 
Type Culture Collection (10/21/91) under Accession Number 
75127. All restrictions upon the availability of the 
deposited material will be irrevocably removed upon 
granting of a U.S. patent based on the subject 

15 application. 

Isolation and Characterization of TGF-g Type III 
Receptor 

As described herein, two separate strategies were 
pursued for the isolation of the TGF-^ type III receptor 

20 CDNA. In one approach, monoclonal antibodies were 

generated against the type III receptor protein and used 
to purify the receptor, which was then subjected to 
microsequencing. (See Example 1) Microsequencing of 
several peptides resulting from partial proteolysis of 

25 the purified receptor produced four oligopeptide 
sequences, which were used to construct degenerate 
oligonucleotides. The degenerate oligonucleotides were 
used either as primers in a cloning strategy using the 
polymerase chain reaction (PCR) or as probes in screening 

30 CDNA libraries. Although this strategy did not prove to 
be productive, the oligopeptide sequences were useful in 
verifying the identity of the receptor clones isolated by 
the second strategy. 



wo 93/09228 



PCT/US92/09326 



-7- 



In the second approach to isolating TGF-/5 receptor- 
encoding clones, an expression cloning strategy was used 
in COS cells; direct visualization of receptor positive 
cells was used to isolate receptor cDNAs. (See Example 
2) In this approach, a cDNA library was constructed from 

5 A-10 cells, a rat vascular smooth muscle cell line which 
expresses all three TGF-fi receptors (type I, II and III). 
COS cells transfected with cDNA components of this 
library in a vector carrying the cytomegalovirus (CMV) 
transcriptional promoter and the SV40 origin of repli- 

10 cation were screened to identify cells expressing sub- 
stantially higher than normal levels of TGF-^ receptor. 
One transfectant expressing such high levels of a TGF-^ 
binding protein was identified and the original pool of 
expression constructs from which it was derived was split 

15 into subpools, which were subjected to a second round of 
screening. Two further rounds of sib-selection resulted 
in isolation of one cDNA clone (R3-0F) with a 2.9 kb 
insert which induced high levels of TGF-^-binding 
proteins in approximately 10% of cells into which it was 

20 introduced. The specificity of the TGF-^ binding was 
validated by showing that addition of a 200-fold excess 
unlabeled competitor TGF-^1 strongly reduced binding of 
125 I-TGF-^ to transfected cells. 

The R3-0F cDNA encoded an open reading frame of 817 

25 amino acid residues, but did not contain a stop codon. 
R3-0F was used as a probe to isolate a full-length cDNA 
from a rat 208F library. The resulting clone, R3-0FF, is 
6kb in length and encodes a protein of 853 amino acids, 
which is colinear with clone R3-0F. The nucleotide 

30 seguence of R3-0FF is shown in Figure 1, along with the 
translated amino acid sequence. 



wo 93/09228 



PCr/US92/09326 



-8- 



Characterization of the receptor encoded by R3-0FF 
was carried out, as described in Example 3. Results . 
showed three distinct TGF-^ binding protein species of 
TGF-/? on the surface of mock-transfected COS cells, which 
is in accord with results reported by others. (Massague, 

5 J. et al., Ann. NY Acad. Sci. 593; 59-72 (1990)). These 
included the two lower molecular weight type I and II 
receptors (65 and 85 kD) and the higher molecular weight 
type III proteoglycan, which migrates as a diffuse band 
of 280-330 kd. Enzymatic removal of the proteoglycan 

10 yielded a core protein of approximately 100 kd. Binding 
to all three receptor types is specific in that it was 
competed by 200-fold excess of unlabeled TGF-^1. 

Transfccting the isolated cDNA caused a two-fold 
increase in expression of the type III receptor. When a 

15 cell lysate derived from COS cells transfected with clone 
R3-0FF was treated with deglycosylating enzymes, the 
heterogeneous 280-330 kd band was converted to a protein 
core which co-migrates with the type III protein core 
seen in parental AlO cells. Importantly, the recombinant 

20 protein core migrated differently from the endogenous COS 
cell type III protein core. 

These observations were confirmed and extended using 
stably transfected cells expressing the type III cDNA. 
L6 rat skeleton muscle myoblasts do not express any 

25 detectable type III mRHA and no endogeneous surface type 
III receptor (Massague et al^, 1986; Segarini et al., 
1989) . These cells were transfected with the isolated 
CDNA in the vector pcDNA-neo. Cell clones stably 
expressing this clone in both the forward and reverse 

30 orientations with respect to the CMV promoter were 
isolated and analyzed by ligand binding assay. 



wo 93/09228 



PCT/US92/09326 



-9- 



Introduction of either the full-length clone R3-OFF 
or the partial clone R3-0F in the forward orientation ^ 
resulted in expression of type III receptor. L6 cells 
transfected with the cDNA clones in the reverse orien- 
tation did not express this protein. Importantly, the 

5 apparent size of the protein core of the type III re- 
ceptor in cells transformed with the R3-0F clone is 
smaller than that from R3-0FF transformed cells, 
consistent with the difference in the sizes of the 
protein cores predicted from their nucleic acid 

10 sequences. 

Surprisingly, binding of radio-labeled ligand to the 
type II receptor was increased by 2.5 fold in cells 
expressing the type III cDNA. Binding to the type I 
receptor was unchanged. This apparently specific 

15 up-regulation of ligand-binding to the type II receptor 
was evident in all of the 15 stably transfected L6 cell 
lines analyzed to date. Furthermore, this effect seems 
to be mediated equally well by the full-length clone or a 
truncated clone (R3-0F} that lacks the cytoplasmic domain 

20 of TGF-^ type III receptor was expressed. 

Expression of type III receptor mRNA was assessed by 
Northern blot analysis and RNA blot analysis. Northern 
gel analysis showed that the type III receptor mRNA is 
expressed as a single 6 kb message in several rat 

25 tissues. RNA dot blot analysis of several different 

tissue culture cell lines was also carried out. Cells of 
mouse origin (MEL and YH16) appear to express a smaller 
(-5.5 kb) message for the type III mRNA than those of 
pig, rat and human origin. In all of these cells, 

30 expression or absence of the type III mRNA is consistent 
with the expression or absence of detectable cell surface 



wo 93/09228 



PCr/US92/09326 



-10- 



10 



type III receptors, with the notable exception of the 
retinoblastoma cell lines (Y79, Weri-1, Weri-24, and. 
Weri-27). These cells lack detectable surface expression 
of type III receptor, which confirms an earlier report. 
(Kimchi, A. et al., Science 240:196-198 (1988)). It is 
striking that the type III receptor mRNA is expressed in 
these cells at a level comparable to that of other cells 
that do indeed express type III receptor proteins at 
readily detectable levels. It appears that TGF-^ 
receptor III expression, which is substantial in normal 
retinoblasts (AD12) , has been down-regulated in these 
retinoblastoma tumor cells, perhaps through post- 
transcript ional mechanisms. 

The nucleotide sequence full reading frame along 
with flanking sequences of the full-length cDNA clone 
15 R3-0FF was determined and is presented in Figure 1. The 
reading frame encodes a protein of 853 amino acid 
residues, which is compatible with the 100 kD size 
observed for the fully deglycosylated TGF-^1 type III 
receptor. The identity of the receptor as TGF-^ type III 
was verified by searching for segments of the putative 
transcription product which included the peptide 
sequences determined by microsequencing of the isolated 
type III receptor. (See Example 1) As indicated in 
Figure l, two segments of derived protein (underlined and 
italicized, residues 378-388 and 427-434) precisely match 
with the amino acid sequences of two peptides (I and III) 
determined from direct biochemical analysis of the 
purified type III receptor. 

Further analysis showed that TGF-p type III binding 
protein has an unusual structure for a cytokine receptor. 
Hydropathy analysis indicates that the protein includes a 



20 



25 



30 



wo 93/09228 



PCr/US92/09326 



-11 



N-terminal signal sequence, followed by a long, hydro- 
philic N-terminal region. A 27 residue region of strong 
hydrophobicity (underlined in Figure 1, residues 786-812) 
toward the C-terminus represents the single putative 
transmembrane domain. This suggests that nearly all of 

5 the receptor which is an N-terminal extracellular domain 
is anchored to the plasma membrane near its C-terminus. 
A relatively small C-tenninal tail of 41 residues repre- 
sents the cytoplasmic domain. 

Analysis of related sequences provides few clues to 

10 function of TGF-^ type III protein. Only one other gene 
described to date, a glycoprotein expressed in high 
quantities by endothelial cells and termed endoglin, 
contains a related amino acid sequence. The most 
homologous regions between the sequences of the type III 

15 receptor and endoglin (74%) falls primarily in the 

putative transmembrane and cytoplasmic domains. Similar 
to the general structure of type III receptor, endoglin 
is a glycoprotein which contains a large hydrophilic 
N-terminal domain which is presumably extracellular, 

20 followed by a putative transmembrane domain and a short 
cytoplasmic tail of 47 amino acid residues. The bio- 
logical role of endoglin is still unclear at present, 
although it has been suggested that it may involved in 
cell-cell recognition through interactions of an "RGD" 

25 sequence on its ectodomain with other adhesion molecules. 
Unlike the TGF-^ type III receptor, endoglin does not 
carry GAG groups. 



wo 93/09228 



PCr/US92/09326 



-12- 



Tsolation of TGF-fl Typ* > H Receptor 
The cDNA encoding the type II TGF-/J receptor was 
also isolated, using expression cloning in COS cells. A 
full-length cDNA (designated clone 3FF) was isolated by 
high stringency hybridization from a human HepG2 cell 
5 CDNA library. (See Example 6) Analysis showed that the 
corresponding message is a 5 Kb message which is expres- 
sed in different cell lines and tissues. Sequence 
analysis indicated that the cDNA has an open reading 
frame encoding a core 572 amino acid residue protein. 
10 The nucleotide sequence of the full-length type II TGF-^ 
receptor cDNA clone 3FF is shown in Figure 2; the ammo 
acid sequence is represented in Figure 3. 

The 572 amino acid residue protein has a single 
putative transmembrane domain, several consensus glyco- 
ls sylation sites, and a putative intracellular serine/ 
threonine kinase domain. The predicted size of the 
encoded protein core is -60 kd, which is too large for a 
type I TGF-^ receptor. Instead, crosslinking experiments 
using iodinated TGF-^ and COS cells transiently trans- 
20 fected with clone 3FF shows over-expression of a protein 
approximately 70-80 kd which corresponds to the size of 
type II TGF-^ receptors. Thus, clone 3FF encodes a 
protein that specifically binds TGF-/J and has an ex- 
pressed protein size of 70-80 kd, both characteristic of 
25 the type II TGF-^ receptor. 

Ttees of the Cloned TGF-g R eceptors and Related 
Products 

For the first time, as a result of the work 
described herein, DMAs encoding two of the three high 
30 affinity cell-surface TGF-/9 receptors have been isolated, 
their sequences and expression patterns determined and 



wo 93/09228 



PCT/US92/09326 



-13- 



the encoded proteins characterized. Expression of the 
TGF-/3 type III receptor in cells which do not normally 
express the receptor, followed by ligand binding assay, 
verified that the cloned type III receptor-encoding DNA 
(i.e., either the full-length clone R3-0FF or the partial 

5 clone R3-0F) encoded the receptor. In addition, the work 
described herein resulted in the surprising finding that 
binding of TGF-^ to type II receptors in cells expressing 
the type III DNA was increased by 2.5 fold. 

Additional insight into the role of the TGF-/5 type 

10 III receptor and its interaction with TGF-p type II 

receptor is a result of the work described. For example, 
the role of TGF-^ type III receptor is unclear, but it 
has been proposed that it serves a most unusual function 
of attracting and concentrating TGF-^s for eventual 

15 transfer to closely situated signal-transducing 

receptors. While most cytokines bind to a single cell 
surface receptor, members of the TGF-^ family bind with 
greater or lesser affinity to three distinct cell surface 
proteins. This has raised the question of why these 

20 three receptors are displayed by most cell types and 
whether they subserve distinct functions. Evidence 
obtained to date suggests that the type III receptor may 
perform functions quite different from those of types I 
and II. Thus, type III is substantially modified by GAGs 

25 while types I and II appear to carry primarily the 
N-linked (and perhaps 0-lihked) sidechains that are 
characteristic of most growth factor receptors. In 
addition, variant cells that have been selected for their 
ability to resist TGF-^-induced growth inhibition show 

30 the absence of Type I or Type II receptors while 

continuing to display Type III receptors. Together, 
these data have caused some to propose that the Type I 



wo 93/09228 



PCr/US92/09326 



-14- 



10 



and II receptors represent bona fide signal-transducing 
receptors while the type III receptor, described here, 
plays another distinct role in the cell. 

It remains possible that the type III receptor 
serves a most unusual function of attracting and 
concentrating TGF-^s on the cell surface for eventual 
transfer to closely situated signal-transducing 
receptors. Such a function would be unprecedented for a 
proteinaceous receptor, although heparin sulfate has been 
shown to activate basic FGF by binding to this growth 
factor prior to FGF association with its receptor (Yayon, 
A. et al^. Cell 64; 841-848 (1991)) Parenthetically, 
since the type III receptor also contains large 
quantities of heparan sulfate side-chains, it may also 
bind and present basic FGF to its receptor. 
15 Evidence that is consistent with the role for the 

type III receptor comes from the work with L6 rat 
myoblast cells which is described herein. As described 
above, in L6 cells overexpressing type III receptor, the 
binding of radiolabelled TGF-p to the type II receptor is 
20 increased several fold when compared with that seen with 
parental cells. Further assessment of TGT-p type III 
function and interaction with type II and type I 
receptors will be needed to answer these questions and 
can be carried out using the materials and methods 

25 described here. 

TGF-^ receptors, both type III and type II, can be 
identified in other species, using all or a portion of 
the DNA encoding the receptor to be identified as a probe 
and methods described herein. For example, all or a 

30 portion of the DNA sequence encoding TGF-^ type III 

preceptor (shown in Figure 1) or all or a portion of the 



wo 93/09228 



PCr/US92/09326 



-15- 



DNA sequence encoding TGF-^ type II receptor (shown in 
Figure 2) can be used to identify equivalent sequences in 
other animals. Stringency conditions used can be varied, 
as needed, to identify eqpiivalent sequences in other 
species. Once a putative TGF-^ receptor type III or type 

5 Il-encoding sequence has been identified, whether it 
encodes the respective receptor type can be determined 
using known methods, such as described herein for 
verification that the cDNA insert of full-length clone 
R3-0FF and the cDNA insert of partial clone R3-0F encode 

10 the type III receptor. For example, DMA isolated in this 
manner can be expressed in an appropriate host cell which 
does not express the receptor mRNA or the surface 
receptor (e.g., L6 rat skeleton muscle myoblasts) and 
analyzed by ligand binding (TGF-^ binding) assay, as 

15 described herein. 

Also as a result of the work described herein, 
antibodies (polyclonal or monoclonal) specific for the 
cloned TGF-^ type III or the clones TGF-^ type II 
receptor can be produced, using known methods. Such 

20 antibodies and host cells (e.g., hybridoma cells) 

producing the antibodies are also the subject of the 
present invention. Antibodies specific for the cloned 
TGF-^ receptor can be used to identify host cells 
expressing isolated DNA thought to encode a TGF-/5 

25 receptor. In addition, antibodies can be used to block 
or inhibit TGF-^ activity. For example, antibodies 
specific for the cloned TGF-^ type III receptor can be 
used to block binding of TGF-)9 to the receptor. They can 
be administered to an individual for whom reduction of 

30 TGF-^ binding is desirable, such as in some fibrotic 
diseases (e.g., of skin, kidney and lung). 



wo 93/09228 



PCT/US92/09326 



16 



The method of the present invention can be used for 
diagnosis of disorders involving abnormal binding of . 
TGF-^ to TGF-fi type III receptors and/or TGF-^ type II 
receptors, such as fibrotic diseases. Abnormal binding 
of TGF-^ to TGF-^ type III receptor or TGF-^ type II 

5 receptor at a cell surface may be measured, resulting in 
a test binding value, which is compared to an appropriate 
control binding value. Control binding values can be 
obtained using control cells known to have abnormal 
binding of TGF-^ to its receptors or control cells which 

10 are normal cells (e.g., evidence TGF-^ binding to the 
TGF-^ receptor is within physiological levels). Control 
values are obtained by determining the extent to which 
TGF-p binds the appropriate receptor (i.e., TGF-^ type 
III receptor or TGF-^ type II receptor) ; such values can 

15 be obtained at the time the test binding value is 
determined or can be previously determined (i.e., a 
previously determined standard) . A test binding value 
similar to the control binding value obtained from 
abnormal cells is indicative of abnormal binding of TGF-^ 

20 to TGF-^ type III receptor or TGF-^ type II receptor. A 
test binding value similar to the control binding value 
obtained from normal cells is indicative of normal 
binding of TGF-^ to TGF-^ type III receptor or TGF-A type 
II receptor. 

25 DNA and RNA encoding TGF-^ type III receptor and DNA 

and RNA encoding TGF-p type II receptor are now 
available. As used herein, the term DNA or RNA encoding 
the respective TGF-^ receptor includes any 
oligodeoxynucleotide or oligodeoxyribonucleotide sequence 

30 which, upon expression, results in production of a TGF-^ 
receptor having the functional characteristics of the 



wo 93/09228 



PCT/US92/09326 



-17- 



TGF-^ receptor. That is, the present invention includes 
DNA and RNA which, upon expression in an appropriate host 
cell, produces a TGF-^ type III receptor which has an 
affinity for TGF-^ similar to that of the TGF-p type III 
receptor on naturally occurring cell surfaces (e.g., it 

5 shows comparable affinities for all TGF-fi isotypes) • 
Similarly, the present invention includes DNA and RNA 
which, upon expression in an appropriate host cell, 
produces a TGF-p type II receptor which has an affinity 
for TGF-^ similar to that of TGF-^ type II receptor on 

10 naturally occurring cell surfaces (e^g., it has a 

distinctive affinity for each member of the TGF-^ family 
of ligands similar to that of the naturally occurring 
TGF-^ type II receptor) . The DNA or RNA can be produced 
in an appropriate host cell or can be produced 

15 synthetically (e.g., by an amplification technique such 
as PCR) or chemically. 

The present invention also includes the isolated 
TGF-)9 type III receptor encoded by the nucleotide 
sequence of full-length R3-0FF, the isolated TGF-^ type 

20 III receptor encoded by the nucleotide sequence of 

partial clone R3-0F, the isolated TGF-p type II receptor 
encoded by the nucleotide sequence of full-length clone 
3FF and TGF-^ type III and type II receptors which bind 
TGF-^ isotypes with substantially the same affinity. The 

25 isolated TGF-^ type III and type II receptors can be 

produced by recombinant techniques, as described herein, 
or can be isolated from sources in which they occur 
naturally or synthesized chemically. As used herein, the 
terms cloned TGF-^ type III and cloned TGF-^ type II 

30 receptors include the respective receptors identified as 



wo 93/09228 



PCr/US92/09326 



-18- 



10 



described herein, and TGF-^ type III and type II 
receptors (e.g., from other species) which exhibit - 
substantially the same affinity for the TGF-p isotypes as 
the respective receptors. 

AS described previously, cells in which the cloned 
TGF-p type III receptor is expressed bind TGF-p in 
essentially the same manner as do cells on which the type 
III receptor occurs naturally. Further analysis of 
ligand interactions with the cloned TGF-/J type III 
receptor, based upon site-directed mutagenesis of both 
TGF-^ and the receptor, can be carried out to identify 
residues important for binding. For example, DNA having 
the sequence of Figure 1 can be altered by adding, 
deleting or substituting at least one nucleotide, in 
order to produce a modified DNA sequence which encodes a 
modified cloned T6F-/J type III receptor. The functional 
characteristics of the modified receptor (e.g., its 
TGF-^-binding ability and association of the binding with 
effects normally resulting from binding) can be assessed, 
using the methods described herein. Modification of the 
cloned TGF-fi type III receptor can be carried out to 
produce, for example, a form of the TGF-p type III 
receptor, referred to herein as soluble TGF-/J receptor, 
which is not membrane bound and retains the ability to 
bind the TGF-^ isotypes with an affinity substantially 
25 the same as the naturally-occurring receptor. Such a 
TGF-/> type III receptor could be produced, using known 
genetic engineering or synthetic techniques; it could 
include none of the transmembrane region present in the 
naturally-occurring TGF-^ type III receptor or only a 
small portion of that region (i.e., small enough not to 



15 



20 



30 



wo 93/09228 



PCT/US92/09326 



19- 



interf ere with its soluble nature) . For example, it can 
include amino acids 1 through 785 of the TGF-^ type III 
sequence of Figure 1 or a portion of that sequence 
sufficient to retain TGF-^ binding ability (e.g., amino 
acids 24-785, which does not include the signal peptide 

5 cleavage site present in the first 23 amino acids) . A 
soluble TGF-p type II receptor (e.g., one which does not 
include the transmembrane and cytoplasmic domains) can 
also be produced. For example, it can include amino 
acids 1 through 166, inclusive, of Figure 3 or a 

10 sufficient portion thereof to retain TGF-^ binding 

ability substantially the same as that of TGF-^ type II 
receptor. 

The TGF-^ type III receptor and /or type II receptor 
can be used for therapeutic purposes. As described above, 

15 the TGF-p family of proteins mediate a wide variety of 
cellular activities, including regulation of cell growth, 
regulation of cell differentiation and control of cell 
metabolism. TGF-^ may be essential to cell function and 
most cells synthesize TGF-^ and have TGF-^ cell surface 

20 receptors. Depending on cell type and environment, the 
effects of TGF-^ vary: proliferation can be stimulated 
or inhibited, differentiation can be induced or 
interrupted and cell functions can be stimulated or 
suppressed. TGF-^ is present from embryonic stages 

25 through adult life and, thus, can affect these key 
processes throughout life. The similarities of a 
particular TGF-^ (e*g*i TGF-^1) across species and from 
cell to cell are considerable. For example, the amino 
acid sequence of a particular TGF-^ and the nucleotide 

30 sequence of the gene which encodes it regardless of 



wo 93/09228 



PCT/US92/09326 



-20- 

source, are essentially identical across species. This 
further suggests that TGF-^ has a critical role in 
essential processes. 

Specifically, TGF-p has been shown to have anti- 
inflamnatory and immune suppression capabilities, to play 

5 an important role in bone formation (by increasing 

osteoblast activity) , inhibit cancer cell proliferation 
in culture, and control proliferation of glandular cells 
of the prostate. As a result, it has potential thera- 
peutic applications in altering certain immune system 

10 responses (and possibly in modifying immune-mediated 
diseases); in treating systemic bone disease (e.g., 
osteoporosis) and conditions in which bone growth is to 
be enhanced (e.g., repair of broken bones) and in con- 
trolling growth and metastasis of cancer cells. In 

15 addition, TGF-^ appears to play a role in determining 
whether some cell types undergo or do not undergo mito- 
sis. In this respect, TGF-^ may play an important role 
in tissue repair. Some diseases or conditions appear to 
involve low production or chronic overproduction of 

20 TGF-p. (For example, results of animal studies suggest 
that there is a correlation between the over production 
of TGF-^ and diseases characterized by fibrosis in the 
lung, kidney, liver or in viral mediated immune 
expression. ) 

25 Clearly, TGF-^ has key roles in body processes and 

numerous related potential clinical or therapeutic 
applications in wound healing, cancer, immune therapy and 
bone therapy. Availability of TGF~fi receptor genes, the 
encoded products and methods of using them in vitro and 

30 in vivo provides an additional ability to control or 



wo 93/09228 



PCr/lJS92/09326 



regulate TGF-^ activity and effect in the body. For 
example, the TGF-^ type II or type III receptor encoded 
by the type II or the type III receptor genes of the 
subject invention can be used, as appropriate, to alter 
the effects of TGF-fi (e.g., to enhance the effect of 

5 TGF-^ in the body or to inhibit or reduce (totally or 
partially) its effects) . It is also possible to admin- 
ister to an individual in whom TGF-^ bound to TGF-^ type 
III receptor, such as soluble TGT~fi type III receptor. 
The present invention provides both a TGF-p agonist and a 

10 TGF-p antagonist. For these purposes, DNA gene encoding 
the entire TGF-^ type II or type III receptor, the 
encoded type II or type III receptor or a soluble form of 
either receptor can be used. Alternatively, antibodies 
or other ligands designed based upon these sequences or 

15 specific for them can be used for this purpose. 

Knowledge of the amino acid sequences of TGF-^ type 
III and type II receptors makes it possible to better 
understand their structure and to design compounds which 
interfere with binding of the receptor with TGF-^. It 

20 makes possible identification of existing compounds and 
design of new compounds which are type III and/or type II 
receptor antagonists. 

Cells expressing the type III and/or type II recep- 
tors of the present invention can be used to screen 

25 compounds for their ability to interfere with (block 

totally or partially) TGF binding to the receptors. For 
example, cells which do not express TGF-^ type III 
receptor (e.g., L6 rat skeleton muscle myoblasts) but 
have been modified to do so by incorporation of the type 

30 III cDNA in an appropriate vector can be used for this 



wo 93/09228 



PCr/US92/09326 



-22- 



10 



purpose. A compound to be assessed is added, for 
example, to tissue culture dishes containing type III. 
expressing cells, along with labeled T6F-^. As a 
control, the same concentration of labeled TGF-/J is added 
to tissue culture dishes containing the same type of 
cells. After sufficient time for binding of TGF-^ to the 
receptor to occur, binding of labeled TGF-fi to the cells 
is assessed, using knovm methods (e.g., by means of a 
gamma counter) and the extent to whcih it occurred in the 
presence and in the absence of the compound to be 
assessed is determined. Comparison of the two values 
show whether the test compound blocked TGF-^ binding to 
the receptor (i.e., less binding in the presence of the 
compound than in its absence is evidence that the test 
15 compound has blocked binding of TGF-/> to the TGF-/J type 

III receptor) . 

Alternatively, a cell line expressing the TGF-^ 
receptor or cells expressing microinjected TGF-p receptor 
RNA, can be used to assess compounds for their ability to 
20 block TGF-^ binding to the receptor. In this embodiment, 
a compound to be assessed is added to tissue culture 
dishes containing the cell line cells expressing 
microinjected TGF-^ receptor RNA, along with TGF-/J. As a 
control, TGF-^ alone is added to the same type of cells 
25 expressing microinjected endothelin receptor RNA. After 
sufficient time for binding of TGF-p to the receptor to 
occur, the extent to which binding occurred is measured, 
both in the presence and in the absence of the compound 
to be assessed. Comparison of the two values shows 
whether the compound blocked TGF-^ binding to the 
receptor. The TGT-p type III and type II receptors can 



30 



wo 93/09228 



PCT/US92/09326 



-23- 

also be used to identify TGF-^-like substances, to purify 
TGF-p and to identify TGF-^ regions which are respons-ible 
for binding to the respective receptors. For example, 
the type III receptor can be used in an affinity-based 
5 method to identify substances which bind the receptor in 
a manner similar to TGF-p. 

The invention will now be illustrated by the 
following examples, which are not intended to be limiting 
in any way. 

10 EXAMPLES 

Materials and methods used in Examples 1-5 are 
described below. 

Materials 

The following is a description of materials used in 

15 the work described herein. 

Recombinant human TGF-^1 was provided by Rik Derynck 
of Genentech. C0S-M6 cells were provided by Brian Seed 
of the Massachusetts General Hospital and Alejandro 
Aruffo of Bristol-Myers-Squibb. Heparitinase was pro- 

20 vided by Tetsuhito Kojima and Robert Rosenberg of MIT. 
LLC-PKj^ cells were a gift of Dennis Ausiello of the 
Massachusetts General Hospital. YH-16 cell were a gift 
of Edward Yeh of the Massachusetts General Hospital. 3-4 
cells were a gift of Eugene Kaji of the Whitehead 

25 Institute for Biomedical Research. All other cell lines 
were purchased from ATCC and grown as specified by the 
supplier, except as noted. 



wo 93/09228 



PCr/US92/09326 



24- 



Expression Cloning 

Construction of cDNA Library and Generation of • 

Plasmid Pools 

10/ig polyadenylated mRNA was prepared from AlO cells 

5 by the proteinase-K/SDS method (Gonda et al. , Molec. 

cell. Biol. 2; 617-624 (1982)). Double stranded cDNA was 
synthesized and linkered to nonpalindromic BstXl adaptors 
as described by Seed, B. and A. Aruffo, Proc. Natl. Acad. 
Sci. PSA 84:3365-3369 (1987). Acaptored cDNA was size- 

10 fractionated on a 5 to 20% potassium acetate gradient, 
and inserts greater than 1 kb were ligated to the plasmid 
vector pcDNA-1, and electroporated in the E. coli 
MC1061/P3, yielding a primary library with a titer of 
>lo'' recombinants. A portion of the cDNA was plated as 

15 pools of -1x10^ recombinant bacteria colonies grown on 15 
cm petri dishes with Luria-broth agar containing 7.5 
mg/ml tetracycline and 12.5 mg/ml ampicilliri. Cells were 
scraped off the plates in 10 mis of Luria-broth, and 
glycerol stocks of pooled bacteria were stored at -70«C. 

20 The remaining bacteria was used to purify plasmid DNA 
using the alkaline lysis mini-prep method (Sambrook, J. 
et al.. Molecular Cloning: A Laboratory Manual, 2d Ed. 
(Cold Spring Harbor, NY, Cold Spring Harbor Laboratory 
Press (1989)}. 

25 COS Cell Transfections and Binding Assay 

Plasmid pools (each representing -ixlO* clones) were 
transfected into C0S-M6 (subclone of COS-7 cells) by the 
DEAE-dextran/chloroquine method described by Seed, B. and 
A. Aruffo, Proc. Natl. Acad. Sci. USA 84:3365-3369 

30 (1987). Forty-eight hours after transfection, cells were 



wo 93/09228 



PCr/US92/09326 



-25- 



incubated with 50 pMizsi-TGF-^l (100 to 200 Ci/mmol) for 
4 hours at 4«C), autoradiographic analysis of transfected 
cells was performed using NT-B2 photographic emulsion 
(Kodak) essentially as described (Gearing, D.P. et al. , 

5 EMBO J« 8:3667-3676 (1989)). After development of 

slides, cells were air-dried and mounted with mounting 
media and a glass coverslip. Slides were analyzed under 
an Olympus OM-Tl inverted phase-contrast microscope using 
a dissection trans-illuminator for darkfield illumi- 

10 nation. 

Subdivision of Positive Pool 

Of 86 pools screened, one pool (#13) was identified 
as positive and a glycerol stock of bacteria corres- 
ponding to this pool was titered and 25 pools of 1000 

15 clones were generated and miniprep plasroid from these 

pools were transfected into COS cells as described above. 
Several positive pools of 1000 were identified, and one 
was replated as 25 plates of 100 colonies. A replica was 
made of this positive plate and harvested. Once a 

20 positive pool was identified, individual colonies were 
picked from the corresponding roaster plate and grown 
overnight in 3 ml liquid culture. A 2-dimensional grid 
representing the 100 clones was generated and a single 
clone, R3-OF, was isolated. 

25 Cloning of R3-0FF 

A 208F rat fibroblast library in lambda ZAP II 
(Stratagene) was screened at high stringency with clone 
R3-0F insert, and several clones with -6kb inserts were 
isolated, one of which is referred to as R3-OFF. 



wo 93/09228 



PCT/US92/09326 



-26- 



10 



15 



20 



DNA Sequencing and Sequence Analysis 
Double-stranded DNA was sequenced by the dideoxy. 
chain termination method using Sequenase reagents (United 
States Biochemicals) . Comparison of the sequence to the 
data bases was performed using BLAST (Altschcul, S.F. et 
al., J. Mol. Biol. 215:403-410 (1990)). 

lodination of TGF-^ 

TGF-/J1 was iodinated using the chloramine-T method 
as described (Cheifetz, S. and J.L. Andres, J. Biol. 
Chem. 263:16984-16991 (1988)). 



Chemical Cross-Linking 

Transfected COS cells grown on 10 cm dishes or 
subconfluent L6 and A-10 cells grown on 3.5 cm dishes 
were incubated with i2 5i-TGF-pi in binding buffer 
(Frebs-Ringer buffered with 20 mM Hepes, pH 7.5, 5 mM 
MgSO., 0.5% BSA), washed 4 times with ice-cold binding 
buffer without BSA, and incubated for 15 minutes with 
binding buffer without BSA containing 60ng/ml disuc- 
cinimidyl subcrate at 4»C under constant rotation. 
Crosslinking was terminated by addition of 7* sucrose in 
binding buffer. Cells were scraped, collected and 
pelleted by centrifugation, then resuspended in lysis 
buffer (10 mM Tris, pH 7.4, 1 mM EDTA, pH 8.0, 1% 
Triton-X 100, 10 /ig/ml of pepstatin, lOpg/ml leupeptin, 
25 10 itg/ml antipain, 100 /ig/m; benzamidine hydrochloride, 
100 ,ig/ml soybean trypsin inhibitor, 50 ^g/ml aprotonin, 
and 1 mM phenylmethylsulfonyl fluoride) . Solubilized 
material was analyzed by 7% SDS-PAGE and subjected to 



wo 93/09228 



PCT/US92/09326 



-27- 



autoradiographic analysis by exposure to X-AR film 
(Kodak) at -70^C. 

Enzymatic Digestion 

Digestion of solubilized TGF-b receptors with 
5 chondroitinase and heparitinase was performed as 

described (Cheifetz, S. and J.L. Andres, J. Biol. Chem. 
263:16984-16991 (1988); Segarini, P.R. and S.M, Seyedin, 
J. Biol, Chem, , 263 : 8366-8370 (1988). 

Generation of Stable Cell Lines 

10 . L6 myoblasts were split 1:10 into 10 cm dishes and 

transfected the following day by the calcium phosphate 
method (Chen, C, and H. Okayama, Molec, Cell. Biol. 
7:2745-2752 (1987)) with clones R3-0F or R3-0FF in the 
forward and reverse orientations in the vector pcDNA-neo 

15 (InVitrogen) . Cells were subjected to selection in the 
presence of G418 (Geneticin, GIBCO) for several weeks 
until individual colonies were visible by the naked eye. 
These clones were isolated and amplified. 

RNA Blot Analyses 

20 Rat tissue polyadenylated mRNA was prepared by the 

lithium chloride/urea method (Auffrey, C. and F. Raugeon, 
Eur. J, Biochemistry 107:303-313 (1980), followed by 
oligo-dT cellulose chromatography (Aviv and Leder, 1972) . 
Polyadenylated mRNA from cell lines was prepared by the 

25 proteinase K/SDS method (Gonda, T-J. et al., Molec. Cell. 
Biol. 2:617-624 (1982)). Samples of mRNA were resolved 
by electrophoresis on 1% agarose-2.2M formaldehyde gels, 
blotted onto nylon membranes (Biotrns, ICN) and incubated 



wo 93/09228 



PCr/US92/09326 



28 



With the 2.9 kb insert of clone Re-OF labeled with ssp by 
random priming as probe (Sambrook, J. et al.. Molecular 
Cloning: A Laboratory Manual, 2d Ed., Cold Spring 
Harbor, NY, Cold Spring Harbor Laboratory Press, (1989). 
5 Hybridizations were performed at 42 -C in hybridization 
buffer containing 50% formamide overnight, and blots were 
washed at 55 'C in 0.2X SSC, 0.1% SDS, before exposure to 
X-AR film at -70«»C. 

Example 1 . Production of Anti-Typ e III Receptor Protein 
Antibodies and Microsecruenc ino and Micro- 
seouencing of Peptides Resulti ng from Partial 
Proteolysis of Purified Type III Receptor 
Initially cellular proteoglycans were purified from 
human placenta and then subjected to enzymatic deglycosy- 

15 lation with heparitinase and chondroitinase. Protein 
cores in the molecular weight range of 100-130 kilo- 
daltons were further purified by preparative gel electro- 
phoresis; these should include the type III receptor. 
This partially purified material was used as an immunogen 

20 in mice. After screening 850 hybridoma lines developed 
from immunized mice, three lines were found to produce 
antibodies that specifically recognized and immuno- 
precipitated a deglycosylated polypeptide species of 
100-120 kD. This species could be radiolabelled by 

25 incubation of whole cells with "'l-TGF-p followed by 

covalent cross-linking. Its size is consistent with that 
of the protein core previously reported for the type III 
receptor. (Massague, J., Annu. Rev . Cell. Biol. 
6:597-641 (1990)) 



wo 93/09228 



PCr/US92/09326 



-29- 



Monoclonal antibody 94 was used to purify the type 
III receptor from rat liver by af f inity-chromatography. 
The purified receptor was subjected to partial proteo- 
lysis and the resulting peptides were resolved by high 
5 pressure liquid chromatography. Several peptides were 
subjected to microsequencing and yielded the following 
oligopeptide sequences: 

Peptide I: ILLDPDHPPAL (SEQ ID NO. 5) 

Peptide II: QAPFPINFMIA (SEQ ID NO. 6) 

10 Peptide III: QPIVPSVQ (SEQ ID NO. 7) 

Peptide IV: FYVEQGYGR (SEQ ID NO. 8) 

These peptide sequences were used to construct 
degenerate oligonucleotides that served either as primers 
In a cloning strategy using the polymerase chain reaction 

15 (PCR) or as probes in screening cDNA libraries. While 
this strategy was not productive, the oligopeptide 
sequences proved useful in verifying the receptor clones 
isolated by the second, alternative strategy (see Example 

- 2). 

20 Example 2 . Expression Cloning of the Ty pe III Receptor 
cDNA 

An expression cloning strategy in COS cells, a 
procedure which takes advantage of the considerable 
amplification of individual cDNAs in transfected COS 
25 cells was used as an alternative means to isolate TGF-^ 
receptor clones. Such amplification is mediated by SV40 
large T antigen expressed by the- COS cells interacting 



wo 93/09228 



PCr/US92/09326 



-30- 



10 



15 



with a SV40 origin of replication in the cDNA vector. 
Gearing, D. et al^, embO J. 8; 3667-3676 (1989); Lin, • 
H.Y., et al:., Proe. Natl. A cad. Sci. 88;3185-3189 
(199la); Lin, H. Y. et al^, Science , in press (1991); 
Mathews, L. S. and Vale, W. W . , Cell 65:973-982 (1991). 

The strategy involved the construction of a cDMA 
library from A-10 cells, a rat vascular smooth muscle 
cell line that expresses all three high-affinity TGF-fi 
receptors. The resulting cDNAs were inserted into the 
vector pcDNA-1, which carries the CMV transcriptional 
promoter and the SV40 origin of replication. The result- 
ing library was then divided into pools of 10,000 inde- 
pendent recombinants each and DNA from each pool was 
transfected into 1.5 x 10*COS-7 cells grown on glass 
f laskettes by means of DEAE-dextran transfection proce- 
dure. Aruffo, A. and Seed, B., Proc. Hatl. Acad. Sci., 
U.S.A. 84:8573-8577 (1987); Gearing, D. et ali, IMBO J. 
8:3667-3676 (1989); Mathews, L. S. and Vale, W. W., Cell 
^:973-982 (1991). The transfected cells were cultured 
for 48-60 hours and then exposed to radiolabelled TGF-^1 
for four hours. Following this treatment, the glass 
slides carrying these cells were washed extensively and 
fixed. These slides were dipped in liquid photographic 
emulsion and examined by darkfield microscopy. While all 
of the receptor genes cloned to date by this procedure 
25 have undetectable or low constitutive levels of expres- 
sion in COS cells, we were hindered by the fact that 
untransfected COS cells already express substantial 
amounts of type III TGF-^ receptor. Such expression, 
estimated to be approximately 2 x 10* receptor molecules 
per cell, might well have generated an unacceptably high 
level of background binding. However, since the 
detection procedure involves visualizing radiolabelled 



20 



30 



wo 93/09228 



PCT/US92/09326 



31- 



ligand-Wnding on individual cells, it was hoped that 
identifying occasional cells expressing substantially 
higher levels of vector-encoded receptor would be 
possible. This hope was vindicated in the initial 
experiments. 

5 After screening nearly one million cDNA clones in 

this manner, a glass slide containing 20 positive trans- 
fectants was identified. The original pool of expression 
constructs from which one such trans feet ant was derived 
was split into 25 subpools of 1000 clones each and these 

10 were subjected to a second round of screening. Two 

further rounds of sib-selection resulted in the isolation 
of a cDNA clone (R3-0F) with a 2.9 kb insert that induced 
high levels of TGF-^ -binding proteins in approximately 
10% of COS cells into which it was transfected. 

15 The specificity of this binding was validated by 

showing that addition of a 200-fold excess of unlabeled 
TGF-0 competitor strongly reduced binding of ***I-TGF-/3 
to transfected cells. By taking into account a 
transfection efficiency of 10% and the high background of 

20 endogenous receptor expression, we calculated that the 
level of total ^**I-TGF-^ binding to each glass slide of 
cells transfected with this cDNA clone (Figure IC) was 
- only 2-fold above the level seen with mock transf ectants 
(data not shown). Nonetheless, this marginal increase in 

25 ligand-binding was sufficient to identify rare trans- 
fectants amidst a large field of cells expressing this 
background level of receptor. 

The R3-or cDNA encoded an open reading frame; of 836 
amino acid residues of which the 3' most 18 were encoded 

30 by vector sequence, clearly indicating that clone R3-0F 



wo 93/09228 



PCT/US92/09326 



-32- 



was an incomplete cDNA insert which ended prematurely at 
the codon specifying alanine 818 (Figure 4). R3-0F was 
used as a probe to isolate a full-length cDNA from a rat 
208F lambda phage library. This clone, termed R3-0FF, 
5 was 6 kb in length and encoded a protein of 853 amino 
acids; its sequence was co-linear with that of clone 
R3-0F. 

Example 3 , characterizat ion of the Product of the Full 
len gth Clone R3-0FF 

10 Characterization of the product of the full length 

clone R3-0FF was undertaken in order to determine which 
of the three TGF-^ receptors it specified. To do so. COS 
transfectants were incubated with radioiodinated TGF-^, 
chemical crosslinker was added and the labelled receptors 

15 were resolved by polyacrylamide gel electrophoresis. 

Labelling of cell surface TGF-^ receptors in this 
way resulted in the detection of three distinct species 
on the surface of COS cells, as extensively by others 
(Massague, J. et al^, Ann. NY Acad. Sci. 593:59-72 

20 (1990) . These included the two lower molecular weight 
type I and II receptors (65 and 85 kD) and the higher 
molecular weight type III proteoglycan, which migrated as 
a diffuse band of 280-330 kd. Enzymatic treatment of the 
proteoglycan with chondroitinase and heparitinase yielded 

25 a core protein of approximately 100 kd. Binding to all 
three receptor types was specific, in that it was 
completed by 200-fold excess of unlabeled TGF-^1. 

Transfecting the R3-0FF cDNA caused a two-fold 
increase in expression of the type III receptor. When a 

30 cell lysate derived from COS cells transfected with clone 



wo 93/09228 



PCT/US92/09326 



-33- 



R3-orF was treated with deglycosylating enzymes, the 
heterogenous 280-330 kd band was converted to a protein 
core which co-migrated with the type III protein core 
seen in untransf ected AlO cells. Importantly, the 
recombinant protein core migrates differently from the 

5 endogenous COS cell type III protein core. 

These observations were confirmed and extended in 
experiments using stably transfected cells expressing the 
R3-0FF cDNA. L6 rat skeleton muscle myoblasts normally 
do not express detectable type III mRNA or endogenous 

10 type III receptor (determined by radiolabelled 

ligand-binding assay) . Such cells were transfected with 
the isolated cDNA in the vector pcDNA*neo. Cell clones 
stably expressing this clone in both the forward and 
reverse orientations with respect to the CMV promoter 

15 were isolated and analyzed by ligand-binding assay. 

Introduction of either the full length clone R3-0FF 
or the partial clone R3-0F in the forward orientation led 
to the de novo expression of the type III receptor. L6 
cells transfected with the cDNA in reversed orientation 

20 did not express this protein. The apparent size of the 
protein core of the type III receptor in cells 
transfected with the R3-0F clone is smaller than that 
expressed by R3-0FF transfected cells, consistent with 
the difference in the sizes of the protein cores 

25 predicted from the respective nucleic acid sequences 
(Figure 1) . 

Unexpectedly, the amount of radio-labelled ligand 
corss-linked to the type II receptor is increased by 2.5 
fold in cells expressing the type III cDNA, while the 
30 amount cross-linked to the type I receptor remained 



wo 93/09228 



PCr/US92/09326 



-34- 



unchanged. This apparent specific up-regulation of 
ligand-binding to the type II receptor could be detected 
with all of the 15 stably transfected L6 cell lines 
analyzed so far. This effect seems to be mediated by the 
5 truncated clone R3-OF which lacks the cytoplasmic domain 
as well as by the full-length clone R3-0FF. 

Example 4 . Bv pression o ^ Type III Receptor 

Northern blot analysis demonstrated that the type 
III receptor mRNA is expressed as a single 6 kb message 

10 in several rat tissues. The level of mRNA expression in 
the liver was less than in other tissues, a result 
expected from earlier surveys of various tissues using 
radioiodinated TGF-^1. Based on this information, it 
appears that clone R3-0FF, with a -6 kb cDNA insert, 

15 represents a full length rat type III cDNA clone. 

cells of mouse origin (MEL and YH16) appear to 
express a smaller (-5.5 kb) message for the type III mRNA 
than those of pig, rat and human origin. In all of these 
cells, expression or absence of the type III mRNA is 

20 consistent with the expression or absence of detectable 
cell surface type III receptors with the notable 
exception of the retinoblastoma cell lines (¥79, Weri-1, 
Weri-24, and Weri-27). These cells have previously been 
shown to lack detectable surface expression of type III 

25 receptor, a result confirmed by our own unpublished work. 
It is striking that the type III receptor mRNA is 
expressed in these cells at a level comparable to that of 
other cells that do indeed express type III receptor 
proteins at readily detectable levels. At this moment, 

30 we can only conclude that TGF-/> receptor III expression. 



wo 93/09228 



PCr/US92/09326 



which is substantial in normal retinoblasts (AD12) , has 
been down-regulated in these retinoblastoma tumor cells, 
perhaps through post-transcriptional mechanisms. 

Example 5 . Sequence Analysis of the Full-Length Type 

5 III cDNA 

The full-length cDNA clone (R3-orF) , described in 
Example 4, was subjected to sequence analysis. The full 
reading frame along with flanking sequences is presented 
in Figure 1. This reading frame encodes a protein of 853 

10 amino acid residues, which is compatible with the 100 kD 
observed for the fully deglycosylated TGF-^ type III 
receptor. 

Two segments of derived protein sequence (underlined 
and italicized, residues 378-388 and 427-434) precisely 

15 match those determined earlier from direct biochemical 
analysis of the purified receptor protein. This further 
secured the identity of this isolated cDNA clone as 
encoding the rat type III receptor. 

This TGF-^ binding protein has an unusual structure 

20 for a cytokine receptor. Hydropathy analysis indicates a 
N-terminal signal sequence, followed by a long, 
hydrophilic N-terminal region (Kyte, J. and R. F. 
Doolittle, J. Mol. Biol. 157:105-132 (1982)). A 27 
residue region of strong hydrophobicity (underlined, 

25 residues 786-812) toward the C-terminus represents the 
single putative transmembrane domain. This suggests that 
nearly all of the receptor is composed of an N-terminal 
extracellular domain that is anchored to the plasma 
membrane near its C-terminus. A relatively short 



wo 93/09228 



PCT/US92/09326 



-36- 



C-tenninal tail of 41 residues represents the putative 

cytoplasmic domain. 

The clone R3-0F was also analyzed and found to be a 
truncated version of R3-0FF, with an identical open 
reading frame but whose last encoded residue is alanine 

5 818 (Figure 1) . 

In R3-0FF there are six consensus N-linked 
glycosylation sites and 15 cysteines (indicated in Figure 
1) . There is at least one consensus glycosaminoglycan 
addition site at serine 535 (Bernf ield, M. and K. C. 

10 Hooper, tn«. w.Y. Acad. Sci. in press (1991) , and 

numerous Ser-Gly residues that are potential sites for 
GA6 conjugation. A consensus protein kinase C site is 
also present at residue 817. 

Only one other gene described to date, a 

15 glycoprotein expressed in high quantities by endothelial 
cells and termed endoglin (Gougos and Letarte, 1990), 
contains a related amino acid sequence, overall, there 
is -30% identity with the type III receptor over the 
entire 645 amino acid residue length of endoglin. The 

20 most homologous regions between the sequences of the type 
III receptor and endoglin (74% identity) falls primarily 
in the putative transmembrane and cytoplasmic domains. 
Similar to the general structure of type III receptor, 
endoglin is a glycoprotein which contains a large 

25 hydrophilic and presumably extracellular N-terminal 

domain followed by a putative transmembrane domain and a 
short cytoplasmic tail of 47 amino acid residues. The 
biological role of endoglin is unclear, though it has 
been suggested that it may involve cell-cell recognition 

30 through interactions of an "RGD" sequence on its 



wo 93/09228 



PCr/US92/09326 



-37- 

ectodomain with other adhesion molecules. Unlike the 
TGF-^ type III receptor, endoglin does not carry GAG 
groups. 

Ecniivalents 

Those skilled in the art will recognize, or be able 
to ascertain using not more than routine experimentation, 
many equivalents to the specific embodiments of the 
invention described herein. Such equivalents are 
intended to be encompassed by the following claims. 



wo 93/09228 



PCr/US92/09326 



38 



tnttmalienal AppUeitien No: PCTI 



\ MICROORGANISMS 

6 X2 I 



A. loiimncATio* OP oi^osrr • 



•ISM •! tfmM»r MMMfl « 

American Type Culture Collection 



A««i»M •! OoeHiaint po«i* « 

12301 Parklawn Drive 
Rockville, Maryland 20852 USA 



21 October 1991 



B. ADBmOHAL 



In respect of those designations in which a European Patent is sought, 
the Applicant hereby informs the European Patent Office under 
European Rule 28(4) that, until the publication of the mention of the 
grant of the European Patent or until the date on which the European 
Application has been refused or is withdrawn or is deemed to be 
withdrawn, the availability of the biological material deposited with 
the American Type Culture Collection under Accession No. 
shall be effected only by the issue of a sample to an expert nominated 
by the requester in acco rdance with Eur opean Rule 28(5) > . 



C. OiSICMATf e STATtS FOR WMICM 



INOICATIOMS AlU VAOI • W H»« iadtotow «• AM tor i 



Europe (EP) 
Australia 
Canada 
Japan 



, SirARATt nJIINlSHme of INOICATIOMS • btmnt H iiM««»llc»M» 



f AutlMiuM OAmO 



Q Th« *«t« •« f»e«iit «f»m tha BP»lic«nU ir» lnt«fiwii«ft«J 



(AiitborUM omctr) 



Ferm fCT/RO/l** Wanusnr 



PCr/US92/09326 



-39- 



CLAIMS 

Isolated DNA encoding TGF-fi receptor of vertebrate 
origin or DNA which hybridizes thereto and encodes 
TGF-^ receptor of vertebrate origin. 

Isolated DNA of Claim 1 wherein the TGF-fi receptor 
is TGF-p type III receptor or TGF-^ type II 
receptor . 

Isolated DNA of Claim 2 which is of mammalian 
origin. 

Isolated DNA of murine or human origin encoding 
TGF-^ type III receptor or DNA which hybridizes 
thereto. 

Isolated DNA of Claim 4 having the nucleotide 
sequence of Figure 1 or a portion thereof sufficient 
to encode TGF-^ type III receptor. 

Isolated DNA of murine or human origin encoding 
TGF-^ type II receptor or DNA which hybridizes 
thereto. 

Isolated DNA of Claim 6 having the nucleotide 
sequence of Figure 2 or a portion thereof sufficient 
to encode TGF-fi type II receptor. 

Isolated TGF-^ type III receptor of mammalian 
origin. 



wo 93/09228 



PCr/US92/09326 



- 40 - 



9. Isolated TGr-^ type III receptor of Claim 8 having 
the amino acid sequence of Figure 1 or a 
substantially similar amino acid sequence. 

10. Isolated TGF-^ type II receptor of mammalian origin. 

5 11. Isolated TGF-fi type II receptor of Claim 10 having 
the amino acid sequence of Figure 3 or a 
substantially similar amino acid sequence. 

12. Recombinant TGF-^ type III receptor of mammalian 
origin. 

10 13. Recombinant TGF-^ type III receptor of Claim 8 
having the amino acid sequence of Figure 1 or a 
substantially similar amino acid sequence. 

14. Recombinant TGF-^ type II receptor of mammalian 
origin. 

15 15. Recombinant TGF-^ type II receptor of Claim 10 
having the amino acid sequence of Figure 4 or a 
substantially similar amino acid sequence. 

16. Soluble TGF-^ receptor. 

17 . Soluble TGF-^ receptor of Claim 16 which is soluble 
20 TGF-p type III receptor. 

18. soluble TGF-p type III receptor of Claim 17 in which 
the amino acid sequence is amino acids 1 through 



wo 93/09228 



PCr/US92/09326 



-41- 



785, inclusive, of Figure 1 or a substantially 
similar amino acid sequence. 

19. Soluble TGF-^ receptor of Claim 16 which is soluble 
TGF-p type II receptor. 

5 20. Soluble TGF-^ receptor of Claim 19 in which the 

amino acid sequence is approximately amino acids 1 
through 166, inclusive, of Figure 3, or a 
substantially similar amino acid sequence. 

21. An antibody which specifically recognized TGF-fi type 
10 III receptor of mammalian origin. 

22. An antibody of Claim 21 which is a monoclonal 
antibody. 

23. An antibody which specifically recognizes soluble 
TGF-p type III receptor of mammalian origin. 

15 24. An antibody which specifically recognizes soluble 
TGF'fi type II receptor of mammalian origin. 

25^ A method of altering TGF-^ binding to TGF-p type II 
or type III receptor on the surface of a cell, 
~ comprising combining soluble TGF-0 type II or type 
20 III receptor with the cell, under conditions 

appropriate for binding of the soluble TGF-^ 
receptor and TGF-^. 



wo 93/09228 



PCT/US92/09326 



-42- 



26. The method of Claim 25 wherein TGF-fi binding is 
inhibited. * 

27. A method of altering TGF-/> binding to TGF-/J type III 
receptor on the surface of a cell comprising 

5 combining the cell with DNA encoding TGF-^ type III 

receptor in an appropriate expression system which 
expresses TGF-p type III receptor, under conditions 
appropriate for expression of TGT-fi type III 
receptor and binding of TGF-^ with TGF-^ type III 

10 receptor. 

28. A method of regulating the effect of TGF-^ in a 
mammal, comprising administering to the mammal a 
TGF-^ receptor selected from the group consisting 
of: TGT-fi type III receptor, TGF-^ type II 

15 receptor, soluble TGF-^ type III receptor, soluble 

TGF'P type II receptor, TGF-p bound to TGF-^ type 
III receptor or a combination thereof, in sufficient 
quantity to alter binding of TGF-^ to TGF-^ type III 
receptor, binding of TGF-^ to TGF-/I type II receptor 

20 both. 



wo 93/09228 



PCr/US92/09326 



-43- 



10 



15 



29. TGF-^ receptor according to any one of Claims 8 to 
20, for use in therapy. 

30. An antibody according to any one of Claims 21 to 24, 
for use in therapy. 

31. Use of receptor according to any one of Claims 
8 to 20, for the manufacture of a medicament for 
altering (e.g. inhibiting) TGF-fi binding to TGF^fi 
type II or type III receptor on the surface of a 
cell. 

32. Use of a TGT-fi receptor selected from the group 
consisting of: TGF-fi type III receptor, TGF-fi type 
II receptor, soluble TGF-^ type III receptor, 
soluble TGF-fi type II receptor, TGF-^ bound to TGF-p 
type III receptor or a combination thereof, for the 
manufacturing of a medicament for use in regulating 
the affect of TGF-^ in a mammal. 

33. A method of assessing the ability of a compound to 
interfere with TGF-^ binding to the TGF-^ type III 
receptor, comprising the steps of: 

a) combining: 



2) 
3) 



1) 



mammalian cells which express the TGF-^ 

type III receptor; 

labeled TGF-p; and 

a compound to. be assessed; 



wo 93/09228 



PCT/US92/09326 



-44- 



b) maintaining the product of (a) under conditions 
sufficient for TGF-/> to bind to the TGF-^ type 
III receptor; 

c) determining the extent of binding of TGF-p to 

5 TGF-^ type III receptors in the presence of the 

compound to be assessed; and 

d) comparing the determination made in (c) with 
the extent to which binding of TGF-^ to the 
TGF-A type III receptor occurs in the absence 

jQ of the compound to be assessed, 

wherein if TGF-^ binding to the TGF-^ type III 
receptor occurs to a lesser extent in the presence 
of the compound to be assessed than in the absence 
of the compound to be assessed, the compound to be 

15 assessed interferes with TGF-^ binding to TGF-/J type 

III receptors. 

34. A method of Claim 33 wherein the cells which express 
the TGF-p type III receptor are a cell line. 

35. A method of Claim 34 wherein the cells which express 
20 the TGF-^ type III receptor are cells modified to 

express the TGF-^ type III receptor. 

36. A method of Claim 35 wherein the cells modified to 
express the TGF-^ type III receptor are cells which 
have incorporated into them TGF-/J receptor cDNA in 

25 an appropriate vector or microinjected TGF-p 

receptor RNA. 



wo 93/09228 



PCT/US92/09326 



-45- 



37. A method of assessing the ability of a coinpound to 
interfere with TGF-^ binding to the TGF-^ type II 
receptor comprising the steps of: 

a) combining: 

5 1) mammalian cells which express the TGF-^ 

type II receptor; 

2) labeled TGF-^; and 

3) a compound to be assessed; 

b) maintaining the product of (a) under conditions 
10 sufficient for TGF-^ to bind to the TGF-^ type 

II receptor; 

c) determining the extent of binding of TGF-^ to 
TGF-^ type II receptors in the presence of the 
compound to be assessed; and 

15 d) comparing the determination made in (c) with 

the extent to which binding of TGF-^ to the 
TGF-i8 type II receptor occurs in the absence of 
the compound to be assessed, 
wherein if TGF-^ binding to the TGF-^ type II 
20 receptor occurs to a lesser extent in the presence 

of the compound to be assessed than in the absence 
of the compound to be assessed, the compound to be 
assessed has interfered with TGF-^ binding to TGF-^ 
type II receptor. 

25 38. A method of Claim 37 wherein the cells which express 
the TGF'fi type II receptor are a cell line. 



39. 



A method of Claim 38 wherein the cells which express 
the TGF-^ type II receptor are cells modified to 
express the TGF-p type II receptor. 



wo 93/09228 



PCr/US92/09326 



10 



15 



20 



-46- 



40. A method of Claim 39 wherein the cells modified to 
express the TGF-^ type II receptor are cells which 
have incorporated into them TGF-^ receptor cDNA in 
an appropriate vector or microinjected TGF-p 
receptor SNA. 

41. A method of detecting abnormal binding of TGF-^ 
TGF-^ type III receptors of TGF-/J type II receptors 
at a cell surface, comprising: 

a) determining the extent of binding of TGF-/> to 
TGF-^ type III receptors or TGF-/> type II 
receptors by cells in a sample obtained from an 
individual in whom binding is to be assessed 
thereby producing a test binding value; and 

b) comparing the results of (a) with the extent to 
which binding occurs at the cell surface in 
control cells which are cells known to have 
abnormal binding of TGF-p to TGF-p type III 
receptors or TGF-^ type II receptors resulting 
in a control binding value, 

wherein a test binding value similar to the control 
binding value is indicative of abnormal binding of 
TGF-p to TGF-fi type III receptor or TGF-^ type II 
receptor. 



wo 93/09228 



PCr/US92/09326 



1/7 

FIGURE lA 

"240 

CAGGAGGTGA AAGTCCCOGG CGGTCCGGAT GGCGCAGTTG CACTGOGCTG CTGAGCTCGC -180 

GGCC6CCTGC GCACACTGGG GGGACTCGCT TCGGCTAGTA ACTCCtOCAC CTCGCGGCGG -120 

ACGACCGGTC CTGGACACGC TGCCTGCGAG GCAAGTTGAA CAGTGCAGAG AAGGATCTTA - 60 

AAGCTACACC CGACTTGCCA CGATTGCCTT CAATCTGAAG AACCAAAGGC TGTTGGAGAG - 1 

ATG GCA GTG ACA TCC CAC CAC ATG ATC CCG GTG ATG GTT GTC CTG ATG 48 

Met Ala Val Thr Ser Hie His Met lie Pro Val Met Val Val Leu Met 16 

AGC GCC TGC CTG GCC ACC GCC GGT CCA GAG CCC AGC ACC OGG TGT GAA 96 

Ser Ala Cye Leu Ala Thr Ala Gly Pro Glu Pro Ser Thr Arg Cye Glu 32 

T 6 

CTG TCA CCA ATC AAC GCC TCT CAC CCA GTC CAG GCC TTG ATG GAG AGC 144 

Leu Ser Pro lie Asn Ala Ser Hie Pro Val Gin Ala Leu Met Glu Ser 48 
# 

TTC ACC GTT CTG TCT GGC TGT GCC AGC AGA GGC ACC ACC GGG CTG CCA 192 

Phe Thr Val Leu Ser Gly Cys Ala Ser Arg Gly Thr Thr Gly Leu Pro 64 
+ & 

AGG GAG GTC CAT GTC CTA AAC CTC CGA AGT ACA GAT CAG GGA CCA GGC 240 

Arg Glu Val His Val Leu Asn Leu Arg Ser Thr Asp Gin Gly Pro Gly 80 

CAG CGG CAG AGA GAG GTT ACC CTC CAC CTG AAC CCC ATT GCC TCG GTG 288 

Gin Arg Gin Arg Glu Val Thr Leu Hie Leu Asn Pro He Ala Ser Val 96 

CAC ACT CAC CAC AAA CCT ATC GTG TTC CTG CTC AAC TCC CCC CAG CCC 336 

His Thr His His Lys Pro He Val Phe Leu Leu Asn Ser Pro Gin Pro 112 

CTG GTG TGG CAT CTG AA6 ACG GAG AGA CTG GCC OCT GGT GTC CCC AGA 384 

Leu Val Trp His Leu Lys Thr Glu Arg Leu Ala Ala Gly Val Pro Arg 128 

CTC TTC CTG GTT TCG GAG GGT TCT GTG GTC CAG TTT CCA TCA GGA AAC 432 

Leu Phe Leu Val Ser Glu Gly Ser Val Val Gin Phe Pro Ser Gly Asn 144 

+ # 

TTC TCC TTG ACA GCA GAA ACA GAG GAA AGG AAT TTC CCT CAA GAA AAT 480 

Phe Ser Leu Thr Ala Glu Thr Glu Glu Arg Asn Phe Pro Gin Glu Asn 160 

GAA CAT CTC GTG CGC TGG GCC CAA AAG GAA TAT GGA GCA GTG ACT TCG 528 

Glu His Leu Val Arg Trp Ala Gin Lys Glu Tyr Gly Ala Val Thr Ser 176 

TTC ACT GAA CTC AAG ATA GCA AGA AAC ATC TAT ATT AAA GTG GGA GAA 576 

Phe Thr Glu Leu Lys He Ala Arg Asn He Tyr He Lys Val Gly Glu 192 



SUBSTITUTE SHEST 



wo 93/09228 



PCT/US92/09326 



in 

FIGURE IB 



GAT 
Asp 


CAA 
Gin 


GTG TTT CCT CCT ACG TGT AAC ATA GGG AAG 
Val Phe Pro Pro Thr Cys Asn He Gly Lys 

fi 


AAT TTC CTC 
Asn Phe Leu 


TCA 
Ser 


624 
208 


CTC 
Leu 


AAT 
Asn 


TAC CTT GCC GAG TAC CTT 
Tyr Leu Ala Glu Tyr I*eu 


CAA 
Gin 


CCC AAA 
Pro Lys 


GCC 
Ala 


GCC GAA GGT 
Ala Glu Gly 


TGT 
Cys 


672 
224 


GTC 

Val 


CTG 

Leu 


CCC AGT CAG CCC CAT GAA 
Pro Ser Gin Pro His Glu 


AAG 
Lys 


GAA 
Glu 


GTA 
Val 


CAC 
His 


ATC ATC GAG 
He He Glu 


TTA 
Leu 


720 
240 


ATT 
He 


ACC 
Thr 


CCC A6C TCG AAC CCT TAC 
Pro Ser Ser Asn Pro Tyr 


A6C 
Ser 


GCT 
Ala 


TTC 
Phe 


CAG 
Gin 


GTG GAT ATA 
Val Asp He 


ATA 
He 


768 
256 


6TT 
Val 


GAC 
Asp 


ATA CGA CCT GCT CAA CAG GAT CCC 
He Arg Pro Ala Gin Glu Asp Pro 


GAG 
Glu 


GTG 
Val 


GTC AAA AAC 
Val Lys Asn 


CTT 
Leu 


816 
272 


GTC 
val 


CTG 
Leu 


ATC TTG AAG TGC AAA AAG TCT 
He Leu Lys Cys Lys Lys Ser 
fi 


GTC 
Val 


AAC TGG 
Asn Trp 


GTG ATC AAG 
Val He Lys 


TCT 
Ser 


864 
288 


TTT. 
Phe 


GAC 
Asp 


GTC AAG GGA AAC TTG AAA GTC 
Val Lys Gly Asn Leu Lys Val 


ATT 
He 


GCT 
Ala 


CCC 
Pro 


AAC AGT ATC 
Asn Ser He 


GGC 
Gly 


912 
304 


TTT 
Phe 


GGA 
Gly 


AAA GAG AGT GAA CGA TCC 
Lys Glu Ser Glu Arg Ser 


ATG 
Met 


ACA 
Thr 


ATG 
Met 


ACC 
Thr 


AAA TTG GTA 
Lys Leu Val 


AGA 
Arg 


960 
320 


GAT 
Asp 


GAC 
Asp 


ATC CCT TCC ACC CAA GAG AAT CTG ATG AAG 
He Pro Ser Thr Gin Glu Asn Leu Met Lys 


TGG GCA CTG 
Trp Ala Leu 


GAC 
Asp 


1008 
336 


AAT 
Asn 


GGC 
Gly 


TAC AGG CCA GTG ACG TCA TAC ACA ATG 
Tyr Arg Pro Val Thr Ser Tyr Thr Met 


GCT 
Ala 


CCC GTG GCT 
Pro Val Ala 


AAT 
Asn 


1056 
352 


A6A 
Arg 


TTT 
Phe 


CAT CTT CGG CTT GAG AAC 
His Leu Arg Leu Glu Asn 


AAC 
Asn 


GAG 
Glu 


GAG 
Glu 


ATG 
Met 


AGA GAT GAG 
Arg Asp Glu 


GAA 
Glu 


1104 
368 


GTC 
Val 


CAC 
His 


ACC ATT CCT CCT GAG CTT CGT ATC 
Thv- Tie Pro Pro Glu Leu Ara He 


CTG 
Leu 


CTG 
Leu 


GAC CCT GAC 
ASP Pro ASP 


CAC 
His 


1152 
384 








peptide 1 






CCG 
Pro 


CCC 
Pro 


GCC CTG GAC AAC CCA CTC 
Ala Leu Asp Asn Pro Leu 


TTC 
Phe 


CCA 
Pro 


GGA GAG GGA AGC CCA 
Gly Glu Gly Ser Pro 


AAT 
Asn 


1200 
400 



GGT GGT CTC CCC TTT CCA TTC CCG GAT ATC CCC AGG AGA GGC TGG AAG 1248 

Gly Gly Leu Pro Phe Pro Phe Pro Asp He Pro Arg Arg Gly Trp Lys 416 

GAG GGC GAA GAT AGG ATC CCC CGG CCA AAG CAG CCC ATC GTT CCC AGT 1296 

Glu Gly Glu Asp Arg He Pro Arg Pro Lys Gin Pro H e Val Pro Ser 432 

peptide 2 



SWeSTITUTE SHEET 



wo 93/09228 



PCT/US92/09326 



3/7 
FIGURE IC 

GTT CAA CTG CTT CCT GAC CAC CGA GTA CCA GAA GAA GTG CAA GGG GGC 1344 
Val Gin Leu Leu Pro Asp His Arg Glu Pro Glu Glu Val Gin Gly Gly 448 

GTG GAC ATC CCC CTG TCA GTC AAA TGT GAC CAT GAA AAG ATG GTC GTG 1392 

Val Asp He Ala Leu Ser Val Lys Cys Asp His Glu Lys Met Val Val 464 

& 

GCT GTA GAC AAA GAC TCT TTC CAG ACC AAT GGC TAC TCA GGG ATG GAG 1440 

Ala Val Asp Lys Asp Ser Phe Gin Thr Asn Gly Tyr Ser Gly Met Glu 480 

+ 

CTC ACC CTG TTG GAT CCT TCG TGT AAA GCC AAA ATG AAT GGT ACT CAC 1488 

Leu Thr Leu Leu Asp Pro Ser Cys Lys Ala Lys Met Asn Gly Thr His 496 

& # 



TTT GTT CTC GAG TCT CCC CTG AAT GGC 
Phe Val Leu Glu Ser Pro Leu Asn Gly 



TCG ACC CCG GAT GGT GTG GTT TAC TAT 
Ser Thr Pro Asp Gly Val Val Tyr Tyr 

CCG TCC CCT GGG GAT AGC AGT GGC TGG 
Pro Ser Pro Gly Asp Ser Ser Gly Trp 

+++ 

GAG TCA GGC GAT AAT GGA TTT CCT GGA 
Glu Ser Gly Asp Asn Gly Phe Pro Gly 

GCC CCC CTG AGC CGA GCT GGA GTG GTG 
Ala Pro Leu Ser Arg Ala Gly Val Val 



CAG CTG AGG AAT CCC AGT GGC TTC CAG 
Gin Leu Arg Asn Pro Ser Gly Phe Gin 

ACC TTC AAC ATG GAG CTG TAT AAC ACA 
Thr Phe, Asn Met Glu Leu Tyr Asn Thr 

CCA GGG GTC TTC TCT GTG GCA GAG AAC 
Pro Gly Val Phe Ser Val Ala Glu Asn 

TCT GTC ACC AAG GCT GAC CAA GAT CTG 
Ser Val Thr Lys Ala Asp Gin Asp Leu 



TGT GGT ACT CCA CAT CGG AGG 1536 

Cys Gly Thr Arg His Arg Arg 512 
6 

AAC TCT ATT GTG GTG CAG GCT 1584 

Asn Ser He Val Val Gin Ala 528 

CCT GAT GGC TAT GAA GAC TTG 1632 

Pro Asp Gly Tyr Glu Asp Leu 544 

GAC GGG GAT CAA GGA GAA ACT 1680 

Asp Gly Asp Glu Gly Glu Thr 560 



GTG TTT AAC TGC AGC TTC CGG 1728 

Val Phe Asn Cys Ser Leu Arg 576 
# & 

GGC CAG CTC GAT GGA AAT GCT 1776 

Gly Gin Leu Asp Gly Asn Ala 592 

# 

GAC CTC TTT CTG GTG CCC TCC 1824 

Asp Leu Phe Leu Val Pro Ser 608 

GAG CAT GTT TAT GTT GAG GTG 1872 

Glu His Val Tyr Val Glu Val 624 

GGA TTC GCC ATC CAA ACC TGC 1920 

Gly Phe Ala He Gin Thr Cys 640 



wo 93/09228 



PCr/US92/09326 



4/7 
FIGURE ID 

TTT CTC TCT CCA TAC TCC AAC CCA GAC AGA ATG TCT GAT TAC ACC ATC 1968 

Phe Leu Ser Pro Tyr Ser Asn Pro Asp Arg Met Ser Asp Tyr Thr lie 6&6 

ATC GAG AAC ATC TGT CC6 AAA GAC GAC TCT GTG AAG TTC TAC AGC TCC 2016 

lie Glu Asn He Cye Pro Lys Asp Asp Ser Val Lys Phe Tyr Ser Ser 672 

AAG AGA GTG CAC TTT CCC ATC CCG CAT GCT GAG GTG GAC AAG AAG CGC 2064 

Lys Arg Val His Phe Pro He Pro His Ala Glu Val Asp Lye Lys Arg 688 

TTC AGC TTC CTG TTC AAG TCT GTG TTC AAC ACC TCC CTG CTC TTC CTG 2112 

Phe Ser Phe Leu Phe Lys Ser Val Phe Asn Thr Ser Leu Leu Phe Leu 704 

CAC TCC GAG TTC ACT CTG TCC TCC AGG AAG AAG GGC TCC CTC AAG CTC 2160 

His Cye Glu Leu Thr Leu Cys Ser Arg Lys Lys Gly Ser Leu Lys Leu 720 
& & 

CCG AGG TCT 6TC ACT CCT GAC GAC GCC TGC ACT TCT CTC GAT CCC ACC 2208 

Pro Arg Cys Val Thr Pro Asp Asp Ala Cys Thr Ser Leu Asp Ala Thr 736 

ATG ATC TGG ACC ATC ATC CAG AAT AAG AAG ACA TTC ACC AAC CCC CTG 2256 

Met He Trp Thr Met Met Gin Asn Lys Lys Thr Phe Thr Lys Pro Leu 752 

GCT GTC CTC CTC CAG GTA GAC TAT AAA GAA AAT GTT CCC AGC ACT AAC 2304 

Ala Val Val Leu Gin Val Asp Tyr Lys Glu Asn Val Pro Ser Thr Lys 768 

GAT TCC ACT CCA ATT CCT CCT CCT CCT CCA CAG ATT TTC CAT GGC CTC 2352 

Asp Ser Ser Pro He Pro Pro Pro Pro Pro Gin He Phe His Gly Leu 784 

GAC ACG CTC ACC GTC ATC GGC ATT CCA TTT GCA GCA TTT GTC ATC G6A 2400 
Asp Thr Leu Thr Val Met Glv He Ala Phe Ala Ala Phe Val He Civ 800 

GCC CTC CTC ACG GGC GCC TTG TGG TAC ATC TAC TCC CAC ACA CCG GAG 2448 
Ala Leu Leu Thr Glv Ala Leu Tro Tvr He Tvr Ser His Thr Gly Glu 816 

ACA GCA CCA AGG CAG CAA GTC CCT ACC TCC CCG CCA GCC TCC GAG AAC 2496 
Thr Ala Arg Arg Gin Gin Val Pro Thr Ser Pro Pro Ala Ser Glu Asn 832 
$ € 

AGC AGC GCG GCC CAC AGC ATC GGC AGC ACT CAG AGT ACC CCC TCC TCT 2544 
Ser Ser Ala Ala His Ser He Gly Ser Thr Gin Ser Thr Pro Cys Ser 848 

AGC AGC AGC ACA GCC TAGGTGGACA GACAGACGCC CGCCCACCGC AGCCAGGGCA 2599 
Ser Ser Ser Thr Ala*** 653 



dUSSTITUTE SHEST 



wo 93/09228 



PCT/US92/09326 



FIGURE IE 



GGGCCCGATG CCAGTGCTGC GTGTCCACAG TCAGAAGTCT TGATCTGGGC TCCCTGTAAA 2659 
GAAAGAGTGA ATTTCAGTAT ACAGACAGCC AGTTCTACCC ACCCCTTACC ACGGCCCACA 2719 
TAAATGTGAC CCTGGGCATC TGTCACACGA AAGCTAAGCT GGTGGCCTTC CCCACCAGCC 2779 
CCTCGCAGGA TGGGGGTTTC AATGTGAAAC ATCTGCCAGT TTTGTTTTGT TTTTTTAATG 2839 
CTGCTTTGTC CAGGTGTCCA AACATCCATC ATTTGGGGTG GTCTGTTTTA CA6AGTAAAG 2899 
GAGGCGGT6A AGGGACGTCA GCTAGTGTGT AGAGCCAAGG GGAGACAGCT AGGATTCTCG 2959 
CCTAGCTGAA CCAAGGTGTA AAATAGAAGA CACGCTCC 2997 



8y@®TITUTE 8HEgT 



wo 93/09228 



PCr/US92/09326 



6/7 

' FIGURE 2 



AGTTTCCTGT 
CGCGCAC6GA 
CCX^GACTCCT 
GCCGGCCTCC 
GCACATCTGC 
GAGCGCAGCC 
TCTATGAOGA 
TGTGGCCGCT 
CCGCACX5TTC 
CGGTGCAGTC 
CCACCTGTGA 
ATCTGTGA6A 
CGAG AACAT A 
ATGACTTXAT 
AAAAAAAAGC 
GTGCAATGAC 
ACTTGTTGCT 
CTGGGAGTT6 
CCGGCAGGAG 
TCATGGAGTT 
GACATCAGCT 
GCCCATTGAG 
ATAAGGCCAA 
GTCAAGATCT 
CATCTTCTCA 
CGGCTGAGGA 
GCCTTCCAOG 
CAGCTGGGAG 
CTCACCTCCA 
GTGCACAGGG 
CTGCTGCCT6 
CTGTGGATGA 
GCTCCAGAAG 
CAAGCAGACC 
CTCGCTGTAA 
TCCAAGGTGC 
GAGAGATCGA 
GCAXCCAGAT 
GAGGCCCGTC 
GCATCTGGAC 
AAGACGGCTC 
ATGTCCAAAG 



TTCCCCCGCA 
GCGACGACAC 
GTGCAGCTTC 
AGGCCCCTCC 
GCTGCCGGCC 
AGGGGTCCGG 
GCAGCGGGGT 
GCACATCGTC 
AGAAGTCGGT 
AAGTTTCCAC 
CAACCAGAAA 
AGCCACAGGA 
ACACTAGAGA 
TCTGGAAGAT 
CTGGTGAGAC 
AACATCATCT 
AGTCATATTT 
CCATATCTGT 
AAGCTGAGTT 
CAGCGAGCAC 
CCAOGTGTGC 
CTGGACACCC 
GCTGAAGCAG 
TTCOrTATGA 
GACATCAATC 
GCGGAAGACG 
CCAAGGGCAA 
GACCTGCGCA 
CAGTGATCAC 
ACCTCAAGAG 
TGTGACTTTG 
CCTGGCTAAC 
TCCTAGAATC 
GATGTCTACT 
TGCAGTGGGA 
GGGAGCACCC 
GGGCGACCAG 
GGTGTGTGAG 
TCACA6CCCA 
AGGCTCTCGG 
CCTAAACACT 
AGGCTGCCCC 



GCGCTGAGTT 

CCCCGCGOGT 

CCTCGGCCX;C 

TGGCTGGCGA 

CGGCGOGGGG 

GAAGGCGCOG 

CTGCCATGGG 

CTGTGGAOGC 

TAATAAOGAC 

AACTGTGTAA 

TCCTGCATGA 

AGTCTGTGTG 

CAGTTTGCCA 

GCTGCTTCTC 

TTTCTTCATG 

TCTCAGAAGA 

CAAGTGACAG 

CATCATCATC 

CAACCTGGGA 

TGTGCCATCA 

CAACAACATC 

TGGTGGGGAA 

AACACTTCAG 

GGAGTATGCC 

TGAAGCATGA 

GAGTTGGGGA 

CCTACAGGAG 

AGCTGGGCAG 

ACTCCATGTG 

CTCCAATATC 

GGCTTTCCCT 

AGTGGGCAGG 

CAGGATGAAT 

CCATGGCTCT 

GAAGTAAAAG 

CTGTGTCGAA 

AAATTCCCAG 

ACGTTGACTG 

GTGTGTGGCA 

GGAGGAGCTG 

ACCAAATAGC 

TCTCACCAAA 



GAAGTTGAGT 
GCACCCGCTC 
CGGGGGCCTC 
GCGGGCGCCA 
TCOGGAGAGG 
TCCGTGCGCT 
TOGGGGGCTG 
GTATCGCCAG 
A TGATA CTCA 
ATTTTGTGAT 
GGAACTGCAG 
GCTGTATGGA 
TGACCCCAAG 
CAAAGTGCAT 
TGTTCCXGTA 
ATATAACACC 
GCATCAGCCT 
TTCTACTGCT 
AACOGGCAAG 
TOCTGGAAGA 
AAOCAGAACA 
AG6TOGCTTT 
AGCAGTTTGA 
TCTTGGAAGA 
GAACATACTC 
AACAATACT6 
TACCTGAOGC 
CTCCCTOGCC 
GGAGGCCCAA 
CTOGTGAAGA 
GCGTCTGGAC 
TGGGAACTGC 
TTGGA6AATG 
GGTGCTCTGG 
ATTATGAGCC 
AGCATGAAGG 
CTTCTGGCTC 
AG7GCTGGGA 
GAACGCTTCA 
CTCGGAGGAG 
TCTTATGGGG 



GTTGGCGAGG 
GAGTCACTCG 
GGGACAGGAG 
CCCGOGCCTC 
CATCTGGCCC 
GCGCGGCGCG 
GGGGGCTCGG 
CTCAGGGGCC 
CAOGATCCCA 
CTGAC AACAA 
GTGAGATTTT 
CATCACCTCC 
GAAAGAATGA 
CTCCCCTACC 
TATGAAGGAA 
GCTCTGATGA 
AGGAATCCTG 
OCTGCCACCA 
ACCGCGrrAA 
ACX;CGGAAGC 
TGACOGCTCT 
CAGAGCTGCT 
GCTGAGGTCT 
GACAGTGGCA 
CAGAGAAGGA 
CAGTTCCTGA 
GCTGATCACC 
GGCATGTCAT 
OGGGGGATTG 
GATGCCCATC 
ACGACCTAAC 
CCTACTCTGT 
AAGATACATG 
CTGAGTCCTT 
GAAATGACAT 
TCCATTTGGT 
ACAACGTGTT 
AACCACCAGG 
CCACGACCCA 
GTGAGCTGGA 
AAGATTCCTG 
CAGGCTGGGC 



SUBSTITUTE SHEET 



wo 93/09228 



PCT/US92/09326 



7/7 



FIGURE 3 



MGRGLLRGLW 
CKFCDVRFST 
CHDPKLPYHD 
EEYNTSNPDL 
WETGKTRKLM 
GKGRFAEVnC 
HENILQFLTA 
GSSLARGIAH 
SLRLDPTLSV 
ALVLWEMTSR 
PSFWLNHQGI 
SCSEERIPEO 



PLHIVLWTRI 
CONQKSCMSN 
FILEDAASPK 
LLVIPQVT6I 
EFSEHCAIIL 
AKLKQNTSEQ 
EERKTELGKQ 
LHSDHTPCGR 
DDLANSGQVG 
CNAVGEVKOY 
QMVCETLTEC 
GSLNTTK 



ASTIPPHVQK 
CSITSICEKP 
CIKKEKKXPG 
SLLPPLGVAI 
EDDRSDISST 
FETVAVKIFP 
yWLITAFHAK 
PKMPIVHRDL 
TARYMAPEVL 
EPPFGSKVRE 
WDHDPEARLT 



SVNNDMIVTD 
QEVCVAVWRK 
ETFFMCSCSS 
SVIIIFYCYR 
CANNINHNTE 
YEEYASWKTE 
GNLQEYLTRH 
KSSNILVKND 
ESRMNLENAE 
HPCVESMKDN 
AQCVAERFSB 



NNGAVKFPQL 
NDENITLETV 
DECNDNIIFS 
VNHQQKLSST 
LLPIELDTLV 
KOIFSDINLK 
VISWEDLRKL 
LTCCLCDFGL 
SFKQTDVYSM 
VLRDRGRPEI 
LEHLDRLSGR 



SUBSTITUTE SHEET 



INTERNATIONAL SEARCH REPORT 



PCT/US 92/09326 



tOASSIiraCATlIOWOFSmiECrftlATTiSa O? scvctoI dnslfleaflca syabob oppiy, tKltetto aU)Q 



Aca>rd6i8 to QntcnatlooQl Pascal OassfficotlOD (SIPQ or to toib Notoal QossfflcaiiBa and IDPC 

Int.Cl. 5 C12N15/12; C12N5/10; C12P21/08; 

A61K37/02 



C07t(13/00 



0. IFIDSUD5 SEAaCKQSO} 



ClasziflcoSioD System 



Int.Cl. 5 



C12N 



C07K 



A61K 



DccniDQntatiim Seonttsd oaba thoo MbiEQBD IDocsmaitasioa 
to tfao ISjctoDt that sncb !I>oa»cc3tS5 ora QadicSod la tbo IFIolds Soorcbod" 



M. POOMEWgS CON SlBDlSaiSa) TO BE imSVAWT^ 

Chatloa 0? IDoaQacat, U ^tb Mlcotloa, ^C7o c 



, ftf tbo rciGvoiit posso^ ^ 



EP,A,0 369 861 (ONCOGEN LTD PARTNERSHIP, 
US) 

23 May 1990 
see the t/hole document 

BIOCHEMICAL AMD BIOPHYSICAL RESEARCH 
COMMUMICATIOMS. 

vol. 179, no. 1, 30 August 1991, DULUTH, 
MINNESOTA US 
pages 373 - 385 

O'GRADY P;HUANG SS;HUAWG JS; 'Expression 
of a new type high raolecular weight 
receptor (typ® V receptor) of transforming 
growth factor beta in normal and 
transformed cells.' 
see the whole document 




*> Spcdol catqgoriss cited docamots : 

"A" doonaeot dcflnlOQ tb« gakoiQl stnto of tbe art ^iTblcb Is nox 

conddood to bo d pca^calap roEsvoEEeo 

earliGT doanpcat bat ptAUsbed 00 07 oftor tbo intoiiatioQal 



*V docoiDGnt iDay tbro? doobb oa sriortty doiin(s) o? 
vThlcb is dtod to ostnUisb tbo pubUcadon dato of aootb:? 
dtaSioQ or otbc? s^cdol rosoa (as spcdflcd) 
. ''Cr docuEsstt fc^GRiss to ^ oml disdoraro, nso, onhibitloo c? 

0tbG7&lQQBS 

'HP' doauQcnt pnblisfacd prfor to tbQ IntcT&atlDiial flUng doto bm 
late? tboD tbo priori^ data doieicd 



li^ docomQit pobilsbtd oftt? tbo IntCTOtffisial flUsQ data 
or Briorlty dmo ood nm la conflict critb tbo oanHcatloa bet 
dt3 to midcntoad tbo prladplo ox tbcosy nndoTlyisio tbo 

^ docfflocat porticalar rdcvaacG; tbo dolmcd iovcatisa 
comet bo cnstdcrcd oovol or constot bo coosidoc:! to 
iBvolvo aa laKsstlvo stop 

dccamcal o? portlcDlar rolovoDeo; tbo d nlBT c rl tavcatlfea 
CQUko) b3 ooasldacd to involvQ aa tavostlvo stop tJbcat tt&o 
docasBcat Is eomMnod witb ooo or coro othc? socb dwss- 
cxats, sucb ooDbtaatioa bdag obvioos to a psTSOQ sUUsd 
tatboon. 

"^f' deeoccat G3C3DbQi' itf tbo sacQ patoat foioily 



BV. OEainiFIICATEON 



Doto tbo Actual ComplciioB cf tfao Satematlo&al Scardi 

20 JANUARY 1993 



SatSBotlooa] Seorcbtag Aatbort^ 

EmOPEAN PATENT OPPICE 



Dato of MaUtag vS tbls SattamatloBal Soareb Qispofft 

1 2. 02, 93 



SlQoataro Aiitborizcd Offlcc? 

S.A. NAUCHE 



Iat«utlfiB>] A^fUculoD No 



PCT/US 92/09326 



UL DOCUMENTS COfWOTJOSD TO BE SHEVANT (CONTINUED FROM THE SECOND SHEET) 



Otttte or Docomm; with ioilotiflQ, wbm wrapriit^ 



Uraat to QaiM Na 



P.X 



P.X 



P.X 



CELL. 

vol. 68, no. 4, 21 February 1992, 
CAMBRIDGE, NA US 
pages 775 - 785 

LIN HY;WANG XF;NG-EATON E; WEINBERG 
RA;LODISH HF; 'Expression cloning of the 
TGF-beta type II receptor, a functional 
transnenbrane serine/threonine kinase.' 
see the whole document 

CELL. 

vol. 67, no. 4, 15 November 1991, 
CAMBRIDGE, NA US 
pages 785 - 795 

LOPEZ-CASILLAS F;CHEIFETZ S;DOODY J;ANDRES 
JL;LANE WS;NASSAGUE J; 'Structure and 
expression of the menbrane proteoglycan 
betaglycan, a component of the TGF-beta 
receptor system.' 
see the whole document 

CELL. 

vol. 67, no. 4, IS November 1991, 
CAMBRIDGE, NA US 
pages 797 - SOS 

WANG XF:LIN HY;NG-EATON E; DOWNWARD 
J;LODISH HF; WEINBERG RA; 'Expression 
cloning and characterization of the 
TGF-beta type III receptor.' 
see the whole document 

JOURNAL OF BIOLOGICAL CHEMISTRY. 
(MICROFILMS) 

vol. 265, no. 33, 25 November 1990, 
BALTIMORE, MD US 
pages 20533 - 20538 

CHEIFETZ S;HERNANDEZ H;LAIHO M:TEN DUKE 
P;IWATA KK;MASSAGUE J; 'Distinct 
transforming growth factor-beta (TGF-beta) 
receptor subsets as determinants of 
cellular responsiveness to three TGF-beta 
isoforms.' 

JOURNAL OF BIOLOGICAL CHEMISTRY. 
(MICROFILMS) 

vol. 263, no. 32, 15 November 1988, 
BALTIMORE, MD US 
pages 16984 - 16991 

CHEIFETZ S.; ANDRES J.L.; MASSA6UE J; 'The 
transforming growth factor-beta receptor 
type III is a membrane proteoglycan. 
Domain structure of the receptor. ' 



-/- 



1-3,6,7, 

10,11, 

14-16, 

19.20, 

24-26, 

28-31 



37-41 • 

1-5.8,9, 

12,13, 

16-18, 

21-23, 

25-36,41 



1-5,8,9, 

12.13, 

16-18, 

21-23, 

25-36,41 



iDtradoul Appllrirttm No 



PCT/US 92/09326 



m. DOCUMENTS CONSIDEXED TO BE SELCVANT 



(CONnNUED FROM THE SECOND SHEET) 



Calegoiy* 



OtitioD of DoooiMnt, with IndiotioD, wbcrc ippnpiiate, of the rdevut pasagcs 



nrimnttoCUiBNoL 



ANNALS OF THE NEW YORK ACADEMY OF SCIENCES 
vol. 593. 1990, NEW YORK, US; 
pages 59 - 72 

MASSAGUe, J. ET AL; 'TGF-Beta recepotrs 
and TGF-beta binding proteoglycans : 
recent progress in identifying their 
functional properties.' 



ram PCT/lSAAao imtam iM) iJtmmy IMS) 



ANNEX TO THE INTERNATIONAL SEARCH REPORT 
ON INTERNATIONAL PATENT APPLICATION NO. ^5 



9209326 
66669 



Thtv 



^tbcpaCEBtftLiidtriDcmbevsRiatiiic to the patent docu^^ tbe abovr-maADed utoutiooal MRt mmt. 

Tbe B^p«ui Pitent Officr b m no way &ible for tb« partkulan ivfaidi are merdr fJva for tbe piirpo« of EafonnatieB. 2 0/0 1/93 



Pateat 
cHcd ID 



report 



PublicadoD 
date 



neoriMr^a} 



PohficidM 



EP-A-0369861 



23-05-90 



CA-A- 2002011 



14-05-90- 



Si Far MR ditiBt abMt ikB ■ 



: ate OfficalJovMl af Ite Enrercu Fatat OSbb. Na. l«/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 appHcant. 

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

□ BLACK BORDERS 

□ IMAGE CUT OFF AT TOP, BOTTOM OR SIDES 
□TfADED text or DRAWING 

1^ 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.