Mitogenic action of osteogenic growth peptide (OGP) Role of amino and carboxy-terminal regions and charge

Biochimica et Biophysica Acta, 1178 (1993) 273-280 © 1993 Elsevier Science Publishers B.V. All rights reserved 0167-4889/93/$06.00

273

BBAMCR 13440

Mitogenic action of osteogenic growth peptide (OGP): role of amino and carboxy-terminal regions and charge Zvi Greenberg a, Michael Chorev a Andras Muhlrad b Arye Shteyer c M a l k a N a m d a r a, N o r a M a n s u r a and Itai Bab a a Bone Laboratory, Faculty of Dental Medicine, The Hebrew University of Jerusalem, Jerusalem (Israel), h Department of Oral Biology, Faculty of Dental Medicine, The Hebrew Unicersity of Jerusalem, Jerusalem (Israel), c Department of Oral and Maxillofacial Surgery, Faculty of Dental Medicine, The Hebrew University of Jerusalem, Jerusalem (Israel) and a Department of Pharmaceutical Chemistry, Faculty of Medicine, The Hebrew Unic'ersity of Jerusalem, Jerusalem (Israel) (Received 20 April 1993)

Key words: Osteogenic growth peptide; Growth factor; Histone H4; Osteoblast; Fibroblast; Binding protein

We have recently reported the discovery of a 14-amino-acid osteogenic growth peptide (OGP). In vivo OGP increases bone formation and trabecular bone density. Physiologically it is found in serum complexed to an OGP binding protein (OGPBP). In vitro OGP has a biphasic effect on osteoblastic MC 3T3 E1 and fibroblastic NIH 3T3 cell proliferation; at low concentrations (0.01-1.0 and 1.0-100.0 pM, respectively) it is highly stimulatory with an inhibition at higher doses. To assess possibilities of labeling synthetic OGP to obtain radio- or fluorescent ligands, OGP analogues were extended at the N- or C-termini with Cys or Cys(S-NEtSucc) or the OGP Tyr-10 replaced by 3-I(Tyr). All analogues with N-terminal modifications, as well as the [CyslS]OGP-NH2 retained the OGP-like dose-dependent effect on proliferation of the MC 3T3 E1 and NIH 3T3 cells, although the magnitude of stimulation was lower, approx. 2/3 that of the native-like synthetic OGP. The [CysmS(S-NEtSucc)]OGP-NH2 and [3-I(TyrI°)]OGP shared only the inhibitory activity of OGP. This suppression is further shared by a number of other positively and negatively net charged, but not net neutral, peptides. Both N-terminal-modified analogues displayed a decreased binding activity to the OGPBP. All analogues except reverse OGP, [3-I(Tyrl°)]OGP and [CyslS(S-NEtSucc)]OGP-NH2 reacted with anti-OGP antibodies. These data are not only important for labeling purposes but suggest a respective role for the OGP Nand C-terminal regions in binding to the OGPBP and putative OGP receptor. It appears that the OGP proliferative activity represents the net effect of stimulation specific to the OGP structure and nonspecific inhibition associated with the peptide's high positive net charge.

Introduction It has been established that regenerating bone marrow induces an osteogenic response in distance skeletal sites [1-5]. This activity is mediated by factors released into the circulation by the healing tissue [6-8]. We have recently characterized one of these factors, a positively charged 14-amino-acid peptide named osteogenic growth peptide (OGP), identical to the Cterminus of histone H4 [9]. Native-like synthetic O G P (sOGP) increases bone formation and trabecular bone density when administered in vivo in rats. Immunoreactive O G P in high abundance is present physiologically in the serum, mainly in the form of an OGP-

Correspondence to: I. Bab, Bone Laboratory, Faculty of Dental Medicine, The Hebrew University of Jerusalem, P.O. Box 1172, Jerusalem 91010, Israel.

O G P binding protein (OGPBP) complex. Modifications introduced to the O G P termini suggest a role for the N-terminal region in the O G P - O G P B P complex formation. A marked increase in serum-bound and unbound O G P accompanies the osteogenic phase of postablation marrow regeneration and associated systemic osteogenic response. In vitro O G P is a potent mitogenic promoter of osteoblastic and fibroblastic ceils [9]. In osteoblastic MC3T3 E1 cells and in NIH 3T3 fibroblasts, O G P stimulates proliferation dose dependently at 10 -1410-t3 M and 10-11-10-l0 M, respectively, followed by an inhibition at higher concentrations. This inhibitory effect is shared by a synthetic peptide presenting the reverse sequence of O G P (designated retro-OGP or rOGP), r O G P fails to stimulate cell proliferation in vitro or bone formation in vivo. The present study reports the effect of modifications of the O G P molecule at the N- and C-termini and at

274 Tyr-10. These modifications were introduced in order to evaluate possibilities of labeling synthetic OGP to obtain either radio- or fluorescent ligands. This work was presented at 14th Annual Meeting of the American Society for Bone and Mineral Research [10]. Materials and Methods

Cell cultures. Tissue culture ingredients were purchased from Biological Industries (Beit Haemek, Israel). Osteoblastic MC3T3 El (the gift of Dr. H. Kodama, Tohoku Dental University, Fukushima, Japan) and fibroblastic NIH 3T3 cells were maintained in alpha minimal essential medium supplemented with 10% fetal calf serum (FCS) and subcultured twice a week. Cells for experiments were derived from maintenance cultures at confluency. Cells in the test cultures were seeded in 16-mm multiwell dishes (Nunc, Roskilde, Denmark) at 2- 104 cells per well and incubated in the same medium at 37°C in CO2/air. For the initial 46 h the medium was supplemented with 10% FCS and 0.25% nucleoside-ribonucleosides. During this time period the cell content resumed its seeding values. This was followed by an additional 2-h incubation under serum free conditions prior to the addition of peptides. These conditions did not have a recordable effect on the cell number. The different peptides were preincubated with 4% BSA (Sigma, St. Louis, MO, USA; cat no. A-7030) for 30 min at 37°C, and then added to the cultures. Cell counts were carried out after 48 h using a hemocytometer. In either cell system, the control cultures showed an approx. 2-fold increase in cell number during the 48-h treatment with BSA. Synthetic peptides. Substance P (acetate salt), adrenocorticotropic hormone fragment 1-14 (human), Lys-bradykinin (acetate salt), poly-L-lysine (hydrobromide, MW 1000-4000), osteocalcin fragment 7-19 (human), leucine enkephalin (acetate salt), poly-glycine (MW 2000-5000) and poly-L-glutamic acid (MW 200015 000, sodium salt) were purchased from Sigma. sOGP, rOGP, N-Ac[Cys°]OGP and [CystS]OGP NH 2 were prepared according to the standard solidphase peptide-synthesis methodology [11]. Side chains were deprotected and the peptide was cleaved from the resin using the HF procedure. The crude synthetic peptides were initially purified on a Sephadex G15F 3 x 35 cm column, eluted with 50% (v/v) aqueous acetic acid. Further purification was accomplished on a Waters DeltaPrep 3000 HPLC instrument equipped with a PrePak 1000 module (Millipore, Milford, MA, USA). The cartridge was pumped with 2 1 5-33.5% acetonitrile gradient at a flow rate of 100 ml/min. Peptide composition was confirmed by amino-acid sequence determination. The identity between the synthetic and native OGP was further established by mass

spectroscopy and retention time on reverse-phase HPLC. N-Acetyl[Cys°[S(N-Ethylsuccinimidyl)]-OGP (Ac [Cys°(S-NEtSucc)]OGP) and [CyslS(S-2(N-Ethylsuc cinyl))]OGP-Carboxamide ([Cys15(S-NEtSucc)]OGP NH 2) were prepared by incubating N-Ac[Cys°]OGP or [Cys151OGP-NH2, respectively, with a 10-fold molar excess of N-ethyl-maleimide (NEM) in 0.5 M PBS (pH 7.0), for 1 h at room temperature. The S-NEtSucc analogues were purified on HPLC Vydac protein C18 reverse-phase column (The Separation Group, Hesperia, CA, USA) eluted with 2 1 0-40% acetonitrile gradient at a flow rate of 100 ml/min. Partially purified acidic fibroblast growth factor (aFGF) from bovine brain was the gift of Dr. I. VIodavsky, Department of Radiation and Clinical Oncology, Hadassah University Hospital (Jerusalem, Israel). Radioiodination. Radioiodination of sOGP was carried out as described previously [9] in a reaction mixture containing sOGP, 40 mCi/ml Na~251 (Nuclear Research Center, Negev, Israel) and 100 tzg//zl Iodogen (Pierce, Rockford, IL, USA). The unreacted iodide was removed using a reverse phase Sep-Pak C18 cartridge (Waters Associates, Milford, MA, USA). The Na1251 was washed out with 100 mM phosphate buffer (pH 7.5), followed by elution of the radioiodinated peptide with 0.1% TFA in acetonitrile/H20 (1:1). The monoiodinated sOGP was purified on a reversephase HPLC Vydac Protein C4 column (The Separation Group) eluted with 30 ml linear gradient of 1525% acetonitrile in water containing 0.1% TFA at a flow rate of 1.0 ml/min. Binding assay. Testing the binding of the synthetic OGP analogues to the OGPBP was done as previously reported [9]. [3-125I(Tyrm)]sOGP, 103 cpm/ml, was incubated for 30 min at 37°C with 1:200 dilution in PBS of fresh serum obtained from 300 g male Sabra rats. The incubation was carried out in the absence of unlabeiled peptides or in the presence of 0.2/zmol/ml of the respective synthetic analogue. The bound and unbound [3-~zSI(Tyrl°)]sOGP were separated using a cathodic native PAGE system consisting of 15% polyacrylamide gel and 1 mM phosphate buffer (pH 6.0), as the electrophoresis buffer. The gels were dried and autoradiographed by exposure to Agfa Curix RP2 film. Competitive ELISA. Anti-OGP IgG fraction was prepared as described previously [9]. In ELISA this antibody preparation reacted specifically with the OGP C-terminal region as revealed from its binding to Ac[Cys°(S-NEtSucc)]OGP and failure to bind to [CysIS(S-NEtSucc)]OGP-NH2, [3-I(Tyrl°)]sOGP and rOGP [9]. Dynathech microtiter plates were coated with 100 /zl of 10/xg/I sOGP in 0.5 M sodium carbonate buffer (pH 9.5), overnight at 4°C, blocked with 1% gelatin in PBS at 37°C for 1 h and incubated with 100 /zl anti-

275 OGP IgG for 2 h at room temperature. Antibody binding was detected by the binding of anti-rabbit IgG-alkaline phosphatase conjugate and visualized with p-nitrophenyl phosphate substrate in diethanolamine buffer (pH 8.4). In competitive ELISA the anti-OGP IgG was preincubated for 30 min at 37°C with the respective synthetic OGP analogue. Results

The synthetic OGP analogues used in this study are shown in Table I. These analogues were tested for their (i) proliferative effect on osteoblastic and fibroblastic cells; (ii) complex formation with the OGPBP and (iii) binding to the anti-OGP antibodies. In line with previous observations [9] sOGP markedly stimulated the number of MC3T3 E1 osteoblasts and NIH 3T3 fibroblasts, with a peak effect at 10 -13 M and 10 -11 M peptide concentration, respectively. Higher concentrations induced a dose dependent inhibition in the respective cell systems (Figs. 1, 2 and 6). This pattern of osteoblast and fibroblast reaction was similar to that evoked by aFGF in the same culture systems (Fig. 3). However, unlike the higher sensitivity to sOGP shown by the osteoblasts as compared to fibroblasts, the peak effect of aFGF in the NIH 3T3 cells was seen at a lower concentration than in the MC3T3 E1 osteoblasts. The rOGP was designed to address the issue of specificity to the nonpalindromic nature of the sequence. It maintains the net charge, hydrophobicity and relative side chain order of the OGP. In both the osteoblastic and fibroblastic cell lines this analogue failed to exhibit the OGP-like stimulation of cell proliferation but shared the OGP inhibitory effect (Figs. 2, 4 and 5). This inhibition, approximately 40% with reference to control cultures incubated without any exogenously added peptides, showed a dose response relationship between 1 0 - 1 4 - 1 0 -9 M and 1 0 - t 2 - 1 0 -7 M in the MC3T3 E1 and NIH 3T3 cells, respectively (Figs. 4, 5).

We have previously shown that the binding properties of [3-125I(Tyrl°)]sOGP to the OGPBP are similar to those of the sOGP. However, the [3-125I(Tyrl°)] sOGP does not bind to the anti-OGP antibodies [9]. The [3-I(Tyrl°)]sOGP was tested herein for its proliferative activity in order to evaluate the relevance of using the [3-125I(Tyr HI)]sOGP as a radioligand in future studies involving the putative OGP receptor. The data indicate that the effect of the [3-I(Tyrl°)]sOGP is similar to that of the rOGP, namely, it not only failed to stimulate cell proliferation, but induced a moderate dose-dependent inhibition, particularly in the MC3T3 E1 osteoblasts (Figs. 2, 4 and 5). The Cys and Cys(S-NEtSucc) modifications of the OGP termini were introduced to examine the role of these regions in the OGP mechanism of action and test further possibilities for labelling approaches. Both the N-Ac[Cys°]OGP and [CyslS]OGP-NH 2 shared the OGP proliferative effect in the MC3T3 E1 and NIH 3T3 cells. However, although the peak effects in the two cell types were found at the same concentrations as those of the sOGP, their magnitude was consistently somewhat lower (Fig. 1). The mitogenic activity of the Ac[Cys°(S-NEtSucc)]OGP was the lowest, though still significant. The [CyslS(S-NEtSucc)]OGP-NH2 did not stimulate proliferation at all but shared some of the inhibitory effect of the rOGP, [3-l(Tyrt°)]sOGP and the high concentration of sOGP (Fig. 2). The OGP primary structure contains one Lys and two Arg residues (Table I), conveying a high positive net charge to the molecule. This charge was kept unchanged in the synthetic OGP analogues tested. Thus, one plausible explanation for the inhibitory effect on cell proliferation shared by the native-like sOGP and all other analogues is a nonspecific cellgrowth suppression induced by the net positive electrical charge. To test the feasibility of this approach, the effect of the sOGP and its analogues was compared to that of other small peptides which vary in the nature and magnitude of their charge. Like the OGP ana-

TABLE I

Primary structure of synthetic OGP analogues sOGP rOGP [3-I(Tyr t0)]sOGP

I

.

A-L-K-R-Q-G-R-T-L-Y-G-F-G-G

.

14

1

.

G-G-F-G-Y-L-T-R-G-Q-R-K-L-A

.

14

I

.

A-L-K-R-Q-G-R-T-L-Y-G-F-G-G

.

14

I 3-I

N-Ac~(Cys°)]OGP Ac[Cys°(S-NEtSucc)]OG P

0 0

Ac-C-A-L-K-R-Q-G-R-T-L-Y-G-F-G-G

.

14

Ac-C-A-L-K-R-Q-G-R-T-L-Y-G-F-G-G

.

14

l $-NEtSucc

[(CystS)]OGP-NH 2 [CYsls(S-NEtsucc)]OG P'NH 2

1

.

A-L-K-R-Q-G-R-T-L-Y-G-F-G-G-C-NH

2

I

.

A-L-K-R-Q-G-R-T-L-Y-G-F-G-G-C-NH

z

I $-NEtSucc

15 15

276 55

40

B

A

?O

lm

30

45

X O .J -I I,M

20

0

10

L~L

C

,~

10

,

!~3

10

:

'.,,

10

'

!~

25

I......-41

C

10

113

I

10

ill

ig

i

10

I

10

i 7

10

[OGP or OGP analogue](M} Fig. 1. Effect of Cys-extended synthetic OGP analogues on proliferation of osteoblastic MC3T3 E1 cells (A) and NIH 3T3 fibroblasts (B). Test cultures were set as described in text. Symbols: ( • ) , sOGP; (11), N-Ac[Cys°]OGP; (D), [C-'yslS]OGP-NH2 . Data are mean :t: S.E. of triplicate cultures representing three repetitive experiments.

logues, all the net charged peptides, either positive or negative, suppressed the MC3T3 E1 and NIH 3T3 cell numbers dose dependently (Figs. 4, 5). The osteoblastic cells showed a higher sensitivity resulting in 20-35% inhibition compared to 5-23% revealed by the fibroblasts. The uncharged or slightly charged peptides had no effect in either cell system (Fig. 6).

It was shown previously that like the OGP, the [Cys]5(S-NEtSucc)]OGP-NH2 readily forms a complex with the OGPBP. The rOGP and Ac[Cys°(S NEtSucc)]OGP displayed a substantially lower OGPBP binding activity [9]. These observations were now extended to show that like Ac[Cys°(S-NEtSucc)]OGP, the N-Ac[Cys°]OGP competed for binding with the 50

35

B

A

40

O O3 .-I ,.J

IJU

(.)

15

20

-@ 10 C

10

10

10

10

~

C

t I.t ~

I

10

/.l

10

I

I

I 9

10

I

I7

10

[sOGP or OGP analogue](M) Fig. 2. Effect of Cys(S-NEtSucc)-extended synthetic OGP analogues on proliferation of osteoblastic MC3T3 E1 cells (A) and NIH 3T3 fibroblasts (B). Test cultures were set as described in text. Symbols: ( - - • - - ) , sOGP; (---o---), rOGP; ( - - o - - ) , [3-I(Tyr]°)]sOGP; ( - - o - - ) , Ac[Cys°(S NEtSucc)]OGP; (---©---), [CyslS(S-NEtSucc)]OGP-NH 2- Data are mean :t: S.E. of triplicate cultures representing three repetitive experiments.

277 25

65 B 60

?

2o

55

O

50

o

/

.J ,J

15

45

(

40

I

10

II

C

I0

1

I

I

100

1000

35

I ~

C

I

I

I

i

i

.1

1

10

100

[FGF](ng/ml)

Fig. 3. Effect of aFGF on proliferation of osteoblastic MC3T3 E1 cells (A) and NIH 3T3 fibroblasts (B). Test cultures were set as described in text. Data are mean :t: S.E. of triplicate cultures representing three repetitive experiments.

O G P B P less than the sOGP or the C-terminal-modified analogues (Fig. 7). rOGP, [3-I(Tyrm)]sOGP and [CyslS(S-NEtSucc)] O G P - N H 2 do not react with the anti-OGP antibodies

35

tO

A

25

~ 1II

°~

JO r¢-

[9]. The other analogues tested herein all display binding to the antibody. The results suggest that the sOGP has the highest binding affinity and Ac[Cys°(S NEtSucc)]OGP the lowest (Fig. 8).

•

15

i

I

m

r"

0 x_

ft. -5

-15

A_,5 10

.13

10

I

I 11

10

[,

I g

10

!,~ , 10

,_. 10

,

1. 10

,

,_~ 10

[Peptide](M) Fig. 4. Inhibition of osteoblastic MC3T3 E1 cell proliferation by net charged peptides. (A) Synthetic OGP analogues. Symbols: (---o---), rOGP; ( - - e - - ) , [3-I(Tyrl°)}sOGP. (B) Reference peptides. Symbols: (---e---), substance P; ( - - e - - ) , poly-L-lysine; ( - - o - - ) , bradykinin; (---o---), poly-L-glutamic acid. Test cultures were set as described in Materials and Methods. Percent inhibition was calculated using control cultures incubated free of exogenously added peptides in chemically-defined medium containing 4% BSA. Data are mean+ S.E. of triplicate cultures representing three repetitive experiments.

278

B

40

1I

t-

30 i!

o

1

e~ re-

/

I)

//

20

m

.?

e"

10 L_

? -10

-13

I

10

L-11

10

I

|-9

I

1.7

10

a-13

I

10

10

/11

i

10

I g

10

I

1_7

10

[Peptide](M) Fig. 5. Inhibition of fibroblastic NIH 3T3 cell proliferation by net charged peptides. (A) Synthetic OGP analogues. Symbols: (---e---), rOGP; ( - - e - - ) , [3-1(TyrZ°)JsOGP. (B) Reference peptides. Symbols: (---e---). substance P; ( - - e - - ) , poly-L-lysine; ( - - © - - ) , bradykinin; (---©---), poly-L-glutamic acid. Test cultures were set as described in Materials and Methods. Percent inhibition was calculated using control cultures incubated free of exogenously added peptides in chemically-defined medium containing 4% BSA. Data are mean + S.E. of triplicate cultures representing three repetitive experiments.

50

20

B

A

I~

40

15

X

¢q

_J ,..1 u,l

30

t.--4t !,5 C 10

i

~-,3 ~ 10

~.,, 10

~

L.9 10

20

L-.4t , G 10

~

A, . 10

'

'* 10

~

'7 10

rPeptidel(M) Fig. 6. C h a l l e n g e of o s t e o b l a s t i c M C 3 T 3 E1 cells (A) a n d N I H 3T3 fibroblasts (B) with u n c h a r g e d or slightly c h a r g e d p e p t i d e s . T e s t c u l t u r e s w e r e set as d e s c r i b e d in M a t e r i a l s a n d M e t h o d s . Symbols: ( - - • - - ) , s O G P ; (---e---), poly-glycine; ( - - e - - ) , osteocalcin; ( - - o - - ) , leucine e n k e p h a l i n ; (---O---), A C T H 1 - 14. D a t a are m e a n + S.E. of t r i p l i c a t e c u l t u r e s r e p r e s e n t i n g t h r e e repetitive e x p e r i m e n t s .

279

be

a

d .e

Fig. 7. Autoradiographs of cathodic native PAGE demonstrating binding of synthetic OGP analogues to OGPBP. Lane a, [312"SI(Tyr1°)]sOGP alone; lanes b-g, {3-z~I(Tyrl°)]sOGP preincubated with fresh rat serum in absence of synthetic OGP analogues (b) or presence of 208 nmol/ml sOGP (c), N-Ac[Cys°]OGP (d), Ac[Cys°(S-NEtSucc)]OGP (e), [CyslS]OGP-NH2 (f), [CystS(S NEtSucc)]OGP-NH 2 (g). See Materials and Methods for experimental details.

Discussion

In the present study the synthetic OGP analogues were tested for three activities, i.e., their effect on cell proliferation, OGP-OGPBP complex formation and binding to the anti-OGP antibodies. The sOGP displayed the highest activity in all three categories. The other analogues can be divided into three groups.

0.9

0.8

0.7

05

~.

~

0

04

[

0"30

I

2

|

I

4

,

I

6

i

I

.8

i

I

10

[sOGP or OGP analoguel(ng/rnl) Fig. 8. Competitive ELISA showing binding of synthetic OGP analogues to anti-OGP antibodies. ( • ) , sOGP; ( • ) , Ac[Cys°]OGP; (0), Ac[Cys°(S-NEtSucc)IOGP; (121), ICysIs]OGP-NH 2. See Materials and Methods for experimental details.

Group 1 includes the N-Ac[Cys°]OGP, [CysIS]OGPNH 2 and Ac[Cys°(S-NEtSucc)]OGP which display all the OGP activities, but in most instances their effects are of a lower magnitude. These analogues may therefore be considered for the preparation of radio- or fluorescent ligands by using [35S]Met, [35S]Cys or respective conjugates obtained with maleimide derivatives [12], which can also be used for the design of additional antigenic and solid-phase affinity preparations [13,14]. Group 2 is comprised of the [3l(Tyr t°)]sOGP and [CysIS(S-NEtSucc)]OGP-NH 2. Like OGP these molecules fully retain the binding activity to the OGPBP and thus provide tools for studying the OGP-OGPBP complex formation. Actually, they have already been employed successfully in such experiments [9]. Group 3 consists of the rOGP that does not share any of the OGP activities, except for the inhibition of cell proliferation at high concentrations, indicating, at least, that the nonpalindromic nature presented by the OGP backbone is essential for these activities. Maintenance of side chain order and the net positive charge are not sufficient to maintain the rest of the OGP-like activities. The retention of the OGP proliferative effect by [Cys151OGP-NH2 suggests that unlike some other bioactive peptides, such as bradykinin and angiotensin II [15-19], the free C-terminus does not have a role in this activity. However, the complete loss of agonistic activity following modification of the thioi in Cys-15 by N-ethylmaleimide, or monoiodination of Tyr-10 indicate that the C-terminal region, at least from Tyr-10 to Gly-14, has a limited latitude for structural modifications without affecting the interaction of OGP with its putative receptor. The importance of this region is further stressed by the retention of the OGP stimulation of cell proliferation following a similar modification introduced at the N-terminus. Like OGP, other regulatory peptides of which the C-terminal region is critical for bioactivity, such as ACTH, gastrin, tachykinins, bradykinin, cholecystokinin and angiotensin II, show a high prevalence of Phe, Gly and Tyr in this region [15,20]. It is also of interest that a Chou-Fasman analysis [21,22] of OGP predicts a high incidence for a/3-turn conformation in proximity to the C-terminus consisting of the Tyrl°-Gly~LPhe12-Gly~~. Similarly, /3-turn conformations were predicted in a large number of regulatory peptides as a critical characteristic of their bioactive conformation [23]. In the OGP, a conformation similar to the bioactive one may also pertain to the peptide's interaction with the antiOGP antibody preparation inasmuch as this interaction is also inhibited by the Cys(S-NEtSucc) extension of the C-terminus and iodination of Tyr-10. Another region showing a high probability for a /3-turn conformation consists of the LysS-Arg4-Gln 5Gly 6 sequence in proximity to the OGP N-terminus.

280 Since the N-, but not the C-terminal modifications partially inhibited the O G P - O G P B P complex formation, it is suggested that the N-terminal region may be of relevance to this interaction. W h e n considered together, the present results, in particular the partial inhibition of the O G P activity following extension of the C-terminus by Cys, and to a greater extent by Cys(S-NEtSucc), may suggest a role for the O G P B P in the interaction of the O G P with its putative receptor. All of the O G P analogues tested, including the native-like O G P , induce a m o d e r a t e d o s e - d e p e n d e n t inhibition of osteoblastic and fibroblastic cell proliferation. T h e most p r o m i n e n t feature c o m m o n to these analogues is the high positive net charge conveyed by the Lys and two Arg residues. Therefore, we considered the possibility that the inhibitory activity of O G P at high concentrations, as well as that o f its analogues even at lower concentrations, is associated with this net charge. In the absence of available neutral O G P analogues, we tested the in vitro proliferative effect of several small, cationic, anionic and neutral (or slightly charged) peptides. N o n e of these peptides is structurally related to O G P and n o n e stimulated the osteoblastic or fibroblastic cell proliferation. Still, all the cationic and anionic peptides induced a m o d e r a t e inhibition in the cell n u m b e r reminiscent of the antiproliferative effect of the nonmitogenic O G P analogues. At least the cationic peptides are known to cause cell death in a m e c h a n i s m that involves synergism with h y d r o g e n peroxide [24]. It thus a p p e a r s that the suppression of cell n u m b e r seen at high O G P concentrations is associated with the peptide's high positive net charge. It is further suggested that the overall e n h a n c e m e n t / i n h i b i t i o n pattern of the O G P dose response relationship represents at each c o n c e n t r a t i o n a net effect consisting of a structure-related specific mitogenic activity and the nonspecific charge induced cell death. In conclusion, we d e m o n s t r a t e a differential respective role for the O G P N- and C-terminal regions in the O G P - O G P B P complex formation, O G P mitogenic activity and recognition by a n t i - O G P antibodies. In addition, our data suggest an association between the O G P electrical net charge and its biphasic effect on cell proliferation.

Acknowledgement Part of work was d o n e towards a Ph.D. at the H e b r e w University o f Jerusalem (Z.G. and N.M.). This

study was s u p p o r t e d by grants from the Israel Ministry of Science and T e c h n o l o g y and the Basic Research Foundation, T h e Israel A c a d e m y of Science and Humanities.

References 1 Bab, I., Gazit, D., Massarawa, A. and Sela, J. (1985) Calcif. Tissue Int. 37, 551-555. 2 Foldes, J., Naparstek, E., Statter, M., Menczel, J. and Bab, 1. (1989) J. Bone Miner. Res. 4, 643-646. 3 Gazit, D., Karmish, M., Holzman, L. and Bab, I. (1990) Endocrinology 126, 2607-2613. 4 Einhorn, T.A., Simon, G., Devlin, V.J., Warman, J., Sidhu, S.P. and Vigorita, V.J. (1990) J. Bone Joint Surg. Am. 72, 1374-1378. 5 Mueller, M., Schilling, T., Minne, H. and Ziegler, R. (1991) J. Bone Miner. Res. 6, 401-41{). 6 Bab, I., Gazit, D., Muhlrad, A. and Shteyer, A. (1988) Endocrinology 123, 345-352. 7 Gazit, D., Shteyer, A. and Bab, I. (1989) Connect. Tissue Res. 23, 153-161. 8 Lippiello, L., Chavda, D. and Connolly, J. (1992) J. Orthop. Res. 10, 145-148. 9 Bab, I., Gazit, D., Chorev, M., Muhlrad, A., Shteyer, A., Greenberg, Z., Namdar, M. and Kahn, A. (1992) EMBO J. 11, 18671873. 10 Greenberg, Z., Chorev, M., Muhlrad, A., Shteyer, A., Namdar, M. and Bab, I. (1992) J. Bone Miner. Res. 7 (Suppl. 1), $227. 11 Barany, G. and Merrifield, R.B. (1979) in The Peptides, Vol. 1 (Gross, E. and Meierhofer, J., eds.), pp. 1-27, Academic Press, New York. 12 Chorev, M., Caulfield, M., Roubini, E., McKee, R.L., Gibbons, S.W., Leu, C.T., Levy, J.J. and Rosenblatt, M. (1992) Int. J. Peptide Protein Res. 40, 445-455. 13 Kitagawa, T. (1981) in Enzyme Immunoassay (Ishikawa, E., ed.), pp. 81-89, Igaku-Shin Medical Publishers, Tokyo. 14 O'Sullivan, M.J. and Marks, V. (1981) Methods Enzymol. 73, 147-166. 15 Rudinger, J. (1971) in Drug Design, Vol. 2 (Ariens, E.J., ed.), pp. 319-419, Academic Press, New York. 16 Riniker, B. and Schwyzer, R. (1961) Helv. Chim. Acta 44, 674-676. 17 Schwyzer, R. and Turrian, H. (1960) in Vitamins and Hormones, Vol. 18 (Harris, R.S. and Ingle, D.J., eds.), pp. 237-288, Academic Press, New York. 18 Schwyzer, R. (1961) Helv. Chim. Acta 44, 667-674. 19 Stewart, J.M. (1968) Fed. Proc. 27, 63-66. 20 Meierhofer, J. (1977) in Peptide Hormones, Vol. 12 (Clarke, F.H., ed.), pp. 158-169, Academic Press, New York. 21 Chou, P.Y. and Fasman, G.D. (1978) Adv. Enzymol. 47, 45-148. 22 Ross, A.M. and Golub, E.E. (1988) Nucleic Acids Res. 16, 1801-1812. 23 Bakalkin, G.Y., Rakhmaninova, A.B., Akparov, V.K., Volodin, A.A., Ovchinnikov, V.V. and Sarkisyan, R.A. (1991) Int. J. Peptide Protein Res. 38, 505-510. 24 Ginsburg, I., Gibbs, D.F., Schuger, L., Johnson, K.J., Ryan, U.S., Ward, P.A. and Varani, J. (1989) Free Radic. Biol. Med. 7, 369-376.