JOURNAL OF BONE AND MINERAL RESEARCH

Volume 10, Number 5, 1995

Blackwell Science, Inc.

Osteogenic Growth Peptide Regulates Proliferation and

Osteogenic Maturation of Human and Rabbit Bone
Marrow Stromal Cells

D. ROBINSON,' I. BAB,' A N D Z. NEVO'

ABSTRACT
The recently discovered osteogenic growth peptide (OGP) has been shown to regulate proliferation in fibroblastic
and osteoblastic cell lines derived from rats and mice and also alkaline phosphatase activity in the latter was found
to be affected. In vivo the OGP enhances bone formation and trabecular bone density. The results of the current
study indicate that the OGP is also a potent regulator of marrow stromal cells from man and rabbit, as well as
rabbit muscle fibroblasts. The main OGP activity in both marrow systems is a marked stimulation of alkaline
phosphatase activity and matrix mineralization. In the rabbit-derived cell culture this enhancement is accompa-
nied by a reciprocal inhibition of proliferation. On the other hand, the human cells show a concomitant increase
of both parameters. The proliferative effect of the OGP is similar to that of growth hormone (GH) and basic
fibroblast growth factor (bFGF). The combined activity of the OGP with GH is smaller than that of each of the
polypeptides alone. The OGP and bFGF potentiate each other. Of the three polypeptides tested, OGP is the most
potent enhancer of alkaline phosphatase activity and mineralization. bFGF has no influence on these character-
istics of osteogenic maturation. The OGP maturational activity is unaffected by either GH or bFGF. These data
suggest that the marrow stromal cells serve as targets for the OGP that mediate the OGP-induced increase in
osteogenesis. The effect on the human cells implies a role for the OGP in clinical situations where the osteogenic
potential of bone marrow is involved.

INTRODUCTION teonal remodeling and repair processes of bone and bone

(OGP) iS a recently dis-
T HE OSl'EOGENIC GROWTH PEPTIDE
covered factor isolated from the osteogenic phase of
postablation marrow regeneration.(") Physiologically, it is
present in high abundance in the blood, mainly in the form
of a complex with a binding protein. The circulating levels
of the O G P are highly increased during the postablation
intramedullary bone formation and the associated systemic
osteogenic Adm'mistration of O G P in vivo
enhances bone formation and increases trabecular bone
mass. In vitro, the O G P stimulates the proliferation and
alkaline phosphatase activity in osteogenic cell lines. In
addition, it is mitogenic to fibroblasts.(2,')
The bone marrow stroma of man and other species con-
tains the osteogenic precursors that give rise to osteoblasts
during physiologic endosteal (trabecular and cortical) and os-
marrow.('-'') It is therefore of direct relevance to assess the
potential regulatory role of the OGP in marrow stromal cell
systems, particularly those derived from human subjects who
also exhibit a systemic osteogenic reaction after extensive in-
sult to the marrow.") Accordingly, the present experiments
were designed to test the in vitro effect of the OGP compar-
atively on marrow stromal cells obtained from man and rabbit.
MATERIALS AND METHODS
Materiais
Tissue culture media and specific ingredients were pur-
chased from Biological Industries (Beit Haemek, Israel).
Native-like synthetic O G P (sOGP), Ala-Leu-Lys-Arg-Gln-
Gly-Arg-Thr-Leu-Tyr-Gly-Phe-Gly-Gly, and a synthetic pep-

'Department of Chemical Pathology, Sackler Medical School, Tel Aviv University, Tel-Aviv, Israel.
'Bone Laboratory, Faculty of Dental Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel.

690
OGP REGULATION OF HONE MARROW STROMAL CELLS
tide presenting the reverse sequence of OGP (designated line phosphatase or mineral were carried out o n day 10
retro-OGP o r rOGP). Gly-Gly-Phe-Gly-Tyr-Leu-Thr-Arg-
Gly-Gln-Arg-L.ys-Lcu-Ala, were prepared as previously de-
scribed'2.') according t o the standard solid-phase peptide syn- Histochernistty
thesis methodology.("" Basic fibroblast growth factor (bFGF)
from bovine brain was the gift of Dr. D. Gospodarowicz,
Cancer Research Institute, UCSF. Recombinant bovine by fast blue, the hydrolysis product of naphthol-AS-phos-
growth hormone (GH) was from BioTechnology General phate.'"') Mineralized matrix was stained using alizarin red S
(Rehovot, Israel). Other reagents were from Sigma Chemical for the calcium.'2''
Co. (St. Louis, MO).
Cell Ciiltures
Bone marrow cells were isolated from adult male New
Zealand White rabbits, I-year-old or older. A cell suspen-
sion of unsclected marrow cells was prepared essentially as
described before.'"-'" In short, the femoral bones were sep-
arated from euthanizcd animals, and the epiphyseal ends
were removed. The marrow was then flushed out from the
diaphyscal shaft with phosphate buffered saline (PBS). Ap-
proximately 3 mI of sample containing 5.4 x 10x cells was
obtained from each femur. A suspension consisting mainly
of single cells was prepared by passing the marrow sample
several times through a tapering syringe needle. The cells
were seeded in 35 mm dishes at 3 X 10'-5 X 10' cellsicm'
and incubated in a-minimal essential medium (a-MEM)
containing 0.2.5';6 nucleoside-ribonucleosides at 37°C in 5%:
C02-air. As before, the medium was supplemented with
15% fctal calf serum (FCS), 25 ngiml of bFGF, 0.4 pgiml of
dcxamcthasone, and SO pg/ml of ascorbic After 2
days. most of the cells were washed off leaving only the
adherent cell layer. From this time onward the medium was
changed twice a week for 2 weeks. At this stage, when the
cultures showed multiple foci o f confluency, the cells were
harvested by trypsinization and pooled. Each femoral di-
iiphysis yielded -3 X l o 7 adherent cells. Human bone
marrow was retrieved from the fernoral shaft o f subjects,
55-ycars-old or older. undergoing total hip replacement
arthroplasty for the treatment of severe localized joint dis-
orders. The marrow tissue was manipulated, and primary
cultures were set in a manner similar to that described
earlier for the rabbit material. I n the human system, how-
ever, 3 wccks were required for the cultures to attain t o ii
state of confluency suitable for cell harvesting. Rabbit fi-
broblasts were harvested as previously from in vitro out-
growths originating in muscle explants.' "') First passage
cells were used t o test the effect of growth regulators.
Accordingly, the harvested cells were resuspended and sub-
cultured in the same medium at 7.3 X 10' cellsicm'. After
2 days the medium was removed, the cultures were washed
with PBS prior t o the addition of polypeptides in serum-free
defined cell culture medium (DCCM)-1 synthetic medium.
The different pcptides were preincubated with fatty acid-
free bovine serum albumin (BSA) for 30 minutes at 37°C:
and then added to the cultures. The final concentration of
BSA in the medium was 4%. The polypeptide containing
medium was replaced every 2 days. Cell counts. using a
Coulter Counter (Model D Industrial) with an orifice of 100
pm, and histochemical staining of sister cultures for alka-
69 I
after passaging.
Cells expressing alkaline phosphatase activity were stained
Quantitarive assessment of lristoclzeniical staining
The percent culture area positively stained for alkaline
phosphatase and overall alizarin red S staining intensity
were determined in situ using an image analyzer with :I
semiautomatic object counting module (Java Video Analy-
sis Software, Jandell Corporation, Corte Madera, CA) :is
previously The analysis was done on images
obtained at X40 magnification in a Model IMT Olynipus
Inverted Research Microscope, and the images were digi-
tized using a frame grabber (Imaging Technologies PC
Vision Plus Video Digitizing Board #3023, Jandcll Scien-
tific Hardware and Accessories for JAVA. Corte Madera,
CA). Specifically, we measured the arca occupied by alka-
line phosphatase positive cells as the fractional number of
total culture area. Mineral calcium was determined bkised
on a standard curve prepared by reacting known amounts of
CaCI, (20-400 pg) with alizarin red S.
RESULTS
Proliferative effect of OGP
In rabbit marrow stromal cells, the overall proliferation
was slightly accelerated at 1 0 M peptide concentration,
followed by a marked inhibition at highcr concentrations
(Fig. IA). The peak of inhibition was seen at 10 "-10 ' M
sOGP. The decrease in overall cell number was acconipa-
nied by a reciprocal, - 10-fold increase in the proportion o f
culture occupied by alkaline phosphatase positive cells (Fig.
IB), while in rabbit muscle fibroblasts the peak of prolifer-
ative activity of the sOGP was at 10 ~' M (Fig. IC). Thus, in
the rabbit cell systems the sOGP became progressively less
effective at concentrations higher than 10 ' M. Human
marrow stromal cells also responded to the sOGP dose
dependently. However, in this system the sOGP concentra-
tion optimum for the proliferative effect and alkaline phos-
phatase positive cell number was the same, 10 "'-10 " M
(Fig. 2). The 3-fold increase in the relative number o f
alkaline phosphatase-positive cells in human marrow stro-
ma1 cells, was smaller compared with the rabbit m;ii-row
stromal cell culture (compare Figs. I B and 2B). Unlike the
overall cell number, which at higher concentrations was
inhibited towards control Icvcls, the fraction of alkaline
phosphatase-positive cells remained at suboptimal, how-
ever suprabasal, levels (Fig. 2B). In all three cell systems
studied, cultures challenged with rOGP showed baseline
levels of the overall cell counts and alkaline phosphatase-
positive cell number at all concentrations tested (Figs. 1 and
2). Figure 3 illustrates the overview of a control and an
692 ROBINSON ET AL.

30 ~

A A T -
30 50
ii 20
ei
I
Y 30
I?
J
10
w
0

0 10 4 1 ' I ' I I I I I

70
60
a
w 50
a 40
a
8
20 30

0 10
C
c)
0
ii 30 Effect of O G P on number (A) and alkaline Dhos-
\ ,

ei phatase activity (B) of human bone marrow stromal cells.

I Continuous line, sOGP; dashed line, rOGP. Control cul-
Y tures (c) were incubated in medium supplemented with
I?
J
20 BSA alone. Data are mean 2 SD of triplicate cultures per
condition representing three repetitive experiments.
w
0

10
A B

FIG. 1. Effect of O G P on rabbit bone marrow stromal

cells and muscle fibroblasts. (A) Number of total bone
marrow stromal cells per square centimeter. (B) Alkaline
phosphatase activity of bone marrow stromal cells ex-
pressed as a percentage of culture dish area stained posi-
tively for alkaline phosphatase (total culture area equals
100%). (C) Number of muscle fibroblasts. Continuous line,
sOGP; dashed line, rOGP. Control cultures (c) were incu-
bated in medium supplemented with BSA alone. Data are
mean 2 S D of triplicate cultures per condition representing FIG. 3. Human bone marrow stromal cell cultures stained
three repetitive experiments. for alkaline phosphatase. (A) Control culture. (B) sOGP-
treated culture.

OGP-treated culture plates stained for alkaline phosphatase.

Figure 4 illustrates the same plates under higher magnifica-
tion, showing much more intensive and extensive staining for
alkaline phosphatase in the OGP-treated cells. The comput-
erized measurements of the alkaline phosphatase positively
stained area show in the control (plate A in both Figs. 3 and 4)
10.5% and in the OGP-treated cells (plate B in both Figs. 3
and 4) 79.3%.
Effect of OGP on mineralization
In both the human and rabbit marrow stromal cell cul-
tures, the sOGP concentration that optimally enhanced cell
number (lo-' M) markedly stimulated matrix mineraliza-
tion, expressed as a 7- to 10-fold increase in the amount of
alizarin red S staining (Fig. 5).
OGP REGULAIION OF BONE MARROW STROMAL CELLS
FIG. 4. Enlarged areas from Figs. 3A and 3B. (A) Cells
from conlrol cultures (magnification X400). (B) Cells from
sOGP-treated cultures (magnification X 100).
C o o p m t i o n of OGP with otlier polypeptide growth
r q ulu tors
These experiments were conducted in the human marrow
stromal cell system using- the optimal 10 '' M .IiolvDeDtide
i l
HUMAN
7-
CNTL sOGP
FIG. 5. Effect of O G P on calcium accumulation in cul-
tures of bone marrow stromal cells. sOGP concentration
was lo-' M. Control cultures (CNTL) were incubated in
medium supplemented with BSA alone. Data are mean 5
SD of triplicate cultures representing three repetitive
experiments.
concentration. In this system, either the sOGP, hFGF, or
GH alone showed a -2-fold stimulation of the overall cell
number. The sOGP in combination with the hFGF had :I
3-fold stimulatory effect, similar t o the influcncc of I+GF
together with GH. However, when the sOGP was tested
with GH, the effect was substantially smaller than that o f
sOGP alone, -135% over the baseline control (Fig. 6). O f
the three polypeptides tested, the sOGP had the highest
stimulatory effect on the relative number of alkaline phos-
phatase positive cells and matrix mineral content. Trcat-
ment with GH alone had a smaller influence on either
parameter, and cultures challenged with bFGF remained at
control levels. The activity of the sOGP was not altered in
the presence of bFGF or GH. Similarly, the addition of
bFGF had no effect on the GH activity (Fig. 7).
DISCUSSION
So far, the OGP proliferative and maturational (ostco-
genic) activities were tested only in immortalized cell
Now the in vitro identification of freshly isolatcd
marrow stromal cells a s targets for the OGP strongly sug-
gests that these cells function in vivo in the same capacity
and mediate the OGP-induced enhancement of matrix svn-
L thesis and mineralization.'*' That these cells are capablc o f
694
40 m amount of mineralized matrix seen here in vitro and for the
CNTL OGP GH FGF GH FGF
+ +
sOGP sOOP
FIG. 6. Combined effect of OGP, GH, and bFGF on
proliferation of human bone marrow stromal cells. Concen-
tration of all polypeptides was 10 ' M. Control cultures
(CNTL) were incubated in medium supplemented with
BSA alone. Data are mean +- S D of triplicate cultures per
condition representing three repetitive experiments.
" phosphatase activity. As in MC3T3-EI osteoblasts,") this
CNTL OOP OH FOF 0,H FQF
aOGP SOOP
FIG. 7. Combined effect of OGP, GH, and bFGF on
alkaline phosphatase activity and calcium accumulation in
cultures of human bone marrow stromal cells. Concentra-
tion of all polypeptides was lo-' M. Control cultures
(CNTL) were incubated in medium supplemented with
BSA alone. Data are mean 5 SD of triplicate cultures per
condition representing three repetitive experiments.
producing a mineralized, physiologic-like matrix in the ab-
sence of phosphate-rich compounds, such as P-glycerophos-
phate, was shown previously in long-term culture^.(*'^^^'
The results presented here indicate that the O G P is capable
of stimulating the number of the marrow-derived osteo-
genic cells represented by cells with alkaline phosphatase
activity. Since alkaline phosphatase catalyzes calcifica-
tion,"" this stimulation probably accounts for the increased
ROBINSON ET AL.
enhancement of trabecular bone formation and mineraliza-
tion reported previously in vivo."' It is still unclear, how-
ever, whether the O G P regulates the alkaline phosphatase
directly or alternatively stimulates mitogenesis in alkaline
phosphatase-positive osteoprogenitor cells that still retain a
potential for proliferation.
In the three stromal cell systems studied, namely rabbit
and human marrow cells and rabbit fibroblasts, the O G P
had basically a biphasic effect on cellular activity. This
mode of action of the O G P was reported before in serum-
free, BSA-supported stromal cell cultures, and it was sug-
gested that the O G P high electrical charge is involved in the
FGF
suppression of cell proliferation.",') This conclusion is not
+ supported by the present findings in bone marrow stromal
GH
cell cultures in as much as high concentrations of rOGP,
which has the same charge as the sOGP, did not inhibit the
cellular activity. An alternative mechanism, suggested in the
case of dual activity of other growth factors, was related to
the stereological interference of excess ligand with binding
to the receptor.(2"'
There are two main differences between the responses of
the human and rabbit marrow stromal cells to the OGP.
The first involves the optimal O G P dose for the stimulation
of proliferation. In the rabbit marrow cells a substantially
lower concentration is required for this effect as compared
with the rabbit muscle fibroblasts and human marrow stro-
ma1 cells. In vitro stromal cell systems constitutively contain
endogenous O G P that comprises an autocrineiparacrine
circuit. The peptide's levels are higher in osteogenic than
nonosteogenic cell cultures (data not shown). Thus, the
marrow cells derived from young adult rabbits may produce
more endogenous O G P than the nonosteogenic rabbit fi-
broblasts and the human marrow cells obtained from eld-
erly individuals. In this event, lower doses of exogenous
O G P would be sufficient to stimulate proliferation. This
explanation, of course, does not rule out other consider-
ations, such as differences in the number of the putative
O G P receptors, the ligand-receptor affinity, and the basic
proliferative potential of the cells. The second difference is
seen in the relationship between proliferation and alkaline
FfF
OH relationship in rabbit marrow stromal cells was reciprocal.
A similar reciprocity is also seen in osteogenic cells allowed
to mature in long term cultures or exposed to agents such as
transforming growth factor beta (TGF-P) and thyroid hor-
m ~ n e . ( ~ ~ -The' ' ) human cells showed a conjoint prolifera-
tive and maturational responses to the OGP, reminiscent of
the osteogenic cell response to calcitonin and bone mor-
phogenetic pr0tein-2.'"~~'' This difference may reflect the
occurrence of two distinct mechanisms for the O G P signal-
ing as in the case of TGF-P(34-'f') with different concentra-
tion optima in each of the cell types studied.
F G F and GH are potent modulators of proliferation and
activity of osteogenic cells. Either of these polypeptides may
be functionally associated with other regulatory mole-
~ u l e s . ' ~ ~ .Furthermore,
'~) in bone cell cultures, the F G F
presents an autocrirdparacrine circ~it.('"~~") It was there-
fore of interest to test the activities of F G F and G H in the
marrow-derived osteogenic cells and their combined effect
OGP REGULATION OF BONE MARROW STROMAL CELLS
with the OGP. As shown before in other bone cell sys-
tem"'.3''. L1.12) both the FGF and G H stimulated the prolif-
eration of the marrow stronial cells. Thc enhancement in
cell number was similar t o that induced by the OGP in
terms o f magnitude and polypeptide concentration. The
cooperation of thc O G P with each of the other polypeptides
had a different pattern. suggesting a different type mecha-
nism of involvement. I t has been shown previously that GH
~ ' stimulatory effect of the
promotes o ~ t e o g e n e s i s . ' ~The
O G P on this process, measured as the number of alkaline
phosphatase-positive cells and matrix mineralization, is
considerably higher and remains unaltered in the presence
of G H or FGF. As in other bone cell systems,("' the FGF
did not promote osteogenic parameters in the bone marrow
stromal cells.
Finally, when conaidcred together with the OGP-induced
systemic enhancement o f bone formation in experimental
animals, the stimul;ition of proliferation and osteogenesis in
the human marrow cells suggests a role for the O G P in
clinical situations where the osteogenic potential of bone
marrow is involved,'"" including orthopaedic and oral sur-
gery and particularly osteoporosis.
ACKNOWLEDGMENTS
This study was supported by grants (to IB) from the Basic
Rcscarch Foundationithe Israel Academy of Sciences and
Humanities, and the Israel Ministry of Health.
REFERENCES
I. Bab I , Gazit D. Muhlrad A. Shteycr A 1988 Regenerating bone
marrow produces ;I potent growth-promoting activity to osteo-
genic cells. Endocrinology 123:345-352.
2. H;ih I,Gazit D, ('horev M, Muhlrad A, Shteycr A, Greenherg
Z. Namdar M, K a h n A 1902 Histone H4-related osteogenic
growth peptide (OGP): ii novel circulating stimulator of osteo-
hlastic activity. EMBO J I1:1807-1873.
3. Grcenherg Z, C h o w M, Muhlrad A. Shteyer A. Namdar M,
Mansur N, Hah I lY93 Mitogcnic action o f ostcogenic growth
pcptidc (OGP): role of amino and carhoxy terminal regions
and charge. Biochim I3iophys ,4cta 1 l78:273-280.
4. Bnb I,Gazit D, Ma\sarawa A. Scla J I985 Removal of tihial
miirrow induces increased formation of hone and cartilage in
rat ni;indihul;ir condyle. Ciilcit Tissue I n t 3 7 5 5 1-555.
5. Foldcs J, Naparstck E, Statter M, Menczel J. Bah 1 l98Y
Ostcogcnic rcsponsc to marrow aspiration: increased 5erum
ostcocalcin and alkaline phosphatase in h u m a n hone marrow
donors. J Bone Miner Rcs 4643-646,
0 . Gazit D, Karmish M, Holznian L, Bab I IYYO Regenerating
marrow induces \ystemic incrcaw in osteo- and chondrogcne-
sis. Endocrinology 126:2607-2613.
7. Ashton BA, Allen TI). Howlctt CR. Eaglesom CC, Hattori A,
Owen M E IOXO F'c)rni:ition of Iwnc nnd cartilage by marrow
strom;d cells in diffusion chambers in vivo. Clin Orthop 151:
204-307.
8. Ashton BA, E,aglesoni CC, Bah I, Owen ME, IOXJ Distrihution
of fibroblastic colony-forming cells in rabbit bone marrow and
assay of their ostcogenic potential by a n in vivo diffusion
chamber method. Calcif Tissue Int 3683-80.
9. Bah I,Ashton BA, Syftcstxd GT, Owen M E 19x4 Assessment
6Y5
of an iri L , ~ L Y Idiffusion chamher method a s a quantitative a\say
for osteogenesis. Calcif Tissue Int 36:77-82.
10. Bab I , Ashton BA, Gazit D, Marx G, Williamson MC. Owen
M E 1986 Kinetics and differentiation of marrow s1roni;il cclls
in diffusion chamhers iri vivo. J Cell Sci 84139-151.
1 I. Bab I,Passi-Even L, Ciazit D, Sekelcs E, Ashton BA, I'eyI;in-
Ramu N, Ziv I.Ulmansky M I Y X X Ostcogenesis in irr i . i i ' o
diffusion chamber cultures of humen marrow cells. Hone
Miner 4373-386.
12. Benayahu D, Kletter Y, Zipori D. Wicntrouh S IYXO Bone
marrow-derived stromal cell line expressing ostcohlnstic phc-
notypc in rirro and ostcogenic capacity CI i > i i . o . J Cell Ph!siol
1401-7.
13. Fried A, Bcnayahu D, Wicntrouh S lOY3 Marrow strom;i-
derived ostcogenic clonal cell lines: putative stages in o\tco-
blastic differentiation. J Cell Physiol 15S472-482.
14. Bah I, €%horn T A 190.1 Regulatory role o f ostcogcnic growth
polypeptides in bone forniation and hemopoicsis. Crit Kcv
Eukaryot Gene Exp 3 3 - 4 6
15. Beresford JN 1989 Osteogcnic stem cells and the strtrmal
system of bone and marrow. Clin Orthop 24U:270-280.
16. Barany G, Merrifield R B lY79 Solid phase pcptidc synthc\i\.
In: Gross E, Meienhofcr J (eds.) The Pcptidcsil. Aciitlcniic
Press, New York. pp, 1-284.
17. Grigcriadis AE. Heersch JNM, Auhin J E 19x0 Diffcrcnti:ition
of muscle, fat, cartilage, and hone from progenitor cells p r c w i t
in a hone derived clonal cell population: effect o f dcxamcth;i-
sonc. J Cell Hiol 106:213Y-2151.
18. Noff D. Pitaru S. Savion N 1989 Basic fibroblast growth l:ictor
enhances the capacity of hone marrow cells to form hone lihc
nodules in virro. FEBS 250:61Y-h21.
l Y . lwata 1-1 l98Y Model for cartilage repair in ostcoarthritis:
chondrocyte differentiation with bone morphogcnctic protein
and the effect of NSAID therapy. In: Muir H. Hiroha1;i K.
Shichikawa K (cds.) Mechanisms of Articular Cartilage 1)aiii-
age and Repair in Ostconrthritis. Hogrcfc Kc Hubcr I'uhlislicrs.
Toronto, pp. 56-61,
2 0 Edelson JD, Shannon JM, Mason RJ IY88 Alkaline phos-
phatasc: a marker o f alveolar type I I cell differentintion. Am
Rev Resp Dis 138:1268-1275.
21. Garvey W, Fathi A, Higelow F, Jinicnez c', Carpenter B IOXY
Modified alizarin red S and modified silver nitrate methock I o i -
calcium determination. J Histotcch 12925-227.
22. Robinson D, Segal M, Halpcrin N. Nevo Z 1092 Ncuropcpti-
dcrgic innervation of intramuscular hcmangioma. Exp Mol
Pathol 56: 180- IYO.
23. Howlett CR, Cave J. Willianison M, Farmer J. Ali SY. Hah I.
Owen M E 1986 Mineralization in / i r rirro cultures of r:ihhit
marrow stromal cells. Clin Orthop 213:25 1-203.
24. Pfeilschifter J, D i d I,Pilz U, Brunottc K, Naumanii A, Zicglcr
R l Y Y 3 Mitogenic responsiveness o f human hone cells irr i.i/ro
to hormones and growth factors dccrctiscs with age. J Ikmc
Miner Res 8:7(17-717.
25. Weiss MJ, Cole DE, Ray li, Whytc MP, L;iffcrty MA, Mulivor
RA, Harris H 1989 First idcntification of a gene defect lor
hypophosphatasia: evidence that alkaline phosphatasc acts in
skeletal mineralization. Connect Tissue Res 21:9Y -100.
26. Tallarida RJ IYYO Further characterization of ii control n i c d c l
for ligand-receptor interaction: phase plane geometry, stability
and oscillation. Ann Biomed Eng 18671-684.
27. Ccntrella M, McCarthy TL, Canalis E IYX7 Tr;insforming
growth factor /3 is ii hifunctional regulator of replication and
collagcn synthesis in osteohlast-enriched cell cultures lrom
fetal rat bone. J Biol Chcm 2623286') -2874.
28. Noda M, Rodan G A 1987 Type p transforming growth t x t o r
(TGF-/3) regulation o f alkaline phosphat;isc expression ;ind
696
other phcnotype-related mRNAs in osteoblastic rat osteosar-
coma cells. J Cell Physiol 1333426-437.
29. Pfeilschifter J, D’Souza SM, Mundy GR 1987 Effect of trans-
forming growth factor p on osteoblastic osteosarcoma cells.
Endocrinology 121:2 12-2 18.
30. Sato K, Han DC, Fujii Y , Tsushima T, Shizume K 1987 Thy-
roid hormone stimulates alkaline phosphatase activity in cul-
tured rat osteoblastic cells (ROS 17/23) through 3,5,3’-triiodo-
L-thyronine nuclear receptors. Endocrinology 1201873-1881.
31. Fedarko NS, Bianco P, Vetter U, Gehron-Robey P 1990 Hu-
man bone cell enzyme expression and cellular heterogeneity:
correlation of alkaline phosphatase enzyme activity with cell
cycle. J Cell Physiol 144115-121.
32. Farley JR, Hall SL, Tarbaux NM 1989 Calcitonin (but not
calcitonin gene-related peptide) increases mouse bone cell
proliferation in a dose-dependent manner and increases mouse
bone formation alone and in combination with fluoride. Calcif
Tissue Int 45214-221.
33. Yamaguchi A. Katagiri T, lkeda T, Yoshiki S, Wozney JM,
Rosen V, Wang EA, Kahn AJ, Suda T, Yoshiki S 1991 Re-
combinant human bone morphogenetic protein-2 stimulates
osteoblastic maturation and inhibits myogenic differentiation
in vitro. J Cell Biol 113:681-687.
34. Wrana JL, Attisano L, Carcamo J, Zentella A, Doody J, Laiho
M, Wang XF, Massague J 1992 TGF beta signals through a
hctcromeric protein kinase receptor complex. Cell 71: 1003-
1014.
35. Chcn RH, Ebner R, Derynck R 1993 Inactivation of the type I1
receptor reveals two receptor pathways for the diverse TGF-P
activities. Science 2601335-1338.
36. Ebncr R, Chen RH, Shum L, Lawler S, Zioncheck TF, Lee A,
Lopez AR, Derynck R 1993 Cloning of a type I TGF-P recep-
tor and its effect on TGF-P binding to the type 11 receptor.
Science 2 6 0 1344-1348.
37. Noda M, Vogel R 1989 Fibroblast growth factor enhances type
ROBINSON ET AL.
beta 1 transforming growth factor gene expression in ostco-
blast-like cells. J Cell Biol 1092529-2535.
38. Bouillon R 1991 Growth hormone and bone. Horm Res
36(S~ppl1):49-55.
39. Globus RK, Pattcrson-Buckcndahl P, Gospodarowicz D I988
Regulation of bovine bonc cell proliferation by fibroblast
growth factor and transforming growth factor beta. Endocri-
nology l23:98 -1 05.
40. Globus RK, Plouet J, Gospodarowicz D 1989 Cultured bovine
bone cells synthesize basic fibroblast growth factor and store it
in their extracellular matrix. Endocrinology 124:1539-1547.
41. Slootweg MC, van Bud-Offers SC, Herrmann-Erlec MPM,
van der Mcer JM, Duursma SA 1988 Growth hormone is
mitogenic for fetal mouse osteoblasts but not for undifferenti-
ated bone cells. J Endocrinol 116Rll-R13.
42. Rodan SB, Wcsolowski G, Thomas KA, Yoon K, Rodan GA
1989 Effects of acidic and hasic fibroblast growth factors on
osteoblastic cells. Connect Tissue Res 20283-288.
43. Maor G, Hochberg Z, von der Mark K, Heinegard D, Silber-
man M 1989 Human growth hormone enhances chondrogen-
esis and osteogenesis in tissue culture system of chondropro-
genitor cells. Endocrinology 3251239-1245.
Address reprint requests to:
Pro$ Zvi Nevo
Department of Clinical Biochemistry
Sackler School of Medicine
Tel Aviv Universiq
Ramat Aviv, Tel Aviv 69978
Israel
Received in original form Septcmber 7, 1993; in revised form
December 14, 1994; accepted December 15, 1994.