| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Endocrine Research Unit, Division of Endocrinology, Metabolism, and Nutrition, Department of Internal Medicine (T.T., F.G., S.K., B.L.R., C.A.C.), and the Department of Biochemistry and Molecular Biology (T.C.S.), Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905
Address all correspondence and requests for reprints to: Cheryl A. Conover, Ph.D., Mayo Clinic, 200 First Street SW, 5–194 Joseph, Rochester, Minnesota 55905. E-mail: conover.cheryl{at}mayo.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2 (transcription factors involved in commitment to
osteoblast and adipocyte pathways, respectively), alkaline phosphatase,
type I collagen, and osteocalcin (markers of the osteoblast lineage),
or lipoprotein lipase and adipsin (markers of the adipocyte lineage)
and did not change alkaline phosphatase activity or type I collagen and
osteocalcin protein relative to total protein production. In contrast,
IGFs significantly increased type I collagen expression in
differentiated hMS(3–4) cells as well as mature osteoblasts and
promoted lipid accumulation in differentiated adipocytes. In summary,
hMS(3–4) cells express essential components of the IGF system and
respond to IGF treatment with increased proliferation. There was no
evidence for IGFs directly modulating the commitment of hMS(3–4) cells
to either osteoblast or adipocyte pathways, and their effects on
differentiation within these lineages were dependent on the stage of
cell maturation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Compared with the wealth of data on mature OB, much less is known about the effect of IGFs on marrow stromal cells, the multipotential precursor cells that can differentiate into OB or into other mesenchymal types, including adipocytes (15, 16, 17, 18). This is important to assess because a decrease in the number and the differentiation potential of bone marrow precursor cells (19) and/or alteration of the shunt of these precursor cells between OB and adipocyte lineages (20) have been related to postmenopausal osteoporosis and aging-induced bone loss. In a rodent model of stromal cells, IGF-I was effective in stimulating DNA synthesis, whereas its effects on differentiation were limited to stimulating type I collagen (Col I) expression (21). On the other hand, a recent study in primary cultures of human marrow stromal cells concluded that both IGF-I and IGF-II exerted proliferative effects, but inhibited collagen production (22). These discrepancies underline the limitations that species differences, heterogeneity of primary cultures, and serum-induced differentiation before the experiments introduce in the interpretation of IGF effects. Furthermore, neither study took into account IGF-binding proteins (IGFBPs) that have the potential to modulate IGF action, as previously shown in OB cells (1, 23), nor did they assess the role the IGFs may play in the shunting between OB and adipocyte differentiation.
We have recently developed and characterized human marrow stromal (hMS) cell lines (24, 25). These cell lines are conditionally immortalized with a mutant temperature-sensitive simian virus 40 large T antigen (SV40LTA) (26, 27, 28). When cultured at the restrictive temperature for SV40LTA, these cell lines regain a normal phenotype and provide a clonal human cell model that can be induced to differentiate either into osteoblasts that produce matrix or into adipocytes that accumulate neutral lipids droplets in their cytoplasm (24, 29). In this study, we identify these bipotential hMS cells as important targets of IGF action and define specific regulation of IGFBP production, cell proliferation, commitment, and differentiation in response to exogenous IGFs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,25-Dihydroxyvitamin D3, [3H]thymidine,
and [-32P]deoxy (d)-CTP were obtained from NEN Life Science Products (Boston, MA). L-Ascorbic acid
phosphate was purchased from WAKO Chemicals USA, Inc. (Richmond, VA).
Kits for the measurement of osteocalcin (OC) and procollagen protein
were generous gifts from Metra Biosystem (Mountain View, CA). IGFBP
antibodies were provided by Diagnostics Systems Laboratories, Inc. (Webster, TX; IGFBP-3 antiserum) and Chiron Corp. (Emeryville, CA; IGFBP-4, IGFBP-5, and IGFBP-6 antisera).
IGFBP-2 antibodies were obtained from Upstate Biotechnology, Inc. (Lake Placid, NY). The complementary DNA (cDNA) clones used
and their providers were: human IGFBP-1 and IGFBP-3, Dr. D. R.
Powell (Baylor College of Medicine, Houston, TX); and human IGFBP-2,
-4, -5, and -6, Dr. S. Shimasaki (Whittier Institute, La Jolla,
CA).
Cell culture
The cell line hMS(3, 4) was established in our laboratory from
human marrow stromal cells that were isolated by negative
immunoselection (25) and conditional immortalization with a mutant
temperature-sensitive SV40LTA (26, 27, 28, 29). At 34 C, the permissive
temperature for SV40LTA (24, 26, 27, 28), the large T antigen is active,
and the cells rapidly proliferate. At 39.5 C, the restrictive
temperature for SV40LTA, the large T antigen is inactive, the cells
regain a normal phenotype, proliferation is inhibited, and hMS(3, 4)
cells differentiate into either osteoblasts or adipocytes depending on
the appropriate conditions (24, 29).
The hMS(3, 4) cells were routinely cultured at 34 C in MEM
containing 10% (vol/vol) heat-inactivated FBS (HI-FBS), 0.2 µg/ml
geneticin, and 1% stock penicillin (10,000 U/ml)-streptomycin (10,000
µg/ml), as described previously (24). Medium was changed twice a
week. For experiments, cells were plated at a density of 2 x
104 cells/cm2. After 48 h at 34 C, cells
were washed twice in PBS and incubated at 39.5 C for a further 24
h in serum-free medium plus 0.1% (wt/vol) BSA to inactivate the
SV40LTA.
Normal human osteoblast-like (hOB) cells were cultured from trabecular bone explants obtained at the time of orthopedic procedures from patients who had no evidence of metabolic bone disease, as previously reported (30, 31).
Affinity cross-linking
[125I]IGF-I monolayer binding and chemical
cross-linking were performed as previously described (4). Briefly,
hMS(3, 4) cells were washed and incubated with
[125I]IGF-I (1 x 106 cpm/well) in the
presence or absence of unlabeled IGF-I (125 ng/ml) or insulin (100
µg/ml) for 2.5 h at 15 C. Cells were then washed and incubated
with 10 mM disuccinimydil suberate for 15 min at 15 C. The
reaction was stopped by the addition of Tris-EDTA quench buffer. Cells
were then solubilized, and cross-linked complexes were analyzed by 10%
SDS-PAGE under reducing conditions.
Northern blot analysis
Total cellular RNA was extracted using the RNA STAT-60 kit,
following the manufacturer’s instructions. Total RNA (20 µg) was
separated on a 1.5% agarose gel containing 2.2 mol/liter formaldehyde
and transferred to nylon membranes (Nytran, Schleicher & Schuell, Inc., Keene, NH). Relative loading and integrity of the RNA were
assessed by UV shadowing. The membranes were UV autocross-linked,
prehybridized at 43 C for 6 h, and hybridized at 43 C overnight
with 106 cpm/ml 32P-labeled cDNA probe (31, 32). Human IGFBP cDNA probes were labeled with [32P]dCTP
using a random primed DNA labeling kit (New England Nuclear Corp.,
Boston, MA). Filters were washed twice in 6 x SSPE (0.15
M NaCl, 0.01 M NaH2PO4,
1 mM EDTA, pH 7.4)-0.1% SDS at room temperature,
twice in 2 x SSPE-0.1% SDS at 42 C, and once in 0.1 x
SSPE-0.1% SDS at 60 C. Membranes were stripped according to the
manufacturer’s instructions for sequential probing for the six
IGFBPs.
Cell-conditioned medium
hMS(3, 4) cells were washed twice and incubated at 39.5 C in
MEM containing 0.1% (vol/vol) HI-FBS and 0.1% (wt/vol) of BSA with
the indicated experimental additions for 48 h. The conditioned
media were collected, centrifuged to eliminate cell debris, aliquoted,
and stored at -80 C.
Western ligand blot analysis
Western ligand blotting was performed as described previously
(31, 32, 33, 34). Aliquots of hMS(3, 4)-conditioned media (50 µl) were
analyzed by SDS-PAGE using a 7.5–15% gradient under nonreducing
conditions. The separated proteins were electroblotted onto
nitrocellulose filters, and the IGFBPs were identified by incubation
with [125I]IGF-I and -II at 4 C overnight and
visualization by autoradiography.
Immunoprecipitation analysis
hMS(3, 4) cell-conditioned media (1 ml) were incubated with
primary antibody (1:100 final dilution) and Protein G Plus/Protein
A-agarose (Oncogene Science, Inc., Uniondale, NY)
overnight at 4 C. Immunoprecipitated proteins were washed and analyzed
by Western ligand blotting.
Cell-free IGFBP-4 protease assay
IGFBP-4 protease activity was measured as previously described
(31, 33, 34). Aliquots of hMS(3, 4)-cell-conditioned media (50 µl)
were incubated at 37 C with [125I]IGFBP-4 without or with
5 nM IGF-II in a microfuge tube. Reaction products were
analyzed by SDS-PAGE and autoradiography.
Cell proliferation
[3H]Thymidine incorporation was assessed as
described previously (24, 31, 35), with minor modifications.
Preliminary experiments in hMS(3, 4) cells indicated that maximum
[3H]thymidine incorporation occurred 48–72 h after a
mitogenic stimulus (data not shown). Subconfluent cells were incubated
in 0.1% HI-FBS plus experimental additions for 72 h at 39.5 C
(the restrictive temperature for the SV40LTA), and 1 µCi
[3H]thymidine was added for the last 24 h.
To assess changes in cell numbers, cells were harvested by trypsinization and counted using an electronic cell counter (Coulter Electronics Ltd., Hialeah, FL).
Apoptosis
hMS(3, 4) cells were treated as described for the cell
proliferation studies. At the end of the 72-h incubation with and
without IGFs, cells were stained with Hoechst 33342 (36). Total and
apoptotic cells were identified and captured by digital camera and
quantitated using UTHSCSA Image Tool (http://ddsdx.uthscsa.edu). Three
fields for each of quadruplicate measurements for each condition were
analyzed in two separate experiments.
Cell differentiation
Differentiation of hMS cells was studied in a medium containing
10% HI-FBS, 10-8 M dexamethasone,
10-8 M 1,25-dihydroxyvitamin D3,
10 mM ß-glycerol phosphate, and 100 µM
L-ascorbate phosphate. We previously demonstrated that the
hMS cell lines have n equal opportunity to differentiate along either
the OB or adipocyte pathway in this differentiation medium (29). Cells
were refed every 3 days with differentiation medium and fresh IGFs
throughout the treatment interval unless otherwise specified.
Semiquantitative RT-PCR
Total cellular RNA was extracted using the RNA STAT-60 kit
following the manufacturer’s instructions. cDNA were synthesized from
2 µg total RNA in a 20-µl reaction mix containing 4 µl 5 x
incubation buffer for AMV reverse transcriptase; 50 pmol
poly(deoxythymidine)15 primer; 20 nmol each of dATP, dCTP,
dGTP, and dTTP; 20 U ribonuclease inhibitor; and 20 U AMV reverse
transcriptase. The reaction time was 1 h at 42 C.
Aliquots of 1 µl cDNA were amplified in a 25-µl PCR mixture that
contained 2.5 µl 10 x Expand high fidelity PCR buffer
(Roche Molecular Biochemicals, Indianapolis, IN) with
MgCl2 15 mM; 5 pmol 5'- and 3'-oligoprimers;
2.5 nmol each of dATP, dCTP, dGTP, and dTTP; 0.25 µl
[-32P]dCTP (10 µCi/µl); and 0.35 U Expand high
fidelity Taq DNA polymerase. Each cDNA sample was amplified
in duplicate PCRs for each gene. Amplifications were performed in a
GeneAmp 9600 thermal cycler (Perkin-Elmer Corp., Norwalk,
CT), for the following cDNAs: Cbfa1, bone/liver/kidney alkaline
phosphatase (AP), OC, Col I, peroxisome proliferator-activated receptor
2 (PPAR2), plain lipoprotein lipase
(LPL), and adipsin. The housekeeping gene glyceraldehyde phosphate
dehydrogenase (GAPDH) was amplified as a control for RNA loading of RT
and variation in cDNA synthesis efficiency. Primer sequences and
amplification profiles used for all of these genes were reported
previously (25, 27, 29). Amplification of LPL and Col I was performed
for 35 and 30 cycles, respectively, with the following temperature
profile: denaturation at 94 C (30 sec), annealing at 55 C (60 sec), and
extension at 72 C (120 sec). Amplifications of AP and OC were performed
for 30 cycles, and those of PPAR2 and adipsin were
performed for 35 cycles, with denaturation at 94 C (30 sec), annealing
at 55 C (30 sec), and extension at 72 C (30 sec). All PCR ended in a
7-min incubation at 72 C.
PCR products were analyzed as described previously (25). Briefly, 9-µl samples were electrophoresed on a 1.5% (wt/vol) agarose gel containing 0.01% (wt/vol) ethidium bromide. Visualized PCR product bands were sliced from gel, and radioactivity within gel slices was quantitated using an LS600 scintillation counter (Beckman Coulter, Inc., Fullerton, CA). Quantification of PCR product was normalized to GAPDH PCR product. The different gene products were purified using a Wizard PCR Preps DNA kit (Promega Corp.). For sequence analysis, approximately 150 ng of each purified cDNA fragment were added to 3.2 pmol of either 5'- or 3'-primer and analyzed in both directions in an automated DNA sequence analyzer.
Bone-related proteins
The AP enzyme activity was quantitated in cell lysates by
spectrophotometric measurement of p-nitrophenol 1-h release
at 37 C (37). Total cellular protein values were measured from cell
lysate by the Bradford method (Bio-Rad Laboratories, Inc.,
Hercules, CA). Col I (Prolagen-C, Metra Biosystem) and OC (Novocalcin,
Metra Biosystem, Mountain View, CA) proteins were measured in
hMS(3, 4)-conditioned medium samples using an enzyme-linked
immunosorbent assay.
Cytoplasmic lipid droplet formation
Cytoplasmic inclusions of neutral lipids were assessed by Oil
Red O staining, as described previously (24, 29). Cells were observed
using a Nikon Diaphot inverted microscope and a
Nikon 35-mm camera (Nikon, Melville, NY). The
percentage of Oil Red O-positive cells was determined by counting cells
in 30 contiguous fields/well after a random start. Results were
normalized to cell number per well.
Statistical analysis
All values are expressed as the mean ± SEM.
ANOVA was used for dose- and time-dependent differences and when
comparing multiple groups with a single control.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, cross-linking of
[125I]IGF-I to hMS(3, 4) cells and SDS-PAGE of the cell
lysate under reducing conditions identified one major band of
approximately 130 kDa. Labeling of this protein band was displaceable
by unlabeled IGF-I or insulin at high concentrations, consistent with
the -subunit of the type I IGF receptor (4). In addition, we
observed a faint 40- to 50-kDa band displaceable by unlabeled IGF-I but
not by insulin, indicating very low levels of cell-associated IGFBPs
(4). Cross-linking of [125I]IGF-II to hMS(3, 4) cells
also demonstrated low expression of the type II IGF receptor (data not
shown). However, it is generally accepted that both IGF-I and -II exert
their anabolic effects through the type I receptor (1, 38, 39). These
data indicate that hMS(3, 4) cells, with documented IGF receptors, are
potential targets for IGF stimulation.
|
) showed that hMS(3, 4) cells
constitutively express IGFBP-3, -4, -5, and -6. No expression of
IGFBP-1 or IGFBP-2 was evidenced under these conditions (data not
shown). In addition, treatment of hMS(3, 4) cells for 48 h with
IGF-I or an IGF-I analog (LR3-IGF-I) that has similar affinity for the
type I IGF receptor (43) had no apparent effect on IGFBP expression.
This experiment was repeated with similar results.
|
). The identities of these hMS IGFBPs
were confirmed by immunoprecipitation with specific IGFBP antibodies
(data not shown). No IGFBP-5 protein was detected, and a 34-kDa IGFBP
appeared to be contributed by 0.1% HI-FBS in the medium. In contrast
to negligible effects on messenger RNA (mRNA) levels, IGF treatment had
marked effects on IGFBP protein levels in the medium. Treatment of
hMS(3, 4) cells with IGF-I or IGF-II resulted in elevated IGFBP-3 and
loss of detectable IGFBP-4 in the conditioned medium. On the other
hand, high concentrations of insulin, which bind type I IGF receptors
but do not bind IGFBPs, had no effect on IGFBP levels. Treatment of
cells with LR3-IGF-I likewise had no effect on IGFBP-3 levels, but
resulted in a loss of IGFBP-4. As treatments had no effect on IGFBP
mRNA expression that could account for these changes in IGFBP protein
levels (see Fig. 2
), the data suggest specific posttranslational
regulation of IGFBP-3 and -4 by IGFs in these cells, similar to what
has been reported for human OB cells and normal human fibroblasts (31, 33, 42). Indeed, the loss of IGFBP-4 induced by IGF-I and IGF-II was
demonstrated by cell-free assay to be due to IGF-dependent proteolysis
of 24-kDa IGFBP-4 into 18- and 14-kDa fragments (Fig. 4). Although treatment of hMS(3, 4) cells
with LR3-IGF-I resulted in a decrease in medium IGFBP-4, LR3-IGF-I did
not directly induce IGFBP-4 proteolysis in a cell-free assay.
Conditioned medium from hMS(3, 4) cells also exhibited constitutively
active IGFBP-5 proteolysis (data not shown), which probably accounts
for the failure to detect intact IGFBP-5 protein by Western ligand
blot. Thus, hMS(3, 4) cells express specific IGFBPs, which, in turn,
can be modulated by IGFs.
|
|
).
At this concentration, the mitogenic activities of the IGFs and their
analogs were not significantly different. Similarly, dose-response
curves did not reveal statistically significant differences between
IGFs and their analogs (data not shown). Although IGFs have been shown
to have antiapoptotic effects in a number of cell systems (46),
treatment of hMS(3, 4) cells with 10 nM IGF-I or -II had no
suppressive effect on apoptosis in these cells (data not shown). Thus,
hMS(3, 4) cells respond to mitogenic stimulation by IGFs.
|
. Gene
expression of Cbfa1, a transcription factor involved in commitment to
the osteoblast lineage (29, 47, 48, 49), was not affected by treatment with
10 nM des(1, 2, 3)IGF-I or des(1, 2, 3, 4, 5, 6)IGF-II. This was true when
measured over a period of 30 min to 6 days (data not shown). Expression
of osteoblast marker genes, Col I, AP, and OC, was not altered by
treatment with 10 nM des(1, 2, 3)IGF or des(1, 2, 3, 4, 5, 6)IGF-II. Gene
expression of PPAR2, the transcription factor involved
in commitment to the adipocyte lineage (29, 50), also was not affected
by treatment with 10 nM IGF. Furthermore, expression of
adipocyte markers LPL and adipsin up to 6 days was not altered. Similar
results were obtained with 10 nM IGF-I or IGF-II and
increasing doses of LR3-IGF-I (data not shown). These experiments were
repeated, and the results were confirmed under serum-free conditions
(data not shown).
|
). Furthermore, there was a differential
response to IGF-I and LR3-IGF-I in hOB cells, indicating an influence
of IGFBPs.
|
, treatment with 10 nM
des(1, 2, 3)IGF-I or des(1, 2, 3, 4, 5, 6)IGF-II for 3–9 days did not affect the AP
activity of hMS(3, 4) cells. Similar data were obtained with IGF-I and
-II (data not shown). Type I procollagen and OC production normalized
to total protein varied during culture, but was not affected by 10
nM des(1, 2, 3)IGF-I or des(1, 2, 3, 4, 5, 6)IGF-II for up to 15 days of
treatment (Fig. 9). Equivalent results
were obtained when cells were treated with IGF-I, IGF-II, or LR3-IGF-I
(data not shown).
|
|
). Similar results were obtained when
cells were treated with 10 nM IGF-I (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The IGFs are potent mitogenic factors for a variety of cell types (1, 51). We demonstrate that they are mitogenic for human marrow stromal cells as well, with the potential for providing an increased number of progenitors available for new units of bone formation. These findings are consistent with the study by Canalis (6) showing that in organ culture of fetal rat calvaria, the increase in thymidine uptake was observed both in the periosteum, which is rich in osteoprogenitors, and in the central area, which is rich in mature osteoblasts. Interestingly, treatment of hMS(3, 4) cells with IGFs or their analogs induced equivalent increases in the proliferation rate, suggesting that under our in vitro conditions, endogenous IGFBPs did not modulate the effects of IGFs on cell proliferation. This is probably due to experimental conditions, i.e. stimulating the cells in fresh serum-free medium conditions, allowing the IGFs to bind to their receptors before any release of IGFBPs would occur. Furthermore, our cells have very low levels of cell-associated IGFBPs that could potentially modulate IGF bioactivity. IGF-I has also been shown to stimulate [3H]thymidine incorporation in murine TC-1 stromal cells and primary cultures of human stromal cells grown under conditions that promote osteogenic differentiation (21, 22).
The effects induced by the IGFs on differentiation were more complex.
Differentiation of marrow stromal cells into mature cells is under the
control of transcriptional factors that regulate the commitment and
early differentiation of these progenitors to the different lineages.
Cbfa1 is a gene expressed early during differentiation, and its product
serves as a transcriptional activator of the commitment to the
osteoblastic lineage (47, 48, 49). Similarly, PPAR2 is an
early response gene that is involved in commitment to the adipocyte
pathway (50). As assessed by semiquantitative RT-PCR, IGFs induced no
change in the gene expression of Cbfa1 or PPAR2,
suggesting that the IGFs do not participate in the commitment to the
osteoblast or to the adipocyte pathway. Investigating the effects of
the IGFs on early osteoblast differentiation, we observed no changes in
AP, Col I, and OC gene expression. We could rule out a negative effect
being due to IGFBPs because similar results were obtained using
LR3-IGF-I, des(1, 2, 3)-IGF-I, or des(1, 2, 3, 4, 5, 6)-IGF-II, analogs of the IGFs
possessing a similar affinity to the IGF receptor but a very low
affinity to the IGFBPs largely present in FBS (43, 44, 45). At the protein
level, IGFs had no effect on AP activity, but increased Col I and OC
levels in the conditioned medium. However, this increase in Col I and
OC disappeared after normalization to total protein level, suggesting
that the IGFs exert a posttranscriptional effect on overall protein
synthesis (7) without specifically triggering the differentiation of
human marrow stromal cells along the osteoblast pathway.
Recently, Langdhal and colleagues reported inhibitory effects of IGF-I and IGF-II on Col I protein levels in primary cultures of human marrow cells, whereas IGF-I decreased and IGF-II increased AP activity (22). Methodological differences probably explain the differing results. These researchers used primary cultures of human marrow stromal cell aspirates with a heterogeneous cell population within each bone marrow sample and, consequently, variability between samples. Also, the results were not normalized to total protein. Using primary rodent bone marrow stromal cells, Tanaka and Liang observed an increase in Col I gene and protein levels and no difference in AP activity with IGF-I treatment (21). However, IGF-I was added after 10 days in culture in the presence of 10% FBS, suggesting that part of the population may have already differentiated. Furthermore, these researchers reported in the same model either no effect (21) or an increase in osteopontin mRNA levels in response to IGF-I treatment (52), suggesting variability and/or heterogeneity of their cell population and/or confounding effects of IGFBPs in the FBS.
Unlike hMS(3, 4) cells, mature hOB cells significantly increased Col I expression in response to IGF treatment, with IGF analogs being more effective than the native IGFs. A greater abundance of cell-associated IGFBP in hOB cells compared with hMS cells may account for this differential response. In addition, treating hMS(3, 4) cells for 4 days with IGF analogs after 15 days in culture conditions promoting their differentiation provided a similar increase in Col I gene and protein expression. Overall, these data suggest that in mature osteoblasts, the IGFs induce an increase in collagen synthesis through both transcriptional and translational mechanisms. These data are in agreement with the initial observation by Canalis (6) and support the hypothesis that the effects of IGFs on differentiation are dependent on the maturation and origin of the cells (53).
Treatment of hMS(3, 4) cells with IGFs did not modulate gene expression of LPL or adipsin, early differentiation markers of the adipocyte lineage. These results are consistent with previous studies showing a lack of IGF effect on differentiation of a preadipocyte 3T3-F442A cell line without a prestimulation with GH triggering the early stages of differentiation (54). In the adipocyte lineage, it is known that IGFs promote adipocyte differentiation predominantly by acting on adipocytes at a late stage of differentiation (54, 55, 56). Our observation that in long term experiments with culture conditions triggering the early stages of differentiation the IGFs were able to increase lipid droplet accumulation provides further evidence that the stage of cell maturation plays a critical role in the response to IGFs during adipocyte differentiation. We previously reported that addition of insulin with dexamethasone to hMS cell cultures did not increase the number of Oil Red O-positive cells above that with dexamethasone alone (24). These observations support the hypothesis that IGFs could increase lipid accumulation by acting through their specific receptors.
IGFBPs are important determinants of IGF action at the cellular level (40). hMS(3, 4) cells produce a pattern of IGFBPs similar to what our group previously described in mature hOB cells and SV40-transformed hOB cells (23, 31, 57). Thus, hMS(3, 4) cells in culture constitutively expressed and secreted IGFBP-3, -4, -5, and -6. Treatment with IGF-I or LR3-IGF-I had no effect on steady state mRNA levels for these IGFBPs. However, treatment of hMS(3, 4) cells with IGF-I or IGF-II increased IGFBP-3 medium levels and induced the loss of detectable IGFBP-4. LR3-IGF-I or insulin treatment could not duplicate the increase in IGFBP-3, suggesting receptor-independent regulation as previously reported in human fibroblasts (32, 42) and other cells (58, 59). Whereas IGF-I and -II did not alter IGFBP-4 mRNA steady state levels, they induced a dramatic decrease in IGFBP-4 protein levels related to activation of a specific IGFBP-4 protease, similar to what we and others previously described in hOB cells (31, 57, 60). Interestingly, treatment of hMS cells with LR3-IGF-I, an IGF analog that does not bind IGFBP-4 and does not induce proteolysis in a cell-free assay, was as effective as IGF-I and -II in producing proteolysis of secreted IGFBP-4. These data are in contrast to those from other cell systems, and they suggest that LR3-IGF-I may be acting indirectly to induce the primary activator of IGFBP-4 proteolysis. We are in the process of determining whether LR3-IGF-I treatment up-regulates IGF expression in these cells. Our results for IGFBP expression in hMS(3, 4) cells differ from those reported for the murine marrow stromal cell line, TC-1 (61, 62, 63). In this mouse cell model, IGF-I induced an increase in IGFBP-4 and IGFBP-5 at a posttranscriptional level and an increase in IGFBP-3 at both transcriptional and posttranscriptional levels, underlining the known species differences in the IGF system. The significance of IGFBPs in the marrow stroma is unknown, but they have been proven to be critical determinants of IGF action in other systems as well as possessing intrinsic growth regulatory potential.
In summary, we demonstrated that human marrow stromal cells express important components of the IGF system and are potential targets for IGF regulation. Whereas the IGFs exert mitogenic effects on human marrow stromal cells, their effects on differentiation are dependent on the stage of maturation of the cells. The IGFs do not directly modulate the genes determining either commitment to the osteoblast or the adipocyte lineages or early differentiation, but they enhance the differentiated function of mature osteoblasts and adipocytes. These data further support an important role for IGFs in bone formation and suggest that IGF therapy might be especially useful in bone diseases related to an insufficiency of osteoprogenitor cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 11, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Giustina, G. Mazziotti, and E. Canalis Growth Hormone, Insulin-Like Growth Factors, and the Skeleton Endocr. Rev., August 1, 2008; 29(5): 535 - 559. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kucic, I. B. Copland, J. Cuerquis, D. L. Coutu, L. E. Chalifour, R. F. Gagnon, and J. Galipeau Mesenchymal stromal cells genetically engineered to overexpress IGF-I enhance cell-based gene therapy of renal failure-induced anemia Am J Physiol Renal Physiol, August 1, 2008; 295(2): F488 - F496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. S Miller, J. T Bronk, T. Nishiyama, H. Yamagiwa, A. Srivastava, M. E Bolander, and C. A Conover Pregnancy associated plasma protein-A is necessary for expeditious fracture healing in mice J. Endocrinol., March 1, 2007; 192(3): 505 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Celil and P. G. Campbell BMP-2 and Insulin-like Growth Factor-I Mediate Osterix (Osx) Expression in Human Mesenchymal Stem Cells via the MAPK and Protein Kinase D Signaling Pathways J. Biol. Chem., September 9, 2005; 280(36): 31353 - 31359. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Jadlowiec, D. Dongell, J. Smith, C. Conover, and P. Campbell Pregnancy-Associated Plasma Protein-A Is Involved in Matrix Mineralization of Human Adult Mesenchymal Stem Cells and Angiogenesis in the Chick Chorioallontoic Membrane Endocrinology, September 1, 2005; 146(9): 3765 - 3772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Canalis The Fate of Circulating Osteoblasts N. Engl. J. Med., May 12, 2005; 352(19): 2014 - 2016. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Abou Samra, N. H. Baba, N. Torbay, L. Dib, and G. E.-H. Fuleihan High Plasma Leptin Is Not Associated with Higher Bone Mineral Density in Insulin-Resistant Premenopausal Obese Women J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2588 - 2594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Matsunuma, T. Kawane, T. Maeda, S. Hamada, and N. Horiuchi Leptin Corrects Increased Gene Expression of Renal 25-Hydroxyvitamin D3-1{alpha}-Hydroxylase and -24-Hydroxylase in Leptin-Deficient, ob/ob Mice Endocrinology, March 1, 2004; 145(3): 1367 - 1375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-L. Cheng, J.-S. Shao, N. Charlton-Kachigian, A. P. Loewy, and D. A. Towler Msx2 Promotes Osteogenesis and Suppresses Adipogenic Differentiation of Multipotent Mesenchymal Progenitors J. Biol. Chem., November 14, 2003; 278(46): 45969 - 45977. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Devlin, Z. Du, V. Buccilli, V. Jorgetti, and E. Canalis Transgenic Mice Overexpressing Insulin-Like Growth Factor Binding Protein-5 Display Transiently Decreased Osteoblastic Function and Osteopenia Endocrinology, October 1, 2002; 143(10): 3955 - 3962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Khalil, E. A. Lucas, S. Juma, B. J. Smith, M. E. Payton, and B. H. Arjmandi Soy Protein Supplementation Increases Serum Insulin-Like Growth Factor-I in Young and Old Men but Does Not Affect Markers of Bone Metabolism J. Nutr., September 1, 2002; 132(9): 2605 - 2608. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Welniak, R. Sun, and W. J. Murphy The role of growth hormone in T-cell development and reconstitution J. Leukoc. Biol., March 1, 2002; 71(3): 381 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ivanovski, H.R. Haase, and P.M. Bartold Expression of Bone Matrix Protein mRNAs by Primary and Cloned Cultures of the Regenerative Phenotype of Human Periodontal Fibroblasts Journal of Dental Research, July 1, 2001; 80(7): 1665 - 1671. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.05
(significant difference between IGF-I and LR3-IGF-I).
