| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Pathology, University of Washington, School of Medicine, Seattle, Washington 98195, USA
b Division of Molecular Genetics, Institute of Life Science, Kurume University, Kurume, Fukuoka, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: tooth cementum • CGF • epidermal growth factor • cell cycle phase • gingival fibroblast • mesenchymal cells
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have reported that the mineralized matrix of tooth cementum contains a 14 kDa polypeptide growth factor referred to as cementum-derived growth factor (CGF) (13–15). This molecule is similar to insulin-like growth factor I (IGF-I), a growth factor that promotes the proliferation as well as differentiation of many cell types (16). The CGF is mitogenic to fibroblasts and osteoblastic cells present in connective tissues adjacent to the cementum (15). In the cementum matrix, the CGF is sequestered along with other growth factors such as fibroblast growth factors 1 and 2 and epidermal growth factor (EGF), and adhesion molecules such as collagens, bone sialoprotein, and osteopontin (17, 18). These molecules are likely to influence the outcome of CGF action because regulation of cellular activities often involves the combined action of more than one growth factor. For example, most cell types require more than one mitogenic stimulus to escape from G1 arrest, enter S phase, and complete cell division (3, 5, 6). The outcome can also be affected by the extracellular matrix, which determines whether cells divide or differentiate (19, 20).
The CGF is a poor mitogen for fibroblasts; however, its mitogenic activity, even at suboptimal concentrations, is synergistically potentiated by EGF and serum (14, 15). By using gingival fibroblasts as target cells, we previously examined the biochemical events induced by the CGF and EGF in order to identify the reactions responsible for optimal mitogenic response. We observed that the CGF induces many classical, immediate, and early mitogenic signaling events and that the type and pattern of events induced are characteristic of the CGF (14). We have extended these studies to examine later G1 phase events involved in cell cycle progression, specifically, cyclin E levels and the activity of Cdk associated with this cyclin. We chose to examine these molecules because cyclin E plays a major regulatory role during the G1 phase of the cell cycle and in adhesion-dependent cell cycle progression, and overexpression of cyclin E diminishes serum dependence for growth (6–12, 21–23). The cyclin E levels and activity of its Cdk are up-regulated during mitogenic stimulation and differentiation (24, 25); because these are mutually exclusive processes, cyclin E and its Cdk may represent a convergence point for proliferation, differentiation, and other biological responses induced by growth factors such as the CGF. Our objectives were to determine whether changes in the levels and activity of these two signaling molecules correlate with the mitogenic response induced by CGF and whether they contribute to the synergism in mitogenic activity between CGF and EGF.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-32P]ATP (specific activity 6000 Ci/mmol), and an enhanced chemiluminescence (ECL) detection kit were purchased from Amersham Corp., Arlington Heights, Ill. Human recombinant EGF was obtained from Sigma Chemical Company (St. Louis, Mo.) and H1 from Boehringer-Mannheim (Indianapolis, Ind.). Mouse monoclonal anti-cyclin E antibodies HE12 and HE111 and highly specific affinity-purified polyclonal antibodies to p21 (C-19) and p27 (N-20) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, Calif.). G-418 and protein assay kit were the products of Gibco-BRL (Gaithersburg, Md.) and Bio-Rad (Hercules, Calif.), respectively. High-pressure liquid chromatography (HPLC) columns TSK CM-5W and C18 reverse phase columns were purchased from Toso Haas (Montgomeryville, Pa.) and Vydac (Hesperia, Calif.), respectively. Control LXSN (LTR Xho1 SV40 Neomycin) and LXSN-cyclin E cDNA-retroviral expression vectors containing neomycin phosphotransferase gene for G418 resistance were constructed by one of us (T.H.N.; see ref 21).
Purification of CGF
CGF was purified from bovine cementum as described previously (14, 15). Briefly, cementum tissue was harvested from bovine teeth, extracted in 1.0 M acetic acid containing proteinase inhibitors, and fractionated by heparin affinity chromatography. Proteins eluted by 0.6 M NaCl were separated by HPLC using TSK CM-5PW cation exchange and narrow bore C18 reverse phase columns. CGF obtained in this manner contained two major protein bands migrating with Mr 14,000–16,000 and 18,000–20,000, representing unglycosylated and N-glycosylated forms, respectively (15).
Assay for DNA synthesis
Human gingival fibroblasts were plated in 48-well tissue culture plates at a subconfluent cell density of 2 x 104 cells/well, incubated in serum-free Dulbecco's modified Eagle's medium (DMEM) for 48 h, and then added with 10 ng/ml CGF (this represents 
20% of optimal concentration when used alone) (14, 15), 5 ng/ml EGF, or both. Cells exposed to serum-free DMEM and 10% fetal bovine serum (FBS) served as negative and positive controls, respectively. DNA synthesis was measured as [3H]thymidine uptake, which was added at 5 µCi/ml concentration for the time periods indicated (15).
Stable transfection of cells
Fibroblasts were transfected with LXSN (control) and LXSN-human cyclin E cDNA-retroviral expression vectors as described elsewhere (21). Selection of infected cells was carried out in DMEM supplemented with 10% FBS and 800 µg/ml of G418 for 2 wk (21).
Immunoprecipitation of cyclin E
Approximately 2 x 105 of cells were lysed by sonication in lysis buffer containing 1% Triton X-100, 10 mM Tris/HCl (pH 7.4), 50 mM each of NaCl and NaF, 5 mM EDTA, 0.1 mM NaVa3O4, 0.5 mM phenylmethanesulfonyl fluoride, and 10 mM ß-glycerophosphate (21). After centrifugation, cell lysate containing 100 µg protein was added with anti-human cyclin E antibody (2 µl of antiserum or 5 µg of HE111) (21). The immunoprecipitates were collected on protein A-Sepharose or protein G-agarose beads.
Histone H1 kinase assay
Immunoprecipitates obtained using anti-human cyclin E antibodies were washed twice with lysis buffer and then with kinase buffer containing 50 mM HEPES (pH 7.5), 10 mM MgCl2, 1 mM dithiothreitol, 2.5 mM EDTA, 10 mM ß-glycerophosphate, 50 mM NaF, and 0.1 mM NaVa3O4 (21). They were incubated with kinase buffer containing 1 µg of histone H1, 20 µM of ATP substrate, and 10 µCi of [-32P]ATP for 30 min at 30°C (21). Reaction products were separated by 12.5% NadodSO4-polyacrylamide gel electrophoresis (SDS-PAGE), and the gels were stained, dried, and exposed to Kodak X-omat AR film.
Western blot analysis
Proteins (50 µg) were separated by SDS-PAGE under reducing conditions. They were transferred to a nitrocellulose membrane and blocked with 5% nonfat dry milk. The membranes were first incubated with rabbit polyclonal antibodies against human cyclin E, p21 or p27, then with horseradish peroxidase-conjugated secondary antibodies, and immunoreactive proteins were detected by ECL.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fig. 1).
When cells were exposed first to EGF and CGF was added 3 h later, DNA synthesis was 75% as much as that in cells incubated with both growth factors for the entire period ( Fig. 1). It decreased further as CGF was added later, and was 55% when CGF was added after 12 h ( Fig. 1). However, when CGF was added first and EGF added 3 h later, DNA synthesis was <30% of the maximum ( Fig. 1). When EGF was added after 12 h, there was very little DNA synthesis. DNA synthesis was lowest with either CGF or EGF alone. In the next experiment, both growth factors were added at 0 time; at the times indicated, the medium was withdrawn and replaced with fresh medium containing either CGF or EGF alone. Removal of CGF even after 1.5 h resulted in reduction of the DNA synthesis to levels similar to negative controls containing no growth factors ( Fig. 2).
In contrast, cultures from which EGF was removed after 1.5 h retained 70% of maximum DNA synthesis and EGF was not necessary after 12 h. As the cells were exposed to CGF + EGF for longer periods, the level of DNA synthesis approached positive controls.
|
|
These results indicate that signaling reactions induced concurrently by both EGF and CGF are necessary for optimal mitogenic stimulation. In preliminary experiments, we found that D1 cyclin levels in cells exposed to CGF, EGF, both growth factors, or 10% FBS increased by three- to fivefold, but there were no significant differences among the four treatments in the kinetics or magnitude of the increase in D1 (data not shown). Nor was there a correlation between the levels of cyclin D1 and DNA synthesis. Therefore, we compared cyclin E levels in these cells. Fibroblasts contain three isoforms of cyclin E migrating in SDS-PAGE-gels with 52, 50, and 40 kDa (21), and these three molecules were recognized by the anti-cyclin E antibody in Western analysis. The level of the 50 kDa species increased 2.3-, 3.1-, and 1.7-fold in cells treated with CGF, CGF + EGF, and 10% FBS, respectively, after 12 h ( Fig. 3).
Levels were lower and returned to 0 time value 4 h later in cells challenged with CGF alone, but were higher and remained elevated for 20 h in CGF + EGF and 10% FBS treatments ( Fig. 3). In cells exposed to EGF alone, cyclin E levels did not increase and, after 8 h, decreased.
|
We examined the activity of cyclin E-associated H1 kinase after 8 h, when the cyclin E levels began to increase. The results showed that the H1 kinase activity increased 12 h after addition of CGF, EGF, or FBS; at this time the activity was two- to threefold as much as the levels at 8 h ( Fig. 4).
Activity decreased 4 h later in EGF-only cells. However, in cultures exposed to CGF + EGF and 10% FBS, increased levels of kinase activity were sustained for up to 20 h ( Fig. 4). We also examined the levels of p21 and p27 in these cultures; p21 was not detectable except at the 4 h point and p27 levels showed no significant differences among the various treatments (data not shown).
|
These data indicated that sustained high levels of cyclin E and Cdk associated with cyclin E may contribute to the high mitogenic response obtained in the presence of CGF + EGF and FBS. To examine this possibility, human cyclin E was overexpressed in the gingival fibroblasts by transfection with cyclin E-cDNA. The transfected cells expressed fivefold as much cyclin E as the cells transfected with vector alone (data not shown, see Fig. 5).
Nevertheless, addition of growth factors to cyclin E overexpressing cells increased cyclin E levels by an additional two- to threefold ( Fig. 5, top panel; shown for 16 h time point). The H1 kinase activity of the LXSN-cyclin E cells also increased in all treatments (from 1.6- to 2.3-fold as much as control4; Fig. 5, bottom panel). To determine whether the increase in cyclin E-associated H1 kinase activity correlates with mitogenic response, we measured DNA synthesis by these cells. When no growth factors were added, the basal level of DNA synthesis by the cyclin E-transfected cells was higher than control cells transfected with the vector alone. The synthesis increased in response to CGF and EGF; however, maximum levels of DNA synthesis required the presence of both growth factors ( Fig. 6).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Figs. 1, 2). The EGF did not increase cyclin E levels, whereas cyclin E levels and H1 kinase activity associated with this cyclin increased in the presence of CGF. The increase occurred in the presence of CGF alone, but the increase was greater and higher levels were maintained for up to 24 h in cells exposed to CGF + EGF or 10% FBS. The increased cyclin E levels were associated with a parallel change in H1 kinase activity, which increased after 12 h; the overall activity was higher and remained elevated for up to 20 h in CGF + EGF and 10% FBS treatments. These data indicate that sustained higher levels of cyclin E and Cdk associated with it, presumably the Cdk2, are necessary for optimal mitogenic stimulation by CGF and EGF. This conclusion is supported by the fact that cyclin E overexpressing cells manifest higher levels of DNA synthesis in the absence of added growth factors and in response to CGF or EGF.
The increased levels of cyclin E and E-Cdk activity alone appear insufficient for optimal mitogenic response, because cyclin E overexpressing cells with constitutively higher Cdk activity do not enter into S phase in the presence of CGF or EGF alone. The H1 kinase activity of cells exposed to EGF alone for 12 h is also comparable to positive controls; however, DNA synthesis in these cells is low. The synthesis is also low in cells exposed to EGF first, later to CGF, and to (CGF+EGF) for 1.5 h, where the kinase activity is likely to reach levels comparable to positive controls ( Figs. 1, 2). These results indicate that factors other than cyclin E and its Cdk are necessary for optimal DNA synthesis. The D1 cyclin does not appear to be the needed additional factor because the level of this cyclin increases similarly in all treatments and does not correlate with the degree of DNA synthesis (data not shown). It is likely that cyclin A and Cdk associated with cyclin A are two additional factors necessary for optimal DNA synthesis (26, 27, 28); however, we did not perform experiments to examine this possibility.
Recently, Winston et al. (26) reported that in Balb/c mouse fibroblasts, PDGF and serum factors alone cause only a marginal increase in cyclin E, but together evoke a synergistic increase, which is aided by down-regulation of p27 and by sequestration of the p27 by cyclin D1-Cdk4. Modulation of CKI levels is also a feature of T cells exposed to interleukin 1 (29) and in adhesion-dependent increase in Cdk activity (22). Inhibition of E-Cdk by CKI may explain why H1 kinase activity decreases faster in human gingival fibroblasts exposed to EGF alone ( Fig. 4); however, this does not appear to be the case, because p21 and p27 are barely detectable in these cells. p21 and p27 levels do not differ among the four treatments and do not correlate with mitogenic activity (data not shown). However, our experiments do not rule out the role of other CKIs such as p15INK4b, p16INK4a, p18INK4c, and p57kip2, since we did not examine their presence.
The target cells for the CGF are likely to be the cementoblasts as well as other cells in close proximity to the cementum. These include gingival fibroblasts, periodontal ligament cells, and possibly alveolar bone cells. The CGF is mitogenic to all these cell types (15). The alveolar bone cells mitogenically respond to CGF in a manner similar to the cyclin E overexpressing cells; these cells respond to CGF better than fibroblasts, they do not manifest synergism, and the effect of CGF and EGF is additive (15). A greater proportion of these cells also complete cell division in the presence of CGF alone. It is conceivable that such differential responses may determine which cells respond to CGF and other growth factors and whether the cells grow or differentiate. For example, CGF alone, which is an IGF-I-like molecule and weakly mitogenic, may promote cell proliferation in the presence of other factors available during inflammation, such as serum or the EGF, and the proliferation response may predominate during early stages of wound healing. However, during later stages of healing, or under healthy conditions when additional factors are not available and extracellular matrix is in place, the CGF may promote the differentiation of these cells (19, 20, 30, 31). These activities may involve cyclin E and its Cdk, whose levels increase during cell division as well as differentiation (24, 25).
Although a greater number of fibroblasts respond to both CGF plus EGF, a proportion of cells complete the cell cycle in either CGF and EGF alone, in normal and cyclin E overexpressing cells.3 This indicates that fibroblast cultures contain populations of cells that complete S phase with EGF alone, CGF alone, and with both growth factors. Although we have not attempted to isolate such cell populations, these cells, if present, are likely to be associated with different roles in healthy tissues and during wound healing (18).
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: EGF, epidermal growth factor; Cdks, cyclin-dependent kinases; CKI, cyclin-dependent kinase inhibitor (or inhibitors); CGF, cementum-derived growth factor; ECL, enhanced chemiluminescence; DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; PDGF, platelet-derived growth factor; IGF, insulin-like growth factor; HPLC, high-pressure liquid chromatography.
3 The mitogenic response measured as [3H]thymidine uptake in the presence of CGF, EGF, and CGF+EGF was 5.1 ± 0.4, 23 ± 1.5, and 123.6 ± 33.4% as much as that in the presence of 10% FBS (mean±SD of three different experiments each consisting of triplicates). In a parallel experiment, cell cycle analysis by FACS showed that the percent of (G2+S) phase cells in medium containing no serum, CGF, EGF, CGF + EGF, and 10% FBS were 10, 10, 16, 31, and 22, respectively. The (G2+S) phase cells in cyclin E overexpressing cells for these treatments were 18, 21, 25, 29, and 25%, respectively.
4 The magnitude of increase in cyclin E-associated H1 kinase activity in the cyclin E overexpressing cells was not as high as cyclin E (1.6-to 2.3-fold vs. 5-fold). We believe this is because of compensatory overexpression of p21 in these cells (data not shown).
Received for publication January 2, 1998. Accepted for publication March 25, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D.D. Bosshardt Are Cementoblasts a Subpopulation of Osteoblasts or a Unique Phenotype? J. Dent. Res., May 1, 2005; 84(5): 390 - 406. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. DONNELLAN and R. CHETTY Cyclin E in human cancers FASEB J, May 1, 1999; 13(8): 773 - 780. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or EGF (5 ng/ml, •) was added later at the times indicated. Labeling was begun 22 h after the addition of the first growth factor and DNA synthesis was measured after 6 h. Data are mean ± SD of triplicates from one representative experiment. The percent of [3H]thymidine uptake relative to when both CGF and EGF were present for the entire period is presented. DNA synthesis in the presence of 10% FBS (
), EGF alone (
), and CGF (
) alone for the entire incubation period is also indicated. Similar results were obtained when labeling was done for 0–48 h after addition of the first growth factor.
, EGF alone; 
, CGF alone. Results were similar when labeling was done for 0–48 h.
