Modulation of mitogenic activity of fibroblast growth factors by inorganic polyphosphate*

The proliferation of normal human fibroblast cells was enhanced by the addition of inorganic polyphosphate (poly(P)) into culture media. The mitogenic activities of acidic fibroblast growth factor (FGF-1) and basic fibroblast growth factor (FGF-2) were also enhanced by poly(P). A physical interaction between poly(P) and FGF-2 was observed, and FGF-2 was both physically and functionally stabilized by poly(P). Furthermore, poly(P) facilitated the FGF-2 binding to its cell surface receptors. Because poly(P) is widely distributed in mammalian tissues, it may be a spontaneous modulator of FGFs.

Na-PO 4 buffer (orthophosphate) was used. The pH of the Na-PO 4 buffer was adjusted to 7.0 by mixing the same concentrations of Na 2 HPO 4 and NaH 2 PO 4 solution. MTS cell proliferation assay kit was from Promega. Human recombinant FGF-2 was from Toyobo (Japan). Anti FGF-2 antibody was from Santa Cruz Biotechnology, Inc. (USA).
Assay for Cell Proliferation-Cells were seeded to 96 multiwell plates at 5x10 3 cells/well (100 µl/well) and cultured in Eagle's minimal essential medium (E-MEM) (for NHDF) or Dulbecco's modified Eagle's minimal essential medium (D-MEM) (for Balb/c 3T3) containing 10 % FBS for 24 hours. After cells had adhered, the medium was replaced with E-MEM or D-MEM without FBS, and cells were further incubated for 48 hours. The medium was replaced again with an appropriate media described in the text and figure legends. After incubation at 37 °C, cell number was directly counted using hematocytometer after trypcinization, or was evaluated by MTS assay (9). For MTS assay, the medium was replaced with 100 µl of E-MEM (without Phenol Red), and 25 µl of mixture of 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfophenyl)-2H-tetr azolium (MTS) (Promega) and phenazine methosulfate (PMS) solution (2 µg/ml MTS, 0.92 µg/ml PMS) was added to each well. After incubation for 75 min at 37 °C, the absorbance at 490 nm of each well was measured. The cell number was quantified by means of the bio-reduction activity of viable cells Preparation of Short-chain [ 32 P]-poly(P)-Long chain [ 32 P]-poly(P) was synthesized by using by guest on July 9, 2020 http://www.jbc.org/ Downloaded from purified E. coli polyphosphate kinase and purified as described (10). To prepare short-chain [ 32 P]-poly(P) (average chain length of around 65 phosphate residues), the long-chain [ 32 P]-poly(P) was hydrolyzed by 20 mM HCl for 2 min. at 95 °C, and the hydrolyzed poly(P) was separated by 15 % polyacrylamide gel electrophoresis together with poly(P) type 65 as a maker. The conditions of gel electrophoresis were the same as those described in the legend of Fig. 3. The gel was stained by 0.05 % toluidine blue containing 5 % glycerol. The portion of the gel corresponding to the position of poly(P) type 65 was cut, and short-chain [ 32 P]-poly(P) was isolated from the gel. Isolated gel was soaked in H 2 O and shaken at room temperature for 2 hours. After brief centrifugation, the supernatant was collected, and short-chain [ 32 P]-poly(P) was recovered from the gel.
Assay for binding between FGF-2 and high-affinity surface receptors-The assay procedure was basically followed as described (11). Confluent Balb/c 3T3 cells were plated onto 60-mm dishes at 35,000 cells/cm 2 in D-MEM with 10 % FBS at 37 °C. After 24 hours, the medium was changed to D-MEM with 2 % FBS with 50 mM sodium chlorate to remove heparin sulfate proteoglycan (HSPG). Cells were further incubated for 72 hours at 37 °C and were then washed once with a binding buffer (D-MEM, 25 mM HEPES) at 4 °C. A fresh binding buffer was added (4 ml/well), and cells were incubated at 4 °C for 10 min. FGF-2 was added at the indicated concentrations, and cells were incubated at 4 °C for 2.5 hours. At the end of the binding period, cells were placed on ice and washed three times with ice-cold binding buffer. Cell by guest on July 9, 2020 http://www.jbc.org/ Downloaded from surface receptor-bound FGF-2 was extracted with two washes (one 5-min wash and one rapid wash, 0.5 ml each/dish) at room temperature using 2 M NaCl in 20 mM sodium acetate (pH 4.0). Washed solutions (0.5 ml x 2) containing the extracted cell surface receptor-bound FGF-2 were collected, and 5 ml of 8 % bovine serum albumin was added to each sample as a carrier protein.
Trichloroacetic acid was added to the final concentration of 5 %, and precipitated proteins were pelleted by centrifugation. After removal of the supernatant, the protein pellet was neutralized by 1 N KOH and dissolved by SDS-PAGE sample buffer. FGF-2 was separated by 15 % SDS-PAGE and visualized by western blotting. Enhancement of Mitogenic Activity of FGFs by Poly(P) -One possible mechanism of growth stimulation by poly(P) is that poly(P) enhances the activity of growth factors that were released into the culture medium from the cultured cells themselves. Therefore, in this study, we focused on the possibility of modulation of FGF activity by poly(P). Since FGFs have been implicated as autocrine growth factors (12), it is possible that poly(P) somehow modulates the biological activity of FGFs.

Stimulation of Cell Growth by Adding Poly(P) into
If poly(P) does in fact modulate the mitogenic activity of FGF, then cell proliferation would be enhanced by co-treatment with poly(P) and FGF rather than by single treatment with poly(P) or FGF. To examine whether the co-treatment of poly(P) and FGF is effective in stimulating mitogenic activity of FGFs, Balb/c 3T3, NHDF and HGF, whose growth could be dependent on FGF, were treated with FGF-1 or FGF-2 in combination with poly(P). As shown in Fig. 2, cell growth was slightly stimulated by both FGF-1 and FGF-2, and, moreover, the levels of growth stimulation by both FGFs became greater in the presence of poly(P). Cell growth in the medium containing FGF-1 and poly(P) was 1.6 to 1.9-times higher than that in the medium containing only FGF-1. Similarly, cell growth in the medium containing FGF-2 and poly(P) was 1.3 to 1.7-times higher than that in the medium containing only FGF-2. Since the levels of growth stimulation by co-treatment of poly(P) and FGFs is greater than sum of stimulation levels of single treatment with poly(P) and FGFs in Balb/c 3T3 and NHDF, poly(P) seems to augment the mitogenic activities of FGFs.
However, in HGF, the level of growth stimulation by co-treatment with FGF-2 and poly(P) is not greater than sum of stimulation levels of single treatment with poly(P) and FGFs. From these results, it is difficult to rule out the possibility that poly(P) and FGFs independently functions on cell growth stimulation. To show the direct evidence of FGF modulation by poly(P), the physical and functional interactions between poly(P) and FGFs has been examined.

Physical Interaction Between FGF-2 and Poly(P) -Physical interaction between
FGF-2 and poly(P) was examined by a gel shift assay as described in the legend of Fig.   3. When FGF-2 was incubated with short chain [ 32 P]-poly(P), FGF-2 bound to the poly(P) dependent on its concentration, and poly(P) formed a complex with FGF-2 and the complex almost remained at the origin of the gel (Fig. 3). This result shows poly(P) does bind to FGF-2. Furthermore, it may possible to roughly estimate the stoichiometry between poly(P) and FGF-2. However, it may also be possible that poly(P) somehow facilitates oligomer formation of FGF-2 as in case of heparin-like glycosaminoglycans (HLGAG) (13), and a FGF-2 oligomer requires only a few phosphate molecules of poly(P) for its binding. The modulation of mitogenic activity of FGF-2 may result from the oligomer formation of FGF-2 initiated by poly(P).

Stabilization of FGF-2 by Poly(P) -
To further examine the effect of poly(P) on FGF-2, the stability of FGF-2 with or without poly(P) was evaluated. Fig. 4A shows the degradation of FGF-2 in culture media in the presence or absence of poly(P). Intact FGF-2 clearly remained after 24 hours of incubation with poly(P) (Fig. 4A, lane 17), whereas intact FGF-2 was not observed after 24 hours of incubation without poly(P) (Fig. 4A, lane 9) or with Na-PO 4 buffer (Fig. 4A, lane 25). The half-life of FGF-2, which was calculated from the intensity of the bands of Fig. 4A was 13.7 hour when FGF-2 was incubated with poly(P), whereas it was only 4.7 hour when FGF-2 was incubated with Na-PO 4 buffer (Fig. 4B).
Since FGF-2 was physically stabilized by poly(P), stability of the biological activity of FGF-2 was also examined using BALB/c 3T3 cells. Since FGF-2 by guest on July 9, 2020 http://www.jbc.org/ Downloaded from almost loses its mitogenic activity within 24 hours of incubation at 37 °C, the residual activity after preincubation (24 hours at 37 °C) of FGF-2 was examined in the presence and absence of poly(P). As shown in Fig. 4C, cells cultured in a medium containing FGF-2 that have been preincubated with poly(P) (medium #2) maintained the same population levels as that of cells cultured in a medium containing FGF-2 without preincubation (medium #3). On the other hand, a medium containing FGF-2 that had been preincubated without poly(P) (medium #4) showed slight proliferation activity but the activity decreased to the same level as that of media without FGF-2 (media #5 and #6) after 83 hours of incubation. This means that FGF-2 that had been preincubated at 37 °C was stabilized by poly(P) and maintained its proliferation activity at the same level as that of FGF-2 that had not been preincubated. In addition, the highest proliferation activity was observed in the medium containing FGF-2 and poly(P) without preincubation. This is consistent with the result shown in Fig.   2, indicating enhancement of mitogenic activity of FGF-2 by poly(P). These results indicate that poly(P) stabilizes the biological activity of FGF-2.

Facilitation of FGF-2 Binding to Its Receptors by Poly(P) -In order to further
examine the effect of poly(P) as an FGF-2 modulator, we analyzed whether poly(P) not only stabilizes FGF-2 but facilitates FGF-2 binding to FGF receptors.
The biological activity of FGF-2 is mediated by interaction with high-affinity cell surface receptors (14)(15)(16). In addition to binding to receptors, FGF-2 binds by guest on July 9, 2020 http://www.jbc.org/ Downloaded from to HSPG on the cell surface. Since many studies have indicated that the binding to HSPG facilitates FGF-2 receptor binding and activation (17)(18)(19)(20), we removed the cell surface HSPG by sodium chlorate treatment in order to observe the direct effect of poly(P) on FGF-2 and its receptor binding (11).
Sodium chlorate-treated cells were incubated with FGF-2, and the receptor-bound FGF-2 was collected and analyzed by western blotting (Fig. 5A), and intensities of the bands were quantified and plotted in Fig. 5B. The amount of FGF-2 that bound to FGF receptors in the presence of poly(P) was more than twice that in the absence of poly(P). This clearly indicates that poly(P) facilitates the binding between FGF-2 and FGF receptors. Poly(P), FGF-2, and FGF receptors may form a trimolecular complex on the cell surface.
There may also be direct interactions between poly(P) and FGF receptor that facilitate FGF-2 binding and receptor dimerization as in the case of heparin (17)(18)(19)(20).
In order to rule out the possibility that stabilization of FGF-2 by poly(P) occur during incubation with binding buffer or elution buffer, stability of FGF-2 during binding assay was evaluated. FGF-2 was incubated with poly(P) or Na-PO 4 buffer in the binding buffer or in the elution buffer at 4 °C. The amount of FGF-2 remaining in the buffer was shown in Fig. 5C. Since almost the same amount of FGF-2 was detected in buffers with poly(P) or Na-PO 4 up to 5.5 hours exposures of the binding buffer (Fig. 5C, lanes 1-4) and up to 30 min by guest on July 9, 2020 http://www.jbc.org/ Downloaded from exposures of the elution buffer (Fig. 5C, lanes 5 and 6), there is no stabilization effect of poly(P) on FGF-2 during this binding assay.

DISCUSSION
With regard to the mechanism of growth stimulation by poly(P), a similar effect has also been reported by adding heparin to culture media. Heparin and HLGAG are well-known potent modulators of acidic fibroblast growth factor (FGF-1) and basic fibroblast growth factor (FGF-2), and they potentiate the mitogenic activity of both FGFs (17)(18)(19)(20). Heparin sulfate stabilizes FGFs and binds to a site on the receptor and at least one site on the growth factor.
Several models propose an important role for heparin sulfate not only in facilitating FGF-2 binding to its receptor tyrosine kinase but also in promoting signaling via formation of receptor dimers. Such dimers are capable of transphosphorylation of the cytoplasmic domain of the receptor, leading to the generation of phosphotyrosine, that is an important initiator of the intracellular signaling pathway (17)(18)(19)(20). Poly(P) may also facilitate FGF-2 binding to its receptors and promote signaling through the same binding sites of heparin sulfate, FGF-2, and its receptors. However, poly (P) and heparin have completely different chemical structures, besides both molecules are negatively charged cellular polymers. It is likely that the mechanism for modulation of FGF activity by poly(P) is different from that of heparin. Based on our results, the level of growth stimulation by poly(P) is higher than that by heparin whose by guest on July 9, 2020 http://www.jbc.org/ Downloaded from concentration is enough for maximum growth stimulation (21) (Fig. 2). This also suggests that the binding sites between poly(P) and FGFs could be different.
Further analyses are needed to elucidate the detailed mechanism of interaction between poly(P) and FGF-2.
One hypothesis is that poly(P) has biological functions for controlling the activity of FGF in vivo. Since poly(P) is widely distributed in mammalian tissues (5), it is possible that degradation of FGF in tissues that have been injured is prevented if there were a mechanism for local regulation of poly(P) concentration. Furthermore, it is also possible that poly(P) interacts with other