JOURNAL OF CELLULAR PHYSIOLOGY 143:129-132 (1990)

Acidic Fibroblast Growth Factor (HBGF-1) Stimulates DNA Synthesis in Primary Rat Hepatocyte CuItu res KEITH A. HOUCK, REZA ZARNEGAR, STEPHANIE j. M U G A , AND GEORGE K. MICHALOPOULOS' Box 3 4 3 2 , Department of Pathology, Duke University Medical Center, Durham, North Carolina (R.Z., S.I.M., C.K.M.); Cenentech, Inc., Department of Molecular Biology, South San Francisco, California (K.A.H.1

Acidic fibroblast growth factor (aFGF) stimulated DNA synthesis in primary rat hepatocyte cultures in a dose-dependent manner with maximal effect at 10-50 ng ml-'. This activity was dependent on the presence of heparin at a concentration of 10-50 kg.ml-'. Insulin interacted synergistically with aFGF, as it did with epidermal growth factor (EGF). The response to aFCF was only 50% that found with ECF. T h e disparity was not due to different kinetics of DNA synthesis, since the peak response for both growth factors occurred at 36-72 h r after plating of the hepatocytes. The potential relevance of this novel hepatocyte mitogen to normal and pathological liver growth is discussed.

Identification of substances capable of stimulating proliferation of hepatocytes is a n important step in elucidating the mechanisms of liver regeneration and hepatocarcinogenesis. The involvement of humoral factors in transmitting the regenerative stimulus has been demonstrated by the parabiosis experiments of Moolten and Bucher (1967) and by ectopic transplantation of liver slices (Grisham et al., 1964) or hepatocytes (Jirtle and Michalopoulos, 1981). Determination of specific factors that directly induced hepatocyte growth was possible following development of a primary culture system capable of supporting serum-free DNA synthesis and mitosis of rat hepatocytes (McGowan et al., 1981; Michalopoulos et al., 1982). To date, only a limited number of defined mitogens for hepatocytes have been found. First and foremost is EGF, which is capable of inciting DNA synthesis in hepatocytes both in vivo (Bucher et al., 1978) and in vitro (McGowan et al., 1981; Tomita et al., 1981; Hasegawa et al., 1982). Likewise, TGF,, which binds the EGF receptor, is a hepatocyte mitogen and is produced by some hepatoma cell lines (Luetteke and Michalopoulos, 1985, Meade and Fausto, 1989). Serum and platelet-poor plasma from normal and partially hepatectomized rats (Michalopoulos et al., 1984) as well a s plasma from humans with fulminant hepatic failure (Gohda et al., 1988) displays hepatocyte growth-stimulatory activity. The activity in rat serum was shown to be due to the synergistic effect of a large (>100,000) and a low-molecular-weight ( = G O O ) compound. The high-molecular-weight polypeptide, hepatopoietin A (HPTA), has a molecular weight of 105,000. HPTA was recently purified and demonstrated to be a heterodimer of 70,000 and 35,000 M , (Zarnegar and Michalopoulos, 1989). This factor may be closely related to the one from human plasma andlor to another hepatotrophic 8 1990 WILEY-LISS, INC.

factor partially purified from rat platelets (Nakamura et al., 1986). No other complete hepatocyte mitogens have been described, although several hormones are known to amplify the mitogenic effect of EGF, including norepinephrine (Cruise et al., 1985) and vasopressin (Russell and Bucher, 1982). Recently, we investigated the ability of another wellcharacterized polypeptide to initiate DNA synthesis in primary cultures of r a t hepatocytes. Acidic fibroblast growth factor (aFGF), originally purified from normal brain, is a n anionic, 17-kDa protein mitogenic for a variety of cells primarily of mesodermal and neuroectodermal origin (Gospodarowicz et al., 1987). Kan et al. (1988) recently established its mitogenic activity on the human hepatoblastoma-derived cell line HepG2. We report here the ability of aFGF to operate as a complete mitogen for normal hepatocytes in serum-free primary culture with a dependence on heparin for its activity.

MATERIALS AND METHODS Materials Recombinant human aFGF was a gift from Dr. Thomas Maciag (Red Cross Laboratory, Bethesda, MD) and J. Anthony Thompson (GTI, Gaithersburg, MD). EGF was obtained from Collaborative Research, Inc. (Lexington, MA). Collagenase (type 1)was purchased from Worthington Biochemical Corporation (Freehold, NJ). The medium used was minimal essential media (MEM) with Earle's salts and nonessential amino acids from Grand Island Biological Co. (Grand Island, NY). Nuclear track emulsion (NTBS) and other autoradioReceived August 1, 1989; accepted December 6,1989.

'To whom reprint requestsicorrespondenceshould be addressed.

130

HOUCK ET AL. 60

6007 I I

X

0

C

500

U

-

F

,,or

1 I

i

C

.E

c

40 9)

n -0 1

I L 0

200

0

100

I

/

n

0

2o

0

I

0

0

0

0

10

20

30

40

50

60

70

0.01

0.1

1.0

10

100

Heparin ( p g m1-l)

80

HBGF I (ng mi-')

Fig. 1. Dose-dependent stimulation of hepatocyte DNA synthesis by aFGF. Primary cultures of rat hepatocytes were treated with the indicated amount of aFGF in the presence of 10 p g m l heparin and DNA synthesis determined as described under Materials and Methods. EGF (10 ng.ml-') was used in the absence of aFGF and heparin for comparison and datum is expressed as DPM per culture. The effect of TGF, (0.5 ng.rn1-l) on a single dose of aFGF (26 ng.ml- ') is shown. Points represent the mean of triplicate determinations and error bars the standard deviation (SD).

'

graphic chemicals were from Eastman Kodak (Rochester, NY). All other reagents, including heparin, were from Sigma Chemical Company (St. Louis, MO).

Fig. 2. Effect of heparin on hepatocyte DNA synthesis induced by aFGF. Hepatocyte cultures were stimulated with 10 ng.ml ' aFGF and varying concentrations of heparin. Points represent the mean of triplicate determinations and error bars the SD. In the absence of heparin, the monitored counts were 47.2 x 10'.

thesis reached a peak a t 10-50 ngam1-l aFGF. This corresponds fairly well to the dose response for EGF in terms of potency (McGowan et al., 1982; Sand and Christoffersen, 1987); however, the maximum response to aFGF was only 40% that of EGF. Less than 40% of EGF incorporation was measured when DNA synthesis was assayed in 12-hr pulses, as shown in Figure ,4. TGF,, a potent inhibitor of EGF-induced hepatocyte DNA synthesis (Carr e t al., 19861, also blocked I'HIthymidine uptake stimulated by aFGF. Figure 1 demonstrates that the effect of aFGF a t 26 ng.rn1-I was completely abolished by TGF, a t 0.5 ng-ml-'.

Hepatocyte isolation and culture Hepatocytes were isolated from male Fisher 344 rats (100-250 g) by a modification of the two-step collagenase perfusion technique of Seglen (1976). Concisely, livers were perfused in situ through the vena cava with Interaction of a F G F with heparin and insulin 250 ml of calcium-free buffer followed by the buffer The data in Figure 1were obtained in the presence of with 5.7 mM calcium and 0.5 mg-ml-' collagenase 50 pg/ml-' heparin; aFGF had little activity in its ab(Jirtle et al., 1981).The digested liver was minced with sence. Figure 2 clearly shows the dose dependency #of scissors in ice-cold buffer, strained through nylon aFGF on heparin for its activity. A heparin concentrismesh, and centrifuged three times at 50g to yield tion of 5-50 pg.rn1-l was optimal, with doses above greater than 95% parenchymal hepatocytes, a s judged 100 pg-ml-' significantly suppressing the mitogenic by morphological and histochemical criteria (Michalo- effect of aFGF. poulos, et al., 1982). Hepatocytes were plated at Figure 3 shows that, as it did with EGF, insulin in100,000 cells ml-' in 1 ml of MEM with Earle's salts, teracted synergistically with aFGF. For both EGF and nonessential amino acids, pyruvate (1 mM), gentami- aFGF, insulin (lop7 M), which had very little activity cin sulfate (50 pg.ml-'), insulin (lop7 M), and 2.5% on its own, stimulated a n approximate doubling of the fetal calf serum (FCS). Tissue culture dishes, without l3H1thymidine incorporation induced by either growth or with (for autoradiography) glass coverslips, were factor alone. There is a small, but insignificant, addiprecoated with rat tail collagen (Strom and Michalop- tive effect with maximally effective doses of EGF and oulos, 1982). The plating medium was changed after 2 aFGF. h r to the medium described above but without FCS and with L3Hlthymidine (5 pCi.ml-'1 and the growth facKinetics of proliferation in response to a F G F tors to be tested. Cultures were incubated for 48 h r a t The inability of aFGF to reach the zenith of prolifer37°C in 7% C 0 2 and then harvested and processed as ation prompted by EGF was puzzling. A potential sopreviously described in liquid scintillation counting or lution could be a n altered time course of response; i.e., autoradiography (Houck e t al., 1988). aFGF required much longer than EGF to provoke S phase. To this end, hepatocytes were pulsed for 12 hr RESULTS with [3H]thymidine from 0-96 h r after plating. Figure Effect of a F G F on hepatocyte DNA synthesis 4 shows the above hypothesis to be incorrect, since the The ability of aFGF to stimulate DNA synthesis in time course for response to aFGF mirrors t h a t of EGF, primary cultures of rat hepatocytes in a dose dependent but with a truncated peak of incorporation at 48-72 hr. manner is illustrated in Figure 1.The DNA synthesis DISCUSSION measured was cumulative for a total of 46 h r (from Acidic FGF is a member of a class of heparin-binding 2-48 h r in culture). Measured by both incorporation of ['Hlthymidine into trichloracetic acid (TCA)-precip- growth factors that are mitogenic for a wide range of itable material and nuclear labeling index, DNA syn- neuroectoderm- and mesoderm-derived cell types (Gos-

STIMULATION OF HEPATOCYTE GROWTH BY aFGF

,-, 350 m

131

to inherent differences between hepatocytes. Precedence exists for such a response-a perivenous subpopulation of glutamine synthetase-positive hepatocytes respond differently than glutamine synthetase-negative cells to a variety of hepatocyte growth stimuli (Gebhardt e t al., 1986). However, this particular subpopulation is too small to account for the observed disparity. As with many cell types, hepatocytes require the presence of heparin for a mitogenic response to aFGF. The potentiation of aFGF activity by the anionic glycosaminoglycan heparin is well known but incompletely understood. aFGF activity in neural (Wagner Control EGF HBGF I EGF/HBGF I and D’Amore, 1986) and endothelial (Thornton et al., Fig. 3. Interaction of HBGF and EGF with insulin in hepatocyte 1983) cells is markedly stimulated by heparin. aFGF cultures. EGF (10 ng.ml ’) and aFGF (10ngml-’1 with heparin (10 binds heparin (Maciag et al., 1984), which may generbg.ml-’) were used in primary hepatocyte cultures without and with insulin (1 x M). Vertical bars represent the mean of triplicate ate a more stable tertiary structure, resulting in greater biological activity (Schreiber et al., 1985) or a cultures 2SD. prolonged biological half-life (Damon et al., 1989). In vitro, heparin protects aFGF from proteolytic degrada1000, 1 tion (Rosengart et al., 1988). However, heparin is apparently not required in all experimental systems, a s demonstrated by the mitogenic activity of aFGF alone on normal human keratinocytes (Rosengart et al., 1988). Heparin binding may also be critical to tissue localization of aFGF, since it has been shown to be associated with heparin-like glycosaminoglycans in basement membranes (Folkman et al., 1988). h The physiological significance of the ability of aFGF 0 3 e-~,/-o-o~o\~\o to act as a hepatocyte mitogen is unknown. aFGF is the z principal mitogen from neural tissue for human endo‘ 0 a n thelial cells (Maciag et al., 19841, but neither the poly0 peptide nor the mRNA, for i t has been detected in liver 0 12 24 36 48 60 72 84 96 108 (Jaye et al., 1986). Liver efficiently clears intraveTime of Harvest (Hr) nously injected [lz5I1-aFGF, and this uptake was increased by simultaneous injection of heparin (RosenFig. 4. Rate of DNA synthesis in hepatocyte cultures treated with et al., 1989). However, physiological sources of gart EGF and aFGF. Two hours after plating, cultures were treated with aFGF relevant to liver have not been identified. ( 0 )or aFGF (50 ng.rn1-I) with heparin (50 kg.ml ‘1 EGF (10ngml Of potential relevance to carcinogenesis is the find( 0 ) .Cultures were pulsed for 12 h r with [’H lthymidine and harvested at the end of the labeling period for determination of DNA synthesis. ing that aFGF is one of a six-member gene family of Data is displayed as incorporation of I’HIthymidine over that seen in which four are oncogenes discovered in various transcontrol cultures not treated with EGF or aFGF. Symbols are the mean forming assays: int-2, hst/KS3, FGF-5 (Thomas, 1988), of triplicate determinations tSD. and FGF-6 (Marics et al., 1989).Notably, both hst/KS3 and FGF-5 have been associated with hepatomas. The sixth gene family member, basic fibroblast growth facpodarowicz et al., 1987) as well a s some ectoderm-de- tor, can be transformed into a n oncogene by the addirived ones (Shipley et al., 1989). The potential rele- tion of a hydrophobic leader sequence to the normally vance of heparin-binding growth factors to regulation “leaderless” polypeptide (Rogeliji et al., 1988). Finally, of the growth of liver parenchymal cells is becoming Kan et al. (1988) showed that the human hepatoblasincreasingly clear with the purification of a potent hep- toma-derived cell line, Hep-G2, both expresses aFGF arin-binding hepatocyte mitogen, HPTA, from rabbit mRNA and responds mitogenically to the polypeptide. and human sera (Zarnegar and Michalopoulos, 1989) On the basis of these intimations, the role of aFGF in or rat platelets (Nakamura et al., 19871, and our dem- regulation of normal and pathological hepatocyte onstration that aFGF also stimulates hepatocyte DNA growth deserves thorough examination. synthesis. HPTA and aFGF are clearly different polypeptides since HPTA is a much higher molecular ACKNOWLEDGMENTS weight protein (105 kDa compared with 17 kDa) and This work was supported by grants CA-35373, CAdoes not require for heparin for activity (Zarnegar and Michalopoulos, 1989). In addition, HPTA can stimulate 30241, CA-43632, and EPA CR 814344 from the Nahepatocyte growth in vitro to the same degree as EGF, tional Institutes of Health. K.A.H. was supported by while aFGF is much less effective. The reason for the NIEHS training grant ES-07031. We are grateful to inability to stimulate hepatocyte growth fully is un- Dr. Thomas Maciag and Dr. J. Anthony Thompson for clear. It does not appear to be related to the kinetics of providing the aFGF and for stimulating discussions. induction of DNA synthesis. It is possible that hepato- We also thank Debbie Regan for excellent technical cytes may have a heterogeneous response to aFGF due assistance.

2ool

1

132

HOUCK ET AL.

LITERATURE CITED

sis in primary cultures of adult rat hepatocytes in a defined medium: Effects of epidermal growth factor, insulin, glucagon, and cyclic-AMP. J . Cell. Physiol., 108t353-363. Bucher, N.L.R., Patel, U., and Cohen, S. (1978)Hormonal factors and liver growth. Adv. Enzyme Regul., 16:205-213. Mead, J.E., and Fausto, N. (1989) Transforming growth factor alpha Carr, B.I., Hayashi, I., Branum, E.L., and Moses, H.L. (1986) Inhibimay be a physiological regulator of liver regeneration by means of tion of DNA synthesis in rat hepatocytes by platelet-derived type a n autocrine mechanism. Proc. Natl. Acad. Sci. USA, 86: 1558-1562. transforming growth factor. Cancer Res., 442230-2234. Cruise, J.L., Houck, K.A., and Michalopoulos, G.K. (1985) Induction Michalopoulos, G.K., Cianculli, H.D., Novotny, A.R., Kligerman, of DNA synthesis in cultured rat hepatocytes through stimulation A.D., Strom, S.C., and Jirtle, R.L. (1982)Liver regeneration studies of alpha-1 adrenoreceptor by norepinephrine. Science, 227: with rat hepatocytes in primary culture. Cancer Res., 42t4673749-751. 4682. Damon, D.H., Lobb, R.R., D'Amore, P.A., and Wagner, J.A. (1989) Michalopoulos, G.K., Houck, K.A., Dolan, M.L., and Luetteke, N.C. Heparin potentiates the action of acidic fibroblast growth factor by (1984)Control of hepatocyte replication by two serum factors. Canprolonging its biological half-life. J . Cell. Physiol., 138t221-226. cer Res., 44:4414-4419. Folkman, J., Klagsbrun, M., Sasse, J., Wadzinski, M., Ingber, D., and Moolten, C.G.D., and Bucher, N.L.R. (1967) Regeneration of the rat Vlodavsky, I. (1988) A heparin-binding angiogenic protein-basic liver: Transfer of humoral agents by cross circulation. Science, 158: 272-274. fibroblast growth factor-is stored within basement membrane. Am. J. Pathol., 130:393-400. Nakamura, T., Teramoto, H., and Ichihara, A. (1986)Purification and Gebhardt, R., Cruise, J., Houck, K.A., Luetteke, N.C., Novotny, A., characterization of a growth factor from rat platelets for mature Thaler, F., and Michalopoulos, G.K. (1986) Differential effect of parenchymal hepatocytes in primary cultures. Proc. Natl. Acad. Sci. USA, 83:6489-6493. growth factors on growth stimulation and phenotypic stability of glutamine synthetase-positive and -negative hepatocytes in pri- Rogeliji, S., Weinberg, R.A., Fanning, P., and Klagsbrun, M. (1988) mary culture. Differentiation, 33:45-55. Basic fibroblast growth factor fused to a single peptide transforms Gohda, E., Hirohito, T., Nakayama, H., Hirono, S., Sakiyama, 0.. cells. Nature (Lond.), 331:173-175. Takahashi, K., Miyazaki, H., Hashimoto, S., and Daikuhara, Y. Rosengart, T.K., Johnson, W.V., Friesel, R., Clark, R., and Maciag, T. (1988) Purification and partial characterization of hepatocyte Heparin protects heparin-binding growth factor I from proteolyt.ic inactivation in vitro. (1988)Biochem. Biophys. Res. Commun., 152: growth factor from plasma of a patient with hepatic failure. J. Clin. 432-440. Invest., 81t414-419. Gospodarowicz, D., Ferrara, N., Schweigerer, L., and Neufeld, G. Rosengart, T.K., Kuperschmid, J.P., Maciag, T., and Clark, R.E. (1987)Structural characterization and biological functions of fibro(1989) Pharmacokinetics and distribution of heparin-binding growth factor I (endothelial cell growth factor) in the rat. Circ. Res., blast growth factor. Endocr. Rev., 8:95-109. 64:227-234. Grisham, J.W., Leong, G.F., and Hold, B.V. (1964)Heterotopic partial autotransplantation of rat liver. Cancer Res., 24t1474-1501. Russell, W.E., and Bucher, N.R.L. (1982)Vasopressin modulates liver Hasegawa, K., Watanabe, K., and Koga, M. (1982)Induction of mitoregeneration in the Brattleboro rat. Am. J . Physiol., 245:G321-324. sis in primary cultures of adult rat hepatocytes in primary cultures. Sand, T.E., and Christoffersen, T. (1987) Temporal requirement for Biochem. Biophys. Res. Commun., 104:259-265. epidermal growth factor and insulin in the stimulation of hepatoHouck, K.A., Cruise, J.L., and Michalopoulos, G. (1988) Norepinephcyte DNA synthesis. J . Cell. Physiol., 131:141-148. rine modulates the growth-inhibitory effect of TGFa in primary rat Seglen, P.O. (1976) Preparation of isolated rat liver cells. Metho'ds Cell Biol., 13:29-83. hepatocyte cultures. J. Cell. Physiol., 135:551-555. Jaye, M., Howk, R., Burgess, W., Ricca, G.A., Chiu, I.M., Ravera, Schreiber, A.B., Kenney, J., Kowalski, W.J., Friesel, R., Mehlman, T., M.W., O'Brien, S.J., Modi, W.S., Maciag, T., and Drohan, W.N. and Maciag, T. (1985) Interaction of endothelial cell growth factor (1986) Human endothelial cell growth factor: Cloning, nucleotide with heparin: Characterization by receptor and antibody recognition. Proc. Natl. Acad. Sci. USA, 82t6138-6142. sequence, and chromosome location. Science, 233.541-545. Jirtle, R.L., and Michalopoulos, G.K. (1982) Effects of partial hepa- Shipley, G.D., Keeble, W.W., Hendrickson, J.E., Coffey, R.J. Jr., and Pittlekow, M.R. (1989) Growth of normal human keratinocytes and tectomy on transplanted rat hepatocytes. Cancer Res., 42:30003004. fibroblasts in serum-free medium is stimulated by acidic and basic Jirtle, R.L., Michalopoulos, G., McLain, J.R., and Crowley, J. (1981) fibroblast growth factor. J. Cell. Physiol., 138511-518. Transplantation system for determining the clonogenic survival of Strom, S.C., and Michalopoulos, G. (1982) Collagen as a substrate for parenchymal hepatocytes exposed to ionizing radiation. Cancer cell growth and differentiation. Methods Enzymol., 82544-555. Res., 41t3512-3518. Thomas, K.A. (1987) Fibroblast growth factors. FASEB J., lt434440. Kan, M., DiSorbo, D., Hou, J., Hoshi, H., Mansson, P.E., and McKeehan, W.L. (1988) High and low affinity binding of heparin-binding Thomas, K.A. (1988) Transforming potential of fibroblast growth factor genes. TIBS, 13r327-328. growth factor to a 130-kDa receptor correlates with stimulation and inhibition of growth of a differentiated human hepatoma cell. J. Thornton, S.L., Mueller, S.N., and Levine, E.M. (1983) Human endoBiol. Chem., 263:11306-113 13. thelial cells: Use of heparin in cloning and long-term serial cultivation. Science, 222623-625. Luetteke, N.C., and Michalopoulos, G.K. (1985) Partial purification and characterization of a hepatocyte growth factor produced by rat Tomita, Y., Nakamura, T., and Ichihara, A. (1981) Control of DNA hepatocellular carcinoma cells. Cancer Res., 45:6331-6337. synthesis and ornithine decarboxylase activity by hormones and Maciag, T., Mehlman, T., Friesel, R., and Schreiber, A.B. (1984) Hepamino acids in primary cultures of adult rat hepatocytes. Exp. Cell arin binds endothelial cell growth factor, the principal endothelial Res., 135:363-371. cell mitogen in bovine brain. Science, 225:932-935. Wagner, J.A., and D'Amore, P.A. (1986) Neurite outgrowth induced Marics, K., Adelaide, J., Raybaud, F., Mattei, M.-G., Coulier, F., by an endothelial cell mitogen isolated from retina. J. Cell Biol., Planche, J., de Lapeyriere, O., and Birnbaum, D. (1989)Character103t1363-1367. ization of the HST-related FGF.6 gene, a new member of the fibro- Zarnegar, R., and Michalopoulos, G.K. (1989) Purification and biologblast growth factor gene family. Oncogene, 4t335-340. ical characterization of human hepatopoietin A, a polypeptide McGowan, J.A., Strain, A.J., and Bucher, N.L.R. (1981) DNA synthegrowth factor for hepatocytes. Cancer Res., 49:3314-3320.

Acidic fibroblast growth factor (HBGF-1) stimulates DNA synthesis in primary rat hepatocyte cultures.

Acidic fibroblast growth factor (aFGF) stimulated DNA synthesis in primary rat hepatocyte cultures in a dose-dependent manner with maximal effect at 1...
463KB Sizes 0 Downloads 0 Views