Novel Therapeutic Strategies for Leiomyomas: Targeting Growth Factors and Their Receptors

Environmental Health Perspectives Volume 108, Supplement 5, October 2000

Novel Therapeutic Strategies for Leiomyomas: Targeting Growth Factors and Their Receptors

Romana A. Nowak

Department of Obstetrics, Gynecology and Reproductive Biology, Harvard Medical School and the Center for Uterine Fibroids, Brigham and Women's Hospital, Boston, Massachusetts, USA

Abstract
Leiomyomas (fibroids) are benign smooth-muscle cell (SMC) tumors of the uterus and are the most common pelvic tumors in women. These tumors occur primarily during the reproductive years and are the most common indication for hysterectomy in women. Unfortunately the only effective treatments for leiomyomas and the associated abnormal uterine bleeding are surgical, involving either hysterectomy, myomectomy, or hysteroscopic removal of the tumors. The goal of this paper is to discuss recent research findings that support the idea of using therapeutic compounds that block the actions of specific growth factors as therapeutic agents for treatment of leiomyomas and abnormal uterine bleeding. Most of the studies were carried out using cell cultures of leiomyoma or myometrial SMCs. Primary cultures of SMCs provide a system for investigation of the roles of growth factors and their receptors in proliferation of normal myometrial and leiomyoma SMCs. Several growth factors have been shown to be present and to have regulatory roles in the proliferation of uterine SMCs. Bioassay and Western blotting of fast protein liquid chromatography fractions of tissue extracts identified platelet-derived growth factor, heparin-binding epidermal growth factor, hepatoma-derived growth factor, and basic fibroblast growth factor in normal myometrium and fibroid tumors. The presence of heparin-binding growth factors suggests a possible focus for therapeutic agents. RG13577 (a heparinlike compound) and halofuginone (an alkyloid) reversibly inhibited DNA synthesis of normal myometrial and leiomyoma cells without toxic effects. Pirfenidone, a known antifibrotic drug, inhibited DNA synthesis and synthesis of collagen type I mRNA in normal and fibroid cells, and decreased collagen type III mRNA only in normal myometrial cells. Another hopeful therapeutic candidate, interferon-

, significantly inhibited growth factor-stimulated proliferation in both normal and leiomyoma cells. These results suggest that future nonrgical treatments for leiomyomas may include compounds that block the actions of specific growth factors that regulate proliferation and collagen production by uterine SMCs. Key words: anti-angiogenic, fibrosis, growth factors, heparin, interferons, leiomyomas, smooth muscle, transforming growth factors. -- Environ Health Perspect 108(suppl 5):849-853 (2000).

http://ehpnet1.niehs.nih.gov/docs/2000/suppl-5/849-853nowak/abstract.html

This article is based on a presentation at the conference on Women's Health and the Environment: The Next Century--Advances in Uterine Leiomyoma Research held 7-8 October 1999 in Research Triangle Park, North Carolina, USA.
Address correspondence to R.A. Nowak, University of Illinois, 1207 W. Gregory Dr., ASL 310, Urbana, IL 61801 USA. Telephone: (217) 244-3902. Fax: (217) 333-8286. E-mail: ranowak@uiuc.edu
Received 23 February 2000; accepted 31 July 2000.

Uterine leiomyomas, or fibroids, are well-vascularized, benign, monoclonal tumors of the myometrium characterized by an increase in smooth-muscle cell (SMC) proliferation and an abnormally great amount of collagen deposition (1). They are relatively common tumors, with an incidence reported between 20 and 25% of women. They occur predominantly in women during their reproductive years. Current therapies are limited primarily to surgery.

Heterogeneous cytogenetic abnormalities and specific translocations have been associated with fibroids (2). One example is the chromosomal 12:14 translocation that results in abnormal expression of the high mobility group IC gene in the leiomyoma (3). The loss of a specific gene, tuberous sclerosis 2, has also been shown to result in the development of leiomyoma (4). However, the pathogenesis of uterine fibroid tumors is still not well understood. Comparison of leiomyomas with other types of mesenchymally derived lesions might suggest similar pathophysiologies and some possible avenues for treatment. Keloid tumors, for example, are benign, monoclonal, fibroblast tumors characterized also by increased proliferation and high collagen production (5). Vascular SMCs that contribute to restenosis after balloon angioplasty are also monoclonal proliferations that produce high amounts of collagen (6,7). There is abundant evidence that the dysregulation of growth factors and receptors for growth factors are important in the growth of keloid and vascular SMC lesions. This brief overview will address recent studies in uterine fibroids of growth factors and receptors that may be useful targets for controlling leiomyoma growth.

A growth factor shown to be involved in wound repair and implicated in fibrotic disease is transforming growth factor (TGF)-ß. The TGF-ß ligand system turns on several processes, including increases in extracellular matrix production and decreases in collagenase production (8,9). Our laboratory examined the expression of two of the TGF-ß isoforms, ß1 and ß3, in matched normal myometrium and leiomyoma of excised tissues of surgical patients. Results from six of these patients are shown (Figure 1). Consistently, TGF-ß3 was elevated in leiomyomas and was virtually undetectable in myometrium of any of the matched pairs examined. Densitometric analysis from a total of 12 patients revealed that in leiomyomas, expression of TGF-ß3 was 5-fold higher than in corresponding autologous myometrium. TGF-ß1 expression did not display a discernable pattern; in the leiomyomas it was the ß3 isoform of TGF that was elevated. We then examined the proliferative responses to TGF-ß1 and -ß3 isoforms of leiomyoma and normal myometrial cells that were established as primary cultures from surgical specimens. In myometrium, TGF-ß1 inhibited cell proliferation as measured by DNA synthesis, whereas there was no inhibitory effect seen in leiomyoma cells (Figure 2). The inhibitory effect of TGF-ß1 on normal myometrial cells was lost in the presence of an antibody that recognizes all TGF-ß isoforms. TGF-ß3 caused a similar inhibition of DNA synthesis in myometrial cells, and either did not affect or slightly increased DNA synthesis in leiomyoma cells (data not shown). Antibody to TGF-ß abolished these effects.

Figure 1. Northern blot analysis of leiomyoma (L) and myometrial (M) tissue RNA extracts for TGF-ß3 (3.6 kb mRNA), TGF-ß1 (2.5 kb mRNA) and -tubulin (2.2 kb mRNA). Results are shown for L and M from six different patients. Tissues from a total of 12 different patients were analyzed. Densitometric analysis of the relative levels of TGF-ß1 and TGF-ß3 mRNAs were made after correcting for possible differences in sample loading by normalizing to -tubulin expression. TGF-ß3 mRNA levels were 5-fold higher in L than in autologous M, whereas TGF-ß1 mRNA levels were similar between the two tissues.

Figure 2. The effect of TGF-ß1 on DNA synthesis by cultured leiomyoma and myometrial SMCs. All three concentrations of TGF-ß1 (0.1, 1.0, and 10.0 ng/mL) caused a significant reduction in DNA synthesis by myometrial SMCs (p < 0.05, bars with letter a are different from control) but had no effect on leiomyoma SMCs. Concomitant administration of a TGF-ß immunoneutralizing antibody (Ab) prevented the inhibitory effect of TGF-ß.

Myometrial and leiomyoma SMCs, when placed in culture, produce collagen in great amounts. A stimulatory effect of TGF-ß on collagen production by cultured uterine SMCs is therefore difficult to demonstrate directly. To investigate the effect of TGF-ß on collagen production, we attempted to block autocrine- or paracrine-regulated collagen production with immunoneutralizing antibody. Steady-state levels of new collagen mRNA were measured after incubating leiomyoma and normal myometrial cells with antibody that recognizes all isoforms of TGF-ß (Figure 3). The two primary collagens made by the cells are collagen types I and III. Production of collagen types I and III were both inhibited 45-65% by anti-TGF-ß antibody, compared to untreated control cultures or cultures treated with nonspecific mouse IgG. The data in summary show that TGF-ß regulates proliferation and collagen production in uterine SMCs, and that leiomyomas display two defects in the TGF-ß system: a significantly elevated expression of TGF-ß3, and an abnormal growth inhibitory response to TGF-ß.

Figure 3. The effect of a TGF-ß1-3 immunoneutralizing antibody (TGF-ß AB) on steady-state levels of collagen type I and type III mRNAs in cultured leiomyoma (L) and myometrial (M) SMCs. Cells were treated for 24 hr with either no antibody (control), a nonspecific mouse IgG (NS AB), or the TGF-ß AB. Levels of collagen type I and type III mRNAs were normalized to -tubulin expression. Treatment with the TGF-ß AB caused a significant suppression in mRNA levels of both collagen type I and type III for both cell types.

Proliferative responses of cultured leiomyoma cells may then be used as an assay for prospective therapeutics against fibroids. Pirfenidone (Figure 4) is one such compound, an antifibrotic drug currently in phase II clinical trials for the treatment of pulmonary fibrosis (10). Because it is an antifibrotic drug, its effects in our culture system were of great interest to us. Pirfenidone caused a dose-dependent inhibition of serum-stimulated DNA synthesis in both normal myometrial and leiomyoma SMCs (Figure 5). The inhibition was significant at each concentration of the compound tested. Live and dead cell counts were made of cultured cells after 7 days of treatment with pirfenidone to show that cell numbers were diminished without an increased percentage of dead cells at all but the highest concentration tested (data not shown). Levels of lactate dehydrogenase activity were also not affected in either leiomyoma or normal myometrial cells (11). Pirfenidone caused a dose-dependent decrease in the production of mRNA for collagen type I in both leiomyoma cell cultures and matched normal myometrial cell cultures (Figure 6) but a decrease in collagen type III steady-state mRNA only in cultures of normal myometrial cells, not in leiomyoma cells (except at the highest pirfenidone concentration). The effect of this antifibrotic drug in leiomyoma cells is specific only for collagen type I. The potential development of pirfenidone as a therapy for treatment of uterine leiomyomas should be seriously considered, as it has been reported to be well tolerated by patients taking it orally (12).

Figure 4. The chemical structure of pirfenidone (5-methyl-1-phenyl-2(1H)-pyridone).

Figure 5. Effect of pirfenidone on myometrial and leiomyoma cell proliferation. Concentrations of pirfenidone tested were 0, 0.01, 0.1, 0.3, and 1.0 mg/mL. The effect of the various doses of pirfenidone on thymidine incorporation is expressed as a percent of control (0 mg/mL pirfenidone). The serum stimulation of DNA synthesis was significantly inhibited in a dose-dependent manner in both myometrial (p < 0.001) and leiomyoma (p < 0.005) cells. Each bar represents the mean ± SD of 24 wells. Bars bearing different letters (a vs b) are significantly different from one another. Reproduced from Lee et al. (11) with permission of The Endocrine Society.

Figure 6. Effect of pirfenidone on collagen type I (1) messenger RNA expression in myometrial and leiomyoma cells. Lanes 1,6: 0 mg/mL pirfenidone (P); lanes 2,7: 0.01 mg/mL P; lanes 3,8: 0.1 mg/mL P; lanes 4,9: 0.3 mg/mL P; lanes 5,10: 1.0 mg/mL P. Two mRNA transcripts (4.8 and 5.8 kb) were detected for collagen type I. Differences in sample loading were corrected by normalization to -tubulin (2.2 kb). The graph shows densitometric analysis after normalization. Each bar represents the mean ± SD with asterisk (*) representing p < 0.05. Reproduced from Lee et al. (11) with permission of The Endocrine Society.

Although leiomyomas bear the hallmarks of fibrotic disease, they also appear to be a disease of abnormal blood vessel structure or function. Dye injection of uterine fibroids reveals a rich and complex network of blood vessels (13). One would suppose a massive production of growth factors to develop so rich a vasculature for the tumor. Additionally, the presence of the fibroid appears to exert dramatic effects on the vasculature of its surrounding environment, as a significant dilation of veins occurs in the local endometrium (2). Worthwhile subjects of study, then, are angiogenic growth factors that may be produced by fibroids. Our laboratory has focused on characterization and identification of heparin-binding growth factors produced by fibroid tumors (as many of the known angiogenic growth factors are heparin binding). Leiomyomas and normal myometrial tissue from surgical patients were homogenized and their protein extracts were run through mini-heparin columns. Fast protein liquid chromatography (FPLC) fractions were then eluted over linear sodium chloride gradients and tested in bioassays using 3T3 fibroblasts and human normal myometrial cells (14). Figure 7 shows an elution profile acquired in the myometrial cell bioassay. Four peaks of mitogenic activity were detected both for normal myometrial tissue extracts and for leiomyoma extracts. Western blots determined that peak I (0.4 M salt) contained platelet-derived growth factor (14). Basic fibroblast growth factor (bFGF) was detected in peak IV (1.7 M salt) and also weakly in peak III (14).

Figure 7. Elution profile: SMC bioassay. Myometrial smooth muscle cell mitogenic profile of leiomyoma and myometrial protein extracts. Protein extracts from leiomyoma (A) and myometrial (B) samples were applied individually to an FPLC heparin Sepharose column apparatus. The fractions were eluted over a linear sodium chloride gradient from 0.2 to 2.0 M. These fractions were then applied to a cell proliferation assay, measuring tritiated thymidine incorporation into myometrial SMCs. Profiles from five other patients yielded similar results. Reproduced from Mangrulkar et al. (14) with permission of The Society for the Study of Reproduction.

The levels of bFGF mRNA were compared by Northern blot analysis and RNase protection assay in both myometrial and leiomyoma tissue RNA samples. Levels of bFGF mRNA were higher in some, but not all, leiomyoma tissue specimens. Immunohistochemistry showed that bFGF protein was present in only low amounts in myometrium, being concentrated around bundles of SMCs (Figure 8). In contrast, immunostaining of leiomyoma tissue sections showed bFGF protein was deposited heavily in extracellular matrix surrounding the SMCs. Thus, the extensive matrix of leiomyoma tumors appears to act as a reservoir of bFGF. To examine expression of fibroblast growth factor (FGF) receptor, surgical patients were divided into those in the menstrual/early proliferative phase, those in the periovulatory/early secretory phase (days 17-21), and those in late secretory phase. Specific immunohistochemical staining for FGF type I receptor showed similar amounts of receptor expression in uterine sections of women with and without fibroid tumors in the menstrual/early proliferative and late secretory phases. However, in the periovulatory/early secretory phase, there was a dramatic suppression of FGF receptor in the stromal cells of the endometrium of normal patients. In patients with leiomyoma, suppression of FGF receptor did not occur (15). It is possible the lack of receptor suppression in leiomyoma patients is due to increased bFGF production by the fibroids.

Figure 8. Immunohistochemical staining of leiomyoma and myometrial tissue sections using an antibody to basic FGF. Leiomyoma tissue sections were incubated with (A) 5 µg/mL nonspecific rabbit IgG; (B)1:75 dilution of rabbit anti-basic FGF. Myometrial tissue sections were incubated with (C) nonspecific rabbit IgG; (D) 1:75 dilution of rabbit anti-basic FGF. Immunostaining from five additional patients revealed similar results. Final magnification = 400. Reproduced from Mangrulkar et al. (14) with permission of The Society for the Study of Reproduction.

Elution peak II was found by Western blots to contain two known growth factors of similar molecular weight: heparin-binding epidermal growth factor (HBEGF, 38K MW) and hepatoma-derived growth factor (HDGF, 40K MW) (data not shown). In cell proliferation assays, we have found HDGF was poorly mitogenic for SMCs but was a potent mitogen for endothelial cells. HDGF warrants further examination as an angiogenic factor in the uterus. In contrast, HBEGF is a strong mitogen for SMCs but is not mitogenic for vascular endothelial cells.

It is not premature at these early stages of research to consider inhibitors of heparin-binding growth factors potential therapeutic agents against fibroids. One candidate inhibitor is heparin itself. Numerous vascular biological studies have shown that heparin or heparinlike compounds added to SMCs in culture inhibit their proliferation (16). We tested a polyanionic sulfated heparinlike compound, RG13577. This compound is known to inhibit the binding of heparin-binding growth factors to vascular SMCs and to inhibit dimerization of the FGF receptor (17). A specific receptor for RG13577 has been demonstrated, but not yet identified for its natural ligand, on vascular SMCs. We examined the effect of RG13577 on proliferation of normal myometrial and leiomyoma SMCs in culture. The compound inhibited DNA synthesis and cell number (without increased cell death measured by trypan blue exclusion) in both normal cells and leiomyoma cells (data not shown). The inhibition was more significant in leiomyoma than in normal smooth muscle. We found the inhibitory effects were reversible with cell washes and further incubation without RG13577. The mechanism of inhibition by this compound is not known.

Another potential therapeutic agent is halofuginone, an alkyloid used as a coccidiostat in poultry (Figure 9). At high doses, halofuginone causes tissue abnormalities in chickens that make their skin fragile and easily torn. The compound is shown to inhibit collagen production by fibroblasts and vascular SMCs (18,19). We tested the effect of halofuginone on cell proliferation of leiomyoma and normal myometrial SMCs. Cell counts revealed significant inhibition of cell proliferation in leiomyoma (fibroid) SMCs (Figure 10). The ED₅₀ was between 7.5 and 10.0 ng/mL. Similar inhibition was seen for myometrial SMCs (data not shown).

Figure 9. The chemical structure of halofuginone.

Figure 10. Effect of increasing concentrations of halufuginone (H) on proliferation of leiomyoma (fibroid) SMCs. Cells were plated and treated with halofuginone (0, 1, 5, 10, 25, 50, 75, and 100 ng/mL) for a period of 7 days, after which cell counts were performed. Medium was replaced after 4 days with fresh treatments. Results are the means of three experiments.

All the potential therapeutic agents investigated, pirfenidone, RG13577, and halofuginone, inhibit SMC proliferation and inhibit collagen production as well. Another family of proteins with similar effects on other cells is the interferons (IFNs). IFNß inhibits cell proliferation in the cancer cell line MVP2 (20). IFNs also block both the synthesis and action of FGF in certain tumor cells, and inhibit collagen production in skin fibroblasts (21). In our proliferation assay system, IFN inhibited in a dose-dependent fashion the serum-stimulated proliferative responses of normal myometrial cells from six surgical patients (Figure 11). When myometrial cells were stimulated to proliferate with several concentrations of bFGF, 100 U/mL IFN significantly inhibited this DNA synthesis also, even though maximal proliferation was stimulated by the lowest bFGF concentration (22). Serum-stimulated proliferation of matched leiomyoma cells from the patients was also inhibited by addition of IFN (22), as was bFGF-stimulated proliferation. These results are consistent with a recently reported case study that described a patient treated for hepatitis with IFN for 6 months (23). Before treatment was initiated, a large fibroid was detected in her uterus. After the 6 months of treatment, the fibroid had shrunk 50% in diameter and did not show any increase in size up to 17 months after cessation of treatment. We are hopeful that complete clinical trials will show the efficacy of IFN treatment against fibroid tumors.

Figure 11. The inhibitory effect of IFN- on bFGF-induced DNA synthesis in cultured human myometrial SMCs. Cells were treated with either medium + 10% fetal bovine serum or with various concentrations of bFGF (0.5, 1.0, 5.0, and 10 ng/mL) to compare mitogenic effects. Cells cultured in serum were also treated with IFN- at 10, 100, and 1,000 U/mL, whereas cells cultured in the various concentrations of bFGF also were treated with 100 U/mL IFN-. All concentrations of bFGF caused a significant increase (p < 0.05) in DNA synthesis that was inhibited by concomitant treatment with IFN-. Reproduced from Lee et al. (22) with permission of The Society for the Study of Reproduction.

In conclusion, we hypothesize that leiomyoma tumors would be good targets for a variety of compounds that inhibit the action of specific growth factors. In particular, the targeting of heparin-binding growth factors and of the TGF-ßs may prove to be an effective nonsurgical alternative for treatment of leiomyomas and their associated abnormal uterine bleeding. We must await the outcome of clinical trials to see if this proves to be the case.

REFERENCES AND NOTES

1. Buttram VC, Reiter RC. Uterine leiomyomata: etiology, symptomatology, and management. Fertil Steril 36:433-445 (1981).

2. Stewart EA, Nowak RA. Leiomyoma-related bleeding: a classic hypothesis updated for the molecular era. Hum Reprod Update 2:295-306 (1996).

3. Schoenberg Fejzo M, Ashar HR, Krauter KS, Powell WL, Rein MS, Weremowicz S, Yoon S, Kucherlapati RS, Chada K, Morton CC. Translocation breakpoints upstream of the HMGIC gene in uterine leiomyomata suggest dysregulation of this gene by a mechanism different from that in lipomas. Genes Chromosomes Cancer 17:1-6 (1996).

4. Howe SR, Everitt JL, Gottardis MM, Walker C. Rodent model of reproductive tract leiomyomata: characterization and use in preclinical therapeutic studies. Prog Clin Biol Res 396:205-215 (1997).

5. Tredget EE, Nedelec B, Scott PG, Ghahary A. Hypertrophic scars, keloids, and contractures. Surg Clin North Am 77:701-730 (1997).

6. Zalewski A, Shi Y. Vascular myofibroblasts: lessons from coronary repair and remodeling. Arterioscler Thromb Vasc Biol 17:417-422 (1997).

7. Chung IM, Schwartz SM, Murry CE. Clonal architecture of normal and atherosclerotic aorta: implications for atherogenesis and vascular development. Am J Pathol 152:913-923 (1998).

8. Massague J. The transforming growth factor-ß family. Annu Rev Cell Biol 6:597-641 (1990).

9. Border WA, Noble NA. Transforming growth factor ß in tissue fibrosis. N Engl J Med 331:1286-1292 (1994).

10. Iyer SN, Wild JS, Schiedt MJ, Hyde DM, Margolin SB, Giri SN. Dietary intake of pirfenidone ameliorates bleomycin induced lung fibrosis in hamsters. J Lab Clin Med 125:779-785 (1995).

11. Lee B-S, Margolin SB, Nowak RA. Pirfenidone: a novel pharmacological agent that inhibits leiomyoma cell proliferation and collagen production. J Clin Endocrinol Metab 83:219-223 (1998).

12. Raghu G, Mageto Y, Lurton J, Limond M. Treatment of idiopathic pulmonary fibrosis (IPF) with a new antifibrotic drug- pirfenidone [Abstract]. Am J Respir Crit Care Med 155:A741 (1997).

13. Sampson JA. The blood supply of uterine myomata. Surg Gynecol Obstet 14:215-230 (1912).

14. Mangrulkar RS, Ono M, Ishikawa M, Takashima S, Klagsbrun M, Nowak RA. Isolation and characterization of heparin-binding growth factors in human leiomyomas and normal myometrium. Biol Reprod 53:636-646 (1995).

15. Anania CA, Stewart EA, Quade BJ, Hill JA, Nowak RA. Expression of the fibroblast growth factor receptor in women with leiomyomas and abnormal uterine bleeding. Mol Hum Reprod 3:685-691(1997).

16. Horiuchi A, Nikaido T, Zhai Y-L, Ito K, Orii A, Fujii S. Heprin inhibits proliferation of myometrial and leiomyomal smooth muscle cells through the induction of -smooth muscle actin, calponin h1 and p27. Mol Hum Reprod 5:139-145 (1999).

17. Benezra M, Ben-Sasson SA, Regan J, Chang M, Bar-Shavit R, Vlodavsky I. Antiproliferative activity to vascular smooth muscle cells and receptor binding of heparin-mimicking polyaromatic anionic compounds. Arterioscler Thromb 14:1992-1999 (1994).

18. Nagler A, Miao H-Q, Aingorn H, Pines M, Genina O, Vlodavsky I. Inhibition of collagen synthesis, smooth muscle cell proliferation, and injury-induced intimal hyperplasia by halofuginone. Arterioscler Thromb Vasc Biol 17:194-202 (1997).

19. Elkin M, Ariel I, Miao H-Q, Nagler A, Pines M, deGroot N, Hochberg A, Vlodavsky I. Inhibition of bladder carcinoma angiogenesis, stromal support, and tumor growth by halofuginone. Cancer Res 59:4111-4118 (1999).

20. Granstein RD, Flotte TJ, Amento EP. Interferons and collagen production. J Invest Dermatol 95:75S-80S (1990).

21. Buszello H. Antiproliferative effects of four different cytokines on renal carcinoma cell lines. Anticancer Res 15:735-738 (1995).

22. Lee B-S, Stewart EA, Sahakian M, Nowak RA. Interferon- is a potent inhibitor of basic fibroblast growth factor-stimulated cell proliferation in human uterine cells. Am J Reprod Immunol 40:19-25 (1998).

23. Minakuchi K, Kawamura N, Tsujimura A, Ogita S. Remarkable and persistent shrinkage of uterine leiomyoma associated with interferon alpha treatment for hepatitis. Lancet 353:2127-2128 (1999).

Last Updated: October 4, 2000