Conditioned Media from a Renal Cell Carcinoma Cell Line Demonstrates the Presence of Basic Fibroblast Growth Factor

0022-5347 /93/1503-0997$03.00/0

THE JOURNAL OF UROLOGY Copyright © 1993 by AMERICAN UROLOGICAL ASSOCIATION, lNC.

Vol. 150, 997-1001, September 1993 Printed in U. S. A.

CONDITIONED MEDIA FROM A RENAL CELL CARCINOMA CELL LINE DEMONSTRATES THE PRESENCE OF BASIC FIBROBLAST GROWTH FACTOR JACK H. MYDLO*, JOSEPH ZAJAC AND RICHARD J. MACCHIA From the Department of Urology, State University of New York Health Science Center at Brooklyn and the Brooklyn Veteran's Administration Hospital, Brooklyn, New York

ABSTRACT

In a previous report, we demonstrated the isolation and purification of a heparin binding growth factor from human renal carcinoma, and suggested that this growth factor may play a role in the neovascularity and growth of the tumor. In this report, we demonstrate that the growth of the renal cell carcinoma cell line RC29 is stimulated by the addition of exogenous fibroblast growth factor (FGF) , epidermal growth factor (EGF) and transforming growth factor alpha (TGFa) . Also, media conditioned by this cell line was able to stimulate growth of the A431 vulvar tumor cell line, known for its high concentration of EGF receptors, 3T3 fibroblasts, human umbilical vein (HUV) cells and RC29 cells. Using heparin-sepharose chromatography and then SDS polyacrylamide gel electrophoresis (PAGE), we were able to demonstrate several proteins in the conditioned media of the RC29 cell line. Using Western blot analysis, we detected that at least one of the proteins expressed in this conditioned media was FGF and that it belongs to the basic, not acidic, family of fibroblast growth factors. These findings suggest that renal tumors may express growth factors that may play a direct role in maintaining their unrestricted proliferation. KEY WORDS: neovascularization; growth substances; carcinoma, renal cell

The role of growth factors in tumor etiology is a subject of extensive scrutiny. The interaction of cells in a paracrine, endocrine and even autocrine manner has been described. 1• 2 Previous investigators have demonstrated the role of growth factors in angiogenesis, and have thus implicated this as a possible mechanism by which tumors expressing them can become invasive and metastasize. 3-5 Metastasizing renal cell carcinoma has been a frustrating clinical problem by virtue of its unresponsiveness to radiother­ apy, hormonal therapy, or chemotherapy. The fact that this hypervascular lesion acts in a more aggressive manner than its hypovascular counterpart, the papillary adenocarcinoma, sug­ gests the importance of vascularity and of angiogenic growth factors. 6 This, in addition to their stimulatory effect on prolif­ eration, implicates growth factors as playing a major role in the etiology of certain neoplasms. 7 Transforming growth factor alpha (TGFa) is a growth factor derived from the cleavage of a known precursor. 8 It can stimu­ late anchorage-independent cell proliferation and induce angi­ ogenesis, and is found largely in embryologic and neoplastic tissues. Recently, it has been demonstrated to exist in non­ neoplastic tissue such as keratinocytes. 9• 10 The epidermal growth factor receptor (EGFR) is an ubiquitous growth factor and has been well characterized. Its expression has been docu­ mented in many tissues. It not only binds to epidermal growth factor (EGF) , but to TGFa as well. • In a previous report, we purified a heparin binding growth factor from renal cell carcinoma and demonstrated that it is homologous to basic FGF (bFGF) . It was also shown to be angiogenic. 13 Subsequent to that report we demonstrated that five samples of stage I renal cell carcinoma expressed more TGFa and epidermal growth factor receptor (EGFR) both at the mRNA and protein level than did normal autologous tis11

12

Accepted for publication February 8, 1993. * Requests for reprints: Department of Urology, SUNY Health Sci­ ences Center at Brooklyn, Box 79, 470 Clarkson Avenue, Brooklyn, New York 11203.

sue. 14 It is conceivable, therefore, that an autocrine mechanism exists by which renal cell carcinoma cells can produce their own angiogenic growth factors which can then bind its own membrane receptors and stimulate proliferation. 15 In this report, the stimulation of the A431 vulva cancer cell line, which has a high concentration of EGF receptors, 16 by the RC29 cell line suggests that the latter cell line is producing either EGF or TGFa, as well as bFGF, based on Western blot analysis, and that a possible autocrine mechanism may exist in this particular type of tumor cell line. This may represent the mechanism of growth of renal cell carcinoma in vivo. MATERIALS AND METHODS

All cell culture reagents, antibiotics, fetal calf serum and Hydrofluor scintillation cocktail were purchased from GIBCO Laboratories, Grand Island, New York Tritiated thymidine and Il25-protein A were purchased from New England Nuclear (NEN) , Boston, Massachusetts. Heparin-sepharose was pur­ chased from Pharmacia, Piscataway, New Jersey. The chemi­ cals for the gel electrophoresis and for silver staining were from BioRad, Richmond, California. All chemical reagents used in the extraction and purification of the growth factor were pur­ chased from Fisher Scientific Co., Fairlawn, New Jersey. Pu­ rified acidic and basic FGF, EGF and TGFa were purchased from Collaborative Research, Boston, Massachusetts. Molecu­ lar weight standards included chymotrypsin A (25,000) , ovalbumin, (43,000) and bovine serum albumin, (67,000) and were purchased from BRL, Gaithersburg, Maryland. Cell cultures and growth factor assay, BALB/c 3T3 cells and human umbil­ ical vein (HUV) endothelial cells were obtained from American Type Culture Collection, Rockville, Maryland and were grown in Dulbecco's Modified Eagles Medium (DME) containing 4.5 gm./1. glucose, 1% 1-glutamine, 50 U/ml. penicillin and 50 µg./ ml. streptomycin and 10% fetal calf serum (FCS) . The vulva tumor cell line A431 was obtained from the Human Tumor Cell Laboratory, Sloan-Kettering Institute, Rye, New York and was grown in minimal essential media (MEM) , 15% FCS, 100 µg.

997

998

BASIC FGF FROM RENAL CELL CONDITIONED MEDIA

ECGF (endothelial cell growth factor), 1% 1-glutamine and penicillin G (10 U/ml.).1 7 An established renal cell carcinoma cell line, RC29, passage number 200, was a generous gift from Dr. Joseph Michaeli, Memorial Sloan-Kettering Cancer Center, New York, New York.18 This was passaged with 10% FCS and EGF until growing in log phase, and was then grown without FCS and EGF, collected, dialysed and run on heparin-sepharose chro­ matography. The eluted fractions were tested for mitogenic activity and then run on SDS/PAGE. The cells to be tested for mitogenic activity were plated in 12 well plastic plates in concentrations of 10,000 cells per cm.2 on day 1 and incubated at 37C and 5% C0 2. Cells were refed by 0.1% FCS to achieve a quiescent state on day 2, and on day 3 50 µl. of each elution fraction was added to the well in triplicate. Following incubation with tritiated thymidine (4µCi/ml., 6.7/ mmol., NEN, for 24 hours) the cells were washed twice with Hanks' balanced salt solution. The removal of unincorporated thymidine was accomplished by the addition of cold 5% tri­ chloroacetic acid (TCA) for 30 minutes followed by a 5-minute 5% TCA wash. The cells were lysed in 0.3N NaOH for 1 hour and added to 5 cc of Hydrofluor scintillation cocktail. Counts were performed by a Packard model 3330 scintillation counter; 10% FCS and 1% FCS were used as maximum and minimum stimulation controls, respectively, so that we could gauge the relative activity of the proteins tested. The fractions that were eluted in the void (before the addition of NaCl) were collected, as were the fractions that tested positive for mitogenic activity, and the individual peaks of these protein fractions were tested for mitogenic activity. Con­ ditioned media from the established renal carcinoma cell line RC29 was collected, dialysed and stored for later use on the incorporation of tritiated thymidine. In addition, conditioned media from the HUV endothelial cells, A431 cells and 3T3 cells were also collected and stored for further use. All conditioned medias were run separately on heparin-sepharose chromatography, eluted with O to 3M. NaCl, and the individual peaks of protein were tested for mitogenic activity. Protein elution was monitored by absorbance at 280 nm., and concentrations were determined using an LKB Ul­ traspec 4050 spectrophotometer at 280 nm., utilizing the prin­ ciple of protein-dye binding method of Bradford using the kit from BioRad.19 The protein fractions were dialyzed against distilled water in dialysis tubing of 6,000 d pore size. The fractions were stored at 4C, and 50 µl. aliquots were pipetted into each well in triplicate, and the incorporation of tritiated thymidine was measured. The rest of the fractions were then frozen and lyophilized, reconstituted with 50 µl. of distilled water, and added to 25 µl. of l x sample buffer. After heating for 1 minute, they were run on 12.5% SDS/PAGE using a BioRad protein gel electrophoresis tank for 6 hours. Western blot analysis was performed using antibodies to acid FGF (aFGF) and bFGF. After SDS/PAGE, transfer to nitro­ cellulose was performed by the method of Towbin et al., 20 as modified by Burnette.21 Briefly, nitrocellulose transfer was performed at 0.6 A for 6 hours using an HSI tank (Hoeffer Scientific Instruments, San Francisco, California) and power supply. The nitrocellulose was then incubated with 10% Car­ nation Instant Milk in Tris buffer overnight at 4C and then incubated with a 1:200 dilution of antisera in 10% CIM solution for 1 hour. After a final rinse using Tris saline and 0.05% NP40, the nitrocellulose was air-dried and exposed to film (Kodak XRAS) at -70C for 12 hours. Purified aFGF, bFGF and anti­ sera to these factors from Collaborative Research were used as controls. RESULTS

The schema of the experiments are outlined in figure 1. RC29 cells were grown in normal culture media until they reached

RC29 CELLS

LJ �

-,-------- D ... /II

OTHER CELLS

CONDITIONED MEDIA

"

D BLOT

SCINTILLATION

FIG. 1. Schema of experiment. RC29 cells are passaged with 10% FCS until log phase and are then grown without FCS. After fourth passage, media is collected, purified using heparin-sepharose chroma­ tography, tested on other cell lines, and run on SDS/PAGE.

ADDITION OF EGF, FGF, AND TGFa TO THE RC29 CELL LINE Counts per minute (cpm) in 1,000's

4 0 r------------------------,

R029 Cella - EGF

- FGF

RC29 CELLS

Effil TGFa

- 0. 1'l, FOS

(iiiii] 10'l, FOS

FIG. 2. Addition of EGF, FGF and TGFa to RC29 cell line demon­ strates stimulation of cells by all three growth factors; 0.1% FCS and 10% FCS are controls.

log phase and, after the fourth passage, were placed in minimal media without FCS or exogenous growth factors. The condi­ tioned media was then collected, tested for mitogenic activity, purified and run on SDS/PAGE for Western blot analysis. Addition of EGF, FGF, and TGFa was seen to stimulate the proliferation of the RC29 cells (fig. 2). These three growth factors were also able to stimulate 3T3 cells, HUV cells and A431 cells with a similar response (data not shown). Addition of the conditioned media from the HUV endothelial cells had no stimulatory effect on either the 3T3, HUV, A431, or RC29 cell lines (data not shown). However, conditioned media from the RC29 cell line stimulated all four cell lines (fig. 3). One percent and 10% FCS were tested on all four cell lines and are displayed as the average minimal and maximal stimulation. The elution profile off the heparin-sepharose chromatogra­ phy column showed some mitogenic activity in the void (very beginning of the elution profile), but these fractions were not heparin-binding and therefore could not be purified using our technique. The majority of the mitogenic protein fractions eluted between 1.3 and 2.4 M. NaCl (fig. 4). Further purification with heparin-sepharose chromatography on the RC29 conditioned media followed by SDS/PAGE finally revealed a single band at 18,000 d (fig. 5, lane 6). There was no mitogenic activity seen with the conditioned media from the

999

BASIC FGF FROM RENAL CELL C O N D ITIONED MEDIA

ADDITION OF CONDITIONED MEDIA TO 3T3, HUV, A431, AND RC29 CELLS. Counts per minute ( cpm) in 1,000's

so �-----------------------. 26

67K 43 K

20 16 10

25K

15 0

- HUV

l!lllll- :JTS

1\111 R029

[fill A431

0 10 \II FOS

ffiilll 0.1\\\ FOS

Fm. 3. Addition of conditioned media to 3T3, HUV, A431 and RC29 cells reveals stimulation of fibroblasts, endothelial cells and epithelial cells, suggesting that media consists of numerous growth factors in addition to bFGF. Last two columns represent average mitogenic effect of all cell lines by 0.1 % FCS and 10% FCS, respectively.

ELUTION PROFILE

OF CONDITIONED MEDIA M lt@g11 n 1c Ac 11 1v1111 IOB1il ca1 0 8n s 1t yl 1 20 �__:..______:;___;._______-'-_________-, lM NIICl

10

16

20

26

30

Proteins Fractloni

31!

40

1

2

3

4

5

6

Fm. 5. SDS/PAGE after heparin-sepharose chromatography. Lanes 1, 2 and 3 represent molecular weight markers chymotrypsin A, ovalbumin and bovine serum albumin. Lanes 4 and 5 demonstrate numerous growth factors present after first and second runs. At third run, a thin faint band is seen at 18,000 d, which is consistent with molecular weight of bFGF (lane 6).

4il

FIG. 4. Elution profile of conditioned media demonstrates that it elutes between 1.3 and 2.4M. NaCl.

other cell lines, nor was any evidence of proteins seen on SDS/ PAGE. When neutralizing antisera to bFGF were added to the media, there was a 70% reduction in mitogenic activity, sug­ gesting the presence of other growth factors besides bFGF (data not shown). On Western blot analysis there was binding to the bFGF antisera using the RC29 conditioned media (fig. 6, lane 3). Using antisera to aFGF there was no binding, suggesting no crossover and good specificity of the antibodies (lane 4). Lanes 1 and 2 are chymotrypsin and control bFGF, respectively. Using serial dilutions of the purified RC29 conditioned media on 3T3 cells, the dose response curve was seen to be 1.0 ng./ ml. at half maximal stimulation, and 10 ng./ml. at maximal stimulation (fig. 7). Amounts greater than this showed a slight decrease in activity, suggesting saturation of cell membrane receptors, leading possibly to down regulation. D ISCUSSION

In the present work we have demonstrated the stimulation of an established renal cell carcinoma cell line by the addition of FGF, EGF and TGFa. In addition, we have demonstrated that the malignant renal cancer cell line RC29 produces growth factors that probably bind to the EGF receptor of the A431 cell line, (either EGF or TGFa) as well as a heparin binding growth

1

2

3

4

Fm. 6. Western blot using antisera to bFGF demonstrates binding to purified bFGF (control) in lane 2, as well as binding to purified heparin binding growth factor obtained from RC29 conditioned media, seen in lane 3. Lane 4 is aFGF, which shows no binding to bFGF antisera, thus demonstrating excellent specificity.

1000

BASIC FGF FROM RENAL CELL CONDITIONED MEDIA

DOSE RESPONSE CURVE counts per minute ( opm) ( T housands)

40 .-------------------------,

30

--····---------··---- ---··-·--··--··----··----------··---·-- ---

20 -----··----·----·-·--·-------·------ ---r=--c::-·---·· -----

---·--··--·

-·----

O L-------------------0.001

0.01

0. 1

1.0

� Conditoned Media

10

ng/ml

100

B 10% FCS

1000

llEJl 0.1% FCS

FIG. 7. Dose response curve of purified conditioned media demon­ strates half-maximal stimulation at 1.0 ng./ml. and maximal stimula­ tion at 10 ng./ml. Further addition of growth factor showed a slight decrease in stimulation, suggesting saturation of receptors and possible down regulation.

factor that binds to the FGF receptor on 3T3 fibroblasts and HUV cells. This corroborates our previous work, in which we identified a heparin binding angiogenic growth factor from renal cell carcinoma and subsequently demonstrated an in­ creased expression of TGFa and EGF receptor expression at both the mRNA and protein level. 14 Since, in the present study, the RC29 renal carcinoma cell line is itself stimulated by the addition of EGF and FGF, and also produces its own growth factors, this suggests a possible autocrine mechanism by which this tumor cell line can proliferate, a theory proposed by Sporn and Todaro. 15 Using cell lines rather than pathological tumor specimens has obvious advantages, chiefly in that homogenity is ascer­ tained. Pathological specimens suffer from the fact that there are differing concentrations of red blood cells, white blood cells, endothelial cells and so forth which may account for the given findings. On the other hand, there are also disadvantages in using established cell lines, since there are acquired character­ istics as well as lost characteristics from the original tumor cell line. Thus, our findings may not necessarily apply to renal cell carcinoma in vivo. Although there are numerous renal carci­ noma cell lines available for study, we chose the RC29 cell line because of easy access (same institution), easy passage and easy growth ability. Other cell lines will be looked at in subsequent experiments. Polymerase chain reaction (PCR)/in-situ hybrid­ ization studies may improve our understanding of growth fac­ tors from original tumor tissues in the future. 22 The fact that addition of the RC29 conditioned media was able to stimulate epithelial cells as well as fibroblasts suggests that the purified sample still contained other growth factors besides bFGF. This can be evidenced by the numerous bands seen in the first few purifications over heparin-sepharose (fig. 5, lanes 4, 5). Further evidence suggesting the presence of other growth factors was demonstrated when neutralizing antibody to bFGF was added to the conditioned media. This resulted in a 70% reduction in mitogenic activity. If only bFGF were present, one would expect a complete cessation of mitogenic activity. Fibroblast growth factors are a family of polypeptides that exist in both acidic and basic forms, are angiogenic and are found in many tissues. The acidic form usually has a lower affinity to heparin and elutes with 0.9 to 1.1 M. NaCl compared

with 1.3 to 1.8M. NaCl for the basic FGF. The acidic form is also found in the central nervous system tissues, whereas the basic form is found in most other tissues. 2 Although basic and acidic FGF have a 60% crossover homol­ ogy, these antibodies have been used in previous experiments looking at FGF from kidney, prostate and omentum, and show good specificity without cross-reactivity (data not shown). Thus, this confirms that the faint band we see in the last run over heparin-sepharose chromatography is indeed bFGF (fig. 5). The fact that we never demonstrated cross-reactivity of the antisera on Western blot analysis further verifies the specificity of these antibodies. The dose effect curve of the purified protein demonstrated a half maximal stimulation at 1.0 ng./ml. and maximal stimula­ tion at 10 ng./ml., with only a slight decrease in activity with additional protein. This suggests a possible saturation of growth factor receptors with down regulation. It is of interest that epithelial cells are usually not thought of as target tissues for heparin binding growth factors, but recent reviews have reported that increasing numbers of epi­ thelial cells are now considered mitogenically responsive to these substances. 23 · 24 As mentioned previously, a noted feature of renal cell carci­ noma is its rich vascularity. Folkman described the concept of angiogenesis, which is necessary for tumors to grow, invade and eventually metastasize.3 -5 The kidney is rich in angiogenic growth factor activity, and this may play a role in its aggressive nature. Baird et al. demonstrated the presence of an angiogenic polypeptide in bovine kidney, which is similar, if not identical, to bFGF. 25 In addition, Gomella et al. demonstrated the expres­ sion of transforming growth factor alpha in human adult kid­ ney. 26 Singletary et al. looked at the stimulation of several malig­ nant human tumor specimens by the addition of EGF. These included melanomas, sarcomas, lung, gynecological, breast, renal cell and gastrointestinal tumors, and found 50% in 81% of the 186 tumors and by over 100% in 54% . 23 The stimulation of the renal cell carcinoma cell line by EGF and the increased EGF receptor expression we reported previ­ ously may also have some therapeutic implications, since cer­ tain cell lines that overexpress the EGF receptor can be inhib­ ited by monoclonal antibodies to the EGFR when the tumors are grown in nude mice. 2 7· 28 The concept of developing a mono­ clonal antibody to the growth factors and/or their receptors produced by these cell lines and resulting in diminished growth holds promise for the future in cancer therapy. More promise may be seen with the development of antisense nucleotides. 29 In summary, we have demonstrated the stimulation of a renal cell carcinoma cell line by the addition of exogenous FGF, EGF and TGFa. We have also demonstrated that the conditioned media of this cell line stimulates the growth of 3T3 fibroblasts, A431 cancer cells and human umbilical vein endothelial cells as well as the renal cell cancer cells themselves. This finding suggests that a possible autocrine mechanism may exist by which these cells can self-proliferate in vitro, and that this mechanism could play a role in the unrestricted cancer cell growth and rich vascular network development seen in renal cell carcinoma in vivo. REFERENCES 1. Mydlo, J. H., Heston, W. D. W. and Fair, W. R.: Characterization of a heparin binding growth factor from adenocarcinoma of the kidney. J. Urol., 140: 1575, 1988. 2. Gospodarowicz, D., Neufield, G. and Schweiger, L.: Fibroblast growth factor: a review. Mol. Cell Endocrinol., 46: 187, 1986. 3. Folkman, J.: The vascularization of tumors. Sci. Am., 238: 58, 1976. 4. Folkman, J. and Cotran, R.: Relation of vascular proliferation to tumor growth. Int. Rev. Exp. Pathol., 16: 207, 1976. 5. Folkman, J.: How is blood vessel growth regulated in normal and

BASIC F G F FROlvi RENAL CELL C O NDITIONED MEDIA

6. 7. 8. 9.

10.

11.

12. 13. 14.

15. 16.

17.

18.

neoplastic tissue ? The HA Clowes Memorial Award Lecture. Cancer Res., 46: 467, 1986. Mydlo, J. H. and B ard, R. H.: An analysis of papillary adenocar­ cinoma of the kidney. Urology, 30: 529, 1987. Goustin, A. S., Leof, E. B., Shipley, G. D. and Moses, M. L.: Growth factors and cancer. Cancer Res., 46: 1015, 1986. Derynck, R., Roberts, A. B., Winkler, M. E., Chen, E. Y. and Goedde!, D. V.: Human transforming growth factor alpha: pre­ cursor and structure expression in E. Coli. Cell, 3 8 : 287, 1984. Rosenthal, A., Lindquist, P . B., Bringman, T. S., Goedde!, D. V. and Derynck, R.: Expression in rat fibroblasts of a human transforming growth factor alpha results in transfo rmation. Cell, 46: 301, 1987. Coffey, R. J., Derynck, R., Wilcox, J. N., Bringman, T. S., Goustin, A. S., Moses, M. L. and Pittelkow, M. R.: Production and autoinduction of transforming growth factor alpha in human keratinocytes. Nature, 3 2 8 : 817, 1987. Derynck, R., Goedde!, D . V., Ullrich, A., Gutterman, J. U., Wil­ liams, R. D., Bringman, T. S . and Berger, W. H.: Synthesis of messenger RNA's for transforming growth factors alpha and beta and the epidermal growth factor receptor by human tumors. Cancer Res., 47: 707, 1987. Cohen, S.: The stimulation of epidermal proliferation by a specific protein (EGF). D ev. Biol., 1 2 : 394, 1965. Mydlo, J. H., Bard, R. H . and Freed, S. Z.: Detection of tumor angiogenesis factor in adenocarcinoma of the kidney. Urology, 2 7 : 446, 1986. Mydlo, J. H., Michaeli, J., Cordon-Cardo, C., Goldenberg, A. S., Heston, W. D. W . and Fair, W. R.: Expression of transforming growth factor alpha and epidermal growth factor receptor mRNA in neoplastic and non-neoplastic human kidney tissue. Cancer Res., 49: 3407, 1989. Sporn, M. B. and Todaro, G. J.: Autosecretion and malignant transformation of cells. N. Engl. J. Med., 3 0 3 : 878, 1980. Van de Viger, M., Kumar, R. and Mendelsohn, J.: Ligand-induced activation of A431 cell epidermal growth factor receptors occurs primarily by an autocrine pathway that acts upon receptors on the surface rather than intracellularly. J. Biol. Chem., 2 5 : 7503, 1991. Sunada, M., Peacock, J. and Mendelsohn, J.: Ligand-induced in­ ternalization of epidermal growth factor receptors by A431 cells decreases at high cell densities in culture. Growth Factors, 5: 45, 1991. Ebert, T., Bander, N., Finstad, C., Ramsawak, R. D . and Old, L.

19. 20.

21.

22.

23.

24.

25.

26. 27.

28.

29.

1001

J.: Establishment and characterization of human renal cancer and normal kidney cell lines. Cancer Res., 50: 5531, 1990. Bradford, M. A.: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding. Anal. Biochem., 7 2 : 248, 1976. Towbin, H., Staehlin, T. and Gordon, J.: Electrophoretic transfer of proteins from polyacrylarnide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA., 76: 4350, 1979. Burnette, W. N.: Western blotting: electrophoretic transfer of protein from sodium dodecyl sulfate polyacrylamide gels to un­ modified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem., 1 12 : 195, 1981. Baldini, A., Ross, M., Nizetic, D., Vatcheva, R., Lindsay, E., Leh­ rach, A. and Siniscalco, M.: Chromosomal assignment of human YAC clones by fluorescence in-situ hybridization: use of single­ yeast-colony PCR and multiple labeling. Genomics, 14: 181, 1992. Singletary, S. E., Baker, F. L., Spitzer, G., Tucker, S. L., Tomasovic, B., Brock, W. A., Ajani, J. A. and Kelly, A. M.: Biological effects of epidermal growth factor on the in vitro growth of human tumors. Cancer Res., 4 7: 403, 1987. Hirata, Y., Uchihashi, M., Fujuta, T., Matsukura, S., Motoyama, T., Kaku, S. and Koshimisu, K.: Characteristics of epidermal growth factor on human tumor cell lines. Endocrinol. Jpn., 3 0 : 601, 1983. Baird, A., Esch, F., Bohlen, P., Ling, N. and Gospodarowicz, D.: Isolation and partial purification of an endothelial cell growth factor from bovine kidney: homology with basic fibroblast growth factor. Regul. Pept., 1 2 : 201 , 1985. Gomella, L. G., Sargent, E. R., Wade, T. P., Anglard, P., Linehan, W. M. and Kasid, A.: Expression of transforming growth factor alpha in human adult kidney. Cancer Res., 46: 6972, 1989. Masui, H., Moroyama, T. and Mendelsohn, J.: Mechanism of antitumor activity in mice for anti-epidermal growth factor receptor monoclonal antibodies with different isotypes. Cancer Res., 46: 5592, 1986. Masui, H., Kawamoto, T., Sato, J . D., Wolf, B., Sato, B. and Mendelsohn, J.: Growth inhibition of human tumor cells in athymic mice by anti-epidermal growth factor receptor monoclo­ nal antibodies. Cancer Res., 44: 1002, 1984. Tortora, G., Yokozaki, H., Pepe, S., Claire, T. and Cho-chung, Y.: Differentiation of HL-60 leukemia by type I regulatory subunit anti-sense oligodeoxynucleotide of cAMP-dependent protein ki­ nase. Proc. Natl. Acad. Sci. USA, 88: 2011, 1991.