1983;43:1972-1979. Published online May 1, 1983.Cancer Res Jaroslava Halper and Harold L. Moses Epithelial Tissue-derived Growth Factor-like Polypeptides
Updated Version http://cancerres.aacrjournals.org/content/43/5/1972
Access the most recent version of this article at:
Departments of Cell Biology and Pathology, Mayo Clinic, Foundation and Medical School, Rochester, Minnesota 55905
ABSTRACT
SW-13 cells, derived from a human adenocarcinoma of theadrenal cortex, formed only a few small colonies when suspended in soft agar at low cell densities. The number and sizeof colonies increased dramatically following stimulation withserum-free medium conditioned by SW-13 cells, indicating the
possibility of autostimulation in these malignant cells. Evidenceis presented suggesting that SW-13 cells form progressivelygrowing soft agar colonies upon stimulation by epithelial tissue-derived growth factor-like polypeptides. Both acid-ethanol ex
tracts and conditioned media from three human carcinoma celllines (A431, D562, and A549) caused similar increases in colonynumber and size of SW-13 cells. Extracts from 26 of 32 freshlyexcised human carcinomas and five freshly excised nonneoplas-
tic human kidneys and one human lung stimulated soft agargrowth of SW-13 cells as well. None of the nine extracts fromnonepithelial human solid malignant tumors stimulated SW-13
cells. However, a benign nonepithelial tumor (uterine leiomyoma)caused a low level of soft agar growth of SW-13 cells. Cell
extract from A204 human sarcoma cells and both conditionedmedium and acid-ethanol cell extract from A375 human melanoma cells lacked SW-13 activity, whereas medium conditionedby A204 cells stimulated soft agar growth of SW-13 cells.
Chemical and physical treatment data indicated that the epithelialtissue-derived growth factor-like substances are acid- and heat-
stable polypeptides with disulfide bonds. The major peak of thisactivity had an apparent molecular weight of 20,000 to 22,000and was clearly separable from transforming growth factorsreported previously which stimulate colony formation by non-transformed mouse AKR-2B and rat NRK cells. The major peaksof SW-13, NRK, and AKR-2B activity could be separated byhigh-performance liquid chromatography. This SW-13 activityinduced irreversible anchorage-independent growth of SW-13cells and an increase in DNA synthesis as measured by [3H]-
thymidine incorporation.
INTRODUCTION
The production of a growth factor(s) by malignant cells andtheir capability of responding to these factors may be responsiblefor the decreased requirement of malignant cells for exogenousfactors for growth in vitro (10). This phenomenon has been called"autocrine secretion" and "autostimulation" (23). The first en
dogenous factor identified as capable of inducing soft agargrowth in anchorage-dependent cells was sarcoma growth fac-
' This investigation was supported by USPHS Grant CA27217 awarded by the
National Cancer Institute, Department of Health and Human Services. Presentedin part at the 66th Annual Meeting of the Federation of American Societies forExperimental Biology in New Orleans, April 1982 (8).
2 Recipient of NIH Training Grant AM 07147.3To whom requests for reprints should be addressed.
Received August 16. 1982; accepted January 21,1983.
tor, a polypeptide with an apparent molecular weight of 10,000isolated by DeLarco and Todaro (6) from medium conditioned bymurine sarcoma virus-transformed mouse 3T3 cells. Since then,numerous similar factors, generally called TGFs," have been
recognized and isolated from conditioned medium of Kirstenmurine sarcoma virus- and Abelson murine leukemia virus-trans
formed cells (10, 28), chemically transformed murine cells (13),and human cell lines (12). They were also found intracellularly inboth virally and chemically transformed rodent cells (20), in solidhuman malignant neoplasms (15), in nonneoplastic human adulttissues (15), and nonneoplastic mouse tissues (both adult andembryonic) (17,19, 27). TGFs are acid- and heat-stable polypeptides of low molecular weight (6,000 to 23,000). They are mito-genie and reversibly confer the transformed phenotype on non-
transformed cells (6, 13). They induce nontransformed mouseand rat cells to become anchorage independent (6, 13, 20), anin vitro characteristic which correlates well with tumorigenicity invivo (9).
Recent studies from our laboratory have shown that differentTGFs present in medium conditioned by chemically transformedmouse cells (26), in normal human platelets (4), in human solidneoplastic and nonneoplastic tissues (15), and in mouse embryos(17) can be separated on the basis of their indicator cell specificity. Using nontransformed mouse AKR-2B, rat NRK, and EGF-
receptorless NR6 cells as indicators, several peaks of activitywhich differ in their apparent molecular weight and which areindicator cell specific can be detected when acid-ethanol extracts
from a particular tissue are analyzed by gel exclusion chromatography.
Human malignant cells have not been used extensively asindicator cells for detecting growth factor-like substances pro
duced by neoplastic cells. It has not been determined withcertainty that autostimulation occurs in human cancer cell linesand that, if it does, it plays any significant role in malignantgrowth. In the present study, SW-13 cells, derived from a human
small cell adenocarcinoma of the adrenal cortex, were examinedfor the production of and response to growth factor-like substances. SW-13 cells did not grow well in soft agar at low cell
densities but formed progressively growing colonies in soft agarupon stimulation by medium conditioned by SW-13 cells. Poly
peptides extracted from human epithelial malignant cell lines,from medium conditioned by these cells, from a variety of freshlyexcised human epithelial solid neoplasms and nonneoplastictissues, but not from nonepithelial cancers induce anchorage-independent growth of SW-13 cells. These SW-13 cell-stimulating factors are separable by molecular sieve and high-performance liquid chromatography from TGFs inducing anchorage-independent growth of nontransformed mouse AKR-2B and rat
Cell Culture. SW-13 cell line, derived from a human cell adenocarci-
noma of the adrenal, was obtained from the American Type CultureCollection (Rockville, Md.) and has been described previously (11). Themouse AKR-2B and rat NRK cells have a nontransformed morphology,
are nontumorigenic in nude mice, and are anchorage dependent (6,14).A431 human epidermoid carcinoma, A549 human bronchioloalveolarcarcinoma, A204 human rhabdomyosarcoma, and A375 human melanoma cell lines were a generous gift from Dr. G. J. Todaro and Dr. J. E.DeLarco and were described previously (7). The D562 human epidermoidcarcinoma cell line was kindly provided by Dr. R. E. Scott who obtainedit from the American Type Culture Collection. This line was describedpreviously (16). All cell lines were grown in McCoy's Medium 5a (Grand
Island Biological Co., Grand Island, N. Y.) supplemented with 10% FBS(Reheis Chemical Co., Phoenix, Ariz.). All cell lines were used for experiments within 10 to 15 passages of the frozen stock from which theywere replaced periodically. Cells were regularly examined after Hoechst33258 staining to ensure they remain free of Mycoplasmes. Cells wereincubated at 37°in a humidified atmosphere of 5% CO2 and 95% air.
Collection of Serum-free Conditioned Medium. Cells were grown in490-sq cm roller bottles (Corning Glassworks, Corning, N. Y.). Approxi
mately 50 roller bottles of cells were used for each cell line examined.Cells were grown in 25 ml of McCoy's Medium 5a supplemented with
10% FBS. When 90% confluent, the bottles were rinsed 3 times for 1 hreach with serum-free Dulbecco's modification of Eagle's medium (Grand
Island Biological Co.) supplemented with penicillin G (100 units/ml) andstreptomycin (100 ng/m\). Afterwards, the cells were placed in serum-free Dulbecco's modification of Eagle's medium for 24 hr, and 1.25 to
1.4 liters of medium were then collected. The cells were scraped with arubber policeman from the walls of the bottles and frozen at -70° until
used as described below. The yield of cells ranged from 1.7 to 22 g withan average of 8.3 g. The serum-free conditioned medium after centrifu-
gation at 17,680 x g for 3 hr was extensively dialyzed against 1% aceticacid (5 changes of 50 volumes at 24-hr intervals) in Spectrapor 3 dialysis
tubing with a cutoff of M, 3,500 (Spectrum Medical Industries No.132720), lyophilized to dryness, and stored at -20° until used.
Tissue and Cell Extraction. Freshly excised human solid neoplasticand nonneoplastic tissues and human malignant cells grown in culturewere extracted by a modification (17) of the acid-ethanol procedure
reported by Roberts ef a/. (20).Soft Agar Colony Stimulation Assay. SW-13 cells and, in some
experiments, clones of nontransformed mouse AKR-2B and rat NRKcells were used as indicators for growth-stimulating activity in soft agar.
Solidified base layers of 1 ml of 0.8% agarose (Sea Plaque; MarineColloids Division, FMC Corp., Rockland, Maine) in McCoy's Medium 5a
with 10% FBS were overlaid with 1 ml of upper layer of 0.4% agarosein McCoy's Medium 5a with 10% FBS, 7.5 x 103 indicator cells, and
appropriate quantities of sample in 35-mm Petri dishes (Falcon No. 3001 ;
Falcon Labware Division, Oxnard, Calif.). The dishes were incubated at37°in a humidified atmosphere of 5% CO2. SW-13 cells were analyzed
at 14 to 21 days, and AKR-2B and NRK cells were analyzed at 7 to 14
days with a Bausch and Lomb Omnicon Feature Analysis Stem Model IIor a Quantimet 800 image analyzer (Can-.bridge Instruments, Ltd., Mon-
sey, N. Y.). Colonies of a diameter greater than 60 Mm (in the case ofSW-13 and NRK cells) or 50 urn (in the case of AKR-2B and occasionally
NRK cells) were counted. The background (unstimulated) counts obtained with SW-13 cells generally ranged from 10 to 30 colonies.
Molecular Sieve Chromatography. Twenty to 40 mg of a lyophilizedacid-ethanol extract were dissolved in 3 ml of 1 M acetic acid and clarifiedat 1000 x g for 30 min. The supernatant was then applied to a 1.6- x87-cm Bio-Gel P-60 column equilibrated with 1 M acetic acid. The columnwas eluted at 22°with 1 M acetic acid at a flow rate of 7.5 ml/hr with
3.0 to 3.5-ml fractions being collected. The fractions were lyophilized todryness and stored at -20° until used. On several occasions, aliquots
of every second fraction were used for protein determination with a dye-
binding assay (3).Physical and Chemical Treatment. One-mg aliquots of protein from
either the crude acid-ethanol extract or from pooled material from Bio-Gel P-60 fractions were dissolved in 1 ml of Dulbecco's PBS, pH 7.2.One aliquot was heated to 56° in a water bath for 30 min. Another
aliquot was subjected to a 1-hr treatment with a final concentration of
0.065 M dithiothreitol in 0.1 M NH4HCO3 in PBS, pH 7.2, at roomtemperature to test the effect of a reducing agent. To test trypsinsensitivity, an aliquot was incubated with 100 ^g of trypsin (crystallinetrypsin, type 3; Worthington Biochemicals, Freehold, N. J.) for 3 hr at37°,to which 200 ng of soybean trypsin inhibitor (Sigma Chemical Co.,
St. Louis, Mo.) were added at the end of the incubation time. As acontrol, 200 fig of soybean trypsin inhibitor were preincubated with 100tig of trypsin for 2 hr at 22° and then the sample was added andincubated for an additional 3 hr at 37°.After each treatment, the volume
of each aliquot was brought to 4.0 ml with PBS, dialyzed against 1%acetic acid (3 changes of 50 volumes), and lyophilized to dryness. Then,the samples were dissolved in McCoy's Medium 5a with 10% FBS and
tested for colony-stimulating activity in the soft agar growth assay.
Hormones and Growth Factors. The following hormones and growthfactors were added to the soft agar assay to test for SW-13 colony-stimulating activity: insulin; hydrocortisone; dexamethasone; and lysine-vasopressin (all from Sigma Chemical Co.); and FGF and multiplication-
stimulating activity (both from Collaborative Research, Inc., Waltham,Mass.). EGF was purified from male mouse salivary glands by the methodof Savage and Cohen (21). TGF from normal human platelets wasobtained from C. Childs from our laboratory (4). Mouse embryo TGF wasprovided by J. Proper from our laboratory (17). TPA was obtained fromthe IIT Research Corp., Chicago, III., through the National Cancer InstituteCarcinogenesis Research Program and dissolved in acetone at concentrations 1 mg/ml. When used in the soft agar assay, up to final maximalconcentration of TPA (1 ^g/ml medium), only 0.1 % acetone was present.
Stimulation of DMA Synthesis in Quiescent Cells. A modification ofthe [3H]thymidine incorporation method described previously was used(14). The level of [3H]thymidine incorporation was determined in SW-13
cells as they reached their saturation density. Resting cells were stimulated to grow by changing to fresh medium containing either 10% FBSor 10% FBS with 200 ^g/ml of acid-ethanol extract from a renal cellcarcinoma. [3H]Thymidine incorporation was determined 24, 48, 72, and
96 hr after medium change. Cells were pulsed for 60 min with 1.0 nC\[mef/?y/-3H]thymidine (6.7 /¿Ci/mmo!; New England Nuclear, Boston,
Mass.) per ml medium, and the incorporation into acid-precipitable ma
terial was determined as described previously (14).High-Performance Liquid Chromatography. A 1 x 25-cm Altex Ul-
trasphere ODS column (Altex, Berkeley, Calif.) with a 4.6-mm x 4.5-cm
precolumn was used. A sample containing 1 to 4 mg protein in 200 p\0.025% trifluoroacetic acid in water (v/v) was injected and eluted with alinear propanol gradient in 0.025% trifluoroacetic acid (v/v). A lineargradient of 20% n-propyl alcohol to 50% n-propyl alcohol over 75 min
was used. The effluent was monitored at 254 nm. The flow rate was 1ml/min at room temperature. Three-mi fractions were collected. Aliquotswere lyophilized and stored at -20° until further used.
RESULTS
Stimulation of SW-13 Cell Colony Formation in Soft Agarby Medium Conditioned by SW-13 Cells. SW-13 cells do not
grow well in soft agar when plated at low cell densities, such as7.5 x 103 cells/ml, and start forming large colonies only atdensities higher than 2.5 x 104 cells/ml (Chart 1). Upon stimula
tion of SW-13 cells at low cell density by serum-free mediumconditioned by SW-13 cells, the number and size of colonies
increased dramatically (Table 1). The appearance of the stimulated and unstimulated cells was identical to that illustrated inFig. 1, A and B. Cell extracts were not used because of technical
difficulties due to poor adaptation of SW-13 cells to roller bottlesand serum-free medium.
Stimulation of SW-13 Colony Formation by Conditioned
Media and Cell Extracts from Human Neoplastia Cell Lines.Several cell lines derived from human malignant neoplasms wereexamined for the presence of intra- and extracellular SW-13colony-stimulating activity. As shown in Table 1, both acid-ethanol cell extracts and serum-free conditioned media from 3
carcinoma cell lines (D562, A431, and A549) and serum-freemedium conditioned by a sarcoma cell line A204 had markedsoft agar colony-stimulating activity on SW-13 cells. The re-
Table 1Stimulation of SW-13 cell colony formation in soft agar by conditioned media and
cell extracts from various human neoplastic cell lines
Soft agar colonies(»60-iim diameter)
at 1000 ¿igprotein/ml"Human cell line
10' 10*
Ce//s / ml
10*
Chart 1. Soft agar colony formation as a function of SW-13 cell density. Softagar assays were performed with increasing concentrations of cells. The numberof colonies per dish greater than 60 ¿imin diameter was quantitated after 14 daysof incubation.
SW-13 (small cell carcinoma of adrenal)
Conditioned mediumA431 (epidermoid carcinoma of vulva)
Conditioned mediumCell extract
D562 (epidermoid carcinoma of nasopharynx)Conditioned mediumCell extract
A549 (bronchioalveolar carcinoma of lung)Conditioned mediumCell extract
Fig. 1. TGF stimulation of growth in soft agar in SW-13 cells. A, untreated SW-13 cells after 14 days in soft agar; B, SW-13 cells treated with acid-ethanol extractfrom human nonneoplastic kidney after 14 days in soft agar. Phase contrast, x 150.
spense was dose dependent (data not shown). However, theA204 cell extract and both the cell extract and serum-free
conditioned medium from the A375 melanoma cell line did notstimulate soft agar growth of SW-13 cells.
Effects of Acid-Ethanol Extracts from Human Solid Cancerson Soft Agar Growth of SW-13 Cells. Forty freshly excised
human solid cancers and one freshly excised human solid benigntumor were acid-ethanol extracted, dialyzed, lyophilized, dissolved in medium, and tested for SW-13 colony-stimulatingactivity. As shown in Table 2, thirty-two of these 41 neoplasmswere carcinomas. Twenty-six of the 32 carcinomas (80%) stimulated colony formation of SW-13 cells in soft agar. Colonie,
gastric, and renal carcinomas appeared to have the most potentcolony-stimulating activity, while lung, ovarian, and squamous
cell carcinomas from sites other than lung had moderate activity,and breast carcinomas had the least activity, causing only mildor no increase in colony number. However, all of these 5 breastcarcinomas tested were scirrhous with a high content of stromaand a low content of carcinoma cells. The response of SW-13cells was dose dependent, since the number and size of coloniesincreased with increasing protein concentration (Chart 2). The 8remaining cancers were tumors distinct from carcinomas andincluded 4 sarcomas, 2 melanomas, one lymphoma, and onechordoma, none of which stimulated soft agar growth of SW-13carcinoma cells (Table 2). In addition to these tumors, one benignneoplasm, a uterine leiomyoma, was examined for SW-13 colony-stimulating activity as well and was found to have low levels
of activity (Table 2).Effects of Acid-Ethanol Extracts from Nonneoplastic Hu
man Kidney and Lung. Portions from nonneoplastic humankidneys and a nonneoplastic human lung were extracted withacid-ethanol and tested for SW-13 cell colony-stimulating activity
(Fig. 1, A and B). The results in Table 3 show that all 5 kidneyscontained considerable activity; the lung extract was less active.An acid-ethanol extract from mouse embryos (17) was effectivein stimulating SW-13 colony formation as well.
Physical and Chemical Treatment of Cell Extracts andConditioned Media. Samples representing all 3 active groups(I.e., human carcinoma cell lines, freshly excised carcinomas,
and freshly excised epithelial nonneoplastic tissues) were subjected to physical and chemical treatment described under "Materials and Methods." Their activity was retained or somewhatenhanced by a 30-min exposure to 56°.Trypsin treatment de
creased their activity, whereas the activity was fully retained withtrypsin preincubated with soybean trypsin inhibitor. Dithiothreitol,a reducing agent, abolished their activity completely. Theseresults indicate that these substances with SW-13 colony-stimulating activity are heat-stable polypeptides requiring disulfide
bonds for their action.Effects of Growth Factors, Hormones, and TPA on SW-13
Colony Formation. Only the preparation of FGF from all of thegrowth factors and hormones tested stimulated soft agar growthof SW-13 cells (Table 4). Since the FGF used in these experi-
10
10 10»
fig Protein / ml
10»
Chart 2. Stimulation of soft agar growth of SW-13 cells by acid-ethanol extractsfrom carcinoma of colon (O), ovary (•).and stomach (A). The response of SW-13cells was dose dependent, since the number and size of colonies increased withincreasing protein concentration. Concentrations of 10 to 1000 ¡igprotein per mlpresent in acid-ethanol extracts were used.
Table 2Stimulation of SW-13 cell colony formation in soft agar by acid-ethanol extracts from freshly excised human
from normal human platelets1085"548;Õ°ao"0e°/°dcf' Background subtracted.6 At 1000 ng FGF per ml.c At 3.3 ng which gave maximal number of colonies." No activity at 1 to 1000 ng/ml.8 No activity at 1 to 1000 ng/ml.' No activity at 1 to 100 »ig/ml.
ments was probably not completely pure, the possibility thatsome contaminating compound was responsible for stimulatingsoft agar growth cannot be excluded. However, the dose response obtained with 50% maximal stimulation at less than 10ng/ml and some stimulation with FGF concentrations as low as1 ng/ml makes this latter possibility unlikely. TPA was also foundto stimulate SW-13 colony formation with an optimal concentra
tion of about 3 ng/ml; higher concentrations gave less colonystimulation, but the number of colonies remained above thecontrol level (Chart 3).
Irreversible Effects of Acid-Ethanol Extracts. When coloniesof SW-13 cells stimulated by an acid-ethanol extract were trans
ferred from soft agar into monolayer culture for several passagesand then back into soft agar, SW-13 cells started forming nu
merous large colonies even without addition of any stimulatoryextracts. This colony formation occurred with high efficiency,with at least 50% of plated cells forming colonies (data notshown). Similarly, the capability for anchorage-independentgrowth was irreversibly induced when the acid-ethanol extractfrom a renal cell carcinoma was added to SW-13 in monolayer
culture for 7 to 10 days at concentrations of 200 u.g/m\.Stimulation of DNA Synthesis in Quiescent SW-13 Cells.
Changing to fresh medium with 10% FBS and 200 ^g/ml of acid-ethanol extract from a renal cell carcinoma resulted in markedstimulation of thymidine incorporation into quiescent SW-13 cells.If fresh medium containing only 10% FBS was added, thisincrease in thymidine incorporation was much lower (Chart 4).This marked increase in thymidine incorporation lasted up to 3days after the medium was changed. Then, the cells with orwithout the added acid-ethanol extract returned rapidly into the
resting state.
10 10*ng / ml
103
Charts. Stimulation of soft agar growth of SW-13 cells by FGF (O) and TPA(•).Both FGF and TPA were used in concentrations from 1 to 1000 ng/ml.
auS
20I
10u„,-•%ft \• j' •V/^./N—X56789
Days
Chart 4. Stimulation of DNA synthesis in quiescent SW-13 cells. Cells wereseeded in 35-sq mm dishes, and the medium was changed 5 days later. Cells werestimulated to synthesize DNA by changing to fresh McCoy's Medium 5a containing
10% FBS without (•)or with 200 ^g/ml of acid-ethanol extract from a renal cellcarcinoma (O). This was followed by a 1-hr pulse with [3H]thymidine every 24 hr
for 3 days.
Bio-Gel P-60 Chromatography. Acid-ethanol extracts from
A431 cells, 2 colonie carcinomas, a squamous cell carcinoma, arenal cell carcinoma, a breast carcinoma, and a nonneoplastichuman kidney were further analyzed by gel exclusion chroma-tography on a Bio-Gel P-60 column. Soft agar growth-stimulatingactivity was followed by using SW-13 cells, together with non-transformed mouse AKR-2B and rat NRK cells as indicator cells.As shown in Charts 5 and 6, SW-13 activity eluted after the
major portion of protein and resolved into several peaks. Themajor peak consistently had an apparent molecular weight of20,000 to 22,000 and was followed by one to 3 minor peaks of
Chart 5. Bio-Gel P-60 chromatography of acid-ethanolextract from a renal cellcarcinoma. The column was loaded with 38 mg of the acetic acid-soluble proteinand eluted with 1 Macetic acid. An aliquot (one-halffraction for SW-13 cells, one-third fraction for AKR-2B and NRK cells) of every second fraction was tested forstimulation of colony formation (•)using SW-13 (A), AKR-2B (B), and NRK cells(C) as indicator cells. Protein was determined on every second fraction using thedye-binding assay. The markers were: carbonic anhydrase, M, 29,000; RNase,M, 14,000; and insulin,M, 6,000.
lower apparent molecular weight between 7,000 and 12,000. Inall instances, AKR-2B activity resolved characteristically into a
peak of apparent molecular weight of 14,000. This peak wasclearly separable from SW-13 peaks (Chart 5). Sometimes, as inthe case of the nonneoplastic kidney, AKR-2B colony-stimulating
activity was also present in an additional peak of higher molecularweight partially overlapping with the major SW-13 peak (Chart
6). NRK activity resolved into one or 2 peaks and followed 2patterns. The renal cell carcinoma contained one peak of NRKactivity, which was separable from SW-13 activity but identicalwith the AKR-2B cells (Chart 5). Nonneoplastic kidney resolved
into 2 peaks of NRK activity, the major one overlapping with themajor peak of SW-13 activity and a minor one of apparent
molecular weight of 8000.High-Pertormance Liquid Chromatography. The major peak
of SW-13 activity was separable from AKR-2B and NRK activitywhen active fractions from the P-60 chromatography werepooled and further analyzed by high-performance liquid chro
matography using a linear gradient of n-propyl alcohol (Chart 7).SW-13 activity eluted between 22 and 26% n-propyl alcohol,whereas the major peaks of AKR-2B and NRK colony-stimulatingactivity appeared at 30 to 32% n-propyl alcohol. The NRK activity
was markedly enhanced when EGF was added to the assay ata concentration of 2 ng/ml.
X DISCUSSION
The results presented in this study suggest that SW-13 humancarcinoma cells exhibit autostimulatory activity. SW-13 cells donot grow well in soft agar when plated at low cell densities;however, the number and size of SW-13 colonies increasedramatically when an extract of medium conditioned by SW-13cells is added to the soft agar assay. This indicates that SW-13cells release into serum-free medium a factor(s) which markedlyenhances the growth in soft agar of SW-13 cells. This phenomenon may explain why colony formation absent at low celldensities appears at high SW-13 cell densities where the re
leased endogenous factor(s) would reach higher concentrations.This is also in accordance with the findings of Todaro et al. (25)that a critical concentration of TGFs released by tumor cells isnecessary for anchorage-independent growth. Such autocrine
activity of malignant cells could explain their relative independence from growth factors and hormones present in tissue fluids(23).
The findings presented in this study indicate that SW-13 cellschiefly respond to epithelial tissue-derived factor(s). Twenty-sixof 32 (80%) carcinoma extracts stimulated anchorage-independent growth of SW-13 cells, whereas none of the 9 noncarcinoma
malignant neoplasms (mostly sarcomas and melanomas) stimulated SW-13 cells. The only exception was a uterine leiomyoma,
29K 14K 6K
o1C•tiüSoft
Agar201510n30to1015105A.
SW13 CtlltA/
I
i ALAA.-A. V,s--.,.,-v.B.
ÄKR-2BCells,\.\!"\r
VC.Nf?K Ce/(*I*L
lh y\...A20 30 40 50
Fraction number60
Chart 6. Bio-Gel P-60 chromatography of acid-ethanolextract from a nonneoplastic human kidney. One-quarter fraction of a 20-mg protein concentrate wastested for stimulation of colony formation using SW-13 cells (A)and NRK cells (C)as indicator cells. One-quarter fraction of a 37-mg protein concentrate from thesame kidney was used to stimulate soft agar growth of AKR-2B cells (B).
Chart 7. High-performance liquid chromatography on an Altex Ultrasphere ODScolumn. A lyophilized sample (4 mg of protein) of pooled P-60 fractions 23 to 27from a renal cell carcinoma (Chart 5) was dissolved in 200 ¡¿of 0.025% trifluoroa-cetic acid in water (v/v) and loaded on an Altex Ultrasphere ODS column (1 x 25cm). A linear gradient of 20% n-propyl alcohol to 50% n-propyl alcohol containing0.025% trifluoroacetic acid over 75 min at a flow rate of 1 ml/min was used. One-sixth fraction was used to stimulate soft agar growth of SW-13 (A). AKR-2B cells
(B), and NRK cells (C) with or without the addition of 2 ng EOF per ml.
which stimulated low levels of soft agar growth of SW-13 cells.
Both cell extracts and conditioned media from 3 carcinoma celllines stimulated soft agar growth of SW-13 cells. The A204
rhabdomyosarcoma cell extract and cell extract and conditionedmedium from A375 melanoma cells were inactive. However, theA204 conditioned medium did stimulate colony formation by SW-
13 cells. Human nonneoplastic epithelial organ extracts (kidneyand lung) also had SW-13 colony-stimulating activity. Extracts
from kidneys had considerably more activity than did extractsfrom lung, an observation which may be accounted for by themuch lower content of stroma in kidney relative to lung. It is ofinterest that FGF was also shown to stimulate colony formationby SW-13 cells, since the FGF used in these experiments was
extracted from bovine pituitaries, an epithelial organ.The effect of TGFs on SW-13 cells was found to be irreversible.
The SW-13 cells acquired the ability to grow well in soft agar at
low cell densities following treatment with and removal of theTGF. This is in contrast with the reversible effects of TGFs uponnontransformed NRK and AKR-2B cells (6, 13). However, theirreversible effect of TGFs on SW-13 cells is in accordance with
the studies of Colburn and Gindhart (5). They demonstrated the
irreversible conversion of anchorage-dependent epidermalmouse JB6 cells to anchorage-independent cells following treat
ment with TGF or TPA. The JB6 cells are thought to representa late preneoplastic stage of progression with the TGF or TPApromoting the conversion of the cells to neoplastic cells. SinceSW-13 cells are derived from a malignant tumor, it is likely thatthe mechanism of TGF action on SW-13 cells is similar to that
of TGF on JB6 cells causing progression of the cells to a moreneoplastic state. The mechanism of TGF action is poorly understood at the present time; however, evidence does exist thatTGF may act in a manner similar to tumor promoters (24).
The factor(s) which stimulates colony formation by SW-13
cells shares many of the properties of the TGFs describedpreviously (6, 13, 17, 20). These factors are heat- and acid-
stable polypeptides with disulfide bonds and have the propertyof stimulating anchorage-dependent cells to grow in soft agar in
addition to their ability to stimulate DNA synthesis and causemorphological changes in cells in monolayer. However, the fac-tor(s) stimulating SW-13 colony formation appears to be separate
and distinct from the factors which stimulate colony formationby AKR-2B and NRK cells as reported by Nickell er a/. (15).Separation of the factor(s) stimulating SW-13 cells from those
stimulating the other cell types was achieved by both molecularsieve chromatography and high-performance liquid chromatog
raphy. In addition, it was possible in a number of circumstancesto separate on molecular sieve chromatography the factor whichstimulates AKR-2B cells from that which stimulated NRK cells.
This means that many of the human tissues and neoplasmscontain at least 3 separate and distinct growth factor-like poly
peptides capable of stimulating different cell types to grow insoft agar. The presence of more than one type of growth factorin a given tissue or cell type is well known. For example, thesalivary gland is a source for both nerve growth factor and EGF(21), while platelets have been shown to contain the well-knowncationic platelet-derived growth factor (1) and a separate TGF
(4). The production of multiple growth active peptides by celllines is a well-known phenomenon (22).
Roberts ef a/. (18) have reported that murine sarcoma virus-
transformed mouse cells contain 2 separate and distinct TGFsand have presented evidence indicating that both factors maybe required for the expression of the transformed phenotype (2).It is possible that a complex interaction of 2 or more of theendogenous factors described in the present study may play arole in certain stages of neoplastic development and in themaintenance of the transformed phenotype. This possibility remains to be established as does the relationship of these endogenous growth factor-like polypeptides to other, better character
ized growth factors.
REFERENCES
1. Antoniades, H. N. Human platelet-derived growth factor (PDGF): purificationof PDGF-I and PDGF-II and separation of their reduced subunits. Proc. Nati.Acad. Sei. U. S. A., 78: 7314-7317,1981.
2. Anzano, M. A., Roberts, A. B., Meyers, C. A., Komoriya, A., Lamb, L. C.,Smith, J. M., and Sporn, M. B. Synergistic interaction of two classes oftransforming growth factors from murine sarcoma cells. Cancer Res., 42:4776-4778, 1982.
3. Bradford, M. M. A rapid and sensitive method for the quantitation of microgramquantities of protein utilizing the principle of protein-dye binding. Anal.Biochem., 72: 248-254,1976.
4. Childs, C. B., Proper, J. A., Tucker, R. F., and Moses, H. L. Serum contains aplatelet-derived transforming growth factor. Proc. Nati. Acad. Sei. U. S. A., 79:
5312-5316,1982.5. Colburn, N. H., and Gindhart, T. D. Specific binding of transforming growth
factor correlates with promotion of anchorage independence in EGF receptor-less JB6 cells. Biochem. Biophys. Res. Commun., 102: 799-807,1981.
6. DeLarco, J. E., and Todaro, G. J. Growth factors from murine sarcoma virus-transformed cells. Proc. Nati. Acad. Sei. U. S. A., 75: 4001-4005,1978.
7. Gierd, D. J., Aaronson, S. A., Todaro, G. J., Arnstein, P., Kersey, J. H., Dosik,H., and Parks, W. P. In vitro cultivation of human tumors: establishment of celllines derived from a series of solid tumors. J. Nati. Cancer Inst., 57: 1417-1423, 1973.
8. Halper, J., and Moses, H. L. Stimulation of growth in soft agar of SW-13human carcinoma cells by factors released by these cells by extracts fromnormal and neoplastic human tumors. Fed. Proc., 41: 854,1982.
9. Kahn, P., and Shin, S.-l. Cellular tumorigenicity in nude mice: test of associationamong loss of cell-surface fibronectin, anchorage independence, and tumor-forming ability. J. Cell Biol., 82: 1-16, 1979.
10. Kaplan, P. L., Anderson, M., and Ozanne, B. Transforming growth factor(s)production enables cells to grow in the absence of serum: an autocrine system.Proc. Nati. Acad. Sei. U. S. A., 79: 485-489, 1982.
11. Leibovitz, A., McCombs, W. B., Ill, Johnston, D., McCoy, C. E., and Stinson,J. C. New human cancer cell culture lines. I. SW-13, small-cell carcinoma ofthe adrenal cortex. J. Nati. Cancer Inst., 57: 691-697,1973.
12. Marquardt, H., and Todaro, G. J. Human transforming growth factor: production by a melanoma cell line, purification, and initial characterization. J. Biol.Chem., 257: 5220-5225, 1982.
13. Moses, H. L., Branum, E. L., Proper, J. A., and Robinson, R. A. Transforminggrowth factor production by chemically transformed cells. Cancer Res., 47:2842-2848,1981.
14. Moses, H. L., Proper, J. A., Volkenant. M. E.. Wells, D. J., and Getz, M. J.Mechanism of growth arrest of chemically transformed cells in culture. CancerRes., 38:2807-2812, 1978.
15. Nickell, K. A., Halper, J., and Moses, H. L. Transforming growth factors insolid human malignant neoplasms. Cancer Res., 43: 1966-1971, 1983.
16. Peterson, W. D., Jr., Stulberg, C. S., and Simpson, W. F. A permanentheteroploid human cell line with type B glucose-6-phosphate dehydrogenase.Proc. Soc. Exp. Biol. Med., 736: 1187-1191,1971.
17. Proper, J. A., Bjornson, C. L., and Moses, H. L. Mouse embryos containpolypeptide growth factor(s) capable of inducing a reversible neoplastic phen-
Epithelial Tissue-derived TGFs
otype in nontransformed cells in culture. J. Cell. Physiol., 770:169-174,1982.
18. Roberts, A. B., Anzano, M. A., Lamb, L. C., Smith, J. M., Frolik, C. A.,Marquardt, H., Todaro, G. J., and Sporn, M. B. Isolation from murine sarcomacells of a new class of transforming growth factors potentiated by epidermalgrowth factor. Nature (Lond.), 295: 417-419,1982.
19. Roberts, A. B., Anzano, M. A., Lamb, L. C., Smith, J. M., and Sporn, M. B.New class of transforming growth factors potentiated by epidermal growthfactor: isolation from non-neoplastic tissues. Proc. Nati. Acad. Sei. U. S. A.,78:5339-5343.1981.
20. Roberts, A. B., Lamb, L. C., Newton, D. C., Sporn, M. B., DeLarco, J. E., andTodaro, G. J. Transforming growth factors: isolation of polypeptides fromvirally and chemically transformed cells by acid-ethanol extraction. Proc. Nati.Acad. Sei. U. S. A., 77: 3494-3498, 1980.
21. Savage, C. R.. Jr., and Cohen, S. Epidermal growth factor and a new derivative.Rapid isolation procedure and biological and chemical characterization. J. Bid.Chem., 247: 7609-7611, 1972.
22. Shields, R. Growth factors for tumours. Nature (Lond.), 272: 670-671, 1978.23. Sporn, M. B., and Todaro, G. J. Autocrine secretion and malignant transfor
mation of cells. N. Engl. J. Med., 303: 878-880, 1980.24. Todaro, G. J., DeLarco, J. E., and Sporn, M. B. Retinoids block phenotypic
cell transformation produced by sarcoma growth factor. Nature (Lond.), 276:272-274,1978.
25. Todaro, G. J., Fryling, C., and DeLarco, J. E. Transforming growth factorsproduced by certain human tumor cells: polypeptides that interact with epidermal growth factor receptors. Proc. Nati. Acad. Sei. U. S. A., 77: 5258-5262,1980.
26. Tucker, R. F., Volkenant, M. E., Branum, E. L., and Moses. H. L. Comparisonof intra- and extracellular transforming growth factors from nontransformedand chemically transformed mouse embryo cells. Cancer Res., 43: 1581-1586, 1983.
27. Twardzik, D. R., Ranchalis, J. E., and Todaro, G. J. Mouse embryonic transforming growth factors related to those isolated from tumor cells. Cancer Res.,42:590-593,1982.
28. Twardzik, D. R.. Todaro, 'G. J.. Marquardt, H., Reynolds, R. H., Jr., and
Stephenson, J. R. Transformation induced by Abelson murine leukemia virusinvolves production of a polypeptide growth factor. Science (Wash. D. C.),276:894-897, 1982.
![Exome-wide mutation profile in benzo[a]pyrene-derived post-stasis and immortal human mammary epithelial cells](https://static.documents.pub/doc/80x56/6351df259ff726561f0930d3/exome-wide-mutation-profile-in-benzoapyrene-derived-post-stasis-and-immortal-human.jpg)
