[CANCER RESEARCH 48, 3737-3741, July 1, 1988]

Characterization of a Low-Molecular-Weight Growth Inhibitor Formed byDensity-inhibited, Tumorigenic V79 Chinese Hamster Cells1

Tatsuhiko Koga, Shuji Nakano,2 Ichiro Ichinose, Hidenori Yantada, and Yoshiyuki Niho

First Department of Internal Medicine, Faculty of Medicine [T. K., S. N., 1.1., Y. NJ, and Faculty of Pharmaceutical Science [H. YJ, Kyushu University, 3-1-1 Maidashi,Fukuoka 812, Japan

ABSTRACT

A novel type of low-molecular-weight growth-inhibitory factor respon

sible for the density inhibition of tumorigenic V79 Chinese hamster cellshas been purified, if not homogenously, by a series of reverse-phase andgel filtration high-performance liquid chromatography. The factor is anacid-stable, heat-labile substance distinct from antiproliferative nucleo-

side analogues or polyamines and has a molecular weight of approximately 2000. The biological activity of this inhibitor was enhanced nearly5-fold by trypsin treatment, thereby suggesting that the inhibitor may be

a precursor peptide which becomes an oligopeptide with intense biologicalactivity by proteolysis, or that trypsin treatment allows resultant smallmolecules to efficiently transfer across the cytoplasmic membrane. Thisinhibitor reversibly inhibits the growth of a broad spectrum of cell typesfrom neoplastic and nonneoplastic cells from various species. These datasuggest that this inhibitor is primarily a growth-regulatory molecule

common to mammalian cells and may play an important role in regulatinggrowth of both normal and neoplastic cells.

INTRODUCTION

Proliferation of normal cells in vitro is stringently controlledby a variety of regulatory mechanisms (1-6). Conversely, neoplastic cells are insensitive to these regulatory programs andunable to shift into the quiescent state after cell-to-cell contact,thus resulting in a high saturation density (7-9). However, evenneoplastic cells decrease in their active growth potential as theybecome confluent (density-inhibited state). This growth characteristic of neoplastic cells in the density-inhibited state seemsto be mainly due to soluble inhibitory factors, as these cellscontinued to proliferate, in a pile-up fashion, when the mediumis frequently changed, as opposed to normal cells where frequent medium changes have little effect on proliferation of thecells at the density-inhibited state. Correspondingly we foundthat the medium conditioned by exposure to density-inhibited,tumorigenic V79 cells reversibly inhibits the growth of sparse,proliferating cultures of the same cell line (10, 11).

These observations have prompted us to search for thegrowth-inhibitory substance in the V79-conditioned medium.We found that this inhibitory activity is apparently mediatedby soluble inhibitors produced endogenously and released intothe medium at the time of density inhibition. An approximateMr 2000 low-molecular-weight inhibitor has been identified byultrafiltration and gel filtration (12). Our results suggest thatgrowth regulation by an endogenous inhibitor is also opérantin tumorigenic V79 cells. In addition, several human tumor celllines, which secrete specific growth factors, have been shown tosecrete growth-inhibitory factors (13-15). All these observations suggest that the proliferaiivc potential of neoplastic cellsmay be augmented both positively and negatively and thatmechanisms involved in the tumor cell growth are too complicated to be explained by the simple extrapolation of the en-

Received7/28/87;revised2/10/88;accepted3/30/88.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1These studies were supported by <¿rant-inAid for Scientific Research

62570290 from the Ministry of Education, Science, and Culture, Japan, and bya grant from the Uehara Memorial Foundation.

2To whom requests for reprints should be addressed.

hanced positive growth potential. To understand such a complex nature of proliferation mechanisms inherent in tumor cells,identification and characterization of factors involved in thegrowth regulation of tumor cells are of paramount importance.

In this paper, we further purified and characterized a partiallypurified low-molecular-weight growth inhibitor derived fromV79 cells, which accounts for the major growth-inhibitoryactivity of the V79-conditioned medium. This inhibitor appearsto be an oligopeptide which reversibly inhibits growth of a widevariety of cell types of neoplastic and nonneoplastic cells, beinga common regulator for growth of mammalian cells, regardlessof tumorigenicity.

MATERIALS AND METHODS

Cell Culture and Preparation of Conditioned Medium. Cells weregrown at 37°Cin a humidified atmosphere of 5% COi and 95% air inDulbecco-modified Eagle's medium (Nissui Pharma Co., Tokyo, Japan)supplemented with 5% heat-inactivated PCS3 (Gioco Lab., Chagrin

Falls, OH), with the exception of human cells which required 10%PCS. V79 cells, established from Chinese hamster lung fibroblasts,were demonstrated to be tumorigenic by s.c. inoculation of 10 cellsinto athymic BALB/c-nu/n« mice. BALB/c 3T3 A31-714 cells, originally isolated by T. Kakunaga (16), were recloned in our laboratory andwere found to be strictly sensitive to density-dependent inhibition ofcell division. This cell line was used as target cells for measurement ofthe inhibitory activity of fractionated materials, as described (12). HeLais a human cervical carcinoma cell line. The cell line KSE-1, a humanesophageal carcinoma, and US 11. a human tongue squami HIs cellcarcinoma which we established, were found to be tumorigenic by s.c.inoculation of IO7cells into athymic nude mice. The KSE-1 cell line in

particular has been fully characterized (17). HSF is a human adult skinfibroblast cell line maintained in our laboratory.

For preparation of conditioned medium from density-inhibited V79cells, V79 cells were grown to confluence in 100-mm dishes (Falcon).Since medium conditioned in the presence of 0.2% serum containedonly approximately 10% of the inhibitory activity, the conditionedmedium was prepared by exposing 10 ml of culture medium supplemented with 5% PCS to the confluently grown V79 cell culture for 24h, as soon as the culture became confluent. The medium was thencollected, passed through a 0.4-^m membrane filter, and stored at 4*C

until use. Unconditioned medium was prepared in parallel from thesame fresh medium and incubated under the same conditions, but inthe absence of cells.

Reverse-Phase Chromatography. The conditioned medium was precipitated by the addition of an equal volume of acetonitrile. Thesupernatant, obtained after centrifugation at 4000 rpm for l S min, wasconcentrated by an evaporator. Gel filtration of 30 ml of concentratedmaterials was carried out at 4°Con a Sephadex G-2S column (SS x 3.2

cm) equilibrated and eluted with 0.1% TFA as described (12). Low-molecular-weight growth-inhibitory fractions ranging between 500 and2000 were combined and lyophilized. The redissolved residue with Viovolume of distilled water was applied to a reverse-phase Lobar columnchromatography (Lichroprep RP-18; Merck). Chromatography wasdone on Pharmacia fast protein liquid chromatography equipment,using distilled water containing 0.1 % TPA with a stepwise gradient ofacetonitrile containing 0.1% TFA. Every fraction was monitored by a

3The abbreviations used are: PCS, fetal calf serum; TFA, trifluoroacetic acid;HPLC, high-performance liquid chromatography; cAMP, cyclic AMP; TGF,transforming growth factor; TIP, tumor cell growth-inhibiting factor.

3737

Research. on November 21, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

OLIGOPEPTIDE GROWTH REGULATOR IN V79 CELLS

UV100 fluorescent stream detector (Hitachi) at 240 nm, and 2.0 ml ofeach fraction were lyophilized, redissolved in complete medium, andassayed for growth-inhibitory activity.

Purification of the Inhibitor by HPLC. The major inhibitory peaks(Fig. 1, Pool A) from the Lobar column were combined, lyophilized,resuspended in 0.5 ml of distilled water, and injected onto a TSK geloctadecyl silane 120A reverse-phase column (Toyosoda). Chromatog-raphy was carried out with Hitachi HPLC equipment, using an isocraticelution of distilled water containing 0.1 % TFA. The column eluate wasmonitored simultaneously with absorbance at 210 and 280 nm. Eachpeak was lyophilized, resuspended in complete medium, and assayedfor inhibitory activity. By repeating the rechromatography, a singlefraction containing the inhibitory activity from a reverse-phase HPLCwas separated from other impurities included in the fraction, and thispeak fraction was lyophilized, resuspended in 100 pi of distilled water,applied to a column (7.5 x 600 mm) of TSK gel 3000 SW (Toyosoda)for gel filtration HPLC, equilibrated, and eluted with 10% acetic acidat room temperature. The column eluate was monitored with a UVdetector set at 280 nm at a flow rate of 0.4 ml/min. Each peak waslyophilized, redissolved in complete medium, and assayed for growth-inhibitory activity, as described below.

Assays of Growth-inhibitory Activity. Inhibitory activities of fractionated materials were determined by measurements of both cell growthand DNA synthesis in 3T3 cells, as described (12). To determine cellgrowth, replicate plates (35 mm; Falcon 3001) inoculated with 2 x IO4

cells were incubated for 18 h for attachment. Medium was then replacedwith complete medium containing each of the lyophilized and dissolvedfractions, and the incubation continued for an additional 48 h. At thattime, the cell number was determined by detaching the cells with 0.05%iryps in and 0.02% EDTA in calcium- and magnesium-free phosphate-buffered saline and counting them in a particle data counter (Model IDCoulter Counter). For the DNA synthesis, a double labeling assay wasperformed. Briefly, 3T3 cells were inoculated into replicate dishes (35mm; Falcon) at a density of 2 x IO4cells per dish and then incubatedfor 18 h in medium containing 0.01 pCi of [l4C]thymidine (specific

activity, 61 mCi/mmol) per ml. After this prelabeling, the medium wasreplaced with complete medium containing each lyophilized and redissolved fraction. The cells were then continuously labeled with 1.0 ¿iCiof [3H]thymidine (specific activity, 50 Ci/mmol) per ml for 6 h. Afterlabeling, the radioactivities of both 3H and 14C incorporated intotrichloroacetic acid-insoluble materials were determined simultaneously in a scintillation counter (Aloka Model 701). Since the radioactivity of I4C indicates the DNA content of the cycling cells at the timeof the measurement of DNA synthesis, the resulting 3H/14C ratios can

be considered as DNA synthesis per DNA and, therefore, the specificsynthetic activity of DNA (18). In the assay of every peak in HPLC,the inhibitory coefficient, defined as a reciprocal of the percentage ofcell growth, was calculated.

Stability of the Inhibitor. Aliquots from the fraction containing

5g

Ô 50

S

20 30 40 50 60

Fraction number

Fig. 1. Reverse-phase Lobar column chromatography of concentrated low-molecular-weight fractions obtained from Sephadex G-25 gel filtration chromatography. The elution sequence of acetonitrile was a 0 to 15% stepwise gradientfor 600 ml. One-fifth volume of each fraction was lyophilized, resuspended infresh medium, and tested for growth-inhibitory activity on 3T3 cells, as describedin "Materials and Methods." The fractions containing the growth-inhibitory

activity were divided into 2 pools (A and B). OD, absorbance.

inhibitory activity from reverse-phase chromatography were tested fortheir sensitivity to trypsin and heat. Aliquots were treated with trypsin(No. T-2395; Sigma, St. Louis, MO) at a concentration of '/iooweightof lyophilized aliquots for 2 h at 37°C,and then an equal activity of

soybean trypsin inhibitor (No. T-9003, Sigma) was added. After lyoph-ilizing and redissolving in complete medium, aliquots were assayed forcell growth. For the controls, the same amounts of trypsin and trypsininhibitor were mixed, lyophilized, and equally assayed. For determination of heat sensitivity, lyophilized aliquots were redissolved indistilled water, heated at 95°Cfor 30 min, immediately refrigerated,

and lyophilized for growth-inhibitory activity.Elution Patterns on Reverse-Phase HPLC. Thymidine, thymine,

dTTP, and cAMP (all from Sigma Chemical Co., St. Louis, MO) wereadjusted to approximately the same concentration of the purified inhibitor by absorbance at 260 nm, and the elution patterns were comparedon reverse-phase HPLC.

RESULTS

Reverse-Phase Chromatography. The low-molecular-weightinhibitor has been shown to be the major factor responsible forthe density inhibition of tumorigenic V79 cells (12). To furtherpurify this low-molecular growth inhibitor, preparations with agrowth-inhibitory activity from Sephadex G-25 chromatography were fractionated over a reverse-phase Lobar column (Fig.1). There were three major peaks containing the inhibitoryactivity. Fractions in the first inhibitory peak eluting at 0%acetonitrile contained enormous amounts of salts, and theinhibitory activity was toxic. The remaining two peaks elutedafter the salts, and their inhibitory activities were not cytotoxic.The fractions in the second inhibitory peak eluting at approximately 5% acetonitrile were designated Pool A (Fractions 29to 34), and the fractions in the third inhibitory peak eluting atapproximately 8% acetonitrile were Pool B (Fractions 38 to42). The inhibitory activities measured by cell proliferationlargely paralleled those determined by DNA synthesis. However, in Pool B, the inhibitory activity was more evident whenmeasured by DNA synthesis. This was expected as Pool B hadbeen purified to apparent homogeneity and was found to bethy mid ine, as determined by physicochemical analyses (19).This endogenously produced cold thymidine is considered tocompete with [3H]thymidine, resulting in a decrease in incorporation of [3H]thymidine. When unconditioned medium was

similarly processed for gel filtration and subsequently for reverse-phase chromatography, mostly salts were isolated, andno significant peaks corresponding to Pool A appeared.

Reverse-Phase and Gel Filtration HPLC. When Pool A fromthe reverse-phase Lobar column was chromatographed on aTSK gel octadecyl silane 120A reverse-phase HPLC columnusing a 2% isocratic elution, the inhibitory activity was observedin a single small peak (Fig. 2). The potent inhibitory fractions(Peak A) were collected by repetitive reverse-phase HPLC andinjected onto a gel filtration HPLC (Fig. 3). Although severalsmall peaks appeared, these peaks contained no inhibitoryactivity. Inhibitory activity was demonstrated in the fractionwith no discernible UV absorbance at 280 nm. The molecularweight of this inhibitor appears to be approximately 2000,according to the elution data of several markers (Fig. 3). Although there was no evidence of homogeneity, the inhibitor wasconsidered to have been purified, as Pool B proved to bethymidine, with the same HPLC procedures (19).

Physicochemical Analyses. As shown in Table 1, Pool A wasan acid-stable (pH 3) but heat-labile compound not inactivatedby trypsin. However, it should be noted that the activity wasenhanced nearly 5-fold by trypsin treatment (Table 1). Thisexperiment was repeated, with reproducible results. These data

3738

Research. on November 21, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

OLIGOPEPTIDE GROWTH REGULATOR IN V79 CELLS

_ 1.5

- 1.0

i

10 20 30 40 50

elution time (mm)

60

Fig. 2. Reverse-phase HPLC of the partially purified inhibitor (Pool A). Theconcentrated Pool A was chromatographed using a 2% isocratic acetonitrileelution in 0.1 % TFA at a flow rate of 0.8 ml/min, and the effluent was monitoredsimultaneously at 210 and 280 nm. Each peak was lyophilized, resuspended infresh medium, and tested for growth-inhibitory activity on 3T3 cells. Arrowsindicate the fraction containing the inhibitory activity. OD, absorbance.

Table 1 Effects of heat or trypsin on activity of the growth-inhibitory fraction(Pool A)

Each treatment was performed on an aliquot (2 ml) of Pool A obtained fromreverse-phase Lobar column chromatography (Fig. 1). The amount of trypsin wasone-hundredth of the weight of Pool A. Treatment with trypsin was terminatedby the addition of soybean trypsin inhibitor, adjusted by the trypsin activity.

TreatmentControl

mediumPool A (no treatment)Pool A, heated (95'C, 30 min)

Control medium + trypsin +TI»

Pool A, trypsin treated (37'C,

2h)Cell

no.(X10-')6.91

±0.24'

3.79 ±0.196.79 ±0.407.31±0.370.66

±0.05%of

cell growth100

54.8 ±2.795.4 ±5.8

105.8 ±5.49.6

±0.7

" Mean ±SD of triplicate plates.* TI, trypsin inhibitor.

peak A

cAMP

Thymine

Thymidme

M W

(kilodalton)

500

100

50

10

elution time

dTTP

20 40

Fig. 3. Gel nitration HPLC of inhibitory peak (Peak A) obtained from reverse-phase HPLC. The inhibitory peak on reverse-phase HPLC (Peak A) was concentrated by lyophilization and injected into TSK gel 3000 SW. The flow rate was0.4 ml/min using 10% acetic acid, and the effluent was monitored at 280 nm.Each fraction eluting every 2 min (0.8 ml) was lyophilized, resuspended in freshmedium, and tested for growth-inhibitory activity on 3T3 cells, as described in"Materials and Methods." The inhibitory activity was shown as an inhibitory

coefficient, and inhibitory fractions were marked by the bidirectional arrow.Molecular weight (M. W.) markers indicated were gluconate dehydrogenase (M,290,000), lactate dehydrogenase (M, 142,000), anolase (M, 67,000), and lysozyme(M, 14,000). OD, absorbance.

3739

elution time (min)

Fig. 4. Comparison of elution patterns of the purified inhibitor (Peak A) withvarious synthetic nucleic acid compounds including cAMP, thymidine, thymine,and dTTP. The chromatography was performed with a flow rate of 0.8 ml/minusing a 2% isocratic acetonitrile elution in 0.1% TFA.

indicate that this inhibitor appears to be a small poiypeptidewith a trypsin-susceptible peptide bond in the primary structure.To distinguish Peak A from the inhibitory nucleic acid compounds (20, 21), Peak A was compared with cAMP, thymidine,and dTTP in elution pattern, using a 2% acetonitrile isocraticelution on reverse-phase HPLC (Fig. 4). We found that PeakA differs from these compounds in the elution pattern.

Comparison with Spermine. We also compared the inhibitor(Peak A) with spermine, the most hydrophobic among thepolyamines, because polyamines have been shown to have cy-totoxic effects on mammalian cells (22). Initially, we measuredthe concentration of polyamines in the conditioned medium,using an enzymatic method, but polyamines were not detected.Therefore, spermine was injected onto reverse-phase HPLC,using the same procedure as was used for the purification ofPeak A. As shown in Fig. 5, spermine showed no UV absor-

Research. on November 21, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

OLIGOPEPTIDE GROWTH REGULATOR IN V79 CELLS

bance but was detected by the ninhydrin reaction. The resultshows that Peak A differed from spermine, as deduced by theelution rate on reverse-phase HPLC. Altogether, these resultsindicate that the inhibitor may be an oligopeptide distinct fromantiproliferative nucleoside analogues and polyamines.

Effects of Pool A on Various Cell Lines. Pool A was testedfor its effect on other normal and tumor cells from variousspecies. Table 2 shows representative examples of the effect ona variety of cell lines from rodents to humans. Although thereare considerable differences in growth suppression among thesecell lines, this inhibitor is species nonspecific and effective forall cell lines tested, with either normal and malignant pheno-types. Growth inhibition became more evident when the cellswere incubated for longer periods of time (Table 2). However,the inhibitory effects were reversible at least within 72 h,indicating that a toxic factor was not included in Pool A. Thedifferential inhibitory effect of this inhibitor among various celllines appears not to depend on the growth rate of the cells,because there was little difference in growth inhibition betweenrapidly growing V79 cells (doubling time, 11 h) and slowlygrowing human skin fibroblasts (36 h).

DISCUSSIONThe culture medium of density-inhibited, tumorigenic V79

cells contains several molecular weight classes of growth-inhib-

peak A

S0o l

spermine

F;oo

fninhydrin \\ reaction /

O 10 20 30 40

elution time (min)Fig. 5. Comparison of elution patterns between Peak A and spermine. The

chromatography was done using a 0% isocratic acetonitrile gradient. Peak A wasmonitored at 210 nni, and spermine was detected by ninhydrin reaction. ++,strongly positive; +, positive; ±,slightly positive; -, negative. OD, absorbance.

Table 2 Growth-inhibitory activities of Pool A on various cell linesThe cells were inoculated at a density of 2 ~ 5 x IO4 cells per 35-mm dish.

After 18 h, the lyophilized Pool A, obtained from reverse-phase Lobar columnchromatography (Fig. 1), was resuspended in fresh medium and equally dispensedinto replicate dishes inoculated with various cells. After 48 to 72 h, the cellnumber was determined as described in "Materials and Methods."

CelllineExperiment

1(48-htreatment)V793T3HST-1HSFExperiment

2(72-htreatment)KSE-1HeLaHST-1HSFCell

no.(xControl24.6

±0.15°6.91

±0.240.54±0.050.50±0.0432.7

±1.2723.0±0.0916.8±0.4610.6±0.3910~!)Pool

A22.0

±0.43.79±0.190.37

±0.040.36±0.0222.0

±1.1420.9±0.964.34±0.443.74±0.04%of

cellgrowth89.4

±1.654.8±2.768.5±7.472.0±4.067.2

±3.590.7±4.125.8±3.535.3±0.3

°Mean ±SD of triplicate plates.

itory factors (12). Specifically, the major factor responsible forthe inhibitory activity of the conditioned medium has beendemonstrated in the low-molecular-weight fractions (Mr 500 to2000). This partially purified inhibitor was subsequently separated into two distinct inhibitory fractions (Pools A and B),using reverse-phase chromatography. Using HPLC procedures,these inhibitory fractions were purified, if not homogenously.The more hydrophobic inhibitor (Pool B) proved to be thymi-dine, as determined by physicochemical analyses, and the partial involvement of thymidine in the inhibitory activity of conditioned medium has been ascertained (19). A similar analysisof the hydrophilic inhibitor (Pool A) revealed that this inhibitoris an acid-stable, heat-labile oligopeptide distinct from antiproliferative nucleic acid compounds or polyamines with a molecular weight of approximately 2000. This compound reversiblyinhibits the growth of a broad spectrum of cell types fromnormal to malignant cells, suggesting that this inhibitor isprimarily a growth-regulatory molecule common to all mammalian cells regardless of the tumorigenicity.

The unique property of this molecule is that the inhibitoryactivity is markedly enhanced by trypsin treatment, therebyindicating the presence of trypsin-susceptible peptide bonds.These data also suggest that the inhibitor is a complex of theactive molecular part and a carrier molecule which may maskthe activity of the active part and that these complexes arelinked together by trypsin-susceptible peptide bonds. The cleavage of biologically active polypeptide from the preformed andinactive precursor is a well-known process occurring in theproduction of insulin from proinsulin in pancreatic B-cells (23).Likewise, this inhibitor might be a precursor peptide similar toproinsulin and become a small peptide (oiigopeptide) withintense biological activity by proteolysis. Another explanationis that it is much easier for the smaller molecule to transferacross the cytoplasmic membrane, such as via gap junctions,which limit the size of molecules to a molecular weight below1500. It is unlikely that the crude inhibitor (Pool A) containsboth growth inhibitor and trypsin-sensitive growth stimulators,because no remarkable growth-stimulating fractions were detected in the further purification steps of Pool A. Moreover,since the inhibition assay was performed under growth-stimulating conditions, including serum factors, the influence of thepresence or absence of growth-stimulating factors on the assayis probably minimal. A similar activation was noted with TGF-ß,a highly ubiquitous molecule produced by a variety of celltypes in an inactive form that is irreversibly activated by acidtreatment (24, 25).

As yet, no growth-inhibitory factors in this molecular rangehave been noted in neoplastic cells. In another aspect, cell-

specific, endogenous growth inhibitors were isolated from cellsor tissue extracts as regulators of cell growth. These inhibitorshave been referred to as chalones, and many types were described (26). Recent recognition of the low-molecular nature of

lymphocyte, granulocyte, and epidermal chalones has accelerated progress in their purification (27-29). Although molecularsizes of these chalones are close to the inhibitor we isolated,their biochemical and biological properties are dissimilar inboth cell specificity and trypsin susceptibility.

Like chalones, polyamines are found in most tissues andinhibit the proliferation in vitro of most cell types to varyingdegrees, especially when cultured in fetal calf serum (30, 31).In addition, a close relationship between small chalones andpolyamines has been suggested (32). Therefore, we investigatedthe relation of our inhibitor to the polyamines but detected nopolyamines in the conditioned medium. Moreover, the elutionstudy on reverse-phase HPLC revealed that this inhibitor is

3740

Research. on November 21, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

OLIGOPEPTIDE GROWTH REGULATOR IN V79 CELLS

much more Hydrophobie than spermine, the most hydrophobiccompound among polyamines, and thus can be readily distinguished from the polyamines.

Several growth-inhibiting soluble factors have been isolatedfrom the conditioned medium of density-inhibited cultures ofcells with a contact-mediated growth control. Wang et al.identified a growth inhibitor, a A/r 13,000 polypeptide, fromSwiss 3T3 fibroblasts (4, 33, 34). Wells and Malluci (35)isolated a growth inhibitor with a character of heat-labile,nondialyzable protein from mouse embryo fibroblasts. Sharmaand Gehring (36) reported a protease-resistant, heat-stable, low-molecular-weight inhibitor (Mr 2000) from chicken embryofibroblasts. Holley et al. have purified another class of thegrowth inhibitor, a A/r 24,000 glycoprotein, from African greenmonkey BSC-1 cells (5, 37), although this inhibitor was laterfound to be similar, if not identical, to TGF-ß(38) and to haveboth stimulatory and inhibitory effects, depending on the cellsand experimental conditions (25). Biological and physicochem-ical comparison of these inhibitors with the inhibitor from V79cells indicates that our inhibitor is a novel type of growthinhibitor distinct from these previously described growth inhibitors closely associated with contact-mediated growth control.

Todaro and coworkers reported the partial purification ofTIF-1 and TIF-2 from a human rhabdomyosarcoma cell line(13, 14). These preparations inhibit the growth of a line ofhuman tumor cells but stimulate the proliferation of normalhuman cells. Therefore, TIP, TGF-0, and BSC-1 inhibitor maybe a related but not identical polypeptide which may share thesame receptor and thus show similar biological properties. Sincethe low-molecular-weight growth inhibitor isolated from theV79 cell has the property of a peptide, it is tempting to considerthat this inhibitor is an active part of these relatively high-molecular-weight inhibitors. However, since this inhibitor invariably inhibits the growth of both normal and tumor cells andhas no stimulatory effect, it appears to be another importantclass of growth regulator which may have a critical role inregulating growth of mammalian cells.

ACKNOWLEDGMENTS

We thank Professor T. Imoto and M. Ohara for advice and J. Aonoand F. Nagata for technical assistance.

REFERENCES

1. Stoker, M. G. P., and Rubin, H. Density-dependent inhibitor of cell growthin culture. Nature (Lond.), 215: 171-172, 1967.

2. Holley, R. W. Control of animal cell proliferation. J. Supramol. Struct., /3:191-197, 1980.

3. Lieberman, M. A., and Glaser, L. Density-dependent regulation of cellgrowth; an example of a cell-cell recognition phenomenon. J. Membr. Biol.,0*1-11, 1981.

4. Steck, P. A., Blenis, J., Voss, P. G., and Wang, J. L. Growth control incultured 3T3 fibroblasts. II. Molecular properties of a fraction enriched ingrowth inhibitory activity. J. Cell Biol., 92: 523-530, 1982.

5. Holley, R. W., Armour, R., and Baldwin, J. H. Density dependent regulationof growth of BSC-1 cells in cell culture; growth inhibitor formed by the cells.Proc. Nati. Acad. Sci. USA, 75:1864-1866, 1978.

6. YVhittenberger, B.. Raben, D., Lieberman, M. A., and Glaser, L. Inhibitionof growth of 3T3 cells by extract of surface membranes. Proc. Nati. Acad.Sci. USA, 75: 5457-5461, 1978.

7. Abercrombie, M. Contact inhibition and malignancy. Nature (Lond.), 281:259-265, 1979.

8. Pardee, A. B. A restriction point for control of normal animal cell proliferation. Proc. Nati. Acad. Sci. USA, 71: 1286-1290, 1974.

9. Pardee, A. B., Dubrow, R., Hamlin, J. L., and Kletzien, R. F. Animal cellcycle. Annu. Rev. Biochem., 47: 715-750, 1978.

10. Nakano, S., Yamagami, H., and Takaki, R. Enhancement of excision repair

efficiency by conditioned medium from density-inhibited cultures in V79Chinese hamster cells: evidence for excision repair as an error-free repairprocess. Mutât.Res., 62: 369-381, 1979.

11. Nakano, S., Koga, T., Nagafuchi, S., Nakamura, M., Kounoue, E., Niho, Y.,and Takaki, R. The relationship between molecular and cellular repair frompotentially lethal damage induced by .Y mi-lliyl .V niiro .Y mlrosoguanidincin Chinese hamster V79 cells. Mutât.Res., ¡46:271-276, 1985.

12. Koga, T., Nakano, S., Nakayama, M., Kounoue, E., Nagafuchi, S., Niho, Y.,and Yamada, H. Identification and partial purification of a low-molecular-weight growth inhibitor formed by density-inhibited, tumorigenic V79Chinese hamster cells. Cancer Res., 46:4431-4437, 1986.

13. Iwata, K. K., Flyling, C. M., Knott, W. B., and Todaro, G. J. Isolation oftumor cell growth-inhibiting factors from a human rhabdomyosarcoma cellline. Cancer Res., 45: 2689-2694, 1985.

14. Fryling, C. M., Iwata, K. K., Johnson, P. A., Knott, W. B., and Todaro, G.J. Two distinct tumor cell growth-inhibiting factors from a human rhabdomyosarcoma cell line. Cancer Res., 45: 2695-2699, 1985.

15. Levine, A. E., McRae, L. J., Hamilton, D. A., Brattain, D. E., Yeoman, L.C., and Brattain, M. G. Identification of endogenous grow'h-inhibitoryfactors from a human colon carcinoma cell line. Cancer Res., 45: 2248-2254, 1985.

16. Kakunaga, T. A quantitative system for assay of malignant transformationby chemical carcinogens using a clone derived from Halb/3'1'3. Int. J. Cancer,

12:463-473, 1973.17. Matsuoka, H., Sugimachi, K., Ueo, H., Kuwano, H., Nakano, S., and

Nakayama, M. Sex hormone response of a newly established squamous cellline derived from clinical esophageal carcinoma. Cancer Res., 47: 4134-4140, 1987.

18. Painter, R. B., and Young, B. R. Radiosensitivity in ataxia teleangiectasia. Anew explanation. Proc. Nati. Acad. Sci. USA, 77: 7315-7317, 1980.

19. Koga, T., Nakano, S., Ichinose, I., Nakayama, M., Nagafuchi, S., Iwakiri,R., Niho, Y., and Yamada, H. Purification and characterization of growthinhibitors from density-inhibited, tumorigenic V79 cell cultures. Proc. Am.Assoc. Cancer Res., 28: 258, 1987.

20. Lockshin, A., Mendoza, J. T., Giovanella, B. C., and Stehlin, J. S., Jr.Cytotoxic and biochemical effects of thymidine and 3-deazauridine on humantumor cells. Cancer Res., 44: 2534-2539, 1984.

21. Pastan, I. H., Johnson, G. S., and Anderson, W. B. Role of cyclic nucleotidesin growth control. Annu. Rev. Biochem., 44:491-522, 1975.

22. Katsuta, H., Takaoka, T., Nose, K., and Nagai, Y. Effects of polyamines onthe proliferation of mammalian cells in tissue culture. Jpn. J. Exp. Med., 45:345-354, 1975.

23. Grant, P. T., and Coombs, T. L. Proinsulin, a biosynthetic precursor ofinsulin. Essays Biochem., 6:69-92, 1970.

24. Goustin, A. S., Leof, E. B., Shirley, G. D., and Moses, H. L. Growth factorsand cancer. Cancer Res., 46: 1015-1029, 1986.

25. Moses, H. L. Transforming growth factor, type ,<:an indirect mitogen and agrowth inhibitor. In: R. Baserga, P. Foa, D. Metcalf, and E. E. Polli (eds.),Biological Regulation of Cell Proliferation, pp. 41-48. New York: RavenPress, 1987.

26. Patt, L. M., and Houck, J. C. The incredible shrinking chalone. FEBS Lett.,120:163-170, 1980.

27. Houck, J. C., Kanagalingam, K., Hunt, C., Attallah, A., and Chung, A.Lymphocyte and fibroblast chalones: some chemical properties. Science(Wash. DC), 196: 896-897, 1977.

28. Paukovits, W. R., Hinterberger, W., and Paukovits, J. B. The granulocytechalones—a specific inhibitor of granulopoiesis: molecular weight and chemical nature. Oncology, 34: 187-189, 1977.

29. Gonzalez, R., and Verly, W. G. Purification of an inhibitor of DN A synthesisin mammary cells. Eur. J. Cancer, 14:689-697, 1977.

30. Byrd, W. J., Jacobs, D. M., and Amoss, M. S. Synthetic polyamines addedto cultures containing bovine sera reversibly inhibit in vitro parameters ofimmunity. Nature (Lond.), 267: 621-623, 1977.

31. Webber, M. M., and Chaproniere-Rickenberg, D. Spermine oxidation products are selectively toxic to fibroblasts in cultures of normal human prostaticepithelium. Cell Biol. Int. Rep., 4: 185-193, 1980.

32. Allen, J. C., Smith, C. J., Curry, M. C., and Gaugas, J. M. Identification ofa thymic inhibitor (chalone) of lymphocyte transformation as a sperminecomplex. Nature (Lond.), 267: 623-625, 1977.

33. Hsu, Y. M., Barry, J. M., and Wang, J. L. Growth control in cultured 3T3fibroblasts: neutralization by a monoclonal antibody. Proc. Nati. Acad. Sci.USA, 81: 2107-2111, 1984.

34. Hsu, Y. M., and Wang, J. L. Growth control in cultured 3T3 fibroblasts. V.Purification of an A/, 13,000 polypeptide responsible for growth-inhibitoryactivity. J. Cell Biol., 102: 362-369, 1986.

35. Wells, V., and Mallucci, L. Properties of a cell growth inhibitor produced bymouse embryo fibroblasts. J. Cell. Physiol., 117: 148-154, 1983.

36. Sharma, C. P., and Gehring, H. A low molecular weight growth inhibitorsecreted in cultures of chicken embryo fibroblasts. Biochem. Biophys. Res.Commun., 139: 1243-1249, 1986.

37. Holley, R. W., Boehlen, P., Fava, R., Baldwin, J. H., Kleeman, G., andArmour, R. Purification of kidney epithelial cell growth inhibitors. Proc.Nati. Acad. Sci. USA, 77: 5989-5992, 1980.

38. Tucker, R. F., Shipley, G. D., Moses, H. L., and Holley, R. W. Growthinhibitor from BSC-1 cells closely related to the platelet type .) transforminggrowth factor. Science (Wash. DC), 226: 705-707, 1984.

3741

Research. on November 21, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

1988;48:3737-3741. Cancer Res   Tatsuhiko Koga, Shuji Nakano, Ichiro Ichinose, et al.   Cells

HamsterFormed by Density-inhibited, Tumorigenic V79 Chinese Characterization of a Low-Molecular-Weight Growth Inhibitor

  Updated version

  http://cancerres.aacrjournals.org/content/48/13/3737

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/48/13/3737To request permission to re-use all or part of this article, use this link

Research. on November 21, 2020. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from