96 Biochimica et Biophysica Acta, 947 (1988) 96-100 Elsevier

BBA 10217 BBA Report

M o d u l a t i o n of the urokinase receptor in h u m a n co lon cell l ines

by N , N - d i m e t h y l f o r m a m i d e

Douglas Boyd, G e r m a i n e F loren t , Genes io M u r a n o and Michae l Bra t t a in

BristoI-Baylor Laboratory, Baylor College of Medicine, Houston, TX (U.S.A.)

(Received 3 February 1988)

Key words: Urokinase receptor; N,N-Dimethylformamide; Receptor modulation

The present study documents the effect of the planar, polar differentiation promoter N,N-dimethylfor- mamide (DMF) on urokinase binding to colon carcinoma cells. Exposure of the colon carcinomacell lines to the agent resulted in enhanced specific binding of radioactive urokinase to all cells tested. Insulin binding to the cells was, however, unaffected by DMF. A DMF exposure period of 45 h was required to observe maximum urokinase binding to two representative cell lines FET and RKO. Optimal stimulation of both cell lines occurred with 0.8% DMF. Scatchard analysis revealed the dissociation constants to be unchanged by the agent with the increased binding of radioactive plasminogen activator reflecting an up-regulation of binding sites. In this regard, the cell line RKO upon exposure to DMF, displayed approx. 700000 receptors/cell, the highest value published, to date, for any cell line.

A common feature of a number of malignancies including colon cancer is the elevated level of the plasminogen activator, urokinase [1-3]. The func- tion of the plasminogen activator in malignancy is not entirely clear at the present time. Although a role in extracellular matrix degradation has been proposed [4,5], other functions such as growth factor activation [6] and growth stimulation through direct mechanisms [7], have been put for- ward. Some of the actions of urokinase may be mediated via its interactions with specific binding sites [7]. Indeed, specific receptors have been de- tected on a number of normal and transformed

Abbreviations: BSA, bovine serum albumin; DMF, N,N- dimethylformamide; PBS, phosphate-buffered saline (30 mM sodium phosphate/150 mM sodium chloride, pH 7.4); Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.

Correspondence: M. Brattain, Bristol-Baylor Laboratory, Baylor College of Medicine, Houston, TX 77030, U.S.A.

cell lines which recognize the high, (55 kDa) but not the low molecular weight (33 kDa) form of urokinase [8,9].

There are few studies, to date, documenting the modulation of the urokinase rec.eptor. Treatment of U937 monocytes with phorbol ester increased the binding of a urokinase fragment by more than 10-fold [10]. Similarly, elevated cellular receptor was observed after exposure of Friend erythro- leukemia cells to hexamethylene bisacetamide [11]. In an earlier comparison, untransformed fibrob- lasts expressed 40 000 sites/cell, whilst their trans- formed counterparts displayed only 2500 recep- tors [12].

We recently demonstrated the presence of a single class of urokinase binding sites (unpub- lished data) on six human colon carcinoma cell lines which resembled, in terms of specificity stud- ies and dissociation constant, that found on other cell types, including U937 monocytes [13]. This study documents the effect of the differentiation

0167-4889/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)

promoter, N, N-dimethylformamide (DMF) on the expression of urokinase receptor on colon carcinoma cell lines. DMF was the agent of choice as it is a potent modulator of certain biochemical events in transformed colon cells. Exposure to this agent results in (a) the appearance of fibronectin- like material [14], (b) increased secretion of carcinoembryonic antigen [15] and (c) repression of c-myc expression [16]. In the current investiga- tion, treatment of the cells with this agent resulted in a 80-210% increase in the number of binding sites. For the cell line RKO, 700 000 receptors/cell were observed following treatment with DMF. N,N-dimethylformamide was purchased from Fisher Scientific, Fair Lawn, NJ. Iodogen and immobilized p-aminobenzamidine were obtained from Pierce Chemicals, Rockford, IL. Hepes, in- sulin, and bovine serum albumin (RIA grade) were supplied by Sigma Chemicals, St. Louis, MO. Amersham, Arlington Heights, IL, provided the carrier free Na125I.

The colon carcinoma cell lines 'CBS', 'GEO', 'FET', 'HCT 116', 'HCT 116b' and 'RKO' utilized in these studies have been described elsewhere [17,18]. All cells lines were maintained at 37 °C in McCoy's 5A medium supplemented with 5% fetal bovine serum (Gibco, Grand Island, NY). Pre- confluent cells were subcultured using 0.05% (w/v) (HCT 116, HCT 116b, RKO) or 0.125% trypsin (CBS, GEO, FET) in the presence of 3 mM EDTA.

For the urokinase or insulin receptor studies, colon cells were passed into 16 or 35 mm (o) dishes (Falcon). To counter the cytostatic effects of dimethylformamide, plating numbers were dou- bled in all experiments, except 'time-course stud- ies', in which 1% DMF was present. Unless stated to the contrary, the agent was added 48 h after plating, and the ceils were grown for an additional 3-4 days prior to receptor assays.

Urokinase (55 kDa) was purified from 'Winkinase' by affinity chromatography on a p- aminobenzamidine column as described elsewhere [19]. Bound urokinase was eluted using a 0.1 M glycine buffer (pH 2.2) and dialyzed extensively against PBS. Purified high-molecular weight plasminogen activator was used in receptor assays.

Urokinase, (25-50 #g) was incubated at 4°C for 10 rain in an 'Iodogen' coated tube with 0.3 mCi Na125I [20]. Radiolabelled protein was re-

97

solved from free iodide by gel filtration on a Pharmacia PD10 column.

Insulin (25 /zg) was reacted at room tempera- ture with 0.25 /~mol chloramine-T and 2 mCi Na125I [21] and the reaction was terminated after 90 s with 1/~mol of sodium metabisulfite. Radio- iodinated protein was separated from free iodide by gel filtration.

Urokinase receptor assays were carried out as follows. Near (80-100%) confluent cultures were washed twice with binding buffer (McCoy's 5A medium, 20 mM Hepes, 1 mg/ml BSA, pH 7.4) and incubated at 37°C for 30 min with 2.0 nM radioactive urokinase in the presence or absence of a 20-fold excess of unlabelled like-competitor. The cells were washed three lines with PBS con- taining 1 mg/ml BSA (PBS/BSA) and lysed with a solution containing 1% Triton X-100, 10% glycerol, 20 mM Hepes, (pH 7.4). Non-specific binding was corrected for, by subtracting the binding observed in the presence of competitor from that observed in its absence.

Saturation analysis was performed in an identi- cal fashion, but with a range (0.5-10 nM) of radioactive urokinase concentrations.

The insulin receptor assay was essentially that of Cuatrecasas [22]. Confluent cultures were washed twice with binding buffer (identical to that used for the urokinase receptor assays) and in- cubated with 2.0 nM radioactive insulin at 22°C for 2 h. The cells were washed three times with PBS/BSA and lysed as described for the iarokinase assay. The specific binding of ligand was defined as the difference between that observed in the absence and presence of a 100-fold excess of inert competitor.

Colon carcinoma cell lines express varying levels of urokinase receptor (Table I). DMF increased the amount of radioligand specifically bound to the colon cells by 80-210%. Insofar as the effect of the agent was reversible upon its removal, it is unlikely that the modulation of urokinase binding represents a toxic manifestation. The HCT 116b cell line was the weakest responder, manifesting an 80% increase over pre-treatment levels. In con- trast, the cell line CBS, which, similarly, expresses low pre-treatment levels displayed a 2.9-fold in- crease in the amount of proteinase bound. Thus, no correlation exists between response to the

98

TABLE I

UROKINASE BINDING CAPACITIES ON COLON CELL LINES GROWN WITH ( + ) OR WITHOUT ( - ) DMF

Colon cells propagated in 16 mm dishes were exposed to 1% DMF for 3-4 days. The 80-100% confluent cells were washed (twice) with binding buffer and incubated at 37°c for 30 min with a range (0.5-10 nM) of radioactive urokinase concentra- tions. Cells were washed (three times) with PBS/BSA and lysed with 1% detergent. Non-specific ligand interaction was assessed with a 20-fold excess of unlabelled competitor. The data were plotted by the Scatchard method [23] to yield binding capacities for each cell line. The results are expressed as average values + S.D. of three separate experiments.

Cell line Urokinase binding sites (receptors/cell)

- DMF + DMF % increase

CBS 6965+ 1408 20198+ 2482 188 GEO 10033+ 3251 24079+ 1003 141 FET 13877+ 5430 26366+ 2775 90 H C T l l 6 106075+ 4649 328832+42429 209 HCT l16b 133488+48000 240278+40046 80 RKO 300280+51270 690644+90083 130

chemical agent and pre-exposure levels. It is un- likely that the enhanced binding of urokinase is a result of reduced occupation of the binding sites

with endogenous proteinase as previous experi- ments (unpublished data) had revealed the inef- ficacy of acid treatment in altering the amount of urokinase bound to control (no DMF) cultures.

In two control experiments, the cell lines HCT 116 and GEO demonstrated no sensitivity to the agent with respect to the amount of insulin bound. In the absence of DMF, the cell lines GEO and HCTll6 bound specifically 2164 ___ 480 and 4279 + 223 dpm/106 cells, respectively.

Cell lines representative of the group (i.e., one low (FET) and one high (RKO) binding of the plasminogen activator) were used in subsequent characterization studies. In both cases, there was a distinct lapse between addition of the DMF and enhanced plasminogen activator binding. For FET ceils, a marginal effect of the agent was seen after 15 h (Fig. 1A). Binding of the urokinase was raised by 50% after 1 day and maximal stimula- tion was observed after 45 h. A minimum of 25 h was, however, required for enhanced binding to RKO cells (Fig. 1B), with the maximum effect apparent after 45 h. The observations that a pro- longed period of time is a pre-requisite for en- hanced binding of proteinase argues against the

,,, 20"0 I

~>_~. 15.o Z.jU')

I r _ . o

o.o

Om--~ 5.0 < E

A

t, / / ~- / / t / / / /

/ /

A / A / A / A / A ~ / A / A / A / /1 / /1 t

/

/ t /

/ / f I i I / / /

" I 71 7 400 /

w i ~_~_ aoo

2

~ 100 n"

INCUBATION TIME (h)

B

/

I f I

@

d t- / / / , / / / / / 1 /

40

/

t / t / / / / t / t

/ / / / /

i / I i I I I f I

r / / / / /

,t,1,

Fig. 1. Kadiolabelled urokinase binding to colon cells: a time-course study. At approx. 80% confluency, the medium on FET (A) and RKO cells (B) was replaced with either fresh medium (control-open column) or fresh medium containing 1~ DMF (hatched column). The cultures were analyzed for l~l-labelled plasminogen activator binding at the indicated time points. The data are representative.

of three experiments and are expressed as the mean value with error bars indicating the S.D. values.

possibility that the results herein are a conse- quence of non-specific membrane effects of the chemical agent.

FET and RKO cells exhibited similar dose dependencies to the agent (Fig. 2). A level of 0.4% DMF which was without effect on the prolifera- tion rate, stimulated a 60% increase in the amount of urokinase specifically bound to FET cells. DMF (0.6%) enhanced the binding of the plasminogen activator to RKO cells by 90%. Both FET and RKO cells were maximally stimulated at 0.8% DMF.

All cell lines were subjected to saturation analy- sis and the binding data were manipulated to yield Scatchard [23] plots (Fig. 3). It is evident that the difference in the amount of ligand bound reflects an increase in the number of binding sites for DMF-treated cells (Table I). The dissociation con- stants for all colon cells tested were unchanged by the agent (1.6-2.9 nM).

Our findings of the DMF modulation of the plasminogen activator binding sites are similar to those documented for hexamethylene bisaceta-

~ 300

Z ~ 200

~ 100

0.1 0.2 0.4 0.6 0.8 1.0

DMF CONCENTRATION (%)

Fig. 2. The DMF concentration dependency of uroldnase binding to FET and RKO cells. After 3-4 days of treatment with various DMF concentrations, FET (o) and RKO (A) cells (80-100% confluent) were used in urokinase binding assays. Ceils were enumerated at each concentration of DMF, and the amount of radioligand specifically bound was normalized to cell number. The ligand binding data were expressed as a percent increase over control (DMF absent cultures, 100%). The results depict three separate experiments with error bars indicating the S.D. values. In the absence of DMF, FET and RKO cultures bound specifically; 8902+2136 and 175284+

24700 dpm of 125 I-labelled urokinase/106 cells.

99

0.04 LU ttl IZ u.. 0.03 O z

0,02 rn

0.01

A°

50 100

0.32 LU UJ u_ 0.24 a z ~ 0,16 II1

0.08

B,

300 600 900

125 I-LABELED UROKINASE BOUND SPECIFICALLY

(tMOLE / 106 CELLS)

Fig. 3. Scatchard analysis of 125I-labelled urokinase binding to DMF-treated colon carcinoma cells. Cultures of 80-100% con- fluent FET (A) or RKO (B) cells grown with (A) or without (o) 1% DMF were subjected to saturation analysis using a range (0.5-10 nM) of radioactive urokinase concentrations. Non-specific binding was corrected for, by running parallel incubations in the presence of a 20-fold excess of unlabeUed urokinase. The specific binding data were plotted by the Scatchard method [23]. The experiment was performed at least

three times.

mide-treated Friend erythroleukemia cells in- asmuch as the increased binding of radioligand reflects an increase in receptor number rather than an altered affinity for the urokinase. The up-regu- lation of urokinase receptor in our colon cells probably does not represent a homeostatic re- sponse to an initial depletion of endogenous urokinase, since control experiments showed that co-incubation of cells with DMF and urokinase failed to abrogate the enhancing effects of the former. Further studies will, however, be neces- sary to clarify the mode of action of this agent in urokinase receptor up-regulation. In this regard, it is improbable that the effects of DMF on urokinase receptor represents a general response to reduced proliferation. Retinoic acid which re- duced the growth rate of the FET cell fine, was found to have little effect on the amount of plas- minogen activator bound (retinoic acid exposed cultures bound 12% less urokinase than untreated ceils).

At present, with the exception of binding specificity and dissociation constant, little is known about the urokinase receptor or, for that matter, its regulation. Consequently, characteriza-

100

t ion of the recep tor is a p r ior i ty and will require its i sola t ion and pur i f icat ion. Thus, the observa- t ion that D M F is an effective s t imulus for elevat- hag the number of urokinase b ind ing sites ha colon cells may prove useful in such studies. In this respect, the R K O cells line, expressed 700000 receptors af ter exposure to this agent. In teres t ing is that this value is a factor of 2.5-fold higher than that publ i shed for un t rea ted GM1380 f ibroblasts , which to the best knowledge of the authors , is the highest value for the urokinase receptor pub l i shed to da te [13]. Accordingly , D M F - t r e a t e d R K O cells m a y be ideal s tar t ing mater ia l for pur i f ica t ion and charac ter iza t ion studies of the urokinase receptor .

This work was suppor t ed b y a N I H G r a n t C A 34432.

References

1 Markus, G., Takita, H., Camiola, S., Corasanti, J., Evers, J. and Hobika, G. (1980) Cancer Res. 40, 841-848.

2 Markus, G., Camiolo, S, Kohga, S., Madeja, J. and Mittle- man, M., (1983) Cancer Res. 43, 5517-5525.

3 Gelister, J., Jass, J., Mahmoud, M., Gaffney, P. and Boulos, P. (1987) Br. J. Surg. 74, 460-463.

4 Sheela, S. and Barrett, J. (1982) Carcinogenesis 3, 363-369. 5 Liotta, L. Goldfarb, R., Brundage, R., Siegel, G., Ter-

ranova, V. and Garbisa, S. (1981) Cancer Res. 41, 4629-4636.

6 Keski-Oja, J., Lyons, R. and Moses, H. (1987) J. Cell Biochem. Suppl. 11A:60.

7 Kirchmeier, J., Wojta, J., Christ, G. and Binder, 13. (1987) FASEB J. 1, 125-128.

8 Fibbi, G., Dini, G., Pasquali, F. Pucci, M. and Del Rosso, M. (1986) Biochim. Biophys. Acta 885, 301-308.

9 Needham, G., Sherbet, G., Farndom, J. and Harris, A. (1986) Br. J. Cancer 55, 13-16.

10 Stoppelli, M., Corti, A., Soffientini, A., Cassani, G., Blasi, F. and Assoian, R. (1985) Proc. Natl. Acad. Sci. USA 82, 4939-4943.

11 Del Rosso, M., Pucci, M., Fibbi, G. and Dini, G. (1987) Br. J. Haematol. 66, 289-294.

12 Del Rosso, M., Dini, G. and Fibbi, G. (1985) Cancer Res. 45, 630-636.

13 Plow, E., Freaney, D., Plescia, J. and Miles, L. (1986) J. Cell Biol. 103, 2411-2420.

14 Marks, M., Ziober, B. and Brattain, M. (1986) Cancer Res. 46, 5248-5258.

15 Hager, J., Gold, D., Barbosa, J., Fligiel, S., Miller, F. and Dexter, D. (1980) J. Natl. Cancer Inst. 61, 439-445.

16 Mulder, K., Rickling, S. Levine, A. and Brattaln, M. (1986) Fed. Proc. 45, 1707.

17 Brattain, M., Brattain, D., Fine, W., Khaled, F., Marks, M., Kimball, P., Arcolano, L. and Danbury, B. (1981) Oncodev. Biol. Med. 2, 355-366.

18 Brattaln M., Levine, A., Chakrabarty, S., Yeoman, L. Will- son, J. and Long, B. (1984) Cancer Metastases Rev. 3, 117-191.

19 Stump, D., Thienpont, M. and Collen, D. (1986) J. Biol. Chem. 261, 1267-1273.

20 Eaton, D. and Baker, J. (1983) J. Cell Physiol. 117, 175-182. 21 Hunter, W. (1073) in Handbook for Experimental Im-

munology, (Weir D.M., ed.), pp. 1-36, Blackwell Press, Oxford.

22 Cuatrecasas, P. (1971) Proc. Natl. Aead. Sci. USA 63, 1264-1268.

23 Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672.