EXPERIMENTAL CELL RESEARCH 1$+&133-136

(1990)

SHORT NOT Stimulation of Lysosomal Enzyme Secretio LEONARD The Wistar Institute

of Anatomy

WARREN

and Biology, Philadelphia,

Levels of extracellular lysosomal enzymes are relatively high in tumors and especially so at their periphery. By degrading the intercellular matrix, these and other nonlysosomal enzymes could facilitate invasion and metastasis by tumor cells. Using a rapid assay, we have shown that cells transformed by a variety of agents can be stimulated in culture by several growth factors to secrete lysosomal enzymes. These factors have little or no stimulatory activity on their nontransformed counterparts. The basal rate of secretion of Nacetylglucosaminidase (NAGA) and the efficiency of the stimulus are greater in transformed cells in log phase of growth. These observations suggest that altered or increased responsiveness to paracrine and autocrine growth factors not only may be responsible for the persistent division of malignant cells but also may 0 isgo Academic PPZSS. 1~. promote their invasiveness. -

INTRODUCTION

The lysosomal enzymes can degrade proteins, carbohydrates, lipids, and nucleic acids both inside and outside of the cell [ 11. Normally, these scavenging enzymes are secreted slowly. However, when mouse cells such as F9, PyS-2, and other cells in culture are bathed in a medium containing buffered, isotonic sucrose solution (0.25 M) and sodium chloride (0.15 M), as much as a third of their total lysosomal enzyme content, as measured by N-acetyl-D-glucosaminidase activity, is discharged within 30 min, linearly with time. The cells remain fully viable [a]. With this procedure, we have studied the mechanism of secretion and various agents affecting the secretory rate [Z]. In this report, we show that factors such as epidermal cell growth factor (EGF), b’asic fibroblast growth factor (b-FGF), transformation growth factor (TGF-/3), thrombin, and tumor necrosis f,actor (TNF) stimulate secretion, depending on the

Pennsylvania

19104

state of growth of the cell. The actions of growth and mitogenic agents may account for the long-known, elevated levels of extracellular lysosomal enzymes in tumors [3-5] which are believed to participate in the degradation of extracellular, matrix materials, thereby facilitating tumor invasiveness and formation of metastasis t6, 71.

MATERIALS

AN

All chemicals used were of reagent grade: tumor necrosis factor (TNF), human recombinant, SA 107 u/mg protein from Cetus Corp. (Emoryville, CA); thrombin from bovine plasma and epidermal growth factor (EGF) from mouse submaxillary glands (Sigma, St. Louis, MO); basic fibroblast growtb factor (b-FGF) from bovine brain (Calbiochem, La Jolla, CA); transformation growth factor (TGF) from R & D Systems (Minneapolis, MN). Cells were seeded in Costar plastic, six-well plates with wells 35 mm in diameter (1.8 X lo5 ceils per well) and cultured in DMEM medium containing 10% bovine calf serum as described [2]. For the assay, cells were washed three times with 3 ml cold TMsucrose (T&Cl, IO &, [pH 7.01, MgCl,, 1 m&f, sucrose, 250 m&f) and incubated with either 0.5 ml TMsucrose or 0.5 ml TMsucrose containing 150 mM NaCl with or without various factors. Assays were carried out in duplicate or triplicate. After gentle rocking for 30 min at 37°C NAGA was assayed in the medium and in cells after homogenization by tbep-nitrophenyl assay. For a more detailed description of the procedure, see Ref. [2]. The extent of secretion is expressed as the percentage of the total enzyme in the cells secreted in 30 min. RESULTS

When PyS-2 cells wer h after plating (5?,,Z = stimulated by several growth factors as well as tumor necrosis factor and thrombin (Fig. 1). Cell growth was determined by the amount of protein per plate. From the time of plating to 24 h of culturing when secretion was first measured, protein per plate increased 2%fol Protein increased 2.1, 1.5, and LO + O.&fold. (n = 4) during 24-48, 48-72, and 72-96 h, respectively, of cdture; i.e., growth had ceased between the third and fourth day, at which time the rate of secretion declined

133

0024-4827&O $3.00 Copyright 0 1990 by Acatkmi~ Press, Inc. All rights of reproduction in any form reserved.

134

SHORT 50

EGF

40

h TNF

1rhrombin bFGF

I-

l( )-

I-

1

2

3

4

DEiYS

FIG. 1. Secretion of N-acetylglucosaminidase by PyS-2 cells. Cells were plated and cultured as described under Materials and Methods, and each day thereafter duplicate cultures were assayed for NAGA. Factors were added to wells containing TMsucrose-NaCl, 150 mM; EGF, 10 rig/ml; b-FGF, 20 rig/ml; TGF-0, 1 pg/ml; thrombin, 0.5 u/ml; TNF, 100 u/ml. The specific activity of NAGA of the whole cell remained constant over 4 days, 572 k 51 nmol substrate hydrolyzed per mg protein per 30 min. Mean rt SEM, n = 8.

by half and there was virtually no response to factors. A similar decline in secretion rate of F9 cells in culture with age (Day 1 to Day 4) has been observed [2]. The concentration of each factor used was that which in previous studies had stimulated secretion maximally. EGF and TNF were no more effective together than was EGF alone. Transferrin had no effect on the rate of secretion. EGF, b-FGF, and TNF boiled for 3 min lost their stimulatory activity. In these experiments, only NAGA was measured since this enzyme is generally regarded as a marker, typical of all lysosomal enzymes. Previous work in this laboratory has shown that protease and several glycosidases are secreted in parallel with NAGA [2]. Secretion depended on the presence of both sucrose and sodium ions; little secretion occurred in the absence of sodium ions, even when sucrose was present at a concentration of 0.5 M (2X isoosmolar). Factors added to cells incubated in sucrose solution without NaCl (TMsucrose) elicited little or no response. The factors had no effect on NAGA activity when added directly to assay tubes.

NOTE

The rapidity of the response of cells to the factors is remarkable. Maximum stimulation of the washed cells occurred within the 30-min incubation period and increases in secretion were observed after only 15 min (data not shown). Serum was omitted from the assay vessels because it was found to have no effect in this and previous studies [2]. Preincubation of cell cultures with factors for 1.5 h before the assay did not increase secretion over that of cells exposed to factors during the 30min assay only. The immediacy of the response to growth factor would suggest that change in the distribution of mannose-6-phosphate receptors within or on the surface of the cell is not involved. Further, secretion by cells induced by the sucrose-NaCl stimulus was not affected by the presence of glucose-6-phosphate or mannose-6-phosphate in the medium in concentrations ranging from l-10 mM (data not shown). Two sets of control and transformed mouse and rat cells in culture were used in studies showing that some transformed cells are considerably more responsive to EGF, b-FGF, TGF-P, thrombin, and TNF than are their control counterparts. Contact-inhibited, nontumorigenie, Balb/c3T3 cells were barely affected by these factors while those cells transformed by DNA coding for Ki-ras oncogene product or polyoma virus were stimulated to various degrees (Fig. 2). In an experiment using baby rat kidney cells (BRK) transfected and immortalized with Ad-2 ElA plasmids but not transformed [B], the factors had no significant effect on secretion 1 day after cells were plated (Table 1). Four days after plating, secretion was sharply reduced and again there was little significant response to the factors. Upon transformation of cells containing

I””

.

BALB

c

3T3

KiBALB

3T3

Py3T3

t

FIG. 2. Secretion of N-acetylglucosaminidase by control and transformed Balb/c3T3 cells stimulated by various factors. Fortyeight-hour cultures in six-well Falcon plates were processed as described under Materials and Methods. Concentrations of factors: (1) b-FGF, 20 rig/ml; (2) TGF-P, 1 pg/ml; (3) thrombin, 0.5 u/ml; (4) TNF, 100 u/ml; (5) EGF, 10 rig/ml. The total enzyme activities of Balb/c3T3, KiBalb3T3, and Py3T3 cells are 212 + 21, 151 f 18, and 136 nmol substrate hydrolyzed per 30 min per well, respectively. The corresponding “zero percent” baseline secretion values for the three cells are 8.5 f 0.6%, 7.7 f 0.9%, and 9.0% (mean + SEM). Balb/c3T3, KiBalb3T3, n = 8; Py3T3, n = 1.

SHORT

TABLE

1

Secretion of NAGA by Control and Transformed Cells; Percentage of Total Cellular Enzymes (NAGA) Secreted BRK-Ad-s?ElA (control)

Control + b-FGF + Thrombin + EGF

Q

Q4

Day 1

Day 4

Day 1

Day 4

Day 1

Day 4

PyMT Day 1

9.3 8.0 9.7 8.2

1.4 1.4 1.5 1. 7

15.3 26.0 21.1 16.3

9.7 13.3 10.8 8.7

5.5 10.1 16.5 8.6

4.9 5.0 5.2 7.1

6.8 9.9 11.0 11.4

ATote. Cells were plated, cultured, and assayed as described under Materials and Methods and in Ref. [2]. The control: TMsucroseNaCl, 150 mM. b-FGF, 20 rig/ml; thrombin 0.5 u/ml; EGF, 10.0 ng/ ml. Incubation 30 min, 37°C. Control cells: baby rat kidney cells immortalized with Ad-P-ElA; transformed cells: D, and D, are cell lines containing ElA transformed by T24 Ha-ras-1; PyMT is a line of cells containing ElA transformed by DNA coding for Polyoma middle T antigen [7].

EIA sequence with T24 Ha-ras (D2 and Q4), 1 day after p:ating the cells were responsive to the factors to different extents. These cells were positive in focus assays and were tumorigenic in syngeneic rats [8]; however, invasiveness of the cells was not tested. Again, total secretion fell and responsiveness to factors was greatly reduced by Day 4 when cell protein per plate had increased less than 20% in the previous 24 h. However, the ability of transformed cells to secrete NAGA in 4-day cultures was considerably greater than that in controls. Transformed cells co-transfected with plasmids containing DNA coding for Ad-2 EIA and polyoma middle T antigen (ElA+PyMT) [7] were also stimulated by various factors.

DISCUSSION While certain control, nontransformed cells used in this study barely responded to various factors (Fig. 2, Table l), secretion by PyS-2, a differentiated and nontumorigenic cell 191, was stimulated. It would appear that the lysosomal secretory response is not strictly confined to malignancy, but is more dependent on the growth state of the cell. However, PyS-2 cells are derived from F9, a malignant mouse stem cell [9] and could still retain some malignant characteristics of the progenitor. The ability of various factors, similar to autocrine and paracrine growth factors produced by tumors [lo-141, to stimulate secretion of lysosomal enzymes may explain the elevated enzyme levels found in the intercellu-

135

NOTE

lar spaces of tumors 13, 4, 15183 an tumor-bearing humans and animals [S lysosomal enzymes in the i~vasiv~~es~ been diseussed [4]. There is also detailed information on the association of extracellular c an array of degradative enzymes incl activator [ZO], with invasiveness. In of plasminogen activator and th man fibroblasts in culture is stimulate and EGF [21] ~ Through networking and crossed signals, unexpected consequences may follow the sti~~~atio~ of a specific receptor by an agonist such as a growth factor. A cascade of signals may be initiated that not o’nly affects growth but other processes, such as secretion, as well. It has been shown that the secretion of lysosomal enzymes is dependent upon c-AMP and G protein ~~~~:~ioning [2j common to several signal trams ction systems within the cell. Tumor necrosis factor, wbicb increases the lestimulates vel of c-AMP in human flbroblasts [22] .dependent their growth [23] could affect many c-A processes, including the secretion of lysosomal enzymes observed here. Malignant cells at the ~eripbe~~ of a tumor are usually those that are actively and ~ers~st~~~ly dividing and invading and, as the results of t&is study show, dividing cells are the ones most stim~~.ated to secrete. The same signal transduction pathways may be utilized for cell secretion and proliferation. Further, lactic acid, accumulating at the periphery of the tumor, results in a lowering of pH [23]. This may permit lysosomal acid hydrolases with low pH optima to operat lar substances more effectively in local could also alter the surface structure a mobility of cells. These are re isites for invasiveness and the formation of metastases. This work was supported by Grant CA19130 from the United States Public Health Service and the American Cancer Society, Grant RDPlSH. I acknowledge the excellent technical assistance of Rosalinda Espiritu, Chestine Bogan, and Elizabeth Tecza. I also thank Marie Lennon for secretarial service and Shirley Peterson for editorial review. Dr. H. E. Ruley generously provided the cultures of control and transformed rat kidney cells.

1.

Holtzman,

2. 3.

Warren, L. (1989) J. Biol. Chem. 264,8835-8842. Sylven, B., and Bois-Svennsson, I. (1965) Cc;ncer .Res. 28, 458468. Poole, A. R. (1970) Cancer Res. 30,2252-2259.

4. 5.

E. (1989) Lysosomes,

Plenum,

New York/London.

Poole, A. R. (1973) in Lysosomes in Biology and Pathology. (Dingle, J. T., Ed.), Vol. 3, pp. 303-337, Nor~b-~o~la~~~ Amsterdam.

136 6. 7. 8. 9. 10. 11. 12. 13. 14.

SHORT Bernacki, R. J., Niedbala, M. J., and Korytnyk, W. (1985) Cancer Metastasis Rev. 4,81-101. Alhadeff, J. A. (1989) CRC Grit. Rev. OncollHematol. 9, 37-107. Ruley, H. E. (1983) Nature (London) 304, 602-607. Lehman, J. M., Speers, W. C., Swartzendruber, D. E., and Pierce, G. B. (1974) J. Cell Physiol. 84, 13-18. GulIino, P. M., Clark, S. H., and Grantham, F. H. (1964) Cancer Res. 24, 780-798. Defalco, J., and Todaro, G. J. (1978) Proc. Natl. Acad. Sci. USA 75,4001-4005. Heldin, C. H., Betsholtz, C., Claesson-Welsh, L., and Westermark, B. (1987) Biochim. Biophys. Acta 907,219-244. Nicolson, G. L. (1988) Biochim. Biophys. Actu 948, 175-224. Weinberg, R. A. (1987) in Oncogenes and Growth Factors. (Bradshaw, R. A., and Prentis, S., Eds.), pp. 11-16, Elsevier Science, Amsterdam.

Received January 19,199O Revised version received May 2, 1990

NOTE 15. 16. 17.

18. 19. 20. 21. 22. 23.

Sylven, B., and Bois-Svensson, I. (1964) Histochemie 4, 135 149. Sylven, B. (1968) Eur. J. Cancer 4,463-474. Barrett, A. J. (1968) in Cartilage in Comprehensive Biochemistry (Florkin, M., and Stotz, E. H., Eds.), Vol. 26B, pp. 425-474, Elsevier Scientific, Amsterdam. Sylven, B., and Bois, I. (1960) Cancer Res. 20, 831-841. Liotta, L. A. (1986) Cancer Res. 46, l-7. Moscatelli, D., and Rifkin, D. B. (1988) Biochim. Biophys. Acta 948,67-85. Eaton, D. L., and Baker, J. B. (1983) J. Cell Biol. 9’7, 323-328. Zhang, Y., Lin, J.-X., Yip, Y. K., and Vilcek, J. (1988) Proc. Natl. Acad. Sci. USA 85,6802-6805. Vilcek, J., Palombella, V. J., Henriksen-DeStefano, D., Swenson, C., Feinman, R., Makoto, H., and Masafumi, T. (1986) J. Exp. Med. 163,632-643.