Increase in proliferative markers after inhibition of ... - Europe PMC

Proc. NatL. Acad. Sci. USA Vol. 78, No. 8, pp. 5005-5008, August 1981

Cell Biology

Increase in proliferative markers after inhibition of transglutaminase [isopeptide/E-(y-glutamyl)lysine/protein crosslinks/cystamine]

P. J. BIRCKBICHLER, G. R. ORR, M. K. PATTERSON, JR., E. CONWAY, AND H. A. CARTER Biomedical Division, The Samuel Roberts Noble Foundation, Ardmore, Oklahoma 73401

Communicated by Renato Dulbecco, May 4, 1981 ABSTRACT Cystamine inhibited transglutaminase activity (R-glutaminyl-peptide:amine y-glutamyltransferase, EC 2.3.2.13) of proliferating WI-38 cells in a dose-dependent manner over the concentration range 0.005-0.25 mM when added to the culture medium. The e-(y-glutamyl)lysine content in the cells was decreased and several proliferation markers were enhanced. "Nonmitotic" cells were stimulated by cystamine (about 25% of that observed with 10% fetal bovine serum) to undergo DNA synthesis with subsequent increases in nuclei number. Numerous other disulfides, thiols, and amines were ineffective when added to culture medium. The findings are supportive of the concept that growth control involves a relationship between isopeptide crosslinks and proliferation.

of isopeptide bonds, because reversal has not been readily apparent (8). Recently an enzyme capable of isopeptide cleavage has been observed, but there was no evidence that cleavage occurred prior to proteolytic breakdown (21). Since our model was proposed, a class ofTGase requiring Mg2+ATP in addition to Ca2+ and capable of raising and lowering isopeptide formation has been suggested (22, 23). The data reported here show that inhibition of TGase by cystamine has a growth-promoting effect. This result supports the hypothesis that crosslinking by TGase plays a role in cell growth regulation. Cystamine, an active site inhibitor (15), was tested, because a maximal proliferative response should be demonstrated when TGase activity is inhibited.

The biological role of protein crosslinks, especially those involving a covalent bond between peptide-bound glutamine and peptide-bound lysine, has been defined primarily in extracellular terms. Thus, fibrin clot stabilization (1-3), seminal fluid viscosity (4), and hair follicle (5) and epidermal (6) protein insolubility are correlated with the formation of an E-(y-glutamyl)lysine isopeptide bond between glutamine and lysine. We demonstrated the presence of the isopeptide bond in mouse L cell plasma membranes (7) and subsequently showed the occurrence ofthe enzyme transglutaminase (TGase; R-glutaminylpeptide:amine y-glutamyltransferase, EC 2.3.2.13), which catalyzes the formation of the bond (8) in eukaryotic cells maintained in culture (9). Recently, a wider intracellular occurrence of the isopeptide has been observed (10-12), and other investigators have suggested that TGase and protein crosslinking may be essential for receptor-mediated endocytosis (13, 14), erythrocyte membrane stiffening (15), terminal differentiation of keratinocytes (16), and lymphocyte blastogenesis (17). We have suggested that the level of protein crosslinks in the cell plasma membrane correlates with the proliferative state of the cell (9, 18) and have proposed a model (12) that depicts an increase in isopeptide bonds as the cell moves from a proliferative to a nonproliferative state (Fig. 1, reaction 1). For cells to remain in the proliferative state, isopeptide bonds must remain at low levels. One mode by which this might occur would be modulation at the enzyme level, either through the amount or activity of enzyme or intracellular Ca2+ concentrations (Fig. 1, reaction 3). Enzyme activity regulation via Ca2+ flux in erythrocytes has been demonstrated (15, 19). Another mode ofcontrol might be by substrate competition. For example, a polyamine could compete with the endogenous substrate, the E-amino group ofpeptide-bound lysine, to form a non-lysine isopeptide bond (Fig. 1, reaction 2). We, as well as others, have supplied evidence for a substrate function of putrescine, spermidine, and spermine in the TGase reaction (18, 20). This model emphasizes formation

MATERIALS AND METHODS WI-38 human lung cells (American Type Culture Collection, CCL 75) were inoculated into T-75 or T-25 flasks (Coming) or Leighton tubes (Bellco Glass) fitted with glass coverslips (10.5 x 35 mm) at a density of20,000-30,000 cells per cm2 and maintained by using modified McCoy's 5a medium (5.6 mM glucose) supplemented with 10% fetal bovine serum (Reheis, Kankakee, IL). Medium was removed and replaced with fresh medium every 48 hr unless otherwise specified. Cells for homogenization were washed three times with cold balanced salt solution (9), scraped lightly into cold calcium-free homogenization buffer (250 mM sucrose/0.2 mM MgSOJ5 mM Tris-HCl, pH 7.4) and homogenized with a mechanically driven Potter-Elvehjem homogenizer. TGase activity was determined by measuring the amount of 0.5 mM [1,4-'4C]putrescine (New England Nuclear, 0.7 ,uCi/ ,umol; 1 Ci = 3.7 x 1010 becquerels) incorporated into trichloroacetic acid-insoluble material after incubation of cell preparation and 10 mg of N,N-dimethylated casein (9). Nuclei were enumerated by using the citric acid/crystal violet method (24), isopeptide was quantitated after exhaustive proteolytic digestion (10), and protein was estimated by the method of Lowry et al. (25). [3H]Thymidine (New England Nuclear, 50-80 Ci/mmol) was added to the culture medium at a final concentration of 0.1 uCi/ ml. After incubation for 1 hr at 37°C, the medium was removed and the cell layer was washed with phosphate-buffered saline (137 mM NaCl/2.7 mM KCI/8 mM Na2HPO4/1 mM KH2PO4, pH 7.4) containing 4 mM thymidine. For thymidine incorporation the cell layer was extracted with 10% trichloroacetic acid for 30 min at room temperature, the liquid was removed, and the residue was extracted again with 10% trichloroacetic acid for 1 hr at 4°C. The residue was dissolved in 0.5 M NaOH by incubation for 2 hr at 37°C. Aliquots were taken for liquid scintillation counting and protein determination. Counting efficiency was 39%. The cpm were converted to dpm by the ex-

The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Abbreviation: TGase, transglutaminase. 5005

5006

Cell Biology: Birckbichler et al.

Proc. Natl. Acad. Sci. USA 78 (1981) NON-PROLI FERATION

PROLI FERATION HN

H2N-C=O

D I

IO

NH2

HNH 2N-C=O +

HN

Cystamine/-Ca++

H2N-C=O

NH2

NH2

H2N-C=O

A| C=O

(-

~'t HN

R-NH

!

R-N-C=O

HN (EC)

C=O

C=O

NH

NH

C=O

2

(EC) NH2

H

C C=O

NH2

H2N-C=O

FIG. 1. Model depicting isopeptide control by proliferating and nonproliferating cells. State A, lower level of isopeptide under proliferating conditions. State B, stabilization of cell by isopeptide formation under nonproliferating conditions. State C, maintenance of low isopeptide by control at substrate level. A substrate competitor (e.g., putrescine), vies for the carboxamide site, producing a non-lysine isopeptide. State D, maintenance of low isopeptide level by control at enzyme level. Inhibition of enzyme activity by direct inhibition of enzyme-e.g., cystamine or low levels of calcium. (EC) depicts the enzyme complex, consisting of TGase with an essential sulfhydryl group and calcium. ($C) depicts inhibition of the enzyme complex. (Taken from ref. 12 with permission.)

ternal standardization method using H number. Cells on coverslips were prepared for autoradiography by the method of Stanulis et aL (26). Aquasol (New England Nuclear) was used as the scintillator and slides were exposed overnight. Autoradiograms were analyzed microscopically by determining the percentage of cells with labeled nuclei (five or more grains) in random fields throughout the coverslip. At least 400 cells were counted on each coverslip and coverslips were prepared in duplicate. Cystamine dihydrochloride, histamine dihydrochloride, thymidine, diamide, monodansylcadaverine, and dithiothreitol were obtained from Sigma; L-cystine, from H & M Chemical (Santa Monica, CA); oxidized glutathione, from Schwarz/ Mann; methylamine hydrochloride, from Fisher; 2-mercaptoethanol, from Eastman Kodak; and.N,N-dimethylated casein, from Calbiochem. Statistical comparisons were determined by Student's t test. RESULTS AND DISCUSSION Overnight incubation of WI-38 cultures with cystamine led to a reduction in homogenate TGase activity. The degree of inhibition was dependent on the amount of cystamine added to culture medium over the range 0.005-0.25 mM (Fig. 2). Increasingly higher concentrations of cystamine did not result in substantially more TGase inhibition until the cystamine concentration exceeded 1 mM (10 mM = 98%). However, at these higher cystamine levels, the cells appeared to be irreversibly damaged as expressed by (i) trypan blue uptake, (ii) detachment of the cells from the culture flask, and (iii) failure to grow upon cystamine removal. Fifty to 60% inhibitions ofTGase, however, were obtainable without adverse cellular effects. Similar results were obtained when conditioned medium was used in iplace of fresh medium. Admixture experiments and studies with [3S]cystamine ruled out inhibition of enzyme activity as a result

of soluble cystamine carry-over to the enzyme assay mixture. Culture exposure to cystamine for longer than 24 hr did not greatly enhance the degree of TGase inhibition. Several proliferation markers and the cellular isopeptide content were examined after 0.1 mM cystamine addition to culture medium of preconfluent WI-38 populations (Table 1). Because of the several time points and variations in values from experiment to experiment, the data are expressed as the value found in cystamine-treated cells divided by that found in parallel untreated cells. [3H]Thymidine incorporation into acid-insoluble 75

° 50 |

X/

-05

g254 0.25

0.5 mM CYSTAMINE

0.75

1.0

2. Concentration dependence. of cystamine inhibition of hufibroblast TGase. Cultures were incubated in serum-supplemented modifed McCoy's 5a medium with and without cystamine. After 18-24 hr the medium was removed, and cells were homogenized and assayed for TGase activity. FIG.

man

Cell Biology: Birckbichler et al. Table 1. Proliferative markers after cystamine treatment Ratio (cystamine to control) Culture 1-2 days with 3'-7 days with state Marker cystamine cystamine Preconfluent, Thymidine 1.52 ± 0.19 (4)* 1.45 ± 0.15 (9)t Nuclei 1.06 ± 0.14 (4) 1.21 ± 0.07 (8)t Protein 0.98 ± 0.04 (4) 1.17 ± 0.05 (9)t 0.78 ± 0.01 (4)t 0.61 ± 0.04 (9)t Isopeptide Confluent

Thymidine

1.32 ± 0.14 (8)* 1.34 ± 0.06 (3)t 1.01 ± 0.06 (5) 1.10 ± 0.06 (5) 1.03 ± 0.02 (8) 0.98 ± 0.04 (5) 0.61 0.63 Isopeptide WI-38 cells in T-75 or T-25 flasks were maintained in medium supplemented with 10% fetal bovine serum. Cystamine (0.1 mM) was added to preconfluent cells 1-3 days after subculture and to confluent cells 7 days after subculture. Values are mean ± SEM, with the number of experiments in parentheses. *, P < 0.05; t, P < 0.01; t, P < 0.001 as compared to 1.0.

Nuclei Protein

material was higher in cystamine-treated cells as early as 18-24 hr after cystamine addition. Incubation with cystamine for 3-7 days resulted in cultures containing more nuclei and protein than comparable untreated cultures (Table 1). Thus, addition of cystamine at levels sufficient to give about 50% inhibition of TGase activity increased cell proliferation 20-30% in normal proliferating cells. The e-Qy-glutamyl)lysine isopeptide level was lower in preconfluent WI-38 cells exposed to cystamine for 24 hr compared to untreated cultures (Table 1). Exposure for 72 hr or longer led to a larger divergence in the isopeptide level ofthe two cultures. Because it is unlikely that cystamine catalyzes isopeptide hydrolysis (8, 21), the data are consistent with inhibition of isopeptide formation in cultures exposed to cystamine. The early increased thymidine incorporation after cystamine addition was not restricted to preconfluent cultures. Increased thymidine incorporation was also evident by 18-24 hr after cystamine addition to confluent cells (Table 1). Incubation with 0. 1 mM cystamine for more than 72 hr did not lead to consistently increased nuclei and protein ratios compared to untreated cultures. Nevertheless, isopeptide was reduced in confluent cells exposed to 0. 1 m M cystamine. If cystamine was acting as a mitogen in enhancing proliferation, then addition of cystamine to nonmitotic cells should also effect a positive proliferation response. When preconfluent "99nonmitotic" cells produced by serum deprivation were exposed to 0. 1 mM cystamine for 24 hr, they showed more [3H]thymidine incorporation into acid-insoluble material and more labeled nuclei after autoradiography than did cells not exposed to cystamine (Table 2). While cystamine was not as mitogenic as 10% fetal bovine serum, it was nonetheless mitogenic to WI-38 cells as evidenced by differences in the number of nuclei between cystamine-treated and untreated cells 72 hr after cystamine addition. The time course of DNA synthesis in unstimulated, 10% fetal bovine serum-stimulated,, and cystamine-stimulated populations of WI-38 cells was similar. DNA synthesis started around 12 hr, reached a maximum at about 24 hr, and declined thereafter. Thus the difference in thymidine incorporation between cystamine-treated and untreated cells was not the result of an alteration in the transit time of cells through S phase. Thymidine incorporation was increased to the same extent when cystamine was added to either preconfluent or confluent cells grown on 10% serum or "nonmitotic" cells maintained on 0.5% serum, suggesting that cystamine does not enhance the effect of some serum factor.

~Proc. NatL Acad. Sci. USA 78

(1981)

5007

Table 2. Enhancement of proliferative markers in "nonmitotic" cells after cystamine treatment

['H]Thymidine

Addition

incorporation, dpm, per coverslip

Nuclei

Labeled

10-3

nuclei, %

X

None 253 ±15 102 ±9 1.2 ±0.1 Fetal bovine serum, 10% 1203 ± 44'* 212 ± 5* 13.0 ± 2.5* 390 ± 43t 128 ± 7* 2.6 ± 0.2* Cystamine, 0.1 mM WI-38 cells were maintained on glass coverslips in Leighton tubes for 4 days in medium supplemented with 10% fetal bovine serum. "Nonmitotic" cell populations were obtained by removing the medium, replacing it with medium containing no fetal bovine serum, and incubating for 72 hr. Medium was removed and cells were incubated with medium containing 0.5% fetal bovine serum and the additions listed. Cells for measurement of thymidine incorporation and percent labeled nuclei were incubated for 24 hr-, whereas cells for nuclei enumeration were incubated for 72 hr. Values are mean ± SEM of three experiments with samples run in duplicate. * , P < 0.001; t, P < 0.02; t, P < 0.05 as compared to sample receiving no addition.

The observation that both confluent (Table 1) and "nonmitotic" (Table 2) cells showed [3H]thymidine stimulation after cystamine addition but only "nonmitotic" cells showed proliferation suggests that cell crowding may play a role in growth control with consequences different from serum reduction. The number of cells remaining in the G2 period of the cell cycle for at least 24 hr after DNA synthesis has been shown to increase with cell density (27). Similarly, several reports have demonstrated that DNA synthesis can occur independent of cell di-

vision (28-32). In an effort to determine if cystamine inhibition of TGase

activity was specific or a general effect of disuffide exchange, confluent WI-38 cells were incubated with several disulfide and thiol reagents (Table 3). Cystamine was the only material examined that produced a significant inhibition of TGase activity and an increase in DNA synthesis. Diamide (a thiol-oxidizing agent) and oxidized glutathione, while lowering the TGase activity slightly, showed no concentration dependency. Concentrations of diamide greater than 0.5 mM were cytotoxic. Table 3. TGase activity after treatment with various disulfides, thiols, and amines TGase activity, Labeled Conc., units/mg Addition mm protein nuclei, % None 21.4 ± 1.7 (4) 9.6 ± 1.2

Disulfide/thiol Cystamine Cystine Diamide Glutathione Dithiothreitol

2-Mercaptoethanol

0.1 0.5 0.05-0.5 0.01-0.1 0.1 0.1

12.1 ± 0.5 (4)* 23.2 ± 1.2 (3) 17.8 ± 0.5 (5) 18.6 ± 1.6 (3) 23.2, 21.9 23.8, 24.0

20.0 ± 1.4* 13.7 ± 1.0 11.6 ± 1.9 ND ND ND

Amine

Dansylcadaverine Methylamine

0.05 23.9 ± 1.2 (4) 9.9 ± 1.1 0.5 22.9 ± 1.6 (3) 9.3 ± 0.7 Histamine 0.5 ND 24.4, 25.7 Confluent WI-38 cells were treated with medium supplemented with 10% fetal bovine serum and the additions listed. After incubation for 24 hr at 37"C the cells were processed for TGase activity and autoradiography. Values are mean ±- SEM, with number of determinations in parentheses. ND, not determined. * , P < 0.005 as compared to sample receiving no addition.

5008

Cell Biology: Birckbichler et aLPProc. Nad Acad. Sci. USA 78 (1981)

Because cystamine also contains primary amino groups in addition to the disulfide bond, several primary amines were examined for their ability to inhibit TGase activity. When added to culture medium, dansylcadaverine, methylamine, and histamine did not lower WI-38 TGase activity (Table 3). Fluorometry and studies with radiolabeled methylamine showed this was not the result of impaired amine transport into the cell. These amines were active in inhibiting [14C]putrescine incorporation into dimethylcasein when added directly to an assay mixture. The mechanism of cystamine inhibition cannot be ascertained with certainty from these studies. The failure of dansylcadaverine, methylamine, and histamine to inhibit TGase activity when added to culture medium suggests the cystamine effect did not reside in the amine region of the molecule. Cystamine stimulated proliferation of mouse lymphoma and mitogen-activated mouse spleen lymphocytes, whereas several other disulfide compounds were ineffective (33). It would be inappropriate to assume that cystamine is specific for TGase (34-36), but the results with the various disulfide and thiol reagents suggest an affinity of WI-38 TGase for cystamine and appear to be compatible with inhibition via disulfide exchange at the active site thiol group of TGase as proposed by Siefring et al. (15). Further support for such an interaction comes from our preliminary studies showing binding of [3S]cystamine to purified guinea pig liver TGase primarily through disulfide exchange (unpublished data). The observation that only about 50% of the WI-38 TGase activity can be inhibited without irreversible cellular damage is intriguing. Confluent WI-38 cells have TGase activity equally distributed between soluble and particulate fractions. The question arises whether one of these fractions is preferentially inhibited by cystamine. Efforts to establish such a relationship have been unsuccessful. Soluble and particulate TGase activity were both inhibited when cystamine was added to an assay mixture. It is possible, however, that homogenization alters the normal configuration of the enzyme and opens sites not available to cystamine in the intact cells in a manner akin to trypsin treatment (9, 18). Whether cystamine is a biological control molecule for isopeptide formation intracellularly is presently unknown; however, its use as an inhibitor of TGase activity has provided a means to study the effects on cell replication produced by inactivation of the enzyme. Inhibition of TGase activity was accompanied by decreased cellular isopeptide content and increases in proliferative markers. Preliminary findings indicate that, with essentially total recovery, greater than 90% of the total cellular isopeptide was found in a plasma membrane-rich fraction. A similar fraction isolated from a parallel sample treated with cystamine had 70% less isopeptide (unpublished data). The data therefore support the model (Fig. 1) showing a relationship between isopeptide bonds and proliferation. 1. Matacic, S. S. & Loewy, A. G. (1968) Biochem. Biophys. Res. Commun. 30, 356-362. 2. Pisano, J. J., Finlayson, J. S. & Peyton, M. P. (1968) Science 160,

892-893.

3. Lorand, L., Rule, N. G., Ong, H. H., Furlanetto, R., Jacobsen, A., Downey, J., Oner, N. & Bruner-Lorand, J. (1968) Biochemistry 7, 1214-1223. 4. Williams-Ashman, H. G., Notides, A. C., Pabalan, S. S. & Lorand, L. (1972) Proc. Nati. Acad. Sci. USA 69, 2322-2325. 5. Harding. H. W. J. & Rogers, G. E. (1971) Biochemistry 10, 624-630. 6. Abernethy, J. L., Hill, R. L. & Goldsmith, L. A. (1977) J. Biol Chem. 252, 1837-1839. 7. Birckbichler, P. J., Dowben, R. M., Matacic, S. & Loewy, A. G. (1973) Biochim. Biophys. Acta 291, 149-155. 8. Folk, J. E. & Finlayson, J. S. (1977) Adv. Protein Chem. 31, 1-133. 9. Birckbichler, P. J., Orr, G. R., Conway, E. & Patterson, M. K., Jr. (1977) Cancer Res. 37, 1340-1344. 10. Birckbichler, P. J., Carter, H. A., Orr, G. R., Conway, E. & Patterson, M. K., Jr. (1978) Biochem. Biophys. Res. Commun. 84, 232-237. 11. Matacic, S. & Loewy, A. G. (1979) Biochim. Biophys. Acta 576, 263-268. 12. Birckbichler, P. J. & Patterson, M. K., Jr. (1980) Prog. Clin. Bol. Res. 41, 845-855. 13. Davies, P. J. A., Davies, D. R., Levitzki, A., Maxfield, F. R., Milhaud, P., Willingham, M. C. & Pastan, I. H. (1980) Nature (London) 283, 162-167. 14. Levitzki, A., Willingham, M. & Pastan, I. (1980) Proc. Natl Acad. Sci. USA 77, 2706-2710. 15. Siefring, G. E., Jr., Apostol, A. B., Velasco, P. T. & Lorand, L. (1978) Biochemistry 17, 2598-2604. 16. Rice, R. H. & Green, H. (1977) Cell 11, 417-422. 17. Novogrodsky, A., Quittner, S., Rubin, A. L. & Stenzel, K. H. (1978) Proc. Natl Acad. Sci. USA 75, 1157-1161. 18. Birckbichler, P. J. & Patterson, M. K., Jr. (1978) Ann. N.Y. Acad. Sci. 312, 354-365. 19. Lorand, L., Weissmann, L. B., Epel, D. L. & Bruner-Lorand, J. (1976) Proc. Natt Acad. Sci. USA 73, 4479-4481. 20. Folk, J. E., Park, M. H., Chung, S. I., Schrode, J., Lester, E. P. & Cooper, H. L. (1980) J. Biol Chem. 255, 3695-3700. 21. Fink, M. L., Chung, S. I. & Folk, J. E. (1980) Proc. Natl Acad. Sci. USA 77, 4564-4568. 22. Loewy, A. G. & Matacic, S. S. (1981) Biochim. Biophys. Acta 668, 167-176. 23. Loewy, A. G., Matacic, S. S., Rice, P. & Stern, J. (1981) Biochim. Biophys. Acta 668, 177-185. 24. Absher, M. (1973) in Tissue Culture: Methods and Applications, eds. Kruse, P. F., Jr. & Patterson, M. K., Jr. (Academic, New York), pp. 395-397. 25. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.

(1951) 1. Biol Chem. 193, 265-275. 26. Stanulis, B. M., Sheldon, S., Grove, G. L. & Cristofalo, V. J. (1979) J. Histochem. Cytochem. 27, 1303-1307. 27. Macieira-Coelho, A. (1973) Front. Matrix Biol 1, 46-77. 28. Smith, B. J. & Wigglesworth, N. M. (1972) J. Cell Physiol 80, 253-260. 29. Hatzfeld, J. & Buttin, G. (1975) Cell 5, 123-129. 30. Thompson, L. H. & Lindl, P. A. (1976) Somatic Cell Genet. 2, 387-400. 31. Wissinger, W. & Wang, R. J. (1978) Exp. Cell Res. 112, 89-94. 32. Van Wijk, R., Zoutewelle, G. & van de Poll, K. W. (1979) Cell Biol Int. Rep. 3, 447-452. 33. Broome, J. D. & Jeng, M. W. (1973)J. Exp. Med. 138, 574-592. 34. Griffith, 0. W., Larsson, A. & Meister, A. (1977) Biochem. Biophys. Res. Commun. 79, 919-925. 35. Lebo, R. V. & Kredich, N. M. (1978) J. Biol Chemr. 253, 2615-2623. 36. Namboodiri, M. A. A., Favilla, J. T. & Klein, D. C. (1979) Biochem. Biophys. Res. Commun. 91, 1166-1173.