Identifying inhibitors of queuine modification of tRNA in cultured cells

ANALYTICAL

BIOCHEMISTRY

Identifying

171,346-35

1 (1988)

Inhibitors of Queuine Modification

of tRNA in Cultured

GIRIJA MURALIDHAR,ERICD.UTZ,MARK S.ELLIOTT,*JON ANDRONALDW.TREWYN

Cells’

R. KATZE,*

Department ofPhysiological Chemistry. The Ohio State University. Columbus, Ohio 43210, and *Department of Microbiology and Immunology, University of Tennessee Health Sciences Center, Memphis, Tennessee 38163 Received July 23, 1987 Altered queuine modification of tRNA has been associated with cellular development, differentiation, and neoplastic transformation. Present methods of evaluating agents for their ability to induce queuine hypomodification of tRNA are tedious, time-consuming, and not readily amenable to examining cell-type or tissue specificity. Therefore, a rapid, small-scale assay was developed to identify agents that alter queuine modification of tRNA in cultured cells. Monolayer cultures (2 cm*) of Chinese hamster embryo cells depleted of queuine for 24 h were evaluated for their ability to incorporate [3H]dihydroqueuine into acid precipitable material (tRNA) in the presence and absence of potential inhibitors. Known inhibitors of the queuine modification enzyme tRNA-guanine ribosyltransferase (e.g., 7-methylguanine, 6-thioguanine, and 8-azaguanine) were very effective in blocking incorporation of the radiolabel, and the dose-dependent results exhibited small standard deviations in independent experiments. The data indicate that the method is rapid, reliable, and potentially useful with a variety of cell types. 8 1988 Academic Press. Inc. KEY WORDS: tRNA modification; queuine; bioassay; cell culture; purines; nucleic acid chemistry.

Queuine,2 a highly modified analog of guanine, is found exclusively in the first (wobble) position of anticodon of tRNA isoacceptors for aspartic acid, asparagine, histidine, and tyrosine (1,2). Queuine is inserted for guanine at this position by a base exchange reaction catalyzed by the enzyme tRNA-guanine ribosyltransferase (EC 2.4. 2.29) (3,4). This post-transcriptional modification is dependent on the availability of queuine in the cell. Since mammals cannot synthesize queuine, they must obtain it from their diet or gut flora (5), while mammalian cells grown in culture obtain queuine from the sera utilized to supplement most culture media (6). ’ Supported by Grant R8 107 14-02 (R.W.T.) from the Environmental Protection Agency and Grants CA-20919 (J.R.K.) and P30-CA-16058-14 (OSUCCC) from the National Cancer Institute, Department of Health and Human Services. ’ Queuine, 7-{ { 5-[( l.S,4S.SR)-4,5-dihydroxy-2-cyclopenten-I-yl-amino]methyl}}-7-deazaguanine. 0003-2697/88 $3.00 Copyright

0 1988 by Academic

It is well documented that queuine metabolism in cancer cells is altered and that tumor cell tRNA is often queuine hypomodified (5,7- 10). Inhibition of queuine modification of tRNA has been implicated in promotion of carcinogenesis in vitro (11,12) and in the maintainance of the malignant phenotype in vivo (8). In addition, present knowledge of queuine metabolism indicates that queuine modification of tRNA may be essential for normal cellular development, differentiation, and function (13-l 7). Since tRNA plays a pivotal role in the translation of the genetic code, it is conceivable that structural changes, such as those involving queuine modification, could lead to alterations in codon-anticodon recognition and, thereby, alterations in protein synthesis. Therefore, it is essential to establish the queuine status of tRNA in cells and to identify potential inhibitors of queuine modification. Currently, there are two basic approaches to determine the queuine status of tRNAs 346

Press. Inc.

QUEUINE

MODIFICATION

and/or to test compounds for their ability to induce queuine hypomodification of tRNA. The first approach is to isolate tRNA from cells with or without treatment with compounds of interest and to assay the queuine status of the tRNAs in vitro using the tRNAguanine ribosyltransferase isolated from some source (3,7). This reaction is based on the fact that once queuine is inserted into tRNAs, the reaction is essentially irreversible. However, guanine in the wobble position can be exchanged by radiolabeled guanine, queuine, or dihydroqueuine, and the incorporation of radiolabel into tRNA then gives an indirect measure of queuine hypomodification. Another variation of this approach is to take advantage of differences in chromatographic patterns of tRNAs containing queuine or guanine. With this method, tRNA is aminoacylated with aspartic acid, asparagine, histidine, or tyrosine, and the tRNA isoacceptors are separated using reversed-phase chromatography (6). An extension of this method involves treating tRNA samples with periodate or cyanogen bromide which oxidizes the cyclopentenediol ring of queuine and thereby changes the elution pattern (18). With either variation of this assay method, there is the requirement for large-scale isolation of tRNA, which is a multi-step process, and the subsequent techniques are very time-consuming. The second approach involves purification of tRNA-guanine ribosyltransferase from cells or tissue (e.g., rat liver, rabbit reticulocytes, or Escherichia coli) and assaying its activity in the presence of potential inhibitors in vitro (4,19,20). The most obvious difficulty with this procedure is the isolation and purification of the enzyme, since a relatively pure enzyme preparation is required. Moreover, compounds of interest could have a different effect depending on the source of the enzyme, which makes it difficult to apply the findings from this assay to other systems. With increasing knowledge about the involvement of tRNA structural changes in neoplasia and cell differentiation, it will be crucial to have convenient, small-scale assay

OF

347

tRNA

protocols to evaluate compounds for their ability to induce queuine hypomodification of tRNA. We report here a new analytical method for screening compounds that inhibit queuine modification of tRNA in cultured cells. MATERIALS

AND

METHODS

Cell culture. Pregnant Chinese hamsters (Cambridge Diagnostics, Cambridge, MA) were sacrificed and primary embryo cultures were prepared and stored at -70°C for later use as previously described (2 1). Frozen ChH cells were thawed to room temperature and seeded in Eagle’s minimum essential medium (MEM)3 (GIBCO, Grand Island, NY) supplemented with 5% fetal bovine serum (FBS) (Sterile Systems, Logan, UT). The cells were then incubated at 37°C in a 4% CO?humidified atmosphere for 24 h after which they were fed with fresh medium and incubated until they reached confluency. [3H]Dihydroqueuine incorporation assay. ChH cells at passage two or three were seeded at a density of 10,000 per well of 2-cm2 fourwell plates (Nunc) in a total volume of 0.5 ml of MEM containing 5% FBS. Once the cells reached confluency (usually 3 days after seeding), each well was fed with 0.2 ml of MEM containing 5% charcoal-treated FBS (charcoal removes queuine from the serum; see Ref. (6)) for a period of 24 h. Following this treatment, the cells were fed with MEM supplemented with 5% charcoal-treated FBS, test compound at the desired concentration, and 0.25-0.30 PM [3H]dihydroqueuine (3.5 Ci/mmol) (prepared by catalytic reduction of queuine with tritium gas by NEN Research Products, Boston, MA) in a total volume of 0.1 ml. After incubating at 37°C for an appropriate period (O-6 h), the medium containing radiolabeled reduced queuine was aspirated, and the cell monolayer was washed once with calcium and magnesium-free ’ Abbreviations dium; FBS, fetal acid.

used: MEM, minimal bovine serum: TCA.

essential metrichloroacetic

MURALIDHAR

348

phosphate-buffered saline. The cells were lysed using 0.2 ml of lysis buffer [ 140 mM NaCl, 1.5 InM MgC12, 10 mM Tris-HCl (pH 8.6), and 0.5% NP-401 after which the cell lysate was maintained at 4°C for at least f h. Radiolabelled material incorporated into tRNA was precipitated with an equal volume of 10% ice-cold trichloroacetic acid (TCA), and the acid-insoluble material was collected on glass fiber filters with three additional washes with ice-cold 5% TCA. Filters were then washed one time with 95% ethanol, and radioactivity was measured by liquid scintillation counting. Purine analogs. Queuine was isolated from bovine amniotic fluid (Irvine Scientific, Santa Ana, CA) using published protocols (22,23). All guanine and hypoxanthine analogs, except 1-methylguanine and 7-methylguanine, were purchased from Sigma Chemical Co. (St Louis, MO), whereas l-methylguanine and 7-methylguanine were from Fluka AG (Buchs, Switzerland). RESULTS

Incorporation of radiolabeled dihydroqueuine into tRNA of ChH cells under different conditions was monitored for 4 h, and the results are shown in Fig. 1. Untreated

FIG. 1. [3H]Dihydroqueuine incorporation into ChH cells treated with queuine or methylated purines. Queuine-depleted ChH cells were treated with 0.3 pM [3H]dihydroqueuine in MEM containing 5% charcoaltreated FE%Splus no addition(O), 0.5 pM queuine (O), 10 pM I-methylguanine (A), or 10 pM 7-methylguanine (A). Each point represents a mean of triplicate samples. See Materials and Methods for additional details.

ET AL.

control ChH cells starved for queuine for 24 h were capable of incorporating the reduced queuine into tRNA as evidenced by the increasing radioactivity in acid-insoluble material. Similar results were obtained when the cell lysate was treated with proteinase K prior to TCA precipitation, but the combination of RNase A and P reduced the incorporation of precipitable radiolabel to background levels (data not shown). When unlabelled queuine or 7-methylguanine was added to the cells along with radiolabeled dihydroqueuine, there was also no detectable incorporation of radiolabel (Fig. 1). Another methylated purine, lmethylguanine, did not inhibit the uptake or incorporation of radiolabeled reduced queuine into TCA precipitable material. The reaction was linear for 4 h in the untreated control cells (Fig. I), and in separate experiments, linear incorporation of [3H]dihydroqueuine was observed for up to 6 h (data not shown). The dose dependence of the queuine and 7-methylguanine inhibition observed in Fig. 1 was then evaluated. At concentrations of 5 and 0.5 PM, queuine was capable of inhibiting [3H]dihydroqueuine incorporation into tRNA very efficiently with the degree of inhibition being essentially 100% (Fig. 2). Even when the concentration of queuine was decreased to 0.05 PM, the incorporation of radiolabeled reduced queuine was still less than 10% of control. However, as the concentration of 7-methylguanine was decreased from 5.0 to 0.05 PM, there was an increase in the incorporation of radiolabeled dihydroqueuine into tRNA, starting from less than 10% to up to 80% of control (Fig. 2). As shown in Fig. 3, two known substrates for tRNA-guanine ribosyltransferase, 6-thioguanine and 8-azaguanine, were also very potent inhibitors in the small-scale assay, with almost 90% inhibition at the highest concentration tested (500 PM) and approximately 40% inhibition at the lowest concentration (5 PM). While 8-azaguanine inhibited the incorporation of [3H]dihydroqueuine into TCA precipitable material very effec-

QUEUINE

MODIFICATION

FIG. 2. Inhibition of [3H]dihydroqueuine incorporation into ChH cell tRNA by queuine and 7-methylguanine. Queuine-depleted ChH cells were fed with 0.25 pM [‘Hldihydroqueuine in MEM containing 5% charcoal-treated FBS plus queuine or 7-methylguanine at 0.05, 0.5, or 5.0 FM. Incorporation of radiolabel was measured in duplicate cultures after a 4-h incubation period as described under Materials and Methods. Mean values (*SD) from three independent experiments are presented as percentage of control.

tively at the concentrations tested, its hypoxanthine analog had lesser effects, with only 50% inhibition at 500 PM (Fig. 3). The possibility of using the small-scale assay to screen compounds as inhibitors of queuine modification of tRNA was examined with a variety of natural and synthetic

OF tRNA

349

purine analogs. Results presented in Fig. 4 indicate that among the naturally occurring compounds, the only one that inhibited dihydroqueuine incorporation into tRNA appreciably was hypoxanthine (80% inhibition); xanthine and uric acid had very little inhibitory capacity at the concentration tested (50 FM). Allopurinol and oxypurinol were not particularly effective in blocking the incorporation of ]3H]dihydroqueuine, with about 40 and 20% inhibition, respectively. However, among the synthetic analogs of hypoxanthine, 6-mercaptopurine was almost as effective as hypoxanthine in blocking incorporation of radiolabel, whereas 6-ethylmercaptopurine had only a minimal effect. Another purine analog tested, 2,6-diaminopurine, inhibited the incorporation by approximately 60%. DISCUSSION

In this paper we report a novel, rapid method for identifying inhibitors of queuine modification of tRNA. The only known metabolic fate of queuine (or dihydroqueuine) is incorporation intact into tRNA (22), and the enzymatic incorporation of 7-deazaguanine analogs such as queuine and dihydroqueuine is essentially irreversible ( 19). It was demon-

I +

i

(H)

FIG. 3. Inhibition of [‘Hldihydroqueuine incorporation into ChH cell tRNA by 6-thioguanine, I-azaguanine, and 8-azahypoxanthine. Conditions are as described in the legend to Fig. 2, but with the purine analogs included at 5,50, and 500 PM. A queuine (0.5 pM) control was also included in each experiment to verify 100% inhibition.

FIG. 4. Inhibition of [‘Hldihydroqueuine incorporation into ChH cell tRNA by purine analogs. The analogs evaluated at a concentration of 50 pM were (A) hypoxanthine, (B) xanthine, (C) uric acid, (D) allopurinol, (E) oxypurinol, (F) 6-mercaptopurine, (G) 6-ethylmercaptopurine, and (H) 2,6-diaminopurine. Conditions are as described in the legends to Figs. 2 and 3.

350

MURALIDHAR

ET AL.

strated previously that dihydroqueuine is long as the results are calculated as percentneither a substrate for the purine salvage en- age of control. zymes nor is it converted to the nucleoside or The purine antimetabolites 6-thioguanine nucleotide level in mammalian cells other and 8-azaguanine have been reported to be than by incorporation into tRNA and subse- excellent substrates for tRNA-guanine riboquent degradation (turnover) of that macrosyltransferase in vitro and in vivo (4,19,20), molecule (24). Although some mammalian and these agents proved to be very good incells have a specialized queuine salvage hibitors in the small-scale assay (Fig. 3). The pathway, essentially all dihydroqueuine me- 8-aza analog of hypoxanthine was much less tabolites could be accounted for via process- effective. The two guanine analogs are used ing through tRNA (24,25). Therefore, it was routinely to select for mutants lacking the possible to assess [‘Hldihydroqueuine modipurine salvage enzyme hypoxanthine-guafication of tRNA in the current study by nine phosphoribosyltransferase, and since monitoring incorporation of radiolabel into other (undefined) cellular changes have been TCA precipitable material, since it was veri- implicated in the mutant selection procedure fied that the radiolabeled macromolecule (27,28), effects on the queuine modification was sensitive to RNase digestion. in the anticodon wobble position of tRNA It was reported previously that 7-methylcould be involved. Examining such possibiliguanine, but not 1-methylguanine, induces ties should be facilitated by the assay procequeuine hypomodification of tRNA in ChH dure reported here. cells under conditions leading to in vitro Using the small-scale [3H]dihydroqueuine transformation (26). The earlier work re- incorporation assay to screen for inhibitors quired the isolation of tRNA from cells in of queuine modification of tRNA was also 8- 16 150~cm* flasks for each treatment, fol- demonstrated (Fig. 4), but certain incongruilowed by assaying for radiolabeled guanine ties with previously published reports are apincorporation by the tRNA-guanine ribosylparent. Hypoxanthine and 6-mercaptoputransferase from E. coli. Data from Fig. 1 rine were not expected to be inhibitors based demonstrate essentially the same result using on in vitro studies with the tRNA-guanine our small-scale assay. As expected, queuine, ribosyltransferases from rabbit reticulocytes which is reportedly a lo- to 50-fold better (19) and rat liver (4) but obviously, both substrate for incorporation into tRNA than were very effective inhibitors in cultured dihydroqueuine (5,19), was an effective in- ChH cells (Fig. 4). While this could represent hibitor in the assay as well (Fig. 1). cell-type differences for analog insertion, it is The dose dependence of the inhibition ob- also possible that hypoxanthine and 6-merserved in Fig. 1 was evaluated, and with the captopurine alter queuine transport (29) or queuine concentration fivefold lower than that cellular metabolites of these agents are dihydroqueuine (0.05 versus 0.25 PM), some involved. In this regard, hypoxanthine (at 20 incorporation of radiolabel (ca. 5% of con- pM) is not an inhibitor of queuine transport trol) was obtained (Fig. 2). Significantly in human skin fibroblasts (29). more 7-methylguanine was required to inIt is interesting that hypoxanthine, 6-merhibit the incorporation of [3H]dihydrocaptopurine, 6-thioguanine, 8-azaguanine, queuine, but the inhibition was clearly and 2,6-diaminopurine have all been implicated in inducing the differentiation of mudependent on the concentration of 7-methylrine erythroleukemia cells and/or human guanine included (Fig. 2). The reproducleukemia cells in vitro ibility of the small-scale assay procedure can promyelocytic also be seen in Fig. 2, where the standard (13,30,31), and all of these purine analogs deviations for three totally independent ex- were good inhibitors of [3H]dihydroqueuine incorporation in the small-scale assay (Figs. 3 periments are depicted. Therefore, the small (2 cm*) monolayer cultures are sufficient so and 4). Furthermore, xanthine and uric acid

QUEUINE

MODIFICATION

do not induce the differentiation of murine erythroleukemia cells (30), and they were not inhibitors in the ChH cell assay. Of more direct relevance, ‘I-methylguanine has been shown to enhance the chemical transformation of ChH cells in vitro ( 12) and it was an excellent inhibitor in the same cells (Figs. 1 and 2). While the use of cultured cells in the small-scale assay does not allow one to pinpoint the site of action of inhibitors on queuine metabolism (i.e., whether they are altering transport, salvage, or insertion), it has the distinct advantage of being rapid and amenable to a variety of cell types. For example, we have already applied the technique to monolayer cultures of human skin fibroblasts (M. S. Elliott and J. R. Katze, unpublished data) and, with minor modifications, to suspension cultures of human leukemia cells (D. S. Gibboney and R. W. Trewyn, submitted for publication). Therefore, it should be possible to apply the bioassay procedures directly to studies of neoplasia and cell differentiation, and if necessary, the specific site of action of the inhibitors could be elucidated subsequently.

8. 9. IO.

Il. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

ACKNOWLEDGMENTS We thank Edith F. Yamasaki and Dawn E. Patrick for expert technical assistance.

23. 24.

REFERENCES 1. Harada, F.. and Nishimura, S. (1972) Biochemistry l&301-308.

2. Nishimura. S. (1983) Prog. Nucleic Acid Rex Mol. Biol. 28, 49-80. 3. Katze, J. R., and Farkas, W. R. (1979) Proc. Natl. Acad. Sci. USA 76, 327 l-3275. 4. Shindo-Okada. N.. Okada. N., Ohgi, T., Goto, T., and Nishimura. S. (1980) Biochemistry 19, 395-400. 5. Katze, J. R., Beck. W. T., Cheng, C. S., and McCloskey, J. A. (1983) Recent Results Cancer Rex 84, 146-159. 6. Katze, J. R. (1978) Biochem. Biophys. Rex Commun. 84,527-535. 7. Okada. N., Shindo-Okada. N., Sato. S., Itoh, Y.,

25 26. 27. 28.

29. 30. 31.

OF tRNA

351

Oda. K., and Nishimura. S. (1978) Proc. Natl. Acad. Sri. USA l&4247-425 1. Katze, J. R., and Beck, W. T. (1980) Biochem. Biophys. Rex Commun. 96, 3 13-3 19. Randerath, E., Agrawal. H. P.. and Randerath, K. (1984)CancerRes.44, 1167-1171. Emmerich. B., Zubrod. E.. Weber, H., Maubach, P. A., Kersten. H.. and Kersten W. (1985) Cancer Res. 45,4308-43 14. Elliott. M. S.. Katze, J. R., and Trewyn, R. W. ( 1984) Cancer Rex 44, 32 15-32 19. Muralidhar, G., and Trewyn, R. W. (1987) Cancer Rex 47,2440-2444. Kretz, K. A.. Katze. J. R.. and Trewyn, R. W. ( 1987) Mol. Cell. Biol. 7, 36 13-3619. Schachner. E.. and Kersten, H. (1984) J. Gen. Microbiol. 130, 135-144. Shindo-Okada, N.. Terada, M.. and Nishimura. S. (1981) Eur. J. Biochem. 115, 423-428. Jlnel, G.. Michelson. U., Nishimura. S., and Kersten. H. (1984) EMBO J. 3, 1603-1608. Schachner. E., Aschhoff. H. J., and Kersten. H. ( 1984) Eur. J. Biochem. 139,48 I-487. White, B. N. (1974) Biochim. Biophys. Acfa 353, 283-29 I. Farkas, W. R.. Jacobson, B. K., and Katze, J. R. (1984) Biochim. Biophys. Acta 781, 64-75. Okada, N., and Nishimura, S. (I 979) J. Biol. Chem. 254,306 I-3066. Trewyn, R. W., and Kerr, S. J. ( 1978) Cancer Res. 38,2285-2289. Katze, J. R., Gtindiiz., Smith. D. L., Cheng, C. S., and McCloskey, J. A. (1984) Biochemistry 23, 1171-l 176. Reyniers, J. P., and Farkas. W. R. (1983) Anal. Biothem. 130,427-430. Gtindtiz. U.. and Katze. J. R. (I 984) J. Biol. Chem. 259, 1110-1113. Giindiiz, U., and Katze, J. R. (1982) Biochem. Biophys. Res. Commun. 109, 159-167. Elliott, M. S., and Trewyn, R. W. (I 982) Biochem. Blophys. Res. Comm. 104, 326-332. Hsie, A. W., Brimer, P. A., Machanoff, R., and Hsie, M. H. (1977) Mutat. Rex 45, 271-282. Gallagher. R. E.. Ferrari, A. C., Zulich, A. W., Yen, R. C.. and Testa, J. R. (1984) Cancer Res. 44, 2642-2653. Elliott, M. S., Trewyn, R. W., and Katze, J. R. (1985) Cancer Res. 45, 1079-1085. Gusella, J. F., and Housman. D. (1976) Cell 8, 263-269. Schwartz, E. L., Blair, 0. C., and Sartorelli, A. C. ( 1984) Cancer Res. 44, 3907-39 10.