Dependence of depurination of ... - Wiley Online Library

BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL Pages285-291

Vol. 39, No. 2, May 1996

DEPENDENCE OF DEPURINATION OF OLIGORIBONUCLEOTIDES BY RICIN A-CHAIN ON DIVALENT CATIONS AND CHELATING AGENTS Anton Gliick and Ira G. Wool*

Department of Biochemistry and Molecular Biology, The University of Chicago, Chicago, IL 60637, U.S.A. Received March 6, 1996 Received after revision March 11, 1996

Summary: Ricin A-chain is a cytotoxic RNA N-glycosidase that inactivates eukaryotic ribosomes by depurinating the adenosine at position 4324 in 28S rRNA. The enzyme retains its specificity when a synthetic oligoribonucleotide (a 35-mer) that mimics the structure at the site of action is the substrate. However, covalent modification by ricin A-chain of the oligoribonucleotide but not of ribosomes, depends on the simultaneous presence of a divalent cation and a chelating agent. keywords: ricin A-chain, oligoribonucleotide, RNA protein interaction

Introduction Ricin A-chain (RA 1) is a ribotoxin that inactivates eukaryotic ribosomes by depurinating the adenosine at position 4324 in 28S rRNA (1,2). This single covalent modification accounts entirely for the toxicity of RA. The toxin retains its specificity when a synthetic oligoribonucleotide, a 35-mer, that mimics the structure at the site of action in ribosomes is the substrate (cf. Fig. 1) (3). The reaction is less efficient with the oligoribonucleotide than with ribosomes: The kcat with ribosomes is 1777 min -1 and the Kin 2.6 pM (4), whereas the kcat with the oligonucleotide is 0.06 rain -1 and the K m 16.9 /aM (5). The activity of RA with oligoribonucleotides has routinely been assessed in buffer containing 3 mM MgC12 and 2.5 mM EDTA (6,7). The oligoribonucleotides, synthesized in vitro with T7 RNA polymerase and purified by polyacrylamide gel electrophoresis, are renatured in 10 mM MgC12. Addition of the renatured RNA to the reaction mixture reduces the MgC12 concentration to 3 mM and EDTA is added to further reduce the concentration of magnesium to 0.5 mM which was thought to be optimal for RA activity (Y. Endo, unpublished data). In the course of experiments aimed at identifying the RNA recognition elements for RA we became suspicious of these assumptions and were led to reinvestigate the dependence of RA activity on magnesium. The unexpected finding was that RA IAbbreviations used: RA, ricin A-chain; EDTA, ethylenedialninetetraacetic acid; EGTA, (ethylenedioxy)diethylenedinitriolotetraacetic acid. *to whom correspondence should be addressed 1039-9712/96/02028547505.00/0 285

Copyright © 1996 by Academic Press Australia. All rights of reproduction in anyform reserved.

Vol. 39, No. 2, 1996

BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL

A

oo°:°o,.o,oo°./~

u-. ¢--13 e-u oAe-o*u,

I I I 1 1 .111 tl UAGue,uccc,t~ o .u~, °uA

,. °" ~-u

~o,,',%

B

,o°;..... ...~ ..... ~ u ,Ill. I c ~_ UCOCUO UA ~ ' u uu4156 c*°c°•

a,°

, ° \ u~u= • /u u ° =. .- c u c = Bua¢ 4250 cUL~ ~ -- t~-- 4 3 0 0 =_¢ G° ec *-u 4319"°uAc ° ua-¢ c • A~S*~CA*--U-% u° uc ~

¢

RA

+

Ac •

-

rlcin depunnatzon g-g s~t~ ,u =o,;,,°¢~,~° .~_%=,°.o ......~ \%°~¢ c | ~ I, II[ll

¢ ~* . . . . . . . s~e . ~ , Z =r = ° , , ~ = . . . . . . i u:°~ u cleavage G•A~ e= c c¢ - • 4200 c• .~ °*" % o ~ % c~ou c, ~ u.,s~,\~uf I UCU.~' % - 4 5 5 0 £ Gu uc XuCA 4400 cA ~j u

,s

C G AGAG A U G G 15 A c

A

UC-G c G-C U-A

c-Q

c-c UA-U

~-Uo~3,

G 5'ppp G

=o xCu

Figure 1. The ricin/0t-sarcin domain of 28S rRNA. A, a portion of rat 28S rRNA with the ricin/sarcin domain (the lower part of the stem of the domain is unpaired in this early version of the secondary structure); the sites of RA and ot-sarcin action are indicated and a stretch of 12 bases that is near universal is underlined. B, a RNA oligoribonucleotide (a 35-mer) that mimics the toxin domain and has in addition three 5' guanosine residues that derive from the T7 promoter used to transcribe the RNA.

had no enzymatic activity unless a divalent cation and a chelating agent capable of forming a complex with the cation were present in the reaction mixture.

Materials

and

Methods

The oligoribonucleotide was prepared with phage T7 RNA polymerase, a synthetic DNA template, and the four nucleoside triphosphates (supplemented with 5'[c~-32p]ATP) and purified by polyacrylamide gel electrophoresis (8,3). The oligoribonucleotides were renatured by heating at 90 °C for 1 rain in 10 mM Tris-HC1 (pH 7.6), 50 mM NaCI, and cooled to 0 °C. RA (purchased fi'om Inland Laboratories as a 100 mM solution in phosphate buffered saline containing 50% glycerol), divalent cation, chelating agent, and water, were added to the renatured RNA to give the following concentrations: 3 mM Tris-HC1 (pH 7.6), 15 mM NaC1, and the cation and chelating agent concentrations specified in the figures. Incubation was at 35 °C for the times indicated and the reaction was stopped by the addition of sufficient sodium dodecyl sulfate to give a final concentration of 0.5%. The oligoribonucleotide and 15 pg of added carrier tRNA were precipitated with 300 mM NaC1 and 2.5 volumes of ethanol. The RNA was dissolved in 5 gl of water and 25 pl of a solution of aniline and acetic acid (sufficient to give a final concentration of 1 and 2.8 M respectively) was added: the sample was then incubated for 10 rain at 40 °C. (Aniline treatment induces strand scission at the site of depurination, creating fragments that can be separated by gel electrophoresis.) The aniline-treated RNA was precipitated with ethanol and 300 mM NaC1; dissolved in 15 ~t of 89 mM Tris-HC1 (pH 8.3), 89 mM boric acid, 2.5 mM EDTA, 0.025% bromphenol blue, and 7 M urea; and separated by electrophoresis for 2 h at 2 kV in 20% polyacrylamide gels containing 7 M urea and visualized by autoradiography.

286

Vol. 39, No. 2, 1996

BIOCHEMISTRYand MOLECULARBIOLOGY INTERNATIONAL

Results and Discussion Although a magnesium concentration of 0.5 mM was thought to be optimal for RA activity no enzymatic activity was observed at any magnesium concentration in the absence of EDTA (cf. Fig. 2A, lane 2 for an example); moreover EDTA did not support RA activity in the absence of magnesium (Fig. 2A, lane 3). Thus there is a requirement for both divalent cation and chelating agent (Fig. 2A, lane 5); an excess of magnesium ions over EDTA is tolerated (Fig. 2A, lane 4) although it does not improve the activity of the enzyme; on the other hand an excess of EDTA abolishes catalysis by RA (Fig. 2A, lane 6).

A

B

m M - M g C I2

0

5

0

I0

5

5

5

mM-C.aCI2

0

5

0

10

5

5

5

mM-EDTA

0

0

5

5

5

to

5

mM-EDTA

0

0

5

5

5

I0

5

C

D

mM_MgCI2

0

5

0

I0

5

5

5

raM-CaCI2

0

5

0

I0

5

5

5

mM-EGTA

0

0

5

5

5

tO

5

mM-EGTA

0

II

5

5

5

I0

5

I , I 21 31 , [ nl 61 71

I , I 2131,1,1617

Figure 2. The dependence of RA activity on divalent cation and chelating agent. The RNA oligonucleotide (cf. Fig. 1B) was incubated with RA and different combinations and different concentrations of magnesium, calcium, EDTA, and EGTA; the combinations and the concentrations of each are designated. In lanes 7, RA was added but the RNA was not treated with aniline (aniline causes strand scission at the site of depurination). The concentration of RNA was 5 nM; of RA 2.5 ~M; and incubation was at 35 °C for 60 min.

287

Vol. 39, No. 2, 1996

BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL

Similar results were obtained when calcium replaced magnesium (Fig. 2B). Catalysis required the simultaneous presence of both calcium and EDTA, neither alone was sufficient; and, once again, the enzyme tolerated excess calcium but not excess EDTA. It is noteworthy that catalysis by RA in the presence of EDTA is significantly more efficient with calcium than with magnesium (compare lanes 5 in Figs. 2A and 2B). To determine whether the divalent cation and the chelating agent had dependent but separate functions we substituted EGTA for EDTA. EGTA chelates calcium but not magnesium whereas EDTA binds both. No RA activity was observed with magnesium and EGTA (Fig. 2C, especially lane 5) whereas efficient depurination was seen with calcium and EGTA (Fig. 2D, lanes 4 to 6). Another unexpected finding: neither an excess of EGTA nor an excess of calcium inhibits RA (Fig. 2D, lane 4 and 6). We conclude that it is a complex, or the simultaneous presence, of divalent cation and chelating agent that is essential for RA enzymatic activity; and that depurination by RA is more efficient with calcium and EGTA than with either magnesium or calcium and EDTA. It seemed possible that the chelating agent and the divalent cation were affecting either the structure of the RNA substrate, or the structure of RA, or removing or inactivating an inhibitory contaminant in the toxin preparation, or all of these. We have not been able to identify the molecular species affected no less how the effect is mediated. There is no requirement for calcium nor for EGTA for RA action when ribosomes are the substrate rather than synthetic oligoribonucleotides. Efficient depurination of A4324 in 28S rRNA in reticulocyte lysate ribosomes never exposed to a chelating agent occurs with concentrations of RA as low as 0.5 nM (results not shown). This suggests that the effect of the complex is on the structure of the synthetic oligoribonucleotide used in the in vitro assay. But the effect of pokeweed anti-viral protein, which has the same mechanism of action as RA, on the identical oligoribonucleotide is not influenced by calcium and EGTA.

Moreover, calcium and EGTA are not required for

cleavage of oligoribonncleotide substrates by o~-sarcin, a ribonuclease that specifically cleaves the phosphodiester bond at G4325 in 28S rRNA and that also retains its specificity on the small RNA oligonucleotide (cf. Fig. 1) (10,3); indeed, calcium and EGTA reduce o~-sarcin cleavage by 50% (results not shown). These results now favored an effect on RA. But pretreatment of RA, or of the oligoribonucleotide, or of both, with calcium and EGTA, followed by removal of the cation-chelator complex by ultra filtration through a Microcon 3 filter unit, did not yield a preparation of either RA or RNA that was competent in the reaction (results not shown). However, if calcium and EGTA were added to the reaction mixture containing pmtreated RA or

288

Vol. 39, No. 2, 1996

BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL

A 100 80

g =

60-

=

40-

o

J 20-

"g--

I

I

I

I

0

20

40

60

80

100

Time (min)

100

0 80-

g .2

6040-

C3 20-

w

I

I

I

I

2

4

6

8

I

10

12

Ricin A-chain (IxM) F i g u r e 3. A comparison of the effect of magnesium and EDTA, and of calcium and EGTA, on the depurination of R N A by RA. In A, the time course; in B, the dependence on the concentration of RA. For the comparison the concentrations of cation and o f chelating agent were: 3 m M MgCI 2 and 2.5 mM EDTA ( 0 ) ; or 5 m M CaCI 2 and 5 m M E G T A (o). In A, the concentration of the R N A oligonucleotide (cf. Fig. 1B) was 3 pM and of R A 2.5 laM; in B, the R N A concentration was 3 laM and incubation was for 20 min at 35 °C. After incubation and aniline treatment the R N A was resolved by polyacrylamide gel electrophoresis and the radioactivity of the bands was determined with Fuji Bas-III imaging plates at a Fuji Bas 2000 workstation; the radioactivity was used to calculate the percentage of the R N A that had been depurinated.

289

Vol. 39, No. 2, 1996

BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL

Table 1

Kinetic parameters for the depurination by RA of A4324 in 28S rRNA in ribosomes or its analog in an oligoribonucleotide

Substrate

K m (jaM)

kcat (min -1)

Ribosomes

2.6

1777

700,000

Mg and EDTA

16.9 +/- 5.4

0.06 +/-0.01

3.5 +/- 1.3

Ca and EGTA

126.9 +/- 20.4

3.75 +/- 0.27

29.5 +/- 5.2

kcat/Km x 10.3 (min -1 ~M -1)

RNA oligonucleotide:

The data for ribosomes are from (4); for the oligoribonucleotide in Mg and EDTA from (5); and for the otigoribonucleotide in Ca and EGTA from (9).

RNA efficient depurination occurred (results not shown). Thus, the effect of the calcium and EGTA is not imprinted on either RA or on the RNA; nor does the complex appear to remove an inhibitory substance from the RA preparation; rather calcium and EGTA are necessary to support catalysis. Although we are unable to provide an explanation of the effect, addition of calcium and EGTA to the reaction mixture has materially improved the efficiency of the assay of RA activity (Fig. 3). In the conditions used earlier (3 mM Tris-HC1 (pH 7.6), 15 mM NaC1, 3 mM MgC12, and 2.5 mM EDTA) the depurination of the oligoribonucleotide substrate by RA did not reach a plateau in 90 min and less than 40% of the substrate was utilized; with calcium and EGTA (3 mM Tris-HCl (pH 7.6), 15 mM NaCI, 5 mM CaC12, 5 mM EGTA) the reaction is appreciably complete in 10 min and 80% or more of the substrate is depurinated (Fig. 3A). Moreover, less RA is required to achieve these results: with calcium and EGTA, 1 jaM RA catalyzes

290

Vol. 39, No. 2, 1996

BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL

depurination of approximately 80% of the substrate, whereas in the previous conditions 10 pM RA depurinates only about 50% of the RNA (Fig. 3B). Calcium and EGTA substantially improve the kinetics of the reaction catalyzed by RA (Table 1). The kcat is increased from 0.06 min -1 to 3.75 rain -1 despite a concurrent increase in K m from 16.9 to 126.9 laM; the kcat/Km with calcium and EGTA is 29.5 as compared to 3.5 in the former conditions, an improvement of almost an order of magnitude. Even in these improved conditions the kcat for the reaction with oligoribonucleotides is pitifully small, 3.75 min -1, compared to the reaction with ribosomes which is 1777 min q (4) - a difference of 2.5 orders of magnitude. Perhaps, the most significant practical consequence of the improvement in the reaction conditions is that they support enzyme turnover which will now permit certain types of analyses of the mechanism that were not possible before.

Acknowledgments The work was supported by a grant from the National Institutes of Health (GM33702). We are grateful for the advise of our colleagues Yuen-Ling Chan and Alexander Munishkin.

References (1) (2) (3)

Endo, Y., and Tsurugi, K. (1987) J. Biol. Chem.'262, 8128-8130. Endo, Y., Mitsui, K., Motizuki, M., and Tsurugi, K. (1987) J. Biol. Chem. 262, 5908-5912. Endo, Y., Chan, Y.-L., Lin, A., Tsurugi, K., and Wool, I. G. (1988) J. Biol. Chem. 263, 7917-7920. (4) Endo, Y., and Tsurugi, K. (1988) J. Biol. Chem. 263, 8735-8739. (5) Gltick, A., Endo, Y., and Wool, I. G. (1994) Nucleic Acids Res. 22, 321-324. (6) Endo, Y., Gltick, A., and Wool, I. G. (1991) J. Mol. Biol. 221, 193-207. (7) GliJck, A., Endo, Y., and Wool, I. G. (1992) J. Mol. Biol. 226, 411-424. (8) Milligan, J. F., Groebe, D. R., Witherell, G. W., and Uhlenbeck, O. C. (1987) Nucleic Acids Res. 15, 8783-8798. (9) Glfick, A., and Wool, I. G. (1995) J. Mol. Biol., in press. (10) Endo, Y., and Wool, I. G. (1982) J. Biol. Chem. 257, 9054-9060.

291