Biochimie 70 (1988) 827-831 (~) Socitt6 de Chimie biologique/Elsevier, Paris

Short

827

communication

Note on the activation of the heme-stabilized translational inhibitor of reticulocyte lysates by oxidized giutathione*

Concepci6n PALOMO and Jos6 Manuel SIERRA**

Centro de Biologia Molecular, CSIC-UAM, Cantoblanco, 28049 Madrid, Spain

(Received 29-7-1987, accepted 23-12-1987)

Summary m The heme-controlled translational inhibitor (HCI) of reticulocyte lysates can be activat- ed either by a lack of. by heme or, in the presence of heme, by oxidized glutathione (GSSG) and various oxidative processes. The latter activation can be prevented or reversed by NADPH or NADPH gener- ators, such as glucose-6-phosphate (G-6-P). Since reticulocyte lysates contain a very active GSSG reduc- tase, it was conceivable that GSSG acts by draining lysate NADPH via the reaction GSSG + NADPH + H+Z2 GSH + NADP ÷. However, removal of lysate GSSG reductase by its corresponding antibody has no effect on the activity of GSSG. This supports previous observations with lysates depleted of GSSG reductase by affinity chromatography and supports the notion that GSSG activates HCI in a more direct fashion. The role of NADPH generation in maintaining HCI in its inactive, pro-HCI form is further supported by the observation that the addition of anti-lysate G-6-P dehydrogenase antibody leads to activation of HCI in reticulocyte lysates.

heme-stabilized inhibitor / protein synthesis / rabbit reticulociZes / oxidized glutathiose

Introduction

The activation of the heme-controUed transla- tional inhibitor (HCI) of reticulocyte lysates by oxidized glutathione (GSSG) has been studied in several laboratories in an effort to elucidate its mode of action (for reviews see [1, 2]). Since, as reported earlier (see [3] and references therein), the presence of NADPH is required to maintain HCI in its inactive pro-HCI form and since GSSG can drain the NADPH stores via the reac- tion catalyzed by GSSG reductase, an enzyme present in large amounts in reticulocyte lysates, it appeared possible that GSSG activates HCI indirectly via NADPH depletion. With use of

lysates freed of GSSG reductase by affinity chro- matography, Jackson [1] showed that GSSG acts in a more direct fashion. Yet the mode of action of GSSG is still unknown. We only know that GSSG does not activate partially purified pro- HCI, so that some lysate component(s) is (are) necessary for its effect [3, 4].

The purpose of the present note is to report that we have confirmed Jackson's observation with use of a different and more specific method of removal of lysate GSSG reductase, namely with its corresponding antibody. We also report that removal of iysate G-6-P dehydrogenase with its antibody le~ds, as expected, to HCI acti- vation.

*Dedicated to the memory of David Vtizquez **Author to whom correspondence should be addressed. Abbreviations:elF-2: eukaryotic initiation factor-2; Met-tRNai: eukaryotic initiator methionyl-tRHa; HILl: heine-controlled translational inhibitor (an eIF-2akinase); G-6-P: glucose-6-phosphate; I ~ P : fructose-l,6-bisphosphate; GSSG" oxidized glu- talhione; EDTA: ethylenediaminetetraacetic acid.

828 C. Palomo and J.M. Sierra

Materials and methods

Assays GSSG reductase was assayed spectrophotometrically by following the oxidation of NADPH by GSSG at 340 nm. The G-6-P dehydrogenase assay followed the reduction of NADP by G-6-P under similar conditions. The protein concentration was determin- ed using the method of Lowry et al. [5] with bovine serum albumin as the standard. The assay for protein synthesis and elF-2a kinase activity was as de-scribed earlier [3].

Preparations GSSG reductase [6-8] and G-6-P dehydlrogenase [9-11] were purified from a rabbit erythrocyte lysate essentially as described for other sources. The lysate from 56 rabbits, prepared as described previously [3], was made 1.0 mM in 0.1% EDTA in 2-mercapto- ethanol and fractionated with ammonium sulfate. A 30-55% saturation cut contained the bulk of the G-6-P dehydrogenase, whereas most of the GSSG reductase activity was in a 55-70% saturation frac- tion. The latter was dissolved in a buffer containing 20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 5% glycerol, 0.1% 2-mercaptoethanol, 1 mM ¢-amino-caproic acid, and the enzyme further purified by D E A E - cellulose chromatography, heating (8 min at 65oC), phosphocellulose chromatography, affinity chroma- tography on Cibacron-blue (Sigma) and, finally, chromatography on hydroxyapatite. For purification of G-6-P dehydrogenase, the 30-55% saturation ammonium sulfate fraction was dissolved in the same buffer except that its pH was 7.3 and that it was made 20/.~M in NADP. After dialysis vs a similar buffer that contained only 5 p,M NADP, the enzyme was

l . , , ivop, . ,_ , - cellulose chromatographies followed by affinity chro- matography on 2' ,5 '-ADP-Sepharose. Both en- zymes were electrophoretically homogeneous (not shown). They were kept at -70oC until used.

Anti-GSSG reductase antibody was prepared by injecting rats subcutaneously with the purified enzyme (0.12 mg/Kg) evenly suspended in 0.25 ml of a buffer containing 138 mM NaCI, 2.7 mM KCI, 8.1 mM Na2HPO4, and 1.5 mM KHaPO4, along with 0.25 ml of complete Freund's adjuvant. A second dose (0.08 mg/kg) was injected 3 weeks later with incomplete Freund's adjuvant. A final intravenous injection (0.08 mg/kg) was administered during the 5th week and the animals were bled a week later. The serum was fraetionated by chromatography on pro- tein A-Sepharose (Pharmacia, 1 ml of gei/ml of serum) equilibrated in 10 mM Tris-HCl, pH 7.9, containing 150 mM NaCI. The column was tho- roughly washed with this buffer and the retained immunoglobulins were cluted with 0.58% (v / v) ace- tic acid, 150 mM NaCI. The eluate was neutralized with 1.0 M Tris-base and concentrated to about 1 / 8 of its original volume by vacuum dialysis vs 20 mM

Tris-HCl, pH 7.3, containing 20 mM KCI. The anti- body was kept frozen at -70°C. G-6-P dehydrogen- ase antibody was similarly prepared except that guinea- pigs were used instead of rats and the final injection of antigen was given subcutaneously.

Results and Discussion

Lysates deprived of GSSG reductase

As shown in Fig. I, removal of GSSG reductase had no effect on the inhibition of protein synthe- sis caused by GSSG (Fig. 1A). In either case the inhibition was not only prevented by N A D P H (Fig. 1B) but also reversed by G-6-P (Table I). In agreement with these results, GSSG promot- ed phosphorylation of the elF-2a subunit equally well, whether the lysates contained (Fig. 2, track 2) or lacked (Fig. 2, track 8) GSSG reductase and this effect was prevented or re- versed by either G-6-P (Fig. 2, tracks 3 - 5 , 9 -11 ) or N A D P H (not shown). Essentially simi- lar results were obtained using lysates chromato- graphed through 2 ' , 5 ' -ADP-Sepha rose (data not shown).

L ysates depleted of G-6-P dehydrogenase

For the removal of G-6-P dehydrogenase, the lysates were preincubated with the antibody and then passed through a protein A-Sepha rose column. Fig. 3 shows the resulting activation of HCI as reflected in the increased phosphory- lation of the eIF-2t~ subunit (Fig. 3, tracks 1A, 1B). This was true even though the enzyme ac- tivity, albeit ma:kedly reduced (see legend to Fig. 3), was not completely removed. As expect- ed, while prevented by dithiothreitol (Fig. 3, track 3A), N A D P H (Fig. 3, track 4A) or the simultaneous addition of G-6-P plus G-6-P dehy- drogenase (Fig. 3, track 4B), the activation of HCI by anti-G-6-P dehydrogenase antibody was not prevented by G-6-P alone (Fig. 3, tracks 2A, 2B).

These results obtained with lysates depleted of G-P-6 dehydrogenase are consistent with the notion that the G-6-P dehydrogenase system is responsible for maintaining HCI in its inactive, pro-HCI form. This supports earlier evidence [3, 12. 13] for the regulatory role of N A D P H in polypeptide chain initiation. Furthermore, these results suggest that G-6-P controls the level of eIF-2a kinase in standard lysates mainly because of its NADPH-generat ion ~apacity.

Activation of the herne-stabilized inhibitor 829

Table I. Inhibition of protein synthesis by GSSG in reticulocyte lysates deprived of GSSG reduetase and its rever- sal by G-6-P.

Addit ions Buffer Preincubation of lysate with

GSSG G-6-P non-specific (mM) (0.2 mM) immunoglobuhn

([14C]Leu;

GSSG reductase antibody cpm x 10-3)

- - 21.3 17.7 19.1 0.03 - 11.0 10.1 10.8 0.03 at 0 min 21.4 20.5 21.0 0.03 at 5 min 21.1 18.7 21.8

0.12 - 9.1 8.1 8.2 O. 12 at 0 min 20.3 20.3 19.9 O. 12 at 5 min 20.9 21.3 21.2

Hemin containing lysates were preincubated for 40 min at 25oC in the presence of either buffer, non-specific immunoglobulin (1.4 mg/ml of lysate) or GSSG reduetase antibody (1.47 mg/ml of lysate) and then centrifuged for 10 rain in a microfuge. Glutathione reductase activity in these lysates was 0.95, 1.04 and zero units/nil of lysate, respectively. Translation mixtures (33/zl) containing 15 v,M hemin, 15 t~l of the preineubated lysate and the additions shown were incubated for 60 rain at 30°C. Values represent [~4C]Leu incorporated into 2/.d of reaction mixture.

,.a-

I o , r - -

x 2 E c a .

. . . .J

L . - - . J

A .,, B

0.05 010 0.15 0.2 0./+ 0.6 0.8

GSSG (mH) NAI3PH (mH)

2

O

E r a

- t l = 0~

..._1

L . . J .,a-

Fig, 1. A. Inhibition of protein synthesis by GSSG in reticuiocyte lysates deprived of GSSG reductase. Heroin-containing lysates were preineubated for 40 rain at 25oC in the absence (e) or presence (o) of GSSG reductase antibody (1.57 mg/ml of lysate) and then centrifuged for 10 min in a microfuge. The GSSG reductase activity in control lysates was 1.02 units/ml, in the anti- body-treated lysates it was nil. Samples (32 ~1) containing 14.7/zM hemin, 14/zl of lysate and GSSG at increasing concentrations were incubated for 60 min at 30°C. B. Reversal of GSSG inhibition by NADPH. Lysate was preincubated in the absence (It----I, o) or presence (A--A, I ) of GSSG reductase antibody (1.47 mg/ml of lysate) as in (A). The GSSG reductase activity in control lysates was 0.97 units/ml, in the antibody-treated lysates it was nil. Samples (33 v,l) containing 15 tzM hemin, 15 t~l of lysate, 0.12 mM GSSG (I-----I~ A--A), and increasing concentrations of NADPH, were incubated for 60 min at 30°C. The clam of [14C]leueine incorporated/2 V,I sample are plotted against the GSSG (A) or NADPH (B) concentration.

830 C, Palomo and J.M. Sierra

1 2 3 4 5 6 7 8 9 1 0 1 1 1 2

O Q ' --,~ qllp a

A 1 2 3 4

B 1 2 3 4

e e ~ W W

~ W ~ O

e e q b i ~ t -

C I 2 3 4

Fig. 2. elF-2a kinase in reticulocyte lysates deprived of GSSG reductase. Heroin-containing lysates were pre- incubated for 20 min at 25oC in the absence (tracks 1-6) or presence (tracks 7-12) of GSSG reductase antibody (2.26 mg/ ml of lysate) and then centrifuged for 10 min in a micro- fuge. The GSSG reductase activity in control lysates was 0.47 units/ml, in the antibody-treated lysates it was nii. Translation mixtures (33 #1) containing 15/zM hemin, 15 #l of lysate, 0.12 mM GSSG (tracks 2-6 , 8-12), 0.23 mM G-6-P added at 0 rain (tracks 3, 9), 5 rain (tracks 4, 10) or l0 min (tracks 5, 11) o1 incubation, and 0.45 mM FDP (tracks 6, 12), were incubated for 60 rain at 30°C. [14C]Leu incorporated into 2 g.l aliquots was (cpm x 10-3): 19.7, 7.9, 24.6, 26.8, 25.1, 7.9, 16.1, 7.7, 23.8, 25.0, 21.8 and 7.8 in samples 1-12, respectively, elF-2a kinase activity was assayed on 3/~l aliquots as previously described [3]. The arrow indicates the position of the eIF-2~ subunit.

Acknowledgments

This investigation was supported by grants from the Comisi6n Asesora de Investigaci6n and the Fondo de Investigaciones Sanitar;as (Spain). C.P. was the recipient of a fellowship fIom the Fondo de Investi- gacmnes Sanitarias. We thank Prof. Severo Ochoa for discussions and revisien of the manuscript.

References

1 Jackson R.J, (i982) in: Protein Biosynthesis in Eukaryotes (P~reT,-Bercoff R., ed.), Vol. 41, N A T O Advanced S~udy Institute Series, Series

Fig. 3. Activation of HCI in G-6-P dehydrogenase-deficient reticulocyte lysates. Lysates (0.15 ml)were made 16.5/zM in hemin, preincubated for 150 min at 25°C with 4 mg of either G-6-P dehydrogenase antibody (panels A, B) or non- specific immunoglobulin (panel C), and passed through a protein A-Sepharose column (0.3 ml of gel). The G-6-P dehydrogenase activity of iysates treated with the specific antibody was 0.16 units / ml (A) and 0.07 units / mi (B), that of those treated with non-specific immunoglobulin was 0.80 units/ml. Translation samples (33 /~l), containing 14 ttM heroin and all the components of the translation system, were incubated for 60 min at 30°C either without further additions (tracks 1A, 1B, IC) or with additions as follows: 0.9 mM G-6-P (tracks 2A, 2B, 4B, 2C), 0.9 mM dithiothrei-. tol (tracks 3A, 3C), 0.9 mM NADPI-t (tracks 4A, 4C), 1.5 x 10-2 units of G-6-P dehydrogenase (tracks 3B, 4B). elF- 2,, kinase activity was assayed on 3 ttl aliquots. The arrow indicates the position of the elF-2~ subunit.

A: Life Sciences, Plenum Press, New York, pp. 363 - 4 1 8

2 0 c h o a S. (1983)Arch. Biochem. Biophys. 223, 325-349

3 Palomo C., Vicente O., Sierra J. M. & Ochoa S. (1985) Arch. Biochem. Biophys. 239, 497-507

4 Jackson R.J. , Campbell E .A . , Herber t P. & Hunt T. (1983) Eur. J. Biochem. 131 ,289-301

5 Lowry O. H. , Rosebrough N. J. , Farr A. L, & Randall R. J. (1951)J. Biol. Chem. 193,265-275

6 Worthington D.J. & Rosemeyer M.A. (1974) Eur. J. Biochem. 48, 167-177

7 Mannervik B., Jacobsson K. & Boggaram V.

Activation o f the heine-stabilized inhibitor 831

(1976) FEBS Lett. 66, 221--224 S Zanetti G. (1979) Arch. Biochem. Biophys. 198,

241-246 9 De Flora A., Morelli A., Benatti U. & Giuliano

F. (1975)Arch. Biochem. Biophys. 169,362-363 10 Dao M.L., Watson J.J., Delaney R. & Johnson

B.C. (1979)J. Biol. Chem. 254, 9441-9447

11 Schroeder L., Christoff M. & Levy H.R. (1984) Biochim. Biophys. Acta 784, 48-52

12 Hunt T., Herbert P., Campbell E.A., Delidakis C. & Jackson R.J. (1983) Eur. J. B/ochem. 131, 303-311

13 Jackson R.J., Herbert P., Campbell E.A. & Hunt T. (1983) Eur. J. Biochem. 131,313-324