THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1984 by The American Society of Biological Chemists, Inc.

Vol. 259, No. 4, Issue of March 10, pp. 2803-2809.1984 Printed in U.S.A.

Vanadate Inhibits the ATP-dependent Degradation of Proteins in Reticulocytes without Affecting Ubiquitin Conjugation*

(Received for publication, May 10,1983)

Keiji Tanaka, Lloyd Waxman$, and Alfred L. Goldberg From the Department of Physiology and Biophysics, Harvard Medical School, Boston, Massachusetts 02115

Reticulocytes contain a nonlysosomal, ATP-depend- ent system for degrading abnormal proteins and nor- mal proteins during cell maturation. Vanadate, which inhibits several ATPases including the ATP-dependent proteases in Escherichia coli and liver mitochondria, also markedly reduced the ATP-dependent degrada- tion of proteins in reticulocyte extracts. At low concen- trations (KI = 50 MM), vanadate inhibited the ATP- dependent hydrolysis of [3H]methylcasein and dena- tured ‘261-labeled bovine serum albumin, but it did not reduce the low amount of proteolysis seen in the ab- sence of ATP. This inhibition by vanadate was rapid in onset, reversed by dialysis, and was not mimicked by molybdate.

Vanadate inhibits proteolysis at an ATP-stimulated step which is independent of the ATP requirement for ubiquitin conjugation to protein substrates. When the amino groups on casein and bovine serum albumin were covalently modified so as to prevent their conju- gation to ubiquitin, the derivatized proteins were still degraded by an ATP-stimulated process that was in- hibited by vanadate. In addition, vanadate did not reduce the ATP-dependent conjugation of ‘2SI-ubiqui- tin to endogenous reticulocyte proteins, although it markedly inhibited their degradation.

In intact reticulocytes vanadate also inhibited the degradation of endogenous proteins and of abnormal proteins containing amino acid analogs. This effect was rapid and reversible; however, vanadate also reduced protein synthesis and eventually lowered ATP levels in the intact cells. Vanadate (10 mM) has also been reported to decrease intralysosomal proteolysis in hep- atocytes. However, in liver extracts this effect on ly- sosomal proteases required high concentrations of van- adate (KI = 500 @f) and was also observed with mo- lybdate, unlike the inhibition of ATP-dependent pro- teolysis in reticulocytes.

Reticulocytes, and probably other mammalian cells, contain a soluble, ATP-dependent proteolytic system (1, 2). This nonlysosomal degradative pathway appears responsible for the selective breakdown of abnormal polypeptides, as may result from mutations (3), biosynthetic errors (3), or postsyn- thetic damage (4), and also for the selective elimination of many normal proteins during reticulocyte maturation into

* This work was supported by a research grant from the National Institute of Neurological Diseases and Stroke. The costs of publica- tion of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertise- ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$Recipient of a career development award from the Eli Lilly Foundation.

erythrocytes (5-7). Of particular biochemical interest is the function of ATP in this process (2, 8). Hershko, Rose, and colleagues have proposed that, in this pathway, ATP is re- quired for the covalent linkage of the polypeptide, ubiquitin, to protein substrates, in order to enhance their susceptibility to cytoplasmic proteases (2, 9, 10). In this conjugation reac- tion, ATP is essential for the enzymatic activation of the carboxyl-terminal glycine on ubiquitin to a form which is then linked by an isopeptide bond to t-amino groups on lysine residues in proteins. According to this model, free amino groups must be available on proteins for them to be substrates for ATP-dependent proteolysis. However, we recently showed that if the amino groups were covalently modified so as to prevent their conjugation to ubiquitin, these proteins were still degraded by an ATP-stimulated process (8). Thus, in reticulocytes, ATP seems to serve at least two distinct func- tions in the soluble pathway for protein breakdown, one requiring and one independent of ubiquitin.

When the original ubiquitin model was proposed, there was no precedent for direct involvement of ATP in the function of proteolytic enzymes. However, since that time, ATP-de- pendent proteases have been purified from Escherichia coli (11-13) and rat liver mitochondria (14), and the ATP require- ment for these enzymes has been shown to be responsible for the energy requirement for protein breakdown in intact bac- teria (15) and mitochondria (16). These novel proteases con- tain an ATPase activity that is essential for proteolysis (11, 12). In fact, inhibition of their ATPase activity with vanadate or quercetin reduces proteolysis in parallel with the fall in ATP hydrolysis (13, 17).

If a similar ATP-dependent protease existed in rabbit re- ticulocytes, it could help account for the ATP requirement for protein breakdown that is independent of ubiquitin con- jugation. To test this possibility, we investigated whether the ATPase inhibitor, vanadate (18), also decreased ATP-depend- ent proteolysis in reticulocyte extracts and intact reticulo- cytes. These studies demonstrate that vanadate is a very potent inhibitor of this pathway and that it acts on an ATP- requiring step which is independent of ubiquitin.

MATERIALS AND METHODS’

RESULTS

Effect of Vanadate on Proteolysis in Reticulocyte Extracts- As shown in Fig. 1, addition of vanadate (100 PM) to reticu-

The “Materials and Methods” are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 83M-1280, cite the authors, and include a check or money order for $1.60 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

2803

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2804 Inhibition of ATP-dependent Degradation of Proteins

0 50 IO0 -

Hours VonadatehM)

FIG. 1 (left). Effect of vanadate on [CHs-'H] casein hy- drolysis in reticulocyte extracts with (c-".) or without (0- - -0) 5 mM ATP. Vanadate (100 g M ) was added either initially (X- - -X) or after the reaction had proceeded in the presence of ATP for 60 min (X-X), as indicated by the arrow. FIG. 2 (right). Effects of different concentrations of vana-

date on [CH3-SH]casein hydrolysis in reticulocyte extracts. Assays were conducted in the presence (U) or absence (-1 of 5 mM ATP.

TABLE I Effect of vanadate addition and removal on protein breakdown in

reticulocyte extracts ATP was added at the final concentration of 5 mM and vanadate

at 100 #M. To remove the vanadate, 1 ml of reticulocyte extract was pretreated with vanadate (100 g ~ ) and dialyzed at 4 "C against 50 m M Tris-HC1, pH 7.8, containing 8 mM KCl, 5 mM MgCI,, 1 m M dithiothreitol, and 20% glycerol. 50 g1 of the reticulocyte extracts were used to assay for breakdown of radioactive substrates and 0.5 ml in the assay for the breakdown of endogenous protein.

Hydrolysis of

I X ~ - ~ S ~ Endogenous ['HICasein proteins

-ATP +ATP -ATP +ATP -ATP +ATP %/h %/h nmol Tyr/h

Control 2.8 13.4 0 10.2 0.28 8.4 Vanadate 2.7 6.3 0 2.4 0.24 3.3 Vanadate then 2.9 13.9 0 7.5

dialysis

locyte extracts dramatically inhibited the ATP-dependent hydrolysis of [3H]casein. At this concentration, vanadate re- duced proteolysis nearly to the levels seen in the absence of ATP. This inhibition was rapid in onset and was seen when vanadate was added either initially or after the reaction had been allowed to proceed for 60 min. When different concen- trations of vanadate were added in the presence of 5 mM ATP, half-maximal inhibition of proteolysis was obtained at a concentration of approximately 50 ~ L M (Fig. 2). By contrast, vanadate did not affect the degradation of ["]casein mea- sured in the absence of ATP.

Vanadate caused a similar marked inhibition of the hydrol- ysis of other proteins, such as denatured '251-albumin (Table I). Prior studies indicated that many proteins within the reticulocytes are also substrates for this degradative system when these cells mature into adult erythrocytes (5-7). The net degradation of reticulocyte proteins during reticulocyte maturation can be easily estimated by measuring release of tyrosine from endogenous cell proteins (6). Table I shows that this breakdown of cell proteins in lysates was almost com- pletely dependent on the presence of ATP. Addition of van- adate inhibited this ATP-dependent process by 60%, but it had little or no effect on the very low rate of proteolysis occurring without ATP.

The inhibition by vanadate of certain ATPases, e.g. the (Na+-K+)-ATPase, is reversible (28). To test whether vana- date inhibited protein degradation reversibly, the reticulocyte

TABLE 11 Effects of various ATPase inhibitors on the degradation of [CH3-3HJ

casein and denatured '261-BSA in reticulocyte extracts In these experiments ATP stimulated casein breakdown 5-fold.

The breakdown of BSA was almost totally dependent on the presence of ATP (Table I).

Inhibition of hydrolysis rate

Inhibitor (mM) [3H]Casein '"1-BSA: -ATP +ATP +ATP

% %

Vanadate (0.1) 0 64 69 Molybdate (0.1) 6 0 0 Quercetin (0.1) 6 26 38 Ouabain (0.1) 6 0 0 Azide (5.0) 0 0 10

extracts were treated with vanadate and then were dialyzed overnight and reassayed. As shown in Table I, dialysis to remove vanadate restored almost completely activity against [3H]casein and most of the activity against "51-BSA.2

To examine the specificity of this action of vanadate, the effects of molybdate and various other ATPase inhibitors were studied (Table 11). Molybdate had no effect on proteol- ysis, even though it is a chemical homolog of vanadate. Ouabain, a specific inhibitor of the membrane (Na+-K+)- ATPase (29), and azide, an inhibitor of the membrane-bound mitochondrial ATPase (30) had little or no effect on proteol- ysis with or without ATP. By contrast, quercetin, which, like vanadate, is a potent inhibitor of several soluble ATPases (31), reduced ATP-stimulated proteolysis, although less effec- tively than vanadate. The two ATPase inhibitors effective against ATP-stimulated proteolysis in reticulocytes, vanadate and quercetin, are also the only ones that inhibit the ATP- hydrolyzing proteases from E. coli (13, 17) and liver mito- chondria (14).

By themselves, these findings do not prove involvement of an ATPase in proteolysis, since vanadate can also inhibit many other ATP-utilizing enzymes which involve a phospho- rylated intermediate (32). Possibly, vanadate may reduce pro- teolysis by inhibiting the ATP-dependent conjugation of ubi- quitin to protein substrates (2, 9, 10). To test this possibility, we examined whether vanadate also decreased the hydrolysis of proteins whose amino groups were covalently modified so as to prevent their conjugation to ubiquitin (8). We have recently reported that such proteins are still degraded by an ATP-stimulated process (8). Accordingly, as shown in Table 111, after methylation, acetylation, carbamylation, or succi- nylation of 13H]casein and denatured lz5I-BSA, their hydrol- ysis is still stimulated 2-4-fold by ATP (although the magni- tude of this ATP effect was less than that with normal substrates (8)). Vanadate strongly inhibited the ATP-stimu- lated breakdown of the substrates lacking free amino groups (Table 111), just as it affected the breakdown of unmodified [3H]casein or lZ5I-BSA. Thus, vanadate seems to inhibit a ubiquitin-independent, ATP-activated step in protein degra- dation.

By themselves, these experiments do not rule out the ad- ditional possibility that vanadate also decreases the conjuga- tion of ubiquitin to substrates. Subsequent experiments, therefore, measured the incorporation of exogenous 1251-~bi- quitin into reticulocyte proteins after DEAE-chromatography to remove endogenous ubiquitin from the extracts ( i e . to obtain fraction I1 (9)). As shown in Fig. 3, Iz5I-ubiquitin was incorporated covalently into many different reticulocyte pro-

' The abbreviation used is: BSA, bovine serum albumin.

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Inhibition of A TP-dependent Degradation of Proteins 2805

TABLE I11 Effect of vanadate on the ATP-stimulated degradation of proteins

with free or blocked amino groups in reticulocyte extracts Vanadate was added at the final concentration of 100 p ~ . These

chemical modifications blocked greater than 98% of the amino groups on casein and BSA (8) and prevented ubiquitin stimulation of pro- teolysis (8). The data represent the ATP-stimulated hydrolysis, which was determined by subtracting the values obtained in the absence of ATP from the activity with ATP. In the absence of ATP, vanadate had no effect on the basal hydrolysis of modified or unmodified Droteins.

ATP-stimulated hydrolysis

TABLE IV Effect of vanadate on ATP and ubiquitin-&pendent proteolysis in

fraction ZI The assay of proteolysis in fraction I1 was similar to that with

reticulocyte extracts except that 216 pg of fraction I1 and 1 pg of denatured '%I-BSA (50,000 cpm) were used for the assay. Ubiquitin was added at 5 pg/assay and ATP at a final concentration of 5 mM.

'"I-BSA hydrolysis Inhibition -VOI +vo4 by vo4 Addition

%fh % None 1.11 1.03 7 Ubiquitin ATP

0.99 0.88 11 5.29 3.67 31

ATP + ubiquitin 20.87 13.38 36 %/3 h %f3 h

Control 39.9 3.0 35.7 7.8 Methylation 8.4 0.9 2.1 0 Acetylation 22.2 2.1 11.1 3.0 Carbamylation 10.5 0.6 10.2 2.7 Succinylation 9.3 0.3 3.6 0.6

14PK 2$lK 30K 43K 67K 94K I t t t +

TABLE V Effect of vanadate on r3HJcasein hydrolysis by various proteases Papain, trypsin, and chymotrypsin were obtained from Sigma, and

the Ca2+-activated protease (type 11) was purified from bovine lung (42). Purified E. coli protease La (16) and partially purified mito- chondrial ATP-dependent protease (13) were used. The assay of protease was performed in the presence or absence of vanadate (100 CM) or molybdate (100 PM).

4 '2sI-Ubiquitin Origin

FIG. 3. Effect of vanadate on the formation of covalent bonds between '"I-ubiquitin and endogenous proteins in frac- tion 11. The assay of '%I-ubiquitin conjugation was performed by densitometry of radioautographs, as described under "Materials and Methods." A, with ATP (5 mM); B, with ATP (5 mM) and vanadate (100 p ~ ) , C, without ATP. Pattern of proteins stained with Coomassie brilliant blue. Marker proteins used and their molecular weights: phosphorylase b ( M , = 94,000), bovine serum albumin ( M , = 67,000); ovalbumin ( M , = 43,000); carbonic anhydrase ( M , = 30,000); soybean trypsin inhibitor ( M , = 20,100), a-lactalbumin (M. = 14,400).

teins, and this process was totally dependent on the presence of ATP, in accord with previous reports (9). Addition of vanadate had no effect on the amount of '%I-ubiquitin con- jugated to cell proteins. Nevertheless, in fraction 11, vanadate significantly inhibited the hydrolysis of lz5I-BSA (Table IV). This process was stimulated by ubiquitin in the presence of ATP and also by ATP in the absence of ubiquitin (Table IV). These findings together suggest that vanadate, like hemin (8, 33, 34), acts on an ATP-requiring step that occurs later in this pathway than does the conjugation of ubiquitin to protein substrates.

Effects of Vanadate on Other Proteases-To test the speci- ficity of the action of vanadate on the ATP-dependent path- way, we investigated whether it affected the activity of various

-Migration 4

ProteaseS Inhibition of hydrolysis

Vanadate Molvbdate %

Papain 0.5 0 Trypsin 6.0 0 Chymotrypsin 3.0 0 Ca2+-activated protease 9.5 0 ATP-dependent protease from 45.0

ATP-dependent protease from 68.0 0 rat liver mitochondria

E. coli (protease La)

well characterized proteases. As shown in Table V, addition of vanadate or molybdate at a final concentration of 100 PM had very little or no effect on the hydrolysis of [3H]casein by several serine proteases, including trypsin and chymotrypsin, or by typical thiol proteases, such as papain or the mammalian Ca2+-activated protease. By contrast, this concentration of vanadate caused a marked inhibition of the ATP-dependent proteases from E. coli and rat liver mitochondria, as reported previously (13, 14, 16, 17), and this effect was not obtained with molybdate.

Recently, Seglen and Gordon (35) reported that a much higher concentration of vanadate (10 mM) reduced protein degradation in isolated rat hepatocytes and suggested that this agent directly inhibited a lysosomal thiol protease. To test this possibility, we investigated the effects of different concentrations of vanadate on the hydrolysis of [3H]casein in liver extracts. As shown in Fig. 4, proteolysis in rat liver extracts was much higher at pH 5.0 where lysosomal hydro- lases are active than at pH 7.9. At pH 5.0, vanadate caused a marked inhibition of proteolysis which was maximal (75%) at 1 mM. At concentrations below 100 p ~ , vanadate had no effect on this process, although these low concentrations did inhibit ATP-dependent proteolysis in reticulocytes (Fig. 2). At pH 7.9 in liver extracts, proteolysis was not stimulatable by ATP (data not shown), and vanadate did not inhibit this process, even at very high concentrations (10 mM).

Molybdate, a homolog of vanadate, also inhibited hepatic proteolysis at pH 5.0, in a similar fashion as vanadate. A specific inhibitor of lysosomal thiol proteases (Ep-475) (36), also markedly inhibited hydrolysis of [3H]casein at pH 5.0 but not at pH 7.9, and the addition of vanadate and Ep-475 together did not produce any greater inhibition than did either

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2806 Inhibition of ATP-dependent Degradation of Proteins

A I \

0 001 01 I 1 0 0 001 0 1 I I O -I

Vanodate(mM) Molybdate ( m M ) FIG. 4. Effects of vanadate and molybdate on [3H]casein

hydrolysis in liver extracts. A 10% homogenate of rat liver was made with 0.25 M sucrose, 20 mM KCI, and 10 mM Tris-HC1, pH 7.4, in a Dounce type A homogenizer and centrifuged at 1000 X g for 10 min. The postnuclear supernatant was frozen and thawed twice, and used as the liver extract. Proteolytic activity was determined by incubating extracts (100 pug of protein) in a volume of 200 pl, which contained 75 mM (Tris-HC1, pH 7.9 (X-X) or Na-acetate, pH 5.0 ( O " - O ) , 0.1 mM EDTA, 0.5 mM dithiothreitol, and 20 pg of [3H] casein (30,000 cpm). After incubation for 1 h, acid-soluble radioactiv- ity was determined as described under "Materials and Methods." Left, concentration-dependent effect of vanadate; right, concentration- dependent effect of molybdate.

agent alone (data not shown). Thus, at high concentrations, vanadate and molybdate probably are affecting the lysosomal enzymes cathepsin B, H, or L, as suggested earlier. However, this effect clearly differs in many ways from the specific action of vanadate on the ATP-dependent proteolytic pathway in reticulocytes.

Effect of Vandate on Intact Cells-In reticulocytes, the nonlysosomal ATP-dependent pathway is involved in the elimination of various proteins as these cells become eryth- rocytes (5-7) , and also in the rapid degradation of highly abnormal proteins (1-3, 37). Additional experiments tested whether vanadate can inhibit these processes in intact cells. As shown in Fig. 5, the breakdown of endogenous proteins in reticulocytes was linezr for about 3 h (as measured by the net production of free tyrosine). The addition of vanadate (0.5 mM) caused a strong inhibition of this process. In cell extracts, the effect of vanadate on proteolysis was found to be reversible (Table I). When cells exposed to vanadate for 1 h were washed and resuspended in vanadate-free medium, the rate of protein degradation returned almost completely after a single lag period. This rapid restoration of proteolysis indicates that vanadate is not decreasing proteolysis by irreversibly damag- ing the cells.

The inhibition in the intact cells required higher concen- trations of vanadate than in the reticulocyte extract (Fig. 6), perhaps because it must compete for entry into the cell via the anion carrier (18, 32). This effect was maximal at 1 mM. Several observations further indicated that vanadate reduced proteolysis by its effect on the soluble ATP-dependent system, and not on the lysosomal route. In the reticulocytes, the maximal inhibition obtained with vanadate (80-90%) was similar to that seen when ATP production was blocked with an uncoupler of oxidative phosphorylation, 2,4-dinitrophenol, and an inhibitor of glycolysis, 2-deoxyglucose (Fig. 6). Fur- thermore, addition of vanadate to cells already depleted of ATP caused no further reduction in the breakdown of cell proteins. Finally, Ep-475, a highly selective inhibitor of lyso- somal thiol proteases which can enter intact cells (36), had no effect on protein breakdown in the reticulocytes (Fig. 6), in accord with previous conclusions (6).

To test whether vanadate also could inhibit the selective breakdown of abnormal proteins in intact cells, the reticulo-

- E 2 300. c x

ul W

0 - E

c 2 0 0 - v c

z 73 0

x

0 0 5 I O 1 5 2 0 2 5 3.0 Hours

FIG. 5. Effect of vanadate on protein breakdown in reticu- locytes, measured by release of tyrosine from cell protein: time course and reversibility. Reticulocytes were incubated in the presence (M) or absence (0-43) of vandate at a final concen- tration of 100 p ~ . The amount of acid-soluble tyrosine in cells and in the medium was determined at the times indicated, and the amount present initially was subtracted from each value. After the reaction had proceeded for 1 h in the presence of vanadate, as indicated by the arrow, the cells were washed twice with Krebs-Ringer-phosphate buffer and reincubated in a vanadate-free medium (X- - -X).

I 0 0.01 0.1 I I O

V a n a d a t e (mM) FIG. 6. Concentration dependence of the effect of vanadate

on protein breakdown in reticulocytes. Protein was estimated by measuring the release of tyrosine from cell protein as described in the legend to Fig. 5, except that different concentrations of vanadate were used in Krebs-Ringer-phosphate buffer containing either 10 mM glucose (o"-o), 10 mM glucose and 100 p M Ep-475 (X-X), or 0.2 mM 2,4-dinitrophenol (DNP) and 20 mM 2-deoxyglucose (DG) (-).

cytes were incubated with the valine analog 2-amino-3-chlo- robutyrate. The newly synthesized analog-containing proteins were rapidly hydrolyzed in accord with earlier findings (l), but within 15 min, vanadate (0.5 mM) dramatically this proc- ess (Fig. 7) . Thus, vanadate seems to be a useful reagent for studies of the ATP-dependent proteolytic pathway in intact reticulocytes.

Additional experiments were undertaken to determine whether treatment with vanadate was highly toxic to the cells and whether the fall in proteolysis might be secondary to a depletion of cellular ATP levels (Fig. 8). Exposure to vanadate for 1 h did promote a small degree of hemolysis (4%) of the reticulocytes (Fig. 8C), but this effect was not evident until 30 min of treatment, when proteolysis was already signifi-

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Inhibition of ATP-dependent Degradation of Proteins 2807

1

Mlnutes

FIG. 7. The effect of vanadate on the degradation of pro- teins containing 2-amino-3-chlorobutyrate in place of valine in reticulocytes. Breakdown of proteins synthesized previously in the presence of the valine analog was measured in the absence (o”-o) or presence (&”O) of vanadate (0.5 mM).

0 5 4 3 2

I C

4

2

0

A P T i

“

C

Q”

15 30 45 6 0 Minutes

FIG. 8. Effect of vanadate on ATP content, protein synthe- sis, and hemolysis in reticulocytes. A, effect of vanadate on the intracellular ATP concentration. Cells were incubated at 37 “C with- out (-) or with ( O “ 4 ) vanadate (0.5 mM). A t the indicated times, trichloroacetic acid was added at a final concentration of IO%, and ATP concentration in the acid-soluble fraction was analyzed as described under “Materials and Methods.” The data are the mean & S.D. of 6 determinations. B, effect of vanadate on protein synthesis in reticulocytes. C , effect of vanadate on the lysis of reticulocytes.

cantly reduced. Addition of vanadate also did not significantly affect the ATP content of the cells initially, but after 45 min of exposure, ATP levels gradually decrased (Fig. 8A). Thus, the rapid inhibition of protein breakdown in uiuo (Figs. 5 and 7) did not result from either cell lysis or a reduction of intracellular ATP concentation. One interesting finding was that vanadate dramatically reduced protein synthesis in these

cells (Fig. 8). Whatever the mechanism of this effect? it cannot be responsible for the fall in proteolysis, since com- plete inhibition of protein synthesis does not reduce the breakdown of normal or abnormal proteins in reticulocytes (1, 3, 6 ) . These experiments indicate that although vanadate may affect multiple cellular processes, the rapid, reversible inhibition of proteolysis in these cells probably involves the same mechanism observed in the cell-free system.

DISCUSSION

Specific inhibitors of the soluble ATP-dependent pathway for protein breakdown should prove very useful in attempts to study this pathway and to clarify its physiological signifi- cance. The present findings indicate that vanadate, at low concentration (KI = 50 PM), is a potent inhibitor of this process in reticulocyte extracts. This effect is not due to an inhibition of ubiquitin conjugation (Tables 111 and IV and Fig. 3), and vanadate therefore seems to be affecting another ATP-activated process (8). A major goal for future research will be to identify the precise enzymatic step that vanadate blocks in this pathway (see below). In addition, vanadate may be useful in short term studies with intact cells, but caution seems appropriate in such experiments because of the multiple effects of vanadate in vivo (Fig. 8) including its ability to inhibit protein ~ynthesis,~ and with time, to reduce ATP levels and eventually cause some cell lysis.

The marked inhibition of proteolysis in the reticulocyte extracts by vanadate resembles the inhibition found with the ATP-dependent proteases from E. coli (11-13) and rat liver mitochondria (14). For example, protease La from E. coli shows a similar Kr (50 p ~ ) and also responds to quercetin but not to molybdate (Table 11). Also, in intact mitochondria, vanadate at low concentrations prevents energy-dependent proteolysis (16), apparently by inhibiting the ATP-hydrolyz- ing protease found in the mitochondrial matrix (14). Thus, the ATP-dependent pathways for protein degradation in re- ticulocytes and those in prokaryotic cells and mitochondria may involve similar mechanisms. However, in the mammalian cytosol, unlike in E. coli (17) or mitochondria (16), this process seems to involve ubiquitin (2, 9, 26), and it has been argued (2) that ubiquitin conjugation accounts for the energy require- ment for proteolysis in reticulocytes. However, our recent studies indicate that in reticulocytes ATP acts at multiple steps in this pathway (8)4 and that an energy-requiring reac- tion, which is sensitive to hemin is necessary for the break- down of protein substrates, whether or not they contain ubiquitin (8).

By analogy to protein breakdown in bacteria and mitochon- dria, this unidentified step could be an ATP-hydrolyzing protease. Additional evidence for this conclusion is the pres- ent finding that the ATPase inhibitors, vanadate and to a lesser extent quercetin, inhibit this process (Table 11) in a similar fashion as hemin (8, 33, 34). Despite much effort, however, we have thus far failed to isolate an ATP-hydrolyz- ing protease from these cells, although we have purified sev- eral novel proteases that appear to be involved in this pathway (38, 39).4 Recently we have succeeded in isolating from these extracts a fraction capable of carrying out ATP-dependent proteolysis independently-of ~biqui t in .~ In this fraction, both

Since this manuscript was completed Ranu has reported that vanadate inhibits peptide chain initiation in reticulocyte extracts (Ranu, R. S. (1983) Proc. Natl. Acad. Sci. U. S. A . SO, 3148-3152).

K. Tanaka, L. Waxman, and A. L. Goldberg manuscript in prep- aration.

K. Tanaka, L. Waxman, and A. L. Goldberg submitted for pub- lication.

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2808 Inhibition of A TP-dependent Degradation of Proteins

proteolysis and ATP hydrolysis are inhibited by vanadate as well as by hemin. However, real proof that these inhibitors are acting on an ATP-hydrolyzing protease will require its complete purification and characterization.

Whatever its precise site of action, this effect of vanadate on the reticulocyte degradative sytem differs qualitatively from its effect on hepatocytes. Our findings with liver homog- enates confirm the original suggestion of Seglen and Gordon (35) that in hepatocytes, vanadate at high concentrations can inhibit protein breakdown within lysosomes. Here we have shown that at millimolar concentrations, vanadate and mo- lybdate inhibit the acid-optimal proteases. These agents seem to be acting on thiol proteases, since a similar inhibition was obtained with Ep-475, which specifically binds to the active sites of thiol proteases (36). In these liver extracts, only lysosomal proteases (which are the dominant proteolytic sys- tem in these preparations) were sensitive to vanadate and no effect was seen below 100 PM, the concentration which caused a strong inhibition of the nonlysosomal pathway in the retic- ulocyte extracts. In the liver extracts, proteolysis was not affected by ATP, either at neutral or acid pH. By contrast, in reticulocytes the vanadate-sensitive alkaline pathway is ATP- dependent and is not affected by molybdate. Thus, by these criteria, vanadate seems to be a much more specific inhibitor of the cytoplasmic system in reticulocytes.

The experiments of Seglen and Gordon (35) utilized intact cells and measured the degradation of long-lived intracellular proteins. This process is generally believed to occur within lysosomes (2, 35, 40). Although we have demonstrated direct inhibition of the lysosomal proteases by vanadate, their: in vivo findings may also result from indirect actions of this compound, since intralysosomal proteolysis is decreased by inhibition of protein synthesis and ATP production (2, 4, 40, 41). In intact hepatocytes, there is strong evidence that deg- radation of many proteins also occurs by a nonlysosomal route; for example, the degradation of short-lived polypeptides and proteins with abnormal structures requires energy and is not affected by inhibitors of lysosomal function (2, 40, 41). Although this process thus seems to resemble the ATP- dependent pathway in reticulocytes, we and others have not succeeded in establishing soluble ATP-dependent proteolytic systems from liver or other mammalian cells. It therefore remains unclear whether the nonlysosomal process in cells other than reticulocytes is sensitive to low concentrations of vanadate and whether this inhibitor will be of general use in studies of intracellular protein breakdown.

REFERENCES 1. Etlinger, J. D., and Goldberg, A. L. (1977) Proc. Natl. Acad. Sci.

2. Hershko, A., and Ciechanover, A. (1982) Annu. Rev. Biochem.

3. Goldberg, A. L., and St. John, A. C. (1976) Annu. Rev. Biochem.

4. Goldbere. A. L.. and Boches. F. S. (1982) Science (Wash. D. C.)

U. S. A. 74, 54-58

5 1,335-364

45,747-803

2 1 5 , rlO7-1109

Eur. J. Biochem. 109 , 405-410 5. Muller, M., Dubiel, W., Rathmann, J., and Rapoport, S. (1980)

6. Boches, F. S., and Goldberg, A. L. (1982) Science (Wash. D. C.) 2 15.978-980

7. Speiser, S., and Etlinger, J. D. (1982) J. Biol. Chem. 257,14122-

8. Tanaka, K., Waxman, L., and Goldberg, A. L. (1983) J. Cell Biol.

9. Hershko, A., Ciechanover, A., Heller, H., Haas, A. L., and Rose, I. A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,1783-1786

10. Ciechanover, A., Heller, H., Elias, S., Haas, A. L., and Hershko, A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 1365-1368

11. Chung, C. H., and Goldberg, A. L. (1981) Proc. Natl. Acad. Sci. U. S. A. 78,4931-4935

12. Charette, M. F., Henderson, G. W., and Markovitz, A. (1981) Proc. Natl. Acad. Sci. U. S. A. 78 , 4728-4732

13. Larimore, F. S., Waxman, L., and Goldberg, A. L. (1982) J. Biol. Chem. 257,4187-4195

14. Desautels, M., and Goldberg, A. L. (1982) J. Biol. Chem. 257,

15. Olden, K., and Goldberg, A. L. (1978) Biochim. Biophys. Acta

16. Desautels, M., and Goldberg, A. L. (1982) Proc. Natl. Acad. Sci.

17. Waxman, L., and Goldberg, A. L. (1982) Proc. Natl. Acad. Sci. U.

18. Macara, I. G. (1980) Trends Biochem. Sci. 5,92-94 19. Goodno, C. C. (1979) Proc. Natl. Acad. Sci. U. S. A. 7 6 , 2620-

20. Boches, F. S., Waxman, L., and Goldberg, A. L. (1984) Arch.

21. Rice, R. H., and Means, G. E. (1971) J. Biol. Chem. 246 , 831-

22. Moore, A. T., Williams, K. E., and Lloyd, J. B. (1977) Biochem.

23. Hirs, C. H. W. (1967) Methods Enzymol. 11, 199-203 24. Waalkes, T. P., and Udenfriend, S. J. (1957) J. Lab. Clin. Med.

25. Deluca, M., and McElroy, W. D. (1978) Methods Enzymol. 5 7 ,

26. Ciechanover, A., Elias, S., Heller, H., Ferber, S., and Hershko, A.

27. Laemmli, U. K. (1970) Nature (Lord.) 227 , 680-685 28. Bowman, B. J., and Slayrnan, C. W. (1979) J. Biol. Chem. 254,

29. Cantley, L. C. (1981) Curr. Top. Bioenerg. 11, 201-237 30. Pedersen, P. L. (1975) Bioenergetics 6 , 243-275 31. Lang, D. R., and Racker, E. (1974) Biochim. Biophys. Acta 333 ,

32. Ramasarma, T.. and Crane, F. L. (1981) Curr. TOD. Cell. Reml.

14127

96,1580-1585

11673-11679

542,385-398

U. S. A. 7 9 , 1869-1873

S. A. 79,4883-4887

2624

Biochem. Biophys., in press

832

J. 164,607-616

50,733-735

3-15

(1980) J. Biol. Chem. 255, 7525-7528

2928-2934

180-186

20,247-301 -

33. Etlinger, J. D., and Goldberg, A. L. (1980) J. Biol. Chem. 255, 4563-4568

34. Haas, A. L., and Rose, I. A. (1981) Proc. Natl. Acad. Sci. U. S. A.

35. Seglen, P. O., and Gordon, P. B. (1981) J. Biol. Chem. 256,7699-

36. Hanada, K., Tami, M., Yamaguchi, M., Ohmura, S., Sawada, J.,

37. McKay, M. J., Daniels, R. S., and Hipkiss, A. R. (1980) Biochem.

38. Kirscbner, R. J., and Goldberg, A. L. (1983) J. Biol. Chem. 258 ,

39. Boches, F. S., Klemes, Y., and Goldberg, A. L. (1980) Fed. Proc. 39,1682

40. Hopgood, M. F., and Ballard, F. J. (1980) Protein Degradation in Health and Disease, Ciba Foundation Symposium 75, pp. 205- 215, Excerpta Medica, Amsterdam

41. Neff, N. T., DeMartino, G. N., and Goldberg, A. L. (1979) J. Cell. Physiol. 101,439-458

42. Waxman, L. (1981) Methods Enzymol. 80, 664-680

78 , 6845-6848

7701

and Tanaka, N. (1978) Agric. Biol. Chem. 42,523-527

J. 188,279-283

967-976

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Inhibition of A TP-dependent Degradation of Proteins

"Vanadate lnh lb i ts the ATP-Dependent Pathway for Protein Degradation i n Reticulocytes'

SUpp lmnt l ry Mater ia l to

K e i ~ i Tanaka, Lloyd Yaman and Alfred L. Goldberg

MATERIALS AND u m m s

0-Casein. 8SA. -d im m lybda te . and ATP (d i rod iun sa l t ] *ere purchaLed f m Sigma Chmical Cmpany. and sodim vanadate frrm Fisher Scienti f ic Cmpmy. Ha lZ51, [3Hlformaldeh~de. [3HlHaBH4 and L-[4.5-3Hlleudne (59.2 C i t m l ) were obtained frm New

England Nuclear Corporation.

6 N HC1 and then boiled to eliminate pol-ric species o f VloOz8-6, h i c h are yel low in color. The concen tmt im o f Vanadate was detenined spectmphotmetv ica l ly - 2925

Stock r o l u t i o n r (0.1 M ) of vanadate were dissolved i n rater. adjusted to pH 10 M i t h

"1 u - 1 L191).

Prepamtion of ReticulocYte Extmcts

Rabbit ret iculocytes were obtained after phenylhydrazine treatmnt of m l e rabbits.

1 s described previously (1). The washed c e l l s were l y u d by the addi t ion of 1.5 vo1-s o f rater containing 1 rn d i t h l a t h r e i t o l . The lysate was centr i fuged at 4O.W rg for 90 nin,

8 n)l KC1. 5 n)l "JC12. 0 . 5 M d i t h l o t h r e i t o l and 20% glycerol. and S t o w at -2OY u n t i l and the Supernatant was dialyzed overnight against 54 M Tris-HCl buf fer pH 7.8 containing

"Sed.

A s w of P m t w l y t i c A c t i v i t y i n E x t r a c t s P v o t W l y t l C a c t i v i t y was determined by incubating 50 p l Of the r e t i c u l m y t e extvacts

In a v o l w of 100 ut. which contained M 1)1 Trls-HCl pH 8.0. 5 M W 1 2 . 5 M ATP, 1 1111 d i t h i o t h r e i t o l , and 20 ug i3ic)*-cauin (25.WO cpn) 0- 10 "9 1251-8% (50.W CPn). After

tr ichloroacetic acid containing BSA (1 W.1) as 1 carrier, and ac id-so lub le rad ioact iv i ty incubation fov 3 hr a t 3 1 Y . the reactions r r e terminated by addition o f 0.7 a1 10%

*as measured. Oegradatlon o f these rubstrater to acid-soluble mtevial Occurved a t l i n e a r r a t e s fay

up t o 3 h r l (8). T h i s p ? ~ ~ e s l re f l ec ts t he complete h y d m l r r i r of the protein rubstrater. without measurable 101s of label f m prete inr by d e n t h y l a t i o n 07 deiodination (PO). ['HlMe-ca$ein #as prepared by reductive methylation (21). BSA was iodinated with Na lZ5l by the chlarmine-T method (22) and then denatured by reduction and c a r b o x m i d a r t h y l a t i o n

(23) t o enhance i t s p r o t e o l y t i c r u r e p t i b i l i t y . The deta i led methods far modif icat ion of the f ree amino groups O f 13Hple-cauin and denatured 1251-8SA by IuCCiWl l t ion.

acetylation, mcthylation and carbawlat ion e r e reported previously ( 8 ) . Oegradation o f endogenous ret iculocyte proteins was maswed by following the

production Of f r e e t y m r i n c frm ce l l p ro te ins as described previously (61. 0.5 nl Of

r e t i cu locy te ex t rac ts con ta ldng 0.5 dl cycloheximide was incubated at 37-C. At the s ta r t

of the incubation and a f t e r 3 h r r . t he aaun t O f tyrosine released w a s determined f l u o r m e t r i c a l l y (24 ) . Production o f ty ros ine frm ret iculOcyte pmteinr was l i nea r f o r 3 h r s i n the presence O f ATP.

Prote in Breakdown in Intact Ret iculocytes

suspended i n four w 1 m 1 Of Krebs-Ringer phosphate (KRP) buffer containing 10 n)l glucose

and 0.5 d cycloheximide. The c e l l suspension (1 m1) was incubated with gentle shaking a t 37T. The reaction was terminated by addi t ion O f an equal v o l m O f Cold 2m

To nearure the re lease of tyror lne from ce l l p ro te in (4.6). the washed c e l l s were

trichlDmaCetiC acid. and the m p l e r were incubated a t 4'C overnight. After

centrifugation. the supernatant (1 m1) was assayed for f ree tymr ine. a s described above.

To detenine the degradatlon Of abnonnal proteins (1) . ret iculocytes *em suspended

amino acids. but without leucine and valine. After 5 mln at 37-C i n a shaking ra te r ba th ,

i n f o u r v o l m s o f KRP buffer containing 10 nW glucose and plasm Concentrat ions of the

the va l ine analog. 2-anino-3-chlombutyric acid (Calbiochm), was added at a f i n a l concentration Of I M. After 5 min mre, 2 uCi/ml of [3Hlleucine (0.02 el) was added. The

20 mln. and then were washed three t imcr I n ice-cold KRP buffer containing IO nW glucose

c e l l s were allawed to incorporate the analog and the radioact ive leucine into proteins for

H i t h shaking a t 37% O ~ p l i ~ a t e I m p l e l (0.5 m1) were rmored per iod ica l ly and added t o an

and 10 n)l unlabeled leucine. The ce l l s were resuspended I n t h i s same buf fer and incubated

equal volune Of cold 20% t r l ch lomdcet ic ac ld . To ~ d l ~ u l a t e the percent degradation of

labeled proteins at each t ime min t . the amount Of acid-soluble radioact iv i ty generated # d l

dl r lded by t o t a l r a d i o a c t i v i t y I n i t i a l l y i n c o r p o r a t e d i n t o p r ~ t e i n ( { .e . . acid-insoluble

materi411 at the beginning of the experiment. and this r a t i o w a s m u l t i p l i e d by 100.

ATP was measured by the f i re f l y luc i fe r in - luc i fense reac t ion (25) w i th the procedure recamended by the manufacturer (Analytical Lminercence Laboratory, Inc.). In t h i s a l l ay .

vanadate did not af fect the estimation of ATP.

O e t r m i n d i a n of Pmtein synthesis and H m o l y r i s

re t icu locytes which contained 10 dl glucose. plasm Concentrat ion$ of amino acids. and 5 Protein synthesis was detemined by incubating 0.25 nl Of a 20s suspension a f

uCi tn l of [3Hlleucine i n the absence or p ~ e l e n c e of Vanadate (1M1 U n l . At the indicated times. the -action *as terminated by addition of 0.25 ml of 20s t l i C h l o r W c e t l c acid. The mid- inso lub le ra te r i a l was washed 10°F t ines w i th 101 t r i ch lo roacet ic acid. dissolved i n

0.1 N NaOH containing 0.11 50s. and the rad ioac t i v i t y Of the no lub i l i led prote ins was

detenined.

a t t he i nd i ca ted t i r es t hey e r e pel le ted by cent r i fugat ion (500 19 10 "in a t C Y ) . and the

concentration Of hrmoglobin i n the nediun was determined frm the absorbance at 540 m. The p e r r e n t o f h e n l y r i r was calculated ill the mount of hmglob in re leased in to the

Redim divided by t o t a l m u m of h m g l o b i n t h a t can be released upon c e l l l y s i s w i t h

water. and t h i s r a t l o was m u l t i p l i e d by 100.

1251-Ubiquit~n Conpmtion

For the asmy O f h a a l y r i r . t h e c e l l s were incubated under the wmc conditions, but

Ubiqui t in -5 Purif ied to hamgeneity from rabbit erythmcytes, as dercnbed by Ciechanaver ct. (26) and then labeled with lZ5I WIth chloramine-T (22). To remve free

was prepared. as dercrlbcd previously ( 8 ) . u b i w i t i n fmn ce l l ex tvac ts . f rac t lon I1 (0.5 I4 KC1 eluate frm the DEAE-cellulose pH 7.0)

TO malure UbiUYit ln COnJugation t o p m t e l n r (9, 10). 100 ,l reaction r n ~ x t ~ r e s containing 5C r*l Tnr-HC1 pH 8.0. 1 d4 d i t h l o t h r e i t o l . 5 OH MgC12. 5 111 ATP, 2.6 U Q

for dO mln. PmtelnS conjugated with lZbl-ubiquitln were analyzed by e lect ropharer i l on 1251-ub?quitin ( 3 . 3 x Id5 cpm), and 1UO uy O f f r a c t i o n I 1 p m t e i n were lncubated a t 37OC

electmphorcr ls, the gels were stained w t h C o r n s l i e b r l l l i a n t blue, dr ied and auto-

polyacrylamide Slab gels (12.51) I n the presence of sodium dodecyl sulfate (27). After

radiographed with Kodat x-ray film. The density Of the autoradlogPams was rmmtared with

an LW 2202 Ultrascan Laser senr i tomtcr .

oata Presentation

*hich agreed w i t h i n 10 per cent. S l m i l a r r e w l t s were seen in a t least three separate exper lmntr u l l n g different preparat ranr o f re t lcu locyter .

All WintS shown i n Figwes and Tables are the average o f at least three detemnnationr by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from

K Tanaka, L Waxman and A L Goldbergwithout affecting ubiquitin conjugation.

Vanadate inhibits the ATP-dependent degradation of proteins in reticulocytes

1984, 259:2803-2809.J. Biol. Chem. 

  http://www.jbc.org/content/259/5/2803Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/259/5/2803.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 3, 2020

http://ww

w.jbc.org/

Dow

nloaded from