medlspmdttbgispprPfhPpritc

tlrplfc

7

Experimental Cell Research 246, 443–450 (1999)Article ID excr.1998.4331, available online at http://www.idealibrary.com on

The Effects of Superoxide and the Peripheral BenzodiazepineReceptor Ligands on the Mitochondrial Processing of

Manganese-Dependent Superoxide DismutaseGary Wright and Vernon Reichenbecher1

Department of Biochemistry and Molecular Biology, Marshall University School of Medicine,

1542 Spring Valley Drive, Huntington, West Virginia 25755

clsdptndpdobpbtc

(etosmSafhn[

dmadpiitfs

The mitochondrion imports and processes the vastajority of the proteins that constitute its structural

lements and metabolic pathways. To study mitochon-rial precursor processing in the context of the cellu-

ar environment, we employed the baculovirus expres-ion system to overexpress the prototypical precursorrotein, human manganese-dependent superoxide dis-utase (hMn-SOD). It was found that superoxide pro-

uced by hyperoxic culture conditions (95% O2 atm) orhe redox cycling agent paraquat caused a lesion ofhe import/processing of precursor hMn-SOD in theaculovirus model. The oxidation of key sulfhydrylroups as a component of the mitochondrial process-ng lesion was implicated by the observation that theulfhydryl reducing agent dithiothreitol was com-letely effective in preventing the block of hMn-SODrocessing induced by paraquat. Interestingly, the pe-ipheral benzodiazepine receptor (PBzR) agonistsK11194, Ro5-4864, and protoporphyrin IX were all

ound to enhance mitochondrial processing of theMn-SOD precursor protein, suggesting a role for theBzR in the regulation of mitochondrial import ofroteins. Collectively, our results suggest a possibleedox-regulated mechanism of mitochondrial proteinmport that may lead to less efficient precursor pro-ein uptake by mitochondria under severely oxidizingonditions. © 1999 Academic Press

INTRODUCTION

The mitochondrion is a semiautonomous organellehat is central to cellular metabolism. It converts cel-ular reducing power into the ubiquitous energy cur-ency, adenosine triphosphate (ATP), via oxidativehosphorylation. The process of oxidative phosphory-ation produces the majority of cellular ATP and is, byar, the largest consumer of molecular oxygen at theellular level. The enzymes involved in the citric acid

1 To whom correspondence should be addressed. Fax: (304) 696-

a253.

443

ycle, fatty acid oxidation, and steroidogenesis are allocated in the mitochondrion. The vast majorities oftructural and catalytic proteins within the mitochon-rion (98%) are encoded in the nucleus and are im-orted from the cytosol. Most proteins are directed tohe organelle by a short, 10 to 30-amino-acid, N-termi-al leader sequence that is recognized by the mitochon-rial import apparatus. The internalization of thesereproteins is then accomplished via an energy-depen-ent process that is mediated by the TOM (translocaseuter membrane) and TIM (translocase inner mem-rane) protein complexes [for review, see 1]. Once im-orted, the preprotein is converted to the mature formy the proteolytic removal of the leader sequence andhe chaperone-mediated folding of the protein into theorrect tertiary structure.The manganese-dependent superoxide dismutase

Mn-SOD) is a prototypical mitochondrial protein. It isncoded in the nucleus, synthesized as a precursor inhe cytosol, converted to mature form with the removalf its 23-amino-acid N-terminal presequence, and as-embled into active enzyme with the incorporation of aanganese ion in the mitochondrial matrix [2]. Mn-OD is of particular interest, having been found to becritical component of the cellular anti-oxidant de-

ense system [3], a putative tumor suppressor [4], andighly induced by several cytokines including tumorecrosis factor-a [5], interleukin-1 [6], and endotoxin7].

The recent putative identification of the mitochon-rial multiple conductance channel (MCC) as the inneritochondrial protein import pore may provide valu-

ble insight into factors that may influence mitochon-rial protein uptake [8, 9]. The MCC was discovered inatch clamp studies of the electrical activity of thenner mitochondrial membrane [10] and has been stud-ed as an ion channel until its identification as a po-ential protein import pore. Several factors have beenound to influence MCC electrical activity. For in-tance, the peripheral benzodiazepine receptor (PBzR)

gonists were reported to potentiate its open time [11].

0014-4827/99 $30.00Copyright © 1999 by Academic Press

All rights of reproduction in any form reserved.

Ept

taictmcficSsocccz

siapsdcp

twd3Smpcci(pp(t

tpwbeaSgfat

er(oStwusat

crm(ua

cpccphafttstleapmfr(retw2ehtcpqgqd

a

444 WRIGHT AND REICHENBECHER

xtrapolation of these observations suggests that theeripheral benzodiazapines might influence the pro-ein uptake activity of the import machinery.

We have previously shown that paraquat, which ishought to act through the generation of superoxidenion, specifically inhibits the mitochondrial process-ng and functional expression of hMn-SOD in insectells [12]. Other reports also document a posttranscrip-ional down-regulation of Mn-SOD expression in mam-alian cells. For example, in rat lung tissue hyperoxic

onditions or treatment with pertussis toxin wereound to increase Mn-SOD mRNA levels. Despite thenduction of the Mn-SOD gene and a several-fold in-rease in Mn-SOD mRNA by these treatments, Mn-OD activity was seen to drop by about 50% during theame time interval [13, 14]. Because paraquat, hyper-xia, and pertussis toxin are all associated with in-reased superoxide production, the above observationsould suggest a role for the superoxide radical in de-reasing mitochondrial import/activation of this en-yme.Here we have used the baculovirus-driven expres-

ion of precursor hMn-SOD to explore the potentialnfluence of oxidizing and sulfhydryl reducing agentss well as several PBzR agonists on the mitochondrialrocessing of precursor hMn-SOD. The results are con-istent with a redox-regulated mechanism of mitochon-rial protein import. Moreover, the results support theontention that the MCC is an inner mitochondrialrotein import pore.

MATERIALS AND METHODS

Cell culture. Spodoptera fungiperda (Sf-9) insect cells were rou-inely maintained at 27°C in monolayer cultures. Grace’s mediumas supplemented with 0.35 g/L NaHCO3, 3.3 g/L lactalbumin hy-rolysate, and 10% fetal calf serum (FCS). Cells were passaged every–4 days by plating 1 3 106 cells per 100-mm tissue culture dish.f-9 cells were stored in freezing solution (20% FCS and 10% di-ethyl sulfoxide in Grace’s medium) under liquid nitrogen. All ex-

eriments were performed in standard Grace’s medium unless indi-ated otherwise. Hyperoxic cell conditions were generated by placingell culture plates in airtight plasticware containers that were mod-fied to accept Millipore Millex-GV Filters at entry ports for 100% O2

5 cc/min). Small exit ports were located opposite the oxygen entryorts. For experiments Sf-9 cells were inoculated with about 10laque-forming units per cell with wild-type or hMn-SOD expressinghMn-SOD-AcNPV) baculovirus described previously [12]. Virus ti-er was determined by the end-point dilution method [15].

Immunoblot analysis. Upon completion of the experimental pro-ocol cells were collected by gentle centrifugation (600g). The cellellet was suspended in ddH2O on ice, sonicated, and mixed quicklyith 3 parts standard gel loading buffer (100 mM Tris z HCl, 10%-mercaptoethanol, 4% SDS, 0.2% bromophenol blue, and 20% glyc-rol, pH 6.8). Samples were immediately heated to 100°C for 3–5 minnd either loaded directly onto the gels or stored frozen at 220°C.amples were elecrophoresed at 100 V on 12% SDS–polyacrymideels for 6–8 h. Gels were blotted onto nitrocellulose filters purchasedrom Millipore using a Bio-Rad trans-blot semidry transfer cell runt 20 V for 18–25 min. The blot was incubated for 1 h at room

emperature in blocking solution (5% nonfat dry milk in Tris-buff- i

red saline (TBS), pH 7.5, containing 0.1% Tween 20). The blot wasinsed briefly in TBS and incubated in the primary antibody solution1:500 dilution of sheep anti-full-length hMn-SOD horseradish per-xidase-conjugated antibody solution purchased from the Bindingite, Birmingham, UK, 3% BSA in TBS, pH 7.4) for 2 h at roomemperature. The blot was then washed four times (10 min each)ith TBS–0.1% Tween 20 solution. Immunoblots were visualizedsing the enhanced chemiluminescence Western blotting detectionystem (Amersham) and quantified by densitometric scanning (Im-ge Quant, Molecular Dynamics). The results are expressed in arbi-rary units.

Cell viability. Viability was assessed based upon the ability ofells to exclude trypan blue dye. Sf-9 cells were pelleted (800g) andesuspended in PBS (pH 7.4) with gentle vortexing. An aliquot of 300l of this cell suspension was added to 700 ml of trypan blue solution

0.2% w/v trypan blue in PBS, pH 7.4) and mixed gently. The areander the coverslip was allowed to fill via capillary action, and viablend nonviable cells were counted.

RESULTS

Paraquat treatment in the absence of manganeseofactor, during the 2- to 4-day postinoculation (p.i.)eriod in which the majority of the recombinant pre-ursor hMn-SOD is synthesized and processed in theontrol cells, leads to the accumulation of precursorrotein (27 kDa) and a concomitant decrease in matureMn-SOD protein levels (Fig. 1). Activation of Mn-SODctivity with manganese cofactor (1 mM MnCl2) wasound to significantly reduce the effects of paraquatreatment, with notably less precursor and more ma-ure form hMn-SOD obtained at 4 days p.i. These re-ults implicate paraquat-induced superoxide genera-ion at or in close proximity to the mitochondrion as aikely mechanism in the reduced hMn-SOD processingfficiency induced by paraquat treatment. Hyperoxia isnother treatment known to lead to mitochondrial su-eroxide production [16]. Similarly to paraquat treat-ent, hyperoxic culture conditions (.95% O2) were

ound to lead to an accumulation of precursor and aeduction of mature hMn-SOD protein at day 4 p.i.Fig. 2). Like paraquat, hyperoxia was not seen toesult in measurable cellular toxicity in trypan bluexclusion tests (Fig. 3) and also as evidenced by the facthat total precursor plus mature hMn-SOD expressionas well maintained during the treatment period (Fig.). The presence of manganese cofactor was seen to beffective in preventing the processing block induced byyperoxic culture conditions. By comparison, the elec-ron transport chain blockers rotenone and potassiumyanide decreased hMn-SOD expression without therecursor/mature form ratio characteristic of the para-uat- or hyperoxia-induced processing block (Fig. 4),reatly reducing the likelihood that the effects of para-uat or hyperoxia were due to nonspecific mitochon-rial damage and deenergization.The thiol reducing agent dithiothreitol (DTT) was

ssessed for protective effects against the paraquat-

nduced hMn-SOD processing block. The addition of

1aotcoeadpmttb

(ttthdaa

iicecassmaamamrsg

gSmmwabim0crt

445MITOCHONDRIAL PROCESSING OF PRECURSOR Mn-SOD

0 mM DTT 2 h after the addition of 1 mM paraquatt day 2 p.i. was seen to completely prevent the blockf hMn-SOD processing induced by the paraquatreatment (Fig. 5). This concentration led to slightellular toxicity as evidenced by the lower total levelsf recombinant hMn-SOD found at 4 days p.i. How-ver, it was found that treatment with 5 mM DTTlso resulted in the reduction of the paraquat-in-uced processing lesion without any sign of lowerroduction of hMn-SOD during the 2- to 4-day treat-ent period. Treatment of control cells not exposed

o paraquat with these two concentrations of DTT ledo the reduction of the precursor/mature ratio toelow basal control levels.

FIG. 1. The effect of Mn-SOD activity, induced with 1 mM man-anese supplementation, on paraquat-induced inhibition of hMn-OD processing. hMn-SOD-AcNPV infected cells were cultured inedium with (lanes 1 and 3) and without (lanes 2 and 4) a 1 mManganese chloride supplementation. These cultures were treatedith 1 mM paraquat (lanes 1 and 2) or vehicle control (lanes 3 and 4)t 48 h p.i., and cells were harvested for SDS–PAGE, at 92 h p.i. Thislot and the densitometric analysis are representative of four exper-ments. The statistical analysis of the pooled ratios for each experi-

ental groups is as follows: paraquat 1 manganese (lane 1) ratio,.31 SE 6 0.04; paraquat (lane 2) ratio, 0.68 SE 6 0.08; manganeseontrol (lane 3) ratio, 0.27 SE 6 0.06; and untreated control (lane 4)atio, 0.22 SE 6 0.03. The paraquat treated group ratio (lane 2) washe only significantly higher value with P , 0.05.

The previous observation that PBzR ligands o

PK11195, Ro5-4864, and protoporphyrin IX) poten-iate the electrical activity of the MCC [10], nowhought to be the protein import channel [8, 9], led uso explore their effects on the processing of precursorMn-SOD. Treatment with PK11195 and Ro5-4864uring the 2- to 4-day p.i. period during which largemounts of precursor hMn-SOD were synthesizednd processed by the Sf-9 insect cells was seen to

FIG. 2. The effect of hyperoxic culture conditions on the process-ng of hMn-SOD. hMn-SOD-AcNPV infected Sf-9 cells were culturedn a .95% oxygen atmosphere from 48 to 84 h p.i. and compared toells cultured in a standard air atmosphere. Lanes 1 and 2 containxtracts derived from hyperoxia-treated cells. Lanes 3 and 4 containells that were cultured under standard conditions. Lanes 1 and 3re control inoculated cells; lanes 2 and 4 were cultured in mediumupplemented with manganese chloride (1 mM). The blot is repre-entative of three independent experiments. The results of densito-etric analysis that accompany the blot are shown. The statistical

nalysis of the pooled ratios of precursor to mature form hMn-SOD iss follows: the hyperoxic group ratio, 0.57 SE 6 0.08; hyperoxic 1 1M manganese supplementation ratio, 0.30 SE 6 0.06; normaloxia

tmosphere ratio, 0.16 SE 6 0.03; normaloxia with manganese (1M) supplementation ratio, 0.20 SE 6 0.07. The hyperoxic group

atios were statistically greater (P , 0.05) than those of hyperoxiaupplemented with manganese or the normal atmosphere controlroups. No significant difference was found between the ratios of any

ther groups as assessed by the Student–Newman–Keuls method.

e7lmtlddesSRl

SRhtmaHSp1Petad

onccbtal5em

mrcatim

b(chi

446 WRIGHT AND REICHENBECHER

nhance hMn-SOD precursor processing (Figs. 6 and). Ro5-4864 and PK11195 were seen to lower theevel of precursor hMn-SOD in a dose-dependent

anner while the amount of mature hMn-SOD de-ected either increased or remained unchanged. Theow basal precursor/mature hMn-SOD ratio madeetection of an increase in the processing efficiencyifficult, so that some overexposure of blots was nec-ssary to visualize the 27-kDa precursor band. De-pite this, however, a clear trend for increased hMn-OD processing was evident, especially in the case ofo5-4864, where consistently decreased precursor

evels were accompanied by increased mature hMn-

FIG. 3. Sf-9 cell viability, as measured by trypan blue exclusion,f hyperoxic culture conditions in medium with and without manga-ese (1 mM) supplementation. The hMn-SOD-AcNPV-infected Sf-9ells were cultured in a .95% oxygen atmosphere or under standardulture conditions from 48 to 84 h p.i. Cell viability was assayedased upon the ability to exclude dye in Hank’s balanced salt solu-ion containing 0.2% trypan blue at 84 h p.i. The data are expresseds a percentage of the viability of control hMn-SOD-AcNPV inocu-ated cells cultured in a standard air atmosphere. The n value 5 4 to, and the data are expressed as means 6 SE. No significant differ-nce was found in cell viability between the various culture treat-ents.

FIG. 4. Western blot analysis of the effect of treatment with theitochondrial electron transport chain blocking agents KCN and

otenone on hMn-SOD processing. Cells were treated with the indi-ated concentrations of electron transport chain blockers at 48 h p.i.nd harvested at 92 h p.i. for SDS–PAGE analysis. Untreated (con-rol) and paraquat (1 mM) treated cell cultures are also shown. Themmunoblot shown is representative of three independent experi-

tents.

OD levels (Fig. 6). Only the highest concentration ofo5-4864 (50 mM) was toxic, resulting in lower totalMn-SOD production by 4 days postinoculation. Inhe case of protoporphyrin IX, the results of treat-ent (not shown) were inconclusive concerning the

ffects of this PBzR ligand on hMn-SOD processing.owever, treatment of hMn-SOD-AcNPV inoculatedf-9 cell cultures with a high-end dose response toaraquat (0.3–3 mM), plus and minus the presence of

mM protoporphyrin IX, provided evidence thatPIX stimulates mitochondrial precursor processingfficiency. At every concentration of paraquat, PPIXreatment was seen to reduce precursor hMn-SODnd raise the mature hMn-SOD levels detectable at 4ays p.i. compared to the corresponding paraquat-

FIG. 5. The effect of the sulfhydryl reducing agent DTT on thelock of hMn-SOD processing induced by 1 mM paraquat. Paraquat1 mM) was added to the cells at 2 days p.i. with the indicatedoncentrations of DTT added shortly thereafter (2 h). The cells werearvested for SDS–PAGE and Western blotting at 84 h p.i. The blot

s representative of three independent experiments.

reated control cultures (Fig. 8).

ttpcpemrwiSatSoa

optedcdt

thgmpdmq

gscc

Pwdw(

Rwdae

447MITOCHONDRIAL PROCESSING OF PRECURSOR Mn-SOD

DISCUSSION

The robust expression of precursor hMn-SOD usinghe baculovirus expression system afforded the oppor-unity to explore several factors that might affect therocessing of cytosolic precursor proteins by the mito-hondrion in the context of the intact cell. We havereviously shown that precursor hMn-SOD is correctlyxpressed using the baculovirus system, with correctitochondrial processing and proper localization of the

ecombinant protein in Sf-9 cell mitochondria [12]. Itas found that the redox-cycling agent paraquat inhib-

ted the mitochondrial processing of precursor hMn-OD. It was also observed that recombinant Mn-SODctivity was dependent upon manganese supplementa-ion of the growth medium, with the recombinant hMn-OD imported as an inactive apoprotein in the absencef its cofactor, Mn21 ion [12]. Mitochondrial proteins

FIG. 6. Immunoblot analysis of the effect of the PBzR ligand,o5-4864, on hMn-SOD processing. Sf-9 cells were treated 2 days p.i.ith the indicated concentrations of Ro5-4865 and harvested at 4ays p.i. for SDS–PAGE and Western blotting. Vehicle control cellsre shown in lane 5. The blot is representative of three independentxperiments.

re thought to incorporate cofactors as a terminal step s

f their assembly, when the tertiary structure of therotein is “recognizable” to the cofactor [17]. The “ac-ivation” of hMn-SOD by manganese was exploitedxperimentally. Thus by the relatively simple intro-uction of a 1 mM manganese supplementation of theulture medium, the introduction of a large superoxideismutase activity was effected in a mitochondrial-argeted fashion.

The present results confirm our previous observa-ions that paraquat inhibits the processing of precursorMn-SOD into the mature form. The activation of man-anese superoxide dismutase with manganese supple-entation was found to significantly attenuate the

araquat-induced mitochondrial precursor processingefect. This directly confirms the presumed involve-ent of the superoxide radical in the effects of para-

uat treatment.Hyperoxia is well known to increase mitochondrial

eneration of superoxide [16]. Even in basal circum-tances the mitochondrial electron transport chain in-ompletely reduces ;2–3% of the molecular oxygen itonsumes to the superoxide radical. Furthermore, any

FIG. 7. Immunoblot analysis of the effect of the PBzR ligand,K11195, on hMn-SOD processing. Sf-9 cells were treated 2 days p.i.ith the indicated concentrations of PK11195 and harvested at 4ays p.i. for SDS–PAGE and Western blotting. Sf-9 cells not infectedith baculovirus (noninoculated) and untreated inoculated cells

control) are shown in lanes 1 and 6, respectively. The blot is repre-

entative of three independent experiments.

ic“esecfSdoott

oc

1leosZTcapccrdolsz

(dcmptpsdaecflbl[ruf

dbfptqplmopcrcm

htvi1lpm1p

448 WRIGHT AND REICHENBECHER

ncrease in PO2 leads to a directly proportional in-rease in the amount of superoxide that is seen toleak” from the electron transport chain [18]. The gen-ration of this superoxide, and its subsequent conver-ion to other reactive oxygen species, is widely consid-red to be the major deleterious effect of hyperoxygenoncentrations in vivo and in vitro. As expected, it wasound that hyperoxic culture conditions inhibited hMn-OD processing in a fashion similar to paraquat-in-uced inhibition. Specifically, the activation of super-xide dismutase activity blocked the inhibitory effectsf both treatments. Neither hyperoxic nor paraquatreatment was associated with overt cellular toxicity inhe time frame studied.

The potential significance of the finding that hyper-xia leads to the inhibition of precursor hMn-SOD pro-

FIG. 8. The effect of the PBzR ligand, protoporphyrin IX, onMn-SOD processing inhibited by a range of paraquat concentra-ions. Cells were treated with the reagents at 2 days p.i. and har-ested for analysis at 4 days p.i. Lane 1, untreated hMn-SOD-AcNPVnoculated control cells; lane 2, 3 mM paraquat treated cells; lane 3,mM paraquat treated cells; lane 4, 0.5 mM paraquat treated cells;

ane 5, 0.3 mM paraquat treated cells; lane 6, cells treated witharaquat (3 mM) in the presence of 1 mM protoporphrin IX; lane 7, 1M paraquat 1 1 mM protoporphrin IX; lane 8, 0.5 mM paraquat 1mM protoporphrin IX; lane 9, 0.3 mM paraquat 1 1 mM protopor-

hrin IX; lane 10, wild-type baculovirus inoculated control cultures.

essing is highlighted by the work of Clerch et al. [13, o

4]. This group documented a posttranslational col-apse of Mn-SOD expression in the lung tissue of ratsxposed to hyperoxia that closely correlated with thebserved lung dysfunction. This occurred in the ab-ence of changes of other antioxidant enzymes (i.e., Cu,n-SOD) and despite the induction of Mn-SOD mRNA.he reliance of the functional expression of Mn-SOD onorrect mitochondrial processing clearly suggests it aslikely site for a translational block of Mn-SOD ex-

ression in lung tissue. The exact nature of the pro-essing lesion that is observed in insect cells is notertain, but in the case of Mn-SOD it represents aelevant consideration for the cellular response to oxi-ative stress. The posttranslational block of expressionf a key antioxidant enzyme presents an obvious di-emma for the cellular response against oxidativetress, which consists of increasing such protective en-ymes.The observation that total hMn-SOD protein levels

precursor plus mature hMn-SOD) were maintaineduring paraquat or hyperoxia exposure indicates thatell health and energization were unaffected by theseaneuvers, in that cellular resources were sufficient to

roduce large quantities of precursor protein duringhe 2- to 4-day p.i. treatment period. In fact, the ex-ression of hMn-SOD appeared to be exquisitely sen-itive to any treatment aimed at lowering mitochon-rial oxidative phosphorylation or energy production,s evidenced by the dose-dependent loss of hMn-SODxpression in the presence of the two electron transporthain blockers, KCN and rotenone. This probably re-ects the fact that protein synthesis has been shown toe one of the cellular parameters most sensitive to ATPevels [19]. These results, together with previous data12] showing that FCCP treatment leads to similaresults in this system, emphasizes that mitochondrialncoupling or deenergization offers a poor explanationor the hMn-SOD processing defect.

It was found that prevention of the paraquat-in-uced precursor-processing defect was accomplishedy concurrent dithiothreitol treatment. DTT is an ef-ective thiol-reducing agent that is commonly usedharmacologically to assess the influence of proteinhioester linkages. The complete inhibition of the para-uat-induced processing lesion by DTT suggests arominent role for protein sulfhydryl oxidation in thisesion. Thus a simple view can be put forward that

aintains that paraquat treatment leads to oxidationf critical sulfhydryl groups (e.g., cysteine residues) ofrotein constituents of the mitochondrial import pro-essing apparatus. Consistent with this hypothesis, aecent report indicates that treatments to alkylate orrosslink SH groups on the cytosolic face of the innerembrane of plant mitochondria resulted in inhibition

f precursor protein import [20]. Reduction of SH

gm

ceacttbeeclmtta[tcdsb

tpfiMfzbdatnerPc

gtPmmpyH3cetSpl

qbrnamcd

cbtmessm

1

1

449MITOCHONDRIAL PROCESSING OF PRECURSOR Mn-SOD

roups with DTT enhanced precursor uptake by theseitochondria.The idea that a mitochondrial parameter, in this

ase precursor uptake and processing, may be influ-nced by changes in the redox status of the cell isttractive from the perspective that it allows the majorellular consumer of oxygen, the mitochondrion, to de-ect and respond to changes in oxygen tension. Thushe possibility arises that the superoxide-induced inhi-ition of recombinant hMn-SOD processing may beffected through a redox-sensing mechanism that op-rates to modulate mitochondrial protein import ac-ording to cellular redox status. Because cells contain aarge oxygen gradient in vivo [21], a redox mechanism

ay explain the finding that the subsarcolemmal mi-ochondrial subpopulation (which occupies a region ofhe cell with a relatively high PO2) imports proteins at

lower rate than the intermyofibrillar mitochondria22]. In this scenario pathologically oxidizing condi-ions, acting through this physiological mechanism,ould inappropriately signal the mitochondria to shutown protein import. If the inhibition of import isevere enough, Mn-SOD activity would be expected toe impacted despite the induction of its gene.The suggestion that the likely physiological role of

he MCC is mitochondrial protein import sheds newerspective on the previous work of others that identi-ed parameters that “modulate” the probability of theCC open state. One class of compounds that has been

ound to influence MCC activity is the peripheral ben-odiazepine receptor ligands. Several ligands haveeen found to bind selectively to this outer mitochon-rial membrane receptor, though it has not yet beenscribed a definite physiological function. They includehe benzodiazepines Ro5-4864 and diazepam; the phe-ylquinoline carboxamide, PK11195; and the endog-nously found protoporphyrin IX. The bezodiazepineeceptor ligands protoporphyrin IX, Ro5-4864, andK111195 potently increase MCC activity in patchlamp recordings [11].

The effects of peripheral benzodiazepine receptor li-ands on hMn-SOD processing were consistent withhe stimulation of mitochondrial import/processing.K11195 and Ro5-4864 were both found to lead toore efficient processing of precursor hMn-SOD to theature form in a dose-dependent manner. Protopor-

hyrin IX (PPIX) treatment appeared to be less potent,ielding more ambiguous results (data not shown).owever, the use of a high-range dose–response (0.5–.0 mM) paraquat treatment to raise the ratio of pre-ursor to mature hMn-SOD, plus or minus the pres-nce of PPIX provided the most compelling evidencehat PPIX’s general influence was to enhance hMn-OD processing (Fig. 8). At every concentration ofaraquat, PPIX improved the processing of hMn-SOD,

eading to less unprocessed precursor and greater

uantities of mature form at 4 days p.i. Peripheralenzodiazepine receptor ligands have been ascribed aegulatory role in cholesterol transport across the in-er membrane space, mitochondrial steroidogenesis,nd heme transport across the inner mitochondrialembrane [23–25]. This is the first report that impli-

ates this receptor complex’s involvement in mitochon-rial protein uptake.The data presented suggest that mitochondrial pre-

ursor processing is not constitutive, but instead maye influenced by many factors, including redox condi-ions and the PBzR. These studies also point to theitochondrial precursor processing apparatus as an

arly and susceptible cellular target of reactive oxygenpecies attack. Studies are currently under way totrengthen these assertions and to precisely define theechanism by which processing is effected.

REFERENCES

1. Lill, R., Nargany, F., and Neupert, W. (1996). Biogenesis ofmitochondrial proteins. Trends Cell Biol. 6, 56–61.

2. Wispe, J. R., Clark, J. C., Burhans, M. S., Kropp, K. E., Korf-hagen, T. R., and Whitsett, J. A. (1989). Synthesis and process-ing of the precursor for mangano-superoxide dismutase. Bio-chim. Biophys. Acta 994(1), 30–36.

3. Wispe, J. R., Clark, J. C., Dey, C. R., Neuman, J., Glasser, S. W.,Crapo, J. D., Chang, L. Y., and Whitsett, J. A. (1992). HumanMn-superoxide dismutase in pulmonary epithelial cells oftransgenic mice confers protection from oxygen injury. J. Biol.Chem. 267, 23937–23941.

4. Church, S. L., Grant, J., Ridnour, L., Oberley, L., Swanson, P.,Meltzer, P., and Trent, J. (1993). Increased manganese super-oxide dismutase expression suppresses the malignant pheno-type of human melanoma cells. Proc. Natl. Acad. Sci. USA 90,3113–3117.

5. Wong, G. H. W., and Goeddel, D. V. (1988). Induction of man-ganous superoxide dismutase by tumor necrosis factor: Possibleprotective mechanism. Science 242, 941–944.

6. Masuda, A., Longo, D. L., Kobayashi, Y., Appelia, E., and Op-penheim, J. J. (1988). Induction of mitochondrial manganesesuperoxide dismutase by interleukin 1. FASEB J. 2, 3087–3091.

7. Visner, G. A., Dougall, W. C., Wilson, J. M., Burr, T. A., andNick, H. S. (1990). Regulation of manganese superoxide dis-mutase by lipopolysaccharide, interleukin-1, and tumour necro-sis factor. Role in the acute inflammatory response. J. Biol.Chem. 265, 2856–2864.

8. Lohret, T. A., and Kinnally, K. (1995). Targeting peptides tran-siently block a mitochondrial channel. J. Biol. Chem. 270(27),15950–15953.

9. Lohret, T. A., Jensen, R. E., and Kinnally, K. W. (1997). Tim23,a protein import component of the mitochondrial inner mem-brane, is required for normal activity of the multiple conduc-tance channel, MCC. J. Cell Biol. 137(2), 377–386.

0. Kinnally, K. W., Campo, M. I., and Tedeshi, H. (1989). Mito-chondrial channel activity studied by patch-clamping mito-plasts. J. Bioenerg. Biomembr. 21(4), 497–506.

1. Kinnally, K. W., Zorov, D. B. Antonenko, Y. N. Snyder, S. S.,McEnery, M. W., and Tedeschi, H. (1993). Mitochondrial ben-

zodiazepine receptor linked to inner membrane ion channels by

1

1

1

1

1

1

1

1

2

2

2

2

2

2

RR

450 WRIGHT AND REICHENBECHER

nanomolar actions of ligands. Proc. Natl. Acad. Sci. USA 90(4),1374–13782.

2. Wright, G., Reichenbecher, V., Green, T., Wright, G. L., andWang, S. (1997). Paraquat inhibits the processing of humanmanganese-dependent superoxide dismutase by SF-9 insect cellmitochondria. Exp. Cell Res. 234, 78–84.

3. Clerch, L. B., and Massaro, D. (1993). Tolerance of rats tohyperoxia. Lung antioxidant enzyme gene expression. J. Clin.Invest. 91, 499–508.

4. Clerch, L. B., Neithardt, G., Spencer, U., Melendez, J. A., Mas-saro, G. D., and Massaro, D. (1994). Pertussis toxin treatmentalters manganese superoxide dismutase activity in lung. Evi-dence for lung oxygen toxicity in air-breathing rats. J. Clin.Invest. 93, 2482–2489.

5. O’Reilly, O., Miller, L., and Luckow, V. (1993). “BaculovirusExpression Vectors: A Laboratory Manual,” Vol. 1, Freeman,New York.

6. Freeman, B. A., and Crapo, J. D. (1981). Hyperoxia increasesoxygen radical production in rat lungs and lung mitochondria.J. Biol. Chem. 256, 10986–10992.

7. Robinson, K., and Lemire, B. (1996). Covalent attachment ofFAD to the yeast succinate dehydrogenase flavoprotein re-quires import into mitochondria, presequence removal, andfolding. J. Biol. Chem. 271, 4055–4060.

8. Turrens, J. F., and Boveris, A. (1980). Generation of superoxideanion by the NADH dehydrogenase of bovine heart mitochon-

dria. Biochem. J. 191, 421–427.

9. Brown, G. C. (1992). Control of respiration and ATP synthesisin mammalian mitochondria and cells. Biochem. J. 284, 1–13.

0. von Stedingk, E. M., Pavlov, P. F., Grinkevich, V. A., andGlaser, E. (1997). Mitochondrial protein import: Modification ofsulfhydryl groups of the inner mitochondrial membrane importmachinery in Solanum tuberosum inhibits protein import.Plant Mol. Biol. 35(6), 809–820.

1. Mainwood, G. W., and Rakusan, K. (1982). A model for intra-cellular energy transport. Can. J. Physiol. Pharmacol. 60(1),98–102.

2. Takahashi, M., and Hood, D. A. (1996). Protein import intosubsarcolemma and intermyofibrillar skeletal muscle mito-chondria. J. Biol. Chem. 271, 27285–27291.

3. Papadopoulos, V., Mukhin, A. G., Costa, E., and Krueger, K. E.(1990). The peripheral-type benzodiazepine receptor is func-tionally linked to Leydig cell steroidogenesis. J. Biol. Chem.265, 3772–3780.

4. Krueger, K. E., and Papadopoulos, V. (1990). Peripheral-typebenzodiazepine receptors mediate translocation of cholesterolfrom outer to inner mitochondrial membranes in adrenocorticalcells. J. Biol. Chem. 265, 15015–5021.

5. Taketani, S., Kohno, H., Furukawa, T., and Tokunaga R.(1995). Involvement of peripheral-type benzodiazepine recep-tors in the intracellular transport of heme and porphyrins.

J. Biochem. 117, 875.

eceived February 18, 1998evised version received September 30, 1998