Identification of a thermolabile component of the human neutrophil NADPH oxidase A model for chronic granulomatous disease caused by deficiency of the p67 phox cytosolic component

Identification of a thermolabile component of

the human neutrophil NADPH oxidase. A model

for chronic granulomatous disease caused by

deficiency of the p67-phox cytosolic

component.

R W Erickson, … , T L Leto, J T Curnutte

J Clin Invest. 1992;89(5):1587-1595. https://doi.org/10.1172/JCI115753.

Mild heating of human neutrophils inactivates the respiratory burst oxidase, producing a defect in superoxide production and bacterial killing comparable to that seen in patients afflicted with chronic granulomatous disease (CGD). We have now investigated the

mechanism and specificity of this inactivation by examining the effect of mild heating on the known oxidase components: the membrane-bound subunits of the cytochrome b558 (gp91-phox and p22-(gp91-phox) and the two cytosolic oxidase factors (p47-(gp91-phox and p67-(gp91-phox).

Heating (46 degrees C for 7.5 min) caused intact neutrophils to lose greater than 85% of their capacity to produce superoxide, a defect which was localized to the cytosolic, but not the membrane, fraction. Complementation studies with CGD cytosols deficient in either p47-phox or p67-p47-phox suggested that the defective component of heat-inactivated cytosol was p67-phox. This was confirmed by experiments showing that recombinant p67-phox, but not p47-phox, exhibited lability at 46 degrees C and completely reconstituted oxidase activity of heat-treated cytosol. These studies indicate that mild heating of either intact neutrophils or normal neutrophil cytosol results in a selective inactivation of p67-phox, providing a model oxidase system for the extremely rare p67-phox-deficient form of CGD.

Research Article

Find the latest version:

Identification

of

a

Thermolabile

Component

of the

Human

Neutrophil

NADPH

Oxidase

AModel for Chronic Granulomatous Disease Caused by Deficiencyof thep67-phox Cytosolic Component

Richard W. Erickson,* StephenE.Malawista,t MichaelC.

Garrett,'

Gretchen VanBlaricom,$ Thomas L.

Leto,*

and John T. Cumutte*

*Department of Molecular and Experimental Medicine, The ScrippsResearchInstitute, LaJolla, California 92037;tDepartment of Internal Medicine, Yale University School ofMedicine, New Haven, Connecticut 06510;and§Laboratory ofHostDefenses,

National Institute ofAllergy and Infectious Diseases, National Institutes ofHealth, Bethesda, Maryland 20892

Abstract

Mildheating of humanneutrophilsinactivates therespiratory

burstoxidase,producingadefect insuperoxide productionand bacterial killingcomparable to thatseenin patientsafflicted withchronic granulomatous disease(CGD). We havenow in-vestigated the mechanism andspecificityof thisinactivation by examining the effect of mildheatingonthe known oxidase com-ponents: the membrane-boundsubunits ofthecytochrome

_b558

(gp9l-phox andp22-phox) and the twocytosolicoxidase fac-tors (p47-phox and p67-phox). Heating (460C for 7.5 min)

caused intact neutrophilsto lose >85% of their

capacity

to

produce superoxide, a defect which was localizedtothe

cyto-solic,butnotthemembrane, fraction.

Complementation

studies with CGD cytosolsdeficient ineither p47-phoxor p67-phox

suggested that the defective component of heat-inactivated cy-tosol wasp67-phox.Thiswasconfirmedby experiments show-ing thatrecombinant p67-phox, but notp47-phox, exhibited lability at460Candcompletelyreconstituted oxidaseactivity

of heat-treated cytosol. These studies indicate that mildheating

of either intactneutrophilsornormalneutrophil cytosolresults ina selective inactivation ofp67-phox,providingamodel oxi-dase system for theextremelyrarep67-phox-deficientform of CGD.(J.Clin. Invest.1992.89:1587-1595.) Keywords: cyto-kineplast *heating*phagocyte*

p47-phox.

superoxide Introduction

Controlledheating ofhuman

neutrophils

onsurfacesproduces anucleate, granule-poor cytoplasmic fragments termed cyto-kineplasts

(CKP),'

which retain the chemotactic and

phago-Thiswork has beenpublishedin part in abstract form(1990.Clin. Res. 38:435a).

Addressreprintrequests to Dr.Curnutte, Departmentof Molecular andExperimental Medicine,SBR 12, TheScrippsResearchInstitute,

10666 North Torrey Pines Road, La Jolla, CA 92037.

Receivedfor publication 31May 1991andinrevisedform20 No-vember1991.

1.Abbreviationsused in this paper: CGD, chronicgranulomatous dis-ease; CKP, cytokineplast; GTPyS, guanosine

5'-043-thiotriphos-phate);phox, phagocyteoxidase (usedtodesignate proteincomponents of the phagocyte respiratory burst oxidase); PIPES,

1,4-piperazine-diethanesulfonic acid.

cytic function of the parent cells, but lack the capability of undergoing a respiratory burst (1-6). Subsequent studies (7)

have shown that the loss of respiratory burst activity in

cyto-plastsis directly linked to the length oftime neutrophils in suspensionarepreexposedtoelevatedtemperatures(45°C). In theneutrophil,the NADPH oxidase is responsible for the oxi-dativeburst, directing thetransportof electrons from NADPH

tomolecularoxygenaccordingtothefollowing reaction:

NADPH+ 202-- NADP+₊ 20- ₊ H+

This respiratory oxidase isamulticomponentenzymecomplex that is associated with the plasma membrane after activation of the cellbyavariety of stimuli (8-10). The absence ofa

respira-toryburst in CKP preparations functionally resembles

neutro-philsof patients suffering from the inherited disorderchronic

granulomatousdisease (CGD), in which afflicted individuals

areincapable of initiatingorsustaining the neutrophil

respira-toryburst (1 1). The study of this disease has helped identifyat

least four oxidase components, two of which are membrane-bound andtwoof whicharecytosolicintheresting neutrophil. InX-linked inheritance, themostcommonform of CGD (55% of CGDcases[12]), genetic alterations ingenelocusXp21.1 of

theX-chromosome resultinthe complete lack of the 91-kD

trans-membrane glycoprotein subunit (gp9l-phox[for

phago-cyteoxidase]) of the heterodimeric cytochrome b558 (13, 14). This mode ofinheritance is also marked by a concomitant loss of the cytochrome 22-kD subunit, p22-phox (15-17), although thegeneencoding this protein has been localizedto

chromo-some 16(gene locus 16q24 [18]). Similarly, mutations involv-ing the p22-phox gene (18), involvinvolv-ing- 5%of CGD patients

( 12), also leadtothe absence ofboth cytochrome subunits(17).

Incertainrarevariantforms of CGD, mutationsin the genefor

eithergp9l-phox orp22-phox lead to a cytochrome complex, which is quantitatively and spectrally normal but is nonfunc-tional(19-22). Of the remaining 40% of CGDcasesinwhich cytochrome b558 is functionally normal, a complete absence of

oneofthe two known cytosolic components of the oxidase is

seen: p47-phox (35% of CGD cases), encoded on gene locus 7ql1.23, and p67-phox (5% of CGD cases), encoded on gene locus1q25 (12, 23-31). Given the apparent similarity of respi-ratory burst oxidase dysfunction in CKP and CGD

neutro-phils,weexplored the possibilitythatheatedneutrophils might

suffer from anacquired defect in one or more oxidase

compo-nents.

In this reportwe show that mild heating inactivates the neutrophil respiratory burst oxidase through a mechanism

in-volving the specific thermolability of the p67-phox cytosolic component. The rarity ofp67-phox-deficient CGD patients

has madeit difficult to study the functional role of p67-phox

J.Clin. Invest.

©TheAmerican Society for Clinical Investigation, Inc. 0021-9738/92/05/1587/09 $2.00

andto preparep67-deficientcytosolforusein cell-free

comple-mentation studiestoidentifythemoleculardefect inCGD

pa-tients. Wenowdescribeamethod which both enables the

pro-duction ofa model p67-phox-deficient cytosol system and demonstrates the mechanism by which the respiratory burst oxidase is disabled by mild heattreatment.

Methods

Allreagents used wereofthe bestgradecommercially available, and wereobtained fromsources previouslydescribed (32).

Preparation of neutrophils. Blood was obtainedinmost cases by venipuncture from normal donorsand CGD patients with their in-formedconsent.Neutrophilswerepreparedusing acid-citrate-dextrose asthe anticoagulant, dextran to sediment erythrocytes, and Ficoll-Paque(PharmaciaLKB,Uppsala,Sweden)density-gradient centrifuga-tiontoseparate mononuclear cellsfromneutrophilsaspreviously de-scribed(33). Forexperiments requiring purified cytosolicand mem-branefractions, neutrophilswereobtained byleukapheresis ofdonors andpatients (after acquiring their informed consent)andpurifiedas reportedbefore(34). Normal donors wereadministered dexametha-sone(4 mgorally12and 2 hbefore procedure)toincrease neutrophil yields. Neutrophilsweremaintainedat4VCatall stagesofpurification afterdextransedimentation.

Intact neutrophil functional studies. Purified neutrophils were heated at460C for7.5 minand assayedfor O° productionasdetailed below, whilecontrolneutrophilswerekeptonice.Degranulationwas

assessedbyquantitatingthe extracellular releaseof vitamin

B12-bind-ing protein fromphorbol12-myristate 13-acetate(PMA)-treated neu-trophilsusingapreviously describedmethodfor measuringthebinding of[57Co]vitamin B12tothisprotein(35, 36). Neutrophil bactericidal activitywasmeasuredinvitrousingStaphylococcusaureusstrain502A asoutlined earlier (36, 37).

Neutrophilfractionation. Unstimulated neutrophilsweredisrupted bynitrogen cavitation andpurifiedmembrane andcytosolfractions were prepared as previously described (32-34, 38) and stored at -70°C.Membranefractionswerebroughtto 1.25X 109cell

equiva-lents/ml (- 5 mgprotein/ml) in5 mM1,4-piperazinediethanesulfonic

acid (PIPES),50 mMKC1, 1.75 mMMgCl2, 1.5 mMNaCl,0.5 mM ATP, 0.63mM EGTA, 0.34Msucrose,pH7.3;cytosolic materialwas preparedat9X I07cellequivalents/ml (- 2 mgprotein/ml)in 10 mM PIPES, 100 mM KCI, 3.5 mMMgCl2, 3 mM NaCl, pH 7.3 (also termed "relaxbuffer") supplementedwith 1 mM ATP and 1.25 mM EGTA.Sodiumdeoxycholate-solubilizedmembraneswereprepared accordingto apublishedmethod(32)forusein thecell-free assay of superoxide production.

Heat treatment of neutrophils and cellfractions. Purified neutro-philsweresuspendedat 108cells/mlin25°Cphosphate-bufferedsaline (PBS)containing 138 mMNaCl,2.7 mMKCI,8.1 mM Na2HPO4, 1.47mM KH2PO4,pH 7.3. One volumeof this suspensionwasthen addedto3 volofPBSpreequilibrated to46°Cinasiliconized glass vessel equipped with amagnetic stirrer. Aliquots were removed at

various times and stored on icefor 15 min before assay forsuperoxide production.

Purified neutrophilcytosol was removed from storage at -70°C and allowed to come to room temperature before heating. Cytosols weretransferred to siliconized glass tubes and heated in a 46°C water bathwith constantagitation forup to 20 min.Aliquotswereremoved attimedintervals,transferred to ice for 5 min, and then allowedto

equilibrateat roomtemperature beforeanalysisin thecell-free oxidase activation system.

Purified membrane vesicles were allowedtothaw at room tempera-ture, and then transferred to asiliconizedglass tube for heat treatment. Single-volume aliquotswereremoved atvarious times,anddiluted with 5 vol relaxbufferat4°C.Samples were then incubated on ice for 10 minbefore analyzing oxidaseactivityin the presence of normal cytosol.

Whole-cellsuperoxide production.Therateofgeneration of super-oxidebyintactneutrophilswasdeterminedat370C inaThermoMax kineticmicroplatereaderequippedwitha550±1nmfilter (Molecular DevicesCorp.,MenloPark, CA)aspreviously described (39).Inthis assay(totalvolume 0.25ml), cytochromecreduction inapair of reac-tions(oneof which contained 15Mgofsuperoxidedismutase [SOD])

wascontinuouslymeasuredat550nmafter stimulation ofthe neutro-phil suspension byPMAat afinalconcentration of 200ng/ml.

Maxi-mumreaction velocitiesweredeterminedusingSoftMaxkinetic analy-sis software(version 2.01,Molecular DevicesCorp.).The SOD-inhibit-ablerateofsuperoxide productionwascalculated by subtractingthe

velocityoftheSOD-containingreaction from that measured in its

companion reaction without SOD. Absorbance changes were con-verted to nanomoles °2 as described previously for this 550-nm filter(39).

Cell-freeassayofoxidaseactivation.The kinetics of activation of NADPHoxidasewerestudiedin acell-freesystemusingmembranes and cytosol from unstimulated neutrophils as previously described

(40).Inthis assay system, theoxidase is activatedat250C bythe addi-tion of SDS in the presence of NADPH(0.16mM)andcytochromec

(0.1 mM).Reaction mixturesof 0.15mltotal volume wereincubated in 96-well Linbro microtitration plates (Flow Laboratories, Inc.,

McLean, VA)andcontained cytosol (0-2X 106cellequivalents)and eithermembranevesicles(4X 105cellequivalents)or deoxycholate-solubilizedmembranes(1.25X 106cellequivalents). The optimal con-centration of SDS foractivation of the oxidasewas40MM for

experi-ments using deoxycholate-solubilized membranes and 90 MM for assaysusingmembranevesicles.Again,measurements ofSOD-inhibit-ablecytochrome creduction were obtainedon aThermomaxplate

reader,withsubsequentdetermination of maximal velocitiesby Soft-Maxsoftwareandconversiontonanomoles °2 asdescribedabove.

Complementationof CGDpatientandheat-treatedcytosol frac-tionswasaccomplished by adding equalvolumesoftestcytosols (2

X 106cellequivalents)tothecomplete cell-freeassayusing deoxycho-late-solubilizednormal membranes.This volumeofnormalcytosol (2

X 106cellequivalents)wassufficienttogenerate NADPHoxidase

activ-ity>40 nmol_Oj/minper107cellequivalentsmembranes. All other parametersofthecell-freeassay systemremainedasdescribedabove.

Thedeterminations of CGDpatientandheat-treatedcytosolrate ordercharacteristicsweremade in thecell-free oxidasesystem, as

previ-ously reported (34, 41).Eitherrelaxbufferor an excess(3.6 X 106cell

equivalents)oftestcytosolwasaddedtoreaction mixtures (0.15 ml)

containingnormalcytosolinvaryingamounts(3.6x I05to1.6 x106

cellequivalents).Thesecytosolmixtureswereassayed forsuperoxide

production using deoxycholate-solubilizednormal membranes(1.25

X 106cellequivalentsperreaction). Forkineticstudies ofheated or

p67-phox-deficient cytosol reconstituted with recombinant oxidase factors(seebelow),asimilarcell-freeassaywasemployed. Again,

vary-ingamounts(3.6XI0 to1.6X 106cellequivalents)ofeitherheated or

p67-phox-deficientCGD_cytosolwereaddedto0.15 mlreaction

mix-turescontainingrecombinantp47-orp67-phoxpresentat aconstant

stoichiometryof1.7

Mg

of recombinantproteinper 106 cellequivalents

cytosol.Assayswerecarriedoutusingdeoxycholate-solubilizednormal membranes(1.25 X 106 cellequivalents perreaction) as described above.

Preparation, heat treatment, and cell-freeassay ofpreactivated NADPHoxidase. Purifiedneutrophilsweresuspendedat5X _I07cells/

mlin PBS(supplementedwith 0.5 mMMgCl2,0.9 mM CaCI2,and 7.5 mMglucose)andstimulated with 200ng/mlPMAfor 10 minat37°C. Theseactivated_neutrophilswerewashedtwice with PBS and resus-pendedat108cells/mlin 0.34 Msucrosebufferedwith 10 mMHepes, pH7.3.This suspensionwassonicated for three consecutive bursts of 10seach,usingaSonifier celldisruptormodel W185(Heat

Systems-Ultrasonics,Inc.,Plainview,NY)at aconstantpower of 50 W.A

low-speedcentrifugation (500g, 10min, 4°C)wasperformedtosediment

remainingintact cells anddebris, and theresultingsupernatantwas

removedto wetice for immediateuse.Thispreactivatedoxidasewas

was added to a5.0-ml polypropylene test tube prewarmed at either 37 or460C.Heated incubationswerecarriedoutwithconstantagitation ofthe suspension, removing0.07-mlaliquots directly to iced microcen-trifuge tubesat1 min intervals. NADPH oxidaseactivitywasmeasured asdescribedabovefor the cell-free activation assay with thefollowing modifications: no SDSwasused, the total assay volume wasincreased to0.75 ml, and reactionswere runinsemi-microcuvettesutilizinga

Uvikon model 810 dual-beam spectrophotometer (Kontron Electron-ics,Inc., RedwoodCity, CA). Reactions containing3X 106cell equiva-lentsactivated oxidase(0.03ml) were initiated by the additionof0.16 mM NADPH.TheSOD-inhibitable reduction ofcytochrome c was

measuredcontinuouslyat550nmwith theSOD-containing reactionin the reference beam. Maximum reaction velocitiesweredeterminedby visualestimation. Incertain experiments, recombinantp47- and p67-phox (5 _ggof either [or both]) wereaddedto reactionscontaining preactivated oxidase (3 X 106cell equivalents)thatwas heat-inacti-vated(460C,7.5 min) todetermine whetheroxidase activitycould be restored. The completereactionmixture was then allowedtoincubate

at250C for3 minbeforeinitiationof superoxideproduction bythe additionofNADPH.

Recombinant protein studies. Both recombinant p47-phox and p67-phoxwerederived from baculovirusconstructs(BV/p47and BV/ p67)asdescribedelsewhere(42).Eachoftheseeukaryotic expression vectorscontains the complete coding sequencesofone ofthese two cytosolic oxidasecomponents,enabling productionof complete native proteinslacking anyadditional fusionprotein sequences. The recombi-nantproteins werepurified from baculovirus-infected insectcell (Sf9) lysates 3-4 d postinfection, using ion-exchange chromatography. Theseproteins exhibitedthesamemolecularweightsasthenative

neu-trophil proteinsandwereestimatedtobe>90% purebySDS-PAGE. The pure proteinsweredilutedto0.2 mg/mlwith relaxbuffer and heated for various time periodsat46°C, asdescribed above. These solutions contained 0.1 mMdithiothreitoltoinhibit oxidative damage of proteins.Thecell-freeassay system usedto testtheactivityof these recombinantfactorswasessentiallythesame asthatdescribedabove, with modifications(42).Briefly,these reactions(0.1 mlfinal volume) contained variableamountsofpurerecombinant p47-phoxor p67-phox, I05cellequivalentsofneutrophil cytosol,and 5XI05cell equiva-lentsofdeoxycholate-solubilized membranes. Theoxidase reactions wereactivated by addition of 40MMarachidonic acid instead of SDS. Inexperimentsinvolvingtherestoration ofheat-treatedcytosol with therecombinantfactors, thecytosolwasheatedfor7.5 minat46°C before use inareaction containing 106cellequivalents ofthe heat-treated cytosol, 5 X I05cell equivalents ofdeoxycholate-solubilized membranes, andvariableamounts(0-1,000 ng)of therecombinant cytosolic factors.

SDS-PAGE andWesternblotting.Heat-treatedneutrophil-derived

cytosol, heated normalcytosol, andrecombinant cytosolic proteins weresubjectedtodiscontinuousSDS-PAGEprior to electrophoretic transfertoZeta-Probe(BioRad Laboratories, Richmond, CA)or nitro-celluloseblottingmembranes,accordingto ageneral method(43,44).

Ananti-peptide antibodyraised inarabbitagainstamino acidresidues 447-460(DEP K ES E K AD AN NQ)of thep67-phoxcomponent of theneutrophilNADPHoxidasewasusedastheprimary antibody for the cytosolimmunoblots.Therecombinant proteins were probed with goatantibodiesraised against these purified factors as described elsewhere(42).Inboth cases, the boundprimary antibodieswere de-tected with appropriate species-specific secondary antibody-alkaline phosphatase conjugates (BioRad Laboratories) using 5-bromo-4-chloro-3-indolyl phosphateandp-nitrobluetetrazoliumassubstrates. Dataanalysisand curvefitting. Allvariances shown are standard deviations, unlessotherwise noted. Theexponential relationship

be-tweencytosolconcentration and maximalratevelocityin thecell-free assay systemwasdetermined bytwomethods: (a) nonlinearregression analysis power curve fitting using Enzfitter(Elsevier-Biosoft; Cam-bridge, UK)dataanalysis software;and(b) least-squareslinear regres-sionsoflog-logtransformplotsofmaximum reactionvelocityvs.

nor-malcytosolconcentration(34). Enzfitter powercurveanalyseswere

basedon theequation:V=K(C'),where V=maximalrate velocity, K =constant,C=cytosol concentration,and P = rate order.

Results

Thecontrolledheating of intact neutrophils at temperatures

ranging from 39to470C has been shown to decrease the

capa-bilityofthese cells toproduce superoxide (7, 45).For the

exper-iments described in thisreport,460Cwaschosenastheoptimal

temperatureforinactivating the respiratoryburstonthe basis

of extensiveempirical trials. Neutrophils preincubatedat460C forvarying lengths of timeweresubsequently assayed for their ability to produce superoxide in response to stimulation by

PMAat370C (Fig. 1). Superoxide production remained largely unaffected after 2 min at460C, and then abruptly declined

overthenext3 min. By 8 minanalmostcomplete abolishment ofsuperoxide production wasobserved. Controlexperiments showed that preincubationat460C forup to 10min didnot

induce neutrophils to produce superoxide in the absence of PMA, nordid it affect their viability as measured bytrypan

bluedye exclusion (> 95% viable cells).This patternof

heat-in-duced inactivation ofrespiratory burst activity is comparable

tothatpreviously described (7).

To obtain abroader understanding of the effect of mild heating on neutrophils,wealsostudiedtwootherfacets of

neu-trophil function: bacterial killing and degranulation. Table I

summarizes these results andcompares the functional

capabili-ties of heatedneutrophilstothose obtained from CGDpatients with either the standard or variant form ofthe disease. A strong resemblence is seen betweenthese heated neutrophils and their

CGDcounterparts,particularlyinthe reducedcapacityofboth

heated andCGD neutrophils to kill S. aureus.

One of themanypossible mechanismstoexplainthe46°C inactivation ofthe respiratory burst was that theNADPH oxi-daseitself,or oneof itssubunits,wasdamaged by theheating.

To test thispossibility, purified membrane and cytosol

frac-tions from46°C-treated neutrophils disrupted by nitrogen

cavi-14

o a 0

-0 0

°L

0

00C.

x0,

CC

0 _L

C/)~

0 2 4 6 8 10

Time at 46 degrees (minutes)

Figure1. Inhibition of intact neutrophil superoxide production by preincubationat460C.Neutrophilswereheatedat2.5X I07cells/ml in PBS andaliquotsremovedat30-sintervals.Superoxide production afterstimulation with 200ng/mlPMA wasassayedasdescribed in Methodsusingakineticmicroplatereader and computer software that calculated the maximalratesof absorbancechangeat550nm.

TableI. Selected Functional Properties ofHeated Neutrophils

Neutrophils Functional

assay Heated CGD VariantCGD

percentcontrol activity

Superoxide

production 5.8±4.2 (n=4) 0.4±0.8(n=12) 5.8 (n =2)

Bacterialkilling:

1 h 44.6±15.6 (n=3) 50.9±35.9 (n=12) 48.8(n=2)

2 h 60.9±34.5 (n=3) 68.5±32.1(n=12) 45.7(n =2) Degranulation 45.3±21.0 (n=3) Notdetermined 57.6 (n = 2)

Purifiedneutrophilswereincubatedat4VC(control)or460C for 7.5min.

Functionalassays wereperformedasoutlined in Methods, and the resultsare

presentedaspercentageof (unheated) control activity. 14CGDpatients (none ofwhom weretakinginterferon-yatthetime)were also studied. 12 of these

patients hadacompleteabsenceofrespiratoryburstactivity (duetomutations ingp9 I -phox, p22-phox,p47-phox,andp67-phox) while2hada variantform of X-linked CGDinwhich therewas aresidual level(-5%ofnormal)of

°2-production owingto a95%deficiency of cytochromeb. The control level of

°2-productionwas 149.2±33.8nmol/minper IO' cells (n=14). Normalbacterial killing (expressedas thepercentcolony-forming units killed by incubation with

controlneutrophils)was46.7±18.8%(n=7)after 1 h ofincubation,and

66.4±19.0% (n=7)after 2 h of incubation. Controlneutrophilsreleased

42.0±22.7%(n =7) of theirtotalvitaminB,2-binding protein intothe extracel-lularmedium(degranulation)after30 minofexposure to PMA.

tationwereanalyzedinacell-free oxidase activationsystem. As

shown in Table II, membranesfromboth unheated (control) and heat-treated (46°) neutrophils showed normal activity

when activated with SDS inthepresenceofcontrol cytosol. In contrast,cytosolic fractions fromheatedneutrophilsshoweda

dramatic reduction in activitywhen comparedtothat observed with untreated cytosol. In control experiments, membrane

fractions fromheat-treatedneutrophilsshowed normalactivity in thecell-freesystemwhen tested atconcentrationsranging above and below that shown in TableII(datanotshown).In

parallel experiments performedwithCKP,therewas asimilar lossofcytosol oxidase activitywhen assayed in the cell-free oxidase activationsystem(datanotshown).

Giventherelatively selective inactivation ofthecytosolic fraction,we nextinvestigatedwhether the effect could be

dupli-catedby 46°Ctreatmentofcytosol and membranefractions

obtained from unheatedneutrophils. Asshown inFig. 2, the heated membrane fraction showed only mildly compromised

oxidase activitywhen assayedwith controlcytosol, while the heat-treatedcytosolic fractiondemonstratedatime-dependent

lossofoxidaseactivitywhenassayed usingcontrolmembrane

vesicles.Baseduponthesedata,wefoundthatwecould

repro-ducibly generate cytosols inhibited by >95% by heating at

46°C for20 min.Addition of10,uMguanosine

5'-0-(3-thiotri-phosphate) (GTPyS) failed to restore any activity in these heated cytosolpreparations,anindicationthat lossofoxidase

activityin46°C-treatedcytosol isnotduetothermaldamage of this essential nucleotide cofactor (data not shown). A func-tionaldefect wasthus shown tobe inducible in (previously)

normalcytosol,indicatingtheinactivation by heating ofa

cyto-solic oxidasecomponent(orcomponents).

Determination of the heat-labile cytosolic oxidase compo-nent wasaccomplished by cytosol complementation studiesin

the cell-freeassay system (Table III). Cytosolsprepared from CGDpatientswith identical modes of inheritanceare

incapa-ble ofassociatingwith each otherin the presence of normal

membranes to form activatable oxidase, whereas cytosol frac-tions prepared from patients with different cytosolic oxidase component deficiencies are capable of complementation in the cell-free oxidase activation system (26, 40). Normal cytosol that was heat-treated at 460C for 20 min was assayed for com-plementation with four different cytosol proparations from CGD patients lacking either p47-phox (patients CGD1 and CGD2) or p67-phox (patients CGD3 and CGD4). No oxidase activity was detectable in reactions combining heated cytosol withp67-phox-deficientcytosol,suggesting that the p67-phox component ofheatedcytosol is functionally inactive. In con-trast, heated cytosol was capable of complementing p47-phox-deficient cytosols, demonstrating a retention of p47-phox func-tion in the heat-treated cytosol.

Kineticstudies were undertaken to determine further func-tional similarities between heated normal and p67-phox-defi-cient CGD cytosols. We have previously reported (34, 38, 41) that the order of the oxidase-activating reaction with respect to normal cytosol concentration is 2.52±0.09 SE (n=7), afinding which implies that at least three kinetically independent solu-ble components (or component complexes) are required for the formation ofactiveoxidase. In these same studies, when oxi-dase was activated at varying concentrations of normal cytosol inthe presence of an excess of p47-phox (or

p67-phox)-defi-cient cytosol, the yield of active oxidase no longer varied as the 2.5 power of normal cytosol concentration. Instead, oxidase activity increased in direct proportion to the concentration of normal cytosol, indicating that the defective cytosols were con-tributing all but one of the required kinetic components. The results of similar experiments performed with cytosol from

460C-heated

neutrophils or heat-treated cytosol are shown in Fig. 3, where log-log plots of normal (unheated) cytosol con-centration vs. maximum reaction velocity are shown. The slope of each line defines the order of the oxidase-activating reaction with respect to normal cytosol concentration. In the absence of heat-inactivated cytosols, normal control cytosol showed the

expected slope

of - _{2.5. The presence of}excess heat-treated cytosol altered the dependence of oxidase yield on

TableI. Cell-free Activation of NADPH Oxidase Using

Membranes and CytosolDerivedfrom Control and 46°C-treated IntactNeutrophils

Sourceof fraction

Cytosol Membrane Superoxideproduced

nmolOjminper10' cell equivalentsmembranes

Control Control 38.5

Control 460C 28.7

460C Control 4.5*

460C 460C 2.9

Experiments were carried out as described in Methods. Neutrophils were incubated at460Cor4°C(control) for 7.5minbeforedisruption and preparation of cytosol and membrane fractions. Each reaction

(0.15ml) contained 4X IO'cellequivalents of membrane vesicles (fromeither control or460C-treatedneutrophils) and 2 X 106 cell

O :t

ip 4._

V. -W

0

o3 < L.

CL C

0

G) 0

x °

o

L C.

0 0

CL 0

V _QE

1004

80

60

40

.l

0 2 4 6 8 10 12 14

Time

at 46 degrees

(minutes)

Figure 2.Effect of460C heatingonoxidase function in membranes andcytosol.Cytosolicand membranefractions fromuntreated nor-malneutrophilswereobtained bynitrogen cavitationasdescribedin Methods. Thesefractionsweresubjectedto46°Cheat treatment, and aliquotswereremoved atthe timesindicated foranalysisof superox-ideproductionin thecell-free oxidase activationsystem. In the upper panel,heat-treated membranes(3.2x 105cellequivalents)were as-sayed for oxidaseactivity withnormaluntreatedcytosol (1.8X 106 cellequivalents).Inthe lowerpanel,heatedcytosol(2.5 X 106cell equivalents)wasassayed foractivity usingnormaluntreated deoxy-cholate-solubilizedmembranes(6.3X I05cellequivalents).Control cytosolicandmembranefractionswerekeptat room temperaturefor thedurationof the time coursepriortodetermination of oxidase activity.Controlactivitywasdefinedasthat observed when cytosol (lowerpanel) or membranes (upper panel) incubated at 25°C for 14 min weresubstituted for the 46°C-treatedfractionin thecell-free as-say. Controlactivitieswere63.3±12.3 and 73.8±18.2 nmol0/min per107cellequivalentsmembranesforthe upper and lowerpanels, respectively. The results shown are the means±SDofthree experi-ments, eachperformed with cytosoland membranesfromadifferent donor.

untreated cytosol concentration from 2.5 order to approxi-mately first order. Thesefindingsindicate that heated cytosols appeartosuffer from a defect in one kinetic component, much like cytosols obtained from CGD patients lacking either

p47-phoxorp67-phox.Coupled with the data presentedin Tables II and III, these results strongly suggest that heat treatment of neutrophils selectively inactivates the p67-phox cytosolic oxi-dase component.

Thespecificthermolabilityofp67-phoxwasdirectly demon-stratedbythe 46°C heat treatment of the purified recombinant cytosolic oxidase factorsrp47-phoxandrp67-phox (Fig. 4). The effects of heat treatment on the two recombinant factors were

assessed under conditions in which the reconstituted oxidase activitywasthemostsensitivetothe addition of these exoge-nousfactors. Previous work has shown that thetwo

recombi-nantfactors (whencombined with plasma membranes) donot supportoxidase activation without the addition of whole

cyto-sol,indicatingtherequirementof other solublecomponent(s) (42). The oxidase activity of reactions (0.1 ml) containing 5

X 105 cell equivalents of membranes and a suboptimum

amountofwhole cytosol

(IO'

cell equivalents)wasaugmented

by>20-foldby the addition of1

gg

of each ofthe recombinant cytosolic factors (under these conditions,>95%ofcytochrome

creductionwasinhibitableby superoxide dismutase).

Pretreat-mentof these recombinant cytosolic factorsat460C revealed that thep67-phoxcomponent wasmarkedly sensitivetoheat. While rp47-phox showed onlyaslight decrease in its abilityto supportoxidase activity after preincubationat460C foraslong

as 10

min,

heattreatmentof rp67-phox resulted ina

progres-sive lossof its activity with time (Fig. 4). By 6

min,

>90% of rp67-phoxactivitywaslost. These inactivation kinetics closely resemble those observed with cytosol treated in thesame man-ner(Fig. 2), suggesting that there isadirect thermal effecton

p67-phox that doesnotinvolve the participation of other

fac-tors inwhole cytosol.Westernblottingof heatedcytosolorthe

pure recombinant proteins confirmed that the heat-treated p67-phox hadretainedthesameapparent molecularweightas

intactp67-phox (data not shown). It is therefore unlikelythat

the heat inactivationprocessinvolved proteolysis.

Thestrongestevidenceindicating that p67-phox istheonly oxidase component substantially affected by this mild heat

treatment wasobtained from experiments in which the

recom-binantcytosolic factorswereusedtoreconstitute the activity of heat-treated normal cytosol (Table IV). Normal cytosol was

heated at 46°C for 7.5 min, yielding activities that were

re-Table III. Complementationamong Various TypesofCGD

Cytosols DeficientinOxidase Components and460C-treated

Cytosol

Superoxide production

Cytosol2

Cytosol1 Normal 46°C CGD, CGD2 CGD3 CGD4

nmolOIminper 107 cellequivalents membranes

Buffer 44.1 0.4 0.2 0.8 0.1 0.1

460C 0.4 13.3 21.1 0.6 0.1

CGD, 1.4 2.1 18.3 25.0

CGD2 5.1 39.0 33.2

CGD3 0.1 1.0

CGD4 0.4

NADPHoxidase was activated in a cell-free system asdescribedin Table IIexcept that each 0.15-mlreaction contained amixture of 22

ul(2 X 106cell equivalents)ofcytosol 1 (orrelaxbuffer, where indicated)and22ul (2 x 106cellequivalents)of cytosol 2. The heat-treatedcytosolwasprepared byheatingnormalcytosolfor 20 min

at46°C.Patients1and 2(CGD,andCGD2)weredeficient in

p47-phox, patients3 and4(CGD3andCGD4)were deficientinp67-phox.

The results shown arerepresentativeof sixexperiments,each per-formedusingheatedcytosolfromadifferent normal donor.Atotal of sevendifferentp47-phox-deficient patientsand two different

tosol from Heated Cells Heated Cytosol couldexplainthe well-established heat lability of NADPH

oxi-dase in its activated state (46). Intheseexperiments,

neutro-m=0.84 1.6 m=0.82 phils stimulated withPMAand thensonicatedwereusedas a

source of activated oxidase. As shownin Fig. 6, this type of

1.2 0 oxidase

preparation

proved

tobeeven moresensitiveto

heat-ing

at

460C

than either intact

neutrophils

ornormal

cytosol

0.8

(Figs.

1 and

2).

The levelof

activity

of

preactivated

oxidasewas 4Control y Control reduced from 22.0±3.2 to 1.4±0.6 nmolO/minper 107 cell

m=2.47 m=2.51 equivalents(n= 3),after heatingat460Cfor7.5min.

Recombi-0.4 _{nantp47- and p67-phox failed}to augment the activity of this heatedpreactivated oxidase: additionof5,000ng rp47-phox or 6.4 6.6 6.8 7.0 6.4 6.6 6.8 7.0 rp67-phox resulted in

1.3±0.3

and

1.2±0.3

nmolO/min per

107cellequivalents, respectively(n=3). The addition of 5,000 log (Unheated

cy'tosol

concentration) ng of both recombinant oxidase factors also failed to reconsti-Effect of460C-treatedcytosols on the order of the oxidase- tuteactivity(1.21±0.28nmolO-/minper

I07

cell equivalents reaction with respect to normal unheated cytosol concen- [n = 3]).Unlike the situationobserved withcytosol and mem-he cell-free oxidase assays were carried out as described in branes from unstimulated neutrophils, 460C heating of preac-using normal membranes at 1.25 X 106 cell equivalents per tivated NADPH oxidase produces a defect that cannot be re-action andvarying concentrations of normal (unheated) paired by the addition of rp47- and rp67-phox.

cytosol

as

shown.

The

reactions were

pertormed

in

the

absence(C)or

presence (-) of an excess amountofheat-treated cytosol (3.6X 10'

cellequivalentsper 0.1 5-mlreaction).The heat-treatedcytosol used in theexperimentdepicted in the left panel wasderivedfrom heated neutrophils,whereasthat used in therightpanel was prepared by heating normal cytosol at460Cfor 20 min. The slopes of the lines (m)reflectthe order of the activationreactionwith respecttonormal (unheated) cytosol concentration.Theresults in theleft panelare

representative oftwoseparateexperiments;those in therightpanel

areselectedfrom 10differentexperiments.Forthese 10 determina-tions,theslopesin the absence and presence of heatedcytosolwere

2.40±0.08and 0.98±0.29,respectively.

Discussion

Wehave demonstrated that heat treatment of either normal neutrophils or their cytosol selectively inactivates a single

NADPHoxidasecomponent, giving risetoacytosolbearinga

closefunctionalresemblancetothatderived froma rareclass of

CGDpatients suffering fromaninherited loss of p67-phox. In

thisregard, heat inactivation of neutrophil cytosolcanprovide

1c

ducedby>92%whenassayedwithnormal deoxycholate-solu-bilized membranesinthe cell-free system. Theaddition of in-creasing amounts ofrp47-phox failed to restore the oxidase activity ofheatedcytosol,while almostcompleterecovery(to activity levels closetothose observed withunheatedcytosol)

was seenwith theaddition of phox. The quantity of

rp67-phox neededtoattainnearlyfull restoration ofactivity (200-400 ng) wasclose to that required for full reconstitution of

p67-phox-deficientCGDcytosol(42)and consistent with the level ofp67-phox estimatedto bepresentinnormalneutrophil cytosol (- 100 ng/106 cell equivalents) (42). Recombinant p67-phoxalsoreconstitutedthe kineticactivityofheated cyto-solasmeasuredby therate order constantrelating cytosol

con-centrationtooxidaseactivity.As showninFig. 5,heated cyto-solsupplementedwithrp67-phoxshoweda meanrate orderof 2.12±0.32,avaluesimilarto those observed with normal cyto-sol (2.44±0.08) and with p67-phox-deficient cytosol supple-mentedwithrp67-phox (2.42±0.07).The levelofactivityin the reconstituted reaction mixtureswasless than thatseenin

ex-periments 1and 2 inTable IV wherenearnormalactivitywas obtained. Thisapparently isdue tovariations in the specific activitiesofdifferentpreparationsofrp67-phox. The reconsti-tutionofboth oxidaseactivityandnormal kinetic behavior in heatedcytosolby rp67-phox providesstrongevidence that the defectinoxidaseactivityproduced bymildheatingofnormal neutrophils is duetothespecific thermolability ofone of the oxidasecomponents,p67-phox.

The susceptibility of p67-phoxto mild heating led us to

investigate whether the loss of function of this component

80

860

Heated Component

0 ₄₀ _ rp47phox

e.

te

A

-Yv rp67phox

XCL 2

V) 0-2

20

*

0 2 4 6 8 10

Time at 46 degrees (minutes)

Figure4.Effect of460Cheatingonrecombinant cytosol oxidase fac-tors.Purerecombinantp47-phox (e)orp67-phox (v)washeatedat

460Cat30-s intervals for indicatedtimes, asdescribed in Methods. Theactivities of theseproteinsweretestedinthecell-freeoxidase

activationsystemusing1 _ggeach ofrp47-phoxandrp67-phox (either heatedorunheated), 105cellequivalents ofwholenormalcytosol,

and 5X I05cellequivalentsofdeoxycholate-solubilizedmembranes.

The datapointsareaveragesofduplicate reaction mixtures (assayed

simultaneously) obtained inarepresentative experiment.Inthetwo

suchexperiments performed,theaverageactivities of reactions

sup-plemented with unheatedrp47-phoxorrp67-phoxwere20.2 and 15.9

nmol O2/minperI07cellequivalent membranes, respectively.The

averageactivityof reactions thatlacked recombinant factorswas0.8

nmolO/minper 107cellequivalentsmembranes. The kinetics of therp67-phoxinactivationwerenearlyidentical in thetwo

experi-ments:at 2min,54.6% and62.6% ofcontrolactivity;at4min,22.5% and 19.1%;at6min7.3%and7.9%;andat8min,8.3% and 6.1%.

Cyt

% 1.6

0

c _1.2 0

E

g 0.8

x %- 0.4

0

0.0

Figure 3. 1 activatingi

TableIV. Reconstitution of460C-treatedCytosolActivity by Recombinantp67-phox

Superoxide production

Reaction mixture

constituents Quantity Expt.I Expt.2 Expt.3

ng percentcontrol activity

Heatedcytosol 5.1 7.3 0.7

Heated cytosol 200 10.4 9.2 0.7

plusrp47phox 1 400 15.8 6.9 0.6

pur4ox1000 21.9 8.2 0.4

Heated cytosol 200 78.1 96.6 42.9 400 87.9 116.9 44.1 plusrp67-phox}

1000 95.6 121.5 70.3

NADPHoxidase was activated in thecell-freesystemasdescribed in Methodsusing reaction mixtures (0.1 mlin totalvolume) which contained deoxycholate-solubilizedmembranes(5x I05cell

equiva-lents), normalcytosol heatedat460Cfor 7.5 min(106cell equiva-lents), and recombinantcytosolfactors(rp47-phoxorrp67-phox)in theamountsindicated in the table. Resultsfromthree separate

ex-perimentsareshown,eachperformedwithadifferentpreparationof cytosol.Superoxide productionisexpressedasthe percentage of controlactivity,which is definedasthat measuredin the presenceof

106cellequivalents ofunheatedcytosol. * Themeancontrolactivity

forthe threeexperimentswas29.2±4.3 nmol0/min/0I cell

equiv-alentsmembranes.

amodel system for the oxidase defectinthis type of CGDin a

manner analogous to the 560C inactivation ofthe complement systemin serum.

Thebehavior ofheatedneutrophilsin cellfunction studies isconsistent with the selective inactivation ofanoxidase

com-ponent. The lossofrespiratoryburstfunction in heatedcells

parallels the observed defect in thekillingof S. aureus,a

corre-lation to be expectedgiventhe central roleof the respiratory

burstinthemicrobicidalactivityofphagocytes. Interestingly,

heated neutrophils also demonstrated a reduced capacity to

degranulate. Although this effect may beindependent ofthe

lossoftheneutrophil'srespiratoryburstactivity,the

finding

of a similar 50% reduction in degranulation in two _patients

with variantX-linkedCGD suggests there may bealinkage of

someneutrophil functionstotherespiratoryburstoxidase(or its components). Although the heat treatment of neutrophils (orneutrophil cytosol)representsa useful model for the oxi-dasedefect in thep67-phox-deficientform ofCGD,it should notbe construed as astrictparadigm forthisdisease.

Thefindingthatp67-phox functionis lost by mildheating

in vitro raises the question ofwhether this critical oxidase com-ponentmightbesimilarly affected during periods of high fever

invivo. Even though the present studies do not directly address thisissue,wehave observed thatnormalcytosolcan beheated

at 41°C for upto 30 min with no demonstrable decrease in

activityin the cell-free oxidase activation system. Based on this

findingit does not appear that the thermolability of p67-phox

seen inourpreparations at 46°C has physiological relevance. What effect extended incubations of several hours and longer under fever conditions might have upon neutrophil oxidase function remainsto beinvestigated.

Elucidation ofthe mechanism bywhichp67-phoxis dis-abledbyheattreatmentand how this alteration then interferes withoxidaseactivation willrequire furtherstudies.

Proteolytic

degradation ofp67-phoxdoesnotappeartoberesponsiblein

lightof thefindingthat heat-inactivated p67-phox(inboth

re-combinant and nativeforms)is stillrecognizable by

immuno-blottingand hasan unalteredapparent molecular weight. In thisrespect, heatedcytosol differsmarkedlyfrom

p67-phox-deficient CGDcytosol,which demonstratesacompletelack of

p67-phoxby immunoblotting.Becauseheatingthepure

recom-binantp67-phox byitself results in defective oxidasefunction,

a heat-induced conformational change in p67-phox may be involved.

The baculovirus-derived recombinant NADPH oxidase

components,rp47-phoxand rp67-phox, have provideda

power-fulmeans toinvestigatedirectly defects in the respiratory burst NADPH oxidase. We have shown that the inactivation of NADPHoxidase by controlled heating doesnotappear tobe duetothe action ofheat shock proteins,ashad beenpreviously

suggested (45). Instead, loss of oxidase activity canbe attrib-utedtothethermolabilityofp67-phox alone,since

recombi-nantp67-phox demonstrated thesamesensitivitytoheatthat

was observed with whole neutrophil cytosol. These

experi-mentshave providedadirect and independentmeansoftesting

the kinetically derived model proposing thermal damagetoa

single soluble oxidasecomponent.In that heating results in the selective inactivation ofp67-phox, and in that the oxidase activ-ity of heated cytosolcanbecompletely reconstituted by

recom-binant p67-phox, this system should be useful in exploring functional relationships involving genetically engineered

struc-1.6 t)

4-W

a

1.2

N

0

E E 0.8

.0

0

0o0.4

0

0.0

6.4 6.6 6.8

log (Cytosol Concentration)

7.0

110

00%

.05 0*

03

-2

2

0

QR 0

n.

Lo

0 1 2 3 4 5 6 7 8 9 10

Time at Incubation Temperature

(minutes)

Figure 6. Effect of heatingonactivated oxidase obtained from stimu-latedneutrophils. Preactivated NADPH oxidasewaspreparedas

noted underMethods, and subjectedtoheattreatment at37(v) and 460C(v). Asacontrol,aportion of the oxidase preparationwaskept

onice(v). Aliquotswereremovedto wetice after theindicated times ofheating and promptly assayed for activity. Reactionswereinitiated

with the addition of 0.16 mM NADPH, and the superoxide dismu-tase-inhibitable reduction ofcytochromec wasmeasuredusinga

dual-beamspectrophotometer. Themeancontrolactivityforthe threeexperiments performedwas22.0±3.2 nmol0/minperI07 cell equivalents.

tural variants ofp67-phox. Furthermore,thefindingthat rp67-phoxrestores normalkinetic behaviortobothheatedcytosol

andp67-phox-deficientCGDcytosolindicates that thistypeof

reconstitution system maybe useful forstudying the intricate

kinetic details ofthe mechanismbywhichtheNADPHoxidase

becomes activated.

NADPHoxidaseinits activatedstateshowsastriking

sensi-tivityto mildheating,both at370C(Fig. 6 and reference46) and (especially)at460C. The mechanismresponsiblefor this

inactivation, however,isnotclear. The failure ofrp67-phoxto restoreactivity doesnotruleoutthepossibilitythatp67-phox thermolabilityis stillacausalorcontributoryfactor. For

exam-ple, rp67-phoxmaynotbe abletoexchangewithheat-damaged p67-phox inthe activated oxidase. Alternatively,the

quater-nary structureof this form of the oxidasemaybedisruptedat

46°Cand contributetothe loss ofenzymeactivityinaddition to any specific effects on p67-phox. Additional experiments

aimed atdeducingthe mechanismbywhichthepreactivated

oxidase isdamaged by 46°C heatingcouldprovide insightsinto the functional roles of the various oxidase componentsinthe activeenzyme.Inabroadersense,the heat inactivation ofboth

normalcytosolandpreactivatedoxidaseprovidesapotentially valuable model system for studying the role that p67-phox playsboth inactivatingNADPHoxidase in normalphagocytes

and in causingthe respiratory burst defect in CGDpatients lackingthisprotein.

The authorsacknowledgeJames 0.Tolleyforhis skill inperforming Westernblotsdescribedinthismanuscript,Dr.IyadAlosachie for

pro-vidingtherabbit antiserumcontainingthep67-phox anti-peptide

anti-body, and Tercia L. Ku-Balai for herexpertise in carryingoutthe neutrophilfunctionalstudies.

This workwassupportedin partbygrantsfrom the National

Insti-tutesof Health(AI-24838, RR-00833, AR-10493,andAR-07107) and

from the ArthritisFoundation(Dr. Malawista).Dr.Curnuttewas an

EstablishedInvestigatorof theAmerican HeartAssociationatthe time

this workwasperformed.

References

1.Keller,H.U., andM.Bessis. 1975. Chemotaxisand phagocytosis in anu-cleatedcytoplasmic fragments ofhumanperipheral blood leukocytes.Nouv. Rev. Fr.Hematol. 15:439-446.

2.Malawista,S. E.,and A. deBoisfleury-Chevance. 1982.Thecytokineplast:

purified, stable,and functionalmotilemachinery fromhuman blood

polymor-phonuclear leukocytes.J.Cell Biol. 95:960-973.

3.Malawista,S.E., G.VanBlaricom,and S. B.Cretella.1985. Cytokineplasts from human bloodpolymorphonuclear leukocytes:lackof oxidaseactivity and extendedfunctionallongevity. Inflammation.9:99-106.

4.Dyett,D.E., S.E.Malawista,G. VanBlaricom,D. A.Melnick,and H. L. Malech. 1985. Functionalintegrityof cytokineplasts: specific chemotacticand

cappingresponses. J. Immunol. 135:2090-2094.

5. Dyett,D.E., S.E.Malawista,P. H.Naccache,and R. I.Sha'afi. 1986.

Stimulatedcytokineplastsfrom humanpolymorphonuclear leukocytes mobilize calcium andpolymerizeactin: cytoplastsmadeincytochalasinBretainadefect in actinpolymerization.J.Clin.Invest.77:34-37.

6.Malawista,S.E.,andG. VanBlaricom. 1986. Phagocytic capacity of

cyto-kineplasts fromhumanblood polymorphonuclear leukocytes. Blood Cells(Berl).

12:167-173.

7.Malawista,S. E., and G.VanBlaricom. 1987. Cytoplasts made from

hu-manbloodpolymorphonuclearleukocytes withorwithout heat: preservation of bothmotile functionandrespiratoryburst oxidaseactivity.Proc.Nat!.Acad. Sci. USA. 84:454-458.

8.Cross,A.R.,and0. T. G. Jones. 1991.Enzymic mechanisms ofsuperoxide

production.Biochim.Biophys.Acta. 1057:281-298.

9.Segal,A.W.1989.The electron transportchainofthemicrobicidaloxidase ofphagocyticcellsandits involvementin themolecular pathology of chronic granulomatous disease.J.Clin.Invest.83:1785-1793.

10.Babior,B.M.,and R. C.Woodman. 1990.Chronic granulomatous

dis-ease.Semin.Hematol. 27:247-259.

11.Smith,R.M.,and J. T.Curnutte. 1991.Molecularbasisofchronic

granu-lomatousdisease.Blood. 77:673-686.

12.Clark,R.A.,H. L.Malech,J. I.Gallin,H.Nunoi,B. D.Volpp, D. W.

Pearson,W. M.Nauseef,and J. T.Curnutte. 1989.Genetic variants ofchronic

granulomatous disease: prevalence ofdeficiencies oftwocytosoliccomponentsof the NADPHoxidasesystem. N.Engl.J.Med. 321:647-652.

13.Dinauer,M.C.,S.H.Orkin,R.Brown,A.J.Jesaitis, and C.A.Parkos.

1987.Theglycoproteinencodedby the X-linked chronic granulomatous disease

locusisacomponentoftheneutrophilcytochrome bcomplex. Nature(Lond.).

327:717-720.

14.Teahan, C.,P.Rowe,P.Parker,N. Totty,andA.W.Segal. 1987. The

X-linked chronicgranulomatous diseasegene codes for the B-chain of cy-tochrome b-245. Nature(Lond.).327:720-721.

15.Parkos, C. A., R.C. Allen, C.G.Cochrane,and A. J.Jesaitis. 1987.

Purifiedcytochromebfromhumangranulocyteplasma membrane is comprised oftwopolypeptideswith relative molecularweightsof91,000and22,000.J.Clin.

Invest.80:732-742.

16.Segal,A.W. 1987. Absenceofbothcytochromeb-245subunitsin

X-linked chronicgranulomatousdisease.Nature(Lond.).326:88-91.

17.Parkos,C. A.,M.C.Dinauer,A. J.Jesaitis,S.H.Orkin,andJ. T.

Cur-nutte.1989. Absenceof both the 9lkD and 22kDsubunits ofhumanneutrophil cytochromeb intwogeneticforms of chronicgranulomatousdisease.Blood.

73:1416-1420.

18.Dinauer,M.C.,E.A.Pierce,G.A.P.Bruns,J. T.Curnutte,and S. H. Orkin. 1990. Humanneutrophil cytochromeblightchain(p22-phox):gene

struc-ture, chromosomallocalizationandmutationsincytochrome-negative

autoso-malrecessive chronicgranulomatousdisease.J.Clin.Invest.86:1729-1737. 19.Hurst,J.K.,T. M.Loehr,J. T.Curnutte,and H. Rosen. 1991. Resonance Raman and electronparamagneticresonancestructuralinvestigations of neutro-philcytochromeb55g.J.Biol. Chem.266:1627-1634.

20.Okamura,N.,S. E.Malawista,R. L.Roberts,H.Rosen,H. D.Ochs,B. M.

Babior,and J. T.Curnutte.1988.Phosphorylation oftheoxidase-related48kD

phosphoprotein familyin the unusualcytochromenegativeandX-linked

cy-tochrome-positivetypes of chronicgranulomatous disease. Blood. 72:811-816.

21.Dinauer,M.C., J. T.Curnutte,H.Rosen, and S. H. Orkin. 1989.A

missense mutation in theneutrophil cytochromebheavychainin cytochrome-positiveX-linked chronicgranulomatousdisease. J.Clin.Invest. 84:2012-2016.

'I--...

\

_i

*-i~~~~~~~~~~i~~

.

22. Dinauer, M.C.,E.Pierce,R.W.Erickson, T. Muhlebach, H. Messner, S. H.Orkin,R.Seger, and J. T. Curnutte. 1991. Point mutation in the cytoplasmic

domain oftheneutrophil p22-phoxcytochrome bsubunit is associated with a nonfunctional NADPH oxidase and chronic granulomatous disease. Proc.Natl.

Acad. Sci. USA. 88:11231-11235.

23. Segal, A. W., P. G. Heyworth, S. Cockcroft, and M. M. Barrowman. 1985.

Stimulatedneutrophils from patientswith autosomalrecessiveCGD fail to phos-phorylate aMr44000protein. Nature (Lond.). 316:547-549.

24.Curnutte, J. T., R. L.Berkow, R. L. Roberts, S. B.Shurin,and P. J.Scott. 1988.Chronicgranulomatous diseasedue to adefectin thecytosolic factor

re-quired for nicotinamideadenine dinucleotidephosphateoxidase activation.J. Clin. Invest.81:606-610.

25. Volpp, B. D., W. M.Nauseef, and R.A.Clark. 1988. Twocytosolic neutrophil oxidasecomponentsabsentinautosomalchronic granulomatous

dis-ease.Science(Wash. DC).242:1295-1297.

26.Nunoi,H., D.Rotrosen, J. I.Gallin,and H. L. Malech. 1988. Two forms

ofautosomalCGDlackdistinctneutrophil cytosolicfactors. Science(Wash. DC).

242: 1298-1301.

27. Lomax, K.J., T. L. Leto, H. Nunoi, J. I.Gallin,and H. L.Malech.1989. Recombinant 47-kilodalton cytosol factor restores NADPHoxidaseinchronic granulomatousdisease. Science(Wash.DC). 245:409-412.

28. Volpp, B. D., W. M.Nauseef,J. E.Donelson, D. R. Moser, and R.A.

Clark. 1989.Cloning ofthecDNAandfunctionalexpression ofthe47-kilodalton cytosoliccomponentofhumanneutrophil respiratoryburst oxidase. Proc. Natl.

Acad.Sci. USA. 86:7195-7199.

29. Leto, T. L., K. J. Lomax, B. D. Volpp, H.Nunoi,J. M.G.Sechler,W. M.

Nauseef,R. A.Clark,J.I.Gallin,and H. L.Malech. 1990.Cloning ofa67-kD

neutrophil oxidase factor with similarityto a noncatalytic region of p60.

Science(Wash.DC). 248:727-730.

30.Franke,U., C.Hsieh,B. E.Foellmer, K. J. Lomax, H. L. Malech, and T. L. Leto. 1990. Genesfortwoautosomal recessive forms of chronic granulomatous diseaseassigned to lq25 (NCF2) and 7ql 1.23 (NCF1). Am. J. Hum. Genet. 47:483-492.

31. Casimir, C. M., H. N. Bu-Ghanim, A. R. F. Rodaway, D. L. Bentley, P. Rowe, and A. W. Segal. 1991. Autosomal recessive chronicgranulomatous

dis-easecaused by deletion at a dinucleotide repeat. Proc.Nat!.Acad. Sci. USA. 88:2753-2757.

32.Curnutte,J. T., R. Kuver, and B. M. Babior. 1987. Activation of the

respiratoryburstoxidase in a fully soluble system from human neutrophils.J. Biol. Chem.262:6450-6452.

33. Curnutte,J. T., R. Kuver, and P. J. Scott. 1987. Activation of the

neutro-phil NADPH oxidase in a cell-free system: partialpurificationofcomponents and

characterizationof theactivationprocess. J. Biol. Chem.262:5563-5569.

34. Curnutte, J. T., P. J. Scott, and B. M.Babior.1989.Functional defectin neutrophil cytosols from two patients with autosomal recessive cytochrome-posi-tive chronic granulomatous disease. J. Clin. Invest. 83:1236-1240.

35.Gottlieb, C., S. Lau, L. R. Wasserman, and V. Herbert. 1965.Rapid

charcoal assay for intrisic factor(IF),gastric juiceunsaturatedB12binding capac-ity,antibody to IF, and serum unsaturatedB12 binding capacity. Blood. 25:875-884.

36.Babior,B.M., and H. J. Cohen. 1981.Measurementofneutrophil

func-tion:phagocytosis, degranulation,therespiratory burst,andbacterialkilling.In

Leukocyte Function.M.J.Kline, editor. Churchill Livingstone,NewYork. 1-38. 37. TheInternational ChronicGranulomatousDiseaseCooperativeStudy Group. 1991. A controlledtrial of interferongammatopreventinfectionin

chronicgranulomatous disease. N. Engl. J.Med.324:509-516.

38.Babior,B.M., R. Kuver, and J. T. Curnutte. 1988. Kinetics of activation

oftherespiratory burst oxidaseinafully solublesystem from humanneutrophils.

J. Biol.Chem.263:1713-1718.

39. Mayo, L. A., and J. T. Curnutte. 1990. Kineticmicroplateassayfor

super-oxideproductionbyneutrophilsandotherphagocyticcells.MethodsEnzymol.

186:567-575.

40.Curnutte,J.T., P. J.Scott,and L. A.Mayo.1989.Cytosoliccomponents

oftherespiratory burst oxidase: resolution of fourcomponents, twoof whichare

missingin complementary typesofchronicgranulomatousdisease. Proc.Nat!.

Acad.Sci. USA. 86:825-829.

41. Okamura, N., B. M.Babior, L. A. Mayo, P. Peveri, R. M. Smith, and J. T. Curnutte. 1990. Thep67-phoxcytosolic peptide oftherespiratoryburstoxidase fromhumanneutrophils: functionalaspects. J.Clin.Invest.85:1583-1587.

42.Leto, T. L., M. C. Garrett, H.Fujii,and H.Nunoi.1991.Characterization of neutrophilNADPHoxidase factors p47-phoxandp67-phoxfrom recombi-nantbaculoviruses.J. Biol.Chem.266:19812-19818.

43.Laemmli,U. K.1970.Cleavageof structuralproteins duringtheassembly

ofthe headof bacteriophageT4.Nature(LondJ.227:680-685.

44. Harlow, E., and D. Lane. 1988. Antibodies: A Laboratory Manual. Cold

SpringHarborLaboratory, ColdSpringHarbor, NY. 471-510.

45.Maridonneau-Parini,I., J. Clerc, and B. S. Polla. 1988. Heat shock in-hibits NADPH oxidase in humanneutrophils. Biochem.Biophys.Res.Commun. 154:179-186.

46.Light,D.R., C.Walsh, A. M. O'Callaghan, E. J. Goetzl, and A.I.Tauber. 1981. Characteristicsofthecofactor requirements forthesuperoxide-generating

NADPH oxidase of human polymorphonuclear leukocytes. Biochemistry.