the lung: Inhibition of its synthesis leads
body and mitochondrial defects
MARTENSSON*, AJEY JAINt, WILLIAM FRAYERt,AND
*Departments of Biochemistry andtPediatrics,CornellUniversity Medical College, 1300 York Avenue, New York, NY 10021
ABSTRACT Mice treated with buthioninesulfoximine, an inhibitor ofglutathione synthesis, showedstriking alterations ofmorphology of lungtype2celllameilar bodies (swelling and disintegration) and mitochondria (degeneration) and of lung capillary endothelial cells (mitochondrial swelling). These ef-fects probably may be ascribed toglutathione deficiency; ad-ministration of glutathione monoester protects
againstthem. Measurements of arteriovenous plasma glutathione levels across thelung indicate that thenet uptake ofglutathione by thisorgan is substantial. Thus,glutathione exported from the liver to theblood plasma is utilized by the lung which, like liver, kidney, andlymphocytes (and unlike skeletal muscle), exhibits a high overall rate ofglutathione turnover. Intraperitoneal Hijection of glutathione into buthionine sulfo treated mice leads to very high levels of plasma glutathione without significant increase in theglutathiolelevelsofliver,lung, and lymphocytes; on the other hand, admirationof glutathione monoester leads to markedlyincreased tissue and mitochon-drial levels ofglutathione. Admirationofglutathione mo-noester (in contrast to glutathione) to control mice also in-creasesmitochondrialglutathione levels. Thefindings indicate that glutathione is required for mitochondrial integrity and that itprobably also functionsinthe processing and storage of surfactant in lamellar bodies. The morphological changes observed aftertreatmentwithbuthioninesulfoximine and their prevention by glutathione monoester as well as finding on glutathione metabolism indicate that this tripeptide plays an important role in the lung. Thepreviously observed failure of buthioninesulfoximine-treated mice to gain weightismainly due to glutathione deficiencyintheintestinal mucosa.
Previousstudies (1) on the effectsof inhibition of glutathione
(GSH) synthesisonskeletalmuscleprovided evidencethat
GSH functions in muscle under normalconditions, (it) the levels of GSH in muscle are greatly in excess of those
required, and (iib) GSH isrequired for mitochondrial
integ-rity.It wasalso found thatadministration of GSHmonoester
(incontrast tothatofGSH) led to maintenance of a significant level of GSH inmusclemitochondria duringarelatively long
periodof administration ofbuthioninesulfoximine(BSO),an
Inthepresentworkwehave extended thisapproachtothe
lung,which likekidney and liver (and in contrast to skeletal muscle) exhibits ahighrate of turnoverof GSH. Wefound evidenceofsignificantcellular damage to lung and to a lesser extent to lymphocytes after prolonged inhibition of GSH
synthesis produced byadministrationofBSO. Interestingly,
wefound striking evidence of lamellar body damage in type 2cells anddegenerationofmitochondriain these cells and in
capillaryendothelial cells. Thesechangeswerepreventedby administration ofGSH monoester. Administration of GSH
monoester prevented the marked decline in tissue GSH levels
found after administration of BSO and led to levels of
mitochondrialGSH in thelung and liver that were higher than thosefound in untreated controls.
Materials. Mice (male, 28-32 g, Swiss-Webster; Taconic
Farms) were maintained on Purina Chow. L-Buthionine-[S,R]-sulfoximine (BSO)(2-4), L-buthionine sulfone(4), and GSH monoisopropyl (glycyl) ester (5) wereprepared as de-scribed. Hypaque
(50%16)waspurchased fromWinthrop-Breon Labs (New York). An aqueous solution of GSH
monoisopro-pylesterwaspreparedin the 1/2(H2SO4) form andadjustedto
pH6.5-6.8 by cautiousaddition of NaOHimmediatelyprior to use. Instudies inwhich this compound wasgiven,control experiments were carried out in which mice were given equimolaramountsofGSH,isopropanol, andNa2SO4.
Methods. BSO wasadministered tomiceas stated below
(see Figs. 1 and 2 and Table 1). L-Buthionine sulfone was given to mice in equimolar amounts in parallelstudies.The tissues were obtained as follows. Mice wereanesthetized by intraperitoneal injection of a mixture of xylazine (5 or 10
mg/kgof body weight) and ketamine (100mg/kg);within 3
min, the thoracicandperitoneal cavities wereopened. The lung andliver were perfused from
theright ventricle with 10 ml of cold saline after
clampingof the venous heart input. Perfusion was completed within 2 min as the lung turned
white and theliver became light brown. Theleft inferiorlobe
ofthe lung was excised, rinsed with cold saline, blotted, weighed,andhomogenized in 5 vol of5%sulfosalicylicacid. Apieceof the left lobe ofthe liver wasexcisedandprocessed
as described for the lung. The tissue homogenates were
centrifuged for5minat10,000 x gat
40C,and the supernatant solutions wereanalyzedfor GSH.
Lymphocytes were obtained from whole blood (0.7 ml)
drawn from the right ventricle into a heparinized syringe containing0.7,umol ofEDTAand 14 amoleach of L-serine and sodium borate (6). Samples obtained from two mice were
with2.6mlof coldsaline and then
gradientconsistedoftwophases:anupperphasecontaining3 mlofamixtureof10partsof33.9%Hypaque and 24 parts of
10 parts of
50%oHypaque and 30 parts of
lymphocyteswereseparated fromgranulocytes by differential centrifugation; theywerecentrifugedat300x gfor 30minat
40Crather than 800x gasrecommended(7)so as todecrease
contamination by platelets. The lymphocyte
suspensions(which contained about
10%monocytes) were washed by centrifugationwith cold salinecontaining1mMEDTA,20 mM
L-serine,and 20 mMsodium borate. Contaminating
erythro-cytes were lysed by suspension of the cells in 0.15 mM ammonium chlorideat
370Cfor 5 min(6,8).Thelymphocytes
Abbreviations:GSH, glutathione; BSO, buthionine sulfoximine.
5296 Thepublicationcostsof this article weredefrayedin partbypagecharge payment.Thisarticlemustthereforebehereby marked "advertisement" inaccordance with 18U.S.C.§1734 solelytoindicatethisfact.
Proc. Natl.Acad. Sci. USA86 (1989) 5297 werewashed twice with saline (containing EDTA, serine, and
sodiumborate),counted, and lysed bytwocycles of
thawing in 0.1-0.2 mlof 4.3%sulfosalicylic acidasdescribed (6, 9). After centrifugation, the supernatant solutions were immediatelyanalyzedfor total GSH.
Determinations of total GSH were also carried out on plasma obtained by immediate centrifugation of 0.2 ml of
wholeblood[from the right ventricle (see above) or by direct aspiration (0.05 ml) from the left ventricle] at10,000 x gat
40C for 1 min (10, 11). A portion of the clear supernatant
solutionwasimmediatelymixedwithsulfosalicylicacid(final concentration, 4.3%) and then centrifuged for 5 minat 40C. The supernatant solution obtainedwas usedfor determina-tion of total GSH. The presence of hemoglobin in thesamples
was examined (12, 13). The minute amount ofhemoglobin
found indicated that the presence of erythrocytes could accountfor <0.3%of the observed GSH content(14).
GSH determinations carried out as described above on
control samples oflung,liver,andlymphocyteswere,within
experimental error, identical to values obtained on mice
killed byspinal cord transection.Theanesthesiaprocedureis advantageous in that significantly larger amounts of blood maybeobtainedthanareobtained afterspinalcord transec-tion. However, the perfusion was completed within 5 min; afterlongerperiods ofanesthesia,evenwithone-thirdof the doses statedabove, lower GSH levelswerefound.
Total GSH was determined by the glutathione disulfide reductase-5,5-dithiobis(2-nitrobenzoate)
(11). For plasma, values that include mixed disulfides be-tween GSH and cysteine or proteins were alsodetermined
afterreduction with5 mMdithiothreitol(9). BSOwas
deter-minedbyuseofaDurrum(model 500) aminoacidanalyzer.
The activities of -glutamylcysteine synthetase, y-glutamyl transpeptidase,andcitratesynthase (15,16)weredetermined
as stated (1). Protein was determined (17), and electron
microscopy(1)was performed as previouslystated. The mitochondrial fractionsofliver and lungwere sepa-ratedbytheprocedurepreviously given (1, 16, 18, 19)except that heparin was omitted in the homogenization step. The
fractions obtained by thisprocedurewerechecked by elec-tronmicroscopy. The fractions from livercontainedvirtually uncontaminated mitochondria; those from lung contained predominantly mitochondriaand about 5% lamellarbodies.
Effect ofAdministration of BSOonGSH Levels.
Adminis-tration of BSO led torapid decline ofthelevels of GSH in the lung, lymphocytes (Fig. 1), liver, and plasma (Fig. 2). The initialrapid rates ofdecline,expressedasthetimerequiredfor a50% decrease in the level, were about 25 and 45 min for lymphocytes andlung,respectively.Thedeclineintheplasma
GSHlevel,asexpected (20, 21),wasclosely
(ti,2= about 65min). Thedeclineof the GSH levels in the liver wasbiphasic;thesecond slower rate,asdiscussed
previously (1, 18), isthoughttoreflect loss of GSH from the
mitochondria. The slower laterratesof GSHdisappearance
were0.01%, 0.02%, 0.05%,and0.05% ofthecorresponding
initialratesforlymphocytes, lung, liver, andplasma,
respec-tively.Thedisappearance of GSHfrom
exhibited bothrapidand slowphasesbutseemedtobe more
complexthanin liver. The dataobtainedonGSH levels in the lungareprobablyafunction of severalfactors, includingthe cellularheterogeneity ofthis organ, theratesof cellularexport ofGSH, and the intracellular levels of BSO. Although
rela-tively high overall levels ofBSO were maintained in these
experiments,theintracellularlevelofBSO in the variouslung
celltypesisnotknown.BSOwasrapidly cleared fromplasma
theurine. Preliminarystudies inwhich single doses of BSO
were administeredindicate that the time requiredfora
loo -J 0 z 0 (-IL 0 0I~ (I)
6~~~~~~~~~~~~0 100 200 300 400 500 HOURS
FIG. 1. Effect oftreatmentwith BSOontheGSHleveloflungand lymphocytes. Mice were given BSO twicedaily by
intraperitonealinjection(4mmol/kg)at9a.m.andat6p.m.for21days.BSOwasalso givenin thedrinkingwater(20mM).Controlswereinjectedwithan
equivalentvolumeof saline andweregiventapwater todrink. Control valuesof GSHwere1.69±0.12,umol/g(lung)and 1.39 + 0.15nmol per106cells
(lymphocytes).After0.25hr, 0.5 hr, 1 hr, 2 hr, 1
day,2 days,4days,7days,15days,and 21daysof BSO treatment, the GSH values(mean ±SD;n=3-5)forlungwere,respectively,1.31±0.16,
0..91±0.10,0.54 ±0.06,0.47±0.05,0.43 ± 0.05, 0.42 ± 0.04, 0.39 ± 0.04, and 0.38 ± 0.04
lymphocyteswere0.97 ±0.11,0.51±0.08, 0.42±0.05,0.36±0.05,0.27+ 0.04,0.25 ±0.04,0.24± 0.04,0.18 ±0.02,0.16±0.02,and 0.14±0.03nmol per106cells.
(e)GSH levels ofasingle
decrease in tissue BSO level is about
50,and 120 min for
rapidly,and the levels of GSH increased
GSH return werelower than the initialrates of GSH
disap-pearance, and theGSH values"overshot"thecontrolvalues asobserved in the heart
rapidrates at which GSH levels increased after
discontinuation of BSOtreatment
precludedstudies on the effects of amino acid
aswasdone in studiesonmuscle
(1). However,studiesonfed and fastedmice showed thatadministrationeitherof
the GSH levelsof
lymphocytestoabout the same extent. Administration of
a-keto-glutarate)alone alsoincreased liver and
lungGSH levels in fed mice.Administration of
glutaminedidnotincrease GSH levels in these tissues orin
respectively;the valuefor liver is 82.3 + 1.5
100 -J 0 z 0 LL 0 0R I-n,. 80 60 40 20 E (n
ad1.0 0.5 0 100 200 300 400 500 HOURS
FIG. 2. Effect ofasingledose(Inset)ormultipledoses of BSO at9 a.m.and 6 p.m.(each dose,2mmol/kg)onGSHlevelsofliver (A)andplasma (A). ControlvaluesofGSHwere8.26±0.06
(liver)and 58.3±4.8AM(venousplasmafromrightventricle).After 0.25hr,0.5hr,1hr,2hr.1day,4days (only liver),9days,15days, and 21daysofBSO treatment, the GSH values(mean±SD;n=3-5) for liverwere,respectively, 6.61 ±0.28,5.78 ± 0.25,4.54 ± 0.21, 2.48±0.09,1.10±0.08,0.83±0.05,0.66±0.04,0.59±0.03,and 0.52±0.02
50.7 ±2.9,46.6± 3.1,33.2 ± 3.2,12.9±1.5, 3.50± 0.50,2.92±
AsM.AplasmaBSO value ofcloseto
zero wasfoundat2 hr.
y-rglutamyltranspeptidase activities of lymphocytes, lung,kidney, andliver are,respectively,41.2 ± 1.2(mean ±
SD;n = 5), 25.0 + 0.3 26,300 ± 3700(22), and1.92 ±0.05
nmol/hrper mgofprotein (18); seealso refs. 20and 23-27. Utilization of GSH by the Lung. Previous studies on
ne-phrectomized mice led to the conclusion that about two-thirdsofthearterialbloodplasmaGSHisused
and that about one-third of the plasma GSH is used by extrarenal
-t-glutamyltranspeptidase activity (21). On the basis of studiesontheisolatedperfusedratlung, Berggrenet al.(24) concludedthat thelungin additiontothekidneymay utilizeplasmaGSH and thatutilization of GSHisprobably mediatedby extracellular breakdown andresynthesisrather than by direct
studies,which lead to similarconclusions,make itpossibletoconsidera quantita-tive estimate of the amount of GSH used by the lung; interestingly, thelungsuse moreGSHthando thekidneys. Plasma from blood drawn fromtheright ventriclehasGSH
,M(Fig. 1),and levels that include GSH
mixed disulfidesof 62.1 ± 5.0
AM.Incontrast, GSH levels
ofplasmataken from theleft ventriclewerefoundtobe 22.1 ± 2.3
pAM,and levels thatinclude GSH mixed disulfides of 29.9 ± 7.5
nmol/ml,whenmultiplied bythecardiacoutput[=6.3ml of
(28)],indicates that the loss of GSH from plasma during transit through the lungisabout 200 nmolper min. This value is greater than has been estimated for
Morphological Changes Associated with Administration of
microscopicstudies ofthe liver and
after 3 weeks oftreatmentwith BSO showednoabnormalities.
lung capillaryendothelial cells
(Fig. 3).Intype 2cellsthere was a
significantdecreaseinthenumberof mitochondria and
degeneration.After treatmentwithBSO the lamellar bodiesseemedtodominate thetype 2
cells;thelamellar bodies became
appearedto occupy more space than in the controls.
Swellingof thelamellar bodieswasassociatedwith
organelles.There was a decrease in the numberofmicrovilli on the alveolar
surfaceofthe type 2
Bluntingof themicrovilli of the
morphologywere not seenin
givenbuthionine sulfonenor were
treated with BSO and GSHmonoester.
mice treated with BSO and GSH. Determinations of citrate
synthase activitywerecarriedout onthemitochondrial
comparedwith values of0.094 + 0.007 for controls. Mice treated with BSOfor3weeks and then allowedto recoverfor 8 weeks didnot show the
synthase activityof themitochondrial frac-tion of
lungwasaboutthesame asthatof untreatedcontrols.
Effect ofAdministrationof GSH Monoester and of GSHon GSHLevels.When miceweretreated with BSOfor 9
there was as
expecteda marked decline in the levels of cellular GSHandof
experiment 1).When GSH monoester was
experiment 3),theliver GSHlevelswere about thesame as
controls,andthe GSHlevel oflivermitochondriawassomewhat greaterthan thelevelsin controls. Administration of GSH monoester led to
higherlevelsofGSH in the
mitochondrial fraction obtained from
lunghad GSH levels
higherthanthose in thecontrols
BSO and GSHmonoester were
higherthan thosefound for the BSO-treated mice
experiment 2).When GSHwasgiven
(about80 timesgreaterthan the control
values),but there was
error)oftheGSH levels of liver
lung,ofthemitochondrial fractions obtainedfrom these
tissues,and of the
slighteffect or no effect on liver and
findingswere made in the present studies on
lungGSH levels thatwere
only30% greater than those of controlsafter2 hrand wereabout thesame as the
controls after 4 hr. Administration of GSH monoester to untreated mice ledtolevels of livermitochondrial GSHthat were
equimolaramountsof GSH didnotaffect livermitochondrial GSHlevels.
tThearteriovenous difference inplasmaGSH level acrosstherat
lungis smaller thanwasfound hereforthemousebecause there is
alowerplasmalevel of GSH in theratrightventriclethan in thatof themouse.EstimatesofGSHutilizationbyratlungand
basedon arteriovenous differences inplasmaGSH levels across
these organs andcardiacoutput(31)indicatethatkidneyand
Proc. Natl. Acad. Sci. USA86(1989) 5299 f TZ
..li tt- .*!
FIG. 3. Representativeelectronmicrographsof thelungsof control mice(aandb)andmice treated with BSO for 21days (candd).cand d show lamellarbody(lb)swellinganddisintegration (one arrow),mitochondrial (m) degeneration (2 arrows),endothelial cell(ec)mitochondrial swelling,andthickeningof the basal membrane(bm). [aandc=x5400(horizontalbar in the lowerrightcorner=0.60
/Am);b and d= x17,800
(horizontal bar in the lowerrightcorner=0.18Am).] DISCUSSION
Theproductionof cellular GSHdeficiency bytreatmentwith BSO hasadvantagesovervarious methodsofGSHdepletion
(compare ref.20). Thus, decreased GSH levelscanbe main-tained for relatively long periods, facilitating detection of structural and other abnormalities. Itseemssignificantthat the
effects of BSO on tissue and mitochondrial GSH levels,
skeletalmuscledegeneration (1),morphological changesin the
lung, anddepression ofweight gain (1, §)arevirtually
com-pletely preventedby giving GSHmonoesterbutnotby
GSH. Although BSOishighly effective as aninhibitor,
rela-tively highdosesmustbegiventomaintain effective cellular
levels,andBSOisonly slowlytransportedintosome organ-elles[e.g.,mitochondria(18)]andtissues[e.g.,brain(20,
led us to examine theeffect of BSO administration onintestinal GSH levels. After 14 days of treatment with BSO (oral and intraperitoneal),thelevels of GSH inmousejejunalmucosa,colon mucosa, and pancreas were about4% of the controls. Electron microscopy showed microvillus and mitochondrial degeneration andvacuolization of theepithelialcellsofjejunalmucosa.Previous studies showed thatratjejunal mucosalcells have high levels of GSH(33).Itisprobablethat thenormalabsorptivefunctions of the gutrequireGSH.
§Theeffect of BSO on weight gain notedpreviously (1)was
con-firmed in the presentstudies; this effect is largely prevented by administration ofGSHmonoesterbut isnotaffectedby adminis-tration of GSH. ThefindingthatmicegivenBSOdevelopdiarrhea
Table 1. Effect of administration of GSH monoester and of GSH on GSH levels GSH levels
Mitochondrial, Mitochondrial, Lymphocyte
Total, nmol/mgof Total, nmol/mg of total, nmol Plasma Exp. Treatment* ,Umol/g protein Jtmol/g protein per106cells total,
1 None 8.26 6.4 1.69 4.7 1.39 58.3
2 BSO 0.98 3.8 0.36 0.86 0.20 0.90
3 BSO + GSH monoester 8.29 10.8 0.71 10.0 0.45 43.0
4 BSO + GSH 1.10 3.9 0.42 0.87 0.24 5080t
*Mice were treated with saline (experiment 1) and with BSO (experiments 2-4) for 9 days. The BSO was given intraperitoneally(4mmol/kg)at10 a.m.and6 p.m. Inexperiment 3, GSHmonoisopropylester (5mmol/kg;1-1.3ml) was injectedintraperitoneallyas anisosmolarsolution (pH 6.5-6.8) twice daily (8 a.m. and 4 p.m.) for 9 days. In experiment 4, mice wereinjectedwithGSH,isopropanol,andNa2SO4in amountsequimolartotheester given inexperiment3.Mice (4-5 perexperiment) weresacrificedat 11 a.m. on day 9,and GSHdeterminations were done as described. Data are given asmeans;SDs were + 0.04-0.08fortissues and 0.2-480 for plasma.
tTheGSH levels in plasma (from rightventricle)were25,100 and 12,080
,.Mat 50min and 120 min, respectively, after GSH administration.
Themorphological changes observed here after treatment with BSO and their prevention by GSH monoester strongly suggestthat GSH isrequired forlung mitochondrialintegrity and for otheraspects of lung function. GSHprobably plays arole in theprocessingand storageoflung surfactant. The
highlevel of GSH in the alveolarepithelial liningfluid and related studies on this interesting finding (34) suggest that surfactant and GSH areofmajorimportancein the protection of thelungagainsttoxiccompoundsin theinhaledair.Export of GSHfrom such cellsasphagocytes,type 2cells,
lympho-cytes, and fibroblasts may contribute to the GSH of the alveolarliningfluid. The presentstudies revealanadditional aspectof theinterorgantransport of GSH in which plasma GSH(whosemajorsourceis theliver)isutilizedby the lung. Administrationof GSH ledtoextraordinarily highplasma GSHlevels(Table 1) but not to significantincreases in the GSH levels oflung,liver,orlymphocytes.Thesefindingsare in accord with the view that there is normally little or no transport of intact GSH into cells.AdministrationofGSHto control mice doesnotleadtoappreciableincreases in tissue GSH levels,andthe small increases thatareobserved most
likelymaybe ascribedtoextracellulardegradation,transport of theproducts,andintracellularsynthesis ofGSH(compare
ref. 20). Underphysiological conditions many cells export
GSH; the reverse of this process has not been observed. Theseconsiderations donotexcludethelikely possibilitythat administered GSH mightprotectagainst extracellularly ap-plied toxicity and that it can serve as a good source of cysteinefor GSHsynthesis.The substantialincrease in GSH levelsfound afteradministrationof GSHmonoesterconfirms
that thiscompound iswelltransportedand converted
intra-cellularlytoGSH(5, 32).Thehighlevelsof GSH foundinthe
mitochondrial fractions oflungandliver afteradministration ofGSH monoestersuggest thatthisagentmaybeusefulfor
protectionagainst various typesoftoxicity.
Wethank Dr. Donald A.Fischman,Mrs. LeeCohen-Gould,and Mr. James Dennis for valuable advice and help in the electron microscopystudies and Dr.Carl G. Becker forproviding equipment forleukocyte counting.This workwassupportedin partbyfunds from the National Institutes of Health(Grant2 R37DK-12034).J.M.,
on leave from the Department ofClinical Chemistry, University Hospital,Link6ping,Sweden, acknowledgesgenerous supportfrom the Swedish MedicalResearchCouncil (05644-06B),ToreNilsson Research Fund, Throne-Holst Foundation, Sweden-American Foundation,SwedishSocietyofMedicine,Medical Research Coun-cil of theSwedishLife InsuranceCompanies, S-O Liljedahl Foun-dation, Ollie and Elof Ericssons Foundation, andthe Albert and NannaSkantze Foundation.
1. Mfrtensson,J. &Meister, A. (1989) Proc.Natl.Acad. Sci. USA86, 471-475.
2. Griffith, 0. W., Anderson, M. E. & Meister, A. (1979) J. Biol. Chem. 254, 1205-1210.
3. Griffith, 0. W. & Meister, A. (1979) Proc. Natl. Acad. Sci. USA 76, 2715-2719.
4. Griffith, 0. W. (1982) J. Biol. Chem. 257, 13704-13712.
5. Anderson, M.E., Powrie, F., Puri, R. N. &Meister, A.(1985)
Arch.Biochem.Biophys. 239, 538-548.
6. MArtensson,J. (1986)Metabolism35,118-121.
7. English, D. & Andersen, B.R. (1974)J. Immunol. Methods 5, 249-252.
8. Boyum,A.(1968) Scand.J.Clin.Lab. Invest. 21,97-102.
9. Mirtensson, J. (1987)J.Chromatogr. 420, 152-157.
10. Anderson,M.E. &Meister,A.(1980)J.Biol. Chem.255,9530-9533. 11. Anderson, M. E. (1985) MethodsEnzymol.113,550-551.
12. Cross, C. E., Watanabe,T. T.,Hasegawa,G.K.,Goralnik,G.N., Roertgen,K.E.,Kaizu, T., Reiser,K.M.,Gorin,A. B.&Last, J.A.(1979) Toxicol.Appl.Pharmacol. 48,99-109.
13. Marklund, S. (1979)Clin. Chim.Acta92,229-234.
14. Smith,J. R.(1974)J.Lab.Clin. Med.83, 444-450. 15. Srere,P. A. (1969)MethodsEnzymol. 13, 3-11.
16. Robinson, J.B., Jr, & Srere, P. A. (1985)J. Biol. Chem. 260,
17. Bradford, M.M.(1976) Anal. Biochem.72,248-254.
18. Griffith,0.W.& Meister,A.(1985) Proc.Natl. Acad.Sci. USA82,
19. Nedergaard, J. & Cannon,B.(1979)MethodsEnzymol.55, 3-28.
20. Meister,A.& Anderson, M. E. (1983)Annu. Rev.Biochem. 52, 711-760.
21. Griffith, 0.W.&Meister,A.(1979)Proc.Nat!.Acad.Sci.USA 76, 5606-5610.
22. Orlowski,M.&Wilk,S.(1975)Eur.J. Biochem.53, 581-590. 23. Tate, S. S.&Meister,A.(1985)MethodsEnzymol. 113,400-404. 24. Berggren,M.,Dawson,J.& Moldeus,P.(1984) FEBS Lett. 176,
25. Dethmers,J. K.&Meister,A.(1981)Proc.Nat!. Acad.Sci.USA 78, 7792-7796.
26. Novogrodsky, A., Tate, S. S. & Meister, A. (1976)Proc. Nat!.
Acad. Sci. USA 73, 2414-2418.
27. Dawson, J.R., Vahikangas, K., Jernstr6m, B. & Moldeus, P.
28. Foster,H.L.,Small, J. D.&Fox,J.G., eds. (1983) TheMouse in BiomedicalResearch, NormativeBiology, Immunology,and
Hus-bandry (Academic,NewYork), Vol.III.
29. Meister,A.(1983)in FunctionsofGlutathione-Biochemical, Phys-iologicalandToxicologicalAspects, eds. Larsson, A.,Orrenius,S., Holmgren,A.&Mannervik,B.(Raven,NewYork),pp. 1-22.
30. Anderson, M.E., Bridges, R.J. & Meister, A. (1980) Biochem. Biophys.Res.Commun.96,848-853.
31. Altman, P. L. & Dittmer,D.S., eds. (1974) BiologyDataBook (Fed.Am.Soc. Exp.Biol., Bethesda, MD), 2ndEd., Vol.3. 32. Puri,R.N. &Meister,A.(1983)Proc.Nat!.Acad. Sci. USA80,
33. Cornell,J.S. &Meister,A.(1976)Proc.Nat!.Acad.Sci.USA73, 420-422.
34. Cantin, A.M., North, S.L., Hubbard, R.L. & Crystal, R.G. (1987)J.Appl.Physiol. 63,152-157.