Mechanisms of methylene blue stimulation ofthe hexose monophosphate shunt inerythrocytes.

E N Metz, … , P Balcerzak, A L Sagone Jr

J Clin Invest. 1976;58(4):797-802. https://doi.org/10.1172/JCI108531.

The response of the hexose monophosphate shunt in erythrocytes was studied with theionization chamber-electrometer apparatus to measure continuously 14CO2 derived from14C-labeled substrates. The effect of methylene blue at high (0.1 mM) and low (1 muM)concentrations was evaluated under different gas mixtures; air, carbon monoxide, and 6%carbon monoxide in air. The latter gas mixture results in nearly 100% carboxyhemoglobinbut provides a physiologic partial pressure of oxygen. The extent to which pentose isrecycled through the shunt in response to methylene blue stimulation was examined withradioactive glucose substrates labeled on the first, second, and third carbon positions.Generation of hydrogen peroxide after stimulation of erythrocytes with methylene blue wasevaluated by the catalase-aminotriazole trapping technique, [14C]formate oxidation, andoxidation of reduced glutatione. Stimulation of the shunt with 1 muM methylene blue wasmarkedly impaired in the absence of oxyhemoglobin, but stimulation with 0.1 mM methyleneblue was only slightly impaired under the carbon monoxide-air mixture. The higherconcentration of methylene blue produced evidence of hydrogen peroxide generation of allthree techniques. Despite the evidence for the involvement of oxygen, oxyhemoglobin, andhydrogen peroxide in the response to methylene blue, cells containing methemoglobininduced by sodium nitrite or from a patient with congenital methemoglobinemia respondednormally to methylene blue in the absence of oxygen. These experiments indicate […]

Research Article

Find the latest version:

http://jci.me/108531-pdf

Mechanisms of Methylene Blue Stimulation of the Hexose

Monophosphate Shunt in Erythrocytes

EARL N. METZ, STANLEYP. BALCERZAK, and ARTHURL. SAGONE,JR.

From the Department of Medicine, Ohio State University College of Medicine,Columbus, Ohio 43210

A B S T R A C T The response of the hexose monophos-phate shunt in erythrocytes was studied with the ioniza-tion chamber-electrometer apparatus to measure continu-ously "4CO2 derived from "4C-labeled substrates. The ef-fect of methylene blue at high (0.1 mM) and low (1AhM) concentrations was evaluated under different gasmixtures; air, carbon monoxide, and 6% carbon monox-ide in air. The latter gas mixture results in nearly 100%carboxyhemoglobin but provides a physiologic partialpressure of oxygen. The extent to which pentose is re-cycled through the shunt in response to methylene bluestimulation was examined with radioactive glucose sub-strates labeled on the first, second, and third carbon posi-tions. Generation of hydrogen peroxide after stimulationof erythrocytes with methylene blue was evaluated by thecatalase-aminotriazole trapping technique, [14C1 formateoxidation, and oxidation of reduced glutatione. Stimula-tion of the shunt with 1 AM methylene blue was mark-edly impaired in the absence of oxyhemoglobin, butstimulation with 0.1 mM methylene blue was onlyslightly impaired under the carbon monoxide-air mix-ture. The higher concentration of methylene blue pro-duced evidence of hydrogen peroxide generation by allthree techniques. Despite the evidence for the involve-ment of oxygen, oxyhemoglobin, and hydrogen peroxidein the response to methylene blue, cells containingmethemoglobin induced by sodium nitrite or from apatient with congenital methemoglobinemia respondednormally to methylene blue in the absence of oxygen.These experiments indicate that the reactions inducedby methylene blue in erythrocytes are more complexthan generally thought and that high concentrations areassociated with production of peroxide.

Dr. Sagone is a Scholar of the Leukemia Society ofAmerica.

This work was presented in part at the annual meetingof the Midwest Section, American Federation for ClinicalResearch, Chicago, November 1973.

Received for publication 29 December 1975 and in rev-isedform 17 May 1976.

INTRODUCTIONThe redox dye, methylene blue, is a standard agent foruse in the study of the hexose monophosphate shunt(HMPS)1 pathway (1-5). This compound functionsas an intermediate in the transfer of electrons from pyri-dine nucleotides to a suitable electron acceptor, therebystimulating the HMPSin a variety of cell systems. Inthe erythrocyte, methylene blue is reduced to leuko-methylene blue primarily by an NADPH-dependentdiaphorase. The diaphorase is reduced via oxidation ofNADPHthat, in turn, stimulates the HMPS, and leu-komethylene blue transfers electrons to an acceptor suchas methemoglobin. This series of reactions can be uti-lized clinically for the reduction of ferric to ferrousheme iron in patients with acquired methemoglobinemia.Leukomethylene blue may also react with oxyhemo-globin or molecular oxygen resulting in increased oxy-gen uptake by erythrocytes.

We have investigated further the mechanisms bywhich methylene blue stimulates the HMPS in intacthuman erythrocytes. By using a range of concentrationsof methylene blue and different gas mixtures it is ap-parent that stimulation of the HMPSby methylene bluemay occur by more than one mechanism and that thepresence of oxygen is not required. On the other hand,methylene blue in high concentrations may react withoxyhemoglobin-producing oxygen radicals that react,in turn, with reduced glutathione and stimulate theHMPSvia the glutathione peroxidase-reductase reac-tion. This mechanism might account for the toxicity ofmethylene blue when used in therapeutic excess (6).

The use of ["CIglucose stubstrates labeled on differentcarbon atoms and the use of ionization chamber-elec-trometer apparatus to record continuously the productionof '4CO2 is also a convenient technique for quantitatingthe degree to which pentose is recycled through the

l Abbreviations iused in this paper: GSH, reduced gluta-thione; HMPS, hexose monophosphate shunt; NEM, N-ethylmaleimide; RBC, erythrocytes.

The Journal of Clinical Investigation Volume 58 October 1976- 797-802 797

HMPSat various levels of stimulation with methyleneblue.

METHODSPreparation of erythrocytesVenous blood was collected in heparinized tubes from

normal volunteers and a patient with congenital methemo-globinemia (NADH diaphorase deficiency) and was cen-trifuged at 1,000 g for 10 mim. The plasma and buffy coatwere removed and the erythrocytes were washed once in 5vol of saline and resuspended in a pH 7.4 buffer containing145 mMsodium, 5 mMpotassium, 20 mMglycylglycine,5 mMglucose, and 145 mMchloride. As reported previ-ously, pH values in this incubation system were main-tained at 7.3 or greater for more than 3 h (7). Leukocyteand platelet contamination was less than 1,000/mm3 and10,000/mm3, respectively.

Metabolic studies3-4 ml of packed erythrocytes (RBC) were resuspended

in buffer to a final volume of 10 ml in a 50-ml triple-headeddistilling flask to which was then added 5 ,Ci of radioactivesubstrate. All radioactive reagents were obtained fromAmersham/Searle Corp., Arlington Heights, Ill.

The inlet of the flask was connected to a gas cylinder con-taining either compressed air, carbon monoxide (CO), or6% CO in air. The latter gas mixture produces nearly 100%0carboxyhemoglobin while maintainiing a physiologic Po2(approximately 140 mmHg). The outlet arm of the flaskwas connected to a 275-ml Cary-Tolbert ionization chamberan(i a Cary Mvlodel 401 vibrating reed electrometer (CaryInistrumenits. fonrovia. Calif.). The third arm of the flaskwas covered with a rubber stopper through which reagentscould be addedl or samples withdrawx n through a spinalneedle. The use of the ionization chamber-electrometerapparatus for continuous monitoring of "CO2 produced bycell suspensions, and our modificationis of this apparatus,have been described in detail elsewhere (7-9). A duplicatesystem was used so that 1"CO2 derived from ["C]glucosecould be measured simultaneously from both control andthe experimental flasks. The incubation flasks were main-tained at 37'C throughout these experiments and werestirred continuously. After base-line "CO2 production wasestablished, agents (in buffer) were added through thecenter well to the experimental flasks and an equal volumeof buffer alone (0.1-0.3 cm') to the control flasks. C02production was calculated from the millivolt reading oncesteady-state conditions were reestablished in the experi-mental chambers and compared to the corresponding valuefrom the control flasks. C02 production was calculated aspreviously described and expressed in micromoles of C02per milliliter of RBCper hour (7).

In some experiments, the response of the HMIPS wasstudied in RBC with methemoglobin. MIethemoglobin was

produce(d bv incubation of 1 vol of packed cells with 1 volof isotonic sodium nitrite for 30 mni at room temperature.The packedl cells were then xwashed three times with 8 volof isotonlic saline and resuspended in glycylglycine buffer.Controls for thlese experiments were RBCs incubated insaline at room teml)erature for 30 min and washed similarlyto the nitrite-treated cells.

Studies to detect peroxide generationThree methods were used to detect the generation of

hydrogen peroxide in RBC incubated with methylene blue.

Glutathione stability. RBC were washed thoroughly andresuspended in glycylglycine buffer with or without glucoseand with methylene blue. RBC reduced glutathione (GSH)concentrations were then determined before and after 2 or 4h of incubation at 37'C. RBC GSH values were measuredby the 5,5'-dithiobis-2-nitrobenzoic acid method of Beutleret al. (10).

Catalase inhibition techntique. RBC were incubated for30 min with N-ethylmaleimide (NEM) in amounts requiredto bind all intracellular GSH (2-3 umol NEM/ml RBC).The cells were then washed with isotonic saline and incu-bated for an additional hour with or without methyleneblue (0.1 mM) in the presence of 50 mMaminotriazole.In the presence of aminotriazole. hydrogen peroxide andcatalase form an irreversible complex. Hydrogen peroxidegeneration can be detected by serial determinations of cata-lase (11). After the second incubation, stroma-free hemo-lysates were prepared in a concentration equal to a 1: 1,500dilution of packed RBC. Duplicate 1-ml aliquots of thehemolysate were incubated with sodium perborate sub-strate for 3 min and residual perborate was titrated with1 N potassium permanganate according to the method ofTarlov and Kellermeyer (12). This system provides a quali-tative estimate of peroxide generation and results are ex-pressed as percent fall in catalase activity during a 60-minincubation with methvlene blue.

['4C]Forioate oxidation. Since '4C-formate is oxidizedto "CO2 in the presence of H202 and catalase, this systemwas also used to detect H202 (13). RBC were incubated inmetabolic flasks as previously described, except that ["C]-formate wvas used as a radioactive substrate rather thanj`QC]glucose. Since there is good evidence that H202 ispreferentially reduced by glutathione peroxidase in RBC,["C]formate oxidation was also studied in cells treated withNEMto block cellular GSH (14).

RESULTS

Effect of CO iW. air on, the HMPSof utnstimtutlated andmiiethlylence blue-stimulated RBC. The rate of C02 pro-duction of RBC suspensions incubated with 6% CO inair compared with air or CO alone is illustrated in Fig.1. As previouslv reported, unstimulated RBC suspen-sions incubated uinder CO have a rate of 1"CO2 produc-tion wlhich is 40% of the value under air, suggestingthat approximately 60% of baseline HMPSactivity isrelated to reactioiis involving oxygen (15). In the pres-ence of 6% CO in air, the rate of C02 production of0.055 jumol/nml RBC per h was significantly lower thanthe value for RBC suspensions incubated under air(P < 0.05). These data indicate that a major portionof oxvgeni-depenident HMPS activitv of uinstimulatedRBC stispelnsions is related to reactions involving oxy-lhenmoglobin. and lnot nmolecular oxygen.

In RBC suspensions incubated under air, the additionof metlhylene blue in low concentrations (1 AMI) resultedin a prompt increase in "'CO2 production (Fig. 1). Un-der CO, a barelv detectable increase in "CO2 productionoccurred confirming the anaerobic conditions of our

experiments and an apparent requirement for oxygen inmethvlene blue stimulation of the HMPS. Under CO inair, metlhvlene blue (1 ,LM) resulted in only a small in-

798 E. N. Metz, S. P. Balcerzak, and A. L. Sagone, Jr.

0.6001

0.500-I~~~~I

Ip

TABLE I[1-'4C]Glucose Oxidation with Low, Intermediate, and High

Concentrations of Methylene Blue under AirCompared with 6% Carbon Monoxide in Air

oe 0.080-

t 0.06-

HOURS

FIGURE 1 The effect of carbon monoxide on the oxidationof [1-"C]glucose in RBC incubated with methylene blue(MB). C02 production was calculated for steady-state con-ditions at base line and after methylene blue stimulation(1 AM) under air (13 experiments), under 6%o carbonmonoxide in air (6 experiments), and under 100% carbonmonoxide (7 experiments). The curves are constructedfrom the original continuous recordings using the meanvalues for the three experimental conditions. The bracketsindicate +1 SD.

crease in the COa production to 0.110 Amol/ml RBCper h and was in contrast to the value of 0.580 Asmolwith air alone (P < 0.01). These data indicate thatstimulation of the HMPSby low concentrations ofmethylene blue is primarily related to a reaction involv-ing oxyhemoglobin. This requirement for oxyhemoglobinwas less apparent when the HMPSwas stimulated withhigh concentrations of methylene blue (0.1 mM).In contrast to experiments with lower concentrationsof methylene blue, 0.1 mMmethylene blue resulted in amarked stimulation of [1-'4Clglucose oxidation underair with CO and was 87% of the value for air alone(Table I). Stimulation of [1-14Clglucose oxidation byan intermediate concentration of methylene blue (10 1AM)was significantly impaired as a result of incubation un-der air with CO and was only 36% of the value ofsimilar suspensions under air.

Effect of methemoglobin. Since methylene blue iseffective in the transfer of electrons to methemoglobin,the rate of 14CO2 production by sodium nitrite-treatedcells was examined in the presence of methylene blue.The rate of 14C02 production was studied in the samethree gas mixtures to determine if methylene blue stimu-lation of the HMPScould occur in the absence of oxy-gen. As seen in Fig. 2, the rate of "4CO2 production bynitrite-treated cells incubated with 1 AMmethylene blueunder air was similar to untreated controls. In contrastto control RBC, however, the stimulation of 14CO2 pro-

Methylene blue concentration

None I ;sM 10 pM 0.1 mM

AirMean 0.087* 0.58 2.60 2.90SD :10.016 40.09 ±0.25 40.38

(13) (13) (3) (8)

Air and COMean 0.055 0.110 0.95 2.50SD 40.007 ±0.01 ±t0.11 ±0.25

(6) (5) (3) (5)

* Results are expressed as micromoles of CO2 produced permilliliter RBC per hour. Numbers in parentheses refer tonumber of experiments.

duction of nitrite-treated cells by methylene blue wasunaffected by the absence of oxygen. Likewise, methyleneblue stimulation of the HMPSwas unaffected in nitrite-treated cells which were incubated under the mixture ofCOin air. These data indicate that methemoglobin can actas an electron acceptor for methylene blue in the absenceof oxygen. It is of interest that incubation of nitrite-treated cells with methylene blue under CO resulted incomplete reduction of methemoglobin to carboxyhemo-globin in 3-4 h and that completion of this reactionwas accompanied by a prompt fall in HMPSactivity.

The HMPSactivity of the RBCfrom a patient withcongenital methemoglobinemia (approximately 33%)

0.60 -

m0

t 0.40E

0

0E O020-

AIR CO CO+AIR

FIGURE 2 Oxidation of [1-"C]glucose by nitrite-treatedand normal RBC incubated with methylene blue (1 FM).The values represent the mean increase in 14CO2 productionafter the addition of methylene blue. Three paired experi-ments were carried out under each gas mixture. Nitrite-treated cells are indicated by the hatched bars, control cellsby the solid bars.

Methylene Blue Stimulation of the Hexose Monophosphate Shunt 799

TABLE I IHMPSStimulation with Methylene Blue under Air and

Carbon Monoxide in RBCfrom a Patient withCongenital Methemoglobinemia

Gas mixture

Ca'rbonAir monoxide

Base line 0.078* 0.046Methylene blue (1 ,uM) 0.716 0.630

* Results are expressed as micromoles of CO2 per milliliterRBCper hour.

were also studied to determine the similarity to nitrite-treated cells (Table II). The rate of '4CO2 production ofunstimulated suspensions incubated under air and theresponse to methylene blue were similar to normal RBC.The rate of 14CO2 production of unstimulated RBC sus-pensions also was reduced by incubation under CO. Theaddition of methylene blue, however, resulted in aprompt increase in 14CO2 production similar to nitrite-treated cells.

Peroxide generation in RBCincubated with methyleneblue. Evidence for hydrogen peroxide generation inRBC incubated with methylene blue was demonstratedby the following experiments.

A significant fall in GSHoccurred in RBC incubatedwithout glucose in the presence of 0.1 mMmethyleneblue (Fig. 3). Methylene blue at a concentration of 1 gM

80

C.)

E 60-8N

40E

9 40

6--z

zo 20-UIU)CD

0 I.OAM 0.1 mM L.OALM O.mM2 HXURS 4 HOURS

FIGURE 3 Effect of methylene blue on reduced glutathione.The bars represent RBC GSH concentration (±SD) inmilligrams per 100 milliliters. The open bars indicate sam-ples incubated with glucose and the hatched bars representcells incubated without supplemental glucose. The methyleneblue concentrations were 1 ,uM and 0.1 mMas indicated.

did not affect GSHconcentrations. GSHconcentrationsdid not fall with either concentration of the drug whenthe incubations were supplemented with glucose.

A marked fall in catalase occurred in NEM-treatedcells in the presence of the 0.1 mMmethylene blue andaminotriazole (Fig. 4). No fall occurred in the presenceof the 1M concentration.

Oxidation of formate proved to be the most sensitiveindicator of peroxide generation in response to methyl-ene blue. "4CO2 derived from [14Clformate increased 3times the base-line value after addition of methyleneblue (1 pM) and 40 times the control value in thepresence of the higher concentration (0.1 mM). Bind-ing of GSHwith NEMdiverts peroxide to catalase and,as expected, preincubation of RBC with NEM (3.5Amol/ml RBC) greatly augmented formate oxidation.Under these circumstances 1 pM methylene blue pro-duced a 40-fold increase in formate oxidation which in-

00-

80

LU

O40

I-60\

0I-z0

o40

z

a.)

20-

00 2 4

HOURS INCUBATIONFIGuRE 4 Hydrogen peroxide production by methylene blue(0.1 mM) determined by the catalase-aminotriazole trap-ping technique. The scale on the left represents the percentof initial catalase activity remaining after incubation. RBCwere pretreated with NEMin a concentration sufficient tobind intracellular GSH (2-3 umol/ml RBC). The solidline shows the effect of methylene blue, 0.1 mM, the upper-most dashed line shows the catalase content of control in-cubations, and the middle line shows the effect of NEMwithout methylene blue.

800 E. N. Metz, S. P. Balcerzak, and A. L. Sagone, Jr.

n

creased to approximately 250 times the base-line valueafter addition of 0.1 mMmethylene blue.

Recycling of pentose with methylene blue stimulation.The use of [1-14C]glucose as a substrate provides asemiquantitative estimate of HMPSactivity in RBCand is a useful and convenient method for measuringrelative changes in HMPSactivity with various experi-mental manipulations (3, 7). In fact, however, oxida-tion of the first carbon atom yields quantitative data re-garding the HMPSonly under base-line conditions. Itis well known that when activity of the HMPSis in-creased, pentose is recycled via transketolase and "CO2can then be measured using glucose labeled in the sec-ond carbon position. The contribution of recycled pen-tose to total shunt activity increases progressively as theshunt is stimulated with methylene blue. These data aresummarized in Fig. 5. Under base-line conditions, no"CO2 was detected from glucose labeled in the thirdposition and oxidation of the second carbon was eitherabsent or barely detectable. With addition of 1 uLMmethylene blue, C0a production, calculated from oxida-tion of [1-'4C]glucose, increased from 0.09 to 0.57 4mol/ml RBC per h. At this level of HMPSactivity, 14C02production from [2-14C]glucose was readily detectableat a rate of 0.19 Amol/ml RBC per h, or 33% of thevalue from [1-14C]glucose. Furthermore, oxidation of thethird carbon atom of glucose was now 30% of the valuefrom [2-14C]glucose. With additional stimulation with0.1 mMmethylene blue, oxidation of [1-"C]glucosegrossly underestimated total HMPSactivity. At thishigh level of shunt activity, oxidation of the secondcarbon was 74% of that from [1-14C]glucose and oxida-tion of the third carbon atom was again increased pro-portionally and equaled 75% of the value from [2-14C]-glucose.

DISCUSSIONIt is well established that addition of methylene blue toa suspension of RBC evokes a marked stimulation ofHMPSactivity. This stimulation increases progessivelywith increased concentrations of methylene blue andreaches a maximum at about 0.1 mM(4). It is clearfrom these data, however, that there are qualitativedifferences in the reactions stimulated by methylene blueat various concentrations. For example, stimulation ofthe HMPSby a low concentration of methylene blue(1 IUM) is dependent upon the presence of oxyhemoglo-bin but hydrogen peroxide was not produced in amountsdetectable by the aminotriazole trapping system nor wasthere any detectable oxidation of reduced glutathione.However, data derived from formate oxidation afterNEMblockade suggest that even low levels of HMPSstimulation are associated with generation of smallamounts of peroxide. High concentrations of methyleneblue (0.1 mM) oxidize glutathione and produce enough

0.70

N, 0.50u

m'EN

0O 0.30

0.70-

0.50-

0.30

£0.10 0.10 1I

UNSTIMULATED METHYLENEBLUE METHYLENEBLUE1.OpM 0.1mM

FIGURE 5 Recycling of pentose with HMPS stimulation.The bars indicate "4CO2 production (±+SD) from [1-4GC]glu-cose (open bars), [2-14C]glucose (hatched bars), and [3-"CC]glucose (solid bars) after stimulation with methyleneblue at the concentrations noted. Numbers in parenthesesat the bottom of each bar note the number of experiments.

peroxide to be detected by both formate oxidation andaminotriazole trapping. This production of hydrogenperoxide required, of course, the presence of oxygen,but in the presence of methemoglobin it was possibleto dissociate shunt activity and oxygen utilization.

These observations help to clarify some of the ap-parent contradictions concerning the effect of methyleneblue. For example, Jacob and Jandl found that methyl-ene blue stimulation of the HMPSwas not impaired bysulfhydryl blockade and concluded therefore that methyl-ene blue stimulation of the shunt was not associatedwith hydrogen peroxide generation (5). Roth et al. ar-rived at a similar conclusion based upon the fact thatcells depleted of GSHby incubation with nitrogen mus-tard responded normally to methylene blue while stimu-lation with ascorbate was impaired (16). In contrast,Tephly et al. showed that activation of the HMPSwithmethylene blue and glucose induced the oxidation ofmethanol and ethanol in intact cells by a reaction requir-ing catalase and concluded that this reaction must re-sult in the generation of hydrogen peroxide (17). Thedifferences in peroxide generation which we have dem-onstrated with high and low concentrations of methyleneblue would seem to resolve this contradiction. Tephlymade his observations utilizing a high concentration ofmethylene blue while Jacob and Jandl employed a muchlower dose.

These experiments may also help clarify the inter-relationship of HMPSstimulation with methylene blue,oxygen consumption, and methemoglobin reduction.Experiments in which RBCwere exposed to a mixtureof air and 6% carbon monoxide demonstrate clearly thatoxyhemoglobin is a more effective electron acceptor in

Methylene Blue Stimulation of the Hexose Monophosphate Shunt 801

the methylene blue reaction than molecular oxygen inintact RBC. This may relate to the altered electronspin state of oxygen liganded to hemoglobin (18). Fur-ther, the ability of methemoglobin to serve as an elec-tron acceptor for the methylene blue reaction in theabsence of oxygen suggests that the concept of oxygenconsumption during stimulation of the HMPSis, in asense, artifactual. We would suggest that the reactionbetween methylene blue and oxyhemoglobin results inthe release of oxygen radicals and that the observedoxygen consumption is really related to the generationof these radicals or the reoxygenation of hemoglobin.Although a substantial portion of HMPSactivity inRBCappears to be dependent upon oxygen, it also seemsapparent that this activity represents a reaction to thepresence of oxygen, or active radicals, rather than arequirement for oxygen.

The effectiveness of methylene blue in reducingmethemoglobin is an established clinical fact. In addi-tion, the experiments of Jacob and Jandl showed thatstimulation of the HMPSwith methylene blue in vitroresulted in accelerated methemoglobin reduction (5).In those experiments, however, the concentration ofmethemoglobin did not appear to have any influence onthe degree of HMPSstimulation. Our data suggest thatmethemoglobin and oxyhemoglobin are equally effectiveas electron acceptors from methylene blue and that rateof activity of the HMPSis governed primarily by theconcentration of methylene blue, NADP+, and theHMPSenzymes rather than by the terminal acceptor.Smith and Thron reached a similar conclusion (19).They noted that reduction of methemoglobin by methyl-ene blue was more rapid under anaerobic conditionsthan under air, an observation best explained by a com-petition between oxyhemoglobin and methemoglobin forelectrons. This interpretation might also be a possibleexplanation for those instances of acute hemolytic anemiaprecipitated by toxic doses of methylene blue. When oxy-hemoglobin is the only available electron acceptor, it ispossible that there are sufficient oxygen radicals gener-ated to result in RBC injury.

ACKNOWLEDGMENTSThe authors are indebted to Ms. Rosemarie Husney forher excellent technical assistance throughout these studiesand to Mrs. Adele Klinger for assistance in the preparationof the manuscript.

This work was supported in part by a grant from theNational Institutes of Health, AM15649.

REFERENCES1. Brin, M., and R. H. Yonemoto. 1958. Stimulation of

the glucose oxidative pathway in human erythrocytes bymethylene blue. J. Biol. Chem. 230: 307-317.

2. Szeinberg, A., and P. A. Marks. 1961. Substances stim-

ulating glucose catabolism by the oxidative reactions ofthe pentose phosphate pathway in human erythrocytes.J. Clin. Invest. 40: 914-924.

3. Murphy, J. R. 1960. Erythrocyte metabolism. II. Glu-cose metabolism and pathways. J. Lab. Clin. Med. 55:286-302.

4. Davidson, W. D., and K. R. Tanaka. 1972. Factors af-fecting pentose phosphate pathway activity in humanred cells. Br. J. Haematol. 23: 371-385.

5. Jacob, H. S., and J. H. Jandl. 1966. Effects of sulfhydrylinhibition on red blood cells. II. Glutathione in the regu-lation of the hexose monophosphate pathway. J. Biol.Chem. 241: 4243-4250.

6. Goluboff, N., and R. Wheaton. 1961. Methylene blueinduced cyanosis and acute hemolytic anemia complicat-ing the treatment of methemoglobinemia. J. Pcdiatr.58: 86-89.

7. Sagone, A. L., Jr., E. N. Metz, and S. P. Balcerzak.1972. Effect of inorganic phosphate on erythrocyte pen-tose phosphate pathway activity. Biochim. Biophys. Acta.261: 1-8.

8. Davidson, W. D., and K. R. Tanaka. 1969. Continuousmeasurement of pentose phosphate pathway activity inerythrocytes. An ionization chamber method. J. Lab.Clin. Med. 73: 173-180.

9. Chaudhry, A. A., A. L. Sagone, Jr., E. N. Metz, andS. P. Balcerzak. 1973. Relationship of glucose oxidationto aggregation of human platelets. Blood. 41: 249-258.

10. Beutler, E., 0. Duron, and B. M. Kelly. 1963. Improvedmethod for the determination of blood glutathione. J.Lab. Clin. Med. 61: 882-888.

11. Cohen, G., and P. Hochstein. 1964. Generation of hydro-gen peroxide in erythrocytes by hemolytic agents. Bio-chemistry. 3: 895-900.

12. Tarlov, A. R., and R. W. Kellermeyer. 1961. The hemo-lytic effect of primaquine. XI. Decreased catalase ac-tivity in primaquine-sensitive erythrocytes. J. Lab. Clin.Med. 58: 204-216.

13. Klebanoff, S. J., and S. H. Pincus. 1971. Hydrogenperoxide utilization in myeloperoxidase-deficient leuko-cytes: a possible microbicidal control mechanism. J.Cliii. Invest. 50: 2226-2229.

14. Cohen, G., and P. Hochstein. 1961. Glucose-6-phosphatedehydrogenase and detoxification of hydrogen peroxidein human erythrocytes. Science (Wash. D. C.). 134:1756-1757.

15. Sagone, A. L., Jr., S. P. Balcerzak, and E. N. Metz.1975. The response of red cell hexose monophosphateshunt after sulfhydryl inhibition. Blood. 45: 49-54.

16. Roth, E. F., Jr., R. L. Nagel, G. Neuman, G. Vander-hoff, B. H. Kaplan, and E. R. Jaffe. 1975. Metaboliceffects of antisickling amounts of nitrogen and nor-nitrogen mustard on rabbit and human erythrocytes.Blood. 45: 779-788.

17. Tephly, T. R., M. Atkins, G. J. Mannering, and R. E.Parks, Jr. 1965. Activation of a catalase peroxidativepathway for the oxidation of alcohols in mammalianerythrocytes. Biochemn. Pharmacol. 14: 435-444.

18. Spiro, T. G., and T. C. Strekas. 1974. Resonance Ramanspectra of heme proteins. Effects of oxidation and spinstate. J. Am. Chew. Soc. 96: 338-345.

19. Smith, R. P., and C. D. Thron. 1972. Hemoglobin,methylene blue and oxygen interactions in human redcells. J. Pharmacol. Exp. Ther. 183: 549-558.

802 E. N. Metz, S. P. Balcerzak, and A. L. Sagone, Jr.