Relationship between the Rate of H+ Transport and Pathways of Glucose Metabolism by Turtle Urinary Bladder

Relationship between the Rate of H

+

Transport

and Pathways of Glucose Metabolism by Turtle

Urinary Bladder

L. H. Norby, John H. Schwartz

J Clin Invest. 1978;62(3):532-538. https://doi.org/10.1172/JCI109157.

The urinary bladder of the fresh-water turtle acidifies its contents by actively transporting H+ ions across the luminal membrane. It is known that the H+ transport system is dependent upon oxidative metabolism and the substrate glucose; however, the specific biochemical events resulting in H+ translocation have not been identified.

This study examines the relationship between active H+ transport and a specific oxidative pathway of glucose metabolism, the pentose phosphate shunt. To investigate this

relationship the metabolic and transport rates were simultaneously measured under several well-defined conditions. When H+ transport was inhibited by either the application of an opposing pH gradient or by acetazolamide, glucose metabolism by the pentose phosphate shunt declined. Conversely, stimulation of H+ transport by either imposing a more favorable pH gradient or by CO₂ addition resulted in an increase in pentose phosphate shunt

metabolism. Glycolytic activity, in contrast, was invariant with the maneuvers which altered the rate of H+ transport. Additional experiments localized pentose phosphate shunt enzyme activity to the mucosal fraction of the bladder which is the cell layer responsible for acid secretion. The finding that the rate of glucose metabolism by the pentose phosphate shunt is related to the rate of H+ transport suggests but does not prove that the pentose phosphate shunt may be an important metabolic pathway for H+ transport by the turtle […]

Research Article

Find the latest version:

Relationship between the

Rate

of H+

Transport

and Pathways of Glucose

Metabolism

by

Turtle

Urinary

Bladder

L. H. NORBY andJOHN H. SCHWARTZ, Department of Nephrology, Walter Reed

Army Institute of Research, Washington, D. C. 20012

ABSTR ACT The urinary bladder of the fresh-water turtle acidifies its contents by actively transportingH+

ions across the luminal membrane. It isknown that the

H+ transport system is dependent upon oxidative metabolism and the substrate glucose; however, the specific biochemical events resulting in H+ transloca-tion have not been identified.

Thisstudyexaminestherelationship between active

H+transport and a specific oxidative pathway ofglucose metabolism, the pentose phosphate shunt. To investi-gate this relationship the metabolic and transportrates were simultaneously measured under several well-de-fined conditions. When H+transport was inhibited by either theapplication ofanopposing pH gradient or by acetazolamide, glucose metabolism by the pentose phosphate shunt declined. Conversely, stimulation of

H+ transport by either imposing a more favorable pH gradient or by CO2 addition resultedinan increase in

pentose phosphate shunt metabolism. Glycolytic activ-ity, in contrast, was invariant with the maneuvers which altered the rate ofH+ transport. Additional

ex-periments localized pentose

phosphate

shunt enzyme activitytothemucosal fraction of the bladderwhichis

thecell layer responsible for acidsecretion.The

finding

that the rate ofglucose metabolism by the pentose phos-phate shunt is related to the rate ofH+ transport sug-gests but does not prove that the pentose phosphate shunt may be an important metabolic pathway forH+

transport by the turtle urinary bladder.

This work waspresented inpartbeforethe Federation of

AmericanSocietiesfor Experimental Biologyon 15April1976,

Anaheim,Calif.

Dr. Norby's present address is Department of Medicine, UniversityofIowaCollegeof Medicine, Iowa City, Iowa 52240.

Dr.Schwartz's present addressisThorndike Memorial Lab-oratory,RenalSection, Department ofMedicine,BostonCity Hospitaland Boston University School of Medicine, Boston, Mass.02118.

Receivedfor publication22April1977andinrevisedform

24April 1978.

INTRODUCTION

Several investigators have shown an association be-tween ion transport and metabolic activity in urinary epithelia (1-8). These studies have been primarily cerned with the ratio of ion transported to total 02 con-sumed or

CO2

produced. There have been few studies in urinary epithelia that simultaneously measure and compare the activity ofa specific metabolic pathway and the rate of ion transport (9). Inthe present study

weexamined the relationship between glucose metab-olismby the pentose phosphate shunt andH+secretion inthe isolated turtle urinary bladder. In this epithelium, both the transport and metabolicrates can be carefully measured under well-defined conditions (10). Previous work has shown that (a)H+transport by turtle bladder is primarily dependent upon oxidative metabolism (11), (b) glucose is preferredover pyruvate as a metabolic substrate (11, 12), and (c) the rate ofH+ transport is

linearly related to the rate of glucose metabolism (8).

In addition, preliminary studies from our laboratory demonstrated that

[14C]CO2

production from

[1-14C]glu-cose is greater than from [6-14C]glucose by spon-taneously

acidifying

turtle bladders (13). These ob-servations suggested the possibility of coupling be-tween pentose phosphate shunt metabolism and H+

transport in isolated turtle urinary bladder.

To investigate this possibility we have measured the rate of glucose metabolism by the pentose phosphate

shuntduring manipulationsintherate ofH+transport. WhenH+transportwasstimulated, glucose metabolism by

the

pentose phosphate shunt increased. After in-hibition of acidification, pentose phosphate shunt

metabolism declined. Glycolytic activity, in contrast,

wasunaffectedbychanges inthe rate ofH+ transport. From these data it was concluded that the rate of

glu-cose metabolismby the pentose phosphate shunt was

related to the rate ofH+ transport. Additional studies localized pentose phosphate shunt enzyme activity to

TheJournalofClinicalInvestigation Volume 62

September

1978 * 532 -538

only the mucosal fraction ofthe bladder which is the

cell layer responsible for acidification.

METHODS

Pairedhemibladders were carefully removed from the turtle

and mountedinidenticalLucite (DuPont de Nemours & Co., Inc., Wilmington, Del.) chambers that provided a

sur-face area of8 cm2. The tissues were initially bathed in a

substrate-freeRinger's solution containing in millimoles per

liter: Na, 115;K, 3.5;Ca, 1.0;Mg, 1.0;Cl, 119.5; and HPO4,

1.5. The osmolality was 230 mosmol per kg H20. The pH was adjusted to 5.9 with 0.1 N HC1to minimize trapping of

[14C]CO2as_[14C]HCO-.All

_bathing

media contained0.1mg/ml

ofpenicillin andstreptomycin.Thesolutionswerestirred and oxygenatedwith airthat had been passed through3 M KOH traps to removeall detectable CO2. In one series of experi-mentsthe serosal solutionwasbubbledwith 2.5% CO2 in air

(MathesonGas Products, East Rutherford, N. J.).

Bladders that failedto maintain apotential differencegreater

than20 mVduring the 1sth werediscarded. Thereafter the bladderswerecontinuously short-circuited.H+transport was

measured as the short-circuitcurrent aftersodium transport was inhibited with0.1 mM ouabain. The technique has

pre-viously been validated as an accurate measure ofH+ trans-port rate in turtle bladder (8, 14, 15). After the acidification

ratehad stabilized for2h, the bathingmedia were changed to Ringer's solution containing 5 mM glucose. One hemi-bladder from each pair wasexposed to 10 ,uCi of [1-14C]glu-cose and the other to 10,uCi of[6-14C]glucose (New

Eng-land Nuclear,Boston, Mass.).[14C]CO2production was

meas-ured bypassingthe gaseffluent from the chambers through

5ml HyarnineHydroxide(NewEnglandNuclear) that was later

quantitativelytransferred for liquid scintillationcounting. A

preliminary controlexperimentwith a synthetic membrane

demonstrated 96% recovery of [14C]CO2 from HC1 titration

of['4C]HCO-.Initialexperimentsdemonstrated that [14C]CO2 productionby bladderswaslinearoverthe timeinterval of30 min to 3 h afterexposure to [14C]glucose. The protocol for subsequent studies consisted ofa 30-mincontrolperiod dur-ingwhich therate of[14C]CO2production and H+ transport were continuously and simultaneouslymeasured.After

con-trolobservations therateofH+transport wasalteredandthe

samemeasurements were obtained during a 30-min

experi-mental period. At the end ofeach experiment the exposed tissues wereremoved from the chambers and dry weightswere

obtained. H+transport isexpressedas micromoles per gram

dry weight per hour. The metabolicrates are expressed as

micromoles of glucoseutilized per gram dry weight perhour. The bladder weightsinthese experiments averaged 10±1 mg. To estimate the relative pathways ofglucose metabolism

weassumed that [14C]CO2 evolution from[6-14C]glucose orig-inates from the glycolyticpathwayand that [14C]CO2

evolu-tion from [1-14C]glucose results from both glycolysis and pentosephosphateshunt metabolism.An estimateof[14C]CO2

produced by the latter pathway is obtained by subtracting

[14C]CO2 of[6-14C]glucose from that of[1-14C]glucose. This

simpleand convenientapproach haspreviouslybeen usedto

makequalitativeestimatesofchangesinglucosemetabolism

intoad bladder (9, 16-18).Tovalidate the quantitative aspect

ofthis simplermethodwealso reestimated thepathwaysof

glucose metabolism accordingto amethod ofKatzand Wood (19). Thistechnique requiresindependentmeasurements of

[I4C]C02production and glucoseutilization. The utilization measurements were obtained by incubating an additional

large segmentfrom each hemibladder under similar condi-tions and determining the rate at which[14C]glucose

disap-peared from the medium as indicated by the loss of radio-activity. Overthe timecourseof these experiments, 10-15%

of the added substratewas utilized. The assumptions

under-lying the method of Katz and Wood have been described in

detail elsewhere (19).

Because the turtle bladder consists of a mesothelium, smooth muscle, connective tissue, and mucosal epithelial cells, it was considered important todemonstratepentose phos-phate shunt enzymeactivityin the mucosal cell fraction of the tissue which is the fraction responsible for acidification. In these experiments,bladders were removed from the animal, washed in coldRinger's solution, and placed in a large petri

dish. The mucosal surface was lightly scraped withaglass

microslide.Microscopicexaminationof Papanicolaou-stained

scrapings showedpredominantlyepithelial cells.Afterthree washes in coldRinger's solution containing 10,.tm NADP+, the bladder scrapings were ultrasonicatedandfreeze thawed. The mixture was then centrifuged at 4°Cand 7,000 rpm for 45 min. The supernatefractionwas removed and assayedfor

glucose 6-phosphate dehydrogenase activity by a

spectro-photometric method (20).Thereaction measured therateof

NADP+ reduction at pH 7.6with 1 mMglucose 6-phosphate

as substrate. In theabsence ofsupernate orsubstrate there

was no change in absorbance. The reaction velocity was a linear function of the amountofsupernatefraction added.The

extinction coefficient of NADP was assumed to be 6.22

A/4molNADPH per ml. The proteinconcentrationof supemate

was measuredaccording to Lowry et al. (21). Enzyme activity

is expressed as IU per gram protein where 1 IU is equivalent to1 ,umol ofNADP+reduced permin.

All results are given as the means+SEM. Statistical analysis was by paired t test and differences were considered sig-nificant at P < 0.05.

RESULTS

Timecourse of

["1C]glucose

decarboxylatiotn duritng

spontaneous

acidification.

Before examining the

re-lationship between H+ transport (JH)1 and pathways

ofglucose metabolism it was considered importantto

demonstrate that [14C]CO2 productionwas constant in

the absence of any alteration in JH. This insures that

differences in['4C]C02 production during control and

experimental periods reflect the influence ofthe

ex-perimental maneuver and are not simply the result of spontaneouschangesinisotopeequilibriumor

metab-olism with time. In these studies 12 bladders were

treated with ouabain and incubated in substrate-free Ringer's solution. Afterthe acidification rate was stable

for2h, the bathing medium waschangedto one con-taining5 mMglucoseand 10 ,uCiof labeled substrate.

[14C]CO2 wascollected insequential 15-min intervals

for3hafter isotope addition. The results are presented in Fig. 1. Totalcounts of[14C]CO2increased in a linear

fashionwith time.The equation for the regression line wasy = 172x - 1439 (r= 0.99). Insubsequent studies, control andexperimentalmeasurementswereobtained

during the time interval when[14C]CO2 production- was

IAbbreviations used in this paper: G6P, glucose-6-phos-phate; G6PD, glucose-6-phosphate dehydrogenase; JH, H+

transport.

30

20

0

o 2

vj a.

x 0

10_y

30 60 90 120 150

TIME AFTER ISOTOPE ADDITION (MIN)

FIGURE 1 Timnecourse of[14C]glucosedecarboxylation

clur-ingspontaneous acidification. Results are means±SEM.

shownto beconstantinthesetime-controlexperiments.

These results are in close agreement with those

pre-viously published by Beauwens and Al-Awqati who also demonstrated constanicy of

CO,

production by turtle bladder when the rate ofspontaneous

acidifica-tioni was stable (8).

Chaniges in glucose miietabolism (lurinig inihibitiont of

JH bi acetazolamidle. Table I shows the effects of 0.1 mM acetazolamide on JHand glucose metabolism. H+ secretion averaged 65

_gmol/g

dry wt per h during the

control period and declined to a value not different

from zeroafteracetazolamide. Complete inhibition of

JH by0.1 mM acetazolamide isin agreement with pre-viousreports (10, 22).[1-14C]Glucoseutilization during the control periodwas2.72

_gumol/g

drywtper h and de-creasedsignificantly after acetazolamideto 1.76

_,umol/g

dry wt per h. In contrast, [14C] C02 production from [6-14C]glucose

didl

notsignificantly changeafter

inhibi-TABLE I

Chaniges in Glucose Metabolism (lurinigInihibitiont of

J" b! Acetazolamide

[6-'4C]Glu-[1-_4C]Glucose cose C-1-C-6 JH p.mtol/g(Irq/1wt/li

Control 2.72 0.76 1.96 65

Acetazol-amiiide 1.76 0.72 1.04 4

A-SE 0.96+(0.20* 0.04±0.03 0.92+(0.21* 61±12

Acetazolaimidle, 0.1 m11X, was adldedl to the serosal solution

after the conitrol periodl. (ni =6)

*_P<0.05.

tion of JH by acetazolamide. Subtracting [14C]CO2 pro-duction of [6-14C]glucose from that of [1-14C]glucose

gives an estimate of glucose metabolism by nonglycolytic

pathways, the principle one of which is the pentose phosphate shunt (16, 17).Afteracetazolamide this value declinedfrom 1.96 to 1.04

Amol/g

drywtper h. These results show that inhibition of JH by acetazolamide is associated with a decline in glucose metabolism by

the pentose phosphate shunt.

Effects of varyintg mucosal pH oil JH and1C glucose

mletabolisml.

The rate ofnetacidification by isolated

turtle bladder is a linear function of the mucosal

solu-tionl

to cell pH gradient (10). The back leak ofH+ is

negligible over themucosalpH range of 4.5-7.0, hence the net rate of acidification approximates the rate of

active JH(8, 10). Inthe next series ofexperiments we

determined the effect of inhibition of JH by mucosal

acidification on pentosephosphate shunt metabolism.

The results are presented in Table II. In these eight experiments, mucosal pH waslowered from 5.9inthe controlperiod to4.8duringexperimental period.In

re-sponse to the reduction in mucosal pH, JH

decreasedl

from 70 to 7

,uinmol/g

dry wt per h.[1-14C]Glucose utiliza-tion declined from 3.11 to 1.92 umol/g dry wt per h whereas[6-14C]glucose metabolismwasnot significantly changed. Pentose phosphate shunt metabolismi,

es-timatedasthe difference of[14C]CO2 from [1-14C]minus

[6-14C]glucose, decreased significantly during

inhibi-tionl

of JH. These results are

simiiilar

to those obtained

with

acetazolamiiide andcl

provide additional support for

a relationship between pentose phosphate

shunit

metabolism and urinary acidification.

In a separate group of bladders, mucosal pH was

increasedfrom 5.9 in the control period to 7.5 except

for abrief interval at the endofthe

experimental

pe-riod when it was lowered to 4.9 to remove any[14C]CO2

trapped

as

[14C]HCOJ.

TheseresultsareshowninTable

III. In

response

to the increase in mucosalpH, JH ill_

creasedfrom63 to 105,umol/g drywtper h.

[1-14C]Glu-cose utilization increased from2.91 to 4.24

,umol/g

dry

TABLE 1

Chaniges in Glucose Metabolisml clurintg Inihibitioniof JHbl Applied pH

Graclient

[6-.4C]Glu-[1-'4C]Glucose cose C-1-C-6 JH

/iLmIol/g(Iry/ittil

Conitrol 3.11 0.82 2.30 70

pH

Gra(li-enit 1.92 0.69 1.22 7

A+SE 1.19+(0.37* 0.13+(0.12 1.08±0.23* 63±6*

Mlucosal pH was loweredl fromii 5.9 to 4.8 after the conitrol

period.(n =₈₎

*P_<

0.05.

TABLE III

Chlcanges in Glucose Metabolism dlurinig Stimitulationtof JH

bil

MucosalAlkalintization

[6-'4C]Glu-[1-'4C]Glucose cose C-1-C-6 JH ,Ltmollgdrywtlh

Conitrol 2.91 0.64 2.27 63

Alkaliiniza-tioni 4.24 0.72 3.52 105 A+SE 1.22±0.51* 0.08±0.07 1.25±0.42* 42±6* Mucosal pH was iniereased fromii 5.9 to 7.5 after the conitrol period. (n =8)

*P _<0.05.

wtper h.

[6-'4C]Glucose

utilizationwas nlot statistically

affectedby theincreaseinl mucosal pH. Glucose

metab-olism by the penitose phosphate shunt increased sig-nificantly duriingstimulationi of JH.

Effect ofCO., enth(anicemlenit of

JH

onl glucose

mrletab-olism. Therelationshipbetween JHandglucose

metab-olism by the pentose phosphate shunt was also

ex-aminedinagroup of bladders before and after exposure to exogenous

CO,.

Previous work has shown that the

availability ofCO2 is rate limliting forJH in

substrate-replete bladders with nio tranisepithelial pH gradient

(10). In these seveni

experimiients

the serosal gas phase

waschangedfromii

CO,-free

air to2.5%CO2inairdurinig the experimnenital periodl. The results are shown in

Table IV. JHinicreasedl fromii 0.77 to 1.23

,tumol/g

drywt

per h after additioni of

exogenious

CO2. The inicrease inl tranisport rate was associated with a marked

incre-miienit

in

[1-14C]glucose

utilization. Therate of

[6-14C]glu-cose utilizationi did notchanige signiificantly after

addi-tioni of exogenous CO2. The fraction ofglucose metab-olized by the pentose phosphate shunt increased significantly after enhancement of JH by CO2addition. Calculationtofpentose

_p)hosphate

shunit mrletabolism by a method ofKatz anti( Wood (19). Use of absolute yields of[14C]C02 to evaluate the pathway of glucose

TABLE IV

Chatnges inl GlucoseMletabolisml dlurinig Stimlulation

of ]"by CO2

[6-'4C]Glu.

[1-_4C]Glucose cose C-1- C-6 H

ILmolIg(drywtlh

Conttrol 3.84 0.86 2.97 77

2.5% CO2 6.71 1.55 5.16 123

A±SE 2.87±(0.50* 0.69±0.37 2.17±0.66* 46±7* The serosal gas phase was chaniged to 2.5% CO2 after the

conitrol periodl.(n =₆₎

*_P<0.05.

mnetabolismiifails to conisider recycling of hexose phos-phates, imietabolismof glucose by nontriose phosphate pathways, and incomplete oxidation of triose phos-phates to CO. Katz and Wood have provided a method which conisiders these possibilities in estimating the relativepathways of glucose metabolism from [14C]CO2 evolutioni studies (19). Table V shows the results of the presenit studies analyzed according to this tech-nique. The specific yield refers to that fraction of utilized glucose which appears as [14C]CO2. In hemibladders exposed to[1-14C]glucose,the specific yieldof[14C]CO2 increasedduring stimulation of JH and decreased when acidification was inhibited. In contrast, the specific yield of [14C]CO2 from bladders exposed to [6-14C]glucose

was invarianit with maneuvers which significantly al-teredJH.The changes in thecontribution of the pentose phosphate shunt to glucose metabolism were similar to the changes in the specific yield of [14C]CO2 from [1-14C]glucose. During stimulation of JHby 2.5% CO2, pentosephosphateshuntmetabolism increased from29 to 59%. A similar response was seen after JH was

in-creasedbyraisingmucosalpH.WhenJH wasinhibited

by acetazolamnide, calculatedpentosephosphateshunt

mnetabolismii declined from 35 to 10%. The response ofthepentose phosphate shunt metabolism to

inhibi-tion ofJH by an imposed pH gradient was virtually

identical to that of acetazolamide.

TABLE V

Chaniges inlGlucose Metabolism by Penttose Phosphate Shunit(PS)durinig Alterations inJH

Specific yield Specificyield

[1_-4C]glucose [6_'4C]glucose PS JH

% ymollg dry wtlh

Conitrol 0.66+(0.05 0.24+0.03 29+6 77±10

Additioni

CO2 0.86±0.03* 0.26+0.03 57+7* 123±13

Cointrol 0.65+0.10 0.27+0.03 28±+11 63±+10

NMucosal

pH 7.5 0.77+0.09* 0.28+0.02 42+14* 105+16

Conitrol 0.17±0.09 0.24+0.04 35±+11 65+4

Acetazol-amnide 0.43±0.05* 0.25+0.03 10+2* 4+2

Control 0.76±0.08 0.24+0.04 42±10 70+4 Mlucosal

pH4.9 0.43±0.05* 0.25+0.04 10+2* 7+3* Paired hemiiibladders were exposed to either [1-14C]- or

[6-'4C]glucose aind ['4C]C02 production was determnined

before aind after alterations in JH. The percent contribution ofPS to glucose metabolisimi was calculated according to a

methodof Katz and Wood where specific yield refers to that fraction of utilized glucose which appears as [14C]C02 (19).

(it=₈₎

*_P<0.05.

Demonstration of pentose phosphateshunt dehydro-genase activity inmucosalhomogenates. Toexclude the possibility that changes in [14C]CO2 production

resulted from changes in pentose phosphate shunt

metabolism innonacid secreting portions of the blad-der, we measured the activity and distribution of glucose6-phosphate dehydrogenase (G6PD)inmucosal andnonmucosalfractions of the bladder.G6PDactivity

of 16 single-bladder homogenates averaged 346+44 IU per g protein. Turtle erythrocyte G6PD activity was 1.0±0.2 IU per g Hb. Less than 2% of G6PD activity in mucosal homogenates could be accounted for by erythrocyte contamination. G6PD activity was

notfound innonmucosal bladder fractions.

In a second group of eight bladders the reaction

velocity was analyzed as a function of substrate

(glu-cose-6-phosphate,

G6P) concentration. A double

re-ciprocal plot of the results is presentedin Fig. 2. The calculated Michaelisconstant(Km G6P)was 0.0001and maximal enzyme activity was 248 IU per g protein.

This Km is less than physiologic concentrations of the substrateG6P. It isalso lower than the KmG6Pof1 mM

previously reported forratgastric mucosa(23). Inthis

gastric acidsecretingtissuethe pentose shunt metab-olism accounts for a large fraction of02 consumption (23). In additional studies the range ofpH optimum

for turtle bladder G6PD was found to be 7.1-8.0.

DISCUSSION

Comparedto ourunderstanding of theenergy require-ments for sodium transport, the energetics of JH in

0.0i

/V

0.005

Vmox - 248 I U PER G PROTEIN

Km(G 6P)= 0.1mM

4.0

1/ (G6P)MM

FIGURE2 Double reciprocal plot ofreaction velocity as a

function ofG6P concentration. The reaction velocity meas-ures the rate ofNADP+ reduction. The equation for the

re-gression lineisy=0.000394(±0.00011)x + 0.004(+0.0003) (r=0.99,n =_8).TheKm. 0.1±0.02 mM,was estimatedfrom theslopeand intercept ofthe regression line with the ordinate.

urinary epithelia havenotbeen asthoroughlydefined. In a recent study Beauwens andAl-Awqati were able to

demonstrate tight coupling between JH _{and glucose}

oxidation in isolated turtle bladder (8). Other investiga-tions have shown that urinary acidification by turtle bladder is primarily dependent upon aerobic

metab-olism (11)and thatglucose is the preferred metabolic

substratefor support of H+ secretion (11, 12). The sub-strate specificity for glucose and dependence upon

aerobicmetabolism suggest thatglucosesupports JHby

turtle bladder through some mechanism other than

glycolytic metabolism orthe Krebs cycle.

The majoralternativepathway for glucose oxidation

is the pentose phosphate shunt.Therefore, we sought todetermine ifthis pathway was important in urinary

acidificationby isolated turtle bladder. Inthe present

studyit is demonstrated that therateofpentose

phos-phate shunt metabolism by isolated turtle bladder is

relatedtothe rateof JH. Inhibitionofurinary acidifica-tionby two different maneuvers resulted in identical decreases inglucose metabolism by thepentose phos-phate shunt. Conversely, whenH+ secretion was

stim-ulated either byaddingexogenous CO2orbyimposing a morefavorable pH gradient pentose phosphate shunt

me-tabolismincreased. Glycolyticactivity,incontrast, was not

statisticallychanged bythe maneuverswhichaltered the

rateof acidification andpentosephosphate shunt metab-olism.Because of the magnitude ofchangein[14C]CO2 productionfrom [1-14C]glucoseasopposedto thatfrom [6-14C]glucose with changes in the acidification rate, weconsidereditunlikely thatthese results represented factors other thanchangesinpentosephosphateshunt

metabolism.To confirm this impression, wealso used

the more quantitative method of Katz and Wood to

estimate the pathways of glucose metabolism (19).

These results (Table V) support the conclusion that therateofpentosephosphate shuntmetabolismvaries with the rate of JH. Furthermore, the possibility that changesin[14C]CO2production resulted fromchanges

inpentosephosphate shuntmetabolism in nonacid se-cretory portionsof the bladderisexcludedbythe

local-ization ofpentosephosphate shuntenzyme activityto

mucosal epithelial cells.

Thus it wouldappear thatthe previously described

linear coupling between glucose metabolism and JH

(8)probably represents a coupling betweenJHand

glu-cose oxidation by the pentose phosphate shunt. Al-thoughit isattractive toconcludethatJHis dependent upon pentoseshuntmetabolism, it is also possible that

thesetwo apparently coupled activities of the bladder

epithelial cells are not functionally dependent upon

one another, but only linked by a common variable alteredby the maneuvers used to change H+ transport. The most likely factor common to both functions that could indirectly link these rates would be intracellular pH. For example, the alteration in H+ transport may

changeintracellular pH which then changes the rate of pentose shunt metabolism. This particular possibility, however, isimprobable. Intracellular pH does not

ap-preciably change after lowering mucosal pH or after

addition ofCO2 or acetazolamide (24). Also, the range

of pH optimum for G6PD is well within- the ranige

of change in intracellular pH of turtle bladder after these

maneuvers.

A dependence of JH upon pentose phosphate shuint metabolism may not be unique to the turtle bladder. Sernkaand Harris have presented evidence suggesting that acid secretion by ratgastric mucosais dependent

upon glucose metabolism by this pathway (23). Dies

and Lotspeich have made a similar proposal for acid-ification by rat kidney (25). The present study, how-ever, represents the first report of simultaneous

meas-urements of the rate of metabolisnm by this specific pathway and the rate ofJH. In each group of experi-mentsthe rate ofpentosephosphate shuntmetabolism varied with the rate of JH. This finding providles

ad-ditional support for the idea that pentose phosphate

shuntmetabolismisinsomewaycoupledtothe cellular process of acid secretion.

The nature of the coupling between the penitose

shunt pathway for glucose metabolismand active

J'

is

not readily apparent. The oxidative anid

decarboxyla-tion reacdecarboxyla-tions catalyzed by G6PD are

probably

niot di-rectly related to the transport step because these eni-zymes arelocalized inthecytosol ofthis tissue2anid in

othertissues (26). Perhapsthe pentose shuntproduces

a specific substrate that is required for JH. The miiajor end product ofthis pathway is NADPH. The

depend-ence ofJH upon NADPH could be indirect, that is,

NADPH isutilizedfor the synthesis ofsomeotherclass

of substrate not present in the in vitro system. One

possibility is that fatty acids are important metabolic

fuels for JH by the turtle bladder. The function of the pentose shunt would be to produce reducing equiva-lents, NADPH, which are required for the biosynthesis offattyacids.Althoughthispossibilitywas not directly examined inthe present study, we view it as

unilikely

becauseexogenousfattyacids fail tostimiiulateJHin

sub-strate-depleted bladders and the

metabolismii

of fatty

acids is notcoupledto the rate of JH (27). Aniother

ex-planationi

is thatthe endproduct ofthe pentose

phos-phate shunt, NADPH, provides the source of

protonis

for

translocationi

across the luminal

miiembrane.

This proposal is similar to the one offered by Dies

and(l

Lot-speich(25) and

represenits

a

mlodificationi

of theConway

redox pump theory (28) or the

NMitchell chemiiiosmiiotic

pumllp

theory (29).

Accordinig

to this

scheimia

NADPH

is oxidized by a luminial bounied

cytochrome

or its

equivalent resulting in H+ secretioni inito the lumien and OH-

geinerationi

in the cell.

2

Unpublished

_{observations.}

This latter proposal requires that 1 mol ofNADPH be conisumiied for each mole of H+ tranisported. The

comiipleteoxidation of each mole of glucose by the

peni-toseshunit results in theformiiationof 12imol of NADPH. Thus, the rate of JHshotil(Iben1o miiore thani 12timiiesthe

rate of glucose imietabolized by the penitose shunit. In thisstucdv the ratio ofchanigein the rate Of JH to chanlge

inpenitoseshuniitmetabolisimiof glucosevariedlbetween

14.7 anid 66.5, with a miiean valuie of 44.2±10.1. It has been previously reported that metabolism of exogenous glucose accounits forone third or less of total CO2 pro-duced by the isolated turtle bladder (8). If the bulk

of enidogenous substrate utilized is glucose derived

from tissue glycogen aned this glucose is metabolized

proportionately at the samiie rate by the pentose shunt as exogenious glucose, this ratio miiaybe as low as 14. This estimated value approximates the 12:1 ratio re-quired by our last proposal. Therefore, the rate of NADPH productionand utilization as a source of pro-tons for trainsport could account for the bulk of the

acidificatioin rate.

REFERENCES

1. Maffly, R. H.,anid I. S. Edelimnani. 1963. Ihe coupliingof short-circuit currenit to metal)olisminlthe urinarybladl(ler of the toad.J. Geni. Phmt1siol. 46: 733-754.

2. Schwartz, J. H.,anid P. R. Steinlmetz. 1971. C(2

require-menitsfor H'secretion1l the isolate(d tturtle uriniary blad-der. Amii. J. Pi1.ilsiol. 220: 2051-2057.

3. Coploni,N.S., anidR.H.Mafflv. 1972. The effect of ouabain oni sodiumtranisportanid metabolism of the toa(l bladder.

Biochimn.Bioph1 s.Acta. 282: 250-254.

4. Nellans, H. N.,and A. L. Finn. 1974. Oxygen

consumiip-tion anii( so(liumln transport in the toad uriniary bladder.

Am71.J. Ph/n.siol. 227: 670-675.

5. Al-Awqati, Q.,R.Beauwenis, anidA.Leaf' 1975.Couplinig

ofso(liumii tranisport to respirationi ill the toad( bladder. J. Membr. Biol. 22: 91-105.

6. Maffly,R.H.,H.Steele,W.Walker,andlN.Coploni. 1973.

Respirationiand(activesodliumiitransport.Am. Soc.NTephrol.

72. (Abstr.)

7. Beauwenis,R.,anid Q. Al-Awqati.1976.Further studiesoni couplinig betweeni sodiumii tranisportandrespirationiintoa(I

uriniary bladder.Am.j. Phil1siol.231: 222-227.

8. Beauwenis, R., aindQ. Al-Awqati. 1976. Active H+

tranis-port inl the turtle urinjary bladder. J. Gent. P/lilsiol. 68: 421-439.

9. Kirchberger, M\. A., D. G. Martin, A. Leaf, and(IG. W. G.

Sharp. 1968. Theeffect ofaldosteronie oni glucose

metab-olismiiinltoadblacdcder.Biochim.Biophys.Acta. 165: 22-31. 10. Steiinmiietz, P. R. 1974. Cellular mechanisms of urinarv

acidlificationi.Phll.siol. Rev. 54:889-956.

11. Schwartz, J. H., and(i P. R. Steinmiiietz. 1977. Mletabolic

eniergy and PCO2 as (leternlilcaints of H' secretioni by turtleurinary bladder.Am. J. P/lzmsiol. 2: F145-F149.

12. _Al-Awqalti,Q.,L.Norby, A. Mueller,and P.R.Steinmiiietz.

1976. Characteristics ofstimiuEllationl of H' traniisport by

aldosteromme inl turtle turiniary bladder. J. Cliti. Ivcest.

58: 351-358.

13. Norby,L.H.,andJ.H.Schwvartz. 1975.Roleofthepenitose shuntinl urinarv aci(lificationi.KidieijIot.8: 486. (Abstr.) 14. Schwartz, J. H. 1976. 11H cnrrenit responise to CO2 and(1

carbonic anhydrase inhibition in turtle bladder. Aml.J.

Physiol.231: 565-572.

15. Steinmetz, P. R., R. S.Omachi,and H. S. Frazier. 1967.

Independence of hydrogenionsecretionand transportof other electrolytes in turtle bladder. J. Cliti. Invest. 46:

1541-1548.

16. Kirchberger,M.A., L. C.Chen,and G. W. G. Sharp. 1971. Furtherstudies on the effectofaldosterone on glucose metabolism in toad bladder. Biochim. Biophys. Acta. 241: 861-875.

17. Beckman,B., G. Beat, R.Leaf,P.Witkum, and G. W. G. Sharp. 1974. Evidence for anapical membrane effect in theregulation ofthepentose monophosphateshunt

path-way in toad bladder studies with amiloride. Biochim.

Biophys. Acta. 332: 350-357.

18. Spooner,P.M.,andI. S. Edelman. 1976. Effectsof aldo-sterone on Na+ transport in the toad bladder.Biochimit. Biophys.Acta. 444: 653-662.

19. Katz, J., and H. G.Wood. 1963.The use of C1402 yields from glucose-l- and-6-C'4 for the evaluation of the path-waysof glucosemetabolism.J.Biol. Chem.238:517-524.

20. Beutler, E. 1971. Red cellmetabolism: a manual of

bio-chemical methods.Grune & Stratton Inc., New York. 21. Lowry,0.H., N. J.Rosebrough,A. L. Farr, and R. J.

Ran-dall. 1951. Protein measurement with the Folin phenol reagent.J.Biol. Chem. 93:265-275.

22. Schwartz,J. H., S. Rosen,andP.R.Steinmetz. 1972. Car-bonixanhydrasefunction and theepithelialorganization ofH+secretion inturtle urinary bladder. J. Cliii.

Itnvest.

51:2653-2662.

23. Sernka, T.J.,

anid

J. B. Harris. 1972. Pentose phosphate

shuntandgastricacid

secretioin

intherat.Amti. J.Physiol.

222: 25-32.

24. Steinmetz, P. R. 1969.Acid-base relationsin

epitheliumii

of turtlebladder: site of active step in acidification and

role ofmetabolic

CO2.J. CliGu. Intvest.

48: 1258-1265. 25. Dies, F.,andW. D.Lotspeich. 1967.

Hexose-imonophos-phate

shunit

in thekidney acid-base andelectrolyte

im-balance.

Am,l.

J.

_Phiysiol.

212: 61-71.

26. Krebs, H. A., and L. V. Eggleston. 1976. Regulation of

pentosephosphate cycleinrat liver.

Itn

Advancesin

En-zymeRegulation.G.Weber, editor.

Pergamoni

PressInc., Elmsford, N.Y.421-434.

27. Al-Awqati,Q. 1977.Effectofaldosteroneon thecoupling between H+ transport and glucose oxidation. J. Clin.

Itivest. 60: 1240-1247.

28. Conway, E. J. 1953. The biochemistry ofgastric

secre-tion. Charles C.Thomas Publisher, Springfield, Ill.185pp. 29. Mitchell,P. 1961.Couplingofphosphorylationtoelectron

and hydrogen transfer by a chemi-osmotic type of

mechanism.

Nature

(Loud(I.).

191: 144-148.