AfM

HI

R

moqvoamfsplimoafae

omtpittbrwtpl

E

o(

Biochemical and Biophysical Research Communications 259, 606–610 (1999)

Article ID bbrc.1999.0785, available online at http://www.idealibrary.com on

0CA

cetate and Formate Uptake into Vesicles Isolatedrom the Basolateral Region of the Plasma

embrane of Ovine Parotid Acinar Cells

a-Van Nguyen1 and R. Brian Beechey2

nstitute of Biological Sciences, University of Wales, Aberystwyth, SY23 3DD, Wales, United Kingdom

eceived April 5, 1999

source of carbon and hydrogen for energy production(eottac

E

waiiaM

ccw

ttse

Cb

po

aid1stf

[ms

The transport of acetate and formate into plasmaembrane vesicles derived from the basolateral face

f the ovine parotid acinar cell has an absolute re-uirement for an anion to be present within the intra-esicular space: bicarbonate, formate, acetate, propi-nate, and butyrate support the uptake of eithercetate or formate. A pH gradient across the vesicleembrane, pHi 7.4, pH0 5.5, enhances the uptake of

ormate, but not acetate. There is no direct relation-hip between the rate of exchange and the degree ofrotonation of formate or acetate in the extravesicu-

ar medium. The process is saturable and can be inhib-ted by a range of functional group reagents. When

annitol is the main external osmoticum, the uptakef acetate and formate is still rapid; thus, no other ionsre involved in the process apart from the externalormate or acetate and the intravesicular anion. Thisctivity could play a major role in the provision ofnergy in ruminant tissues. © 1999 Academic Press

In ruminant animals the mechanisms for the uptakef acetate into the peripheral organs and tissues are ofajor metabolic significance. The main end products of

he rumen fermentation of ingested forage are acetic,ropionic and butyric acids (1, 2). These are absorbednto the systemic system mainly through the wall ofhe rumen. In the sheep, acetate comprises $95% ofhese anions; the concentrations in arterial and venouslood are 0.8 and 1.4 mM respectively (3, 4). They areapidly absorbed in the peripheral organs and tissues,here the propionic acid is converted almost quantita-

ively into D-glucose by well-established gluconeogenicathways (3). The main function of acetate and to aesser extent butyrate is to provide the major metabolic

1 Present address: Medical Centre, University of Rochester, 601lmwood Avenue, Box 611, Rochester, NY 14642-8611.

2 To whom correspondence should be addressed at Physiology Lab-ratory, University of Liverpool, Liverpool, L69 3BX, UK. Fax: 1440) 151 794 5337. E-mail: rbeechey@liv.ac.uk.

606006-291X/99 $30.00opyright © 1999 by Academic Pressll rights of reproduction in any form reserved.

4). It has been estimated that 80% of the metabolicnergy that is utilized in sheep is derived from thexidation of acetate and to a much lesser extent bu-yrate (3). Essig and co-workers have demonstratedhat the half-life of acetate in the blood of sheep ispproximately 3 min (5, 6). Clearly, acetate plays aentral role in the overall metabolism of the ruminant.

XPERIMENTAL

Collection of parotid glands. Sheep of mixed genetic background,ere stunned electrically and exsanguinated by colleagues workingt the abattoir. The parotid glands were removed, trimmed of adher-ng fatty and connective tissues, wrapped in aluminum foil andmmediately dropped into liquid nitrogen. This procedure was usu-lly completed within 3–5 min. Prior approval was obtained from theeat Hygiene Service.

Preparation of plasma membrane vesicles from parotid acinarells. Basolateral membrane vesicles were prepared using the pro-edure described previously (7). Routinely, approximately 6–9 g wett of frozen tissue was used in each preparation.The final pellets (2.5–4 mg protein) were suspended in 500 ml of

he solution to be internalized. The internalized solutions all con-ained 100 mM mannitol, 0.1 mM magnesium sulfate, 0.02% (w/v)odium azide, 20 mM buffer of the required pH and 200 mosM anion,ither sodium or potassium salts.

Estimation of protein. Protein was assayed by its ability to bindoomassie blue according to the Bio-Rad assay technique, usingovine g-globulin (30–150 mg protein) as the standard.

Assays of ouabain-sensitive K1-activated phosphatase and alkalinehosphatase activities. These were performed as described previ-usly (8).

Purification of sodium and potassium polygalacturonic acids withhigh molecular mass. Sodium and potassium polygalacturonic ac-

ds, 85–90% purity were purchased from Sigma. Solutions, 10% (w/v) inistilled water were placed in Visking dialysis tubing, cut off point2–14 kDa, and dialyzed against distilled water containing 0.02% (w/v)odium azide at 4°C. The dialysate was changed hourly, at least fiveimes. The dialyzed solution was transferred to round bottom flasks andreeze dried. The final products were white powders.

Measurements of uptake of acetate and formate. The transport of1-14C]acetate or [14C]formate into sheep parotid acinar basolateral

embrane vesicles was measured using a standard rapid filtrationtop technique as described by Shirazi et al. (9).

The data are presented as representative experiments. Each datape

ap

R

atnrsmBaamvTbaimsnpwv1tabv

asmshmbffu(r

tvttptfi

ipp

avwcFm[lsttttitwtpav

iV1

TABLE 1

Rb

MGPBFAPBPLTOS

laagNTa

Vol. 259, No. 3, 1999 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

oint was performed either in triplicate or quadruplicate, and isxpressed as mean 6 SEM.

Radiolabeled compounds. Sodium salts of [14C]formate specificctivity, 2.07 GBq/mmol and [1-14C]acetate 2.11 GBq/mmol wereurchased from Amersham International plc. (UK).

ESULTS

Uptake of acetate and formate into ovine parotidcinar basolateral membrane vesicles: trans-stimula-ion by intravesicular anions. The ovine parotid aci-ar basolateral membrane vesicles were loaded with aange of isotonic solutions that contained 200 mosMalt, 100 mM mannitol, 20 mM Hepes/Tris, pH 7.4, 0.1M magnesium sulfate and 0.02% (w/v) sodium azide.oth sodium and potassium salts were tested. Thebility of the vesicles loaded with different internalnions, to support the uptake of either acetate or for-ate (1 mM) from the extravesicular medium was in-

estigated. It can be seen from the data presented inable 1 that acetate moved across the vesicle mem-rane at very significant initial rates. These rates wereffected greatly by the nature of the anions in thentravesicular medium. Vesicles preloaded with for-

ate, acetate, propionate and bicarbonate demon-trated rapid initial rates of acetate uptake, 15–36mol.min21.mg21 protein, from the reaction medium, atH 7.4, inside and outside. However, vesicles loadedith phosphate, sulfate, thiocyanate, gluconate, pyru-ate and lactate, showed lower rates of acetate uptake,–2 nmol.min21.mg21 protein. These results indicatehat there was an absolute requirement for specificnions to be present within the intravesicular spaceefore the uptake of formate or acetate into the intra-esicular space was observed.There was a low, minimal uptake of both formate

nd acetate when the vesicles were pre-loaded witholutions that contained 300 mM mannitol or 100 mMannitol plus polygalacturonic acid, 200 mM with re-

pect to Na1 or K1. This indicates that neither internalydroxyl ions nor the presence of the internal, imper-eable polyanionic polygalacturonate species, nor the

uffer and azide anions can stimulate the uptake oformate or acetate. There was no specific requirementor either Na1 or K1. The rates of acetate/formateptake were unaffected when the major osmolytesNa1 or K1) in the external reaction medium wereeplaced by mannitol.

If the uptakes of both acetate and formate were dueo non-specific diffusion of the protonated acid into theesicles then differences in the composition of the in-ravesicular osmoticum would not be expected to affecthe rates of internalization of formate and acetate. Thereliminary conclusion to be drawn from these data ishat the vesicle membranes contain a protein(s) thatacilitates the movements of acetate and formate an-ons across the membrane; possibly in exchange for the

607

ntravesicular anions. The preferred substrates for thisotential antiporter mechanism are formate, acetate,ropionate, butyrate, and bicarbonate.

Intravesicular location of the radiolabeled acetatend formate associated with the ovine parotid acinaresicles. The osmolarity of the incubation mediumas varied from 250 to 700 mosM by changing the

oncentration of mannitol, as described in the legend toig. 1. The vesicles were incubated in the reactionedia for 30 min (to reach equilibrium) with 1 mM

1-14C]acetate or [14C]formate, and the content of radio-abel in the vesicles was determined. The data pre-ented in Fig. 1 show that there was an inverse rela-ionship between the uptake of acetate or formate, andhe osmolarity of the reaction medium. On extrapola-ion to infinite concentration of mannitol, i.e., zero in-ernal volume, the lines pass through the origin. Thismplies that the major proportion of the acetate, andhe formate associated with the vesicles is locatedithin an osmotically active compartment; in this case

he lumen of the vesicles. There is no evidence for theresence of a significant amount of either of thesenions being associated with the external face of theesicle membrane.

The time courses of anion-stimulated acetate uptakento ovine parotid basolateral membrane vesicles.esicles were preloaded with 100 mM mannitol and00 mM solutions of potassium salts: bicarbonate, for-

Effects of Different Intravesicular Anions on the Initialates of Acetate and Formate Uptake in Basolateral Mem-rane Vesicles Derived from the Ovine Parotid Acinar Cells

Intravesicularanion

(200 mosM)

Formate uptake(nmol z min21 zmg21 protein)

Acetate uptake(nmol z min21 zmg21 protein)

annitol 1.72 6 0.08 1.04 6 0.01luconate 2.21 6 0.37 1.13 6 0.01olygalacturonate NA 0.84 6 0.07icarbonate 9.68 6 0.22 15.76 6 1.64ormate 18.32 6 1.51 14.94 6 0.30cetate 17.76 6 0.42 36.52 6 1.44ropionate 16.12 6 0.89 32.76 6 3.61utyrate 22.00 6 0.32 NAyruvate 3.52 6 0.16 1.99 6 0.24actate 4.56 6 0.55 1.79 6 0.13hiocyanate 3.68 6 0.17 1.66 6 0.14rthophosphate 2.60 6 0.08 1.60 6 0.04ulfate 4.36 6 0.29 1.74 6 0.11

Note. Ovine parotid acinar basolateral membrane vesicles were pre-oaded with 100 mM mannitol and 200 mosM anions of potassium saltss described under Experimental. The reaction was initiated by theddition of 100 ml incubation medium containing 100 mM potassiumluconate, 20 mM Hepes/Tris, pH 7.4, 0.1 mM MgSO4, 0.02% (w/v)aN3), and either 1 mM formate or acetate to the vesicle suspension.he reaction was quenched with 1 ml of an ice-cold stop buffer. Valuesre presented as means 6 SEM for three assays. NA, not assayed.

mctTaewnmact

wTiTdiiithmerb

tration gradient to impede the entry of acetate. How-eabtmtpt

pmbgimas2antaatoabi

blmaiiiTatm

fcc2Vc0[ooV

Vol. 259, No. 3, 1999 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ate, acetate, or gluconate, internal pH 7.4. The vesi-les were incubated at 39°C for the indicated times inhe reaction medium described in the legend to Fig. 2.he radiolabel associated with the vesicles was thenssayed. It can be seen that there is a rapid initialntry of acetate into vesicles that are preloadedith acetate, formate or bicarbonate, 38, 28, 21,mol.min21.mg21 protein respectively. After approxi-ately 20 s the content of radiolabel peaked at levels

round 3 nmol.mg21 protein, within a minute it de-lined and after 5 min reached an equilibrium levelhat was approximately 1 nmol.mg21 protein.

There was no movement of acetate into vesicles thatere preloaded with gluconate, a non-permeant anion.hus, different intravesicular anions cause variations

n the rates, and the time courses of acetate uptake.his emphasizes that there is negligible non-specificiffusion of acetate across these membranes in eitheronized or protonated forms. In the acetate-loaded ves-cles there is a high concentration gradient of anion,nside to outside 100:1. If the acetate diffused acrosshe vesicle membrane as the protonated species theigh concentration of acetate inside the vesicles (100M) would tend to exit down its concentration gradi-

nt, the vesicles would then shrink and the content ofadiolabel decrease. In the cases of the formate andicarbonate loaded vesicles there was no such concen-

FIG. 1. Effect of osmolarity of the external medium on acetate orormate uptake by ovine parotid acinar basolateral-membrane vesi-les. Basolateral membrane vesicles were pre-loaded with a mediumontaining 100 mM potassium acetate or formate, 100 mM mannitol,0 mM Hepes/Tris, pH 7.4, 0.1 mM MgSO4 and 0.02% (w/v) NaN3.esicles were incubated for 30 min at 39°C in a reaction mediumontaining 100 mM potassium gluconate, 20 mM Mes/Tris, pH 5.5,.1 mM MgSO4, 0.02% (w/v) NaN3 and 1 mM [1-14C]acetate (F) or

14C]formate (‚) and mannitol in concentrations to give the indicatedsmolarity. The reaction was quenched after 30 min and the uptakesf acetate or formate were measured as described under Experimental.alues are presented as means 6 SEM for three assays.

608

ver, it can be seen from Fig. 2 that the time course ofcetate entry is similar for formate, acetate and bicar-onate loaded vesicles. These observations are difficulto reconcile with the nonfacilitated transmembraneovements of acetate, formate and bicarbonate as pro-

onated species (10). Similar results were noted in ex-eriments where the time course of formate entry intohe vesicles was investigated (data not shown).

Evidence for the presence of a K1 channel in the ovinearotid acinar cell basolateral membrane. The move-ent of formate into the intravesicular space mediated

y the combined action of a valinomycin-mediated K1-radient and an anion-channel was investigated. Ves-cles were loaded with an isotonic medium containing

annitol as the major osmoticum. These were added ton incubation medium that contained 100 mM potas-ium gluconate, 1 mM [14C]formate, 100 mM mannitol,0 mM Hepes/Tris, pH 7.4, 0.1 mM magnesium sulfatend 0.02% (w/v) sodium azide, with and without vali-omycin, 30 mg/mg protein. The uptake of formate intohe vesicles was followed over a period of 5 min. In thebsence of valinomycin, formate entered the vesicles atslow initial rate, 168 pmol formate.min21.mg21 pro-

ein. This rate was enhanced four fold by the presencef valinomycin. Thus, formate can enter the vesicles inn electrogenic manner, (electrical compensation muste achieved by the movement of a cation into the ves-cle since there are no anions in the lumen of the vesicle

FIG. 2. Time-course of acetate uptake into ovine parotid acinarasolateral membrane vesicles. Vesicles were pre-loaded with a so-ution containing 100 mM mannitol, 20 mM Hepes/Tris, pH 7.4, 0.1

M MgSO4, 0.02% (w/v) NaN3 and 100 mM potassium salts of eithercetate (ƒ), formate (X), bicarbonate (E), or gluconate (F). The ves-cles were incubated at 39°C for 2 min. Uptake of acetate wasnitiated by the addition of 100 ml of an incubation medium contain-ng 100 mM potassium gluconate, 100 mM mannitol, 20 mM Hepes/ris, pH 7.4, 0.1 mM MgSO4, 0.02% (w/v) NaN3 and 1 mM [1-14C]-cetate. At appropriate times, the reaction mixture was quenched byhe addition of 1 ml ice-cold stop-buffer. Results are presented aseans 6 SEM for four assays.

ttuowtca

swo[triteoh

mafbtrcc

eul

anions, each having different substrate specificitiesafaseuati

vwemtsMtbmsvumcpbmut

aw5tmt7fatasai

csisoi

sb

tl(0pasoarp

Vol. 259, No. 3, 1999 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

hat can exchange with the formate). However, even inhe presence of valinomycin, the measured rates ofptake of formate were too low to account for the ratesf entry of formate and acetate into vesicles pre-loadedith bicarbonate, formate, acetate, propionate and bu-

yrate (see Table 1). Consequently, we conclude thathannels are not involved in the movement of formate,nd acetate into these vesicles.

Effects of extravesicular pH on the rates of anion-timulated uptake of formate and acetate. Vesiclesere pre-loaded with either 100 mM potassium acetater 100 mM potassium formate. The uptake of1-14C]acetate or [14C]formate was measured over aime period of 10 s in a reaction medium with pHanging from 5.5 to 7.4. It can be seen from the resultsn Fig. 3 that acetate uptake is relatively insensitive tohe external pH when the internal anion is either ac-tate or formate. The profile is different for the uptakef formate. The initial rate of formate entry is en-anced with a decrease in the external pH.In this experiment the concentrations in the reactionedia of the protonated formate/acetate increases by

pproximately one hundred fold as the pH diminishesrom 7.4 to 5.5. The rate of uptake of formate increasesy a factor of 2.5 and there is no significant increase inhe rate of uptake of acetate. Clearly, there is no directelationship between the rates of anion uptake and theoncentration of the protonated anions. This is notompatible with a diffusion-limited mechanism.

The data in Fig. 3 are reproducible. One possiblexplanation for the differential sensitivity to pH of theptakes of acetate and formate is that there are at

east two proteins that facilitate the uptake of these

FIG. 3. Effect of extravesicular pH on acetate-stimulated ace-ate, and formate-stimulated formate uptakes. Vesicles were pre-oaded with a solution containing either 100 mM potassium acetateE, F) or -formate (‚, *), 100 mM mannitol, 20 Hepes/Tris, pH 7.4,.1 mM MgSO4 and 0.02% (w/v) NaN3. The vesicle suspension wasreincubated for 2 min at 39°C. The reaction was initiated by theddition of 100 ml of incubation medium containing 100 mM potas-ium gluconate, 100 mM mannitol, either 20 mM Mes/Tris, pH 5.5–6r 20 mM Hepes/Tris, pH 6.5–7.4, 0.1 mM MgSO4, 0.02% (w/v) NaN3

nd 1 mM [1-14C]acetate (‚, E) or [14C]formate (F, *). After 10 s theeaction was quenched with 1 ml of ice-cold stop buffer. Values areresented as means 6 SEM for four assays.

609

nd pH profiles. Another model is that the binding oformate to the external substrate site leaves an ioniz-ble group exposed. This group has a pK value thatenses the change in pH over the range used in thisxperiment; resulting in the differences in the rates ofptake which are illustrated in Fig. 3. The binding ofcetate to the protein renders this group inaccessible tohe external pH, and hence the relative pH insensitiv-ty of the movement of acetate into the vesicle.

Kinetics of acetate-stimulated acetate uptake. Theariations of the rates of acetate and formate uptakesere plotted as a function of the concentration of thexternal anion in the reaction medium, from 0.1 to 100M. The internal concentration of acetate was main-

ained constant at 100 mM. The uptake of acetate wasaturable, the system followed classical Michaelis–enten kinetics. We conclude that the uptakes of ace-

ate and formate are facilitated processes, not limitedy diffusion. The Km values for external acetate for-ate are 43 and 32 mM, respectively. When fully

aturated the acetate-uptake system has a maximumelocity of 2 mmol.min21.mg.21 protein. The formate-ptake system has a maximum velocity of 1.8mol.min21.mg.21 protein. These are high rates. Theyompare well with the Vmax 5 2.1 mmol.min21.mg.21

rotein reported by Harig (11) for the bicarbonate/utyrate “exchange” in vesicles from the apical plasmaembrane of human colonocytes. In contrast, the val-

es for the Km reported here are some tenfold higherhan that reported by Harig et al. (11).

The effects of inhibitors on anion-stimulated acetatend formate uptake. Pretreatment of the vesiclesith 1 mM N-ethylmaleimide, pH 7.4, resulted in a0% inhibition of the uptake of acetate; this indicateshat a thiolate anion may be involved in the uptakeechanism. The degree of inhibition was insensitive to

he external pH in the assay medium, either pH0 5.5 or.4. However, for the formate-stimulated uptake oformate the inhibition by 1 mM N-ethylmaleimide wasffected by the pH of the reaction medium, at pH 5.5he inhibition was negligible, but when the uptake wasssayed with pH0 7.4 an inhibition of 60% was ob-erved. The effect of 500 mM mercuric ions on acetatend formate uptake was also investigated, 80–90%nhibition was observed.

Pretreatment of the vesicles with 1 mM dicyclohexyl-arbodiimide resulted in a 55% inhibition of acetate-timulated uptake of acetate. This process was alsonhibited by 1 mM phenyl glyoxal, but the acetate-timulated uptake of formate was not affected. Thesebservations are not characteristic of a diffusion lim-ted process.

The effects of 0.5 mM DIDS and SITS on the acetate-timulated uptake of acetate were investigated. Incu-ation for up to 3 h at 39°C with 0.5 mM SITS had no

effect on the uptake activity; while 0.5 mM DIDSrtc

tp

D

tfpuvbtsmt

an(ftootnficua

o

here rule out the participation of a monovalent cation-d

tpaapscatppa

R

1

1

Vol. 259, No. 3, 1999 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

esulted in approximately 30% inhibition. The con-rol values during this incubation period remainedonstant.

These phenomena collectively support the conclusionhat the uptake reactions are the consequence of arotein mediated reaction(s).

ISCUSSION

We have shown that there is a rapid facilitated up-ake of both acetate and formate into vesicles isolatedrom the basolateral plasma membrane of the ovinearotid acinar cells. The extent and the velocity of theptake are modulated by the nature of the intra-esicular anion. The mechanisms involved are satura-le, they can be inhibited by reagents that are knowno modify the activity of proteins. These observationsuggest that the anion-dependent uptakes of both for-ate and acetate be mediated by a protein, located in

he basolateral membrane of the acinar cells.There are many reports of the movement of acetate

nd formate across bilayer membranes being due to theon facilitated diffusion of the protonated anions, e.g.10). However, we reject this mechanism as the basisor the observed uptakes reported here. The kinetics ofhe processes, the differential responses of the uptakesf acetate and formate to (i) pH see Fig. 3 (ii) the naturef the internal anions (Table 1 and Fig. 2), (iii) inhibi-ors (iv) the relationship between the pH of the exter-al medium and the rates of uptake of both acetate andormate, are all phenomena that can not be rational-zed in terms of a spontaneous, diffusion limited pro-ess. They collectively support the conclusion that theptakes of formate and acetate are the consequences ofprotein mediated mechanism.The requirement for an internal anion for the uptake

f acetate/formate and the effects of pH on reported

610

ependent mechanism.The most feasible mechanism for the uptake of ace-

ate and formate is via an anion exchange protein. Weropose that in vivo the complementary anion is met-bolically generated bicarbonate. This postulatedcetate-bicarbonate exchange system would have a ca-acity to enable acetate to fulfil its role as the majorource of energy for the maintenance of the secretaryells, and to provide the driving force for the flux ofnions through the acinar cells; from the systemic sys-em into the saliva. This is in marked contrast to thehysiological roles envisaged for most anion-exchangeroteins, i.e., to facilitate intracellular pH homeostasisnd to regulate changes in cell volume.

EFERENCES

1. Bergman, E. N., Reid, R. S., Murray, M. G., Brockway, J. M., andWhitelaw, F. G. (1965) Biochem. J. 97, 53–58.

2. Barcroft, J., McAnally, R., and Phillipson, A. T. (1944) J. Exp.Biol. 20, 120–129.

3. Bergman, E. N. (1990) Physiol. Rev. 70, 567–590.4. Blaxter, K. L. (1962) The Energy Metabolism of Ruminants,

Hutchinson & Co., London.5. Annison, E. F., and Lindsay, D. B. (1961) Biochem. J. 78, 777–

785.6. Essig, H. W., Norton, W., and Johnson, B. C. (1961) Proc. Soc.

Exp. Biol. Med. 108, 194–197.7. Vayro, S., Kemp, R., Beechey, R. B., and Shirazi-Beechey, S. P.

(1991) Biochem. J. 279, 843–848.8. Tarpey, P. S., Wood, I. S., Shirazi-Beechey, S. P., and Beechey,

R. B. (1995) Biochem. J. 312(Pt. 1), 293–300.9. Shirazi, S. P., Beechey, R. B., and Butterworth, P. J. (1981)

Biochem. J. 194, 797–802.0. Bakker, E. P., and Van Dam, K. (1974) Biochim. Biophys. Acta

339, 285–289.1. Harig, J. M., Ng, E. K., Dudeja, P. K., Brasitus, T. A., and

Ramaswamy, K. (1996) Am. J. Physiol. 34, G415–G422.