Gut, 1992,33,554-559

Evidence of hydrogen ion secretion from the humangall bladder in vitro

J N Plevris, P C Hayes, D J Harrison, I A D Bouchier

AbstractGail bladder bile is more acid that hepatic bileand this has been attributed to bicarbonateabsorption by the gall bladder epithelium. Theaim of this study was to investigate in vitro theacid base changes that occur across the humangall bladder mucosa. Fresh gall bladder tissuewas obtained at cholecystectomy and placed inan Ussing Chamber and perfused with Ringer-Krebs glucose bicarbonate solution. The via-bility of the gall bladder was assessed bymeasuring the potential differences across theepithelium and by the morphology of theepithelial cells at the end of the experiments.Aliquots from the solutions were taken at two,45 and 70 minutes and pCO2, hydrogen ion andbicarbonate concentrations were measured. Inthe mucosal side of the chamber a consistentand significant decrease was observed fromtwo minutes to 70 minutes in bicarbonateconcentration while pCO2 and hydrogen ionconcentrations significantly increased. Thedegree ofinflammation correlated weil with theability for acidification, the more inflamedthe tissue the less its ability to acidify. Whenthe gail bladder was exposed to amilorideor sodium free solution acidification wasabolished in the mucosal side. When tissuemetabolism was irreversibly inhibited byexposure to formaldehyde, hydrogen ionconcentration and pCO2 were significantlydecreased in the mucosal side of the chambercompared with the viable gail bladder. Thehuman gail bladder is capable of secreting acidand this may be an important mechanism forpreventing calcium precipitation and gail stoneformation.

Department ofMedicine, RoyalInfirmary of Edinburgh,and Department ofPathology, University ofEdinburgh, ScotlandJ N PlevrisP C HayesI A D Bouchier

Department ofPathologyD J HarrisonCorrespondence to:Dr J N Plevris,Department of Medicine,Royal Infirmary of Edinburgh,Lauriston Place, EdinburghEH8 9YW.Accepted for publication11 July 1991

Gall stone disease is a common cause ofmorbidity and mortality and cholocystectomyrepresents a significant percentage of majorupper abdominal operations in western society.Although considerable research has been con-centrated on the pathogenesis of gall stoneformation, this has often focused on the bio-chemical changes that occur in the bile. It is onlywithin the last 10 years that the importance of theevents in the gall bladder and in particular thepossible contributions of the gall bladder mucosato lithogenesis has been recognised.The gall bladder mucosa has one of the highest

rates of water absorption in the body and an 80to 90% decrease in the initial volume of bileis achieved within the gall bladder. This is

achieved by coupling of active sodium transportand passive water absorption resulting in iso-tonic fluid absorption. ' Fluid transport is subjectto physiological regulation and is higher in thedaytime.2 3 Electrolyte transport has been exten-sively investigated and chloride is activelyabsorbed in exchange for bicarbonate.4 Potas-sium moves passively from the mucosa to theserosa according to electrochemical gradients,5and calcium is also absorbed and its distributionacross the gall bladder epithelium is consistentwith passive transport.6 Its contribution to gallstone formation is increasingly recognised. Mostof the organic components of bile - namely, bileacids, lecithin, bilirubin, and cholesterol areabsorbed to a limited degree by the human gallbladder mucosa.578 There is more recentevidence that the human gall bladder has notonly absorptive but also secretory properties.Fluid absorption can be reversed to net secretionwith feeding in chronic inflammation or with theuse of pharmacological agents (prostaglandins,prostacyclin, gut hormones, etc).9 An increasingnumber of biliary proteins are known to besecreted from the gall bladder mucosal cells suchas mucin, immunoglobulins, etc.-'>2 In man aswell as in other species there is decline in the pHof the gall bladder bile compared with hepaticbile. Initially this was thought to be the result ofbicarbonate reabsorption by the gall bladdermucosa.' 5 Studies in rabbit, guinea pig, andnecturus gall bladders have produced evidencefor mucosal hydrogen secretion during sodiumreabsorption, probably representing a sodium/hydrogen exchange.'3-'7 In a more recent study ithas been suggested that the canine gall bladderhas the ability to secrete hydrogen ions in vivo. "There are, however, no studies on the humangall bladder mucosa dealing with the mech-anisms of bile acidification and it is not clearwhether this is because ofhydrogen ion secretionor bicarbonate reabsorption. Acid secretion bythe gall bladder has important implications forgall stone formation because the majority of gallstones contain calcium carbonate and changes inthe pH of the bile are of critical importance ininfluencing the calcium solubility in bile. '1The present study was undertaken to investi-

gate the role of the gall bladder epithelium in theacidification process of bile in vitro using freshviable human gall bladder mucosa obtained atcholecystectomy.

Methods

Part of this work has been presented at the British Society of Gastroenterologv Studies were done using fresh human gallAutumn Meeting, Sheffield 1988, and has been published as an Abstract inGut (October 1988). bladder mucosa obtained at the time of elective

554

on February 2, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.33.4.554 on 1 A

pril 1992. Dow

nloaded from

Evidence ofhydrogen ion secretionfrom the human gall bladder in vitro

cholecystectomy usually for gall stone disease.Patient's sex, age, and type of gall stones, ifany, were recorded. The gall bladder tissue wasretrieved within 15 minutes of the ligation of thecystic artery. A circular piece of the gall bladderwall (d=12 mm) was removed using a tissuepunch and placed in Ringer-Krebs glucosebicarbonate solution (0 9% NaCl (0-154 M),1-15% KC1 (0-154 M), 1-22% CaC12 (0-11 M),2-11% KH2PO4 (0154 M), 3-8% MgSO4. 7H20(0-154 M), 1-3% NaHCO3, 5 4% glucose (0 3M)) at 4 C and preoxygenated with 95% 02. 5%CO2. To minimise tissue hypoxia, this was doneas soon as the gall bladder was removed.The gall bladder tissue was transferred to the

laboratory and within five minutes the mucosawas stripped by blunt dissection from themuscular part of the wall, rinsed with Ringer-Krebs bicarbonate glucose (RKGB) solution toremove debris and bile and mounted in anUssing Chamber. The Ussing Chamber con-sisted of two 2 ml volume compartments each ofwhich communicated with a glass tube abovethrough two plastic tubes (inlet and outlet) (Fig1). The gall bladder tissue separated the twocompartments creating a mucosal compartmentat the mucosal site and a serosal compartment onthe opposite side. The term 'serosa' is usedthroughout to indicate the non-luminal surfaceof the mucous membrane. Both compartmentswere filled with 10 ml RKGB through the glasstubes (2 ml) in each compartment ofthe chamberand 8 ml at each glass tube). The mucosalcompartment was sealed while 95% 02/5% CO2(21/min) was bubbled through to the serosalcompartment only. Preliminary experimentshad shown that PO2 between mucosal and serosalcompartments did not differ significantly if95%02/5% CO2 was bubbled into the serosal com-partment only and this was preferred in order toavoid disruption of the 'unstirred water layer' inthe mucosal side.s0-' The whole system wascovered by a thermostatic water jacket to main-tain a constant temperature of 37°C throughout

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

Time (minutes)Figure 1: Potential difference across the human gall bladder mucosa; (A) represents a viablegall bladder with an initial potential difference of 7 mV. Oxygen deprivation or immersion informaldehyde resulted in a drop in the potential difference across the epithelium but recoveredwhen oxygen was reintroduced. (B) represents a gall bladder with an initial potential differenceofless than 2 mV. Although potential difference recovered after 15 minutes, oxygen deprivationresulted in a non-reversible fall in the potential difference. Gall bladders with initial potentialdifference <2 mV were excludedfromfurther study.

the experiment. A pair of silver/silver chloridematrix 1 mm electrodes (Clark ElectromedicalInstruments, Reading, Berks, UK) monitoredthe potential difference across the two sites ofthe tissue and used as an index of viability of thetissue. Each study lasted for 70 minutes. Onemillilitre of the solution was removed through amicropipette from each compartment of theUssing Chamber at two minutes, 45 minutes and70 minutes and was immediately analysed in an1302 pH/blood gas analyser (InstrumentationLaboratory System, Lexington, MA, USA) forPO2, pCO2, hydrogen ion ([H+]), and bicarbon-ate ([HCO3-I) concentrations. Gall bladderswhich were macroscopically grossly distortedor damaged were excluded from furtherstudy.

EXPERIMENTSForty gall bladders were studied. The experi-ments were divided into three groups.

First groupFive gall bladders were studied. The effect ofoxygen deprivation and immersion in formalde-hyde 4% on the resting potential difference wasobserved. 95% 02/5% CO2 was bubbled at asteady flow rate of 21/min from time 0 to 20minutes, then stopped for 10 minutes from time20 to 30 minutes and subsequently continued atthe same flow rate until the end of the experi-ment. Thereafter the gall bladder mucosa wasexposed to 4% formaldehyde for two minutes,rinsed with RKGB solution and remounted tothe Ussing Chamber and studied for further10 minutes. The potential differences weremonitored continuously throughout the experi-ment.

Second groupTwenty five gall bladders were studied. Twentyone were processed according to the standardprotocol to maintain viability and four wereimmersed in formaldehyde 4% for two minutesbefore being mounted on the Ussing Chamber.The hydrogen ion concentration, bicarbonateconcentration pCO2 and P02 were measured atthe beginning (two minutes) at 45 minutes andthe end of the experiments (70 minutes). Theresting transepithelial potential difference wasmonitored throughout. The aim was to study theacidification capacity of the viable gall bladdercompared with non-viable tissue.

Third groupTen gall bladders were studied. In the first setof experiments involving six gall bladders themucosal bathing solution was replaced bysodium free Krebs-Ringer solution (NaCl andNaHCO3 removed) and the analytes mentionedabove were measured. In a second set ofexperiments involving four gall bladders 2 mMamiloride (a specific Na+/H+ inhibitor) wasintroduced into the mucosal compartment andits effect on acidification was observed after 45minutes.

E

U,00._

(U0a)cn

555

on February 2, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.33.4.554 on 1 A

pril 1992. Dow

nloaded from

Plevris, Hayes, Harrison, Bouchier

A *7-19'O[ T

0Ecc0+-

m

c

0

0._

U

00

I

0C0

016'a

-I

5

0

-5

-101

-11-3*-151

0

EcC

0

~~~~~~a)

CB

.m-71

(0

0

.0

(0

C.)

554-l

B

0

-1

1-33

*0-74

TL'aI

EN0C,0.

| MucosalSerosal

C

05r-

0

-05 F

- 1

0-34

-0-49*

-1S5F -1-45**

_Z-} - l- |}ZEViable Non-viable Viable Non-viable Viable Non-viable

gall bladder gall bladder gall bladder gall bladder gall bladder gall bladder

Figure 2: Differences in hydrogen ion concentration (A), bicarbonate concentration (B), andpCO2 (C) between 70 and two minutes in the mucosal and serosal sideofthe viable and non-viable gall bladders (*p<0 05, +p<0-03, + +p<0-02, **p<0.01).

PATHOLOGYA small sample of the tissue under study wasfixed in 10% buffered formalin before eachUssing Chamber experiment. The rest of thetissue was also fixed in 10% buffered formalin atthe end of the experiment. An experiencedhistopathologist examined each gall bladderblindly on two occasions and reported on thedegree of cholecystitis and the viability ofthe tissue according to cell morphology at thebeginning and end of the experiments. Thedegree of cholecystitis was graded from 1 (mild)to 3 (severe) according to the appearance of themucosa, muscle layer thickness, presence ofRokitansky-Aschoff sinuses and degree of in-flammatory process.22 The morphology changeswere recorded as 0 (healthy looking cells), 1(mild morphological changes such as celloedema, presence of granules, vacuolation), 2(moderate but without evidence of cell necrosis).The gall bladders used in the above experi-

ments were selected according to the followingcriteria: (a) satisfactory macroscopic appearancewith no obvious ulceration or fibrosis at thebeginning of the study and no obvious damageduring the study; (b) expression of a restingpotential difference more than 2 mV of serosapositive; (c) data from mucosa with severechanges in cell morphology were excluded fromfurther analysis.

STATISTICAL ANALYSISData are expressed as mean standard error ofmean (SEM). The data were not normally distri-

14

12

10

8

6

4

2

0

T

T

uraue Z (n = 6)

TGrade 3 (n = 6)

Degree of cholecystitis

buted and were analysed using non-parametrictests; Mann-Whitney test was used for unpaireddata and Wilcoxon's test for paired data. A pvalue of 0-05 was taken as significant.

ResultsIn the first set of experiments the gall bladders(five) documented a mean potential difference of75± 3 5 mV serosa positive after five minutes inthe chamber. This potential difference remainedquite stable throughout the ensuing 15 minutes.There was a significant drop in the potentialdifference when 95% oxygen 5% CO2 was dis-continued, but when reapplied the potentialdifference recovered to levels similar to thosebefore oxygen deprivation indicating thathypoxia had a direct effect on reducing theresting potential difference. When those gallbladders were immersed in formaldehyde 4%and remounted on the Ussing chamber, how-ever, there was a dramatic and irreversible dropin the potential difference indicating non-viability of the tissue (Fig 1). The P02 wasmeasured in both compartments at two minutes,45 minutes, and 70 minutes. Although 95% 02/5% CO2 was bubbled in the serosal compartmentonly, there was no difference in the P02 betweenthe mucosal and serosal compartment (25-3 (2-1)pKa v 23-5 (2-2) pKa respectively) which showsthat the diffusion of oxygen through the gallbladder mucosa was adequate. It was thoughttherefore unnecessary to oxygenate the mucosalcompartment directly, as this would disrupt the'unstirred water layer'. All gall bladders showedevidence of chronic cholecystitis (three grade 1,two grade 2) without any significant changes incell morphology at the beginning and end of theexperiments.

In the second set of experiments 25 gallbladders were studied (18 women, seven men;13 contained cholesterol stones, nine pigmentstones, three biliary sludge). Four of those wereimmersed in formaldehyde and used as controlsand from the remaining 21, six showed evidenceof grade 3 chronic cholecystitis and were notincluded in the analysis outlined below. Therewas a significant increase in the hydrogen ionconcentration from two minutes to 45 minutes

Figure 3: Differences inhydrogen ion secretion withhistology (grade I (mild) tograde 3 (severe) cholecystitis)*p<0.01C0

4-C.)-

o'.- E0

~0I

-7n, . -gi I 17

556

2 r-

.6

*

on February 2, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.33.4.554 on 1 A

pril 1992. Dow

nloaded from

Evidence ofhydrogen ion secretionfrom the human gall bladder in vitro

Figure 4: Differences inhydrogen ion secretion withcell morphology changes(grade 0: healthy cells, grade1: mild morphologicalchanges suggesting 'sickcells', grade 2: moderatechanges without evidence ofnecrosis) *p<001.c

0

a)

L-

c E0 CC,-0

.1

and to 70 minutes observed in the mucosalcompartment (p<001) while in the serosal com-partment there was a significant decrease inhydrogen concentration (p<0c05). By contrastin the non-viable gall bladder (those immersed informaldehyde) there was a significant drop inhydrogen concentration on the mucosal sitewithout any significant change of hydrogen con-centration in the serosal side (Fig 2a). In theviable gall bladder bicarbonate concentrationsignificantly decreased in the mucosal compart-ment from two minutes to 45 and 70 minutes(p<002) while on the serosal side there was asignificant increase of bicarbonate concentrationfrom two minutes to 70 minutes. In the non-viable gall bladder an increase of bicarbonateconcentration was seen in both compartments(Fig 2b). In the viable gall bladder there was asignificant increase in pCO2 between two and 70minutes on the mucosal site (p<003) while inthe serosal compartment a significant decrease inPCO2 was observed (p<005). In the non-viablegall bladder there was a drop in pCO2 in bothcompartments being significant in the mucosalside (p<001) (Fig 2c).In the viable gall bladder the mean potential

difference did not significantly change through-out the experiments. There was, however, asignificant difference in the potential differencebetween the viable and the non-viable gallbladders studied (6&3 (2 6) to 1L3 (0-9), p<0.01)at five minutes. The potential difference of thenon-viable gall bladders rapidly declined to 0.When the acidification ability of the 21 gallbladders was plotted against the degree ofchronic cholecystitis there was a progressive andstatistically significant decrease of hydrogen ionsecretion from grade 1 (mild) to grade 3 (severe)cholecystitis (Fig 3).The same was true when hydrogen ion secre-

tion was plotted against the degree of cellmorphology changes graded from 0 to 2 indicat-ing that gall bladder epithelia with impairedviability had a tendency to secrete less hydrogenions (Fig 4).There were no differences in the acidification

capacity of the gall bladders with sex or type ofgall stones, although epithelia from gall bladderswith pigment stones had a higher ability toabsorb bicarbonate (A[HCO3]=-1 68 (0 49)mmol/l (pigment) v 3-8 (0 74) mmol/l (choles-terol), p<0 01).

In the third set of experiments (10) RKGBsolution was replaced with sodium free isotonicsolution in the mucosal side. No significant

14 F-

12

10

8

6

4

2

0

T

T T

Grade 0 (n = 5) Grade 1 (n = 7) Grade 2 (n = 3)

Degree of cell morphology changes

075

0 025

-1 _-0-7 __

I)A [H+] A [Hco03- A lPCo2I(nmol/l) (mmol/l) (mm Hg)

Amiloride 2 (mmol/l) (mucosal side) (n = 4)Figure 5: Effect ofmucosal application ofamiloride (2 mM)on hydrogen ion secretion.

differences in the [H+], [HCO3 ] and pCO2were observed in any site of the Ussing chamber,between two and 70 minutes of the experiment.Similarly, the use of amiloride (2 mM) in themucosal side abolished acidification (Fig 5).

DiscussionThe Ussing chamber method has been adaptedand used by many investigators since its intro-duction by Ussing and Zerhan in 1951.23 It is avalid method of keeping a biological preparationviable during the period of investigation and hasbeen used to study electrophysiological, secre-tory, and absorptive properties of tissues invitro. Tissues previously used include frog skin,intestine, bladder, and animal and human gallbladder.

For years it has been known that gall bladderbile is more acid compared with hepatic andthe difference in the pH has been attributed tobicarbonate absorption.' 5 In this study we haveshown that fresh viable human gall bladdermucosa is capable of acidifying physiologicalsolutions in vitro. This acidification is a functionof viable tissue; it was reduced in more inflam-med gall bladders or when the epithelial cellsbecame 'sick' during the study and was lost whenthe mucosa was non-viable. Acidification wasabolished when the gall bladder epithelium wasexposed to. sodium free solution or in the pres-ence of high concentration of amiloride in themucosal compartment. In our study wherehuman gall bladders were used to investigateacid secretion in vitro, the gall bladder epith-elium appeared capable of increasing thehydrogen concentration in mucosal side withsimultaneous decrease of hydrogen concentra-tion in the serosal side which suggests thathydrogen ions were transferred from the serosalto mucosal side of the tissue. The concomminantdecrease of bicarbonate concentration withincreased pCO2 in the mucosal side indicates thatthis acidification is not simply the result ofbicarbonate reabsorption. It appears thathydrogen ions produced by the mucosal cellsreact with bicarbonate to form Co2 and waterwith a decrease in bicarbonate ions. Theobserved differences in pCO2 between the viableand non-viable gall bladders could be explainedon the basis that a non-viable gall bladdermucosa loses its ability to secrete acid thereforeless hydrogen ions are available to react withbicarbonate and also that the production of CO2from the mucosal cell metabolism is depressed.

Studies on gall bladders of other animal

557

2r 12

111

1

-9

on February 2, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.33.4.554 on 1 A

pril 1992. Dow

nloaded from

Plevris, Hayes, Ham'son, Bouchier

species such as rabbit, guinea pig, necturus, anddog have shown that these tissues are capable ofacidifying the mucosal solutions in vitro andthere is evidence that a sodium/hydrogen anti-port is present at the apical site of the epithelialcell. In our experiments the use of sodium freesolutions and amiloride, which is a specificsodium/hydrogen inhibitor, abolished acidifica-tion which suggests that hydrogen secretion inthe human gall bladder depends upon a sodium/hydrogen antiport.A potential problem, with Ussing chamber

studies is related to the viability of the tissuereceived at cholecystectomy. The ligation ofcystic artery which is part of the operativeprocedure usually takes place between five to 20minutes before the removal of the gall bladder.The effect ofhypoxia during this period has beenstudied24 and about 70% of the gall bladdersexamined immediately after the operationshowed evidence of mitochondrial change whichwas attributed to anoxia and mechanical damage.The human as well as animal gall bladder,however, is a durable organ which can rapidlyrecover from hypoxia and can be preservedviable for in vitro experiments for up to fourhours. ' The gall bladder has been used for manyyears as a model to study ion transport across theepithelia. In this study all the gall bladdersshowed a transmural potential difference ofmorethan 2-3 mV (range 2-3-11-9 mV) serosa posi-tive. This potential difference remained stablethroughout the period of experiment and we, aswell as other workers,25 27 were able to show thatanoxia resulted in a drop at the transmuralpotential diffference reflecting a reducingviability of the tissue which was reversible if theanoxic period was less than 10 minutes. Similarresults were produced after immersion informaldehyde with a permanent drop of thepotential difference indicating loss of viability.The continuous transpotential difference moni-toring therefore has been used as a reliablemethod of monitoring the viability of the tissueand allow further study of the mucosal function.The transmural potential difference of the gallbladders under study was in accordance with theliterature. Ross and others25 carried out studieson the electrical properties of the human gallbladder and on 46 gall bladders the mean trans-potential difference was 7-6 mV while otherinvestigators26 later showed that inflamed gallbladders have a lower resting transpotentialdifference which is dependent upon the degree ofinflammation, the variation in time between theligation of cystic artery, and the transfer to thelaboratory and the methodology used for study-ing the electrical properties of the tissue. Forinstance, tissues which are clamped at the edgesas in the Ussing chamber technique suffer fromedge damage which reduces the transpotentialdifference.27

Another potential problem with physiologicalstudies using tissues from routine cholocystec-tomy is that most are not histologically normal.Although it is not appropriate to extrapolate tonormal gall bladders in respect of hydrogen ionsecretion, the fact that we could show changes inthe acidification capacity with histology, themore inflamed gall bladders being less capable

to secrete acid, implies that hydrogen ion secre-tion is a function of the normal human gallbladder mucosa. The acidification capacity wasimpaired when the cell morphology wasabnormal and was abolished in the non-viablegall bladder; these suggest that hydrogen ionsecretion is a function of a viable tissue. There isrecent supportive evidence from animal andhuman studies that the normal gall bladdersecretes acid. Moore et al showed that the caninegall bladder is capable of secreting hydrogenions. I Preliminary in vivo studies from the samegroup have shown that the diseased human gallbladder is associated with decreased acid out-put.2829 Their conclusions, however, were notbased on studies in human gall bladder mucosadirectly, but were inferred from biochemicalanalysis of gall bladder biles from laparotomy.Our data would support the hypothesis by

Moore and colleagues that reduced gall bladderhydrogen ion secretion is associated with gallstone formation. It is postulated that theobserved pH changes in our experiments withdiseased gall bladders represent only a fractionof the capacity which the normal human gallbladder epithelium might possess to secrete acid.More studies on fresh normal tissue are requiredto test this suggestion.

It is now accepted that the sequence of eventsin the process of gall stone formation is super-saturation of bile, nucleation, precipitation, andsubsequent growth to stone sized aggregates.30Supersaturation of bile with cholesterol ispresent in the vast majority of patients with gallstone disease3"; but 40 to 80% of normal indi-viduals may have supersaturated bile withoutever forming gall stones.3233 Therefore theprocess of nucleation is important and thisdepends upon the presence of certain nucleatingagents or the absence of the naturally occuringinhibitors of crystal formation.Calcium bilirubinate or mucous glucoproteins

could serve as nucleating factors34 while a bileprotein which is a cholesterol crystal formationinhibitor has been proposed by Holzbach as agall stone formation protective protein.35Calcium has long been implicated in the patho-genesis of cholesterol and pigment gall stones.Pigmented gall stones are predominantly com-posed of the calcium salts of carbonate,bilirubinate, phosphate, and long chain fattyacids and to a lesser extent carbonate.3638Calcium carbonate also precipitates on to thesurface of the cholesterol gall stones and ispresent in most cholesterol gall stones.8 Theregulation of calcium concentration in the gallbladder bile is therefore of critical importance.It is postulated that acid secretion may bebiologically important because a reduction in thepH of the gall bladder bile effectively lowers thebicarbonate and may reduce the risk of insolublecarbonate formation. As a result, the concentra-tion of ionised calcium is increased in bile. Thegall bladder epithelium, however, has the abilityto absorb calcium and can reduce its concentra-tion in bile by more than 50%8; bile acids alsobuffer the remaining ionised calcium. Conse-quently less calcium is available to forminsoluble salts.39The mechanism of acid secretion also merits

558

on February 2, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.33.4.554 on 1 A

pril 1992. Dow

nloaded from

Evidence ofhydrogen ion secretionfrom the human gall bladder in vitro 559

further investigation. In this study we haveshown that it is sodium dependant, therefore it isclosely linked with the concentrating ability ofthe gall bladder because it is known that water ispassively absorbed consequent upon the absorp-tion of sodium from the epithelial cell. It is,however, necessary to identify other factorswhich may influence acid secretion and furtherstudies are under way to investigate this.

In conclusion we have shown in this study thatthe viable human gall bladder is capable ofsecreting acid. This acid secretion probablyoccurs through an apical Na+/H+ exchange atthe mucosal site of the gall bladder epithelial cell.This may represent a protective mechanismagainst calcium precipitation. Further studiesare needed to investigate more fully the processand identify those factors which regulate acidsecretion.We would like to thank Dr P W Flatman, Dept Physiology foradvising on the Ussing Chamber method, Mr B White for histechnical assistance and the surgeons and staff of 9/10 and 7/8operating theatres, RIE for providing the gall bladders. We wouldalso like to thank Mrs V Campbell for typing this manuscript. DrJ N Plevris is a recipient of William Gibson Fellowship, Faculty ofMedicine, University of Edinburgh. This study was partlysupported by a grant from the Lothian Health Board.

1 Diamond JM. Cook CF, ed. Transport mechanisms in thegallbladder. Handbook of physiology: alimentary canal.Washington DC: American Physiological Society 1968:245 1-82.

2 Diamond J. The mechanism of isotonic water transport. ] GenPhysiol 1964; 48: 15-42.

3 Diamond J. Transport of salt and water in rabbit and guineapig gall bladder.J Gen Physiol 1964; 48: 1-14.

4 Heintze K, Petersen KU, Olles P, Saverymuttu SH, Wood JR.Effects of bicarbonate on fluid and electrolyte transport bythe guinea pig gallbladder: a bicarbonate-chloride exchange.J Memor Biol 1979; 45: 43-59.

5 Rose RC. Absorptive functions of the gallbladder. In: JohnsonLR, ed. Physiology of the gastrointestinal tract. Vol 2. NewYork: Raven Press, 1981; 1021-2.

6 Rege RV, Nahrwold DL, Moore EW. Absorption of biliarycalcium from the canine gallbladder; protection against theformation of calcium-containing gallstones. ] Lab Clin Med1987; 110: 381-6.

7 Jacyna MR, Ross PE, Bakar MA, Hopwood D, Bouchier IAD.Characteristics of cholesterol absorption by human gall-bladder: relevance to cholesterolosis.J Clin Pathol 1987; 40:524-9.

8 Conter RL, Roslyn JL, Porter-Fink V, DenBesten L. Gall-bladder absorption increases during early cholesterolgallstone formation. AmJ Surg 1986; 151: 184-91.

9 Wood JR, Svanvik J. Gallbladder water and electrolytetransport and its regulation. Gut 1983; 24: 579-93.

10 Wahlin T, Thornell E, Jivegard L, Svanvik J. Effects of intra-luminal prostaglandin E2 in vivo on secretory behaviour andultrastructural changes in mouse gallbladder epithelium.Gastroenterology 1988; 95: 1632-5.

11 Lee SP, LaMont JT, Carey MC. Role of gallbladder mucushypersecretion in the evolution of cholesterol gallstones.J Clin Invest 1981; 67: 1712-23.

12 Reuben A. Biliary proteins. Hepatology 1984; 4: 465-505.13 Whitlock RT, Wheeler HO. Hydrogen ion transport by

isolated rabbit gallbladder. Am]J Physiol 1969; 217: 310-6.

14 Cremiaschi D, Henin S, Meyer G. Stimulation by HCO3 ofNa+ transport in rabbit gallbladder. jMembrBiol 1979; 47:145-70.

15 Heintz K, Petersen KU, Wood JR. Effects of bicarbonate onfluid and electrolyte transport by guinea pig and rabbitgallbladder: stimulation of absorption. 7 Membr Biol 1981;62: 175-81.

16 Weinman SA, Reuss L. Na+/H+ exchange at the apicalmembrane of the necturus gallbladder. J Gen Phvsiol 1982;80: 299-319.

17 Altenberg G, Reuss L. Apical membrane Na+/H + exchangein necturus gallbladder epithelium. ] Gen Phvsiol 1990; 95:369-92.

18 Rege RV, Moore EW. Evidence for H+ secretion by the invivo canine gallbladder. Gastroenterology 1987; 92: 281-9.

19 Rege RV, Moore EW. Pathogenesis of calcium-containinggallstones. ] Clin Invest 1986; 77: 21-6.

20 Diamond JM. A rapid method for determining voltage con-centration relations across membranes. j Phvsiol 1965;183(1): 83-100.

21 DiamondJM. Non-linearosmosis.] Phvsiol 1965;183: 58-82.22 Symmers W S-C. Systemic pathology. Vol 3. 2nd Ed.

Edinburgh: Churchill Livingstone, 1978: 1304-30.23 Ussing HH, Zerhan K. Active transport of sodium as the

source of electric current in the short-circuited isolated frog-skin. Acta Physiol Scand 1951; 23: 110-27.

24 Hopwood D, Kouroumalis E, Milne G, Bouchier IAD.Cholecystitis. A fine structural analysis. jPathol 1980; 130:1-13.

25 Ross RC, Gelarden RT, Nahrwold DL. Electrical propertiesof isolates human gallbladder. Am ] Phvsiol 1973; 224:1320-6.

26 Nahrwold DL, Ross RC, Ward SP. Abnormalities in gall-bladder morphology and function in patients withcholelithiasis. Ann Surg 1976; 184: 415-21.

27 Jacyna MR. Aspects of mucosal tunction in the human gall-bladder. 1986: 108-10 (Thesis). University of Dundee,Scotland.

28 Moore EW, Shiffman ML. Evidence of H+ ion secretion bythe human gallbladder in vivo. Gastroenterologv 1988; 95:A573.

29 Shiffman ML, Moore EW. Defective acidification leads toCaCo3 supersaturation of gallbladder bile in patients with alltypes of gallstones. Gastroenterology 1988; 94: A591.

30 Smith BF, Lamont JT. The sequence of events in gallstoneformation. Lab Invest 1987; 56: 125-6.

31 Admirand WH, Small DM. The physiochemical basis ofcholesterol gallstone formation in man. ] Clin Invest 1968;47: 1043-52.

32 Holzbach RT, Marsh M, Olsweski M, Holan K. Cholesterolsolubility in bile. Evidence that supersaturated bile isfrequent in healthy man. ] Clin Invest 1973; 52: 1467-79.

33 Hofman AF, Grundy SM, Lachin JM, Lan SP, Baum RA,Hanson RF, et al. Pretreatment biliary lipid composition inwhite patients with radiolucent gallstones in the NationalCooperative Gallstone Study. Gastroenterology 1982; 83:738-52.

34 Burnstein MJ, Ison RG, Petrunka CN, Taylor RD, StrasbergSM. Evidence of a potent nucleating factor in the gallbladderbile of patients with cholesterol gallstones. Gastroenterology1983; 85: 801-7.

35 Holzbach RT, Kibe A, Thiel E, Howell JH, Marsh M,Hermann RE. Bilieary proteins. Unique inhibitors ofcholesterol crystal nucleation in human gallbladder bile.] Clin Invest 1984; 73: 35-45.

36 Sutor DJ, Wilkie LI. Calcium in bile and calcium salts ingallstones. Clin Chim Acta 1977; 79: 119-27.

37 Been JM, Bills PM, Lewis D. Microstructures of gallstones.Gastroenterologv 1979; 76: 548-55.

38 Wosiewitz U. Limy bile and radiopaque, calcified gallstones: acombined analystical, radiographic and micromorphologicexamination. Pathol Res Pract 1980; 167: 273-86.

39 Moore EW, Celic L, Ostrow JD. Interactions between ionisedcalcium and sodium Taurochelate: bile salts are importantbuffers for prevention of calcium containing gallstones.Gastroenterology 1982; 83: 1079-89.

on February 2, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.33.4.554 on 1 A

pril 1992. Dow

nloaded from