Mechanisms of histamine stimulated secretion in

Transcription

1 Gut, 1981, 22, Mehanisms of histamine stimulated seretion in rabbit ileal muosa* B D LNAKER,t J S MKAY,+ N B HGGS, AND L A TURNBERG From the Department of Mediine, Hope Hospital (University of Manhester Shool of Mediine), Salford, Lanashire SUMMARY Histamine is present in high onentrations in the intestine and we investigated the possibility that it might have a role here in intestinal transport. When added to the basal side of rabbit ileal muosa in vitro histamine (10-4M) indued a short-lived inrease in eletrial potential differene and short iruit urrent. t inhibited net hloride absorption but did not influene sodium transport. Alkali seretion, measured by a ph stat tehnique, was inhibited, suggesting that biarbonate seretion was redued. Both the eletrial and ion flux responses to histamine were bloked by the H 1 reeptor bloker diphenhydramine, but not by the H2 reeptor bloker imetidine. The presene of speifi H, histamine reeptors was further supported by shifts in the dose-response urve to histamine by four different onentrations of diphenhydramine. Calulation of a pa2 value from these 'Shild' plots provided a figure of 7 85, whih is similar to that for H1 reeptors in other tissues. Aminoguanidine, a histaminase bloker, had no eletrial effets alone but shifted the histamine dose response urve to the left. These studies indiate that histamine inhibits hloride absorption and alkali seretion, possibly by influening a hloride/biarbonate exhange proess, through speifi muosal H1 reeptors. Enhanement of histamine effets by a histaminase inhibitor suggests that histaminases are present in the intestinal muosa and supports the possibility of a role for endogenous histamine in influening ion transport. The observations indiate a mehanism by whih absorption might be impaired in diseases in whih histamine is liberated loally in the intestine. The intestinal muosa ontains a rapidly turning over pool of histamine whih appears to be derived largely from non-mast ell soures. -3 ts physiologial role here, however, is unertain, although in systemi mastoytosis, in whih histamine is liberated loally in high onentration, diarrhoea is a feature.3 t is possible therefore that histamine may at as an intestinal seretogogue and this possibility was examined in rabbit ileal muosa in vitro. *Part of this work was presented at a meeting of the European Soiety for Clinial nvestigation (Marh 1979) and the British Soiety of Gastroenterology (September 1979). tpresent address: Department of Mediine, Warrington General Hospital, Warrington WA5 QG. tpresent address: Department of Gastroenterology, Royal Liverpool Hospital, Liverpool L69 3BX. Address for orrespondene: Professor LA Turnberg, Department of Mediine, Hope Hospital (University of Manhester Shool of Mediine), Salford M6 8HD, Lanashire, UK. Reeived for publiation 5 May Methods Using an in vitro tehnique previously desribed in detail,5 segments of stripped ileal muosa were mounted in Perspex flux hambers. Tissues were bathed on both sides in buffer maintained at 37 C and of omposition Na 146, K 4 2, Cl 125 8, HCO3 266, H2,PO4 02 HPO4 1 2, Ca 1 2, Mg 1 2, gluose 10 mmol/l, and ph 7 4. The buffer was stirred and oxygenated by a 95% oxygen 5% CO., gas lift system. n the experiments using a biarbonate-free buffer, hloride was substituted for biarbonate and 100% O., was bubbled through the buffer. Transmuosal potential differene was measured via saturated KC in agar bridges and alomel eletrodes using a high impedane digital voltmeter. Sodium hloride in agar bridges to silver/silver hloride eletrodes were used to transmit the short-iruit urrent. This was adjusted initially eah minute and later every five minutes. Experiments onduted later were performed using

2 Histamine and ion tiranspo 9t an automati voltage lamp apparatus, whih ontinuously adjusted the short-iruit urrent. Corretions were made for the fluid gap resistane as desribed by Field et al.6 Tissue resistane (R) was alulated from the short-iruit urrent and potential differene and expressed in Ohms per m2. Short-iruit urrent was onverted to net ion flux in pmol/m-2/h by multipliation by a fator of 3 6 x derived from where A equals area AF of exposed tissue and F is Faraday's onstant. on fluxes were determined in tissue pairs obtained from the same animal and with resistanes varying by less than 250%. Usually eight tissues were set up at the same time. Steady state fluxes were usually obtained 15 minutes after addition of isotopes and thus flux measurements were started 30 minutes after isotope addition to ensure equilibration. Fluxes were measured during the 15 minute period before addition of drugs (ontrol flux) and for 15 to 30 minutes starting 15 minutes after addition of drugs when equilibrium onditions had been reestablished. Control fluxes were also measured in some experiments during four onseutive 15 minute periods. Unidiretional fluxes from muosa to serosa and from serosa to muosa were measured in eah tissue pair after the addition of 05pCi22Na and 2 5pCi36C to the muosal side of one tissue and the serosal side of its paired tissue. One millilitre samples were taken from muosal and serosal solutions and replaed by ml warm unlabelled buffer. Calulation of fluxes was performed as desribed previously5 and fluxes were determined under short-iruit onditions with minimal interruption to read the spontaneous potential differene. With the use of a ph stat tehnique7 net alkali seretion was determined on the muosal side of the tissue bathed in the biarbonate free buffer and bubbled with 100o% oxygen. The serosal bathing solution was a biarbonate buffer bubbled with 95% oxygen 5% CO.,. The muosal solution was maintained at ph of 7 4 by ontinuous titration with 5pmol/l hydrohlori aid titrant. This was arried out with an Autotitrator l and Autoburette ABU 13 (Radiometer, Cophenhagen) whih very aurately deliver the small volumes of hydrohlori aid needed to maintain the ph onstant as measured by ombined eletrodes GK2321C and GK2332C. The volume of titrant needed to maintain the ph onstant during a 30 minute ontrol period and one hour after addition of histamine was reorded at five minute intervals. From the amount of aid added it is possible to alulate the rate of alkali (presumably biarbonate) sereted. Potential differene and short-iruit urrent were also measured during this experiment. An attempt was made to identify speifi histamine reeptors by studying the effet, on a histamine dose-response urve, of four different onentrations of the speifi H, reeptor antagonist diphenhydramine and alulating a pa., value from these dose response urves" 9. pax values, a measure of drug antagonism, are an aepted basis for reeptor lassifiation based on the hypothesis that agonists and antagonists ompete for reeptors aording to mass law'0. The same pax values should be obtained with similar reeptors in different tissues with the same agonist/antagonist ombination9. pax is defined as the negative logarithm of the molar onentration of the antagonist whih will redue the effet of a multiple dose of the agonist (x) to that of a single dose. More simp!y, pa., is the negative logarithm of the molar onentration of antagonist whih halves the sensitivity of the preparation to the agonist". These values are dependent on ontat time between antagonist and tissue8. A fixed lime of 14 minutes was taken in these experiments to allow omparison with published data on pa., in other tissues. For dose response and agonist/antagonist interation studies only one dose of agonist or antagonist was used in any single piee of muosa to avoid persistene of effets and the possibility of tahyphylaxis. Aminoguanidine, a speifi inhibitor of the enzyme responsible for histamine atabolism, was used to determine whether histaminase was present in intestinal muosa. Tissues were pretreated with 10-4M aminoguanidine (a dose whih ompletely inhibits histaminase") and histamine dose response, using maximal perentage potential differene inreases, were determined and ompared with results from ontrol tissues. Statistial analysis was made using Student's t test for paired and unpaired samples and variability is expressed as mean ; one standard error of the mean. Materials used were histamine aid phosphate (Sigma Chemial Corporation), diphenhydramine (Park Davies Ltd), imetidine (Smith, Kline & Frenh Ltd), and aminoguanidine (Sigma Chemial Corporation). Results 965 Histamine added to the aerosol side of ileal muosa signifiantly inreased potential differene, shortiruit urrent, and tissue resistane, the hanges in potential differene and short-iruit urrent being maximal at two minutes post histamine (Fig. l). Potential differene returned to ontrol values within 10 minutes but resistane remained raised for

3 Tirme minutes Fig. Changes in potential differene, short-iruit u rrent, and resistane of ileal muosa afier histamine 10-4M added to rabbit ileal muosa bathed in gluose buiffer (mean ± SEM shown). *Signifiantly liffierent fiom ontrol response (p < 0001). 50- *-E._ - a.r_ * a 0.*h 30 - o u a. x 10 - o -0 E 0a 0 2: a. -10 o-----o Serosal (n=6) x-x Muosal (n=6) -7 id n Conentration of histaimine (molar) Linaker, MKay, Higgs, and Turnbelg at least 30 minultes after histamine was added. The eletrial response was reproduible and doserelated. A dose response urve (Fig. 2) demonstrated that the maximal response was ahieved at a onentration of 10-4M and the smallest dose at whih a signifiant response was noted was 10-5M. Histamine in onentrations greater than 10-3M failed to inrease potential differene or shortiruit urrent. Histamine added to the muosal side of tissues failed to produe an eletrial response (Fig. 2). The H2 reeptor bloker imetidine in onentrations between 10-2 and 10-4M did not influene eletrial values in ontrol tissues, nor did it influene the eletrial response to histamine. However, equimolar onentrations of diphenhydramine, an H reeptor bloker, ompletely prevented the ---q Fig. 2 Dose response urves for histamnine added to the serosal and miiosal side of ileal muiosa. Muosal appliation was ineffetive. Signifiantli' liffelrent, p<0l0/, *P<0.OOl Gut: first published as /gut on 1 November Downloaded from (N i, 50 u L- o.6 40a 3- ui 30 Control Bloker Hi Diphenhydromine (n=7) H2 Cimetidine (n=5) H2 Histamine 10-4M (serosol side) * 20 * 25 * Time (minutes) Hl Fig.3 Effets of diphenlhydranmine (10-4M) and imetidine (10-4M). on short-iriuit urrent hanges indu(ed by histamine 10-4M(gluosefree buffer). Diphen- hydramine onmpletely prevented the response to histamine, while inietidine was without effet. on 9 November 2018 by guest. Proteted by opyright.

4 Histamine and ion tr-ansport histamine-indued eletrial response (Fig. 3). n addition, diphenhydramine lowered potential differene and short-iruit urrent in tissues bathed in gluose-free buffer but not in those in gluoseontaining buffer. The influene of four different onentrations of diphenhydramine on the dose response urve to histamine is shown in Fig. 4, the urve being shifted to the right by the inreasing onentrations of H bloker. The shift to the right of the onentration of histamine required to produe a half maximal ontrol eletrial response was used to alulate a PA2 value as desribed in the legend to Figs. 4 and 5 and shown in Fig. 5. From this graph a pa.2 value of 7 85 was alulated, whih is similar to pa., values for H reeptors in other tissues. 5x x x104 Conentrotion of histamine (molar) Fig. 4 Dose response urves for histatnine-in(dle(l inr-eases in potential ditffrene with and without diphenhydratnine (DPH) in rabbit ileuem. The maximltm int-ease in potential differene obtainable with histatnine (2-2 m V) was taken as and all other r-eslults r-elated to that value. The lose response ttrive to histaitiine is inreasingi' shilted to the l ight b V 10-7, 5 x /0-7, 10-6 and 10-i'M diphenh),dratnine. The extent to whih the dlose of' histamyiine reqttired fbr a half inaximal ontrol rlesponse is shitiedl bi, the (liffer-ent onentr-ations o' H bloker indliated at A, B, C, D, and E was used to alitlate the pa2 value as shown in Fig. 5. n the histaminase inhibition studies, aminoguanidine 10-4 M had no eletrial effet when given alone. However, in tissues pretreated with aminoguanidine, histamine provoked an eletrial response at lower onentrations than in ontrol tissues. The urve was shifted to the left by aminoguanidine pretreatment (Fig. 6), suggesting that histaminases are present in the tissues. ) j pa Log diphenhydrmine onentration (molar) Dose ratio Log (dose ratio-i) Diphenlhydramyrine -Log (from Fig. 4) onentration diphenhydrainne (o1101) onentration BA- 30 Log 2 00= CA =475 Log 375=057 5 x DA 6 50 Log 5-50= EA 125 Log Fig. 5 Plot oflog (dose ratio-) vs-log antagonist onentration to determine pa2 (Shildplots). ON FLUXES 967 Preliminary studies with ontrol tissues bathed in gluose buffer alone showed that after 30 minutes' equilibration following addition of isotopes there was no signifiant hange in sodium, hloride, or residual ion fluxes during four onseutive 15 minute periods (n=8). Steady state fluxes were ahieved by 15 minutes after addition of histamine. Fluxes in ontrol tissues and during the periods minutes after histamine are shown in the Table (A). Net hloride absorption was signifiantly redued (P < 0 01) due to a redution in muosa to serosa flux (P<0-01). t should be noted that these fluxes were measured at a time when the transient eletrial responses had largely resolved. There was no obvious effet on sodium transport. The alulated residual ion flux, whih represents the unmeasured ions making up the total net ioni movement inferred from the short iruit urrent, was signifiantly redued after histamine (P <001). This result is ompatible with a redued seretion of biarbonate ions (or inreased seretion of hydrogen ions). Alkali seretion, measured by the ph stat tehnique, onfirmed that this residual ion flux was likely to be due, at least in part, to movement of biarbonate or hydrogen ions. As shown in Fig. 7, alkali seretion was redued after histamine from

5 C E C4 a).,!tn 'a, -4. U) } 8 -. ~0 C0 U) Arninoguanidine 10 x-x Pretreated(n=8) o --- o Control (n=6) ~~~~~~~~00 * o Linaker, MKay, Higgs, and Turnberg.0., 00 n 0 l Conentration of histamine (moles) 105 B 10-4 Fig. 6 Dose response urves.for histamine in ontrol and aminogiuanidine (10-'M) pretreated tissues. The ur-ve is shifted to the left by aninogutanidine. (Mean ± SEM shown.) *Signifiantly different fr-om ontr ol Eresponse (p < 0 001). Table onfluxes and short-iruit urrent in rabbit ilealmuosa: responses to histamine (A); histamine in diphenhydraminepretreated tissues (C); diphenhydramine alone (B) Na C- SCC JRnet Jm?ls Jsm Jnet J)ms Jsn1i Jnet A Histamine 10-4M (n =9) Control ±1 0 3 ± ±0 4 ±0 7 Histamine t t 4-4t -1 Ot =0 8 12±'1 ±0 7 : ±0-4 ±09 B Diphenhydramine (10-4M) alone (n=11l) Control ±0 8 ±0 7 ±0 6 ±0 5 ± ±0-8 Diphenhydramine ±10 ±08 ±08 ±06 ± ±0 3 ±1 C Histamnine 10-4 after liphlenlsydramp1ine 10-4M (n 9) Control (no drug) t.o6 ±0 6 -t03 -f0-3 ±0 4 ± Histamine = l-0 6 n eah instane ontrol fluxes for the 15 minutes before addition of drugs and responses to histamine or diphenhydramine were measured in the periods 15 to 30 minutes after drug addition. n study (C) histamine was added 15 minutes after diphenhydramine and fluxes measured 15 to 30 minutes after histamine. Fluxes and short-iruit urrent are in AMmol/m-2/h Jms=flux from muosa to serosa. Jsm=flux from serosa to muosa. Jnet=net flux. J,tnet=alulated residual ion flux. Mean±SEM; *P < 0-02; tp < 0-01; +P < ±0:3 to 1 4±0 3 tmol/m-2/hl (n=8, P<0 01). Although the flux data for the 15 minute period immediately after histamine addition are not presented, net hloride absorption was redued (4-1 to 1 5 pmol/m-2/h, P < 0 02, n = 9) beause of a derease in muosa to serosa flux (12-2 to 9 8 pmol/.0.0 J* m-2/h, P<0-01, n=9). The time ourses for the effet of histamine on alkali seretion (Fig. 7) and on hloride flux are thus similar. on fluxes were measured in response to histamine after pretreatment with diphenhydramine (10 --4M). The hanges in hloride fluxes indued by histamine /.4 t /4 /.4 4

6 Histamine and ion transport C.E u , a Un - 1~~\ 1 l H istamine. > a Time mirutes) Fig. 7 Effets of histamine (5 x 1F4M) ont alkali seretion (jumol/m-2/h) in rabbit ileum measured by ph stat tehnique. (Mean SEM shown.) (N=8.) Signifiantli' differ ent from ontlrol esponse (tp < 0 05, < 0Q01, 0P < (Net and lndir etionalfluxes of Cl-, obtained in a differ-ent set of tissues, during the same time per iod, ar-e shown for omparison.) were shown to be ompletely prevented by this H bloker, although a slight fall in short-iruit urrent persisted (Table, C). Diphenhydramine (10-4M) alone did not influene sodium or hloride fluxes and a fall in residual ion flux was not statistially signifiant (Table, B). The derease in short-iruit urrent indued by diphenhydramine in tissues in gluose buffer was not signifiantly different from the slight fall in short-iruit urrent seen in ontrol tissues over the same time period. Disussion jc,net iims i ilm The presene of histamine reeptors in the musulature of the small intestine has been supported by several investigations'3-'5 but their presene in intestinal muosa has not been well desribed. nfusions of histamine into the mesenteri arterial system of dogs provoked marked seretion into the lumen, but it was suggested that this was due to leakage of fluid through a apillary network whose permeability had beome markedly inreased.'6 Reent studies'7 with a number of histamine reeptor antagonists, however, have supported the idea that histamine reeptors may be involved in intestinal ion transport, at least in rabbit ileum. nterpretation of the latter studies is dependent on the speifiity of the histamine reeptor antagonists used (see below) but the urrent studies lend additional weight to the 969 onept of speifi H reeptors in rabbit ileal muosa. The eletrial responses to histamine were shown to be speifially bloked by diphenhydramine in a dose-related manner. The alulated value of 7 85 for pa2 is very similar to the values alulated for histamine/diphenhydramine relationships in other tissues (8 1, 8 0, and 7 8 in guinea-pig ileal musle ; 7-8 in guinea-pig trahea;9 and 7 8 in guineapig lung9) supporting the interpretation that speifi H reeptors are involved. The residual ion flux hanges were shown, by the ph stat tehnique, to be due, at least in part, to a redution in biarbonate seretion. The residual ion flux hange was larger than the hange in titratable alkali but this is hardly surprising beause the former was a alulated figure derived from the sum of Na and Cl fluxes and short-iruit urrent and the latter a diret measure in a different group of tissues at a different time. t is thus not possible to judge whether all or only a part of the hange in residual ion flux in these studies ould be attributed to biarbonate (of H +) movement hanges. The results presented here thus support the possibility of a hloride-biarbonate exhange whih is inhibited by histamine through an H reeptor but additional effets on other unmeasured ions annot be exluded. t is oneivable that the hanges in potential differene and short-iruit urrent were due not to effets of histamine on epithelium but rather to effets on submuosal tissue suh as the musularis muosae. The speifi H 1 blokade demonstrated here ould then theoretially represent a demonstration of H 1 reeptors on non-epithelial tissues. However, this seems somewhat unlikely, partiularly as steady ioni flux hanges, reahed 15 to 30 minutes after histamine, were shown to be speifially bloked as well. The observation by Fromm and Halpern"7 that H reeptor blokade alone signifiantly redued residual ion flux and biarbonate seretion suggests the possibility that endogenous histamine may be ating on the muosa under the basal onditions of their study. However, a omparison of their and our results reveals some disrepanies. Firstly, we ould not demonstrate a flux hange in response to H reeptor blokade alone. This may be due to our having used diphenhydramine, a bloker reputed to have minimal loal anaestheti ativity ompared with the pyrilamine20 primarily used by Fromm and Halpern. t is oneivable that the response to the latter agent was related to its loal anaestheti ativity2' rather than its H reeptor bloking ativity. They report a similar response with diphenhydramine but only in a onentration 10 times higher than that used in our study. Seondly,

7 970 Linaker, MKay, Higgs, and Turnberg their response to H l reeptor blokade is not in keeping with that expeted simply from antagonism to the ations of histamine demonstrated here. Our results would suggest that blokade of any endogenous histamine ativity whih may have been present should have enhaned biarbonate seretion (and Cl absorption) rather than the reverse. While the eletrial and flux hanges indued by histamine have been shown to be prevented by the H 1 bloker used in our experiments, this and other blokers may also have an ation independant of H1 bloking ativity, partiularly in high onentrations. Our observations that aminoguanidine, a histaminase inhibitor, enhanes the effet of histamine suggests that histaminases are present in intestinal muosal preparations and indeed they have been loalised in intestinal villous ells22. This observation supports the possibility that histamine may have a physiologial role in this site. The lak of effet of aminoguanidine alone suggests that it is unlikely that endogenous prodution of histamine is ourring to an appreiable extent in this in vitro preparation in the basal state. Preliminary studies of eletrial responses to histamine in rabbit jejunal and oloni muosa and in human oloni muosa suggest that these regions of the intestine respond in a similar manner (unpublished observations). t is of interest to ontrast the effets of histamine on the stomah with those on the intestine. n the former its ativity seems to be mediated through H2 reeptors23, while in the intestine it is lear that H reeptor mediated ativity is responsible for its effet on transport. These data do not, of ourse, demonstrate a physiologial role for histamine in intestinal transport but they do provide a possible mehanism by whih absorption may be impaired in irumstanes where histamine is liberated loally in the intestine. Suh may be the ase, for example, in intestinal allergies and systemi mastoytosis. We are grateful to the Medial Researh Counil and the North West Regional Health Authority for finanial support during this projet. Referenes 'Douglas WW, Feldberg W, Paton WDM, Shahter M. Distribution of histamine and substane P in the wall of the dog's digestive trat. J Physiol 1951; 115: Lorenz W, Matejka E, Shmal A et al. A phylogeneti study on the ourrene and distribution of histamine in the gastro-intestinal trat and other tissues of man and various animals. Comp Gen Pharmaol 1973; 4: Douglas WW. Autaoids. n: Goodman LS, Gilman A, eds. The phar.maologial basis of therapeutis. New York: MaMillan, 1975: Avery Jones F, Gummer JWP, Leonard Jones JE. n: Clinial gastr-oenter-ology. Oxford: Blakwell, 1968; Corbett CL, saas PET, Riley AK, Turnberg LA. Human intestinal ion transport in vitro. Gut 1977; 18: Field M, Fromm D, MColl. on transport in rabbit ileal muosa. 1. Sodium and hloride fluxes and short iruit urrent. Am J Physiol 1971; 220: Durbin RP, Heinz E. Eletromotive hloride transport and gastri aid seretion in the frog. J Gen Physiol 1958; 41: Shild HO. pa, a new sale for the measurement of drug antagonism. Br J Pharmaol 1974; 2: Arunlakshana 0, Shild HO. Some quantitative uses of drug antagonists. Br J Pharmaol 1959; 14: Shild HO. Reeptor lassifiation with speial referene to B-adrenergi reeptors. n: Rang HP, ed. Drug r-eeptors. London: MaMillan, 1973; "Reuse JJ. Comparisons of various histamine antagonists. Br J Pharmaol 1958; 3: Waton NG. Studies on mammalian histidine dearboxylase. Br J Phar-maol 1956; 11: Hill SJ, Young JM, Marrian DH. Speifi binding of 3H-mepyramine to histamine Hl reeptors in intestinal smooth musle. Nature 1977; 270: "Ash ASF, Shild HO. Reeptors mediating some ations of histamine. Br J Phar-maol 1966; 27: Bareiha 1, Roha-E-Silva M. Ourrene of H2 reeptors for histamine in the guinea pig intestine. Biohem Pharmaol 1975; 24: Lee JS, Silverberg JW. Effet of histamine on intestinal fluid seretion in the dog. Am JPhysiol 1976; 231 : '7Fromm D, Halpern N. Effets of histamine reeptor antagonists on ion transport by isolated ileum of the rabbit. Gastr-oenterology 1979; 77: '8Marshall PB. Some hemial and physial properties assoiated with histamine antagonism. Br J Pharmaol 1955; 10: Wilbrandt W. Zur frage des Wirkungsmehanismus der Antihistamin substanzen. Helv Physiol Pharmaol Ata 1950; 8: Douglas WW. Autaoids. n: Goodman LS, Gilman A, eds. The pharnmaologial basis of therapeutis. New York: MaMillan, 1975: Murdoh Rithie J, Cohen PJ. Loal anaesthetis. n: Goodman LS, Gilman A, eds. The pharmaologial basis of therapeutis. New York: MaMillan, 1975: Shakir KMM, Margolis S, Baylin SB. Loalization of histamninase (diamine oxidase) in rat small intestinal muosa-site of release by heparin. Biohem Phar-maol 1977; 26: Blak JW, Dunan WAM, Durrant CJ, Ganellin CR, Parsons EM. Definition and antagonism of histamine H2 reeptors. Nature (Lond) 1972; 236: