CHAPTER 4

MEMBRANE POTENTIAL AND CONDUCTANCE CHANGES EVOKED BY VIP AND CARBACHOL IN SMALL INTESTINAL CRYPTS.

In 1864 Bernard published the first evidence for the neural regulation of intestinal ion and fluid transport when he demonstrated that the removal of the extrinsic sympathetic innervation of the gut resulted in fluid secretion and diarrhoea (Bernard, 1864). Later acetylcholine became the first neurotransmitter substance to be identified in the gastrointestinal tract (Dale, 1937), although the neural control of intestinal transport was not firmly established physiologically until 1978 when Hubel reported that electrolyte secretion across the rabbit ileal mucosa could be evoked by electrical field stimulation (EFS) of the tissue in vitro (Hubel, 1978).

Similarly when the intrinsic cholinergic nerves of the guinea-pig were stimulated by EFS an increase in Cl^- secretion was evoked across the guinea-pig ileum (Cooke, 1983) which was partially diminished by the muscarinic antagonist atropine (Cooke, 1984a), suggesting that acetylcholine, released by neural depolarisation, was partially responsible for the increase in short-circuit current (I_sc) across the intestinal mucosa. This observation indicated that other neurotransmitters were involved in mediating the secretory response following EFS. VIP containing nerves have been shown to be abundant near to the crypt and villus epithelium (Schultzberg et al., 1980) and VIP has also been shown to be released following EFS (Gaginella et al., 1981) making VIP a strong candidate to modulate intestinal Cl^- transport in conjunction with acetylcholine.

Receptors for VIP have been demonstrated to be present in the basolateral membrane of enterocytes isolated from rat jejunum and rabbit ileum (Dharmsathaphorn et al., 1983) and guinea-pig small intestine (Binder et al., 1980), suggesting that VIP has a direct action on the epithelium in stimulating an increase in I_sc across guinea-pig small intestine (Cooke, 1984a). Muscarinic receptors of similar characteristics have also been found on both crypt and villus enterocytes isolated from rat small intestine (Isaacs et al., 1982).

VIP and acetylcholine appear to modulate small intestinal transport via different intracellular signalling pathways. VIP has been reported to elevate intracellular cAMP levels in the intact small intestinal epithelium and also in isolated enterocytes (Binder et al., 1980; Schwartz et al., 1974). Other agents, such as cholera toxin, also evoke intestinal secretion through an increase in cAMP, suggesting a central role for the cAMP signalling pathway in modulating intestinal ion transport (for review see O'Loughlin & Grant Gall, 1989). Several lines of evidence that the effect of acetylcholine is mediated by an increase in intracellular free Ca²⁺. In contrast carbachol, a metabolically stable muscarinic agonist, has been demonstrated to elevate intracellular Ca²⁺ levels in chicken villus enterocytes (Chang et al., 1986). Both the removal of extracellular Ca²⁺ (Hardcastle et al., 1984) and the addition of TMB-8, an inhibitor of intracellular Ca²⁺ mobilisation (Donowitz et al., 1986), have both been shown to abolish carbachol-induced elevations in I_sc across the small intestinal mucosa.

Most of our understanding of the cellular mechanisms by which VIP and carbachol stimulate fluid and electrolyte secretion come from studies of colonic carcinoma cell lines. Carbachol has been demonstrated to evoke Cl^- secretion across confluent T84 cell monolayers (Dharmsathaphorn & Pandol, 1986) which has been attributed to the activation of both K⁺ and Cl^- conductances (Worrell & Frizzell, 1991; Devor & Duffey, 1992) mediated by an increase in intracellular free Ca²⁺ (Devor et al., 1991). VIP has also been shown to evoke an increase in I_sc across T84 cell monolayers that is mediated via an increase in cAMP levels (Mandel et al., 1986b).

Discrepancies between the model provided from observations from colonic carcinoma cell lines and that afforded by observations from the small intestinal epithelium suggest that studies at the cellular level are important to our understanding of the mechanisms of small intestinal secretion. Measurements of whole-cell currents in freshly dissociated guinea-pig villus enterocytes revealed no increase in membrane conductance in response to secretagogues that elevate intracellular Ca²⁺ or cAMP levels (Sepulveda et al., 1991). The effects of secretagogues upon membrane conductance in isolated mammalian crypt enterocytes, however, have yet to be investigated. The results presented in this chapter are consistent with the presence of K⁺ and Cl^- conductances in the small intestinal crypt epithelium that are modulated by VIP and carbachol.

Electrophysiological recordings and data acquisition and analysis were performed as described in sections 3.2.3. and 3.2.4. respectively. All recordings presented in this chapter were obtained from the mid-region of the isolated small intestinal crypt.

4.2.1. CRYPT PREPARATION.

For electrophysiological experiments the crypt enterocytes were prepared by the method described in section 3.2.1..

4.2.2. SOLUTIONS.

For current-clamp recordings patch pipettes were filled with KCl-rich solutions containing 100 m g/ml nystatin prepared as described in 3.2.2.. In experiments conducted with "low" Cl^- (60mM) in the pipette solution, 85mM KCl was substituted with K gluconate. Stock solutions of carbachol (100mM) and porcine VIP (100m M) were prepared in distilled water and stored frozen. Stock solutions of atropine were prepared freshly by dissolving in Hanks medium to a final concentration of 1mM, taking precautions to prevent exposure to light. Cl^- channel blockers 5-nitro-2-(3-phenylpropylamino)-benzoic acid (NPPB) and 9-anthracene carboxylic acid (9-AC) were prepared as stocks in DMSO at 100mM concentration and stored frozen. Forskolin was prepared as a 20mM stock, frozen in DMSO. Precautions against light-sensitive reactions were adopted when using Cl^- channel blockers. All bath and pipette solutions were filtered through Dynaguard 0.2m m syringe filters prior to use.

Final concentrations of all agents were achieved by the appropriate dilutions with normal Hanks medium. All errors quoted are standard errors of the mean of n observations. The composition of all pipette and extracellular solutions used are given in table 4.1.

Table 4.1. Composition of solutions.

All concentrations are given in mM and solutions were titrated to pH 7.2 with Tris. Solutions containing 0.5mM and 20mM K⁺ were made by equimolar replacements of NaCl with KCl. The 0Ca²⁺ Hanks solution was prepared by omitting CaCl₂ from normal Hanks medium.

Muscarinic stimulation has been shown to evoke Cl^- secretion across the guinea-pig small intestinal mucosa (Carey et al., 1987). An effect of muscarinic activation on crypt E_m may therefore be expected provided that the small intestinal crypt is a site of fluid and electrolyte secretion. The effect of carbachol, a specific muscarinic receptor agonist, on crypt E_m is shown in Fig.4.1A. The recording shown is taken from a crypt with an initial E_m value of -57 mV whereupon addition of 100m M carbachol a 15mV hyperpolarisation was observed which was sustained for the duration of agonist application, and was readily reversed upon carbachol washout. The carbachol induced hyperpolarisation was repeatable with no apparent reduction in the magnitude of the response after 168 seconds. Hyperpolarisations were observed in response to carbachol addition in 21 out of 22 other crypts tested. The effect of carbachol was totally suppressed in the presence of 10m M atropine (not shown). The deflection of E_m in response to a constant current pulses was decreased during carbachol application, indicating an increase in membrane conductance.

The earliest detectable effect of carbachol took place after a delay of less than 2 seconds, and the maximal change in E_m occurred within 6 ± 1 s from the onset of the hyperpolarisation. An estimation of the rate at which E_m changed after carbachol addition gave a mean value of 3.3 ± 1 mV/s, and E_m recovered within 10 ± 2 s following carbachol washout (values quoted are means ± SEM, n=9). The magnitude of the carbachol-evoked hyperpolarisation was concentration-dependent with an IC₅₀ of around 2m M (Fig.4.5).

To investigate the ionic basis of the carbachol-induced hyperpolarisation, changes in extracellular K⁺ were made in the presence and absence of carbachol. Fig. 4.1B shows the effect of changing [K⁺]_o first to 0.5 and then 20 mM under control conditions, which produced the hyperpolarisation and depolarisation of E_m reported earlier (Fig.3.1). The subsequent addition of carbachol produced a sustained hyperpolarisation during which the magnitude of the E_m deflections in response to changes in extracellular [K⁺]_o were clearly potentiated. The effect of Cl^- replacement during carbachol action was little different from that observed under control conditions (Fig.3.1). On prolonged incubation with low Cl^- medium a potentiation of the carbachol-dependent hyperpolarisation was observed (results not shown).

Effect of carbachol on the membrane potential on the mid-region of isolated crypts. A. Recording of crypt E_m bathed in normal Hanks medium during carbachol addition. Current pulses of constant amplitude were passed throughout the recording. The two record segments were separated by the time indicated. B. E_m recording of a separate crypt showing the effect of changes in extracellular K⁺ concentration before and during carbachol application. The duration of ion substitutions and carbachol additions are indicated by the horizontal bars. The initial value of E_m is given for each recording.

It is well documented that muscarinic activation leads to the mobilisation of both intra- and extracellular Ca²⁺ and the activation of protein kinase C (Berridge, 1993). To investigate the possible involvement of such a signalling cascade in mediating the action of carbachol, the effect of removal of extracellular Ca²⁺ upon the action of carbachol was explored. Fig.4.2A shows a representative recording of the effects of extracellular Ca²⁺ removal upon the hyperpolarisation induced by carbachol. Removal of extracellular Ca²⁺ had little effect on E_m by itself in this particular experiment (although it produced a small depolarisation in several other experiments) but the carbachol-dependent hyperpolarisation, while initially of the same magnitude, became transient in the absence of extracellular Ca²⁺. The subsequent replacement of normal Hanks medium immediately after perfusion with Ca²⁺-free Hanks medium containing carbachol, led to a transient hyperpolarisation. In 6 experiments the initial hyperpolarisation produced by 100m M carbachol was 15 ± 2 and 12 ± 1 mV in normal Hanks and Ca²⁺-free medium respectively. At the steady-state the corresponding values were 10 ± 1 mV hyperpolarised and 6 ± 3 mV depolarised (means ± SEM). The mean hyperpolarisation evoked by replenishment of extracellular Ca²⁺ following carbachol washout was 12 ± 1 mV. Further experiments were performed to establish the Ca²⁺-dependency of the carbachol-induced hyperpolarisation by the addition of the Ca²⁺-ionophore A23187 to the bathing medium. Fig.4.2B shows the effect of addition of 5m M A23187 to the bathing medium of a crypt with an initial E_m of -55mV. The Ca²⁺ ionophore evoked a 25mV hyperpolarisation of E_m concomitant with an increase in membrane conductance. The effect of carbachol upon the whole-cell current elicited by a ramp potential in the small intestinal crypt is shown in Fig.4.6. The addition of carbachol evoked an increase in outward current accompanied by a shift in the reversal potential from -39mV to -57mV.

Dependence of carbachol-induced hyperpolarisation upon extracellular Ca²⁺. A. Continuous recording of E_m from a crypt bathed in Hanks medium with the duration of the addition of carbachol and Ca²⁺ replacements indicated by the horizontal bars. The solution identified as 0Ca²⁺ Hanks was prepared by omitting CaCl₂ from normal Hanks medium. The two traces shown belong to the same continuous recording. B. Effect upon crypt E_m of the addition of 5m M A23187 to the bathing medium. The duration of solution changes are as indicated by the horizontal bars. The initial value of E_m is given for each recording. Scale bars for E_m are given for each recording.

4.3.3. Effects of phorbol esters upon crypt membrane potential and conductance.

The possible involvement of the protein kinase C (PKC) signalling pathway in mediating conductance changes in response to carbachol addition was determined by the addition of the phorbol ester phorbol 12-myristate 13-acetate (PMA) to the bathing medium. Fig.4.3A shows the effect upon crypt E_m and membrane conductance of addition of 1m M PMA to a crypt continuously bathed in Hanks medium. In 5 separate experiments the addition of 1m M PMA resulted in a maximal depolarisation of 9 ± 1.4 mV within 5-10 minutes of application which was associated with an increase in conductance. In three separate experiments 10-20 minutes of exposure to 1m M PMA E_m became more negative approaching the value of E_m observed prior to phorbol ester addition (-5 ± 7 mV relative to initial E_m). Application of 100nM PMA resulted in no apparent change in E_m (n=3).

Actions of phorbol esters and VIP upon crypt membrane potential and conductance. A. Effect of addition of 1m M PMA to a crypt bathed with Hanks medium. B. Effect of addition of 50nM VIP to the bathing medium. Current pulses of constant amplitude were passed throughout both recordings. The duration of agonist additions are indicated by the horizontal bars. The initial value of E_m is given for each recording. Scale bars are representative of both recordings.

VIP has also been shown to evoke an increase in I_sc across the guinea-pig small intestinal mucosa (Cooke, 1984a). Experiments were therefore performed to test whether VIP has a direct action on the small intestinal epithelium. Fig.4.3B shows a continuous recording of crypt E_m in which 50nM VIP was added to the bathing medium for the duration indicated by the solid bar. After a delay, VIP evoked a 30mV depolarisation, associated with an increase in membrane conductance. The effect of VIP was reversible with a recovery of E_m occurring within 7 minutes after washout. The dose-dependency of the effect of VIP is shown in Fig.4.5, where the concentration giving a half-maximal depolarisation was 30 nM. The response to carbachol addition was more rapid in onset than that seen following VIP addition. The earliest detectable effect of VIP took place after a delay of 24 ± 3 sec, and the maximal change in E_m occurred within 43 ± 8 s from the start of the effect (means ± SE from 6 experiments). An estimation of the rate at which changes in E_m took place in response to VIP gave a mean value of 1 ± 0.3 mV/s. The mean recovery time for E_m following VIP washout was 940 ± 370 s.

To determine whether the VIP-induced depolarisation was in part due to the activation of a Cl^- conductance, changes in E_m in response to VIP addition were determined with both low (60 mM) and high (145 mM) concentrations of Cl^- in the patch pipette (equimolar replacement of Cl^- with gluconate^-). Depolarisations induced by the addition of 100 nM VIP to the bath were significantly smaller (P < 0.05, unpaired student's t-test) with 60 mM Cl^- in the pipette (24 ± 2 mV, n=4) than with 145 mM Cl^- (32 ± 1, n = 4). E_Cl values of -23 mV and 0.4 mV respectively were calculated assuming complete equilibration of the intracellular milieu with the pipette solution.

The co-operative effects of 100 nM VIP and 100 m M carbachol upon E_m are shown in the upper panel of Fig.4.4. In the presence of VIP a marked depolarisation was observed, which recovered to the resting value of E_m within 6 minutes of agonist washout. Subsequent addition of carbachol produced a hyperpolarisation which was rapidly reversed upon washout. The simultaneous addition of carbachol and VIP induced a brief initial hyperpolarisation before a depolarisation which was larger than that produced by VIP by itself. Recovery to the resting E_m took place within 15 minutes of agonist washout. The large depolarisation and conductance increase was repeatable. The deflections in E_m in response to a constant train of current pulses indicate that the increase in membrane conductance evoked was larger in the presence of both agonists than in response to the addition of either agonist added individually.

Effects of vasoactive intestinal polypeptide (VIP), carbachol (CCH) and forskolin (FSK) on crypt membrane potential. The upper trace shows a continuous recording of crypt E_m. During the time indicated by the horizontal bars the crypt was successively perfused with Hanks medium containing 100 nM VIP, 100 m M carbachol (CCH) and 100 nM VIP + 100 m M CCH. Current pulses of constant amplitude were applied throughout the recording. The lower trace shows the effect upon E_m of the addition of 20 m M forskolin to a different crypt. Initial E_m values are as indicated by the arrows.

Dose response relationships for the effect of VIP (open circles) and carbachol (closed circles) on crypt E_m. Data are expressed as the percentage of the maximal change recorded the highest concentration used and are expressed as means ± SE from three separate experiments for VIP and from a single experiment for carbachol.

VIP has been shown to evoke secretion across the small intestine associated with increased levels of intracellular cAMP (Schwartz et al., 1974). Thus we employed forskolin, a known activator of adenylate cyclase (Seamon et al., 1981), to see if it would reproduce the depolarisation evoked by VIP. Fig.5.4B shows the effect of addition of 20 m M forskolin to a crypt with an initial E_m of -44mV, where a sustained 28 mV depolarisation was evoked which was slow to recover following washout. In five separate experiments 20m M forskolin produced a mean 23 ± 3 mV depolarisation. The effect of 20m M forskolin upon crypt whole-cell current elicited by a ramp potential is shown in Fig.4.6A. Forskolin addition shifted the reversal potential in response to a ramp potential from -39mV to -11mV, associated with a large increase in both inward and outward currents at negative and positive holding potentials respectively.

Experiments were performed to determine the sensitivity of the hyperpolarisation evoked by 100m M carbachol to quinine. The K⁺ conductance of the unstimulated crypt appears to be potently inhibited by quinine with an IC₅₀ of 200m M (Fig.3.3). The continuous recording of crypt E_m in Fig.4.6B shows an initial 8mV hyperpolarisation evoked by the addition of 100m M carbachol, followed by a 19mV depolarisation in response to the addition of 200m M quinine in the continued presence of 100m M carbachol. The magnitude of the depolarisation evoked by quinine is evidently greater in the absence of carbachol (23mV) than in its presence (14mV). The 9mV difference is similar to the size of the initial hyperpolarisation evoked by carbachol, suggesting that the carbachol-induced hyperpolarisation is relatively insensitive to quinine. Hyperpolarisations evoked by carbachol in the presence and absence of quinine were of similar magnitude in two separate experiments.

Effect of carbachol upon crypt whole-cell currents and sensitivity of the carbachol-induced hyperpolarisation to quinine. A. Current-voltage relationship showing the perforated-patch membrane currents recorded from an isolated crypt (initial E_m of -39mV) elicited by a ramp potential (middle right) in the presence and absence of the agonists carbachol and forskolin at the times indicated. B. Effect of 200m M quinine upon the hyperpolarisation evoked by 100m M CCH in the mid-region of the crypt. Continuous recording of crypt E_m where the effect of the successive additions of Hanks medium containing 100m M CCH, 100m M CCH in the presence of 200m M quinine and finally 200m M quinine before washout with Hanks medium are shown. The duration of all solution changes are as indicated by the horizontal bars. Initial E_m values are as indicated by the arrows.

The sensitivity of the VIP-induced depolarisation to the Cl^- channel blocker 5-nitro-2-(3-phenylpropyl-amino)-benzoic acid (NPPB) was tested, an inhibitor known to be effective in a number of transporting epithelia (Greger, 1990). Fig.4.7A shows the effect of 100 m M NPPB upon crypt E_m. NPPB induced a slow 10 mV hyperpolarisation that was maximal within 100s, and which did not readily wash out. In 4 separate experiments 100 m M NPPB induced a mean 7 ± 2 mV hyperpolarisation. The recording in Fig.4.7B shows the effect of 100 m M NPPB addition upon a crypt depolarised in the continuous presence of 30 nM VIP. The Cl^- channel blocker fully reversed the VIP-induced depolarisation, leading to a hyperpolarisation of the crypt membrane. In four separate experiments the addition of 100 m M NPPB after the application of VIP led to a mean hyperpolarisation of 12 ± 4 mV (mean ± SEM). The addition of the Cl^- channel blocker 9-anthracene carboxylic acid at 100m M concentration evoked hyperpolarisations of 4 and 6 mV respectively in unstimulated small intestinal crypts.

The effect of addition of 100nM VIP to crypts preincubated with 100m M NPPB is shown in Fig.4.7B. In the recording shown VIP addition induced a rapid 17mV hyperpolarisation. In three separate experiments addition of 100nM VIP after preincubation with 100m M NPPB resulted in a mean 10 ± 3 mV hyperpolarisation of E_m.

Effects of NPPB and VIP on crypt membrane potential. Recordings of E_m from three different isolated crypts are shown. The effects of the addition of (A) 100m M NPPB alone, (B) 30nM VIP followed by 100m M NPPB and (C) 100nM VIP after preincubation with 100m M NPPB on crypt E_m are shown. The duration of agonist and blocker additions are as indicated by the horizontal bars. Initial E_m values are indicated by arrows.

4.4.1. Overview.

The location of the cells mediating the secretion of fluid and electrolytes within the small intestinal epithelium is uncertain. Sepulveda and co-workers reported no change in whole-cell current in response to the addition of forskolin, 5-HT or carbachol in guinea-pig villus enterocytes, although the possible dialysis of intracellular signalling mediators could not be excluded (Sepulveda et al., 1991). However, microelectrode studies in the rat small intestine revealed that PGE₂, 5-HT and acetylcholine decreased V_a in cells from both villus and crypt regions by the apparent activation of a Cl^- conductance (Stewart & Turnberg, 1989). The membrane conductance changes evoked by VIP and carbachol are consistent with the presence of functional receptors in the epithelium of the mid-region of isolated small intestinal crypts and these findings support the contention that the crypt epithelium is a site of fluid and electrolyte secretion. The observation that VIP and carbachol have opposite effects on crypt membrane potential, however, suggests that different conductance pathways may be modulated by the two agonists, probably via the activation of distinct signalling pathways.

4.4.2. Second messenger pathways mediating the response to carbachol.

The inhibition of the carbachol response by the specific muscarinic antagonist atropine implicates muscarinic receptors in mediating the dose-dependent hyperpolarising action of carbachol on the small intestinal crypt epithelium. Muscarinic receptors of the M1, M3 and M5 subtypes (Wess et al., 1990) are coupled via a G-protein to the activation of phospho-inositidase C, an enzyme that mobilises DAG and IP₃ from phosphatidyl inositol bisphosphate, leading to the activation of PKC and the mobilisation of both intra- and extracellular Ca²⁺ (Berridge & Irvine, 1989; Berridge, 1993). Muscarinic receptors with similar binding characteristics have previously been reported to be present on both crypt and villus enterocytes isolated from the rat small intestine (Isaacs et al., 1982).

The initial phase of the carbachol-induced hyperpolarisation appears to be independent of the presence of extracellular Ca²⁺, whilst the sustained hyperpolarisation phase showed a marked dependence upon extracellular Ca²⁺. These observations are consistent with the rapid release of Ca²⁺ from intracellular stores and the subsequent entry of extracellular Ca²⁺ that have been reported to occur in other cell types, including the T84 colonic cell line (Reinlib et al., 1989). The results are also consistent with the actions of the inhibitor of intracellular Ca²⁺ mobilisation TMB-8 (Donowitz et al., 1986) and extracellular Ca²⁺ removal (Hardcastle et al., 1984), which have both been shown to abolish carbachol-induced elevations in I_sc across the small intestinal mucosa. The concentration dependence of the carbachol effect on crypt membrane potential is similar to that reported for its effect on I_sc across the guinea-pig ileum (Carey et al., 1987). The similar hyperpolarisation induced by the Ca²⁺-ionophore A23187 provides further evidence for the role of elevations in intracellular Ca²⁺ in mediating the action of carbachol, although caution must be applied to the interpretation of this result, as high concentrations of intracellular Ca²⁺ have been demonstrated to activate the phospholipase A₂ signalling cascade (Dieter et al., 1988) leading to the mobilisation of arachidonic acid metabolites.

Carbachol has also been reported to increase the DAG content of microvillus membranes in isolated rabbit enterocytes (Cohen et al., 1991). The exposure of small intestinal crypts to the phorbol ester PMA, a membrane-permeant activator of PKC, evokes a slow increase in crypt membrane conductance, suggesting a possible functional role for the PKC signalling pathway in the regulation of electrolyte transport across the crypt epithelium.

4.4.3. Second messenger pathways mediating the response to VIP.

VIP addition evoked a depolarisation of E_m that could be emulated by the addition 20m M of forskolin, an established activator of adenylate cyclase (Seamon et al., 1981). Consistent with this observation, VIP has been reported to increase cAMP levels in the intact small intestinal epithelium (Schwartz et al., 1974), in isolated guinea-pig enterocytes (Binder et al., 1980) and in T84 cell monolayers (Mandel et al., 1986a). The effect of VIP on crypt E_m is dose-dependent, with an IC₅₀ of 30nM, which compares with half-maximal effective concentrations for the binding of VIP to isolated guinea-pig isolated enterocytes and the VIP-evoked increase in I_sc across guinea-pig small intestine of 12 nM and 10 nM respectively (Binder et al., 1980; Cooke et al., 1984a). The current body of evidence points to the presence of a functional VIP receptor in the crypt basolateral membrane, coupled to adenylate cyclase via the guanine nucleotide regulatory protein G_s, a protein that is abundant in isolated rat crypt enterocytes (Couvineau et al., 1992).

4.4.4. Evidence for the activation of K⁺ channels by VIP and CCH.

The rapid and reversible hyperpolarisation evoked by carbachol shows a biphasic dependence upon extracellular Ca²⁺ (Fig.4.2.) and is associated with an increase in membrane conductance (Fig.4.1). Carbachol also increases outward current (Fig.4.6.) and the magnitude of the deflections of E_m in response to extracellular K⁺ replacement experiments (Fig.4.1.). The results are consistent with the presence of a Ca²⁺-activated K⁺ conductance in the crypt basolateral membrane, a feature common to many fluid-secreting epithelia (Latorre, 1989). A Ca²⁺-activated K⁺ conductance has also been reported in isolated rat colonic crypts (Bohme et al., 1991) and in the T84 cell line (Devor et al., 1990), although no evidence for a carbachol-activated K⁺ conductance has been obtained from patch-clamp studies of guinea-pig villus enterocytes (Sepulveda et al., 1991).

The apparent insensitivity of the carbachol-induced hyperpolarisation to 200m M quinine, the concentration at which quinine half-maximally depolarises the unstimulated crypt, indicates that distinct K⁺ channels probably underlie the spontaneous and carbachol-activated K⁺ conductances present in the small intestinal crypt. The insensitivity of the carbachol-induced hyperpolarisation to quinine argues against the involvement of large conductance Ca²⁺-activated "maxi-" K⁺ channels in mediating the response to carbachol (at physiological membrane potentials), as the maxi-K⁺ channel is known to be blocked by extracellular quinine at sub-millimolar concentrations (Segal et al., 1990).

Hyperpolarisations of the small intestinal crypt membrane potential by VIP after preincubation with NPPB are consistent with the presence of a cAMP-activated K⁺ conductance. If VIP does additionally activate a K⁺ conductance, this effect is masked by its predominant action upon a Na⁺ and/or Cl^- conductance. Although a cAMP-activated K⁺ conductance has been reported to be present in rabbit colonic crypts (Loo & Kaunitz, 1989) and in T84 cell monolayers (Mandel et al., 1986b), such a conductance has been reported to be absent from the HT29 colonic carcinoma cell line (Bajnath et al., 1991), although this may be due to de-differentiation in this particular cell line.

4.4.5. Evidence for the activation of a Cl^- conductance by VIP in the small intestinal crypt.

In isolated small intestinal crypts VIP evokes a marked depolarisation of the crypt membrane associated with a decrease in membrane resistance. Such a depolarisation is consistent with the predominant activation of a Cl^- conductance, but could also be explained by an increase in Na⁺ (or non-selective cation) conductance. Two lines of evidence are consistent with the activation of a Cl^- conductance by VIP. Firstly, the depolarisation can be reversed by the addition of the established Cl^- channel blocker NPPB to the bathing medium. Such a high concentration of NPPB (100m M) was used to compensate for any diffusion gradient that may have been established between the bathing medium and the lumen. NPPB has been widely reported to be a potent inhibitor (IC₅₀ 0.9m M) of the intermediate conductance outwardly-rectifying epithelial Cl^- channel (Tilmann et al., 1991) and recently 10m M NPPB has also been demonstrated to inhibit potently cAMP-activated (CFTR) Cl^- currents in pancreatic duct cells (Gray et al., 1993). In addition NPPB has been shown to potently inhibit forskolin and PGE₂ evoked Cl^- secretion across rat colon (Diener & Rummel, 1989; Greger et al., 1991). Secondly, the magnitude of the VIP-induced depolarisation appears to be dependent upon the concentration of Cl^- used in the patch pipette, suggesting that E_m is driven towards E_Cl by the action of VIP. This would indicate that efficient dialysis occurs between the pipette solution and the intracellular milieu across the nystatin-perforated patch, although this has not been measured directly.

The VIP-induced depolarisation was strongly potentiated by carbachol and the increase in membrane conductance was greater in the presence of both agonists than when either agonist was added separately. If the principal action of carbachol is on a K⁺ conductance then it would be expected to oppose the action of VIP, decreasing the magnitude of the VIP-induced depolarisation. This observation suggests that either carbachol allows VIP to activate a Cl^- conductance not normally accessible to its intracellular mediators, or that VIP activation renders Cl^- channels, normally unresponsive to carbachol, sensitive to its intracellular mediators.

4.4.6. Evidence for the activation of a Cl^- conductance by PKC and Ca²⁺.

The addition of 1m M of the PKC activator PMA consistently evoked a slow, transient depolarisation associated with an increase in membrane conductance, consistent with the activation of a Cl^- conductance. The magnitude of the PMA-induced depolarisation (9 ± 1.4 mV, n=5) was smaller than that evoked by VIP (32 ± 1, n=4) or forskolin (23 ± 3 mV, n=5). Alternatively the PMA-induced depolarisation could be interpreted as the activation of a Na⁺ or cation non-selective conductance. Prolonged incubation with PMA resulted in a secondary hyperpolarisation and an increase in conductance that could either be interpreted as the activation of a K⁺ conductance and/or the inhibition of a Na⁺ or Cl^- conductance. Studies in the HT29 (Bajnath et al., 1992) and T84 (Kachintorn et al., 1992) colonic carcinoma cell lines have revealed that phorbol esters initially stimulate and subsequently inhibit a Cl^- conductance. This indicates that muscarinic stimulation may result in the potentiation of a Cl^- conductance through the activation of PKC, in addition to the increase in basolateral K⁺ conductance through an elevation in intracellular [Ca²⁺]_i. Whether a rise in intracellular free Ca²⁺ also results in the activation of a Cl^- conductance sensitive to its mediators cannot be answered from these experiments. If such an effect occurs, however, the overall effect on membrane potential seems to be dominated by the change in K⁺ conductance. This would be in agreement with results obtained in T84 colonic carcinoma cells, where carbachol evokes Cl^- secretion through preactivated Cl^- channels by increasing basolateral K⁺ permeability only (Dharmsathaphorn & Pandol, 1986).

4.4.7. Evidence for the presence of a cAMP-activated Cl^- conductance in the small intestinal epithelium.

Several studies have demonstrated that in the CF intestine both cAMP and Ca²⁺-stimulated Cl^- secretion are defective (Berschneider et al., 1988; Taylor et al., 1987). In contrast Ca²⁺-stimulated Cl^- secretion is unaffected in CF airways (Willumsen & Boucher, 1989), consistent with the finding that Ca²⁺ calmodulin-dependent protein kinase activates Cl^- channels in both normal and CF airway epithelial cells (Wagner et al., 1991). Similarly Cl^- channels activated by Ca²⁺ have been found to be absent from the apical membrane of polarised monolayers of colonic carcinoma cells (Anderson & Welsh, 1991).

The relative inaccessibility of the crypt lumen impedes the localisation of putative agonist-activated Cl^- conductance pathways to the apical membrane. The slow hyperpolarisation of E_m induced by NPPB in unstimulated crypts would be consistent with the presence of a basal Cl^- (or Na⁺) permeability pathway in the apical membrane of the small intestinal crypt epithelium. The presence of such a conductance would be consistent with the non-Nernstian dependence of E_m upon extracellular K⁺, a resting membrane potential that is considerably depolarised with respect to the predicted value of E_K and the slow hyperpolarising action of the Cl^- channel blocker NPPB.

Protein kinase A has been demonstrated to activate small conductance Cl^- channels in various epithelia (Gray et al., 1988), which have now been associated with the expression of the CFTR gene product (Anderson et al., 1991, Bear et al., 1992). A cAMP-activated conductance probably reflecting the presence of phosphorylation-activated Cl^- channels has been described in Necturus enterocytes (Giraldez et al., 1988a, 1989). Cl^- channels activated by cAMP have also been shown to be present in the T84 colonic carcinoma cell line (Tabcharani et al., 1990). It may be speculated that the activation of such small conductance channels underlies the depolarisations evoked by VIP, forskolin and PMA in small intestinal crypts. This argument gains some credence from recent work showing high levels of expression of the CFTR gene product in the crypt region of the rat small intestine (Trezise & Buchwald, 1992). It has been reported that PKC strongly potentiates the PKA-mediated activation of Cl^- channels associated with the expression of the CFTR gene product, whilst PKC alone has only a weak effect upon channel activity (Tabcharani et al., 1991). This may explain the synergistic effect of VIP and carbachol on crypt membrane conductance (Fig.4.4) if we assume the respective activation of PKC and PKA by these agonists. Depolarisations have also been reported to be evoked in isolated colonic crypts by forskolin (Bohme et al., 1991), PGE₂ (Siemer & Gogelein, 1992) and VIP (Greenwald & Biagi, 1992), all agents known to increase cAMP levels in the intestinal epithelium. However, the ionic basis of this depolarisation is unclear, as Siemer and Gogelein (1992) have demonstrated the activation of a cation non-selective conductance by PGE₂ , whilst Bohme and co-workers (1991) postulated that forskolin potentiated a NPPB-sensitive Cl^- conductance whose functional expression depended upon the age of the colonocyte. Whether the depolarising action of forskolin and VIP upon the small intestinal crypt is in part mediated by the activation of a Na⁺ (or cationic) conductance cannot be determined from the data presented here, although the regional variation of the response to forskolin and carbachol is discussed further in chapter 6.

4.4.8. Conclusions.

Muscarinic and VIP receptor stimulation evokes conductance changes in isolated small intestinal crypts consistent with the activation of K⁺ and Cl^- conductance pathways. The present data supports the contention that the crypt compartment mediates agonist-evoked fluid and electrolyte secretion by the small intestinal epithelium. The variation in the response to agonists along the crypt axis and the identity of the channel underlying the carbachol-activated basolateral K⁺ conductance are addressed in later chapters. More direct experimental evidence is required to conclusively establish the location and regulation of the Cl^- channels modulated by agonists in the small intestinal crypt epithelium.

Return to top
Return to table of contents
Return to science homepage
References
Next chapter