EFFECT OF SECRETAGOGUES UPON CRYPT VOLUME AND ION TRANSPORT.

The agonist-induced activation of K⁺ and Cl^- conductance pathways is an established feature of Cl^--secreting epithelia (Petersen & Gallacher, 1988). In the Petersen model of epithelial Cl^- secretion Cl^- ions are proposed to be accumulated above electrochemical equilibrium in the cell by a basolateral cotransport mechanism, and thus, following the agonist-induced activation of an apical Cl^- conductance pathway, Cl^- will leave the cell down its electrochemical gradient. The efflux of K⁺ ions, taken up into the cell by the Na⁺-K⁺ ATPase (and in some cells by the NaK2Cl cotransporter) and recycled across the basolateral membrane via K⁺-selective ion channels, is believed to accompany Cl^- due to a commensurate increase in basolateral K⁺ conductance (Petersen & Gallacher, 1988). In this model of exocrine secretion there is no requirement for the net loss of KCl and water from the cytosol and therefore cell volume changes were not predicted.

However, investigators measuring cell volume changes in response to cholinergic agonists in dissociated eccrine clear cells (Suzuki et al., 1991) and isolated salivary acinar cells (Foskett & Melvin, 1989) have demonstrated that muscarinic stimulation results in as much as a 20-30% decrease in cell volume that occurs within minutes of agonist addition. Manipulations of the extracellular K⁺ and Cl^- concentrations and K⁺ and Cl^- channel inhibitors have been used to show that the volume decrease induced by cholinergic agonists is due to the net loss of K⁺ and Cl^- ions from the cytosol (Suzuki et al., 1991). Thus the dramatic cell shrinkage that is evoked by agonists in isotonic media may be a characteristic feature of exocrine acinar cells.

The activation of K⁺ and Cl^- conductance pathways by VIP and carbachol in small intestinal crypts, as inferred from membrane potential changes, raises the possibility that these agonists might also modulate crypt cell volume. To address this issue viable small intestinal crypts have been isolated from guinea-pig small intestine and changes in crypt volume in response to VIP and carbachol have been estimated from the analysis of crypt photomicrographs taken at timed intervals. The data presented in this chapter indicates that both agonists evoke a reversible decrease in crypt volume that is apparently due to the net loss of K⁺ and Cl^- ions from the cytosol.

5.2.1. Crypt preparation.

The preparation of crypt enterocytes for crypt volume measurements was as described in section 3.2.1. except for a minor modification. After an initial wash in Hanks medium, the isolated crypts were not washed in DMEM containing 2mM DTT, but resuspended in normal Hanks medium and maintained on ice.

5.2.2. Solutions.

Stock solutions of carbachol (100mM), porcine vasoactive polypeptide (VIP, 100m M) , NPPB and 9-anthracene carboxylic acid (9-ACA) were prepared as described in section 4.2.2.. Final concentrations of NPPB and 9-ACA were prepared so that the final concentration of DMSO was <0.1% (v/v). High K⁺ (20mM) Hanks medium was prepared by the equimolar substitution of NaCl with KCl. For the composition of solutions see table 7.1. The osmolarity of all solutions was measured with a freezing point depression osmometer (3MO Advanced Instruments, Massachussetts, USA), and if necessary was adjusted to 298 ± 5 mosmole/l by the addition of D-mannitol.

Changes in the bathing solution during experiments to measure cell volume were made by the successive addition of two 5ml aliquots of the replacement solution to a chamber containing approximately 1ml of bathing medium, removing excess solution by means of a vacuum pump. All measurements quoted are the means ±SE from n observations.

An individual isolated crypt was selected by its cylindrical morphology and birefringence when observed under phase contrast optics with an inverted microscope (Wild, Germany, total magnification 320x). The crypts were aliquoted into a perspex chamber (total volume of 1ml), the bottom of which was made from a glass coverslip. Adherence of the crypts was achieved by precoating the glass surface with 0.1% PEI for 4 hours. Crypts were allowed to adhere to the coverslip by pre-equilibration for 15 minutes in Hanks medium and the chamber was subsequently washed with 10 ml of Hanks medium to remove loosely attached cellular material. All solutions and the chambers containing crypts were pre-equilibrated at 25° C for 15 minutes before the start of the experiment. Photomicrographs of the selected crypt were taken at timed intervals before and after solution changes were effected. Zero time was defined as the time of solution change. Photomicrographs of the crypts were analysed for total area, mean height (h) and mean width (2r) with the aid of a Quantimet 520 image analysis system (Cambridge Instruments, U.K.). Volume was estimated, assuming a cylindrical morphology, as Ò.h.r², and normalised to the point previous to solution changes. All crypt volume measurement experiments were carried in a room maintained at 25° C.

The addition of secretagogues to crypts during electrophysiological experiments appeared to be accompanied by a discernible change in crypt size. This phenomenon is illustrated in Fig.5.1., which shows a crypt immediately before and 8 minutes after the addition of 100 nM VIP to the bathing medium (Figs. 6.1A and B respectively). Crypt volume was then measured by the analysis of timed photographic images to quantify this volume change in response to secretagogues. Fig.5.2 shows the time course for the mean change in apparent crypt volume in response to the addition of 100 nM VIP to the bathing medium. There was a marked volume reduction following VIP addition with a maximal shrinkage being attained within 2-5 minutes of agonist addition. This "secretagogue-induced volume decrease" (SVD) reached a stable level, a decrease of 27% of the initial volume in the experiments illustrated, with no indication of volume recovery even after 24 min in the continued presence of the agonist. A similar sustained reduction in crypt volume was observed when challenging crypts with 100 m M carbachol (Fig.5.3, middle panel).

The reduction in crypt cell volume elicited by VIP or carbachol was apparently reversed upon agonist washout. This is shown in the upper and middle panels of Fig.5.3.. VIP induced an SVD of around 26% within 8 min of addition. Subsequent washout led to a progressive recovery of initial volume within 20 min of VIP removal. The addition of carbachol reduced crypt volume by around 31% with a gradual recovery of crypt volume also being observed upon removal of the agonist. The SVD induced by the simultaneous addition of both 100nM VIP and 100m M carbachol was not significantly different from that produced by the addition of either agonist individually (data not shown).

If SVD is due to the passive loss of K⁺ and Cl^- ions from the cytoplasm down their respective electrochemical gradients, then it should be prevented by the elimination of one of these gradients or by the inhibition of one of the conductive pathways. Experiments were performed in which the concentration of extracellular K⁺ was increased from 6 to 20 mM to determine the effect of this manoeuvre upon the secretagogue-induced SVD. The upper two panels of Fig.5.3 show that increasing extracellular K⁺ from 6 to 20 mM abolished SVD induced by both VIP and carbachol. Increasing extracellular K⁺ in the absence of secretagogues did not produce a sizeable change in crypt volume (Fig.5.3, bottom panel).

The sensitivity of SVD to Cl^- channel blockade was tested using the inhibitors anthracene-9-carboxylic acid (9-AC) and 5-nitro-2-(3-phenylpropyl-amino)-benzoic acid (NPPB), which both evoke changes in crypt E_m (Fig.4.7). Fig.5.4 illustrates the effect of 100nM VIP upon mean crypt volume in the absence and presence of 100m M NPPB. Within 8 minutes of VIP addition a maximal SVD had taken place with a 24% decrease in crypt volume, which subsequently recovered following agonist washout. However the addition of 100 nM VIP in the presence of 100 m M NPPB failed to elicit any detectable SVD (Fig.5.4., upper panel). Addition of 100 m M NPPB alone (Fig.5.4, middle panel) led to only minor changes in volume. The decrease in crypt volume evoked by both secretagogues was partially blocked by 100m M 9-AC, with maximal SVD values of 7 ± 2 and 5 ± 2 % induced by the addition of VIP and carbachol respectively in the presence of 100 m M 9-AC (results not shown). A dose-response relationship for the NPPB-induced inhibition of crypt SVD evoked by VIP was determined (Fig.5.5.), NPPB producing a half-maximal inhibition at approximately 25m M.

Agonist-evoked fluid secretion is believed to occur as a result of the activation of basolateral K⁺ channels and apical Cl^- channels, as both Cl^- and K⁺ are believed to be accumulated above equilibrium in the cytoplasm of exocrine cells (Petersen & Gallacher, 1988). It has been suggested that the loss of cytosolic KCl through conductive pathways selective for these ions is balanced by the uptake of Na⁺, K⁺ and Cl^- ions via the parallel operation of a Na⁺-K⁺ ATPase and a cotransport mechanism (Petersen, 1986). In this scheme there is no necessity for the net loss or uptake of ions, which would be associated with osmotically obliged water fluxes, and therefore substantial changes in cell volume are not predicted by this model. In the present experiments however, both carbachol and VIP, when added at concentrations that produce maximal changes in membrane potential and conductance, cause a decrease in crypt volume of between 20-30% within 4-8 minutes of agonist addition. Similarly secretagogue action in other exocrine cells has been shown to decrease cell volume. Muscarinic stimulation evokes a cell shrinkage in dissociated eccrine clear cells (Suzuki et al., 1991) and in isolated salivary acinar cells (Foskett & Melvin, 1989), and the action of VIP on a colonic carcinoma cell line has been shown to be accompanied by a decrease in intracellular water content (Mandel et al., 1986a).

Elevating extracellular K⁺ from 6 to 20mM abolished the SVD induced by both carbachol and VIP. Such an increase in extracellular K⁺ has been demonstrated to depolarise the resting membrane potential by approximately 18mV in the unstimulated crypt (Fig 3.1), which may prevent SVD either by decreasing the electrical driving force for Cl^- exit and/or the concentration gradient for K⁺ efflux. If we assume intracellular K⁺ and Cl^- activities of 125 and 40 mM respectively (taken from Sullivan & Field, 1991) then we can calculate the changes in the electrochemical gradients for Cl^- and K⁺ efflux from the Nernst equation (Emf_K = E_m - E_K), taking E_m as -49 mV and -31 mV for 6 and 20 mM [K⁺]_o respectively (see chapter 3). The calculated electrochemical potential for K⁺ exit is reduced from 28 mV to 16 mV by such an elevation in [K⁺]_o. An equivalent calculation for Cl^- gives driving forces for Cl^- efflux of 15 and -3 mV respectively. We can therefore conclude that the elimination of the electrochemical gradient for Cl^- exit is the most likely explanation for the blockade of SVD at high [K⁺]_o. Muscarinic stimulation has been shown to evoke a decrease in intracellular Cl^- activity to less than half its value in the unstimulated salivary acinar cell, with the change in [Cl^-]_i quantitatively accounting for the observed changes in cell volume (Foskett, 1990). Therefore intracellular Cl^- activity may effectively limit the extent of the SVD following the agonist-induced activation of membrane conductance pathways.

The interpretation that the depolarisation of E_m evoked by VIP is predominantly due to the activation of a Cl^- conductance is supported by the observation that VIP stimulates a decrease in crypt volume, as the activation of a Na⁺ permeability would be expected to swell, rather than to shrink the crypt. The inhibition of SVD by the established Cl^- channel blockers NPPB and 9-AC is also consistent with the activation of a Cl^- conductance pathway. Two observations concerning the decrease in crypt volume induced by secretagogues remain to be explained. Carbachol appears to exert its main effect via the activation of a K⁺ conductance and yet appears to be as effective as VIP in reducing crypt volume. This may be accounted for by the presence of a constitutively active Cl^- conductance, consistent with the observation that both NPPB and 9-AC evoke hyperpolarisations in unstimulated small intestinal crypts. If this is the case it is therefore surprising that in the absence of secretagogues NPPB does not induce a prolonged swelling of the crypt, rather than the transient increase in volume observed. Alternatively it is possible that carbachol does indeed increase membrane Cl^- conductance by the mobilisation of PKC (Fig.4.3.), as evidence for the presence of a Ca²⁺-activated Cl^- conductance, present in other fluid and electrolyte secreting cells (Cliff et al., 1990; Randriamampita et al., 1988; Wagner et al., 1991), is lacking. However the potentiation of a Cl^- conductance by carbachol was not detectable in electrophysiological recordings, because any change in Cl^- conductance was masked by a much larger effect on K⁺ conductance. The second unexplained feature of SVD in small intestinal crypts is the absence of a regulatory volume increase in crypts maintained in the presence of secretagogue. Such an absence of volume recovery following isosmotic cell shrinkage might, in principle, maintain intracellular K⁺ and Cl^- concentrations favourable to sustained fluid and electrolyte secretion.

If secretagogues do stimulate fluid and electrolyte secretion in small intestinal crypts then fluid would be predicted to flow into the crypt lumen, independently of the volume of the cells. The shrinkage evoked by VIP and carbachol would suggest that if fluid is indeed flowing into the lumen, this finds its way to the mouths of the crypts and does not contribute to expansion of the luminal space. This appears to be confirmed by direct observation which fails to detect the appearance of any fluid-filled spaces in secretagogue-stimulated crypts (see Fig 5.1B).

No regulatory volume increase takes place in the continued presence of secretagogue in small intestinal crypts. It is tempting to speculate that in the new steady-state in the continued presence of agonist a basolateral Na⁺(K⁺)2Cl^- cotransport system has been activated to keep pace with the exit of Cl^-, K⁺ and Na⁺ ions through specific channel conductance pathways for K⁺ and Cl^- and the ATP-dependent Na⁺ pump respectively. It could be argued that a new steady state is reached with shrunken cells having an increased rate of apical Cl^- exit, via an agonist-activated Cl^- conductance, and an enhanced basolateral uptake of Cl^- via an activated cotransport mechanism. Several observations are consistent with these proposals: firstly, VIP is known to produce a sustained effect on Cl^- secretion across the intact small intestinal epithelium and in colonic carcinoma epithelial cell layers (Dharmsathaphorn et al., 1985; Schwartz et al., 1974), in accordance with a new steady state being reached with increased efflux and uptake of the ion. Secondly, under unstimulated conditions the Na⁺(K⁺)2Cl^- cotransporter may be inactive or functioning at low rate, but its activity may be increased following VIP stimulation. This view would be consistent with the rapid volume recovery initiated by VIP removal, where the inactivation of conductance pathways following secretagogue washout allows the activated cotransporter to mediate the net uptake of ions and effect crypt volume recovery. This argument assumes that the inactivation of conductance pathways after the removal of VIP occurs at a faster rate than that of the cotransporter. Eccrine clear cells were also observed to recover their volume after agonist washout, although the mechanism for this volume gain was not investigated (Suzuki et al., 1991).

The increased activity of the cotransporter during secretagogue action might be secondary to increases in second messenger levels, or alternatively the cell shrinkage could itself be the mechanism leading to the activation of a cotransport mechanism. Foskett calculated that the fall in intracellular Cl^- activity following muscarinic stimulation would increase the thermodynamic favourability for Cl^- entry via a cotransport mechanism, and suggested that a lower [Cl^-]_i may be necessary to facilitate Cl^- entry during sustained secretion. In this way the crypt volume decrease may play an important role in increasing the efficiency of secretion.

In the human airway epithelium Ó₁-adrenergic stimulation was reported to increase furosemide-sensitive Cl^- transport (Liedtke, 1989), by a mechanism later determined to be NaCl cotransport (Liedtke, 1992). VIP has been demonstrated to stimulate a bumetanide-sensitive basolateral NaK2Cl cotransport mechanism in T84 cell monolayers (Dharmsathaphorn, 1985). It is tempting to speculate that such a cotransport mechanism is present in the small intestinal crypt epithelium, since villus and crypt units isolated from rat duodenum have been demonstrated to possess a bumetanide-sensitive ⁸⁶Rb⁺ uptake mechanism proposed to be a NaK2Cl cotransporter, although the regulation of this transport system was not investigated further (McNicholas et al., 1992).

Both VIP and carbachol induce a sustained decrease in crypt volume that is consistent with the net loss of KCl through separate K⁺ and Cl^- conductance pathways, as inferred by the inhibition of SVD by Cl^- channel blockers and elevations in extracellular K⁺ and the changes in membrane potential and conductance that are elicited by both agonists. The results presented here support the contention that the crypt enterocyte is a site of fluid and electrolyte secretion in the small intestinal epithelium, with both Cl^- uptake and efflux pathways stimulated following agonist addition.

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