Submucosal enteric neurons of the cavine distal colon are sensitive to hypoosmolar stimuli

Abstract

Key points: Neurons of the enteric submucous plexus are challenged by osmolar fluctuations during digestion and absorption of nutrients. Central neurons are very sensitive to changes in osmolality but knowledge on that issue related to enteric neurons is sparse. The present study focuses on investigation of osmosensitivity of submucosal neurons including potential molecular mediating mechanisms. Results show that submucosal neurons respond to hypoosmolar stimuli with increased activity which is partially mediated by the transient receptor potential vanilloid 4 channel. We provided important information on osmosensitive properties of enteric neurons. These data are fundamental to better explain the nerve-mediated control of the gastrointestinal functions during physiological and pathophysiological (diarrhoea) conditions.

Abstract: Enteric neurons are located inside the gut wall, where they are confronted with changes in osmolality during (inter-) digestive periods. In particular, neurons of the submucous plexus (SMP), located between epithelial cells and blood vessels may sense and respond to osmotic shifts. The present study was conducted to investigate osmosensitivity of enteric submucosal neurons and the potential role of the transient receptor potential vanilloid 4 channel (TRPV4) as a mediator of enteric neuronal osmosensitivity. Therefore, freshly dissected submucosal preparations from guinea pig colon were investigated for osmosensitivity using voltage-sensitive dye and Ca²⁺ imaging. Acute hypoosmolar stimuli (final osmolality reached at ganglia of 94, 144 and 194 mOsm kg^-1 ) were applied to single ganglia using a local perfusion system. Expression of TRPV4 in the SMP was quantified using qRT-PCR, and GSK1016790A and HC-067047 were used to activate or block the receptor, respectively, revealing its relevance in enteric osmosensitivity. On average, 11.0 [7.0/17.0] % of submucosal neurons per ganglion responded to the hypoosmolar stimulus. The Ca²⁺ imaging experiments showed that glia responded to the hypoosmolar stimulus, but with a delay in comparison with neurons. mRNA expression of TRPV4 could be shown in the SMP and blockade of the receptor by HC-067047 significantly decreased the number of responding neurons (0.0 [0.0/6.3] %) while the TRPV4 agonist GSK1016790A caused action potential discharge in a subpopulation of osmosensitive enteric neurons. The results of the present study provide insight into the osmosensitivity of submucosal enteric neurons and strongly indicate the involvement of TRPV4 as an osmotransducer.

Keywords: TRPV4; enteric nervous system; osmosensitivity; submucosal enteric neurons.

Figure 1:. Schematic display of the two different recording paradigms.

A: interval recording with a first recording period of 2 s under isoosmolar conditions and three following recordings (2 s each) under hypoosmolar conditions with 8 s in between the recordings. B: continuous recording lasting for 12 s with 2 s under isoosmolar conditions following 10 s under hypoosmolar conditions.

Figure 2:. Averaged action potential frequency in the three recording periods after application of an isotonic or a hypotonic solution in cells defined as responders.

Action potential frequency was significantly higher after application of the 94 mOsm/kg HEPES-solution compared to isotonic conditions (Wilcoxon signed rank test). n = 7/21/99. Horizontal lines within data points indicate median values.

Figure 3:. Temporal distribution of action potential frequencies after application of an isoosmolar or hypoosmolar solution.

Each symbol represents a certain neuron. Baseline activity was compared to every other recording period under isoosmolar as well as hypoosmolar conditions (Friedman test, p <0.0001). Comparison of mean action potential frequency between iso- and hypoosmolar in one recording period was performed with Dunn’s multiple comparison test and significant differences are indicated by asteriks. n = 7/21/33. Horizontal lines within data points indicate median values.

Figure 4:. Comparison of the neuronal responses to three hypoosmotic stimuli.

Action potential frequency of the neurons responding to the 194 mOsm/kg HEPES-solution was significantly lower compared to that of the neurons responding to 94 and 144 mOsm/kg HEPES-solution (Kruskal Wallis test; exact p values were calculated with Dunn’s multiple comparison test and significant differences are indicated by asteriks; p < 0.01). N numbers indicate guinea pigs/ganglia/tested neurons (osmosensitive neurons). Horizontal lines within data points indicate median values.

Figure 5:. Ca²⁺- response after application of a hypoosmolar HEPES-solution.

A: fluorescence of the calcium indicator Fluo-4 before (left) and after (right) application of a 94 mOsm/kg HEPES-solution onto a ganglion of the SMP (recorded with a framerate of 2Hz). B: change of fluorescence of the cell responding in A) over a time course of one minute during recording period 3.

Figure 6:. Latency of Ca²⁺- response and action potential discharge after hypoosmolar stimulation of enteric neurons.

Increase in [Ca²⁺]_in started significantly earlier than the action potential discharge observed in experiments using the VSD technique. Unpaired t test. For Ca²⁺- experiments: n = 3/6/31, for VSD experiments n = 2/3/13. Horizontal lines within data points indicate median values.

Figure 7:. Relative mRNA expression of TRPV4 in kidney, bladder, submucosa and muscle.

Ordinary one way ANOVA (p < 0.001). Comparison of individual specimen by Dunn’s multiple comparison test. n = 3 to 6 guinea pigs. Horizontal lines within data points indicate median values.

Figure 8:. Effect of specific TRPV4 blocker HC-067047 on the neuronal response to hypoosmolar stimulation.

A: percentage of responding cells after hypooosmolar stimulation with and without HC-067047; Mann-Whitney test. B: action potential frequency in responding cells after hypoosmolar stimulation with and without HC-067047. Mann-Whitney test. n = 3/8/19 (94 mOsm/kg), n= 3/10/7 (94 mOsm/kg + HC-0670479). Horizontal lines within data points indicate median values.

Figure 9:. Neuronal activity pattern after GSK1016790A and after hypoosmolar stimulation.

A: ganglion stained with Di-8-ANEPPS and exposed to 20 μM GSK1016790A and 94 mOsm/kg consecutively. White arrow indicates a cell responding to both stimuli. B: traces recorded from the cell displayed in A. The response to GSK1016790A was measured in the 4 s following the agonist application while the response to hypoosmolar stimulus during period 10–12 s of the interval recording paradigm.