Co-Cultures Provide a New Tool to Probe Communication Between Adult Sensory Neurons and Urothelium

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Co-Cultures Provide a New Tool to Probe Communication Between Adult Sensory Neurons and Urothelium Lauren M. O’Mullane,* Janet R. Keast†,‡ and Peregrine B. Osborne‡,§ From the Pain Management Research Institute and Kolling Institute, University of Sydney at Royal North Shore Hospital, St. Leonards, New South Wales, Australia

Purpose: Recent evidence suggests that the urothelium functions as a sensory transducer of chemical, mechanical or thermal stimuli and signals to nerve terminals and other cells in the bladder wall. The cellular and molecular basis of neuro-urothelial communication is not easily studied in the intact bladder. This led us to establish a method of co-culturing dorsal root ganglion sensory neurons and bladder urothelial cells. Materials and Methods: Sensory neurons and urothelial cells obtained from dorsal root ganglia and bladders dissected from adult female Sprague-Dawley® rats were isolated by enzyme treatment and mechanical dissociation. They were plated together or separately on collagen coated substrate and cultured in keratinocyte medium for 48 to 72 hours. Retrograde tracer labeling was performed to identify bladder afferents used for functional testing. Results: Neurite growth and complexity in neurons co-cultured with urothelial cells was increased relative to that in neuronal monocultures. The growth promoting effect of urothelial cells was reduced by the tyrosine kinase inhibitor K252a but upstream inhibition of NGF signaling with TrkA-Fc had no effect. Fura-2 calcium imaging of urothelial cells showed responses to ATP (100 ␮M) and activation of TRPV4 (4␣-PDD, 10 ␮M) but not TRPV1 (capsaicin, 1 ␮M), TRPV3 (farnesyl pyrophosphate, 1 ␮M) or TRPA1 (mustard oil, 100 ␮M). In contrast, co-cultured neurons were activated by all agonists except farnesyl pyrophosphate. Conclusions: Co-culturing provides a new methodology for investigating neurourothelial interactions in animal models of urological conditions. Results suggest that neuronal properties are maintained in the presence of urothelium and neurite growth is potentiated by an NGF independent mechanism. Key Words: urinary bladder; urothelium; ganglia, spinal; sensory receptor cells; coculture techniques SPECIALIZED epithelial cells in the urothelium lining the lower urinary tract respond rapidly to environmental changes by altering their structure and chemistry, and releasing signaling molecules to influence nearby cells.1,2 Neuroactive substances (ATP and nitric oxide) released by urothelial cells in response to stretch and chemical stimulation were suggested to communicate with sensory nerve

terminals in normal bladder functions, eg activation of afferents during bladder filling, and pathological states, eg bladder hyperexcitability and pain.1– 4 Further progress in understanding the molecular basis of functional interactions of the urothelium and bladder innervation has been hindered by the complexity of the 2 tissue types and the technical difficulty of directly

0022-5347/13/1902-0001/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION

http://dx.doi.org/10.1016/j.juro.2013.01.048 Vol. 190, 000, August 2013 RESEARCH, INC. Printed in U.S.A.

AND

Abbreviations and Acronyms ATP ⫽ adenosine triphosphate [Ca2⫹]in ⫽ intracellular calcium CGRP ⫽ calcitonin gene-related peptide DAPI ⫽ 4=, 6-diamidino-2phenylindole dihydrochloride DRG ⫽ dorsal root ganglion FPP ⫽ farnesyl pyrophosphate Fura-2 ⫽ fura-2-acetoxymethyl ester KSFM ⫽ keratinocyte serum-free medium MEM ⫽ minimum essential medium NGF ⫽ nerve growth factor TK ⫽ tyrosine kinase TRP ⫽ transient receptor potential Tuj1 ⫽ ␤-tubulin isotype III Accepted for publication January 16, 2013. Study received institutional animal care and use committee approval. Supported by National Institute of Diabetes and Digestive and Kidney Diseases Award R01DK069351, and National Health and Medical Research Council of Australia Project Grant 1003512 and Senior Research Fellowship 632903. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIDDK or the National Institutes of Health. * Current address: Department of Pharmacology, University of Sydney, New South Wales 2006, Australia. † Correspondence: Department of Anatomy and Neuroscience, University of Melbourne, Victoria 3010, Australia (telephone: ⫹61 3 8344 5804; e-mail: [email protected]). ‡ Equal study contribution. § Current address: Department of Anatomy and Neuroscience, University of Melbourne, Victoria 3010, Australia.

www.jurology.com

1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

measuring interactions in vivo. Many populations of sensory nerves innervate the bladder, expressing different transmitters, neuropeptides and other markers, and urothelial cells are also heterogeneous.2,3 Functional studies showed that numerous channels and receptors are expressed by neurons and urothelial cells, limiting the interpretation of experiments using conventional pharmacological or global gene deletion approaches. Cultures containing 2 distinct cell types (co-cultures) provide a powerful way to examine the mechanisms of intercellular communication. This approach has been adopted for a number of neuronal targets, such as skin, tracheal epithelium, cardiac muscle and vascular smooth muscle.5– 8 To our knowledge this method has not yet been applied to studies of the urothelium and its sensory nerve supply. We previously refined a culture method to study the physiological, pharmacological and growth properties of isolated adult rat sensory neurons.9,10 Methods of culturing isolated urothelial cells from adult rodents are also well established.11 We established co-cultures of bladder sensory neurons and urothelial cells, isolating each cell type from adult rats, using 2 types of output measures, including analysis of neuronal growth and live cell imaging, to determine the functional expression of key receptors and channels. The outcome of our study was establishment of a viable co-culture system and identification of a robust, growth promoting effect of urothelial cells on sensory neurons.

MATERIALS AND METHODS Animals Female Sprague-Dawley rats (Animal Resources Centre, Perth, Australia) 8 to 12 weeks old were used according to procedures approved by the institutional ethics committee and the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes (National Health and Medical Research Council, Australia). Before tissue removal, rats were anesthetized (sodium pentobarbitone 80 mg/kg intraperitoneally) and decapitated. The estrus cycle was monitored but not controlled for since no effect was detected.

DRG Cell Culture DRGs containing bladder projecting sensory neurons12,13 were removed from spinal levels L6/S1 and neurons were isolated and cultured.10 Briefly, ganglia collected in Tyrode solution (130 mmol/l NaCl, 20 mmol/l NaHCO3, 3 mmol/l KCl, 4 CaCl2, 2 mmol/l MgCl2, 10 mmol/l HEPES and 1 mmol/l glucose with antibiotic-antimycotic) were dissociated in collagenase type I (Worthington Biochemical, West Chester, Pennsylvania)/0.25% trypsin (Invitrogen™) at 37C for 1 hour. They were washed and triturated before centrifugation at 900 rpm for 10 minutes in 15% bovine serum albumin (Sigma-Aldrich®). The pellet was resus-

pended in Neurobasal®-A containing B27 supplement, glutamine and antibiotic-antimycotic (Invitrogen), and plated on polyornithine (500 ␮g/ml0 (Sigma-Aldrich)/ laminin (5 ␮g/ml) (Invitrogen) coated glass coverslips. All cultures were incubated in 5% CO2 and used 48 to 72 hours after isolation.

Urothelial Cell Culture Urothelial cell isolation and culture were based on a published protocol.14 Dissected bladders with urothelium uppermost were pinned flat to a Sylgard® plate in Krebs solution (6.90 mmol/l NaCl, 0.37 mmol/l KCl, 0.27 mmol/l MgSO4, 0.17 mmol/l NaH2PO4, 2.10 mmol/l NaHCO3, 0.99 mmol/l glucose and 0.18 mmol/l CaCl2 in 5% CO2 in O2). After washing with MEM and incubating at 37C for 3 to 4 hours in MEM/dispase (0.25 mg/ml0 (Sigma-Aldrich), the urothelium was removed by gentle scraping with a scalpel blade. It was resuspended in 0.25% trypsin/ethylenediaminetetraacetic acid solution (Invitrogen) and incubated at 37C for 30 minutes with gentle trituration every 10 minutes. The cell suspension was centrifuged at 1,500 rpm for 15 minutes in a tube containing MEM/10% fetal bovine serum (Invitrogen). The pellet was resuspended in KSFM (Invitrogen), centrifuged, resuspended in KSFM and plated on collagen coated (250 ␮g collagen/ml 0.02N acetic acid) (Sigma-Aldrich) glass coverslips.

Urothelial-Neuronal Co-Cultures As described, pilot studies were done to determine whether KSFM medium was most suitable for co-culture experiments, of which most used the following protocol. Urothelial and DRG cells from the same animal were isolated using the protocols described for monoculture but substituting KSFM for the final resuspension. Urothelial cells were plated 45 minutes after neurons on collagen coated coverslips and maintained for 16 hours at 37C before washing in KSFM. They were incubated until use. Some cultures were treated with NGF (10 ng/ml) (Sigma), K252a (100 nM) (Biomol Research Laboratories, Plymouth Meeting, Pennsylvania) or recombinant rat TrkA/Fc chimera (4 ng/ml) (R&D Systems®). These concentrations are effective in cultures of adult rodent neurons.15 For some qualitative immunocytochemical studies co-cultures were maintained for 72 hours (2), used neurons and urothelial cells from different rats (3) or grew urothelial cells 24 hours before adding neurons from a second rat (2).

Bladder Afferent Neuron Retrograde Labeling Bladder projecting DRG neurons were labeled by injecting fluorescent tracer in the bladder 1 to 2 weeks before tissue removal.16 DiI (2 mg/ml in dimethyl sulfoxide) was chosen because the excitation/emission wavelengths (549/565 nm) are compatible with Fura-2 live cell calcium imaging. With the rat under anesthesia (60 mg/kg ketamine and 10 mg/kg xylazine intraperitoneally) the bladder was exposed via an abdominal incision and DiI (total 10 to 20 ␮l) was microinjected at 4 sites. The musculature and skin were sutured separately with silk. Tissues were removed for culture 1 to 2 weeks later.

Immunocytochemistry Cultures were fixed (4% paraformaldehyde in 0.1M phosphate buffer) and processed for immunocytochemistry (see

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142T1 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184

3

Antibodies and labels used to visualize cultured cells Antibody or label Anti-Tuj1 Anti-CGRP Cytokeratin: 17 20 Cy2 conjugated IgG: Anti-mouse Anti-goat Alexa Fluor® 488 phalloidin DAPI

Purpose

Working Dilution

General neuronal marker Peptidergic nociceptor sensory neurons marker

1:200 1:2,000

Mouse Goat

Sigma-Aldrich Biogenesis, Poole, United Kingdom

Basal urothelial cell marker Urothelial umbrella cell marker

1:1,000 1:100

Mouse Mouse

DakoCytomation, Campbellfield, Australia DakoCytomation

Visualizing cells binding anti-Tuj1 and anti-cytokeratin antibodies Visualizing cells binding anti-CGRP antibody Labeling filamentous actin, showing all cell types Labeling nuclei of all cell types

1:400

Donkey

1:2,000

Donkey

Jackson Immunoresearch Laboratories, West Grove, PA Jackson Immunoresearch Laboratories

1:40

Not applicable

Invitrogen

1 ␮g/ml

Not applicable

Sigma-Aldrich

table 1), as described previously.9 Digital analysis was performed on 8-bit monochrome images captured using a BX-51 fluorescence microscope (Olympus®). Neurite initiation was defined as neurons with 1 or greater Tuj1⫹ neurites longer than the soma diameter. Total neurite length and neurite branch points were quantified using HCA-Vision (CSIRO, North Ryde, Australia). Parameters measured in 30 or greater neurons per culture and cultures from 3 rats or greater were used for statistical analysis.

Cultured Cell Calcium Imaging Cells were loaded with 5 ␮M Fura-2 (Anaspec, Fremont, California) in 0.02% Pluronic® at room temperature for 45 minutes in culture medium.17 [Ca2⫹]in was measured as the ratio of fluorescence emitted by alternating 340/380 nm excitation. Isolated neurons and groups of 5 to 10 urothelial cells were imaged on coverslips in a superfusion chamber using an IX-71 microscope (Olympus), Polychrome V monochromator (TILL Photonics, Rochester, New York) and an iXonEM⫹ camera (Andor™ Technology). Cells were constantly superfused at room temperature with 140 mmol/l NaCl, 4 mmol/l KCl, 10 mmol/l HEPES, 2 mmol/l MgCl2, 2 mmol/l CaCl2 and 10 mmol/l glucose, pH 7.4. Pharmacological agents were applied with a Perfusion Fast Step SF77B system (Warner Instruments, Hamden, Connecticut). Fura-2 ratios were calculated using Andor iQ 1.10.2. Farnesyl pyrophosphate, phorbol 12,13-didecanoate and icilin were obtained from Enzo Life Sciences, Plymouth Meeting, Pennsylvania. All other agents were obtained from Sigma-Aldrich. Pharmacological agents were tested on neurons in which depolarization by high K⫹ (30 mM) increased [Ca2⫹]in by greater than 20% over baseline.17

Statistical Analysis SPSS®, version 20 or Prism® 6.0b was used for statistical analysis. Planned Holm-Sidak multiple comparisons were used for neurite initiation data. Two-way mixed model ANOVA was used for separate comparisons of total neurite length and neurite branch points. The factors were culture type (2 levels, including neuronal culture and neuron/urothelial co-culture) and neuron class (2 levels, including CGRP positive and negative). All factorial combi-

Host Species

Source

nations were randomly assigned to 1 or more culture plates obtained from 1 rat, which were analyzed as within subject or matched data using repeated measurements ANOVA.

RESULTS Structural Features DRG and urothelial monocultures. The growth pattern of L6-S1 DRG neurons cultured on lamininpolyornithine was similar to that in previous studies of adult rat DRG cultures.9 Neurons grown without neurotrophic factor showed diverse morphology after 48 hours. Most neurons had grown 1 or more neurites with a morphology ranging from highly complex branched neurites to a simple structure with few branches (fig. 1, a). Glial cells were closely associated with the soma and primary neurites of most neurons or they were in small clumps distant from neurons (fig. 1, b). NGF (10 ng/ml for 48 hours) stimulated exuberant outgrowth of neurites from many neurons (fig. 1, c and d). Urothelial cell monocultures resembled those in previous studies.11,14 Single or small islands of cells (fewer than 10 cells per island) predominated after 24 hours in culture. By 72 hours larger areas of closely associated urothelial cells had formed (fig. 1, e and f), including cells that were cytokeratin 17⫹ (marker of basal cells) or cytokeratin 20⫹ (marker of umbrella cells) (fig. 1, g and h).2 Co-cultures. In pilot studies urothelial cells adhered poorly to neuronal substrate (laminin-polyornithine) and did not grow in neuronal culture medium (Neurobasal A) with few cells surviving after 24 hours. However, neurons adhered to the collagen substrate used for urothelial cell culture and grew in the urothelial culture medium, KSFM. Since neurons and urothelial cells isolated from the same or different rats grew equally well, as described, in

123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 F1 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184

4

185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 C 201 O 202 L 203 O 204 R 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220F2 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245F3 246

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

a

b

c

d

e

f

g

h

Figure 1. Structural and immunohistochemical properties of neuronal and urothelial monocultures. DRG monoculture was grown on polyornithine-laminin substrate for 48 hours (a). Five neurons are shown, immunostained for ␤-tubulin (green areas) with contrast slightly enhanced to enable better visualization of neurites. DAPI nuclear stain (blue areas) shows glial cell distribution but it is not readily visible in neurons due to immunolabeling intensity. DRG neuron grown and labeled as in a but at higher magnification to show typical glial cell distribution (arrows) (b). DRG neurons grown and labeled as for a but with 10 ng/ml NGF in culture (c and d). Arrows indicate 2 neuronal somata (c). Urothelial monocultures grown for 48 hours on collagen and stained with phalloidin-fluorescein isothiocyanate to show filamentous actin (e). Urothelial monocultures grown on collagen for 72 hours (f). Urothelial monoculture grown for 72 hours (g and h), immunolabeled for cytokeratin 17 or 20 (red areas) (g) and co-stained for phalloidin-fluorescein isothiocyanate (green areas). Scale bar indicates 60 (a), 30 (b to f) and 20 ␮m (g and h).

subsequent experiments we used the former approach to minimize animal use. There was great diversity in the types of neuronal growth and in the spatial relationships between neurons and urothelial cells (fig. 2). Most neurons grew long, branched neurites (fig. 2, a). However, the exuberance of neurite growth and proximity to urothelial cells did not appear to correlate, which suggests that the trophic effect of urothelium could be due to a factor(s) secreted into the medium. Many neurons located close to islands of urothelial cells extended neurites to envelope them (fig. 2, b to e). Varicose neurites were often closely opposed to urothelial cells (fig. 2, b, c and e). Some appeared to form terminations but others traversed islands of urothelial cells without terminating (fig. 2, f to h). These features were comparable in co-cultures in which urothelial cells and neurons were from different rats and they were independent of the timing of neurons added to culture. Neurons and urothelial cells containing retrograde tracer dye had structural features similar to those of unlabeled cells (fig. 2, i to k). Neurite initiation was low (mean ⫾ SEM 15% ⫾ 10%) in DRG monocultures grown for 48 hours under urothelial culture conditions or in the absence of exogenous NGF. This increased to a mean of 78% ⫾ 4.8% when neurons were cultured with urothelial cells under identical conditions (fig. 3, a). The structure of peptidergic unmyelinated bladder afferents

is regulated by NGF derived from the urothelium and other tissues in normal and diseased bladders.18 To determine whether the neurotrophic effect of urothelial cells depended on TK signaling, which is required for downstream signaling by the NGF receptor TrkA,19 we tested the broad spectrum inhibitor K252a (100 nM). This significantly reduced neurite initiation to a mean of 34% ⫾ 13% in co-cultures (fig. 3, a). We next used a recombinant TrkA-Fc chimera (4 ng/ml)15 as an NGF decoy to determine whether upstream neuronal NGF/TrkA receptors were activated in co-cultures. In this experiment identification of the neuronal phenotype revealed no significant effect of TrkA/Fc treatment on neurite initiation in co-cultured peptidergic and nonpeptidergic neurons (p ⫽ 0.546 and 0.171, respectively, each 4 df). However, in a positive control monoculture treatment with TrkA/Fc attenuated neurite initiation in peptidergic neurons treated with NGF (fig. 3, c). Neither TrkA/Fc nor NGF had a significant effect on the proportion of CGRP⫹ neurons in cocultures and there was no effect on neurite initiation in nonpeptidergic neurons in monoculture (fig. 3, b and d). Analysis of untreated control co-cultures showed that the neurotrophic effect of urothelial cells was caused by increased neurite initiation in CGRP⫹ neurons but not in CGRP– nonpeptidergic neurons (fig. 3, c and d). Briefly, co-culturing with urothelial cells preferentially stimulated neurite ini-

185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 C 200 O 201 L 202 O R 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 C 281 O 282 L 283 O 284 R 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308

a

b

c

d

e

f

g

h

i

j

k

5

Figure 2. Structural and immunohistochemical properties of neuronal and urothelial co-cultures. Note 48 to 72-hour co-cultures of cells isolated from controls (a to h). Neurons were immunolabeled for ␤-tubulin (red areas), urothelial cells were identified with phalloidinfluorescein isothiocyanate (green areas) and nuclei were labeled with DAPI (blue areas) (a and b). Urothelial cells were digitally recolored blue and DAPI was recolored green to facilitate visualization of closely associated nerve fibers (red areas), while in some cases nuclear staining was omitted for simplicity (c to h). Low magnification reveals typical co-culture (a). Arrows indicate examples of neuronal somata. Glial cell nuclei shows unstained cytoplasm. Neurites enveloping urothelial cell island (b), enveloping small cluster of urothelial cells but not approaching other single urothelial cells growing nearby (c), and enveloping clusters and single urothelial cells (d). Dense networks of neurites associated with urothelial cells show potential termination sites (e and f). Neurites traveling near urothelial cells but forming few or no close association sites with urothelial cells (g and h). Co-cultures from rats in which bladder was previously microinjected in vivo with DiI (red areas) (i to k), which is apparent in many neurons (i and j) and urothelial cells (j and k). Neurons were labeled with ␤-tubulin (green areas) (i and k). Urothelial cells were labeled with phalloidin-fluorescein isothiocyanate (j). Scale bar represents 50 (a), 30 (b, j and k) and 20 ␮m (c to i).

tiation in peptidergic sensory neurons. This depended on tyrosine kinase signaling but not on NGF/TrkA. Further analysis of neurite outgrowth revealed no significant difference in the effect of co-culturing on peptidergic and nonpeptidergic sensory neurons. There was a significant main effect of culture type (neuronal culture or neuronal/urothelial co-culture) on total neurite length and the number of branch

points (fig. 3, e and f). However, we detected neither a main effect of neuronal class nor significant interaction. Co-Cultured Neuron and Urothelial Cell Functional Characterization Sensory neurons and urothelial cells in co-culture were also assessed using ratiometric Fura-2 imaging to detect agonist induced changes in [Ca2⫹]in. Neu-

247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 C 280 O 281 L 282 O 283 R 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308

b

10

c

F

/F

re

G

kA Tr

N

G

n+

F

u

+

cu co

n

52 K2

+ re

Tr

a

ltu re kA /F c

0

co

cu

ltu

n+

u

n

co

cu

cu

ltu

ltu

re

re

0

20

ltu

20

30

N

40

40

+

60

50

cu

P < .001 P = .01

80

+

neurite initiation (% freq)

100

CGRP-IR neurons (% of total)

a

peptidergic neurons (CGRP-IR) p < .001 p < .001 p < .001

40

20

non-peptidergic neurons (CGRP -ve) 60

40

20

0

F /F c

ronal recordings were restricted to neurons with a soma diameter of less than 30 ␮m. Rapid local superfusion with high K⫹ (30 mM) transiently increased [Ca2⫹]in in 53 DRG neurons in co-culture (mean ⌬F 0.94 ⫾ 0.05 AU). These Ca2⫹ transients typically peaked during the 60-second period of high K⫹ superfusion and returned to baseline within 2 to 3 minutes of washout. Figure 4 shows typical recordings and figure 5 shows group data. Ca2⫹ transients were induced by selective agonists of TRPV120 (1 ␮M capsaicin), TRPV421,22 (10 ␮M 4␣-PDD) and TRPA120 (100 ␮M mustard oil). Ca2⫹ transients were also induced by a cannabinoid that is a nonselective TRPA1 agonist (30 ␮M ⌬9THC). Agonists of TRPV323 (1 ␮M farnesyl pyrophosphate, 0 of 3) and TRPM822 (1 ␮M icilin, 0 of 6) did not induce Ca2⫹ transients in these recordings. The purinoceptor agonist ATP (100 ␮M) induced small Ca2⫹ transients. Only 2 of 6 neurons showed a

Co-cultured DRG neurons

Tr

kA

G

N

+

ltu cu

high K+

1.5

Capsaicin, 1μM

1.5

G

u

600 400 200 0

+

-

f

+

-

peptidergic neurons (CGRP-IR)

n culture 25

.5

n+u co-culture 0

p = .009

0 0

10 Time (min)

20 high K+

1.5

10

Capsaicin, 1μM

1.5

1

5 0

0

-

+

high K+

4α-PDD, 10μM

.5

0

-

0 0

10 Time (min)

peptidergic neurons (CGRP-IR) high K+

1.5

20

mustard oil, 100μM

0 1.5

high K+

10 Time (min)

20

Δ9-THC

ΔF

1

ΔF

1 .5

.5

0

0 0

c

10 Time (min)

20

0

10 Time (min)

20

Co-cultured urothelial cells ATP, 100μM 4α-PDD, 10μM

1.5

1.5

ATP, 100μM Capsaicin, 1μM mustard oil, 100μM

1 ΔF

1 ΔF

Figure 3. Enhanced neurite initiation and growth in sensory neurons co-cultured with urothelial cells (n⫹u). Data were analyzed by Holm-Sidak multiple comparison test (a to c). n, neuron. Group data indicate co-culturing significantly increased mean neurite initiation, as shown as percent frequency of neurons with neurites (10 df) (a). Neurite initiation was significantly decreased in co-cultures after 24-hour treatment with 100 nM K252a (10 df). Boxplots show that percent frequency of CGRP immunoreactive (IR) neurons was not significantly affected by co-culturing and 24-hour treatments with 10 ng/ml NGF or 4 ng/ml NGF plus TrkA/Fc (4 df) (b). Group data show that co-culturing significantly increased mean neurite initiation in peptidergic, CGRP immunoreactive sensory neurons but not in nonpeptidergic neurons (16 df) (c). Neurite initiation in co-cultured sensory neurons was not significantly affected by 24hour treatment with 4 ng/ml TrkA/Fc. In neuronal monocultures 10 ng/ml NGF for 24 hours significantly increased neurite initiation in peptidergic neurons but not nonpeptidergic neurons. This NGF effect was reduced by TrkA/Fc. Image analysis of neurite growth reveals that co-culturing significantly increased mean total neurite length (F[1,6] ⫽ 15.67) and number of branch points (F[1,6] ⫽ 14.62, 2-way repeated measurements ANOVA, main effect of culture type) (d). ve, vehicle. Peptidergic and nonpeptidergic neurons showed no significant difference in neurite length (F[1,6] ⫽ 2.844, p ⫽ 0.143) (e) or branch points (F[1,6] ⫽ 2.513, p ⫽ 0.164, main effect of neuron class) (f). Summary data are shown as mean ⫾ SEM.

20

1

.5

+

10 Time (min)

Co-cultured bladder afferent neurons

b

15

20

ΔF

n+u co-culture

p = .008

4α-PDD, 10μM

1

ΔF

N

n+

1 .5

ΔF

n culture 800

Neurite branch points (freq)

e

high K+

F

+

n

Tr

-c co

re

ul tu re kA /F c

T r GF kA /F c

ltu

+ N

n+

G

u

F

+

n

cu

N

re

c /F

kA

Tr +

co

-c

ul

tu

re

0

+

Neurite initiation (% freq)

60

d Neurite initiation (% freq)

c

Total neurite length (µm)

309 310 311 312 313 314 315 316 317F4 318F5 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

ΔF

6

.5

.5

0

0 0

10 Time (min)

20

0

10 Time (min)

20

Figure 4. Fura-2 ratiometric imaging of [Ca2⫹]i in co-cultured sensory neurons and urothelial cells. Example plots show ⌬F (change in peak 340 nm/380 nm fluorescence emission ratio of Fura-2 from baseline) with time in minutes, illustrating agonist effect on, a, unlabeled DRG neurons (a), bladder afferent neurons labeled with DiI retrograde tracer dye (b) and urothelial cells (c). All agonists were locally applied by rapid superfusion. High K⫹ was applied at 30 mM concentration in isosmotic solution. All other agonists were applied at concentrations shown. Data are from single neurons and from greater than 5 contiguous urothelial cells within larger cell islands.

309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 C 355 O 356 L 357 O 358 R 359 360 361 362 363 364 365 366 367 368 369 370

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

Co-cultured DRG neurons

a

(4) (5)

1.5 (31)

(10)

(6)

F

1.0 0.5

(6) (8) (6) (5)

(3)

FP P M) D (1 D M M (10 ) O 9 - (10 M ) TH 0 m C ( M) en 30 th o l M) i c (1m illi M AT n ( 1 ) , M -M P ( eA 1 0 0 ) TP M (1 ) 0 M ) 4

ca

ps

-P

ai

ci

n

H

ig

(1

h

K+

0.0

Co-cultured bladder afferent neurons (6)

b 1.5 (34)

(13)

1.0

(9)

(13)

(3)

F

(4) 0.5

)

M

0

m

(1 TP

eA

Ca2⫹ transients were induced in co-cultured bladder afferents by capsaicin (1 ␮M), 4␣-PDD (10 ␮M) and mustard oil (100 ␮M). The nonselective TRPA1 agonist ⌬9-THC (30 ␮M) also induced Ca2⫹ transients and the TRPM8 agonist menthol (1 mM) induced Ca2⫹ transients in 2 of 4 neurons (⌬F 0.77 and 0.13 AU, respectively). ATP (100 ␮M) reliably induced Ca2⫹ transients in urothelial cells in co-culture (figs. 4 and 5). These responses were measured in islands of 5 or more urothelial cells. Time to peak and the amplitude of Ca2⫹ transients were variable. In some recordings the responses peaked early and decayed during agonist application, while others plateaued or were still increasing at the end of agonist application. The P2X2/3 selective agonist agonist ␣,␤,MeATP had no effect on [Ca2⫹]in in 4 recordings from co-cultured urothelial cells. Only urothelial cells that responded to ATP were used to evaluate TRP channel expression. The TRPV4 agonist 4␣-PDD (10 ␮M) induced amplitude Ca2⫹ transients (mean 1.8 ⫾ 0.65 AU) after a delay in 4 of 8 recordings from co-cultured urothelial cells. Ca2⫹ transients were not induced in recordings from co-cultured urothelial cells by any of the other TRP agonists tested.

-M

m

,

4

M

)

) ol th en

TH 9-

M

(1

0 (3 C

(1 O

D D -P

M

M

) 0 (1

(1 n ci ai

00

M

) M

K h ig H ps ca

)

0.0

c

DISCUSSION

Co-cultured urothelial cells 1.5 (7)

(19) F

1.0 0.5 (5)

(4)

(7)

(5)

(6)

) 0 (1

TP eA

,

-M

en m

M

M m (1

th

ol

(1 O M

)

) 00

M 0 (1

D D -P

M

)

M ) (1

FP P 4a

ca

ps

ai

P

ci

n

(1

(1

00

M

M

)

)

0.0

AT

371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

7

Figure 5. Calcium transients activated by high potassium and agonists in co-cultured sensory neurons and urothelial cells. Plots of group data show agonist effect on ⌬F (change in peak 340 nm/380 nm fluorescence emission ratio of Fura-2 from baseline) in unidentified DRG neurons (a), retrograde labeled bladder neurons (b) and urothelial cells (c). Data are shown as mean ⫾ SEM. Values in parentheses indicate number of replicates.

Ca2⫹ transients in response to the selective P2X1/3 agonist ␣,␤,MeATP (⌬F 0.20 and 0.32, respectively). Recordings were also made from sacral bladder afferents co-cultured with urothelial cells (figs. 4 and 5), in which high K⫹ induced Ca2⫹ transients (mean 0.84 ⫾ 0.05 AU) in 48 DiI labeled neurons.

The urothelium is suggested to interact with the bladder sensory innervation and function in sensory transduction.2 The molecular mechanism of this interaction has been difficult to define. We report a co-culture method in which neurons and urothelial cells grown together retain structural, immunohistochemical and functional properties previously identified in acutely dissociated preparations. To our knowledge this offers a new approach for understanding neuro-urothelial communication in the healthy bladder and in animal models of clinical conditions, eg cystitis, overactive bladder and spinal cord injury. While this opens up many opportunities, it is also important to assess whether urothelial cells and/or neurons in these co-cultures undergo plastic changes during the culture period. The strong neurotrophic effect of urothelial cells was similar to the trophic effects of keratinocytes on sensory neurons.8,24 Our preliminary attempts to identify the factor responsible showed that TK but not TrkA signaling was involved. A study of neonatal sensory neurons also identified a K252a sensitive neurotrophic effect of keratinocytes that was reduced by steroidogenesis inhibitors.24 A possible role for urothelial derived steroids in our adult neuron cultures would be of great interest in the context of bladder modulation.

371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

8

433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

Functional responses to agonists were detected in small diameter, presumed unmyelinated neurons25 and urothelial cells in co-culture by measuring changes in [Ca2⫹]in with Fura-2 ratiometric imaging. The robust Ca2⫹ transients induced by high K⫹ in most neurons suggest that hyperpolarized resting membrane potential and neuron excitability were maintained in co-culture. This is consistent with evidence that Ca2⫹ transients activated by high K⫹ in rat lumbar DRG sensory neurons are caused by membrane depolarization and the subsequent influx of extracellular Ca2⫹ through voltage-activated calcium channels.17 Agonists of TRPV1, TRPA1 and TRPV4 also induced Ca2⫹ transients in sacral DRG neurons and bladder afferents in co-culture but these channels are permeable to calcium and a broad range of monovalent cations. Our experiments using the TRPV1 agonist capsaicin were in agreement with electrophysiological data suggesting that 70% of rat bladder afferents are capsaicin sensitive.12 Functional expression of TRPA1 in bladder afferents has not been previously demonstrated in primary cultures but it is inferred from immunohistochemistry, in situ hybridization and physiological studies.26 The TRPA1 agonist used in our experiments, mustard oil, can activate TRPV1 at millimolar concentrations but it is selective for TRPA1 at the micromolar concentration used.20 We are unaware of prior evidence of TRPV4 expression in bladder afferents but strong evidence suggests that TRPV4 expressed by urothelial cells regulates urinary tract function.26 TRPV4 is a thermo-TRP but the functioning of TRPV4 in mechanosensation may have greater functional relevance in bladder afferents.27 Neuronal Ca2⫹ transients activated by ⌬9-THC, which increases [Ca2⫹]in in DRG sensory neurons by activating TRPV1 and TRPA1, may be clinically relevant since desensitization of these TRPs by canna-

binoids induces peripheral antinociception and antihyperalgesia.28 Only 2 of 10 recordings of bladder afferents showed a response to the TRPM8 agonist menthol but this channel is mostly expressed in the A␦ class of DRG neurons,26 which have a soma diameter of greater than 30 ␮m and were excluded from our recordings. To our knowledge TRPV3 has not been reported in bladder afferents and FPP, a selective agonist of TRPV3,23 had no effect. Ca2⫹ transients in co-cultured urothelial cells were consistently activated by ATP, which was previously used as a positive control in calcium imaging studies of mouse urothelial cells.22 They were not consistently activated by ␣,␤,MeATP, which selectively activates P2X receptor homomers, or P2X1 or P2X2 heteromers.29,30 Consistent with a recent functional characterization of TRP channels in mouse urothelial cells,22 we detected TRPV4 channels in rat urothelial cells maintained in co-culture but did not detect TRPV1, TRPA1 or TRPM8. FPP did not detect TRPV3, which is detected in mouse urothelium.22 Further experimentation is required to determine the functional expression of TRPV3 in rat urothelium.

CONCLUSIONS We refined and characterized a method of co-culturing lumbosacral sensory neurons and bladder urothelium. The presence of urothelial cells promoted neuron growth. This new approach will be valuable to probe neuro-urothelial communication in health and disease.

ACKNOWLEDGMENTS Dr. Lori Birder assisted with the co-culture concept. Dr. Mitchell Quinlivan assisted with neural growth experiments.

REFERENCES 1. Birder LA: Urothelial signaling. Handb Exp Pharmacol 2011: 202: 207.

cular smooth muscle cocultures. Am J Physiol Heart Circ Physiol 2001; 280: H1722.

ment of sensory neurons in a coculture model. Eur J Neurosci 2007; 26: 113.

2. Khandelwal P, Abraham SN and Apodaca G: Cell biology and physiology of the uroepithelium. Am J Physiol Renal Physiol 2009; 297: F1477.

6. Koizumi S, Fujishita K, Inoue K et al: Ca2⫹ waves in keratinocytes are transmitted to sensory neurons: the involvement of extracellular ATP and P2Y2 receptor activation. Biochem J 2004; 380: 329.

9. Wanigasekara Y and Keast JR: Nerve growth factor, glial cell line-derived neurotrophic factor and neurturin prevent semaphorin 3A-mediated growth cone collapse in adult sensory neurons. Neuroscience 2006; 142: 369.

4. Birder LA and de Groat WC: Mechanisms of disease: involvement of the urothelium in bladder dysfunction. Nat Clin Pract Urol 2007; 4: 46.

7. Liu Z, Liu H and Li Z: Formation of neuromuscular junctions and synthesis of sensory neuropeptides in the co-cultures of dorsal root ganglion and cardiac myocytes. Cell Mol Neurobiol 2008; 28: 939.

10. Xu S, Cheng Y, Keast JR et al: 17beta-estradiol activates estrogen receptor beta-signalling and inhibits transient receptor potential vanilloid receptor 1 activation by capsaicin in adult rat nociceptor neurons. Endocrinology 2008; 149: 5540.

5. Damon DH: NGF-independent survival of postganglionic sympathetic neurons in neuronal-vas-

8. Ulmann L, Rodeau J-L, Danoux L et al: Trophic effects of keratinocytes on the axonal develop-

11. Truschel ST, Ruiz WG, Shulman T et al: Primary uroepithelial cultures. A model system to analyze

3. Arms L and Vizzard MA: Neuropeptides in lower urinary tract function. Handb Exp Pharmacol 2011; 202: 395.

433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494

CO-CULTURES TO PROBE COMMUNICATION BETWEEN SENSORY NEURONS AND UROTHELIUM

495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556

umbrella cell barrier function. J Biol Chem 1999; 274: 15020.

lations of rat dorsal root ganglion neurons. Neuroscience 2008; 153: 279.

12. Yoshimura N, Seki S, Erickson KA et al: Histological and electrical properties of rat dorsal root ganglion neurons innervating the lower urinary tract. J Neurosci 2003; 23: 4355.

18. Steers WD and Tuttle JB: Mechanisms of Disease: the role of nerve growth factor in the pathophysiology of bladder disorders. Nat Clin Pract Urol 2006; 3: 101.

13. de Groat WC and Yoshimura N: Afferent nerve regulation of bladder function in health and disease. Handb Exp Pharmacol 2009; 194: 91.

19. Ernsberger U: Role of neurotrophin signalling in the differentiation of neurons from dorsal root ganglia and sympathetic ganglia. Cell Tissue Res 2009; 336: 349.

14. Kullmann FA, Artim D, Beckel J et al: Heterogeneity of muscarinic receptor-mediated Ca2⫹ responses in cultured urothelial cells from rat. Am J Physiol Renal Physiol 2008; 294: F971. 15. Kalous A, Nangle MR, Anastasia A et al: Neurotrophic actions initiated by proNGF in adult sensory neurons may require peri-somatic glia to drive local cleavage to NGF. J Neurochem 2012; 122: 523. 16. Forrest SL, Forrest SL, Keast JR et al: Expression of receptors for glial cell line-derived neurotrophic factor family ligands in sacral spinal cord reveals separate targets of pelvic afferent fibers. J Comp Neurol 2008; 506: 989. 17. Lu S and Gold MS: Inflammation-induced increase in evoked calcium transients in subpopu-

20. Everaerts W, Gees M, Alpizar YA et al: The capsaicin receptor TRPV1 is a crucial mediator of the noxious effects of mustard oil. Curr Biol 2011; 21: 316. 21. Mochizuki T, Sokabe T, Araki I et al: The TRPV4 cation channel mediates stretch-evoked Ca2⫹ influx and ATP release in primary urothelial cell cultures. J Biol Chem 2009; 284: 21257.

9

24. Ulmann L, Rodeau JL, Danoux L et al: Dehydroepiandrosterone and neurotrophins favor axonal growth in a sensory neuron-keratinocyte coculture model. Neuroscience 2009; 159: 514. 25. Gabella G: Structure of the intramural nerves of the rat bladder. J Neurocytol 1999; 28: 615. 26. Skryma R, Prevarskaya N, Gkika D et al: From urgency to frequency: facts and controversies of TRPs in the lower urinary tract. Nat Rev Urol 2011; 8: 617. 27. Everaerts W, Nilius B and Owsianik G: The vanilloid transient receptor potential channel TRPV4: from structure to disease. Prog Biophys Mol Biol 2010; 103: 2. 28. Akopian AN, Ruparel NB, Jeske NA et al: Role of ionotropic cannabinoid receptors in peripheral antinociception and antihyperalgesia. Trends Pharmacol Sci 2009; 30: 79.

22. Everaerts W, Vriens J, Owsianik G et al: Functional characterization of transient receptor potential channels in mouse urothelial cells. Am J Physiol Renal Physiol 2010; 298: F692.

29. Ford APDW and Cockayne DA: ATP and P2X purinoceptors in urinary tract disorders. Handb Exp Pharmacol 2011; 202: 485.

23. Bang S, Yoo S, Yang TJ et al: Farnesyl pyrophosphate is a novel pain-producing molecule via specific activation of TRPV3. J Biol Chem 2010; 285: 19362.

30. Yu W, Hill WG, Apodaca G et al: Expression and distribution of transient receptor potential (TRP) channels in bladder epithelium. Am J Physiol Renal Physiol 2011; 300: F49.

495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556