SEARCH

SEARCH BY CITATION

Keywords:

  • urinary bladder;
  • myofibroblasts;
  • purinergic receptors;
  • immuohistochemistry

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

OBJECTIVE

To identify particular purinoceptor subtypes by immunohistochemical labelling, as a layer of suburothelial myofibroblasts has been identified in the urinary bladder, and these cells respond to exogenous ATP by generating an intracellular Ca2+ transient, but the particular purinoceptor that responds to ATP is unclear.

MATERIALS AND METHODS

Tissue sections and isolated cells from the urothelial layer of the guinea-pig bladder were used. Preparations were labelled with primary antibodies to the intermediate-filament protein, vimentin, or the purinoceptors P2X3, P2Y1, P2Y2, P2Y4 and P2Y6. For single-labelling we used a secondary antibody tagged with the fluorescent marker Cy3, and for double-labelling also a secondary antibody tagged with fluorescein isothiocyanate or Cy2. Images were examined using a confocal microscope, with an argon (488 nm) or helium-neon (543 nm) laser.

RESULTS

Vimentin-labelling was confined to the suburothelial layer and appeared as discrete signals. Isolated cells labelled with vimentin and strongly for the P2Y6 antibody. There was weaker staining for P2X3, P2Y2 and P2Y4, but none to P2Y1. With frozen sections there was P2Y6 labelling in the urothelial and suburothelial layer.

CONCLUSION

The predominant purinoceptor in suburothelial myofibroblasts, from these labelling studies, is the P2Y6 subtype. However, there was weaker labelling to other subtypes, suggesting multiple receptor subtypes or heterogeneity of receptor subunits. The consequences of there being multiple purinoceptor subtypes in the suburothelial space with respect to sensory signalling are discussed.


Abbreviations
FITC

fluorescein isothiocyanate.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

A layer of suburothelial myofibroblasts has been identified in human and guinea-pig bladder [1,2]. These cells have several interesting characteristics that include: apposition to suburothelial nerves [1]; intense labelling for the intermediate filament protein vimentin and the gap junction protein connexin43 [2]; and a response to exogenous ATP that includes a transient rise of intracellular [Ca2+] and a subsequent depolarization through generation of a Ca2+-dependent Cl inward current [3,4].

ATP is released from urothelium when exposed to stretch by physical or chemical methods [5,6], and it was hypothesized that this is the initial cellular step in the excitation of bladder afferents as the bladder fills with urine [7]. The presence of purinergic (P2X3) receptors on suburothelial afferents and a depression of the micturition reflex in P2X3 knockout mice lend weight to this hypothesis [8,9].

The closeness of the myofibroblast layer to both the urothelium and nerves, their potential ability to act as a functional electrical syncitium, and their response to ATP place them in an ideal setting to act as modulators of this sensory process. However, their response to purinergic receptor agonists is complex, suggesting the presence of multiple purinergic receptors in suburothelial signalling. The myofibroblast response to ATP was not suppressed by immediate pre-exposure to α,β methylene-ATP, which would suggest the absence of functional P2X1/3 signalling [4]. Mediation via a P2Y receptor was suggested because UTP and ADP were equally able to evoke cellular responses, and the rise in intracellular [Ca2+] was not due to changes of membrane potential, but rather caused the subsequent depolarization. Mediation by a P2Y1 receptor was excluded by the inability of the specific antagonist MRS2179 to block the ATP response [4]. However, because there are no other specific P2Y receptor agonists or antagonists it was not possible to identify further the P2Y receptor subtype. The objective of the present report was to identify P2Y receptor subtypes by immunohistochemical labelling of isolated guinea-pig myofibroblasts.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Guinea-pigs were killed according to guidelines prescribed by the Animals (Scientific Procedures) Act, 1986, the bladder removed immediately, and placed in a Ca2+-free medium (mM): NaCl, 105.4; NaHCO3, 22.3; KCl, 3.6; MgCl2, 0.9; NaH2PO4, 0.4; HEPES, 19.5; glucose 5.4; sodium pyruvate, 4.5, pH 7.4. Full-thickness, longitudinal strips were dissected from one half of the bladder, placed in embedding medium (Cryo-m-Bed, Bright, UK) and stored in liquid N2. For tissue immunohistochemistry, 15 µm frozen sections were cut, collected on poly-l-lysine-coated slides and fixed with 4% (w/v) paraformaldehyde in PBS for 30 min at room temperature.

To prepare isolated cells, the methods were identical to those used previously to generate myofibroblast preparations [3,4]. Briefly, the urothelial sheet was peeled away from the other half of the bladder by blunt dissection. The sheet was soaked in Ca2+-free medium at room temperature for 5 min before transferring it to a disrupting Ca2+-free solution, and stirred at 37 °C. The disrupting solution had the same ionic composition as above, with the following additions: collagenase type-I (1.0 mg/mL, Worthington Biochemical Corp, Lakewood, NJ, USA); hyaluronidase type I-S (0.25 mg/mL) and type-III (0.25 mg/mL), trypsin inhibitor type-IIS (0.45 mg/mL) and BSA (2.5 mg/mL). After 10 min treatment the solution was discarded by gentle aspiration. The residual was transferred to an uncoated culture dish and resuspended in RPMI-1640 culture medium (Invitrogen, UK) with 4% (w/v) fetal bovine serum, and allowed to settle for ≈ 2 h at 37 °C in a 5% CO2 atmosphere. This final step was important to allow the cells to adhere to the culture plate so that the cells could be washed and exposed to antibody solutions without them washing away. For immunohistochemistry the culture medium was discarded after the 2-h period, and cells fixed in paraformaldehyde as above.

For immunocytochemistry, tissue slices and cells were washed in PBS, and nonspecific sites blocked by adding PBS with 1% BSA at room temperature for 1 h. Single-labelling was done by applying primary rabbit polyclonal antibodies for P2X3 (1 : 1000 dilution), P2Y1, P2Y2, P2Y4, or P2Y6 (all P2Y at 1 : 200 dilution) in PBS with 1% BSA for 48 h at 4 °C. After a thorough wash in PBS secondary labelling was carried out with goat-anti-rabbit IgG conjugated to Cy3 (1 : 200 dilution), applied at room temperature for 2 h. Double-labelling for purinoceptors and vimentin was also carried out in representative samples. The procedure was similar to that above except that the purinoceptor primary antibody and a monoclonal antibody to vimentin (1 : 100 dilution) were added together. The secondary antibodies were again goat-anti-rabbit IgG-Cy3 conjugate for purinoceptors, and for vimentin either goat-anti-mouse IgG Cy2 (1 : 200 dilution) or goat-anti-mouse IgG fluorescein isothiocyanate (FITC, 1 : 50). Negative controls were carried out by omitting the primary antibody from the first incubation stage. All chemicals were from Sigma Co (Poole, UK), unless otherwise stated.

Cells were visualized under a laser-controlled fluorescence microscope (Radiance 2100, Bio-Rad, Hercules, CA, USA) using a × 20 objective (NA 0.5). Excitation of FITC or Cy2 was from an argon laser (488 nm) and emitted light was collected through a band pass filter (515 ± 15 nm). Cy3 was excited from a helium-neon laser (excitation 543 nm) and light collected via a band-pass filter (>570 nm). Transmitted light images were collected via a visible-light band-pass filter. Autofluorescence from the tissue sections was recorded at 543–570 nm, and resulted predominantly from excitation with the helium-neon laser.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Figure 1 shows a section of the urothelial/suburothelial layer labelled with the antibody to vimentin, and a secondary fluorescent probe Cy3 (showing red). Discrete vimentin-labelled elements were distributed throughout the suburothelial layer, but staining was absent in the urothelium (Fig. 1A). The urothelium is visible as a faint background. In addition, the vimentin-labelled cells were somewhat more concentrated immediately below the urothelium itself, compared to deeper towards the detrusor layer. Fig. 1B shows green autofluorescence coloration, due mainly to collagen, and this was also abundant in the suburothelium, but absent from the urothelium itself. The close association of these two elements is useful to attribute a particular marker to the suburothelium. Fig. 1C shows an overlay of the images in A and B, and shows that the vimentin labelling is in the same region as the autofluorescence.

image

Figure 1. Vimentin labelling of guinea-pig bladder wall. A: vimentin labelling using Cy3 as a secondary antibody (Ab), in red: the label shows as discrete elements. uro; urothelium; suburo, suburothelium. The arrow marks the boundary between the two layers and runs approximately vertically. B: autofluorescence (green) from the same section. C: superposition of images A and B.

Download figure to PowerPoint

Samples of cells used for subsequent purinoceptor immunolabelling showed vimentin labelling, indicating that they were cells derived from the suburothelial layer. There were no images from any of the negative controls. Figure 2A shows paired transmission light and fluorescence (>570 nm) images of a cell labelled with the antibody to P2Y6 receptors. The image shows intense staining in one cell, a second cell observed in the transmitted light image showing no staining. However, lack of P2Y6 staining was unusual; of 54 cells visualized there was intense staining in 52 (96%). Figure 2B shows another example of cells, this time double-labelled for P2Y6 and vimentin, using Cy3 and FITC, respectively, as fluorescent probes. The overlay shows the two markers were not exactly co-located, as they occur at different locations in the cell, but were present throughout the cell images.

image

Figure 2. A: Two isolated myofibroblasts indicated by arrows in the transmission image (i). P2Y6 labelling (ii) of the same cells with Cy3 secondary antibody (Ab). In this and subsequent images, yellow arrows indicate cells showing Ab labelling, and white arrows no labelling. B: a single isolated myofibroblast: transmission image (i); labelled with vimentin Ab (ii, FITC secondary Ab); labelled with P2Y6 antibody (iii, Cy3 secondary Ab); and superposition of the three images (iv).

Download figure to PowerPoint

By contrast, Fig. 3A shows no P2Y1 labelling, and this was never seen in any cell (Fig. 3A). The remaining panels of Fig. 3 show that there was weak labelling of P2X3 (Fig. 3B), P2Y2 (C) and P2Y4 (D) antibodies, although it was present in similar numbers of cells as for P2Y6 labelling.

image

Figure 3. Purinoceptor labelling of isolated myofibroblasts; in each case transmitted light and fluorescence images (Cy3 secondary antibody, Ab) are shown side-by-side. For colour of arrows see Fig. 2. A: P2Y1 Ab. B: P2X3 Ab. C: P2Y2 Ab. D: P2Y4 Ab.

Download figure to PowerPoint

The co-location of vimentin and P2Y6 antibodies on isolated cells suggests that P2Y6 staining should be present in the suburothelial layer. Figure 4 shows that P2Y6 labelling was present as discrete locations in this layer. Fig. 4A shows a transmitted light image where the urothelium and suburothelium are marked. The white line delineates the two layers, columnar cells mark the urothelium, and the suburothelial layer contains a more random array of cells. Fig. 4B shows that P2Y6 staining is present throughout the section, with particular intense staining on the apical surface of the urothelium. However, there is also staining in the suburothelial layer, most dense in the area immediately below the urothelium, and becoming more punctate in deeper layers. Fig. 4C shows autofluorescence of the section, confined to the suburothelial region. Fig. 4D is an overlay of the three images and shows the association of the less dense and punctate P2Y6 staining with the autofluorescence. The intense spot in Fig. 4C and D is probably a lipofuscin droplet.

image

Figure 4. P2Y6 labelling of guinea-pig bladder wall. A: transmitted light section. B: P2Y6 labelling of the same section using Cy3 as a secondary antibody, C: autofluorescence (green) from the same section. D: superposition of the three images. The white lines show the approximate course of the urothelial/suburothelial division.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The study identified P2Y receptors on vimentin-staining cells from the urothelial/suburothelial layer of the bladder wall, with the most intense staining for the P2Y6 subtype. Our previous work showed that vimentin labelling is confined to the suburothelium and closely localized with connexin43 labels, and indicative of myofibroblasts [2], suggesting these cells are a separate population from the epithelial cells of the urothelium. Furthermore, we think that contamination from associated urothelial cells is small because the latter show intense staining for P2Y1, P2Y2 and P2Y4 receptors [10], a feature that was not obvious in our observations. In particular, P2Y1 antibodies showed no staining in our preparations, and this agrees with the inability of a P2Y1 receptor antagonist to block ATP-induced responses in isolated cells [4]. The small amount of P2X3, P2Y2 and P2Y4 receptor labelling may be due a minor presence of these subtypes, or to some heterogeneity in individual receptor units. The same cells used in these experiments respond to exogenous ATP by generating transient intracellular Ca2+ changes and depolarizations, and have a spindle-shape immediately after isolation [3,4]. Their rounded appearance in Figs 2 and 3 was due to the procedures required to generate single cells suitable for plating on the culture dish, while awaiting antibody labelling. This final step was necessary to allow cells to adhere to the culture dish so that they could withstand the staining procedures without them washing away. In all other respects they retained the physiological phenotype of the suburothelial myofibroblast. We also show intense P2Y6 staining in the urothelial layer itself (Fig. 4). Previous studies [10] did not report results using antibodies to this receptor subtype and the significance of this finding is unknown at present.

In our previous studies we measured intracellular Ca2+ transients in response to exogenous application of adenine and uridine nucleotides [3,4]. This showed that ATP, UTP and ADP at 10–100 µm are effective in generating transients of equivalent magnitude, and that ATP was effective down to ≈ 0.1 µm. Moreover ATP-dependent transients could still be generated in the presence of α,β methylene-ATP, excluding functional P2X3 receptors due to their rapid desensitization by this analogue [11]. P2Y6 receptors show a potency in the order UTP > ADP > ATP [12,13] which will not entirely explain our data, as ATP in the lower concentration range should not elicit a response. It is possible that the nucleotide binding profile for P2Y6 receptors is different in these suburothelial myofibroblasts, or that the similar potency of the above nucleotides may reflect activation of the different P2Y receptor subtypes in these cells. The lack of selective subtype antagonists does not resolve the question of the functional P2Y receptor in intact cells, and may require the use of short interfering RNA techniques to specifically ‘knock-down’ specific P2Y receptor genes.

The observation that multiple types of purinoceptors are located in the suburothelial space increases the complexity of the putative signalling pathway of bladder fullness. The simplest scheme was that ATP released from the urothelium by distension, or possibly other noxious stimuli, excited bladder afferents, which conveyed the signal to the CNS [9]. The detection of a myofibroblast layer closely apposed to unmyelinated nerves in the suburothelium immediately next to the urothelium [1] also raised the possibility of interaction between these two cell types. The additional observations that myofibroblasts also responded to ATP by generating intracellular Ca2+ and membrane potential transients [3], and that they are coupled by low-conductance gap junctions [2], suggested that they might also be involved in the sensory pathways. In principle, myofibroblasts could alter the gain of the sensory pathway by modulating afferent activity in one of two ways: by physically contracting and thus mechanically stimulating afferents, or by releasing agents that act in a paracrine stimulatory fashion. Evidence for the latter process was reported for cell types such as endothelial, HeLa and glioma cells, where exogenous P2Y receptor agonists evoke ATP release from the same cells, suggesting an autocatalytic excitatory process [14,15].

The detection of multiple purinoceptors in the suburothelial space shows several targets that could modulate the sensory pathway, either from pathological processes or therapeutic manipulation. Alteration of the variable element of a sensory pathway, i.e. the myofibroblast layer in the sensation of bladder fullness, permits an approach to achieve high sensitivity in its regulation.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

We are grateful to St Peter’s Trust and Detrol research grants for financial assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES