Expression of Vesicular Glutamate Transporters in Sensory and Autonomic Neurons Innervating the Mouse Bladder

Abstract

Purpose

VGLUTs, which are essential for loading glutamate into synaptic vesicles, are present in various neuronal systems. However, to our knowledge the expression of VGLUTs in neurons innervating the bladder has not yet been analyzed. We studied VGLUT₁, VGLUT₂ and VGLUT₃ in mouse bladder neurons.

Materials and Methods

We analyzed the expression of VGLUT₁, VGLUT₂ and calcitonin gene-related peptide by immunohistochemistry in the retrograde labeled primary afferent and autonomic neurons of BALB/c mice after injecting fast blue in the bladder wall. To study VGLUT₃ we traced the bladder of transgenic mice, in which VGLUT₃ is identified by enhanced green fluorescent protein detection.

Results

Most bladder dorsal root ganglion neurons expressed VGLUT₂. A smaller percentage of neurons also expressed VGLUT₁ or VGLUT₃. Co-expression with calcitonin gene-related peptide was only observed for VGLUT₂. Occasional VGLUT₂ immunoreactive neurons were seen in the major pelvic ganglia. Abundant VGLUT₂ immunoreactive nerves were detected in the bladder dome and trigone, and the urethra. VGLUT₁ immunoreactive nerves were discretely present.

Conclusions

We present what are to our knowledge novel data on VGLUT expression in sensory and autonomic neurons innervating the mouse bladder. The frequent association of VGLUT₂ and calcitonin gene-related peptide in sensory neurons suggests interactions between glutamatergic and peptidergic neurotransmissions, potentially influencing commonly perceived sensations in the bladder, such as discomfort and pain.

The bladder is profusely innervated by sympathetic, parasympathetic and primary afferent neurons.^1,2 The main neurotransmitters involved in autonomic functions are noradrenaline and acetylcholine, which are produced by sympathetic and parasympathetic neurons, respectively. The presence of these neurotransmitters in neurons and nerve fibers is usually indicated by the identification of associated molecules, such as TH, the norepinephrine transporter-1 (noradrenergic) and choline acetyltransferase or the vesicular acetylcholine transporter (acetylcholinergic). Like nonvisceral primary afferent neurons,³ visceral primary afferent neurons are classically described as glutamatergic and often capable of co-expressing various molecules, such as neuropeptides, various types of receptors⁴ and even autonomic neuron markers (TH).⁵

The earliest morphological study of the glutamatergic nature of bladder DRG neurons was reported more than 12 years ago, when Keast and Stephensen observed immunohistochemically detectable glutamate in approximately 50% of rat bladder DRG neurons.⁶ For use in neurotransmission glutamate is taken up into synaptic vesicles by VGLUTs. Different types of VGLUTs were recently identified, resulting in the reliable immunohistochemical characterization of various subpopulations of glutamatergic neurons in the central and peripheral nervous systems.⁷ In the latter system VGLUTs were characterized in rodent nonvisceral⁸ and colorectal primary afferent neurons,⁹ and in some LSC neurons after peripheral nerve injury.¹⁰

To our knowledge no analysis of the presence of VGLUTs in DRG and autonomic neurons projecting to the bladder has been available to date. We present a comprehensive immunohistochemical study of the expression of the 3 known VGLUTs in DRG, LSC and MPG neurons innervating the mouse bladder.

MATERIALS AND METHODS

Male BALB/c mice (Taconic, Germantown, New York) at age 7 to 8 weeks and 129S6/SvEvTac-C57Bl/6 mice¹¹ at age 6 weeks were used in all experiments. All research protocols adhered to the Guide for the Care and Use of Laboratory Animals and were approved by the University of Pittsburgh and University of California-San Francisco institutional animal care use committees.

None of the currently available commercial antibodies against VGLUT₃ efficiently label DRG neurons. Therefore, we also used a transgenic mouse expressing eGFP under the control of VGLUT₃ regulatory sequences that reliably identify VGLUT₃ expressing neurons in these mice.¹¹

The bladders of 5 BALB/c male mice and 3 VGLUT₃-eGFP male mice at age 6 weeks were injected with the fluorescent retrograde neuronal tracer fast blue (2% in saline) (EMS-Chemie, Gross-Umstadt, Germany) according to previously reported surgical procedures.⁹

At 12 days after fast blue injection the mice were deeply anesthetized using sodium pentobarbital (60 mg/kg intraperitoneally) (Ovation Pharmaceuticals, Deerfield, Illinois) and perfused via the ascending aorta with previously described fixative mixtures.⁹ Three naïve BALB/c mice were similarly anesthetized and perfused. Thoracolumbar (T8-L1) and lumbosacral (L6-S2) DRGs, and the MPG, LSC and bladder were dissected out and processed according to previously described immunohistochemistry protocols.^8,9 The table lists the antibodies used. Further details on these antibodies were published previously.⁹

We performed single or double staining experiments using TSA™ Plus and/or indirect immunofluorescence techniques.⁹ Briefly, sections were incubated with 1 and 2 antibodies for single and double staining, respectively, and processed first using the tyramide signal amplification technique. Additional tyramide signal amplification or indirect immunofluorescence protocols were used in double staining experiments. Nonspecific staining by secondary antibody was tested in a few sections by omitting the primary antibody. CGRP, VGLUTs and eGFP antiserum were previously described in detail.^8,9

Immunofluorescence sections were examined using an Eclipse E600 microscope (Nikon, Tokyo, Japan) provided with a Retiga™ 2000 R Fast CCD camera using IPLab (Scanalytics, Vancouver, British Columbia, Canada). Colocalization was analyzed with a FluoView FV 1000 confocal microscope (Olympus®). Image resolution, brightness and contrast were optimized using Adobe® Photoshop® CS3. Because confocal imaging of fast blue was not possible due to the lack of appropriate filters, some MPG images were composed by merging separate optical (fast blue) and confocal (other markers of interest) photomicrographs.

Retrograde traced neuron profiles were quantified in every fifth section of thoracolumbar and lumbosacral DRGs (5 to 8 sections per ganglion). The percent of VGLUT expressing bladder neuron profiles was determined by comparing the total number of fast blue positive neurons to those expressing each marker. Cell body diameters of a representative sample of each type of quantified neuron profile (more than 7 and up to 48 neurons per type) were measured using ImageJ (http://rsbweb.nih.gov/ij/). These data were used to correct raw counts based on the Abercrombie correction factor.¹² Only neuron profiles showing immunostaining intensity higher than 2 orders of magnitude were counted.

Data are presented as the mean ± SEM. They were statistically analyzed using the Student t test for independent samples with p <0.05 considered significant.

RESULTS

Retrograde tracing from the bladder revealed a discrete number of fast blue positive neuron profiles of different sizes per section in thoracolumbar and lumbosacral DRGs, which were easily differentiated from fast blue negative neuron profiles (figs. 1 and 2).

Figure 1.

VGLUT expression in DRGs innervating bladder. Optical immunofluorescence photomicrographs show sections of S1 (A to C), L6 (D to F), T10 (G to I) and T11 (J to L) DRGs incubated with VGLUT₁ (B), VGLUT₂ (E and H) and eGFP-VGLUT₃ (K) antiserum. Merged images (C, F, I and L) show retrograde labeled bladder DRG neurons containing fast blue (A, D, G and J) (purple areas). Green areas represent VGLUT₁, VGLUT₂ or eGFP-VGLUT₃-IR neurons. White areas represent VGLUT and fast blue colocalization. Double arrowheads indicate VGLUT₁ (A to C), VGLUT₂ (D to I) and eGFP-VGLUT₃ (J to L) IR bladder neuron profiles. Arrows indicate bladder neuron profiles lacking VGLUT₁, (A), VGLUT₂ (G) and eGFP-VGLUT₃ (J)-like immunoreactivity. Arrowheads indicate nontraced neuron profiles expressing VGLUT₁ (B), VGLUT₂ (E) and eGFP-VGLUT₃ (K). Scale bars represent 50 μm (G to I, and A to F and J to L).

Figure 2.

VGLUT-CGRP co-expression patterns in DRGs innervating bladder. Optical immunofluorescence photomicrographs show sections of T10 (A to D), L6 (E to H) and S2 (I to L) DRGs incubated with VGLUT₁ (B), VGLUT₂ (F) and eGFP-VGLUT₃ (J) antiserum. In merged images (D, H and L) red areas represent retrograde labeled bladder DRG neurons containing fast blue (A, E and I). Green areas represent VGLUT₁, VGLUT₂ or eGFP-VGLUT₃-IR neurons. Blue areas represent CGRP-IR neurons. White areas represent VGLUT plus fast blue plus CGRP colocalization. Orange areas indicate VGLUT plus fast blue. Purple areas indicate CGRP plus fast blue. Double arrowheads indicate VGLUT₁ (A to D), VGLUT₂ (E to H) and eGFP-VGLUT₃ (I to L) IR bladder neuron profiles lacking CGRP-like immunoreactivity. Black double arrowheads indicate bladder neuron profiles expressing CGRP (A to D) and lacking VGLUT-like immunoreactivity (I to L). Double arrows indicate VGLUT₂-IR bladder neuron profiles co-expressing CGRP (E to H). White arrowheads indicate nontraced VGLUT₁ (B), VGLUT₂ (F) and eGFP-VGLUT₃ (J) IR DRG neuron profiles lacking CGRP-like immunoreactivity. Black arrowheads indicate nontraced VGLUT₂-IR neuron profiles co-expressing CGRP (E to H). Scale bar represents 50 μm (L).

All VGLUTs were expressed in mouse DRG neurons projecting to the bladder, although in different proportions (fig. 1). An average of 32.4% ± 3.6%, 94.1% ± 3.0% and 18.1% ± 1.3% of bladder DRG neuron profiles expressed VGLUT₁, VGLUT₂ and eGFP-VGLUT₃, respectively. Further analysis revealed that VGLUT expression differed between thoracolumbar and lumbosacral DRGs. VGLUT₁ was present in a greater proportion of thoracolumbar than lumbosacral bladder DRG neuron profiles (mean 38.9% ± 4.2% vs 25.9% ± 3.6%, p <0.05, fig. 1, A to C). Most VGLUT₁-IR neuron profiles appeared to be medium to large (fig. 1, A to C). VGLUT₂ was detected in similar high proportions in thoracolumbar and lumbosacral bladder DRG neuron profiles of all sizes (mean 93.8% ± 2.0% and 94.3% ± 4.1%, respectively, fig. 1, D to I). Occasional bladder DRG neurons lacking VGLUT₂ were also found (fig. 1, G to I). eGFP-VGLUT₃ was expressed in a greater proportion of thoracolumbar than lumbosacral small bladder DRG neuron profiles (mean 27.9% ± 2.0% vs 8.3% ± 0.6%, p <0.001, fig. 1, J and L). Finally, in addition to bladder neurons, abundant VGLUT₂ or eGFP-VGLUT₃-IR and several VGLUT₁-IR fast blue negative neurons were detected in all DRGs (figs. 1 and 2).

Colocalization analysis showed that only VGLUT₂ was co-expressed with CGRP in bladder DRG neuron profiles (fig. 2). Thus, while no VGLUT₁ bladder DRG neuron profiles co-expressed CGRP (fig. 2, A to D), a mean of 54.7% ± 3.4% of thoracolumbar and 52.2% ± 9.9% of lumbosacral bladder DRG neuron profiles co-expressed VGLUT₂ and CGRP (fig. 2, E to H). We also detected fast blue negative VGLUT₂-IR neuron profiles that co-expressed CGRP as well as CGRP only bladder DRG neurons (fig. 2, A to D, F and I to L). As observed for VGLUT₁, virtually no eGFP-VGLUT₃ bladder DRG neuron profiles co-expressed CGRP (fig. 2, I to L). This lack of co-expression with CGRP was also noted for fast blue negative VGLUT₁ and eGFP VGLUT₃-IR neuron profiles (fig. 2, A to D and I to L).

We also observed retrograde labeling of LSC and MPG bladder neurons (fig. 3, A, D, G, J, M and P). In the LSC no fast blue positive neurons expressed VGLUT₁ or VGLUT₂ (fig. 3, A to F). Nevertheless, some VGLUT₂-IR fibers and what appeared to be VGLUT₂-IR synaptic varicosities were detected in the LSC (fig. 3, D to F). Similarly, fast blue positive neurons in the MPG lacked VGLUT₁-like immunoreactivity (fig. 3, G to I), which was only associated with a few varicose profiles (fig. 3, H and I). In contrast, occasional VGLUT₂-IR bladder MPG neurons were found (fig. 3, J and L), along with many VGLUT₂-IR fibers and varicosities (fig. 3, K, L, N, O, Q and R). Some varicosities with a basket-like appearance surrounded nonbladder MPG neurons (fig. 3, P and R).

Figure 3.

VGLUT expression in LSCs and MPGs innervating bladder. Optical (A, D, G, J, M and P) and confocal (B, C, E, F, H, I, K, L, N, O, Q and R) immunofluorescence photomicrographs show LSC (A to F) and MPG (G to R) sections after incubation with VGLUT₁ (B and H) and VGLUT₂ (E, K, N and Q) antiserum. Merged images (C, F, I, L, O and R) show retrograde labeled bladder DRG neurons containing fast blue (purple areas) (A, D, G, J, M and P). Green areas represent VGLUT₁ and VGLUT₂-IR neurons. White areas represent VGLUT plus fast blue colocalization. Arrows indicate bladder neuron profiles in LSC (A and D) and MPG (G, J and P) lacking VGLUT₁ and VGLUT₂-like immunoreactivity, respectively. White arrowheads indicate VGLUT₁ or VGLUT₂-IR varicosities in LSC (E) and MPG (H, K and Q). Note higher magnification (I) of arrowhead (H). Black arrowheads indicate VGLUT₂-IR fibers in LSC (E) and MPG (K and Q). VGLUT₂-IR varicosities (arrowhead and arrow) surrounding bladder neuron profiles lacking transporter were detected (F and inset). Small double arrows indicate VGLUT₂-IR bladder neuron profiles (J and L). Note higher magnification (M to O). White double arrowhead in panel shows VGLUT₂-IR basket surrounding nontraced neuron profiles lacking transporter (Q and R, inset). Scale bars represent 50 (A to Q) and 10 (insets) μm.

In the bladder VGLUT₁-like immunoreactivity was commonly observed in the muscular and to a lesser extent in the submucosal layers of the bladder (fig. 4, A and B). In contrast, VGLUT₂-IR nerves formed a profuse fiber network spanning all layers of the bladder wall throughout the dome and trigone, and the urethra (fig. 4, C to G). Colocalization analysis demonstrated that VGLUT₁ and CGRP co-expression in bladder nerve fibers was virtually non-existent (fig. 5, A to C). In contrast, VGLUT₂ and CGRP were often colocalized in nerve fibers in the bladder (fig. 5, D to L).

Figure 4.

Confocal immunofluorescence photomicrographs show transverse (A to D) and sagittal (E to G) sections of bladder dome (A to D), trigone (E) and urethra (F and G) after incubation with VGLUT₁ (A and B) and VGLUT₂ (C to G) antiserum. White double arrowheads indicate VGLUT₁ (A and B) and VGLUT₂ (C and D) IR nerve fibers in bladder muscular layer. White arrowheads indicate VGLUT₁ (B) and VGLUT₂ (C and D) IR nerve fibers in bladder submucosal layer. Arrows indicate VGLUT₂-IR nerve fibers in bladder trigone (E). Black arrowheads (F and G) indicate VGLUT₂-IR fibers penetrating urethra. White arrowheads indicate VGLUT₂-IR nerve fibers innervating urethra (F and G). Scale bars represent 100 μm (A to D and E to G).

Figure 5.

Confocal immunofluorescence photomicrographs show transverse sections of bladder after co-incubation with VGLUT₁ (A), VGLUT₂ (D, G and J) and CGRP (B, E, H and K) antiserum. In merged images green areas represent VGLUT₁ or VGLUT₂-IR fibers (C, F, I and L). Purple areas represent CGRP-IR fibers. White areas represent VGLUT plus CGRP colocalization. Arrowheads indicate VGLUT₁ (A) and VGLUT₂ (D) IR nerve fibers lacking CGRP-like immunoreactivity in bladder dome. Arrows indicate CGRP-IR nerve fibers lacking VGLUT-like immunoreactivity in bladder dome muscular layer (B and H). Double arrows indicate VGLUT₂ (G to I) and CGRP (J to L) IR nerve fibers in bladder mucosal (inset 1) and muscular (inset 2) layers (F) at higher magnification. Scale bars represent 50 μm (D to F) and 25 (A to C and G to K) μm.

DISCUSSION

Although sensory neurons are often glutamatergic,⁶ information on the types of VGLUTs involved in the synaptic vesicle uptake of glutamate in such neurons has only recently received attention.⁸ Evidence of VGLUT expression in visceral sensory neurons was first reported by Tong et al in ganglia innervating the rat stomach.¹³ Subsequently, VGLUT expression was confirmed in the peripheral projections of vagal and DRG neurons innervating the gut of the guinea pig, rat and mouse.⁹ More recently, we reported that more than 95% of mouse colorectal DRG neurons express VGLUT₂ as well as VGLUT₁ but the latter in a much lower proportion.⁹ Similarly, the current study shows that mouse bladder DRG neurons also predominantly express VGLUT₂ and to a lesser extent VGLUT₁ or VGLUT₃.

Like the colorectum, the bladder is innervated by lumbar splanchnic and pelvic nerves with cell bodies in thoracolumbar and lumbosacral DRGs, respectively. We found that bladder sensory neurons expressing VGLUT₁ or VGLUT₃, unlike those synthesizing VGLUT₂, are more abundant in thoracolumbar than in lumbosacral DRGs. An increasing number of reports have revealed significant differences in the function and neurochemical expression of various molecules in afferent neurons comprising the 2 innervation pathways.⁵ For example, 66% of mouse thoracolumbar bladder neurons express mRNA of TRP, subfamily A, member 1, while expression in lumbosacral bladder neurons is sparse at about 2.5%.¹⁴ In contrast, TH is expressed in threefold (colorectum) and fivefold (bladder) greater proportions in lumbosacral DRGs.⁵ The differences in VGLUT₁ and VGLUT₃ expression between thoracolumbar and lumbosacral bladder neurons that we report could imply differences in neuronal physiology. This was already reported in the mouse colorectum, in which higher TRPV₁ expression in thoracolumbar than in lumbosacral DRGs corresponds to a stronger response to applied capsaicin at colorectal thoracolumbar nerve terminals.¹⁵

We found that about half of the 94% of retrograde traced bladder DRG neurons expressing VGLUT₂ co-expressed the peptidergic marker CGRP. These values correlate with previous studies in rats showing that up to 69% of DRG neurons innervating the bladder express CGRP.⁴ In the mouse colorectum more than 80% of all retrograde traced DRG neurons co-expressed VGLUT₂ and CGRP,⁹ in accordance with CGRP expression in most DRG neurons innervating the rodent colorectum.⁴ Accordingly, a large proportion of visceral glutamatergic sensory neurons may also synthesize and potentially co-release CGRP. In contrast, in DRGs innervating nonvisceral tissues only about 31% of the approximately 65% VGLUT₂-IR neurons colocalize CGRP,⁸ suggesting potentially variable levels of co-release of glutamate and CGRP by DRG neurons that inner-vate different tissues.

VGLUT₁ and VGLUT₃-IR bladder neurons did not co-express CGRP. However, since virtually all neurons innervating this organ synthesized VGLUT₂, it is highly likely that VGLUT₁ and/or VGLUT₃ is co-expressed with VGLUT₂ in its non-peptidergic subpopulation, as described. Co-expression of more than 1 type of VGLUT was first reported for some neurons in the central nervous system¹⁶ and later confirmed for VGLUT₁ and VGLUT₂ in nonvisceral primary afferent neurons and their projections.⁸ Our study in the mouse colorectum⁹ and our current bladder series further support the supplementary contribution of different VGLUTs to glutamatergic physiology in sensory neurons.

TRPV₁ is a nonselective cation channel activated by pH, heat and capsaicin that is strongly implicated in bladder nociception.¹⁷ TRPV₁ is expressed in many peptidergic bladder DRG neurons in the rat¹⁸ and mouse.^14,15,18 Considering the prominent VGLUT₂ expression in peptidergic bladder DRG neurons, VGLUT₂ undoubtedly coexists with TRPV₁. In support of this notion, VGLUT₂ co-expression with TRPV₁ was previously noted in nerve fibers terminating in the mouse rectum.¹⁹ Interestingly, TRPV₁ and glutamate receptors also colocalize in nonvisceral small diameter primary afferent neurons and the induction of c-Fos expression in the spinal dorsal horn induced by subcutaneous capsaicin injection has been prevented by concomitant subcutaneous administration of ionotropic or metabotropic glutamatergic receptor antagonists.²⁰ Peripheral release of glutamate, likely from primary afferent nerve terminals, was noted in the hind paw skin of rats exposed to intraplantar injection of the irritant formalin.²¹ Moreover, local injection of glutamate induces nociceptive behavior in rats²² and humans.²³ Thus, we speculate that TRPV₁-glutamate receptor interaction at central and peripheral nerve terminals, potentially driven by glutamate released from primary afferent neurons, could have a role in the physiopathology of bladder sensation and pain.

As suggested, bladder nociception could be influenced by the peripheral release and action of glutamate. In recent years botulinum toxin A has been used to treat bladder pain associated with interstitial cystitis. Although botulinum toxin A was originally described as a potent inhibitor of acetylcholine release in the neuromuscular junction,²⁴ it is increasingly accepted that botulinum toxin A also blocks transmitter release at nonacetylcholinergic synapses. Thus, it was hypothesized that botulinum toxin A desensitizes peripheral afferent nerves by inhibiting vesicular release of adenosine triphosphate and the release of peptides such as CGRP and substance P. Reduction of the axonal expression of TRPV₁ from urothelial and suburothelial nerve endings was also proposed.²⁵

In humans synaptic vesicle protein 2, a high affinity receptor for botulinum toxin A, and synaptosomal-associated protein 25, one of the soluble NSF-attachment protein receptors involved in vesicle fusion before neurotransmitter release that is essential for the blocking action of botulinum toxin A, was found in at least half of bladder peptidergic nerve fibers.²⁶ In rodents most nonvisceral,⁸ colorectal⁹ and bladder (current study) DRG neurons/nerves that express VGLUT₂ and often synthesize peptides most likely have the soluble NSF-attachment protein receptor machinery for synaptic neurotransmitter release. Therefore, we speculate that in these neurons/nerves botulinum toxin A could also affect glutamatergic synaptic vesicle docking, limiting glutamate release and, thus, contributing to the analgesic effect of the toxin. In fact, the benefits of botulinum toxin were successfully explored in animals with nonvisceral neuropathy.²⁴

Finally, we detected a few MPG neurons projecting to the bladder that also expressed VGLUT₂, while no retrograde traced neurons in the LSC showed positive immunostaining for any VGLUT. Thus, in normal conditions DRGs as well as the MPG may be a source of VGLUT₂-IR fibers in the bladder. Interestingly, we recently reported that a subpopulation of LSC neurons up-regulates VGLUT₂ upon injury to the pelvic nerve.¹⁰ In that study we did not trace the neurons. Thus, to our knowledge it remains to be established whether some LSC neurons that up-regulate VGLUT₂ project to the bladder or colorectum. Likewise, we speculate that MPG neurons that do or do not project to visceral organs may up-regulate VGLUT₂ after injury to their postganglionic axons.

CONCLUSIONS

We present an immunohistochemical account of VGLUT expression in sensory and autonomic neurons innervating the mouse bladder. The finding of a preponderant expression of VGLUT₂, as recently also noted in the mouse colorectum,⁹ suggests a relevant role for this transporter in glutamatergic neurotransmission in neurons innervating pelvic visceral organs, from which the principal conscious sensations that arise are discomfort and pain. Glutamate, which is contained in the central terminals of primary afferents, spinal interneurons and terminals of fibers that descend from the medulla oblongata,²⁷ is well established as important for nociceptive transmission from pelvic organs.^1,28 Thus, complex interaction was noted between spinal ionotropic (NMDA and AMPA) receptors in the spinal processing of nociceptive input from the irritated lower urinary tract.²⁹ Likewise, a facilitatory role in bladder primary afferent processing was attributable to metabotropic glutamate receptor 5 in normal conditions and during lower urinary tract inflammation.²⁸ However, the role of peripheral glutamate release in the bladder is unclear. Thus, the intravesical application of metabotropic glutamate receptor antagonists has no effect on bladder contractility and pelvic nerve afferent firing in normal rats.³⁰ Studies in animal disease models are needed to es tablish the potential involvement of peripheral glutamate/glutamatergic receptors in the bladder.

Technical specifications of primary and secondary antibodies, and fluorophores

	Dilution	Reference/Source
Host species + target antigen:
Rabbit anti-VGLUT1	1:4,000 (tyramide signal amplification)	Kawamura Y et al: J Neurosci 2006; 26: 2991
Guinea pig anti-VGLUT2	1:8,000 (tyramide signal amplification)	Miyazaki T et al: Eur J Neurosci 2003; 17: 2563 + Brumovsky et al8–10
Rabbit anti-EGFP	1:4,000 (tyramide signal amplification)	Molecular Probes® (catalogue No. A-11122)
Rabbit anti-CGRP	1:8,000 (indirect fluorescence)/1:40,000 (tyramide signal amplification)	Sigma® (product A8198)
Secondary antibodies/fluorophores:
Horseradish peroxidase-donkey anti-guinea pig	1:200	Jackson ImmunoResearch®, West Grove, PA (code No. 713-035-003)
Horseradish peroxidase-donkey anti-rabbit	1:200	Jackson ImmunoResearch (code No. 711-035-152)
Tetramethylrhodamine isothiocyanate-donkey anti-rabbit	1:400	Jackson ImmunoResearch (code No. 711-025-152)
Fluorescein tyramide	1:700	PerkinElmer® (product No. NEL741)

Abbreviations and Acronyms

CGRP

calcitonin gene-related peptide

DRG

dorsal root ganglion

eGFP

enhanced green fluorescent protein

IR

immunoreactive

LSC

lumbar sympathetic chain

MPG

major pelvic ganglion

TH

tyrosine hydroxylase

TRP

transient receptor potential cation channel

TRPV₁

TRP, subfamily V, member 1

VGLUT

vesicular glutamate transporter