Transient Receptor Potential Vanilloid Type 1 Channel (TRPV1) Immunolocalization in the Murine Enteric Nervous System Is Affected by the Targeted C-terminal Epitope of the Applied Antibody

Abstract

The expression of transient receptor potential vanilloid type 1 channel (TRPV1) in the enteric nervous system is still the subject of debate. Although a number of studies have reported that TRPV1 is limited to extrinsic afferent fibers, other studies argue for an intrinsic expression of TRPV1. In the present study, reverse transcriptase PCR was employed to establish the expression of TRPV1 mRNA throughout the gastrointestinal tract. Using two antibodies directed against different epitopes of TRPV1, we were able to show at the protein level that the observed distribution pattern of TRPV1 is dependent on the antibody used in the immunohistochemical staining. A first antibody indeed mainly stained neuronal fibers, whereas a second antibody exclusively stained perikarya of enteric neurons throughout the mouse gastrointestinal tract. We argue that these different distribution patterns are due to the antibodies discriminating between different modulated forms of TRPV1 that influence the recognition of the targeted immunogen and as such distinguish intracellular from plasmalemmal forms of TRPV1. Our study is the first to directly compare these two antibodies within the same species and in identical conditions. Our observations underline that detailed knowledge of the epitope that is recognized by the antibodies employed in immunohistochemical procedures is a prerequisite for correctly interpreting experimental results.

The transient receptor potential vanilloid type 1 channel (TRPV1), also known as the capsaicin receptor or the vanilloid receptor subtype 1, was first described by Caterina et al. (1997) and has since then been proven to play a central role in pain mechanisms (reviewed in Vennekens et al. 2008). TRPV1 is a cation channel that reacts to noxious stimuli, such as low pH (<5.9) and hyperthermia (>43C). These activation properties also make it an interesting mediator in inflammation. Next to this direct channel activation, TRPV1 is potentiated by inflammatory factors through changes in phosphorylation or, alternatively, through increased expression. Therefore, TRPV1 has long been assigned a proinflammatory role. However, evidence is emerging that TRPV1 also has protective functions (Alawi and Keeble 2010; Devesa et al. 2011). In gastrointestinal (GI) pathology, activation of TRPV1 aggravates intestinal inflammation through the release of substance P (McVey and Vigna 2001; McVey et al. 2003; Engel et al. 2011). The sensitization of afferent nerves via TRPV1 affects GI motility in inflammatory conditions (De Schepper, De Man, et al. 2008; De Winter et al. 2009) and promotes inflammation- and stress-related hypersensitivity (De Schepper, De Winter, et al. 2008; van den Wijngaard et al. 2009). Despite these clear indications that TRPV1 is involved in GI functioning, its expression in the enteric nervous system (ENS) remains a matter of debate. Indeed, the effects of TRPV1 on GI functioning have mostly been linked to its expression in extrinsic nerve fibers. Regardless of GI region or species, TRPV1 is widely known to be expressed in afferent nerve fibers (Guo et al. 1999; Yiangou et al. 2001; Poonyachoti et al. 2002; Anavi-Goffer and Coutts 2003; Chan et al. 2003; De Jonge et al. 2003; Ward et al. 2003; Schicho et al. 2004; Kadowaki et al. 2004; Faussone-Pellegrini et al. 2005; Matsumoto et al. 2011). Some of these studies also reported an intrinsic expression of TRPV1 in the GI tract of several mammalian species (Akiba et al. 2001; Kulkarni-Narla and Brown 2001; Anavi-Goffer et al. 2002; Poonyachoti et al. 2002; Anavi-Goffer and Coutts 2003; Chan et al. 2003; Faussone-Pellegrini et al. 2005), as further corroborated by a physiological study in the mouse by Penuelas et al. (2007). Others, however, were unable to establish TRPV1 expression in enteric neurons. In a study in guinea pig, the observed TRPV1 expression was dependent on the antibody used during the immunohistochemical procedure (Anavi-Goffer and Coutts 2003). One of the antibodies directed against TRPV1 did not give rise to TRPV1 immunoreactivity (IR) in myenteric perikarya. We speculated that discrepant results might result from the use of different antibodies and investigated the differences in immunolocalization of TRPV1 in the GI tract of the mouse, using two different, commercially available antibodies. Like the study by Anavi-Goffer and Coutts (2003) in guinea pig, we report that the two different antibodies specific for TRPV1 resulted in two distinct distribution patterns of TRPV1 throughout the murine GI tract. Whereas the results of the first antibody confirmed earlier findings that TRPV1 is mainly expressed in extrinsic nerve fibers, the second antibody resulted in an extensive intracellular expression of TRPV1 throughout the ENS.

Materials and Methods

Tissue Preparation

All animal handling and housing procedures were conducted in accordance with European directive 86/609/EEC. Tissues were obtained from adult C57BL/6J mice and Wistar rats (Janvier, Le Genest St Isle, France). Animals were housed in a 12/12-hr light/dark cycle at a constant temperature of 22C and were provided with ad libitum access to a standard pellet diet and water. Animals were sacrificed by cervical dislocation followed by exsanguination. The GI tract was removed and washed in Krebs solution (117 mM NaCl, 5 mM KCl, 2.5 mM CaCl₂·2H₂O, 1.2 mM MgSO₄·7H₂O, 25 mM NaHCO₃, 1.2 mM NaH₂PO₄·2H₂O, and 10 mM glucose; pH 7.4). Tissue samples for PCR analysis were snap-frozen in liquid nitrogen and stored at −80C until RNA isolation. Tissue parts for cryosectioning were fixed for 2 hr in Zamboni’s fixative, washed in 0.01 M phosphate-buffered saline (PBS), and rinsed using the Llewellyn-Smith procedure (Llewellyn-Smith et al. 1985). After overnight incubation in 20% sucrose at 4C, these tissues were embedded in OCT (Pelko Int.; Torrance, CA) and sectioned at 12 µm. Cryostat sections were thaw-mounted on poly-L-lysine–coated slides. Whole mounts were prepared by opening the organs and pinning them out on Sylgard-coated Petri dishes. Tissues were then fixed with 4% paraformaldehyde for 2 hr and washed with 0.01 M PBS prior to further dissection. Next, the mucosa/submucosa was separated from the external musculature. The circular muscle layer of the intestinal and gastric musculature and the inner part of the esophageal striated muscle layer were (partially) removed under a stereomicroscope. For submucous plexus whole-mount preparations of the intestine, the mucosa was carefully scraped from the submucosa under a stereomicroscope.

RNA Isolation and Reverse Transcriptase PCR

Total RNA was isolated as described previously (Van Op den Bosch et al. 2007). In summary, the frozen tissue samples (see earlier) were mechanically homogenized, after which RNA was isolated using Trizol reagent (Life Technologies; Gaithersburg, MD). Concentration and purity of the isolated RNA were analyzed using a Nanodrop ND-1000 system (Nanodrop; Wilmington, DE). Then, 5 µg isolated RNA was further purified with the Turbo DNA-free kit (Life Technologies), and 1 µg DNase-treated RNA was reverse transcribed with the Transcriptor First Strand cDNA synthesis kit (Roche; Mannheim, Germany) on an MJ Mini Cycler (Bio-Rad; Hercules, CA). Primers were synthesized by Biolego (Nijmegen, the Netherlands). The forward primer (FP) sequence was 5′-CTTGTGGAGGTGGCAGA TA-3′, and the reverse primer (RP) sequence was 5′-TTCCA CAGGCCGATAGTAAG-3′. This primer set spans the exons 5, 6, and 7 (based on NCBI reference sequence NM_001001445.1) and yields an amplicon of 505 bp (TRPV1α) or 475 bp (TRPV1β). The reaction mix for RT-PCR contained (in µl) 10 RNAse-free water, 12.5 HotstarTaq Master Mix (Qiagen; Hilden, Germany), 0.25 FP (10 mM), 0.25 RP (10 mM), and 2 cDNA in a total volume of 25 µl. After denaturation at 95C for 15 min, 40 cycles were performed consisting of a denaturation step at 94C for 60 sec and an annealing step at 56C for 30 sec, followed by elongation at 72C for 90 sec. The RT-PCR procedure was ended by a final extension phase at 72C for 10 min. Amplification products were separated using a 2% agarose gel with 10 mg/ml ethidium bromide (Amresco; Solon, OH) and visualized under UV illumination (G:Box; Syngene, Cambridge, UK).

Immunohistochemistry

The immunohistochemistry procedures were performed at room temperature and, unless indicated otherwise, washing steps with PBS were conducted between incubations. Cryosections were dried at 37C. Next, the slides were sequentially incubated with avidin and biotin (Life Technologies Europe B.V.; Ghent, Belgium), each for 10 min. Then, 50 mM glycine was applied for 15 min to block excess free aldehydes. Cryosections were next permeabilized and blocked with 0.1 M PBS (pH 7.4) containing 0.05% thimerosal (PBS*), 10% normal horse serum (NHS), and 1% Triton X-100 for 1 hr. A mixture of primary antibodies raised in different species was applied and the antibodies were dissolved overnight in PBS* with 10% NHS and 0.1% Triton X-100. After a rinsing step, the cryosections were incubated with the appropriate secondary antibodies dissolved in PBS* with 1% NHS. Whole-mount preparations were permeabilized with PBS* containing 10% NHS and 1% Triton X-100 for 2 hr, followed by an overnight incubation with the first primary antibody in PBS* with 10% NHS and 0.1% Triton X-100. After rinsing, the corresponding secondary antibody was applied for 3 hr in PBS* with 1% NHS. The second primary antibody was again incubated overnight in PBS* with 1% NHS, followed by rinsing and incubation with the appropriate secondary antibody in PBS* with 1% NHS. Biotinylated secondary antibodies, whenever used, were incubated for 2 hr, followed by staining with a streptavidin-conjugated fluorophore for 3 hr in PBS. After staining, tissue was mounted in citifluor (Citifluor Ltd.; London, UK) and stored at 4C awaiting (confocal) fluorescence microscopy analysis. The specificity of the anti-TRPV1 antibodies has been demonstrated in earlier reports (Tominaga et al. 1998; Guo et al. 1999). Next to the expected ~90-kDa western blot band of TRPV1, Guo et al. (1999), who used the anti-TRPV1 raised in guinea pig, described an additional band at ~65 kDa and interpreted it as aspecificity at the time. However, this band is likely to be the VR.5′sv N-terminal splice variant, which was not described until a year later (Schumacher et al. 2000). We have performed an independent antigen profiling of mouse TRPV1 (Open Biosystems; Rockford, IL) and a protein BLAST (http://blast.ncbi.nlm.nih.gov/) to reveal the putative epitope regions of mouse TRPV1. These analyses confirmed the specificity of the immunogen sequences. To eliminate aspecific binding of the secondary antibodies, negative controls, in which the primary antibodies were omitted, were performed. All primary and secondary antibodies used in this study are listed in Table 1.

Table 1.

Primary and Secondary Antibodies

Primary Antibodies
Antigen	Source	Dilution	Supplier
TRPV1	Guinea pig	1:1000	Millipore, Temecula, CA; AB5566
TRPV1	Rabbit	1:100	Enzo Life Sciences, Lausen, Switzerland; SA-564
Peripherin	Rabbit	1:1000	Millipore, Temecula, CA; AB1530
PGP	Guinea pig	1:300	Abcam, Cambridge, UK; ab10410
Calretinin	Goat	1:5000	Swant, Bellinzona, Switzerland; CG1
nNOS	Goat	1:300	Abcam, Cambridge, UK; ab72428
CGRP	Goat	1:5000	Abcam, Cambridge, UK; ab36001
SP	Guinea pig	1:1000	Abcam, Cambridge, UK; ab10353
VIP	Goat	1:100	Santa Cruz Biotechnology, Santa Cruz, CA; sc6170
Secondary Antibodiesa
Source	Target	Conjugate	Dilution
Donkey	Guinea pig	Cy3	1:1000
Goat	Rabbit	Cy3	1:8000
Donkey	Rabbit	Cy3	1:800
Donkey	Guinea pig	FITC	1:300
Donkey	Goat	FITC	1:200
Donkey	Guinea pig	Biotin	1:1000
Streptavidin conjugate
FITC			1:2000

CGRP, calcitonin gene–related peptide; nNOS, neuronal nitric oxide synthase; PGP, protein gene product 9.5; SP, substance P; TRPV1, transient receptor potential vanilloid type 1 channel; VIP, vasoactive intestinal peptide.

^aAll from Jackson Immunoresearch (West Grove, PA).

Results

RT-PCR Reveals TRPV1 mRNA throughout the Mouse GI Tract

mRNA expression of TRPV1 was investigated in the esophagus, stomach, ileum, and proximal colon of three adult mice. RT-PCR yielded a positive result at 505 bp in all investigated tissues, as well as in the positive control brain tissue (Fig. 1). Non–reverse-transcribed DNase-treated mRNA samples did not yield an amplicon, indicating the absence of genomic DNA. Negative control samples with water instead of cDNA also lacked amplification.

Figure 1.

mRNA expression of transient receptor potential vanilloid type 1 channel (TRPV1) in the mouse gastrointestinal tract. TRPV1 mRNA was expressed in all examined regions of the gastrointestinal tract (esophagus, stomach, ileum, and proximal colon). Brain tissue was used as a positive control, whereas the negative control did not yield a signal.

Different Anti-TRPV1 Antibodies Reveal a Distinct Localization of TRPV1

To evaluate TRPV1 protein expression in the ENS, we used two antibodies targeting TRPV1; that is, one raised in guinea pig (Gp) and one raised in rabbit (Rb) (see Table 1). Cryosections and whole-mount preparations of the mouse GI tract were made. Both Gp–anti-TRPV1 and Rb–anti-TRPV1 resulted in stained structures in the enteric nerve plexuses at all investigated levels of the GI tract (esophagus, stomach, ileum, and proximal colon) (Fig. 2). However, the staining patterns of these antibodies were found to be strikingly different. Epitope and sequence analysis showed that both antibodies were directed against highly specific parts of the C-terminus epitope of TRPV1. Gp–anti-TRPV1 revealed an intracellular granular IR in the soma of virtually all enteric neurons at all levels of the mouse GI tract. Rb–anti-TRPV1 IR was clearly present in nerve fibers in the enteric plexuses (Fig. 2), but the extent to which it stained neuronal cell bodies depended on the GI region. The esophagus harbored most Rb–anti-TRPV1–immunoreactive neurons (Fig. 2a–c), being about twice the number of immunoreactive intrinsic neurons seen in the myenteric plexus of the stomach (Fig. 2d–e) and the proximal colon (Fig. 2j–l), where approximately 1 in 10 neurons were labeled with Rb–anti-TRPV1. In the ileum, cellular IR for Rb–anti-TRPV1 was rare (Fig. 2g–i). The submucosal plexus of both ileum and proximal colon did not contain any Rb–anti-TRPV1–positive neuronal cell bodies. The aforementioned antibody-dependent distribution pattern of TRPV1 was not limited to mouse tissue because similar results were obtained in the rat ileum (Fig. 3). We have used two neuronal markers: a pan-neuronal marker protein gene product 9.5 (PGP) and an antibody directed against the intermediate filament peripherin, which is not a pan-neuronal marker in the adult ENS but often better delineates neuronal morphology within a ganglion. We were able to confirm that the Gp–anti-TRPV1 staining is indeed present in enteric neurons (Fig. 4a–d). The Rb–anti-TRPV1–immunoreactive nerve fibers extensively contacted enteric neurons in ENS ganglia (Fig. 4e–f). No co-localization was observed between both markers for TRPV1 and glial (S100 protein, glial fibrillary acidic protein [GFAP]) or macrophage (CD169) markers (Fig. 5). Gp–anti-TRPV1–immunoreactive neuronal perikarya co-stained with calretinin and neuronal nitric oxide synthase (nNOS) (Fig. 6a–c). In the submucous plexus, again co-localization with calretinin could be observed (Fig. 6d). In addition, all neurons expressing vasoactive intestinal peptide (VIP) also expressed intracellular Gp–anti-TRPV1 IR. Moreover, the subcellular distribution of VIP and TRPV1 was comparable (Fig. 6e–g). Fibers stained with Rb–anti-TRPV1 were observed in the primary and secondary myenteric plexus and in the submucous plexus (Fig. 7) and showed co-localization with calcitonin gene–related peptide (CGRP) in both the myenteric and the submucous plexus (Fig. 7a–f). Partial co-localization with substance P was observed in the submucous plexus but not in the myenteric plexus (Fig. 7g–l).

Figure 2.

Transient receptor potential vanilloid type 1 channel (TRPV1) immunoreactivity (IR) in the enteric nervous system. Gp–anti-TRPV1 (FITC, green) and Rb–anti-TRPV1 (Cy3, red) stain enteric plexuses throughout the gastrointestinal (GI) tract. Gp–anti-TRPV1 IR is observed in nearly all perikarya, whereas Rb–anti-TRPV1 immunoreactive perikarya are rare and mostly observed in the upper GI tract (esophagus-stomach). Rb–anti-TRPV1 IR, however, is prominently present in fibers throughout the GI tract. Scale bar: 20 µm.

Figure 3.

Expression of transient receptor potential vanilloid type 1 channel (TRPV1) in the rat. Staining of the myenteric plexus of the rat ileum shows that the mainly cellular stain by Gp–anti-TRPV1 versus the mainly fibrous stain by Rb–anti-TRPV1 is not confined to the mouse species. Scale bar: 20 µm.

Figure 4.

Neuronal expression of transient receptor potential vanilloid type 1 channel (TRPV1). Cryosections (a, b) and whole mounts (c–f) of the mouse gastrointestinal tract show co-localization of Gp–anti-TRPV1 with peripherin (a–d). The neuronal fibers stained with Rb–anti-TRPV1 are closely associated with enteric neurons as shown by co-staining with PGP (e–f). Scale bar: 20 µm.

Figure 5.

Transient receptor potential vanilloid type 1 channel (TRPV1) is not expressed in glia or macrophages. Glial markers S100 and GFAP do not co-localize with Gp– or Rb–anti-TRPV1 (a, b). Cell bodies immunopositive for Gp–anti-TRPV1 do not contain the macrophage marker CD169 (c). Scale bar: 20 µm.

Figure 6.

Neurochemical coding of transient receptor potential vanilloid type 1 channel (TRPV1)–expressing enteric neurons. The perikarya immunoreactive for Gp–anti-TRPV1 do not belong to a specific neuronal subtype. Both neuronal nitric oxide synthase (nNOS)–immunopositive (a, b) and CalR-immunopositive (c) enteric neurons show immunoreactivity for Gp–anti-TRPV1 in the myenteric plexus. In the submucous plexus, Gp–anti-TRPV1 stains CalR-immunopositive (d) and vasoactive intestinal peptide (VIP)–immunopositive neurons (e–g). Note the comparable subcellular distribution of VIP and Gp–anti-TRPV1. Scale bar: 20 µm.

Figure 7.

Expression of Rb–anti–transient receptor potential vanilloid type 1 channel (TRPV1) in extrinsic fibers. Confocal images showing that Rb–anti-TRPV1–immunopositive neuronal fibers co-localize with calcitonin gene–related peptide (CGRP) in both the myenteric and the submucous plexus (a–f). Partial co-localization with substance P is found in the submucous plexus (g–i) but not in the myenteric plexus (j–l). Scale bar: 20 µm.

Discussion

The expression of TRPV1 in the ENS remains a matter of debate. It is well established in various species that TRPV1 is expressed in extrinsic nerve fibers innervating the GI tract (Yiangou et al. 2001; Poonyachoti et al. 2002; Anavi-Goffer and Coutts 2003; Chan et al. 2003; De Jonge et al. 2003; Ward et al. 2003; Kadowaki et al. 2004; Schicho et al. 2004; Faussone-Pellegrini et al. 2005; Matsumoto et al. 2011). However, the expression of TRPV1 in intrinsic enteric neurons is still controversial. Myenteric neurons were found to be TRPV1 immunoreactive in the guinea pig colon and ileum (Anavi-Goffer et al. 2002; Anavi-Goffer and Coutts 2003), in the rat GI tract (Akiba et al. 2001; Anavi-Goffer et al. 2002; Anavi-Goffer and Coutts 2003), and in the porcine ileum (Kulkarni-Narla and Brown 2001; Poonyachoti et al. 2002). Chan et al. (2003) reported TRPV1 expression in submucosal and myenteric neurons in the human rectum. Likewise, both mRNA and protein expression of TRPV1 was demonstrated in submucosal and myenteric neurons and in neuronal fibers of the human stomach (Faussone-Pellegrini et al. 2005). Penuelas et al. (2007) suggested neurally mediated capsaicin-induced contractions in an isolated preparation of the mouse colon. Other studies, however, were unable to confirm this intrinsic expression in the rodent GI tract (Patterson et al. 2003; Ward et al. 2003; Schicho et al. 2004; Suckow and Caudle 2008; Matsumoto et al. 2011) or the human colon (Yiangou et al. 2001). Using RT-PCR, we have shown the presence of TRPV1 mRNA in the mouse GI tract, which suggests intrinsic expression of TRPV1, although others demonstrated that TRPV1 mRNA disappears from the rat GI tract after vagotomy and celiac ganglionectomy (Schicho et al. 2004). This latter finding suggests an axonal distribution of mRNA in extrinsic fibers with local translation, a mechanism that has been described in neurons of the dorsal root ganglia (reviewed in Lin and Holt 2008). Using immunohistochemical methods, we observed extrinsic as well as intrinsic expression of TRPV1, but our findings depend on the type of antibody used. When using the anti-TRPV1 antibody raised in guinea pig against the immunogen YTGSLKPEDAE VFKDSMAPGEK (mouse aa817–839), only a granular intrinsic TRPV1 IR was visible in perikarya throughout the mouse GI tract. Gp–anti-TRPV1 IR was found in nearly all enteric neurons, and co-staining with neurochemical markers confirmed IR in perikarya belonging to both excitatory and inhibitory myenteric neurons and to secretomotor/vasodilator submucous neurons (Qu et al. 2008; Fantaguzzi et al. 2009). On the other hand, the anti-TRPV1 antibody raised in rabbit against the immunogen EDAEVFKDSMAPGEK (mouse aa825–839) yielded only sporadic intrinsic expression, which was mainly confined to the upper GI tract, but instead showed IR in fibers coexpressing CGRP and SP. Based on previous studies in the mouse by our group and others and the limited IR in perikarya, these fibers are presumed to originate extrinsically in the spinal and vagal ganglia (De Jonge et al. 2003; Ward et al. 2003; Matsumoto et al. 2011). The intracellular staining by Gp–anti-TRPV1 hints at a splice variant because this splicing affects translocation to the membrane (Hellwig et al. 2005). In mouse, two TRPV1 isoforms have been identified: TRPV1α, the most common form, and TRPV1β, a splice variant, which lacks 10 amino acids in the N-terminal region and does not form functional channels but may act as a negative regulator of TRPV1 activity (Xue et al. 2001; Wang et al. 2004). Our RT-PCR results did not reveal a second band, although our primer set spanned the spliced region. This absence could be explained by the fact that TRPV1β is hard to detect with RT-PCR (Wang et al. 2004); however, because both antibodies recognize the C-terminus, the antibody binding is probably not influenced by an N-terminal deletion. Consequently, TRPV1β is unlikely to underlie the variation in staining. Intracellularly located TRPV1, however, may be a functional form of the receptor. The presence of functional TRPV1 on the membrane of the endoplasmic reticulum has been described (Olah et al. 2001). TRPV1 on the endoplasmic reticulum and the Golgi apparatus has been found to be involved in the regulation of Ca²⁺ homeostasis, which has been established in non-excitable cells as well as in spinal neurons (Olah et al. 2001; Liu et al. 2003; Turner et al. 2003; Wisnoskey et al. 2003; Kárai et al. 2004; Gallego-Sandín et al. 2009). Intracellular TRPV1 localization on the Golgi apparatus is also in keeping with the subcellular distribution being similar to VIP, which we observed in our experiments, because VIP is known to localize to the Golgi apparatus (Johansson 1983; Turner et al. 2003). This localization is further in accordance with the findings of Guo et al. (1999), who described TRPV1 in the Golgi apparatus of dorsal root ganglion neurons using a Gp–anti-TRPV1 antibody aimed at the same epitope. These different staining patterns might hence result from conformational differences or variation in the phosphorylation state between intracellular and plasmalemmal TRPV1. Differences in interacting molecules also influence epitope recognition by antibodies. The C-terminal region of TRPV1 contains several modulatory regions, including phosphorylation sites at T705, S801, S775, and S821, as well as binding sites for calmodulin (aa768–802) and phosphatidylinositol 4,5-bisphosphate (aa778–793) (Bhave and Gereau 2004; Messeguer et al. 2006). Functional differences between TRPV1 in the plasmalemmal and intracellular TRPV1 have been attributed to different regulatory interactions (Gallego-Sandín et al. 2009), but the regulation of intracellular TRPV1 is still largely unknown.

We screened the use of the two antibodies in previous studies of the ENS and found that the Rb–anti-TRPV1 antibody was used in seven studies, five of which described an extrinsic expression of TRPV1 in mice, rats, and guinea pigs (Patterson et al. 2003; Ward et al. 2003; Kadowaki et al. 2004; Schicho et al. 2004; Matsumoto et al. 2011), whereas the other two studies (originating from the same group) described an intrinsic expression of TRPV1 in the guinea pig ENS (Anavi-Goffer et al. 2002; Anavi-Goffer and Coutts 2003). These different findings might be related to species differences. To our knowledge, the Gp–anti-TRPV1 antibody has been used only in two studies on pig ENS, which showed IR in enteric neurons (Kulkarni-Narla and Brown 2001; Poonyachoti et al. 2002). Some studies also used anti-TRPV1 raised in goat (supplied by Santa Cruz Biotechnologies) or anti-TRPV1 raised in rabbit (supplied by Affinity Bioreagents; now Thermo Scientific, Billerica, MA), but the exact epitopes of these antibodies are not disclosed. Our study is the first to compare the Rb–anti-TRPV1 and Gp–anti-TRPV1 antibodies within the same species and in identical conditions, allowing direct comparison.

In conclusion, we demonstrated the intracellular expression of TRPV1 in the perikarya of enteric neurons and argue that the discrepant results of studies dealing with TRPV1 expression in the ENS result from differences in antibody specificity discriminating between different modulated forms of TRPV1 in the enteric neurons. More generally, our findings show that it is highly recommended to correlate epitope data with protein modifications and splice variants. However, some commercial suppliers still regard epitope and/or immunogen data as confidential, which might hamper correct data interpretation.