Lipopolysaccharide-induced Pulpitis Up-regulates TRPV1 in Trigeminal Ganglia

Abstract

Tooth pain often accompanies pulpitis. Accumulation of lipopolysaccharides (LPS), a product of Gram-negative bacteria, is associated with painful clinical symptoms. However, the mechanisms underlying LPS-induced tooth pain are not clearly understood. TRPV1 is a capsaicin- and heat-gated nociceptive ion channel implicated in thermosensation and hyperalgesia under inflammation or injury. Although TRPV1 is expressed in pulpal afferents, it is not known whether the application of LPS to teeth modulates TRPV1 in trigeminal nociceptors. By assessing the levels of protein and transcript of TRPV1 in mouse trigeminal ganglia, we demonstrate that dentinal application of LPS increases the expression of TRPV1. Our results suggest that the up-regulation of TRPV1 in trigeminal nociceptors following bacterial infection could contribute to hyperalgesia under pulpitis conditions.

Introduction

Dental pain is the most frequent etiology causing patients to visit dental clinics. Tooth pain during or following treatment is also the main cause of preventing them from visiting dental clinics regularly. Therefore, tooth pain needs to be managed appropriately before, during, and after dental treatment. However, management of tooth pain is often problematic, and the development of mechanism-based therapeutic approaches is hampered because of the unclear neurobiological mechanisms of tooth pain.

Dental caries is an infectious disease involving a wide spectrum of oral bacteria. As the disease progresses beyond the enamel layer and reaches dentin, bacteria and their products gain access to pulp through dentinal tubules, and trigger a cascade of pathological changes (Love and Jenkinson, 2002). Pulpal inflammation induced by bacteria and their products frequently results in tooth pain. Although not all pulpitis patients suffer from pain (Michaelson and Holland, 2002), clinical studies have shown the correlation between the pulpal invasion by Gram-negative bacteria and their product, lipopolysaccharide (LPS), and increased thermal sensitivity and painful symptoms in individuals with pulpitis (Schein and Schilder, 1975; Hahn et al., 1993; Khabbaz et al., 2000; Jacinto et al., 2005). However, the mechanism of LPS-induced tooth pain is not completely understood.

Transient receptor potential vanilloid 1 (TRPV1) is a receptor of capsaicin, a pungent substance found in chili peppers, and can also be activated by protons and a noxious level of heat (> 43oC). TRPV1 is enriched in nociceptive sensory neurons and plays a critical role in inflammatory pain (Szallasi et al., 2007). Since TRPV1 is reported to be expressed in pulpal afferents in humans and rats (Ichikawa and Sugimoto, 2001; Morgan et al., 2005; Park et al., 2006; Kim et al., 2009), it is the best candidate for the induction of tooth pain and thermal hyperalgesia in pulpitis. Previous studies demonstrated that up-regulation of TRPV1 in sensory ganglia following inflammatory peripheral insult is one of the mechanisms contributing to inflammatory hyperalgesia (Szallasi et al., 2007). However, it is not known whether LPS-induced pulpitis modulates the expression of TRPV1 in trigeminal ganglia (TG). To examine this, we assessed TRPV1 expression using Western blot and real-time reverse-transcription polymerase chain-reaction (RT-PCR) methods in TG following dentinal application of LPS. We performed these experiments in mice in an effort to establish a mouse model of pulpitis for the use of genetically engineered mice in future mechanistic studies.

Materials & Methods

Experimental Animals

All procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and under a University of Maryland-approved Institutional Animal Care and Use Committee protocol. All experiments were carried out on adult male C57BL/6 mice (weighing from 22 to 30 g). In total, 56 mice were used in this study.

Preparation of a Cavity in Mouse Molars

Each mouse was anesthetized with sodium pentobarbital (intraperitoneal, 50 mg/kg) and fixed on a mouse stereotaxic frame (Stoelting, Kiel, WI, USA), with bars immobilizing the zygomatic arch on both sides of the head to achieve maximal stability during the drilling procedure. The abdomen of the mouse faced up, and the head was further immobilized by means of a tooth bar fixing the upper incisors and snout. The mouth was kept open by retraction of the lower incisors by means of a rubber retractor fixed on the surgery platform. For visualization of mouse molars, we performed the procedure using a surgical microscope (X500, Motic, Richmond, BC, Canada). The cavity was prepared by means of a sterile new fissurotomy bur (SS White Burs Inc., Lakewood, NJ, USA; diameter ~ 0.3 mm at the tip) driven by a high-speed drill in the presence of a continuous flow of water on the tooth. The entire layer of enamel overlying the mesial slope of the intermediate cusp and mesial fossa of the maxillary left first molar was removed. We used fine paper points to dry the tooth and inspected it carefully to detect any sign of exposure of the pulp horn, such as bleeding from tooth pulp. These methods allowed us to expose the dentinal tubule to tooth pulp with minimal damage.

Dentinal Application of LPS and Retrograde Labeling Dye

For retrograde labeling of pulpal afferents, we placed crystals of Fluorogold (Fluorochrome, Englewood, CO, USA) into the cavity following the preparation as described above. To minimize any potential effects of pulpal injury, we applied Fluorogold to the surface of dentin without exposing the pulp. Then the cavity was sealed with a light-cured resin (Maxcem, Kerr, Orange, CA, USA) to prevent leakage. To apply LPS, we used a fine paper point (~ 0.5 mm) as a carrier to deliver liquid to the cavity. To exclude the effects of contaminating oral bacteria, we did not expose the tooth pulp, but applied LPS to freshly cut dentin. LPS (derived from E. coli; Sigma, St. Louis, MO, USA; 10 mg/mL) or saline was applied to the cavity. We were careful not to overfill the cavity, which was then allowed to dry. This wet-and-dry procedure was repeated 5 or 6 times. After the applied substances dried, the cavity was sealed with a light-cured resin. Softened food was provided following the surgery, and the mice showed no signs of sickness. The changes of body weight for 7 days following surgery were not significantly different between saline- and LPS-treated groups (data not shown).

Immunohistochemistry of Teeth

Mice were sacrificed 4 hrs following the dentinal application of saline or LPS to the maxillary 1^st molar. The mice were transcardially perfused with 4% paraformaldehyde. The maxillae were dissected and decalcified with 10% EDTA for 5 to 7 days at 4°C. Tissue was cryoprotected with 30% sucrose, cryosectioned at 30-µm intervals, and collected on slides. To label inflammatory cells within dental pulp, we stained the CD45 that is expressed in all nucleated cells of hematopoietic origin (Thomas, 1989). Immunohistochemistry was performed with a rat antibody against CD45 (AbD serotec, Raleigh, NC, USA; 1:1000), and a goat secondary antibody conjugated with Dylight-488 (Jackson ImmunoResearch, West Grove, PA, USA).

Immunohistochemistry of TG

Mice were perfused transcardially. TG were dissected, cryoprotected, and sectioned at 10- to 12-µm intervals. Four alternative sets of sections were collected. The conventional procedures of immunohistochemistry were performed with antibodies against TRPV1 raised in goat antiserum (Santa Cruz Laboratories, Santa Cruz, CA, USA; 1:2000) and secondary antibody conjugated with FITC (Jackson ImmunoResearch). The images were taken by fluorescence microscopy (Zeiss Axiovert, Carl Zeiss MicroImaging, Thornwood, NY, USA). The Fluorogold signal was identified under gold filter (11006v2, Chroma, Bellows Falls, VT, USA). We used Image J to calculate the surface area of the cells. To generate size-distribution plots, we measured 99 fluorogold-labeled and 600 unlabeled neurons from 3 ganglia. We followed the criteria for neuronal size classification (small, < 300 µm²; medium, 300 to 600 µm²; large, > 600 µm²) (Ichikawa et al., 2006).

Western Blot Analysis

To confirm the specificity of TRPV1 antibody, we obtained protein lysates from HEK293 cells heterologously transfected with cDNA encoding TRPV1 (a positive control), untransfected naïve HEK293 cells (a negative control), TG from a wild-type C57/Bl6 mouse and a TRPV1-null mutant mouse (Jackson Laboratory, Bar Harbor, ME, USA), and a Sprague-Dawley rat (Harlan Laboratories, Indianapolis, IN, USA). HEK293 cells were cultured and transfected following methods described previously (Chung et al., 2008). Western blot analysis was performed as previously described (Lee and Ro, 2007). We used primary antibodies for TRPV1 (1:500, Santa Cruz #P-19) and β-actin (1:5000, Sigma), and secondary antibodies conjugated to horseradish peroxidase (1:5000). Bands were visualized with ECL Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ, USA). Densitometry of each band was performed, and the protein level for TRPV1 was normalized to that of β-actin, an internal control, in the same sample. The TRPV1/actin ratio was normalized again to the mean value of the saline control group and expressed as ‘relative intensity’.

Real-time Quantitative RT-PCR

We extracted total RNA from TG by using Trizol (Invitrogen, Carlsbad, CA, USA) and further purified it with an RNeasy kit (Qiagen, Valencia, CA, USA). Reverse transcription was carried out with the Superscript first-strand synthesis kit (Invitrogen). PCR amplification was performed with SYBR Green supermix (Fermentas, Hanover, MD, USA) in a PCR program with 10-minute denaturation at 95°C followed by 45 cycles at 95°C for 15 sec, 60°C for 20 sec, and 68°C for 30 sec in an Realplex real-time thermocycler (Eppendorf, New York, NY, USA). The specific PCR primers for TRPV1 and GAPDH, an internal control, were as follows: TRPV1 (NM_001001445), 5′-CCTGCAT TGACACCTGTGAA-3′ (forward), 5′-AGTCGGTTCAAGGGTT CC A-3′ (reverse); and GAPDH (M32599.1), 5′-CAA TGC ATCCTGCACCACCAA-3′ (forward), 5′-GTCATTGAGAGCA ATGCCAGC-3′ (reverse). The change of mRNA level was assessed by relative quantification methods. We obtained the ratios between TRPV1 and GAPDH to calculate the relative abundance of mRNA transcripts in each experimental sample. In each PCR run, the TRPV1/GAPDH ratio of sample from the LPS-treated group was normalized to the ratio of sample from the saline-treated group, which was referred to as ‘relative expression’.

Results

To demonstrate the expression of TRPV1 in mouse pulpal afferents, we retrogradely labeled pulpal afferents by applying Fluorogold to the maxillary first molars and performed immunohistochemical staining of TG using antibody against TRPV1. Pulpal afferents were identified by the discrete gold-colored particles within cytoplasm (Fig. 1a). Fluorogold-labeled afferents were composed mainly of medium- (23%) to large-sized (48%) neurons (Fig. 1). Such size distribution is consistent with the results previously found in rats (Ichikawa and Sugimoto, 2001; Kvinnsland et al., 2004). When we co-labeled TRPV1, 10% of pulpal afferents (10/99) were TRPV1-positive, similar to the percentage in rat pulpal afferents (8%) (Ichikawa and Sugimoto, 2001; Kvinnsland et al., 2004). The TRPV1-positive pulpal afferents were mostly small- (4/10) to medium-sized (3/10), and the mean surface area was 496 ± 85 µm² (mean ± SEM, n = 10). These results suggest that TRPV1 is expressed in a subpopulation of mouse pulpal afferents.

Figure 1.

The expression of TRPV1 in trigeminal ganglia (TG) pulpal afferent neurons. (A) Fluorogold-labeled primary afferents in a TG section. Scale bar: 40 µm. (B) Immunohistochemical staining with TRPV1 antibody. (C) A merged image of A and B. The arrow indicates a TRPV1-positive pulpal afferent neuron. (D) Size distribution of primary afferents labeled by fluorogold.

To verify whether the dentinal application of LPS evokes pulpitis, we examined the number of immune cells within tooth pulp demonstrated by labeling with CD45, a pan-leukocyte marker (Thomas, 1989). LPS or saline was applied to maxillary first molars, and adjacent second molars that had not been drilled served as internal controls. We counted ameboid or round cells with strong membrane-delimited CD45 staining (presumably activated macrophages or recruited leukocytes) in the pulp under the cusps of 1^st or 2^nd molars (Figs. 2A, 2B). The recruitment of inflammatory cells within the tooth pulp was significantly greater in the LPS-treated group than in the saline-treated group. However, a similar increase was not observed in adjacent second molars in both the LPS- and saline-treated groups (Fig. 2C). These results suggest that dentinal application of LPS to mouse molars can evoke localized inflammatory reaction within tooth pulp.

Figure 2.

Dentinal application of LPS evoked immune responses in the pulp. (A,B) Immunohistochemical staining of CD45 in decalcified mouse molars treated with saline (A) or LPS (B). Scale bar, 50 µm. Arrows indicate examples of cells that were counted. (C) Number of inflammatory cells per unit area (mean ± SEM; n = 4; *p < 0.05, Student’s t test).

Next, we examined whether the dentinal application of LPS induced up-regulation of TRPV1 in TG by examining the protein levels 1, 3, and 7 days following the application of LPS or saline. Initially, we verified the specificity of TRPV1 antibody. The primary antibody detected a common band in recombinant TRPV1, mouse TG, and rat TG, and the band of corresponding size was absent in samples from naïve HEK293 or TG of TRPV1-null mutants (Fig. 3A). These results confirmed that the antibody specifically recognized TRPV1 in our conditions. When we compared the relative expression of TRPV1 in TG from saline- and LPS-treated groups, the expression of TRPV1 was significantly greater by ~2-fold in LPS group after 1 day of application. However, this difference disappeared at days 3 and 7 (Figs. 3B, 3C). Such up-regulation of TRPV1 was not observed in contralateral TG (p > 0.2, n = 4), suggesting that the effects of LPS were localized but not systemic. Since pulpal afferent terminals are significantly affected by even minor tooth injury, such as tooth drilling without pulp exposure (Byers and Närhi, 1999), we investigated whether our manipulation affected TRPV1 expression. The application of saline following tooth drilling without exposure of pulp did not significantly alter the expression of TRPV1 in TG compared with naïve TG (Fig. 3D).

Figure 3.

Relative expression of TRPV1 protein in TG following the dentinal application of LPS and saline. (A) Verification of specificity of TRPV1 antibody. HEK 293, Human Embryonic Kidney (HEK) 293 cells; TRPV1 +/+, wild-type; TRPV1 -/-, TRPV1-null mutant mice. (B) Representative Western blot of lysate of TG on days 1, 3, and 7 following the pulpal application of saline (S) and LPS (L) with the antibody against TRPV1 or β-actin. (C) The relative intensity of TRPV1 in ipsilateral TG in saline- (white) and LPS-treated (black) groups (n = 4; *p < 0.05, Student’s t test). (D) Comparison of the expressions of TRP channels in TGs from naïve animals with those from saline-treated animals on different days. The naïve, day 1, day 3, and day 7 samples were run in the same gel (n = 4; p > 0.7, Kruskal-Wallis one-way ANOVA on rank).

As an independent measurement of TRPV1 expression, we also examined the effects of LPS application on TRPV1 mRNA in TG by real-time RT-PCR. Consistent with the results in Western blot analysis, TRPV1 mRNA levels were significantly up-regulated in the LPS-treated group compared with those in the saline-treated group at day 1, but not at day 3 (Fig. 4). These results strongly suggest that dentinal application of LPS up-regulates TRPV1 in TG.

Figure 4.

Relative expression of TRPV1 mRNA in TG following the dentinal application of LPS or saline. Relative ratio of TRPV1/GAPDH between saline (white)- and LPS (black)-treated groups on day 1 (A) and day 3 (B) (n = 4; *p < 0.05, Mann-Whitney rank-sum test).

Discussion

In this study, we demonstrated that dentinal application of LPS up-regulates TRPV1 in mouse TG. We successfully established methods for labeling pulpal afferents from mouse molars and evoking pulpal inflammation using LPS. We expect that these methods will allow us to take advantage of genetically modified mice to study mechanisms of tooth pain and plasticity of pulpal afferents in the future.

Bacterial products applied to freshly cut dentin have been shown to induce pulpal inflammation in experimental animals. Dentinal application of LPS has been shown to induce pulpal inflammation in monkey molars and ferret canines (Warfvinge et al., 1985; Chattipakorn et al., 2002). In this paper, we demonstrated that application of LPS to dentinal surfaces induced pulpal inflammatory responses in mouse molars. Importantly, we established non-invasive methods for evoking pulpal inflammation in mouse molars. This approach allows us to minimize any potential effects of pulpal injury and to study the consequences of LPS application separately from unknown sources of inflammation.

Our results show that dentinal application of LPS significantly up-regulates TRPV1 mRNA and protein in mouse TG after 1 day compared with the saline-treated group. Recently, LPS was shown to increase capsaicin-induced neuropeptide release from rat TG neurons, suggesting that LPS sensitizes the function of TRPV1 in trigeminal nociceptors (Ferraz et al., 2011). Therefore, pulpal invasion of LPS enhances both the function and expression of TRPV1 in TG, which could contribute to hyperalgesia in pulpitis, especially during an acute time-course. The mechanisms underlying LPS-induced up-regulation of TRPV1 are unclear, and multiple mechanisms might be involved. LPS binds to CD14 and Toll-like receptors-4 (TLR-4) in the cell membrane of immune and non-immune cells, which results in the secretion of chemokines and cytokines (Palsson-McDermott and O’Neill, 2004). These factors could, in turn, affect sensory neuronal afferents. Alternatively, LPS can directly modulate pulpal afferents. TLR4 is expressed in a subpopulation of dorsal root ganglia neurons that contain substance P (Shanley et al., 2011). Moreover, CD14 and TLR4 are expressed in pulpal afferent terminals and trigeminal nociceptors, mainly in a TRPV1-positive subpopulation (Wadachi and Hargreaves, 2006; Diogenes et al., 2011; Ferraz et al., 2011). Further experiments are necessary to clarify the relative contributions of these mechanisms.

When one considers the paucity of dental afferents in TG and the extent of TRPV1 expression in pulpal afferents, the TRPV1 up-regulation observed in this study cannot be fully explained exclusively by that in pulpal afferents, but apparently involves non-pulpal afferents as well. This may not be due to the systemic effects of LPS, since LPS-induced recruitment of inflammatory cells was well-localized to the affected molar without spreading to adjacent teeth, and the contralateral TG did not show significant change in TRPV1 expression. Although mechanisms underlying the up-regulation of TRPV1 in non-dental afferents following LPS-induced pulpitis are unclear, multiple players, including glial cells, gap junction proteins, or chemical mediators such as neurotrophic factors, might be involved (Thalakoti et al., 2007; Tarsa et al., 2010).

In conclusion, we demonstrated that dentinal application of LPS up-regulated the transcript and protein of TRPV1 in TG, which could contribute to tooth pain in acute pulpitis conditions. Further studies about the effects of LPS on other nociception-related molecules in trigeminal nociceptors and molecular mechanisms underlying such regulation may be beneficial for better understanding of the pathophysiology of tooth pain due to bacterial infection.