Male, 6-10 weeks old C57BL/6 mice obtained from the Laboratory Animal Service of the Autonomous University of Aguascalientes were used in this study. Animal protocols followed the guidelines from the Mexican norm (NOM-062-Z00-1999) and were approved by the Ethics Committee Concerning Animals in Teaching and Research at the Autonomous University of Aguascalientes (CEADI-UAA 5/03/2018). Mice were randomly divided into two groups of five and maintained at an ambient temperature of 23-25°C, with 12 light/12 dark cycles, and fed ad libitum (Nutricubo, Purina USA).

Prior to colitis induction, mice were fasted overnight (16-18 hours). Animals were anesthetized by isoflurane inhalation, and colitis was induced via intrarectal administration of DNBS (100 μl, 2-6 mg; dissolved in 50% ethanol (EtOH; v/v)), deposited with a polyethylene catheter, 3 cm into the colon, as previously described (13). The control group received only 50% EtOH (v/v) intrarectal administration. Four days post-DNBS application, the mice were euthanized with a sodium pentobarbital overdose (200 mg/kg).

In order to discern the inflammation severity caused by DNBS, the following parameters were recorded daily: body weight, stool consistency, and fecal blood. The difference between the daily recorded weights and the initial body weight (prior to DNBS administration, taken as the 100% weight) was determined to be the % weight loss induced by colitis. As previously described, stool consistency was determined (see Table 1 ), and fecal blood was detected using Hema screen test strips (14).

Upon completion of the study, mice were euthanized, the distal colon was exteriorized, dissected free, rinsed with cold PBS, and photographed for posterior serosal-surface damage analyses according to the validated criteria described by Morris and collaborators (15), where 0 = no bleeding, no ulcers or edema; 1 = mild bleeding, mild edema, or mild mucosal erosion; 2 = moderate edema, ulcers or bleeding erosions; 3 = severe ulceration, erosions, edema, tissue necrosis, and perforation. The damage analyses (scoring) were conducted by two independent individuals blinded to the treatment of the mice. The dissected colon was divided into three 1 cm segments and used as follows.

The first piece of colon was fixed in 10% formalin for 24 hours (Sigma-Aldrich, Inc. St. Louis, USA), dehydrated in graded alcohols, cleared in xylene, and embedded in paraffin wax. Colon cross-sections (7 μm thick) stained with hematoxylin and eosin (H & E) were visualized under a light microscope (Axioscope 40, Carl Zeiss, Germany) and photographed for further analysis.

A second colon section was employed for the myeloperoxidase (MPO) assay, a well-established assay for determining neutrophil presence and activity (16). This assay is commonly utilized to determine the level of inflammation in colon tissue in different colitis models (17, 18). To perform this assay, a 1 cm distal colon section was weighed and homogenized (D1000 homogenizer, Benchmark Scientific MA, USA) in 0.5% hexadecyltrimethylammonium bromide (HTAB) buffer made in potassium phosphate (50 mM, pH 6.0). Homogenates were centrifuged at 13,400 g, 10 min at 4°C, and supernatants were aliquoted and stored. A total volume of 100 μl of each supernatant was added to 2.9 mL of O-dianisidine solution (16.7 mg/mL of O-dianisidine dihydrochloride and 0.5% hydrogen peroxide in PBS). Absorbances at wavelength of 450 nm were measured using a microplate reader (iMark, Biorad, Hercules CA, USA) during a time-lapse of 60 s. Data were expressed as units per gram of colon (U/mg), where one enzymatic unit is defined as the amount of myeloperoxidase that can hydrolyze 1 μmol hydrogen peroxide in 60 s at room temperature.

Finally, from the third colon segment, RNA was isolated, and expression levels were determined, by semi-quantitative RT-PCR, for the following transcripts: tumor necrosis factor-alpha (TNF-α), interleukin-1beta (IL-1β), and interleukin-6 (IL-6).

Using TRIzol-Chloroform (Invitrogen, Carlsbad, CA, USA), RNA was extracted from dissected 1 cm colon sections. Total RNA obtained from all colon samples was quantified with a Nanodrop 2000 spectrophotometer (Thermo Scientific, U.S.A), and RNA quality was calculated based on the absorbance ratios at 260/280 and 260/230 nm; the accepted ratio range was 1.9-2.0 (19). The NanoDrop software automatically calculated the nucleic acid concentration, of which 1 μg total RNA was reversed transcribed with the iScript kit (Biorad, Hercules CA, USA) according to the manufacturer’s instructions. From this, we amplified 1 μl of cDNA/25 μl reaction volume (containing: 1X PCR buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 0.2 mM of forward and reverse primer, and 0.5 U DNA Taq Polymerase (Life Technologies, TX, USA)). Transcript glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the internal control, and the expression of other transcripts was normalized to it. Each transcript amplification was conducted in triplicate, and the results shown are the average of these three amplifications. Amplification conditions were as follows: 3 min denaturation at 94°C, 35 amplification cycles (denaturation: 30 s at 95°C; annealing: 30 s at 55°C; and extension: 30 s at 72°C). Following the completion of the final cycle, the samples were incubated for an additional 5 min period at 72°C for terminal elongation. All primers were purchased from Integrated DNA Technologies, their sequences are listed in Table 2 .

A total of 5 μl of each PCR product were separated on a 1.5% agarose gel, stained with 0.1% ethidium bromide, photo-documented (miniBis Pro, DNR Bio-Imaging System), and the densitometry of the visible band determined using Image J 1.43 (NIH) software, following the user´s manual (20). As aforementioned, all products were normalized with the housekeeping control, GAPDH.

For DRG isolation, control and DNBS mice were anesthetized with isofluorane inhalation. Subsequently, intracardiac perfusion with Hank’s balanced salt solution (HBSS) allowed the clearing of systemic blood. Following this, laminectomy was performed to dissect the DRGs from levels thoracic 8 to lumbar 1 (T8-L1), known to contain colon projecting nociceptive DRG neurons (21, 22). DRGs were collected in cold HBSS for primary culture. The ganglia were dissociated by enzymatic treatment using a mixture of Trypsin (1 mg/mL) and collagenase D (2.8 mg/mL) and incubating for 30 minutes at 37°C. After this incubation period, enzymes were removed from the tissue by washing with HBSS and 1 ml of F12 medium (Sigma Aldrich^®) with 100 U/ml penicillin and 0.1 mg/ml streptomycin supplemented with 10% fetal bovine serum was added. Disaggregated neurons were plated on round coverslips previously covered with poly-D-lysine (20 μg/ml) in 24-well dishes in a humid, 37°C incubator with 95% O2 and 5% CO2 atmosphere for 18-24 h before further manipulation. Cultured DRG neurons from naïve mice were incubated with human TNF-α solution (0.1 μg/mL), TNF-α/ATP (100 µM) or F-12 medium alone for 17-18 hours before electrophysiological studies (23).

The perforate-patch current clamp recordings with 0.24 mg/ml amphotericin B (Sigma-Aldrich) were performed with borosilicate pipettes with resistance between 2-5 MΩ on small neurons (≤ 40 pF capacitance), which are known to exhibit nociceptive properties (24). DRG neurons were superfused with an external solution containing (mM): 140 NaCl, 5 KCl, 1 MgCl₂, 2 CaCl₂, 10 HEPES, and 10 D-glucose, pH adjusted to 7.4 with NaOH. The composition of the internal pipette solution was (in mM): 110 K-gluconate, 30 mM KCl, 10 HEPES, 1 MgCl₂, and 2 CaCl₂ with pH adjusted to 7.25 using KOH.

After euthanizing the animals, the DRGs (T8-L1) from control and DNBS mice were dissected and immediately embedded in O.C.T. on dry ice. Cryosections (20 μm) were adhered on SuperFrost Plus slides (Thermo Fisher Scientific), fixed for 1 hour in ice-cold 10% formalin in 1 X PBS, washed 3x 10 minutes in PBS, and permeabilized for 30 minutes in 0.2% Triton X-100 (Sigma-Aldrich). Following this, excess Triton was washed off 3x 5 minutes with PBS, and tissues were blocked for 2 hours in 10% NGS (normal goat serum, ThermoFisher). Incubation with Alexa Fluor 568-labeled isolectin IB4 (cat # I21412; Thermo Fisher) and P2X3 antibody (1:500; cat # ab10269; Abcam) was performed overnight at a temperature of 4°C. Finally, a 1-hour incubation with Alexa Fluor 488- conjugated secondary antibody was performed at room temperature (approximately 21°C). A P2X3-positive nociceptive neuron subgroup found in the DRG has been previously identified to also bind this isolectin; thus, both markers were hereby employed with the purpose of identifying these cells (24–26). Fluorescent emission was visualized and images were captured using an Olympus FluoView 1200 confocal microscope; colocalization of P2X3 and IB4 analysis was performed as described previously (27). Colon samples of 3 separate animals per treatment group (control and DNBS) were processed as described, and representative images (projections of z stacks) were presented in the results. Using FIJI (Image J), the corrected total cell fluorescence (CTCF) was calculated to determine the intensity of the signal between experimental groups. To do so, the integrated density and the area, as well as the background noise, were measured, and the CTCF was calculated as equal to the integrated density - (area of selected cell x mean fluorescence of background readings). CTCF values from the DNBS-treated group were normalized to the control group and expressed as normalized CTCF.

Western blotting was used to examine the expression of P2X3 receptor and GAPDH expression in DRG (T8-L1) neuron protein extract. A total of six mice per group (control vs DNBS) were used for DRG harvesting. Total ganglia (T8-L1) from each mouse were pooled together, snapped frozen on dry ice, and later homogenized in 300 μl of lysis buffer (50 mM Tris, pH 7.4; 150 mM NaCl; 1 mM EDTA, pH 8.0, and 1% Triton X-100) supplemented with protease (P8340 Sigma-Aldrich^®) and phosphatase (P5726 and D0044 Sigma-Aldrich^®) inhibitors. Homogenates were centrifuged at 14,000 rpm, 15 min at 4°C; the supernatant was isolated and stored at -80°C. Protein quantification was conducted with the Micro BCA protein assays kit (Thermo Scientific, Rockford, IL). A total of 15 μg of protein was separated in a 10% SDS-PAGE and transferred onto PVDF membranes (Biorad, Hercules CA, USA) overnight at 25 V and at 4°C. Non-specific antibody binding was blocked by incubating the membrane for at least 24 hours with 5% nonfat dry milk in 1X Tris Buffer solution containing Tween 20 (TTBS). Excess milk was washed off with 1X TTBS, and membranes were then incubated with anti-P2X3 (1:1000; Abcam #ab10269) and GAPDH (1:10,000; Cell Signaling) overnight at 4°C. The following day, the primary antibody was removed, excess washed off with 1X TTBS, and membranes were then incubated with the corresponding secondary antibody at room temperature for 1 hour (1:10,000; Jackson Immunoresearch). After incubation, membranes were washed with 1X TTBS, and immunoreactivity was detected using Clarity™ Western ECL substrate (BioRad) and documented with MicroChemi 4.2 (Bio-imagine Systems). Densitometric analysis was performed using LI-COR Image Studio Software.

DRG cells were incubated with P2X3 receptor antagonist, A-317491, 30 min before the application of ATP or TNF-α, unless otherwise mentioned.

F12 medium, Fetal Bovine Serum (FBS), L-glutamine, penicillin, streptomycin, ATP, A-317491, human TNF-α, and all salts were purchased from Sigma-Aldrich (Toluca, MX). ATP stock solution (100 mM) was made using deionized distiller water and stored frozen; these were diluted to obtain the desired final concentration in external solution prior to use, and the pH was adjusted to 7.4 with NaOH.

Data values are presented as the mean ± standard error of the mean (S.E.M.). The number of performed experiments is designated as “n.” Cells from at least five different mice were used for each experimental protocol. To test statistical differences between two data sets, we used the unpaired Student’s t-test. One-way ANOVA with Bonferroni’s post hoc test was used for multiple comparisons. Data were considered statistically different when P ≤ 0.05 in GraphPad Prism version 8 (GraphPad Software, La Jolla, CA, USA).

The presence of colitis is generally confirmed using body weight loss, diarrhea, rectal bleeding, and macroscopic damage scores. As expected, colitis induction produced significant body weight loss in the DNBS group ( Figure 1A ), reaching 14% after four days. Furthermore, diarrhea and rectal bleeding scores were significantly higher (P ≤ 0.001) for the DNBS-treated group than for the control group ( Figures 1B , C ). Four days after DNBS application, the mice were euthanized, and their colons were dissected, scored, and processed. Visual inspections of the colon revealed erosions, adhesions, colon wall thickness, and petechial hemorrhage in all animals treated with DNBS ( Figure 1D ), resulting in a macroscopic damage score 2.9 times greater in DNBS-treated mice compared to controls (P = 0.001; Figure 1E ). In addition, histological analysis of the distal colon revealed occasional epithelial erosions and extensive immune cell infiltration in DNBS colons, neither of which were present in the controls ( Figure 1G ). Neutrophil infiltration, assessed indirectly via MPO tissue activity content, is commonly accepted as a positive indicator of acute inflammation (17, 18). DNBS administration resulted in colon MPO activity ~17-fold greater, compared to the control group (7.0 ± 2.3 U/mg vs 0.4 ± 0.1 U/mg; P < 0.05; n = 5; Figure 1F ). Taken together, these results indicate that DNBS- treated mice exhibited significant signals of acute colitis.

The colonic inflammatory process induced by DNBS in mice was further reflected by a statistically significant, 1.5-fold increase in TNF-α mRNA expression compared to the control group (P < 0.05; Figure 2A ). However, even when there was an increase in IL-1β and IL-6 mRNA expression in colons of DNBS-treated mice compared to controls, this difference was not statistically significant ( Figures 2B, C ). These results indicate that colonic inflammation in this acute model is influenced primarily by TNF-α overexpression.

To study the effect of acute colitis on neuronal excitability, we cultured T8-L1 DRGs from DNBS-treated and control mice. The current clamp recording trace shown in Figure 3A is representative of DRG membrane potential at rheobase (minimum current stimulus required to elicit a single action potential) and twice rheobase current injection. The rheobase was significantly reduced (41%) by DNBS treatment (DNBS: 21.7 ± 6.0 pA vs Control: 52.8 ± 10.8 pA; P = 0.035; Figure 3B ). However, mean action potential discharge at twice rheobase was similar between the DNBS and control groups ( Figure 3C ). Neuronal resting membrane potential was alike in both treatment groups (control: -45.6 ± 1.29 mV vs DNBS: -46.7 ± 1.22 mV), nor input resistance (control: 880.69 ± 87.18 MΩ vs DNBS: 926.37 + 103.7 MΩ). The mean cell diameter for both control and DNBS neurons was similar (control: 19.04 ± 1.13 pF vs DNBS: 19.5 ± 0.95 pF). These results indicate that colonic inflammation does not affect the passive membrane properties of sensory neurons in the DNBS-treated group, but it does affect neuronal excitability properties, making them more hyperexcitable.

In order to determine whether TNF-α is capable of producing similar changes in excitability in DRG neurons from DNBS-treated mice, patch-clamp recordings were executed from dissociated T8-L1 DRG neurons incubated overnight with TNF-α (0.1 μg/ml). Like DRG neurons from DNBS-treated mice, the rheobase was significantly decreased in TNF-α-incubated neurons (representative recordings shown in Figure 4A ), which was reduced by 54.3% (TNF-α: 27.1 ± 5.7 pA, n = 7; vs control: 50 ± 7.6 pA, n= 7; P = 0.03; unpaired student t-test; Figure 4B ). The number of action potentials elicited at twice rheobase was almost the same between neurons incubated with TNF-α and neurons incubated with a culture medium. To test whether P2X3 receptor activation precedes the potentiation of DRG excitability by TNF-α, neurons were co-incubated with ATP (100 μM), TNF-α (0.1 μg/ml) and the P2X3 receptor antagonist A-317491 (1 μM). The antagonist was applied 30 min prior to ATP and TNF-α. The effects of TNF-α on the rheobase of neurons were inhibited by pre-incubation with A-317491 ( Figure 5A ). The mean rheobase of control neurons (ATP alone) was 61.3 ± 9.8 pA, while TNF-α-incubation decreased this to 28.8 ± 5.1 pA, a response that was blocked by pre-incubation with A-317491 followed by TNF-α stimulation, resulting in rheobase of 73.3 ± 18.5 pA ( Figure 5B ). Between these groups, there were no differences in the number of action potentials elicited with a stimulus twice rheobase ( Figure 5C ). Taken together, these results indicate that TNF-α enhances DRG neuron excitability, potentially through modulation of P2X3 receptors.

The electrophysiological data described above suggest that an inflamed colon may cause direct sensitization of primary afferent neurons through direct modulation of P2X3 receptors. This direct modulation could be a result of DRG neuron P2X3 receptor overexpression. To test this hypothesis, we carried out immunofluorescence and western blot assays. It is well known that the majority of P2X3+ DRG neurons in the mouse are non-peptidergic neurons that bind to isolectin B4 (IB4) (28–30). Representative immunofluorescence images are shown in Figures 6A, B . Immunofluorescent analysis of P2X3 receptors in T8-L1 DRGs revealed that compared to controls, the P2X3-IR measured as Corrected Total Cell Fluorescence (CTCF) was markedly increased in DNBS-treated mice compared to control mice (P < 0.05; Figures 6A – C ). Western blot analysis confirmed this finding, as P2X3 protein levels in T8-L1 DRGs significantly increased in mice with DNBS-induced colitis (P <0.05; Figure 6D ). These results reveal an upregulation of P2X3 receptor in DRG neurons that innervate the colon of DNBS-treated mice.