Animals
All animal experiments were conducted in accordance to ARRIVE guidelines and approved and curated protocols by the Institutional Animal Care and Use Committee (IACUC) of the University of Texas at Dallas. We used males and females to determine if they possess different mechanisms that drive the development of diet sensitization. Mice were housed (4–5 per cage) in a temperature-controlled facility kept around 21 °C and 50% humidity and maintained on a 12-h light/dark cycle (lights on from 6:00am to 6:00 pm). Animals had ad-libitum access to their respective diet and water and were 6-weeks old at the beginning of the experiment (males: 19.734 g ± 0.444; females: 15.634 g ± 0.254). Pirt-GCaMP3 mice were a generous gift from Xinzhong Dong (John Hopkins University) while WT C57BL/6J mice (stock no. 000664) were purchased from Jackson Laboratory49. Pirt-GCaMP3 mice have a genetically encoded calcium indicator, GCaMP3, expressed under the control of the Pirt promoter, expressed in 95% of sensory neurons49. All Pirt-GCaMP3 mice used were heterozygotes for Pirt-GCaMP3. They were bred in-house for behavior, IHC, and calcium imaging experiments.
Diet
Animals were randomly assigned either control (chow) (laboratory chow diet with 5.3% kcal from fat, 54.7% kcal from carbohydrates, and 18.0% kcal from proteins (LabDiet ProLab RMH 1800)) or high-fat diet (HFD) (adjusted calories diet with 60.3% kcal from fat (92.5% saturated, 2% monounsaturated, 5.5% polyunsaturated), 21.4% kcal from carbohydrates, and 18.3% kcal from proteins (TD.08500, Teklad Custom Diet, Envigo)) with consideration to ensure the weights between the diet groups were no different prior to diet initiation. Littermates were randomly placed on their respective diet at age 6 weeks for 9 weeks. Animals were weighed weekly (Monday) between 8:00am and 1:00 pm for 8 weeks on their respective diets, and diets were replenished weekly after weighing the animals in a controlled environment (HFD males: n = 14; Chow males: n = 14; HFD females: n = 15; Chow females: n = 14).
Fasting glucose
Fasting glucose levels were performed by 14 h overnight removal of food (starting at 6 pm). Since mice are coprophagic, they were placed in a fresh cage with fresh bedding, a new empty food hopper, and used nest (to control for stress levels) and glucose testing was performed the next morning at 8:00am in a quiet environment. The mice were allowed five minutes to become calm after the lid of their cages were removed. Once mice were calm, the tail was clipped ~ 2 mm, and blood was quickly taken from the tail tip and assessed using an Alpha Trak 2 Veterinary Blood Glucose Monitoring Meter Kit week 0, 4, and 8 (HFD males: n = 7; Chow males: n = 7; HFD females: n = 8; Chow females: n = 8). Mice with fasting glucose levels greater than 250 mg/dL were considered hyperglycemic which no animals used for this experiment reached38.
Serum collection and analysis
Tail blood was collected week 8 or 9 (the day before tissue collection) on diet with 300µL EDTA-lined capillary tubes (Sarstedt, Cat#101093-992) after a ~ 2 mm tail clip was taken which took ~2 min for each mouse. After each capillary tube was filled, they were placed on ice until centrifugation at 15,000 rpm at 4 °C for 15 min to isolate serum. 20µL serum aliquots were stored in 0.6 mL tubes and immediately stored at − 80 °C until analysis. Serum/Fatty Acid Detection Kit (Zenbio, Cat#sfa-1) was used to detect serum NEFA levels (Week 8: HFD males: n = 5; Chow males: n = 3; HFD females: n = 6; Chow females: n = 5) (Week 9: HFD males: n = 5; Chow males: n = 5; HFD females: n = 5; Chow females: n = 5). Serum aliquots were randomized such that the experimenters were blinded to sex and dietary conditions during serum analysis.
Drugs
PGE2 (Cayman Chemical, Cat#14010) was reconstituted in 100% ethanol at a concentration of 1 mg/mL and stored at − 20 °C. On the day of injection, a concentration of 0.005 mg/mL of PGE2 was made in sterile 1× phosphate-buffered saline (PBS, pH 7.4) right before injections took place. A stock of 10 mM capsaicin (Sigma-Aldrich, Cat#M2028) was made in 100% ethanol and stored at 4 °C. On the day of calcium imaging capsaicin was diluted to 0.00025 mM in bath solution, described below (see Calcium Imaging).
Injections
Intraplantar injections were administered to the left hind paw (ipsilateral). A Hamilton syringe (Hamilton, Cat#80501) was used with a disposable 30-gauge ½ inch needle (BD, Cat#305106). Mice were covered with a red towel (to minimize stress) with the left paw lifted, and PGE2 was injected at 100 ng/20 µL into the paw pad of the second digit for all animals. After injections, mice were placed in the behavior rack until the time points were collected.
Behavior
Mice were placed in an elevated rack containing separate compartments with a wire mesh bottom for habituation and behavioral testing. Each compartment was 11.43 cm in length and 5.08 cm in width and was separated by a transparent plexiglass divider. Mice are first acclimated to the behavior room in their own home cage for 30 min, then they were habituated to the behavior rack for 1–2 h for three days before to PGE2 administration and habituated for 1–2 h along with measuring baseline sensitivity for 1 and 2 days before PGE2 administration). On the day of PGE2 administration, the mice were left in the rack to habituate for 1–2 h to become calm. Mechanical sensitivity was assessed using flexible von Frey filaments (Stoelting, Cat#58011) to evaluate the ipsilateral hind paw based on the up-down experimental paradigm13,51,52. 2 baseline measurements were taken 1 to 2 days before the day of injections, and again, before PGE2 administration or the day of PGE2 administration. The average was taken between the 2 baseline measurements for each animal. One animal displayed heightened baseline/naïve pain sensitivity (i.e., < 0.4 g) and was excluded from subsequent analyses. Paw withdrawal threshold was assessed 3 h, 24 h, and 48 h post PGE2 injection during week 8 on diet and were done between the hours of 10:00 am and 5:00 pm in a controlled, quiet environment (HFD males: n = 10; Chow males: n = 11; HFD females: n = 11; Chow females: n = 10). Mice were randomly placed into each compartment such that the experimenter was blinded to dietary conditions during behavioral experiments.
Tissue collection
Mice were anesthetized with isoflurane and then euthanized by decapitation. Lumbar DRGs (L3–L5) were collected on week 9 on diet. If tissues were used for IHC, animals were intracardially perfused with 1x PBS followed by 4% PFA made in 1x PBS. DRGs were then post-fixed in 4% PFA for 4 h and cryoprotected using 30% sucrose (Sigma-Aldrich, Cat#S0389) diluted in 1× PBS in at 4 °C and replaced with 30% sucrose every 24 h until the tissue sank. Once the tissue had sunk, it was embedded in optimal cutting temperature (OCT) (Thermo Fisher Scientific, Cat#50-363-773) compound and stored at either − 20 or − 80 °C until sectioning. DRGs used for calcium imaging were collected fresh. and cultured based on the protocol below.
Tissue isolation
Cultured primary DRG neurons from thoracic and lumbar (T12–T13, L1–L6) were used for calcium imaging experiments. DRGs were dissected and placed in chilled HBSS solution. The neurons were then digested in collagenase A (1:1; A (Sigma-Aldrich, Cat#10103586001):HBSS (Gibco, Cat#14-170-112)) for 20 min at 37 °C, collagenase D (1:1:10%; D (Sigma-Aldrich, Cat#1188866001):HBSS:papain (Sigma-Aldrich, Cat#10108014001) for 20 min at 37 °C, and then placed in a Trypsin Inhibitor solution (1:1:1; Trypsin (Sigma-Aldrich, Cat#10109886001):BSA: Media) for trituration. Media is DME/F-12 1:1 (1) with 2.50 mM l-Glutamine and 15 mM HEPES buffer (HyClone, Cat#SH30023) supplemented with 10% Fetal Bovine Serum (HyClone, Cat#SH30088.03) and 1% Penicillin Streptomycin (Fisher Scientific, Cat#15070063). After trituration, cells were filtered through a 70 μm cell strainer (Corning, Cat#CLS431751), pelleted, buffer removed, and then resuspended in 200 μL of Media. Plates used were 35 mm Petri dish, 10 mm Microwell, and No. 1.5 cover glass (MatTek Corporation, Cat#P35GC-1.5-C) that were coated in facility with a 2 µg/mL poly-D lysine solution (Sigma-Aldrich, Cat#P0899). Cells were plated in a 200 μL bubble on the center of the plate to rest in an incubator set at 37 °C with 5% CO2 for 2 h before filling the rest of the well with Media. Cells were used the next day for calcium imaging.
Calcium imaging
DRGs were cultured as described below and imaged the day after plating. Cells were imaged using either GCaMP3 or fura-2 AM (Invitrogen, Cat#F1221). If fura-2 AM was used the dish was loaded with 0.005 mg/mL in loading buffer made in HBSS (Gibco, Cat#14-170-112) containing 0.25% bovine serum albumin (endotoxin free) (Sigma-Aldrich, Cat#A9576) and 2 mM CaCl2 for 1 h at 37 °C. Then fura-2 AM was esterified in bath solution (125 mM NaCl (Fisher Scientific, Cat#S271-500), 5 mM KCl (Fisher Scientific, Cat#P217-500), 10 mM HEPES (Sigma-Aldrich, Cat#H4034), 1000 mM CaCl2 (Sigma-Aldrich, Cat#21115, 1,000 mM MgCl2 (Fisher Scientific, Cat#M35-500), and 2,000 mM glucose (Sigma-Aldrich, Cat#G7528)) at a pH of 7.4 ± 0.05 and mOsm of 300 ± 5 for 30 min at 37 °C. If GCaMP3 was used for imaging, cells were placed in bath and imaged 30 min after being placed in bath. A previous study revealed comparable activity levels between Pirt-GCaMP3 DRG neurons and fura-2 AM stained DRG neurons, so subsequent analyses graphed both data sets together49. Capsaicin at 0.00025 mM was applied on cells for 20 s followed by a 120 s wash and then a 10 s application of 50 mM KCl as a positive control. Only neurons were used in analysis which were defined as any cell with a 20% response change from baseline change to 50 mM KCl cells not meeting this requirement were excluded from further analysis. At least 16% change from baseline was counted as a response to 0.00025 mM capsaicin. All experiments were run using the MetaFluor Fluorescence Ratio Imaging Software on an Olympus TH4-100 apparatus. Imaging was done on a 40 × oil objective on the FITC channel. Ratios of bound (excitation wavelength: 340 nm) to unbound (excitation wavelength: 380 nm) fura-2 AM (emission wavelength: 510 nm) were recorded real time and GCaMP3 intensity changes were recorded real time. All imaging data was exported in a Microsoft Excel spreadsheet and analyzed using the algorithms in a Microsoft Excel. Latency to peak was measured by the amount of time it took to reach the peak response to capsaicin and the magnitude response was calculated by the percentage increase from baseline response (prior to capsaicin application) to peak response to capsaicin. Individual cells were excluded if the baseline value was underneath a value of 0.45 or exceeded a value of 1.75 for F340nm/F380nm ratio.
Immunohistochemistry (IHC)
Lumbar (L3-L5) DRGs were cryosectioned at 18 microns and mounted onto a charged slide. Sections were dried on a slide warmer before being stored at − 20 °C. On the day of IHC, slides were dried on a slide warmer for ten minutes and a hydrophobic border was drawn around the sections using a PAP pen (Fisher Scientific, Cat#23-769-533) to ensure the reagents stay on the slide. The slides were subsequently fixed with 4% paraformaldehyde (PFA) (Fisher Scientific, Cat# F79-500) diluted in 1x PBS in the fume hood for five minutes. After three changes of wash buffer (1x PBS containing 0.05% Tween-20 (Sigma-Aldrich, Cat#P1379)), antigen retrieval was performed using three changes of heated 10 mM citrate buffer (2.94% trisodium citrate Sigma-Aldrich, Cat# C8532), 0.05% Tween-20, diluted in ddH2O) (pH = 6.0 ± 0.05) for five minutes each. A combination permeabilization/blocking buffer (2% heat-inactivated normal goat serum (Gibco, Cat#16210-072) for IB4+ and ATF-3 antibodies or 5% heat-inactivated normal goat serum for CD68, NeuN, and GFAP antibodies , 1% bovine serum albumin (BSA; VWR, Cat#97061-416), 0.1% Triton X-100 (Sigma-Aldrich, Cat#X100), 0.05% Tween-20, and 0.05% sodium azide (Sigma, Cat#RTC000068) diluted in 1x PBS) was applied for 2 h followed by the appropriate primary antibody cocktail application overnight at 4 °C (Table 1). When applying isolectin GS-IB4, biotin-xx conjugates, CaCl2 was added at the same concentration to activate IB453,54. After three changes of wash buffer, the slides were incubated in the appropriate secondary antibodies cocktail for 2 h at room temperature (Table 2). Slides were washed before and after incubating the tissues in 4′,6-diamidino-2-phenylindole (DAPI) for five minutes. Coverslips were mounted onto slides using Gelvatol and cured overnight at room temperature. Sealant was applied around the border of the coverslip to prevent tissue dehydration, and slides were stored at 4 °C until imaging. See Table 1 and 2 for more information on the antibodies used for this study.
Image acquisition and analysis
Each animal was randomly assigned a letter (A-Z) when sectioning, and images were acquired using a Zeiss Axiobserver 7 Epifluorescent Microscope for DRGs by a blinded experimenter. 2 experimenters analyzed these images and were blinded to sex and dietary conditions throughout the entirety of imaging analysis. ATF3 quantification was presented as the number of ATF3+ cells in the DRG that was calculated through a manual count of the number of ATF3+ cells divided by the area of the DRG. The area of the DRG was measured by drawing a region of interest around the DRG, excluding the axons of the DRG as much as possible through ImageJ Version 2.0.0. In addition, ATF3 quantification was further presented as the percentage of IB4 expressing ATF3 that was calculated as a percentage of the manual count of the number of IB4+ cells expressing ATF3 in the IB4+ cell population in the DRG. (HFD males: 6; Chow males: 4; HFD females: 6; Chow females: 3). CD68 quantification was presented as the number of CD68 + cells in the DRG divided by the total area of DRG. The number of CD68+ cells was quantified through a manual threshold adjustment with ImageJ Version 2.0.0. for each DRG. (HFD males: 6; Chow males: 4; HFD females: 5; Chow females: 3). GFAP quantification was presented as the percentage of neurons surrounded by activated satellite glial cells, GFA+ cells, divided by the total number of neurons, NeuN + cells. GFAP + cells were counted and included in the analysis if it surrounded the neurons. (HFD males: 6; Chow males: 4; HFD females: 5; Chow females: 3). Up to 4 images from each animal were acquired, analyzed, and averaged. There was at least a biological replicate of three animals for each group. All representative images were taken on an Olympus FluoView 3000 RS laser scanning confocal microscope.
Statistical analysis
All data are represented as mean ± SEM. All graphs and analyses were performed using GraphPad Prism 9.3.1 (GraphPad, San Diego, CA, USA). Weights were analyzed using a two-way ANOVA with Sidak’s post hoc test to observe the influence of diet and sex on the data and compare the differences between each group (HFD males: n = 14; Chow males: n = 14; HFD females: n = 15; Chow females: n = 14). Assessment of fasting glucose levels at each timepoint was analyzed using two-way mixed analysis with Sidak’s post hoc test since fasting glucose was not recorded at every time point for each animal. Behavioral data was analyzed using a two-way ANOVA with Sidak’s post hoc test to observe the influence of time, diet, and sex on the data and the area over the curve (AOC) were analyzed using a two-way ANOVA with Sidak’s post hoc test to analyze the effect of diet and sex. Calcium imaging analysis were analyzed using two-way ANOVA with Sidak’s post hoc test to analyze effect of diet and sex on the trends and form quantitative comparisons between groups. Fisher’s exact test analysis was used to compare the proportions of cells responding to capsaicin between HFD and chow within sex (HFD males: n = 6 mice, 195 cells; Chow males: n = 9 mice, 151 cells; HFD females: n = 8 mice, 239 cells; Chow females: n = 7 mice, 292 cells). For the IHC experiments, each group had at least three animals with up to 4 replicates obtained and averaged for each animal. IHC analysis was then analyzed using three-way ANOVA with Sidak’s post hoc multiple comparison to observe the interactions between diet, sex, and PGE2 injection. NEFA fold induction analysis was analyzed using a three-way ANOVA with Sidak’s post hoc multiple comparison to observe the influence of time, diet, and sex on the data and the interactions between each factor. Statistical values for each analysis are also provided in Supplementary Information. All experiments were performed with experimenters blinded to condition. p values *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.
