Altered sensory neuron activity in a mouse model of post-burn pain and itch

Abstract

Burn injury induces pain and is frequently accompanied by persistent itch during wound healing. However, the underlying neural mechanisms remain poorly understood. Here, we developed a mouse model of post-burn pain and itch and examined changes in primary sensory neuron activity using in vivocalcium imaging of trigeminal ganglion (TG) neurons. To induce scald burn injury, anesthetized mice were exposed to boiling water on the cheek skin. Following injury, mice exhibited both spontaneous pain-related behaviors (wiping) and itch-related behaviors (scratching). Pain-related behaviors peaked on day 1 and returned to baseline within 5 days, whereas itch-related behaviors peaked on day 7 and persisted for up to 28 days. In vivo calcium imaging revealed a significant increase in the number of TG neurons exhibiting spontaneous activity on days 1 and 7 post-injury compared to baseline. While the proportion of capsaicin-sensitive neurons remained unchanged after scald burn, the proportion of chloroquine-sensitive neurons was reduced on day 1 and partially recovered on day 7. These findings suggest that enhanced spontaneous activity in primary sensory neurons may contribute to post-burn pain and itch. This model appears to be useful to investigate the neural mechanisms underlying sensory dysfunction following burn injury.

Introduction

Burn injury induces pain and is frequently accompanied by persistent itch during the wound-healing process. Chronic post-burn pain and itch represent major unmet challenges in burn care, affecting a large proportion of patients and substantially impairing quality of life¹. The prevalence of itch among burn patients has been reported to range from 76–93% at hospital discharge², while chronic post-burn pain affects approximately 25–36% of patients^3–5. Both conditions contribute to significant physical discomfort, sleep disturbance, and psychological distress^1,6,7. Moreover, ongoing pain and itch can exacerbate stress responses that negatively impact wound healing, leading to prolonged recovery and increased healthcare costs^8–12.

Despite their high prevalence and clinical significance, effective treatments for post-burn pain and itch remain limited. Current therapeutic options often provide incomplete relief and are frequently associated with adverse effects or safety concerns^7,13. These limitations underscore the need for a deeper mechanistic understanding of the neural processes that drive sensory dysfunction following burn injury.

Primary sensory neurons play a central role in the detection and transmission of noxious and pruritic stimuli. Alterations in the excitability and functional properties of these neurons have been implicated in various chronic pain and itch conditions^14,15. However, how burn injury affects sensory neuron activity over time, and whether distinct neuronal populations contribute to pain versus itch during the healing process, remain poorly defined. In particular, the relationship between spontaneous sensory neuron activity and the emergence of persistent post-burn pain and itch has not been directly examined in vivo.

In this study, we developed a mouse model of post-burn pain and itch. Using in vivo calcium imaging of trigeminal ganglion (TG) neurons, we investigated changes in spontaneous neuronal activity and stimulus-evoked responses following scald burn injury. This approach allowed us to directly assess how burn injury alters the functional properties of primary sensory neurons during distinct phases of pain and itch. Our findings provide insight into the neuronal mechanisms underlying post-burn sensory dysfunction and establish an experimental tool for evaluating targeted therapeutic strategies.

Methods

Mice:

All mouse procedures were approved by Institutional Animal Care and Use Committees of the University of Miami (protocol number: 21–111) and were performed in accordance with the guidelines for the Care and Use of Laboratory Animals¹⁶. C57BL/6J mice were obtained from the Jackson Laboratory (Bar Harbor, ME) and maintained as mating colonies within our facility. Mice were group housed (2–5 per cage) in a 12–12 light-dark cycle and had access to food and water ad libitum. Adult male mice were randomly assigned to experimental groups.

Scald burn induction:

For all mice, a small area of fur on the cheek was shaved to prepare for scald burn induction. After two or three days, the mice were anesthetized, and their shaved cheek skin was pressed against a heat-resistant sheet with a 4 mm-diameter window. This allowed a precise patch of skin to be exposed to boiling water for eight seconds. In contrast, control mice had room temperature water applied for the same duration. To transiently mitigate pain associated with burn injury, mice were treated with the short-acting analgesic ketoprofen (5 mg/kg, subcutaneous) 30 minutes before injury and on post-injury days 1 and 2. Ketoprofen treatment did not significantly alter spontaneous wiping behavior on days 1 or 3 compared with vehicle-treated controls (day 1: 45.0 ± 3.3 vs. 43.9 ± 2.7; day 3: 24.4 ± 1.3 vs. 25.1 ± 1.7; n = 8–11 per group). On postburn days 1, 3, 5, 7, 10, 14, and 21, the skin was digitally photographed, and wound areas were measured using Fiji image analysis software¹⁷. Changes in wound area were expressed as percentages of initial wound area and was compared postburn days.

Behavior tests:

Behavior tests were conducted according to our previously established protocols with slight modifications^18,19. Before the injury, all mice were habituated to the behavioral recording environment and handling. Behavioral observations took place prior to the burn and then on postburn days 1, 3, 5, 7, 10, 14, 21, and 28. On these days, each mouse was placed in the testing arena for 120 minutes, during which their behavior was recorded. The numbers of scratch bouts and wiping behavior were counted in 5-minute bins by a trained observer blinded to the experimental condition^18,19. One scratch bout was defined as 1 or more rapid back-and-forth hindpaw motions directed toward and contacting the injection site, ending with licking or biting of the toes or placement of the hindpaw on the floor^18,19. A wipe was defined as a singular motion of the ipsilateral, but not bilateral, forelimb, beginning at the caudal extent of the injected cheek and proceeding in a rostral direction^18,19. The inner aspect of the ankle and/or forelimb typically contacted the cheek with the paw closed. After behavioral testing, mice were deeply anesthetized and euthanized by transcardinal perfusion.

Histochemistry:

Under anesthesia, mice were perfused with 20 mL PBS followed by 20 mL 4% paraformaldehyde¹⁶. The skin was dissected, fixed in 4% paraformaldehyde followed by 30% sucrose, frozen in optimal cutting temperature (OCT) compound (Tissue-Tek, Sakura Finetek, Torrance, CA), and cut in 16-μm sections on a cryostat²⁰. The sections were stained with hematoxylin and eosin (H&E) or toluidine blue using standard procedures.

in vivo Ca²⁺ imaging:

Mice were placed in a stereotaxic frame under isoflurane anesthesia. Following craniotomy, An adeno-associated virus (AAV; AAV-syn-GCaMP7f; Addgene; Douglas Kim & GENIE Project; Addgene viral prep #104488-AAV1; RRID:Addgene_104488; 2.2 × 10¹³ vector genomes/ml²¹) was unilaterally injected into the TG [coordinates: anterior-posterior (AP) −0.35 mm, medial-lateral (ML) ±1.73 mm, dorsal-ventral (DV) −6.1 mm]. A total volume of 3 μl was carefully delivered over 4–5 minutes using a glass needle and plunger. Mice underwent in vivo Ca²⁺ imaging two to three weeks later. In some experiments, AAV was administered to mouse pups at P2-P5 as previously described²². Briefly, five microliters of AAV (AAV-CAG-GCaMP6s; Addgene; a gift from Douglas Kim & GENIE Project; Addgene viral prep # 100844-AAV9; RRID:Addgene_100844; 2.6 × 10¹³ vector genomes/ml²³) was injected subcutaneously in the nape of the neck. These mice were then used for in vivo imaging starting at least 10 weeks post-injection.

For in vivo Ca²⁺ imaging in the TG, mice were placed on a custom stage with a heatpad. The head was stabilized in a stereotaxic mask under urethane anesthesia (1.5 g/kg). Following craniotomy, half of the cerebral tissue was aspirated using a customized glass pipette to expose the TG²⁴, and the head angle was adjusted for optimal visualization of the ganglion. Images were acquired with single-photon confocal microscopy (Leica SP5 or Nikon ECLIPSE FN1), collecting Z-stacks of 10 frames (512×512 pixels in the x-y plane) every 7.5–9.0 seconds using a 5X dry lens (N.A. 0.15) or 4X dry lens (N.A. 0.2). Green fluorescence was detected after excitation with a 488 nm solid diode laser and emission collected at 500–550 nm. After recording spontaneous Ca²⁺ activity for 30 minutes, PBS, chloroquine (3 mM), capsaicin (10 μM), and high-K⁺ solution (100 mM) were directly applied to the TG. Mice were then deeply anesthetized and euthanized by cervical dislocation.

Data processing involved suite2P for motion correction and alignment²⁵. Regions of interest (ROI) were defined using Cellpose²⁶. Fluorescence intensities were normalized to baseline values (F/F₀). Spontaneously active neurons were defined as cells exhibiting one or more calcium transients in the absence of stimulation. Cells were considered responsive if their fluorescence increased by more than three standard deviations above the resting level following chemical application. Only cells exhibiting responses to high potassium were included in subsequent analyses.

Statistical analysis:

Data are presented as mean ± SEM. For comparison among more than two groups, a one-way RM ANOVA followed by Dunnett’s multiple comparisons test or a one-way ANOVA followed by Tukey’s multiple comparisons test was used. Statistical significance was set at p < 0.05 for all tests. All statistical analyses and graphs were made using GraphPad Prism10 (GraphPad Software, San Diego, CA).

Results

Induction of a scald burn injury on the cheek skin.

To establish a mouse model of post-burn pain and itch, a scald burn injury was induced on the cheek skin (Fig. 1A). Targeting the cheek enabled dissociation of spontaneous itch-related behaviors, assessed as hindlimb scratching, from pain-related behaviors, assessed as forelimb wiping^18,27. Following scald burn injury, the cheek wounds gradually healed, with wound closure occurring approximately 2–3 weeks post-injury (Fig. 1B).

Figure 1.

Temporal changes in burn size following scald injury. (A) Representative images show shaved mouse skin before (day 0) and after (days 1, 5, and 14) scald burn injury. Scald burn injury was induced by exposing a 4 mm area of the shaved cheek skin to boiling water for 8 sec. (B) The time course of burn area progression was quantified using Fiji image analysis software (n =11). Burn size were shown as percentage of initial wound size.

Histological changes in the cheek skin were examined by H&E staining before injury (day 0; Fig. 2A) and at multiple time points after injury (days 1, 5, and 14; Fig. 2B–D). A marked increase in inflammatory cell infiltration was observed on day 1 post-injury. Re-epithelialization was finished on day 14.

Figure 2.

Hematoxylin and eosin staining of skin sections following scald burn injury. Representative images show H&E-stained samples taken before the injury (day 0 (A)) and after the injury at different time points: days 1 (B), 5 (C), and 14 (D). Labels indicate the epidermis (E), hair follicle (HF), dermis (D), and subcutaneous fat (SF). A red arrow marks the wound edge.

Behavioral changes following scald burn injury.

Scald burn injury induced distinct temporal patterns of pain- and itch-related behaviors. Pain-related behaviors peaked on day 1 and returned to baseline within 5 days, whereas itch-related behaviors peaked on day 7 and persisted for up to 28 days. In contrast, control mice exposed to room-temperature water did not exhibit changes in either pain- or itch-related behaviors.

Increased spontaneous activity in TG neurons following a scald burn injury.

We next tested whether spontaneous pain- and itch-related behaviors are associated with increased spontaneous activity in primary sensory neurons. To address this, we performed in vivo Ca²⁺ imaging to monitor neuronal activity in TG neurons. The proportion of TG neurons exhibiting spontaneous activity was significantly increased on day 1 after scald burn injury, corresponding to the peak of pain-related behaviors (Fig. 4; Supplementary video 1). Although the proportion of spontaneously active neurons on day 7, corresponding to the peak of itch-related behaviors, was lower than that observed on day 1, it remained significantly elevated compared with baseline levels (Fig. 4; Supplementary video 2–3).

Figure 4.

An increase in the number of spontaneously activated TG neurons following scald burn injury. (A) The proportion of spontaneously active neurons was significantly higher on day 1 (blue bar) and day 7 (green bar) compared to baseline (day 0; white bar) (n=4). Error bars are SEM. *p < 0.05, **p < 0.01, ****p< 0.0001; one-way ANOVA followed by Tukey’s multiple comparisons test. F(2, 9)=32.72, p<0.0001. (B) Heatmaps illustrate the activity of individual spontaneous active neurons prior to injury (day 0), as well as on day 1 and day 7 post-injury.

Differential changes in itch- and pain-sensitive neuronal populations after scald burn injury

We next examined whether the relative proportions of pain- and itch-sensitive sensory neurons were altered following scald burn injury. To identify these populations, capsaicin, a TRPV1 agonist that activates a subset of pain-sensitive neurons, and chloroquine, an MrgprA3 agonist that activates a subset of itch-sensitive neurons, were applied directly to TG neurons during in vivo imaging. The proportion of chloroquine-sensitive neurons was markedly reduced on day 1 after injury and showed significant recovery by day 7 (Fig. 5A; Supplementary video 4–6). In contrast, the proportion of capsaicin-sensitive neurons was not significantly changed on either day 1 or day 7 compared with baseline (Fig. 5B; Supplementary video 7–9).

Figure 5.

A reduction in the proportion of chloroquine-sensitive TG neurons following scald burn injury. (A) There was a statistically significant decrease in chloroquine-sensitive neurons on day 1 (blue bar) and day 7 (green bar) compared to baseline (day 0; white bar) (n = 4). Error bars are SEM. *p < 0.05, ***p < 0.001; one-way ANOVA followed by Tukey’s multiple comparisons test. F(2, 9)=18.58, p=0.0006. (B) In contrast, the proportion of capsaicin-sensitive neurons remained unchanged after scald burn injury (n= 4). Error bars are SEM. one-way ANOVA followed by Tukey’s multiple comparisons test. F(2, 9)=0.06972, p=0.9332.

Discussion

In this study, we demonstrate that scald burn injury to the cheek skin induces both pain- and itch-related behaviors with distinct temporal profiles. Pain-related behaviors peaked on day 1 after injury and resolved within 5 days, whereas itch-related behaviors emerged later, peaked on day 7, and persisted for up to 28 days. Using in vivo calcium imaging, we further identified a significant increase in the number of TG neurons exhibiting spontaneous activity during both the pain-dominant (day 1) and itch-dominant (day 7) phases. Together, these findings establish a mouse model that captures key behavioral and neuronal features of post-burn pain and itch and provides a useful tool for investigating the peripheral mechanisms underlying post-burn sensory dysfunction.

The temporal dissociation between pain and itch observed in this model is consistent with clinical observations. Human studies have reported that burns involving the face, head, and neck are strong predictors of post-burn itch²⁸, while these regions are often associated with lower pain intensity compared to burns at other body sites²⁹. Our cheek scald burn model similarly exhibits prolonged itch with relatively transient pain, suggesting that anatomical location may influence the balance between pain and itch following burn injury. It is possible that burn injury at other body sites may produce more persistent pain states, an important consideration for future model development and comparative studies.

We observed a greater number of spontaneously active TG neurons during the pain-dominant phase compared to the itch-dominant phase. This pattern is consistent with the population coding model of pain and itch signaling³⁰, in which itch is proposed to arise from activation of a restricted subset of spinal neurons that respond to both pruritogenic and noxious stimuli, whereas noxious stimulation recruits a broader population of nociceptive neurons, including pruritogen-responsive neurons, to encode pain. Accordingly, a larger population of spontaneously active sensory neurons during the pain-dominant phase may bias sensory processing toward pain rather than itch. However, an important caveat is that spontaneous neuronal activity observed after burn injury may reflect injury-induced tissue damage or inflammation, rather than directly driving pain or itch behaviors.

We also observed a marked reduction in the proportion of chloroquine-sensitive (MrgprA3-expressing) neurons shortly after burn injury, with partial recovery by day 7, whereas the proportion of capsaicin-sensitive neurons remained unchanged. The mechanisms underlying this selective and transient reduction remain unclear. One possibility is that the relatively small calcium signals evoked by chloroquine make MrgprA3-expressing neurons particularly sensitive to burn-induced alterations in excitability. Alternatively, burn injury may transiently suppress MrgprA3-mediated signaling without eliminating itch-responsive neurons. Further studies will be necessary to clarify these possibilities.

Increased spontaneous activity in primary sensory neurons is a hallmark of peripheral sensitization and has been implicated in multiple chronic pain and itch states. Consistent with this notion, patients with chronic post-burn itch exhibit enhanced itch evoked by histamine application³¹. Such heightened itch responsiveness may arise from increased spontaneous firing of primary sensory neurons following burn injury, thereby lowering activation thresholds and amplifying pruritic signaling.

In summary, this study establishes a clinically relevant mouse model of post-burn pain and itch and identifies dynamic changes in primary sensory neuron activity following burn injury. These findings advance our understanding of peripheral sensory mechanisms in post-burn conditions and provide a foundation for future studies aimed at identifying targeted therapies for post-burn pain and itch.

Supplementary Material

Supplementary Files

This is a list of supplementary files associated with this preprint. Click to download.

• Suppl4.mp4

• Suppl6.mp4

• Suppl7.mp4

• Suppl5.mp4

• Suppl9.mp4

• Suppl8.mp4

• Suppl3.mp4

• Suppl1.mp4

• Suppl2.mp4

Figure 3.

Pain- and itch-related behaviors following scald burn injury. Postburn mice exhibited spontaneous scratching (filled circle) and wiping behavior (open circle; n=11). In the control group (CNT), scratching is shown as filled squares and wiping as open squares (n=6). Error bars are SEM. *p < 0.05, **p < 0.01, ****p < 0.0001; one-way RM ANOVA followed by Dunnett’s multiple comparisons test. Burn scratching; F(8, 80)=2.682, p=0.0115. Burn wiping; F(8, 80)=60.38, p<0.0001.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding authors upon reasonable request.