Histamine H4 receptor and TRPV1 mediate itch induced by cadaverine,a metabolite of the microbiome

Abstract

Cadaverine is an endogenous metabolite produced by the gut microbiome with various activity in physiological and pathological conditions. However, whether cadaverine regulates pain or itch remains unclear. In this study, we first found that cadaverine may bind to histamine 4 receptor (H4R) with higher docking energy score using molecular docking simulations, suggesting cadaverine may act as an endogenous ligand for H4R. We subsequently found intradermal injection of cadaverine into the nape or cheek of mice induces a dose-dependent scratching response in mice, which was suppressed by a selective H4R antagonist JNJ-7777120, transient receptor potential vanilloid 1 (TRPV1) antagonist capsazepine and PLC inhibitor U73122, but not H1R antagonist or TRPA1 antagonist or TRPV4 antagonist. Consistently, cadaverine-induced itch was abolished in Trpv1 ^−/− but not Trpa1 ^−/− mice. Pharmacological analysis indicated that mast cells and opioid receptors were also involved in cadaverine-induced itch in mice. scRNA-Seq data analysis showed that H4R and TRPV1 are mainly co-expressed on NP2, NP3 and PEP1 DRG neurons. Calcium imaging analysis showed that cadaverine perfusion enhanced calcium influx in the dissociated dorsal root ganglion (DRG) neurons, which was suppressed by JNJ-7777120 and capsazepine, as well as in the DRG neurons from Trpv1 ^−/− mice. Patch-clamp recordings found that cadaverine perfusion significantly increased the excitability of small diameter DRG neurons, and JNJ-7777120 abolished this effect, indicating involvement of H4R. Together, these results provide evidences that cadaverine is a novel endogenous pruritogens, which activates H4R/TRPV1 signaling pathways in the primary sensory neurons.

Introduction

The gut microbiome is composed of microorganism and serves as a key regulator of host physiology, brain function and behavior through immune, neuronal, and metabolic pathways.^1,2 Certain microbes possess the ability to synthesize and/or modulate various metabolites (e.g. neurotransmitters), which may directly or indirectly affect neuronal activity in the peripheral and central nervous system. ³ The metabolites of gut microbiome, including lipids, short-chain fatty acids, bile acids, neurotransmitters, vitamins and peptides, etc., have been identified as important regulators in human health and the pathogenesis of many different diseases, such as obesity, diabetes, cardiovascular diseases, chronic liver disease, inflammatory bowel disease and cancers. ⁴ Recently, emerging evidence indicates that gut microbiome is closely associated with the regulation of chronic pain and itch, serving as a promising target for treatment of chronic pain or itch.^5–10 Intriguingly, the biogenic amines are a class of metabolites of microbial activity in the gastrointestinal tract, and they may serve as neurotransmitters, precursors to hormones, modulators of immune response, as well as regulators of physiological functions.^11,12 Some key prominent monoamines, including histamine, serotonin, dopamine, epinephrine and norepinephrine, are critically involved in neurotransmission and regulation of mood and behavior. Polyamines differ significantly in structure and metabolism with monoamines, and they play pivotal roles in cellular growth, proliferation, differentiation and apoptosis, as well as the development of nociception. ¹³

Cadaverine (Pentane-1,5-diamine) is a biogenic polyamine generated by gut microbiome through enzymatic decarboxylation of l-lysine, with various chemical and biological activity in living organisms. ¹⁴ Cadaverine seems to be distinct from other polyamines in structure and has independent biosynthetic pathways, plays an important role in the regulation of cell survival and tumorigenesis.^15–18 Cadaverine was found to accumulate in significant quantities in various food products, including cheese, fish and fish-based products, ¹⁹ the consumption of high concentration of cadaverine may cause severe toxicological reactions. ²⁰ Elevated levels of cadaverine, as observed in the urine of patients with defects in lysine metabolism, indicating disruptions in amino acid degradation pathways involved. In addition, cadaverine was identified as the primary source of urinary putrescine. ²¹ Interestingly, cadaverine had also been proven to be present in the central nervous system, ²² indicating it is a possible neuromodulator. Cadaverine is secreted primarily by Gram-negative bacteria in the gut. ²³ Elevated concentrations of cadaverine synthesized by lysine decarboxylase enzyme is instrumental in conferring protection to the human body against acidic stressors. ²⁴ Despite cadaverine exhibiting a relative lower toxicity compared with histamine, its interaction with the amine oxidases may slow down the metabolism and enhance the toxicity of histamine, ¹⁹ which is primarily facilitated through its competitive inhibition of diamine oxidase and histamine N-methyltransferase.^25–28

Histamine, a classic endogenous monoamine, is well-known as an inflammatory mediator that can induce itch (pruritus) and pain.^29,30 Histamine exerts pronociceptive effects primarily by stimulating unmyelinated C fibers, whose cell bodies located in the dorsal root ganglia (DRG) or trigeminal ganglia (TG). To date, there are four types of G-protein-coupled receptors (GPCRs) that can bind histamine, including H1R, H2R, H3R, and H4R.^29,31 The activation of H1 and H4 receptors located at peripheral nerve terminals by mast cell-derived histamine are particularly important for signaling transduction of itch. ³² Histamine binding to the receptors H1 and H4 activates the transient receptor potential vanilloid-1 (TRPV1) channel via the enzymatic action of phospholipase C (PLC), leading to histamine-dependent pruritus (histaminergic itch). ³³ The H4 receptor is a member of histamine receptors family, which has a relative higher genetic similarity to the H3 receptor (37% homology) compared to the H1 and H2 receptors (less than 30% homology). ³⁴ Recent studies indicate that the H4 receptor is critically involved in the pathogenesis of inflammation, immunity and itch in murine models.^35,36 Previous studies demonstrated that intradermal injection of 4-methylhistamine, an H4 receptor agonist, was able to induce acute itch in mice. ³⁵ Both immunohistochemistry and single-cell RT-PCR analysis demonstrated the expression of H4 receptor in DRG neurons in humans and rats.^37,38 Activation of H4 receptor can increase intracellular calcium ion level in DRG neurons. ³⁷ Intriguingly, recent research provided a clue that cadaverine may act as an endogenous agonist for the H4 receptor (with an EC50 of 1.1 μM). ³⁹ However, whether cadaverine play a role in itch sensation and the downstream signaling remain to be fully elucidated.

In the present study, we tested the hypothesis that cadaverine may act as a new ligand for H4R and induce acute itch through activation of H4R and TRPV1 in mice. We first found that cadaverine showed similar binding characteristics with histamine when docking to H4R. We further confirmed that cadaverine dose-dependently provoked acute itch in mice, and the activation of both H4R and TRPV1 were required for cadaverine-induced itch. Moreover, H4R was essential in cadaverine-induced itch and the hyperexcitability of dissociated DRG neurons. In conclusion, our findings demonstrated that gut-derived metabolite cadaverine may serve as a new endogenous pruritogen and the H4R/PLC/TRPV1 signaling axis played a crucial role in cadaverine-induced acute itch in mice.

Materials and methods

Animals

Male C57BL/6 wild-type (WT) (6-8 weeks old) mice were purchased from the Shanghai SLAC Laboratory Animal CO., LTD (Shanghai, China). For knockout experiments, TRPA1 knockout and TRPV1 knockout mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA). All animals were maintained under a controlled environment with constant temperature (22 ± 2°C) and humidity (40-60%), and a 12-h light cycle, with free access to food and water. All animal experiment protocols in this study were reviewed and approved by the Animal Care and Use Committee of Nantong University (Reference: S20210305-025) and were performed following the guidelines of the International Association for the Study of Pain.

Drugs and administration

Cadaverine (Cat#C8561), Compound 48/80 (Cat#C2313), chlorpheniramine maleate (Cat#C3025) were all purchased from Sigma-Aldrich (St Louis, MO, USA). JNJ-7777120 (Cat# HY-13508), HC-030031 (Cat# HY-15064), HC-067047 (Cat#HY-100208), Capsazepine (Cat# HY-15640), U-73122 (Cat# HY-13419) were obtained from MCE (MedChemExpress, New Jersey, USA). Morphine hydrochloride was obtained from China Northeast Pharmaceutical Group Shenyang No. 1 Pharmaceutical CO., Ltd (Shenyang City, Liaoning Province, China). Naloxone hydrochloride was obtained from China Sinopharm Group Guorui Pharmaceutical CO., Ltd (Huainan City, Anhui Province, China).

Cadaverine (10-800 μg), JNJ-7777120 (200-600 μg), HC-030031 (100 μg), capsazepine (30-100 μg), HC-067047 (300 μg) and U73122 (300 μg) were intradermally injected into the nape region of the mice using a 26G needle. The volume of intradermal injection in the nape region was 50 μL. Chlorpheniramine (1-10 mg/kg), JNJ-7777120 (1-30 mg/kg), naloxone (10 mg/kg), morphine (10 mg/kg) were intraperitoneal injection using a 26G needle. Compound 48/80 was administered intraperitoneally to mice on four consecutive days at doses of 125, 300, 625, and 1000 μg/kg. The volume of intraperitoneal injection was 200 μL. Specific information relating to timing and doses are indicated in the results section or figure legends.

Neck model of acute itch in mice

A neck model of acute itch was established as previously reported,^40,41 mice were shaved at the nape and were habituated to the testing environment 2 days before experiments. On the day of behavioral testing, mice were placed in individual acrylic boxes on a metal grid rack, allowing for at least 40 min of acclimatization. Following transient anesthesia with isoflurane, 50 μL of drugs were intradermally injected into the nape of the neck of the mice using a 26G needle. After injection, the mice were promptly relocated to their boxes, with their activities video-recorded for a duration of 30 min. These recordings were later analyzed offline to quantify the scratching behavior. A single scratching behavior was defined as the mouse lifting its hind leg off the ground, scratching the skin behind the ear or on the back, and then placing the paw back to the ground.

Cheek model in mice

The cheek model was employed to distinguish the itch and pain behavior of mice. ⁴² Mice were shaved at the cheek 2 days before experiments. On the day of behavioral testing, mice were placed in individual acrylic boxes on a metal grid rack, allowing for at least 30 min of acclimatization. Following transient anesthesia with isoflurane, cadaverine (100-300 μg) was injected into the cheek of the mice using a 26G intradermal injection needle. The volume of intradermal injection in the cheek region was 10 μL. After injection, the mice were promptly relocated to their boxes, with their activities video-recorded for a duration of 30 min. We separately quantified wiping and scratching behaviors. A single instance of wiping behavior is characterized by the mouse lifting its forelimb towards the cheek, engaging in a wiping motion for a duration ranging from one to several seconds, followed by the retraction of the forelimb.

Primary dorsal root ganglia neuron culture

Dorsal root ganglion neurons were dissociated and prepared from adult mice (6-8 weeks) using a similar protocol as previously described.^43,44 Briefly, mice were sacrificed after anesthesia with isoflurane. All the cervical, thoracic and lumbar DRGs were rapidly removed and placed in ice-cold oxygenated balanced Dulbecco’s Modified Eagle Medium (DMEM, Gibco, Grand Island, NY). The DRGs were then digested with collagenase D (0.6 U/ml; Roche, Mannheim, Germany) and dispase II (3.0 U/ml; Roche, Mannheim, Germany) for 35–40 min at 37°C. The ganglia were then triturated with fire-polished Pasteur pipettes. The dispersed cells were resuspended in F12 (Biological Industries, Beit HaEmek, Israel) medium with 10% FBS (Gibco, Waltham, MA, United States) and 1% penicillin/streptomycin (Biosharp, Hefei, China) and plated on coverslips coated with polyornithine (BBI lifescience, Shanghai, China). Cell cultures were maintained in regular 95% air and 5% CO2 at 37°C in an incubator.

Calcium imaging

Primary cultured DRG neurons were used for calcium imaging after incubation overnight. Primary cultured DRG neurons were loaded with 1 μg/mL Fura-2 a.m. (1:1000, Thermo Fisher) and 0.01% F-127 (w/v; Invitrogen) for 30 min in a 37°C incubator with 5% CO₂. Images were captured via a CCD camera (PCO, Germany) integrated with an inverted microscope (Nikon, Japan). The excitation of Fura-2 was facilitated by an alternating light source (PTI, USA) emitting wavelengths of 340 nm and 380 nm. In chambers equipped with a custom four-channel perfusion valve control system, neurons were infused with calcium imaging buffer (CIB) (130 mM NaCl, 5.6 mM KCl, 2.6 mM CaCl₂, 1.2 mM MgCl₂, 10 mM HEPES, and 5.6 mM d-glucose at pH 7.4) to baseline, then infused the test solution, and finally with 56 mM KCl for 1 min and infused with CIB to baseline.

Molecular docking

The molecular docking analysis was performed using Discovery Studio Software platform in Shanghai University. The 3D structure of cadaverine and histamine were obtained from PubChem (CID: 273 and 774, respectively), the 3D structures of murine and human H4R are available in UniProt (ID: Q91ZY2 and Q9H3N8, respectively), and they were used for CDOCK (A small molecule-protein docking algorithm) analysis, which is based on energy minimization and simulated annealing techniques. The study selected a suitable grid size and center that covers all possible binding regions in the receptor (H4R) based on the ligand binding mode of 7vv6, and adjusted Exhaustiveness to a high value of 32 to increase search space and precision and reduce the probability of false positives and false negatives. The TYL06 force field was used for the molecular docking study, which is a molecular force field specifically designed for protein-small molecule interactions. Compared with other force fields, TYL06 force field adopts a second-order moment approximation method in handling charge distribution and considers the dynamic process of hydrogen bond formation, which can more accurately calculate the charge interaction between drug molecules and receptors and describe the hydrogen bond interaction between drug molecules and receptors. The study adjusted the number of optimal molecular docking conformations generated by the search algorithm to 12, which broadened the search space and improved the probability of finding the optimal solution. The study employed a global optimization search algorithm based on genetic algorithms, which often produces multiple optimal solutions with very subtle differences, and represents different poses or orientations that the ligand can adopt within the binding site.

Electrophysiological recordings

DRG neurons were recording 16–24 h after dissociation as described previously.^44,45 Small diameter DRG neurons (<25 μm) were chosen for whole-cell patch clamp recording in the current-clamp mode at room temperature. Data were acquired using a Multiclamp 700B amplifier (Molecular Devices Corporation, Sunnyvale, CA, United States) driven by a personal computer in conjunction with an A/D and D/A board (DigiData 1550B series interface, Molecular Devices Corporation). DRG neurons were recorded with fire-polished, borosilicate glass patch pipettes (5–8 MΩ), which were pulled from borosilicate glass capillaries (Sutter Instrument, Novato, CA, United States) using a Sutter P-1000 puller (Sutter Instrument, Novato, CA, United States). The bath solution consisted of 140 mM NaCl, 3 mM KCl, 2 mM MgCl₂, 2 mM CaCl₂, 10 mM HEPES, pH 7.3. The pipette solution consisted of 30 mM KCl, 110 mM potassium gluconate, 0.5 mM EGTA, 5 mM HEPES, and 3 mM Mg-ATP, pH 7.3. The action potential (AP) was evoked by a series depolarizing step currents with long or short time durations (1000 ms or 2 ms). The long current steps were used to test AP rheobases and count evoked AP numbers. The short current steps were used to trigger single action potential that was used to measure the shape properties of APs. A series of ramp currents (100-300 pA, 1000 ms) were also used to study the excitability of neurons under suprathreshold stimuli. Signals were low-pass filtered at 5 kHz, sampled at 20 kHz and analyzed offline.

Statistical analysis

Data were analyzed using Graphpad Prism 8.0 (GraphPad Software Inc., USA), Origin 7.5 (OriginLab, USA) and Fitmaster v2X65 software (HEKA Electronik, Germany). All data were expressed as mean ± standard error of the mean (S.E.M.). Statistical analysis was performed by the Student’s t test, one-way and two-way analysis of variance (ANOVA) followed by Dunnett multiple comparison tests. p < .05 was considered statistically significant.

Results

Cadaverine is a potential ligand of H4R

The structure of histamine consists of an imidazole ring, a carbon skeleton and an amino group (Figure 1(a)). In contrast, cadaverine is a compound with a straight-chain skeleton of five carbon atoms, with an amine group at each end of the chain (Figure 1(a)). To evaluate the possible interaction between cadaverine and H4R, we first used molecular docking to study the interaction mechanisms of cadaverine with murine and human H4R, while histamine was performed the same analysis as a comparison. The 3D structure of the histamine H4 receptor was modeled based on predictions from AlphaFold (UniProt ID: Q91ZY2 for mouse and Q9H3N8 for human). This study employed a global optimization search algorithm based on genetic algorithms, which often produces multiple optimal solutions with slight differences, indicating that the ligand can adopt different poses or orientations within the binding site. Our results showed that both murine and human histamine H4 receptors have a pocket that can accommodate either cadaverine or histamine (Figure 1(a), Figure 2(a)). For the murine histamine H4 receptor, CYS98, SER101 and ASN352 form hydrogen bonds with cadaverine (Figure 1(b), (c)), which shared two same hydrogen bonds with histamine (Figure 1(d), (e)). The binding free energies are −25.7442 kcal/mol and −3.62145 kcal/mol, respectively, indicating that the complexes were in a table conformation. To assess the energy impact of individual residues on binding affinity, we used the Volume Arranged Molecular Shape (VAMS) method. Mutation of ASN352 in the cadaverine-murine histamine H4 receptor complex led to the greatest energy increase (Figure 1(f)), suggesting its importance in stabilizing the complex. Similarly, mutation of TRP318 in the histamine-murine histamine H4 receptor complex caused the greatest energy increase (Figure 1(g)). Mutations in VAL64, SER101, TRP318, and ASN352 in both complexes also resulted in energy increase.

Figure 1.

Molecular docking analysis of cadaverine and histamine to mouse H4R. (a) The murine H4R have a binding pocket that can accommodate cadaverine or histamine. (b), (c) CYS98, SER101 and ASN352 of H4R form hydrogen bonds with cadaverine. (d), (e) CYS98 and ASN352 of H4R form hydrogen bonds with histamine. (f), (g) The energy impact of key residues mutation on binding affinity of cadaverine-murine H4R complex or histamine-murine H4R complex.

Figure 2.

Molecular docking analysis of cadaverine and histamine to human H4R. (a) The human H4R have a binding pocket that can accommodate cadaverine or histamine. (b), (c) CYS98, SER101, ASN350 and SER351 of H4R form hydrogen bonds with cadaverine. (d), (e) CYS98, SER101, and ASN350 of H4R form hydrogen bonds with histamine. (f), (g) The energy impact of key residues mutation on binding affinity in the cadaverine-human H4R complex or histamine-human H4R complex.

Similarly, for the human histamine H4 receptor, CYS98, SER101, ASN350 and SER351 form hydrogen bonds with cadaverine (Figure 2(b), (c)), while CYS98, SER101, and ASN350 also form hydrogen bonds with histamine (Figure 2(d), (e)). The binding free energies are −9.71949 kcal/mol and −5.53632 kcal/mol, respectively, indicating stable conformations of the complexes. Moreover, mutation of ASN350 in both the cadaverine-human H4R and histamine- H4R complexes resulted in the greatest energy increase (Figure 2(f), (g)), suggesting its significant role in stabilizing the complex. Likewise, mutations in ASP61, VAL64, SER101 and TRP316 resulted in energy increase (Figure 2(f), (g)).

Based on the above results, we also used molecular docking to assess whether cadaverine can interact with the other histamine receptors besides H4R. The results showed that among the murine histamine receptors, cadaverine showed highest docking energy score with H4R (−25.7442 kcal/mol), the other three basically have little interaction with cadaverine (Figure 3(a)). Among the human histamine receptors, cadaverine may be able to bind to either H4R or H2R (−9.71949 and −18.5571 kcal/mol, respectively) (Figure 3(b)), but not H1R or H3R. Thus, these results indicate conserved binding sites and interaction modes of cadaverine and the murine or human H4R receptor.

Figure 3.

Molecular docking analysis of cadaverine and histamine receptors. (a) Molecular docking analysis of cadaverine and murine histamine receptors. (b) Molecular docking analysis of cadaverine and human histamine receptors.

Cadaverine is sufficient to induce acute itch in mice

Considering the pivotal role of histamine as a well-known itch mediator, ⁴⁶ we next investigated whether cadaverine has itch-inducing effects in vivo. We first assessed its efficacy in evoking acute itch responses by intradermal injection of cadaverine in mice. The dose-dependent scratching behaviors were induced by the i.d. injection of cadaverine (10-800 μg in 50 μL saline) into the nape of the neck in mice (p < .0001, Figure 4(a), (b)). Behavioral results showed that scratching behavior commenced at an intradermal injection of 30 μg and peaked at 500 μg cadaverine, while a reduction in scratching was observed for 800 μg compared to 500 μg cadaverine, suggesting an invert-U shape dose-dependent curve. The cheek model demonstrated that intradermal injection of cadaverine could induce both itch-indicative scratching and pain-indicative wiping behaviors in mice. Notably, itch-inducing scratching behaviors were more prevalent than pain-inducing wiping behaviors following intradermal injections of 100 and 300 μg cadaverine (p < .001, Figure 4(c), (d)). We also injected cadaverine into the mice plantar area to determine whether cadaverine induced itch or pain behavior, it is typically deemed painful when mice lick their feet and itchy when they bite their hind feet. The results showed that within 30 min, intraplantar injection of 300 μg cadaverine significantly increased mice biting behavior but not licking behavior compared to control mice (p < .01, Figure 4(e), (f)). We also tested the radiant thermal latency and mechanical pain thresholds after intraplantar injection of cadaverine, and there were no significant differences between two groups (Figure 4(g), (h)). Thus, intradermal injection of cadaverine is sufficient to induce itch behavior in mice.

Figure 4.

Cadaverine-induced itch in mice and its pharmacological characterization. (a) The number of scratching incidents induced within 5 min following the injection of 10-800 μg cadaverine into the nape skin of mice. (b) The total number of scratching behavior events within 30 min in response to intradermal injection of 10-800 μg of cadaverine into the nape of mice (***p < .001 vs Saline, n = 12 or 6, one-way AVOVA following Bonferroni’s test). (c) Scratching behavior induced by intradermal injection of cadaverine in the cheek of mice (***p < .001 vs Saline, n = 12 or 6, one-way AVOVA following Bonferroni’s test). (d) Wiping behavior induced by intradermal injection of cadaverine in the cheek of mice (***p < .001 vs Saline, n = 12 and 6, one-way AVOVA following Bonferroni’s test). (e) Biting behavior induced by intraplantar injection of cadaverine (**p < .01 vs Saline, n = 6, unpaired Student’s t test). (f) Liking behavior induced by intraplantar injection of cadaverine (n = 6, unpaired Student’s t test). (g) The effect of intraplantar injection of cadaverine on mice radiant thermal latency (n = 6, unpaired Student’s t test). (h) The effect of intraplantar injection of cadaverine on mice mechanical pain threshold (n = 6, unpaired Student’s t test).

Pharmacological characterization of cadaverine-induced acute itch in mice

We next elucidate the potential role of histamine receptors in the itch induced by cadaverine in mice. Mice were intraperitoneally injected with the H1R antagonist chlorpheniramine (1 mg/kg and 10 mg/kg), and 30 min later, 300 μg of cadaverine was injected intradermally into the nape of the neck in mice. The results showed that chlorpheniramine did not affect scratching behavior induced by cadaverine in mice (Figure 5(a)). Subsequently, mice were intraperitoneally injected with the histamine H4R antagonist JNJ7777120 (1, 10 and 30 mg/kg), followed by 300 μg intradermal injection of cadaverine into the nape of neck in mice. The results showed that acute itch behavior induced by cadaverine was significantly inhibited by JNJ7777120 (p < .01, Figure 5(b)). Consistently, intradermal co-administration of cadaverine and JNJ7777120 (300 μg and 600 μg) also significantly attenuated cadaverine induced acute itch in mice (p < .05 and p < .01, Figure 5(c)). In addition, intradermal injection of PLC inhibitor U73122 significantly alleviated cadaverine induced itch behavior (p < .001, Figure 5(d)), indicating that PLC may act as an important downstream signaling molecule of cadaverine.

Figure 5.

The role of histamine receptors, PLC, μ-opioid receptor, mast cell and TRP channels in cadaverine-induced itch. (a) Pretreatment with chlorpheniramine did not attenuate the scratching behavior induced by cadaverine (n = 6, one-way AVOVA following Bonferroni’s test). (b)The effect of intraperitoneal administration of JNJ7777120 on cadaverine intradermal injection induced scratching behavior (***p < .001 vs Saline, n = 6, one-way AVOVA following Bonferroni’s test). (c) Co-injection of cadaverine with JNJ7777120 in the nape of mice decreased the scratching frequency (*p < .05, **p < .01 vs Saline, n = 13, 8 and 6, respectively, one-way AVOVA following Bonferroni’s test). (d) The effect of intradermal injection of U73122 on cadaverine induced scratching behavior (***p < .001 vs Saline, n = 6, unpaired Student’s t test). (e) Morphine had no effect on cadaverine-induced itch behavior (n = 6, unpaired Student’s t test). (f) The opioid receptor antagonist naloxone alleviated cadaverine-induced itch behavior in mice (**p < .01 vs Control, n = 6, unpaired Student’s t test). (g) Intradermal coadministration of cadaverine and Naloxone alleviated cadaverine-induced itch behavior in mice (**p < .01 vs Control, n = 6, unpaired Student’s t test). (h) Depletion of mast cells in mice using compound 48/80 attenuated cadaverine-induced scratching behavior (**p < .01 vs WT group, n = 6 and 5, unpaired Student’s t test). (i) Co-injection of TRPV1 antagonist Capsazepine in the nape of mice reduced cadaverine induced scratching behavior (**p < .01 vs cadaverine group, n = 6, one-way AVOVA following Bonferroni’s test). (j) Co-injection of TRPA1 antagonist HC-030031 in the nape of mice did not affect cadaverine induced scratching behavior (n = 6, unpaired Student’s t test). (k) Co-injection of TRPV4 antagonist HC-067047 in the nape of mice did not affect cadaverine induced scratching behavior (n = 6, unpaired Student’s t test). (L) Cadaverine-induced scratching behavior in Trpv1^−/− and Trpa1^−/− mice (**p < .01 vs WT, n = 5, one-way AVOVA following Bonferroni’s test). All data are expressed by mean ± SEM. (M) The gene expression of H4R, μ opioid receptor, TRPV1 and PLC on different subgroups of DRG neurons (scRNA-Seq data collected from Mouse Brain Atlas website https://mousebrain.org/).

Endogenous opioids are involved in the regulation of neurotransmission and immune function, as well as pain and itch. to explore whether opioid receptors are involved in cadaverine-mediated acute itch, the μ-opioid receptor agonist morphine (10 mg/kg) was injected intraperitoneally into mice, followed by intradermal injection of cadaverine (300 μg), we observed that morphine did not significantly affect cadaverine induced scratching behavior (Figure 5(e)). In contrast, pretreated with the μ-opioid receptor antagonist naloxone (10 mg/kg) intraperitoneally markedly attenuated the acute itch behavior induced by cadaverine in mice (p < .01, Figure 5(f)). Consistently, intradermal co-administration of cadaverine and Naloxone (300 μg) also significantly attenuated cadaverine induced acute itch in mice (p < .01, Figure 5(g)).

Mast cells are well-known to play a crucial role in neurogenic inflammation, particularly in the genesis of pain and itch. Next, we pre-treated mice with compound 48/80 in order to deplete mast cells in vivo, following an administration of intradermal injection of cadaverine (300 μg) in mice. The behavioral results demonstrated that mice pretreated with compound 48/80 exhibited a significant reduction in cadaverine-induced scratching behavior in mice (p < .01, Figure 5(h)).

TRPV1 but not TRPA1 is requied for cadaverine-induced acute itch in mice

TRP channels (e.g., TRPA1 and TRPV1) are highly expressed in capsaicin-sensitive nociceptors, and they are crucial for the transmission and processing of various sensory signals, including thermal sensation, pain and itch. ⁴⁷ We further investigated the role of TRP channels in cadaverine-induced itch. Co-intradermal injection of the TRPV1 antagonist capsazepine (100 μg and 300 μg) and cadaverine (300 μg) were performed simultaneously, the results showed that capsazepine dose-dependently attenuated the scratching behavior induced by cadaverine in mice (p < .01, Figure 5(i)). However, treatment of TRPA1 antagonist HC-030031 or TRPV4 antagonist HC-067047 had no significant effect on cadaverine induced itch behavior (Figure 5(j), (k)). Consistently, cadaverine-induced itch behavior was abolished in Trpv1^−/− mice, while there was no significant difference between WT mice and Trpa1^−/− mice (p < .01, Figure 5(l)).

To know whether these receptors or proteins mentioned above have co-expression on DRG neurons, we collected the normalized expression level of these genes (scRNA-Seq data) from the Mouse Brain Atlas website (https://mousebrain.org/) in each subtype of DRG neurons, then draw the heat map based on these data. The heat map showed that H4R and the μ opioid receptor are mainly co-expressed in NP2, NP3, and PEP1 neurons. Additionally, the κ opioid receptor and TRPV1 both co-express with H4R mainly in PEP1 neurons. PLC receptors express in several different subtypes of DRG neurons, and they also co-express with H4R mainly in PEP1 neurons, except PLCB2. These data suggested that opioid receptors, TRPV1, and PLC receptors all have co-expression with H4R in DRG neurons (Figure 5(m)), the interaction between these receptors may be required for cadaverine induced itch.

Cadaverine directly activates primary cultured DRG neurons of mice

To determine whether cadaverine induces itch response by directly activating DRG neurons and the underlying molecular mechanisms, we employed calcium imaging to assess the effect of cadaverine on primary cultured DRG neurons. The results revealed that cadaverine induced an increase of intracellular calcium levels in DRG neurons in a dose-dependent manner (p < .001, Figure 6(a)–(c)), with approximately 3.5% of neurons response to cadaverine at 1 μM and about 4.7% response to cadaverine at 10 μM, while around 17.4% of neurons were response to 100 μM cadaverine (Figure 6(d)). Additionally, we further investigated the subtypes of primary cultured DRG neurons activated by cadaverine. By directly perfusing cadaverine followed by treatments with AITC (a TRPA1 agonist) and capsaicin (a TRPV1 agonist) (Figure 6(e), (f)), we found that out of total 357 cells, 21 neurons responded to cadaverine, AITC, and capsaicin, while 27 out of 29 cadaverine-responsive neurons were response to capsaicin (Figure 6(g)). These results indicated that cadaverine can activate DRG neurons directly and TRPV1 channel could be a downstream effector of H4R.

Figure 6.

Cadaverine induced calcium signal in DRG neurons. (a) Representative images of DRG neurons responses to 1 μM, 10 μM, and 100 μM cadaverine. (b), (c) Cadaverine activated DRG neurons in a dose-dependent manner (***p < .001 vs 1 μM group, ^###p < .001 vs 10 μM group, n = 253, one-way AVOVA following Bonferroni’s test). (d) Percentage of DRG neurons activated by different concentrations of cadaverine (e) Representative images of DRG neurons responses to cadaverine, AITC and capsaicin. (f), (g) The proportion of cadaverine-activated DRG neurons among TRPV1 and TRPA1 positive neurons. All data are expressed by mean ± SEM.

H4R/PLC/TRPV1 signaling mediated cadaverine-induced intracellular calcium increase in primary cultured DRG neurons

To further investigate whether cadaverine acts as a selective agonist of the H4 receptor in DRG neurons, we perfused cadaverine and JNJ7777120 (a H4R antagonist) simultaneously on primary cultured DRG neurons and performed calcium imaging analysis. The results showed that JNJ7777120 treatment significantly inhibited the enhancement of intracellular calcium signaling induced by cadaverine in DRG neurons (Figure 7(a), (d) & (g)). Similarly, a TRPV1 antagonist capsazepine (CPZ) also had significant inhibitory effect on cadaverine-induced calcium signaling (p < .0001 and p < .01, Figure 7(b), (e) & (g)). Additionally, considering the critical role of PLC in transmembrane signal transduction and the connection of G-protein-coupled receptors to intracellular signaling networks, we then perfused cadaverine and a PLC inhibitor (U73122) on primary cultured DRG neurons. Calcium imaging results indicated that U73122 markedly reduced intracellular calcium signaling of DRG neurons induced by cadaverine (Figure 7(c), (f) & (g)). Collectively, these findings support the hypothesis that cadaverine may directly activate neuron by stimulating H4R, promoting TRPV1 channel open through the activation of PLC pathway, thereby increasing the intracellular calcium ions concentration in DRG neurons.

Figure 7.

The effect of H4R, LPC and TRPV1 on cadaverine induced calcium signal in DRG neurons. (a) Representative images of DRG neurons responses to H4R antagonists JNJ7777120 and cadaverine. (b) Representative images of DRG neurons responses to TRPV1 antagonists capsazepine and cadaverine. (c) Representative images of DRG neurons responses to PLC antagonists U73122 and cadaverine. (d) Representative traces of DRG neurons responses to H4 antagonists JNJ7777120 and cadaverine. (e) Representative traces of DRG neurons in response to TRPV1 antagonists capsazepine and cadaverine. (f) Representative traces of DRG neurons in response to PLC antagonists U73122 and cadaverine. (g) The normalized peak response of the DRG neurons in the presence of JNJ7777120, CPZ, and U73122. (***p < .001 vs Control, n = 253, 124, 229 and 286, respectively, one-way AVOVA following Bonferroni’s test). All data are expressed by mean ± SEM.

We next employed TRPV1 knockout mice to determine whether these TRPV1 is required for cadaverine-induced increase of intracellular calcium in the DRG neurons (Figure 8(a)). Calcium imaging results showed that the robust calcium signaling caused by cadaverine and the percentage of responsive cells was significantly reduced in primary cultured DRG neurons from TRPV1 knockout mice (p < .0001, Figure 8(b)–(d)), indicating that TRPV1 plays a crucial role in cadaverine-induced activation of calcium signaling in DRG neurons.

Figure 8.

The effect of TRPV1 deletion on cadaverine induced calcium signal in DRG neurons. (a) Representative images of DRG neurons from Trpv1 ^−/− mice in response to cadaverine, capsaicin and AITC. (b) Representative traces of DRG neurons in from Trpv1 ^−/− mice in response to cadaverine, capsaicin and AITC. (c) The normalized peak response of the DRG neurons from Trpv1 ^−/− mice. (d) The proportion of cadaverine-activated DRG neurons in Trpv1 ^−/− mice. (****p < .0001 vs WT, n = 253 and 113, respectively, unpaired Student’s t-test). All data are expressed by mean ± SEM.

H4R mediated cadaverine-induced hyperexcitability in small diameter DRG neurons

Finally, we performed whole cell patch-clamp recording to determine whether cadaverine altered the electrical activity of small diameter DRG neurons (Figure 9(a), (b)). It was found that the membrane capacitance was comparable in the groups with and without JNJ-7777120 (Table 1). We also did not observe spontaneous firing during the perfusion of cadaverine (data not shown). We found that cadaverine-induced membrane depolarization was accompanied with significant decline of rheobase (from 70.0 ± 7.5 pA to 50.0 ± 5.8 pA, p < .01, Figure 9(c)). Naturally, JNJ-7777120 significantly reduced the decreases of rheobase (20.0 ± 4.7 pA to 4.3 ± 4.3 pA, p < .05, Figure 9(d)). Meanwhile, The one minute-perfusion of cadaverine significantly depolarized the membrane potential of DRG neurons (from −50.1 ± 1.6 mV to −47.4 ± 1.9 mV, Δ membrane potential is 2.7 ± 0.8 mV, n = 14), which was almost completely abolished by JNJ-7777120 (form −51.4 ± 1.5 mV to 51.3 ± 1.3 mV, Δ membrane potential is 0.3 ± 0.7, n = 14, p < .01and p < .05, Figure 9(e), (f)), this was consistent with the results of calcium imaging. Cadaverine comparably increased input resistance of both control and JNJ-7777120 incubating group (the increment was 34.3 ± 5.4% and 32.9 ± 8.0%, respectively, p < .01 and p < .0001, Figure 9(g)). In addition, the firing frequency of the DRG neurons in the control group was significantly increased following cadaverine perfusion. Notably, the increases of the firing frequency became significantly smaller in the presence of JNJ-7777120 (p < .05, p < .001and p < .01, Figure 9(h)). We also use a series of ramp current to measure the excitability of the neurons under suprathreshold stimuli (Figure 9(i), (j)). Similarly, cadaverine also significantly increased the firing frequency of DRG neurons (p < .001, Figure 9(k)).

Figure 9.

The effect of cadaverine on the excitability of small DRG neurons. (a), (b) The representative traces of action potentials evoked by step current injection in the absence or in the presence of JNJ7777120, respectively. The traces evoked by rheobase current were shown in red. (c) The rheobase current to evoke an action potential (^**p < .01 pre-vs post-, paired Student’s t-test, n = 14). (d) The shift of the rheobase current induced by cadaverine (^*p < .05 vs cadaverine group, unpaired Student’s t-test, n = 14). (e) The resting membrane potential of the neurons before and after cadaverine treatment (^**p < .01 pre-vs post-, paired Student’s t-test, n = 14). (f) The shift of the resting membrane potential induced by cadaverine (^*p < .05 vs cadaverine group, unpaired Student’s t-test, n = 14). We defined the delta value bigger than ±2 mV as apparently shift. (g) The membrane resistance of the neurons before and after cadaverine treatment (^**p < .01, ^****p < .0001 pre-vs post-, paired Student’s t-test, n = 14). (h) The number of action potentials evoked by step current stimuli (^*p < .05 vs vehicle group, ^##p < .01, ^###p < .001 vs cadaverine group, two-way ANOVA followed by Bonferroni’s multiple comparisons test, n = 14). (i)&(j) The representative trains of action potentials evoked by ramp current injection. (k) The number of action potentials evoked by ramp current injection (^***p < .001 vs vehicle group, two-way ANOVA followed by Bonferroni’s multiple comparisons test, n = 14). All data are expressed by mean ± SEM.

Table 1.

Effects of in vitro Cadaverine prefusion on the electrophysiological properties in mouse small DRG neurons from control group and JNJ7777120 preincubation group.

											AP Number
	Cadaverine	Rm (MΩ)	RMP (mV)	RB (pA)	AMP (mV)	AHP (mV)	APD50 (ms)	AHP_D50 (ms)	S_rise (mV/ms)	S_decy (mV/ms)	20 pA	40pA	60pA	80pA	100pA	100pA	200pA	300pA
Control (n = 14)	Pre-	1570.91 ± 203.26	−50.05 ± 1.64	70.00 ± 7.49	108.60 ± 1.44	−22.57 ± 1.78	3.95 ± 0.35	71.23 ± 9.36	84.80 ± 5.74	−33.87 ± 2.16	0.00 ± 0.00	0.64 ± 0.43	1.86 ± 0.92	3.57 ± 1.14	4.86 ± 1.41	4.93 ± 1.11	8.86 ± 1.69	10.93 ± 1.99
	Post-	2228.84 ± 3 20.72**	−47.38 ± 1.91*	50.00 ± 5.84**	107.96 ± 2.08	−21.72 ± 1.87*	4.28 ± 0.30***	89.89 ± 15.70	74.77 ± 4.89**	−28.73 ± 1.56*	0.57 ± 0.39**	2.64 ± 0.88***	5.00 ± 1.53***	7.36 ± 1.75***	8.86 ± 2.05***	6.85 ± 1.43***	10.57 ± 1.93***	12.07 ± 2.14
JNJ (n = 14)	Pre-	2029.46 ± 2 90.07	−51.41 ± 1.53	70.00 ± 6.55	107.94 ± 3.39	−22.24 ± 1.18	3.94 ± 0.27	106.21 ± 20.57	81.49 ± 9.57	−35.49 ± 2.46	0.00 ± 0.00	0.21 ± 0.11	0.92 ± 0.30	1.50 ± 0.41	2.50 ± 0.63	3.07 ± 0.80	6.21 ± 1.21	8.14 ± 1.43
	Post-	2804.91 ± 2 90.69***	−51.11 ± 1.34	65.71 ± 7.39	107.93 ± 3.53	−22.01 ± 1.24	4.37 ± 0.29***	120.32 ± 19.69**	74.20 ± 8.73**	−31.57 ± 2.57**	0.07 ± 0.07	0.71 ± 0.37	1.71 ± 0.61	2.93 ± 0.82*	4.00 ± 1.03*	3.86 ± 0.91	7.07 ± 1.37*	9.00 ± 1.64*

Rm: membrane resistance; RMP: resting membrane potential; RB: rheobase; AMP: action potential amplitude; AHP: afterhyperpolarization amplitude; APD50: action potential duration at 50% of AMP; AHP_D50: afterhyperpolarization duration at 50% amplitude; S_rise: maximal rising slope; S_decay: maxmimal decaying slope; All data are represented as mean ± S.E.M *, p < .05; **, p < .01; and ***, p < .001, pre-versus post-, paired Student’s t-test.

Discussion

Biogenic amines are naturally occurring amines that are derived from amino acids, and they play crucial roles in various biological processes, including neurotransmission, inflammation, regulation of blood flow, nausea, vomit, as well as itch.^20,48 In the past decades, researchers have found several different biogenic amines which function as neurotransmitters, including dopamine, norepinephrine, epinephrine, histamine, and serotonin. ¹¹ Some of these amines were also found to act as itch mediators, the most well-known among these in the context of itching is histamine, but other biogenic amines may also be involved. Understanding the role of these mediators is essential for effective anti-itch treatment, particularly in chronic itch.

In the present study, using molecular simulation docking analysis, behavioral testing, pharmacological and genetic manipulations, calcium imaging and patch clamp recording, we identified and characterized cadaverine, can bind to H4R and act as a novel mediator of itch in mice. Our results found that cadaverine and histamine shared some similarity in the binding sites and modes when docking with mouse or human H4R. Intradermal injection of cadaverine in either the nape of the neck or cheek induced acute itch in a dose-dependent manner in mice. We confirmed that inhibition of the μ opioid receptor or mast cell depletion both attenuated cadaverine induced scratching. In addition, the pharmacological blockade or genetic knockout showed that H4R and TRPV1 were both required for cadaverine-induced itch in mice. Furthermore, cadaverine perfusion directly activated cultured DRG neurons in a dose-dependent manner, which was characterized with enhanced calcium signaling, and over 90% cadaverine-positive neurons were also response to capsaicin. Consistently, TRPV1 antagonist capsazepine and PLC inhibitor U73122 both significantly inhibited cadaverine-induced calcium signaling, similar calcium signaling attenuation was also observed in the DRG neurons from TRPV1 knockout mice. Patch clamping recording showed that the perfusion of cadaverine directly enhanced the neuronal excitability of the dissociated small diameter DRG neurons, the application of H4R antagonist JNJ-7777120 significantly reduced this effect. Collectively, our data verified the hypothesis that gut microbiome metabolite cadaverine is a endogenous ligand of H4R and induces itch through H4R/PLC/TRPV1 signaling pathways (Figure 10).

Figure 10.

Schematic diagram of the signaling pathway involved in cadaverine induced itch. The gut microbiome generate cadaverine is a novel ligand for H4R, cadaverine activates DRG neurons through H4R/PLC/TRPV1 signaling pathway, causing calcium influx and hyperexcitability in small diameter DRG neurons, thus inducing itch in mice. Abbreviations: AP, action potential; PLC, phospholipase C.

Itch is a common symptom of many dermatological or systemic diseases. Acute itch helps protection against harmful external threats, but chronic itch is a distressing condition that lasts for more than 6 weeks, significantly impairs patient’s life quality. Based on the underlying pathological mechanisms, itch is often categorized into two main types: histaminergic and non-histaminergic (histamine-independent) itch. ⁴⁶ Understanding the mechanism underlying chronic itch helps in diagnosing and treating various chronic itch conditions. For instance, in cases where antihistamines are ineffective for anti-itch treatment, it might indicate a non-histaminergic mechanism. Non-histaminergic itch is often more challenging to treat due to its diverse and complex mechanisms. With the rapid expansion on itch study methods and novel itch mediators, our knowledge on itch mechanism and treatment have dramatically expanded. In this study, cadaverine was found to be a novel itch mediator, indicating that the elevation of endogenous synthesized cadaverine may be involved in itch sensation, and targeting certain gut microbiota or controlling cadaverine synthesis could be an effective therapeutic strategy for chronic itch.

The microbiome plays a crucial role in various human physiological processes, including metabolic regulation, epithelial development, and immune function. Numerous studies have revealed associations between the microbiome and a range of chronic diseases such as obesity, inflammatory bowel disease, diabetes, allergic rhinitis, and atopic dermatitis (AD),^49–53 although these associations do not directly imply causality. Atopic dermatitis is a prevalent chronic inflammatory skin condition characterized by intense itching, with multiple genetic and environmental factors contributing to its development. Researches indicated that the gut microbiome of AD patients is imbalanced, with a lack of microbial diversity compared to healthy individuals.^54,55 The pathophysiology of AD is complex, involving a predisposition towards Th2 immune responses and defects in the innate immune system. ⁵⁶ Notably, the prevalence of AD is increasing globally.^57–59 Factors predisposing to AD may include smaller family sizes, urban living environments, and western dietary habits, all of which can affect the composition of the skin and gut microbial communities. Numerous cohort studies suggested that abnormalities in the gut microbiome may precede the onset of atopic diseases, ⁵⁵ with AD often being considered the initial manifestation of the atopic march. Infants with AD have been found to have lower abundance of Bifidobacteria and Lactobacilli, and higher levels of Enterobacteriaceae, along with a lack of necessary bacterial diversity,^60–62 according to specific cohort studies. While there isn’t a direct relationship between atopic dermatitis and cadaverine, the presence of bacteria capable of producing cadaverine could potentially influence the skin microbiome and contribute to inflammation or exacerbate existing skin conditions. The potential role of certain microbial metabolites, including cadaverine, in skin inflammation and conditions like atopic dermatitis will be explored in more future studies.

As a natural polyamine with multiple bioactivities, cadaverine is prevalently present in the nervous system and is believed to originate almost entirely from the decarboxylation activities of bacteria in the intestine. ⁶³ In recent decades, microbial metabolomics research has been developed remarkably. A major focus in microbiome research has been on understanding the impact of microbial communities and their metabolic products on human health and diseases. For instance, one study reported that high amounts histamine produced by gut microbiota of many patients with irritable bowel syndrome (IBS) plays a vital role in abdominal pain via acting on H4R, targeting bacterial histamine was regarded as new strategies to treat visceral hyperalgesia. ⁶⁴ Another recent study used large scale functional screening of molecules produced by individual members of a simplified human microbiota and found several biogenic amine metabolites including tryptamine, tryptamine, putrescine, agmatine, spermidine and cadaverine to agonize GPCRs associated with diverse functions within the nervous and immune systems. Among them, cadaverine showed very high selective agonism effect to H4R, suggesting an important role of cadaverine on the histamine signaling pathways. ³⁹ Histamine is an endogenous pruritogenic compound predominantly found in mast cells and basophils, has been shown to occupy a similar pocket on both murine and human H4R through molecular docking analysis, forming hydrogen bonds with the same amino acid residues, such as CYS98 in murine H4R and CYS98, ASN350 in human H4R. This reminds that the activation mechanism of H4R by cadaverine may share some similarities with histamine. Although the molecular structure cadaverine and histamine share very limited similarity, we observed very high similarity in the binding sites and modes of cadaverine and histamine with both the murine and human H4R following molecular docking analysis. At least two identical hydrogen bonds were shared between cadaverine and histamine when binding to H4R. In addition, their binding energy was not determined by a single or two amino acids, several amino acids all have an impact on the change of binding energy, but merely a difference of degree. These results provide evidence that cadaverine may serve as an endogenous ligand of H4R.

Mast cells play a crucial role in the sensation of itch. ⁶⁵ These cells are widely distributed across various tissues in the body, with a pronounced presence on the surfaces of skin and mucous membranes. ⁶⁶ Mast cells contain granules that store various inflammatory mediators, such as histamine, leukotrienes, prostaglandins, and cytokines. ⁶⁷ Upon activation, mast cells expedite the release of these mediators that can affect various processes, including tissue remodeling, immune response, vascular tone, and injury perception. Cadaverine was reported to induce histamine release from mast cells, ⁶⁸ the H4R is highly expressed on mast cells and DRG neurons,^69,70 so cadaverine may activate the histamine H4 receptor on mast cells, facilitates the release of inflammatory mediators and cytokines, and on the other hand directly binds to H4R on DRG neurons, activating TRPV1, thus causing itch sensation. Therefore, cadaverine induced itch is a multi-factor involved complicated effect, the mechanism of cadaverine-mediated acute itch warrants more comprehensive investigation.

Numerous studies have substantiated the critical role of transient receptor potential (TRP) channels in regulating a variety of sensory functions. These include perception of thermosensation, mechanosensation, vision, olfaction, taste, as well as pain and itch.^71,72 Among TRP channels, TRPA1 and TRPV1 are widely distribute in nerves fibers and are well studied in the regulation of itch sensation. The histamine itch and non-histamine itch are mainly mediated by TrpV1 and TrpA1 respectively. ⁷³ For example, under atopic dermatitis or allergic rhinitis conditions, TRPV1 was upregulated and mediates itch-associated scratching. ⁷⁴ TRPA1 was found to be an important target of Zn^{2 + 75} and Zn²⁺/TRPA1/GPR39 axis played a critical role in Zn2+-induced acute and chronic itch in mice. ⁷⁶ Our previous study also reported that the polyamine agmatine induces histamine-independent itch in mice through the activation of ASIC3, but not TRPV1 or TRPA1. In this study, cadaverine induced itch behavior was almost abolished in Trpv1^−/− but not Trpa1^−/−mice, and the pharmacology test of capsazepine, HC030031 and U73122 showed consistent results, indicating that TRPV1 may serve as a downstream of cadaverine induced itch response, which has similar mechanism with histamine. However, the difference is that TRPA1 and TRPV1 are both involved in histamine induced itch, ⁷⁷ while only TRPV1 is required for cadaverine induced itch.

Our calcium imaging and electrophysiological recording also observed the direct activation and enhanced excitability of cultured DRG neurons by cadaverine perfusion, this is consistent with the docking analysis that cadaverine was able to bind to H4R. And the subfamily of neurons activated by cadaverine were lowly overlapped with that activated by AITC but very highly overlapped with that activated by capsaicin. Moreover, the proportion responsive cells to cadaverine in Trpv1^−/−mice was reduced to less than 2%, which indicated that TRPV1 expression DRG neurons are the main contributor of cadaverine induced itch. In addition, our electrophysiology study also observed an unexpected but apparent increases of input resistance after perfusion of cadaverine in both control and JNJ-7777120 incubating group, which suggested that cadaverine might causes the closure of putative ion channels. Furthermore, this effect is H4R independent. Consistently, both the rising slope and decay slope were reduced by cadaverine, which suggested that the influx and efflux were attenuated at the same time (Table 1). However, we also got similar results in JNJ-7777120 incubated neurons, so these results should not be associated with the cadaverine-induced hyperexcitability in DRG neurons.

In conclusion, our study demonstrated that cadaverine from gut microbiota may serve as an endogenous ligand for H4R, cadaverine binds and activate H4R and then activates TRPV1 channel, which contributed to itch signal transduction. Thus, besides H4R/PLC/TRPV1 signaling axis, targeting cadaverine synthesis by intervention of gut microbiota may also provide novel anti-itch strategy.