Protein kinase Cδ mediates histamine-evoked itch and responses in pruriceptors

To determine if PKCδ plays a role in behavioral responses to itch, we assessed scratching responses to histamine and non-histaminergic pruritogens in PKCδ knock-out mice (PKCδ-KO) and their wild type littermates. Mice were injected intradermally at the nape of the neck with one of three pruritogens: histamine (1 mg), chloroquine (CQ) (200 μg), or β-alanine (223 μg). PKCδ-KO mice scratched significantly less than their wild type littermates when injected with histamine (Figure 1A, p < 0.05). On the other hand, chloroquine-induced scratching was not significantly different between WT and PKCδ KO mice (Figure 1B, p = 0.129). There was also no difference in the number of scratch bouts induced by β-alanine (Figure 1C, p = 0.61). These results indicate that PKCδ mediates histaminergic itch, but is not necessary for non-histaminergic itch induced by CQ and β-alanine.

PKCδ is expressed in a variety of tissues, including the brain and peripheral nervous system [19, 21, 23‐30]. To assess whether PKCδ is localized to potential pruriceptive sensory neurons, immunohistochemistry (IHC) was used to characterize the distribution of PKCδ in dorsal root ganglion (DRG) neurons. Antibody specificity was confirmed via western blot using PKCδ-KO DRG and spinal cord tissue. No antibody staining was found corresponding to the 78kD PKCδ band in PKCδ-KO DRG tissue (Figure 2A). This was further confirmed by IHC of knock-out and wild type DRG (Figure 2B). In wild type lumbar DRG, PKCδ was expressed in 43.2% of total neurons labeled with βIII tubulin (567/1314 cells, n = 3 animals) and PKCδ expression was predominantly restricted to small diameter neurons (average diameter 23.3 ± 0.19 μm, min 11.6 μm, max 36.2 μm) (Figure 2C).

We further characterized PKCδ expression in small diameter DRG neurons by immunohistochemical analysis of peptidergic and non-peptidergic markers. Of peptidergic neurons identified by anti-Calcitonin Gene Related Peptide (CGRP+) immunoreactivity, 26.8% expressed PKCδ, while 14.7% of PKCδ-expressing neurons were CGRP-positive (Table 1, Figure 3A-B). PKCδ was also expressed in non-peptidergic neurons identified by isolectin B4 (IB4) binding. Of IB4+ DRG neurons, 61.6% expressed PKCδ and 55.0% of PKCδ+ neurons exhibited IB4 binding (Table 1, Figure 3C-D). These findings indicate that PKCδ is expressed in both peptidergic and non-peptidergic sensory neurons, with greater expression overlap found with non-peptidergic IB4+ neurons.

Behavioral and physiological studies have shown that itch produced by histamine is largely dependent on the non-specific cation channel transient receptor potential vanilloid receptor 1 (TRPV1) [31‐33]. Because PKC has previously been demonstrated to modulate TRPV1 function and may present one potential mechanism by which PKCδ regulates itch, we determined the degree of overlap between PKCδ and TRPV1 expression [34, 35]. We first confirmed the specificity of our antibody directed against TRPV1 using TRPV1 knockout mice (Figure 3E, inset). Our data indicate that 29.1% of PKCδ-positive neurons were also TRPV1-positive, and 48.9% of TRPV1-positive neurons also expressed PKCδ, suggesting a potential functional relationship between PKCδ and TRPV1 (Figure 3E-F; Table 1).

PKCδ is expressed in the brain, spinal cord, and the peripheral nervous system [30, 36‐39]. Therefore, global genetic deletion of PKCδ in our knockout mice makes it difficult to pinpoint where PKCδ functions to modulate histamine-induced scratching. The expression of PKCδ in small diameter sensory neurons suggests that it may mediate histamine-evoked itch by signaling in nociceptive neurons responsive to pruritic agents (i.e., pruriceptors). To determine if PKCδ directly modulates neuronal responses to histamine, calcium imaging was performed on acutely dissociated adult mouse DRG neurons (Figure 4A-B). Of the total sensory neurons treated with histamine, 11.1% of wild type neurons responded to bath application of 100 μM histamine (126/1137 total WT neurons, N = 5 animals), but only 6.7% of PKCδ-KO neurons responded to histamine (47/706 total KO neurons, N = 3 animals), indicating a significant reduction of 39.6% in the proportion of histamine responsive neurons (p < 0.01, χ² test) (Figure 4C). No significant difference in peak calcium responses to histamine was detected between knock-out and wild type cells (WT 33.6 ± 2.8% increase from baseline, n = 126 cells; KO 39.9 ± 6.6% increase from baseline, n = 47 cells, unpaired t-test, p = 0.304) (Figure 4D).

We hypothesized that PKCδ could contribute to neuronal responses to histamine by mediating histamine receptor coupling to TRPV1, or by regulating the normal expression or function of TRPV1. To investigate whether the absence of PKCδ affects the activation of TRPV1 within histamine-responsive neurons, we applied the TRPV1-specific agonist capsaicin after the histamine response. We found that 58.7% of WT histamine-responsive neurons subsequently responded to capsaicin (74/126 total His+ neurons), while only 40.4% of KO histamine-sensitive neurons responded to capsaicin (19/47 total His+ neurons; p < 0.05, χ² test) (Figure 4E). To determine whether the functional expression of TRPV1 is altered in PKCδ-KO neurons, we tested WT and KO sensory neurons for responses to capsaicin. There were no differences in the total number of capsaicin-responsive neurons between KO and WT groups (117/194 (60.3%) WT neurons, 144/259 (55.6%) KO neurons, p = 0.315, χ² test, Figure 4F). Together, these data indicate that the reduction in histamine responses is not due to altered levels of TRPV1 receptors in the KO, and support the idea that PKCδ could modulate TRPV1 downstream of histamine receptor activation.

To further control for the possibility of developmental effects or compensatory mechanisms that may occur with congenital genetic deletion of PKCδ, we performed calcium imaging experiments with the same experimental design using acute pharmacological inhibition in wild type DRG cultures. The δV1-1 peptide inhibitor has been shown to inhibit PKCδ activation in vitro and in vivo by competitively binding to receptor for activated C-kinase (RACK) proteins, which confer PKC isoform substrate specificity [28, 40‐43]. DRG neurons were incubated with the peptide inhibitor δV1-1 for 30 minutes prior to recording. Consistent with our findings in PKCδ-KO neurons, there was a significant reduction in the total number of histamine-responsive neurons treated with peptide inhibitor when compared to scramble peptide-treated control neurons (18.5% (75/405) of scramble-treated neurons vs. 13.1% (69/527) of peptide-treated neurons, N = 4 animals, p < 0.05, χ² test) (Figure 4G). Peak calcium responses induced by histamine were not different between inhibitor- and scramble-treated groups (inhibitor 33.23 ± 2.7% change from baseline, n = 69 cells; scramble 41.0 ± 3.7% change from baseline, n = 75 cells, p = 0.095, unpaired t test) (Figure 4H).

All experiments were conducted in accordance with the National Institute of Health guidelines and received the approval of the Animal Care and Use Committee of Washington University School of Medicine. 8–12 week old male littermate mice were housed on a 12 hour light–dark cycle and allowed ad libitum access to food and water.

PKCδ-KO mice were obtained from Dr. Michael Leitges [70]. These mice were generated using a standard gene targeting approach to insert a LacZ/neo cassette in the first transcribed exon of the PKCδ gene to abolish transcription, resulting in a global knock-out [70]. PKCδ-KO mice were backcrossed on a C57BL/6 background for at least 6 generations prior to use. PKCδ-KO mice were then crossed with wild type C57BL/6 mice to generate heterozygous mice, which were used to generate wild type and KO littermates.

The nape of the neck and upper back were shaved with electric clippers one day prior to behavioral experiments. On the day of experiment, mice were placed in individual plexiglass observation boxes and allowed to acclimate in the presence of white noise for 2 hours. Using gentle restraint, 50 μl consisting of pruritogen dissolved in 0.9% normal saline was injected intradermally at the nape of the neck using a 29½ gauge insulin syringe. The following pruritogen amounts were used: 1 mg histamine (Sigma Aldrich, St. Louis, MO), 200 μg chloroquine (Sigma Aldrich, St. Louis, MO), and 223 μg β-alanine (Sigma Aldrich, St. Louis, MO). A single scratch bout was defined as one or more rapid back-and-forth motions of the hindpaw directed at the injection site, ending with either a pause, licking, or biting of the toes or placing of the hindpaw on the floor. Scratch bouts by the hind-paw directed at the injection site were counted over a period of 30 minutes. Experimenters were blinded to mouse genotype.

For Western blotting, mice were euthanized by swift decapitation and lumbar spinal cord and lumbar DRG were removed. Tissue samples were homogenized in homogenization buffer (20 mM Tris–HCl, pH 7.4, 1 mM EDTA, 1 mM sodium pyrophosphate, 25 μg/ml aprotinin, 25 μg/ml leupeptin and 100 μM phenylmethylsufonyl fluoride) on ice. 7 μg of spinal cord and DRG protein were separated using 4-12% SDS-PAGE, then transferred to nitrocellulose membrane. Membrane was blocked in Odyssey blocking buffer for 1 hour, then incubated in rabbit anti-PKCδ (1:1000, Santa Cruz) and mouse anti-β-Tubulin (1:1000, Sigma-Aldrich) primary antibodies in Odyssey buffer with 0.1% Tween-20 at 4°C overnight. Blots were then washed in TBS-0.1% Tween-20, and incubated for 1 hour at room temperature in secondary antibodies (goat anti-rabbit Alexa Fluor 680 (1:20,000, Sigma Aldrich); goat anti-mouse IR800 (1:20,000, Sigma Aldrich)). Blots were washed and scanned using an Odyssey infrared scanner.

For immunohistochemistry (IHC), mice were deeply anesthetized with a ketamine, xylazine, and acepromazine cocktail, then perfused intracardially with cold PBS followed by 4% paraformaldehyde in PBS. Lumbar DRG were removed and cryoprotected in 30% sucrose. Transverse sections were cut at 18 μm thickness on a cryostat and collected on slides. To determine the percentage of total neurons that express PKCδ, dual labeling was performed with rabbit anti-PKCδ (1:50, Santa Cruz) and mouse anti-β-tubulin (1:1000, Sigma Aldrich) primary antibodies. Briefly, sections were blocked in 2% BSA, 0.1% Milk powder, 0.05% Tween-20 TBS for 1 hr, then incubated in primary antibodies overnight at 4°C. On day 2, slides were washed and incubated in secondary antibodies for 2–4 hours at 4°C (Alexa Fluor 488 Donkey anti-rabbit 1:200, Alexa Fluor 555 donkey anti-mouse 1:200, Invitrogen). Images were obtained using an upright epifluorescent microscope (Nikon 80i, CoolSnapES camera). Labeled neurons were counted in at least 3 randomly selected sections separated by >50 μm per animal. The size distribution of PKCδ+ neurons was determined using ImageJ software to measure cell diameter. The percentage of PKCδ+ neurons that also expressed CGRP or IB4 was determined using dual labeling for PKCδ and CGRP (goat anti-CGRP 1:400, Serotec) or Alexa-568 conjugated to IB4 (1:400, Invitrogen) using the above-described procedures. PKCδ-TRPV1 coexpression was determined using a goat anti-PKCδ antibody (1:50, Santa Cruz) and a rabbit antibody directed against the TRPV1 C-terminus peptide (1:500) [34].

Scratching behavior was evoked with pruritic stimuli applied to the back skin where site directed scratching occurs. We expanded our functional analyses of neuronal physiology to include both thoracic and lumbar DRG. Mice were euthanized rapidly by decapitation and DRG removed and acutely dissociated using previously described methods [71]. Briefly, DRG were incubated in 45U papain/L-cysteine in Hank’s buffered saline solution (HBSS) without Ca²⁺ or Mg²⁺ and with 10 mM HEPES for 20 minutes at 37°C and 5% CO₂. Ganglia were then washed, followed by 20 minute incubation in 1.5 mg/ml collagenase in HBSS + HEPES. Ganglia were then triturated with fire-polished Pasteur pipettes, the dissociated cells were filtered through a 40 μm cell strainer, and were plated on poly-D-lysine and collagen-coated glass coverslips. Cells were incubated overnight at 37°C in 5% CO₂ humidified air in culture medium (Neurobasal A with B27, pen/strep, 2 mM glutamax, 5% fetal bovine serum (Gibco)). All experiments were performed within 24 hours of plating.

Cells were incubated in 3 μg/ml Fura-2 AM (Molecular Probes) for 30 minutes and then incubated for 30 minutes in external solution (in mM): 130 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 30 Glucose, 10 HEPES. For each recording, a coverslip was placed in a perfusion chamber and perfused with external solution at room temperature. Cells were viewed under an inverted microscope (Olympus Optical, Tokyo, Japan) and images were captured with a Hamamatsu Orca camera. SimplePCI Software was used to draw regions of interest (ROI) around Fura-loaded cells prior to recording. The ratio of fluorescence emission at an excitation wavelength of 357 and 380 nm was measured for each ROI. The experimental protocol consisted of a 2 minute baseline followed by 30 second bath application of histamine (100 μM in external solution), >8 minutes of external solution wash, 10 second application of capsaicin (200 nM), >8 minutes of external solution wash, and 10 seconds of KCl (50 mM) followed by wash (<2 minutes) to determine live neurons. A 10% or greater change from baseline 357 nm/380 nm ratio was considered a response to histamine. Capsaicin experiments were performed similarly except with 10 second capsaicin application (200nM). For experiments using PKCδ peptide inhibitor and scrambled peptide, cells were incubated in 100 μM peptide solution (dissolved in external solution) for 30 minutes prior to recording (δV1-1 peptide inhibitor (Myr-SFNSYELGSL-NH2), peptide inhibitor scramble (Myr-GLSFSEYLSN-NH2), Biomatik).