Evaluation of a polymer coated nanoparticle cream formulation of resiniferatoxin for the treatment of painful diabetic peripheral neuropathy

Introduction

Diabetic Peripheral Neuropathy (DPN) is a common long-term complication of diabetes^11,12. DPN with pain, referred to as Painful DPN (PDPN) is reported in 10% to 20% of patients with diabetes and in 40% to 50% of those with DPN^10,22,52. DPN develops as a result of progressive loss of nerve fibers^46,55,56. Symptoms of PDPN include electrical or stabbing sensations, paresthesia, hyperesthesia, burning pain, pins and needles and deep aching pain^6,22,52. Pain impinges upon quality of life leading to anxiety, depression and sleep disturbances^{11,13,30,51,62}. Currently, the American Academy of Neurology recommends pregabalin, gabapentin, duloxetine, amitriptyline, and venlafaxine as first-line treatment with second-line treatment options being sodium valproate, opioid analgesics, and tramadol^13,24,54. Orally administered agents have a significantly elevated risk for adverse effects^{9,23,32,58,60}.

However, DPN pain is poorly responsive to centrally acting neuropathic pain medications in 60% of patients, leading to chronic opioid use. Despite evidence-based guidelines, opioids remain the most frequently prescribed first line pain medication; 74% of neuropathy patients receive opioids but only 35% are eligible with the guidelines^12,16,55. Dependence and over-dose leading to premature death has become an epidemic.

Therefore, topical medications that have less systemic effects are alternatives for patients who are intolerant or have contraindications. Transient Receptor Potential Vanilloid 1 (TRPV1) receptor plays a major role in thermal hyperalgesia in pain and inflammation^14,27,41,50. Recent research has uncovered its involvement in PDPN^37,39,41. Studies have shown using a transgene- and STZ-induced diabetes that TRPV1 expression and function are increased and decreased in hyper or hypoalgesic phenotypes³⁹. TRPV1 agonist, capsaicin applied topically has been shown to relieve pain and improve sensory perception in PDPN^3,5. TRPV1 antagonists can be effective in treating painful conditions including PDPN^7,49,53. Unfortunately, in clinical trials, TRPV1 antagonists caused hyperthermia despite showing analgesia^18–21.

Although intra-epidermal nerve (C and Aδ) fiber loss is seen in DPN, patients with PDPN complain of burning deep aching pain. A set of C mechanoinsensitive (CMi) fibers distributed in the dermis is spontaneously active in neuropathies^47,48. The diode laser fiber type selective stimulation (DLss) technique can selectively access these fibers^33,34,35,54.

Resiniferatoxin (RTX), an ultra-potent TRPV1 agonist, has been extensively studied. RTX fully activates TRPV1 receptor even at femtomolar to picomolar ranges. The potency of RTX has been previously shown by recording TRPV1-mediated membrane currents^8,26,42. Our studies have demonstrated that RTX prevented generation of action potential by inducing “depolarization block” preventing coordinated activation of sodium channels^8,26,42. RTX is well-tolerated; we administered 200 μg/kg intraperitoneally and found no observable side effects using common techniques to test toxicity²⁶. Studies have also demonstrated the effectiveness of RTX by intrathecal administration^8,26. Intrathecal administration of RTX is being tested in a clinical trial for certain terminal cancer pain and intraarticular injections for osteoarthritic pain of the knee^1,2,29. In this study, we have formulated a nanoparticle cream of RTX and demonstrated its effectiveness in small and large animal models of PDPN.

Materials and Methods

Streptozotocin-induced diabetes in rats and mini-pigs

Male and female wild type (WT) Wistar rats (n = 24 per group) purchased form Jackson Lab (USA) were housed in specific pathogen-free barrier animal facility, and rodent laboratory chow and drinking water were provided ad libitum. The rats were starved overnight and were injected intraperitoneally with freshly prepared streptozotocin (STZ; 55 mg/kg; Cayman, USA) in saline (pH 4.5with 0.1 N citrate buffer). Control rats received saline (pH 4.5 with 0.1N citrate buffer). The food was introduced to the cages after 1-hour post injection. Blood glucose levels were measured everyday post injection and the animals that show blood glucose levels greater than 300 mg/dL were considered diabetic.

Male/female mini-pigs (20–25 kg; n = 5; Waldo Genetics, De Witt, NE, USA) were housed in large animal facility. The mini-pigs were fed ad libitum on a commercial finisher diet and had free access to water. STZ was freshly prepared by dissolving in saline (pH 4.5 with 0.1N citrate buffer) at a concentration of 80 mg/mL. The mini-pigs were starved overnight and then injected with freshly prepared STZ (150 mg/kg body weight) via the ear vein (~1 mL/s) within 2 min, while the animal was anesthetized. The blood glucose levels were measured by using a commercial glucometer and blood glucose levels greater than 300 mg/dL were considered diabetic.

Preparation of RTX Cream

We prepared a proprietary cream formulation of RTX that contains RTX encapsulated by an FDA approved copolymer poly-l-lactide coglycolide (PLGA) by solvent evaporation method. An accurately weighed quantity of RTX was dissolved in ethanol. This ethanolic solution was mixed with 1% PLGA (50:50) copolymer dissolved in dichloromethane. The resultant mixture is evaporated on a rotavap and dried to remove the solvents. The resultant PLGA-encapsulated RTX was mixed with an absorption base to produce RTX-cream containing various concentrations of RTX. Placebo cream was prepared by using the same method without RTX.

In vitro release characteristics of RTX from cream

Either 5 mg of pure drug or 350 mg of RTX cream was placed and spread on a Strat-M Membrane (SKBM02560; EMD Millipore, USA). RTX diffusion across membranes was investigated using a Franz diffusion cell. The diffusion of RTX cream across the Strat-M membrane is predictive of diffusion in human skin. The membrane was mounted between the donor and receptor compartment and either 5 mg of pure RTX or 350 mg of RTX cream was placed on the membrane surface inside the donor compartment. The receptor compartment was filled with a buffer solution, pH 7.4 and stirred at 100 rpm. The buffer samples were collected from the receptor compartment at 0, 5, 15, 30, 60, 90 and 120, 180, 240 and 300 min. and analyzed for RTX concentration by measuring the absorbance at 210 nm in a JASCO V670 spectrophotometer. The cumulative amount of RTX diffusion over 5 hr. for each membrane was plotted against time. From the data, a calibration curve was plotted with known concentrations of RTX ranging from 30 μg/L to 1 mg/L in pH 5.5 buffer. Experiments were performed in triplicates and analyzed. The formula given below was used for calculating the cumulative percent release of RTX in phosphate buffered saline pH 5.5.

$$\mathbf{C}\mathbf{u}\mathbf{m}\mathbf{u}\mathbf{l}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{v}\mathbf{e}\mathbf{\%}\mathbf{R}\mathbf{e}\mathbf{l}\mathbf{e}\mathbf{a}\mathbf{s}\mathbf{e}={\mathbf{C}}_{\mathbf{1}}{\mathbf{V}}_{\mathbf{1}}\times \mathbf{C}\mathbf{F}/\mathbf{100}\times {\mathbf{C}}_{\mathbf{2}}$$

where, C₁, Concentration of drug in final solution (g/mL); C₂, Drug content (mg) used for dissolution study; CF, Correction Factor; V₁, Volume of dissolution medium (mL).

Stability testing and efficacy analysis

Accelerated stability analyses of RTX-cream was performed by storing the cream preparations in a Shell SCH10R humidity temperature-controlled stability chamber. The visual appearance and chemical stability of the product were analyzed at the end of 30, 60, ad 90 days of storage at 37 °C at a relative humidity (RH) of 60 ± 5%. At the end of the specific storage period, the efficacy of the cream was evaluated by measuring thermal sensitivities by Hargreaves method. The goal was to maintain the efficacy of RTX-cream close to 100% with an allowable deviation of 10% throughout the shelf-life period. We also determined the concentration of RTX diffused from the cream across the Strat-M using a Franz Diffusion Cell as described above.

Determination of thermal and mechanical sensitivities in control and diabetic rats

Placebo or RTX cream was applied on the plantar areas of animals (n = 8 / group) for different durations and the rats were subjected to pain testing by the Hargreaves (thermal hyperalgesia) or von Frey (mechanical allodynia) method^26,39. The concentrations of RTX in the cream that delays paw withdrawal latency (PWL) and increases paw withdrawal threshold (PWT) were determined. Placebo or RTX-cream applied on the plantar surface was protected by Tegaderm film and licking of the paws was prevented by facial cones. All the experiments were performed in a blinded manner by randomly assigning treated and untreated groups.

Determination of thermal sensitivity in control and diabetic mini-pigs

In mini-pigs, the pain behavior was tested by drawing 5 × 5, 1 cm square grid in the upper thigh region. The diode laser fiber selective stimulation (DLss) techniques were used to test thermal pain sensitivity in animal models and humans^33–35,54. A stepwise increase in laser current was delivered to each cell in the grid using diode laser stimulator with duration of 2s and power 1–5W using Lass 20 (LasMed, CA) similar to that has been used in diabetic patients³⁵. The pain behavior indicated by a clear skin twitch, and/or an abrupt limb withdrawal and/or a characteristic tail twirl in response to laser stimulation is considered a positive painful response. The experimenter was blinded to the type of animal (control or diabetic) and the treatment (placebo or RTX-cream) while testing.

Analysis of effect of RTX-cream on body temperature

Since TRPV1 agonists cause hypothermia and antagonists cause hyperthermia^20,21, change in core body temperature was determined by measuring rectal temperature using a RET-3 rectal probe (Bioseb.com, USA). We tested the effect of RTX-cream application every week by measuring the rectal temperature within 30 minutes of application and continued every 30 min, for up to 3 hours.

Analysis of TRPV1 expression and function

Western blotting

Skin samples and biopsies collected from control and diabetic rats and mini-pigs were lysed and centrifuged at 14000 rpm for 20 min. Lysis buffer composition (in mM) 50 Tris pH 7.5, 2 EDTA, 250 NaCl, 0.5 DL-dithiothreitol (DTT), 1 Na orthovanadate, 1 Phenylmethylsulfonyl fluoride (PMSF), 1% Nonidet P-40 (NP-40), 0.5% Sodium deoxycholate, and complete protease inhibitor cocktail and phosphatse inhibitor cocktail were added to the buffer. The supernatant was aliquoted and flash frozen in liquid nitrogen and stored at −80°C for experiments. Lysate (equivalent to 40 μg) was prepared by denaturing the protein using 6x Laemmli buffer (LB) and was resolved via SDS-polyacrylamide gel electrophoresis (SDS-PAGE). 3-color regular range protein marker was used to show molecular weights (kDa) of proteins. Proteins were resolved by running the gel at 130V in TGS (Tris/Glycine/SDS) buffer. Gels were transferred to nitrocellulose paper at constant current of 350 mA for 90 minutes using Tris/glycine buffer containing 20% methanol. Membranes were blocked in 5% non-fat dry milk powder in 1X PBS-buffered saline containing 0.1% Tween-20 (PBS-Tween) for 1 hour at room temperature (RT) and incubated with the TRPV1 primary antibody or loading control overnight at 4°C. Densitometry measurements of each protein were quantified using Image-J software (NIH)^15,59.

Immunohistochemistry

At the end of the study, the mini-pigs were euthanized by intraperitoneal injection of ketamine and xylazine (8 mg/kg and 10 mg/kg, respectively). Biopsies of upper thigh skin from the pigs were collected and processed for experiments as described^8,38. Paraffin embedded pig skin tissues (Control or RTX treated) were cut and mounted on slides. Paraffin was removed and the tissues were rehydrated with different grades of ethyl alcohol, distilled water and PBS. Tissue sections were blocked in PBS, Triton X-100, and 5% normal goat serum. After washing with 1x PBS, the sections were exposed to rabbit polyclonal TRPV1 antibody (1:1000 dilution cat #NB-100-98897, Novus Biologicals), rabbit monoclonal PGP 9.5 antibody (1:500 dilution cat # ab-108986, Abcam, USA). Then, sections were treated with secondary antibody; goat anti rabbit IgG (1:1000; Thermofisher scientific, USA) and gently washed 3 times with PBS and cover slipped with Vectashield mounting media (Vector Laboratories). Colocalization of TRPV1 and PGP 9.5 in skin tissues was examined by confocal microscopy (Nikon, A1R Ti-E) or fluorescent microscopy (Nikon, Microphot FXA).

In vitro CGRP release measurement in isolated paw skin

To determine the function, we measured TRPV1-mediated CGRP release using commercially available ELISA kits (MyBioSource, USA and Cayman Chemicals, USA). We stimulated the isolated paw skin tissue with capsaicin (1 μM; 90 min. incubation at 37°C) in control and diabetic animal treated with placebo or RTX cream.

Statistical analysis

Statistical analyses were performed using Origin 2020 software (Origin Labs, Northampton, MA, USA). Statistical significance was calculated using one-way ANOVA followed by Fishers Least Significant Difference (LSD) test and data were expressed as mean +/− SEM, P < .05 was considered to be significant. In all figures, statistical significance is labeled the following way: *P <0 .05, **P <0.01, and ***P < 0.001

Results

The scheme in Figure 1 shows the experimental study design. Diabetes was induced in rats (Fig.1A) and mini-pigs (Fig.1B) by intraperitoneal and intravenous injection of streptozotocin (STZ), respectively. Thermal pain sensitivity was measured every day, one week after the STZ administration until the end of the study. Hargreaves method in rats and DLss technique in mini-pigs were used to test thermal pain sensitivity. Placebo or RTX cream was applied on the left and right hind limb paw skin of rats, and left and right upper thigh regions of the mini-pigs, respectively. Pain behavior was assessed by PWL in rats and by muscle twitch and/or sudden withdrawal of the limb and/or tail twirl in mini-pigs. At the end of the study (16 weeks for rats and 5 weeks for pigs), the animals were euthanized, the paw skin was isolated from rats and skin from upper thigh region of mini-pigs by skin punch biopsy. Ex vivo capsaicin-stimulated CGRP release was determined by ELISA after incubating the skin tissue at 37°C. The rest of the skin samples were used for immunoblotting and immunohistochemistry to estimate TRPV1 protein expression.

Figure 1. Experimental study design.

Following the preparation of RTX cream, we determined the diffusion rate of RTX across skin mimetic membrane. In vitro diffusion kinetics across Strat M membrane of pure RTX or from RTX cream was measured as described in the Methods Section. There was about three-fold increase (P < 0.01) in RTX permeation from the cream formulation as compared to pure RTX (Fig. 2) at 180 min and up to 300 min.

Figure 2: In vitro diffusion kinetics of RTX cream.

Cumulative % release of RTX from pure RTX or cream formulation. RTX release across a Strat-M-membrane (human skin mimic) is significantly (P < 0.01) higher over several time points measured using a Frantz diffusion cell.

Next, we analyzed the changes in thermal sensitivity in diabetic rats following the application of either placebo or RTX cream. One week after STZ injection in both control and diabetic groups, the paw withdrawal latency (PWL) was determined by calculating the time taken for a rat to withdraw the hind limb from a radiant heat source of noxious thermal stimuli. As compared to control animals (Fig. 3A), diabetic rats showed a significant (P < 0.01) reduction in PWL. (Fig. 3B; 5.5 ± 1.1s). Daily application of placebo or RTX cream (1 μM RTX) to non-diabetic control animals (male and female) did not exhibit any significant change in PWL (Fig. 3A; 12.5 ± 1s). However, daily RTX cream application increased and normalized the PWL within a week in both male and female rats and lasted over the course of the study (Fig.3B).

Figure 3: RTX cream suppresses thermal hyperalgesia in diabetic rats.

A. There was no significant change in PWL measured following treatment of plantar surface with RTX cream in control rats over a period of time. B. STZ-induced diabetic animals exhibit hyperalgesia, indicated by a significant decrease in PWL, which was completely reversed to normal values following treatment with RTX (1 μM) cream. ** represents statistical significance P < 0.01 for n = 8–12/condition).

We also measured mechanical allodynia by determining the paw withdrawal threshold (PWT). Following STZ administration, within a week the rats became mechanically hypersensitive measured as decrease in PWT using automated von-Frey filaments. There was a significant decrease in PWT in diabetic animals (Fig. 4). After week 1, the PWT decreased significantly (P <0.05) (week 1, 25.4 ± 9.93 %; week 5, 5.8 ± 0.76 %). Following application of RTX to the plantar surface, the PWT reversed gradually and significantly (P <0.05) (week 1, 22.4 ± 10.8 %; week 5, 51 + 5.6 %).

Figure 4: RTX cream reverses mechanical hypersensitivity in diabetic rats.

Following STZ administration, within a week the rats became mechanically hypersensitive measured ad paw withdrawal threshold (PWT) using automated von-Frey filaments. There was significant decrease in PWT in diabetic animals. PWT measured following treatment of plantar surface with RTX cream show a reversal of mechanical hypersensitivity.

In these diabetic rats, the average blood glucose levels were significantly (P < 0.01) higher than control at two weeks post injection (control, 111 ± 11 mg/dL and diabetic, 444 ± 22 mg/dL) (Supplementary Fig. 1A).

Calcitonin gene–related peptide (CGRP) is a pain related peptide which plays a key role in the development of peripheral sensitization and the associated thermal hypersensitivity. CGRP-containing peptidergic neurons terminate at different depths in the epidermis and dermis and transduce thermal pain. Since RTX induces long-term desensitization/depletion of TRPV1 expressing nerve terminals and decrease CGRP release, we isolated the paw skin from the diabetic and control rats that were treated with placebo or RTX cream and determined the basal and TRPV1-mediated CGRP release in response to stimulation by capsaicin. CGRP release was measured using ELISA. Paw skin from diabetic rats showed a significantly (P < 0.01) higher basal CGRP release (531 ± 26 pg/ml/mg) compared to the control (232 ± 8 pg/ml/mg), which was further amplified when stimulated with capsaicin (1395 ± 68 pg/ml/mg). But, in RTX cream-treated diabetic rat skin, capsaicin-stimulated CGRP release was significantly (P < 0.01) lower (486 ± 24 pg/ml/mg) (Fig. 5A), suggesting the effect on sensory nerve terminals.

Figure 5. RTX cream decreases capsaicin-stimulated CGRP release and TRPV1 expression in diabetic rat paw skin in vitro.

A. Basal and capsaicin (1 μM)-stimulated CGRP release in skin samples of control and diabetic rats showing a significant increase in CGRP release in diabetic hyperalgesic animals, which is significantly reduced following RTX cream treatment. ** represents statistical significance P < 0.01 for n = 3/condition. B. TRPV1 expression determined by immunoblotting in paw skin from control and STZ-induced diabetic animals treated with placebo or RTX cream. C. Densitometric ratio between TRPV1 expression and the respective loading control show a significant increase in diabetic hyperalgesic animals, which is significantly reduced following RTX cream application (n = 4 experiments).

Next, we analyzed the expression of TRPV1 protein in the paw skin of placebo or RTX cream-treated control and diabetic rats. The expression of TRPV1 was significantly higher in the placebo-treated diabetic rat paw skin compared to RTX cream-treated paw skin (Fig. 5B). The density of the bands was measured by image J software. The ratio of band intensity between TRPV1 and the loading control beta actin is represented as bar graph in Fig. 5C. Immunoblotting revealed that TRPV1 expression (fold change) was significantly (1.19 ± 0.11) increased in hyperalgesic rats and RTX cream application significantly (0.52 ± 0.12; P < 0.01) decreased TRPV1 expression.

We also evaluated the effect of RTX cream application on the body temperature by measuring the rectal temperature of control rats that either received placebo or RTX cream application. As shown in Supplementary Fig. 2, RTX cream treatment did not cause any change in the rectal temperature compared to placebo treated controls.

Then, we used a large animal (mini-pig) model of diabetes to determine the effectiveness of RTX cream. Following administration of STZ, the blood sugar levels were significantly higher in diabetic mini-pigs (525 ± 22 mg/dl). Control mini pigs had an average blood sugar levels of 103 ± 20 mg/dl (Supplementary Fig. 1B).

We used an infrared diode laser to induce thermal pain stimuli in control and diabetic mini-pigs. RTX cream (10 nM RTX) was applied daily on the upper right thigh region after removing the hair. Upper left thigh region was treated with placebo cream. On the second day onwards, the pain behavior was tested on the upper thigh regions by drawing 5 × 5, 1 cm square grid, labelled as 1a, 1b and 2a, 2b etc., (Fig. 6A and B). Then, a stepwise increase in laser current using diode laser was applied to each cell on the grid. A pain behavior indicated by a clear skin twitch, and/or abrupt limb withdrawal and/or characteristic tail twirl in response to laser stimulation was considered as a positive response. The results show a significant (P<0.05) difference in pain behavior between RTX cream and placebo cream treated sides suggesting the alleviation of pain in a mini-pig model of PDPN (Fig. 6C).

Figure 6. RTX cream application reduces pain behavior in response to laser-induced noxious thermal stimulus in diabetic pigs.

A and B. The tested upper thigh region shown in the grid with red squares indicating a response and blue squares indicating lack of response to increasing laser induced noxious thermal stimulus in control and RTX cream treated regions. C. Pain behavioral measure expressed in percentage in response to a laser heat generated pain stimulus (DLss technique) following application of placebo or RTX (10nM) cream formulation to the upper thigh region of diabetic mini-pigs. * represents statistical significance P < 0.01 for n = 4 mini-pigs

In order to determine the function, we used TRPV1-mediated CGRP release from pig skin tissue samples. It is clear that samples obtained after treatment with RTX cream showed significantly (P < 0.01) lower CGRP release when stimulated with 1 μM capsaicin (1953 ± 139 pg/ml/mg) compared to untreated controls (4817 ± 255 pg/ml/mg) (Fig. 7A).

Figure 7. RTX cream application reduces capsaicin-stimulated CGRP release and TRPV1 protein expression in diabetic mini- pig skin in vitro.

A. Capsaicin-stimulated CGRP release from skin samples of control and diabetic mini-pigs treated with placebo or RTX-cream. ** represents statistical significance P < 0.01. B. TRPV1 expression in skin samples from control and STZ-induced diabetic mice treated with placebo or RTX cream. C. Densitometric ratio between TRPV1 expression and the respective loading control, GAPDH (n = 4 experiments). HEK293 cells and TRPV1 stably expressing HEK293 cells (HEK^TRPV1) were used as negative and positive controls for TRPV1 expression. ** represents statistical significance P < 0.01 for n = 3 /condition)

Then to determine the extent of TRPV1 protein expression, we used Western Blot technique, and quantified the fold change in TRPV1 expression, which shows a decreased TRPV1 (0.067 ± 0.01) expression following treatment with RTX cream compared to placebo treated control (0.63 ± 0.012) (Fig. 7 B and C). P < 0.01

In order to further confirm the expression of TRPV1, we analyzed the skin biopsies of placebo or RTX cream-treated control and diabetic mini-pigs by immunohistochemistry. We determined the expression of TRPV1 in skin biopsies obtained from the regions treated with placebo or RTX cream after completing the study. We stained for both PGP 9.5, a marker for epidermal nociceptive fibers and TRPV1. It is clear from the staining that PGP 9.5 and TRPV1 are co-localized to a degree (Fig. 8A–H). In diabetic hyperalgesic animals there was an increase in staining of PGP 9.5 and TRPV1 as compared to control (Fig. 8A, E, C, G). Following treatment with RTX cream (10 nM RTX) both in control (Fig. 8B, F) as well as in diabetic hyperalgesic animals (Fig. 8D, H), TRPV1 labeling was significantly decreased, suggesting that enhanced TRPV1 levels are responsible for painful behavior and RTX treatment reduced TRPV1 expression and alleviated painful behavior (n=4 separate experiments)

Figure 8: Immunohistochemical staining for TRPV1 in control and diabetic mini-pig skin biopsies.

Immunohistochemical staining of PGP 9.5 (yellow) and TRPV1 (red) in skin biopsy samples collected from control (A, B, E, F) and diabetic (C, D, G, H) mini-pigs. In control animals, RTX cream decreased PGP 9.5 and TRPV1 expression (B, F). In diabetic animals, there is increased expression of PGP 9.5 and TRPV1 (C, G). Treatment with RTX cream decreased the expression of PGP 9.5 and TRPV1 (D, H). Scale bar 50 μm.

In order to determine the long-term stability and effectiveness of the cream formulation, we determined the shelf life of RTX-cream over a period of 90 days. During this period, we did not observe any change in color or odor, when the formulation was stored at 50 °C with relative humidity (RH) of 70 ± 5%. Also, neither the content of RTX in the cream formulation nor its efficacy were altered during this 90-day stability analysis. To analyze the efficacy of the cream formulation, we performed in vivo experiments in diabetic rats. Application of RTX cream stored at 50 °C with RH of 70 ± 5% for 90 days, exhibited the same effectiveness in suppressing the thermal hyperalgesia in the diabetic rats (Supplementary Fig. 3).

Discussion

Pain associated with DPN is a common painful neuropathic condition that affects people with chronic diabetes mellitus. Currently, the therapeutic options are largely dependent on centrally acting drugs that include opioids and nonsteroidal anti-inflammatory drugs (NSAIDs). These oral medications are effective in less than 50% of the patients and have significant side effects that are intolerable to many patients. The chronic treatment with opioids causes addiction and use dependence. On the other hand, NSAIDs are useful with limitations due to ineffectiveness and liver and kidney toxicities. Thus, there is a need for non-opioid novel treatment options to manage PDPN.

Published literature suggests that one of the promising peripheral targets for treating PDPN is TRPV1 ion channel. TRPV1 is implicated in painful condition and strategies to suppress TRPV1 activity is effective to treat pain associated with DPN. TRPV1 receptor is involved in inflammatory thermal hyperalgesia and plays a significant role in PDPN. TRPV1 is expressed in C and Aδ nociceptors, which are heat and mechano-sensitive. These fibers are peptidergic able to release CGRP and distributed in the superficial epidermal layers of the skin. A set of C fibers that is known as silent or mechanoinsensitive C (CMi) fibers are also peptidergic and express TRPV1 and distributed deep in the dermis. In several neuropathies CMi fibers are spontaneously active^36,47,48. The ratio of mechano-sensitive to CMi nociceptors was ∼2:1 in the healthy controls, whereas in painful DPN patients, it was 1:2³⁶. Therefore, targeting TRPV1 expressed in the CMi fibers is a useful strategy. Since CMi fibers are located deeper, higher concentrations of topical agents are needed to access the CMi fibers. It is also possible to develop highly penetrating formulation of TRPV1 agonists or antagonists. Capsaicin, a TRPV1 agonist, topically applied relieves pain and improves sensory perception in humans with DPN^3,17. Similarly, TRPV1 antagonists could be effective in treating painful conditions. Unfortunately, in clinical trials, it became apparent that TRPV1 antagonists, although showed excellent analgesia, but caused sustained hyperthermia^18–21. Studies have shown using a transgenic mouse model of diabetes and STZ-induced diabetes that TRPV1 expression and function are increased and decreased in hyper or hypoalgesic phenotypes, respectively^8,38,39. Ultrapotent TRPV1 agonist resiniferatoxin (RTX) fully activates TRPV1 receptor even at very low concentrations (femto to picomolar range). Ultra-potency of RTX has been demonstrated by recording TRPV1-mediated membrane currents using patch-clamp techniques^26,42. In current-clamp experiments, RTX blocked action potential generation by inducing “depolarization block” (causing ramp-like change in membrane potential and preventing coordinated activation of sodium channels to generate an action potential)⁴². Very low concentrations of RTX induce the desired pharmacological effect by sustained activation of TRPV1 in the short-term and reversible degeneration/depletion of nerve terminals in the longer-term resulting in long-lasting pain relief^8.26. Intrathecal or intraplantar administration of RTX caused TRPV1 expressing central and peripheral nerve terminal depletion/degeneration, respectively and the peripheral nerve terminals regenerated over period of two months^8,26. The suggested mechanisms of RTX actions are depolarization block, desensitization of TRPV1 and/or depletion of TRPV1 expressing nerve terminals. ^25,42–44. Depletion of epidermal nerve fibers may not be a desirable outcome. We also suggest that spontaneously active and TRPV1 expressing CMi fibers, which are implicated in painful neuropathies could be affected by deep penetrating RTX cream formulation.

Although, topical formulations are ideal to treat chronic pain, they are not very effective because they do not penetrate to affect the CMi fibers that are located deeper in the skin. To overcome these hurdles, high concentration of capsaicin, a TRPV1 agonist is applied peripherally to exert the desired pharmacological effect. Eight (8) % capsaicin containing patch (Qutenza) has been approved to treat PDPN in Europe since 2015 and recently approved by FDA in US. Since activation of TRPV1 depolarizes the nerve terminal and generates an action potential that carries the pain, lidocaine, a local anesthetic is applied to numb the area prior to application of capsaicin patch to suppress pain. Therefore, the use of 8% capsaicin to treat painful conditions, which is both cumbersome and painful during application. The use of TRPV1 inhibitors is associated with undesired hyperthermia in humans, therefore, this strategy has not advanced to further clinical trials. Nonetheless the development of novel inhibitors, which do not cause dysthermia is ongoing, until now no new molecules are available to treat chronic pain or have shown efficacy against PDPN.

RTX is a resinous compound and it has poor penetrability across the skin. In this study, to enhance the transcutaneous permeation of RTX, we have prepared a low concentration sustained release RTX nanoparticle cream formulation, which has more than 300 times higher permeability across the human skin mimetic (Strata-M membranes; Fig. 2). Since PLGA is a US FDA approved polymer, its ability to sustain the release of RTX causes a long-term depolarization block induced suppression of TRPV1 activity and thereby alleviating pain in diabetic animals.

We have confirmed its effectiveness of RTX cream in a small (rat) and large (mini-pig) models of PDPN. This nanoparticle cream formulation of RTX enhances the penetrability of RTX, which would have a significant advantage of not inducing any of the adverse effects of capsaicin transdermal patch (Qutenza). Thus, RTX-cream will induce the desired effect and will be cost effective without the need to combine with lidocaine. Published work from our laboratories indicates that nerve terminal depolarization block in the short-term prevents pain during application and nerve terminal depletion in the long-term results in long lasting pain relief^26,42.

Experimental results obtained using Hargreaves and von Frey tests in rats and DLss technique in mini-pigs present the novel findings that RTX cream formulation decreased thermal pain hypersensitivity in the diabetic rats and mini-pigs, without causing any discomfort to the animals. Also, in diabetic animals, TRPV1 protein level was increased in the paw skin preparations and RTX cream application prevented/reversed this increase suggesting that TRPV1 plays a vital role in mediating pain associated with DPN and suppressing TRPV1 activity by RTX is an effective treatment strategy. The sustained release of RTX by PLGA coating could contribute to minimizing the burning effect of TRPV1 agonists. These findings suggest that RTX cream formulation is a viable strategy to suppress pain associated with DPN. Also, our data provide compelling evidence for developing RTX topical formulation as an effective replacement for the use of 8% capsaicin to treat painful conditions, which is both cumbersome to apply, painful during application and induces erythema and pruritis²³.

The preclinical efficacy of RTX nanoparticle cream formulation suggests that it is well tolerated without causing pain during application and any other local adverse effects. Our stability testing shows that RTX cream is stable without loss of its efficacy or physical properties. However, long-term safety, diffusion kinetics, tissue availability, and retention should be studied. Nonetheless, the research presented here demonstrate the effectiveness of a sustained release cream formulation of RTX to alleviate pain associated with DPN. Further studies are planned to validate its safety profiles and advance its clinical use to treat pain associated with DPN in humans.

Data

All of the data pertaining to this study are available in the laboratories of Dr. Thyagarajan and Dr. Premkumar.