BACKGROUND AND PURPOSE:Prolonged exposure to Gαi/o-linked receptor agonists such as opioids can lead to a sensitization of adenylyl cyclases (ACs), resulting in heterologous sensitization or cyclic AMP (cAMP) overshoot. The molecular consequences of cAMP overshoot are not well understood, but this adaptive response is suggested to play a critical role in the development of opioid tolerance and withdrawal. EXPERIMENTAL APPROACH:Genetic reduction of AC1 and simultaneous upregulation of ATP-sensitive potassium channel (KATP) subunits, SUR1 or Kir6.2 were performed using viral vectors in mice. In vitro models utilizing an EPAC2-GFP-cAMP biosensor investigated sensitization of AC in SH-SY5Y neuroblastoma cells and HEK-ACΔ3/6 knockout cells. KEY RESULTS:Reduction of AC1 and upregulation of KATP channels significantly attenuated morphine tolerance and reduced precipitated withdrawal. Acute application of DAMGO decreased the cAMP signal from the EPAC2-GFP-cAMP biosensor, while chronic DAMGO administration resulted in enhanced cAMP production. Inhibition of cAMP overshoot was observed with naloxone (NAL) or pertussis toxin (PTX), as well as co-expression of β-adrenergic receptor kinase C-terminus (βARK-CT). Inhibition of AC1 or exchange protein directly activated by cAMP (EPAC) enhanced potassium channel activity after chronic morphine treatment in a thallium-based assay in SH-SY5Y cells and mouse dorsal root ganglia (DRG) after chronic morphine treatment. CONCLUSION AND IMPLICATIONS:This study presents evidence for investigating further AC1 signalling as a target for opioid tolerance and withdrawal, by increasing EPAC activity and affecting potassium channels downstream of opioid receptors.

Adenylyl cyclase 1 (AC1) plays an integral role in the excitatory signaling in the anterior cingulate cortex underlying chronic pain pathophysiology. Upon chronic nociceptive input, sustained Ca2+/calmodulin (CaM)-stimulated AC1 activity leads to the development of chronic pain. In this study, we characterize our recently reported series of pyrazolopyrimidinone amine analogs that selectively inhibit Ca2+/CaM-stimulated AC1 activity. Lead compounds AC10136A, AC10142A, and AC10172A exhibited potent AC1 inhibition (IC50 = 140-290 nM) and complete selectivity over AC2, AC5, and AC8. All compounds showed high therapeutic indices (>300) and minimal cytotoxicity up to 100 μM. Lead compounds displayed no functional agonism at κ-opioid receptor. Further, compounds prevented and reversed μ-opioid receptor-mediated heterologous sensitization of AC1, highlighting potential utility in mitigating opioid-induced dependence. Further, chronic treatment with lead compounds failed to induce opioid-like adaptations, such as heterologous sensitization, and did not modulate expression of AC1. Mechanistic studies revealed that inhibition is Ca2+/CaM dependent but independent of Gαs, suggesting preferential targeting of the active AC1 conformation. In mice, AC10142A demonstrated analgesic efficacy in the complete Freund adjuvant model of chronic inflammatory pain. Acute and repeated dosing at 48-120 h after complete Freund adjuvant treatment increased mechanical paw withdrawal thresholds and sustained analgesia over this repeated dosing. By targeting Ca2+/CaM-bound AC1, these inhibitors preferentially inhibit increased AC1 activity to its normal activity, demonstrating a state-dependent mechanism ideal for chronic pain treatment. Together, these findings establish AC10136A, AC10142A, and AC10172A as potent, selective, and well-tolerated AC1 inhibitors, with promising therapeutic potential for chronic pain and potentially opioid use disorder. SIGNIFICANCE STATEMENT: Adenylyl cyclase 1 (AC1) plays a key role in chronic pain sensitization and opioid use disorder, making it a promising therapeutic target. Selective pyrazolopyrimidinone amine inhibitors of AC1 potently block Ca2+/calmodulin-dependent activation (with IC50 values of 140-290 nM), exhibit excellent in vitro safety profiles, prevent μ-opioid receptor-mediated AC1 sensitization in vitro, and demonstrate robust antiallodynic effects in vivo, positioning this compound series as a compelling nonopioid strategy for managing chronic pain and potentially addressing opioid use disorder.

Prolonged exposure to Gαi/o receptor agonists such as opioids can lead to a sensitization of adenylyl cyclases (ACs), resulting in heterologous sensitization or cyclic AMP (cAMP) overshoot. The molecular consequences of cAMP overshoot are not well understood, but this adaptive response is suggested to play a critical role in the development of opioid tolerance and withdrawal. We found that genetic reduction of AC1 and simultaneous upregulation of ATP-sensitive potassium (KATP) channel subunits, SUR1 or Kir6.2, significantly attenuated morphine tolerance and reduced naloxone-precipitated withdrawal. In vitro models utilized an EPAC2-GFP-cAMP biosensor to investigate sensitization of adenylyl cyclase in SH-SY5Y neuroblastoma cells and HEKΔAC3/6 knockout cells. Acute application of DAMGO significantly decreased the cAMP signal from the EPAC2-GFP-cAMP biosensor, while chronic DAMGO administration resulted in enhanced cAMP production following AC stimulation. Inhibition of cAMP overshoot was observed with naloxone (NAL), pertussis toxin (PTX), and the neddylation inhibitor, MLN4924 (Pevonedistat), as well as co-expression of β-adrenergic receptor kinase C-terminus (β-ARKCT). After establishment of the AC1-EPAC sensitization in the in vitro models, we found that inhibition of AC1 or EPAC enhanced potassium channel activity after chronic morphine treatment, using a thallium-based assay in SH-SY5Y cells. Similar data were obtained in mouse dorsal root ganglia (DRG) after chronic morphine treatment. This study presents evidence for investigating further AC1 signaling as a target for opioid tolerance and withdrawal, by increasing EPAC activity and affecting potassium channels downstream of opioid receptors.

Abstract During diabetes progression, β-cell dysfunction due to loss of potassium channels sensitive to ATP, known as KATP channels, occurs contributing to hyperglycemia. The aim of this study is to investigate if KATP channel expression or activity in the nervous system was altered in a high-fat-diet-(HFD) fed mouse model of diet-induced obesity. Expression of two KATP channel subunits, Kcnj11 (Kir6.2) and Abcc8 (SUR1), were decreased in the peripheral and central nervous system in HFD mice, which is significantly correlated with mechanical paw withdrawal thresholds. HFD mice had decreased antinociception to systemic morphine compared to control diet (CON) mice, which was expected as KATP channels are downstream targets of opioid receptors. Mechanical hypersensitivity in HFD mice was exacerbated after systemic treatment with glyburide or nateglinide, KATP channel antagonists clinically used to control blood glucose levels. Upregulation of SUR1 and Kir6.2, through an adenovirus delivered intrathecally, increased morphine antinociception in HFD mice,. These data present a potential link between KATP channel function and neuropathy during early stages of diabetes. There is a need for increased knowledge in how diabetes affects structural and molecular changes in the nervous system, including ion channels, to lead to the progression of chronic pain and sensory issues.

Abstract ID 96171Poster Board 524Adenylyl cyclases mediate the production of cyclic AMP (cAMP) and plays a pivotal role in regulating important physiological processes. Adenylyl cyclase 1 (AC1) is robustly activated by Ca2+/calmodulin (CaM), highly expressed in the central nervous system, and has been implicated in chronic pain modulation, alcohol behaviors, and memory and learning processes. However, the absence of a robust neuronal cell model along with technological limits for measuring cAMP have posed challenges in studying AC1, and potential therapeutic interventions. Specifically, most cell-based methods lack essential neuronal properties necessary for mimicking native cellular conditions, leading to limitations in assessing endogenous AC activity and inhibitor responses in a neuronal model. To address this gap, we aimed to establish a neuronal model utilizing SH-SY5Y neuroblastoma cells expressing genetically encoded cAMP biosensors to study endogenous adenyl cyclase activity and evaluate potential inhibitors of AC1 signaling. The initial studies used our novel HEK AC3/6 KO cells stably expressing AC1 (ACΔ3/6KO-AC1) to assess cAMP signaling with several dynamic cAMP biosensors (cAADis). Progress to date includes significant advancements in successful expression of all cAADis sensors in HEK ACΔ3/6KO-AC1 cells and subsequent stimulation of AC activity by forskolin, calcium ionophore (A23187), isoproterenol, and capacitive calcium entry. Both a Neo2 plate reader and Cytation 3-based imaging were utilized for fluorescent measures across different plate configurations, including 96-well and 384-well formats. Studies with constitutively active Gs and Gi-linked receptors (u-opioid and D2 dopamine receptors) were also conducted. Additionally, cAMP overshoot or heterologous sensitization experiments using cAADis cAMP biosensors were successfully executed in the ACΔ3/6KO-AC1 cells. A second series of experiments used the best performing cAADis cAMP biosensors in SH-SY5Y neuroblastoma cells to investigate endogenous AC1 activity and explore potential inhibitors of AC1. SH-SY5Y cells express high levels of AC1 mRNA, however, significant Ca2+/CaM-stimulated cAMP accumulation is not readily observed. We hypothesized that cell-selective analysis of cAMP using the targeted cAADis biosensors in SH-SY5Y cells would improve our overall signal to noise window of AC1 activity. Stimulation of SH- SY5Y cells with calcium ionophore, A23187 revealed a significant increase in the endogenous cAMP response of approximately 20% versus the maximal forskolin response. This study presents promising advancements towards establishing a robust neuronal model expressing genetically encoded cAMP biosensors, to study endogenous AC1 activity and explore potential AC inhibitor responses pertinent to chronic pain treatment.Keywords: Adenyl Cyclase; cAMP Signaling; cAMP Biosensors; Chronic Pain; Neuronal Models; HEK 293 Cells; SH-SY5Y Cells; Calcium Ionophore.Funding: This work is supported by Purdue University, and NIH grant numbers R01DA051876 and R01NS119917.

During diabetes, β-cell dysfunction due to loss of potassium channels sensitive to ATP, known as KATP channels occurs progressively over time contributing to hyperglycemia. KATP channels are additionally present in the central and peripheral nervous systems and are downstream targets of opioid receptor signaling. The aim of this study is to investigate if KATP channel expression or activity in the nervous system changes in diabetic mice and if morphine antinociception changes in mice fed a high fat diet (HFD) for 16 weeks compared to controls. Mechanical thresholds were also monitored before and after administration of glyburide or nateglinide, KATP channel antagonists, for four weeks. HFD mice have decreased antinociception to systemic morphine, which is exacerbated after systemic treatment with glyburide or nateglinide. HFD mice also have lower rotarod scores, decreased mobility in an open field test, and lower burrowing behavior compared to their control diet counterparts, which is unaffected by KATP channel antagonist delivery. Expression of KATP channel subunits, Kcnj11 (Kir6.2) and Abcc8 (SUR1), were decreased in the peripheral and central nervous system in HFD mice, which is significantly correlated with baseline paw withdrawal thresholds. Upregulation of SUR1 through an adenovirus delivered intrathecally increased morphine antinociception in HFD mice, whereas Kir6.2 upregulation improved morphine antinociception only marginally. Perspective: This article presents the potential link between KATP channel function and neuropathy during diabetes. There is a need for increased knowledge in how diabetes affects structural and molecular changes in the nervous system to lead to the progression of chronic pain and sensory issues.

Previous studies show ATP-sensitive potassium (KATP) channel openers can reduce hypersensitivity associated with chronic pain models in rodents, and reduce morphine tolerance. Many agonists of KATP channels are not soluble in physiologically relevant vehicles, requiring adaptation for clinical use. This study compared the antinociceptive activity of novel KATP channel targeting prodrugs, CKLP1, CKLP2, and CF3-CKLP. These prodrugs are activated by endogenous alkaline phosphatase enzymes present in the peripheral and central nervous systems. Analgesic capabilities of intrathecally injected prodrugs were tested in rodent models of spinal nerve ligation (SNL) and complete Freund's adjuvant (CFA) as models for neuropathic and inflammatory pain, respectively. CKLP1 and CKLP2 significantly increased mechanical paw withdrawal thresholds 1-2 hours after intrathecal administration in the SNL model, but all three prodrugs were able to attenuate hypersensitivity up to 7 days after CFA treatment. The reduction of opioid tolerance and opioid-induced hypersensitivity in mice treated chronically with morphine was significantly reduced in CKLP1 and CKLP2 treated animals. Prodrug cleavage was confirmed in mouse spinal cords using liquid chromatography. These studies may aid in the further development of KATP channel prodrugs for use in treatments of chronic pain, opioid tolerance, and withdrawal. SIGNIFICANCE STATEMENT: The cromakalim prodrugs, CKLP1, CKLP2, and CF3-CKLP1 reduced hypersensitivity in inflammatory and neuropathic pain models in male and female mice. CKLP1 and CKLP2 also reduced morphine-induced hypersensitivity in a mouse model of chronic morphine exposure. CKLP2 reduced jumping and rearing behaviors after naloxone-induced precipitated morphine withdrawal. Taken together, CKLP2 demonstrates the potential for development as a non-opioid analgesic drug.

Opioid tolerance, opioid-induced hyperalgesia during repeated opioid administration, and chronic pain are associated with upregulation of adenylyl cyclase activity. The objective of this study was to test the hypothesis that a reduction in adenylyl cyclase 1 (AC1) activity or expression would attenuate morphine tolerance and hypersensitivity, and inflammatory pain using murine models. To investigate opioid tolerance and opioid-induced hyperalgesia, mice were subjected to twice daily treatments of saline or morphine using either a static (15 mg/kg, 5 days) or an escalating tolerance paradigm (10–40 mg/kg, 4 days). Systemic treatment with an AC1 inhibitor, ST03437 (2.5–10 mg/kg, IP), reduced morphine-induced hyperalgesia in mice. Lumbar intrathecal administration of a viral vector incorporating a short-hairpin RNA targeting Adcy1 reduced morphine-induced hypersensitivity compared to control mice. In contrast, acute morphine antinociception, along with thermal paw withdrawal latencies, motor performance, exploration in an open field test, and burrowing behaviors were not affected by intrathecal Adcy1 knockdown. Knockdown of Adcy1 by intrathecal injection also decreased inflammatory mechanical hyperalgesia and increased burrowing and nesting activity after intraplantar administration of Complete Freund’s Adjuvant (CFA) one-week post-injection.

Over 50% of individuals with type 2 diabetes eventually develop diabetic neuropathy. There is a need for increased knowledge in how diabetes affects structural and molecular changes in the nervous system, and how current diabetes interventions affect the progression of chronic pain. Loss of KATP channel activity or function could cause nerve fiber hypersensitivity and consequently increased pain sensitivity due to neuron depolarization. Diabetic neuropathy was induced in male and female C57Bl6 mice over a total of 16 weeks by feeding the animals a high-fat diet (HFD) or corresponding control diet (12330 and 12328, Research Diets Inc., New Brunswick, NJ) for over 16 weeks. Mice on the HFD for at longer than 8 weeks have significantly lower mechanical thresholds and compared to control diet mice. Decreased expression of KATP channel subunits in the spinal cord, and dorsal root ganglia were correlated with decreased mechanical paw withdrawal thresholds. Mice fed a HFD also have decreased analgesia to systemic morphine (0-20 mg/kg, s.c.), which is exacerbated after systemic treatment with glyburide or nateglinide for four weeks, (KATP channel antagonist, glyburide or nateglinide, 50 mg/kg/day, IP or vehicle, 5% DMSO + 0.5% Tween). Upregulation of KATP channels using a viral vector strategy increased paw withdrawal thresholds to systemic morphine. Diabetes affects many adults including those with prediabetes and new-onset diabetes. It is clear that novel biomarkers and pharmaceutical targets are needed in order to (1) confront the demand for new treatments that better manage and (2) identify patients that may develop diabetic neuropathy. Further investigation of KATP channel expression and function during chronic pain syndromes, including diabetic neuropathy, may help to find sufficient treatment options for patients. This work was supported through a University of Minnesota Academic Health Center Faculty Development Grant to AHK and MLG. Funding also provided by the NIH: K01 DA042902, R01 DA051876, and UL1TR002494 (AHK). Over 50% of individuals with type 2 diabetes eventually develop diabetic neuropathy. There is a need for increased knowledge in how diabetes affects structural and molecular changes in the nervous system, and how current diabetes interventions affect the progression of chronic pain. Loss of KATP channel activity or function could cause nerve fiber hypersensitivity and consequently increased pain sensitivity due to neuron depolarization. Diabetic neuropathy was induced in male and female C57Bl6 mice over a total of 16 weeks by feeding the animals a high-fat diet (HFD) or corresponding control diet (12330 and 12328, Research Diets Inc., New Brunswick, NJ) for over 16 weeks. Mice on the HFD for at longer than 8 weeks have significantly lower mechanical thresholds and compared to control diet mice. Decreased expression of KATP channel subunits in the spinal cord, and dorsal root ganglia were correlated with decreased mechanical paw withdrawal thresholds. Mice fed a HFD also have decreased analgesia to systemic morphine (0-20 mg/kg, s.c.), which is exacerbated after systemic treatment with glyburide or nateglinide for four weeks, (KATP channel antagonist, glyburide or nateglinide, 50 mg/kg/day, IP or vehicle, 5% DMSO + 0.5% Tween). Upregulation of KATP channels using a viral vector strategy increased paw withdrawal thresholds to systemic morphine. Diabetes affects many adults including those with prediabetes and new-onset diabetes. It is clear that novel biomarkers and pharmaceutical targets are needed in order to (1) confront the demand for new treatments that better manage and (2) identify patients that may develop diabetic neuropathy. Further investigation of KATP channel expression and function during chronic pain syndromes, including diabetic neuropathy, may help to find sufficient treatment options for patients. This work was supported through a University of Minnesota Academic Health Center Faculty Development Grant to AHK and MLG. Funding also provided by the NIH: K01 DA042902, R01 DA051876, and UL1TR002494 (AHK).

Opioid signaling can occur through several downstream mediators and influence analgesia as well as reward mechanisms in the nervous system. KATP channels are downstream targets of the μ opioid receptor and contribute to morphine-induced antinociception. The aim of the present work was to assess the role of SUR1-subtype KATP channels in antinociception and hyperlocomotion of synthetic and semi-synthetic opioids. Adult male and female mice wild-type (WT) and SUR1 deficient (KO) mice were assessed for mechanical and thermal antinociception after administration of either buprenorphine, fentanyl, or DAMGO. Potassium flux was assessed in the dorsal root ganglia and superficial dorsal horn cells in WT and KO mice. Hyperlocomotion was also assessed in WT and KO animals after buprenorphine, fentanyl, or DAMGO administration. SUR1 KO mice had attenuated mechanical antinociception after systemic administration of buprenorphine, fentanyl, and DAMGO. Potassium flux was also attenuated in the dorsal root ganglia and spinal cord dorsal horn cells after acute administration of buprenorphine and fentanyl. Hyperlocomotion after administration of morphine and buprenorphine was potentiated in SUR1 KO mice, but was not seen after administration of fentanyl or DAMGO. These results suggest SUR1-subtype KATP channels mediate the antinociceptive response of several classes of opioids (alkaloid and synthetic/semi-synthetic), but may not contribute to the “drug-seeking” behaviors of all classes of opioids.

Research presented here sought to determine if opioid induced tolerance is linked to activity changes within the PI3Kγ-AKT-cGMP-JNK intracellular signaling pathway in spinal cord or peripheral nervous systems. Morphine or saline injections were given subcutaneously twice a day for five days (15 mg/kg) to male C57Bl/6 mice. A separate cohort of mice received spinal nerve ligation (SNL) one week prior to the start of morphine tolerance. Afterwards, spinal cord, dorsal root ganglia, and sciatic nerves were isolated for quantifying total and phosphorylated- JNK levels, cGMP, and gene expression analysis of Pik3cg, Akt1, Pten, and nNos1. This pathway was downregulated in the spinal cord with increased expression in the sciatic nerve of morphine tolerant and morphine tolerant mice after SNL. We also observed a significant increase in phosphorylated- JNK levels in the sciatic nerve of morphine tolerant mice with SNL. Pharmacological inhibition of PI3K or JNK, using thalidomide, quercetin, or SP600125, attenuated the development of morphine tolerance in mice with SNL as measured by thermal paw withdrawal. Overall, the PI3K/AKT intracellular signaling pathway is a potential target for reducing the development of morphine tolerance in the peripheral nervous system. Continued research into this pathway will contribute to the development of new analgesic drug therapies.

ABSTRACT IMPACT: Pharmacological activation of KATP channels may provide analgesia and attenuate opioid tolerance and withdrawal OBJECTIVES/GOALS: Our long term goal is to develop therapeutics for the treatment of the overuse of opioids. The objective of this application is to test novel KATP channel-targeting prodrugs in rodent models of neuropathic and inflammatory pain in addition to opioid tolerance after chronic morphine administration. METHODS/STUDY POPULATION: In one study, two different measures for chronic pain were implemented in mice. Male and female mice (n=10) were subjected to spinal nerve ligation (SNL) or intraplantar injection of Complete Freund’s Adjuvant (CFA) to induce neuropathic and inflammatory pain, respectively. Administration of KATP channel prodrugs (60ug, it) attenuated mechanical hypersensitivity after SNL or CFA compared to vehicle (saline). In a separate study, changes in mechanical hypersensitivity were tested while mice undergo chronic morphine treatment (15mg/kg, 2x, 5 days) with administration of the prodrugs. Tolerance was measured as the loss of antinociception, and withdrawal is measured ˜24 hours after the final morphine injection. RESULTS/ANTICIPATED RESULTS: Intrathecal administration of either KATP channel prodrugs significantly attenuated mechanical hypersensitivity after SNL and significantly attenuated mechanical hypersensitivity after CFA in mice. We predict that intrathecal administration of these prodrugs will also attenuate morphine tolerance and withdrawal in mice. This hypothesis is based off our previous data indicating non-water soluble KATP channel agonists produce analgesia and attenuate morphine tolerance in mice. DISCUSSION/SIGNIFICANCE OF FINDINGS: Pharmaceutical strategies to utilize KATP channels for therapeutics have been hindered due to the low solubility and low ability to cross the neurovascular unit. Newly developed, water-soluble KATP channel openers could be useful pharmaceutical strategy to reduce chronic pain, opioid tolerance, and withdrawal in human populations.

Opioid tolerance and opioid induced hyperalgesia during repeated opioid administration and chronic pain are associated with upregulation of adenylyl cyclase activity. Therefore, we investigated if expression of components of the cyclic-AMP pathway were increased within in the nervous system tissues of morphine tolerant mice, and if inhibiting any of these changes could alleviate pain and/or tolerance in mouse models. The objective of this study was to test the hypothesis that a reduction in adenylyl cyclase 1 (AC1) activity would attenuate morphine tolerance and hypersensitivity, and inflammatory pain using murine models. Short-hairpin RNA (shRNA) gene knockdown of Adcy1 in the spinal cord and dorsal root ganglia was accomplished using a lumbar injection of an associated adenovirus viral vector (AAV9-GFP-U6-m-Adcy1-shRNA) and negative controls (AAV9-GFP-U6-m-scrambl-shRNA). Behavioral testing such as open field testing, rotarod testing, burrowing, thermal and mechanical paw withdrawal latencies were tested after injection. Morphine tolerance (15mg/kg, sc, 5 days and 10-40mg/kg escalation over 4 days) and opioid-induced hypersenstivity were also assessed after inoculation. Lumbar intrathecal administration of a vector incorporating adeno-associated virus and short-hairpin RNA against Adcy1 did not affect baseline parameters such as open field testing, rotarod testing, and burrowing testing. Morphine tolerance and withdrawal were attenuated in Adcy1 shRNA mice compared to control vector mice. Chronic exposure to morphine leads to changes in mRNA expression of proteins involved in the cAMP signaling pathway in different areas of the nervous system. Gene knockdown of Adcy1 decreases morphine tolerance and opioid-induced hypersensitivity, which could form the basis for novel therapeutics in the future. This work was supported through K01 DA042902 to AHK and the Purdue University College of Pharmacy to VJW. Opioid tolerance and opioid induced hyperalgesia during repeated opioid administration and chronic pain are associated with upregulation of adenylyl cyclase activity. Therefore, we investigated if expression of components of the cyclic-AMP pathway were increased within in the nervous system tissues of morphine tolerant mice, and if inhibiting any of these changes could alleviate pain and/or tolerance in mouse models. The objective of this study was to test the hypothesis that a reduction in adenylyl cyclase 1 (AC1) activity would attenuate morphine tolerance and hypersensitivity, and inflammatory pain using murine models. Short-hairpin RNA (shRNA) gene knockdown of Adcy1 in the spinal cord and dorsal root ganglia was accomplished using a lumbar injection of an associated adenovirus viral vector (AAV9-GFP-U6-m-Adcy1-shRNA) and negative controls (AAV9-GFP-U6-m-scrambl-shRNA). Behavioral testing such as open field testing, rotarod testing, burrowing, thermal and mechanical paw withdrawal latencies were tested after injection. Morphine tolerance (15mg/kg, sc, 5 days and 10-40mg/kg escalation over 4 days) and opioid-induced hypersenstivity were also assessed after inoculation. Lumbar intrathecal administration of a vector incorporating adeno-associated virus and short-hairpin RNA against Adcy1 did not affect baseline parameters such as open field testing, rotarod testing, and burrowing testing. Morphine tolerance and withdrawal were attenuated in Adcy1 shRNA mice compared to control vector mice. Chronic exposure to morphine leads to changes in mRNA expression of proteins involved in the cAMP signaling pathway in different areas of the nervous system. Gene knockdown of Adcy1 decreases morphine tolerance and opioid-induced hypersensitivity, which could form the basis for novel therapeutics in the future. This work was supported through K01 DA042902 to AHK and the Purdue University College of Pharmacy to VJW.

In humans, intradermal administration of β-alanine (ALA) and bovine adrenal medulla peptide 8–22 (BAM8-22) evokes the sensation of itch. Currently, it is unknown which human dorsal root ganglion (DRG) neurons express the receptors of these pruritogens, MRGPRD and MRGPRX1, respectively, and which cutaneous afferents these pruritogens activate in primate. In situ hybridization studies revealed that MRGPRD and MRGPRX1 are co-expressed in a subpopulation of TRPV1+ human DRG neurons. In electrophysiological recordings in nonhuman primates (Macaca nemestrina), subtypes of polymodal C-fiber nociceptors are preferentially activated by ALA and BAM8-22, with significant overlap. When pruritogens ALA, BAM8-22, and histamine, which activate different subclasses of C-fiber afferents, are administered in combination, human volunteers report itch and nociceptive sensations similar to those induced by a single pruritogen. Our results provide evidence for differences in pruriceptive processing between primates and rodents, and do not support the spatial contrast theory of coding of itch and pain.

Article Figures and data Abstract Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract In humans, intradermal administration of β-alanine (ALA) and bovine adrenal medulla peptide 8–22 (BAM8-22) evokes the sensation of itch. Currently, it is unknown which human dorsal root ganglion (DRG) neurons express the receptors of these pruritogens, MRGPRD and MRGPRX1, respectively, and which cutaneous afferents these pruritogens activate in primate. In situ hybridization studies revealed that MRGPRD and MRGPRX1 are co-expressed in a subpopulation of TRPV1+ human DRG neurons. In electrophysiological recordings in nonhuman primates (Macaca nemestrina), subtypes of polymodal C-fiber nociceptors are preferentially activated by ALA and BAM8-22, with significant overlap. When pruritogens ALA, BAM8-22, and histamine, which activate different subclasses of C-fiber afferents, are administered in combination, human volunteers report itch and nociceptive sensations similar to those induced by a single pruritogen. Our results provide evidence for differences in pruriceptive processing between primates and rodents, and do not support the spatial contrast theory of coding of itch and pain. Introduction The sensations of itch and pain serve a similar purpose, which is to alert the organism of potentially harmful external threats. Although the sensations of itch and pain can elicit different behavioral responses (such as scratching vs. withdrawal, rubbing or guarding), they are closely linked as they appear to be elicited by activity in one and the same type of fiber or functionally similar primary afferents. In mice, pruritogens activate at least three populations of neurons (Liu et al., 2012; Liu et al., 2009; Solinski et al., 2019b). Among these, two neuronal populations express Mas-related G-protein-coupled receptors (Mrgpr), namely Mrgprc and Mrgprd, in a non-overlapping fashion (Zylka et al., 2003), and intradermal administration of their respective agonist, bovine adrenal medulla peptide 8–22 (BAM8-22) and β-alanine (ALA), results in scratching behavior. The third population expresses natriuretic polypeptide b (Nppb) and mediates mast cell-induced itch. Neurons expressing Mrgprc or Nppb also express the histamine (HIS) receptor Hrh1 and likely mediate scratching behavior induced by HIS. In humans, HIS and the nonhistaminergic pruritogens ALA and BAM8-22 each elicit itch and nociceptive sensations (Liu et al., 2012; Sikand et al., 2011a; Sikand et al., 2009; Sikand et al., 2011b). HIS preferentially activates mechanically insensitive C-fiber afferents (C-MIAs) in human (Schmelz et al., 1997) and nonhuman primates (Wooten et al., 2014). In nonhuman primates, ALA preferentially excites QCs, a subtype of mechanoheat-sensitive, polymodal C-fiber (CMH) nociceptor that adapts quickly to noxious heat as opposed to those that adapt slowly (SCs), and C-MIAs are unresponsive to ALA (Wooten et al., 2014). The type of afferent nerve fiber activated by BAM8-22, an agonist of MRGPRX1 in primate (Lembo et al., 2002), is currently unclear. Based on the non-overlapping expression of Mrgprd and Mrgprc in mouse dorsal root ganglion (DRG) neurons and the preferential activation of QC fibers by ALA in nonhuman primate, one might expect that in primate MRGPRD and MRGPRX1 would be similarly expressed in non-overlapping DRG neurons and that ALA and BAM8-22 would activate different neuronal afferent populations. In humans and other primates, multiple types of nociceptive primary afferents including unmyelinated and myelinated nerve fibers appear to have a role in itch sensation (Johanek et al., 2008; Johanek et al., 2007; Namer et al., 2008; Ringkamp et al., 2011; Schmelz et al., 1997; Wooten et al., 2014). Moreover, HIS and nonhistaminergic pruritogens activate distinct sets of nociceptive primary afferents, including mechano-insensitive C fibers (C-MIAs), CMHs, mechanosensitive A-fiber nociceptors (A-MSAs), and spinothalamic projection neurons (Davidson et al., 2007; Johanek et al., 2008; Namer et al., 2008; Schmelz et al., 1997; Wooten et al., 2014). The signaling of pruritic stimuli by nociceptive afferents that are also thought to encode nociceptive mechanical, thermal, and chemical stimuli (Ringkamp et al., 2013) poses a conundrum regarding the neuronal coding of these distinct sensations in the nervous system. Among the models that have been proposed to explain this conundrum (for review, see Carstens et al., 2020; LaMotte et al., 2014), the spatial contrast model posits that neuronal activity from a spatially restricted set of nociceptive afferents is interpreted as ‘itch,’ whereas activity from a spatially larger pool of afferents is interpreted as ‘pain’ (Namer et al., 2008; Namer and Reeh, 2013; Steinhoff et al., 2019). However, it has not been tested in humans whether a combination of pruritogens that activates a greater number and variety of primary afferents and thereby decreases the spatial contrast between activated and silent fibers could lead to reduced itch and/or increased nociceptive sensation. Here, we probed the relationship between the expression of receptors for ALA and BAM8-22 in postmortem human DRGs and the physiological responses of these pruritogens in single cutaneous nerve fibers of nonhuman primate. We found that human DRG neurons co-express both, MRGPRD and MRGPRX1, and also transient receptor potential channels from the vanilloid subfamily V member 1 (TRPV1). Further, injection of ALA and BAM8-22 into the receptive fields (RFs) of nociceptors in nonhuman primate revealed that QCs respond to ALA, and SCs preferentially respond to BAM8-22, with a large degree of overlap. In accompanying psychophysical studies, we studied the effects of individual pruritogens and their combinations and recorded the reported itch, nociceptive sensations, and dysesthesias in human subjects. Since we know at least three pruritogens that can preferentially activate QCs, SCs, and C-MIAs, we additionally tested the hypothesis that the co-activation of the different sets of primary afferents by ALA, BAM8-22, and HIS would enhance itch and/or nociceptive sensations and dysesthesias in humans. In addition, in primates, HIS and cowhage activate largely non-overlapping spinothalamic tract (STT) neurons (Davidson et al., 2012). Therefore, by combining ALA, BAM8-22, and HIS, we tested psychophysically whether activating both types of STT neurons would result in altered itch or nociceptive sensations. Results MRGPRX1 and MRGPRD have overlapping expression patterns in human DRG The representation of the MRGPR families of receptors in humans and nonhuman primates differs from rodents in that they contain a distinct MRGPRX subfamily of receptors that is absent in rodents and that they do not have either the rodent Mrgpra or Mrgprc subfamilies of receptors (Figure 1A; Solinski et al., 2014). Whether MRGPRX1 and MRGPRD are expressed in non-overlapping DRG neurons in primate, similar to what has been previously reported in mouse for Mrgprc and Mrgprd (Zylka et al., 2003), is currently unclear. We therefore investigated the expression of MRGPRX1 and MRGPRD in human DRG using double-label in situ hybridization (ISH). We found that MRGPRX1 and MRGPRD were co-expressed in the same neurons (Figure 1B). Specifically, on average 89.6 ± 1.5% of DRG neurons positive for MRGPRD in a given donor were also positive for MRGPRX1 (380/426 neurons from four donors combined), and 93.9 ± 2.0% (380/407) of DRG neurons positive for MRGPRX1 were also positive for MRGPRD. Previously it has been reported in a Mrgprd reporter mouse line that Mrgprd+ neurons were TRPV1 negative and non-peptidergic CMHs (Rau et al., 2009). If findings in human DRG were comparable to mouse, then one would expect to observe no overlap between TRPV1 and MRPGRD expression. However, we found that MRGPRD neurons nearly always co-expressed TRPV1 (Figure 1C) with 95.2 ± 1.6% (479/510) of MRGPRD+ neurons being also positive for TRPV1. In the mouse, Mrgprc is expressed in a subset of capsaicin-responsive neurons (Liu et al., 2009). Recent ISH data from human DRG show that all MRGPRX1+ neurons co-express NPPB, and that NPPB-expressing neurons are a subset of TRPV1+ neurons (Solinski et al., 2019a). In agreement with these previous findings, we observed that 100% (429/429) of MRGPRX1+ neurons were also positive for TRPV1 (Figure 1D). Of TRPV1+ neurons, 32.5 ± 3.7% (429/1273) and 36.5 ± 4.2% (479/1211) of cells also expressed MRGPRX1 and MRGPRD, respectively (Figure 1E). In parallel with these studies, we characterized the expression of MRGPRD and MRGPRX1 in DRG of macaques. We also found, in this primate, that these two receptors are expressed in a largely overlapping fashion, although with slight differences (Figure 1—figure supplement 1). More specifically, of MRGPRD+ neurons, 78.1 ± 6.1% (199/238) co-express MRGPRX1 while of MRGPRX1+ neurons, only 57.1 ± 11.6% (199/448) co-express MRGPRD (Supplementary file 1). Together our results point to profound species-dependent differences of MRGPR expression patterns between rodents and primates with additional slight differences inside the primate lineage. Figure 1 with 1 supplement see all Download asset Open asset MRGPRX1 and MRGPRD are co-expressed in TRPV1-expressing human dorsal root ganglion (DRG) neurons. (A) Phylogenetic tree of MRGPRs from mouse (m), rat (r), macaque (ma), and human (h). Note that the MRGPRD gene is conserved among rodents and primates, while the Mrgpra and Mrgprc subfamilies are rodent-specific and the MRGPRX subfamily is primate-specific. For clarity, only one murine Mrgpra gene and only one macaque or human MRGPRX gene is shown. (B–D) Representative double in situ hybridization (ISH) images of a field of human DRG with neurons stained for MRGPRD (B, green; C, red), MRGPRX1 (B and D, red), and TRPV1 (C and D, green). Double-positive and single-positive neurons are outlined with white and green dashed lines, respectively. DAPI counterstain is displayed in blue. (E) Venn diagram summarizing the relative expression overlap of MRGPRD (yellow), MRGPRX1 (blue), and TRPV1 (red) in human DRG. Note that MRGPRD and MRGPRX1 are expressed in a largely overlapping population (green, i.e., MRGPRD + MRGPRX1/MRGPRD = 89.6 ± 1.5%; MRGPRX1 + MRGPRD/MRGPRX1 = 93.9 ± 2.0%) in about 1/3 of all TRPV1-positive neurons (MRGPRD + TRPV1/TRPV1 = 36.5 ± 4.2%; MRGPRX1 + TRPV1/TRPV1 = 32.5 ± 3.7%). Expression analysis for all three markers was performed in DRG tissue from four individuals, and data are stated as mean ± standard error of the mean (SEM). Green filled area indicates the overlap in expression of MRGPRD and MRGPRX1. Figure 1—source data 1 Expression of MRGPRX1 and MRGPRD in nonhuman primate DRG. https://cdn.elifesciences.org/articles/64506/elife-64506-fig1-data1-v2.xlsx Download elife-64506-fig1-data1-v2.xlsx BAM8-22 preferentially activates SCs in nonhuman primates To learn more about the neurons that are activated by ALA and BAM8-22, we probed the different functional classes of neurons using teased-fiber recordings from single nerve fibers in the primate Macaca nemestrina (Wooten et al., 2014). Across all CMHs tested, the average number of action potentials (APs) elicited by BAM8-22 was significantly larger than the response to ALA (74.95 ± 8.88 APs vs. 47.44 ± 6.05 APs, respectively; paired t-test: t(65) = −2.142, p=0.0359; Figure 2A). Of the 66 CMHs tested, 1 fiber was unresponsive to either agonist, 26 responded only to BAM8-22, 11 only to ALA, and 28 to both agonists (Figure 2B). Thus, BAM8-22 activated 54/66 CMHs, whereas ALA activated 39/66 fibers. Across all fibers tested, the incidence of responsiveness to BAM8-22 and ALA was significantly different (Χ2(1)=6.44, p=0.0112). The response profiles for ALA and BAM8-22 across the responsive CMHs were indistinguishable, except for the first 10 s following the injection when BAM8-22-induced activity was about twice that of ALA. The duration of action potential activity was similar after BAM8-22 and ALA injection (Figure 2C). Figure 2 Download asset Open asset CMHs respond to bovine adrenal medulla peptide 8–22 (BAM8-22) more vigorously than to β-alanine (ALA) but with a similar time course. (A) The average evoked response of all CMHs to BAM8-22 (blue) was significantly larger than that to ALA (yellow) injection (paired t-test: t(65) = −2.142, p=0.0359). (B) Venn diagram of the number of CMHs responsive to BAM8-22 or ALA or both. (C) The time course of action potential activity (plotted as number of action potentials in 10 s bins) was similar after BAM8-22 and ALA injection, except for the greater response to BAM8-22 within the first 10 s following injection. The average responses of the same populations to vehicle (extracellular fluid [ECF] and BAM8-18) are graphed with gray and black lines, respectively. Error bars represent standard error of the mean (SEM). Gray box marks time of needle insertion and injection. Figure 2—source data 1 Activation of CMHs by ALA and BAM8-22. https://cdn.elifesciences.org/articles/64506/elife-64506-fig2-data1-v2.xlsx Download elife-64506-fig2-data1-v2.xlsx Based on responses to a stepped heat stimulus (49°C, 3 s, Figure 3A), CMHs can be classified into two distinct subtypes, QCs and SCs (Wooten et al., 2014). Briefly, heat responses of QCs exhibit a burst of discharge at the onset of the heat stimulus with the discharge adapting during the stimulus plateau. SCs typically have a slower response onset, do not exhibit a burst at the onset of the stimulus, and the peak discharge occurs during the plateau phase of the stimulus (Figure 3B). QCs and SCs can be formally separated by plotting the time of peak discharge (relative to stimulus onset) against the sum of temperature rise time plus the minimum conduction latency from skin (Figure 3C). For 31/66 CMHs, the time of peak discharge fell above the line of equality (Figure 3C, black line), indicating that the peak discharge occurred during the plateau phase of the stimulus, and these afferents were therefore classified as ‘SC’ (Figure 3C, blue circles). 29 out of 66 CMH fibers exhibited a burst of discharge at the onset and then an adapting response, and these afferents were classified as ‘QC’ (Figure 3C, red circles). For 26/29 QCs, the time of peak discharge fell below or close to the line of equality, indicating occurrence of peak discharge during or at the end of the rising phase of the temperature ramp. For the remaining three QC fibers, the time of peak discharge fell above the line of equality. Another six CMHs, in which the time of peak discharge fell close to the line of equality and which did not exhibit an initial burst discharge typical of a QC response, were labeled as ‘unclassified’ (Figure 3C, gray circles). The filled red and blue circles represent the data points for the specimen heat responses shown in Figure 3B. Figure 3 Download asset Open asset CMH subpopulations, QCs and SCs, exhibit differences in their response to heat stimulation and in the magnitude and time course of their response to β-alanine (ALA) and bovine adrenal medulla peptide 8–22 (BAM8-22). (A) Temperature waveform of the CO2 laser-evoked heat stimulus. The skin was first pre-heated to 38 °C, 3 s baseline temperature and then rapidly raised (rise time ~200 ms) to 49 °C for 3 s. (B) Specimen recording showing the response of an individual QC fiber (red circles) and an SC fiber (blue circles) to the laser-heat stimulus described in (A). The instantaneous discharge frequency is plotted versus time. Each dot represents the occurrence of an action potential (AP). The response in the QC fiber starts during the temperature rise and reaches peak frequency at the end of the ramp (red dashed line). The SC fiber reached the peak instantaneous discharge frequency during the plateau phase of the heat stimulus (blue dashed line). (C) The time of peak discharge for each fiber is plotted against the sum of stimulus rise time + the minimal AP conduction time as measured in response to transcutaneous electrical stimulation from the proximal edge of the receptive field (RF). Data points above the line of equality correspond to fibers in which the peak discharge occurred in the plateau phase of the heat stimulus (SCs, blue circles). Data points falling below the line of equality are from those fibers whose peak discharge occurred during the rising phase of the heat stimulus (QCs, red circles). The filled circles represent the data from the specimen recordings shown in (B) and also of the specimen responses to ALA and BAM8-22 shown in (D) and (E). Gray circles indicate data from fibers that were unclassified. Examples of responses in (D) of a QC fiber and (E) an SC fiber to ALA (top panels) and each to BAM8-22 (bottom panels). Responses to ALA and BAM8-22 are from the same fiber. The instantaneous frequency of each AP is plotted versus the time of its occurrence. The time course of neuronal activity induced by ALA, BAM8-22, and vehicle controls in the population of (F) QC fibers and (G) SC fibers. The average number of APs recorded over 10 s intervals during the 5 min observation period following injection is plotted. In QC fibers, ALA and BAM8-22 produced marked excitation, whereas in SC fibers, only BAM8-22 produced long-lasting activity. Error bars represent standard error of the mean (± SEM). Gray boxes mark time of needle insertion and injection. Figure 3—source data 1 Activation of QCs and SCs by ALA and BAM8-22. https://cdn.elifesciences.org/articles/64506/elife-64506-fig3-data1-v2.xlsx Download elife-64506-fig3-data1-v2.xlsx We found that the two subpopulations for CMHs, QCs and SCs, differed in their responses to injections of ALA and BAM8-22 into their cutaneous RFs. The differences are shown first by the responses of a typical QC and SC fiber in Figure 3D, E and summarized for the two populations in Figure 4. For the specimen QC (Figure 3D, same fiber for which the heat response is plotted in Figure 3B), injection of ALA (top panel) and BAM8-22 (bottom panel) induced a neuronal response for about 5 min, and the response to ALA (81 APs) was larger than to BAM8-22 (50 APs). Vehicle (ECF, BAM8-18) did not produce any activity in this afferent beyond the injection (data not shown). In the SC-fiber (for which the heat response is shown in Figure 3B), ALA administration did not result in activity outlasting the injection (Figure 3E, top panel) and the preceding ECF injection (response not shown) produced a total response of 18 APs. In contrast, injection of BAM8-22 (Figure 3E, bottom panel) produced a vigorous response of 198 APs within 5 min. The preceding BAM8-18 injection only produced activity during the injection (response not shown). Time courses of ALA- and BAM8-22-induced activity across the QC- and SC-populations are shown in Figure 3F and G, respectively. In QCs and SCs, vehicles did not evoke activity beyond the injection period. In QCs, ALA and BAM8-22 caused excitation with a similar time course (Figure 3F). For both agonists, the highest activity was observed immediately following the injection, and the activity decreased throughout the 5 min observation period. At 5 min post injection, the number of APs in QCs over 10 s was 0.55 ± 0.29 and 1.14 ± 0.23 for BAM8-22 and ALA, respectively. In contrast, in SCs, only BAM8-22-induced activity outlasted the injection and was different from the response to vehicle injections (Figure 3G). The highest activity occurred during the first 10 s following the injection, and the response decreased within approximately 3 min to a level similarly seen following vehicle injection. Figure 4 Download asset Open asset QCs and SCs from both male and female monkeys differ in sensitivity and incidence of activation by β-alanine (ALA) and bovine adrenal medulla peptide 8–22 (BAM8-22). (A) For each C-fiber, the response to BAM8-22 was plotted against its response to ALA. Solid red and blue circles indicate QC and SC fibers, respectively, recorded from male animals. Red and blue circles with a gray border indicate data obtained from female animals. Solid gray circles indicate data from C fibers with an unclassified heat response. The diagonal line indicates equal response to both compounds. Vertical and horizontal lines indicate the ‘threshold’ (≥10 action potentials [APs]) for an afferent to be counted as being responsive to an agonist (yellow and blue, respectively, for responses to ALA and BAM8-22). (B) Population responses of QCs and SCs to ALA and BAM8-22. In SCs, intradermal injection of BAM8-22 produces a significantly greater response than ALA (*** p<0.0001). The response to ALA was significantly larger in QCs than SCs (###p<0.0001), whereas the response to BAM8-22 was significantly larger in SCs than QCs (#p=0.002). Data were analyzed with repeated measures ANOVA (RMANOVA) with ‘fiber type’ as between-subjects factor and ‘pruritogen’ as a within-subjects factor (‘fiber type’ × ‘pruritogen’: F(1,58)=29,55; p<0.0001), followed by Scheffe test for post hoc analysis. (C) ALA activated 27/29 QCs and 7/31 SCs. BAM8-22 activated 21/29 QCs and 29/31 SCs. The occurrence of QCs and SCs responding to ALA only (eight and one afferents, respectively) was fairly rare. The number of QCs responding only to BAM8-22 (two units) was also small, whereas the majority of SCs (23/31) only responded to BAM8-22. Of the 31 SCs, one did not respond to either agonist. 19 QCs and 6 SCs responded to both, ALA and BAM8-22. Figure 4—source data 1 Preferrential activation of QCs and SCs by ALA and BAM8-22. https://cdn.elifesciences.org/articles/64506/elife-64506-fig4-data1-v2.xlsx Download elife-64506-fig4-data1-v2.xlsx To investigate further whether responses to ALA and BAM8-22 differed between QCs and SCs, the net response to BAM8-22 was plotted against that to ALA for each individual afferent (Figure 4A). For 20/29 QCs, data points fell below the diagonal line of equal responsiveness, indicating that responses to ALA were larger than responses to BAM8-22. In contrast, for 26/31 SCs, responses to BAM8-22 were greater than those to ALA. In fact, ALA responses in only seven SCs fulfilled the criterion of a positive response (≥10 APs). Responses to ALA and BAM8-22 differed significantly within and across fiber types (Figure 4B, repeated measures ANOVA (RMANOVA) with’ fiber type’ as a between-subjects factor and’ pruritogen’ as within-subjects factor; interaction: F(1,58)=29.55, p<0.001). Post hoc analysis revealed that for SCs (n = 31), responses to BAM8-22 were significantly larger than those to ALA (94.00 ± 12.01 APs vs. 10.32 ± 3.22 APs, p<0.001, Scheffe test). In QCs (n = 29), responses to ALA did not significantly differ from those to BAM8-22 (83.38 ± 8.42 APs vs. 51.97 ± 11.17 APs, respectively; p=0.25). Between populations, responses to ALA were significantly larger in QCs than SCs (p<0.001, Scheffe test), whereas responses to BAM8-22 were significantly larger in SCs than QCs (p=0.021, Scheffe test). ALA activated 27/29 QCs and 19 of these responded also to BAM8-22 (Figure 4C). Two QCs were only responsive to BAM8-22. In contrast, only 7/31 SCs responded to ALA with 6 of these being also responsive to BAM8-22, whereas 23/31 SCs responded only to BAM8-22. One of 31 SCs did not respond to either agonist. Of the six unclassified afferents, three units responded to both agonists, two responded to ALA only, and the remaining only to BAM8-22. Taken together, these findings suggest that, at the doses tested, QCs and SCs are preferentially activated by ALA and BAM8-22, respectively. The data imply that QCs may represent fibers expressing both MRGPRs at the peripheral terminals, but preferentially MRGPRD, whereas SCs may preferentially express MRGPRX1. To summarize, in contrast to our human ISH data that suggested nearly 100% overlap of MRGPRX1 and MRGPRD expression, we found that only about half of the CMH neurons tested responded to both BAM8-22 and ALA. Focusing more specifically on macaques, our combined ISH and electrophysiology data suggest that some CMHs (i.e., QCs) express both MRGPRs, while others (i.e., SCs) are functionally preferentially activated by BAM8-22 and only express MRGPRX1. Furthermore, in accordance with MRGPRD and MRGPRX1 neurons expressing TRPV1, all of these mechanosensitive C-fibers responded to noxious heat regardless of the selectivity of their responses to BAM8-22 or ALA. C-MIAs and A-fiber nociceptors are less frequently activated or unresponsive to ALA and BAM8-22 We also investigated whether BAM8-22 and ALA activate C-MIAs and A-fiber nociceptors that are involved in mediating itch sensation to HIS (Schmelz et al., 1997; Wooten et al., 2014) and cowhage (Ringkamp et al., 2011), respectively. We previously reported that seven C-MIAs were tested with ALA and that none responded (Wooten et al., 2014). Of six C-MIAs tested with BAM8-22 in this study, all were tested with heat and four responded. Two of six C-MIAs (one responsive to heat) responded to BAM8-22. One of the C-MIAs had a 5 min response of 146 APs with 112 APs occurring during the first 30 s following the injection, and the response ending within 5 min. The other C-MIA had a 5 min response of 122 APs, of which 50 APs occurred during the first 30 s after injection. The response lasted for 20 min with a total response of 171 APs. We tested 16 mechanosensitive and 8 mechanoinsensitive A-fiber nociceptors (A-MSA and A-MIAs, respectively). Of the 16 A-MSAs, one responded to ALA (5 min net response of 14 APs), and another to BAM8-22 with 129 APs within the first 30 s after injection but no activity thereafter. Of the eight A-MIAs, none responded to ALA or BAM8-22. These data suggest that A-fiber nociceptors and C-MIAs are likely minor contributors to the sensations caused by these compounds. In humans, a combination of pruritogens does not change the itch, nociceptive sensations, dysesthesias, or skin reactions compared to the effects of one of the component pruritogens given alone In humans, cowhage-induced itch and nociceptive sensations are mediated by activation of nociceptive unmyelinated and myelinated afferents that are also thought to mediate pain from noxious heat and mechanical stimuli (Johanek et al., 2008; Namer et al., 2008; Ringkamp et al., 2013; Ringkamp et al., 2011). How the sensations of itch and pain are mediated by activity in the same set of afferent nerve fibers is currently unclear, and several models have been proposed to solve this puzzle (Carstens et al., 2020; LaMotte et al., 2014). Among these, the ‘spatial contrast’ model posits that locally restricted activation of a small population of pruriceptive nociceptors induces itch sensation, while activation of a broader and greater number of nociceptors induces the sensation of pain (Namer et al., 2008; Namer and Reeh, 2013; Steinhoff et al., 2019). Our present electrophysiological results indicate that ALA and BAM8-22 preferentially activate different sets of cutaneous CMHs in monkey (Figure 4). Previous studies have shown that HIS-induced itch is likely mediated by C-MIAs (Schmelz et al., 1997; Wooten et al., 2014). Therefore, the ‘spatial contrast’ model can be tested experimentally by concurrent intradermal injections of combinations of ALA, BAM8-22, and HIS and comparison of the induced psychophysical responses to those from injection of an individual pruritogen. The majority of subjects reported itch and nociceptive sensations after the administration of a single pruritogen and after the combinations thereof. The number of subjects reporting these sensations did not differ between stimuli (Figure 5A), and the number of subjects experiencing itch or nociceptive sensations did not change when combinations of pruritogens were applied. For the co-administration of BAM8-22 and ALA, the temporal profiles of the different sensations were visibly similar compared to those evoked by BAM8-22 or ALA alone, peaking within the first minute after application and decreasing slowly over approximately 10 min (Figure 5B–D). Similarly, HIS-induced itch peaked within the first minute after injection but appeared to decline much slower over 20 min (Figure 5E). The temporal profile of the itch sensation produced by the triple combination was similar to that of the BAM8-22 + ALA combination, except that the decline of itch sensation appeared to be slower (Figure 5E). The time courses of the nociceptive sensations following HIS were similar to those of the BAM8-22 + ALA combination and the combined application of all three pruritogens (Figure 5F, G). Figure 5 with 1 supplement see all Download asset Open asset In humans, co-injection of β-alanine (ALA) and bovine adrenal medulla peptide 8–22 (BAM8-22) with or without histamine (HIS) does not change the itch or nociceptive sensations compared to the effects of one component pruritogen given alone. (A) The majority of subjects (n = 29) reported itch, pricking/stinging, and burning sensations after administration of each of the pruritogens or their combinations. (B) Magnitude of itch, (C) stinging/pricking, and (D) burning sensations evoked by BAM8-22, ALA, and a combination of BAM8-22 and ALA are plotted for successive 30 s intervals after injection averaged across all 29 subjects. (E) Magnitude of itch, (F) stinging/pricking, and (G) burning sensations evoked by HIS, a combination of BAM8-22 and ALA, and a combination of BAM8-22, ALA, and HIS. For clarity, the standard error of the mean (SEM) is plotted only every 5 min starting with the peak rating for each quality. On the right vertical axis, the locations of three verbal descriptors are shown in corresponde

The ATP-sensitive K+ channel (KATP) is involved in hypersensitivity during chronic pain and is presumed to be a downstream target of mu opioid receptors. Multiple subtypes of KATP channels exist in the peripheral and central nervous system and their activity may be inversely correlated to chronic pain phenotypes in rodents. In this study, we investigated the different KATP channel subunits that could be involved in neuropathic pain in mice. In chronic pain models utilizing spinal nerve ligation, SUR1 and Kir6.2 subunits were found to be significantly downregulated in dorsal root ganglia and the spinal cord. Local or intrathecal administration of SUR1-KATP channel subtype agonists resulted in analgesia after spinal nerve ligation but not SUR2 agonists. In ex-vivo nerve recordings, administration of the SUR1 agonist diazoxide to peripheral nerve terminals decreased mechanically evoked potentials. Genetic knockdown of SUR1 through an associated adenoviral strategy resulted in mechanical hyperalgesia but not thermal hyperalgesia compared to control mice. Behavioral data from neuropathic mice indicate that local reductions in SUR1-subtype KATP channel activity can exacerbate neuropathic pain symptoms. Since neuropathic pain is of major clinical relevance, potassium channels present a target for analgesic therapies, especially since they are expressed in nociceptors and could play an essential role in regulating the excitability of neurons involved in pain-transmission.

ATP-sensitive potassium (KATP) channels are found in the nervous system and are downstream targets of opioid receptors. KATP channel activity can effect morphine efficacy and may beneficial for relieving chronic pain in the peripheral and central nervous system. Unfortunately, the KATP channels exists as a heterooctomers, and the exact subtypes responsible for the contribution to chronic pain and opioid signaling in either dorsal root ganglia (DRG) or the spinal cord are yet unknown. Chronic opioid exposure (15 mg/kg morphine, s.c., twice daily) over 5 days produces significant downregulation of Kir6.2 and SUR1 in the spinal cord and DRG of mice. In vitro studies also conclude potassium flux after KATP channel agonist stimulation is decreased in neuroblastoma cells treated with morphine for several days. Mice lacking the KATP channel SUR1 subunit have reduced opioid efficacy in mechanical paw withdrawal behavioral responses compared to wild-type and heterozygous littermates (5 and 15 mg/kg, s.c., morphine). Using either short hairpin RNA (shRNA) or SUR1 cre-lox strategies, downregulation of SUR1 subtype KATP channels in the spinal cord and DRG of mice potentiated the development of morphine tolerance and withdrawal. Opioid tolerance was attenuated with intraplantar injection of SUR1 agonists, such as diazoxide and NN-414 (100 μM, 10 μL) compared to vehicle treated animals. These studies are an important first step in determining the role of KATP channel subunits in antinociception, opioid signaling, and the development of opioid tolerance, and shed light on the potential translational ability of KATP channel targeting pharmaceuticals and their possible future clinical utilization. These data suggest that increasing neuronal KATP channel activity in the peripheral nervous system may be a viable option to alleviate opioid tolerance and withdrawal.

The management of chronic pain with opioids can cause opioid-induced analgesic tolerance (OIT) and hyperalgesia (OIH), which complicates clinical pain-management treatments. Although G-protein coupled receptors have been studied for years, the intracellular signaling pathways triggered by the activation of µ-opioid receptors are not well known. We sought to determine whether OIT is linked to a decreased activity in the PI3K/AKT intracellular signaling pathway. To assess this pathway, 25 C57BL/6 WT male mice (21 ± 4g) were divided into control (n=10, 1 μL saline) and experimental (n=10, 15 mg/kg of morphine; n=5, spinal nerve ligation (SNL) at the L4 vertebrae with 15 mg/kg of morphine) groups. Injections were given subcutaneously twice a day for a total of five days. Thermal paw withdrawal latency for each group was measured before and 0, 30, and 60 minutes post injection. Mice given morphine developed OIH and OIT after 3 days. The brainstem, spinal cord, dorsal root ganglia, and sciatic nerves were removed and processed for qPCR using primers for AKT1, AKT2, AKT3, PIK3cg v1, PIK3cg v2, PIK3cg v3, PTEN, and nNOS1. Significant increases and decreases in gene expression levels (p≤0.05) were seen in the brainstem of morphine tolerant mice without SNL. There was a decreased expression for AKT1 (p<0.05), AKT2 (p<0.05), PIK3cg v3 (p<0.01), PTEN (p<0.01), and nNOS1(p<0.05), and an increased expression of AKT3 (p<0.05) in the brainstem of morphine-tolerant mice without SNL. Current and future studies will be performed to evaluate the expression levels of cGMP, a downstream target of the PI3K/AKT pathway, in the spinal cord and DRG. Supported by grants from the National Institutes of Health (K01 DA 042902) and the University of Minnesota Undergraduate Research Opportunity Program. The management of chronic pain with opioids can cause opioid-induced analgesic tolerance (OIT) and hyperalgesia (OIH), which complicates clinical pain-management treatments. Although G-protein coupled receptors have been studied for years, the intracellular signaling pathways triggered by the activation of µ-opioid receptors are not well known. We sought to determine whether OIT is linked to a decreased activity in the PI3K/AKT intracellular signaling pathway. To assess this pathway, 25 C57BL/6 WT male mice (21 ± 4g) were divided into control (n=10, 1 μL saline) and experimental (n=10, 15 mg/kg of morphine; n=5, spinal nerve ligation (SNL) at the L4 vertebrae with 15 mg/kg of morphine) groups. Injections were given subcutaneously twice a day for a total of five days. Thermal paw withdrawal latency for each group was measured before and 0, 30, and 60 minutes post injection. Mice given morphine developed OIH and OIT after 3 days. The brainstem, spinal cord, dorsal root ganglia, and sciatic nerves were removed and processed for qPCR using primers for AKT1, AKT2, AKT3, PIK3cg v1, PIK3cg v2, PIK3cg v3, PTEN, and nNOS1. Significant increases and decreases in gene expression levels (p≤0.05) were seen in the brainstem of morphine tolerant mice without SNL. There was a decreased expression for AKT1 (p<0.05), AKT2 (p<0.05), PIK3cg v3 (p<0.01), PTEN (p<0.01), and nNOS1(p<0.05), and an increased expression of AKT3 (p<0.05) in the brainstem of morphine-tolerant mice without SNL. Current and future studies will be performed to evaluate the expression levels of cGMP, a downstream target of the PI3K/AKT pathway, in the spinal cord and DRG. Supported by grants from the National Institutes of Health (K01 DA 042902) and the University of Minnesota Undergraduate Research Opportunity Program.

Mouthfeel refers to the physical or textural sensations in the mouth caused by foods and beverages that are essential to the acceptability of many edible products. The sensory subqualities contributing to mouthfeel are often chemogenic in nature and include heat, burning, cooling, tingling, and numbing. These "chemesthetic" sensations are a result of the chemical activation of receptors that are associated with nerve fibers mediating pain and mechanotransduction. Each of these chemesthetic sensations in the oral cavity are transduced in the nervous system by a combination of different molecular channels/receptors expressed on trigeminal nerve fibers that innervate the mouth and tongue. The molecular profile of these channels and receptors involved in mouthfeel include many transient receptor potential channels, proton-sensitive ion channels, and potassium channels to name a few. During the last several years, studies using molecular and physiological approaches have significantly expanded and enhanced our understanding of the neurobiological basis for these chemesthetic sensations. The purpose of the current review is to integrate older and newer studies to present a comprehensive picture of the channels and receptors involved in mouthfeel. We highlight that there still continue to be important gaps in our overall knowledge on flavor integration and perception involving chemesthetic sensations, and these gaps will continue to drive future research direction and future investigation.