The nociceptive system is a major target of interest for the design of analgesic drugs. When administered at this level of the pain pathway, analgesic agents such as opioids and local anesthetics ultimately act on primary afferent endings by inhibiting presynaptic release of nociceptive signaling molecules, such as substance P and calcitonin gene-related peptide (CGRP) (Julius and Basbaum, 2001; Kondo et al., 2005). A better understanding of which afferents are required for nociception and what constellation of receptors are expressed by these neurons is an important component of designing alternative strategies to selectively inhibit nociception while preserving other somatosensory functions.

Transient receptor potential vanilloid subfamily, member 1 (TRPV1) is expressed by primary afferent nociceptive neurons (Caterina et al., 1997; Szallasi and Blumberg, 1999; Caterina, 2007), and transduces thermal and inflammatory tissue damage stimuli (Karai et al., 2004a; Mishra and Hoon, 2010; Mitchell et al., 2010, 2014). In the search for novel non-opioid analgesics, much attention has been centered on the TRPV1 molecule as an analgesic target (Jancso et al., 1977; Yaksh et al., 1979; Bevan et al., 1992; Urban et al., 2000; Wong and Gavva, 2009; Iadarola and Gonnella, 2013). Genetic ablation or chemical silencing of TRPV1+ afferents has been reported to cause loss of nociceptive responses (Jancso et al., 1977; Karai et al., 2004a; Brown et al., 2005, 2015; Mishra and Hoon, 2010; Brown and Iadarola, 2015; Sapio et al., 2018). However, direct antagonism of the TRPV1 receptor has not led to clinically useful antinociceptive agents, despite promising preclinical rodent studies and demonstrable target engagement in humans (Menéndez et al., 2006; Szallasi et al., 2007; Rowbotham et al., 2011; Quiding et al., 2013). These drugs did not advance in part due to induction of varying degrees of hyperthermia (Wong and Gavva, 2009; Garami et al., 2018; Park et al., 2021), and lack of sustained efficacy (Quiding et al., 2013), as TRPV1 is not the only nociceptive transducing molecule expressed by primary afferents (Goswami et al., 2014). In the complex milieu of tissue damage and inflammation, multiple other algesic mediators remain capable of discharging the primary afferents, even with a high degree of TRPV1 inhibition, which may explain why TRPV1 antagonists are much less analgesic than agonist-mediated chemoaxotomy approaches (Karai et al., 2004a; Iadarola et al., 2018; Sapio et al., 2018). Additionally, patients treated with high concentrations of TRPV1 antagonists become insensitive to noxious hot thermal stimuli, posing a burn risk as they rated temperatures up to 49°C as safe (Rowbotham et al., 2011). Thus, TRPV1+ afferents are the critical transducers of pain associated with tissue damage as well as most forms of warm and hot thermosensation.

Untangling desirable and undesirable pharmacologic effects of TRPV1-based therapies requires clarification of the underlying neurobiology. An interesting observation is that selective lesioning of TRPV1+ nerve endings using the highly potent TRPV1 agonist, resiniferatoxin (RTX), leads to loss of thermal nociceptive sensitivity and with high subcutaneous doses (250 ng intraplantar) can produce near-total insensitivity to tissue-damaging stimuli in the denervated area (Mitchell et al., 2010, 2014). While it is known that TRPV1+ nociceptors comprise several discrete populations of sensory afferent neurons (Sapio et al., 2018), detailed studies of nociceptive cellular neurobiology are needed to elucidate the precise molecular markers of each population. The objective of the present investigations is to delineate the molecular complement of receptors and ion channels in neurons responsible for nociception and analgesic action. Specifically, we hypothesize that we can identify a population of cells using multiplex labeling that contains several nociceptive and analgesic markers, that maybe useful for future studies of pain and analgesia. Delineating these neurons in a quantitative, molecularly-defined fashion enables future studies to directly target these neurons using selective pharmacological analgesic strategies.

This study examined rat primary afferent neurons using multiplex fluorescent mRNA in situ hybridization to understand nociceptive and analgesia mechanisms in the peripheral nervous system. We consider identifying critical neuronal populations essential knowledge for developing potential peripherally acting non-opioid analgesics that do not exhibit central nervous system side effects. The investigation focused on determining co-expression patterns of nociceptive-, and analgesia-transducing molecular markers in DRG neurons. We hypothesized the conjuncture of these two types of molecules in one or more neuronal subpopulations would provide a colocalization matrix that could be used to generate a molecularly informed approach to peripheral analgesia and analgesic drug development.

The populations of nociceptive Trpv1+ neurons were heterogeneous, expressing varying levels of Trpv1 and the μ-opioid receptor Oprm1. Co-expression analysis indicated differing degrees of μ-opioid expression among the two populations of Trpv1+ neurons that we hypothesize subserve thermosensory and tissue-injury nociception. We identified a population of very high expressors of Trpv1 that are absent of Oprm1. In contrast, most low and medium expressors of Trpv1 do express Oprm1, and are the evident candidates for the neuronal population inhibited by μ-opioid agonist analgesics, especially when given intrathecally or epidurally, administration paradigms where the injectate bathes the central projection of dorsal root ganglion axons entering the spinal cord. The co-expression of Trpv1 and Oprm1 receptors within the same cell types provides both cellular and molecular level explanations for the efficacy of opioid agonists. That is, opioid drugs act by inhibiting the very cells responsible for transmitting certain types of nociceptive stimuli from the periphery to the central nervous system. Available evidence suggests that opioid analgesia occurs, in part, through presynaptic inhibition of transmitter release from Trpv1+ DRG afferent terminals in spinal cord and potentially spinal cord μ-expressing neurons (Kondo et al., 2005; Chen and Pan, 2006; Xanthos and Sandkühler, 2014; Che et al., 2021). Co-expression of Trpv1 was never observed with Spp1, a marker for large diameter proprioceptive neurons, a finding that validates that nociception and proprioception are mediated by distinct neuronal populations. Additionally, the lower expressors of Trpv1 co-expressed other markers implicated in pathological pain states. For example, Trpm8, a nociceptive ion channel implicated in cold thermosensation (Bautista et al., 2007) and cold allodynia (Xing et al., 2007) was identified in some lower-expressing Trpv1 sensory neurons. Additionally, the ion channel NaV 1.9 (encoded by SCN11A), whose gain-of-function mutation leads to familial episodic pain (Leipold et al., 2013), was also identified in the low to medium-expressing Trpv1 population. In contrast, the small diameter high Trpv1 expressing neurons were negative for all of these markers, most likely indicating an exclusive role as a dedicated labeled line for thermonociception and/or non-noxious thermosensation (Kobayashi et al., 2005; Craig, 2018). Coexpression analysis is also a critical step for gaining insight into the algesic and analgesic receptors on DRG neurons, and could be important for understanding indirect interactions between painful and analgesic target receptors. In particular, there are established interactions between pain transducing ion channels and opioid receptors (Chew et al., 2019).

The μ-opioid receptor is one of the most clinically important targets for analgesia, with the majority of highly impactful orally and intravenously available analgesic agents acting through this receptor. While there are three canonical opioid receptors (μ, δ, and κ), and a fourth opioid-like paralog (Oprl1), only μ-receptor agonists are used clinically at this time, although extensive medicinal chemical and preclinical development on the other receptors has occurred or is in progress (Calo et al., 2000; Floyd et al., 2009; Bardoni et al., 2014; Viscusi et al., 2021). Additionally, some controversy exists over detection of these receptors, as they are frequently expressed to a low degree and difficult to quantify accurately (Imlach and Christie, 2014; Iadarola et al., 2015; Sapio et al., 2016). In this regard, the present study fills a critical knowledge gap by delineating the co-expression matrix of these receptors in detail. We characterize the expression pattern of all four receptors, and, in particular, the canonical three together. One point of contention is the pharmacological relevance of opioid receptor heterodimers. In order for the heterodimer to form (Jordan and Devi, 1999; Erbs et al., 2015; Gaborit and Massotte, 2021), both receptors would, at a minimum, have to be expressed in the same cell. Our data confirm that co-expression of the μ-, δ- and κ-opioid receptors does indeed occur in some combinations, albeit in a very small population of neurons. This heterogeneity is confirmed by recording and staining from rat DRG neurons, although some differences in proportions of cells between the present study and the in vitro recording investigation are seen and are likely due to methodological differences (Rau et al., 2005). Interestingly, some large cells were observed with high concentrations of μ- and δ-opioid receptors (Figure 1E), consistent with putative mechanosensory nociceptors, based on genetic anatomical tracing studies done in the mouse (Bardoni et al., 2014). These anatomical findings showing colocalization of Oprm1 and Oprd1 may be related to the observed analgesic synergy between μ- and δ-opioid agonists in the periphery (Bruce et al., 2019). Another study using genetic labeling in mice showed that the κ-opioid receptor was frequently expressed in myelinated lanceolate or circumferential nerve endings around hair follicles, and demonstrated that κ-opioid receptor function at these afferents modulated pain and itch (Snyder et al., 2018). In our rat study the majority, although not all κ-opioid receptor expressing neurons, also expressed the μ-opioid receptor indicating that this pair of receptors likely functions in the same cell to modulate some modalities of nociception.

Another opioid-like receptor examined in this study is the nociceptin/orphanin FQ receptor (NOP receptor, encoded by Oprl1). This is the least studied of the four opioid-like receptors, and although it has been proposed to have multiple avenues toward clinical potential (Lambert, 2008), none have been realized to date. One finding that is particularly relevant to pain and analgesia, is the observation that intrathecal administration of nociceptin, the endogenous agonist at the NOP receptor was analgesic, and this analgesia was not naltrexone reversible, suggesting NOP-dependent analgesic actions are distinct from analgesia mediated at other opioid receptors (Ko et al., 2006). In our staining, we identified signal for Oprl1 expression in almost all DRG neurons, which does not exclude its utility as an antinociceptive agent, but suggests very broad function(s), and may result in off target side effects on non-nociceptive fibers. Notably, broad action, or expression of an analgesic target outside of the nociceptive neurons is not necessarily indicative of reduced analgesic potential, but the ubiquity of Oprl1 signal does contradict the supposition that this receptor is involved in endogenous analgesia in a specific fashion.

One of the key nociceptive markers examined in the present study is Trpv1. This receptor is the endogenous receptor for capsaicin, the pungent ingredient in hot peppers and resiniferatoxin, a new analgesic currently undergoing clinical trials (NCT03542838, NCT02522611, NCT00804154, NCT04885972, and others) (Caterina et al., 1997). It has long been appreciated that Trpv1 is expressed in several subclasses of neurons, and more recently these populations have been described to a greater extent in several species (Goswami et al., 2014; Isensee et al., 2014; Usoskin et al., 2015; Sapio et al., 2018; Tavares-Ferreira et al., 2022). The overall pattern of expression of Trpv1 is similar to previous studies, although it appears that the RNAScope procedure, in particular, labels more neurons than most antibody-based approaches (Mitchell et al., 2010, 2014; Goswami et al., 2014; Poulson et al., 2020; Sapio et al., 2020b; Shiers et al., 2020; Hall et al., 2022; Tavares-Ferreira et al., 2022). It is also notable that the percentage of Trpv1+ neurons as assessed by a technique as sensitive as RNAScope is not necessarily identical to the number of capsaicin-responsive neurons. For example, the percentage of neurons killed by capsaicin incubation in culture is estimated to be approximately 37% in one study (Wood et al., 1988), which is also similar to the number of cultured mouse lumbar DRG neurons responding to 300 nM capsaicin (39%) (Teichert Russell et al., 2012). Therefore, it appears that at a certain level of expression is required to produce robust enough calcium influx to compromise cellular integrity (Karai et al., 2004b). That is, approximately half of the neurons detected as Trpv1+ in our study (presumably the most strongly labeled neurons) are likely to be capsaicin responsive.

Generally, it has been observed that there is a small number of densely immunoreactive small diameter Trpv1+ neurons, as well as a larger number of larger diameter and less densely stained neurons (Tominaga et al., 1998; Cavanaugh et al., 2011b; Goswami et al., 2014). Presumably, this encompasses at least two functionally defined TRPV1+ subclasses: the Trpv1+ C-fibers and A-δ fibers (Mitchell et al., 2010, 2014). The molecular identity of the TRPV1+ A-δ fibers is still being determined (Raithel et al., 2018), in part due to the small percentage of A-δ neurons in DRG (Lawson et al., 2019). Our study suggests that the small diameter high expressing Trpv1+ fibers in the rat contain a narrower repertoire of noci-responsive TRP channels, potentially implicating a more exclusive thermal nociceptive role which, at least in the rat, does not appear to be sensitive to opioid-induced analgesia. Differential thermal activation of A-δ and C-fibers suggest that, based on behavioral withdrawal, the C-fiber population is the one that is less sensitive to μ-opioid receptor agonists (Mitchell et al., 2014) and the A-δ neurons are a subpopulation of the medium to low TRPV1 expressing neurons. To some extent, a similar phenomenon has been examined in humans in a study using high dose transdermal capsaicin patch. In this study, sensitivity to noxious (55°C) thermal stimulation was confined to the capsaicin-treated region, while intensity ratings for less noxious (44°C) or non-noxious (38°C) thermal stimuli were reduced distal to the patch-treated skin (Van Neerven and Mouraux, 2020). This is potentially consistent with the idea that higher expressing Trpv1+ fibers are more capsaicin sensitive and primarily involved in thermonociceptive sensation, while the more broadly, polymodal nociceptive fibers may have less Trpv1 but co-express multiple noci-responsive channels. However, notably, this type of experiment may also be influenced by other variables such as receptive field size and central modulatory effects. Further studies will be needed in humans to tease out the function of these Trpv1+ populations.

Another TRP channel Trpa1 was examined in regard to both Trpv1 and analgesic-related opioid receptors. Trpa1 is the receptor for chemical nociception and is responsible for transducing stimuli such as allyl isothiocyanate, the active pungent ingredient in foods such as wasabi (Jordt et al., 2004; Mcnamara et al., 2007). In our previous study using transcriptomics and multiplex fluorescent in situ hybridization, we examined the expression pattern of Trpv1 and Trpa1 in DRG and nodose ganglion sensory afferents (Sapio et al., 2020b). In that study, we found that Trpv1 and Trpa1 were more highly co-expressed in nodose, whereas in the DRG, neurons that were high in either Trpv1 or Trpa1 tended to be low for the other, suggesting a separate, although not exclusive set of sensory pathways for thermal and chemical nociception in the peripheral nervous system. The present study confirms this result (Figure 6). The largely differential expression is also consistent with previous studies showing partial depletion of Trpa1 with Trpv1 agonists (Pecze et al., 2009) or coexpression of these receptors. However, TRPA1 is apparently expressed in low expressing TRPV1 population, which could explain why these neurons are resistant to TRPV1 agonist actions and could also explain the failure to deplete TRPA1-expressing neurons with TRPV1 agonists in some experiments (Isensee et al., 2014; Sapio et al., 2018, 2020b). The functional implication of co-expression of these receptors is being investigated. For example, heteromers have been reported between these two ion channels with different pharmacologic and biophysical properties and have been proposed to be involved in nociceptive sensitization (Fischer et al., 2014; Patil et al., 2020). Cells with both Trpv1 and Trpa1 also tended to have moderate to high amounts of Oprm1 expression (see Figure 3, Quad+ neurons), consistent with their proposed dual nociceptive and analgesia-conferring properties (Akopian, 2011).

Trpm8 is the primary receptor for cold and cold pain (Bautista et al., 2007) and serves as a marker for cold-sensing sensory afferents (Renthal, 2020). We describe populations of high and low Trpm8 expression, where presumably the high Trpm8 expressing cells are those that have primarily been described in previous studies as cold thermosensory afferents (Le Pichon and Chesler, 2014; Jankowski et al., 2017), and the lower levels of Trpm8 in other cells are currently of unknown significance. However, these could be polymodal neurons, or cells that become recruited at colder temperatures and/or during cold allodynia. Notably, while Trpm8+ afferents have been shown to be involved in responding to environmental cooling, a small population of Trpm8+ fibers has been reported to respond to noxious mechanical stimulation (Jankowski et al., 2017), and it has been suggested that in injury states a subpopulation of Trpm8+ fibers may become responsive to additional noxious modalities. Additionally, the processes underlying thermal encoding in the DRG can be complex, with populations of cells responding in a graded fashion to various temperatures (Wang et al., 2018). Furthermore, electrical recordings have shown that cooling-responsive cells comprise only about 2.3% of all DRG neurons (Lawson et al., 2019), whereas the total number of Trpm8+ cells we quantified was much larger than this estimate. While examining the full range of thermal-encoding cells is beyond the scope of this study, the coincidence of expression of Trpv1 and Trpm8, which encode two major encoding ion channels of temperature, may be useful information for future investigations. This is also an indication that we are examining multiple classes of DRG neurons among the Trpm8+ population. This is one area where additional electrical characterization of these subtypes would be informative, as this can elucidate functional response characteristics to further segregate the DRG afferents (Petruska et al., 2000; Rau et al., 2005; Lawson et al., 2019). This is also interesting given that Trpm8 expression level may be related to tuning thermal responsiveness, which is notable given the finding that various channels and/or channel combinations may be involved in thermal coding based on mouse studies (Paricio-Montesinos et al., 2020).

The overlap between Trpm8 and Oprm1 expression in the rat makes sense given reduction in clinical responses to cold sensation upon morphine (a mu-opioid receptor agonist) administration (Cleeland et al., 1996). This is also corroborated by the finding that morphine alleviated cold allodynia in rat models of chronic pain (Erichsen et al., 2005). Both cases support the idea that Trpm8+ cells are inhibited by μ-opioid receptor agonists. The Trpm8+ neurons also appear to be largely Trpv1-, indicating discrete types of peripheral neurons for the detection of hot and cold. This lack of overlap between Trpv1 and Trpm8, is consistent with previous evidence demonstrating that Trpm8+ neurons are (in general) a small-diameter, relatively rare Trpv1- population of sensory afferents (Dhaka et al., 2008; Pecze et al., 2009; Le Pichon and Chesler, 2014; Jankowski et al., 2017). Although we did detect Trpm8+/Trpv1+ neurons, the expression was largely anticorrelated (Figure 6H) indicating that these two receptors are never strongly co-expressed. While developmentally, Trpm8 is co-expressed in a subset of Trpv1-lineage neurons in the mouse (Mishra et al., 2011) pharmacologic ablation of Trpv1+ fibers in adult rodents suggests a further differentiation of a distinct subpopulation of Trpm8+/Trpv1- neurons (Cavanaugh et al., 2009) consistent with our anatomical findings. This is also consistent with ISH and single-cell RNA-Seq in DRG and trigeminal ganglion showing very low levels of Trpv1 in Trpm8+ neurons (Kobayashi et al., 2005; Von Buchholtz et al., 2020). The idea of multiple Trpm8+ populations (Xing et al., 2006), where the Trpv1-coexpressing population may be more nociceptive in nature would potentially explain our results in context of the broader literature. We also cannot rule out an interaction between these input pathways at the level of the spinal cord, where Trpv1-specific and Trpm8-specific input pathways may be processed together in spinal circuits. For example, it is known that there is some interaction between Trpv1-specific and Trpm8-specific behavioral outcomes (Anderson et al., 2014).

The tetrodotoxin-resistant sodium channel subunit NaV1.9 (encoded by Scn11a) has come into focus as a potential nociceptor-specific target for the development of new non-opioid analgesics. Alongside two other sodium channel subunits in this family (Scn9a and Scn10a), Scn11a has been suggested as an interesting and understudied mediator of pain signaling. For example, the related ion channel subunit gene, Scn9a has been extensively studied in rodents and humans where it is strongly linked to pain and nociceptive signaling (Hisama et al., 1993; Dib-Hajj et al., 2005, 2007, 2013; Yang et al., 2018). Similarly, Scn10a has been used extensively as a nociceptive specific marker (Akopian et al., 1996; Dib-Hajj et al., 1999; Stirling et al., 2005; Shiers et al., 2020). The current study focuses on Scn11a largely due to the fact that it has been studied less intensively than Scn9a and Scn10a, and because mutations in this gene are associated with pain insensitivity or episodic pain syndromes in humans (Leipold et al., 2013; Zhang et al., 2013). For example, NaV1.9 has been found to be enriched in cells in the Trpv1 lineage (Goswami et al., 2014) or depleted with TRPV1 agonist treatment (Isensee et al., 2014; Sapio et al., 2018), suggesting specificity for the nociceptive population. This channel has also been implicated in nociception and hyperalgesia in animal models, suggesting potential utility as an analgesic pharmacological target (Priest et al., 2005). NaV1.9 is required for cold-triggered nociception in the mouse (Lolignier et al., 2015), consistent with our observation that Scn11a is found in high-expressing Trpm8+ neurons. However, our staining shows a broad expression pattern of Scn11a consistent with involvement in other nociceptive functions. For example, Scn11a is frequently co-expressed with Trpv1, which is consistent with human mutation studies of NaV1.9 in nociception in which loss or gain of function SCN11A mutations can cause pain insensitivity or episodic pain syndromes (Leipold et al., 2013; Zhang et al., 2013). The phenotype depends on the location of mutation in the channel. Importantly, the finding that gain of function mutations that inhibit transmitter release leads to pain insensitivity suggests that NaV1.9 is expressed in enough nociceptive afferents to be sufficient to support a pain insensitive phenotype. These findings are also corroborated by other studies using single cell RNA-Seq showing that Scn11a is expressed in Trpv1+ cells, but also in several other classes of sensory afferents (Li et al., 2016; Sapio et al., 2018).

In our staining analysis, we examined secreted phosphoprotein 1 (Spp1) positive neurons, which are non-nociceptive large diameter primary afferents that innervate muscle spindles. The Spp1 gene encodes osteopontin (Ichikawa et al., 2000). While the function of osteopontin in neurons has not been determined, it has been suggested as a regulator of myelination, and hypothesized that its expression is related to the maintenance or formation of axons with high conduction velocities (Higo et al., 2010). These neurons are thought to be chiefly implicated in proprioception (Ichikawa et al., 2000; Usoskin et al., 2015; Saito-Diaz et al., 2021). Anatomically, osteopontin immunoreactivity localizes to spiral axon terminals in muscle spindle fibers. In our staining, we found that Spp1+ cells were negative for Trpv1 and Oprm1, but co-expressed Oprl1. These neurons were also generally large, and their staining pattern, showing distinction from Trpv1 expressors, further suggests their non-nociceptive nature in the DRG.

The current study has several limitations that must be addressed in future studies. For the present investigation, experiments are conducted in male rats, and as such may not be fully reflective of the neuroanatomy and neurochemistry in the mouse or across both sexes (Sadler et al., 2022). For behavioral studies, the rat has many advantages compared to the mouse, however, at this point in time, a large set of molecular-genetic rat manipulations are not available for investigation as they are for the mouse nervous system (Ellenbroek and Youn, 2016; Homberg et al., 2017). Thus, the use of multiplex fluorescence represents a concerted effort to assemble the combinatorial molecular-cellular expression patterns in rat that mediate algesia and analgesia in sensory ganglia. The ultimate goal of these investigations is to gain insights into conserved molecular biological phenomena in both sexes and in humans. Future studies will concentrate on direct human investigation in male and female organ donor tissue to corroborate and extend these findings into human health, disease, and pharmacology (Iadarola et al., 2022), as such studies are more likely to yield information that may be more clinically relevant and applicable. As an additional area for future directions, much attention has been paid to the use of single cell and single nucleus RNA-Seq for the determination and characterization of neuronal cell types in the DRG in mouse, monkey and human (Usoskin et al., 2015; Kupari et al., 2021; Tavares-Ferreira et al., 2022). These studies are useful in the determination of cell types but often lack the precision to determine the expression profile of an individual gene. In the existing databases, the opioid receptors are weakly detected, and probably encountered a limit of detection. Additionally, for genes such as Trpv1 and Trpm8, the mouse and monkey databases have shown correlation between Trpv1 and Trpm8, where the cell population with the highest level of one also has the highest level of the other, which is in direct contrast to our findings (Usoskin et al., 2015; Kupari et al., 2021). However, this could be due to multiple technical issues or species differences and is beyond the scope of the current investigation. Finally, another limitation is that this study does not assess the impact of pain and nerve injury on expression and co-expression of these markers. Nerve injury in particular induces strong transcriptional changes and may alter the baseline expression patterns of nociceptive and analgesic targets (Ray et al., 2018; Sapio et al., 2020a). Future studies can expand upon these foundational data to understand pain conditions.

While the field of nociceptive neuroscience has expended substantial effort over many years toward understanding key receptors involved in nociceptive circuits, there is a need for exact delineation of neuronal populations that contribute to clinical pain and pain control as well as identification of molecular signatures which can define these distinct populations. Our study utilized multiplex high sensitivity mRNA in situ hybridization, which allowed us to obtain precise answers, leading to enhanced clarity of these issues. Importantly, our findings suggest that the medium to low expressing Trpv1+ neurons indeed represent the population that transmit nociceptive signaling associated with tissue-damage and this population is coincident with opioid-induced anti-nociception. Deeper analysis of the coexpression matrix of these nociceptive and analgesic target genes provides additional rationale for identifying susceptible neuronal populations for early-stage novel therapeutics development, with either pharmacological agents or other methods such as in vivo gene transfer, by better defining the neuronal populations directly responsible for clinically relevant pain and pain control.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by NIH Clinical Center.

The project was conceptualized by MI, WM, and MS. Experiments were performed by WM, with technical involvement from MS, DM, and TG. Staining and slide scanning was performed by WM with assistance from AM and supervision from DM in fluorescence microscopy and image analysis. Stained sections were analyzed in Adobe Photoshop by WM with assistance from AM and MS. Visualizations, final figures, and formal analysis were generated primarily by MS, with assistance from WM, AM, and MD. Initial drafts of the manuscript were prepared by MS, WM, and MI with editing and suggestions from all of the authors. Funding was obtained by WM, DM, MI, and AM. The project was supervised by MS, AM, and MI. All authors revised and approved the final manuscript.

Funding for this work was supported by the Intramural Research Program of the NIH Clinical Center (1ZIACL090033-09, 1ZIACL090034-09, and 1ZIACL090035-08 to AM), and the NINDS. This work was also supported by a funds from the National Center for Complementary and Integrative Health (1ZIAAT000017-03), and the Office of Behavioral and Social Sciences Research. TG was supported by the Japan Society for the Promotion of Science Overseas Research Fellowship. This research was made possible through the National Institutes of Health (NIH) Medical Research Scholars Program, a public-private partnership supported jointly by the NIH and generous contributions to the Foundation for the NIH from the Doris Duke Charitable Foundation, Genentech, the American Association for Dental Research, the Colgate-Palmolive Company, Elsevier, alumni of student research programs, and other individual supporters via contributions to the Foundation for the National Institutes of Health.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.