NaV1.7 mRNA and Protein Expression in Resident Neurons of the Human Spinal Dorsal Horn Authors: Stephanie Shiers, Nwasinachi A Ezeji, Carla Lima‐Truax, Jeffrey L Krajewski, Geoffrey Funk, Anna Cervantes, Peter Horton, Gregory Dussor, Stephanie Hennen, Theodore J Price Published: 2026 ABSTRACT ABSTRACTThe voltage‐gated sodium channel NaV1.7 is a pain target supported by human genetics, and many compounds have been developed to inhibit NaV1.7 but have disappointed in clinical trials due to cardiovascular effects. Because some recent reports suggest that pharmacological inhibition of NaV1.7 in the rodent spinal dorsal horn can achieve pain relief, we sought to better understand NaV1.7 expression in the human spinal cord. We report that NaV1.7 mRNA is expressed in putative projection neurons (NK1R+, GPR83+) in the human spinal dorsal horn, predominantly in lamina I and II, as well as in deep dorsal horn neurons. NaV1.7 mRNA was also detected in preganglionic parasympathetic and sympathetic neurons, motor neurons in the ventral horn, and ependymal cells lining the central canal. NaV1.7 protein was predominantly found in the central axons of sensory neurons terminating in lamina I–II and colocalized, in part, with presynaptic markers like Bassoon and CGRP. Postsynaptically, NaV1.7 protein was detectable in the soma of motor neurons but was more elusive in dorsal horn populations due to the abundance of presynaptic signal. However, NaV1.7 protein was detected in the axon initial segment of some resident dorsal horn neurons and in axons entering the anterior commissure. Given that projection neurons are critical for conveying nociceptive information from the dorsal horn to the brain, these data support that dorsal horn NaV1.7 expression may play an unappreciated role in pain phenotypes observed in humans with genetic SCN9A mutations.Keywords: dorsal horn, NaV1.7, spinal cord Introduction Loss‐ and gain‐of‐function mutations in the gene encoding voltage‐gated sodium channel (VGSC) NaV1.7 (Baker and Nassar 2020), SCN9A, result in congenital pain insensitivity (Cox et al. 2006) or in extreme pain disorders like primary erythromelalgia (Mann et al. 2019; Yang et al. 2004), paroxysmal extreme pain disorder (Fertleman et al. 2006; Hua et al. 2022), and idiopathic small fiber neuropathy (Faber et al. 2012), respectively. NaV1.7 is predominantly expressed in the peripheral nervous system, including sympathetic neurons and nociceptive sensory neurons in the dorsal root ganglia (DRG) and trigeminal ganglia (Hameed 2019). It is localized to the membrane of sensory neurons innervating the skin, muscle, and other organs, and central axons that terminate in the dorsal horn of the spinal cord (Black et al. 2012; S. Shiers et al. 2020; S. I. Shiers et al. 2021). It is a key regulator of neuronal excitability as it mediates Na+ currents during membrane depolarization and action potential firing (McDermott et al. 2019; Middleton et al. 2022) and is upregulated in pain conditions in both rodents and humans (Alvarez et al. 2021; Black et al. 2008; Hameed 2019; Li et al. 2018; Liu et al. 2021; Nascimento et al. 2022; Sun et al. 2018). The peripheral nociceptive system has been considered as the primary site of action for therapeutic targeting of Nav1.7 (Chew et al. 2019). Several NaV1.7 inhibitors have been developed, and a series of clinical trials have been conducted with mixed reports on pain outcomes and cardiovascular safety (Alles and Smith 2021; Biogen 2021; Dormer et al. 2023; Eagles et al. 2022; Kingwell 2019; McDonnell et al. 2018; Price et al. 2017). While some of the lack of efficacy outcomes could be due to poor pharmacokinetics and/or limited selectivity toward other sodium channels (Kraus et al. 2021), recent studies demonstrate that inhibiting Nav1.7 on sympathetic neurons causes cardiovascular effects that reduce the safety profile of these direct inhibitors of Nav1.7 (Dib‐Hajj et al. 2013; Mulcahy et al. 2024). While mostly absent in the rodent brain (Allen Institute for Brain Science 2022b; Lein et al. 2007) and human cortex (Allen Institute for Brain Science 2022a), NaV1.7 protein is localized to presynaptic axons in both the rodent and human spinal dorsal horn (Black et al. 2012; S. I. Shiers et al. 2021). Recent studies have suggested that NaV1.7 channels in the CNS are potentially important for analgesia. A central analgesic mechanism for NaV1.7 was recently proposed wherein NaV1.7‐null mice that were pain insensitive retained nociceptor firing properties but displayed opioid‐dependent impaired synaptic transmission and neurotransmitter release at sensory neuron presynaptic terminals in the spinal dorsal horn (MacDonald et al. 2021). There is also evidence for NaV1.7 expression and function postsynaptically within the spinal cord. For instance, a recent study demonstrated that dorsal horn neuron Nav1.7 expression is important for spinal cord injury pain (Fu et al. 2023). NaV1.7 mRNA has also been identified in several subsets of mouse spinal cord neurons using single‐cell RNA sequencing (Russ et al. 2021), including Phox2a+ projection neurons that convey pain, itch, and temperature information to the brain through the anterolateral system (ALS) (Bell et al. 2023). Another study using electron microscopy revealed that a substantial proportion of NaV1.7 immunoreactivity was localized to dendrites of mouse spinal dorsal horn neurons, but these authors proposed that the protein originated from presynaptic fibers and was transferred to postsynaptic sites through a mechanism that remains to be elucidated. Given the evidence described above in the mouse, we hypothesized that SCN9A mRNA and NaV1.7 protein might be expressed by human dorsal horn neurons, including projection neurons. Here, we used a variety of techniques to present evidence of NaV1.7 mRNA and protein expression in a diverse subset of human dorsal horn neurons, including putative projection neurons. Our findings support that the functional role of spinal Nav1.7 in putative projection neurons should be investigated, as its expression in these cells may be important for human nociception. Materials and Methods Human Tissue Procurement All human tissue procurement procedures were approved by the Institutional Review Boards (protocol Legacy‐MR‐15‐237) at the University of Texas at Dallas. All tissues were procured from neurologic determination of death organ donors through a collaboration with the Southwest Transplant Alliance (STA). The STA obtained informed consent for research tissue donation from the donor's family. Human spinal cords (lumbar to sacral levels) were recovered using a ventral approach (Valtcheva et al. 2016) from organ donors within 4 h after cross‐clamp, frozen immediately in powdered dry ice, and stored in a −80°C freezer. Details about the procurement process and tissue processing can be found on protocols.io (S. Shiers et al. 2024). Sex was not considered as a biological variable, as several studies have found no differences in Nav1.7 expression in DRG (Chang et al. 2018; Ghovanloo et al. 2024), and all experiments consist of pooled male/female data. Donor demographics and sample information are provided in Table 1. TABLE 1Donor demographics.Donor IDAgeSexCODSpinal levelExperimentUTD‐DN004244FCVA/strokeLumbar 4SCN9A/TACR1SCN9A/GPR83NaV1.7/NeuNNaV1.7/MAP2UTD‐DN008756MHead trauma/GSWLumbar 5/Sacral 1SCN9A/TACR1SCN9A/GPR83NaV1.7/NeuNNaV1.7/MAP2UTD‐DN007740FCVA/strokeLumbar 1SCN9A/TACR1SCN9A/GPR83NaV1.7/NeuNUTD‐DN004445FCVA/strokeLumbar 3SCN9A/TACR1SCN9A/GPR83NaV1.7/NeuNNaV1.7/MAP2UTD‐DN002318FAnoxia/seizureSacral 1NaV1.7/CGRP/BassoonUTD‐DN001424MHead trauma/MVALumbar 4NaV1.7/CGRP/BassoonUTD‐DN001153FCVA/strokeLumbar 5/Sacral 1NaV1.7/CGRP/BassoonUTD‐DN003769MCVA/strokeLumbar 5NaV1.7/CGRP/BassoonUTD‐DN009122MOverdose/cardiovascularLumbar 5NaV1.7/MAP2 Donor demographics. RNAscope In Situ Hybridization Frozen human spinal cords (∼1.5‐cm‐long segments) were gradually embedded in OCT in a cryomold by adding small volumes of OCT over dry ice to avoid tissue thawing. Human spinal cords were sectioned at 20 µm onto SuperFrost Plus charged slides (Fisher Scientific; Cat 12‐550‐15). Sections were only briefly thawed in order to adhere to the slide but were immediately returned to the −20°C cryostat chamber until the completion of sectioning. Three sections from each donor were stained in each experiment, and three to four donors were used in each experiment. The slides were removed from the cryostat, thawed at 37°C for exactly 1 min, and then immediately transferred to 10% formalin (Fisher Scientific; Cat 23–245684) for 15 min at room temperature. The sections were then sequentially dehydrated in 50% ethanol (5 min; Fisher Scientific; Cat 04‐355‐223), 70% ethanol (5 min), and twice in 100% ethanol (5 min each) at room temperature. The slides were air‐dried briefly, and then boundaries were drawn around each section using a hydrophobic pen (ImmEdge PAP pen; Vector Labs). Once the hydrophobic boundaries had dried, the slides were immediately processed for RNAscope in situ hybridization. RNAscope in situ hybridization multiplex version 2 (Advanced Cell Diagnostics; Cat 323100) was conducted on human spinal cord sections using the fresh frozen protocol as described by ACD (acdbio; manual # 323100‐USM with rev date: 02272019). Hydrogen peroxide (ACD; Cat 322381) was applied to each section until fully covered and incubated for 10 min at room temperature. The slides were then washed in distilled water and then incubated one at a time in protease III reagent (ACD; Cat 322381) for 10 s at room temperature. Slides were then washed briefly in 1x phosphate‐buffered saline (PBS, pH 7.4) at room temperature. A positive and negative control was run on single sections from each spinal cord for every RNAscope experiment. Due to the high lipofuscin content and high autofluorescence in human nervous tissues, only two mRNAs of interest (SCN9A [ACD; Cat 562251] with TACR1 [ACD; Cat 310701] and SCN9A with GPR83 [ACD; Cat 1082151]) were stained at a time (in the Cy3 and Cy5 channels), and a third channel (488 channel) was left unstained to reveal background signal. The only exception to this was with the positive and negative control probes, as these contain a premixed combination of three mRNA probes. The positive control probe cocktail (ACD; Cat 320861) contains probes for high‐, medium‐, and low‐expressing mRNAs that are present in all cells (ubiquitin C > peptidyl‐prolyl cis–trans isomerase B > DNA‐directed RNA polymerase II subunit RPB1) and allows us to gauge tissue quality and experimental conditions. All tissues showed a robust signal for all three positive control probes. A negative control probe cocktail (ACD; Cat 320871) against the bacterial DapB gene was used to check the background label due to the experimental assay. Each slide was then placed in a prewarmed humidity control tray (ACD; Cat 321710) containing dampened filter paper (ThermoFisher Scientific; Cat 84784), and the probes were pipetted onto each section until fully covered. This was performed one slide at a time to avoid liquid evaporation and section drying. The humidity control tray was placed in a HybEZ oven (ACD; Cat 321710) for 2 h at 40°C. Following probe incubation, the slides were washed two times in 1x RNAscope wash buffer (ACD; Cat 310091) and then placed in 5x SSC buffer (Sigma; Cat S6639) overnight at room temperature. The following morning, the slides were washed two times in 1x RNAscope wash buffer (ACD; Cat 310091) and placed in the 40°C oven for 30 min after submersion in AMP‐1 reagent. Washes and amplification were repeated using AMP‐2 and AMP‐3 reagents with a 30‐ and 15‐min incubation period, respectively. HRP‐C1 reagent was applied to all sections and then incubated in the oven at 40°C for 15 min. The slides were then washed in 1x RNAscope wash buffer (ACD; Cat 310091). TSA Plus Akoya Dyes in Fluorescin, Cy‐3, and Cy‐5 (Akoya; NEL741001KT, NEL744001KT, NEL745001KT) were prepared at 1:1000 in TSA buffer (ACD; Cat 322809). The Akoya dye assigned to the Channel 1 probe was applied to each section until fully covered and incubated for 30 min in the 40°C oven. The slides were washed and then covered in HRP blocker (ACD; Cat 323110) for 15 min at 40°C. The slides were washed again, and then the same steps were repeated using HRP‐C2 and HRP‐C3 reagents with their assigned Akoya dye. DAPI (Cayman Chemical; Cat 14285) was prepared at 1:5000 in RNAscope wash buffer and was applied to each section for 1 min at room temperature. The slides were then washed in RNAscope wash buffer, air‐dried, and cover‐slipped (Globe Scientific; Cat 1415‐15) with Prolong Gold Antifade mounting medium (Fisher Scientific; Cat P36930). RNAscope In Situ Hybridization Imaging and Analysis One side of each spinal cord section (three sections per donor, three to four donors total) was imaged mosaically on an FV4000 confocal microscope (Evident Scientific), which has autofocus and autostitching capabilities. All mosaic images were autostitched 20x confocal xy images and were not projected z stacks. The acquisition parameters were set based on guidelines for the FV4000 provided by Evident Scientific. The raw image files were analyzed in CellSens (Olympus; v1.18). The True black lipofuscin quencher (used in immunofluorescence) is not compatible with RNAscope. Large globular structures and/or signal that auto‐fluoresced in all channels and were visible in the unstained 488 channel were considered to be background lipofuscin and were not considered to be mRNA signal. Aside from adjusting brightness/contrast, we performed no digital image processing to subtract the background. Images of sections that were tilted were rotated so that laminar boundaries were more distinguishable. The DAPI co‐stain was useful in defining the white versus gray matter boundaries and in visualizing the substantia gelatinosa. The spinal level of each section was determined by comparing the images with a human spinal cord atlas (Sengul et al. 2013) and is indicated in Table 1. Laminar boundaries in the human spinal cord atlas (Sengul et al. 2013) were measured with a ruler and converted to micrometers using the atlas's scalebar. The mosaic images were opened in CellSens, and micrometer guidelines were drawn using the polyline tool for each laminar boundary determined in the atlas. The only boundary exceptions made were with the lamina II boundaries, which were clearly defined by the outline of the substantia gelatinosa but did not always completely align with the atlas. The perimeter of each lamina was traced using the closed polygon tool. The number of DAPI+ nuclei that were positive for each mRNA target was counted within each lamina. The total number of nuclei that were positive for either SCN9A, TACR1, and/or GPR83 is shown in Table 2. It is important to note that DAPI weakly stains the nuclei of larger‐sized neurons, an observation we have frequently noted in our many experiments working with human nervous tissues. In the circumstances where we saw robust mRNA signal but faint or diminished nuclei signal, the image was brightened to confirm that a nucleus was present. For purposes of figure display, this may not be obvious. Graphs were generated using GraphPad Prism version 8.4.3 (GraphPad Software, Inc., San Diego, CA, USA). TABLE 2Number of positive SCN9A, TACR1, and/or GPR83 nuclei within the dorsal horn per donor.Donor IDExperimentSCN9A+ cellsExperimentSCN9A+ cellsUTD‐DN0042 SCN9A/TACR1 1007 SCN9A/GPR83 1596UTD‐DN0087 SCN9A/TACR1 2468 SCN9A/GPR83 1786UTD‐DN0077 SCN9A/TACR1 510 SCN9A/GPR83 531UTD‐DN0044 SCN9A/TACR1 1452 SCN9A/GPR83 1153 Number of positive SCN9A, TACR1, and/or GPR83 nuclei within the dorsal horn per donor. Immunofluorescence and Imaging Human spinal cords were sectioned at 20 µm onto SuperFrost Plus charged slides (Fisher Scientific; Cat 12‐550‐15). Sections were only briefly thawed in order to adhere to the slide but were immediately returned to the −20°C cryostat chamber until completion of sectioning. Three sections from each donor were stained in each experiment, and three to four donors were used in each experiment. A negative control section from each spinal cord was exposed to all reagents except for the primary antibody. Slides were removed from the cryostat and immediately transferred to cold 10% formalin (4°C; pH 7.4) for 15 min. The tissues were then dehydrated in 50% ethanol (5 min), 70% ethanol (5 min), 100% ethanol (5 min), and 100% ethanol (5 min) at room temperature. The slides were air‐dried briefly, and then boundaries were drawn around each section using a hydrophobic pen (ImmEdge PAP pen, Vector Labs). When hydrophobic boundaries had dried, the slides were submerged in blocking buffer (10% Normal Goat Serum, 0.3% Triton‐X 100 in 1x PBS) for 1 h at room temperature. Slides were then rinsed in 1x PBS, placed in a light‐protected, humidity‐controlled tray, and incubated in primary antibodies diluted in blocking buffer overnight at 4°C. A list of the antibodies used can be found in Table 3. The NaV1.7 mouse monoclonal antibody has been knockout validated using immunocytochemistry on mouse cultured DRG neurons and using IHC on the rat brain (Grubinska et al. 2019). We have also shown that this antibody robustly stains human DRG and shows a specific and similar expression pattern to its mRNA in human DRG (S. Shiers et al. 2020). TABLE 3Antibodies used for immunofluorescence.AntibodyVendorCatalog # or Clone #RRIDDilution/conc.Mouse‐anti‐ NaV1.7NeuroMabN68/6AB_21843552 µg/mLRabbit‐anti‐NeuNCell Signaling24307SAB_26511401:500Rabbit‐anti‐CGRPImmunoStar24112AB_5722171:1000Mouse‐anti‐BassoonEnzo Life SciSAP7F407AB_111810582 µg/mLMouse‐anti‐Ankyrin GNeuroMabN106/36AB_106730302 µg/mLChicken‐anti‐MAP2Novus BioNB300‐213AB_21381782 µg/mLGoat‐anti‐rabbit H&L 647ThermoFisherA21245AB_25358131:2000Goat‐anti‐chicken H&L 647ThermoFisherA21449AB_25358661:2000Goat‐anti‐mouse IgG2a 555ThermoFisherA21137AB_25357761:2000Goat‐anti‐mouse IgG2a 488ThermoFisherAB21131AB_25357711:2000Goat‐anti‐mouse IgG1 488ThermoFisherA21121AB_25357641:2000Goat‐anti‐mouse IgG1 555ThermoFisherA21127AB_25357691:2000 Antibodies used for immunofluorescence. The next day, slides were washed in 1x PBS and then incubated in their respective secondary antibody diluted at 1:2000 with DAPI (1:5000; Cayman Chemical; Cat 14285) in blocking buffer for 1 h at room temperature. The sections were washed in 1x PBS and then covered with True Black (20% diluted in 70% Ethanol), a blocker of lipofuscin, for 1 min. Sections were then washed in water, air‐dried, and coverslipped with Prolong Gold Antifade reagent (Cat P36930, Fisher Scientific). One side of each spinal cord section (three sections per donor) was imaged mosaically on an FV3000 or FV4000 confocal microscope (Evident Scientific), which has autofocus and autostitching capabilities. All mosaic images were autostitched 20x confocal xy images and were not projected z stacks. Additionally, 10x, 20x, and 40x xy images were also acquired on either a FV3000 or FV4000 confocal microscope. The acquisition parameters were set based on guidelines for each microscope provided by Evident Scientific. The raw image files were visualized in CellSens (Olympus; v1.18). Images were brightened and contrasted in CellSens, and the negative control was equally adjusted. Results SCN9A mRNA Is Expressed in TACR1+ Putative Projection Neurons in the Human Spinal Dorsal Horn The substance P receptor (protein: Nk1r, mRNA: Tacr1) labels a subset of projection neurons in the rodent spinal dorsal horn that transmit pain and itch information to the parabrachial nucleus in the brainstem (Blomqvist and Mackerlova 1995; Carstens et al. 2010; Mantyh et al. 1997). These neurons are localized to the superficial laminae, primarily lamina I (Carstens et al. 2010; Mantyh et al. 1997), but they are also found in the deep dorsal horn, mostly in lamina V (Brown et al. 1995). It is important to note that the molecular identity of human projection neurons remains unknown, and as such, we refer to the large TACR1/Nk1r‐ and GPR83‐expressing neurons in lamina I and V of the dorsal horn as putative projection neurons. Although Nk1r/Tacr1 is not exclusive to the projection neuron population (Russ et al. 2021), it is commonly used to identify dorsal horn projection neurons (Marshall et al. 1996). We assessed the distribution of SCN9A (NaV1.7) and TACR1 (Nk1r) mRNA in the human spinal dorsal horn using RNAscope. We found that SCN9A and TACR1 mRNAs were detected in cells throughout lamina I–V in the human spinal dorsal horn (Figure 1A–F). In the dorsal horn (laminae I–V), virtually all of the TACR1+ cells were SCN9A+ (92%), but only 45% of SCN9A+ cells were TACR1+ (Figure 1G,H). SCN9A mRNA was also detected in ependymal cells near the central canal (Figure 1I), preganglionic parasympathetic neurons in the sacral parasympathetic nucleus (Figure 1J), motor neurons in the ventral horn (Figure 1K), and preganglionic sympathetic neurons in the intermediolateral column (Figure 1L). Distribution of SCN9A (NaV1.7) and TACR1 (NK1R) mRNAs in the human spinal dorsal horn using RNAscope. (A) Mosaic image of a human spinal cord (lumbar 5/sacral 1) labeled with RNAscope in situ hybridization for SCN9A (red) and TACR1 (green) mRNAs and co‐stained with DAPI (cyan). The 488 channel was left unstained (green) to reveal background autofluorescence and lipofuscin, which is present in all human neurons. Magenta line outlines gray matter. Higher magnification images for each channel are shown for (B) lamina I, (C) lamina II, (D) lamina III, (E) lamina IV, and (F) lamina V. (G) Percentage of SCN9A+ nuclei that coexpressed TACR1 (red bar) and percentage of TACR1+ nuclei that coexpressed SCN9A (blue bar) for each lamina (LI–LV). (H) Percentage of TACR1+ nuclei that were copositive for SCN9A (blue bar) and percentage of SCN9A+ nuclei that were copositive for TACR1 (red bar) for the entire dorsal horn (combined data of H). (I) SCN9A mRNA was also detected in (I) nuclei lining the central canal, likely ependymal cells, (J) preganglionic parasympathetic neurons in the sacral parasympathetic nucleus (SPSy), (K) motor neurons in the ventral horn, particularly Pes9, which contains motor neurons associated with the feet, and (L) preganglionic sympathetic neurons in the intermediolateral column (IML) from a lumbar 1 spinal cord section. LI–LV, lamina I–V; SPSy, sacral parasympathetic nucleus; Pes9, motor neurons of the foot; CC, central canal. Sample size: n = 4. Scale bars: A = 500 µm; B–F = 20 µm; I–L = 50 µm. SCN9A mRNA Is Expressed in GPR83+ Putative Projection Neurons in the Human Spinal Dorsal Horn The G protein‐coupled receptor 83 (GPR83) is expressed by a second projection neuron population found in lamina I–II of the rodent spinal dorsal horn. Both the GPR83 and Tacr1 projection neuron populations show little overlap (∼20% overlap), terminate in different subnuclei of the parabrachial nucleus, and, together, comprise ∼88% of all ALS projection neurons (Choi et al. 2020). However, a recent report with a new transgenic line recapitulates older findings (Cameron et al. 2015), which indicate that Tacr1 is expressed in 90% of all ALS projection neurons (Ma et al. 2025), suggesting that these populations may not be as distinct as previously reported. To further investigate the expression of SCN9A in projection neurons, we conducted RNAscope in situ hybridization in combination with GPR83 (Figure 2A). SCN9A and GPR83 were detected in cells throughout lamina I–V (Figure 2B–F). SCN9A was coexpressed in the majority of GPR83‐expressing cells across lamina I–V (Figure 2G), with aggregate data for the region indicating that 89% of all GPR83‐expressing cells coexpressed SCN9A mRNA (Figure 2H). This distribution closely resembles recently published human spinal cord spatial transcriptomic data (Yadav et al. 2023), in which SCN9A, TACR1, and GPR83 mRNAs were found throughout the dorsal and ventral horns (Figure 3A). SCN9A was also detected in a broad population of excitatory and inhibitory dorsal horn neurons using single‐nucleus RNA sequencing of human spinal cord (Yadav et al. 2023), many of which co‐expressed TACR1 and/or GPR83 (Figure 3B). Distribution of SCN9A (NaV1.7) and GPR83 mRNAs in the human spinal dorsal horn using RNAscope. (A) Mosaic image of a human spinal cord section (lumbar 3) labeled with RNAscope in situ hybridization for SCN9A (red) and GPR83 (green) mRNAs and co‐stained with DAPI (cyan). The 488 channel was left unstained (green) to reveal background autofluorescence and lipofuscin, which is present in all human neurons. Magenta line outlines gray matter. Higher magnification images for each channel are shown for (B) lamina I, (C) lamina II, (D) lamina III, (E) lamina IV, and (F) lamina V. (G) Percentage of SCN9A+ nuclei in that coexpressed GPR83 (red bar) and percentage of GPR83+ nuclei that coexpressed SCN9A (blue bar) for each lamina (LI–LV). (H) Percentage of GPR83+ nuclei that were copositive for SCN9A (blue bar) and percentage of SCN9A+ nuclei that were copositive for GPR83 (red bar) for the entire dorsal horn (combined data of H). LI–LV, lamina I–V; D, dorsal nucleus; CC, central canal. Sample size: n = 4. Scale bars: A = 500 µm; B–F = 20 µm. Spatial and single‐nuclear RNA‐sequencing detection of SCN9A and TACR1 in the human spinal dorsal horn. (A) Normalized spatial transcriptomic gene expression for SCN9A, GPR83, and TACR1 per barcoded spot on aggregated human lumbar spinal cord sections (Yadav et al. 2023). Solid lines mark gray matter boundaries. (B) Dot plot showing the average gene expression for SCN9A, GPR83, and TACR1 for each human spinal cord cluster identified using single‐nucleus RNA sequencing. Data obtained from https://vmenon.shinyapps.io/humanspinalcord/. NaV1.7 Protein is Localized Presynaptically in the Human Spinal Cord NaV1.7 protein staining in the human spinal cord was robustly found in the dorsal rootlet and lamina I and II, as we have previously reported (S. I. Shiers et al. 2021) (Figure 4A,B). NaV1.7 staining gave an axonal and synaptic (neuropil) pattern throughout lamina I–II (Figure 4B). NaV1.7 was detected in the cytoplasm of motor neurons in the ventral horn (Figure 4C), axons in the anterior commissure (Figure 4D), and ependymal cells lining the central canal, corroborating its mRNA expression in those cells (Figure 4D). NaV1.7 protein expression in the human lumbar spinal cord. Mosaic image of NaV1.7 (red) protein staining in the human lumbar spinal cord (lumbar 5), co‐stained with DAPI (blue). White outline demarcates the gray matter. (B) NaV1.7 protein gave an axonal and neuropil (synaptic) pattern in the spinal dorsal horn, mostly localized to lamina I–II (LI–LII). (C) Nav1.7 protein was also detected in the cytoplasm of large motor neurons in the ventral horn and (D) in axons in the anterior commissure (white arrows), as well as the cytoplasm of ependymal cells (cells outlining the central canal). (E) Representative 10x magnification confocal image of NaV1.7 protein (red) co‐stained with the nociceptive presynaptic marker CGRP (blue), the presynaptic active zone marker Bassoon (green), and DAPI (cyan) in LI–LII of the spinal dorsal horn. (F) Overlay image (488, 555, 647, DAPI) of LI–LII in the negative control that was exposed to all of the same reagents except primary antibody and imaged and adjusted to the same settings as shown in E. (G) A higher magnification (100x) confocal view of the area outlined in yellow in the overlay image in C. (F) A digitally magnified view of the area outlined in white in E showing NaV1.7 signal colocalized with CGRP and/or Bassoon (white arrows) or in close proximity to these proteins. LI–LV, lamina I–V; CC, central canal. Sample size: n = 4. Scale bars: A = 1 mm; B–C = 500 µm; D–F = 200 µm; G = 20 µm; H = 10 µm. When colabeled with CGRP, a neuropeptide that we have shown is expressed in all human nociceptors (S. Shiers et al. 2020; S. I. Shiers et al. 2021), and the presynaptic active‐zone protein, Bassoon (Figure 4E), the majority of the NaV1.7 signal appeared to colocalize with, or to be in close proximity to, these presynaptic markers above background signal (Figure 4F). However, higher magnification imaging revealed that not all of the NaV1.7 signal in the superficial laminae appeared to be localized with these presynaptic proteins (Figure 4G,H), which could be indicative of differences in the subcellular localization of these proteins within the presynaptic compartment, traveling DRG primary afferents that have not yet met their termination point within the dorsal horn, or even dendritic or axonal localization of NaV1.7 in resident dorsal horn neurons. Indeed, NaV1.7 was also detected in axons in deeper lamina (LIII–LV), with the highest abundance in lamina V (Figure 4D) and also in axons entering the anterior commissure (near lamina X and the central canal), where dorsal horn projection neurons cross the midline to enter the contralateral spinal thalamic tract. As tracing studies in cats and macaques have demonstrated that neurons projecting to the spinal thalamic tract originate in lamina I and lamina V, it is possible that the non‐presynaptic signal in the superficial lamina as well as the axonal signal in the deep lamina and anterior commissure could represent NaV1.7 protein expressed by projection neurons (Craig 2006; Jones et al. 1987; Willis et al. 1979), a hypothesis we explored with additional experiments. Evidence for Postsynaptic NaV1.7 Protein in the Human Spinal Cord In an effort to identify postsynaptic NaV1.7 immunoreactivity in the spinal dorsal horn, we co‐stained NaV1.7 with a variety of neuronal subcellular protein markers. While Nav1.7 is localized within the cytoplasm and membrane of the cell body of sensory neurons, as well as their peripheral and central axons and terminal synapses, its expression in the CNS may differ. Other VGSCs are known to be expressed in different cellular compartments in the CNS, such as the synapse, dendrites, axon initial segment (AIS), nodes of Ranvier, and axons of neurons throughout the spinal cord and brain (Hanson et al. 2004; Schaller and Caldwell 2003). First, we assessed NaV1.7 in combination with the neuronal soma and nuclear marker, NeuN (Figure 5A), to determine if NaV1.7 was detected in the soma of resident dorsal horn neurons. However, we found no convincing evidence that Nav1.7 was expressed in the cell bodies of neurons in lamina I–V (Figure 5B–F). Next, we assessed if Nav1.7 was coexpressed with the cytoskeletal marker, MAP2, which is primarily localized to dendrites (Figure 6A). We assessed the dorsal horn from three organ donors but did not find evidence for NaV1.7 signal localized to the dendritic compartment of resident dorsal horn neurons (Figure 6B–D). In the anterior commissure, we again observed intensely labeled NaV1.7 axonal fibers crossing to the other side of the spinal cord (Figure 6E), but we could not determine what neurons these axons originated from; however, it is likely they are projection neurons, as this is a major hub for spinal neurotransmission. Investigation of NaV1.7 protein in the soma of resident neurons in the dorsal horn. (A) Mosaic image of a human spinal cord section (lumbar 5–sacral 1) immunolabeled with NaV1.7 (red), NeuN (green, soma and nucleus of neurons), and DAPI (blue, nuclei). Magenta line outlines the gray matter. Representative 40x images of NaV1.7 (red), NeuN (green), and DAPI (blue) in (B) lamina I, (C) lamina II, (D) lamina III, (E) lamina IV, and (F) lamina V and their corresponding negative control that was exposed to all of the same reagents except primary antibody and imaged and adjusted with the same settings of each subregion. NaV1.7 was not detected in the soma of any of the dorsal horn (lamina I–V) neurons. However, it was detected in the cell bodies of motor neurons and in preganglionic parasympathetic neurons. LI–LV, lamina I–V; SPSy, sacral parasympathetic nucleus; Pes9, motor neurons of the foot; CC, central canal. Sample size: n = 4. Scale bars: A = 500 µm; B–F = 100 µm. Evidence for postsynaptic NaV1.7 expression in the human spinal cord. (A) Representative 10x image of NaV1.7 (red), MAP2 (green), and DAPI (blue) staining in the human lumbar dorsal horn. The white outlines lamina I and lamina II. Representative 40x images of NaV1.7 (red), MAP2 (green), and DAPI (blue) in (B) lamina I and (C) lamina II. White arrows point to the MAP2 signal that is localized around resident neurons (appears to be the plasma membrane) but is absent of NaV1.7 signal. The image inset in panel C shows a 100x image of a large, LII neuron with a large apical dendrite that is devoid of NaV1.7 signal. (D) Representative 20x image of NaV1.7‐positive axonal fibers in the deeper lamina around LIV–LV. The white arrow points to NaV1.7 and MAP2 copositive signal (yellow in overlay) that does not have a nucleus and is not a cell body. (E) Representative 20x image of NaV1.7 staining in the anterior commissure (ac), where intensely labeled NaV1.7‐positive axons are highlighted (white arrow). (F) A 100x image of NaV1.7 (red), Ankyrin‐G (green), and DAPI (blue) staining in a motor neuron in the ventral horn. (G) A cropped, zoomed‐in image of NaV1.7 (red), Ankyrin‐G (green), and DAPI (blue) signal in a lamina II dorsal horn neuron. LI–LV, lamina I–V; ac, anterior commissure. Sample size: n = 3–4. Scale bars: A = 200 µm; B and C = 50 µm; D and E = 100 µm; F = 20 µm; G = 5 µm. Interestingly, we noted that NaV1.7 was localized not only to the soma of motor neurons but also to what appeared to be their AIS. Recent work with NaV1.7 in the rodent DRG found that sensory neurons contain an AIS, that it is enriched with NaV1.7, and that its localization there is critical for spontaneous activity in neuropathic pain (Nascimento et al. 2022). Other VGSCs are also known to be enriched in the AIS in the CNS (Leterrier 2018). To confirm our observation in motor neurons, we co‐stained NaV1.7 with the AIS marker, Ankyrin‐G, and found that NaV1.7 colocalized with Ankyrin‐G at the AIS of motor neurons in the human ventral horn (Figure 6F). In the dorsal horn, we also found evidence for NaV1.7 localized to the AIS of resident neurons (Figure 6G), but this staining was sparse and was not detected in all neurons. It is important to note that postsynaptic NaV1.7 signal within LI–LII was difficult to detect due to the intense presynaptic neuropil labeling throughout the region. NaV1.7 expression within resident spinal dorsal horn neurons may have been low, or below the level of detection. However, increasing the imaging parameters to capture this signal would result in oversaturation of the presynaptic label and make it so that cellular structures were no longer distinguishable. As such, we kept the imaging parameters low so that the anatomy could be visualized through the neuropil staining. This may be indicative of why the detection of NaV1.7 postsynaptic label was lower than expected. However, given that we observed robust NaV1.7 mRNA expression in most TACR1+ and GPR83+ neurons and found evidence of NaV1.7 localization at AISs in dorsal and ventral horn neurons, our data support that when the mRNA is translated, it is localized to the membrane of intrinsic dorsal horn axons. This is further supported by NaV1.7 axonal labeling in the anterior commissure. Discussion In the present study, we found evidence for the presence of NaV1.7 expression in intrinsic neurons of the human spinal cord, including in the dorsal and ventral horns. First, we identified that SCN9A (protein: NaV1.7) mRNA was detected in ∼90% of all GPR83+ and TACR1+ (protein: Nk1r) neurons, which are known markers of projection neurons in rodents (Cameron et al. 2015; Carstens et al. 2010; Choi et al. 2020; Ma et al. 2025; Mantyh et al. 1997). In rodents, the Nk1r and GPR83 are expressed by a subset of large‐diameter projection neurons that send projections to higher brain regions like the parabrachial nucleus, and are found predominantly in lamina I and in the deeper dorsal horn around lamina V (Brown et al. 1995; Choi et al. 2020; Marshall et al. 1996; Todd et al. 2000). It is important to note that not all rodent projection neurons are Tacr1+ (Nk1r+), and interneurons as well as motor neurons also express this gene (Russ et al. 2021). Single‐nucleus RNA sequencing recapitulates this expression pattern as SCN9A, GPR83, and TACR1 mRNAs were co‐expressed in a wide range of human spinal neuronal populations, including motor neurons (Yadav et al. 2023). The high co‐expression of SCN9A with known rodent projection neuron markers in LI–LII and LV supports that a subset of the SCN9A, GPR83, and TACR1 co‐expressing neurons are likely projection neurons. Another study detected Scn9a mRNA in a variety of Tacr1 co‐expressing neuron populations in the mouse spinal cord using single‐cell sequencing (Russ et al. 2021). However, only its motor neuron expression could be validated with in situ hybridization (Allen Institute for Brain Science 2022b; Alles et al. 2020). While sensitivity issues could underlie these technical differences, detection of NaV1.7 mRNA and protein in rodent spinal dorsal horn neurons has also been inconclusive, likely due to its unique subcellular localization to the neuronal membrane, which could comprise the soma, dendrites, AIS, nodes of Ranvier, and/or synapse. Indeed, most reports suggest that spinal NaV1.7 protein expression is entirely presynaptic due to its robust neuropil staining pattern within lamina I–II (Black et al. 2012; S. I. Shiers et al. 2021). However, immuno‐electron microscopy detected NaV1.7 protein localized to dendrites of mouse dorsal horn neurons, but it was hypothesized that these proteins originated in the presynaptic compartment and were transferred to postsynaptic sites through an unknown mechanism (Alles et al. 2020). While we also detected intense presynaptic NaV1.7 labeling in lamina I–II of the human spinal cord, not all of its expression was colocalized with presynaptic markers like CGRP and Bassoon. NaV1.7 protein was also localized to axons in the deeper lamina, particularly lamina IV and V, and to axons in the anterior commissure, the white matter tract connecting both sides of the spinal cord, and an important relay for dorsal horn projection neurons transmitting nociceptive information to the contralateral spinothalamic tract (Ku and Morrison 2022). Because axon tracing methods cannot readily be employed in the human spinal cord, we could not identify if these were primary afferents or the axons of intrinsic dorsal horn neurons; however, ascending projections of primary afferents ascend in the dorsal columns and do not cross the midline, so it is exceedingly unlikely that these commissural axons are contributed by sensory neurons. Interestingly, we did not detect NaV1.7 localized to dendrites but instead identified its expression in the AISs of some dorsal horn neurons in lamina I–II and also in motor neurons. Other Nav family members are highly concentrated at the AIS, and their localization there is critical for integration of synaptic currents into action potential generation (Grubb and Burrone 2010; Leterrier 2018). Importantly, neurons are known to increase or decrease the size of their AIS as a means to augment or depress their neuronal excitability in response to changing presynaptic input (Grubb and Burrone 2010; Grubb et al. 2011; Kuba et al. 2010). As sensory neuron hyperexcitability and/or spontaneous activity is a major driver of abnormal nociceptive signaling into the dorsal horn in chronic pain conditions, it is possible that NaV1.7 regulation at the AIS in resident dorsal horn neurons is critical for nociceptive processing in the dorsal horn. A recent study found Nav1.7 expression in mouse dorsal horn neurons after spinal cord injury (Fu et al. 2023). This raises the possibility that the presence of mRNA for SCN9A in human dorsal horn neurons sets the stage for upregulation of protein expression after injury. While loss‐of‐function mutations in SCN9A result in pain insensitivity in both rodents and humans (Gingras et al. 2014; Grubinska et al. 2019; Shields et al. 2018), gain‐of‐function mutations engineered in mice to match human mutations that cause pain disorders do not recapitulate the human pain phenotype (Chen et al. 2021). One potential explanation for this is that the loss‐of‐function phenotype is dependent upon sensory neuron expression of NaV1.7, which is conserved across species, and that the loss of function leads to a conserved loss of action potential generation in nociceptors. On the other hand, the human gain‐of‐function pain phenotype may require both peripheral sensitization and spinal amplification. As there has been little convincing evidence for Nav1.7 expression in intrinsic neurons of the mouse spinal cord in the absence of injury, one hypothesis is that spinal amplification of nociceptive signals does not occur in mice but may occur in humans with gain‐of‐function mutations, given the broad expression of SCN9A in dorsal horn neurons, including putative projection neurons. If this is the case, gain‐of‐function mutations in SCN9A could lead to increased excitability of projection neurons, which would be further amplified by increased nociceptive input from hyperexcitable nociceptors in the periphery. However, this mechanistic hypothesis may not be as straightforward as this because SCN9A mRNA was found in many neurons in the dorsal horn, including inhibitory interneurons, by single‐nucleus sequencing (Yadav et al. 2023). Nevertheless, dysregulated circuit dynamics in the dorsal horn caused by gain‐of‐function mutations in Nav1.7 potentially explain the difference in pain phenotype between humans and rodents. In summary, we offer several compelling pieces of evidence to support the existence of NaV1.7 mRNA and protein expression by intrinsic neurons of the human spinal dorsal horn: (1) SCN9A mRNA is expressed by human resident dorsal horn neurons detected by both RNAscope in situ hybridization, spatial sequencing, and single‐nucleus sequencing; (2) ∼90% of GPR83+ and TACR1+ human resident dorsal horn neurons express SCN9A, a subset of which are likely projection neurons; (3) many SCN9A+ and GPR83 or TACR1 co‐expressing neurons were large diameter and were most abundant in laminar regions (lamina I and lamina V) that are known to be enriched with projection neurons; (4) not all NaV1.7 protein signal was limited to the presynaptic compartment as demonstrated with co‐labeling with CGRP and Bassoon; (5) NaV1.7 was detected in the AIS of some resident dorsal horn neurons; and (6) NaV1.7+ axons were detected in the anterior commissure. Funding This work was supported by NIH grant U19NS130608 and a research grant from Grünenthal. Conflicts of Interest The authors declare no conflicts of interest. J.L.K. and C.L. are employees of Eli Lilly. S.H. is an employee of Grünenthal GmbH. Acknowledgments The authors are grateful to the organ donors and their families for the gift of life and research provided by their organ donation. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon request. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon request.