GABAergic and glycinergic inhibitory interneurons in the spinal dorsal horn (SDH) are key modulators of pain sensation and sensitization.29,82,95 These neurons have recently been classified into multiple subpopulations based on their expression of molecular markers such as parvalbumin, neuronal nitric oxide synthase (nNOS), neuropeptide Y (NPY), enkephalin, acetylcholine, dynorphin (pDyn), or galanin.58,87 Genetic ablation studies suggest that these various inhibitory “microcircuits” play distinct, yet overlapping, roles in somatosensory processing, as the lesion of spinal pDyn- or parvalbumin-lineage neurons results in mechanical allodynia,24,69 whereas the ablation of NPY- or Bhlhb5-lineages produces mechanical or chemical itch, respectively.7,49 Notably, the Bhlhb5 lineage includes a large subpopulation of pDyn interneurons.24,49
Understanding the full genetic signature of these diverse classes of interneurons could inform new strategies to modulate neuronal activity in a cell type–specific manner within spinal nociceptive circuits and thereby alleviate persistent pain. Towards this end, the recent employment of complete, unbiased transcriptional analyses has yielded exciting new insight into the molecular architecture of the SDH.12,36,69 However, the majority of these earlier studies either focused exclusively on adult animals or used tissue samples obtained during a brief developmental window.12,42,72 As a result, little is known about how the pattern of gene expression within defined classes of SDH neurons changes during the life span.
Significant evidence suggests that inhibitory circuits in the dorsal horn are functionally immature at birth.9,50 Nociceptive withdrawal reflexes, which are known to be tightly regulated by synaptic inhibition within the dorsal horn,82,95 exhibit enlarged receptive fields and lower mechanical thresholds during the neonatal period in both humans and rodents.3,28 These same features also characterize immature dorsal horn neurons,27 which show an exaggerated afterdischarge in response to sensory stimulation that gradually disappears during postnatal development as GABAergic signaling strengthens.4 In vitro electrophysiological studies have also documented a host of developmental changes in the properties of GABAergic neurons within the SDH, including changes in short-term synaptic plasticity46 and an age-related decrease in innervation by low-threshold primary afferents.20 Unfortunately, the potential genetic underpinnings of these age-dependent changes in the functional properties of inhibitory dorsal horn interneurons remain unclear.
The goal of this study was to elucidate developmental alterations in gene expression in pDyn-lineage SDH interneurons. This population was chosen due to its known importance for suppressing mechanical pain, in light of the well-documented changes in mechanical sensitivity occurring during early life. The results revealed, for the first time, distinct clusters of genes whose expression is significantly upregulated or downregulated from the neonatal period to adulthood. The data also identify genes that are selectively enriched within spinal pDyn neurons in neonates, juveniles, and adults, including a novel specific marker of pDyn neurons in the dorsal horn circuit. Collectively, these findings serve as a solid foundation for further investigations into the detailed molecular mechanisms governing spinal nociceptive processing, and pain sensitivity, in neonates and adolescents.
2. Materials and methods
Homozygous pDyn-IRES-Cre mice55 (Jackson stock #027958) were bred with homozygous R26-CAG-LSL-Sun1-sfGFP-Myc (Jackson stock #021039) to create offspring heterozygous for both alleles. These are referred to as pDyn-GFP mice. For RNA-sequencing (RNA-seq) experiments, 2 males and 2 females at each of 3 developmental time points (postnatal day 7, 21, or 80) were used. For histology and electrophysiology experiments, pDyn-GFP mice of either sex were used at P6-7, P18-21, or P63-80. For electrophysiology experiments, homozygous pDyn-IRES-Cre mice were bred with homozygous R26-LSL-CAG-tdTomato reporter mice (Ai9; Jackson stock #00709) to generate offspring heterozygous for both alleles. These are referred to as pDyn-tdTomato mice. All animal procedures were approved by the Institutional Animal Care and Use Committee of University of Cincinnati.
2.2. Isolation of spinal cord nuclei and fluorescence-activated nuclear sorting (FANS)
At each of 3 developmental time points (postnatal day 7, 21, or 80), mice were euthanized by IP injection of sodium pentobarbital (30 mg/kg; Vortech Pharmaceuticals; Dearborn, MI). The spinal cord was rapidly dissected in 0.1 M ice-cold RNAse-free phosphate buffered saline (PBS), and the dorsal portion of the lumbar enlargement (approx. L1-L5) was snap-frozen on dry ice and stored at −80°C until processing. Tissue was processed as previously described.12 Briefly, on the day of the experiment, each tissue sample was individually homogenized in 1-mL ice-cold homogenization buffer (HB) containing the following (in mM): 250 Sucrose, 25 KCl, 20 Tris-HCl (pH 8), 5 MgCl2, plus 1 tablet/10 mL cOmplete Mini EDTA-free Protease Inhibitor Cocktail (Roche; Indianapolis, IN), 40 U/mL RNAsin Plus (Promega; Madison, WI), and 1-µM DTT. All subsequent steps were performed on ice. Subsequent filtration of each spinal cord was performed as described in Ref. 12, and liberated nuclei were stained with propidium iodide (PI; Thermo Fisher; Waltham, MA) and stored on ice until FANS.
FANS was performed by the Research Flow Cytometry Core in the Division of Rheumatology at Cincinnati Children's Hospital Medical Center, using a BD FACSAria II cell sorter equipped with a 70-µm nozzle. Two populations of nuclei were recovered from each sample: GFP+/PI+ (ie, nuclei stained with PI and also expressing Sun1-GFP under control of the Pdyn promoter) and GFP−/PI+ (ie, nuclei stained with PI but not expressing Sun1-GFP). Five thousand nuclei of each population were sorted into 300 µL of RNAqueous Micro lysis buffer (Life Technologies). At this point, RNA was used for either qPCR or RNA-seq.
2.3. Quantitative real-time reverse transcription -PCR (qPCR)
RNA isolation was performed following RNAqueous Micro kit (Life Technologies) directions, including optional DNAse treatment. cDNA construction was achieved using SuperScript IV First-Strand Synthesis System (Thermo Fisher) with random hexamer primers. qPCR was then performed on a Quant Studio 3 (Applied Biosystems; Waltham, MA) using Taqman Gene Expression Assays (Thermo Fisher). Taqman assays used: pDyn (Mm00457573_m1), Gal (Mm00439056_m1), Pax2 (Mm01217939_m1), Tlx3 (Mm00658289_g1), GFAP (Mm01253033_m1), and Hprt1 (Mm03024075_m1). Relative gene expression was quantified using the 2−ΔΔCt method with Hprt1 as a reference gene.
2.4. cDNA library construction and RNA sequencing
RNA isolation was performed following RNAqueous Micro kit (Life Technologies) directions, including optional DNAse treatment. SoLo RNA-Seq System (NuGEN; Redwood City, CA) was used to construct cDNA libraries, following manufacturer's directions. Libraries were screened for quality on a Bioanalyzer (Agilent; Santa Clara, CA). No more than 50% of the NuGEN libraries were pooled with other TruSeq libraries to maintain the sequence diversity. Using TruSeq SR Cluster Kit v3, at 15 pM, the libraries were clustered onto the flow cell and sequenced on HiSeq 1000 (Illumina; San Diego, CA) using Illumina standard sequencing primer. At single read 1× 51 bp setting, at least 45 million reads per sample were generated.
2.5. Bioinformatics analysis
RNA-seq data were processed using the TrimGalore toolkit,56 which employs Cutadapt63 to trim low-quality bases and Illumina and NuGEN sequencing adapters from the 3′ end of the reads. Only reads that were 20 nt or longer after trimming were kept for further analysis. Reads were mapped to the GRCm38v73 version of the mouse genome and transcriptome51 using the STAR RNA-Seq alignment tool.21 Reads were kept for subsequent analysis if they mapped to a single genomic location. Gene counts were compiled using the HTSeq tool.2 Only genes that had at least 10 reads in any given library were used in subsequent analysis. Normalization and differential expression were performed using the DESeq262 Bioconductor package43 within the R statistical programming environment.71 For the analysis of GFP vs control nuclei at all 3 time points, age was used as a cofactor in the model. The false discovery rate was calculated to control for multiple hypothesis testing.
To identify genes that change across time in the GFP+ samples, the subset of data corresponding to only the GFP+ samples was analyzed using a likelihood ratio test of a model that included time and sex vs a model that included only sex. The false discovery rate was used to correct for multiple hypothesis testing. To identify subpatterns in the significant genes from this analysis (FDR ≤5%), hierarchical clustering of the genes was performed using a correlation distance and complete linkage. The NbClust R package13 was used to determine the optimal number of clusters.
Heatmap generation showing enriched genes and functional class breakdown was achieved using the pheatmap R package.53 Gene lists of enzymes, ion channels, G-protein-coupled receptors (GPCRs), and transporters were obtained from the IUPHAR database38; neuropeptides from the NeuroPep database94; transcription factors from the Riken Transcription Factor database48; and pain genes from the Pain Genes database.57 Lists were obtained from the most recent updated version of these databases as of August 17, 2018.
2.6. In situ hybridization
Male or female mice were euthanized by IP injection of sodium pentobarbital (30 mg/kg; Vortech Pharmaceuticals) and then transcardially perfused with 0.01 M phosphate buffer (PB) followed by 4% paraformaldehyde (PFA) dissolved in PB. Lumbar spinal cords were dissected out and postfixed for an additional 2 hours in room temperature 4% PFA. Tissue was transferred to RNAse-free 30% sucrose in 0.1 M PBS overnight. Fourteen-micrometer transverse sections were cut on a CM 1850 cryostat (Leica; Wetzlar, Germany) and mounted onto Superfrost slides (Fisher). Slides were dried for 20 minutes at −20°C, and then stored at −80°C.
The RNAscope system (Advanced Cell Diagnostics; Newark, CA) was used per manufacturer's directions for fresh-frozen tissue pretreatment, except for omission of the initial on-slide fixation step. Probe hybridization and detection was then performed using Multiplex Fluorescent Kit v2, following manufacturer's directions. The following RNAscope probes were used: Adra2b (425321), Cartpt (432001), Crhr1 (418011-C2), Gucy2d (425451-C2), Npbwr1 (547181), Pdyn (318771, 318771-C2), and Pnoc (437881). TSA Plus Cyanine 3 and Cyanine 5 systems (PerkinElmer #NEL744E001KT, #NEL745E001KT) were used to visualize fluorescent in situ signals. TSA fluorophores were diluted in RNAscope TSA buffer (Cy3 1:1500, Cy5 1:1000).
Because RNAscope tissue pretreatment destroyed the endogenous Sun1-GFP fluorescence, anti-GFP IHC staining was performed on all tissue sections subjected to in situ hybridization. Slide-mounted sections were blocked for 1 hour at room temperature in blocking buffer composed of PBS containing 0.3% Triton X-100 (PBS-Tx) and 10% normal goat serum (Sigma; St. Louis, MO). Rabbit anti-GFP (Abcam #ab6556; Cambridge, United Kingdom) was diluted 1:500 in blocking buffer and applied overnight at 4°C. Slides were washed 3 times with PBS-Tx, and then AlexaFluor 488 Goat anti-Rabbit IgG (ThermoFisher #A11034) secondary antibody was applied at a dilution of 1:500 in antibody buffer and incubated for 1 hour at room temperature. Slides were washed 3 times with PBS, DAPI from the RNAscope kit was applied for 30 seconds, and then slides were finally coverslipped with VectaShield fluorescent mounting medium (Vector Labs; Burlingame, CA).
2.8. Image acquisition and analysis
Images were captured on a BX63 upright fluorescent microscope (Olympus; Center Valley, PA) using CellSens Dimension Desktop software (Olympus). Z-stack images taken at either ×20 or ×40 magnification were projected using the Extended Focal Imaging option in CellSens. Object detection and counting, background subtraction, and grayscale intensity measurements were all executed in CellSens Dimension Desktop.
All colocalization and quantification analysis was performed on images taken at ×20 magnification. For in situ experiments determining the percentage of colocalized cells, a total of 12 sections from 3 animals were randomly selected for each time point examined. For in situ experiments quantifying the number of mRNA transcripts per cell, a total of 8 to 9 sections from 3 animals were randomly selected for each time point examined. Quantitative in situ determined the number of transcripts per cell, as indicated by the number of in situ signal dots, according to RNAscope recommended guidelines.34,93 Briefly, background subtraction was performed on the Cy5 channel, and then 20 discrete dots were randomly selected as individual regions of interest (ROIs). The total grayscale intensity value for each ROI was measured using CellSens Count and Measure function and used to determine the mean intensity value of a single dot in that tissue section. Next, each Sun1-GFP+ nucleus in a tissue section was selected as an individual ROI in the shape of an ellipse described by the clearly visible GFP ring encircling the nucleus. The number of dots contained within each ROI was calculated by dividing the total Cy5 grayscale intensity of the ROI by the previously established mean intensity value of a single dot. Any cells in which the target probe in situ signal was oversaturated were excluded from these calculations (this occurred only in a small number of cells in each adult tissue section probed for Cartpt mRNA).
2.9. In vitro spinal cord slice preparation
At P60-80, pDyn-tdTomato mice were deeply anesthetized with sodium pentobarbital (60 mg/kg) and perfused with ice-cold dissection solution consisting of (in mM): 250 sucrose, 2.5 KCl, 25 NaHCO3, 1.0 NaH2PO4, 6 MgCl2, 0.5 CaCl2, and 25 glucose continuously bubbled with 95% O2/5% CO2. The lumbar spinal cord was isolated and immersed in low-melting-point agarose (3% in above solution) at 37°C, which was then cooled on ice. Parasagittal slices (350-400 µm) were cut using a vibrating microtome (7000smz-2; Campden Instruments; Loughborough, United Kingdom). Slices were incubated for 15 to 20 minutes in a recovery solution containing (in mM): 92 N-Methyl-D-Glutamine (NMDG), 2.5 KCl, 1.2 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 glucose, 5 Na ascorbate, 2 thiourea, 3 Na pyruvate, 10 MgSO4, and 0.5 CaCl2 and then allowed to recover further in an oxygenated artificial cerebrospinal fluid (aCSF) solution containing the following (in mM): 125 NaCl, 2.5 KCl, 25 NaHCO3, 1.0 NaH2PO4, 1.0 MgCl2, 2.0 CaCl2, and 25 glucose for ≥1 hour at room temperature.
2.10. Patch-clamp recordings from spinal pDyn neurons
After recovery, slices were transferred to a submersion-type recording chamber (RC-22; Warner Instruments; Hamden, CT) and mounted on the stage of an upright microscope (BX51WI; Olympus). Slices were then perfused at room temperature with oxygenated aCSF at a rate of 3 to 6 mL/minute.
Patch electrodes were constructed from thin-walled single-filamented borosilicate glass (1.5 mm outer diameter; World Precision Instruments; Sarasota, FL) using a microelectrode puller (P-97; Sutter Instruments; Novato, CA). Pipette resistances ranged from 4 to 6 MΩ, and seal resistances were >1 GΩ. Patch electrodes were filled with an intracellular solution containing (in mM): 130 K-gluconate, 10 KCl, 10 HEPES, 10 Na-phosphocreatine, 4 MgATP, and 0.3 Na2-GTP, pH 7.2 (295-300 mOsm). Whole-cell patch-clamp recordings were obtained from pDyn-tdTomato neurons in laminae I-II using a Multiclamp 700B amplifier (Molecular Devices; San Jose, CA). After the confirmation of a stable resting membrane potential, Neuropeptide W-23 (10 nM) was bath applied. To avoid residual effects of Neuropeptide W-23, only one pDyn-tdTomato neuron was recorded from each slice. The resting membrane potential before and after perfusion with Neuropeptide W-23 was compared using a paired t test.
Membrane voltages were adjusted for liquid junction potentials calculated using JPCalc software (P. Barry, University of New South Wales, Sydney, Australia; modified for Molecular Devices). Currents were filtered at 4 to 6 kHz through a −3 dB, 4-pole low-pass Bessel filter, digitally sampled at 20 kHz, and stored on a personal computer (ICT) using a commercially available data acquisition system (Digidata 1440A with pClamp 10.4 software; Molecular Devices).
2.11. Data availability
Raw sequence data for all samples in this study have been deposited in the Gene Expression Omnibus (GEO; https://www.ncbi.nlm.nih.gov/geo/) under study accession number GSE128052.
To investigate the transcriptional profile of pDyn-lineage spinal cord nuclei, we adapted the recently developed Isolation of Tagged Nuclei from Specific Cell Types (INTACT) method,65 which involves the selective expression of a Sun1-GFP fusion protein within a population of interest to “tag” nuclei. After mechanical dissociation of the spinal cord, which allows for recovery of intact nuclei while avoiding the prolonged heat and protease treatments required for whole-cell dissociation,1,91 GFP-tagged nuclei were separated from non-GFP nuclei by fluorescence-activated nuclear sorting (FANS) for downstream RNA-seq analysis.12 Although only a fraction of total cellular RNA is located in the nucleus, the composition of mRNA transcripts obtained from nuclear RNA is comparable with that obtained from RNA from whole-cell preparations.5,33
3.1. Isolation of mRNA from spinal pDyn-lineage nuclei and qPCR validation
pDyn-IRES-Cre mice55 were bred with R26-LSL-Sun1-GFP reporter mice to generate offspring (referred to as “pDyn-GFP mice”) which selectively expressed the nuclear Sun1-GFP fusion protein in pDyn-lineage neurons (Fig. 1A). These pDyn-GFP neurons were located primarily in laminae I-II and more sparsely in lamina III, consistent with the previously reported distribution of Pdyn-expressing neurons in the spinal cord.24 In situ hybridization revealed substantial colocalization between Pdyn mRNA and neurons containing GFP-immunoreactive (IR) nuclei in adult animals (Figs. 1B–D). Pdyn mRNA transcripts were detected in 79.80% ± 2.42% of neurons containing GFP-IR nuclei, whereas 76.86% ± 1.84% of neurons expressing Pdyn mRNA were GFP-immunoreactive (n = 12; Fig. 1E).
Intact nuclei from adult pDyn-GFP mice were liberated from the SDH of both left and right sides of the lumbar enlargement. Propidium iodide was used to indiscriminately label all nuclei for visualization and cell-sorting purposes. Sun1-GFP nuclei are clearly identifiable among the harvested nuclei, the majority of which are stained only with PI (Fig. 1F). FANS was used to separate the nuclei harvested from each animal into 2 populations: pDyn-lineage nuclei, indicated by the presence of both GFP and PI fluorescence, and unidentified nuclei comprising the rest of the spinal cord, indicated by PI fluorescence only (Fig. 1G). Five thousand nuclei of each population were collected from each animal, from which RNA was subsequently purified and used for either quantitative PCR or RNA-seq analysis.
Quantitative PCR analysis revealed that GFP+ nuclei were highly enriched in Pdyn mRNA, verifying that the FANS technique effectively separates spinal pDyn-lineage nuclei from non-pDyn nuclei (Fig. 1H). Moreover, GFP+ nuclei were significantly enriched in Gal and Pax2, markers known to be expressed by dynorphin interneurons,49,75 and significantly depleted of excitatory neuron marker Tlx314 and glial marker Gfap (Fig. 1I). This initial validation provides strong evidence that the harvested RNA originated from pDyn-lineage neurons in the dorsal horn.
3.2. Transcriptional profile of pDyn-lineage dorsal horn interneurons in adult mice
For an unbiased screen of the transcriptome of spinal pDyn-lineage neurons, nuclei from 2 male and 2 female adult pDyn-GFP mice (postnatal day 80) were sorted as described above, and the RNA from each sample was isolated for RNA-seq analysis. The resulting data were analyzed for differential gene expression comparing the normalized expression of identified transcripts in GFP+ nuclei (ie, pDyn-GFP) to GFP− nuclei from the same animal. Hierarchical clustering analysis revealed that the transcriptional profile of pDyn-GFP nuclei was distinct from that of unspecified, non-pDyn nuclei (Fig. 2A). As expected, pDyn-GFP nuclei were depleted of markers of nonneuronal cells and excitatory neurons, but enriched in the inhibitory neuron markers Pax2, Slc32a1 (encoding the vesicular GABA transporter VGAT), and Gad2 (Fig. 2B).
Over 21,000 genes were detected in this analysis, and 3137 of these exhibited significant differential expression in adult pDyn-GFP nuclei compared with non-pDyn nuclei, as defined by a q-value (P-value adjusted to correct for false discovery rate) of less than 0.01. Of these, 650 were identified as protein-coding genes with a log2-fold change (log2FC) of +1 or greater, a mean normalized expression of 20 or greater, and a coefficient of variation (CV) among biological replicates less than 65%. The top 50 enriched genes that met this criteria included targets identified in previous immunohistochemical studies, such as Gal, Sstr1 and 2, Nos1, and Bhlhb5 (identified in this study as Bhlhe22),49,70 as well as numerous targets that were not previously known to be expressed by pDyn neurons, including Gucy2d, Adra2b, Npbwr1, Cartpt, and Tacr3 (Fig. 2C; see also supplemental digital content 1, Table ST1, available at http://links.lww.com/PAIN/A831).
Because the non-pDyn population used for comparison in this study is composed of both neuronal and nonneuronal nuclei, it is possible that some genes identified in this list are enriched simply by virtue of being highly expressed in spinal inhibitory neurons. To investigate this further, the top 50 enriched genes identified above were compared with single-cell RNA-seq data obtained by Haring et al.39 in a study that analyzed only spinal cord neurons. Although some genes (such as Pax8, Sfrp1, Eepd1, and Zmat4) were indeed enriched across the majority of GABAergic neuronal clusters identified by Haring et al.,39 many highly enriched genes identified in this study were either selectively enriched in the inhibitory Pdyn-expressing neuronal clusters (such as Gucy2d and Hrh3) or enriched in a combination of Pdyn-expressing clusters and other identified subsets of inhibitory dorsal horn neurons (such as Gal and Sstr2; see supplemental digital content 1, Fig. S1, available at http://links.lww.com/PAIN/A831).
pDyn-GFP nuclei within the dorsal horn are enriched in genes of many functional classes (Fig. 3A), according to established curated databases.38,48,57,94 Forty-five of these genes have established relevance to pain sensation or sensitization57 (Fig. 3B; see also supplemental digital content 1, Table ST2 for list and classification of pain-related genes, available at http://links.lww.com/PAIN/A831). Select enzyme, neuropeptide, and GPCR-encoding genes identified in this analysis were validated using in situ hybridization to confirm the presence of target mRNA transcripts in pDyn-expressing neurons.
3.3. Enzymes expressed by pDyn-GFP neurons
A substantial number of enzymes and enzyme-linked receptors are enriched by at least 2-fold in pDyn-GFP neurons (Fig. 3A, Enzymes). Some of these correspond to defining characteristics of inhibitory neurons (such as Gad1 and Gad2), whereas others such as Nos1, encoding nNOS, have been previously shown to be expressed by GABAergic populations within the spinal cord.56 Nitric oxide (NO) synthesized by nNOS catalyzes the formation of cyclic guanosine monophosphate (cGMP), which plays a key role in the establishment and persistence of pain sensitization.79 Although the role of soluble, NO-sensitive guanylate cyclases in nociceptive processing has been studied extensively in both the SDH and dorsal root ganglia, the potential contribution of membrane-bound guanylate cyclases (that are not activated by NO) remains poorly understood.
The present analysis revealed that Gucy2d, which encodes membrane-bound guanylate cyclase D (GC-D), was the most highly enriched gene detected in adult pDyn-GFP nuclei (+6.81 log2FC; Fig. 3AEnzymes & Enzyme-linked Receptors). Surprisingly, this enrichment surpassed that of Pdyn itself (+6.00 log2FC), suggesting that this transcript is unique to pDyn neurons. In situ hybridization for Pdyn and Gucy2d mRNA (Fig. 4A) supports this hypothesis, as almost all Gucy2d+ cells expressed Pdyn mRNA (93.61% ± 1.24%; n = 12). However, only about half of Pdyn+ cells expressed Gucy2d mRNA (53.07% ± 2.95%; n = 12), indicating that Gucy2d+ neurons represent a subset of dynorphin-expressing neurons in the adult dorsal horn.
GC-D has previously been detected only in a very small subset of olfactory epithelial neurons32 that project to the olfactory bulb to terminate in a small number of necklace glomeruli.47 These neurons also express phosphodiesterase 2 (encoded by Pde2a); moreover, Pde2a expression is restricted to GC-D-immunoreactive neurons in the olfactory epithelium.47,60 However, Pde2a was only moderately enriched in adult spinal pDyn-GFP nuclei (+1.83 log2FC), indicating that although Pde2a is present and may be important in spinal Gucy2d-expressing cells, it is likely not exclusive to this population. Other phosphodiesterases (Pde1a, Pde11a, and Pde9a) are also moderately enriched in pDyn nuclei (+1.32, +1.22, and +1.06 Log2FC, respectively), highlighting the importance of cyclic nucleotide signaling in this population.30
3.4. Neuropeptides expressed by pDyn-GFP neurons
In addition to dynorphin itself, spinal pDyn-GFP nuclei express 10 other neuropeptides whose normalized expression was at least 2-fold enriched compared with non-pDyn nuclei (Fig. 3A; Neuropeptides). Although a large subset of spinal pDyn neurons has previously been shown to express galanin49,75 and nociceptin,10 this study is the first to elucidate the full complement of neuropeptide transcripts expressed by this population. Whether these neuropeptides define unique or overlapping subsets of spinal pDyn neurons remains unclear.
Nociceptin, formerly known as orphanin FQ, is a neuropeptide encoded by Pnoc. Pnoc transcripts have been detected in several regions of the brain, as well as in the dorsal and ventral horns of the rat spinal cord.67 A study in the embryonic mouse dorsal horn likewise demonstrated that Pnoc mRNA is expressed predominantly by inhibitory neurons, including some which coexpress Pdyn.10 Consistent with this finding, this study revealed substantial enrichment (+4.16 log2FC) in pDyn-GFP nuclei. Previous studies have not identified the extent to which the nociceptin and dynorphin populations intersect in the adult mouse spinal cord. In situ hybridization for Pdyn and Pnoc mRNA transcripts indicated that there is substantial overlap between these 2 populations (Fig. 4B). About three-quarters of Pdyn+ cells expressed Pnoc mRNA (76.40% ± 1.52%; n = 12), and the majority of Pnoc+ cells also expressed Pdyn mRNA (72.27% ± 0.82%; n = 12). Furthermore, 72.75% ± 1.71% of Sun1-GFP immunoreactive neurons expressed both Pnoc and Pdyn mRNA (n = 12). Reported effects of exogenous nociceptin administration on pain vary widely and seem to be highly dependent on dosage and route of administration.16,35,37,64 Consequently, the role of nociceptin in pain sensitization in the spinal cord remains unclear.
Other notable neuropeptide transcripts enriched in pDyn nuclei include Gal (encoding galanin; +5.42 log2FC); Cartpt encoding the cocaine- and amphetamine-regulated transcript peptide (+3.38 log2FC); and Penk encoding the opioid peptide enkephalin (+1.99 log2FC).
3.5. G-protein-coupled receptors expressed by pDyn-GFP neurons in the adult dorsal horn
G-protein-coupled receptors perform a variety of cellular functions in both neuronal and nonneuronal cells, and are popular targets for pharmacological modulation. Thirty-four percent of all currently Food and Drug Administration (FDA)-approved drugs target GPCRs, and 19% of all drugs currently in clinical trials in the United States target novel GPCRs that are not currently associated with an FDA-approved drug.41 This study identifies several GPCR-encoding transcripts that are enriched in spinal pDyn-GFP nuclei, including adrenergic receptors Adra1a, Adra2b, and Adra2c, corticotropin-releasing hormone receptor Crhr1, and Npbwr1, the receptor for neuropeptides B and W.
α2-adrenergic receptors (encoded by Adra2a, Adra2b, and Adra2c) are known to act synergistically with both μ-opioid and δ-opioid receptors to evoke analgesia.11,23,36 In addition, the application of the α2-adrenergic receptor agonist clonidine increases spinal GABA release under neuropathic conditions, suggesting that inhibitory neurons are key components of α2-adrenoreceptor-evoked analgesia.42 Previous in situ hybridization experiments reported widespread detection of Adra2a and Adra2c mRNA throughout the dorsal and ventral spinal cord. Although Adra2c seems more strongly expressed in the ventral horn, Adra2b is restricted to a small population of cells in the superficial dorsal horn.80 This study revealed that Adra2b transcripts were enriched in pDyn-GFP nuclei (+3.60 log2FC), which suggested that the previously described Adra2b+ population could correspond to pDyn neurons. Surprisingly, in situ hybridization for Adra2b and Pdyn mRNA showed that only 36.62% ± 1.69% of Pdyn-expressing neurons also expressed Adra2b, and 42.59% ± 1.72% of Adra2b-expressing neurons were positive for Pdyn mRNA. Furthermore, only 30.14% ± 2.03% of pDyn-GFP neurons were positive for both Adra2b and Pdyn mRNA (n = 12; Fig. 4C). Adra2c was also enriched in pDyn-GFP nuclei (+2.38 log2FC), but Adra2a was not (+0.05 log2FC).
Corticotropin-releasing hormone receptor 1 (CRHR1 or CRFR1) is a GPCR whose ligands include neuropeptides involved in the regulation of the HPA axis, such as corticotropin-releasing hormone (CRH) and urocortins 1 to 3.40,45 Previous studies in the rodent spinal cord detected CRH immunoreactivity in afferent terminals located in the superficial laminae of the dorsal horn54,78 and localized the receptor itself to laminae I and II.6 Despite previous in situ hybridization studies showing Crhr1 mRNA expression primarily in laminae III-VIII of the mouse lumbar spinal cord with no detectable signal in laminae I-II, differential gene expression analysis in this study revealed that Crhr1 was enriched in pDyn-GFP nuclei (+2.29 log2FC). In situ hybridization confirmed abundant expression throughout the SDH (Fig. 4D). Notably, 74.46% ± 1.78% of Pdyn-expressing neurons and 81.93% ± 1.83% of Sun1-GFP immunoreactive cells also expressed Crhr1 mRNA (n = 12). Although CRH is primarily associated with stress–response pathways, the documented CRH-evoked release of dynorphin A from spinal cord slices suggests that CRH and its receptor are also involved in pain sensation or analgesia.83
Npbwr1 encodes the receptor for neuropeptides B and W. Previous studies have detected Npbwr1 mRNA expression in the mouse spinal cord, including the dorsal horn,31,89 and in the amygdala, where it is localized almost exclusively to Gad67-expressing inhibitory neurons.66 The present analysis revealed that Npbwr1 was enriched in pDyn-GFP nuclei (+3.46 log2FC), suggesting that the pDyn population may be among those inhibitory neurons that express this receptor. In situ hybridization confirmed the presence of Npbwr1 mRNA in the majority of pDyn-GFP neurons (Fig. 4E), and 54.33% ± 2.54% of Sun1-GFP immunoreactive cells expressed both Npbwr1 and Pdyn mRNA (n = 9). To determine whether the neuropeptide B/W receptor was functionally expressed by pDyn-lineage neurons, the effects of the selective agonist W-2381 on the membrane potential of pDyn-tdTomato dorsal horn neurons were investigated using in vitro patch-clamp recordings from spinal cord slices. The bath application of W-23 (10 nM) resulted in membrane hyperpolarization in 7 of the 11 cells recorded (Fig. 4F). Across the population of pDyn-tdTomato neurons sampled, the resting potential was significantly more hyperpolarized after W-23 application compared with the baseline (n = 11, t = 3.01, P = 0.013, paired t test; Fig. 4G). These data confirm that the Npbwr1 mRNA detected in pDyn-GFP neurons is indicative of protein-level expression of the neuropeptide B/W receptor. Moreover, the membrane hyperpolarization induced by activation of this receptor is consistent with the expected actions of a Gi/o-coupled GPCR.
3.6. Age-dependent gene expression in spinal pDyn interneurons
Given the known developmental alterations in the functional properties of inhibitory neurotransmission within the SDH,9,50 we next sought to elucidate how the transcriptome of the spinal pDyn population changes with age. Nuclei harvested at 3 time points, postnatal day 7 (n = 4; 2 male and 2 female), 21 (n = 3; 2 male and 1 female), and 80 days (n = 4; 2 male and 2 female; same samples used in the adult analysis above), were sorted and RNA-seq data were obtained as described above. We first characterized genes that were enriched in the pDyn population at different ages. Comparable to the 3137 genes that were significantly differentially expressed (q-val <0.01) between GFP+ and GFP− samples at the adult time point, 3409 significantly differentially expressed genes (DEGs) were identified at P7 and 1868 at P21. The reduced number of identified DEGs at P21 may be partially due to the lower number of biological replicates included at this age because 1 of the 4 prepared samples was excluded from analysis due to failed cDNA library construction. Principal component and differential gene expression analysis indicated that the sex of the animal was not a significant source of variation (supplemental digital content 1, Figs. S2 and S3, available at http://links.lww.com/PAIN/A831).
Of the DEGs identified at P7, 341 were protein-coding genes with a log2-fold change (log2FC) of +1 or greater, a mean normalized expression of 20 or greater, and a CV among biological replicates less than 65%. At P21, 573 of the identified DEGs met the same criteria. Comparing the lists of the 50 most highly enriched genes across ages shows that many gene targets remain consistent in their relative enrichment in pDyn-GFP nuclei throughout life (Fig. 5A). Expanding this comparison to include all genes that are enriched by ≥2-fold yields similar results, although several gene targets exhibited such enrichment only at particular ages (Fig. 5B).
Because most DEGs did not exhibit significant changes in enrichment across the 3 developmental time points, examining only the relative enrichment of each gene provided an incomplete picture of age-dependent changes in gene expression within pDyn-GFP neurons. Therefore, additional analyses only considered gene expression in the GFP+ nuclei from each time point, and these samples were compared to each other without comparison to the corresponding GFP− population. A likelihood ratio test was performed to statistically determine which genes were significantly changing in pDyn-GFP neurons in an age-related manner. Of the 21,342 genes detected in this analysis, 3331 showed a significant age-dependent change as determined by a q-value (P-value adjusted for false discovery rate) of less than 0.05 (Fig. 6A). A somewhat more lenient threshold of significance was used for this analysis, compared with the q-val <0.01 cutoff established in the gene enrichment analysis portion of this study, to allow for the detection of more subtle changes in expression. Hierarchical clustering (Fig. 6B) and linkage analysis performed on the significantly changing genes identified 4 clusters, each characterized by a distinct developmental trend in gene expression as illustrated by a representative example from each cluster (Fig. 6C). Cluster A contains 1374 genes whose expression levels increased with age (Fig. 6C; Gucy2d). Cluster B contains 1185 genes whose expression decreased between P7 and P21, and then remained relatively stable between P21 and P80 (Fig. 6C; Grik1). Cluster C is the smallest cluster, containing only 81 genes that exhibited peak expression at P21 while having comparatively lower expression during both infancy and adulthood (Fig. 6C; Htr1a). Finally, Cluster D contains 691 genes. Similar to Cluster B, the genes in this cluster had peak expression shortly after birth that decreased between P7 and P21. However, unlike Cluster B, expression levels of genes in Cluster D continued to decline between P21 and P80 (Fig. 6C; Hcrtr1).
To identify the functional classes of genes that changed most dramatically with age, this list of developmentally regulated genes was narrowed down to those whose expression either increased or decreased by a magnitude of 2-fold or greater between any 2 time points. Only genes that had a CV between biological replicates of less than 65%, and mean normalized expression of 20 or greater, at one or more time points were included. The resulting list of 812 genes was sorted by functional class (Fig. 7) according to established curated databases.38,48,57,94
Most of the genes included in this list were not enriched in pDyn-GFP nuclei compared with GFP− nuclei (Fig. 7), suggesting that the developmental trends identified in the present analysis of pDyn-GFP nuclei may indicate global changes in gene expression occurring across the SDH. However, a small number of genes whose expression increased or decreased by at least 2-fold over the course of development were enriched ≥2-fold in pDyn-GFP nuclei (see supplemental digital content 1, Table ST3, available at http://links.lww.com/PAIN/A831), and may therefore represent attractive potential targets for interventions designed to manipulate the activity of these inhibitory interneurons in an age-dependent manner. Two of these targets were validated with quantitative in situ hybridization to confirm the observed age-related changes in gene expression within pDyn neurons.
Analysis of RNA-seq data showed that Npbwr1, which encodes the neuropeptide B/W receptor (and assigned to Cluster B), was expressed most highly in neonates, then sharply decreased by P21 and remained at this lower level until adulthood (Fig. 8A). This target was enriched in pDyn-GFP nuclei at all 3 ages examined (mean +3.99 log2FC). In situ hybridization experiments were performed on spinal cord sections from P7 (Fig. 8B) and P74 (Fig. 8C) pDyn-GFP mice to determine whether the observed developmental decrease in gene expression was due to an age-dependent reduction in the percentage of pDyn-GFP neurons expressing Npbwr1 and/or the amount of Npbwr1 mRNA transcribed per cell. Colocalization analysis demonstrated that the percentage of pDyn-GFP nuclei that expressed both Npbwr1 and Pdyn mRNA decreased from 67.62% ± 2.48% at P7 to 54.33% ± 2.54% at P74 (n = 9 sections in each group; Fig. 8D). Moreover, quantification of the number of Npbwr1 mRNA signal dots per GFP+ nucleus revealed that within pDyn-GFP cells coexpressing both targets, the mean amount of Npbwr1 mRNA transcribed per nucleus also significantly decreased from 10.79 ± 1.18 to 5.49 ± 0.44 transcripts per nucleus between infancy and adulthood (n = 9 sections in each group; Fig. 8E).
Cartpt was also enriched in pDyn-GFP nuclei at all time points examined (mean +2.48 log2FC), and its expression increased significantly throughout development (Fig. 8F). Quantitative in situ hybridization performed on spinal cord sections from P7 (Fig. 8G) and P74 (Fig. 8H) pDyn-GFP mice revealed that the percentage of pDyn-GFP cells that expressed both Cartpt and Pdyn mRNA significantly increased from 26.64% ± 1.66% to 49.86% ± 1.78% between P7 and P74 (Fig. 8I). The mean number of Cartpt transcripts per nucleus also increased with age in those pDyn-GFP cells expressing both Cartpt and Pdyn, from 3.84 ± 0.69 transcripts per nucleus at P7 to 11.72 ± 1.53 at P74 (P7: n = 8 sections, P74: n = 9 sections; Fig. 8J).
This study presents the first unbiased transcriptomic profile of spinal pDyn-lineage neurons in the neonatal, juvenile, and adult mouse, thereby identifying genes that may govern spinal nociceptive processing at distinct stages of development and yielding new molecular insight into the maturation of a key population of dorsal horn interneurons.
4.1. Enriched genes in spinal pDyn neurons
This analysis revealed a number of enriched genes known to be expressed by spinal pDyn neurons or inhibitory neurons in general, including Gal, Bhlhb5, Nos1, Pax2, Sstr2, Neurod1, 2 and 6, and Pnoc.8,10,49 However, because the spinal pDyn population comprises only about 11% of laminae I-II neurons,8 and certainly a much lower percentage of all SDH cells, some genes may be highly expressed in this population and yet remain undetected when using a whole-tissue approach. Indeed, this study detected and verified mRNA expression of many enriched genes that had not previously been reported in pDyn neurons, including Adra2b, Crhr1, Npbwr1, and Cartpt. Notably, we also identify Gucy2d as a novel and selective marker of a subset of pDyn neurons within the dorsal horn.
Gucy2d encodes guanylate cyclase D (GC-D) in rodents, a membrane-bound guanylate cyclase that has only previously been detected in a small number of olfactory epithelial cells.32 Little is known of its biological function, although the identification of extracellular ligands guanylin and uroguanylin suggest a role in odorant signal transduction.25 However, the intracellular activation of GC-D by bicarbonate85 also suggests a possible role in environmental CO2 sensing. The present analysis did not detect Gucy2a, Gucy2b, Gucy2c, or Gucy2g transcripts in either spinal pDyn-GFP nuclei or non-pDyn nuclei. Low levels of GC-E (encoded by Gucy2e), and GC-F (Gucy2f), were detected in pDyn-GFP nuclei, but these transcripts were not enriched in these interneurons. The substantial presence of GC-D in particular suggests a possible function distinct from other NO-insensitive GCs.
Gucy2d corresponds to the human ortholog Gucy2e and is a pseudogene in humans and most primates.98 Although it is still transcribed, the resulting nonfunctional product is unlikely to play a role in human nociception. Nevertheless, the high specificity of this transcript for pDyn spinal neurons may provide an effective means to genetically manipulate a subset of the pDyn population in a selective manner within the mouse dorsal horn. Future studies are needed to fully characterize Gucy2d-expressing neurons in the SDH and elucidate their functional role in nociceptive processing and pain sensation.
In addition to validating Npbwr1 RNA expression through in situ hybridization, patch-clamp recordings confirmed protein-level expression of the neuropeptide B/W receptor (GPR7; encoded by Npbwr1), as the selective agonist W-23 evoked membrane hyperpolarization in spinal pDyn neurons (Fig. 4G). Although this hyperpolarization may seem unexpected, given previous studies showing that intracerebroventricular or intrathecal (i.t.) injection of W-23 reduces mechanical allodynia under inflammatory and neuropathic conditions,96,97 it should be noted that GPR7 is not exclusively expressed by pDyn neurons. As a result, it remains to be determined whether the analgesic effects of GPR7 activation are localized to the periphery, SDH, and/or supraspinal nociceptive circuits.
Cartpt, which encodes the cocaine- and amphetamine-regulated transcript (CART), is another relatively understudied gene in the spinal cord. First identified in the rat brain due to its dramatic upregulation after administration of cocaine or amphetamine,22 CART is a neuropeptide that reduces appetite86 and potentiates morphine-induced reward behavior.90 Dual-label in situ hybridization studies revealed colocalization of CART and Pdyn mRNA in several hypothalamic nuclei,26 whereas a small number of varicosities in the ventral tegmental area (which receives input from the hypothalamus) were immunoreactive for both prodynorphin and CART peptide.17 CART-immunoreactive fibers and somata have been detected in the SDH.68 Although initial studies reported an increase in thermal pain sensitivity after i.t. administration of CART,68 later studies reported no effect on thermal withdrawal latency in naive animals18 or attenuated thermal hyperalgesia and mechanical allodynia under neuropathic conditions.19 The receptor for CART is currently unknown, thus hampering investigation into its role in spinal nociceptive transmission.
4.2. Developmental changes in pDyn neuronal gene expression
A complete understanding of the neurobiological mechanisms underlying pediatric pain necessitates additional insight into the functional organization of inhibitory synaptic circuits within the immature SDH. Although these inhibitory networks are poorly “tuned” during the neonatal period,9 little was known about how age influences gene expression within spinal GABAergic neurons. The present results identify clusters of genes that are either significantly upregulated or downregulated in spinal pDyn neurons during postnatal development, which could yield new insight into factors that govern spinal nociceptive processing in an age-dependent manner. For example, it is notable that the expression of both the neuropeptide neurotensin (encoded by Nts) and its corresponding receptor NTSR2 (Ntsr2) was significantly higher in neonatal pDyn neurons compared with adulthood. The i.t. administration of neurotensin dampens pain sensitivity,44,84 which is at least partially mediated by the activation of NTSR2.73,76 Furthermore, both cerebellin-1 (encoded by Cbln1) and cerebellin-2 (Cbln2), neuropeptides that reportedly evoke mechanical pain hypersensitivity after spinal injection,74 exhibit higher expression within the pDyn population during the neonatal period compared with later ages. These findings highlight the point that although significant progress has been made in understanding developmental changes in GABAergic and glycinergic transmission within the dorsal horn,4,9,15,50,52 very little is known about age-dependent changes in neuropeptide signaling in the region or its role in regulating pain sensitivity at different stages of life. Similarly, the underlying basis for developmental changes in the firing properties of dorsal horn neurons61,92 remains poorly understood because little is known about how age influences the expression of voltage-dependent and voltage-independent ion channels within distinct neuronal subpopulations. Interestingly, 5 genes encoding subunits of gap junction channels (Gja1, Gjb1, and Gjc1-3) were significantly downregulated in spinal pDyn neurons over the course of postnatal development (Fig. 7).
It should be noted that the overwhelming majority of developmentally regulated genes identified in this study are not enriched in pDyn nuclei compared with the rest of spinal cord nuclei, suggesting that the age-dependent gene expression observed in the pDyn population may reflect widespread changes occurring throughout the SDH. Future studies are clearly needed to elucidate the functional implications of these transcriptional changes for spinal nociceptive signaling and pain sensitivity during postnatal development.
4.3. Limitations of INTACT approach
The INTACT technique is well suited to achieve the major goals of this study, namely to identify genes that distinguish the pDyn neuronal population from other cell types in the spinal cord, and to elucidate how postnatal age influences gene expression in this key population of dorsal horn interneurons. However, despite using a sequencing depth of 45 million reads per sample, our analysis of nuclear RNA may have failed to detect minimally abundant transcripts that would be detectable using whole-cell RNA. In addition, a significant limitation of the chosen approach is that it is unable to identify patterns of gene expression within distinct subpopulations of spinal pDyn neurons. For example, it will ultimately be important to determine the full complements of genes that define inhibitory vs the less prevalent excitatory pDyn neurons8 within spinal nociceptive circuits. The ability to analyze gene expression in distinct subtypes of pDyn neurons in the SDH may prove especially important when examining the effects of nerve or tissue damage on gene expression within this population because changes occurring selectively within a subset of pDyn neurons may be missed when considering the population as a whole.
Fortunately, exploring the diversity of cell types within the pDyn population is now feasible due to the recent development of single-nucleus RNA-seq approaches, which involve massively parallel sequencing on a large number of unidentified spinal nuclei and the subsequent use of established markers to compare user-defined subpopulations.59,77 This would allow for the specific comparison of pDyn neurons to other neurons only (including inhibitory interneurons), whereas the present analysis compares pDyn nuclei with a non-pDyn population consisting of both neuronal and nonneuronal cell types. A single-nucleus approach may thus filter out genes whose reported enrichment in this study can be simply explained by a selective expression within neurons. However, because nonneuronal cells are more abundant in the spinal cord compared with neurons, it is possible that such a strategy would yield an inadequate number of neurons comprising a given population of interest. In this case, a combination of the 2 methods could prove useful, whereby a neuronal-specific promoter could be used to drive Sun1-GFP expression within the population of interest to allow for INTACT-based sorting of the targeted nuclei, followed by single nucleus RNA-seq analysis.
In summary, this study provides the first unbiased molecular profile of an identified subpopulation of developing spinal neurons using nuclear RNA-seq. This profile not only identifies novel enriched genes within pDyn neurons but also elucidates significant age-dependent changes in gene expression. Collectively, these results will inform future studies investigating the role of pDyn dorsal horn neurons in nociceptive signaling at different stages of postnatal development.
Conflict of interest statement
The authors have no conflicts of interest to declare.
The authors thank Dr David Corcoran and the Genomic Analysis and Bioinformatics Shared Resource at Duke University for providing bioinformatic analysis of RNA-seq data. The authors would also like to acknowledge the assistance of the Research Flow Cytometry Core in the Division of Rheumatology at Cincinnati Children's Hospital Medical Center, and Aaron Serafin for his assistance with heatmap generation using R.
All work was supported by the National Institutes of Health (NS100469 to M.L.B.).
Supplemental digital content
Supplemental digital content associated with this article can be found online at http://links.lww.com/PAIN/A831.
Supplemental video content
Video content associated with this article can be found at http://links.lww.com/PAIN/A832.
. Adam M, Potter AS, Potter SS. Psychrophilic proteases dramatically reduce single-cell RNA-seq
artifacts: a molecular atlas of kidney development
. Anders S, Pyl PT, Huber W. HTSeq—a Python framework to work with high-throughput sequencing data. Bioinformatics 2015;31:166–9.
. Andrews K, Fitzgerald M. The cutaneous withdrawal reflex in human neonates: sensitization, receptive fields, and the effects of contralateral stimulation. PAIN
. Baccei ML, Fitzgerald M. Development
of GABAergic and glycinergic transmission in the neonatal
rat dorsal horn
. J Neurosci 2004;24:4749–57.
. Barthelson RA, Lambert GM, Vanier C, Lynch RM, Galbraith DW. Comparison of the contributions of the nuclear and cytoplasmic compartments to global gene expression
in human cells. BMC Genomics 2007;8:340.
. Bell JA, de Souza EB. Functional corticotropin-releasing factor receptors in neonatal
rat spinal cord
. Peptides 1988;9:1317–22.
. Bourane S, Duan B, Koch SC, Dalet A, Britz O, Garcia-Campmany L, Kim E, Cheng L, Ghosh A, Ma Q, Goulding M. Gate control of mechanical itch by a subpopulation of spinal cord
interneurons. Science 2015;350:550–4.
. Boyle KA, Gutierrez-Mecinas M, Polgar E, Mooney N, O'Connor E, Furuta T, Watanabe M, Todd AJ. A quantitative study of neurochemically defined populations of inhibitory
interneurons in the superficial dorsal horn
of the mouse spinal cord
. Neuroscience 2017;363:120–33.
. Bremner LR, Fitzgerald M. Postnatal tuning of cutaneous inhibitory
receptive fields in the rat. J Physiol 2008;586:1529–37.
. Brohl D, Strehle M, Wende H, Hori K, Bormuth I, Nave KA, Muller T, Birchmeier C. A transcriptional network coordinately determines transmitter and peptidergic fate in the dorsal spinal cord
. Dev Biol 2008;322:381–93.
. Chabot-Dore AJ, Schuster DJ, Stone LS, Wilcox GL. Analgesic synergy between opioid and alpha2 -adrenoceptors. Br J Pharmacol 2015;172:388–402.
. Chamessian A, Young M, Qadri Y, Berta T, Ji RR, Van de Ven T. Transcriptional profiling of somatostatin interneurons in the spinal dorsal horn
. Sci Rep 2018;8:6809.
. Charrad M, Ghazzali N, Noiteau V, Niknafs A. NbClust: an R package for determining the relevant number of clusters in a data set. J Stat Softw 2014;61:1–36.
. Cheng L, Arata A, Mizuguchi R, Qian Y, Karunaratne A, Gray PA, Arata S, Shirasawa S, Bouchard M, Luo P, Chen CL, Busslinger M, Goulding M, Onimaru H, Ma Q. Tlx3 and Tlx1 are post-mitotic selector genes determining glutamatergic over GABAergic cell fates. Nat Neurosci 2004;7:510–17.
. Cordero-Erausquin M, Coull JA, Boudreau D, Rolland M, De Koninck Y. Differential maturation of GABA action and anion reversal potential in spinal lamina I neurons: impact of chloride extrusion capacity. J Neurosci 2005;25:9613–23.
. Corradini L, Briscini L, Ongini E, Bertorelli R. The putative OP(4) antagonist, [Nphe(1)]nociceptin(1-13)NH(2), prevents the effects of nociceptin in neuropathic rats. Brain Res 2001;905:127–33.
. Dallvechia-Adams S, Kuhar MJ, Smith Y. Cocaine- and amphetamine-regulated transcript peptide projections in the ventral midbrain: colocalization with gamma-aminobutyric acid, melanin-concentrating hormone, dynorphin, and synaptic interactions with dopamine neurons. J Comp Neurol 2002;448:360–72.
. Damaj MI, Hunter RG, Martin BR, Kuhar MJ. Intrathecal CART (55-102) enhances the spinal analgesic actions of morphine in mice. Brain Res 2004;1024:146–9.
. Damaj MI, Zheng J, Martin BR, Kuhar MJ. Intrathecal CART (55-102) attenuates hyperlagesia and allodynia in a mouse model of neuropathic but not inflammatory pain
. Peptides 2006;27:2019–23.
. Daniele CA, MacDermott AB. Low-threshold primary afferent drive onto GABAergic interneurons in the superficial dorsal horn
of the mouse. J Neurosci 2009;29:686–95.
. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq
aligner. Bioinformatics 2013;29:15–21.
. Douglass J, McKinzie AA, Couceyro P. PCR differential display identifies a rat brain mRNA that is transcriptionally regulated by cocaine and amphetamine. J Neurosci 1995;15:2471–81.
. Drasner K, Fields HL. Synergy between the antinociceptive effects of intrathecal clonidine and systemic morphine in the rat. PAIN
. Duan B, Cheng L, Bourane S, Britz O, Padilla C, Garcia-Campmany L, Krashes M, Knowlton W, Velasquez T, Ren X, Ross S, Lowell BB, Wang Y, Goulding M, Ma Q. Identification of spinal circuits transmitting and gating mechanical pain
. Cell 2014;159:1417–32.
. Duda T, Sharma RK. ONE-GC membrane guanylate cyclase, a trimodal odorant signal transducer. Biochem Biophys Res Commun 2008;367:440–5.
. Elias CF, Lee CE, Kelly JF, Ahima RS, Kuhar M, Saper CB, Elmquist JK. Characterization of CART neurons in the rat and human hypothalamus. J Comp Neurol 2001;432:1–19.
. Fitzgerald M, Jennings E. The postnatal development
of spinal sensory processing. Proc Natl Acad Sci U S A 1999;96:7719–22.
. Fitzgerald M, Shaw A, MacIntosh N. Postnatal development
of the cutaneous flexor reflex: comparative study of preterm infants and newborn rat pups. Dev Med Child Neurol 1988;30:520–6.
. Foster E, Wildner H, Tudeau L, Haueter S, Ralvenius WT, Jegen M, Johannssen H, Hosli L, Haenraets K, Ghanem A, Conzelmann KK, Bosl M, Zeilhofer HU. Targeted ablation, silencing, and activation establish glycinergic dorsal horn
neurons as key components of a spinal gate for pain
and itch. Neuron 2015;85:1289–304.
. Francis SH, Busch JL, Corbin JD, Sibley D. cGMP-dependent protein kinases and cGMP phosphodiesterases in nitric oxide and cGMP action. Pharmacol Rev 2010;62:525–63.
. Fujii R, Yoshida H, Fukusumi S, Habata Y, Hosoya M, Kawamata Y, Yano T, Hinuma S, Kitada C, Asami T, Mori M, Fujisawa Y, Fujino M. Identification of a neuropeptide modified with bromine as an endogenous ligand for GPR7. J Biol Chem 2002;277:34010–16.
. Fulle HJ, Vassar R, Foster DC, Yang RB, Axel R, Garbers DL. A receptor guanylyl cyclase expressed specifically in olfactory sensory neurons. Proc Natl Acad Sci U S A 1995;92:3571–5.
. Grindberg RV, Yee-Greenbaum JL, McConnell MJ, Novotny M, O'Shaughnessy AL, Lambert GM, Arauzo-Bravo MJ, Lee J, Fishman M, Robbins GE, Lin X, Venepally P, Badger JH, Galbraith DW, Gage FH, Lasken RS. RNA-sequencing from single nuclei
. Proc Natl Acad Sci U S A 2013;110:19802–7.
. Guideline on how to quantify RNAscope(r) fluorescent assay results. Technical Note SOP 45-006. Newark, CA: ACDBio, 2017.
. Hao JX, Xu IS, Wiesenfeld-Hallin Z, Xu XJ. Anti-hyperalgesic and anti-allodynic effects of intrathecal nociceptin/orphanin FQ in rats after spinal cord
injury, peripheral nerve injury and inflammation. PAIN
. Hao S, Takahata O, Iwasaki H. Intrathecal endomorphin-1 produces antinociceptive activities modulated by alpha 2-adrenoceptors in the rat tail flick, tail pressure and formalin tests. Life Sci 2000;66:PL195–204.
. Hara N, Minami T, Okuda-Ashitaka E, Sugimoto T, Sakai M, Onaka M, Mori H, Imanishi T, Shingu K, Ito S. Characterization of nociceptin hyperalgesia and allodynia in conscious mice. Br J Pharmacol 1997;121:401–8.
. Harding SD, Sharman JL, Faccenda E, Southan C, Pawson AJ, Ireland S, Gray AJG, Bruce L, Alexander SPH, Anderton S, Bryant C, Davenport AP, Doerig C, Fabbro D, Levi-Schaffer F, Spedding M, Davies JA, Nc I. The IUPHAR/BPS Guide to PHARMACOLOGY in 2018: updates and expansion to encompass the new guide to IMMUNOPHARMACOLOGY. Nucleic Acids Res 2018;46:D1091–106.
. Haring M, Zeisel A, Hochgerner H, Rinwa P, Jakobsson JET, Lonnerberg P, La Manno G, Sharma N, Borgius L, Kiehn O, Lagerstrom MC, Linnarsson S, Ernfors P. Neuronal atlas of the dorsal horn
defines its architecture and links sensory input to transcriptional cell types. Nat Neurosci 2018;21:869–80.
. Hauger RL, Risbrough V, Brauns O, Dautzenberg FM. Corticotropin releasing factor (CRF) receptor signaling in the central nervous system: new molecular targets. CNS Neurol Disord Drug Targets 2006;5:453–79.
. Hauser AS, Attwood MM, Rask-Andersen M, Schioth HB, Gloriam DE. Trends in GPCR drug discovery: new agents, targets and indications. Nat Rev Drug Discov 2017;16:829–42.
. Hayashida K, Peters CM, Gutierrez S, Eisenach JC. Depletion of endogenous noradrenaline does not prevent spinal cord
plasticity following peripheral nerve injury. J Pain
. Huber W, Carey VJ, Gentleman R, Anders S, Carlson M, Carvalho BS, Bravo HC, Davis S, Gatto L, Girke T, Gottardo R, Hahne F, Hansen KD, Irizarry RA, Lawrence M, Love MI, MacDonald J, Obenchain V, Oles AK, Pages H, Reyes A, Shannon P, Smyth GK, Tenenbaum D, Waldron L, Morgan M. Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods 2015;12:115–21.
. Hylden JL, Wilcox GL. Antinociceptive action of intrathecal neurotensin in mice. Peptides 1983;4:517–20.
. Inda C, Armando NG, Dos Santos Claro PA, Silberstein S. Endocrinology and the brain: corticotropin-releasing hormone signaling. Endocr Connect 2017;6:R99–R120.
. Ingram RA, Fitzgerald M, Baccei ML. Developmental changes in the fidelity and short-term plasticity of GABAergic synapses in the neonatal
rat dorsal horn
. J Neurophysiol 2008;99:3144–50.
. Juilfs DM, Fulle HJ, Zhao AZ, Houslay MD, Garbers DL, Beavo JA. A subset of olfactory neurons that selectively express cGMP-stimulated phosphodiesterase (PDE2) and guanylyl cyclase-D define a unique olfactory signal transduction pathway. Proc Natl Acad Sci U S A 1997;94:3388–95.
. Kanamori M, Konno H, Osato N, Kawai J, Hayashizaki Y, Suzuki H. A genome-wide and nonredundant mouse transcription factor database. Biochem Biophys Res Commun 2004;322:787–93.
. Kardon AP, Polgar E, Hachisuka J, Snyder LM, Cameron D, Savage S, Cai X, Karnup S, Fan CR, Hemenway GM, Bernard CS, Schwartz ES, Nagase H, Schwarzer C, Watanabe M, Furuta T, Kaneko T, Koerber HR, Todd AJ, Ross SE. Dynorphin acts as a neuromodulator to inhibit itch in the dorsal horn
of the spinal cord
. Neuron 2014;82:573–86.
. Keller AF, Coull JA, Chery N, Poisbeau P, De Koninck Y. Region-specific developmental specialization of GABA-glycine cosynapses in laminas I-II of the rat spinal dorsal horn
. J Neurosci 2001;21:7871–80.
. Kersey PJ, Staines DM, Lawson D, Kulesha E, Derwent P, Humphrey JC, Hughes DS, Keenan S, Kerhornou A, Koscielny G, Langridge N, McDowall MD, Megy K, Maheswari U, Nuhn M, Paulini M, Pedro H, Toneva I, Wilson D, Yates A, Birney E. Ensembl Genomes: an integrative resource for genome-scale data from non-vertebrate species. Nucleic Acids Res 2012;40:D91–97.
. Koch SC, Tochiki KK, Hirschberg S, Fitzgerald M. C-fiber activity-dependent maturation of glycinergic inhibition in the spinal dorsal horn
of the postnatal rat. Proc Natl Acad Sci U S A 2012;109:12201–6.
. Korosi A, Kozicz T, Richter J, Veening JG, Olivier B, Roubos EW. Corticotropin-releasing factor, urocortin 1, and their receptors in the mouse spinal cord
. J Comp Neurol 2007;502:973–89.
. Krashes MJ, Shah BP, Madara JC, Olson DP, Strochlic DE, Garfield AS, Vong L, Pei H, Watabe-Uchida M, Uchida N, Liberles SD, Lowell BB. An excitatory paraventricular nucleus to AgRP neuron circuit that drives hunger. Nature 2014;507:238–42.
. Lacroix-Fralish ML, Ledoux JB, Mogil JS. The Pain
Genes Database: an interactive web browser of pain
-related transgenic knockout studies. PAIN
. Laing I, Todd AJ, Heizmann CW, Schmidt HH. Subpopulations of GABAergic neurons in laminae I-III of rat spinal dorsal horn
defined by coexistence with classical transmitters, peptides, nitric oxide synthase or parvalbumin. Neuroscience 1994;61:123–32.
. Lake BB, Ai R, Kaeser GE, Salathia NS, Yung YC, Liu R, Wildberg A, Gao D, Fung HL, Chen S, Vijayaraghavan R, Wong J, Chen A, Sheng X, Kaper F, Shen R, Ronaghi M, Fan JB, Wang W, Chun J, Zhang K. Neuronal subtypes and diversity revealed by single-nucleus RNA sequencing of the human brain. Science 2016;352:1586–90.
. Leinders-Zufall T, Cockerham RE, Michalakis S, Biel M, Garbers DL, Reed RR, Zufall F, Munger SD. Contribution of the receptor guanylyl cyclase GC-D to chemosensory function in the olfactory epithelium. Proc Natl Acad Sci U S A 2007;104:14507–12.
. Li J, Baccei ML. Pacemaker neurons within newborn spinal pain
circuits. J Neurosci 2011;31:9010–22.
. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq
data with DESeq2. Genome Biol 2014;15:550.
. Marcel M. Cutadapt removes adapter sequences from high-throughout sequencing reads. Bioinf Action 2011;17:10–12.
. Mika J, Obara I, Przewlocka B. The role of nociceptin and dynorphin in chronic pain
: implications of neuro-glial interaction. Neuropeptides 2011;45:247–61.
. Mo A, Mukamel EA, Davis FP, Luo C, Henry GL, Picard S, Urich MA, Nery JR, Sejnowski TJ, Lister R, Eddy SR, Ecker JR, Nathans J. Epigenomic signatures of neuronal diversity in the mammalian brain. Neuron 2015;86:1369–84.
. Nagata-Kuroiwa R, Furutani N, Hara J, Hondo M, Ishii M, Abe T, Mieda M, Tsujino N, Motoike T, Yanagawa Y, Kuwaki T, Yamamoto M, Yanagisawa M, Sakurai T. Critical role of neuropeptides B/W receptor 1 signaling in social behavior and fear memory. PLoS One 2011;6:e16972.
. Neal CR Jr, Mansour A, Reinscheid R, Nothacker HP, Civelli O, Watson SJ Jr. Localization of orphanin FQ (nociceptin) peptide and messenger RNA in the central nervous system of the rat. J Comp Neurol 1999;406:503–47.
. Ohsawa M, Dun SL, Tseng LF, Chang J, Dun NJ. Decrease of hindpaw withdrawal latency by cocaine- and amphetamine-regulated transcript peptide to the mouse spinal cord
. Eur J Pharmacol 2000;399:165–9.
. Petitjean H, Pawlowski SA, Fraine SL, Sharif B, Hamad D, Fatima T, Berg J, Brown CM, Jan LY, Ribeiro-da-Silva A, Braz JM, Basbaum AI, Sharif-Naeini R. Dorsal horn
parvalbumin neurons are gate-keepers of touch-evoked pain
after nerve injury. Cell Rep 2015;13:1246–57.
. Polgar E, Durrieux C, Hughes DI, Todd AJ. A quantitative study of inhibitory
interneurons in laminae I-III of the mouse spinal dorsal horn
. PLoS One 2013;8:e78309.
. R: a language and environment for statistical computing. Vol. 2018. Vienna, Austria: R Foundation for Statistical Computing, 2013.
. Rosenberg AB, Roco CM, Muscat RA, Kuchina A, Sample P, Yao Z, Graybuck LT, Peeler DJ, Mukherjee S, Chen W, Pun SH, Sellers DL, Tasic B, Seelig G. Single-cell profiling of the developing mouse brain and spinal cord
with split-pool barcoding. Science 2018;360:176–82.
. Roussy G, Dansereau MA, Baudisson S, Ezzoubaa F, Belleville K, Beaudet N, Martinez J, Richelson E, Sarret P. Evidence for a role of NTS2 receptors in the modulation of tonic pain
sensitivity. Mol Pain
. Sandor K, Krishnan S, Agalave NM, Krock E, Salcido JV, Fernandez-Zafra T, Khoonsari PE, Svensson CI, Kultima K. Spinal injection of newly identified cerebellin-1 and cerebellin-2 peptides induce mechanical hypersensitivity in mice. Neuropeptides 2018;69:53–9.
. Sardella TC, Polgar E, Garzillo F, Furuta T, Kaneko T, Watanabe M, Todd AJ. Dynorphin is expressed primarily by GABAergic neurons that contain galanin in the rat dorsal horn
. Mol Pain
. Sarret P, Esdaile MJ, Perron A, Martinez J, Stroh T, Beaudet A. Potent spinal analgesia elicited through stimulation of NTS2 neurotensin receptors. J Neurosci 2005;25:8188–96.
. Sathyamurthy A, Johnson KR, Matson KJE, Dobrott CI, Li L, Ryba AR, Bergman TB, Kelly MC, Kelley MW, Levine AJ. Massively parallel single nucleus transcriptional profiling defines spinal cord
neurons and their activity during behavior. Cell Rep 2018;22:2216–25.
. Schipper J, Steinbusch HW, Vermes I, Tilders FJ. Mapping of CRF-immunoreactive nerve fibers in the medulla oblongata and spinal cord
of the rat. Brain Res 1983;267:145–50.
. Schmidtko A, Gao W, Konig P, Heine S, Motterlini R, Ruth P, Schlossmann J, Koesling D, Niederberger E, Tegeder I, Friebe A, Geisslinger G. cGMP produced by NO-sensitive guanylyl cyclase essentially contributes to inflammatory and neuropathic pain
by using targets different from cGMP-dependent protein kinase I. J Neurosci 2008;28:8568–76.
. Shi TJ, Winzer-Serhan U, Leslie F, Hokfelt T. Distribution of alpha2-adrenoceptor mRNAs in the rat lumbar spinal cord
in normal and axotomized rats. Neuroreport 1999;10:2835–9.
. Shimomura Y, Harada M, Goto M, Sugo T, Matsumoto Y, Abe M, Watanabe T, Asami T, Kitada C, Mori M, Onda H, Fujino M. Identification of neuropeptide W as the endogenous ligand for orphan G-protein-coupled receptors GPR7 and GPR8. J Biol Chem 2002;277:35826–32.
. Sivilotti L, Woolf CJ. The contribution of GABAA and glycine receptors to central sensitization: disinhibition and touch-evoked allodynia in the spinal cord
. J Neurophysiol 1994;72:169–79.
. Song ZH, Takemori AE. Stimulation by corticotropin-releasing factor of the release of immunoreactive dynorphin A from mouse spinal cords in vitro. Eur J Pharmacol 1992;222:27–32.
. Spampinato S, Romualdi P, Candeletti S, Cavicchini E, Ferri S. Distinguishable effects of intrathecal dynorphins, somatostatin, neurotensin and s-calcitonin on nociception and motor function in the rat. PAIN
. Sun L, Wang H, Hu J, Han J, Matsunami H, Luo M. Guanylyl cyclase-D in the olfactory CO2 neurons is activated by bicarbonate. Proc Natl Acad Sci U S A 2009;106:2041–6.
. Thim L, Nielsen PF, Judge ME, Andersen AS, Diers I, Egel-Mitani M, Hastrup S. Purification and characterisation of a new hypothalamic satiety peptide, cocaine and amphetamine regulated transcript (CART), produced in yeast. FEBS Lett 1998;428:263–8.
. Tiong SY, Polgar E, van Kralingen JC, Watanabe M, Todd AJ. Galanin-immunoreactivity identifies a distinct population of inhibitory
interneurons in laminae I-III of the rat spinal cord
. Mol Pain
. Todd AJ. Identifying functional populations among the interneurons in laminae I-III of the spinal dorsal horn
. Mol Pain
. Uchio N, Doi M, Matsuo M, Yamazaki F, Mizoro Y, Hondo M, Sakurai T, Okamura H. Circadian characteristics of mice depleted with GPR7. Biomed Res 2009;30:357–64.
. Upadhya MA, Nakhate KT, Kokare DM, Singh U, Singru PS, Subhedar NK. CART peptide in the nucleus accumbens shell acts downstream to dopamine and mediates the reward and reinforcement actions of morphine. Neuropharmacology 2012;62:1823–33.
. van den Brink SC, Sage F, Vertesy A, Spanjaard B, Peterson-Maduro J, Baron CS, Robin C, van Oudenaarden A. Single-cell sequencing reveals dissociation-induced gene expression
in tissue subpopulations. Nat Methods 2017;14:935–6.
. Walsh MA, Graham BA, Brichta AM, Callister RJ. Evidence for a critical period in the development
of excitability and potassium currents in mouse lumbar superficial dorsal horn
neurons. J Neurophysiol 2009;101:1800–12.
. Wang F, Flanagan J, Su N, Wang LC, Bui S, Nielson A, Wu X, Vo HT, Ma XJ, Luo Y. RNAscope: a novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues. J Mol Diagn 2012;14:22–9.
. Wang Y, Wang M, Yin S, Jang R, Wang J, Xue Z, Xu T. NeuroPep: a comprehensive resource of neuropeptides. Database (Oxford) 2015;2015:bav038.
. Yaksh TL. Behavioral and autonomic correlates of the tactile evoked allodynia produced by spinal glycine inhibition: effects of modulatory receptor systems and excitatory amino acid antagonists. PAIN
. Yamamoto T, Saito O, Shono K, Tanabe S. Anti-hyperalgesic effects of intrathecally administered neuropeptide W-23, and neuropeptide B, in tests of inflammatory pain
in rats. Brain Res 2005;1045:97–106.
. Yamamoto T, Saito O, Shono K, Tanabe S. Effects of intrathecal and i.c.v. administration of neuropeptide W-23 and neuropeptide B on the mechanical allodynia induced by partial sciatic nerve ligation in rats. Neuroscience 2006;137:265–73.
. Young JM, Waters H, Dong C, Fulle HJ, Liman ER. Degeneration of the olfactory guanylyl cyclase D gene during primate evolution. PLoS One 2007;2:e884.