Mechanical hypersensitivity is a debilitating symptom for millions of chronic pain patients. It exists in distinct forms, including brush-evoked dynamic and filament-evoked punctate hypersensitivities. We reduced dynamic mechanical hypersensitivity induced by nerve injury or inflammation in mice by ablating a group of adult spinal neurons defined by developmental co-expression of VGLUT3 and Lbx1 (VT3Lbx1 neurons): the mice lost brush-evoked nocifensive responses and conditional place aversion. Electrophysiological recordings show that VT3Lbx1 neurons form morphine-resistant polysynaptic pathways relaying inputs from low-threshold Aβ mechanoreceptors to lamina I output neurons. The subset of somatostatin-lineage neurons preserved in VT3Lbx1-neuron-ablated mice is largely sufficient to mediate morphine-sensitive and morphine-resistant forms of von Frey filament-evoked punctate mechanical hypersensitivity. Furthermore, acute silencing of VT3Lbx1 neurons attenuated pre-established dynamic mechanical hypersensitivity induced by nerve injury, suggesting that these neurons may be a cellular target for treating this form of neuropathic pain.
Your institute does not have access to this article
Open Access articles citing this article.
Nature Open Access 25 May 2022
Neuroscience Bulletin Open Access 05 March 2022
Nature Communications Open Access 29 September 2021
Subscribe to Nature+
Get immediate online access to the entire Nature family of 50+ journals
Subscribe to Journal
Get full journal access for 1 year
only $4.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Melzack, R. & Wall, P.D. Pain mechanisms: a new theory. Science 150, 971–979 (1965).
Mendell, L.M. Constructing and deconstructing the gate theory of pain. Pain 155, 210–216 (2014).
Braz, J., Solorzano, C., Wang, X. & Basbaum, A.I. Transmitting pain and itch messages: a contemporary view of the spinal cord circuits that generate gate control. Neuron 82, 522–536 (2014).
Lu, Y. et al. A feed-forward spinal cord glycinergic neural circuit gates mechanical allodynia. J. Clin. Invest. 123, 4050–4062 (2013).
Duan, B. et al. Identification of spinal circuits transmitting and gating mechanical pain. Cell 159, 1417–1432 (2014).
Foster, E. et al. Targeted ablation, silencing, and activation establish glycinergic dorsal horn neurons as key components of a spinal gate for pain and itch. Neuron 85, 1289–1304 (2015).
Petitjean, H. et al. Dorsal horn parvalbumin neurons are gate-keepers of touch-evoked pain after nerve injury. Cell Rep. 13, 1246–1257 (2015).
Cui, L. et al. Identification of early RET+ deep dorsal spinal cord interneurons in gating pain. Neuron 91, 1137–1153 (2016).
Price, T.J., Cervero, F., Gold, M.S., Hammond, D.L. & Prescott, S.A. Chloride regulation in the pain pathway. Brain Res. Brain Res. Rev. 60, 149–170 (2009).
von Hehn, C.A., Baron, R. & Woolf, C.J. Deconstructing the neuropathic pain phenotype to reveal neural mechanisms. Neuron 73, 638–652 (2012).
Kuner, R. Spinal excitatory mechanisms of pathological pain. Pain 156 (Suppl. 1), S11–S17 (2015).
Treede, R.D. Gain control mechanisms in the nociceptive system. Pain 157, 1199–1204 (2016).
Truini, A., Garcia-Larrea, L. & Cruccu, G. Reappraising neuropathic pain in humans--how symptoms help disclose mechanisms. Nat. Rev. Neurol. 9, 572–582 (2013).
Koltzenburg, M., Lundberg, L.E. & Torebjörk, H.E. Dynamic and static components of mechanical hyperalgesia in human hairy skin. Pain 51, 207–219 (1992).
Ochoa, J.L. & Yarnitsky, D. Mechanical hyperalgesias in neuropathic pain patients: dynamic and static subtypes. Ann. Neurol. 33, 465–472 (1993).
Miraucourt, L.S., Moisset, X., Dallel, R. & Voisin, D.L. Glycine inhibitory dysfunction induces a selectively dynamic, morphine-resistant, and neurokinin 1 receptor- independent mechanical allodynia. J. Neurosci. 29, 2519–2527 (2009).
Field, M.J., Bramwell, S., Hughes, J. & Singh, L. Detection of static and dynamic components of mechanical allodynia in rat models of neuropathic pain: are they signalled by distinct primary sensory neurones? Pain 83, 303–311 (1999).
Samuelsson, M., Leffler, A.S. & Hansson, P. Dynamic mechanical allodynia: on the relationship between temporo-spatial stimulus parameters and evoked pain in patients with peripheral neuropathy. Pain 115, 264–272 (2005).
Torebjörk, H.E., Lundberg, L.E. & LaMotte, R.H. Central changes in processing of mechanoreceptive input in capsaicin-induced secondary hyperalgesia in humans. J. Physiol. (Lond.) 448, 765–780 (1992).
Campbell, J.N., Raja, S.N., Meyer, R.A. & Mackinnon, S.E. Myelinated afferents signal the hyperalgesia associated with nerve injury. Pain 32, 89–94 (1988).
Lindblom, U. & Verrillo, R.T. Sensory functions in chronic neuralgia. J. Neurol. Neurosurg. Psychiatry 42, 422–435 (1979).
Price, D.D., Bennett, G.J. & Rafii, A. Psychophysical observations on patients with neuropathic pain relieved by a sympathetic block. Pain 36, 273–288 (1989).
Cavanaugh, D.J. et al. Distinct subsets of unmyelinated primary sensory fibers mediate behavioral responses to noxious thermal and mechanical stimuli. Proc. Natl. Acad. Sci. USA 106, 9075–9080 (2009).
Boada, M.D. et al. Fast-conducting mechanoreceptors contribute to withdrawal behavior in normal and nerve injured rats. Pain 155, 2646–2655 (2014).
Xu, Z.Z. et al. Inhibition of mechanical allodynia in neuropathic pain by TLR5-mediated A-fiber blockade. Nat. Med. 21, 1326–1331 (2015).
Malmberg, A.B., Chen, C., Tonegawa, S. & Basbaum, A.I. Preserved acute pain and reduced neuropathic pain in mice lacking PKCgamma. Science 278, 279–283 (1997).
Peirs, C. et al. Dorsal horn circuits for persistent mechanical pain. Neuron 87, 797–812 (2015).
Madisen, L. et al. A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat. Neurosci. 13, 133–140 (2010).
Todd, A.J. Neuronal circuitry for pain processing in the dorsal horn. Nat. Rev. Neurosci. 11, 823–836 (2010).
Light, A.R., Trevino, D.L. & Perl, E.R. Morphological features of functionally defined neurons in the marginal zone and substantia gelatinosa of the spinal dorsal horn. J. Comp. Neurol. 186, 151–171 (1979).
Abraira, V.E. & Ginty, D.D. The sensory neurons of touch. Neuron 79, 618–639 (2013).
Bourane, S. et al. Identification of a spinal circuit for light touch and fine motor control. Cell 160, 503–515 (2015).
Ji, R.R., Berta, T. & Nedergaard, M. Glia and pain: is chronic pain a gliopathy? Pain 154 (Suppl. 1), S10–S28 (2013).
Ray, R.S. et al. Impaired respiratory and body temperature control upon acute serotonergic neuron inhibition. Science 333, 637–642 (2011).
Bourane, S. et al. Gate control of mechanical itch by a subpopulation of spinal cord interneurons. Science 350, 550–554 (2015).
Baba, H. et al. Removal of GABAergic inhibition facilitates polysynaptic A fiber-mediated excitatory transmission to the superficial spinal dorsal horn. Mol. Cell. Neurosci. 24, 818–830 (2003).
Torsney, C. & MacDermott, A.B. Disinhibition opens the gate to pathological pain signaling in superficial neurokinin 1 receptor-expressing neurons in rat spinal cord. J. Neurosci. 26, 1833–1843 (2006).
Edmonds, B., Gibb, A.J. & Colquhoun, D. Mechanisms of activation of glutamate receptors and the time course of excitatory synaptic currents. Annu. Rev. Physiol. 57, 495–519 (1995).
McNicol, E.D., Midbari, A. & Eisenberg, E. Opioids for neuropathic pain. Cochrane Database Syst. Rev. 8, CD006146 (2013).
Due, M.R. et al. Carbamazepine potentiates the effectiveness of morphine in a rodent model of neuropathic pain. PLoS One 9, e107399 (2014).
Yaksh, T.L. 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 37, 111–123 (1989).
Sherman, S.E. & Loomis, C.W. Morphine insensitive allodynia is produced by intrathecal strychnine in the lightly anesthetized rat. Pain 56, 17–29 (1994).
François, A. et al. The low-threshold calcium channel cav3.2 determines low-threshold mechanoreceptor function. Cell Rep. http://dx.doi.org/10.1016/j.celrep.2014.12.042 (2015).
Arcourt, A. et al. Touch receptor-derived sensory information alleviates acute pain signaling and fine-tunes nociceptive reflex coordination. Neuron 93, 179–193 (2017).
Prescott, S.A., Sejnowski, T.J. & De Koninck, Y. Reduction of anion reversal potential subverts the inhibitory control of firing rate in spinal lamina I neurons: towards a biophysical basis for neuropathic pain. Mol. Pain 2, 32 (2006).
Bushnell, M.C., Ceko, M. & Low, L.A. Cognitive and emotional control of pain and its disruption in chronic pain. Nat. Rev. Neurosci. 14, 502–511 (2013).
Bai, L. et al. Genetic identification of an expansive mechanoreceptor sensitive to skin stroking. Cell 163, 1783–1795 (2015).
Cobos, E.J. & Portillo-Salido, E. “Bedside-to-bench” behavioral outcomes in animal models of pain: beyond the evaluation of reflexes. Curr. Neuropharmacol. 11, 560–591 (2013).
Mogil, J.S. Animal models of pain: progress and challenges. Nat. Rev. Neurosci. 10, 283–294 (2009).
Reimer, M., Helfert, S.M. & Baron, R. Phenotyping neuropathic pain patients: implications for individual therapy and clinical trials. Curr. Opin. Support. Palliat. Care 8, 124–129 (2014).
Michael, S.K., Brennan, J. & Robertson, E.J. Efficient gene-specific expression of cre recombinase in the mouse embryo by targeted insertion of a novel IRES-Cre cassette into endogenous loci. Mech. Dev. 85, 35–47 (1999).
Lee, E.C. et al. A highly efficient Escherichia coli-based chromosome engineering system adapted for recombinogenic targeting and subcloning of BAC DNA. Genomics 73, 56–65 (2001).
Liu, P., Jenkins, N.A. & Copeland, N.G. A highly efficient recombineering-based method for generating conditional knockout mutations. Genome Res. 13, 476–484 (2003).
Rodríguez, C.I. et al. High-efficiency deleter mice show that FLPe is an alternative to Cre-loxP. Nat. Genet. 25, 139–140 (2000).
Taniguchi, H. et al. A resource of Cre driver lines for genetic targeting of GABAergic neurons in cerebral cortex. Neuron 71, 995–1013 (2011).
Liu, Y. et al. VGLUT2-dependent glutamate release from nociceptors is required to sense pain and suppress itch. Neuron 68, 543–556 (2010).
Lou, S., Duan, B., Vong, L., Lowell, B.B. & Ma, Q. Runx1 controls terminal morphology and mechanosensitivity of VGLUT3-expressing C-mechanoreceptors. J. Neurosci. 33, 870–882 (2013).
Decosterd, I. & Woolf, C.J. Spared nerve injury: an animal model of persistent peripheral neuropathic pain. Pain 87, 149–158 (2000).
We thank E.J. Cobos, who developed the protocol for measuring brush-evoked dynamic mechanical hypersensitivity. We thank S. Dymecki for the ROSA26-LSL-FSF-hM4Di and ROSA26-Flpe mice, Z.J. Huang and the Jackson laboratory for the SOM-IRES-Cre (SOMCre) mice, and the Allen Brain Institute and the Jackson Laboratory for the ROSA26-lsl-tdTomato mice. We thank W. Knowlton for her assistance with thermal pain measurement and D. Zhou (Shanghai Medviser Co. Ltd.) for his assistance with figure preparation. The work was supported by NIH grants to Q.M. and M.G. (R01 NS086372), to Q.M. (R01 DE018025 and R01 NS072031) and to B.B.L. (R01 DK111401, R01 DK075632, R01 DK096010, R01 DK089044, P30 DK046200-BNORC Transgenic core and P30 DK057521-BADERC Transgenic core). Y.W., L.C. and Y.Z. were supported by Grants from National Natural Science Fund of China (31471027, 31571085 and 81100815, 31300922) and by the 111 Project of China.
The authors declare no competing financial interests.
Integrated supplementary information
Supplementary Figure 1 Generation of Vglut3-ires-Cre (VT3Cre) knock-in mice and characterization of VT3Cre-tdTomato neurons
(a) To generate VT3Cre knock-in mice, a cassette containing the Cre recombinase gene preceded by an internal ribosomal entry sequence (“ires”) was targeted just distal to the stop codon of the endogenous Vglut3 (Slc17a8) allele. The “neo” cassette flanked with the FRT sites was removed by crossing with Flpe deletor mouse line [Rodriguez, C.I. et al. Nat Genet. 25, 139-140 (2000)]. (b) Double staining of tdTomato with two markers (“green”), detected by in situ hybridization (ISH) for GAD1 mRNA (b, top, representative images of 9 sections from 3 mice), a marker for GABAergic inhibitory neurons, and GlyT2 mRNA (b, bottom, representative images of 9 sections from 3 mice), a marker for glycinergic neurons, on spinal sections from adult VT3Cre-tdTomato mice. Right panels in b-d represent higher magnification of the boxed areas in left panels. Note virtually no co-localization, with only ~1% of tdTomato+ neurons showing detectable GAD1 mRNA (7/709) or GlyT2 mRNA (10/720). (c) Spinal sections from adult mice showing tdTomato with SOM mRNA. Arrows indicate co-localization. The table shows quantitative colocalization in different laminae. (d) Colocalization of VT3Cre-tdTomato+ neurons with the Calb2/Calretinin protein. Images are representatives of 12 sections from 3 mice. Arrows indicate co-localization. 45% of adult Calb2+ neurons were labeled by our knock-in VT3Cre, which is different from only 8% of adult Calb2+ neurons labeled by the transgenic VT3::Cre used by Peirs et al. [Peirs, C. et al. Neuron 87, 797-812 (2015)]. Additionally, transgenic VT3::Cre failed to mark neurons exhibiting delayed firing [Peirs, C. et al. Neuron 87, 797-812 (2015)], whereas a majority of type 2 neurons labeled by our knock-in VT3Cre displayed the delayed firing pattern (see Supplementary Fig. 6f). This difference is probably due to the absence of full regulatory elements in the BAC genomic fragment used to drive their transgenic VT3::Cre. Scale bars represent 50 μm in the left panels and 20 μm in the right panels, respectively.
Supplementary Figure 2 Characterization of VT3Cre-tdTomato+ neurons in control versus VT3Lbx1-ablated mice in DRG and brain
(a) No gross loss of VT3Cre-tdTomato+ neurons in the dorsal root ganglia (DRG) or cortex, consistent with lack of Lbx1-Flpe expression in these regions [Braz, J., et al. Neuron 82, 522-536 (2014); Lu, Y., et al. J Clin Invest. 123, 4050-4062 (2013)]. (b) VT3Cre-tdTomato+ neurons in spinal trigeminal nuclei (“Sp5”, Bregma -7.48mm) and supratrigeminal nuclei (“Su5”, Bregma -5.34mm) were lost in VT3Lbx1-ablated mice. Su5 had been previously reported to involve nocifensive responses to painful stimuli applied to oral cavity regions [Duan, B., et al. Cell 159, 1417-1432 (2014)]. No gross loss was observed in the area postrema (AP), the nucleus of the solitary tract (NTS), raphe magnus nuclei (“RMg”, Bregma -5.34mm), raphe pallidus nuclei (“Rpa”, Bregma -5.34mm) or the trapezoid body (“Tz”, Bregma -5.34mm). Scale bars represent 100 μm. Reference schematics were from “Franklin, K.B. & Paxinos, G. The Mouse Brain in Stereotaxic Coordinates. 1996, Academic Press”.
(a) The Rota-Rod assay. No significant difference in falling latencies was detected between VT3Lbx1-ablated (“VT3Lbx1 Abl”; purple circles) and control (blue circles) mice (n = 10 in each group; two-tailed Student's unpaired t test, P = 0.4126, t(18) = 0.8388). (b,c) Touch sensation. No significant difference was detected in response numbers per three trials of brushing (b, n = 11, Control; n = 11, VT3Lbx1 Abl; two-tailed Student's unpaired t test, P = 0.2500, t(20) = 1.185). The quick hindpaw lifting and/or walking away in response to brushing was scored as 1 for this assay, but would be scored as 0 for the dynamic allodynia assay described in Fig. 2. No significant difference in latencies of removing sticky tape was detected (c, n = 9, Control; n = 10, VT3Lbx1 Abl; two-tailed Student's unpaired t test, P = 0.8383, t(17) = 0.2072). (d) No significant difference was detected in the Hargreaves heat test (n = 11 in each group; two-tailed Student's unpaired t test, P = 0.6379, t(20) = 0.4779). (e) No significant difference was detected in hot plate assays (n = 10 in each group; two-tailed Student's unpaired t test for flinch and lick latency at each temperature: P = 0.3770, t(18) = 0.9059; P = 0.4705, t(18) = 0.7372); P = 0.5331, t(18) = 0.6355; P = 0.1897, t(18) = 1.363; P = 0.1201, t(18) = 1.632; P = 0.7778, t(18) = 0.2864. (f) Cold plate assays. No significant difference in latencies of forepaw flinching or hindpaw licking was detected (n = 10 in each group; two-tailed Student's unpaired t test, P = 0.636, t(18) = 0.4815; P = 0.4368, t(18) = 0.7953). (g) No significant difference in behavioral responses to acetone-evoked cooling was detected (n = 10 in each group; two-tailed Student's unpaired t test, P = 0.7621, t(18) = 0.3073). (h) Itch test by nape injection of compound 48/80 (100 μg). No significant difference in evoked scratching bouts was detected (n = 9, Control; n = 10, VT3Lbx1 Abl; two-tailed Student's unpaired t test, P = 0.2335, t(17) = 1.235). (i) Itch test by nape injection of chloroquine (200 μg). No significant difference in evoked scratching bouts was detected (n = 7, Control; n = 8, VT3Lbx1 Abl; two-tailed Student's unpaired t test, P = 0.3681, t(13) = 0.9324). (j) Cold allodynia assays. No significant difference in cold allodynia at day 7 after SNI was detected (n = 5, Control; n = 5, VT3Lbx1 Abl; two-tailed Student's unpaired t test, P = 0.9329, t(8) = 0.8687). Student’s unpaired t test. Data are represented as mean ± SEM.
Supplementary Figure 4 Additional analyses on mechanical hypersensitivity and dorsal horn plasticity in VT3Lbx1-ablated (VT3Lbx1-Abl) mice, SOMLbx1-ablated (SOMLbx1-Abl) mice and their control littermates
(a) Loss of brush-evoked dynamic (brush assay) mechanical hypersensitivity, but not von Frey-evoked punctate hypersensitivity, following SNI in both male and female VT3Lbx1-ablated mice (VT3Lbx1 Abl). (Left, dynamic) Male: n = 8, Control; n = 6, VT3Lbx1 Abl; two-way ANOVA with Bonferroni’s post hoc analysis; F1, 72 = 556.20, P < 0.001. Female: n = 5, Control; n = 5, VT3Lbx1 Abl; two-way ANOVA with Bonferroni’s post hoc analysis; F1, 72 = 510.10, P < 0.001. (Right, punctate) Male: control versus VT3Lbx1 Abl mice; before SNI, two-tailed student’s unpaired t test, t12 = 3.392, P = 0.0053; after SNI, two-way ANOVA with Bonferroni’s post hoc analysis, F5, 72 = 8.35, P = 0.1469. Female: control versus VT3Lbx1 Abl mice; before SNI, student’s unpaired t test, t8 = 2.603, P = 0.0315; after SNI, two-way ANOVA with Bonferroni’s post hoc analysis, F5, 48 = 1.248, P = 0.5823. (b) Mechanical hypersensitivity following intrathecal injection of bicuculline and strychnine in SOMLbx1-ablated mice (SOMLbx1 Abl) and their control littermates. Loss of dynamic hypersensitivity (n = 5, Control; n = 5, SOMLbx1 Abl; two-way ANOVA with Bonferroni’s post hoc analysis, F(1, 56) = 675.40, P < 0.0001;), and loss of punctate hypersensitivity (n = 5, Control; n = 5, SOMLbx1 Abl; before SNI, two-tailed student’s unpaired t test, P < 0.0001; after SNI, two-way ANOVA with Bonferroni’s post hoc analysis, P < 0.0001). (c) Microglia activation following SNI (day 7 after SNI) marked by increased expression of Iba1. An increase in Iba1 expression in the ipsilateral (“Ipsi”) dorsal horn in comparison with the contralateral (“Cont”) dorsal horn both in wild type control mice (student’s unpaired t test, t(16) = 2.831, P = 0.0120, t(16) = 2.831) and in SOMLbx1 Abl mice (two-tailed student’s unpaired t test, t(16) = 3.519, P = 0.0028). No significant (“N.S.”) difference in Iba1 expression in the ipsilateral dorsal horn (“Ipsi”, normalized to contralateral side) was detected between control mice and SOMLbx1 Abl (Student’s unpaired t test, t(16) = 0.6419, P = 0.5300; n = 9 sections, from 3 Control mice; n = 9 sections, from 3 SOMLbx1 Abl mice). *, P < 0.05, **, P < 0.01. Our studies do not rule out unknown molecular changes. Data are represented as mean ± SEM. The scale bar represents 100 μm.
Supplementary Figure 5 Criteria for identifying Aβ-, Aδ- and C-fiber-mediated synaptic inputs in the spinal dorsal horn
(a) Image is our sagittal spinal cord slice preparation with DRG and full length of dorsal root attached. Schematics illustrate how to measure distance from the tip of the stimulation electrode to the tip of the recording electrode (the distance is about 5 mm under our recording conditions), and how to perform dorsal root compound action potential recordings. (b) Summary of thresholds, conduction velocities, and shortest latencies for primary afferent Aβ, Aδ and C fibers determined by dorsal root compound AP recordings (from 9 mice). (c) Image is our sagittal spinal cord slice preparation with DRG and full length of dorsal root. Schematics illustrate how to measure distance from the tip of the stimulation electrode to the entrance of the attached dorsal root (the distance is about 8 mm under our recording condition), and the location of the suction electrode when we perform recordings in the spinal dorsal horn. (d) Typical traces illustrate high frequency following with constant latency to assess monosynaptic inputs or not. (e) Parameters for identifying Aβ/Aδ/C-fiber mediated synaptic inputs in the spinal dorsal horn.
Supplementary Figure 6 Additional characterization of Aβ inputs and outputs in VT3Cre-tdTomato+ neurons
(a) Schematics illustrate how to prepare sagittal spinal cord slices to preserve low-threshold Aβ-fiber inputs onto dorsal horn neurons, based on the relative positions of the second cut to the entrance region of the dorsal root. (b) Low threshold Aβ intensity stimulation-induced inputs/outputs in a type 1 VT3Cre-tdTomato+ neuron, indicating receiving Aβ with feedforward inhibition. Black arrows indicate artifacts. Red arrows indicate eEPSC/eEPSP and black arrowheads indicate eIPSC/eIPSP. The recordings are composed of three repeated traces. (c) Summary of low threshold Aβ intensity-induced inputs/outputs in VT3Cre-tdTomato+ neuron in different laminae under normal recording conditions. Most VT3Cre-tdTomato+ neurons that received Aβ evoked-EPSCs under normal recording conditions also received eIPSCs, indicating feedforward inhibition. (d) Summary of low threshold Aβ intensity-induced inputs/outputs in VT3Cre-tdTomato+ neurons under both control and disinhibition conditions, with or without APV, a NMDAR antagonist (see recording traces in Fig. 5a-c). (e) Schematics showing primary afferent inputs to VT3Cre-tdTomato+ neurons. “V1” and “V2” represent type 1 and type 2 neurons, respectively. “In” represents inhibitory interneurons. “V1?” indicates a type 1 VT3Cre-tdTomato+ neuron or a V1-like neuron not marked by VT3Cre. (f) Percentages of type1 and type 2 VT3Cre-tdTomato+ neurons that displayed distinct firing patterns following current injection.
Supplementary Figure 7 Subtypes of type 1 and 2 neurons based on inputs following increased electrical stimulation intensity
(a,b) Afferent inputs to VGLUT3 lineage neurons following dorsal root stimulations at different electrical intensities under normal ACSF conditions. Red arrows indicate fast Aβ inputs, red arrowheads indicate slow Aβ/Aδ inputs, and blue arrows indicate C fiber inputs. Black arrows indicate stimulation artifacts. Schematics showing primary afferent inputs to 1a, 1b, 2a and 2b neurons, respectively. “In” represents inhibitory interneurons. “HT” indicates electrically “High Threshold”. “HT-Aβ/Aδ” and “HT-Aβ” indicate Aβ or Aδ inputs emerged or elevated following high electric intensity stimulations. In Fig. 5 and Supplementary Figure 6e, we have described that all type 1 neurons receive both fast (not necessarily monosynaptic) Aβ inputs with feedforward inhibition, and here we found that 1b cells additionally received HT-Aβ/Aδ inputs with AP firing. Type 2 neurons receive slow Aβ inputs via an interneuron that receives Aβ inputs with feedforward inhibition (see Fig. 5 and Supplementary Figure 6e), and 2a, but not 2b, additionally received C fiber inputs with AP firing and HT Aβ inputs without AP firing. (c) Schematic illustrates how to estimate latency of the second response (“t2”) when neurons receive more than one input (the actual latency should be ≤ t2). For the second input to this neuron, the latency is ≤ 4.7 ms, shorter than 6 ms (the shortest latency for Aδ inputs, see Supplementary Figure 5b), thereby belonging to Aβ inputs. For those neurons with latencies over 6 ms, we cannot be certain if they are Aβ or Aδ inputs, which are referred to as “Aβ/Aδ” here. All recordings are composed of three repeated traces.
Supplementary Figure 8 Summary of the impact of VT3Lbx1 or SOMLbx1 neuron ablation on Aβ inputs to I–IIo neurons, induced by nerve injury (SNI)
SNI opened Aβ inputs to I-IIo neurons in wild type control littermates (in comparison with “naïve” control mice without SNI). Note the selective loss of nerve injury-induced slow polysynaptic Aβ inputs to I-IIo neurons in VT3Lbx1-ablated mice (“VT3Lbx1 Abl”) following SNI, and loss of both fast and slow inputs in SOMLbx1-ablated mice (“SOMLbx1 Abl”) in comparison with control littermates with SNI.
Low threshold Aβ intensity stimulation-induced inputs/outputs in spinal VT3Cre-tdTomato+ neurons under naïve conditions without SNI or after SNI. Black arrows indicate stimulation artifacts, red vertical dashed lines indicate the 10-ms time point following stimulation, and red horizontal dashed lines indicate baseline. SNI underwent at P18-21, and slice recordings were performed at P27-31. We observed sharp increase in the percentages of VT3Cre-tdTomato+ neurons displaying Aβ-evoked APs, from 4.4% (5/114) in naïve control mice to 41% (7/17; c2 test, χ2 = 17.1429, P < 0.001) following SNI under our recording conditions. It should be pointed out that the latencies of firing the first Aβ-evoked AP are shorter than 20 ms in 57% (4/7) neurons with AP firing (from the stimulation artifact to the peak amplitude of the first AP), whereas under pure disinhibition conditions induced by bicuculline plus strychnine, only 18% (7/38) of neurons displaying latencies shorter than 20 ms (χ2 test, χ2 = 4.7988, P = 0.0285). The shortened latencies might reflect the fact that nerve injury causes both sensitization and disinhibition in dorsal spinal neurons [Foster, E., et al. Neuron 85, 1289-1304 (2015); Petitjean, H., et al. Cell Rep. pii: S2211-1247(15)01134-1(2015); Price, T.J., et al. Brain Res Rev. 60, 149-170 (2009); von Hehn, C.A., et al. Neuron 73, 638-652 (2012); Kuner, R., et al. Pain 156 Suppl 1, S11-17 (2015)]. All recordings are composed of three repeated traces.
Supplementary Figure 10 VT3Lbx1-ablated mice show a loss of slow but not fast Aβ inputs to I–IIo neurons induced by disinhibition with bicuculline plus strychnine
(a,b) Low threshold Aβ intensity stimulation-induced inputs/outputs in I-IIo neurons in control and ablation (“VT3Lbx1 Abl”) mice after disinhibition by bicuculline plus strychnine. Left: Aβ inputs tested by Aβ-eEPSCs at -70 mV; Right: Aβ outputs tested by Aβ-evoked EPSPs/APs at resting membrane potentials. Black arrows indicate stimulation artifacts. Red vertical dashed lines indicate the 10-ms time point following stimulation. Red horizontal dashed lines indicate baseline. Red arrows and arrowheads indicate fast-onset and slow-onset Aβ inputs, respectively. All recordings are composed of three repeated traces. (c) Summary of low threshold Aβ intensity stimulation-induced inputs and AP outputs in I-IIo neurons in both control and VT3Lbx1-ablated mice after disinhibition by bicuculline and strychnine. Note a marked loss of slow-onset inputs (χ2 test; eEPSCs, P < 0.0001, χ2 = 62.49; evoked APs, P < 0.0001, χ2 = 46.98), but not fast-onset inputs (χ2 test; eEPSCs, P = 0.7439, χ2 = 0.106; evoked APs, P = 0.3975, χ2 = 0.7160). Data were collected from 8 control and 15 VT3Lbx1 Abl mice. ***, P < 0.001.
Supplementary Figure 11 Schematic for two pathways linking Aβ inputs from lamina III to projection neurons (PN) in lamina I (1 and 2)
“V1” and “V2” indicate type 1 and type 2 VT3Cre-tdTomato+ neurons. Pathway “1” indicates polysynaptic circuits linking Aβ inputs from lamina III to projection neurons in lamina I (“PN”). This pathway starts with V1 cells, a subset of which coexpresses SOM, and they are connected to PN neurons via interneurons with slow Aβ inputs only, such as V2 or V2-like cells (“A”, to be tested), or via vertical cells (“B”). V2 cells could be connected to vertical cells as well (to be tested). Dashed arrows indicate the uncertainty of direct or indirect connections. SOM+/- represents vertical cells that could belong, or do not belong, to SOM lineage neurons, and they could receive more direct Aβ inputs from lamina III (pathway “2”). “In” represents inhibitory neurons for feedforward inhibition.
Supplementary Figure 12 Effects of intrathecal morphine treatment on afferent inputs to dorsal horn neurons
(a) Summary of morphine effects on low threshold Aβ intensity stimulation-induced AP outputs in I-IIo neurons after SNI or treatment with bicuculline and strychnine (“Bic+Stry”), using the preparation retaining low threshold Aβ inputs (see recording traces in Fig. 7c). (b) Summary of C intensity stimulation-induced inputs/outputs in VT3Cre-negative neurons in vIIi, using the preparation removing low threshold Aβ inputs (see recording traces in Fig. 7d). (c) Effects of morphine on the resting membrane potentials in the superficial dorsal horn neurons. Following bath application of morphine (25 μM), only a small percentage of SOMCre-tdTomato+ neurons in lamina II and randomly picked neurons in lamina I-IIo were hyperpolarized with degrees of change ≥ 5 mV. We conclude that while morphine can directly inhibit a very small subset of dorsal horn neurons, morphine-mediated inhibition might operate mainly in primary afferents.
(a) Schematic showing sagittal dorsal root-spinal cord slice preparation that removed low-threshold Aβ-fiber inputs to dorsal horn neurons. (b) No response to Aβ intensity stimulations in both naïve and SNI mice. (c) C intensity stimulations evoked EPSCs insufficient to fire APs in naïve mice (0/30) (see Fig. 7d), but fired APs in 5/10 neurons in SNI mice (χ2 test; P < 0.0001, χ2 = 17.1429), plus an electrically high-threshold Aβ input without AP firing. Blue and red arrows indicate C fiber inputs and electrically high threshold Aβ inputs/outputs (“HT-Aβ”), respectively. Typical traces shown are representative responses of 5 out of 10 VT3Cre-negative neurons recorded in vIIi. Black arrows indicate stimulation artifact, red vertical dashed lines indicate the 10-ms time point following stimulation. All recordings are composed of three repeated traces.
Supplementary Figure 14 Schematic of spinal pathways mediating dynamic versus punctate mechanical hypersensitivity
VT3Lbx1 neuron-dependent pathway 1 could mediate brush-evoked dynamic allodynia, whereas VT3Lbx1 neuron-independent pathways 2 and 3 could potentially mediate punctate allodynia preserved in VT3Lbx1-ablated mice. VT3Lbx1 neuron-dependent pathway 4 transmits acute light mechanical information. All these pathways are dependent on SOMLbx1 neurons. “V1” and “V2” indicate type 1 and type 2 VT3Cre-tdTomato+ neurons. “SOM?” indicates vIIi neurons receiving C fiber inputs, and these neurons either represent SOMLbx1 neurons or neurons connected with SOMLbx1 neurons based on loss of such inputs in SOMLbx1-ablated mice. “In” indicates inhibitory neurons. “PN”: projection neurons. Solid arrows for monosynaptic inputs, and dashed arrows indicating uncertainty of mono- or poly-synaptic inputs. A series of genetically labeled inhibitory neurons have been shown to play a role in gating LTMR inputs to dorsal horn neurons [Duan, B., et al. Cell 159, 1417-1432 (2014); Foster, E., et al. Neuron 85, 1289-1304 (2015); Petitjean, H., et al. Cell Rep. pii: S2211-1247(15)01134-1(2015); Cui, L. et al. Neuron 91, 1137-1153 (2016).], but their relative contributions to the gating of pathways 1-3 remain to be determined. It should be pointed out that all these three pathways could potentially relay afferent inputs to projection neurons located laminae IV-VI as well (to be determined).
Peirs et al. reported that activation of spinal neurons marked by their transgenic VT3::Cre was able to develop punctate mechanical sensitivity [Peirs, C. et al. Neuron 87, 797-812 (2015)], whereas ablation of adult spinal neurons marked by coexpression of our VT3Cre and Lbx1Flpo (VT3Lbx1 neurons) did not affect the transmission of filament-evoked punctate hypersensitivity induced by nerve injury or strong inflammation (see Fig. 2). We envision several possibilities. Firstly, VT3Lbx1 neurons have a redundant, but non-essential, role in mediating this form of mechanical hypersensitivity. Secondly, there are 14% of VT3Cre-marked spinal neurons that are not ablated (see Fig. 1) and these neurons might mediate punctate mechanical hypersensitivity. Thirdly, considering that their VT3::Cre poorly labeled Calb2+ neurons in comparison with our knock-in mice (Supplementary Fig. 1), it remains to be determined if VT3::Cre might conversely mark spinal neurons not marked by our VT3Cre, and if activation of these neurons generated punctate mechanical hypersensitivity. Future intersectional ablation of spinal neurons coexpressing transgenic VT3::Cre and Lbx1Flpo, as we have done for VT3Lbx1 neurons here, would be informative for distinguishing these possibilities.
Peirs et at. also reported that mechanical hypersensitivity evoked by von Frey filaments is eliminated in Vglut3 conditional knockout mice [Peirs, C. et al. Neuron 87, 797-812 (2015)], which again contrasts with the preservation of this form of hypersensitivity following ablation of adult VT3Lbx1 neurons. Because VGLUT3 is expressed transiently in the dorsal spinal cord during postnatal development, we speculate that maturation of spinal substrates mediating VT3Lbx1 neuron-independent punctate mechanical hypersensitivity might be non-autonomously influenced by VGLUT3-dependent glutamate release during development. Scale bars represent 50 μm.
About this article
Cite this article
Cheng, L., Duan, B., Huang, T. et al. Identification of spinal circuits involved in touch-evoked dynamic mechanical pain. Nat Neurosci 20, 804–814 (2017). https://doi.org/10.1038/nn.4549
Neuroscience Bulletin (2022)
Nature Communications (2021)
MHCII-restricted T helper cells: an emerging trigger for chronic tactile allodynia after nerve injuries
Journal of Neuroinflammation (2020)
Expression of the pacemaker channel HCN4 in excitatory interneurons in the dorsal horn of the murine spinal cord
Molecular Brain (2020)