Reactive oxygen species are second messengers ofneurokinin signaling in peripheral sensory neuronsJohn E. Linleya,1, Lezanne Ooia,b,1, Louisa Pettingera, Hannah Kirtona, John P. Boylec, Chris Peersc, and Nikita Gampera,2

aInstitute of Membrane and Systems Biology, Faculty of Biological Sciences, and cLeeds Institute for Genetics, Health and Therapeutics, Faculty of Medicineand Health, University of Leeds, LS2 9JT Leeds, United Kingdom; and bUniversity of Western Sydney, School of Medicine, Penrith, New South Wales 2751,Australia

Edited* by Bertil Hille, University of Washington School of Medicine, Seattle, WA, and approved April 13, 2012 (received for review January 27, 2012)

Substance P (SP) is a prominent neuromodulator, which is pro-duced and released by peripheral damage-sensing (nociceptive)neurons; these neurons also express SP receptors. However, themechanisms of peripheral SP signaling are poorly understood.We report a signaling pathway of SP in nociceptive neurons: Act-ing predominantly through NK1 receptors and Gi/o proteins, SPstimulates increased release of reactive oxygen species from themitochondrial electron transport chain. Reactive oxygen species,functioning as second messengers, induce oxidative modificationand augment M-type potassium channels, thereby suppressingexcitability. This signaling cascade requires activation of phospho-lipase C but is largely uncoupled from the inositol 1,4,5-trisphos-phate sensitive Ca2+ stores. In rats SP causes sensitization of TRPV1and produces thermal hyperalgesia. However, the lack of couplingbetween SP signaling and inositol 1,4,5-trisphosphate sensitive Ca2+

stores, together with the augmenting effect onM channels, rendersthe SP pathway ineffective to excite nociceptors acutely and pro-duce spontaneous pain. Our study describes amechanism for neuro-kinin signaling in sensory neurons and provides evidence thatspontaneous pain and hyperalgesia can have distinct underlyingmechanisms within a single nociceptive neuron.

G protein coupled receptors | inflammatory pain | intracellular signaling |KCNQ | M current

Excitation of peripheral terminals of sensory neurons is a pri-mary event in somatosensation, including pain. A large pro-

portion of sensory neurons (mostly TRPV1+ damage-sensing or“nociceptive” neurons) produce neuropeptides such as substanceP (SP), which are released in response to nociceptive stimulationboth at spinal cord synapses and in the periphery (1). In the spinalcord SP acts as an excitatory neurotransmitter or cotransmitter (2,3), whereas peripheral SP release is thought to underlie the in-flammatory response known as “neurogenic inflammation” (4).Many peripheral nociceptors also express receptors for SP [neu-rokinin receptors (NKR) 1–3 (NK1–3)]; this expression may sug-gest paracrine or autocrine actions of this peptide. However, thenature of such actions is controversial, and the molecular eventstriggered by NKR in peripheral nociceptors are not well estab-lished. The NKR traditionally are classed as Gq/11-coupled Gprotein-coupled receptors (GPCR) that activate phospholipase C(PLC), with subsequent hydrolysis of membrane phosphatidyli-nositol 4,5-bisphosphate (PIP2) and release of inositol 1,4,5-tri-sphosphate (IP3) and diacylglycerol (DAG) (5, 6). Common down-stream steps of Gq/11 signaling include IP3-mediated release of Ca2+

from intracellular stores and DAG-mediated activation of PKC.However, it was hitherto unknown whether NKR in peripheralsensory neurons couple fully to this signaling cascade, becausea lack of coupling between NKR and Ca2+ release in dorsal rootganglia (DRG) neurons has been noted (ref. 7 and 8, but cf. ref.9). Nevertheless, it is accepted that peripherally acting SP sensi-tizes TRPV1 in nociceptors, an effect underlying inflammatoryhyperalgesia; the SP-induced TRPV1 sensitization was shown tobe mediated by PKC (7, 8). There is no consensus regarding theeffect of SP on sensory neuron excitability and neurotransmitter

release, and both positive (10) and negative (11) feedback effectshave been suggested. In sum, current understanding of the sig-naling cascades and functional significance of the peripheral SPreceptors is incomplete. However, given the robust effects of SPin the CNS, it is likely that SP strongly modulates somatosensa-tion in the periphery as well. Here we have investigated the in-tracellular signaling cascades triggered by NKR in DRG andtrigeminal (TG) nociceptors and identified reactive oxygen spe-cies (ROS) as second messengers in the signaling pathway. Wehave evaluated the effects of NKR on two ion channels importantfor controlling the excitability of nociceptors: TRPV1 andM-type(Kv7, KCNQ) K+ channels. In doing so, we have determined theconsequences of NKR activation for nociceptor excitability, forcalcitonin gene-related peptide (CGRP) release, and for the be-havioral response to noxious thermal stimuli. Crucially, our dataidentify the existence of distinct molecular mechanisms forspontaneous pain and hyperalgesia, which can coexist in a singlenociceptive neuron.

ResultsFunctional Expression of NKR in Mammalian Sensory Neurons. Toquantify the population of nociceptive neurons in rat DRG andTG which express functional NKR, we used the phenomenon ofTRPV1 sensitization by SP (7, 8). We focused our study on thesmall-diameter DRG neurons (20–30 μM), which we character-ized in detail in our previous study (12); 60–70% of these neu-rons express TRPV1 (12). In fura-2 Ca2+ imaging experiments,we used a protocol in which three brief applications of theTRPV1 agonist capsaicin (CAP) at a submaximal concentration(50 nM) were followed by the application of SP or selectiveagonists of NK1 (Sar-met SP), NK2 (β-ala NKA), or NK3(senktide) (all at 1 μM), with a subsequent fourth application of50 nM CAP (Fig. 1A). Finally, 1 μM CAP was applied. The EC50of CAP for rat TRPV1 is in the range of hundreds of nanomoles(13), and at high concentrations repetitive application of CAPproduces variable desensitization of TRPV1 (14). However, suc-cessive application of 50 nM CAP produced no desensitization ofthe response magnitude (Fig. 1B). SP (at a concentration thatactivates all three NKR) and the NK1-specific agonist Sar-metSP each caused ∼twofold augmentation of CAP responses in∼40% of DRG neurons (32/83 and 46/121 for SP and Sar-metSP, respectively). NK2- and NK3-specific agonists also induced

Author contributions: J.E.L., L.O., J.P.B., C.P., and N.G. designed research; J.E.L., L.O., L.P.,H.K., J.P.B., and N.G. performed research; J.E.L., L.O., L.P., H.K., J.P.B., C.P., and N.G.analyzed data; and J.E.L., L.O., L.P., J.P.B., C.P., and N.G. wrote the paper.

The authors declare no conflict of interest.

*This Direct Submission article had a prearranged editor.

Freely available online through the PNAS open access option.1J.E.L. and L.O. contributed equally to this work.2To whom correspondence should be addressed. E-mail: n.gamper@leeds.ac.uk.

See Author Summary on page 9246 (volume 109, number 24).

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1201544109/-/DCSupplemental.

E1578–E1586 | PNAS | Published online May 14, 2012 www.pnas.org/cgi/doi/10.1073/pnas.1201544109

sensitization of CAP responses, although with lower efficacy andaffecting a smaller proportion of DRG neurons (22 and 17%,respectively) (Fig. 1C). These experiments suggest that (i) atleast 40% of small, TRPV1+ DRG neurons express functionalNK1, and (ii) NK2 and NK3 also are expressed in some TRPV1+

DRG neurons, but NK1 predominates; moreover, because thepercentage of SP- and Sar-met SP-responsive neurons is thesame, NK2- and NK3-expressing neurons also express NK1. Asimilar proportion (23/51, 45%) of TRPV1+ TG neurons dis-played sensitization of CAP responses by SP (Fig. 1D). RT-PCRexperiments (Fig. S1) supported the findings that both DRG andTG neurons express all three NKR and that NK1 is the pre-dominant mRNA in both types of ganglia.NKR were suggested to couple to a Gq/11 signaling cascade (5,

6); this coupling involves hydrolysis of membrane PIP2 withsubsequent activation of IP3–Ca

2+ and DAG–PKC pathways.Sensitization of CAP responses by SP was suggested to be me-diated by the action of PKC on TRPV1 (7, 8), suggesting thatDAG is released upon SP application. Therefore, we next de-termined whether the IP3–Ca

2+ branch of the Gq/11 signalingcascade is activated also. First, we transfected DRG neurons

with the PIP2/IP3 optical biosensor PLCδ-PH-GFP to evaluatePIP2 hydrolysis by PLC and the release of IP3. This probe islocalized to the plasma membrane at rest and translocates to thecytosol following cleavage of IP3 from PIP2. In 9/34 (26%) smallDRG neurons we observed significant translocation of the probe,indicating PIP2 hydrolysis; i.e., the application of SP resulted ina 1.9 ± 0.3-fold increase in cytosolic fluorescence in these neu-rons (Fig. 2 A and B). The magnitude of PLCδ-PH-GFP trans-location induced by SPwas comparable to that induced by anotherGq/11 GPCR, the bradykinin (BK) B2 receptor, in similar con-ditions (12). Unexpectedly, however, SP- and NKR-specific ago-nists generally were unable to induce elevations of intracellularcalcium ([Ca2+]i) (Fig. 2C). Only 43/336 (13%) of TRPV1+ DRGneurons responded to 1 μM SP with small Ca2+ transients (Fig.1C), a population significantly smaller than that which displayedSP-induced potentiation of CAP responses (P < 0.001, χ2 test).Similarly, poor coupling between NKR and ER Ca2+ stores wasobserved in TG neurons (Fig. 1D). SP-induced Ca2+ transients inDRG and TG were markedly smaller than those induced by BK(P < 0.001) (Fig. 2C–E). These experiments suggest that (i) NKRactivate PLC (shown by translocation of the IP3-sensitive opticalprobe) and PKC (shown by sensitization of CAP responses); (ii)NKR are functionally disconnected from the ER Ca2+ stores,resulting in a lack of cytosolic Ca2+ rise in response to SP in themajority of NKR-expressing neurons; and (iii) the few neurons inwhich SP does induce Ca2+ transients respond with much weakerCa2+ signals than would be expected from a Gq/11 agonist. The

Fig. 1. Functional expression of NKR in sensory neurons. (A and B) Ca2+

imaging in small-diameter TRPV1+ DRG neurons loaded with fura-2 AM. (A)Capsaicin (CAP, 50 nM) was added to the bathing solution as indicated bythe black bars. Before the fourth application of CAP, an NKR agonist wasadded to the bathing solution as indicated by the orange shaded area. “CAPMAX“ indicates a saturating dose of capsaicin (1 µM). Each trace representsan example of the effect of four different NKR agonists; the key is shown inB. All NKR agonists were applied at a concentration of 1 µM. SP is an agonistof all three NKR; Sar-met SP is an agonist of NK1, β-ala NKA is an agonist ofNK2; and senktide is an agonist of NK3. Data are presented as the fluores-cence ratio (340/380 nm, R) normalized to the initial ratio at time = 0 s (R0).(B) Mean data from A normalized to the size of the first CAP peak response(CAP1). Only cells that showed sensitization of the CAP4 response were in-cluded. ***Significant difference between CAP3 and CAP4 peak response(P < 0.001; paired t test). (C) Proportion of DRG neurons responding witha rise in cytosolic Ca2+ in response to NKR agonists (Left) or with a sensiti-zation of TRPV1 (Right). Significant difference in the proportions of Ca2+

responders and TRPV1 sensitizers is shown by **P < 0.01 and ***P < 0.001(χ2 test). Ca2+ responders: SP, n = 43/336; Sar-met SP, n = 20/435; β-ala NKA;NK2, n = 28/228; senktide, n = 7/218. TRPV1 sensitization: SP, n = 32/83; Sar-met SP, n = 46/121; β-ala NKA, n = 51/233; senktide, n = 36/218. (D) As inC but using TG neurons and fluo-4 as the Ca2+ indicator dye. Ca2+ res-ponders: SP, n = 39/200; Sar-met SP, n = 10/124; β-ala NKA, n = 15/124;senktide, n = 18/124. TRPV1 sensitization: SP, n = 23/51; other NKR agonistswere not tested in TG neurons.

Fig. 2. NKR triggering in sensory neurons induces PLC activation but doesnot induce strong Ca2+ release from intracellular stores. (A and B) Trans-location of the PIP2/IP3 probe PLCδ-PH-GFP in the transfected DRG neuron inresponse to SP. (A) Low-resolution epifluorescence image of the transfectedDRG neuron (Upper Left). (Scale bar, 100 μm.) Other images are confocalmicrographs of the same neuron before (basal), during (SP), and after(Wash) application of 1 μM SP; the neuron shown is representative of 9/34cells tested. (Scale bars, 10 μm). (B) Time course of the cytosolic fluorescenceintensity measurements from the cell shown in A. (C) Sample trace showingthe relative size of Ca2+ transient elicited by SP (1 µM) or BK (1 µM) in DRGneurons measured using fura-2 AM. (D) Mean data from experiments as in Cfor DRG neurons. Number of cells is stated inside bars. ***Significant dif-ference between groups (P < 0.001; unpaired t test). (E) As in D, but for TGneurons measured using Fluo-4.

Linley et al. PNAS | Published online May 14, 2012 | E1579

NEU

ROSC

IENCE

PNASPL

US

majority (39/46) of DRG neurons that responded to SP with Ca2+

rises also responded to CAP. A similar trend was observed in TGneurons, although slightlymoreTGthanDRGneurons respondedto SP with Ca2+ transients (39/200, 19.5%, vs. 43/336, 13%; P ≤0.05, χ2 test) (Fig. 1 C and D).

SP Augments M Current in Sensory Neurons via OxidativeModification. The finding that some Gq/11 receptors do not in-duce release of Ca2+ from the endoplasmic reticulum (ER) isnot unprecedented; for example, in sympathetic neurons mus-carinic acetylcholine (M1) receptors robustly activate PLC andhydrolyze PIP2, but this action does not result in ER Ca2+ re-lease, presumably because of poor spatial coupling between thereceptors and the IP3-sensitive stores (15–17). Nevertheless,these M1 receptors exert a robust excitatory effect in sympatheticneurons by inhibiting M current via PIP2 depletion (18, 19).DRG neurons also express M channels [Kv7.2, Kv7.3, and Kv7.5(20, 21)], and M-channel inhibition produces strong excitatoryeffects in DRG (12, 20, 22, 23). Therefore, we tested whether SPinhibits M current in DRG and TG neurons. We performedpatch-clamp recordings from small DRG and TG neurons(whole-cell capacitance of 28 ± 2 pF, n = 32 and 26 ± 1 pF, n =63, respectively) that were responsive to CAP. Surprisingly, ina large proportion of DRG and TG neurons, SP induced markedaugmentation of M current (Fig. 3A). In 25/43 DRG and 14/33TG neurons SP induced a significant increase in the M-currentamplitude of 59 ± 14% and 81 ± 19%, respectively (P < 0.001;ANOVA). In 1/43 DRG and 4/33 TG neurons SP inhibited Mcurrent to 88% and 32 ± 7% of basal levels, respectively, whereasother neurons showed no response (presumably because theylacked functional NKR) (Fig. 3 A–D). Consistent with previousdata (20), most of the small, CAP-responsive DRG and TGneurons expressed M current (e.g., 17/20 DRG neurons in onecohort). The amplitudes of deactivating current (Ideac) at−60 mV in DRG and TG neurons were 85 ± 8 pA (n = 32) and43 ± 5 pA (n = 63), respectively.Current-clamp recordings showed that SP failed to produce an

excitatory effect in all 17 DRG neurons tested; each fired a singleaction potential (AP) before and after the SP application (Fig.3E). In neurons that responded to SP with M-current augmen-tation, SP induced a moderate hyperpolarization of the restingmembrane potential (compared with the population of neuronsin which SP did not affect M-current amplitude) (Fig. 3 E and F).The threshold for AP firing was not altered significantly, al-though there was a trend toward an increase (P = 0.177) in cellsthat responded to SP with an increase in M current (Fig. 3G).The augmentation of M current induced by SP was slow (>10

min) and was not reversible upon washout, features that arereminiscent of augmentation of recombinant Kv7 channels byoxidative modification caused by H2O2 (24). H2O2 oxidizesa triplet of cysteines in the cytosolic S2–S3 linker of Kv7 chan-nels, an effect reversed by the reducing agent DTT (24). Wetherefore tested if DTT (1 mM) would reverse the M-currentaugmentation by SP, and indeed it did so (Fig. 4 A and B). H2O2also augmented M current in DRG neurons, and, again, thisaction was reversed by DTT (Fig. 4 C and D). In both cases DTTdid not inhibit M current completely but rather returned the Mcurrent amplitude to near the basal level. DTT alone did notaffect M current amplitude; moreover, DTT pretreatment ren-dered SP ineffective in producing M-current augmentation (Fig.4B). These data suggest that M-current augmentation producedby SP is mediated by a mechanism similar to that produced byexternal application of H2O2. Oxidation of Kv7.2/7.3 channelsoverexpressed in CHO cells with H2O2 was shown to produce anacceleration of channel activation and a modest slowing of de-activation kinetics (24, 25). In DRG neurons the kinetics of Mcurrent was difficult to analyze because of contamination withother voltage-gated conductances. We fit the activation kinetics

with a double-exponential function; the fast component wascontaminated with conductances other than M current, but theslow component of activation (τslow) was comparable with thekinetics of recombinant Kv7.2/Kv7.3 channels (24, 25). Consis-tent with our hypothesis that SP induces oxidative modificationof M-channel subunits, SP accelerated current activation (uponvoltage step from −60 to −30 mV), resulting in the decrease inτslow from 160 ± 12 ms to 106 ± 10 ms (n = 16; P < 0.001) (Fig.S2A); a similar effect was produced by H2O2 (Fig. S2B). Nosignificant changes in the deactivation kinetics were observedwith either H2O2 or SP (Fig. S2 C and D).Previous reports suggested that SP could induce the pro-

duction of ROS in immune (26) and epithelial cells (27). Todetermine if this induction also could occur in sensory neurons,we used the ROS-sensitive dye CM-H2DCFDA (Fig. 5A). Thirtypercent of dye-loaded DRG neurons (42/140) responded to 1 μMSP with >10% increase in cellular fluorescence; subsequent ap-plication of 100 μMH2O2 caused an increase in fluorescence in all

Fig. 3. SP augments M current and increases the AP firing threshold insmall-diameter sensory neurons. (A–C) Sample perforated patch-clamprecordings from TG neurons. M current is plotted as the magnitude of thedeactivating tail current when stepping from −30 to −60 mV (Ideac); bathapplication of SP (1 µM) and the M channel inhibitor XE991 (3 µM) is in-dicated by the black bars. Insets show current traces recorded at the timepoints indicated (1–4). (D) Proportion of DRG and TG neurons respondingwith an increase, decrease, or no effect in M current in response to SP. Thenumber of cells is shown within the pie charts. (E) Whole-cell current-clamprecording from a DRG neuron. The effect of SP is shown after 15 min ex-posure. Inset shows the current injection protocol. (F and G) Changes inmembrane voltage (Vm) (F) and AP firing threshold (G) after 15-min bathperfusion of SP (1 µM). Each point represents one experiment (n = 17).**Significant difference between groups (P < 0.01; unpaired t test).

E1580 | www.pnas.org/cgi/doi/10.1073/pnas.1201544109 Linley et al.

neurons in the field. To identify the source of the SP-inducedROS generation in DRG neurons, we used mitochondrially tar-geted fluorescent protein, mt-cpYFP, which is particularly sensi-tive to superoxide anion (O2

•−) but is insensitive to Ca2+, ATP/ADP, and NAD(P)H (28). Importantly, the fluorescent signal isreversible, and brief flashes of fluorescence can be visualized inmitochondria during release of O2

•− (28). Fig. 5B, Upper depictsa DRG neuron successfully transfected with mt-cpYFP. Theprobe was distributed heterogeneously within the cell, suggestingcompartmentalization consistent with a mitochondrial localiza-tion. Onemicromolar SP (1 μM) induced a small but significant (P≤ 0.05) increase in total cellular fluorescence in 5/10 small neu-rons (Fig. 5C; initial fluorescence rundown reflects GFP photo-bleaching because of the high sampling rate necessary to resolveindividual flashes; see below). In three of five neurons we wereable to resolve bright, localized flashes (Fig. 5B) that showed ki-netics similar to the mitochondrial O2

•−flashes reported pre-

viously using the same probe (28). Taken together, theseexperiments strongly suggest that SP induces mitochondrial ROSrelease in a subpopulation of small DRG neurons and reinforceour hypothesis that SP-induced augmentation of M current inDRG and TG neurons is mediated by ROS.Mitochondrial superoxide production is dependent on electron

transport chain (ETC) activity (28); inhibition of the ETC complexIII with antimycin A has been shown to cause a burst of mito-chondrial ROS release (29), whereas the protonophore and mi-tochondrial uncoupler N,N,N,N′,N′-tetramethyl-p-phenylenedi-amine dihydrochloride (FCCP) prevents mitochondrial ROSrelease (28). Therefore, we tested if these compounds would in-

terfere with the ability of SP to augment M current in DRG neu-rons.AntimycinA (25 μM) caused a 25± 10% augmentation of theM-current amplitude in five of nine small DRG neurons (Fig. 6 Aand B), whereas 1 μM FCCP prevented M-current augmentationby SP in nine of nine DRG and four of four TG neurons tested; infact in the presence of FCCP, M current was inhibited in seven ofnine DRG and four of four TG neurons by 47 ± 7% and 82 ± 7%,respectively (P < 0.001) (Fig. 6 C and D). This inhibition probablywas caused by the unmasking of PLC-mediated inhibition of Mcurrent (e.g., by PIP2 hydrolysis and, in some cells, by small Ca2+

transients).

Fig. 4. SP increases M current through oxidative modification. (A) Sampletime course of the effect of SP on M current in a DRG neuron. Note thataugmentation was not washable but was reversed by DTT (1 mM). Insetshows current traces recorded at the time points indicated (1–5). (B) (Left)Mean data from experiments as in A expressed as percent change in Mcurrent (XE991-sensitive fraction of Ideac) from basal state. (Right) In a sepa-rate series of experiments DTT alone was shown to have no effect on Mcurrent, and SP had no effect in the presence of DTT. The number ofexperiments is shown within the bars. (C) Sample time course of the effect ofH2O2 (100 µM) on M current in DRG neurons; other labeling is as in A. (D)Mean data from experiments as in C. ***Significant difference from basal(P < 0.001; one-way ANOVA).

Fig. 5. SP induces intracellular release of ROS in a subset of DRG neurons.(A) Sample recording from DRG neurons loaded with the ROS-sensitive dyeCM-H2DCFDA. Each line represents an individual DRG neuron. Data arerepresented as fluorescence at 488 nm (F) normalized to initial fluorescenceat time = 0 (F0) and corrected for the dye auto-oxidation (see Materials andMethods). (B) (Top) Bright-field (Left), fluorescent confocal 3D recon-struction (Z-stack; Center), and a zoomed-in single-plane fluorescent con-focal micrograph (Right) of cultured DRG neurons transfected with themitochondrial O2

•− sensor mt-cpYFP. (Middle) Sample time course of theeffect of SP (1 μM) on the mt-cpYFP fluorescence within the areas of interest(AOI) depicted by colored boxes in the image on the right. (Bottom) In-dividual frames of the blue AOI during the flash of the O2

•− release (shown isa 5-s sequence recorded at three frames/s). The colors of the traces in thetime course correspond to the colors of the AOIs. (C) Mean time course ofthe normalized (F/F0) total cellular fluorescence of five responsive DRGneurons during the application of 1 μM SP. Error bars are indicated in darkgray. *Significant difference in mean F/F0 at t = 150 s (just before SP appli-cation) and t = 330 s (peak effect) (P < 0.05; paired t test).

Linley et al. PNAS | Published online May 14, 2012 | E1581

NEU

ROSC

IENCE

PNASPL

US

To probe if SP has a direct effect on the mitochondrial ETC inDRG neurons, we measured the oxygen consumption and oxi-dative phosphorylation rates in suspensions of freshly dissociatedDRG using respirometry. Application of SP suppressed the mi-tochondrial oxygen consumption rate significantly, by 13 ± 2.1%(n = 5) (Fig. 6E). This finding was consistent with the previousexperiment in which the complex III inhibitor, antimycin A,mimicked the M current-enhancing effect of SP. We thereforerepeated the respirometry experiment using an extracellular so-

lution containing the complex I inhibitor, rotenone (1 μM), and10 mM succinate in place of glucose. Such conditions excludecomplex I from ETC flux but permit respiration, because elec-trons from succinate oxidation enter the ETC directly via complexII (succinate dehydrogenase). In such conditions, SP still reducedmitochondrial respiration significantly (Fig. 6F). Because ETCcomplexes I and III are the main sources of mitochondrial ROSgeneration (30), it is likely that SP signaling affects complex III.

NKR Couple to Gi/o in Sensory Neurons but Can Couple to Gq/11 WhenOverexpressed. Clearly, the SP-induced signaling cascade in smallsensory neurons differs significantly from the classical Gq/11-mediated signaling cascade; so the question arises: Which Gprotein α-subunit mediates the effect? Pertussis toxin (PTX)-sensitive Gi/o-coupled receptors, such as somatostatin receptors,were reported to augment M current in hippocampal neurons(31, 32) via an unidentified mechanism [although the in-volvement of arachidonic acid metabolites has been suggested(33, 34)]. Thus, we tested if the effect of SP in sensory neuronsalso is mediated by Gi/o. After treatment with 300 ng/mL PTXovernight, SP failed to augment M current in five of five TG and9/10 DRG neurons (Fig. S3A). The proportion of small neuronsresponding to SP with Ca2+ transients also was reduced signifi-cantly after PTX treatment; only 4/148 (3%) DRG and 5/120(4%) TG neurons responded to SP with small Ca2+ elevations,and these elevations were significantly smaller than in controlDRG and TG cultures (P ≤ 0.001 with χ2 test) (Fig S3 B and C).Interestingly, SP-mediated M-current augmentation in DRGneurons was blocked by the PLC inhibitor edelfosine (10 μM):Nine of nine edelfosine-treated neurons showed no augmenta-tion (Fig S3 D). These experiments suggest that in DRG and TGneurons the effect of SP is mediated by Gi/o protein and a PLCisoform that is not coupled to Gq/11 exclusively but also can beactivated by Gi/o [e.g., PLCβ2 or β3 (35)].These results were unexpected, given that in other tissues and

expression systems NKR were shown to couple to Gq/11 (5, 6).Thus, we expressed NK1–3 receptors in CHO cells and testedmajor components of the Gq/11 pathway by (i) confocal imagingof PIP2 hydrolysis and IP3 release using PLCδ-PH-GFP; (ii)confocal imaging of DAG release with the PKCγ-C1-GFP probe;(iii) fura-2 Ca2+ imaging to monitor Ca2+ release; and (iv) patch-clamp recording of recombinant M-channel inhibition by NKR.NK1 (stimulated with 1 μM SP) induced robust translocation ofPLCδ-PH-GFP into the cytosol (Fig. 7A) and PKCγ-C1-GFP tothe plasma membrane (Fig. 7B). Likewise, all NKR inducedrobust Ca2+ transients (Fig. 7C), which were not affected by PTXpretreatment. We also measured the effect of recombinant NK1on recombinant M channels (Kv7.2/7.3 heteromers) and foundthat 1 μM SP acutely inhibited M current by 50 ± 7% (P ≤ 0.001;n = 10) (Fig. 7D). Taken together, these data indicate that, whenoverexpressed in CHO cells, NKR can couple to a classical Gq/11signaling cascade. Although SP signaling in sensory neurons isdifferent, it does induce PIP2 hydrolysis (as evidenced by thePLCδ-PH-GFP translocation and edelfosine blockade of M-current augmentation). This hydrolysis is likely to be spatiallydiscrete from M channels (which are inhibited by PIP2 depletion)and from IP3-sensitive ER stores, because NKR rarely inducedM channel inhibition or ER Ca2+ release. To test if this en-dogenous NKR coupling rule can be overcome, we overex-pressed NK1 in DRG neurons (Fig. 7 E and F) and tested if suchoverexpression would affect NK1 signaling. In DRG neuronsoverexpressing NK1, as in NK1-transfected CHO cells, SP in-duced robust Ca2+ transients in eight of eight neurons (R/R0 =2.07 ± 0.34, n = 8) (Fig. 7E) and also robustly and reversiblyinhibited M current (Fig. 7F). Thus, we conclude that, whenoverexpressed in CHO cells or DRG neurons, NK1 receptorscouple to Gq/11, but endogenous NK1 receptors in sensory neu-rons preferentially signal through Gi/o.

Fig. 6. Augmentation of M current is mediated by mitochondrial ROS. (A)Sample time course of the effect of antimycin A (Ant A; 25 μM) on M currentmeasured using the perforated patch-clamp technique. As with SP and H2O2,augmentation was not washable but was reversed by DTT (1 mM). (B) Meandata from A expressed as percent change in M current (XE991-sensitivefraction of Ideac) from basal state. *P < 0.05 compared with basal state (one-way ANOVA). (C) Sample time course of the effect of FCCP (1 μM) on Mcurrent. (D) Mean data from C expressed as percent change in M current[XE991-sensitive (3 µM) fraction of Ideac] from basal state. The number ofcells is presented inside bars. ***P < 0.001 compared with basal state (one-way ANOVA). (E) Rate of oxygen consumption by intact, isolated rat DRGcells normalized to initial O2 flux. Columns from left to right are basal res-piration (Routine); O2 flux following the addition of 1 μL vehicle (distilledwater; Control), 1 μM SP, and 2 μg/mL oligomycin. All measurements weremade when O2 flux had stabilized and are corrected for nonmitochondrialrespiration. Bars show mean data; n = 5; *P < 0.05 (one-way ANOVA fol-lowed by Bonferroni’s post hoc test) (F) As in E, but in glucose-free mediumcontaining 10 mM sodium succinate and the mitochondrial complex I in-hibitor rotenone (1 μM). Columns from left to right are initial O2 flux; fluxfollowing the addition of 1μM SP; and flux following the addition of 2 μg/mLoligomycin. All data are corrected for nonmitochondrial respiration and arenormalized to initial O2 flux. Bars show mean data; n = 10; *P < 0.05 (one-way ANOVA followed by Bonferroni’s post hoc test).

E1582 | www.pnas.org/cgi/doi/10.1073/pnas.1201544109 Linley et al.

Effects of SP on CGRP Release and Nociception. Excitation of pe-ripheral afferents induces nociceptive signaling to the CNS andalso a local release of neuropeptides. Therefore, we evaluated

the effect of SP on the behavioral manifestations of nociceptionby using pain tests and on CGRP release from intact DRGneurons in culture by ELISA. As expected, both CAP (100 nM)and BK (1 μM) induced robust CGRP release, consistent withthe excitatory effects of both agents. In contrast, 1 μM SP notonly failed to stimulate CGRP release but significantly inhibitedbasal release (Fig. S4), consistent with the moderate hyperpo-larization induced by SP in current-clamp experiments.In a recent study (12) we analyzed excitatory and proalgesic

effects of BK in sensory neurons and concluded that the acute“spontaneous” excitation (and the acute phase of BK-inducedpain) is mediated mostly by the cytosolic Ca2+ transients, whichinhibit M current and activate Ca2+-activated Cl− channels(CaCC) encoded by Ano1/Tmem16a [in peripheral sensoryneurons the [Cl−]i is high, so activation of a Cl− channel pro-duces depolarization (12, 36)]. At the same time, it is postulatedthat thermal hyperalgesia induced by BK is mediated by thesensitization of sensory TRP channels [e.g., TRPV1 (37) andTRPA1 (38)]. Present data suggest that peripheral NKR arelargely uncoupled from Ca2+ transients. Consistently, we did notobserve activation of CaCC by SP (Fig. S5). Nevertheless, NKRdo enhance Ca2+ responses to CAP in DRG, suggesting TRPV1sensitization (Fig. 1). We therefore tested the peripheral effectsof SP on the excitability of nociceptive neurons in vivo andcompared them with the effects of BK. First we confirmed thepreviously characterized (7, 39) hyperalgesic effect induced byplantar injection of SP. In accordance with previous studies, thelatency of hind paw withdrawal at the presentation of a thermalstimulus was decreased significantly after plantar injection of10 μM SP in saline (50 μL; 0.5 nmol per site) (Fig. 8A); thehyperalgesic effect of SP was comparable to that of 10 μM BK(Fig. 8A). Equal concentrations of SP and BK were used, be-cause the potency of BK at B2 receptors and SP at NK1 receptorsare comparable in neurons and both have an EC50 in the lownanomolar range (12, 40). Remarkably, in the nocifensive be-havior test (total time of paw biting, licking, and flinching) BK,but not SP, produced prominent spontaneous pain (Fig. 8B). Thelack of SP-induced spontaneous pain is consistent with previousobservations in humans (41) and with the current study showingpoor coupling of NKR to the IP3–Ca

2+ route of the PLC pathwayalong with a lack of M-current inhibition (and CaCC activation)in DRG neurons.

Fig. 7. Overexpressed NKR couple to Gq/11. (A and B) SP-induced trans-location of the PIP2/IP3 probe PLCδ-PH-GFP (A) or the Ca2+-insensitive DAGprobe PKCγ-C1-GFP (B) in CHO cells transfected with NK1. Upper panels showconfocal micrographs of the transfected cell before (basal) and during theapplication of 1 µM SP. Lower panels show the corresponding surface in-tensity plots (warm colors indicate greater pixel intensity). (C) Calcium im-aging from CHO cells transfected with NK1, NK2, or NK3. Data representmean ± SEM (n = 22–27 as indicated) and are normalized to the fluorescenceratio at time 0 (R/R0). SP (1 µM) was added to the bathing solution as in-dicated by the black bar. (D) Perforated patch-clamp recording from CHOcells overexpressing Kv7.2, Kv7.3, and NK1. (Left) Time course of inhibition ofM current by SP (1 µM) and XE991 (XE, 3 µM) measured as the whole-cellcurrent at 0 mV. (Right) Individual current traces at the time points 1–4;voltage protocol is shown above the traces. (E) Ca2+ imaging of a DRGneuron transfected with NK1. Drugs [1 µM SP; 30 mM KCl (30 K); and 1 µMcapsaicin (CAP)] were added to the bathing solution as indicated by theblack bars. (F) Perforated patch-clamp recording from a DRG neuron trans-fected with NK1. Endogenous M current is plotted as the Ideac upon steppingthe membrane voltage from −30 to −60 mV, 1 µM SP; 3 µM XE991. Insetshows current traces at the time points indicated (1–4).

Fig. 8. Intraplantar injection of SP causes minimal spontaneous pain butrobust thermal hyperalgesia. (A) Thermal hyperalgesia was measured as thetime (s) until hind paw withdrawal following application of a heat source(Hargreaves’ apparatus). Measurements were taken at baseline (before in-jection) and 10, 20, and 30 min after 50 μL intraplantar injection of 10 µM SP,10 µM BK, or vehicle. Mean hind paw withdrawal latencies are shown ± SEM;n = 4. A significant difference from vehicle control is seen at each time point(*P < 0.05; **P < 0.01; two-way ANOVA.) (B) Spontaneous pain was mea-sured as the time spent demonstrating nocifensive behavior (licking, biting,and lifting the hind paw) following a 50-μL intraplantar injection of 10 µMSP (n = 8), 10 µM BK (n = 8), or vehicle (n = 10). ***Significant differencefrom control (P < 0.001; one-way ANOVA).

Linley et al. PNAS | Published online May 14, 2012 | E1583

NEU

ROSC

IENCE

PNASPL

US

DiscussionHere we describe an atypical signaling cascade mediated by SP insensory neurons. We show that (i) nociceptive DRG neuronsexpress functional NKR1–3, but NK1 predominates, and neu-rons expressing NK2 and NK3 most likely also express NK1; allthree receptor types are capable of potentiating CAP responses.(ii) When overexpressed in CHO cells or sensory neurons, NKRcouple to the Gq/11 cascade, but native NKR in DRG neuronssignal predominantly through a PTX-sensitive protein (pre-sumably Gi/o). This signaling involves activation of PLC (po-tentially PLCβ2 or β3, which can be activated by Gβγ binding) andhydrolysis of PIP2, but the magnitude or location of PIP2 de-pletion and IP3 release is not sufficient to produce inhibition ofKv7 channels or release of Ca2+ from the ER. (iii) In sensoryneurons NKR activation results in mitochondrial ROS pro-duction and release because of the modulation of mitochondrialETC (most likely at complex III). (iv) ROS act as second mes-sengers to augment M current via oxidative modification of Kv7channels; this M-channel augmentation has a moderate anti-excitatory outcome (as evident from hyperpolarization and in-hibition of basal CGRP release by SP). (v) SP signaling in DRGand TG neurons generally is similar. However, there is signifi-cantly better coupling between NKR and IP3-sensitive Ca2+

stores in TG neurons. In sum, the signaling cascade used byendogenous NKR in sensory neurons is substantially differentfrom the classical Gq/11 signaling cascade, although it does sharesome features of the classical cascade (e.g., involvement of PLC,TRPV1 sensitization); exogenously expressed NKR do couple tothe typical Gq/11–PLC cascade.Although it was assumed that NKR in DRG neurons couple to

the Gq/11 cascade (5, 6), cross-talk with other G proteins [e.g.,G12/13 (42), Gi/o, and Gs (43)] in expression systems has beensuggested also. We suggest that NKR can couple to multipleclasses of G proteins, thereby providing a level of signaling di-versity dependent on local G protein association. A recent studydemonstrated a functional coassembly between Gq-coupled (se-rotonin 5HT2A receptor 2AR) and Gi-coupled (metabotropicglutamate receptor mGluR2) receptors in cortical neurons,which sets a dynamic Gq–Gi balance of the resulting heteromericreceptor (44). It is conceivable that NKR in DRG may coas-semble with a Gi receptor, producing hybrid Gq–Gi signaling.Such potential coassembly may provide an explanation for Gq/11coupling of overexpressed NK1 in DRG, because there wouldnot be enough endogenous Gi partners for coassembly.There is a substantial difference in NKR signaling toward Kv7

channels in DRG neurons and the well-established action ofmuscarinic and BK receptors, which inhibit M channels either byPIP2 depletion or by Ca2+/calmodulin (or by a combination ofboth), producing acute excitation. Here we show that althoughNKR in DRG neurons do indeed activate PLC, they are in-capable of producing either of the signals inhibiting M channels,instead resulting in mitochondrial ROS release that producesoxidative augmentation of Kv7 channel activity. The reasons forthis remarkable difference in signaling subroutines of PLC-cou-pled GPCRs are yet to be established but most likely includerestricted spatial clustering (microdomains) of receptors andtheir effector molecules and targets, as has been suggestedelsewhere (17). Interestingly SP inhibited M channels in frogsympathetic neurons (45, 46), indicating further species and/orneuronal type-dependent heterogeneity in NKR signaling.NK1 receptor activation has been associated previously with

increased ROS formation, particularly in the respiratory (27) andimmune systems (26). However, neither the underlying mecha-nism nor the source of increased ROS has been elucidated fully.We show that NKR activation in sensory neurons inhibits mi-tochondrial ETC, an action that increases ROS formation andrelease, presumably from complex III. We speculate that SP

exerts an effect analogous to that of antimycin A, but the exactmechanism linking NKR with ETC modulation in sensory neu-rons remains to be elucidated.Previous studies of peripheral NKR signaling resulted in cer-

tain contention, but we believe that our study can resolve at leastsome of the controversies. (i) One study (47) found no immu-nohistochemical evidence for the expression of NK1 in periph-eral sensory neurons; however, we show that functional effects ofSP on peripheral neurons can be demonstrated both in vitro andin vivo. (ii) Although one study (9) reported SP-induced Ca2+

rises in cultured DRG neurons, other reports did not (7, 8). Ourdata suggest that Ca2+ rises induced by SP are small and are notdisplayed by the majority of SP-responsive neurons. (iii) The SP-induced, PKCε-mediated sensitization of TRPV1 was attributedto both NK1 (7) and NK2 (8).We show that DRG and TG ex-press all three NKR isoforms, which produce similar effects,although NK1 predominates. (iv) Both hyperpolarization (48)and excitatory effects (49–51) of SP in DRG have been reported.In our hands SP produced a moderate hyperpolarization becauseof an increase in M current, whereas the depolarizing effect ofSP (49, 50) could have been caused by differences in recordingconditions (i.e., a high concentration of cAMP or cesium fluoridein pipette solutions used in these studies). A study performed onthe acutely dissociated DRG neurons reported an inhibition ofoutward K+ currents by SP in most small neurons, whereas ac-tivation was observed in some others (51). However, during theproteolytic dissociation of DRG neurons, M current is inhibitedby the protease-activated receptor PAR-2 (22) and also by highcytosolic Ca2+ levels (52) that are unavoidable in acutely disso-ciated neurons; therefore, the effect of SP on M current is mostlikely underestimated in acutely dissociated neurons. RecentlySP has been reported to enhance an XE991-sensitive current ina proportion of muscle afferent neurons through an Src kinase-dependent, GPCR-independent mechanism (53). AlthoughM-current enhancement by SP is consistent with the presentstudy, the signaling cascade is not. Thus, we have shown clearlythat SP-mediated enhancement of M current in DRG is PLCdependent and is inhibited by PTX, both supporting a role forGPCRs. Src also is widely reported to have an inhibitory effecton M channels, an effect mediated by phosphorylation of twotyrosines [e.g., Y67 and Y349 in Kv7.3 (54–56)].Finally, the evidence presented above allows us to suggest some

further hypotheses. (i) There is a substantial difference in themessage conveyed to peripheral nerve endings by BK (which isreleased by the inflamed or damaged tissue and signals inflam-mation) and SP (which is released by the sensory nerves them-selves). It is conceivable that in nociceptive neurons NKR, incontrast to BKB2 receptors, lack an acute excitatory component ofthe signaling cascade, in this case resulting in a self-perpetuatingpositive feedback loop. In contrast, the hyperpolarizing augmen-tation of M current induced by SP may provide a negative feed-back loop limiting further release of SP, as suggested earlier (11).However, NKR still induce thermal hyperalgesia by sensitizingTRPV1; this hyperalgesia may help maintain the sufferer’sawareness of the ongoing inflammation. (ii) A significantly largerproportion of TG neurons displays coupling between NKR andIP3-sensitive Ca2+ stores compared with DRG neurons. Accord-ingly, SP inhibitedM current in a larger proportion of TGneurons.This inhibition [possibly in combination with other Ca2+-de-pendent effects, e.g., activation of TMEM16A/CaCC (12)] wouldbe expected to excite trigeminal fibers and induce nociceptivesignaling. Thus, we hypothesize that the NKR action would bemore painful in the orofacial area than in the rest of the body. (iii)Our data clearly demonstrate that spontaneous pain and hyper-algesia can be mediated by distinct molecular mechanisms. Thesefindings are of particular importance for designing future strate-gies for analgesic drug discovery, because spontaneous pain andhyperalgesia are conceptually poorly distinguished at present.

E1584 | www.pnas.org/cgi/doi/10.1073/pnas.1201544109 Linley et al.

Furthermore, although chronic pain, which causes most sufferingto humans, is mostly spontaneous in nature, the major animalmodels for testing analgesic efficacy of drugs are based onhyperalgesia tests (57). Accordingly, despite the rapid progress inour understanding of the mechanisms of hyperalgesia, there hasbeen limited achievement in treatment of pain in humans. Furtherstudies highlighting the mechanistic difference in spontaneouspain and hyperalgesia are needed to provide a new framework foranalgesic drug design.

Materials and MethodsCell Cultures, Transfections, cDNA Constructs and Chemicals. DRG and TGneurons were extracted from 7-d-old rats, enzymatically dissociated, andcultured for at least 24 but not more that 96 h before experiments (as de-scribed in ref. 12); no NGF was added to the culture medium. CHO cells werecultured in DMEM/F12 medium. Plasmids encoding human Kv7.2 and humanKv7.3 (GenBank accession numbers, AF110020 and AAC96101, respectively)were given to us by David McKinnon (State University of New York at StonyBrook, Stony Brook, NY) and Thomas Jentsch (Zentrum fur MolekulareNeurobiologie, Hamburg, Germany) and were subcloned into pcDNA3.1(Invitrogen). The PLCδ-PH-GFP and PKCγ-C1-GFP constructs were kind giftsfrom Tobias Meyer (Stanford University, Palo Alto, CA). Human B2R, NK1,NK2, and NK3 (GenBank accession numbers AY275465, AY462098,AY322545, and AY462099, respectively) were purchased from the MissouriScience and Technology cDNA Resource Center. The mitochondrial O2

•−

sensor, mt-cpYFP, was a kind gift from Heping Cheng (Peking University,Beijing, China). CHO cells were transfected using FuGENE HD transfectionreagent (Roche). DRG neurons were transfected using Amaxa NucleofectorDevice (Lonza) in combination with the rat DRG transfection kit and O-03voltage protocol. Edelfosine, SP, and PTX were from Calbiochem (VWR In-ternational); XE991 was from Tocris; and fura-2 AM was from Invitrogen. Allother chemicals were from Sigma.

Electrophysiology. Perforated patch-clamp recordings were performed asdescribed previously (12, 22). The standard bath solution contained (in mM)160 NaCl, 2.5 KCl, 5 CaCl2, 1 MgCl2, 10 Hepes, and 8 glucose, pH 7.4. Thestandard pipette solution contained (in mM) 90 K-acetate, 10 KCl, 10NaCl, 1CaCl2, 3 MgCl2, 3 EGTA, 40 Hepes, and amphotericin B (250 µg/mL), pH 7.4.M-like current amplitude was measured from the deactivation current eli-cited by 600–800-ms square voltage pulses to −60 mV from a holding po-tential of −30 mV, calculated as the difference between the average of a 10-ms segment taken 20–30 ms into the hyperpolarizing step and the averageduring the last 50 ms of that step and termed “Ideac.” The fraction of Ideacattributable to M current was determined from the XE991-sensitive fractionand was termed “IM.” In current-clamp recordings, APs were generated byinjection of 400-pA current for 1 s from a holding current of 0 pA. The APfiring threshold was determined from a train of current steps delivered in10-pA increments.

Fluorescence Imaging. Ca2+ imaging was performed as described previously(52). For ROS imaging with the CM-H2DCFDA dye, DRG neurons were loadedwith the dye (5 µM) for 30 min and imaged using excitation with 488-nm

light for 10 ms every 10 s. Excitation of the dye resulted in some time-de-pendent autofluorescence, which was monitored for the first 300 s andsubtracted offline using a linear function. To measure mitochondrial O2

•−

release, DRG neurons were transiently transfected with mt-cpYFP and im-aged 48 h later using LiveScan Swept Field Confocal System (Nikon) with an488-nm argon laser at three to five frames/s. Translocation of IP3/PIP2 andDAG probes was recorded using the confocal system as described previously(22); images were analyzed with NIS Elements 3.2 software (Nikon).

Respirometry. Freshly dissociated rat DRG neurons were resuspended in thestandard bath solution (see above) and transferred to the chambers of a high-resolution respirometer (Oxygraph-2K; Oroboros Instruments). The chamberswere sealed, and the oxygen consumption rate was measured at 37 °C bypolargraphic oxygen sensors. The sensors were calibrated each day imme-diately before the experiment. Oxygen flux (picomoles per second per 106

cells) was monitored continuously and allowed to reach a stable value (i.e.,a constant rate of O2 consumption) before the addition of either 1 μL SP(giving a final concentration of 1 μM) or 1 μL vehicle to the chambers. Thenthe O2 flux was monitored until it was stable (typically 10–15 min). Oligo-mycin (2 μg/mL) was then added to inhibit complex V, and, when the fluxwas stable, the uncoupler FCCP was added, in 0.5-μM steps, allowing mea-surement of the maximum electron flux through the electronic transportsystem. Finally, nonmitochondrial respiration was measured after inhibitionof complex III by 2.5 μM antimycin A. In one series of experiments glucose inthe chamber solution was replaced by 10 mM succinate, and 1 μM rotenonewas added to inhibit complex I of the electronic transport system.

Behavioral Assays. Wistar rats (body weight, 150 g) were grouped randomlyand allowed to acclimatize for at least 20 min in a transparent observationchamber before the experiment. The right hind paw of the animal receivedan intraplantar 50-μL injection of 10 µM SP, 10 µM BK, or vehicle. Sponta-neous pain behavior (time spent licking, biting, lifting, and flinching) wasrecorded using a video camera for 30 min. The videos were analyzed by anobserver unaware of treatment allocations. For measurement of thermalhyperalgesia, the change in the latency of withdrawal in response to noxiousheat was recorded using the Hargreaves’ plantar method (Ugo Basile).Baseline thermal latency was measured before a 50-μL injection of 10 µM SP,10 µM BK, or vehicle, and measurements repeated at 10, 20, and 30 minafter the injection. Both ipsilateral (injected) and contralateral (control) hindpaw withdrawal latencies were recorded; neither BK nor SP had an effect onthe contralateral paw. The power settings were kept constant throughoutthe series.

Statistics. All data are given as mean ± SEM. Differences between groupswere assessed by Student’s t test or ANOVA with Dunnett’s (one-wayANOVA) or Bonferroni (two-way ANOVAs) post tests. The differences wereconsidered significant at P ≤ 0.05. The χ2 test was used to determine dif-ferences in the number of cells responding to agonists.

ACKNOWLEDGMENTS. This work was supported by Grants 080593/Z06/Z and080833/Z/06/Z from the Wellcome Trust and by Grants G0700966 andG1002183 from the Medical Research Council (to N.G.).

1. McMahon SB, Bennet D LH, Bevan S (2006) Inflammatory mediators and modulators.Wall and Melzack’s Textbook of Pain, eds McMahon SB, Koltzenburg M. (ElsevierChurchill Livingstone, Edinburgh), 5th Ed, pp 49–72.

2. De Biasi S, Rustioni A (1988) Glutamate and substance P coexist in primary afferentterminals in the superficial laminae of spinal cord. Proc Natl Acad Sci USA 85:7820–7824.

3. Malcangio M, Bowery NG (1999) Peptide autoreceptors: Does an autoreceptor forsubstance P exist? Trends Pharmacol Sci 20:405–407.

4. Richardson JD, Vasko MR (2002) Cellular mechanisms of neurogenic inflammation.J Pharmacol Exp Ther 302:839–845.

5. Meza U, Thapliyal A, Bannister RA, Adams BA (2007) Neurokinin 1 receptors triggeroverlapping stimulation and inhibition of CaV2.3 (R-type) calcium channels. MolPharmacol 71:284–293.

6. Macdonald SG, Dumas JJ, Boyd ND (1996) Chemical cross-linking of the substanceP (NK-1) receptor to the alpha subunits of the G proteins Gq and G11. Biochemistry 35:2909–2916.

7. Zhang H, et al. (2007) Neurokinin-1 receptor enhances TRPV1 activity in primarysensory neurons via PKCepsilon: A novel pathway for heat hyperalgesia. J Neurosci 27:12067–12077.

8. Sculptoreanu A, Aura Kullmann F, de Groat WC (2008) Neurokinin 2 receptor-mediated activation of protein kinase C modulates capsaicin responses in DRGneurons from adult rats. Eur J Neurosci 27:3171–3181.

9. Brechenmacher C, Larmet Y, Feltz P, Rodeau JL (1998) Cultured rat sensory neuronesexpress functional tachykinin receptor subtypes 1, 2 and 3. Neurosci Lett 241:159–162.

10. Tang HB, Li YS, Arihiro K, Nakata Y (2007) Activation of the neurokinin-1 receptor bysubstance P triggers the release of substance P from cultured adult rat dorsal rootganglion neurons. Mol Pain 3:42.

11. Lever IJ, et al. (2003) Basal and activity-induced release of substance P from primaryafferent fibres in NK1 receptor knockout mice: Evidence for negative feedback.Neuropharmacology 45:1101–1110.

12. Liu B, et al. (2010) The acute nociceptive signals induced by bradykinin in ratsensory neurons are mediated by inhibition of M-type K+ channels and activation ofCa2+-activated Cl- channels. J Clin Invest 120:1240–1252.

13. Gavva NR, et al. (2004) Molecular determinants of vanilloid sensitivity in TRPV1. J BiolChem 279:20283–20295.

14. Bhave G, et al. (2002) cAMP-dependent protein kinase regulates desensitization ofthe capsaicin receptor (VR1) by direct phosphorylation. Neuron 35:721–731.

15. Gamper N, Reznikov V, Yamada Y, Yang J, Shapiro MS (2004) Phosphatidylinositol[correction] 4,5-bisphosphate signals underlie receptor-specific Gq/11-mediatedmodulation of N-type Ca2+ channels. J Neurosci 24:10980–10992.

16. Zaika O, Zhang J, Shapiro MS (2011) Combined phosphoinositide and Ca2+ signalsmediating receptor specificity toward neuronal Ca2+ channels. J Biol Chem 286:830–841.

Linley et al. PNAS | Published online May 14, 2012 | E1585

NEU

ROSC

IENCE

PNASPL

US

17. Delmas P, Wanaverbecq N, Abogadie FC, Mistry M, Brown DA (2002) Signalingmicrodomains define the specificity of receptor-mediated InsP(3) pathways inneurons. Neuron 34:209–220.

18. Suh BC, Hille B (2002) Recovery from muscarinic modulation of M current channelsrequires phosphatidylinositol 4,5-bisphosphate synthesis. Neuron 35:507–520.

19. Zhang H, et al. (2003) PIP(2) activates KCNQ channels, and its hydrolysis underliesreceptor-mediated inhibition of M currents. Neuron 37:963–975.

20. Passmore GM, et al. (2003) KCNQ/M currents in sensory neurons: Significance for paintherapy. J Neurosci 23:7227–7236.

21. Rose K, et al. (2011) Transcriptional repression of the M channel subunit Kv7.2 inchronic nerve injury. Pain 152:742–754.

22. Linley JE, et al. (2008) Inhibition of M current in sensory neurons by exogenousproteases: A signaling pathway mediating inflammatory nociception. J Neurosci 28:11240–11249.

23. Crozier RA, Ajit SK, Kaftan EJ, Pausch MH (2007) MrgD activation inhibits KCNQ/M-currents and contributes to enhanced neuronal excitability. J Neurosci 27:4492–4496.

24. Gamper N, et al. (2006) Oxidative modification of M-type K(+) channels asa mechanism of cytoprotective neuronal silencing. EMBO J 25:4996–5004.

25. Linley JE, Pettinger L, Huang D, Gamper N (2012) M channel enhancers andphysiological M channel block. J Physiol 590:793–807.

26. Tanabe T, et al. (1996) Intracellular signaling pathway of substance P-inducedsuperoxide production in human neutrophils. Eur J Pharmacol 299:187–195.

27. Springer J, Pleimes D, Scholz FR, Fischer A (2005) Substance P mediates AP-1 inductionin A549 cells via reactive oxygen species. Regul Pept 124:99–103.

28. Wang W, et al. (2008) Superoxide flashes in single mitochondria. Cell 134:279–290.29. Zorov DB, Filburn CR, Klotz LO, Zweier JL, Sollott SJ (2000) Reactive oxygen species

(ROS)-induced ROS release: A new phenomenon accompanying induction of themitochondrial permeability transition in cardiac myocytes. J Exp Med 192:1001–1014.

30. Turrens JF (2003) Mitochondrial formation of reactive oxygen species. J Physiol 552:335–344.

31. Moore SD, Madamba SG, Joëls M, Siggins GR (1988) Somatostatin augments theM-current in hippocampal neurons. Science 239:278–280.

32. Qiu C, et al. (2008) Somatostatin receptor subtype 4 couples to the M-current toregulate seizures. J Neurosci 28:3567–3576.

33. Schweitzer P, Madamba S, Champagnat J, Siggins GR (1993) Somatostatin inhibitionof hippocampal CA1 pyramidal neurons: Mediation by arachidonic acid and itsmetabolites. J Neurosci 13:2033–2049.

34. Schweitzer P, Madamba S, Siggins GR (1990) Arachidonic acid metabolites asmediators of somatostatin-induced increase of neuronal M-current. Nature 346:464–467.

35. Rebres RA, et al. (2011) Synergistic Ca2+ responses by Gαi- and Gαq-coupled G-protein-coupled receptors require a single PLCβ isoform that is sensitive to both Gβγ and Gαq.J Biol Chem 286:942–951.

36. Funk K, et al. (2008) Modulation of chloride homeostasis by inflammatory mediatorsin dorsal root ganglion neurons. Mol Pain 4:32.

37. Chuang HH, et al. (2001) Bradykinin and nerve growth factor release the capsaicinreceptor from PtdIns(4,5)P2-mediated inhibition. Nature 411:957–962.

38. Wang S, et al. (2008) Phospholipase C and protein kinase A mediate bradykininsensitization of TRPA1: A molecular mechanism of inflammatory pain. Brain 131:1241–1251.

39. Nakamura-Craig M, Smith TW (1989) Substance P and peripheral inflammatoryhyperalgesia. Pain 38:91–98.

40. Laufer R, et al. (1985) Neurokinin B is a preferred agonist for a neuronal substance Preceptor and its action is antagonized by enkephalin. Proc Natl Acad Sci USA 82:7444–7448.

41. Weidner C, et al. (2000) Acute effects of substance P and calcitonin gene-relatedpeptide in human skin—a microdialysis study. J Invest Dermatol 115:1015–1020.

42. Meshki J, et al. (2009) Neurokinin 1 receptor mediates membrane blebbing in HEK293cells through a Rho/Rho-associated coiled-coil kinase-dependent mechanism. J Biol

Chem 284:9280–9289.43. Roush ED, Kwatra MM (1998) Human substance P receptor expressed in Chinese

hamster ovary cells directly activates G(αq/11), G(αs), G(αo). FEBS Lett 428:291–294.44. Fribourg M, et al. (2011) Decoding the signaling of a GPCR heteromeric complex

reveals a unifying mechanism of action of antipsychotic drugs. Cell 147:1011–1023.45. Adams PR, Brown DA, Jones SW (1983) Substance P inhibits the M-current in bullfrog

sympathetic neurones. Br J Pharmacol 79:330–333.46. Bosma MM, Hille B (1989) Protein kinase C is not necessary for peptide-induced

suppression of M current or for desensitization of the peptide receptors. Proc NatlAcad Sci USA 86:2943–2947.

47. Brown JL, et al. (1995) Morphological characterization of substance P receptor-immunoreactive neurons in the rat spinal cord and trigeminal nucleus caudalis.J Comp Neurol 356:327–344.

48. Sculptoreanu A, de Groat WC (2007) Neurokinins enhance excitability in capsaicin-responsive DRG neurons. Exp Neurol 205:92–100.

49. Cang CL, Zhang H, Zhang YQ, Zhao ZQ (2009) PKCepsilon-dependent potentiation ofTTX-resistant Nav1.8 current by neurokinin-1 receptor activation in rat dorsal rootganglion neurons. Mol Pain 5:33.

50. Sculptoreanu A, de Groat WC (2003) Protein kinase C is involved in neurokininreceptor modulation of N- and L-type Ca2+ channels in DRG neurons of the adult rat.

J Neurophysiol 90:21–31.51. Abdulla FA, Stebbing MJ, Smith PA (2001) Effects of substance P on excitability and

ionic currents of normal and axotomized rat dorsal root ganglion neurons. Eur

J Neurosci 13:545–552.52. Gamper N, Shapiro MS (2003) Calmodulin mediates Ca2+-dependent modulation of

M-type K+ channels. J Gen Physiol 122:17–31.53. Lin CC, et al. (2012) An antinociceptive role for substance P in acid-induced chronic

muscle pain. Proc Natl Acad Sci USA 109:E76–E83.54. Gamper N, Stockand JD, Shapiro MS (2003) Subunit-specific modulation of KCNQ

potassium channels by Src tyrosine kinase. J Neurosci 23:84–95.55. Li Y, Langlais P, Gamper N, Liu F, Shapiro MS (2004) Dual phosphorylations underlie

modulation of unitary KCNQ K(+) channels by Src tyrosine kinase. J Biol Chem 279:45399–45407.

56. Jia Q, et al. (2007) Activation of epidermal growth factor receptor inhibits KCNQ2/3current through two distinct pathways: Membrane PtdIns(4,5)P2 hydrolysis and

channel phosphorylation. J Neurosci 27:2503–2512.57. Vierck CJ, Jr. (2006) Animal models of pain. Wall and Melzack’s Textbook of Pain, eds

McMahon SB, Koltzenburg M (Elsevier Churchill Livingstone, Edinburgh), 5th Ed, pp

175–185.

E1586 | www.pnas.org/cgi/doi/10.1073/pnas.1201544109 Linley et al.