Involvement Of CB1Receptors In Electroacupuncture-induced Analgesia

Background:Pain-related diseases account for about80%of the out-patients. Pain could induce severestress response, delay recovery and decrease life quality. Therefore appropriate andeffective analgesic therapies are of great importance in early functional recovery andreducing mortality. But current available analgesic treatments have either limitedeffectiveness or deleterious side-effects. Thus, effective analgesic treatments that canproduce satisfactory pain relief are urgently needed. As a healing art in traditional Chinesemedicine, electroacupuncture (EA) administration is safe, reliable, effective and easy tooperate, and its analgesic effect has been acknowledged worldwide and has been widelyused to alleviate acute and chronic pain, such as postoperative pain, low back pain andchronic headache. The central mechanisms of electroacupuncture induced analgesia havebeen partially proposed through recent laboratory and clinical studies. It has been reportedthat endogenous opioid system, glutamate and its receptors and γ-Amino-butyric acidreceptors are involved in EA induced analgesia. But still it is not clear that which specific neuronal populations mediate this analgesia and which brain regions are involved. Furtherstudies of the mechanisms of EA induced analgesia will guide clinical EA treatments andhelp the popularization and application of Traditional Chinese Medicine all around theworld. CB1receptors are widely distributed in nociceptive pathways and play animportant role in regulating pain signaling in central nervous system. Our previous studiesshowed that activation of CB1receptors mediates the cerebral ischemic tolerance inducedby electroacupuncture pretreatment. These findings indicate that CB1receptors might beinvolved in EA induced analgesia. But as previous studies showed, CB1receptors areexpressed on different types of cells (such as glutamatergic neurons and GABAergicneurons), the consequences of systemic or even local manipulations of CB1receptoractivity reflect a “net” effect of CB1signalling onto these various target cells and exert itseffects in multiple different cells and tissue types, creating a complex phenotype in whichit is difficult to distinguish its direct functions in particular cells. Therefore we used theCre/LoxP system to generate cell-type specific CB1receptor knockout mouse lines toinvestigate the role of CB1receptors on different neuronal populations on EA inducedanalgesia, and provide structural and functional evidences to elucidate the centralmechanisms of EA induced analgesia.Part1CB1receptor is required for low-frequency EA induced analgesiaObjectiveTo investigate if CB1receptor is required for EA induced analgesia by using both CB1receptor antagonist and CB1-KO mice.Method1Effect of immobilization and needle insertion on basal thermal latency in C57BL/6Nmice Male C57BL/6N mice weighing23-28g (n=4) were put into holders with needle insertion(without electrical current). After60minutes of habituation, basal latency was measured,followed by TFL measurement every10minutes for30minutes.2Effect of CB1receptor antagonist on low-frequency EA induced analgesiaMale C57BL/6N mice weighing23-28g (n=20) were randomly assigned to2groups:vehicle group (Ve)(n=12) and rimonabant group (SR)(n=8). Mice in both groups wereput into holders with needle insertion (without electrical current) and received vehicle or3mg/kg rimonabant respectively (i.p.). Basal latency was measured60minutes after theinjection, then2Hz EA was applied at bilateral “Zusanli” and “Sanyinjiao” acupoints for30minutes. TFL was measured every10minutes after the onset of EA, with the electricalstimulation paused temporarily during the TFL assessment.3Effect of CB1receptor antagonist on high-frequency EA induced analgesiaMale C57BL/6N mice weighing23-28g (n=14) were randomly assigned to2groups:vehicle group (Ve)(n=8) and rimonabant group (SR)(n=6). Mice in both groups were putinto holders with needle insertion (without electrical current) and received vehicle or3mg/kg rimonabant respectively (i.p.). Basal latency was measured60minutes after theinjection, then100Hz EA was applied at bilateral “Zusanli” and “Sanyinjiao” acupointsfor30minutes. TFL was measured every10minutes after the onset of EA, with theelectrical stimulation paused temporarily during the TFL assessment.4Effect of low-frequency EA on TFL in CB1-KO miceMale CB1-KO mice (n=7) and their wild-type littermates (WT)(n=9)(weighing23-28g)were put into holders with needle insertion (without electrical current) for60minutes.After basal latency was measured,2Hz EA was applied at bilateral “Zusanli” and“Sanyinjiao” acupoints for30minutes. TFL was measured every10minutes after theonset of EA, with the electrical stimulation paused temporarily during the TFLassessment.5Effect of high-frequency EA on TFL in CB1-KO miceMale CB1-KO mice (n=8) and their wild-type littermates (WT)(n=6)(weighing23-28g)were put into holders with needle insertion (without electrical current) for60minutes. After basal latency was measured,100Hz EA was applied at bilateral “Zusanli” and“Sanyinjiao” acupoints for30minutes. TFL was measured every10minutes after theonset of EA, with the electrical stimulation paused temporarily during the TFLassessment.Results1Immobilization and needle insertion do not affect basal thermal latency in C57BL/6NmiceNo statistical difference was found between TFL on each time point and the basal latency.2CB1receptor antagonist reverses the low-frequency EA induced analgesiaEA at low-frequency induced a long-lasting increase of TFL in Ve group and producedpeak analgesic effect at the end of EA application. The percent increase of TFL in SRgroup during the course of EA was significantly lower than that in Ve group.3CB1receptor antagonist does not affect high-frequency EA induced analgesiaEA at high-frequency induced a long-lasting increase of TFL in both Ve and SR groupsand produced peak analgesic effect at the end of EA application. The percent increase ofTFL in SR group during the course of EA had no statistical difference from that in Vegroup.4Low-frequency EA has no analgesic effect in CB1-KO miceEA at low-frequency induced a long-lasting increase of TFL in WT mice and producedpeak analgesic effect at the end of EA application. There’s no statistical differencebetween TFL on each time point and the basal latency in CB1-KO mice, and the percentincrease of TFL in CB1-KO mice during the course of EA was significantly lower thanthat in WT mice.5High-frequency EA induces analgesic effect in CB1-KO miceEA at high-frequency induced a long-lasting increase of TFL in both WT and CB1-KOmice and produced peak analgesic effect at the end of EA application. The percentincrease of TFL in CB1-KO mice during the course of EA had no statistical differencefrom that in WT mice. ConclusionsBoth low and high-frequency electroacupuncture on bilateral “Zusanli” and “Sanyinjiao”acupoints induce analgesic effect, and the low-frequency EA induced analgesia ismediated by CB1receptor, while high-frequency EA induced analgesia is CB1receptorindependent.Part2Low-frequency EA induced analgesia is mediated by CB1receptors expressed oncortical glutamatergic neuronsObjectiveTo investigate the role of CB1receptors on different neuronal populations onlow-frequency EA induced analgesia by using different lines of conditional CB1knockoutmice.Methods1Effect of immobilization and needle insertion on basal thermal latency inGlu/GABA-CB1-KO miceMale Glu/GABA-CB1-KO mice (n=4) and their wild-type littermates (WT)(n=4)(weighing23-28g) were put into holders with needle insertion (without electrical current).After60minutes of habituation, basal latency was measured, followed by TFLmeasurement every10minutes for30minutes.2Effect of low-frequency EA on TFL in CaMK-CB1-KO miceMale CaMK-CB1-KO mice (n=6) and their wild-type littermates (WT)(n=7)(weighing23-28g) were put into holders with needle insertion (without electrical current) for60minutes. After basal latency was measured,2Hz EA was applied for30minutes. TFL wasmeasured every10minutes after the onset of EA, with the electrical stimulation pausedtemporarily during the TFL assessment. 3Effect of low-frequency EA on TFL in Glu-CB1-KO miceMale Glu-CB1-KO mice (n=8) and their wild-type littermates (WT)(n=7)(weighing23-28g) were put into holders with needle insertion (without electrical current) for60minutes. After basal latency was measured,2Hz EA was applied for30minutes. TFL wasmeasured every10minutes after the onset of EA, with the electrical stimulation pausedtemporarily during the TFL assessment.4Effect of low-frequency EA on TFL in Glu/GABA-CB1-KO miceMale Glu-CB1-KO mice (n=10), GABA-CB1-KO mice (n=9), Glu/GABA-CB1-KO mice(n=7) and their wild-type littermates (WT)(n=9)(weighing23-28g) were put into holderswith needle insertion (without electrical current) for60minutes. After basal latency wasmeasured,2Hz EA was applied for30minutes. TFL was measured every10minutes afterthe onset of EA, with the electrical stimulation paused temporarily during the TFLassessment.5Effect of low-frequency EA on TFL in D1-CB1-KO miceMale D1-CB1-KO mice (n=5) and their wild-type littermates (WT)(n=6)(weighing23-28g) were put into holders with needle insertion (without electrical current) for60minutes. After basal latency was measured,2Hz EA was applied for30minutes. TFL wasmeasured every10minutes after the onset of EA, with the electrical stimulation pausedtemporarily during the TFL assessment.Results1Immobilization and needle insertion do not affect basal thermal latency inGlu/GABA-CB1-KO miceNo statistical difference was found between TFL on each time point and the basal latency.2Low-frequency EA has no analgesic effect in CaMK-CB1-KO miceEA induced a long-lasting increase of TFL in WT mice and produced peak analgesic effectat the end of EA application. There’s no statistical difference between TFL on each timepoint and the basal latency in CaMK-CB1-KO mice, and the percent increase of TFL inCaMK-CB1-KO mice during the course of EA was significantly lower than that in WT mice.3Low-frequency EA has no analgesic effect in Glu-CB1-KO miceEA induced a long-lasting increase of TFL in WT mice and produced peak analgesic effectat the end of EA application. There’s no statistical difference between TFL on each timepoint and the basal latency in Glu-CB1-KO mice, and the percent increase of TFL inGlu-CB1-KO mice during the course of EA was significantly lower than that in WT mice.4Low-frequency EA has no analgesic effect in Glu/GABA-CB1-KO and Glu-CB1-KOmice, and the induced analgesic effect in GABA-CB1-KO mice is lower than that in WTmiceEA induced a long-lasting increase of TFL in WT and GABA-CB1-KO mice and producedpeak analgesic effect at the end of EA application. There’s no statistical differencebetween TFL on each time point and the basal latency in Glu/GABA-CB1-KO andGlu-CB1-KO mice, and the percent increase of TFL at the end of EA in GABA-CB1-KOgroup is lower that in WT group, higher than that in Glu/GABA-CB1-KO andGlu-CB1-KO groups.5Low-frequency EA induces analgesic effect in D1-CB1-KO miceEA induced a long-lasting increase of TFL in both WT and D1-CB1-KO mice andproduced peak analgesic effect at the end of EA application. The percent increase of TFLin D1-CB1-KO mice during the course of EA had no statistical difference from that in WTmice.ConclusionsResult by using CamK-CB1-KO mice indicates that CB1receptors on principal forebrainneurons play an important role in low-frequency EA induced analgesia; then Glu-CB1-KOmice, Glu/GABA-CB1-KO mice and D1-CB1-KO mice were used to explore the CB1receptor cell-type specificity and showed that the analgesic effect of low-frequency EA ismediated by CB1receptors on cortical glutamatergic neurons. Part3CB1receptor-dependent acute inhibition of presynaptic glutamatergic transmissioncontributes to low-frequency EA induced analgesiaObjectiveTo investigate the role of CB1receptor-dependent acute inhibition of excitatoryglutamatergic transmission in low-frequency EA induced analgesia by using subeffectivedose of NMDA receptor antagonist MK-801.Method1Dose response of NMDA receptor antagonist MK-801induced analgesic effectMale WT*CB1mice weighing23-28g (n=31) were randomly assigned to5groups:vehicle group (Ve)(n=6),0.02mg/kg MK-801group (MK0.02)(n=6),0.06mg/kg MK-801group (MK0.06)(n=7),0.18mg/kg MK-801group (MK0.18)(n=6) and0.54mg/kgMK-801group (MK0.54)(n=6). Mice in all the groups were put into holders with needleinsertion (without electrical current) for60minutes, then basal latency was measured,followed by injection of vehicle or different doses of MK-801(i.p.).10minutes later, TFLwas measured every10minutes for30minutes.2Dose response of the effect of NMDA receptor antagonist MK-801on low-frequencyEA induced analgesiaMale WT*CB1mice weighing23-28g (n=41) were randomly assigned to5groups:vehicle group (Ve)(n=12),0.003mg/kg MK-801group (MK0.003)(n=6),0.01mg/kgMK-801group (MK0.01)(n=7),0.03mg/kg MK-801group (MK0.03)(n=8) and0.06mg/kg MK-801group (MK0.06)(n=8). Mice in all the groups were put into holderswith needle insertion (without electrical current) for60minutes, then basal latency wasmeasured, followed by injection of vehicle or different doses of MK-801(i.p.).10minuteslater,2Hz EA was applied for30minutes. TFL was measured every10minutes after theonset of EA, with the electrical stimulation paused temporarily during the TFLassessment. 3Effect of low dose MK-801on EA induced analgesia in CB1-KO miceMale CB1-KO mice and their wild-type littermates (WT)(weighing23-28g)(n=49) wererandomly assigned to4groups: WT mice injected with vehicle (Vehicle+WT)(n=11), WTmice injected with MK-801(MK-801+WT)(n=13), CB1-KO mice injected with vehicle(Vehicle+CB1-KO)(n=11) and CB1-KO mice injected with MK-801(MK-801+CB1-KO)(n=14). All the mice were put into holders with needle insertion (without electrical current)for60minutes. Then basal latency was measured, followed by injection of vehicle or0.003mg/kg MK-801(i.p.).10minutes later,2Hz EA was applied for30minutes. TFLwas measured after the termination of EA.4Effect of low dose MK-801on EA induced analgesia in Glu-CB1-KO miceMale Glu-CB1-KO mice and their wild-type littermates (weighing23-28g)(n=48) wererandomly assigned to4groups: WT mice injected with vehicle (Vehicle+WT)(n=18), WTmice injected with MK-801(MK-801+WT)(n=13), Glu-CB1-KO mice injected withvehicle (Vehicle+Glu-CB1-KO)(n=10) and Glu-CB1-KO mice injected with MK-801(MK-801+Glu-CB1-KO)(n=7). All the mice were put into holders with needle insertion(without electrical current) for60minutes. Then basal latency was measured, followed byinjection of vehicle or0.003mg/kg MK-801(i.p.).10minutes later,2Hz EA was appliedfor30minutes. TFL was measured after the termination of EA.Results1High dose of MK-801(0.54mg/kg) induces analgesic effect while middle dose ofMK-801(0.06mg/kg) has no analgesic effectMK-801at the dose of0.54mg/kg induced an increase of TFL and produced peakanalgesic effect at the time point of TFL4. There’s no statistical difference between TFLon each time point and the basal latency in the other4groups.2Middle dose of MK-801(0.06mg/kg) reverses EA induced analgesia while low dose ofMK-801(0.003mg/kg) has no effect on EA induced analgesiaVehicle and MK-801at the dose of0.003mg/kg,0.01mg/kg and0.03mg/kg induced anincrease of TFL and produced peak analgesic effect at the end of EA application. There’s no statistical difference between TFL on each time point and the basal latency in MK0.06group. And the percent increase of TFL in MK0.06group during the course of EA wassignificantly lower than that in Ve group.3Low dose MK-801(0.003mg/kg) restores the analgesic effect of2Hz EA in CB1-KOmiceEA induced an increase of TFL in Vehicle+WT, MK-801+WT and MK-801+CB1-KOgroups, but did not significantly affect the TFL in Vehicle+CB1-KO. The percent increaseof TFL in Vehicle+CB1-KO group at the end of EA was significantly lower than that inVehicle+WT group, and there’s significant interaction between MK-801administrationand CB1-KO/WT genotype.4Low dose MK-801(0.003mg/kg) restores the analgesic effect of2Hz EA inGlu-CB1-KO miceEA induced an increase of TFL in Vehicle+WT, MK-801+WT and MK-801+Glu-CB1-KOgroups, but did not significantly affect the TFL in Vehicle+Glu-CB1-KO. The percentincrease of TFL in Vehicle+Glu-CB1-KO group at the end of EA was significantly lowerthan that in Vehicle+WT group, and there’s significant interaction between MK-801administration and Glu-CB1-KO/WT genotype.ConclusionsHigh dose of MK-801(0.54mg/kg) induces analgesic effect, middle dose of MK-801(0.06mg/kg) reverses EA induced analgesia and low dose of MK-801(0.003mg/kg)restores EA induced analgesic effect in CB1-KO and Glu-CB1-KO mice. These resultsindicate that CB1receptor-dependent acute inhibition of excitatory glutamatergictransmission contributes to EA induced analgesia. Part4PAG is involved in CB1receptor mediated EA induced analgesiaObjectiveTo explore which brain area is activated by low-frequency EA and if this activation isregulated by the presynaptic CB1receptors located on the terminal of glutamatergicneurons projected from brain cortex.Method1Low-frequency EA induced c-fos expression on PAG in WT miceMale WT*CB1mice weighing23-28g (n=9) were randomly assigned to3groups: homecage group (HC)(n=3), immobilization group (IM)(n=3) and EA group (n=3). Mice inHC group received no treatment before intracardiac perfusion; mice in IM and EA groupswere put into holders with needle insertion (without electrical current) for60minutes,followed by immobilization or2Hz EA for30minutes, then mice were returned to theirhome cages and were sacrificed by intracardiac perfusion90minutes later.2Low-frequency EA induced c-fos expression on PAG in Glu-KO-CB1miceMale Glu-CB1-KO mice (n=12) and their wild-type littermates (n=13)(weighing23-28g)were randomly assigned to4groups: WT mice received only immobilization group(WT+IM)(n=7), WT mice received EA group (WT+EA)(n=6), Glu-CB1-KO micereceived only immobilization group (Glu-CB1-KO+IM)(n=5) and Glu-CB1-KO micereceived EA group (Glu-CB1-KO+EA)(n=7). All the mice were put into holders withneedle insertion (without electrical current) for60minutes, followed by immobilization or2Hz EA for30minutes, then mice were returned to their home cages and were sacrificedby intracardiac perfusion90minutes later.Results1Low-frequency induces c-fos expression on PAGIn EA group, c-fos was predominantly expressed on vlPAG, and in vlPAG, c-fosimmunopositive cells in EA group was significantly higher than that in IM and HC group,and there’s only very faint c-fos expression in HC group.2Low-frequency EA induced c-fos expression disappeares on rostral vlPAG inGlu-CB1-KO mice C-fos was mainly expressed on vlPAG in each group. In rostral vlPAG, c-fos expression ofWT+EA group was significantly higher than that of the other groups (no significantdifference of c-fos expression among the other3groups), and there’s significantinteraction between EA treatment and Glu-CB1-KO/WT genotype.ConclusionsLow-frequency EA increases c-fos expression in vlPAG, and this increased c-fosexpression disappeares in rostral vlPAG in Glu-CB1-KO mice, which indicate that thelow-frequency EA induced vlPAG activation is regulated by the presynaptic CB1receptorslocated on the terminal of glutamatergic neurons projected from brain cortex.Summary1． CB1receptor is required for low-frequency EA induced analgesia.2． The analgesic effect induced by low-frequency EA is mediated by CB1receptorsexpressed on cortical glutamatergic neurons.3． CB1receptor-dependent acute inhibition of excitatory glutamatergic transmissioncontributes to EA induced analgesia.4． vlPAG is one of the critical sites in brain for low-frequency EA induced analgesia.5． The low-frequency EA induced vlPAG activation is regulated by the presynaptic CB1receptors located on the terminal of glutamatergic neurons projected from brain cortex.