The Effect And Mechanism Of HMGB1 In Cocaine-induced Reward Behavior Of Rat

Part Ⅰ The Effect of HMGB1 on Cocaine-induced Reward Behavior of RatObjective: It has been reported that cocaine exposure activates the central immune system,yet the role of neuron-immune axis in cocaine-induced reward memory remains unknown.At the same time,it is known that toll-like receptor 4(TLR4)/ microglial signal can regulate cocaine-induced behaviors,however,the relative mechanism is still not clear.When the nuclear protein high mobility group box-1 protein(HMGB1)is released into the extracellular space,can it be used as the upstream signal of TLR4,or can it regulate the behavior induced by cocaine? Here,we established a model of cocaine administration,together with the conditioned place preference behavior to determine the role of HMGB1 in cocaine-induced reward memory.Methods: The model of conditioned place preference induced by intraperitoneal cocaine was used.The conditioned place preference score was evaluated to detect the reward behavior induced by cocaine;Sucrose preference test(SPT),elevated cross maze and open field test(OFT)were performed to evaluate the behavioral specificity of interventions or drugs;Protein immunoblotting detection and enzyme-linked immunosorbent assay(ELISA)were used to detect changes in HMGB1 expression in the Nucleus Accumbens(NAc)of SD rats.The effects of HMGB1 on cocaine-induced reward memory were determined by virus injection to the NAc.Results:(1)The changes of HMGB1 in extracellular space of nucleus accumbens were detected through the dialysis and ELISA method.Compared with normal saline,after repeated cocaine exposure,HMGB1 levels in extracellular space increased significantly(D7,Control: 6.07 ± 4.01 ng/ml;Cocaine: 18.96 ± 1.96 ng/ml,P < 0.01).(2)The component of synaptic protein in the nucleus accumbens region was analysed through the protein immunoblot assay.Compared with normal saline,after repeated cocaine exposure,the enrichment of HMGB1 was found in the synapse.(3)Conditioned place preferences were induced by cocaine injection.Compared with saline group,the conditioned place preference score of SD rats with cocaine exposure increased significantly(Control: 36.01 ± 33.91 s;Cocaine: 271.9 ± 30.19 s,P < 0.0001).Accordingly,HMGB1 protein level in the nucleus accumbens increased significantly with cocaine exposure(Control: 1.00 ± 0.03;Cocaine: 1.21 ± 0.06,P < 0.05).(4)The immunofluorescence analysis showed that intranuclear HMGB1 in the nucleus accumbens neurons was translocated to outside after cocaine exposure.(5)After overexpression of HMGB1 in the nucleus accumbens neuron,the conditioned place preference score increased significantly(Control:-4.88 ± 38.86 s;HMGB1-OE: 227.00 ± 69 s,P < 0.05).(6)After intraperenteral injection of glycyrrhizin and carbenoxolone to inhibit HMGB1,it was found that 50 mg/kg glycyrrhizin reduced the location preference score from 258.9 ± 22.11 s to-127.5 ± 95.58 s(P < 0.01),20 mg/kg carbenoxolone reduced the location preference score from 258.9 ± 22.11 s to-9.81 ± 29.42 s(P<0.05).However,direct intraperenteral injection of two drugs did not affect the spontaneous activity,anxiety level and natural reward of SD rats.In addition,two drugs did not affect the acquisition of situational fear memory when given before the situational training.(7)Compared with control virus group,the conditioned place preference score of HMGB1 silencing group in the nucleus accumbens significantly reduced(Control: 298 ± 61.9 s;sh HMGB1:-53.59 ± 39.48 s,P < 0.001).(8)After micro-injection of small peptide into NAc area to simulate or block HMGB1 function,it was found that contrast with the Phosphate buffered saline(PBS)group(387.3 ± 50.12 s),the Box A treatment after training reduced the conditioned place preference score significantly(-52.30 ± 42.74 sec),S106-Box B treatment group position also significantly reduced the score(8.240 ± 42.95 s),however,there is no obvious change in C106-Box B treatment group(281.7 ± 110.6 s);After given different doses of cocaine before training,the conditioned place preference score in the S106-Box B treatment group was significantly reduced.(9)Intraperenteral injection of minocycline to supress the activation of microglial cells,the conditioned place preference score of the SD rats was significantly reduced from 300.1 ± 28.38 s to 58.42 ± 59 s(P < 0.05).Conclusion: Cocaine induced SD rats to perform conditioned place preferences.After cocaine exposure,HMGB1 increased in extracellular and synapse region in the nucleus accumbens,and neuronal intranuclear HMGB1 was released to outside.Genetic overexpression and deletion of HMGB1 bidirectionally regulated conditioned place preferences induced by cocaine.Pharmacologcial inhibition of HMGB1 also can specifically reduce the conditioned place preferences induced by cocaine without affecting the spontaneous activity.All above results prompt that HMGB1 plays an important role in reward memory induced by cocaine.Part Ⅱ The Mechanism of HMGB1 in Cocaine-induced Reward Behavior of RatObjective: The main receptors of HMGB1 are TLR4 and the receptor for advanced glycation end products(RAGE),both of which can be expressed on microglia cells.It has been reported that cocaine addiction is related to microglia activation,however,whether the regulation of microglia activity by nucleus accumbens HMGB1 was participated in cocaine-induced reward memory is unknown.This study aims to investigate the effect of HMGB1-microglia or other HMGB1 related signaling on reward memory induced by cocaine and the possible mechanism.Methods: The model of conditioned place preference induced by intraperitoneal cocaine was used.Real-time fluorescent quantitative polymerase chain reaction(policy real-time polymerase chain reaction,Q-PCR)and protein immunoblot assay were used to analysis the protein changes in NAc and m PFC of SD rats.Using coimmunoprecipitation(Co-IP)to further analyze the changes of HMGB1 function.The effect of HMGB1 on the synaptic adaptability induced by cocaine was observed through electrophysiology.The effect and mechanism of HMGB1 on the reward memory induced by cocaine were further clarified by adjusting the training mode of conditioned place preference.Results:(1)After repeated cocaine exposure for three times,HMGB1 protein levels of SD rats in the nucleus accumbens increased to a maximum of 1.21 ± 0.05(P < 0.01),at the same time,HMGB1 m RNA level in the nucleus accumbens did not increase,while there is no change of HMGB1 protein levels in m PFC area during the course of the cocaine exposure.(2)The level of Iba1 protein in the nucleus accumbens of SD rats increased after repeated cocaine administration for three times and kept to increase until the seventh day.The protein level of ED1 was only increased after repeated cocaine administration for seven times.(3)50 mg/kg glycyrrhizin decreased conditioned place preference and activity of microglia cells in the nucleus accumbens after cocaine exposure.30 mg/kg of minocycline decreased conditioned place preference induced by cocaine,but the increased protein level of HMGB1 in the nucleus accumbens with cocaine exposure was not reversed.(4)The protein level of TLR4 in the nucleus accumbens began to increase after a single dose of cocaine,and continued until the drug was repeated seven times.The protein level of TLR2 was only increased after repeated administration of cocaine for seven times.(5)After training on the conditioned place preference of cocaine,the protein level of RAGE in the nucleus accumbens of SD rats was significantly increased to 1.23 ± 0.02(P < 0.0001).(6)The increase of TLR4 and RAGE caused by cocaine exposure and the change of downstream mitogen-activated protein kinase(MAPK)signal can be abolished by suppressing and silencing HMGB1.(7)After the cocaine exposure,silencing HMGB1 reduced the total protein of AMPA receptor subunit Glu A2 in the NAc region,without affecting the total amount of Glu A1,Glu A3 and NMDA receptor subunit Glu N2 A,Glu N2 B.(8)It was found that HMGB1 could combine with Glu A2 and Glu N2 B,and the binding site with Glu A2 was Glu A2843-852.(9)The electrophysiological results showed that the ratio of Ca2+-permeable AMPA receptors(CP-AMPARs)in cocaine group increased to 22.46 ± 2.02 %(P < 0.05).The ratio of CP-AMPARs in the nucleus accumbens of SD rats was significantly reduced to 11.42 ± 3.94 %(P < 0.05)after 20 mg/kg of carbenoxolone administered.The ratio of CP-AMPARs was significantly increased to 48.98 ± 7.51 %(P < 0.01)after 50 mg/kg glycyrrhizin administered.(10)In conditioned place preference model under the condition of multiple scenarios,after completing all training,the conditioned place preference score of cocaine group in the scenario B has not changed.After 50 mg/kg glycyrrhizin or 20 mg/kg carbenoxolone treatment,the conditioned place preference scores in the scenario B were both changed.(11)By changing the multi-context conditioned place preference training mode,the conditioned place preference was first obtained in Context A,then the rats were trained in the unfamiliar Context B and returned to Context A for priming detection.Compared with the normal saline control group,the effect of the training itself was cancelled in the CBX treatment group.Conclusion: HMGB1 regulates the adaptive changes of CP-AMPARs induced by cocaine via combining with Glu A2 and activating the microglia.After cocaine exposure,microglia activation in the nucleus accumbens region depends on the extracellular HMGB1.When the memory associated with cocaine is activated again,microglia activation damages new learning information,so as to maintain the original memory.The above results suggest that HMGB1 is involved in drug-related adverse behavior,and the mechanism may be related to changes in the adaptability of synapses and microglial cells,which also provides a new strategy for the treatment of addiction.