Advertisement

Matrix Metalloproteinase-9 and Synaptic Plasticity in the Central Amygdala in Control of Alcohol-Seeking Behavior

      Abstract

      Background

      Dysfunction of the glutamatergic system has been implicated in alcohol addiction; however, the molecular underpinnings of this phenomenon are still poorly understood. In the current study we have investigated the possible function of matrix metalloproteinase-9 (MMP-9) in alcohol addiction because this protein has recently emerged as an important regulator of excitatory synaptic plasticity.

      Methods

      For long-term studies of alcohol drinking in mice we used IntelliCages. Dendritic spines were analyzed using Diolistic staining with DiI. Whole-cell patch clamp was used to assess silent synapses. Motivation for alcohol in human subjects was assessed on the basis of a Semi-Structured Assessment for the Genetics of Alcoholism interview.

      Results

      Mice devoid of MMP-9 (MMP-9 knockout) drank as much alcohol as wild-type animals; however, they were impaired in alcohol seeking during the motivation test and withdrawal. The deficit could be rescued by overexpression of exogenous MMP-9 in the central nucleus of the amygdala (CeA). Furthermore, the impaired alcohol seeking was associated with structural alterations of dendritic spines in the CeA and, moreover, whole-cell patch clamp analysis of the basal amygdala to CeA projections showed that alcohol consumption and withdrawal were associated with generation of silent synapses. These plastic changes were impaired in MMP-9 knockout mice. Finally, C/T polymorphism of MMP-9 gene at position –1562, which upregulates MMP-9 expression, correlated with increased motivation for alcohol in alcoholics.

      Conclusions

      In aggregate, our results indicate a novel mechanism of alcohol craving that involves MMP-9–dependent synaptic plasticity in CeA.

      Keywords

      Drug addiction, a psychiatric disorder characterized by uncontrolled drug taking and seeking, is thought to be driven by aberrant glutamatergic transmission in the brain reward system (
      • Kelley A.E.
      Memory and addiction: Shared neural circuitry and molecular mechanisms.
      ,
      • Bowers M.S.
      • Chen B.T.
      • Bonci A.
      AMPA receptor synaptic plasticity induced by psychostimulants: The past, present, and therapeutic future.
      ,
      • Lüscher C.
      • Malenka R.C.
      Drug-evoked synaptic plasticity in addiction: From molecular changes to circuit remodeling.
      ,
      • Pascoli V.
      • Terrier J.
      • Espallergues J.
      • Valjent E.
      • O’Connor E.C.
      • Lüscher C.
      Contrasting forms of cocaine-evoked plasticity control components of relapse.
      ). This hypothesis is supported by the observation that chronic alcohol consumption leads to elevated levels of extracellular glutamate and alterations in the expression and localization of various glutamate receptors, including alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptors (
      • Holmes A.
      • Spanagel R.
      • Krystal J.H.
      Glutamatergic targets for new alcohol medications.
      ,
      • Holmes A.
      • Fitzgerald P.J.
      • MacPherson K.P.
      • DeBrouse L.
      • Colacicco G.
      • Flynn S.M.
      • et al.
      Chronic alcohol remodels prefrontal neurons and disrupts NMDAR-mediated fear extinction encoding.
      ). The role of the glutamatergic system in the development of addiction is also strengthened by new therapies of alcoholism that are focused on glutamate receptors (
      • Bowers M.S.
      • Chen B.T.
      • Bonci A.
      AMPA receptor synaptic plasticity induced by psychostimulants: The past, present, and therapeutic future.
      ,
      • Holmes A.
      • Spanagel R.
      • Krystal J.H.
      Glutamatergic targets for new alcohol medications.
      ,
      • Rao P.S.S.
      • Bell R.L.
      • Engleman E.A.
      • Sari Y.
      Targeting glutamate uptake to treat alcohol use disorders.
      ). Despite the great effort to understand the regulation of drug-evoked plasticity of the glutamatergic system, the molecular details are still largely missing.
      Recently, matrix metalloproteinase-9 (MMP-9) has emerged as an important controller of the synaptic plasticity of excitatory synapses (
      • Vafadari B.
      • Salamian A.
      • Kaczmarek L.
      MMP-9 in translation: From molecule to brain physiology, pathology and therapy.
      ). MMP-9 was found to affect surface diffusion of NMDA receptors (NMDARs) (
      • Michaluk P.
      • Mikasova L.
      • Groc L.
      • Frischknecht R.
      • Choquet D.
      • Kaczmarek L.
      Matrix metalloproteinase-9 controls NMDA receptor surface diffusion through integrin beta1 signaling.
      ), NMDAR function (
      • Gorkiewicz T.
      • Szczuraszek K.
      • Wyrembek P.
      • Michaluk P.
      • Kaczmarek L.
      • Mozrzymas J.W.
      Matrix metalloproteinase-9 reversibly affects the time course of NMDA-induced currents in cultured rat hippocampal neurons.
      ), NMDAR-dependent long-term potentiation of synaptic efficacy, activity-driven alterations of morphology of dendritic spines (
      • Szepesi Z.
      • Bijata M.
      • Ruszczycki B.
      • Kaczmarek L.
      • Wlodarczyk J.
      Matrix metalloproteinases regulate the formation of dendritic spine head protrusions during chemically induced long-term potentiation.
      ), and different forms of learning (
      • Vafadari B.
      • Salamian A.
      • Kaczmarek L.
      MMP-9 in translation: From molecule to brain physiology, pathology and therapy.
      ,
      • Huntley G.W.
      Synaptic circuit remodelling by matrix metalloproteinases in health and disease.
      ,
      • Dityatev A.
      • Schachner M.
      • Sonderegger P.
      The dual role of the extracellular matrix in synaptic plasticity and homeostasis.
      ). Moreover, an important insight into the role of MMP-9 in addiction has recently been provided by Smith et al. (
      • Smith A.C.W.
      • Kupchik Y.M.
      • Scofield M.D.
      • Gipson C.D.
      • Wiggins A.
      • Thomas CA
      • et al.
      Synaptic plasticity mediating cocaine relapse requires matrix metalloproteinases.
      ), who found transient increases in MMP-9 activity in the nucleus accumbens during cue-induced cocaine relapse. Increased MMP activity was required for both cocaine- and cue-induced relapse and relapse-associated synaptic plasticity (
      • Smith A.C.W.
      • Kupchik Y.M.
      • Scofield M.D.
      • Gipson C.D.
      • Wiggins A.
      • Thomas CA
      • et al.
      Synaptic plasticity mediating cocaine relapse requires matrix metalloproteinases.
      ). Furthermore, general inhibitors of MMPs attenuated escalating ethanol self-administration (
      • Smith A.W.
      • Nealey K.A.
      • Wright J.W.
      • Walker B.M.
      Plasticity associated with escalated operant ethanol self-administration during acute withdrawal in ethanol-dependent rats requires intact matrix metalloproteinase systems.
      ). The possible function of MMP-9 in alcohol addiction was also supported in human studies by Samochowiec et al. (
      • Samochowiec A.
      • Grzywacz A.
      • Kaczmarek L.
      • Bienkowski P.
      • Samochowiec J.
      • Mierzejewski P.
      • et al.
      Functional polymorphism of matrix metalloproteinase-9 (MMP-9) gene in alcohol dependence: Family and case control study.
      ), who found the polymorphism of the MMP-9 gene (at position –1562), producing higher protein expression, to be more frequent in alcoholics’ families than in control subjects’ families. Despite the growing body of evidence implicating MMP-9 in addiction, the detailed understanding of the function of this metalloproteinase in addiction-related processes is still missing.
      The aim of this study was to examine the role of MMP-9 in the regulation of alcohol addiction–related behaviors. First, we performed a longitudinal study in IntelliCages (NewBehavior AG, Zurich, Switzerland) to test whether depletion of MMP-9 in MMP-9 knockout (MMP-9 KO) mice affects different aspects of alcohol addiction, such as alcohol consumption, motivation to obtain alcohol, and alcohol seeking during withdrawal (
      • Radwanska K.
      • Kaczmarek L.
      Characterization of an alcohol addiction-prone phenotype in mice.
      ). Our experiment showed that MMP-9 KO mice drank as much alcohol as wild-type (WT) animals; however, they were impaired in alcohol seeking during motivation test and withdrawal, suggesting the role of MMP-9 in the regulation of alcohol craving. The impaired alcohol seeking was rescued by the local overexpression of exogenous MMP-9 in the central nucleus of the amygdala (CeA). Furthermore, the lack of MMP-9 prevented functional synaptic plasticity in the CeA, observed after alcohol withdrawal (i.e., an increase in the frequency of silent synapses). Moreover, after distinguishing the animals into groups of low- and high-motivation drinkers, we demonstrated that whereas in the highly motivated WT mice alcohol drinking resulted in an increase of dendritic spine size in CeA, such differences could not be seen in MMP-9 KO mice. Finally, we performed a validation study of the Semi-Structured Assessment for the Genetics of Alcoholism questionnaire and found that in the population of alcoholics MMP-9 gene –1562C/T polymorphism (rs3918242) correlates with increased motivation for alcohol.

      Methods and Materials

      The experiments were performed on adult C57BL/6J KO (MMP-9 KO) mice and their WT siblings (Supplemental Figure S1). The animals were generated and genotyped as described in the Supplemental Methods and Materials and in Vu et al. (
      • Vu T.H.
      • Shipley J.M.
      • Bergers G.
      • Berger J.E.
      • Helms J.A.
      • Hanahan D.
      • et al.
      MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes.
      ). Strain colony was maintained in the animal house of the Nencki Institute of Experimental Biology of Polish Academy of Sciences. All procedures were performed in accordance with the Animal Protection Act in Poland, Directive 2010/63/EU, and were approved by the first Local Ethics Committee (Permission No. 257/2012).
      All details of the experimental procedures are in the Supplemental Methods and Materials.

      IntelliCage Training

      The alcohol trainings in the IntelliCages were performed according to previously published protocols with some minor modifications (
      • Radwanska K.
      • Kaczmarek L.
      Characterization of an alcohol addiction-prone phenotype in mice.
      ,
      • Mijakowska Z.
      • Lukasiewicz K.
      • Ziolkowska M.
      • Lipinski M.
      • Trabczynska A.
      • Matuszek Z.
      • et al.
      Autophosphorylation of alpha isoform of calcium/calmodulin-dependent kinase II regulates alcohol addiction-related behaviors.
      ).

      Dendritic Spines Analysis

      After behavioral training, mice were sacrificed and brain tissue was collected. Brains were cut into 130-μm sections on vibratome (Leica VT 1000S, Leica Biosystems Nussloch GmbH, Wetzlar, Germany) and processed for Dil (Life Technologies, Warsaw, Poland) staining. Z-stacks of dendrites in the CeA were acquired using LSM780 confocal system (Zeiss, Poznan, Poland). Maximum intensity projections of images were analyzed using semiautomatic SpineMagick! Software (

      Ruszczycki B, Wlodarczyk J, Kaczmarek, inventors; Nencki Institute of Experimental Biology, assignee. Method and a system for processing an image comprising dendritic spines. U.S. Patent No. US2014169647 (A1). June 19, 2014.

      ). Data analysis was performed using custom scripts written in Python, using NumPy and SciPy (
      • Oliphant T.E.
      Python for scientific computing.
      ), IPython (
      • Perez F.
      • Granger B.E.
      IPython: A system for interactive scientific computing.
      ), scikit-learn (
      • Pedregosa F.
      • Varoquaux G.
      • Gramfort A.
      • Michel V.
      • Thirion B.
      • Grisel O.
      • et al.
      Scikit-learn: Machine learning in python.
      ), and Matplotlib (
      • Hunter J.D.
      Matplotlib: A 2D graphics environment.
      ).

      Injection of Lentiviral Vector Expressing MMP-9 Into the CeA

      Lentiviral vectors expressing autoactive MMP-9 driven under synapsin-1 promoter (LV-MMP-9) and green fluorescent protein (LV-GFP) were used for bilateral stereotactic injections into the CeA. Two weeks after the surgery mice underwent training in IntelliCages.

      Electrophysiology Experiments

      Mice were housed separately in their home cages, receiving food and water ad libitum. Mice exposed to long-term alcohol protocol received additional bottle containing alcohol (4% and 8% for 2 days each, 10% for 3 weeks). Mice were sacrificed 1 day after the termination of 3-week alcohol exposure (1–day withdrawal group) or after 7 days of withdrawal. Control mice were drinking water at all times. At the end of the training mice were anesthetized with isoflurane and decapitated. The whole-cell patch-clamp analysis of silent synapses was performed on the basal amygdala to CeA projections as previously described (
      • Lee B.R.
      • Ma Y.-Y.
      • Huang Y.H.
      • Wang X.
      • Otaka M.
      • Ishikawa M.
      • et al.
      Maturation of silent synapses in amygdala-accumbens projection contributes to incubation of cocaine craving.
      ,
      • Suska A.
      • Lee B.R.
      • Huang Y.H.
      • Dong Y.
      • Schlüter O.M.
      Selective presynaptic enhancement of the prefrontal cortex to nucleus accumbens pathway by cocaine.
      ).

      Human Subjects and Assessment

      In the present study the Semi-Structured Assessment for the Genetics of Alcoholism interview (
      • Rice J.P.
      • Reich T.
      • Bucholz K.K.
      • Neuman R.J.
      • Fishman R.
      • Rochberg N.
      • et al.
      Comparison of direct interview and family history diagnoses of alcohol dependence.
      ) was used to assess alcohol and family history of alcoholism and more precisely selected addiction criteria, as described by Samochowiec et al. (
      • Samochowiec A.
      • Grzywacz A.
      • Kaczmarek L.
      • Bienkowski P.
      • Samochowiec J.
      • Mierzejewski P.
      • et al.
      Functional polymorphism of matrix metalloproteinase-9 (MMP-9) gene in alcohol dependence: Family and case control study.
      ). The protocol of the study was approved by the local institutional review board, and all the participants provided their written informed consent.

      Statistical Analysis

      All results are expressed as mean ± SEM. The appropriate tests were chosen, taking into account whether data had normal distribution and equal variation. All analyses were conducted using GraphPad Prism, version 7.02 (GraphPad Software, Inc., La Jolla, CA) or SPSS (version 9, SPSS Inc., Chicago, IL). Differences between the experimental groups were considered significant if the type 1 error was less than 5%.

      Results

      MMP-9 KO Mice Have Decreased Motivation for Alcohol and Are Less Persistent in Alcohol Seeking

      To assess the role of MMP-9 in the regulation of alcohol addiction-related behaviors we used MMP-9 KO mice along with their WT littermates (
      • Vu T.H.
      • Shipley J.M.
      • Bergers G.
      • Berger J.E.
      • Helms J.A.
      • Hanahan D.
      • et al.
      MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes.
      ). The mice underwent a long-term alcohol addiction training in the IntelliCages as described in detail before by Radwanska and Kaczmarek (
      • Radwanska K.
      • Kaczmarek L.
      Characterization of an alcohol addiction-prone phenotype in mice.
      ). In this apparatus, the animals are housed in large groups and are provided with food ad libitum; however, to access liquids they have to make first a nose poke [NP] through a gate in a corner and then lick either water or alcohol in water from a dispenser. These responses are automatically recorded. We measured general activity (NPs in the corners) as well as several behaviors related to alcohol addiction such as alcohol consumption, alcohol seeking during motivation test, and withdrawal (Figure 1A, B). The mice were first exposed to water for the cage adaptation. Next, they were trained to perform NPs to get access to dispenser (NP adaptation). The following phase included adaptation to increasing concentration of ethanol (4%, 8%, and 12%) and subsequent long-term alcohol drinking (free alcohol access). During free alcohol access, the animals were tested for motivation to obtain alcohol in two independent tests as measured by the number of responses (NPs) to the alcohol corner performed to receive the alcohol reward and persistence in alcohol seeking during the alcohol nonreward period.
      Figure 1.
      Figure 1Matrix metalloproteinase-9 (MMP-9) knockout (KO) mice have decreased motivation for alcohol. MMP-9 KO mice (n = 11) and their wild-type (WT) littermates (n = 11) were trained in the IntelliCages. (A) Experimental timeline. (B) IntelliCage setup during the tests. During adaptation and long-term ethanol drinking animals had unlimited access to water or alcohol and water. The dashed line during motivation tests indicates increasing number of nose pokes (NPs) that animal needs to perform in combined fixed ratio (FR) and progressive ratio schedule to obtain access to alcohol. The solid line during withdrawal test indicates no access to alcohol. (C) General activity of mice during experiment. There was no effect of mice genotype on alcohol consumption (two-way analysis of variance [ANOVA]: genotype: [F1,18 = 1.18, p > .05], time [F7,126 = 3.720, p < .01]) and number of NPs performed to all corners (all NPs) during the experiment (two-way ANOVA: genotype [F1,17 = 3.78, p > .06], time [F12,204 = 65.43, p < .0001]). (D) Motivation for alcohol (M1 and M2). MMP-9 KO mice have lower motivation for alcohol as indicated by the number of NPs performed to the alcohol corner (reward NPs; two-way ANOVA: genotype [F1,152 = 8.48, p < .01], time [F7,152 = 0.82, p > .05]) as compared with WT mice in two motivation tests. There was no effect of genotype on NPs performed to the water corner (water NPs) during the tests (two-way ANOVA: genotype [F1,20 = 1.41, p = .24], time [F7,140 = 14.23, p < .0001]). (E) Alcohol withdrawal (W1). MMP-9 KO mice performed fewer reward NPs during withdrawal (W1) when compared with WT animals (two-way ANOVA: genotype [F1,19 = 5.51, p < .05], time [F6,114 = 6.039, p < .0001]). There was no difference between the genotypes in NPs performed to the water corner during the test (two-way ANOVA: genotype [F1,19 = 0.47, p = .49], time [F6,114 = 2.211, p < .05]). Day 0 indicates the day before the test. The effect of genotype: *p < .05. CA, cage adaptation; FAA, free alcohol access; M, alcohol motivation test; NPA, nose poke adaptation; R, relapse; W, withdrawal test; 4–12%, increasing concentration of ethanol.
      There were no significant effects of the mouse genotype on overall activity of WT and MMP-9 KO mice as indicated by the number of NPs performed by mice to all corners during all training periods (Figure 1C). Furthermore, there was no effect of the genotype on the initiation of alcohol drinking or later alcohol consumption during the training (Figure 1C).
      However, during the motivation tests, MMP-9 KO mice performed significantly fewer NPs to the alcohol door than did WT animals, indicating decreased motivation for alcohol in KO mice. Such difference was not observed when NPs to the water door were counted (Figure 1D). A similar trend was observed when mice behavior was analyzed during the signaled no-reward period (withdrawal; Figure 1E). Again, MMP-9 KO mice performed fewer NPs to the alcohol door than did WT mice. No difference was observed between the genotypes as far as NPs to water door were concerned.
      Additionally, to further assess whether the phenomena we observed were related to alcohol addiction behaviors and not to possible anhedonia in MMP-9 KO animals, we tested both genotypes for natural reward seeking. We found no difference between genotypes when test for reward seeking was performed in mice trained to drink 5% sucrose (Supplemental Figure S2).
      Thus, this experiment has indicated that MMP-9 does not affect alcohol consumption, but rather specifically affects alcohol-seeking behavior when access to alcohol is limited.

      MMP-9 Is Pivotal for Synaptic Plasticity in the CeA in Alcohol Addiction

      In the following experiments we decided to focus on the CeA, as multiple studies implicated the CeA in the regulation of alcohol seeking upon withdrawal (
      • Roberto M.
      • Gilpin N.W.
      • Siggins G.R.
      The central amygdala and alcohol: Role of γ-aminobutyric acid, glutamate, and neuropeptides.
      ), and our previous work showed that MMP-9 controlled synaptic plasticity underlying appetitive learning specifically in this region (
      • Knapska E.
      • Lioudyno V.
      • Kiryk A.
      • Mikosz M.
      • Go T.
      • Michaluk P.
      • et al.
      Reward learning requires activity of matrix metalloproteinase-9 in the central amygdala.
      ). We tested the density and size of dendritic spines in CeA in MMP-9 KO and WT mice, sacrificed after the IntelliCage training. We observed that MMP-9 KO mice had lower density of dendritic spines in CeA than did WT animals (Supplemental Figure S3A). The spines did not differ as far as the average size was concerned (Supplemental Figure S3B). Furthermore, we extended this analysis and divided the spines into shape categories (long, mushroom, and stubby) and the animals into categories on the basis of their performance in the last behavioral tests (6-day motivation test ended up with 1 day of withdrawal), after which they were sacrificed. Thirty percent of the animals that scored the highest in the test (performing the highest number of NPs to reward per day during the test) were assigned to the high category as previously described (
      • Radwanska K.
      • Kaczmarek L.
      Characterization of an alcohol addiction-prone phenotype in mice.
      ); the rest of animals were assigned to the low category (Figure 2A). We observed that in particular MMP-9 KO-low, but not MMP-9 KO-high, animals had lower density of long spines than did WT mice (Figure 2C). Other spine categories were not affected in MMP-9 KO mice, as far as density is concerned. Furthermore, we revealed that areas of mushroom and long spines were bigger in WT-high animals versus WT-low animals (Figure 2D). We also found a significant positive correlation between motivation and the area of long spines in WT animals (Supplemental Figure S3C). Interestingly, no correlation between spine size and alcohol motivation was observed in MMP-9 KO animals. Thus, our data suggest that decreased spine density and aberrant structural plasticity of dendritic spines in CeA underlay motivation deficits during alcohol drinking.
      Figure 2.
      Figure 2Alcohol addiction–related behavior and dendritic spine dynamics in the central amygdala of wild-type (WT) and matrix metalloproteinase-9 (MMP-9) knockout (KO) mice after long-term alcohol drinking in the IntelliCage system. (A) Persistence in alcohol seeking. Reward nose pokes (NPs) per day performed in the nonreward period (withdrawal 2) and before the test (7 days of baseline). WT high (WT-H) persisted/motivated animals performed more NPs to reward than did WT-low (WT-L) persisted/motivated WT animals during withdrawal (W) and baseline (B) (two-way analysis of variance [ANOVA]: effect of groupL/H [F1,14 = 8.101, p = .012], effect of testB/W [F1,14 = 7.173, p = .018]). KO-high (KO-H) persisted/motivated animals performed more nose pokes to reward than did KO-low (KO-L) persisted/motivated animals during withdrawal (two-way ANOVA: effect of groupL/H [F1,16 = 8.472, p = .01], effect of testB/W [F1,16 = 3.1, p = .097]). (B) Example dendrites of L and H persisted/motivated animals per genotype. (C) Density of central amygdala dendritic spines. MMP-9 KO-L mice have lower density of long spines when compared with WT-L animals. MMP-9 KO-H animals have higher density of dendritic spines when compared with MMP-9 KO-L mice (two-way ANOVA: effect of groupL/H [F1,14 = 1.118, p = .3083], effect of genotypeWT/KO [F1,14 = 5.034, p = .0415], followed by Tukey’s post hoc). No such change was observed in population of mushroom (two-way ANOVA: effect of groupL/H [F1,14 = 0.0294, p = .8662], effect of genotypeWT/KO [F1,14 = 0.2339, p = .6361]) and stubby spines (two-way ANOVA: effect of groupL/H [F1,14 = 4.427, p = .0539], effect of genotypeWT/KO [F1,14 = 1.745, p = .2077]). (D) Area of central amygdala dendritic spines. WT-H animals have bigger long (two-way ANOVA: effect of groupL/H [F1,15 = 13.24, p = .0024], effect of genotypeWT/KO [F1,15 = 0.4613, p = .5074], followed by Tukey’s post hoc) and mushroom spines (two-way ANOVA: effect of groupL/H [F1,15 = 8.069, p = .0124], effect of genotypeWT/KO [F1,15 < 0.0001, p = .9985]) when compared with WT-L animals. No change in size was observed in stubby spines for both WT and KO animals (two-way ANOVA: effect of groupL/H [F1,15 = 0.8535, p = .3702], effect of genotypeWT/KO [F1,15 = 0.0024, p = .9610]). WT-H n = 4, WT-L n = 5, KO-H n = 4, KO-L n = 6. *p < .05, **p < .01.

      Overexpression of Exogenous MMP-9 in the CeA Increases Motivation for Alcohol in MMP-9 KO Mice

      We also decided to test whether decreased motivation for alcohol observed in MMP-9 KO mice can be rescued by the local overexpression of exogenous MMP-9 protein in CeA. Our previous study indicated that local delivery of an endogenous MMP-9 inhibitor, tissue inhibitor of metalloproteinase-1, impairs reward-driven learning (
      • Knapska E.
      • Lioudyno V.
      • Kiryk A.
      • Mikosz M.
      • Go T.
      • Michaluk P.
      • et al.
      Reward learning requires activity of matrix metalloproteinase-9 in the central amygdala.
      ). In the current research we used LV vectors expressing MMP-9 and/ or GFP under synapsin-1 promoter (LV-MMP-9 and LV-GFP) (Figure 3A). Effective expression of exogenous MMP-9 in MMP-9 KO mice was confirmed using gel zymography (Figure 3A). MMP-9 KO mice and their WT siblings were injected into CeA with LVs. Next, the animals went through the alcohol training in the IntelliCages (Figure 3B). Overexpression of MMP-9, as compared with GFP, in MMP-9 KO mice affected neither alcohol consumption nor general activity, measured as the number of NPs performed to all doors in the cage (Figure 3C and Supplemental Figure S4A). However, it increased the motivation for alcohol and alcohol seeking during withdrawal, measured as a number of NPs performed to the alcohol door during the tests (Figure 3D, E). The expression of MMP-9 also slightly but significantly increased the number of NPs to the water door during the motivation test, but not during withdrawal. We observed no effect of MMP-9 overexpression on the behavior of WT mice (Supplemental Figure S4A–C).
      Figure 3.
      Figure 3Exogenous expression of matrix metalloproteinase-9 (MMP-9) in the central nucleus of the amygdala (CeA) increases motivation for alcohol in MMP-9 knockout (KO) mice. Wild-type (WT) and MMP-9 KO mice had lentiviral vectors (LVs) (LV-MMP-9 and LV-green fluorescent protein [GFP] or LV-GFP only) injected into the CeA and were trained to drink alcohol in the IntelliCages. (A) Lentiviral infection in the CeA and microphotography of the CeA transfected with LV-GFP. Scale bar = 500 µm. Gel zymography shows enzymatic activity of MMP-9 and MMP-2 in homogenates from the central amygdala of animals lacking functional MMP-9 (MMP-9 KO) after injection of LVs expressing MMP-9 and GFP or GFP only. (B) Experimental timeline. (C) General activity of mice during the experiment. There was no effect of viral transfection on alcohol consumption by WT mice (two-way analysis of variance [ANOVA] for alcohol consumption: virus [F1,14 = 2.27, p > .05]). There was also no effect of viral transfection on alcohol consumption by KO mice (two-way ANOVA for alcohol consumption: virus [F1,7 = 0.02, p > .05]). (D) Motivation for alcohol (M1 and M2). MMP-9 KO animals with exogenous local expression of MMP-9 (KO-LV-MMP-9, n = 5) in CeA had higher motivation for alcohol than did control MMP-9 KO mice (KO-LV-GFP, n = 4), as indicated by the number of reward nose pokes (NPs) performed during the tests (two-way ANOVA for reward NPs: virus [F1,50 = 28.12, p < .01]). No effect of virus was observed for WT mice on the number of reward NPs (two-way ANOVA for reward NPs: virus [F1,15 = 0.01, p > .05]). WT-LV-GFP n = 8, WT-LV-MMP-9 n = 9. (E) Alcohol withdrawal (W1). KO-LV-MMP-9 mice performed more reward NPs during the nonreward period (W1) when compared with KO-LV-GFP animals (two-way ANOVA: virus [F1,49 = 4.85, p < .05]). There was no difference between WT-LV-GFP and WT-LV-MMP-9 animals in the number of reward NPs during the test (two-way ANOVA for reward NPs: virus [F1,77 = 0.14, p > .05]). Day 0 indicates the day before the tests. For all data mean ± SEM is shown. The effect of the virus: **p < .01. CA, cage adaptation; FAA, free alcohol access; M, alcohol motivation test; NPA, nose poke adaptation; R, relapse; W, withdrawal test; 4–12%, increasing concentration of ethanol.

      MMP-9 KO Mice Are Resistant to Withdrawal-Induced Synaptic Plasticity in CeA

      In the following experiment we wanted to further validate the role of MMP-9 in the CeA in the regulation of functional plasticity. Because the growing body of evidence implicates dynamic changes in the generation of silent synapses (which express stable NMDAR-mediated postsynaptic responses, while AMPA receptor [AMPAR]–mediated responses are minimal) (
      • Dong Y.
      Silent synapse-based circuitry remodeling in drug addiction.
      ) in the regulation of addiction-related behaviors (
      • Boudreau A.C.
      • Wolf M.E.
      Behavioral sensitization to cocaine is associated with increased AMPA receptor surface expression in the nucleus accumbens.
      ,
      • Hanse E.
      • Seth H.
      • Riebe I.
      AMPA-silent synapses in brain development and pathology.
      ,
      • Huang Y.H.
      • Schluter O.M.
      • Dong Y.
      Silent synapses speak up: Updates of the neural rejuvenation hypothesis of drug addiction.
      ,
      • Van den Oever M.C.
      • Goriounova N.A.
      • Li K.W.
      • Van der Schors R.C.
      • Binnekade R.
      • Schoffelmeer A.N.M.
      • et al.
      Prefrontal cortex AMPA receptor plasticity is crucial for cue-induced relapse to heroin-seeking.
      ), we decided to focus on this phenomenon. During the experiment, WT and MMP-9 KO mice had free access to 10% alcohol in their home cages for 3 weeks. Control mice had access only to water. WT and MMP-9 KO mice differed only in the initial alcohol consumption; otherwise, they consumed the same amounts of alcohol (data not shown). Next, the alcohol was removed for either 1 day (1–day withdrawal group) or 7 days (7–day withdrawal group). The mice brains were used for electrophysiological whole-cell recordings. We focused on the excitatory efferent projection from the basal amygdala to CeA to discover that alcohol exposure, as well as extended withdrawal, caused the silent synapse number to increase (Figure 4B). Such changes were not observed in MMP-9 KO mice.
      Figure 4.
      Figure 4Alcohol withdrawal induces glutamatergic silent synapses in the central nucleus of the amygdala (CeA) of wild-type (WT) mice but not in matrix metalloproteinase-9 (MMP-9) knockout (KO) mice. (A) Experimental timeline. (B) Example traces for minimal stimulation protocol in WT mice, showing 5 consecutive sweeps of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptor–mediated currents recorded at –60 mV and +45 mV, respectively. The stimulation strength was reduced to obtain both responses and failures (WT water group, bright gray traces: three responses and two failures for both AMPA and NMDA currents). After prolonged withdrawal, the number of AMPA-silent synapses accumulated, which is represented by no failures for NMDA and four failures for AMPA currents in five consecutive trials (WT 7-day withdrawal [WD7] group, dark gray traces). Scale bars = 20 pA, 20 ms. The bar graph shows quantified number of silent synapses. In WT mice but not MMP-9 KO mice, alcohol consumption and extended withdrawal elevate silent synapse number in the CeA (two-way analysis of variance [ANOVA]: genotype [F1,32 = 6.65, p = .0147], alcohol exposure [F3,32 = 6.47, p = .0015]; WT water group n = 4, 1-day withdrawal [WD1] n = 6, WD7 n = 7). Two-hour presentation of alcohol cue decreases the number of silent synapses in CeA (Tukey’s post hoc test of WT WD7 vs. WT WD7+2h cue: p = .013, WD7+2h cue n = 5). MMP-9 KO mice have an elevated baseline level of silent synapses (Mann-Whitney test WT water vs. KO water p = .016), with no effect of alcohol consumption and withdrawal on this parameter (KO water group n = 5, WD1 n = 4, WD7 n = 5, WD7+2h cue n = 4). (C) Averaged example traces of AMPA receptor– and NMDAreceptor–mediated excitatory postsynaptic currents (EPSCs) (recorded at –60 and +45 mV, respectively), normalized to the amplitude of AMPA receptor traces, showing a relative increase in NMDA currents in the CeA after a prolonged alcohol withdrawal in WT mice. Scale bars = 50 pA, 25 ms. A bar graph shows the summary of AMPA/NMDA EPSC ratios over all experimental groups. Similar to silent synapse number, in only WT but not in MMP-9 KO animals was alcohol able to alter synaptic efficacy (two-way ANOVA: genotype [F1,33 = 4.645, p = .038]). Extended withdrawal from alcohol caused a decrease of AMPA/NMDA ratio, while exposure to alcohol-related cue elevated this parameter (two-way ANOVA: alcohol exposure [F3,33 = 1.495, p = .234]; Tukey’s post hoc test of WT water vs. WT WD7: p = .034; WD7 vs. WT WD7+2h cue: p = .048, WT water group n = 4, WD1 n = 6, WD7 n = 7, WD7+2h cue n = 5; KO water group n = 5, WD1 n = 4, WD7 n = 5, WD7+2h cue n = 4). *p < .05, **p < .01, ***p < .001. FAA, free alcohol access.
      To test whether the increase in silent synapse number correlates with a decrease of synaptic efficacy we measured the ratio between excitatory currents from two main glutamate receptors: AMPAR and NMDAR. Our recordings revealed that in WT mice, long-term alcohol exposure does not change the ratio of AMPAR and NMDAR currents (Figure 4C). After extended withdrawal, however, AMPAR-mediated transmission decreases in relation to NMDAR-mediated currents, which strongly correlates with the increase of silent synapse number. In mice that underwent 7-day withdrawal, 2-hour exposure to an alcohol-associated cue (an empty alcohol bottle) restores the AMPA/NMDA ratio to control levels, suggesting that a potentiation process might have happened. Similarly to silent synapse result, neither alcohol exposure nor extended withdrawal had any effect on MMP-9 KO mice. Therefore, we concluded that the lack of MMP-9 impaired alcohol-dependent synaptic plasticity in the CeA.

      rs3918242 MMP-9 Functional Gene Polymorphism in Humans Is Associated With Higher Motivation for Alcohol

      The C/T polymorphism (rs3918242) at the position –1562 of the MMP-9 gene exerts a significant effect on gene transcription, with the T allele shown to produce higher gene transcriptional activity (
      • Zhang B.
      • Henney A.
      • Eriksson P.
      • Hamsten A.
      • Watkins H.
      • Ye S.
      Genetic variation at the matrix metalloproteinase-9 locus on chromosome 20q12.2-13.1.
      ). Previously, we found a significant preferential transmission of the T allele from parents to alcoholics (59%; p = .046) (
      • Samochowiec A.
      • Grzywacz A.
      • Kaczmarek L.
      • Bienkowski P.
      • Samochowiec J.
      • Mierzejewski P.
      • et al.
      Functional polymorphism of matrix metalloproteinase-9 (MMP-9) gene in alcohol dependence: Family and case control study.
      ). In a case-control study genotype TT and T alleles were significantly more frequent in the alcoholics than in the healthy control subjects (odds ratio = 2.6). Here we extended the studied cohort and examined 167 alcohol-addicted male subjects (33 ± 8.5 years of age) and 119 control male subjects (36 ± 14 years of age). The comparison of the alleles of the MMP-9 polymorphism confirmed that MMP-9 C/T and T/T polymorphisms were significantly more frequent in the alcohol-addicted subjects than in the healthy subjects (Table 1). Furthermore, we investigated which addiction-related behavioral traits are affected by the MMP-9 gene polymorphism. We reviewed a structured Semi-Structured Assessment for the Genetics of Alcoholism interview and focused on the information related to increased motivation to drink alcohol despite harmful effects in addicted patients. The patients carrying allele T would consume alcohol despite its harmful physical effects significantly more frequently than would patients with allele C (question E35B, Table 1). When psychological problems were considered, similarly both groups experienced emotional or psychological difficulties such as feeling jumpy or easily startled (question E37.2) and problems with thinking clearly (question E37.3). Again, addicted subjects with allele T continued to drink alcohol despite harmful effects more frequently (Table 1). At the same time T polymorphism did not affect negative physical side effects of alcohol. The analysis revealed that alcohol addicted subjects that carry allele T or C experience with the same frequency physical health problems such as drowsiness (question E7.C.3), nausea (question E7.C.4), vertigo (question E7.C.4), headaches (question E7.C.5), heart failure (question E7.C.6), and liver or stomach problems (question E35.1/E35.2).
      Table 1Comparison of Genotypes and Alleles of –1562C/T Polymorphism of MMP-9 Gene in the Whole Group of Alcoholic and Healthy Control Subjects in Reference to Motivation Assessment Based on the SSAGA Questionnaire
      GroupnGenotypesP Value
      C/CC/TT/T
      MMP-9 Polymorphism
      Matrix metalloproteinase-9 (MMP-9) polymorphism in alcohol-addicted patients. MMP-9 C/T and T/T polymorphisms are significantly more frequent in the alcohol-addicted patients than in healthy subjects (χ22 = 7.548, p = .023).
      .023
       Alcohol-addicted patients167102623
       Healthy subjects11991271
      SSAGA: Drink Despite Harmful Physical Effect?
      Motivation for alcohol in patients. Alcohol-addicted patients with the T allele have higher motivation to drink alcohol as indicated in Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) questions. The patients would consume the alcohol despite its harmful effect significantly more frequently than would patients with C allele. For the question, “Did you continue to drink alcohol despite knowing it causes physical health problems?” they would reply “yes” significantly more frequently than would patients carrying C allele only (χ22 = 8.119, p = .017). For the question, “Did you continue to drink alcohol despite knowing it causes psychological health problems?” they would reply “yes” significantly more frequently than would patients carrying C allele only (χ22 = 7.442, p = .024).
      .017
       Yes9147413
       No6346170
      SSAGA: Drink Despite Harmful Psychological Effect?
      Motivation for alcohol in patients. Alcohol-addicted patients with the T allele have higher motivation to drink alcohol as indicated in Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) questions. The patients would consume the alcohol despite its harmful effect significantly more frequently than would patients with C allele. For the question, “Did you continue to drink alcohol despite knowing it causes physical health problems?” they would reply “yes” significantly more frequently than would patients carrying C allele only (χ22 = 8.119, p = .017). For the question, “Did you continue to drink alcohol despite knowing it causes psychological health problems?” they would reply “yes” significantly more frequently than would patients carrying C allele only (χ22 = 7.442, p = .024).
      .024
       Yes12369522
       No352861
      a Matrix metalloproteinase-9 (MMP-9) polymorphism in alcohol-addicted patients. MMP-9 C/T and T/T polymorphisms are significantly more frequent in the alcohol-addicted patients than in healthy subjects (χ22 = 7.548, p = .023).
      b Motivation for alcohol in patients. Alcohol-addicted patients with the T allele have higher motivation to drink alcohol as indicated in Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) questions. The patients would consume the alcohol despite its harmful effect significantly more frequently than would patients with C allele. For the question, “Did you continue to drink alcohol despite knowing it causes physical health problems?” they would reply “yes” significantly more frequently than would patients carrying C allele only (χ22 = 8.119, p = .017). For the question, “Did you continue to drink alcohol despite knowing it causes psychological health problems?” they would reply “yes” significantly more frequently than would patients carrying C allele only (χ22 = 7.442, p = .024).

      Discussion

      In the present study we have employed animal models and human clinical data to demonstrate that extrasynaptic protease MMP-9 plays a functional role in alcohol addiction–related behaviors, specifically affecting ethanol seeking during periods of its limited access. In particular, the MMP-9 KO mice, missing the enzyme, displayed lower motivation for alcohol and impaired alcohol seeking during withdrawal. In humans, the C/T polymorphism of MMP-9 gene at position –1562, which upregulates MMP-9 expression, correlated with increased motivation for alcohol in alcoholics. Interestingly, the impaired alcohol seeking in MMP-9 KO mice was associated with decreased density and structural alterations of dendritic spines in the CeA. Furthermore, whole-cell patch-clamp analysis of the basal amygdala to CeA projections of WT mice showed that alcohol consumption, as well as withdrawal, were associated with generation of silent synapses and that this phenomenon was absent in MMP-9 KO mice. Most notably, we were able to rescue the motivation impairment by the overexpression of MMP-9 in CeA.
      To assess influence of MMP-9 levels on alcohol addiction–related behaviors, we used previously validated animal model of alcoholism in the IntelliCage (
      • Radwanska K.
      • Kaczmarek L.
      Characterization of an alcohol addiction-prone phenotype in mice.
      ,
      • Smutek M.
      • Turbasa M.
      • Sikora M.
      • Piechota M.
      • Zajdel J.
      • Przewlocki R.
      • et al.
      A model of alcohol drinking under an intermittent access schedule using group-housed mice.
      ). Owing to the complexity of addiction development, such paradigms require precise monitoring of alcohol consumption, motivation, or persistence. In our studies we traced individual differences between animals living in groups maintaining social interactions. We tested the animals for addiction-related behaviors such as alcohol consumption, motivation to obtain alcohol, and persistence in alcohol seeking. Both WT and MMP-9 KO animals consumed alcohol equally. However, when access to alcohol was restricted and direct behavior (i.e., NP) was required to access it, motivation of the WT animals was significantly higher than that of the MMP-9 KO animals, indicating that MMP-9 does not control pharmacological effects of alcohol, but rather motivation for it. This conclusion is further supported by human studies. We tested if MMP-9 expression levels can be related to the motivation for alcohol in human subjects. Our previous study implicated MMP-9 gene polymorphism (T/C at position –1562) in alcohol addiction (
      • Samochowiec A.
      • Grzywacz A.
      • Kaczmarek L.
      • Bienkowski P.
      • Samochowiec J.
      • Mierzejewski P.
      • et al.
      Functional polymorphism of matrix metalloproteinase-9 (MMP-9) gene in alcohol dependence: Family and case control study.
      ). The current results provide additional support to this finding by indicating that the allele T correlates also with higher motivation to drink alcohol despite adverse consequences within the population of alcoholics.
      A growing body of evidence implicates the CeA as an important hub for many aspects of alcohol dependence (
      • Mijakowska Z.
      • Lukasiewicz K.
      • Ziolkowska M.
      • Lipinski M.
      • Trabczynska A.
      • Matuszek Z.
      • et al.
      Autophosphorylation of alpha isoform of calcium/calmodulin-dependent kinase II regulates alcohol addiction-related behaviors.
      ,
      • Roberto M.
      • Schweitzer P.
      • Madamba S.G.
      • Stouffer D.G.
      • Parsons L.H.
      • Siggins G.R.
      Acute and chronic ethanol alter glutamatergic transmission in rat central amygdala: An in vitro and in vivo analysis.
      ,
      • Nie Z.
      • Schweitzer P.
      • Roberts A.J.
      • Madamba S.G.
      • Moore S.D.
      • Siggins G.R.
      Ethanol augments GABAergic transmission in the central amygdala via CRF1 receptors.
      ,
      • Koob G.F.
      • Volkow N.D.
      Neurocircuitry of addiction.
      ,
      • Caine S.B.
      • Heinrichs S.C.
      • Coffin V.L.
      • Koob G.F.
      Effects of the dopamine D-1 antagonist SCH 23390 microinjected into the accumbens, amygdala or striatum on cocaine self-administration in the rat.
      ,
      • McGregor A.
      • Roberts D.C.
      Dopaminergic antagonism within the nucleus accumbens or the amygdala produces differential effects on intravenous cocaine self-administration under fixed and progressive ratio schedules of reinforcement.
      ,
      • Moonat S.
      • Sakharkar A.J.
      • Zhang H.
      • Tang L.
      • Pandey S.C.
      Aberrant histone deacetylase2-mediated histone modifications and synaptic plasticity in the amygdala predisposes to anxiety and alcoholism.
      ,
      • Thorsell A.
      • Tapocik J.D.
      • Liu K.
      • Zook M.
      • Bell L.
      • Flanigan M.
      • et al.
      A novel brain penetrant NPS receptor antagonist, NCGC00185684, blocks alcohol-induced ERK-phosphorylation in the central amygdala and decreases operant alcohol self-administration in rats.
      ,
      • McBride W.J.
      Central nucleus of the amygdala and the effects of alcohol and alcohol-drinking behavior in rodents.
      ,
      • Funk C.K.
      • Koob G.F.
      A CRF(2) agonist administered into the central nucleus of the amygdala decreases ethanol self-administration in ethanol-dependent rats.
      ,
      • Möller C.
      • Wiklund L.
      • Sommer W.
      • Thorsell A.
      • Heilig M.
      Decreased experimental anxiety and voluntary ethanol consumption in rats following central but not basolateral amygdala lesions.
      ,
      • Hill S.Y.
      Trajectories of alcohol use and electrophysiological and morphological indices of brain development: Distinguishing causes from consequences.
      ,
      • Wrase J.
      • Makris N.
      • Braus D.F.
      • Mann K.
      • Smolka M.N.
      • Kennedy D.N.
      • et al.
      Amygdala volume associated with alcohol abuse relapse and craving.
      ). The CeA, in particular, has a key role in the acute reinforcing actions of drugs of abuse (
      • Koob G.F.
      • Volkow N.D.
      Neurocircuitry of addiction.
      ,
      • Caine S.B.
      • Heinrichs S.C.
      • Coffin V.L.
      • Koob G.F.
      Effects of the dopamine D-1 antagonist SCH 23390 microinjected into the accumbens, amygdala or striatum on cocaine self-administration in the rat.
      ) and alcohol motivation (
      • McGregor A.
      • Roberts D.C.
      Dopaminergic antagonism within the nucleus accumbens or the amygdala produces differential effects on intravenous cocaine self-administration under fixed and progressive ratio schedules of reinforcement.
      ,
      • Moonat S.
      • Sakharkar A.J.
      • Zhang H.
      • Tang L.
      • Pandey S.C.
      Aberrant histone deacetylase2-mediated histone modifications and synaptic plasticity in the amygdala predisposes to anxiety and alcoholism.
      ). It is also thought to play an influential role in the regulation and maintenance of alcohol drinking behaviors (
      • Thorsell A.
      • Tapocik J.D.
      • Liu K.
      • Zook M.
      • Bell L.
      • Flanigan M.
      • et al.
      A novel brain penetrant NPS receptor antagonist, NCGC00185684, blocks alcohol-induced ERK-phosphorylation in the central amygdala and decreases operant alcohol self-administration in rats.
      ,
      • McBride W.J.
      Central nucleus of the amygdala and the effects of alcohol and alcohol-drinking behavior in rodents.
      ,
      • Funk C.K.
      • Koob G.F.
      A CRF(2) agonist administered into the central nucleus of the amygdala decreases ethanol self-administration in ethanol-dependent rats.
      ). Notably, lesions of the CeA block oral self-administration of alcohol (
      • Möller C.
      • Wiklund L.
      • Sommer W.
      • Thorsell A.
      • Heilig M.
      Decreased experimental anxiety and voluntary ethanol consumption in rats following central but not basolateral amygdala lesions.
      ). Moreover, human studies showed that the volume of the amygdala was smaller in individuals addicted to alcohol (
      • Hill S.Y.
      Trajectories of alcohol use and electrophysiological and morphological indices of brain development: Distinguishing causes from consequences.
      ,
      • Wrase J.
      • Makris N.
      • Braus D.F.
      • Mann K.
      • Smolka M.N.
      • Kennedy D.N.
      • et al.
      Amygdala volume associated with alcohol abuse relapse and craving.
      ). Similar observations were done in their adolescent and young-adult offspring, suggesting that lower amygdala volume may be a risk factor for addiction (
      • Hill S.Y.
      • De Bellis M.D.
      • Keshavan M.S.
      • Lowers L.
      • Shen S.
      • Hall J.
      • et al.
      Right amygdala volume in adolescent and young adult offspring from families at high risk for developing alcoholism.
      ). Here, we show that long-term exposure to alcohol relates to aberrant circuitry in the CeA, and this phenomenon is MMP-9 dependent. The physiological data go along with results on morphological plasticity of dendritic spines in the CeA, showing that spines of MMP-9 KO mice are less susceptible for such plasticity. Whereas highly motivated WT mice showed higher spine area of long and mushroom spines, no such phenomenon was observed in mice missing MMP-9. The involvement of MMP-9 in the modulation of dendritic spine morphology has been widely studied (
      • Wang X.
      • Bozdagi O.
      • Nikitczuk J.S.
      • Zhai Z.W.
      • Zhou Q.
      • Huntley G.W.
      Extracellular proteolysis by matrix metalloproteinase-9 drives dendritic spine enlargement and long-term potentiation coordinately.
      ,
      • Michaluk P.
      • Wawrzyniak M.
      • Alot P.
      • Szczot M.
      • Wyrembek P.
      • Mercik K.
      • et al.
      Influence of matrix metalloproteinase MMP-9 on dendritic spine morphology.
      ,
      • Kondratiuk I.
      • Łęski S.
      • Urbańska M.
      • Biecek P.
      • Devijver H.
      • Lechat B.
      • et al.
      GSK-3β and MMP-9 cooperate in the control of dendritic spine morphology.
      ). In particular, it was shown that MMP-9 is both necessary and sufficient to drive spine enlargement and synaptic potentiation (
      • Michaluk P.
      • Wawrzyniak M.
      • Alot P.
      • Szczot M.
      • Wyrembek P.
      • Mercik K.
      • et al.
      Influence of matrix metalloproteinase MMP-9 on dendritic spine morphology.
      ). We propose that the MMP-9–dependent spine remodeling in the CeA drives motivation in alcohol dependence. The importance of MMP-9 to plasticity in the amygdala has been demonstrated by Gorkiewicz et al. (
      • Gorkiewicz T.
      • Balcerzyk M.
      • Kaczmarek L.
      • Knapska E.
      Matrix metalloproteinase 9 (MMP-9) is indispensable for long term potentiation in the central and basal but not in the lateral nucleus of the amygdala.
      ), who showed that long-term potentiation in the CeA is negatively affected by MMP-9 KO. Furthermore, Knapska et al. (
      • Knapska E.
      • Lioudyno V.
      • Kiryk A.
      • Mikosz M.
      • Go T.
      • Michaluk P.
      • et al.
      Reward learning requires activity of matrix metalloproteinase-9 in the central amygdala.
      ) implicated MMP-9 in the CeA as pivotal for reward learning.
      Recently it has been observed that repeated exposure to cocaine induces generation of silent synapses, the form of plasticity that is normally associated with brain development (
      • Dong Y.
      • Nestler E.J.
      The neural rejuvenation hypothesis of cocaine addiction.
      ). Silent synapses are highly abundant in early development, as they form a reservoir of immature synapses, ready for activation via experience-dependent plasticity (
      • Isaac J.T.R.
      • Nicoll R.A.
      • Malenka R.C.
      Evidence for silent synapses: Implications for the expression of LTP.
      ,
      • Liao D.
      • Hessler N.A.
      • Malinow R.
      Activation of postsynaptically silent synapses during pairing-induced LTP in CA1 region of hippocampal slice.
      ). These nascent connections are downregulated by postnatal week 2 in the hippocampus (
      • Durand G.M.
      • Kovalchuk Y.
      • Konnerth A.
      Long-term potentiation and functional synapse induction in developing hippocampus.
      ,
      • Isaac J.T.
      • Crair M.C.
      • Nicoll R.A.
      • Malenka R.C.
      Silent synapses during development of thalamocortical inputs.
      ) and postnatal week 8 in the visual cortex (
      • Huang X.
      • Stodieck S.K.
      • Goetze B.
      • Cui L.
      • Wong M.H.
      • Wenzel C.
      • et al.
      Progressive maturation of silent synapses governs the duration of a critical period.
      ). It has been proposed that formation of new synaptic connections, or transforming existing ones into silent, would prime those synapses for enhanced addiction-related plasticity. Here, in adult mice, we discovered a threefold increase in the number of silent synapses after long-term alcohol exposure and further increase in subsequent withdrawal in WT mice. Thus, it can be suggested that generation of silent synapses in the CeA underlies both rewarding effects of alcohol and increased motivation to obtain alcohol during withdrawal. The increase in silent synapse number was associated with a change in synaptic efficacy, measured by AMPA-to-NMDA excitatory postsynaptic currents ratio. Although on one hand in our results chronic alcohol drinking did not alter the AMPA/NMDA ratio in WT mice, on the other hand long-term withdrawal led to a decrease in this parameter and the alcohol-associated cue elevated it back to baseline level. In line with these data, it has been shown that MMP-9 activity increases in the nucleus accumbens of rats presented with a cocaine-associated cue after a 10-day withdrawal period. This was related to an increase in the AMPA/NMDA ratio of neurons in the nucleus accumbens (
      • Smith A.C.W.
      • Kupchik Y.M.
      • Scofield M.D.
      • Gipson C.D.
      • Wiggins A.
      • Thomas CA
      • et al.
      Synaptic plasticity mediating cocaine relapse requires matrix metalloproteinases.
      ). It is conceivable that proteolytic activity of MMP-9 triggers a cascade leading to increased AMPARs number or function. Our results showed that cue-relapse decreases silent synapse number, which is associated with incorporation of AMPARs into the synapse. The lack of MMP-9 proteolytic activity in KO mice prevented those animals from synaptic adaptations of AMPA and NMDA transmission. The silent synapse number in KO mice, however, slightly elevated in water-drinking control subjects, did not change after alcohol exposure or subsequent withdrawal and cue relapse. This correlates well with animals’ lower motivation for alcohol in our IntelliCage addiction model and proves MMP-9 activity to be crucial for alcohol-induced synaptic plasticity in CeA.
      Finally, we tested whether MMP-9 expression levels can be related to the motivation for alcohol in human subjects. Our previous study indicated polymorphism at the MMP-9 gene (T/C at position –1562), which significantly differentiated control subjects from alcoholic subjects (
      • Samochowiec A.
      • Grzywacz A.
      • Kaczmarek L.
      • Bienkowski P.
      • Samochowiec J.
      • Mierzejewski P.
      • et al.
      Functional polymorphism of matrix metalloproteinase-9 (MMP-9) gene in alcohol dependence: Family and case control study.
      ). The current results provide additional support to this notice by indicating that the allele T correlates also with higher motivation to drink alcohol despite adverse consequences within the population of alcoholics.
      Although our results support the influence of MMP-9 on alcohol addiction–related behaviors, predominantly motivation in humans and animal models, there are limitations of our study that should be considered. In particular, the human studies were performed on a small group of subjects and thus further, more extensive research is desirable.
      To conclude, our study proposes a novel mechanism for alcohol seeking that involves MMP-9–dependent alterations of dendritic spines and generation of silent synapses in CeA.

      Acknowledgments and Disclosures

      This work was supported by a Team Grant (Grant No. Team/2016-1/6 to MS and LK) and a Pomost Grant (Grant No. Pomost/2011-4/7 to KR) from the Foundation for Polish Science and the National Science Centre (Poland) (Grant No. 2015/19/D/NZ4/03701 to AB).
      MS, AB, KR, and LK designed the experiments on animal models; MS, AB, TL, OM, and KM performed and analyzed the experiments on animal models; SL analyzed the data from dendritic spines; AG, JS, and AB designed, performed, and analyzed the experiments on human subjects; and MS, AB, KR, and LK wrote the manuscript.
      The authors report no biomedical financial interests or potential conflicts of interest.

      Appendix A. Supplementary material

      References

        • Kelley A.E.
        Memory and addiction: Shared neural circuitry and molecular mechanisms.
        Neuron. 2004; 44: 161-179
        • Bowers M.S.
        • Chen B.T.
        • Bonci A.
        AMPA receptor synaptic plasticity induced by psychostimulants: The past, present, and therapeutic future.
        Neuron. 2010; 67: 11-24
        • Lüscher C.
        • Malenka R.C.
        Drug-evoked synaptic plasticity in addiction: From molecular changes to circuit remodeling.
        Neuron. 2011; 69: 650-663
        • Pascoli V.
        • Terrier J.
        • Espallergues J.
        • Valjent E.
        • O’Connor E.C.
        • Lüscher C.
        Contrasting forms of cocaine-evoked plasticity control components of relapse.
        Nature. 2014; 509: 459-464
        • Holmes A.
        • Spanagel R.
        • Krystal J.H.
        Glutamatergic targets for new alcohol medications.
        Psychopharmacology. 2013; 229: 539-554
        • Holmes A.
        • Fitzgerald P.J.
        • MacPherson K.P.
        • DeBrouse L.
        • Colacicco G.
        • Flynn S.M.
        • et al.
        Chronic alcohol remodels prefrontal neurons and disrupts NMDAR-mediated fear extinction encoding.
        Nat Neurosci. 2012; 15: 1359-1361
        • Rao P.S.S.
        • Bell R.L.
        • Engleman E.A.
        • Sari Y.
        Targeting glutamate uptake to treat alcohol use disorders.
        Front Neurosci. 2015; 9: 144
        • Vafadari B.
        • Salamian A.
        • Kaczmarek L.
        MMP-9 in translation: From molecule to brain physiology, pathology and therapy.
        J Neurochem. 2016; 139: 91-114
        • Michaluk P.
        • Mikasova L.
        • Groc L.
        • Frischknecht R.
        • Choquet D.
        • Kaczmarek L.
        Matrix metalloproteinase-9 controls NMDA receptor surface diffusion through integrin beta1 signaling.
        J Neurosci. 2009; 29: 6007-6012
        • Gorkiewicz T.
        • Szczuraszek K.
        • Wyrembek P.
        • Michaluk P.
        • Kaczmarek L.
        • Mozrzymas J.W.
        Matrix metalloproteinase-9 reversibly affects the time course of NMDA-induced currents in cultured rat hippocampal neurons.
        Hippocampus. 2010; 20: 1105-1108
        • Szepesi Z.
        • Bijata M.
        • Ruszczycki B.
        • Kaczmarek L.
        • Wlodarczyk J.
        Matrix metalloproteinases regulate the formation of dendritic spine head protrusions during chemically induced long-term potentiation.
        PLoS One. 2013; 8: e63314
        • Huntley G.W.
        Synaptic circuit remodelling by matrix metalloproteinases in health and disease.
        Nat Rev Neurosci. 2012; 13: 743-757
        • Dityatev A.
        • Schachner M.
        • Sonderegger P.
        The dual role of the extracellular matrix in synaptic plasticity and homeostasis.
        Nat Rev Neurosci. 2010; 11: 735-746
        • Smith A.C.W.
        • Kupchik Y.M.
        • Scofield M.D.
        • Gipson C.D.
        • Wiggins A.
        • Thomas CA
        • et al.
        Synaptic plasticity mediating cocaine relapse requires matrix metalloproteinases.
        Nat Neurosci. 2014; 17: 1655-1657
        • Smith A.W.
        • Nealey K.A.
        • Wright J.W.
        • Walker B.M.
        Plasticity associated with escalated operant ethanol self-administration during acute withdrawal in ethanol-dependent rats requires intact matrix metalloproteinase systems.
        Neurobiol Learn Mem. 2011; 96: 199-206
        • Samochowiec A.
        • Grzywacz A.
        • Kaczmarek L.
        • Bienkowski P.
        • Samochowiec J.
        • Mierzejewski P.
        • et al.
        Functional polymorphism of matrix metalloproteinase-9 (MMP-9) gene in alcohol dependence: Family and case control study.
        Brain Res. 2010; 1327: 103-106
        • Radwanska K.
        • Kaczmarek L.
        Characterization of an alcohol addiction-prone phenotype in mice.
        Addict Biol. 2012; 17: 601-612
        • Vu T.H.
        • Shipley J.M.
        • Bergers G.
        • Berger J.E.
        • Helms J.A.
        • Hanahan D.
        • et al.
        MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes.
        Cell. 1998; 93: 411-422
        • Mijakowska Z.
        • Lukasiewicz K.
        • Ziolkowska M.
        • Lipinski M.
        • Trabczynska A.
        • Matuszek Z.
        • et al.
        Autophosphorylation of alpha isoform of calcium/calmodulin-dependent kinase II regulates alcohol addiction-related behaviors.
        Addict Biol. 2017; 22: 331-341
      1. Ruszczycki B, Wlodarczyk J, Kaczmarek, inventors; Nencki Institute of Experimental Biology, assignee. Method and a system for processing an image comprising dendritic spines. U.S. Patent No. US2014169647 (A1). June 19, 2014.

        • Oliphant T.E.
        Python for scientific computing.
        Comput Sci Eng. 2007; 9: 10-20
        • Perez F.
        • Granger B.E.
        IPython: A system for interactive scientific computing.
        Comput Sci Eng. 2007; 9: 21-29
        • Pedregosa F.
        • Varoquaux G.
        • Gramfort A.
        • Michel V.
        • Thirion B.
        • Grisel O.
        • et al.
        Scikit-learn: Machine learning in python.
        J Mach Learn Res. 2011; 12: 2825-2830
        • Hunter J.D.
        Matplotlib: A 2D graphics environment.
        Comput Sci Eng. 2007; 9: 90-95
        • Lee B.R.
        • Ma Y.-Y.
        • Huang Y.H.
        • Wang X.
        • Otaka M.
        • Ishikawa M.
        • et al.
        Maturation of silent synapses in amygdala-accumbens projection contributes to incubation of cocaine craving.
        Nat Neurosci. 2013; 16: 1644-1651
        • Suska A.
        • Lee B.R.
        • Huang Y.H.
        • Dong Y.
        • Schlüter O.M.
        Selective presynaptic enhancement of the prefrontal cortex to nucleus accumbens pathway by cocaine.
        Proc Natl Acad Sci U S A. 2013; 8: 713-718
        • Rice J.P.
        • Reich T.
        • Bucholz K.K.
        • Neuman R.J.
        • Fishman R.
        • Rochberg N.
        • et al.
        Comparison of direct interview and family history diagnoses of alcohol dependence.
        Alcohol Clin Exp Res. 1995; 19: 1018-1023
        • Roberto M.
        • Gilpin N.W.
        • Siggins G.R.
        The central amygdala and alcohol: Role of γ-aminobutyric acid, glutamate, and neuropeptides.
        Cold Spring Harb Perspect Med. 2012; 2: 1-14
        • Knapska E.
        • Lioudyno V.
        • Kiryk A.
        • Mikosz M.
        • Go T.
        • Michaluk P.
        • et al.
        Reward learning requires activity of matrix metalloproteinase-9 in the central amygdala.
        J Neurosci. 2013; 33: 14591-14600
        • Dong Y.
        Silent synapse-based circuitry remodeling in drug addiction.
        Int J Neuropsychopharmacol. 2016; 19 (pyv136)
        • Boudreau A.C.
        • Wolf M.E.
        Behavioral sensitization to cocaine is associated with increased AMPA receptor surface expression in the nucleus accumbens.
        J Neurosci. 2005; 25: 9144-9151
        • Hanse E.
        • Seth H.
        • Riebe I.
        AMPA-silent synapses in brain development and pathology.
        Nat Rev Neurosci. 2013; 14: 839-850
        • Huang Y.H.
        • Schluter O.M.
        • Dong Y.
        Silent synapses speak up: Updates of the neural rejuvenation hypothesis of drug addiction.
        Neuroscientist. 2015; 21: 451-459
        • Van den Oever M.C.
        • Goriounova N.A.
        • Li K.W.
        • Van der Schors R.C.
        • Binnekade R.
        • Schoffelmeer A.N.M.
        • et al.
        Prefrontal cortex AMPA receptor plasticity is crucial for cue-induced relapse to heroin-seeking.
        Nat Neurosci. 2008; 11: 1053-1058
        • Zhang B.
        • Henney A.
        • Eriksson P.
        • Hamsten A.
        • Watkins H.
        • Ye S.
        Genetic variation at the matrix metalloproteinase-9 locus on chromosome 20q12.2-13.1.
        Hum Genet. 1999; 105: 418-423
        • Smutek M.
        • Turbasa M.
        • Sikora M.
        • Piechota M.
        • Zajdel J.
        • Przewlocki R.
        • et al.
        A model of alcohol drinking under an intermittent access schedule using group-housed mice.
        PLoS One. 2014; 9: e96787
        • Roberto M.
        • Schweitzer P.
        • Madamba S.G.
        • Stouffer D.G.
        • Parsons L.H.
        • Siggins G.R.
        Acute and chronic ethanol alter glutamatergic transmission in rat central amygdala: An in vitro and in vivo analysis.
        J Neurosci. 2004; 24: 1594-1603
        • Nie Z.
        • Schweitzer P.
        • Roberts A.J.
        • Madamba S.G.
        • Moore S.D.
        • Siggins G.R.
        Ethanol augments GABAergic transmission in the central amygdala via CRF1 receptors.
        Science. 2004; 303: 1512-1514
        • Koob G.F.
        • Volkow N.D.
        Neurocircuitry of addiction.
        Neuropsychopharmacol. 2010; 35: 217-238
        • Caine S.B.
        • Heinrichs S.C.
        • Coffin V.L.
        • Koob G.F.
        Effects of the dopamine D-1 antagonist SCH 23390 microinjected into the accumbens, amygdala or striatum on cocaine self-administration in the rat.
        Brain Res. 1995; 692: 47-56
        • McGregor A.
        • Roberts D.C.
        Dopaminergic antagonism within the nucleus accumbens or the amygdala produces differential effects on intravenous cocaine self-administration under fixed and progressive ratio schedules of reinforcement.
        Brain Res. 1993; 624: 245-252
        • Moonat S.
        • Sakharkar A.J.
        • Zhang H.
        • Tang L.
        • Pandey S.C.
        Aberrant histone deacetylase2-mediated histone modifications and synaptic plasticity in the amygdala predisposes to anxiety and alcoholism.
        Biol Psychiatry. 2013; 73: 763-773
        • Thorsell A.
        • Tapocik J.D.
        • Liu K.
        • Zook M.
        • Bell L.
        • Flanigan M.
        • et al.
        A novel brain penetrant NPS receptor antagonist, NCGC00185684, blocks alcohol-induced ERK-phosphorylation in the central amygdala and decreases operant alcohol self-administration in rats.
        J Neurosci. 2013; 33: 10132-10142
        • McBride W.J.
        Central nucleus of the amygdala and the effects of alcohol and alcohol-drinking behavior in rodents.
        Pharmacol Biochem Behav. 2002; 71: 509-515
        • Funk C.K.
        • Koob G.F.
        A CRF(2) agonist administered into the central nucleus of the amygdala decreases ethanol self-administration in ethanol-dependent rats.
        Brain Res. 2007; 1155: 172-178
        • Möller C.
        • Wiklund L.
        • Sommer W.
        • Thorsell A.
        • Heilig M.
        Decreased experimental anxiety and voluntary ethanol consumption in rats following central but not basolateral amygdala lesions.
        Brain Res. 1997; 760: 94-101
        • Hill S.Y.
        Trajectories of alcohol use and electrophysiological and morphological indices of brain development: Distinguishing causes from consequences.
        Ann N Y Acad Sci. 2004; 1021: 245-259
        • Wrase J.
        • Makris N.
        • Braus D.F.
        • Mann K.
        • Smolka M.N.
        • Kennedy D.N.
        • et al.
        Amygdala volume associated with alcohol abuse relapse and craving.
        Am J Psychiatry. 2008; 165: 1179-1184
        • Hill S.Y.
        • De Bellis M.D.
        • Keshavan M.S.
        • Lowers L.
        • Shen S.
        • Hall J.
        • et al.
        Right amygdala volume in adolescent and young adult offspring from families at high risk for developing alcoholism.
        Biol Psychiatry. 2001; 49: 894-905
        • Wang X.
        • Bozdagi O.
        • Nikitczuk J.S.
        • Zhai Z.W.
        • Zhou Q.
        • Huntley G.W.
        Extracellular proteolysis by matrix metalloproteinase-9 drives dendritic spine enlargement and long-term potentiation coordinately.
        Proc Natl Acad Sci U S A. 2008; 105: 19520-19525
        • Michaluk P.
        • Wawrzyniak M.
        • Alot P.
        • Szczot M.
        • Wyrembek P.
        • Mercik K.
        • et al.
        Influence of matrix metalloproteinase MMP-9 on dendritic spine morphology.
        J Cell Sci. 2011; 124: 3369-3380
        • Kondratiuk I.
        • Łęski S.
        • Urbańska M.
        • Biecek P.
        • Devijver H.
        • Lechat B.
        • et al.
        GSK-3β and MMP-9 cooperate in the control of dendritic spine morphology.
        Mol Neurobiol. 2016; 54: 200-211
        • Gorkiewicz T.
        • Balcerzyk M.
        • Kaczmarek L.
        • Knapska E.
        Matrix metalloproteinase 9 (MMP-9) is indispensable for long term potentiation in the central and basal but not in the lateral nucleus of the amygdala.
        Front Cell Neurosci. 2015; 9: 73
        • Dong Y.
        • Nestler E.J.
        The neural rejuvenation hypothesis of cocaine addiction.
        Trends Pharmacol Sci. 2014; 35: 374-383
        • Isaac J.T.R.
        • Nicoll R.A.
        • Malenka R.C.
        Evidence for silent synapses: Implications for the expression of LTP.
        Neuron. 1995; 15: 427-434
        • Liao D.
        • Hessler N.A.
        • Malinow R.
        Activation of postsynaptically silent synapses during pairing-induced LTP in CA1 region of hippocampal slice.
        Nature. 1995; 375: 400-404
        • Durand G.M.
        • Kovalchuk Y.
        • Konnerth A.
        Long-term potentiation and functional synapse induction in developing hippocampus.
        Nature. 1996; 381: 71-75
        • Isaac J.T.
        • Crair M.C.
        • Nicoll R.A.
        • Malenka R.C.
        Silent synapses during development of thalamocortical inputs.
        Neuron. 1997; 18: 269-280
        • Huang X.
        • Stodieck S.K.
        • Goetze B.
        • Cui L.
        • Wong M.H.
        • Wenzel C.
        • et al.
        Progressive maturation of silent synapses governs the duration of a critical period.
        Proc Natl Acad Sci U S A. 2015; 112: E3131-E3140

      Linked Article

      • A Glitch in the Matrix: Aberrant Extracellular Matrix Proteolysis Contributes to Alcohol Seeking
        Biological PsychiatryVol. 81Issue 11
        • Preview
          Drug addiction has been characterized as a disorder of maladaptive synaptic plasticity. Irrespective of the mechanism of action of the drug being abused, they all produce similar behavioral endpoints: craving and vulnerability to relapse during periods of abstinence, loss of control over intake, and continued use despite negative consequences. As such, identifying shared neurobiological features across different classes of drugs may identify points of attack that can be used to develop efficacious therapeutics for the treatment of addiction.
        • Full-Text
        • PDF