Advertisement

Anxiety, Stress, and Fear Response in Mice With Reduced Endocannabinoid Levels

  • Imke Jenniches
    Affiliations
    Institute of Molecular Psychiatry, University of Bonn, Bonn, Germany.

    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Svenja Ternes
    Affiliations
    Institute of Molecular Psychiatry, University of Bonn, Bonn, Germany.

    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Onder Albayram
    Affiliations
    Institute of Molecular Psychiatry, University of Bonn, Bonn, Germany.

    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • David M. Otte
    Affiliations
    Institute of Molecular Psychiatry, University of Bonn, Bonn, Germany.

    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Karsten Bach
    Affiliations
    Institute of Molecular Psychiatry, University of Bonn, Bonn, Germany.

    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Laura Bindila
    Affiliations
    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Kerstin Michel
    Affiliations
    Institute of Molecular Psychiatry, University of Bonn, Bonn, Germany.

    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Beat Lutz
    Affiliations
    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Andras Bilkei-Gorzo
    Affiliations
    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author
  • Andreas Zimmer
    Correspondence
    Address correspondence to Prof. Dr. Andreas Zimmer, Institute of Molecular Psychiatry, University of Bonn, Sigmund-Freud-Straße 25, 53111 Bonn, Germany.
    Affiliations
    Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany.
    Search for articles by this author

      Abstract

      Background

      Disruption of the endocannabinoid system through pharmacological or genetic invalidation of cannabinoid CB1 receptors has been linked to depression in humans and depression-like behaviors in mice. The two main endogenous cannabinoids, anandamide and 2-arachidonoyl glycerol (2-AG), are produced on demand from phospholipids. The pathways and enzymes involved in endocannabinoid biosynthesis thus play a major role in regulating the activity of this system. This study investigates the role of the main 2-AG producing enzyme diacylglycerol lipase α (DAGL-α).

      Methods

      We generated and used knockout mice lacking DAGL-α (Dagla−/−) to assess the behavioral consequences of reduced endocannabinoid levels in the brain. We performed different behavior tests to determine anxiety- and depression-related behavioral changes in Dagla−/− mice. We also analyzed expression of genes related to the endocannabinoid system via real-time polymerase chain reaction and used the mitotic marker 5-bromo-2′-deoxyuridine to analyze adult neurogenesis.

      Results

      Dagla−/− animals show an 80% reduction of brain 2-AG levels but also a reduction in cortical and amygdalar anandamide. The behavioral changes induced by Dagla deletion include a reduced exploration of the central area of the open field, a maternal neglect behavior, a fear extinction deficit, increased behavioral despair, increased anxiety-related behaviors in the light/dark box, and reduced hippocampal neurogenesis. Some of these behavioral changes resemble those observed in animals lacking the CB1 receptor.

      Conclusions

      Our findings demonstrate that the deletion of Dagla adversely affects the emotional state of animals and results in enhanced anxiety, stress, and fear responses.

      Keywords

      The endocannabinoid system (ECS) is a retrograde feedback signaling system that plays an important, pro-homeostatic role in the central nervous system. On release into the synaptic cleft, the endogenous cannabinoids can bind and activate presynaptic cannabinoid receptors (
      • Di Marzo V.
      • Bifulco M.
      • De Petrocellis L.
      The endocannabinoid system and its therapeutic exploitation.
      ). Diacylglycerol lipase (DAGL) enzymes are responsible for the generation of 2-arachidonoyl glycerol (2-AG), one of the main endocannabinoids. Two isoforms, termed DAGL-α and DAGL-ß, have been described, which are encoded by the Dagla and Daglb genes, respectively. Whereas DAGL-α is expressed at high levels during neuronal development and in the adult central nervous system, DAGL-β is less abundant in the adult brain (
      • Bisogno T.
      • Howell F.
      • Williams G.
      • Minassi A.
      • Cascio M.G.
      • Ligresti A.
      • et al.
      Cloning of the first sn1-DAG lipases points to the spatial and temporal regulation of endocannabinoid signaling in the brain.
      ). During brain development, both enzymes are present on axonal tracts and—together with the cannabinoid receptor CB1—on growth cones of growing neurons, implicating an important function of the ECS in axon guidance and path-finding (
      • Bisogno T.
      • Howell F.
      • Williams G.
      • Minassi A.
      • Cascio M.G.
      • Ligresti A.
      • et al.
      Cloning of the first sn1-DAG lipases points to the spatial and temporal regulation of endocannabinoid signaling in the brain.
      ,
      • Berghuis P.
      • Rajnicek A.M.
      • Morozov Y.M.
      • Ross R.A.
      • Mulder J.
      • Urbán G.M.
      • et al.
      Hardwiring the brain: Endocannabinoids shape neuronal connectivity.
      ). In the adult brain, DAGL-α expression is restricted to the postsynaptic dendritic compartment (
      • Bisogno T.
      • Howell F.
      • Williams G.
      • Minassi A.
      • Cascio M.G.
      • Ligresti A.
      • et al.
      Cloning of the first sn1-DAG lipases points to the spatial and temporal regulation of endocannabinoid signaling in the brain.
      ). Several reports indicate a function of DAGL-α also during adult neurogenesis in the subventricular zone and hippocampus (
      • Goncalves M.B.
      • Suetterlin P.
      • Yip P.
      • Molina-Holgado F.
      • Walker D.J.
      • Oudin M.J.
      • et al.
      A diacylglycerol lipase-CB2 cannabinoid pathway regulates adult subventricular zone neurogenesis in an age-dependent manner.
      ,
      • Gao Y.
      • Vasilyev D.V.
      • Goncalves M.B.
      • Howell F.V.
      • Hobbs C.
      • Reisenberg M.
      • et al.
      Loss of retrograde endocannabinoid signaling and reduced adult neurogenesis in diacylglycerol lipase knock-out mice.
      ). DAGL-α is also the main enzyme involved in the production of 2-AG in the context of the endocannabinoid-mediated retrograde synaptic suppression, because depolarization-induced suppression of inhibition is completely absent in Dagla−/− mice (
      • Tanimura A.
      • Yamazaki M.
      • Hashimotodani Y.
      • Uchigashima M.
      • Kawata S.
      • Manube A.
      • et al.
      The endocannabinoid 2-arachidomoylglycerol produced by diacylglycerol lipase α mediates retrograde suppression of synaptic transmission.
      ). For a brief period from 2006 until 2008, rimonabant, a CB1 antagonist, was clinically used in Europe under the trade name Acomplia (Sanofi-Aventis, Paris, France) to treat obesity. However, the compound caused serious adverse psychiatric side effects including anxiety and depression (
      • Christensen R.
      • Kruse Kristensen P.
      • Bartels E.M.
      • Bliddal H.
      • Astrup A.
      Effiacy and safety of of the weight-loss drug rimonabant: A meta-analysis of randomised trials.
      ), which led to a withdrawal from the market. These clinical findings validated previous results from rodent studies showing that the genetic or pharmacological blockade of CB1 increases depression- and anxiety-like behavior (
      • Valverde O.
      • Karsak M.
      • Zimmer A.
      Analysis of the endocannabinoid system by using CB1 cannabinoid receptor knockout mice.
      ,
      • Valverde O.
      • Torrens M.
      CB1 receptor-deficient mice as a model for depression.
      ). Depression is a highly prevalent psychiatric disorder with a lifetime risk of approximately 20% and is associated with high levels of morbidity and mortality (
      • Kessler R.C.
      • Chiu W.
      • Demler O.
      • Walters E.E.
      Prevalence, severity and comorbidity of twelve-month DSM-IV disorders in the National Comorbidity Survey Replication (NCS-R).
      ). Additionally, it has been determined to be a risk factor for many illnesses, including obesity and cardiovascular and neurodegenerative diseases (
      • Rumsfeld J.S.
      • Ho M.
      Depression and cardiovascular disease: A call for recognition.
      ,
      • Swaab D.F.
      • Bao A.-M.
      • Lucassen P.J.
      The stress system in the human brain in depression an neurodegeneration.
      ,
      • Bornstein S.R.
      • Schuppenies A.
      • Wong M.-L.
      • Licinio J.
      Approaching the shared biology of obesity and depression: the stress axis as the locus of gene-environment interactions.
      ). A major environmental factor, which elicits anxiety disorders and depression, is stress (
      • Kessler R.C.
      The effects of stressful life events on depression.
      ,
      • Pittenger C.
      • Duman R.S.
      Stress, depression, and neuroplasticity: A convergence of mechanisms.
      ). Repeated homotypic stress leads to an elevation of 2-AG levels in cortical brain regions, which is supposedly mediated through an increased synthesis of 2-AG (
      • Patel S.
      • Hillard C.J.
      Adaptations in endocannabinoid signaling in response to repeated homotypic stress: A novel mechanism for stress habituation.
      ). These mechanisms are thought to be involved in the pathophysiology of depression and fear extinction. In this study, we analyzed the behavioral phenotype of mice with a disruption of Dagla and report the development of increased stress- and depression-related behaviors in Dagla−/− mice. Therefore, we generated Dagla−/− mice via the Cre/loxP System, which allows us to generate constitutive as well as different cell type-specific Dagla−/− mice.

      METHODS AND MATERIALS

      Animals

      The Dagla gene was mutated by inserting two loxP sites (ENSMUSG00000035735 flanking exon 1 allowing Cre-mediated excision of the exon, as depicted in Figure 1A. The recombinant and wild-type alleles were identified by means of Southern analysis after Stu I digestion with a specific probe (primer_fw: 5′-AGGAGGATTGGCCTCTGTT-3′; primer_rev: 5′-TGGACGTCGTGACTTATGGA-3′) as 1.9 kb (mutant) or 3 kb (wild-type) fragments, respectively. To obtain constitutive Dagla−/− mice, homozygous Dagla floxed mice (C57BL/6J genetic background) were bred with Pgk1-Cre mice (
      • Lallemand Y.
      • Luria V.
      • Haffner-Krausz R.
      • Lonai P.
      Maternally expressed PGK-Cre transgene as a tool for early and uniform activation oft he Cre site-specific recombinase.
      ) and subsequently to C57BL/6J animals. Male and female constitutive Dagla deficient mice (Dagla−/−, see Supplemental Information) and wild-type controls (WT) on a C57BL/6J genetic background were group-housed (4–5 animals per cage) in standard cages (light phase for 12 hours, light on at 09:00 am, except for the home cage activity measurements), with free access to food and water. Dagla−/−mice were maintained by homozygous breeding. Behavioral experiments were performed with mixed-sex groups and approved by a Local Committee for Animal Health, the North Rhine-Westphalia State Environment Agency.
      Figure thumbnail gr1
      Figure 1Cloning strategy and validation of Dagla knockout mice. (A) An 8.4-kb fragment flanking the first exon of the Dagla gene was introduced into a minimal vector by homologous recombination. One loxP site (orange) was inserted upstream of exon 1. At the 3′ end of the first exon, a neo-cassette flanked by FRT sites (purple) and with an additional loxP site was inserted into the targeting vector for selection of embryonic stem cells. Subsequently, the neo-cassette was removed by FLP-mediated recombination. To obtain constitutive Dagla−/− mice (C57BL/6J genetic background), homozygous floxed Dagla mice were cross-bred with the Pgk1-Cre mouse line. (B) mRNA levels normalized to GAPDH. Real-time PCR results show a complete loss of Dagla mRNA transcript in Dagla−/− mice (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 5 animals/group, ****p < .0001). (C) Representative DAGL-α immunostainings of (I) wild-type and (II) Dagla−/− mice hippocampi. Staining is completely absent in Dagla−/− brain (scale: 100 µm). (D) 2-AG level is significantly decreased (by 8090%) in cortex, hippocampus, striatum, and amygdala of Dagla−/− mice. AEA levels are decreased in cortex by 60%, hippocampus by 30%, and amygdala by 25% but not in the striatum of Dagla−/− mice. Note that the amount of 2-AG and AEA in the amygdala was measured in a separate experiment. Statistical analysis for eCB measurement in cortex, hippocampus, and striatum: two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 5 animals/group, *p < .05, **p < .01, ****p < .0001. Statistical analysis for eCB measurement in amygdala samples: Student t test, values represent mean ± SEM: n = 10 animals/group, *p < .05, **p < .01, ****p < .0001. AEA, anandamide; ANOVA, analysis of variance; FAAH, endocannabinoid-degrading enzyme fatty acid amide hydrolase; FLP, flippase; FRT, flippase recognition target; DAGL, diacylglycerol lipase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MAGL, monoacylglycerol lipase; PCR, polymerase chain reaction; WT, wild type; 2-AG, 2-arachidonoyl glycerol.

      Behavioral Analysis

      All behavioral experiments were performed during the light-phase at least two times with different cohorts of animals. Anxiety- and stress-related behavior was analyzed in the open-field, zero-maze, light/dark box test and in a fear conditioning paradigm. Depressive-like behavior was assessed by home cage activity, sucrose preference, and forced swim test. Furthermore, we conducted a pup retrieval test to investigate the maternal care behavior and analyzed social preference. For detailed description of the behavioral tests and the pharmacological treatment, see Supplemental Information.

      Extraction and Measurement of Endocannabinoids

      Brains were collected and immediately frozen in liquid nitrogen. For pharmacological studies, mice were intraperitoneally injected with either 20 mg/kg JZL184 or 0.9% saline 2 hours before tissue collection. Striatum, cortex, hippocampi, and amygdala samples were punched from the frozen brain slices (1.0 mm) with the use of a metal matrix for mouse brains (Zivic Instruments, Pittsburgh, Pennsylvania) and cylindrical brain punchers (Fine Science Tools; internal diameter, 1.0 mm). Extraction and quantification of basal endocannabinoid (eCB) concentration was carried out as previously described (
      • Lomazzo E.
      • Bindila L.
      • Remmers F.
      • Lerner R.
      • Schwitter C.
      • Hoheisel U.
      • Lutz B.
      Therapeutic potential of inhibitors of endocannabinoid degradation for the treatment of stress-related hyperalgesia in an animal model of chronic pain.
      ). The obtained eCB values were normalized to the amount of protein.

      Immunohistochemistry

      Brains were perfused with 4% paraformaldehyde in PBS, incubated overnight with 20% sucrose in PBS at 4°C, frozen in dry ice-cooled isopentane, and stored at –80°C. Brain slices (thickness: 18 µm) were stained with anti-mouse DAGL-α antibody (DGLα-Rb-Af380, Frontier Institute Co., Ltd., Hokkaido, Japan). Slices were incubated with 5% donkey serum and 5% bovine serum albumin for 1 hour, primary antibodies overnight at 4°C (final concentration, 0.5 µg/mL) and AF488-labeled rabbit-specific secondary antibodies for 1 hour at room temperature (dilution 1:500, Life Technologies A-21206, Darmstadt, Germany). Tris-buffered saline (TBS) containing 0.3 % Triton-X100 was used as a diluent for the antibodies and TBS as a washing buffer. Stained brain slices were embedded in DAPI Fluoromount-G media (Southern Biotechnology Associates, Inc., Birmingham, Alabama).

      Data Analysis

      Data are presented as mean ± SEM. The numbers of samples are indicated in the individual figures. Statistical significance was assessed by means of Student t test, Mann-Whitney test, or two-way analysis of variance with Bonferroni post-hoc test. Body weight changes were assessed with the use of a two-way repeated-measurement analysis of variance. Significance level was set at p < .05.
      For detailed information on measurement of neuronal proliferation, quantitative real-time polymerase chain reaction, and behavioral experiments see Supplemental Information.

      Results

      Generation of Dagla Knockout Mice

      Mice lacking exon 1 of the Dagla gene were generated through standard gene targeting (Figure 1A). Loss of the Dagla transcript was confirmed by use of TaqMan gene expression analysis with the use of a probe that spans the exon junction between exon 8 and 9. The transcript is completely absent in brain tissue of Dagla−/− mice (t = 7.428, p < .0001; Figure 1B). The deletion of DAGL-α protein was validated by immunohistochemistry staining of cryofixed brain slices. Staining was completely lost in Dagla−/− tissue (Figure 1C). We also found an 80% to 90% reduction of 2-AG levels in cortex, hippocampus, striatum (genotype effect: F1,23 = 124.4, p < .0001), and amygdala (t = 17.69, p < .0001) of Dagla−/− mice (Figure 1D). Furthermore, we discovered a significantly reduced level of anandamide in cortex, hippocampus (genotype effect: F1,23 = 23.63, p < .0001), and amygdala (t = 3.308, p < .01) of Dagla−/− mice (Figure 1D). Transcript levels of genes related to the ECS were unchanged.

      Dagla−/− Mice Have a Distinct Deficit in Maternal Care and Reduced Body Weight

      Mice lacking Dagla are viable and fertile, but we noticed that Dagla−/− female mice neglected their offspring when their nest was disturbed, for example, after cage changes. Indeed, Dagla−/− mice showed a significantly decreased maternal care behavior in the pup retrieval test, because Dagla−/− mice needed significantly more time to recollect the pups from the opposite corner of the home cage into the nest compared to WT controls (U = 8.00, p < .05; Figure 2A). In addition, 1 of the tested Dagla−/− dams did not retrieve their pups at all and 2 dams only retrieved 1 of 3 pups during the test period of 300 seconds. On the other hand, the percentage of pups from each litter with visible milk spots did not differ between genotypes (t = .8044, p = not significant [ns]; Figure 2B). Compatible with this result, we found no significant differences in the body weight of Dagla−/− pups compared with age-matched control mice (F1,12 = .9201, p = ns; Figure 3A), which strongly supports that nursing behavior was unchanged. However, after weaning the body weight of Dagla−/− mice was significantly reduced compared with WT controls (male mice: F1,26 = 20.9, p < .001; female mice: F1,22 = 128.6, p < .0001; Figure 3B, C).
      Figure thumbnail gr2
      Figure 2Maternal care and social preference. (A) Dagla−/− mice show decreased maternal care compared with WT controls in the pup retrieval test. The latency in seconds was recorded to sniff the pups and retrieve them to the nest (Mann-Whitney test, values represent mean ± SEM: n = 7 mice/group, *p < .05). (B) However percentage of pups with milk spots per litter did not differ between genotypes (Student t test, values represent mean ± SEM: n = 5–7 litter, p = ns). (C) In addition, Dagla−/− mice show the same social preference as WT controls. The ratio between the time spent with a partner to the time spent with an empty cage did not differ between both genotypes (Student t test, values represent mean ± SEM: n = 10 animals/group, p = ns). (D) Preference for the partner mouse was significantly higher in both genotypes (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 10 animals/group, ****p < .0001). ANOVA, analysis of variance; ns, not significant; WT, wild type.
      Figure thumbnail gr3
      Figure 3Body weight of male and female Dagla−/− mice was significantly decreased compared with controls. (A) Body weight of Dagla−/− and WT controls before weaning was not different between the genotypes (two-way ANOVA with repeated measurements, Bonferroni post-hoc test, values represent mean ± SEM: n = 7 animals/group). (B) Body weight of male Dagla−/− mice and WT controls: n = 12 animals/group. (C) Body weight of female Dagla−/− and WT controls: n = 14 animals/group. Please note that the SEM was very small for some data points, thus resulting in very small error bars. **p < .01, ***p < .001, ****p < .0001. ANOVA, analysis of variance; WT, wild type.
      To determine if the deficit in maternal care reflects a general disturbance of social behavior, we performed a social preference test (Figure 2C, D) in which we analyzed the time spent with the partner mouse in relation to the time spent with the empty cage. We found no significant differences in that ratio between Dagla−/− and WT mice (t = .4321, p = ns; Figure 2C). The time spent with the gender-matched partner mouse was significantly higher compared with the empty cage in both genotypes (F1,36 = 51.27, p < .0001; Figure 2D).

      Dagla−/− Mice Show an Anxiety- and Depressive-like Phenotype

      Anxiety-like behavior and locomotion was analyzed in the open-field test, in which the time spent in the center of the open-field box is inversely correlated to the state of anxiety (Figure 4A, B). We found no genotype effect in the total distance travelled (F1,23 = 1.986, p = ns), but Dagla−/− mice spent less time exploring the center of the open-field compared with WT controls (genotype effect: F1,20 = 5.751, p < .05). These results indicate that exploratory behavior and locomotor activity was not affected by the Dagla deletion, whereas anxiety levels seemed to be slightly elevated in Dagla−/− mice. To further analyze the anxiety-related phenotype, we applied the light/dark box test (Figure 4C–G). The time spent in the dark area (F1,36 = .0000005, p = ns; Figure 4C), the distance travelled in both compartments (F1,36 = 2.830, p = ns; Figure 4D), and the first entry into the light area did not differ between genotypes (t = 0.4582, p = ns; Figure 4E). However, the number of transitions between the two areas was significantly decreased for Dagla−/− mice (t = 2.708, p < .05; Figure 4F). Furthermore, we observed a decreased number of rearings in both areas for Dagla−/− mice (genotype effect: F1,36 = 21.18, p < .0001; Figure 4G). Anxiety-like behavior was also analyzed in the zero-maze test (Figure 4H, I). Time spent in the open arms did not differ significantly between the genotypes (F1,40 = –12.80, p = ns; Figure 4H). However, we found a significant genotype effect in the distance travelled (F1,40 = 4.739, p < .05; Figure 4I), which was significantly increased in Dagla−/− mice (t = 2.939, p < .01; Figure 4I).
      Figure thumbnail gr4
      Figure 4Dagla−/− mice show increased anxiety-like behavior. (A,B) Exploratory and anxiety-related behavior was analyzed in the open-field test. (A) No significant genotype differences in total distance travelled were observed (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 11 animals/group, p = ns). (B) Dagla−/− mice spent less time exploring the center and therefore displayed an anxiety-like phenotype (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 11 animals/group, *p < .05, **p < .01). (C–G) Anxiety behavior was further analyzed in the light/dark box test. (C) Time spent in the dark area did not differ between genotypes (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 9–11 animals/group, p = ns). (D) Distance traveled in both areas did not differ between genotypes (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 9–11 animals/group, p = ns). (E) The latency to first enter the light area was similar in both genotypes (Student t test, values represent mean ± SEM: n = 9–11 animals/group, p = ns). (F) Number of transitions between both areas was significantly decreased for Dagla−/− mice (Student t test, values represent mean ± SEM: n = 9–11 animals/group, *p < .05). (G) Also, the number of rearings in both areas was significantly decreased in Dagla−/− (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 9–11 animals/group, *p < .05, **p < .01). (H,I) Anxiety-like behavior was also analyzed in the zero-maze test. (H) Distance traveled in the open arms was significantly increased in Dagla−/− mice (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 20–22 animals/group, **p < .01). (I) However, time spent in the open or dark area was not significantly changed in Dagla−/− mice (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 20–22 animals/group, p = ns). ANOVA, analysis of variance; ns, not significant; WT, wild type.
      Mice lacking Dagla showed no changes in their activity in the home cage (F1,13 = .05578, p = ns) and a similar circadian rhythm as WT mice (F95,1235 = .9226, p = ns; Figure 5A). We next evaluated despair behavior, which is often considered a symptom of depression, in the forced swim test. Antidepressant drugs decrease the immobility time in this case. The deletion of Dagla led to a significantly increased immobility time (t = 2.091, p < .05; Figure 5B), which is indicative of an elevated despair behavior. We therefore investigated also another symptom of depression, anhedonia (reduced responsiveness to pleasurable stimuli), in the sucrose preference test (Figure 5C). We observed no significant genotype effect but a clear tendency toward anhedonia (t = 1.789, p = .0822). Treatment with the antidepressant amitriptyline (10 mg/kg) significantly decreased the immobility time of Dagla−/− (vehicle vs. amitriptyline: t = 5.386, p < .0001; Figure 6C) and control mice in the forced swim test (vehicle vs. amitriptyline: t = 2.963, p < .05) and could therefore reverse the depressive-like phenotype of Dagla−/− mice. In addition, after treatment with amitriptyline there was no longer a significant difference between the genotypes detectable (t = .2239, p = ns). To restore 2-AG and anandamide (AEA) levels, we treated the animals either with the MAGL inhibitor JZL184 or the FAAH inhibitor URB597 2 hours before the forced swim test (Figure 6C). Neither JZL184 nor URB597 significantly changed the immobility time of Dagla−/− (JZL184: t = 1.122, p = ns; URB597: t = .3351, p = ns) or WT (JZL184: t = 1.468, p = ns; URB597: t = 1.560, p = ns) mice compared with vehicle controls. However, after both treatments we could not observe a significant difference between the genotypes (JZL184: t = 2.022, p = ns; URB597: t = .6697, p = ns), which was present in the vehicle-treated control group (t = 3.205, p < .01; Figure 6C). We also measured 2-AG and AEA levels in the amygdala after JZL184 treatment (Figure 6A) and found increase of 2-AG in WT (t = 5.812, p < .0001) and Dagla−/− mice (t = 9.398, p < .0001). However, even after the treatment with JZL184, Dagla−/− mice still showed a much lower 2-AG level than vehicle-treated WT mice (t = 12.14, p < .0001; Figure 6A). Thus, MAGL inhibition could not restore the 2-AG level in the amygdala of Dagla−/− mice. Furthermore, after treatment with JZL184 we did not find any differences in AEA level between the genotypes (t = .5918, p = ns; Figure 6B), which we observed in the vehicle control group (t = 3.414, p < .01; Figure 6B). Therefore MAGL inhibition normalized AEA level in the amygdala of Dagla−/− mice.
      Figure thumbnail gr5
      Figure 5Dagla−/− mice display a mild depressive-like phenotype. (A) Locomotor activity and circadian rhythm was analyzed by home cage activity measurement. Dagla−/− mice do not show any changes in circadian rhythm and overall activity (two-way ANOVA repeated measurements, Bonferroni post-hoc test, values represent mean ± SEM: n = 14 animals/group, p = ns). (B) Dagla−/− display depression-like behavior in the forced swim test. The immobility time of Dagla−/− was significantly increased compared with WT controls (Student t test, values represent mean ± SEM: n = 1823 animals/group, *p = .0431). (C) Anhedonia was assessed by sucrose preference. No significant differences between Dagla−/− mice and WT controls were observed (Student t test, values represent mean ± SEM: n = 1621 animals/group, p = .0822). ANOVA, analysis of variance; ns, not significant; WT, wild type.
      Figure thumbnail gr6
      Figure 6Effects of MAGL and FAAH inhibition and antidepresssant treatment in the forced swim test. (A) Treatment with 20 mg/kg JZL184 significantly increased 2-AG levels in the amygdala of WT (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 510 animals/group, ####p < .0001) and Dagla−/− mice (Student t test, values represent mean ± SEM: n = 510 animals/group, ****p < .0001). However, MAGL inhibition could not restore the 2-AG level in Dagla−/− mice (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 510 animals/group, ****p < .0001). (B) Treatment with JZL184 did not significantly change AEA levels in the amygdala in WT or Dagla−/− mice. However, after MAGL inhibition, no difference between the genotypes was observed (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 510 animals/group, p = ns). (C) Antidepressant treatment with 10 mg/kg amitriptyline significantly decreased immobility time of WT and Dagla−/− mice. In contrast, neither treatment with 20 mg/kg JZL184 nor with 0.5 mg/kg URB597 significantly changed the immobility of WT or Dagla−/− mice. However, after treatment with MAGL and FAAH inhibitor we could no longer observe a difference between the genotypes (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 1019 animals/group, #p < .0001, **p < .0001, ####p < .0001). AEA, anandamide; ANOVA, analysis of variance; FAAH, endocannabinoid-degrading enzyme fatty acid amide hydrolase; MAGL, monoacylglycerol lipase; WT, wild type; 2-AG, 2-arachidonoyl glycerol.
      Another hallmark of animal models of depression is a reduced adult neurogenesis. Accordingly, we analyzed adult neurogenesis in the subgranular zone of the dentate gyrus. We found a significantly decreased cell proliferation in the subgranular zone of Dagla−/− mice, as evidenced by a decreased number of BrdU-labeled neurons, on day 1 (t = 3.377, p < .05) after BrdU injection compared with age-matched WT controls (Figure 7). We also analyzed the dorsal and ventral hippocampi separately but found no significant difference between the regions (F1,4 = .005386, p = ns, Supplemental Figure S1). Furthermore, the number of surviving neurons also remained significantly decreased on day 21 (t = 4.198, p < .01) after BrdU injection (Figure 7), indicating a reduced neuronal survival rate. Thus, mice lacking Dagla show several symptoms related to depression.
      Figure thumbnail gr7
      Figure 7Reduced adult neurogenesis in Dagla−/− mice. (A) Neurogenesis in the subgranular zone (SGZ) of the hippocampus of Dagla−/− mice was significantly reduced after 1 and 21 days after BrdU injection (Student t test, values represent mean ± SEM: n = 3 animals/group, *p = .05, **p < .01). (B) Representative microphotograph of BrdU-labeled (red fluorescent) and NeuN-stained (green fluorescent) cells in the dentate gyrus of the hippocampus of Dagla−/− mice and WT controls 1 day and 21 days after BrdU injection. WT, wild type.

      Increased Fear Response and Impaired Fear Extinction of Dagla−/− Mice

      We next investigated if Dagla−/− mice have a deficit in fear extinction, which was reported for CB1−/− mice (
      • Marsicano G.
      • Wotjak C.T.
      • Azad S.C.
      • Bisogno T.
      • Rammes G.
      • Cascio M.G.
      The endogenous cannabinoid system controls extinction of aversive memories.
      ,
      • Myers K.M.
      • Davis M.
      Mechanisms of fear extinction.
      ). Dagla−/− mice showed between-session fear extinction and reached similar base line freezing levels as WT control animals after 5 days of extinction training (Figure 8A). However, they displayed a significantly increased freezing response during the first 3 extinction sessions (F1,15 = 11.30, p < .01) and an impaired within-session extinction (Figure 8B). On extinction day 1, we found a significant main effect of genotype (F1,15 = 6.268, p < .05) and an interaction between genotype and freezing behavior (F2,30 = 6.213, p < .01). Thus, in contrast to WT controls, Dagla−/− animals showed no extinction of the freezing response during the session. For the following extinction trials E2 (Figure 8C) and E3 (Figure 8D), we still found a significant main effect for genotype (E2: F1,15 = 7.885, p < .05; E3: F1,14 = 9.921, p < .01). Concerning the within-session extinction, a sustained freezing behavior was detected on persisting tone presentation (E2: p < .05 after 120 seconds until the end of the session after 180 seconds, E3: p < .01 after 120 seconds). However, on extinction trial E3, a reduction in freezing was also observed in Dagla−/− animals during the last third of the tone presentation. We further analyzed the behavioral response during the conditioning (movement/ jumping as a reaction to the foot shock). We found no significant difference in the reaction to the foot shock (amplitude of jumping response) between the genotypes (t = 1.136, p = ns; Figure 8F). In addition, Dagla−/− mice did not show increased freezing if only presented with the sound cue without a foot shock (F1,36 = .2869, p = ns; Figure 8G). To determine if altered pain sensitivity influenced the outcome of the fear extinction experiment, Dagla−/− mice were also tested in the hot plate test. As depicted in Figure 9A, the pain reaction of Dagla−/− mice was comparable to WT controls (t = 1.677, p = ns). However, we observed an increased jumping response as the second reaction of pain in mice lacking Dagla (Figure 9B).
      Figure thumbnail gr8
      Figure 8Dagla−/− mice display an increased fear response and impaired fear extinction. (A–E) Dagla−/− mice and WT controls were conditioned with a tone-foot shock pairing in the conditioned chamber. All animals were exposed to a 180-second tone in a neutral environment 1, 2, 3, and 6 days after the conditioning procedure. (A) Dagla−/− mice show increased freezing time (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 89 animals/group, **p < .001) and (B) are impaired in within-session extinction of conditioned fear, seen on extinction day 1 (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 89 animals/group, genotype effect *p < .01, interaction **p < .001). (B–D) Dagla−/− mice show increased freezing behavior on extinction days 1, 2, and 3 (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 89 animals/group, *p < .05, **p < .01). (E) On day 6 after conditioning, no difference between both genotypes was observed (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 89 animals/group, p = ns). (F) Behavioral response (jumping) during the conditioning did not significantly differ between Dagla−/− and WT mice (Student t test, values represent mean ± SEM: n = 89 animals/group, p = ns). (G) Dagla−/− mice did not show increased freezing behavior on extinction day 1 if only conditioned with the sound-cue alone (two-way ANOVA, Bonferroni post-hoc test, values represent mean ± SEM: n = 20 animals/group, p = ns). ANOVA, analysis of variance; ns, not significant; WT, wild type.
      Figure thumbnail gr9
      Figure 9Dagla−/− mice do not show increased pain sensitivity but altered pain response in the hot plate test. (A) Acute thermal nociception was analyzed in the hot plate test. The latency of the first pain reaction of Dagla−/− mice is similar to WT controls (Student t test, values represent mean ± SEM: n = 911 animals/group, p = ns). (B) Dagla−/− mice showed increased jumping response in the hot plate test as a second reaction of pain. Latency and kind of pain response for each tested animal are listed in the table. ns, not significant; WT, wild type.

      Discussion

      We demonstrate that mice with a deletion of the DAGL-α enzyme show a phenotype of increased emotional or stress-related behaviors. This is characterized by a reduced exploration of the central area of the open field, a maternal neglect behavior, a fear extinction deficit, increased behavioral despair, increased anxiety-related behaviors in the light/dark box, and reduced hippocampal neurogenesis. Some of these behavioral changes resemble those observed in animals lacking CB1. Together, these findings strongly support the notion that a disruption of endocannabiniod signaling adversely affects the emotional state of animals and results in enhanced anxiety, stress, and fearresponses.
      Dagla−/− animals show an extensive reduction (80–90%) in hippocampal, cortical, amygdalar, and striatal 2-AG levels, thus confirming that DAGL-α is the main 2-AG synthesizing enzyme in the brain. Expression analysis of other ECS-related genes revealed no changes. Interestingly, the levels of AEA were also significantly decreased in the hippocampus, amygdala and cortex, but not in the striatum, of Dagla−/− mice. This finding is in agreement with previous studies that have indicated a correlation of 2-AG and AEA production in the brain (
      • Gao Y.
      • Vasilyev D.V.
      • Goncalves M.B.
      • Howell F.V.
      • Hobbs C.
      • Reisenberg M.
      • et al.
      Loss of retrograde endocannabinoid signaling and reduced adult neurogenesis in diacylglycerol lipase knock-out mice.
      ). It seems that the activity of DAGL-α and thus the level of 2-AG influenced AEA levels in a brain region-specific manner. Although the underlying mechanism is not known, it is noteworthy that the administration of the MAGL inhibitor JZL184 to Dagla−/− mice increased not only 2-AG levels but also normalized the level of AEA (Figure 6B). It is unlikely that MAGL contributes directly to the degradation of AEA, because MAGL deficient mice have normal AEA levels and acute JZL184 treatment does not affect AEA concentrations in WT mice (
      • Schlosburg J.E.
      • Blankman J.L.
      • Long J.Z.
      • Nomura D.K.
      • Pan B.
      • Kinsey S.G.
      • et al.
      Chronic monoacylglycerol lipase blockade causes functional antagonism of the endocannabinoid system.
      ,
      • Nomura D.K.
      • Morrison B.E.
      • Blankman J.L.
      • Long J.Z.
      • Kinsey S.G.
      • Marcondes M.C.G.
      • et al.
      Endocannabinoid hydrolysis generates brain prostaglandins that promote neuroinflammation.
      ). It is more likely that the vastly reduced production of 2-AG in Dagla−/− mice leads to a decrease in AEA synthesis or an increase in AEA degradation through an indirect mechanism.
      In line with previous studies (
      • Gao Y.
      • Vasilyev D.V.
      • Goncalves M.B.
      • Howell F.V.
      • Hobbs C.
      • Reisenberg M.
      • et al.
      Loss of retrograde endocannabinoid signaling and reduced adult neurogenesis in diacylglycerol lipase knock-out mice.
      ), we also observed a significantly reduced body weight in Dagla−/− mice. However, it is still unclear whether this effect is due to decreased food intake or changes in energy metabolism. Treatment with a DAGL-α inhibitor significantly decreased the intake of a high-fat diet in mice (
      • Bisogno T.
      • Mahadevan A.
      • Coccurello R.
      • Chang J.W.
      • Allara M.
      • Chen Y.
      • et al.
      A novel fluorophosphonate inhibitor of the biosynthesis of the endocannabinoid 2-arachidonoylglycerol with potential anti-obesity effects.
      ).
      Behavioral studies performed with CB1−/− mice indicated that a disrupted CB1 signaling promotes anxiety-related behaviors (
      • Martin M.
      • Ledent C.
      • Parmentier M.
      • Maldonado R.
      • Valverde O.
      Involvement of CB1 cannabinoid receptors in emotional behaviour.
      ,
      • Haller J.
      • Varga B.
      • Ledent C.
      • Barna I.
      • Freund T.F.
      Context-dependent effects of CB1 cannabinoid gene disruption on anxiety-like and social behaviour in mice.
      ,
      • Urigüen L.
      • Pérez-Rial S.
      • Ledent C.
      • Palomo T.
      • Manzanares J.
      Impaired action of anxiolytic drugs in mice deficient in cannabinoid CB1 receptors.
      ). The opposite was reported when endocannabinoid levels were increased. Inhibition of the 2-AG degrading enzyme MAGL (
      • Busquets-Garcia A.
      • Pulghermanal E.
      • Pastor A.
      • de la Torre R.
      • Maldonado R.
      • Ozaita A.
      Differential role of anandamide and 2-arachidonoylglycerol in memory and anxiety-like responses.
      ,
      • Sciolino N.R.
      • Zhou W.
      • Hohmann A.G.
      Enhancement of endocannabinoid signaling with JZL184, an inhibitor of the 2-arachidonoylglycerol hydrolyzing enzyme monoacylglycerol lipase, produces anxiolytic effects under conditions of high environmental aversiveness in rats.
      ), or the pharmacologic blockade of the AEA degrading enzyme FAAH, led to reduced anxiety-related behavior. This effect was antagonized by the CB1 antagonist rimonabant, demonstrating an involvement of the CB1 receptor (
      • Moreira F.A.
      • Kaiser N.
      • Monory K.
      • Lutz B.
      Reduced anxiety-like behaviour induced by genetic and pharmacological inhibition of the endocannabinoid-degrading enzyme fatty acid amide hydrolase (FAAH) is mediated by CB1 receptors.
      ). Consistently, we observed a mild anxiety- or stress-related phenotype in Dagla−/− mice in the open-field and light/dark box tests. However, there was no genotype effect in the open arm time of the zero-maze, but an increased distance traveled. Dagla−/− mice thus also showed a behavioral change in the more aversive zero-maze test, but these changes cannot be readily related to anxiety. The increased jumping responses of Dagla−/− mice in the hot plate test also indicate an altered behavior under stressful experimental conditions. Furthermore, it is possible that the increased anxiety- or stress-related behavior in Dagla−/− mice has also contributed to the altered maternal behavior in the pup retrieval test. Dagla−/− mice needed much more time to retrieve a pup back into the nest, but the nursing behavior was unaffected and the body weight of Dagla−/− pups did not differ from WT controls. It thus appears that Dagla−/− dams have no general deficit in maternal behavior but rather behaved differently from WT dams because of the stressful test situation. This hypothesis is supported by the normal behavior of Dagla−/− mice in the social preference test, which indicates that the altered maternal behavior is not due to a generally impaired social interest of the knockout animals. Our hypothesis is also in line with previous studies showing that treatment with rimonabant or deletion of the CB1 resulted in slower pup retrieval (
      • Schechter M.
      • Pinhasov A.
      • Weller A.
      • Fride E.
      Blocking postpartum mouse dam’s CB1 receptors impairs maternal behavior as well as offspring development and their adult social-emotional behavior.
      ,
      • Schechter M.
      • Weller A.
      • Zittel P.
      • Gross M.
      • Zimmer A.
      • Pinhasov A.
      Endocannabinoid receptor deficiency affects maternal care and alters the dam’s hippocampal oxytocin receptor and brain-derived neurothrophic factor expression.
      ), whereas nursing was not affected. The authors of this study (
      • Schechter M.
      • Weller A.
      • Zittel P.
      • Gross M.
      • Zimmer A.
      • Pinhasov A.
      Endocannabinoid receptor deficiency affects maternal care and alters the dam’s hippocampal oxytocin receptor and brain-derived neurothrophic factor expression.
      ) have also suggested that the increased stress sensitivity of CB1−/− mice (
      • Martin M.
      • Ledent C.
      • Parmentier M.
      • Maldonado R.
      • Valverde O.
      Involvement of CB1 cannabinoid receptors in emotional behaviour.
      ,
      • Haller J.
      • Varga B.
      • Ledent C.
      • Barna I.
      • Freund T.F.
      Context-dependent effects of CB1 cannabinoid gene disruption on anxiety-like and social behaviour in mice.
      ) contributed to the deficit in maternal care. Nevertheless, we cannot rule out the possibility that a dysfunctional maternal behavior contributed to the anxiety-, fear-, and stress-related behavioral phenotypes.
      The anxiety- and stress-related behavioral phenotypes are entirely consistent with the observed reduction in endocannabinoid levels in the amygdala and hippocampus, two brain regions essential for normal fear and anxiety behaviors. To determine more precisely how endocannabinoid signaling modulates the neuronal circuits associated with affective behaviors will require further studies with a cell-specific Dagla deletion.
      In addition to these anxiety- and stress-related phenotypes, we also observed that the deletion of Dagla leads to depression-like behaviors. Thus, Dagla−/− mice display an enhanced behavioral despair in the forced swim test, a clear tendency toward anhedonia in the sucrose preference test, and reduced hippocampal neurogenesis but a normal sleep/wake cycle. It is not clear if these behavioral phenotypes can be entirely attributed to the reduced levels of 2-AG or if the reduced AEA levels also contributed. However, neither JZL184 nor URB597 treatment, which both increased AEA concentrations, affected the immobility time in the forced swim test. ECS signaling, in particular the activity of the CB1 receptor, is known to modulate depression-related behavior in rodents and depression in humans. Thus, CB1−/− mice are more sensitive in developing a depressive-like phenotype after chronic stress (
      • Martin M.
      • Ledent C.
      • Parmentier M.
      • Maldonado R.
      • Valverde O.
      Involvement of CB1 cannabinoid receptors in emotional behaviour.
      ,
      • Gorzalka B.B.
      • Hill M.N.
      Integration of endocannabinoid signaling into the neural network regulating stress-induced activation of the hypothalamic-pituitary-adrenal axis.
      ), and the CB1 receptor antagonist rimonabant increased depressive symptoms in humans. Increase of 2-AG via the blockade of MAGL also produced antidepressant and anxiolytic effects and enhanced adult hippocampal neurogenesis (
      • Zhong P.
      • Wang W.
      • Pan B.
      • Liu X.
      • Zhang Z.
      • Long J.Z.
      • et al.
      Monoacylglycerol lipase inhibition blocks chronic stress-induced depressive-like behaviors via activation of mTOR signaling.
      ,
      • Zhang Z.
      • Wang W.
      • Zhong P.
      • Liu S.J.
      • Long J.Z.
      • Zhao L.
      • et al.
      Blockade of 2-arachidonoylglycerol hydrolysis produces antidepressant-like effects and enhances adult hippocampal neurogenesis and synaptic plasticity.
      ). Furthermore, chronic treatment with antidepressants led to a significant increase in 2-AG levels in various brain areas, indicating that 2-AG could be important for the antidepressant effect of these drugs (
      • Smaga I.
      • Bystrowska B.
      • Gawlinski D.
      • Pomierny B.
      • Stankowicz P.
      • Filip M.
      Antidepressants and changes in concentration of endocannabinoids and N-acylethanolamines in rat brain structures.
      ). In addition, previous studies showed a significant reduction in serum 2-AG levels of female patients with major depressive disorders (
      • Hill M.N.
      • Miller G.E.
      • Ho V.
      • Gorzalka B.B.
      • Hillard C.J.
      Serum endocannabinoid content is altered in females with depressive disorders: A preliminary report.
      ,
      • Hill M.N.
      • Miller G.E.
      • Carrier E.J.
      • Gorzalka B.B.
      • Hillard C.J.
      Circulating endocannabinoids and N-acylethanolamines are differentially regulated in major depression and following exposure to social stress.
      ).
      We also investigated stress-related behaviors of Dagla−/− mice by use of the fear extinction paradigm. The levels of 2-AG and AEA rise after the first extinction training in the basolateral amygdala, a brain region critically involved in the acquisition and expression of conditioned fear (
      • Marsicano G.
      • Wotjak C.T.
      • Azad S.C.
      • Bisogno T.
      • Rammes G.
      • Cascio M.G.
      The endogenous cannabinoid system controls extinction of aversive memories.
      ,
      • Myers K.M.
      • Davis M.
      Mechanisms of fear extinction.
      ). Genetic ablation of CB1 receptors or pharmacological blockage, systemically or within the amygdala, prevents the extinction of conditioned fear responses. Fear extinction involves safety learning, which refers to the formation of a new (inhibitory) association between the conditioned stimulus with the non-appearance of the punishment. The second process involves a non-associative component of extinction, namely, the habituation to the conditioned stimulus (
      • Kamprath K.
      • Wotjak C.T.
      Nonassociative learning processes determine expression and extinction of conditioned fear in mice.
      ). CB1−/− mice show distinct deficits in both processes. In particular delayed between-session extinction, also termed long-term extinction, and incapacity in achieving within-session extinction (short-term extinction) (
      • Marsicano G.
      • Wotjak C.T.
      • Azad S.C.
      • Bisogno T.
      • Rammes G.
      • Cascio M.G.
      The endogenous cannabinoid system controls extinction of aversive memories.
      ,
      • Ruehle S.
      • Rey A.A.
      • Remmers F.
      • Lutz B.
      The endocannabinoid system in anxiety, fear memory and habituation.
      ,
      • Kamprath K.
      • Marsicano G.
      • Tang J.
      • Monory K.
      • Bisogno T.
      • Di Marzo V.
      • et al.
      Cannabinoid CB1 receptor mediates fear extinction via habituation-like processes.
      ). It is not entirely clear if and how both of these processes are modulated by the ECS. Studies focusing on repeated homotypic stress showed that 2-AG levels in cortical brain regions were elevated after repeated stressor exposure, whereas glutamate release declined gradually (
      • Rademacher D.J.
      • Meier S.E.
      • Shi L.
      • Ho W.S.
      • Jarrahian A.
      • Hillard C.J.
      Effects of acute and repeated restraint stress on endocannabinoid content in the amygdala, ventral striatum, and medial prefrontal cortex in mice.
      ,
      • Patel S.
      • Hillard C.J.
      Adaptations in endocannabinoid signaling in response to repeated homotypic stress: A novel mechanism for stress habituation.
      ). This mechanism of stress adaptation is supposedly mediated through an increased synthesis of 2-AG (
      • Patel S.
      • Hillard C.J.
      Adaptations in endocannabinoid signaling in response to repeated homotypic stress: A novel mechanism for stress habituation.
      ). Therefore, the impaired fear extinction seen in Dagla−/− mice could be due to a deficit in habituation, which was also shown in CB1−/− mice (
      • Kamprath K.
      • Marsicano G.
      • Tang J.
      • Monory K.
      • Bisogno T.
      • Di Marzo V.
      • et al.
      Cannabinoid CB1 receptor mediates fear extinction via habituation-like processes.
      ).
      The analysis of CB1−/− mice has also shown that this receptor is essential for normal brain development. It is thus difficult to tell whether the behavioral phenotypes are due to the acute disruption of 2-AG biosynthesis or rather a consequence of developmental effects. This uncertainty, which concerns all studies with genetic mouse models, may be addressed with the help of an inducible Cre recombinase (
      • Lewandoski M.
      Conditional control of gene expression.
      ,
      • Feil S.
      • Valtecheva N.
      • Feil R.
      Inducible Cre mice.
      ).
      In conclusion, our results indicate an important role of 2-AG in the modulation of stress-responses. These findings are consistent with the idea that a disrupted endocannabinoid signaling contributes to the development of affective disorders, which is supported by clinical data.

      Acknowledgments and Disclosures

      The authors thank Frank Ativie for his support and helpful comments concerning the analysis of the adult neurogenesis. This work was supported by grants from the Deutsche Forschungsgemeinschaft (DFG) to AZ (SFB645, FOR926 SP6), and to BL (FOR926, CP1). AZ is a member of the Excellence Cluster Immunosensation.
      The authors report no biomedical financial interests or potential conflicts of interest.

      Appendix A. Supplementary Materials

      References

        • Di Marzo V.
        • Bifulco M.
        • De Petrocellis L.
        The endocannabinoid system and its therapeutic exploitation.
        Nat Rev Drug Discov. 2004; 3: 771-784
        • Bisogno T.
        • Howell F.
        • Williams G.
        • Minassi A.
        • Cascio M.G.
        • Ligresti A.
        • et al.
        Cloning of the first sn1-DAG lipases points to the spatial and temporal regulation of endocannabinoid signaling in the brain.
        J Cell Biol. 2003; 163: 463-468
        • Berghuis P.
        • Rajnicek A.M.
        • Morozov Y.M.
        • Ross R.A.
        • Mulder J.
        • Urbán G.M.
        • et al.
        Hardwiring the brain: Endocannabinoids shape neuronal connectivity.
        Science. 2007; 316: 1212-1216
        • Goncalves M.B.
        • Suetterlin P.
        • Yip P.
        • Molina-Holgado F.
        • Walker D.J.
        • Oudin M.J.
        • et al.
        A diacylglycerol lipase-CB2 cannabinoid pathway regulates adult subventricular zone neurogenesis in an age-dependent manner.
        Mol Cell Neurosci. 2008; 38: 526-536
        • Gao Y.
        • Vasilyev D.V.
        • Goncalves M.B.
        • Howell F.V.
        • Hobbs C.
        • Reisenberg M.
        • et al.
        Loss of retrograde endocannabinoid signaling and reduced adult neurogenesis in diacylglycerol lipase knock-out mice.
        J Neurosci. 2010; 30: 2017-2024
        • Tanimura A.
        • Yamazaki M.
        • Hashimotodani Y.
        • Uchigashima M.
        • Kawata S.
        • Manube A.
        • et al.
        The endocannabinoid 2-arachidomoylglycerol produced by diacylglycerol lipase α mediates retrograde suppression of synaptic transmission.
        Neuron. 2010; 25: 320-327
        • Christensen R.
        • Kruse Kristensen P.
        • Bartels E.M.
        • Bliddal H.
        • Astrup A.
        Effiacy and safety of of the weight-loss drug rimonabant: A meta-analysis of randomised trials.
        Lancet. 2007; 370: 1706-1713
        • Valverde O.
        • Karsak M.
        • Zimmer A.
        Analysis of the endocannabinoid system by using CB1 cannabinoid receptor knockout mice.
        Handb Exp Pharmacol. 2005; 168: 117-142
        • Valverde O.
        • Torrens M.
        CB1 receptor-deficient mice as a model for depression.
        Neuroscience. 2012; 204: 193-206
        • Kessler R.C.
        • Chiu W.
        • Demler O.
        • Walters E.E.
        Prevalence, severity and comorbidity of twelve-month DSM-IV disorders in the National Comorbidity Survey Replication (NCS-R).
        Arch Gen Psychiatry. 2005; 62: 617-627
        • Rumsfeld J.S.
        • Ho M.
        Depression and cardiovascular disease: A call for recognition.
        Circulation. 2005; 111: 250-253
        • Swaab D.F.
        • Bao A.-M.
        • Lucassen P.J.
        The stress system in the human brain in depression an neurodegeneration.
        Ageing Res Rev. 2005; 4: 141-194
        • Bornstein S.R.
        • Schuppenies A.
        • Wong M.-L.
        • Licinio J.
        Approaching the shared biology of obesity and depression: the stress axis as the locus of gene-environment interactions.
        Mol Psychiatry. 2006; 11: 892-902
        • Kessler R.C.
        The effects of stressful life events on depression.
        Annu Rev Psychol. 1997; 48: 191-214
        • Pittenger C.
        • Duman R.S.
        Stress, depression, and neuroplasticity: A convergence of mechanisms.
        Neuropsychopharmacology. 2008; 33: 88-109
        • Patel S.
        • Hillard C.J.
        Adaptations in endocannabinoid signaling in response to repeated homotypic stress: A novel mechanism for stress habituation.
        Eur J Neurosci. 2008; 27: 2821-2829
        • Lallemand Y.
        • Luria V.
        • Haffner-Krausz R.
        • Lonai P.
        Maternally expressed PGK-Cre transgene as a tool for early and uniform activation oft he Cre site-specific recombinase.
        Transgen Res. 1998; 7: 105-112
        • Lomazzo E.
        • Bindila L.
        • Remmers F.
        • Lerner R.
        • Schwitter C.
        • Hoheisel U.
        • Lutz B.
        Therapeutic potential of inhibitors of endocannabinoid degradation for the treatment of stress-related hyperalgesia in an animal model of chronic pain.
        Neuropsychopharmacology. 2014; 40: 488-501
        • Marsicano G.
        • Wotjak C.T.
        • Azad S.C.
        • Bisogno T.
        • Rammes G.
        • Cascio M.G.
        The endogenous cannabinoid system controls extinction of aversive memories.
        Nature. 2002; 418: 530-534
        • Myers K.M.
        • Davis M.
        Mechanisms of fear extinction.
        Mol Psychiatry. 2007; 12: 120-150
        • Schlosburg J.E.
        • Blankman J.L.
        • Long J.Z.
        • Nomura D.K.
        • Pan B.
        • Kinsey S.G.
        • et al.
        Chronic monoacylglycerol lipase blockade causes functional antagonism of the endocannabinoid system.
        Nat Neurosci. 2010; 13: 1113-1119
        • Nomura D.K.
        • Morrison B.E.
        • Blankman J.L.
        • Long J.Z.
        • Kinsey S.G.
        • Marcondes M.C.G.
        • et al.
        Endocannabinoid hydrolysis generates brain prostaglandins that promote neuroinflammation.
        Science. 2011; 334: 809-813
        • Bisogno T.
        • Mahadevan A.
        • Coccurello R.
        • Chang J.W.
        • Allara M.
        • Chen Y.
        • et al.
        A novel fluorophosphonate inhibitor of the biosynthesis of the endocannabinoid 2-arachidonoylglycerol with potential anti-obesity effects.
        Br J Pharmacol. 2013; 169: 784-793
        • Martin M.
        • Ledent C.
        • Parmentier M.
        • Maldonado R.
        • Valverde O.
        Involvement of CB1 cannabinoid receptors in emotional behaviour.
        Psychopharmacology. 2002; 159: 379-387
        • Haller J.
        • Varga B.
        • Ledent C.
        • Barna I.
        • Freund T.F.
        Context-dependent effects of CB1 cannabinoid gene disruption on anxiety-like and social behaviour in mice.
        Eur J Neurosci. 2004; 19: 1906-1912
        • Urigüen L.
        • Pérez-Rial S.
        • Ledent C.
        • Palomo T.
        • Manzanares J.
        Impaired action of anxiolytic drugs in mice deficient in cannabinoid CB1 receptors.
        Neuropharmacology. 2004; 46: 966-973
        • Busquets-Garcia A.
        • Pulghermanal E.
        • Pastor A.
        • de la Torre R.
        • Maldonado R.
        • Ozaita A.
        Differential role of anandamide and 2-arachidonoylglycerol in memory and anxiety-like responses.
        Biol Psychiatry. 2011; 70: 479-486
        • Sciolino N.R.
        • Zhou W.
        • Hohmann A.G.
        Enhancement of endocannabinoid signaling with JZL184, an inhibitor of the 2-arachidonoylglycerol hydrolyzing enzyme monoacylglycerol lipase, produces anxiolytic effects under conditions of high environmental aversiveness in rats.
        Pharmacol Res. 2011; 64: 226-234
        • Moreira F.A.
        • Kaiser N.
        • Monory K.
        • Lutz B.
        Reduced anxiety-like behaviour induced by genetic and pharmacological inhibition of the endocannabinoid-degrading enzyme fatty acid amide hydrolase (FAAH) is mediated by CB1 receptors.
        Neuropharmacology. 2008; 54: 141-150
        • Schechter M.
        • Pinhasov A.
        • Weller A.
        • Fride E.
        Blocking postpartum mouse dam’s CB1 receptors impairs maternal behavior as well as offspring development and their adult social-emotional behavior.
        Behav Brain Res. 2012; 226: 481-492
        • Schechter M.
        • Weller A.
        • Zittel P.
        • Gross M.
        • Zimmer A.
        • Pinhasov A.
        Endocannabinoid receptor deficiency affects maternal care and alters the dam’s hippocampal oxytocin receptor and brain-derived neurothrophic factor expression.
        J Neuroendocrinology. 2013; 25: 898-909
        • Gorzalka B.B.
        • Hill M.N.
        Integration of endocannabinoid signaling into the neural network regulating stress-induced activation of the hypothalamic-pituitary-adrenal axis.
        Curr Top Behav Neurosci. 2009; 1: 289-306
        • Zhong P.
        • Wang W.
        • Pan B.
        • Liu X.
        • Zhang Z.
        • Long J.Z.
        • et al.
        Monoacylglycerol lipase inhibition blocks chronic stress-induced depressive-like behaviors via activation of mTOR signaling.
        Neurpsychopharmacology. 2014; 39: 1763-1776
        • Zhang Z.
        • Wang W.
        • Zhong P.
        • Liu S.J.
        • Long J.Z.
        • Zhao L.
        • et al.
        Blockade of 2-arachidonoylglycerol hydrolysis produces antidepressant-like effects and enhances adult hippocampal neurogenesis and synaptic plasticity.
        Hippocampus. 2014; 25: 16-26
        • Smaga I.
        • Bystrowska B.
        • Gawlinski D.
        • Pomierny B.
        • Stankowicz P.
        • Filip M.
        Antidepressants and changes in concentration of endocannabinoids and N-acylethanolamines in rat brain structures.
        Neurotox Res. 2014; 26: 190-206
        • Hill M.N.
        • Miller G.E.
        • Ho V.
        • Gorzalka B.B.
        • Hillard C.J.
        Serum endocannabinoid content is altered in females with depressive disorders: A preliminary report.
        Pharmacopsychiatry. 2008; 41: 48-53
        • Hill M.N.
        • Miller G.E.
        • Carrier E.J.
        • Gorzalka B.B.
        • Hillard C.J.
        Circulating endocannabinoids and N-acylethanolamines are differentially regulated in major depression and following exposure to social stress.
        Psychoneuroendocrinology. 2009; 34: 1257-1262
        • Kamprath K.
        • Wotjak C.T.
        Nonassociative learning processes determine expression and extinction of conditioned fear in mice.
        Learn Mem. 2004; 11: 770-786
        • Ruehle S.
        • Rey A.A.
        • Remmers F.
        • Lutz B.
        The endocannabinoid system in anxiety, fear memory and habituation.
        J Psychopharmacology. 2012; 26: 23-39
        • Kamprath K.
        • Marsicano G.
        • Tang J.
        • Monory K.
        • Bisogno T.
        • Di Marzo V.
        • et al.
        Cannabinoid CB1 receptor mediates fear extinction via habituation-like processes.
        J Neurosci. 2006; 26: 6677-6686
        • Rademacher D.J.
        • Meier S.E.
        • Shi L.
        • Ho W.S.
        • Jarrahian A.
        • Hillard C.J.
        Effects of acute and repeated restraint stress on endocannabinoid content in the amygdala, ventral striatum, and medial prefrontal cortex in mice.
        Neuropsychopharmacology. 2008; 54: 108-116
        • Lewandoski M.
        Conditional control of gene expression.
        Nature Rev Genetics. 2001; 2: 743-755
        • Feil S.
        • Valtecheva N.
        • Feil R.
        Inducible Cre mice.
        Methods Mol Biol. 2009; 530: 343-363

      Linked Article

      • Endocannabinoids and Stress Resilience: Is Deficiency Sufficient to Promote Vulnerability?
        Biological PsychiatryVol. 79Issue 10
        • Preview
          Over the past 2 decades, there has been rapidly growing interest in the role of the endocannabinoid (eCB) system in the regulation of stress and emotional processes. Several lines of converging evidence provide strong evidence that eCB signaling is a key player in these processes. First, genetic ablation or pharmacologic antagonism of the cannabinoid type 1 receptor (CB1R) results in exaggerated neuroendocrine and behavioral responses to acute stress (1). More so, sustained disruption of CB1R signaling produces an array of neurobiological changes consistent with alterations seen after chronic stress or in mood disorders, such as reductions in neurotrophin levels, neurogenesis and dendritic complexity, and increased levels of central neuroinflammation (2,3).
        • Full-Text
        • PDF