Increased Metabotropic Glutamate Receptor 5 Signaling Underlies Obsessive-Compulsive Disorder-like Behavioral and Striatal Circuit Abnormalities in Mice

Abstract

Background: Development of treatments for obsessive-compulsive disorder (OCD) is hampered by a lack of mechanistic understanding about this prevalent neuropsychiatric condition. Although circuit changes such as elevated frontostriatal activity are linked to OCD, the underlying molecular signaling that drives OCD-related behaviors remains largely unknown. Here, we examine the significance of type 5 metabotropic glutamate receptors (mGluR5s) for behavioral and circuit abnormalities relevant to OCD.

Methods: Sapap3 knockout (KO) mice treated acutely with an mGluR5 antagonist were evaluated for OCD-relevant phenotypes of self-grooming, anxiety-like behaviors, and increased striatal activity. The role of mGluR5 in the striatal circuit abnormalities of Sapap3 KO mice was further explored using two-photon calcium imaging to monitor striatal output from the direct and indirect pathways. A contribution of constitutive signaling to increased striatal mGluR5 activity in Sapap3 KO mice was investigated using pharmacologic and biochemical approaches. Finally, sufficiency of mGluR5 to drive OCD-like behavior in wild-type mice was tested by potentiating mGluR5 with a positive allosteric modulator.

Results: Excessive mGluR5 signaling underlies OCD-like behaviors and striatal circuit abnormalities in Sapap3 KO mice. Accordingly, enhancing mGluR5 activity acutely recapitulates these behavioral phenotypes in wild-type mice. In Sapap3 KO mice, elevated mGluR5 signaling is associated with constitutively active receptors and increased and imbalanced striatal output that is acutely corrected by antagonizing striatal mGluR5.

Conclusions: These findings demonstrate a causal role for increased mGluR5 signaling in driving striatal output abnormalities and behaviors with relevance to OCD and show the tractability of acute mGluR5 inhibition to remedy circuit and behavioral abnormalities.

Keywords: Circuit; Constitutive activity; Obsessive-compulsive disorder; Positive allosteric modulator; Striatum; mGluR5.

Figure 1

Type 5 metabotropic glutamate receptor antagonism reduces obsessive-compulsive disorder–like behaviors in Sapap3 knockout (KO) mice. (A) Experimental design for evaluating the effects of MTEP (20 mg/kg, i.p.) in wild-type (WT) and Sapap3 KO mice in the open field (OF) (B–E). Injections of either MTEP or vehicle control (Veh) were given at time = 0 minutes. Drug modulation of behavior was evaluated for each genotype by comparing the 30-minute interval before (pre) and after (post) injection. WT vehicle n = 9, WT MTEP n = 14, KO vehicle n = 12, KO MTEP n = 12. (B) Grooming activity time course shows MTEP reduces basal grooming levels in Sapap3 KO mice but not WT mice. (C) Locomotor trajectories of representative WT and Sapap3 KO mice in the OF during 10-minute periods before and after injections of vehicle or MTEP. Time courses showing (D) center time and (E) total distance traveled in the OF. Time course data are presented as means ± SEMs. (F–J) Boxplots showing anxiolytic effects of MTEP in Sapap3 KO mice. For the elevated zero maze, (F) percentage of time spent in the open area and (G) locomotion (n = 17–18 per group). For the light-dark emergence, (H) latency to enter the brightly lit chamber, (I) percentage of time spent in the brightly lit chamber, and (J) locomotion (n = 15–17 per group). Boxplots showing (K) normalized change in striatal firing rate (FR) relative to preinjection period for striatal units recorded in vivo and (L) summary of units showing a change in firing rate during the postinjection period relative to the preinjection baseline with magnitudes >10%. Veh day 1, n = 201 units/4 mice; MTEP day 2, n = 231 units/4 mice; Veh day 3, n = 207 units/4 mice. Boxplots present median, upper, and lower quartiles, and upper and lower 90%. *p < .05, **p < .01. i.p., intraperitoneal.

Figure 2

Striatal projection neuron (SPN) output is increased and imbalanced in Sapap3 knockout (KO) mice. (A) Representative image showing extracellular stimulating electrode placement in acute brain slice. Fields were imaged 600–650 µm from the tip of the electrode along the path of incoming cortical afferents. The box indicates the location of a typical field of view. (B) Representative raster scans showing fluorescence of Drd1a-tdTomato transgene (left), Fura-2 (middle), and overlay with vector path used for line-scan imaging (right). Scale bar = 25 µm. (C) Kymograph showing that Fura-2 fluorescence decreases at the time of extracellular stimulation (Stim.) (blue arrow). Scale bar = 15 µm. Heat maps of direct SPN (dSPN; “X”) and indirect SPN (iSPN; “O”) event amplitudes (amp.) in the X–Y space showing representative responses to stimulation of cortical afferents (0.6 mA) in brain slices from (D) wild-type (WT) and (E) Sapap3 KO mice demonstrate that the genotype effects on firing properties were broadly distributed in space. Summaries of (F) spike probability and (G) event amplitude demonstrate that SPN-evoked firing rates are increased in Sapap3 KO mice relative to their WT littermates. Summaries of dSPN/iSPN ratios for (H) spike probability and (I) event amplitude demonstrate that the relative balance of striatal output is shifted in favor of the direct pathway in Sapap3 KO mice relative to their WT littermates. WT = 262 dSPNs/197 iSPNs, 6 slices, 3 mice; Sapap3 KO = 381 dSPNs/318 iSPNs, 9 slices, 6 mice. Data are presented as means ± SEMs.

Figure 3

Striatal type 5 metabotropic glutamate receptor signaling increases and unbalances of striatal project neuron (SPN) output in Sapap3 knockout (KO) mice. Heat maps of direct SPN (dSPN; “X”) and indirect SPN (iSPN; “O”) event amplitudes (amp.) in the X–Y space showing representative responses to stimulation (Stim.) of cortical afferents (0.6 mA) in brain slices from Sapap3 KO mice treated with either (A) vehicle (Veh) or (B) MTEP. Summaries of (C) spike probability and (D) event amplitude demonstrate that MTEP decreases the SPN-evoked firing rate in Sapap3 KO mice. Summaries of dSPN/iSPN ratios for (E) spike probability and (F) event amplitude demonstrate that MTEP reverses the striatal pathway imbalance in Sapap3 KO mice. Veh: n= 313 dSPNs/205 iSPNs, 6 slices, 3 mice; MTEP: n= 384 dSPNs/284 iSPNs, 8 slices, 4 mice. Data are presented as means ± SEMs.

Figure 4

Constitutive signaling contributes to ongoing type 5 metabotropic glutamate receptor (mGluR5) activity in Sapap3 knockout (KO) mice. (A) Representative Western immunoblots (IBs) from a littermate pair showing that Sapap3 deletion reduces co-immunoprecipitation (IP) of mGluR5 (monomers approximately 130 kDa, dimers approximately 260 kDa) with the long Homer isoform. Boxplots showing (B) significantly less mGluR5 co-immunoprecipitated with long Homer in KO relative to wild-type (WT) striatal extracts; however, input levels of (C) long Homer and (D) mGluR5 were not significantly different between WT and Sapap3 KO striatal extracts (n = 7–8 per group). (E) Representative traces and (F) summary data showing that MCPG (500 µmol/L) does not affect miniature excitatory postsynaptic current (mEPSC) frequency (left) or amplitude (right) in direct striatal projection neurons (dSPNs) in Sapap3 KO mice (n= 11 for both groups). (G) Representative traces and (H) summary data showing that MTEP significantly increases mEPSC frequency (left) but not amplitude (amp.) (right) in dSPNs in Sapap3 KO mice (n = 9 for both groups). (I) Representative traces and (J) summary data showing that MTEP (concentration 20 µmol/L) does not affect mEPSC frequency (left) or amplitude (right) in dSPNs in WT mice [vehicle (Veh), n= 7; MTEP, n= 9]. (E, G, and I) Scale bars = 25 pA, 200 ms. Boxplots present median, upper and lower quartiles, and upper and lower 90%. *p < .05, **p < .01. (K) Schematic showing that peptide corresponding to C-terminus of mGluR5 (mGluR5-CT; pink) competitively interferes with mGluR5 (black) and Homer (yellow) interactions. (L) Representative responses to 0.2-mA stimulation and (M) summary data demonstrate that mGluR5-CT reduces evoked EPSC amplitude in WT but not KO dSPNs (WT mGluR5-CT, n = 8; WT mGluR5-MU, n = 6; KO mGluR5-CT, n = 6; KO mGluR5-MU, n = 5). Scale bar = 500 pA, 20 ms. MU, mutated; Stim., stimulation.

Figure 5

Acute type 5 metabotropic glutamate receptor (mGluR5) potentiation induces obsessive-compulsive disorder–like behaviors in C57BL/6 mice. (A) Experimental design for evaluating the effects of the mGluR5 positive allosteric modulator (PAM), VU 0360172 (20 mg/kg, i.p.), or vehicle (Veh) on C57BL/6 mice (B, D, and E) and VU 0360172 (20 mg/kg, i.p.) on Grm5 knockout (KO) and wild-type (WT) littermates (C, F, and G) in the open field (OF). Injections were given at time = 0 minutes. Modulation of behavior by PAM was evaluated by comparing the 30-minute postinjection periods between groups. Grooming activity time course data show (B) PAM significantly increases basal grooming levels in C57BL/6 mice (n = 11 for both conditions) and (C) that the grooming effects of PAM depend on mGluR5 signaling (Grm5 KO, n = 7; WT, n = 11). (D) Locomotor trajectories of representative C57BL/6 mice in the OF during 10-minute periods before and after injections of Veh (top) or PAM (bottom). (E) Time course showing the percentage of time C57BL/6 mice spent in the center of the OF arena. (F) Locomotor trajectories of a representative Grm5 KO mouse (top) and WT mouse (bottom) in the OF during 10-minute periods before and after injections of PAM. (G) Time course showing the percentage of time WT and Grm5 KO mice spent in the center of the OF arena. Time course data are presented as means ± SEMs. (H–L), Boxplots showing anxiolytic effects of PAM. For the elevated zero maze, (H) time spent in the open area and (I) locomotion (n = 17–18 per group). For the light-dark emergence, (J) latency to enter the brightly lit chamber, (K) time spent in the brightly lit chamber, and (L) locomotion (n = 21–22 per group). Boxplots present median, upper and lower quartiles, and upper and lower 90%. *p < .05, **p < .01. i.p., intraperitoneal.