Metabotropic glutamate receptor 3 (mGlu3; mGluR3; GRM3) in schizophrenia: Antibody characterisation and a semi-quantitative western blot study

Abstract

BackgroundMetabotropic glutamate receptor 3 (mGlu3, mGluR3), encoded by GRM3, is a risk gene for schizophrenia and a therapeutic target. It is unclear whether expression of the receptor is altered in the disorder or related to GRM3 risk genotype. Antibodies used to date to assess mGlu3 in schizophrenia have not been well validated.ObjectiveTo characterise six commercially available anti-mGlu3 antibodies for use in human brain, and then conduct a semi-quantitative study of mGlu3 immunoreactivity in schizophrenia.MethodsAntibodies tested using Grm3−/− and Grm2−/−/3−/− mice and transfected HEK293T/17 cells. Western blotting on membrane protein isolated from superior temporal cortex of 70 patients with schizophrenia and 87 healthy comparison subjects, genotyped for GRM3 SNP rs10234440.ResultsOne (out of six) anti-mGlu3 antibodies was fully validated, a C-terminal antibody which detected monomeric (~ 100 kDa) and dimeric (~ 200 kDa) mGlu3. A second, N-terminal, antibody detected the 200 kDa band but also produced non-specific bands. Using the C-terminal antibody for western blotting in human brain, mGlu3 immunoreactivity was found to decline with age, and was affected by pH and post mortem interval. There were no differences in monomeric or dimeric mGlu3 immunoreactivity in schizophrenia or in relation to GRM3 genotype. The antibody was not suitable for immunohistochemistry.InterpretationThese data highlight the value of knockout mouse tissue for antibody validation, and the need for careful antibody characterisation. The schizophrenia data show that involvement of GRM3 in the disorder and its genetic risk architecture is not reflected in total membrane mGlu3 immunoreactivity in superior temporal cortex.

A group II metabotropic glutamate receptor 3 (mGlu3, GRM3) isoform implicated in schizophrenia interacts with canonical mGlu3 and reduces ligand binding

“…A transcript encoding a C-terminal variant isoform of mGlu3, GRM3Δ4, was previously identified ( Sartorius et al., 2006 ) and its expression found to be influenced by a schizophrenia-associated GRM3 risk SNP ( Sartorius et al., 2008 ). Here we show, using HEK293T/17 cells, that this transcript is translated, with the resulting mGlu3Δ4 protein concentrated in membranes, including the plasma membrane ( Figures 2 and 3 ), confirming earlier observations ( García-Bea et al., 2016 ; Sartorius et al., 2006 ). We further show that mGlu3Δ4 co-immunoprecipitates with mGlu3 ( Figure 4 ) and its presence decreases the abundance of canonical mGlu3 and the Bmax of the mGlu2/3 antagonist [ 3 H]LY341495 ( Figure 5 ).…”

“…Signal was concentrated in the membrane fraction ( Figure 2 , lane 1), with much weaker immunoreactivity, only clearly detectable for the dimer, in the cytosolic fraction ( Figure 2 , lane 2). The same banding pattern and predominant membrane localization for mGlu3Δ4 was seen following transfection of GRM3Δ4-V5 (data not shown), and when using a custom anti-mGlu3Δ4 antibody reported previously ( García-Bea et al., 2016 ; data not shown). The cytosolic protein GAPDH ( Figure 2 , lanes 3 and 4) and the membrane protein N-cadherin ( Figure 2 , lanes 5 and 6) were used to assess the purity of each fraction.…”

“…GPCRs are thought to function largely if not entirely as dimers rather than as monomers ( Gurevich and Gurevich, 2008 ). As expected, we detected full-length mGlu3 primarily as a dimer (as judged by molecular weight) ( Figure 5(c) ; see also García-Bea et al., 2016 ). Interestingly, mGlu3Δ4 was also detected partly as a dimer ( Figure 2 , lane 1), even though it lacks the transmembrane domains thought to be critical to GPCR dimerization ( Bulenger et al., 2005 ).…”

“… Sartorius et al. (2006) reported a band of the predicted size of the isoform in homogenates of frontal cortex, but García-Bea et al. (2016) were unable to detect it.…”

“…Western blot experiments were carried out as previously described ( García-Bea et al., 2016 ). Briefly, 1 µg total membrane protein was run on 4–20% mini-Protean polyacrylamide gel (Bio-Rad 4561095), in SDS/Tris/glycine buffer (25 mM Tris-HCl, 250 mM glycine, 0.1% SDS) at 100V for 2 h. Proteins were transferred to a PVDF (polyvinylidene difluoride) membrane (25 V overnight) and blocked with 5% skimmed milk in PBST (phosphate buffer containing 0.1% tween 20) for 40 min.…”