The Journal of Neuroscience, January 28, 2004, ():
Postsynaptic Density 95 controls AMPA Receptor Incorporation during Long-Term Potentiation and Experience-Driven Synaptic Plasticity
J. Neurosci. Ehrlich and Malinow
24: 916
Supplemental data
We used single-wavelength two-photon laser scanning microscopy to determine green fluorescence intensities as a measure for expression levels of various PSD-95--GFP constructs. We determined the average fluorescence intensity [arbitrary units (a.u.)] in perinuclear regions of somata of neurons imaged under identical conditions (parallel cultures, expression time, tissue depth, and laser intensity). We find no difference between fluorescence intensities for the expression of PSD-95--GFP with gene gun or Sindbis virus (241.3 ± 42.3 a.u., n = 20 neurons; 219.0 ± 17.5 a.u., n = 25 neurons, respectively; p = 0.60; t test), although there was a larger variability in neurons transfected with gene gun. We also quantified fluorescence intensities for mutant versus wt PSD-95--GFP to ensure that differences in AMPA-current potentiation were not caused merely by differences in expression. We find no difference in fluorescence intensities [C3,5S, 98.9 ± 7.0%, n = 16; AAAA, 103.2 ± 18.2%, n = 8; double, 105.5 ± 9.0%, n = 25 neurons; p [mt] 0.73 in all cases (t test) relative to simultaneously imaged wt PSD-95--GFP = 100%; n = 17-25 neurons per group], consistent with similar expression levels.
Files in this Data Supplement:
- Supplemental Figure 1
-
A, B, Properties of evoked NMDA currents are not altered by the expression of PSD-95--GFP. A, Superposition of evoked NMDA currents (scaled) recorded at +40 mV in the presence of 5 ?M NBQX in the ACSF. B, Latency (4.9 ± 0.2 and 5.0 ± 0.2 msec; control, n = 17; infected, n = 19), time to peak (14.5 ± 0.5
and 15.4 ± 0.5 msec) and time to half decay (61.5 ± 2.5 and 63.8 ± 2.1 msec) were not altered; not all recordings are from pairs of cells. This is consistent with no recruitment of extrasynaptic receptors to synaptic sites and no subunit switch of NMDAR by PSD-95. C, D, Spine size is altered by the expression of PSD-95--GFP, but spine density is not. Two-photon images of CA1 pyramidal neurons expressing DsR-T1 or coexpressing DsR-T1 and PSD-95--GFP from age-matched hippocampal slice cultures were analyzed using the red channel only. Red fluorescence values were corrected for leak from the green channel (see Materials and Methods). C, Spine density was not significantly different when measured on secondary dendrites of similar diameter (range, 0.8-1.9 ?m) in control or PSD-95--GFP-expressing neurons (0.59 ± 0.04 spines/?m, n = 14 segments from four cells; and 0.54 ± 0.05 spines/?m, n = 12 segments from five cells, respectively; t test). D, Spines became slightly larger when PSD-95--GFP was expressed. Relative spine size was measured as total red fluorescence from a spine and normalized for expression levels of DsR-T1 (by average integrated red fluorescence from dendritic segments). The average spine size increased by ~14% when PSD-95--GFP was expressed (0.154 ± 0.012 a.u., n = 242 spines from four neurons vs 0.175± 0.007 a.u., n = 312 spines from three neurons; control and PSD-95--GFP, respectively; KS test). This small increase was not apparent when comparing images by eye.
- Supplemental Figure 2
-
A, Mutations in PDZ1 and PDZ2 (PSD-95AAAA) abolish PDZ-ligand binding, as shown by immunoprecipitation (IP) of GFP fusion proteins from hippocampal slices. Slices were uninfected [control (Con)] or were infected with wt-PSD-95--GFP or PSD-95AAAA--GFP (AAAA) for 2 d. PSD-95-NMDAR complexes were solubilized in buffer containing 50 mM Tris, pH 8.8, 1 mM EDTA, MiniComplete Protease inhibitor, and 1% deoxycholate. Subsequently, samples were diluted in and IPs were formed in buffer containing 50 mM Tris, pH 8.8, 1 mM EDTA, MiniComplete Protease inhibitor, 150 mM NaCl, and 0.1% Triton X-100 using a monoclonal anti-GFP antibody (Roche, Indianapolis, IN). Western blotting was performed with a polyclonal antibody against NR2A/B (Chemicon, Temecula, CA) at a dilution of 1:1000. Successful IP was confirmed by stripping and reprobing the blot with anti-PSD-95 antibody. B, The late component of evoked EPSC at +40 mV (NMDA component) was not altered when mutant forms of PSD-95 were expressed. Data are shown as relative changes in NMDA currents compared with control cells: C3,5S (100 ± 10.8 and 114.8 ± 10.6%; n = 27), AAAA (100 ± 15 and 103.8 ± 21.1%; n = 24), Double (100 ± 21.5 and 106.3 ± 24.5%; n = 21) for control and infected cells, respectively, in paired recordings.
- Supplemental Figure 3
-
Coexpression of PSD-95--GFP and DsRed for 2 d using biolistics resulted in potentiation of AMPAR-mediated EPSCs but did not change rectification. A, AMPA currents increased from 100 ± 11.7 to 179.5 ± 27.7% (control vs transfected; n = 28). This potentiation was not significantly different from that seen with the viral expression of PSD-95--GFP (p = 0.17; KS test). B, Rectification was not significantly different in control and transfected neurons (2.04 ± 0.15, n = 20 vs 2.44 ± 0.25, n = 17; t test).
