Increased expression of the immediate-early gene arc/arg3.1 reduces AMPA receptor-mediated synaptic transmission

Abstract

Arc/Arg3.1 is an immediate-early gene whose expression levels are increased by strong synaptic activation, including synapse-strengthening activity patterns. Arc/Arg3.1 mRNA is transported to activated dendritic regions, conferring the distribution of Arc/Arg3.1 protein both temporal correlation with the inducing stimulus and spatial specificity. Here, we investigate the effect of increased Arc/Arg3.1 levels on synaptic transmission. Surprisingly, Arc/Arg3.1 reduces the amplitude of synaptic currents mediated by AMPA-type glutamate receptors (AMPARs). This effect is prevented by RNAi knockdown of Arc/Arg3.1, by deleting a region of Arc/Arg3.1 known to interact with endophilin 3 or by blocking clathrin-coated endocytosis of AMPARs. In the hippocampal slice, Arc/Arg3.1 results in removal of AMPARs composed of GluR2 and GluR3 subunits (GluR2/3). Finally, Arc/Arg3.1 expression occludes NMDAR-dependent long-term depression. Our results demonstrate that Arc/Arg3.1 reduces the number of GluR2/3 receptors leading to a decrease in AMPAR-mediated synaptic currents, consistent with a role in the homeostatic regulation of synaptic strength.

Figure 1. Recombinant Arc/Arg3.1 Expression Levels and Time Course

(A) Epifluorescent images of slices injected in a single site in CA1. Scale bar: 200 μm for both 4× and 10× objectives. The solid white line indicates the edge of the slice. The dashed lines indicate the different layers of the CA1 region: So: stratum oriens, Sp: stratum pyramidale, Sr: stratum radiatum. (B) Two-photon image of dendrites of a pyramidal cell showing the localization of Arc/Arg3.1-GFP in dendritic spines. Scale bar: 10 μm. (C) The time course of recombinant Arc/Arg3.1 (rARC) expression assessed by western blot. Times shown are hours after virus injection. (D) Relative expression levels of recombinant Arc/Arg3.1 and endogenous Arc/Arg3.1 (eARC) in three samples. The ratio for each example is indicated below the blot. (E) Box and whisker graph showing the recombinant to endogenous Arc/Arg3.1 ratios. The bars indicate the range, the box the IQR, and the line the median of the data.

Figure 2. Arc/Arg3.1 Reduces AMPAR-Mediated Current Amplitudes

(A) AMPAR- and NMDAR-mediated EPSCs from a control cell (left) and an neighboring Arc/Arg3.1-expressing cell (right), recorded simultaneously. Dotted lines indicate control cell amplitudes. Scale bar: 10 pA, 10 ms. (B and C) Arc/Arg3.1 depresses evoked AMPAR- but not NMDAR-mediated synaptic responses. Scatter plots of synaptic amplitudes mediated by AMPARs (B) and NMDARs (C) in Arc/Arg3.1-expressing cells (y axis) and neighboring control cells (x axis). In this and all subsequent scatter plots, each dot represents a simultaneous paired recording. Wilcoxon test significance is shown. Star indicates median. Diagonal line indicates unitary slope (y = x). (D) Representative traces of AMPAR- and NMDAR-mediated currents from a control cell (left) and an neighboring GFP-expressing cell (right). Scale bar: 10 pA, 10 ms. (E and F) Viral expression of GFP does not affect AMPAR- or NMDAR-mediated synaptic currents. Scatter plots comparing amplitudes of synaptic AMPAR (E)- and NMDAR (F)-mediated responses in GFP-expressing cells (y axis) and neighboring control cells (x axis). (G) Arc/Arg3.1 expression does not affect GABAR-mediated synaptic transmission. Scatter plot of the amplitude of GABAR-mediated IPSCs in Arc/Arg3.1-expressing cells (y axis) and neighboring control cells (x axis). (H and I) Amplitudes of AMPAR (H)- and NMDAR (I)-mediated currents in infected neurons normalized to neighboring control neurons for GFP- and Arc/Arg3.1-infected slices. Data are expressed as median and IQR. Mann-Whitney test significance is indicated. (J–L) Arc/Arg3.1 decreases AMPAR mini amplitude, but not mini frequency. Cumulative frequency of amplitudes for all AMPAR mEPSC events (J) (Arc/Arg3.1: dashed line, n = 2764; control: solid line, n = 3720). Kolmogorov-Smirnov test significance is indicated. Average mEPSC amplitudes (K) and frequency (L) per cell, expressed as median and IQR. Mann-Whitney test significance is indicated.

Figure 3. The Effect of Arc/Arg3.1 Is Constant Overdevelopment

(A) Western blot showing the increase of endogenous Arc/Arg3.1 (eARC) with time in culture. The same blot reprobed with tubulin antibody was used as a loading control. (B) Simultaneous paired recordings of AMPAR- and NMDAR-mediated currents from a control cell (left) and a neighboring Arc/Arg3.1-expressing cell (right) in a slice after 3 WIC. Scale bar: 10 pA, 10 ms. (C and D) Scatter plots of amplitudes of evoked responses mediated by AMPARs (C) and NMDARs (D) in Arc/Arg3.1-expressing cells (y axis) and neighboring control cells (x axis) recorded in slices after 3 WIC. (E and F) Comparison of the effect of GFP and Arc/Arg3.1 on AMPAR (E)- and NMDAR (F)- mediated currents in slices after 1 or 3 WIC. Responses of infected neurons are normalized to neighboring control neurons. Mann- Whitney significance is indicated.

Figure 4. Arc/Arg3.1 siRNA Prevents the Reduction of AMPA EPSCs

(A) Western blot of Cos-1 cells transfected with the Arc/Arg3.1 complete coding sequence and siRNA against Arc/Arg3.1 or control siRNA. The same blot reprobed with β-tubulin antibody was used as a loading control. (B and C) Amplitudes of AMPAR (B)- and NMDAR (C)-mediated currents in transfected neurons normalized to neighboring control neurons for control siRNA and Arc/Arg3.1 siRNA-transfected slices. Data are expressed as median and IQR. Mann-Whitney test significance is indicated. (D) Simultaneous paired recordings of AMPAR- and NMDAR-mediated currents from a control cell (left) and a neighboring Arc/Arg3.1 siRNA-transfected cell (right). Scale bar: 10 pA, 10 ms. (E and F) Scatter plots of amplitudes of evoked responses mediated by AMPARs (E) and NMDARs (F) in Arc/Arg3.1 siRNA-transfected cells (y axis) and neighboring control cells (x axis). (G) Simultaneous paired recordings of AMPAR- and NMDAR-mediated currents from a control cell (left) and a neighboring control siRNA-transfected cell (right). Scale bar: 10 pA, 10 ms. (H and I) Scatter plots of amplitudes of evoked responses mediated by AMPARs (H) and NMDARs (I) in control siRNA-transfected cells (y axis) and neighboring control cells (x axis).

Figure 5. The Effect of Arc/Arg3.1 Is Independent of Neuronal Activity

(A) AMPAR-mediated EPSC amplitudes from GFP-expressing or Arc/Arg3.1-expressing neurons under the conditions stated, expressed as median and IQR. NS: Wilcoxon, p > 0.05; *Wilcoxon, p < 0.05. (B) Two-photon image showing Arc/Arg3.1- GFP localization to dendritic spines in slices exposed to 200 μM DL-APV, a treatment that prevents endogenous Arc/Arg3.1 mRNA induction and localization. Scale bar: 10 μm. (C) AMPAR response amplitudes from GFP and Arc/Arg3.1-expressing neurons normalized to neighboring control neurons. Arc/Arg3.1 depresses AMPAR-mediated responses even in the presence of drugs that reduce activity levels. *p < 0.05 in Mann- Whitney test and in Kruskal-Wallis with Dunnett’s posthoc test.

Figure 6. Arc/Arg3.1 Induces AMPAR Endocytosis

(A) Simultaneous paired recordings from control cells (left) and neighboring cells expressing GFP + pep-AP2 (right top) and Arc/Arg3.1 + pep- AP2 (right bottom). Scale bar: 10 pA, 10 ms. (B and C) Scatter plots of AMPAR response amplitudes in cells coexpressing GFP (B) or Arc/Arg3.1 (C) with pep-AP2 (y axis) and neighboring control cells (x axis). (D) AMPAR response amplitudes in GFP, GFP + pep-AP2, and Arc/Arg3.1 + pep-AP2 neurons normalized to responses in neighboring control neurons. Kruskal-Wallis test significance is indicated. (E) Simultaneous paired recordings from a control cell (left) and neighboring cell expressing Arc/Arg3.1(Δ91–100). (F) Scatter plot of AMPAR response amplitudes for cells expressing Arc/Arg3.1(Δ91–100) (y axis) compared to neighboring control cells (x axis). (G) Normalized AMPAR response amplitudes in cells expressing GFP alone or Arc/Arg3.1(Δ91–100). Mann-Whitney test significance is indicated. (H) Image showing the localization of Arc/Arg3.1(Δ91–100) to dendritic spines.

Figure 7. Arc/Arg3.1 Decreases Surface GluR2/3 AMPARs

(A) Simultaneous paired recordings from control cells (left) and neighboring cells coexpressing the GluR2 c-tail (GluR2 ct.) with either GFP (right top) or Arc/Arg3.1 (right bottom). Scale bar: 10 pA, 10 ms. (B and C) Scatter plots of AMPAR response amplitudes for cells coexpressing GFP (B) or Arc/Arg3.1 (C) with GluR2 c-tail (y axis) compared to neighboring control cells (x axis). (D) Normalized AMPAR response amplitudes in cells expressing Arc/Arg3.1, GFP + GluR2 c-tail, and Arc/Arg3.1 + GluR2 c-tail. Kruskal-Wallis test significance is indicated. (E) Examples of western blots showing the total protein lane (20% of the homogenate) and the adjacent biotinylated protein lane (precipitated from the remaining 80% of the same homogenate). Examples of GluR2 and GluR1 western blots from slices expressing GFP or Arc/Arg3.1. Control experiments of the binding to the streptavidin beads in the absence of biotinylation reagent (GFP NO BIOT.), and of intracellular protein biotinylation (PICK1 and β-tubulin). (F) Median and IQR of surface to total protein ratio for GluR2, normalized to the median ratio for GFP-expressing slices. Mann-Whitney significance is indicated. (G) Same as (F) for GluR1.

Figure 8. Arc/Arg3.1 Expression Shares Properties with LTD

(A) LTD recordings in cells with increased levels of recombinant Arc/Arg3.1 (gray squares) and control cells (black circles). Average EPSC amplitude normalized to the mean baseline amplitude. Mann-Whitney significance, comparing Arc/Arg3.1 and control cells, over the last 5 min is indicated. (B) Simultaneous paired LTD recordings from neurons expressing recombinant Arc/Arg3.1 and uninfected neighboring cells. Average absolute amplitudes are shown. Mann-Whitney significance, comparing Arc/Arg3.1 and adjacent control cells, is indicated for the baseline and last 5 min. (C and G) Simultaneous paired recordings of AMPAR-mediated currents from a control cell (left) and a neighboring Arc/Arg3.1-infected cell (right), in slices treated with 1 μM okadaic acid (C) or 50 μM FK506 (G). Scale bar: 10 pA, 10 ms. (D and H) Scatter plots of amplitudes of evoked responses mediated by AMPARs in Arc/Arg3.1-infected cells (y axis) and neighboring control cells (x axis), in slices treated with okadaic acid (D) or FK506 (H). (E and I) Amplitudes of AMPAR-mediated currents in infected neurons normalized to neighboring control neurons for GFP-infected slices and Arc/Arg3.1-infected slices treated with okadaic acid (E) or FK506 (I). Data are expressed as median and IQR. Mann-Whitney test significance is indicated. (F and J) Images showing the localization of recombinant Arc/Arg3.1 to dendritic spines in slices treated with okadaic acid (F) or FK506 (J).

Figure 9. Model for Arc/Arg3.1 Action

The schematic illustrates the scenario in which a region of the dendrite receives a synapse-strengthening stimulus that causes different amounts of potentiation in three neighboring synapses and induces Arc/Arg3.1 mRNA expression and localization to that region. Synapse 1 is 100% potentiated, synapse 2 is not potentiated, and synapse 3 is 50% potentiated. Arc/Arg3.1 preferentially affects synapses with relatively more GluR2/3 content. Consequently, the potentiated-to-nonpotentiated synaptic-strength ratio is increased by Arc/Arg3.1-induced GluR2/3 removal, e.g., synapse 1-to-synapse 2 ratio: initially 1, becomes 2 after LTP and increases to 3 after Arc/Arg3.1 has acted. In addition, total synaptic strength for that dendritic region is homeostatically regulated (initially 6, becomes 9 after LTP, returning to 6 after Arc/Arg3.1’s action).