1. Estimating the Synaptic Current in a Multiconductance AMPA Receptor Model 
Adi Taflia, David Holcman 
Biophysical Journal 
Volume 101, Issue 4, Pages (August 2011)
DOI: /j.bpj
Copyright © 2011 Biophysical Society Terms and Conditions

2. Figure 1
Schematic representation of the synaptic cleft. The synaptic cleft geometry is approximated as a narrow cylinder of height h and the PSD is positioned at the center of the presynaptic terminal. We depicted a vesicle released at a distance r0 inside the active zone (AZ). Diffusing receptors can either bind an AMPA receptor or diffuse away.
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions

3. Figure 2
(A) The synaptic current (computed from Eq. 7) is plotted as a function of the number of glutamate molecules. We use different receptor radii, a = 1.5 nm, 1.8 nm, and 2 nm. For a receptor effective binding radius of a = 1.8 nm, saturation occurs for four vesicles. (B) Comparing the analytical result with Brownian simulations. Glutamate molecules are released from the center of the synapse and we plot the binding probability for different PSD radii. Parameters: 20 receptors are of radius 5 nm, and the binding rate is 105 s−1. The probability is computed from 400 runs. The analytical solution (solid line) is computed using Eq. 13, which uses the homogenized approximation for κ, given by Eq. 26.
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions

4. Figure 3
Geometrical properties of the synaptic current. (A) We decompose the synaptic current IS in a sum of current generated by two, three, and four bound glutamate molecules (IS = I2 + I3 + I4). In the range of 3000–9000 glutamates, the contribution of each configuration is I4 > I3 > I2. Glutamates are released from the center. (B) The synaptic current is plotted as a function of the release distance from the center of the synapse for one, two, and three vesicles. In both graphs, we used a synaptic and PSD radius of 500 nm and 300 nm, respectively, whereas the height is 30 nm.
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions

5. Figure 4
Optimal PSD radius. (A) The mean current and standard deviation are plotted as a function of the PSD size for three different active zones (50 nm,100 nm, and 150 nm). The synaptic radius is 500 nm and the height is 20 nm. (B and C) We present the mean number of AMPARs bound by two (respectively, four) glutamate molecules as a function of the PSD radius. (D) The current is plotted as a function of the active zone (AZ) radius. The PSD size is fixed at 300 nm and each curve represents, respectively, one, two, and three released vesicles. (E) CV versus the PSD size: the CV reaches its minimum when the PSD and the AZ have approximately the same size. The AZ radius is 100 nm and the CV minimum is achieved for a PSD radius of 120 nm. (F) The optimal PSD radius is plotted as a function of the AZ radius.
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions

6. Figure 5
Synaptic current for different synapse radiuses. (A) For a fixed active zone radius (50 nm), we plotted the synaptic current as a function of the PSD radius for four different sizes of synapses 200, 300, 400, and 500 nm (the height is 30 nm). Doubling the size of the synapse leads to a current amplitude that increases from 125 to ∼190 pA for a small PSD radius. (B) The CV is plotted as a function of the PSD radius. (C) The CV is plotted as a function of the PSD radius, for different numbers of released glutamate molecules. There is an optimal PSD size. (D) We compare the CV as a function of the PSD radius curve when the number of released glutamate molecules is distributed according to a Gaussian distribution with a mean 3000 and a standard deviation of 500 (bold line) with a fixed number of 3000 glutamate molecules (dashed line).
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions

7. Figure 6
We plot CV as a function of the PSD radius. (A) The CV is computed for several presynaptic release probabilities: increasing the release probability lowers the CV and this effect is much more pronounced than changing the PSD radius. (B) Relative decrease in the ratio R(p) as a function of the presynaptic release probability. The ratio R quantifies the optimal PSD radius for different release probabilities, defined in Eq. 8.
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions

8. Figure 7
Analysis of a simplified model showing discrete current levels: a random variable has three possible outcome values I1, I2, and I3. The probability function is given by Eq. 9. The model describes a single receptor with three conductivity levels, which depend on the number of bound molecules. Each glutamate particle can bind a receptor with a probability q. We plot the CV as a function of the binding probability q for different values of I3, where the values I1 = 1, I2 = 2 are fixed. Interestingly, when I3 >> I1, I2, CV has an optimal point. The optimal CV as a function of binding probability strongly correlates with the nonlinear cooperative effect of multiple bindings.
Biophysical Journal  , DOI: ( /j.bpj )
Copyright © 2011 Biophysical Society Terms and Conditions