Age-related changes in the spatiotemporal responses to electrical stimulation in the visual cortex of rats with progressive vision loss

Abstract

The Royal College of Surgeons (RCS) rat gradually loses vision due to retinal degeneration. Previous physiological studies have depicted the progressive loss of optical responses in the visual pathway, including the primary visual cortex (V1), over the course of retinal degeneration in the RCS rat. However, little is known about how the excitability of the V1 circuit changes during over the course of the gradual loss of visual signal input from the retina. We elucidated the properties of responses to electrical stimulations directly applied to V1 at different stages of vision input loss in the RCS rat in reference to those of the Long-Evans (LE) rat, using in vivo voltage-sensitive dye imaging. The V1 neuronal network of the RCS rat exhibited an excitatory response comparable to the LE rat. The excitatory response was maintained even long after total loss of the visual signal input from the retina. However, the response time-course suggested that the suppressive response was somewhat debilitated in the RCS rat. This is the first experiment demonstrating the long-term effect of retinal degeneration on cortical activities. Our findings provide the physiological fundamentals to enhance the preclinical research of cortical prostheses with the use of the RCS rat.

Figure 1

Typical ΔF/F image sequences indicating the light-evoked responses in early- (3-week-old), middle- (10-week-old), and end- (99-week-old) stage RCS rats, and LE (9-week-old) rats. Green, red, and blue colours indicate no change, an increase, and a decrease of neural cortical activities compared to pre-stimulation levels, respectively. The leftmost column is the VSD fluorescent image showing the stained area in white. The red rectangles indicate the averaged regions (11 × 11 pixels), which were used for the time-courses (see Methods). The middle-stage RCS rats had a longer response latency than the early-stage rats, and no light-evoked response was observed in the end-stage RCS rats. The LE rats had the shortest response latency in these examples.

Figure 2

The temporal dynamics of light-evoked cortical responses. (A) The normalized time-courses of the typical examples indicated in Fig. 1 [red, early (3 weeks old); green, middle (10 weeks old); black, LE (9 weeks old)]. The vertical dashed lines indicate the response latency. The dark and bright horizontal arrows indicate the rising time and decaying time, respectively. Note that the end-stage RCS rat data were omitted since no obvious responses were evoked by the flashing-light stimulation. (B–D) Comparison of the response latency, rising time, and decaying time between all of the groups (mean ± SEM, *p < 0.05). All the temporal parameters in the RCS rats, especially older ones, tended to be longer.

Figure 3

Typical examples of ΔF/F image sequences showing the responses to electrical stimuli with the same rat indicated in Fig. 1. The stimuli were applied through electrodes inserted into the V1 area (red lines in the leftmost and second leftmost columns indicate the tips of the electrodes). The values of the second column from the left indicate the stimulus intensity. For a better view of the response rising phases, we used log-spaced timepoints. Similar responses were induced in all rats, and neural excitation was even induced in the end-stage RCS rats, although no light-evoked response was found.

Figure 4

The temporal properties of cortical responses to the electrical stimuli. (A) Typical examples of the normalized time-courses at the stimulation point indicated in Fig. 3. Red, green, blue, and black lines indicate the time-courses of the early- (3-week-old), middle- (10-week-old), and end- (99-week-old) stage RCS rats, and the LE rat, respectively. The vertical dashed lines indicate the response latency. (B) The averaged time-course for each group (mean ± SEM). The response latencies appeared to be similar in all groups. On the other hand, the response durations appeared to be different between the RCS rat groups and the LE rat group. (C,D) Comparisons of the latency and the peak-to-trough time (mean ± SEM, *p < 0.05). The RCS rat groups exhibited longer peak-to-trough time.

Figure 5

The stimulus intensity dependence of the response. (A–D) The intensity-response curves for the early- (A), middle- (B), and end-stage RCS (C), and LE (D) rat groups. The symbols and lines are the measured values and fitted lines using a modified Naka-Rushton function (Equation 1). There is a similar trend in all groups. (E–H) Comparisons of Q _th (E), n (F), Q ₅₀ (G), and R _max (H; Equation 1) in each group (mean ± SEM, p > 0.05).

Figure 6

Typical examples of the Gaussian fitting map (Equation 2) at a certain time in each rat shown in Fig. 3. The fitting areas are shown as a colour scale, which was defined as the area exceeding the mean + 5 SD before the pre-stimulus timepoint at each pixel. The top numbers indicate the post-stimulus timepoint. The blue and cyan lines indicate the long and short axes estimated by fitting_, respectively. The black contour line indicates the outline (4σ) of the fitted two-dimensional Gaussian function. The numbers embedded in each panel indicate the long axis length, short axis length, aspect ratio, and angle from the top. The responses spread with time; the LE rat response spread in a circle, while the response of the older RCS rats spread elliptically.

Figure 7

Comparison of the two-dimensional Gaussian function fitting parameters at each timepoint in each group. From the top, the long axis length, short axis length, aspect ratio, and angle are shown. The bottom row illustrates the response maps reconstructed from the averaged parameters (red, early; green, middle; blue, end; black, LE). The response was narrowly distributed, especially in the ML axis of the older RCS rat groups. Data are expressed as the mean ± SEM (*p < 0.05).

Figure 8

GABA-mediated inhibition of the electrically induced cortical response. The normalized time-courses at the stimulation points are shown for the middle-stage RCS (left) and the LE rat (right). In each figure, the solid, dashed, and dotted lines indicate the control, CGP application, and recovery condition, respectively. Overall, with CGP 46381 application, the VSD undershoot responses decreased [trough magnitude relative to the peak value in control/CGP/recovery conditions = 0.51/0.10/0.52 for RCS, and 0.41/0.05/0.30 for LE]. The peak-to-trough times in the CGP conditions were longer than the control and recovery conditions in the LE rats (control/CGP/recovery = 87/133/86 ms). On the other hand, CGP application shortened the peak-to-trough times for the RCS rats (control/CGP/recovery = 202/92/147 ms).