3.2 Patterned light stimulation reduces AP amplitude
without affecting AP latency in MCs (Procedure 4)
The effect produced by the different patterns on AP properties was
studied using the protocol depicted figure 3A. Most of the recorded
neurons were previously submitted to the voltage-clamp protocol shown in
figure 2A. Light stimulation induced a reduction in AP amplitude that
reached a significant effect only for patterns 3, 4, 6, 7 and 8 (Figure
3B-C and supplementary figure 3). The lack of effect of the other
patterns is likely due to the low statistical power that follows the
Bonferroni-Holm correction. Indeed, with the exception of pattern 1, all
patterns had a p-value lower than the α risk (0.05) before the
correction (see https://osf.io/h7g8p/ ). Moreover, exploratory
analyses for which the effect was assessed on all 50 APs showed a
significant decrease in AP amplitude for all tested patterns
(supplementary table 3). Again, the light effect was higher for patterns
that have a longer duty cycle. However, the covariation between the duty
cycle and the AP amplitude reduction was less pronounced than the
covariation between the duty cycle and the light-induced outward current
shown in figure 2. In addition, no clear relationship emerged between
step duration and AP amplitude reduction. The effect of light on AP
amplitude had a moderate covariation with the tissue temperature
increase produced by the different patterns (Figure 4B). Patterned light
did not alter AP latency (Figure 3 C, bottom, and supplementary figure
4). We further performed exploratory analysis to look for other eventual
modifications produced by light on other AP parameters. In particular,
we looked at the full width at half maximum (FWHM) and at the AHP
amplitude. Both of these parameters were reduced by patterned
stimulation (supplementary table 3). To test whether light also alters
APs during more physiological stimuli, we induced neuronal firing with
longer pulses of current (2.5 s). We used only the pattern and power
that produced the largest overall effect in previous tests (pattern 4 at
13mW). As shown in figure 5, light stimulation had a heterogeneous
action on MC population. The firing activity was mildly reduced or
unaffected in around 80% of the recorded neurons and strongly reduced
in the remaining cells. The average frequency modification was -16 ± 30
Hz (N= 15, p=0.013).
Effect of patterned light
stimulation on other neuronal types (procedures 3 and 4)
We next determined whether discontinuous light stimulation also affects
the activity of GCs, PYRs, FSIs and MSNs by using the pattern that
produced the largest effect on MCs (pattern 4 at 13mW power). While
continuous light stimulation induced an outward current in all cell
types (Ait Ouares et al. , 2019), discontinuous stimulation had
this effect only in PYRs, GCs and MC (figure 6A). Interestingly, in one
of the recorded GC, we observed a clear inward current upon light
stimulation (supplementary figure 6) suggesting that light can have an
excitatory action on a small fraction of neurons. Patterned light
reduced AP amplitude in all neuronal types, but this effect was less
pronounced than that produced by continuous stimulation (Figure 6B).
Similar to our observation for MCs, we did not find any effect of
patterned light on AP latency (supplementary figure 7).