The experiments described below were conducted in rat somatosensory cortex, either in barrel cortex in response to whisker stimulation or in forepaw cortex in response to tactile stimulation of the digits.
3.1. Intrinsic optical imaging of vibrotactile stimulation
Baseline functionality and viability of somatosensory cortex prior to the application of INS was assessed through optical imaging of hemodynamic responses in somatosensory cortex. illustrates a normal functional response to vibrotactile stimulation of D2 and D4 of the contralateral forepaw in response to taps delivered by a piezoelectric stimulator (8 Hz, 3 s). Darker pixels in the functional maps indicate activation (). Cortical activation to D4 stimulation was medial and posterior to D2 as emphasized by the dark (D4, ROI2) and light (D2, ROI1) ROIs in the subtraction map between the two conditions (). illustrate signal time courses taken from the D2 (ROI 1), D4 (ROI 2), and control (edge of craniotomy, ROI 3) locations, respectively. Optical responses to D2 stimulation, D4 stimulation, and the no stimulation conditions are plotted in green, blue, and red traces, respectively. As indicated by the larger negative deflections in intrinsic signal magnitude, the time courses of activation () demonstrated preferential activation for stimulation of D2 at the D2 site (ROI 1) and for stimulation of D4 at the D4 site (ROI 2). The intrinsic signals were not focal within the forepaw representation as D2 stimulation activated the D4 ROI and D4 stimulation activated the D2 ROI. The blank condition (no stimulation) produced little response at the D2 and D4 sites, which was comparable to the lack of signal obtained at the control site for all three conditions (). These activation maps are representative of the optical responses obtained in rat barrel cortex and rat forepaw cortex generated by tactile stimulation in our experiments.
Fig. 2 Typical intrinsic imaging response to vibrotactile stimulation of contralateral forepaw digits. (A) Blood vessel map. Black boxes are region of interests where time course data was calculated for D2, D4, and no stimulation conditions. (B & C) (more ...)
3.2. Demonstration of INS induced optical intrinsic signals
Infrared neural stimulation was then examined to determine if pulsed infrared light could induce an optical response in cerebral cortex comparable to that induced by natural sensory stimulation. demonstrates that INS of cortical tissue induces changes in optical reflectance signal of somatosensory cortex. A fiber optic was placed over somatosensory cortex corresponding to the barrel fields (). Stimulation of the cortex with INS (100 Hz, 0.55 J/cm2, λ=1.875 μm, 250 μs pulse width, 500 ms pulse train) evoked changes in optical reflectance at and near the fiber optic location, as illustrated by the activation map shown in . The activated region of cortex (dark pixels) in response to these INS parameters produced a focal region of activation, approximately 1.5 – 2 mm in diameter. As shown in , the time course reaches a peak after 1 s and has a duration of 3 s. The magnitude of the change in reflectance peaked at approximately 0.15%.. No such optical reflectance change was obtained during the Blank condition. Optical signal changes were not observed at sites distant from the INS location (), indicating that the INS-induced signal has high spatial selectivity. In a separate experiment, the same laser conditions used in were used to generate a response in forepaw cortex to demonstrate that INS can evoke optical responses in different cortical areas (). These experiments demonstrated that INS is capable of inducing optical responses in somatosensory cortex, some of which are similar to those obtained with natural tactile stimulation.
Fig. 3 INS evoked intrinsic optical signals in somatosensory cortex. INS evokes intrinsic optical signals in somatosensory barrel field (A–D) and forepaw cortex (E–H) in separate experiments (632 nm). (A & E) Blood vessel maps indicating (more ...)
3.3. Effects of laser repetition rate on INS evoked intrinsic signal
To further establish that the optical signal was indeed induced by INS, we varied repetition rate () to study how the reflectance signal changed in relation to the repetition rate of the laser. Our expectation was that the greater the total light energy applied to the cortex, the stronger the optical reflectance change. Using a 500 ms duration pulse train and 250 μs pulse width, we applied repetition rates of 50, 100, 150, and 200 Hz (). illustrates the location of the fiberoptic (Fiber) as well as a t-map generated via pixel-by-pixel t-tests (p < 0.001) between the 100 Hz laser stimulation and blank conditions (orange colored pixels). As can be seen qualitatively in , an increase in laser repetition rate increased the size of the activation region. Quantitatively, the 200 Hz laser stimulus (aqua blue line) produced the largest optical reflectance change, while a 50 Hz stimulus (pink line) produced the smallest response (). As shown in , the magnitude of the intrinsic signal exhibits an exponential fit with repetition rate (cf. Cayce et al., 2010
Fig. 4 Intrinsic signals produced by different rates of INS. (A) Blood vessel map. Location of fiberoptic is indicated by arrow. Orange pixels indicate significant pixels in t-test between 100 Hz stimulation and blank condition. (B–E) Activation maps (more ...)
3.4. Effects of radiant exposure on intrinsic signal
Threshold is an important aspect of INS to consider when developing the modality as an alternative to electrical stimulation. displays the functional response when the radiant exposure of each pulse was adjusted across imaging runs, and examines the time course of activation for the ROI (red box) shown in . The laser parameters of stimulation used to generate these time courses were 200 Hz, 500 ms duration pulse train of 250 μs pulses at the five different radiant exposures indicated in the figure legend. show the functional maps to 0.14 j/cm2 and 0.48 J/cm2 radiant exposures. As expected, the smallest radiant exposure (0.14 J/cm2 , blue line) resulted in the smallest intrinsic signal magnitude. Intermediate radiant exposures (0.21 and 0.32 and J/cm2, green and red lines) produced intermediate signal size, and the largest radiant exposures tested (0.37 and 0.48 J/cm2, purple and orange lines) resulted in the largest signal amplitudes. The magnitude of the intrinsic signal exhibits a linear fit with radiant exposure (). The area of activation also increased with radiant exposure energy, as shown qualitatively ().
Fig. 5 Increased INS radiant exposure leads to an increase in intrinsic signal magnitude. (A) Blood vessel map showing location of ROI (red box) and fiber location (tip barely in FOV). (B & C) Activation maps from stimulation with 0.14 J/cm2 and 0.48 (more ...)
Thus, within the ranges tested, both increases in laser stimulation rate and radiant exposures resulted in greater optical activation signal, suggesting a consistent and specific effect of laser stimulation on cortical response.
3.5. Effective distance of INS induced effect
The spatial selectivity of INS in cortical tissue was characterized by calculating the time course of the intrinsic signal at five distinct locations. displays the time courses based on distance from the laser stimulation site (200 Hz, 0.55 J/cm2
, pulse width 250 μs, pulse train 500 ms). In , the red box represents the region of interest closest to the optical fiber; the orange and magenta boxes are the most distant regions of interest. The peak signal is largest for the location closest to the fiber optic (, red line) and decreases in amplitude with distance from the stimulation location. This decline in signal size with distance also occurred in other directions as indicated by the comparable signal amplitude of magenta and orange ROIs in . As also shown in , the prominent effects of INS stimulation lies within 1–2 mm of the stimulation site and declines rapidly as a function of distance from the stimulation site (). The spatial temporal aspects of the signal are illustrated in through a time series of optical images taken during the imaging run. The data was temporally binned by 2 to decrease the number of images displayed in the mosaic; therefore, each frame represents 400 ms in time. The stimulus came on at 200 ms and was off at 700 ms after trial start, which indicates frames 1–2 represent the time the laser was turned on. The signal peaks between frames 7 and 8 which corresponds to the time courses displayed in . Furthermore, demonstrates that laser induced intrinsic signal is focal in an area measuring approximately 1 mm in diameter at its peak demonstrating the spatial precision of INS. This spatially limited characteristic makes INS a potentially useful method for studies requiring focal stimulation. The time series of images for 50, 100, 150, and 200 Hz were included as a supplemental figure
to further demonstrate the effects of repetition rate on the evoked signal and to demonstrate the spatial precision of INS.
Fig. 6 Spatial distribution of intrinsic signal in response to INS. (A) Blood vessel map with sampled ROIs overlaid. Color of box in map corresponds to color of time course trace displayed in (B). (B) Intrinsic signal time courses at different distances from (more ...)
3.6. Inhibitory effect of INS stimulation (without tactile stimulation) in somatosensory cortex
In addition to optical imaging, electrophysiological techniques were used to study the effects of INS on neuronal activity. displays the positioning of the electrode and fiber, approximately 1 mm apart. This arrangement was similar for each experiment involving electrophysiological measurements. Units that were responsive to tactile stimulation were isolated to assess cortical function during INS. displays the results of laser stimulation on spontaneous neural activity. Illustrated is a peristimulus time histogram (PSTH) resulting from the irradiation of somatosensory cortex (186 trials with intertrial intervals of 15 s; radiant exposure of 0.019 J/cm2, pulse width of 250 μs, pulse train length of 500 ms). Stimulus onset occurred at time zero and lasted for 500 ms (hashed bar on PSTH). We observed that INS led to a reduction in firing rate that lasted approximately 1.5–2.0 s, followed by a return to baseline levels. This reduction in firing rate was statistically significant, as evaluated by comparing the two seconds prior to stimulation and the two seconds post stimulation onset (paired t-test, α=0.05, −2000ms to 0 ms: p < 0.0046 0.0188, 2000ms to 4000 ms p<9.55e-50). No statistical difference was observed between the two seconds before stimulation and the time region corresponding to 2 – 4 seconds after stimulation offset.
Fig. 7 INS induces an inhibitory neural response and does not alter neuronal response to tactile stimulation. (A) Image of somatosensory cortex corresponding to barrel field showing electrode and fiber placement. Fiber stimulation site to electrode distance (more ...)
3.7. Cortex remains responsive during INS
The physiologic health of the cortex was assessed through electrophysiological recordings of neuronal responses to tactile stimulation. In , tactile stimulation was delivered by a piezoelectric stimulator that deflected contralateral whiskers once at each arrow in the PSTH. During runs in which INS was interleaved with tactile stimulation, consistent, normal neural responses to tactile stimulation were recorded (). The PSTH in represents tactile stimulation alone after INS had been applied demonstrating no loss in functionality. These recordings from tactile stimulation demonstrate that the normal excitatory and inhibitory periods of post-stimulation responses were present. Thus, even after repeated presentation of INS for a period of over 2 hours, normal neuronal tactile responses remained intact indicating that INS does not cause damage to cortex which compromises neuronal activity.
3.8. Stability of INS induced responses
Stability of INS induced responses is important to consider when assessing the stimulation modality’s efficacy for neuroscience applications and eventual translation to clinical studies. To examine the stability of INS induced responses, we produced a sequence of PSTHs by dividing the total number of INS trials into sequential 40 trial epochs. displays a sequence of PSTHs recorded from a single unit in response to INS (radiant exposure 0.055 J/cm2, spot size 850 μm, repetition rate of 200 Hz, pulse width 250 μs, train length 500 ms, 15 s intertrial interval). In most of the histograms, inhibition is most evident during the period from stimulus onset (time = 0 s) to approximately 1 s. In this example, it can be seen that INS induced inhibition is readily apparent from trial 1 to at least trial 120, and although appears to weaken slightly subsequently, the signal is still evident as late as trials 360–400. Paired t-tests between the two seconds following stimulus onset and prestimulus periods indicate that the difference in spike rate is statistically significant (α=0.05) for the summation of all trials () indicating that INS in rat somatosensory cortex has an immediate inhibitory effect on neuronal response, one which appears to remain present over many trials.
Fig. 8 Inhibitory effect of INS on neural activity is consistent over many trials. (A–J) PSTH mosaic of ten segments (40 trials per segment). The laser induced inhibition is strongest in the first 120 trials, then weakens but is evident through trials (more ...)
A separate experiment was conducted to assess the stability of this effect at two separate time points during an experiment. represents an initial PSTH taken over 30 trials separated by ITIs of 30 s. Again a period of statistically significant (paired t-test, −1 s to 0 s: p < 0.01, 1 s to 2 s p<4.08E-5) inhibition of baseline activity was observed within the first 1 s after INS onset (spot size 850 μm, pulse train length 500ms, pulse width 250 μs, radiant exposure of approximately 0.078 J/cm2). The laser was then turned off and the cortex was allowed to rest for 30 minutes after which the experiment was repeated. illustrates the PSTH recorded from single unit activity at the same location using the same laser parameters used to generate . Again, a period of inhibition was obtained within the first 1 s following INS (paired t-test, −1 s to 0 ms: p < 0.03, 1 s to 2 s p<0.00038). This further demonstrates the repeatability of the inhibitory effect.
Fig. 9 Repeatability of the neural INS inhibitory effect. (A) PSTH of single unit in response to INS. (B) A second PSTH from the same unit taken approximately 30 minutes later. Laser parameters: λ = 1.875 μm, repetition rate= 200 Hz, pulse train (more ...)
illustrate two separate single units from experiments performed on individual animals where INS generated an inhibitory response in each PSTH. The repetition rate was set at either 100 and 200 Hz for each PSTH and the radiant exposure was 0.078 J/cm2 and 0.019 J/cm2 respectively, using a 500 ms pulse train and a pulse width of 250 μs. INS induced inhibition was significant (paired t-tests: , 0 to 1 s vs. −1 to 0 s: p < 0.0037, or vs. 1 to 2 s; p < 0.000185; 0 to 2 s vs. −2 to 0 s: p < 0.0188, or 2 to 4 s p<0.0047). In total, statistically significant inhibitory responses were observed in all five animals studied in this fashion; excitatory responses were not observed. Thus, the effect of direct INS stimulation in rat somatosensory cortex appears to be inhibitory, as this effect was seen across all animals, different stimulation parameters, within different epochs of repeated INS stimulation, and during different recording periods within an experiment.