C Fibers

Infrared Laser stimuli were administered at the back of the left hand using a thulium solid-state laser (Themis®, StarMedTec GmbH, Starnberg, Germany) at a wavelength of 1.96 µm. The stimuli were short (1 ms) pulses with a power of 150–600 mJ and a beam diameter of 5 mm. The high power of a laser stimulus produces a very fast heat ramp, which generally activates the terminals of both A_δ and C fibers [13] , [14] . However, since A_δ fiber elicited first pain precedes the C fiber elicited second pain due to the different conduction velocities of the two afferents (∼10 m/s for A_δ and ∼1 m/s for C fibers), it can be easily distinguished when the stimulus is applied at a remote location such as the back of the hand [15] (link). Furthermore, “first pain” is more salient than “second pain” [16] (link). Therefore, in the present study, subjects were asked to rate and choose descriptors for the first sensation they experienced when being stimulated by the laser, which set the focus on the A_δ component.

Adult female Wistar rats (n = 22) were anesthetized with 4% isoflurane and a then maintained at 2% after being placed in a stereotaxic frame. A laminectomy was performed to expose the L4-L5 SC segments, which were then fixed in place by two clamps positioned on the apophysis of the rostral and caudal intact vertebras. The dura matter was then removed. To record wide-dynamic-range neurons (WDR), a silicone tetrode (Q1x1-tet-5mm-121-Q4; Neuronexus, USA) was lowered into the medial part of the dorsal horn of the SC, at a depth of around 500–1100 µm from the dorsal surface (see Fig. 5a for localization of recorded WDRs). We recorded WDR neurons of lamina V as they received both noxious and non-noxious stimulus information from the ipsilateral hind paw.
We measured the action potentials of WDR neurons triggered by electrical stimulation of the hind paw. Such stimulation induced the activation of primary fibers, whose identities can be distinguished by their spike onset following each electrical stimuli (Aβ-fibers at 0–20 ms, Aδ-fibers at 20–90 ms, C-fibers at 90–300 ms and C-fiber post discharge at 300 to 800 ms). When the WDR peripheral tactile receptive fields are stimulated with an intensity corresponding to 3 times the C-fiber threshold (1 ms pulse duration, frequency 1 Hz), a short-term potentiation effect, known as wind-up (WU), occurs that leads to an increased firing rate of WDR neurons^{57 (link),58 (link)}. Because the value of WU intensity was highly variable among recorded neurons within and across animals, we averaged the raster plots two dimensionally across neurons within each group of rats. We further normalized these data so that the plateau phase of the maximal WU effect was represented as 100 percent activity. As WU is dependent on C-fiber activation, it can be used as a tool to assess nociceptive information in the SC and, in our case, the anti-nociceptive properties of OT acting in the vlPAG. We recorded WDR neuronal activity using the following protocol: 40 s of hind paw electric stimulation to induce maximal WU followed by continued electrical stimulation to maintain WU while simultaneously delivering 20 s of vlPAG blue light stimulation (30 Hz, 10 ms pulse width, output ~3 mW), followed by another 230 s of electrical stimulation alone to observe the indirect effects of OT on WU in WDR neurons. Electrical stimulation was ceased after the 290 s recording session to allow the WU effect to dissipate. Following a 300 s period of no stimulation, the ability of the WU effect to recover was assessed by resuming electrical stimulation of the hind paw for 60 s of WU. After another 10 min period without stimulation, we sought to confirm the effects of vlPAG OT activity on WU intensity by injecting 600 nl of the OTR antagonist, dOVT, (d(CH2)5-Tyr(Me)-[Orn8]-vasotocine; 1 µM, Bachem, Germany) into the vlPAG of the rats expressing ChR2, and repeated the protocol described above.
The spikes of each recording unit were collected as raster plots with the vertical axis showing the time relative to electric shock, and the horizontal axis showing the number of electric shocks. Next, the raster plots were smoothed by convolution of the Gaussian distribution (horizontal width = 100 ms, vertical width = 20 ms, standard deviation = 20. The total number of C fiber derived spikes occurring between 90 and 800 ms after each electric shock was counted. The spike counts were smoothed with a moving average window of 21 s and the window containing the maximal activity was defined as ‘100% activity’, which was then used to normalize the activity of each recording unit. Finally, the normalized percent activity from each recording unit was averaged for each experimental condition and plotted in Fig. 5.

The optical frequency comb (OFC) used as the source comb in Fig. 1a is based on an Er-doped fiber femtosecond laser (C-fiber, Menlo Systems). The OFC offers a 60 nm wavelength bandwidth centered at 1550 nm, equivalent to a 4 THz frequency span, with a 100 MHz comb line-to-line spacing. The total comb power is set at 20 mW, while a single comb line is given ~200 nW. For each comb line to be extracted, a narrow-band spectral filter of a 100 MHz bandwidth (FWHM) is devised individually by combining a fiber Fabry-Perot (FFP) filter with a fiber Bragg grating (FBG) filter^{39 (link)}. The filtered-out comb line is then injection-locked to a distributed feedback (DFB) laser diode for power amplification, with a factor of 40–50 dB, without notable frequency shifting or linewidth degradation.