Hunting for origins of migraine pain: cluster analysis of spontaneous and capsaicin-induced firing in meningeal trigeminal nerve fibers

Abstract

Trigeminal nerves in meninges are implicated in generation of nociceptive firing underlying migraine pain. However, the neurochemical mechanisms of nociceptive firing in meningeal trigeminal nerves are little understood. In this study, using suction electrode recordings from peripheral branches of the trigeminal nerve in isolated rat meninges, we analyzed spontaneous and capsaicin-induced orthodromic spiking activity. In control, biphasic single spikes with variable amplitude and shapes were observed. Application of the transient receptor potential vanilloid 1 (TRPV1) agonist capsaicin to meninges dramatically increased firing whereas the amplitudes and shapes of spikes remained essentially unchanged. This effect was antagonized by the specific TRPV1 antagonist capsazepine. Using the clustering approach, several groups of uniform spikes (clusters) were identified. The clustering approach combined with capsaicin application allowed us to detect and to distinguish "responder" (65%) from "non-responder" clusters (35%). Notably, responders fired spikes at frequencies exceeding 10 Hz, high enough to provide postsynaptic temporal summation of excitation at brainstem and spinal cord level. Almost all spikes were suppressed by tetrodotoxin (TTX) suggesting an involvement of the TTX-sensitive sodium channels in nociceptive signaling at the peripheral branches of trigeminal neurons. Our analysis also identified transient (desensitizing) and long-lasting (slowly desensitizing) responses to the continuous application of capsaicin. Thus, the persistent activation of nociceptors in capsaicin-sensitive nerve fibers shown here may be involved in trigeminal pain signaling and plasticity along with the release of migraine-related neuropeptides from TRPV1 positive neurons. Furthermore, cluster analysis could be widely used to characterize the temporal and neurochemical profiles of other pain transducers likely implicated in migraine.

Keywords: capsaicin; cluster analysis; pain; spike; trigeminal nerve.

Figure 1

Experimental setup for recordings from the nervus spinosus, branch of the trigeminal nerve in the hemiskull preparation. (A) Photograph of the hemiskull preparation with preserved nervus spinosus (nerve) innervating a region of the medial meningeal artery (MMA). The nerve was cut and placed into the suction glass electrode. (B) Double immunolabeling of a cryostat cross-section of nervus spinosus with antibodies to a light component of neurofilament (NF-L, red) and to a myelin basic protein (MBP, green) showed the presence of myelinated NF-positive nerve fibers in the sample. Bar = 20 μm.

Figure 2

Spontaneous multiple unit activity in nervus spinosus. (A) An example trace of spontaneous spiking activity recorded from the nervus spinosus. (B) Action potential parameters calculation in average spike. (C) Histograms showing the distribution of positive and negative spike amplitudes in one experiment built from n = 2627 spikes recorded during 15 min. Events below threshold (5 SD of baseline noise) are indicated in gray (n = 4100). (D) Histograms showing the distribution of positive and negative spike duration from the same experiment (n = 2627 spikes).

Figure 3

Capsaicin increases multiple unit activity in nervus spinosus. (A) Example of 200 s long recordings of multiple unit activity (MUA) in the nervus spinosus before and during bath-application of 1, 2 or 10 μM capsaicin. (B) MUA densities in the nervus spinosus before and during 1, 2 and 10 μM capsaicin application (mean ± SE; 10 s bin size; n = 10, 7 and 6, respectively). (C) Histograms quantifying the number of nociceptive spikes in control and after application of 1, 2 or 10 μM capsaicin (n = 10, 7 and 6, respectively). *p < 0.05, **p < 0.01.

Figure 4

Capsazepine prevents the activation of nociceptive firing by capsaicin in nervus spinosus. (A) Example traces of the multiple unit activity stimulated by 1 μM capsaicin in control (top) and in the presence of 50 μM capsazepine (bottom). (B) Histograms showing the average effect of 1 μM capsaicin (n = 10), washout of the agonist (n = 10), effect of 50 μM capsazepine (n = 5) and action of 1 μM capsaicin in the presence of 50 μM capsazepine (n = 5) on the number of nociceptive spikes during 5 min of recordings. *p < 0.05.

Figure 5

Cluster analysis of spikes in nervus spinosus. (A) A typical presentation of spike clusters in 2D in control (spontaneous activity). Positive spike amplitudes are plotted vs. positive spike durations (top) and vs. the negative spike amplitudes (bottom). Each individual dot corresponds to a single spike. Color contours outline individual spike clusters separated on the basis of KlustaKwik methods. 1 and 2 indicate responder and non-responder clusters, respectively. (B) Spike clusters from the same predation as shown on panel (A) in the presence of 2 μM capsaicin. Notice that some units (black and magenta) increased firing during application of capsaicin (responders) whereas others (blue and green) were insensitive to capsaicin (non-responders). Inset indicates the isolated presentation of two clusters (magenta and black) when the negative duration was plotted vs. positive duration which were less distinguishable in other projections. (C) Time course of spike frequency of the units 1 (responder) and 2 (non-responder) before and after addition of capsaicin. (D) Average spike shapes for the clusters 1 and 2 (mean ± SD; responder: n = 9 in control, n = 475 in capsaicin; non-responder: n = 160 in control, n = 97 in capsaicin).

Figure 6

Capsaicin induced firing in isolated trigeminal neurons. (A) Double immunolabeling of cultured trigeminal ganglion neurons with antibodies to a capsaicin receptor transient receptor potential vanilloid 1 (TRPV1; green) and to a light component of neurofilament (NF-L, red). The picture indicates that the majority of TRPV1-expressed trigeminal neurons were NF-L immunonegative. Calibration bar = 20 μm. Color bar indicates percentage of TRPV1 and NF-L positive neurons and respective overlap of labeling. (B,C) Example traces of capsaicin (1 μM, 2 s pulses with 20 s intervals) with stable (B) and reduced amplitudes (C). (D) Multiple firing induced by 1 μM capsaicin in an individual neuron. (E) Analysis of firing showing average and shortest intervals in individual neurons. Notice that three neurons generated firing with 20–30 ms intervals.

Figure 7

Spectral analysis of single unit activity. (A) Examples of spikes within the same cluster in two different time windows before and after capsaicin application. The right trace shows the shape of spikes belonging to the same unit. Red ticks above the trace in (a) indicate spikes of the same cluster whereas the blue frame—the part of the trace shown on the right (b) by higher temporal resolution. Notice that multiple firing in this small time window did not change the spike shape (c) inducing that they all belong to the same cluster. (B) The spectral analysis of spiking activity showing an increase in number of frequent spikes within two clusters (one with relatively high ongoing activity (right) and another one essentially silent prior to capsaicin application (left).

Figure 8

Temporal analysis of capsaicin effect in different spike clusters. (A) The variable time-course of capsaicin effect in different clusters in the nervus spinosus grouped from three different experiments (hemiskulls 1–3) on the basis of the response latency and persistence. Notice the presence of early transient responder (a), late transient responders (b), persistent responders (c) and late persistent responders (e). The width of the window was 500 s. (B) Average spike shapes for presented clusters: (a) n = 826; (b) n = 88; (c) n = 606; (d) n = 451; (e) n = 1080; (f) n = 270; (g) n = 587 events within the cluster. (C) Autocorrelogramms for corresponding spike clusters (5 ms bin). (D) Color spectrograms illustrating the temporal profiles of interspike intervals for corresponding clusters.

Figure 9

Quantification of time profiles of clusters in nervus spinosus differently responding to capsaicin. (A) Histograms showing persistence of responses of all clusters in nervus spinosus to application of three different concentrations of capsaicin. Notice an increase in the number of transient type responses proportional to agonist concentration. (B) Histograms showing the latency of responses capsaicin (74 responding clusters from 23 preparations).