The Interval Between VNS-Tone Pairings Determines the Extent of Cortical Map Plasticity

Abstract

Repeatedly pairing vagus nerve stimulation (VNS) with a tone or movement drives highly specific and long-lasting plasticity in auditory or motor cortex, respectively. Based on this robust enhancement of plasticity, VNS paired with rehabilitative training has emerged as a potential therapy to improve recovery, even when delivered long after the neurological insult. Development of VNS delivery paradigms that reduce therapy duration and maximize efficacy would facilitate clinical translation. The goal of the current study was to determine whether primary auditory cortex (A1) plasticity can be generated more quickly by shortening the interval between VNS-tone pairing events or by delivering fewer VNS-tone pairing events. While shortening the inter-stimulus interval between VNS-tone pairing events resulted in significant A1 plasticity, reducing the number of VNS-tone pairing events failed to alter A1 responses. Additionally, shortening the inter-stimulus interval between VNS-tone pairing events failed to normalize neural and behavioral responses following acoustic trauma. Extending the interval between VNS-tone pairing events yielded comparable A1 frequency map plasticity to the standard protocol, but did so without increasing neural excitability. These results indicate that the duration of the VNS-event pairing session is an important parameter that can be adjusted to optimize neural plasticity for different clinical needs.

Keywords: VNS; acoustic trauma; auditory cortex; interval between stimulations; plasticity; vagal.

Figure 1

Schematic diagram of the VNS-tone pairing procedure. A 0.5 second, 30 Hz train of 100 μs wide biphasic pulses was delivered to the left vagus nerve via a cuff electrode. Rats received VNS paired with a 9 kHz tone during 20 daily pairing sessions. The inter-stimulus interval between VNS-tone pairings was either 8, 30, or 120 sec. Cortical recordings were made 24 hours after the last pairing session.

Figure 2

Frequency map plasticity in primary auditory cortex. Each polygon represents a single electrode penetration. The characteristic frequency (CF) of each site is indicated in kHz. The red color indicates that the value of the CF is between 8 and 16 kHz. (a) Representative frequency map from a naive rat. Representative frequency maps from a rat that received VNS paired with 9 kHz tones with (b) an inter-stimulus interval of 8 sec, (c) an inter-stimulus interval of 30 sec, and (d) an inter-stimulus interval of 120 sec. Map plasticity is greatest with longer inter-stimulus intervals. The scale bar indicates a distance of 0.5 mm. Anterior is shown to the left and dorsal is down.

Figure 3

The percentage of A1 recording sites tuned to each of five one-octave frequency bins. There was a significant shift in tuning between 8 and 16 kHz when VNS was paired with a 9 kHz tone using longer inter-stimulus intervals. Error bars indicate standard error of the mean across rats. Asterisks indicate experimental groups that were statistically significant from the naïve control group (p < 0.05).

Figure 4

Percent of A1 neurons responding to each tone frequency intensity combination for (a) naive control rats and for rats that received VNS-tone pairing (b) every 8 sec, (c) 30 sec, and (d) 120 sec. (e–g) The difference in the percent of A1 neurons that respond to tones between VNS-tone paired rats and control rats. Red indicates a greater percent of A1 neurons that respond in VNS-tone paired rats compared to control rats, while blue indicates a decrease in the percent of A1 neurons that respond in VNS-tone paired rats. White lines delineate the frequency intensity combinations which activate significantly more neurons after VNS-tone pairing (p < 0.01). Black lines delineate significantly decreased responses (p < 0.01).

Figure 5

VNS-tone pairing reorganizes the auditory cortex frequency map. Red indicates the proportion of A1 neurons that respond to 50 dB SPL tones between 8 and 16 kHz. Blue indicates the proportion of A1 neurons that respond to 50 dB SPL tones between 1 and 2 kHz. Error bars indicate standard error of the mean across rats. Asterisks indicate experimental groups that were statistically significant from the naïve control group (p < 0.05).

Figure 6

The average number of A1 action potentials (spikes) evoked by each tone frequency intensity combination for (a) naive control rats and for rats that received VNS-tone pairing (b) every 8 sec, (c) 30 sec, and (d) 120 sec. (e–g) The difference between the number of spikes evoked in experimental rats and control rats reveals the range of tones that evoked a stronger (red) or weaker (blue) response in each of the experimental groups. White lines delineate the frequency intensity combinations which generate significantly more spikes after VNS-tone pairing, while black lines delineate significantly fewer spikes (p < 0.01).

Figure 7

VNS-tone pairing reduces the number of spikes evoked by low frequency tones (1–2 kHz, blue), and does not significantly alter the number of spikes evoked by high frequency tones (8–16 kHz, red). Error bars indicate standard error of the mean across rats. Asterisks indicate experimental groups that were statistically significant from the naïve control group (p < 0.05).

Figure 8

Summary of the effects of VNS-tone pairing on A1 responses. The degree of cortical plasticity was a monotonic function that increased with longer daily pairing session durations.

Figure 9

The percentage of A1 recording sites tuned to each of five one-octave frequency bins. There was a significant shift in tuning between 8 and 16 kHz when VNS was paired with tones 300 times per day, but not when VNS was paired with tones 50 times per day. Error bars indicate standard error of the mean across rats. Asterisks indicate experimental groups that were statistically significant from the naïve control group (p < 0.05).

Figure 10

The short VNS-tone pairing inter-stimulus interval was not sufficient to reverse the neural or behavioral deficits observed following acoustic trauma. (a) Sham therapy and VNS-tone pairing therapy delivered every 8 sec were not sufficient to reduce the percent of A1 responding to low frequency tones. Error bars indicate standard error of the mean across rats. Asterisks indicate experimental noise exposed (NE) groups that were statistically significant from the 30 sec VNS-tone paired group (p < 0.05). (b) The inter-stimulus interval between VNS-tone pairings affected the ability of noise exposed rats to detect 50 ms gaps embedded in continuous background noise.