The striatum is the major input nucleus of the basal ganglia (BG) and contains two major populations of medium spiny projection neurons (Gerfen et al., 1990; Mink, 1996). Striatonigral (direct pathway) neurons express D1 receptors and project to the substantia nigra pars reticulata (SNr) and other BG output nuclei. Striatopallidal (indirect pathway) neurons express D2 receptors and project to the external globus pallidus, which in turn inhibits the SNr. Ever since the connectivity of these pathways was defined, there has been controversy on their contributions to behavior. Given their opposite effects on BG output, an influential model of the striatum casts the direct and indirect pathways in opposing roles with the direct pathway promoting behavior and the indirect pathway suppressing behavior (Albin et al., 1989; Freeze et al., 2013; Kravitz et al., 2010). More recently, however, this view has been questioned by studies that described concurrent activation of these two populations during behavior, suggesting complementary contributions of these two pathways (Cui et al., 2013; Isomura et al., 2013).

The lateral or sensorimotor striatum receives topographically organized innervation from the cortex and thalamus (Hintiryan et al., 2016). Micro-stimulation of different striatal regions results in movements of different body parts (Alexander and DeLong, 1985). In rats, the ventrolateral striatum (VLS) has been implicated in orofacial behaviors, including licking (Mittler et al., 1994; Pisa, 1988a; Pisa, 1988b). Although licking behavior is generated by brain stem pattern generators (Travers et al., 1997), BG output can provide top-down regulation of the pattern generators, either directly or via the nigrotectal projections (Deniau and Chevalier, 1992; Redgrave et al., 1992; Rossi et al., 2016; Shammah-Lagnado et al., 1992; Toda et al., 2017). Stimulation or enhancing dopaminergic signaling in the VLS can generate orofacial behavior, in some cases dyskinetic movements, while dopamine depletion can abolish such behaviors (Delfs and Kelley, 1990; Jicha and Salamone, 1991; Kelley et al., 1988). However, the detailed pathway-specific mechanisms underlying such observations remain largely unknown.

One important characteristic of self-initiated behaviors is that they can be precisely timed relative to some anticipated event like a reward (Gibbon et al., 1984; Pavlov, 1927; Roberts, 1981). The BG have been implicated in such interval timing in the seconds to minutes range (Buhusi and Meck, 2005; Merchant et al., 2013). Manipulation of dopamine activity can bidirectionally influence estimation of the passage of time (Buhusi and Meck, 2005; Meck, 1996) and lesions of the striatum also impair timing behavior (Meck, 2006). Furthermore, striatal activity has also been shown to represent time intervals (Bakhurin et al., 2017; Gouvêa et al., 2015). Although the role of D1-expressing and D2-expressing neurons in timing has been studied using systemic pharmacological manipulations (Frederick and Allen, 1996), it remains unclear how the two pathways of the BG interact to mediate these behaviors.

In this study, we used optogenetic stimulation to examine the contributions of the VLS in both action generation and timing. Using a fixed-time (FT) schedule of reinforcement, combined with occasional peak probe trials, we showed that mice rapidly acquired highly predictable anticipatory licking patterns that reflects reliable internal timing processes (Toda et al., 2017). Using temporally and spatially precise manipulations of neural activity with optogenetics, we studied how direct and indirect pathways in the VLS regulate the initiation, maintenance, and timing of self-initiated licking.

To study the roles of the direct and indirect pathways during self-initiated behavior, we manipulated the activity of cell populations that originate these pathways in the orofacial region of the striatum (Figure 1A & B). We bilaterally injected a Cre-dependent AAV5-DIO-ChR2 virus in the VLS region of the striatum in D1-Cre (Figure 1C) and A2A-Cre (Figure 1D) mice and implanted chronic optic fibers just above this area. Five additional A2A-Cre mice received AAV5-DIO-eYFP virus along with fiber implants. We tracked licking behavior in head-fixed mice performing an interval timing task (Toda et al., 2017). Mice received a 5 μL drop of 10% sucrose solution every 10 s (Figure 1E). Voluntary licking behavior in all mice consisted of bouts of rhythmic protrusion and retractions of the tongue at a relatively fixed rate (4–8 Hz). Well-trained mice reliably generated anticipatory licking behavior before reward delivery (Figure 1F). As a result, the average licking rate gradually ramps up in expectation of reward, and then peaks following reward and shows a sharper decline after consumption (Figure 1G).

Figure 1

Download asset Open asset

We tested the effects of stimulating the direct pathway in the VLS on licking at three different time points during the interval. We stimulated the direct pathway for 1 s before mice normally initiate anticipatory licking, starting at 5 s before reward delivery (Figure 2A). Though licking was sparse at this time point (Figure 2B, top), activation of direct pathway MSNs induced licking (Figure 2B, bottom and Figure 2C, Video 1). We determined the change in lick rate that resulted from laser stimulation by calculating the lick rate during the 1 s laser stimulation period for D1-Cre mice and a control group that received eYFP virus. The normal lick rate obtained for the same time point during trials without laser stimulation was subtracted from this value to quantify the stimulation-induced change in lick rate (Figure 2D). A two-way ANOVA (stimulation frequency x experimental group) revealed a change of licking rate resulting from VLS direct pathway stimulation (main effect of group, F_1,44 = 93.6, p<0.0001) and a significant effect of laser frequency (F_3,44 = 4.64, p<0.01); there was a also an interaction between frequency and group (F_3,44 = 4.84, p<0.01). Post-hoc tests revealed that lick rate was significantly increased by 10 Hz (p<0.01), 25 Hz, (p<0.001), and 50 Hz (p<0.001) stimulation of the direct pathway. We next stimulated the direct pathway at 1 s before reward delivery (Figure 2E), during the time of maximal anticipatory licking (Figure 2F, top). Although there was a ceiling effect on licking at this time point (Figure 2F, bottom), we regularly observed a potentiation of the licking rate during stimulation when compared to trials without stimulation (Figure 2G). A two-way ANOVA revealed significant increase in licking as a result of VLS stimulation (Figure 2H, main effect of group, F_1,37 = 48.82, p<0.0001), but no effect of frequency (F_3,37 = 1.31, p=0.3) nor a significant interaction between frequency and group (F_3,37 = 1.79, p=0.16). Stimulation at the time of reward delivery (Figure 2I) had the most modest impact on licking rate (Figure 2J and K). However, a two-way ANOVA detected a significant effect of stimulation on licking even during consumption (Figure 2L, F_1,36 = 8.65, p<0.01), but did not show a significant effect of stimulation frequency (F_3,36 = 1.27, p=0.3), nor a significant interaction (F_3,36 = 1.73, p=0.18). We directly compared the effects of stimulation on licking as a function of time of laser stimulation during the task. A two-way ANOVA revealed significant effects of both the time of stimulation in the task (Figure 2M, main effect of time, F_3,67 = 9.16, p<0.0001) and laser frequency (F_2,67 = 24.31, p<0.0001), with no significant interaction (F_6,67 = 0.99, p=0.44). Together, these results show for the first time that direct pathway stimulation in the orofacial striatum can result in the generation of licking behavior. The effects of direct pathway stimulation were dependent on the on-going behavioral state of the animals, which can be explained by a ceiling effect.

Figure 2

Download asset Open asset

Video 1

Download asset

Video 2

Download asset

Remarkably, we observed that it was possible for licking to exceed the frequencies normally produced by mice during direct pathway stimulation (Figure 3A). Licking during laser presentation could be shifted rightward toward a frequency of 10 Hz (Video 2). This appeared to be the upper limit of the central pattern generator for licking, as naturally generated licking showed predominant frequencies between 4 and 6 Hz. To explore this effect, we analyzed licking in the frequency domain. We quantified power spectral density (PSD) distributions for licking and observed that licking frequencies are higher during direct pathway stimulation compared to licking produced during anticipation (Figure 3B). Laser stimulation increased the licking frequency to around 10 Hz, which is rarely reached by mice naturally. Varying stimulation frequency also systematically regulated the power of licking around a specific frequency, reflected in the percentage of the PSD distribution occupying that frequency range (Figure 3C).We compared the amplitude of the PSD in 3 non-overlapping licking frequency bands (4-6, 6-8, and 8-10 Hz) as a function of direct pathway stimulation frequency when stimulation occurred 1 second prior to reward. A two-way mixed ANOVA showed that was a significant interaction between stimulation frequency and licking frequency bands (Figure 3D, F_8,58 = 2.54, p < 0.05), and significant main effects of both licking frequency band (F_2,58 = 15.46, p < 0.0001) and stimulation frequency (F_4,58 = 3.89, p < 0.05). Post-hoc analyses indicated that power in the 8-10Hz band was significantly greater than power in the 4-6 Hz band during 10 (p < 0.05) and 25 (p < 0.01) Hz stimulation and that power in the 6-8 Hz band was significantly greater than in the 4-6 Hz band during 50 Hz (p < 0.01) stimulation.

Figure 3

Download asset Open asset

When stimulation occurred 5 seconds prior to reward, there was a significant interaction between licking frequency band and stimulation frequency (Figure 3E, F_8,60 = 6.43, p < 0.0001) in their effects on PSD amplitude. We also found main effects of frequency band (F_2,60 = 37.76, p < 0.0001) and stimulation frequency (F_4,60 = 2.84, p < 0.05). Post-hoc analyses indicated that power in the 8-10 Hz band was significantly greater than power in the 4-6 Hz band during 10 (p < 0.001), 25 (p < 0.001), and 50 (p < 0.001) Hz stimulation and was greater than power in the 6-8 Hz band during 25 Hz stimulation (p < 0.001). In addition, power in the 6-8 Hz band was significantly greater than power in the 4-6 Hz band during 25 (p < 0.05) and 50 Hz (p < 0.05) Hz stimulation. These results indicate that increasing direct pathway stimulation 5 seconds prior to reward also resulted in more licking in the 8 –10 Hz frequency band.

Lastly, when stimulation occurred together with reward, there was a significant interaction between licking frequency band and stimulation frequency (Figure 3F, F_8,56 = 2.59, p < 0.05) in their effects on PSD amplitude. We found main effects of frequency band (F_2,56 = 54.02, p < 0.0001) and stimulation frequency (F_4,56 = 3.3, p < 0.05). Post-hoc analyses indicated that power in the 8-10 Hz band was significantly greater than power in the 4-6 Hz band during 5 (p < 0.01, 10 (p < 0.001), 25 (p < 0.001), and 50 (p < 0.001) Hz stimulation and was greater than licking power in the 6-8 Hz band during 25 (p < 0.01) and 50 (p < 0.01) Hz stimulation. In addition, licking power in the 6-8 Hz band was significantly greater than power in the 4-6 Hz band during 5 (p < 0.05) and 10 Hz (p < 0.05) Hz stimulation. These results indicate that increasing direct pathway stimulation during consumption licking also resulted in more licking in at 8 – 10 Hz. Together, light delivery to the VLS resulted in rightward shift in the frequency domain, above the naturally occurring states that mice normally generate. In addition, direct pathway stimulation could increase the degree to which of licking occupied a given frequency band, reflected in the shape of the PSD function. These results suggest a role for the BG in regulating the frequency and duty cycle of the licking oscillator in different modes corresponding to roughly three frequency bands (4-6 Hz, 6-8 Hz, and 8-10 Hz).

When stimulating the direct pathway at 5 s prior to reward, we observed that licking resulting from stimulation only began after a certain delay (Figure 4A), and that this delay was reduced with higher stimulation frequencies (Figure 4B). A one-way ANOVA revealed a significant effect of stimulation frequency on the latency to initiate licking following the onset of laser stimulation (Figure 4C; F_3,22 = 4.92, p<0.01). The lick bout that resulted from stimulation also lasted longer than the 1 s stimulation period. We counted the number of licks that occurred over the course of one second following the offset of stimulation and compared that value with the number of licks normally detected during the same period in non-stimulation trials. Stimulation would result in some additional licking even after offset of stimulation, but this was not related to the stimulation frequency (Figure 4D; two-way mixed ANOVA; main effect of stimulation, F_1,23 = 18.1, p<0.001; main effect of frequency, F_3,23 = 3.58, p=0.30; interaction, F_3,23 = 1.91, p=0.15). These results show that greater stimulation of the direct pathway reduces licking initiation latency, and that licking continues for a short time beyond the immediate period of laser stimulation.

Figure 4

Download asset Open asset

In contrast to D1 activation, indirect pathway activation in the VLS in well-trained mice resulted in a pronounced suppression of licking. When stimulation occurred 5 s prior to reward delivery (Figure 5A), we observed a reduction in licking rate (Figure 5B and C). We found a significant effect of experimental group on the change in licking rate (Figure 5D), two-way ANOVA, F_1,40 = 12.35, p<0.01), but no effect of frequency (F_3,40 = 2.04, p=0.12) nor an interaction between group and frequency (F_3,40 = 2.2, p=0.1). Although we did regularly observe a suppression of licking activity 5 s prior to reward, the likelihood of licking at this period in the interval was low, reflecting a potential floor effect. When activating the indirect pathway 1 s prior to reward (Figure 5E), during peak anticipatory licking, we also observed a significant impact on licking by indirect pathway stimulation (Figure 5F and G, Video 3). In addition to a strong suppression of licking by indirect pathway stimulation (Figure 5H), two-way ANOVA; significant effect of experimental group F_1,42 = 35.48, p<0.0001), we observed a significant effect of stimulation frequency on licking (F_3,42 = 8.88, p<0.0001), and a significant interaction between group and frequency (F_3,42 = 6.31, p<0.01). Post-hoc tests revealed significant suppression of licking by 25 Hz (p<0.001) and 50 Hz (p<0.001) stimulation. We were also able to suppress consummatory licking that occurred following reward (Figures 5I, J and K). There was a significant effect of experimental group on the change in lick rate (Figure 5L, two-way ANOVA, F_1,42 = 6.48, p<0.05), a significant effect of frequency (F_3,42 = 4.01, p<0.05), and a significant interaction between group and frequency (F_3,42 = 3.38, p<0.05). Post-hoc tests showed that the interaction was mediated by significant suppression of licking by 25 Hz (p<0.05) and 50 Hz (p<0.05) stimulation. The effect of indirect pathway stimulation was also dependent on the time of stimulation. A two-way ANOVA revealed significant effects of both time of stimulation in the task (Figure 5M, F_3,76 = 21.38, p<0.0001) and laser frequency (F_2,76 = 3.19, p<0.05), with no significant interaction (F_6,76 = 2.1, p=0.06). Together, these results reveal a frequency-dependent effect of indirect pathway stimulation on licking behavior. Furthermore, indirect pathway activation suggests less of an effect on licking occurring during consumption than on licking during anticipation.

Figure 5

Download asset Open asset

Video 3

Download asset

When we activated the indirect pathway at 5 s prior to reward delivery, when mice were least likely to lick, we observed a striking rebound of licking following the offset of stimulation (Figures 4B C, 6A B, Video 4). Lick rates during this rebound were higher during the second following laser offset as compared to the same time point in the absence of stimulation (Figure 6C, two-way mixed ANOVA, main effect of stimulation, F_1,24 = 11.08, p<0.01), but there was no main effect of frequency (F_3,24 = 0.17, p=0.9) and no interaction between stimulation frequency and the rebound licking rate (F_3,24 = 0.36, p=0.78). Indirect pathway stimulation also resulted in a reduced latency to initial lick onset following stimulation when compared to non-stimulation trials (Figure 6D, two-way mixed ANOVA, F_1,24 = 16.41, p<0.001; main effect of frequency, F_3,24 = 2.85, p=0.05). There was a significant interaction between trial type and the stimulation frequency, with higher frequencies resulting in lower onset latencies (F_3,24 = 5.73, p<0.01). Post-hoc tests showed that the interaction was mediated by shorter rebound latencies following 25 Hz (p<0.05) and 50 Hz (p<0.001) stimulation. Varying stimulation frequencies resulted in differing effects on the regularity of rebound latencies, indicated by measuring the variance of lick onset (Figure 6E, two-way mixed ANOVA, main effect of frequency, F_3,24 = 6.12, p<0.01; main effect of stimulation, F_1,24 = 3.3, p=0.08; interaction between stimulation frequency and trial type, F_3,24 = 8.63, p<0.001). Post-hoc tests showed greater variance of lick onset time following 5 Hz stimulation (p<0.05), and significantly lower variance following 50 Hz stimulation (p<0.01).

Figure 6

Download asset Open asset

Video 4

Download asset

We observed that the effects of laser stimulation would be altered over the course of the behavioral session, as a function of motivational state. For example, the effects of direct pathway stimulation at 5 s prior to reward (Figure 7A) on changes in lick rate were often dependent on how many trials the mice received (Figure 7B). At higher frequencies, this gradual reduction of laser-generated licking rate was not as evident (Figure 7C). We divided the session into quartiles and quantified the increase of licking rate by stimulation as a function of trial quartile (Figure 7D). A two-way, mixed ANOVA revealed a significant interaction between frequency and quartile (F_9,72 = 2.62, p<0.05) in addition to significant effects of frequency (F_3,72 = 9.5, p<0.001) and trial quartile (F_3,72 = 3.64, p<0.05) on lick rate. Post-hoc tests showed significant reductions in lick rate between the 4^th quartile and the 1^st (p<0.001), 2^nd (p<0.001), and the 3^rd (p<0.01) quartiles under 10 Hz stimulation. We also found that the latency to stimulation-evoked lick onset increased over time. A two-way, mixed ANOVA revealed a significant effect of trial quartile on lick onset latency (Figure 7E, F_3,60 = 3.25, p<0.05) and a significant effect of frequency (F_3,60 = 5.67, p<0.01), but no interaction between trial quartile and stimulation frequency (F_9,60 = 0.38, p=0.9), suggesting a uniform increase in latency across all parameters. Like the effects of stimulation of the direct pathway on latency to lick, the latency to rebound licking following stimulation at 5 s was affected by motivation (Figure 7F). The latency of rebound licking increased as the mouse became sated in the course of a session, but this was most evident at higher stimulation frequencies (Figure 7G, two-way mixed ANOVA, interaction between trial quartile and stimulation frequency F_9,72 = 2.16, p<0.05; main effect of frequency, F_3,72 = 3.88, p<0.05; main effect of trial quartile F_3,72 = 4.04, p<0.05). Post-hoc tests showed that latency to rebound during the 4^th quartile was significantly greater than during the 1^st and 2^nd quartiles under 25 Hz (p<0.05) and 50 Hz stimulation (p<0.01). These results suggest that engagement of lower-level licking generators by the BG can be modulated by motivational signals.

Figure 7

Download asset Open asset

Our task allowed for the interrogation of the roles of the BG pathways in timing, as mice will readily anticipate the arrival of the reward (Figure 8A). We incorporated probe trials on which the reward was omitted to assess internal timing (Figure 8B). All mice generated anticipatory licking bouts during probe trials. Averaging across many trials results in a distribution of licking rates whose peak reflects the expected time of reward delivery, in this task at approximately 10 s. On a fraction of the probe trials we stimulated direct and indirect pathways to examine their contributions to timing.

Figure 8

Download asset Open asset

We found that stimulation of the direct pathway during the probe trial could shift the peak rightward by an entire interval-duration. Thus, stimulation of the direct pathway concurrently with reward did not result in an obvious peak shift (Figure 9A, left). However, as we moved the stimulation time rightward by 3 or 5 s, the peak would move rightward proportionally (Figure 9A, middle and right). Indeed, we found that the peak shift is a function of the start time of direct pathway stimulation (Figure 9B, one-way, RM ANOVA; F_3,9 = 21.68, p<0.001). Control mice did not display any shifts in peak timing (data not shown; one-way RM ANOVA; F_2,8 = 0.7, p=0.5). We also quantified other characteristics of the peak distributions that have been shown to be affected by pharmacological manipulations of the dopaminergic system. We found that direct pathway stimulation reduced the peak width, but we did not observe a significant effect of stimulation time on the width (Figure 9C, two-way mixed ANOVA; main effect of stimulation, F_1,12 = 35.49, p<0.0001; no main effect of stimulation time, F_3,12 = 0.23, p=0.8; no interaction between stimulation time and stimulation, F_3,12 = 1.43, p=0.3). Direct pathway stimulation did not significantly impact the skew of the peak distributions (Figure 9D). These results reveal a critical role of direct pathway activation in resetting the internal timekeeping mechanism. This is the first report of a specific neural correlate of timer resetting. Interestingly, stimulation prior to reward did not impact the peak, suggesting that normally reward receipt may be involved in resetting the timing mechanism.

Figure 9

Download asset Open asset

We also tested the effect of indirect pathway stimulation during probe trials. As observed previously, stimulation of the indirect pathway at 5 s after reward resulted in a rebound in licking following the termination of stimulation. This was also the case during peak probe trials. Mice showed different types of responses following rebound licking. First, rebound licking would result in a brief delay in licking which was followed by a bout of licking resembling a peak response (‘recovery’ pattern, Figure 10A). Moreover, rebound licking could also initiate the peak, so that the peak lasts longer than usual (‘initiation’ pattern, Figure 10B). Occasionally mice also generated a peak without showing a clear rebound following stimulation (‘no-rebound’ pattern, Figure 10C). These different patterns (recovery, initiation, and no-rebound) were equally common (Figure 10D, n = 5, one-way RM ANOVA; F_2,8 = 0.6, p>0.05). Interestingly, peak analysis on these three licking patterns revealed distinct effects of indirect pathway stimulation on interval timing. For each pattern of licking, we compared the mean peak time with the peak time on trials in the same session that did not contain laser stimulation. During recovery-pattern trials, we found a large rightward shift of the peak that averaged approximately 2 s (Figure 10E, two-tailed t-test, t₄ = 6.5, p<0.01). In contrast, we found that when licking was initiated by the rebound licking bout, the peak was not affected (two-tailed t-test, t₄ = 2, p=0.1). On trials that lacked rebound licking, we found a rightward peak shift of approximately 750 ms (two-tailed t-test, t₄ = 4.35, p<0.05). Although initiation-type trials did not show a change in peak timing, we found that these peaks showed increased duration (Figure 10F, two-tailed t-test, t₄ = 4.18, p<0.05). We did not find a significant impact on the skew of the peaks in any trial type (Figure 10G).

Figure 10

Download asset Open asset

Finally, we tested whether behavior following indirect pathway stimulation was altered as the session progressed. The probability of the recovery pattern decreased as the session progressed, whereas the likelihood of detecting no-rebound patterns increased (Figure 10H, two-way, mixed ANOVA; interaction of the effects of trial type and session half, F_1,12 = 7.38, p<0.01). We did not detect a change in the probability of detecting the initiation-type pattern.

We found for the first time that optogenetic activation of direct-pathway neurons in the VLS can initiate licking (Figure 2), whereas activation of the indirect pathway in this region can terminate licking (Figure 4). These results are in accord with previous work implicating the VLS in orofacial behaviors (Mittler et al., 1994; Pisa, 1988a; Salamone et al., 1993). The observed effects are distinct from those following activation of the dorsolateral striatum, which produces locomotion and turning behavior (Bartholomew et al., 2016; Kravitz et al., 2010; Tecuapetla et al., 2014). The acute effects on licking is not surprising given the topographical organization of the sensorimotor striatum and our target in the ventrolateral orofacial region.

Although BG projections to lower brainstem pattern generators are not required for reflexive orofacial and tongue movements (Grill and Norgren, 1978), intact BG are required for goal-directed food seeking behavior (Bignall and Schramm, 1974). It has been shown that a subset of GABAergic projections from SNr to the superior colliculus is critical for the initiation and termination of self-paced licking behavior (Redgrave et al., 1992; Rossi et al., 2016; Toda et al., 2017). A pause in licking can be caused by increased activity in nigrotectal output (Rossi et al., 2016; Toda et al., 2017). The GABAergic output from the SNr can also directly innervate the reticular formation (Deniau and Chevalier, 1992), which contains regions that are known to be involved in the generation of licking behavior (Travers et al., 1997). Direct pathway activation may reduce SNr output and indirect pathway activation may increase SNr output. Consequently, brainstem or tectal targets that receive GABAergic SNr projections may be disinhibited by direct pathway activation and suppressed by indirect pathway activation. We previously showed that opponent classes of SNr projection neurons command downstream position control systems in the midbrain and brainstem to move the body in different directions (Barter et al., 2015b; Fan et al., 2012; Yin, 2014b). The present results suggest a critical role of the direct and indirect pathways in modulating the lower level controllers for orofacial behaviors.

Top-down regulation of lick pattern generator

A variety of evidence suggests that the output nuclei of the BG may function as an integrator (Barter et al., 2015b; Yin, 2014a; Yin, 2017). The rate of change in the integrator output is proportional to the input magnitude. With continuous control of position, a larger striatal signal (spikes per time window) is converted to faster rate of change in position control command (higher movement velocity) as well as greater movement amplitude (Barter et al., 2015a; Bartholomew et al., 2016). For licking behavior, we assume that the basal ganglia output also produces a top-down command from an integrator that influences the operation of the brainstem pattern generator. However, when the behavior in question is generated by pattern generators with relatively fixed innate rhythms, the mechanism is different. In this case, the BG output does not modulate the lick frequency in a continuous manner (Rossi et al., 2016), as it is not directly responsible for the generation of each lick. With continuous modulation, we would expect lick frequency to linearly scale with direct pathway stimulation frequency (Figure 11), but this was not observed. Once activated, the pattern generator appears to operate in several discrete settings, analogous to a fan with settings like off, low, and high. Normally, the lick pattern generator oscillates at 5–6 Hz for anticipatory licking and 6–8 Hz for consummatory licking. When the direct pathway was stimulated, licking could reach a new regime (~10 Hz) which appears to be the limit of the lick pattern generator and possibly also the biomechanical constraints of the orofacial musculature. When stimulation frequency exceeded 10 Hz, the licking frequency cannot increase further.

Figure 11

Download asset Open asset

BG output may regulate the activity of the lower level lick generators. The integrator is filled by direct pathway activation. Because direct pathway projections are inhibitory, the integrator is filled by inhibiting GABAergic nigral neurons, which results in a disinhibition of downstream structures. Greater activation of the direct pathway produces longer suprathreshold activation of the pattern generator (Figure 11B). This could be responsible for the greater licking frequencies that we observed (Figure 11C). The increased power of licking in a given frequency band, indicating longer ‘on time’ or higher duty cycle of the oscillator, suggests the presence of the integrator (Figures 3 and 11). We also observed decreasing latencies to lick onset during higher stimulation frequencies, and that licking could be sustained for some time after the end of stimulation (Figure 4). Artificially driving the direct pathway efficiently fills the integrator, which results in saturation and maximum licking rates even at relatively low stimulation frequencies. However, mice rarely enter this regime on their own, perhaps due to the concurrent activity of the indirect pathway.

The indirect pathway, on the other hand, may play an important role by discharging the integrator, preventing the instantaneous licking frequency from reaching its maximum level in natural licking. Selective activation of the indirect pathway is equivalent to rapidly reversing the BG output signal, bringing or keeping the pattern generators below their activation thresholds and pausing behavior. It is therefore only under conditions of artificially activating the direct and indirect pathways separately that behavior can be described as operating in an all-or-none, go/no-go fashion. Prior observations of coincident activity of the two pathways may reflect coordinated regulation of behavior to keep output within a certain range (DeLong, 1990; Tecuapetla et al., 2014). The balanced relationship between these two pathways may be responsible for the continuous adjustments to specific kinematic details of ongoing movement.

We propose that the BG provide top-down modulation of central pattern generators in the form of duty cycle regulation. Coordinated activity of the direct and indirect pathways are combined to produce a BG output that modulates how long a relatively stereotyped pattern generator is activated. This form of top-down regulation is distinct from the more continuous regulation of position controllers (Barter et al., 2015b; Yin, 2014b).

We observed a reliable rebound of licking following indirect pathway stimulation in the middle of the interval (Figure 6). This was only possible to detect because mice rarely lick at that time point, and the effect was not masked by the delivery of reward or the presence of consummatory licking. In addition, higher frequency stimulation resulted in shorter latency to rebound licking as well as more uniform timing in the start of rebound licking. Indirect pathway stimulation may result in more active nigral output, which results in a large suppression of downstream structures. The release of this inhibition then gives rise to a burst of rebound activity responsible for lick generation, resulting in fast reactivation of licking CPGs above the threshold, and a more consistent latency of the rebound licking.

The effects of direct and indirect pathway stimulation were sensitive to the motivational state, as the latencies to licking, overall licking rates, and the latency to rebound all changed over the course of the session (Figure 7). This could not be explained as simply a reduction in the efficacy of laser stimulation over the course of a single session because stimulation of the direct pathway at higher frequencies could override the effects of reducing motivation. These results can be explained by the possibility that, over time, the activation threshold for licking increases, blunting the influence of descending signals.

The effects of stimulation on licking varied in a manner that was dependent on the time in the task when the stimulation occurred. Direct pathway stimulation had the greatest impact on licking when mice were not licking, in the middle of the interval. On the other hand, stimulation during ongoing licking had a smaller effect. This could be explained by a ceiling effect, as mice could only boost their licking frequency so much once they were already performing the behavior. The opposite trend was observed with indirect pathway stimulation: stimulation at 5 s prior to reward showed a lower change in rate than stimulation during anticipatory licking.

All data generated in this study are included in the manuscript.

1. 1. Albin RL
   2. Young AB
   3. Penney JB

   (1989) The functional anatomy of basal ganglia disorders

   Trends in Neurosciences 12:366–375.

   https://doi.org/10.1016/0166-2236(89)90074-X

   • PubMed
   • Google Scholar
2. 1. Alexander GE
   2. DeLong MR

   (1985) Microstimulation of the primate neostriatum. II. somatotopic organization of striatal microexcitable zones and their relation to neuronal response properties

   Journal of Neurophysiology 53:1417–1430.

   https://doi.org/10.1152/jn.1985.53.6.1417

   • PubMed
   • Google Scholar
3. 1. Bakhurin KI
   2. Goudar V
   3. Shobe JL
   4. Claar LD
   5. Buonomano DV
   6. Masmanidis SC

   (2017) Differential encoding of time by prefrontal and striatal network dynamics

   The Journal of Neuroscience 37:854–870.

   https://doi.org/10.1523/JNEUROSCI.1789-16.2016

   • PubMed
   • Google Scholar
4. 1. Barter JW
   2. Li S
   3. Lu D
   4. Bartholomew RA
   5. Rossi MA
   6. Shoemaker CT
   7. Salas-Meza D
   8. Gaidis E
   9. Yin HH

   (2015a) Beyond reward prediction errors: the role of dopamine in movement kinematics

   Frontiers in Integrative Neuroscience 9:39.

   https://doi.org/10.3389/fnint.2015.00039

   • PubMed
   • Google Scholar
5. 1. Barter JW
   2. Li S
   3. Sukharnikova T
   4. Rossi MA
   5. Bartholomew RA
   6. Yin HH

   (2015b) Basal ganglia outputs map instantaneous position coordinates during behavior

   Journal of Neuroscience 35:2703–2716.

   https://doi.org/10.1523/JNEUROSCI.3245-14.2015

   • PubMed
   • Google Scholar
6. 1. Bartholomew RA
   2. Li H
   3. Gaidis EJ
   4. Stackmann M
   5. Shoemaker CT
   6. Rossi MA
   7. Yin HH

   (2016) Striatonigral control of movement velocity in mice

   European Journal of Neuroscience 43:1097–1110.

   https://doi.org/10.1111/ejn.13187

   • PubMed
   • Google Scholar
7. 1. Bignall KE
   2. Schramm L

   (1974) Behavior of chronically decerebrated kittens

   Experimental Neurology 42:519–531.

   https://doi.org/10.1016/0014-4886(74)90075-2

   • PubMed
   • Google Scholar
8. 1. Buhusi CV
   2. Meck WH

   (2002) Differential effects of methamphetamine and haloperidol on the control of an internal clock

   Behavioral Neuroscience 116:291–297.

   https://doi.org/10.1037/0735-7044.116.2.291

   • PubMed
   • Google Scholar
9. 1. Buhusi CV
   2. Meck WH

   (2005) What makes Us tick? functional and neural mechanisms of interval timing

   Nature Reviews Neuroscience 6:755–765.

   https://doi.org/10.1038/nrn1764

   • PubMed
   • Google Scholar
10. 1. Crowe DA
    2. Zarco W
    3. Bartolo R
    4. Merchant H

    (2014) Dynamic representation of the temporal and sequential structure of rhythmic movements in the primate medial premotor cortex

    Journal of Neuroscience 34:11972–11983.

    https://doi.org/10.1523/JNEUROSCI.2177-14.2014

    • PubMed
    • Google Scholar
11. 1. Cui G
    2. Jun SB
    3. Jin X
    4. Pham MD
    5. Vogel SS
    6. Lovinger DM
    7. Costa RM

    (2013) Concurrent activation of striatal direct and indirect pathways during action initiation

    Nature 494:238–242.

    https://doi.org/10.1038/nature11846

    • PubMed
    • Google Scholar
12. 1. Delfs JM
    2. Kelley AE

    (1990) The role of D1 and D2 dopamine receptors in oral stereotypy induced by dopaminergic stimulation of the ventrolateral striatum

    Neuroscience 39:59–67.

    https://doi.org/10.1016/0306-4522(90)90221-O

    • PubMed
    • Google Scholar
13. 1. DeLong MR

    (1990) Primate models of movement disorders of basal ganglia origin

    Trends in Neurosciences 13:281–285.

    https://doi.org/10.1016/0166-2236(90)90110-V

    • PubMed
    • Google Scholar
14. 1. Deniau JM
    2. Chevalier G

    (1992) The lamellar organization of the rat substantia nigra pars reticulata: distribution of projection neurons

    Neuroscience 46:361–377.

    https://doi.org/10.1016/0306-4522(92)90058-A

    • PubMed
    • Google Scholar
15. 1. Fan D
    2. Rossi MA
    3. Yin HH

    (2012) Mechanisms of action selection and timing in substantia nigra neurons

    Journal of Neuroscience 32:5534–5548.

    https://doi.org/10.1523/JNEUROSCI.5924-11.2012

    • PubMed
    • Google Scholar
16. 1. Frederick DL
    2. Allen JD

    (1996) Effects of selective dopamine D1- and D2-agonists and antagonists on timing performance in rats

    Pharmacology Biochemistry and Behavior 53:759–764.

    https://doi.org/10.1016/0091-3057(95)02103-5

    • PubMed
    • Google Scholar
17. 1. Freeze BS
    2. Kravitz AV
    3. Hammack N
    4. Berke JD
    5. Kreitzer AC

    (2013) Control of basal ganglia output by direct and indirect pathway projection neurons

    Journal of Neuroscience 33:18531–18539.

    https://doi.org/10.1523/JNEUROSCI.1278-13.2013

    • PubMed
    • Google Scholar
18. 1. Gerfen CR
    2. Engber TM
    3. Mahan LC
    4. Susel Z
    5. Chase TN
    6. Monsma FJ
    7. Sibley DR

    (1990) D1 and D2 dopamine receptor-regulated gene expression of striatonigral and striatopallidal neurons

    Science 250:1429–1432.

    https://doi.org/10.1126/science.2147780

    • PubMed
    • Google Scholar
19. 1. Gibbon J
    2. Church RM
    3. Meck WH

    (1984) Scalar timing in memory

    Annals of the New York Academy of Sciences 423:52–77.

    https://doi.org/10.1111/j.1749-6632.1984.tb23417.x

    • PubMed
    • Google Scholar
20. 1. Gouvêa TS
    2. Monteiro T
    3. Motiwala A
    4. Soares S
    5. Machens C
    6. Paton JJ

    (2015) Striatal dynamics explain duration judgments

    eLife 4:e11386.

    https://doi.org/10.7554/eLife.11386

    • PubMed
    • Google Scholar
21. 1. Grill HJ
    2. Norgren R

    (1978) The taste reactivity test. II. mimetic responses to gustatory stimuli in chronic thalamic and chronic decerebrate rats

    Brain Research 143:281–297.

    https://doi.org/10.1016/0006-8993(78)90569-3

    • PubMed
    • Google Scholar
22. 1. Hintiryan H
    2. Foster NN
    3. Bowman I
    4. Bay M
    5. Song MY
    6. Gou L
    7. Yamashita S
    8. Bienkowski MS
    9. Zingg B
    10. Zhu M
    11. Yang XW
    12. Shih JC
    13. Toga AW
    14. Dong HW

    (2016) The mouse cortico-striatal projectome

    Nature Neuroscience 19:1100–1114.

    https://doi.org/10.1038/nn.4332

    • PubMed
    • Google Scholar
23. 1. Isomura Y
    2. Takekawa T
    3. Harukuni R
    4. Handa T
    5. Aizawa H
    6. Takada M
    7. Fukai T

    (2013) Reward-modulated motor information in identified striatum neurons

    Journal of Neuroscience 33:10209–10220.

    https://doi.org/10.1523/JNEUROSCI.0381-13.2013

    • PubMed
    • Google Scholar
24. 1. Jicha GA
    2. Salamone JD

    (1991) Vacuous jaw movements and feeding deficits in rats with ventrolateral striatal dopamine depletion: possible relation to parkinsonian symptoms

    The Journal of Neuroscience 11:3822–3829.

    https://doi.org/10.1523/JNEUROSCI.11-12-03822.1991

    • PubMed
    • Google Scholar
25. 1. Kelley AE
    2. Lang CG
    3. Gauthier AM

    (1988) Induction of oral stereotypy following amphetamine microinjection into a discrete subregion of the striatum

    Psychopharmacology 95:556–559.

    https://doi.org/10.1007/BF00172976

    • PubMed
    • Google Scholar
26. 1. Kravitz AV
    2. Freeze BS
    3. Parker PRL
    4. Kay K
    5. Thwin MT
    6. Deisseroth K
    7. Kreitzer AC

    (2010) Regulation of parkinsonian motor behaviours by optogenetic control of basal ganglia circuitry

    Nature 466:622–626.

    https://doi.org/10.1038/nature09159

    • Google Scholar
27. 1. Mauk MD
    2. Buonomano DV

    (2004) The neural basis of temporal processing

    Annual Review of Neuroscience 27:307–340.

    https://doi.org/10.1146/annurev.neuro.27.070203.144247

    • PubMed
    • Google Scholar
28. 1. Meck WH

    (1996) Neuropharmacology of timing and time perception

    Cognitive Brain Research 3:227–242.

    https://doi.org/10.1016/0926-6410(96)00009-2

    • PubMed
    • Google Scholar
29. 1. Meck WH

    (2006) Neuroanatomical localization of an internal clock: a functional link between mesolimbic, Nigrostriatal, and mesocortical dopaminergic systems

    Brain Research 1109:93–107.

    https://doi.org/10.1016/j.brainres.2006.06.031

    • PubMed
    • Google Scholar
30. 1. Mello GB
    2. Soares S
    3. Paton JJ

    (2015) A scalable population code for time in the striatum

    Current Biology 25:1113–1122.

    https://doi.org/10.1016/j.cub.2015.02.036

    • PubMed
    • Google Scholar
31. 1. Merchant H
    2. Harrington DL
    3. Meck WH

    (2013) Neural basis of the perception and estimation of time

    Annual Review of Neuroscience 36:313–336.

    https://doi.org/10.1146/annurev-neuro-062012-170349

    • PubMed
    • Google Scholar
32. 1. Mink JW

    (1996) The basal ganglia: focused selection and inhibition of competing motor programs

    Progress in Neurobiology 50:381–425.

    https://doi.org/10.1016/S0301-0082(96)00042-1

    • PubMed
    • Google Scholar
33. 1. Mittler T
    2. Cho J
    3. Peoples LL
    4. West MO

    (1994) Representation of the body in the lateral striatum of the freely moving rat: single neurons related to licking

    Experimental Brain Research 98:163–167.

    https://doi.org/10.1007/BF00229122

    • PubMed
    • Google Scholar
34. Book

    1. Pavlov I

    (1927) Conditioned Reflexes

    Oxford: Oxford University Press.

    https://doi.org/10.1007/978-1-4684-3821-5_2

    • Google Scholar
35. 1. Pisa M

    (1988a) Motor functions of the striatum in the rat: critical role of the lateral region in tongue and forelimb reaching

    Neuroscience 24:453–463.

    https://doi.org/10.1016/0306-4522(88)90341-7

    • PubMed
    • Google Scholar
36. 1. Pisa M

    (1988b) Motor somatotopy in the striatum of rat: manipulation, biting and gait

    Behavioural Brain Research 27:21–35.

    https://doi.org/10.1016/0166-4328(88)90106-4

    • PubMed
    • Google Scholar
37. 1. Redgrave P
    2. Marrow L
    3. Dean P

    (1992) Topographical organization of the nigrotectal projection in rat: evidence for segregated channels

    Neuroscience 50:571–595.

    https://doi.org/10.1016/0306-4522(92)90448-B

    • PubMed
    • Google Scholar
38. 1. Roberts S

    (1981) Isolation of an internal clock

    Journal of Experimental Psychology: Animal Behavior Processes 7:242–268.

    https://doi.org/10.1037/0097-7403.7.3.242

    • Google Scholar
39. 1. Rossi MA
    2. Li HE
    3. Lu D
    4. Kim IH
    5. Bartholomew RA
    6. Gaidis E
    7. Barter JW
    8. Kim N
    9. Cai MT
    10. Soderling SH
    11. Yin HH

    (2016) A GABAergic nigrotectal pathway for coordination of drinking behavior

    Nature Neuroscience 19:742–748.

    https://doi.org/10.1038/nn.4285

    • PubMed
    • Google Scholar
40. 1. Rossi MA
    2. Yin HH

    (2015) Elevated dopamine alters consummatory pattern generation and increases behavioral variability during learning

    Frontiers in Integrative Neuroscience 09:37.

    https://doi.org/10.3389/fnint.2015.00037

    • Google Scholar
41. 1. Salamone JD
    2. Mahan K
    3. Rogers S

    (1993) Ventrolateral striatal dopamine depletions impair feeding and food handling in rats

    Pharmacology Biochemistry and Behavior 44:605–610.

    https://doi.org/10.1016/0091-3057(93)90174-R

    • PubMed
    • Google Scholar
42. 1. Shammah-Lagnado SJ
    2. Costa MS
    3. Ricardo JA

    (1992) Afferent connections of the parvocellular reticular formation: a horseradish peroxidase study in the rat

    Neuroscience 50:403–425.

    https://doi.org/10.1016/0306-4522(92)90433-3

    • PubMed
    • Google Scholar
43. 1. Tecuapetla F
    2. Matias S
    3. Dugue GP
    4. Mainen ZF
    5. Costa RM

    (2014) Balanced activity in basal ganglia projection pathways is critical for contraversive movements

    Nature Communications 5:5315.

    https://doi.org/10.1038/ncomms5315

    • Google Scholar
44. 1. Toda K
    2. Lusk NA
    3. Watson GDR
    4. Kim N
    5. Lu D
    6. Li HE
    7. Meck WH
    8. Yin HH

    (2017) Nigrotectal stimulation stops interval timing in mice

    Current Biology 27:3763–3770.

    https://doi.org/10.1016/j.cub.2017.11.003

    • PubMed
    • Google Scholar
45. 1. Travers JB
    2. Dinardo LA
    3. Karimnamazi H

    (1997) Motor and premotor mechanisms of licking

    Neuroscience & Biobehavioral Reviews 21:631–647.

    https://doi.org/10.1016/S0149-7634(96)00045-0

    • PubMed
    • Google Scholar
46. 1. Wang J
    2. Narain D
    3. Hosseini EA
    4. Jazayeri M

    (2018) Flexible timing by temporal scaling of cortical responses

    Nature Neuroscience 21:102–110.

    https://doi.org/10.1038/s41593-017-0028-6

    • PubMed
    • Google Scholar
47. 1. Yin HH

    (2014a) Action, time and the basal ganglia philosophical transactions of the royal society B

    Biological Sciences 369:20120473.

    https://doi.org/10.1098/rstb.2012.0473

    • Google Scholar
48. 1. Yin HH

    (2014b) How basal ganglia outputs generate behavior

    Advances in Neuroscience 2014:1–28.

    https://doi.org/10.1155/2014/768313

    • Google Scholar
49. 1. Yin HH

    (2017) The basal ganglia in action

    The Neuroscientist 23:299–313.

    https://doi.org/10.1177/1073858416654115

    • PubMed
    • Google Scholar

Author details

1. Konstantin I Bakhurin

   Department of Psychology and Neuroscience, Duke University, Durham, United States

   Contribution

   Conceptualization, Formal analysis, Supervision, Investigation, Writing - original draft, Project administration, Writing - review and editing

   Contributed equally with

   Xiaoran Li

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3660-4343

2. Xiaoran Li

   Department of Psychology and Neuroscience, Duke University, Durham, United States

   Contribution

   Conceptualization, Formal analysis, Supervision, Investigation, Writing - original draft, Writing - review and editing

   Contributed equally with

   Konstantin I Bakhurin

   Competing interests

   No competing interests declared

3. Alexander D Friedman

   Department of Psychology and Neuroscience, Duke University, Durham, United States

   Contribution

   Conceptualization, Investigation, Writing - original draft

   Competing interests

   No competing interests declared

4. Nicholas A Lusk

   Department of Psychology and Neuroscience, Duke University, Durham, United States

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

5. Glenn DR Watson

   Department of Psychology and Neuroscience, Duke University, Durham, United States

   Contribution

   Data curation, Formal analysis

   Competing interests

   No competing interests declared

6. Namsoo Kim

   Department of Psychology and Neuroscience, Duke University, Durham, United States

   Contribution

   Data curation, Formal analysis

   Competing interests

   No competing interests declared

7. Henry H Yin

   1. Department of Psychology and Neuroscience, Duke University, Durham, United States
   2. Department of Neurobiology, Duke University School of Medicine, Durham, United States

   Contribution

   Conceptualization, Formal analysis, Supervision, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   hy43@duke.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-1546-6850

• 1,682

  Page views

• 243

  Downloads

• 10

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.