Sugar exerts its potent reinforcing effects via both gustatory and post-ingestive pathways. It is, however, unknown whether sweetness and nutritional signals engage segregated brain networks to motivate ingestion. We found in mice that separate basal ganglia circuitries mediated the hedonic and nutritional actions of sugar. During sugar intake, suppressing hedonic value inhibited dopamine release in ventral, but not dorsal, striatum, whereas suppressing nutritional value inhibited dopamine release in dorsal, but not ventral, striatum. Consistently, cell-specific ablation of dopamine-excitable cells in dorsal, but not ventral, striatum inhibited sugar's ability to drive the ingestion of unpalatable solutions. Conversely, optogenetic stimulation of dopamine-excitable cells in dorsal, but not ventral, striatum substituted for sugar in its ability to drive the ingestion of unpalatable solutions. Our data indicate that sugar recruits a distributed dopamine-excitable striatal circuitry that acts to prioritize energy-seeking over taste quality.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Nature Communications Open Access 03 June 2022
Experimental & Molecular Medicine Open Access 26 April 2022
Nature Neuroscience Open Access 13 January 2022
Subscribe to Journal
Get full journal access for 1 year
only $8.25 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Holman, G.L. Intragastric reinforcement effect. J. Comp. Physiol. Psychol. 69, 432–441 (1969).
Sclafani, A. Post-ingestive positive controls of ingestive behavior. Appetite 36, 79–83 (2001).
de Araujo, I.E. et al. Food reward in the absence of taste receptor signaling. Neuron 57, 930–941 (2008).
Zuker, C.S. Food for the brain. Cell 161, 9–11 (2015).
Stephenson-Jones, M., Kardamakis, A.A., Robertson, B. & Grillner, S. Independent circuits in the basal ganglia for the evaluation and selection of actions. Proc. Natl. Acad. Sci. USA 110, E3670–E3679 (2013).
Everitt, B.J. & Robbins, T.W. Neural systems of reinforcement for drug addiction: from actions to habits to compulsion. Nat. Neurosci. 8, 1481–1489 (2005).
Gerfen, C.R. & Surmeier, D.J. Modulation of striatal projection systems by dopamine. Annu. Rev. Neurosci. 34, 441–466 (2011).
Costa, R.M. Plastic corticostriatal circuits for action learning: what's dopamine got to do with it? Ann. NY Acad. Sci. 1104, 172–191 (2007).
Grillner, S., Robertson, B. & Stephenson-Jones, M. The evolutionary origin of the vertebrate basal ganglia and its role in action selection. J. Physiol. (Lond.) 591, 5425–5431 (2013).
Yin, H.H., Ostlund, S.B. & Balleine, B.W. Reward-guided learning beyond dopamine in the nucleus accumbens: the integrative functions of cortico-basal ganglia networks. Eur. J. Neurosci. 28, 1437–1448 (2008).
Wickens, J.R., Horvitz, J.C., Costa, R.M. & Killcross, S. Dopaminergic mechanisms in actions and habits. J. Neurosci. 27, 8181–8183 (2007).
Palmiter, R.D. Dopamine signaling in the dorsal striatum is essential for motivated behaviors: lessons from dopamine-deficient mice. Ann. NY Acad. Sci. 1129, 35–46 (2008).
Ren, X. et al. Nutrient selection in the absence of taste receptor signaling. J. Neurosci. 30, 8012–8023 (2010).
Roitman, M.F., Wheeler, R.A. & Carelli, R.M. Nucleus accumbens neurons are innately tuned for rewarding and aversive taste stimuli, encode their predictors, and are linked to motor output. Neuron 45, 587–597 (2005).
Hajnal, A. & Norgren, R. Taste pathways that mediate accumbens dopamine release by sapid sucrose. Physiol. Behav. 84, 363–369 (2005).
Ericsson, J. et al. Dopamine differentially modulates the excitability of striatal neurons of the direct and indirect pathways in lamprey. J. Neurosci. 33, 8045–8054 (2013).
Planert, H., Berger, T.K. & Silberberg, G. Membrane properties of striatal direct and indirect pathway neurons in mouse and rat slices and their modulation by dopamine. PLoS One 8, e57054 (2013).
Yang, C.F. et al. Sexually dimorphic neurons in the ventromedial hypothalamus govern mating in both sexes and aggression in males. Cell 153, 896–909 (2013).
Glendinning, J.I., Gresack, J. & Spector, A.C. A high-throughput screening procedure for identifying mice with aberrant taste and oromotor function. Chem. Senses 27, 461–474 (2002).
Kravitz, A.V., Tye, L.D. & Kreitzer, A.C. Distinct roles for direct and indirect pathway striatal neurons in reinforcement. Nat. Neurosci. 15, 816–818 (2012).
Lobo, M.K. et al. Cell type-specific loss of BDNF signaling mimics optogenetic control of cocaine reward. Science 330, 385–390 (2010).
Cui, G. et al. Concurrent activation of striatal direct and indirect pathways during action initiation. Nature 494, 238–242 (2013).
Johnson, P.M. & Kenny, P.J. Dopamine D2 receptors in addiction-like reward dysfunction and compulsive eating in obese rats. Nat. Neurosci. 13, 635–641 (2010).
DeLong, M.R., Crutcher, M.D. & Georgopoulos, A.P. Relations between movement and single cell discharge in the substantia nigra of the behaving monkey. J. Neurosci. 3, 1599–1606 (1983).
Wurtz, R.H. & Hikosaka, O. Role of the basal ganglia in the initiation of saccadic eye movements. Prog. Brain Res. 64, 175–190 (1986).
Sternson, S.M. & Roth, B.L. Chemogenetic tools to interrogate brain functions. Annu. Rev. Neurosci. 37, 387–407 (2014).
Dus, M., Min, S., Keene, A.C., Lee, G.Y. & Suh, G.S. Taste-independent detection of the caloric content of sugar in Drosophila. Proc. Natl. Acad. Sci. USA 108, 11644–11649 (2011).
Burke, C.J. et al. Layered reward signalling through octopamine and dopamine in Drosophila. Nature 492, 433–437 (2012).
Yamagata, N. et al. Distinct dopamine neurons mediate reward signals for short- and long-term memories. Proc. Natl. Acad. Sci. USA 112, 578–583 (2015).
O'Doherty, J.P. The problem with value. Neurosci. Biobehav. Rev. 43, 259–268 (2014).
Shammah-Lagnado, S.J., Costa, M.S. & Ricardo, J.A. Afferent connections of the parvocellular reticular formation: a horseradish peroxidase study in the rat. Neuroscience 50, 403–425 (1992).
This work was supported by US National Institutes of Health grants R01DC014859 and R01CA180030 (to I.E.d.A.), and R01 DK103176, DK084052 and NS48476 (to A.N.v.d.P.), the China Scholarship Council 201206260072 (to W.H.) and FAPESP (Sao Paulo) 2013/09405-3 (to T.L.F.).
The authors declare no competing financial interests.
Integrated supplementary information
Supplementary Figure 1 Microdialysis measurements of extracellular dopamine in ventral and dorsal striatum.
A. Behavioural preparation. Mice are placed in an operant box where a spout containing the artificial sweetener sucralose is connected to a contact lickometer. Lick detection triggers an infusion pump that delivers either glucose or more sucralose into the stomach. Brain microdialysates were collected concomitantly to behavioural performance. B. Histological analyses of brain tissue obtained from the animals that performed the experiments described above. Representative cases for ventral (top) and dorsal (bottom) striatum are shown along with schematic representation of the location of microdialysis probes (VS in red and DS in blue). Data from the microdialysis sessions shown in Figure 1 were replotted to allow for direct comparison of the effects produced by the different stimuli within each striatal sector (in main Figure 1 comparisons were performed between sectors). As before blue stands for DS data and red for VS data. Baseline sampling was performed for 30 min prior to introduction of stimulus (green vertical arrows). C-D. Mice licked sippers containing either sweet (sucralose) or bitter (sucralose + denatonium benzoate) solutions such that in both cases detected licks triggered D-glucose intra-gastric infusions. In C, VS levels were significantly higher during sweet (N=6) intake compared to bitter (N=8) intake (group effect F[1,10]=30.05, *p=0.0003) indicating that taste quality regulates dopamine efflux in VS. Comparisons against baseline levels (each condition involves Taste+Intra-gastric): Sweet+D-Glucose: two-way repeated-measures ANOVA F[9,36]=9.02 Bonferroni p=0.005; Bitter+D-Glucose: F[9,54]=4.45 Bonferroni p=0.0058. In D, both sweet (N=6) and bitter (N=7) intake coupled to D-glucose intra-gastric infusions produced similar increases in dopamine release (group effect F[1,12]=0.03, p=0.86), suggesting that changes in dorsal dopamine levels respond to metabolic consequences. Comparisons against baseline levels: Sweet+D-Glucose F[9,45]=4.5 p=0.0029; Bitter+D-Glucose F[9,63]=4.6 Bonferroni p=0.0057. E-F. Mice licked sucralose-containing sippers such that detected licks triggered intra-gastric infusions of either D-glucose or L-glucose (the nonmetabolizable glucose enantiomer). In E, as expected, similar increases in VS dopamine levels were observed during both L-glucose (N=6) and D-glucose (N=5) sessions (group effect F[1,9]=0.24, p=0.88). Comparison against baseline levels: Sweet+L-Glucose F[9,45]=6.09 Bonferroni p=0.0055. In F, DS dopamine levels were significantly higher during D-glucose (N=6) compared to L-glucose sessions (N=6, group effect F[1,10]=6.77, *p=0.027). Comparison against baseline levels: Sweet+L-Glucose F[9,45]=2.47 Bonferroni p=0.019. G. Correlation between dopamine release levels and corresponding lick counts during the equivalent sampling period. The data show that there was no relationship between lick counts and dopamine release levels in VS, ruling out thus the possibility that bitter-driven suppressions in licking may mediate decreased dopamine efflux (F[1,136]=3.7, p= 0.0565). H. Same analyses for DS (F[1,155]=0.49, p=0.48). I-J. Glucose or sucralose were intra-gastrically infused while animals remained in their home cages, i.e. outside any behavioral context. To mimic infusions observed during behavioral sessions, time stamps of infusions pump triggers were “replayed” during these sessions. The rasters indicate the time of an infusion of a 30μL bolus into the stomach. Glucose infusions produced significantly greater increases in both VS (N=5, interaction sampling time × infusate F[14,56]=3.6, * p=0.0003, I) and DS (F[14,56]=5.5, * p= 0.000002, J) compared to sucralose. These results show in particular that licks are not required for dopamine release during sugar intake, consistent with panels G-H above. In fact, and rather remarkably, sugar-induced dopamine effluxes in animals treated with passive infusions were statistically similar to those observed in animals exposed to the behavioral sessions:
Glucose VS (N=5 both groups): group effect F[1,8]=0.8, p=0.39
Glucose DS (behavioral sessions N=6 and passive infusions N=5): group effect F[1,9]=0.02, p=0.89
Sucralose DS (behavioral sessions N=6 and passive infusions N=5): group effect F[1,9]=0.37, p=0.56.
As anticipated, the only exception refers to when passive intra-gastric sucralose is compared to sucralose behavioral sessions (N=5 both groups, group effect F[1,8]=8.56, p=0.019). This is consistent with our assumption that VS dopamine efflux is stimulated by sweetness in the absence of energy intake. n.s. = non-statistically significant.
Supplementary Figure 2 Controls for non-specific effects produced by intra-gastric sugar administration.
A. Behavioural preparation for measurements of dopamine release during sweet taste intake using microdialysis. Mice are placed in an operant box where a spout containing the artificial sweetener sucralose is connected to a contact lickometer. Lick detection triggers an infusion pump that delivers either glucose or more sucralose into the stomach. Brain microdialysates were collected concomitantly to behavioural performance. B. Dopamine measurements were performed in animals exposed to 14 consecutive days of exposure to sucralose licking associated with D-glucose infusions. Higher dorsal striatal dopamine levels were observed during sucralose licking upon intra-gastric infusions of glucose (glucose-exposed mice, N=6) compared to sucralose licking upon intra-gastric infusions of sucralose (sucralose-exposed mice, N=6, Two-way mixed model ANOVA, Sampling time × group effect F[6,60]=3.2, * p=0.008). Graph displays changes in DA during the 1h intake session after 30min baseline sampling. Raster plot shows lick rates for each (sucralose vs. glucose) session. Onset of licking is shown by upward green arrow. DA = dopamine. When a direct comparison was performed between naïve and experienced animals, we found no statistical differences in terms of sugar- or sucralose-induced dopamine efflux between the two groups, as follows: Glucose-induced changes in DS DA in naïve vs. experienced groups (N=6 both groups): group effect F[1,10]=0.001, p=0.97. Sucralose-induced changes in DS DA in naïve vs. experienced groups (N=6 both groups): group effect F[1,10]=0.54, p=0.47. C. Open field tests reveal no effects of intra-gastric infusates on locomotor activity in a novel arena (number of crossings through the arena’s subregions, N=5 in each group, two-sample t-test t=1.34, p=0.21). D. Neither were detected differences in time spent within the illuminated central area of the arena (t=1.74, p=0.12). E. Mice (N=10) were trained on a goal-directed task in which nose pokes on the active hole produce the delivery of palatable food pellets. Previous to testing on this task, animals were infused with an intra-gastric preload of either D-glucose or sucralose in a randomized within-subjects design. Mice learned reliably to poke the active vs. the inactive hole (F[1,9]=117.79, p=0.0000018). As anticipated, glucose preloads, when compared to sucralose preloads, significantly reduced the number of active pokes (F[1,9]=41.89, p=0.000115). Post-hoc t-tests confirm that animals were significantly more active during this goal-directed task after sucralose compared to glucose preloads (Paired sample t-test, Pokes on active hole: t=6.4, p=0.0001; Pokes on inactive hole: t=3.3, p=0.009). F. However, overall preferences for the active vs. inactive hole were not affected by the content of the preload (paired-sample t-test, t=0.37, p=0.71). G. As a consequence of greater activity after sucralose compared to glucose preloads, number of rewards obtained were greater in the former compared to the latter case (t=9.0, p=0.0000085). TD=Training Day.
Supplementary Figure 3 Effects of artificially increasing dopamine levels or ablating dopamine-excitable neurons in striatum
A. Schematic representation of the behavioural preparation where mice licked bitter-containing sippers such that detected licks triggered intra-gastric infusions of either glucose or sucralose. B. Hungry mice (N=8) licked the bitter solution significantly more when self-infusing glucose compared to when self-infusing sucralose (t=2.62, * p=0.035), resulting in significantly larger intra-gastric glucose volumes (not shown, t=3.17, p=0.016). C. Reverse microdialysis was used to perfuse DS or VS with dopamine during ingestion of sucralose or a bitter solution. D. Dopamine perfusion in DS (N=6) and VS (N=5) resulted in increases in sucralose intake when compared to aCSF perfusions (Left panel; Perfusion main effect F[1,9]=16.76, p=0.003) The effect was similar in both VS and DS (Two-way RM-ANOVA, striatal region × brain perfusion F[1,9]=3.43, p=0.097; dopamine vs. aCSF in DS, t=5.07, Bonferroni * p=0.008; in VS effect was weaker: t=1.33, p=0.51). This is consistent with sweetness-driven dopamine efflux in VS but not DS. Dopamine perfusion in DS and VS resulted in robust increases in bitter intake when compared to aCSF perfusions (Right panel; Perfusion main effect F[1,9]=27.55, p=0.001). Effect was similarly robust in both VS and DS (Two-way RM-ANOVA, striatal region × brain perfusion F[1,9]=0.43, p=0.53; dopamine vs. aCSF in DS, t=3.6, Bonferroni *p=0.03; VS: t=3.7, ** p=0.04). E. Cell-specific ablation of D1r-neurones in DS or VS. Top panel: Brief access test for different sucralose concentrations. [Two-way RM-ANOVA, sweetness × group F[6,57]=8.23, p=0.000002; group effect F[2,19]=1.2, p=0.32;; sweetness effect F[3,57]=95.8, p= 3x10-22]. Cell-specific ablation of D1r-neurones in VS, but not DS, produced a lower intake of 2 mM sucralose [One-way ANOVA group effect F[2,21]=5.95, p=0.01. Ventral lesion vs. Control, Bonferroni *p=0.04; and vs. DS lesion, Bonferroni * p=0.014], whereas at 6mM sucralose VS Casp intake was higher [One-way ANOVA group effect F[2,21]=5.94, p=0.01; ventral lesion vs. WT-Casp Bonferroni #p=0.012 and Ventral lesion vs. DS-Casp Bonferroni p=0.055]. Bottom panel: Masking bitterness. The concentration of the bitter compound denatonium was fixed at 6mM and different concentrations of sucralose were used. [Two-way RM-ANOVA, sweetener concentration × group F[6,57]=5.29, p=0.00021; group effect F[2,19]=1.95, p=0.17; sweetener concentration effect F[3,57]=67.32,, p=9.9x10-19]. Cell-specific ablation of D1r-neurones in VS, but not DS, produced a lower intake of the mixture 6mM denatonium and 6mM sucralose [One-way ANOVA group effect F[2,21]=7.55, p=0.004. Ventral lesion vs. Control, Bonferroni **p=0.012; and vs.DS lesion, Bonferroni ** p=0.007]. F. Effects of cell-specific ablation of D1r-neurones in DS or VS on sugar-driven consumption of unpalatable solutions. Top panel: Licks produced during the ingestion of a bitter mixture that triggers intra-gastric infusions of different D-glucose concentrations. [Two-way RM-ANOVA, glucose concentration × group F[6,57]=2.68, p=0.023; group effect F[2,19]=2.75, p=0.089; glucose concentration effect F[3,57]=25.8, p=1.3x10-10]. Bottom panel: Intra-gastric infusions observed during these sessions [Two-way RM-ANOVA, glucose concentration × group F[6,57]=8.83, p=8.2x10-7; group effect F[2,19]=4.18, p=0.031; glucose concentration effect F[3,57]=45.87, p=3.3x10-15]. Concentrations: 0.5% [One-way ANOVA group effect F[2,21]=7.07, p=0.005; ventral lesion vs. Control, Bonferroni *p=0.007; and vs.DS lesion, Bonferroni * p=0.023];: 10% [One-way ANOVA group effect F[2,21]=7.09, p=0.005; intake in dorsal lesion vs. Control, Bonferroni **p=0.007; and vs.VS lesion, Bonferroni ** p=0.024]; 25% [One-way ANOVA group effect F[2,21]=5.75, p=0.011; dorsal lesion vs. Control, Bonferroni #p=0.034; and vs.VS lesion, Bonferroni #p=0.018]; 50% [One-way ANOVA group effect F[2,21]=2.16, p=0.142]. G. Licks (top) and infusions (bottom) produced during the conditioning sessions (bitter intake paired with intra-gastric infusions of glucose) prior to the second two-bottle test. Cell-specific ablation of D1r-neurones did not alter intake. [Licks: One-way ANOVA Group Effect F[2,21]=2.27, p=0.13; Infusions: One-way ANOVA Group Effect F[2,21]=1.11, p=0.34]. H-K. Neuroanatomical analyses of the sham cell-specific lesions. When Retrobeads were injected into globus pallidus (GP, targeted by D2r-expressing neurons of DS) of DS-CTL mice (area within dotted line in H), strong labelling was observed in DS (I). Similarly, robust labelling was observed in DS (J) when Retrobeads were injected into the contralateral SNr (K), which is exclusively targeted by D1r-expressing neurons of DS (confront vs. Main Figure 2 in which lesioned case is shown). To allow visualization of the relevant anatomical landmarks, images show the Retrobead fluorescence signal overlaid on a bright field image of the same section.
Abbreviations: DS-Casp: Caspase-driven D1r-dependent lesions in DS of D1r-Cre mice (N=7); VS-Casp: Caspase-driven D1r-depedent lesions in VS of D1r-Cre mice (N=7); WT-Casp: Viral delivery of Cre-dependent caspase in DS and/or VS of non-Cre mice (N=8). n.s. = non-statistically significant.
A-C ChR2 expression in striatal D1r-neurones. A. Sagital slice showing D1r-neuronal direct pathway from DS to Substantia Nigra, pars reticulata. B. Magnification of area shown in red region containing D1r-postive EYFP labelled axonal fibres originating in DS and descending towards midbrain. C. Magnification of area shown in red region containing dense terminal fibres from DS in Substantia Nigra, pars reticulata (dotted white line). D-F Slice electrophysiological studies of two D1r-neurones types (equally found in DS and VS) after injecting the construct AAV-EF1a-DIO-hChR2(H134R)-EYFP in striatum of D1-Cre mice D. Representative traces showing one type DS D1r-neuron which responded to blue light pulses (10ms) delivered at various frequencies, as well as to continuous light. However only continuous blue light trains produced sustained increases in firing rate. E. Representative traces showing activation of another type of D1r-neurone by blue light pulses (10ms) delivered at various frequencies and by continuous light trains. Both short pulses at various frequencies and continuous light trains activated these D1r-neurones. However, continuous blue light still produced the stronger excitatory effect. F. A representative trace showing long-lasting activation of a D1r-neurone by continuous blue light activation for 1 minute.
Supplementary Figure 5 Optogenetic activation of striatal D2r-expressing neurons during sweetener intake.
Channelrhodopsin expression in D2-Cre mice after injecting the construct AAV-EF1a-DIO-hChR2(H134R)-EYFP as visualized in a saggital diagram of the mouse brain A. The construct was injected in DS of a D2-Cre mouse. Contrast with injections performed in D1-Cre mice shown in Main Figure 3. As expected, unlike the D1-Cre case, in D2-Cre mice no terminals were visualized in Substantia Nigra reticulata (SNr), whereas they were densely expressed in the Globus Pallidus (GP), revealing thus the prototypical D2-dependent indirect pathway. B. Results of injecting the virus in VS. EYFP expression is robust yet totally contained within the nucleus accumbens (Acc) in VS. Note dense VS terminal fibres in ventral pallidum (VP), likewise the D1-Cre case. Importantly, no expression was detected in mesencephalic regions, showing that the AAV-EF1a-DIO-hChR2(H134R)-EYFP did express in striatal D2r-positive cells. C. Effects of inhibiting D2r-positive neurons in VS by injecting the construct AAV-EF1a-DIO-eArch3.0-EYFP in D2-Cre mice. Although D2r-neurone inhibition in VS slightly increased licks for sweet sucralose (N=5, paired t-test t=3.43, * p=0.026), it did not attenuate the aversive effects produced by adulterating the sweet solution with a bitter toxin. D. Effects of inhibiting D2r-positive neurons in DS. Although D2r-neurone inhibition in DS did not impact on the numbers of licks for sweet sucralose, it did significantly attenuate the aversive effects produced by adulterating the sweet solution with a bitter toxin (N=5, t=9.14, ** p=0.001). However, the intake levels remained well below those observed for the sweet solution so that, unlike stimulation of D1r-neurones, inhibition of D2r-neurones does not annul the aversive effects produced by adulterating the sweet solution with a bitter taste.
A. D1-neurone dorsal striatum (DS) stimulation had no effect on bitter taste perception, nor produced indiscriminate licking. A short-term two-bottle preference test involving the choice between one sweet sucralose and one adulterated bitter solution was performed during D1-neurone stimulation in DS. Both groups, D1-ChR2 (N=6) and Control (N=5) strongly preferred sweet versus bitter solution (one-sample t-test against indifference ratio of 0.5 [red dashed line] both * p=0.001) and no between-group difference was detected (t=1.91, p=0.09). n.s.= statistically non-significant. B. Increasing the duration of licks for sweetener-triggered optical stimulation did not alter the total amount of licks produced (N=6, F[3,15]=1.36, p=0.29). Y-axis represents the average number of licks produced while the laser source was on. C-D. The effects of D1 stimulation of either ventral (VS, N=6) or dorsal (DS, N=5) striatum on glucose self-administration were evaluated using ChR2+ D1-cre mice. First, baseline intake (licks and volumes) was established during a Laser OFF session, such that detected licks triggered glucose intra-gastric infusions. Then, on a subsequent test session, detected licks triggered glucose intra-gastric infusions coupled to laser activation. C. Significant effects of light activation on lick rates (F[1,9]=8.95, p=0.015) were observed. D1-neurone activation in VS significantly increased licking compared to baseline light OFF (t=5.99, Bonferroni * p=0.006), whereas no significant difference was observed for the D1-dorsal group (t=1.25, p=0.28). D. Significant effects were also observed on volumes of glucose self-infused (F[1,9]=9.67, p=0.013). Both groups mice self-administered significantly more glucose during laser ON sessions (VS, t=10.73, Bonferroni p=0.0024; DS, t=9.92, Bonferroni * p=0.003); however, D1-neurone DS activation produced a more marked increase in volumes self-infused when directly compared to D1-neurone VS stimulation (t=6.06, p=0.0002). The fact that DS stimulation increased volume infusions without increasing lick rates is due to the fact that, during laser/infusion activation, additional licks had no programmed consequences (see also Methods). Therefore D1-neurone DS stimulation was associated with greater gut, but not oral, stimulation. E. Intra-gastric volumes infused in sessions during which sucralose licks triggered intra-gastric infusions of either sucralose or glucose, concomitantly to optical stimulation. The graph shows the sessions in which laser source was ON. For both DS (N=7) and VS (N=6), intra-gastric infusions of glucose were associated with significantly lower number of self-infusions (two-way ANOVA, main infusate effect F[1,11]=193.21, * p=0.00000003; striatal region × brain perfusion F[1,11]=3.24, p=0.099). However, D1-neurone DS optical stimulation did produce greater self-infused volumes compared to VS stimulation (two-sample t-test t=5.41, p=0.00021) although this did not hold for sucralose sessions (t=0.26, p=0.79). F-H. The reinforcing properties of D1 stimulation at either VS (N=6) or DS (N=5) were assessed using a self-stimulation Nose Poke paradigm. Mice were placed in an operant box equipped with two slots for nose poking at symmetrical locations on one of the cage walls. One slot was associated with laser pulses upon poking (active slot) while poking the second slot had no programmed consequences. Two training sessions performed on two consecutive days, plus one subsequent extinction (laser off) session on a third day, were completed. F. On the first training session, D1-stimulation in both VS and DS produced significantly more responses on the active side (F[1,9]=10.8, p=0.009), with no significant differences between groups in self-stimulatory rates being detected (group effect F[1,9]=2.69, p=0.13; t-test for the active side t=1.71, Bonferroni p=0.121). G. On the second day of training, as above, D1-stimulation in both VS and DS produced significantly more responses on the active side (F[1,9]=18.71, p=0.002), and no differences between these two groups was found (group effect F[1,9]=2.98, p=0.12; t-test for the active side t=1.77, p=0.11). These results suggest that ventral and dorsal D1-stimulations are equally reinforcing under these self-stimulatory conditions. Interestingly, however, during the extinction session shown in H, significant effects associated with neither side (F[1,9]=4.78, p=0.057) nor group (F[1,9]=2.69, p=0.13 t-test for the active side t=1.67, p=0.13) were found. Control group N=6. Note that controls produced remarkably low stimulatory rates throughout the sessions. I. Open field tests reveal no effects of DS optical stimulation on locomotor activity in a novel arena (number of crossings through the arena’s subregions, N=5, two-sample t-test t=0.87, p=0.43). J. Neither were detected differences in relative time spent within the illuminated central part of the arena (paired-samples t-test t=2.03, p=0.11). K. No effects of non-contingent DS optical stimulation on chow intake were observed in either hungry or sated states, (N=4, two-way repeated-measures ANOVA, main effect of laser source F[1,3]=1.61, p=0.29; laser source × hunger state F[1,3]=2.95, p=0.18) despite clear effects of hunger (F[1,3]=47.82, * p=0.006).
Supplementary Figure 7 Interfering with neural activity downstream of dorsal striatum selectively suppresses sugar ingestion.
A. Schematic representation of optical fibre position in ventral pallidum (VP) and B. Substantia nigra, pars reticulata (SNr). C. Optogenetic activation of VP strongly suppresses artificial sweetener intake (N=6, Laser ON vs. OFF effect, paired t-test t=13.1, Boferroni *p=0.0018). D. No effects are observed upon activation of SNr (N=5, t=0.58, p=0.59). E. When a novel artificial sweetener (Rebaudioside A) is paired to light, low intake levels are observed also during subsequent Laser OFF sessions (N=6, t=2.75, Bonferroni p=0.1). F. Again no effects are observed upon activation of SNr (N=5, t=0.96, p=0.39). Importantly note that when licks shown in E (i.e. upon optical stimulation of VP) are compared to those shown in F (i.e. upon optical stimulation of SNr), a strong reduction in Rebaudioside A intake is observed during VP compared to SNr stimulation (two-way repeated-measures ANOVA laser source × brain region: F[1,9]=7.09, p=0.026; Main brain region effect F[1,9]=55.25, p=0.00004; Comparison during laser ON sessions: two-sample t-test t=5.97, p=0.00021). G. Animals are treated with a glucose intra-gastric preload previous to access to the sucralose solution. Under these conditions optogenetic activation of VP weakly suppresses intake (N=6). t=2.63, Bonferroni *p=0.046, one-tailed). H. However, optogenetic activation of SNr led to suppressed intake when preceded by sugar gut infusions (N=5). t=4.04, Bonferroni *p=0.032). I. When a sucralose gastric preload is used instead, VP activations leads to strong suppression (N=6). t=7.57, Bonferroni *p=0.002). J. As expected, no effects upon SNr activation are observed when sucralose intra-gastric preloads are used (N=6). t=0.009, p=0.99).
Supplementary Figure 8 Combined optogenetic and chemogenetic activation: schematic of experimental design
A. One group of animals (“VS=>VP pathway”) was ChR2-tranfected in VS, and optical fibres placed immediately above the D1r-neurone terminals in VP. B. A second group of animals (“DS=>SNr pathway”) was ChR2-tranfected in DS, and optical fibres placed immediately above the D1r-neurone terminals in SNr. In both groups, hM3D(Gq) was expressed in VP. C. To assess pathway-specificity, one additional group of animals expressed hM3D(Gq) in SNr instead of VP.
Supplementary Figure 9 Analysis of the effects in vivo of optogenetically stimulating the striatal outputs Substantia Nigra, pars reticulata (SNr) and ventral pallidum (VP).
Ai32 mice, which express a channelrhodopsin-2/EYFP fusion protein following exposure to Cre recombinase, were used for ChR2 expression in VP and SNr. The construct AAV-CMV-Cre was injected bilaterally. A. EYFP expression SNr as shown on a coronal brain slice. The image shows that ChR2 expression was contained to SNr boundaries. B. To demonstrate that light pulses increased neuronal activity in vivo in SNr, mice were unilaterally stimulated with light (represented by the fibre tip in blue) whereas the other hemisphere was not stimulated. Animals were in their home cages during this protocol and stimulation side was chosen arbitrarily across animals. Mice were perfused 90 minutes after stimulation period for Fos immunohistochemical analyses. Note Fos-positive cells under blue fibre tip. C. Counts of Fos-positive (Fos+) cells through slices containing SNr regions revealed a significantly greater number of Fos+ cells in the stimulated compared to non-stimulated side (N=5 mice, paired t-test t=6.2, * p=0.003). D-G. Same protocol as above, but for VP. As we show in D, unfortunately, the injections encompassed the VP and neighbouring pre-optic area (POA) in some animals. E. We then proceeded to analyse Fos expression upon unilateral stimulation in both POA and VP. Because fibre tips were localized just above the VP, unilateral light stimulation resulted in ipsilateral increases in Fos expression in VP but not POA, although some neurons expressed ChR2 in POA. F. shows that counts of Fos+ cells through slices containing VP regions revealed a significantly greater number of Fos+ cells in the stimulated compared to non-stimulated side (N=5 mice, paired t-test t=6.1, Bonferroni * p=0.008). G shows that this was not the case for POA in the same animals, as the number of Fos+ counts was similar on both stimulated and non-stimulated sides (t=1.19, p=0.29). n.s.= non-statistically significant. acp = posterior limb of the anterior commissure.
Supplementary Figure 10 Electrophysiological circuit identification of D1r-specific VS=>VP and DS=>SNr pathways, and in vivo assessment of optogenetic effects.
A. Outward postsynaptic currents were evoked in VP neurones by blue light pulses (10ms) delivered at various frequencies (1,5,10,20 Hz) as well as continuous light trains, to VP neurones targeted by Chr2-expressing terminals from D1r-neurones in VS. Whole-cell voltage-clamp recording was performed at the holding potential of -40mV. The GABAA receptor antagonist bicuculine (Bic, 50μM) completely abolished the outward currents evoked by blue light, suggesting that optogenetic activation of VS D1r neurones terminals functionally act on (inhibit) VP neurones via GABAergic mechanisms. B. Representative traces showing outward GABAergic postsynaptic current evoked in SNr neurones by blue light (10ms) pulses delivered at various frequencies (1,5,10,20 Hz) to SNr neurones targeted by Chr2-expressing terminals from D1r-neurones in DS. Recordings performed as in A. C-D. Verification of the ability of optogenetic pulses to VS D1-neurones to inhibit DREADD-induced activity in VP in vivo. We proceeded to analyse Fos expression upon unilateral stimulation in VP (as shown in Figure 8 above) concomitantly to DREADD activation by intra-peritoneal infusions of designer drug CNO. C shows that counts of Fos+ cells through slices containing VP regions revealed a significantly greater number of Fos+ cells in the non-stimulated compared to the light-stimulated side (because activation of D1 GABAergic VS terminals should inhibit VP; N=5 mice, paired t-test t=5.06, Bonferroni * p=0.014). D shows that this was nevertheless not the case for preoptic area (POA) in the same animals, as the number of Fos+ counts was similar on both stimulated and non-stimulated sides (t=0.81, p=0.46). The results therefore suggest that POA is unlikely to have had any influence on the results associated with VP manipulation. n.s.= non-statistically significant.
Supplementary Figure 11 Design receptor activation assessment via in vivo multielectrode array recordings.
In vivo electrophysiological studies after injecting the construct AAV-hSyn-HA-hM3D(Gq)-IRES-mCitrine into VP and SNr of D1-Cre mice. A. Histological sample (left hemisphere) and schematic representation (right hemisphere) of final position of microwire tips in VP. Wires were cut at different lengths to account for non-uniform dorsal-ventral depths across mediolateral extension. B. Trace shows average of firing rate activity from a population of 38 neurons in VP before and after DREADD activation with CNO (n=2). Time t=0 (dashed line) indicates onset of CNO i.p. injection. Within pie-plot, black area represents percentage of units excited after the CNO injection, whereas white area represents the percentage of units that were inhibited. Gray area represents the percentage of units with stable firing rate throughout the experiment. C. Same as in B, but for saline injection. Percentages out of the total 29 units that were excited, inhibited and unaffected are shown. Chi-square tests reveal that relative numbers of cells changing firing rates in CNO case was significantly greater than SAL case χ2[2,N=67]=23.99, p=0.0000062. D. Histological sample (left hemisphere) and schematic representation (right hemisphere) of final position of microwire tips in SNr.Wires were cut at different lengths to account for non-uniform dorsal-ventral depths across mediolateral extension. E-F Same as B-C but for SNr (59/29 units for CNO/SAL respectively, χ2[2,N=88]=10.3, p=0.0058).
Supplementary Figure 12 Descending pathways via which nutrient sensing in basal ganglia may command feeding motor programs.
Separate dorsal (blue lines) and ventral (red lines) circuits may activate a common brainstem oral-motor central pattern generator (CPG) via segregated descending pathways. This model is consistent with the animal’s ability to persist licking an aversive solution whenever this solution is associated with energy gain. Connections ending as triangles represent excitatory/modulatory connections. Connections ending as simple trace represent inhibitory connections, as based on current knowledge of the chemical structure of these pathways. DA=Dopamine, DS=Dorsal striatum, LH=Lateral hypothalamic area, SNc=Substantia Nigra, pars compacta, SNr=Substantia Nigra, pars reticulata, VP=Ventral pallidum, VS=Ventral striatum.
Supplementary Figures 1–12 (PDF 2276 kb)
A mouse expressing the blue light-sensitive depolarizing channel ChR2 in dopamine-excitable D1r-expressing cells of dorsal striatum is shown. Upon contacting the bitter taste-containing sipper with the tongue, a blue laser pulse is delivered to the animal's dorsal striatum via the bilaterally implanted optical fibers. When laser source is OFF, the animal quickly interrupts licking, retracts to the opposite corner of the cage, and displays negative taste reactions. In contrast, incessant licking is observed when laser source is ON. (WMV 68220 kb)
About this article
Cite this article
Tellez, L., Han, W., Zhang, X. et al. Separate circuitries encode the hedonic and nutritional values of sugar. Nat Neurosci 19, 465–470 (2016). https://doi.org/10.1038/nn.4224
This article is cited by
Genome Medicine (2022)
Identification of an endocannabinoid gut-brain vagal mechanism controlling food reward and energy homeostasis
Molecular Psychiatry (2022)
Experimental & Molecular Medicine (2022)
Nature Communications (2022)
Nature Neuroscience (2022)