Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Oct 15;12:245.
doi: 10.3389/fnmol.2019.00245. eCollection 2019.

Selective Functional Interaction Between the Lateral Habenula and Hippocampus During Different Tests of Response Flexibility

Affiliations
Free PMC article

Selective Functional Interaction Between the Lateral Habenula and Hippocampus During Different Tests of Response Flexibility

Phillip M Baker et al. Front Mol Neurosci. .
Free PMC article

Abstract

The lateral habenula (LHb) has been shown to play critical roles in a variety of appetitive tasks (e.g., spatial memory and object recognition) that require animals to flexibly respond to changing task conditions. These types of tasks are known to be dependent on hippocampus (HPC) and/or medial prefrontal cortex (mPFC), suggesting that the LHb contributes to the limbic memory circuit. Here we provide new evidence that the LHb and HPC play distinct but complimentary roles in tasks that require flexible responding to changing task conditions. Experiment 1 tested whether the LHb is needed for the performance of a HPC-dependent maze-based spatial delayed alternation task. The importance of interactions between the LHb and HPC to accomplish the same spatial delayed alternation task was tested in Experiment 2 where the LHb and HPC were disconnected both ipsilaterally and contralaterally. Experiment 3 tested LHb's involvement in a standard behavioral economic task that requires flexible responding (maze-based delayed discounting), a task previously shown to rely on HPC. Results of Experiment 1, revealed that LHb inactivation impairs spatial delayed alternation during asymptotic performance but not during initial learning. Importantly, working memory did not appear to be affected as performance remained above chance levels both during initial learning and asymptotic testing. Experiment 2 showed that ipsilateral and contralateral disconnection of the LHb and HPC led to impaired performance on the spatial delayed alternation task. Impairments were not observed after unilateral inactivation of only one structure. Results of Experiment 3 were similar to our previous report of the effects of HPC inactivation: LHb inactivation impaired delayed discounting. All effects could not be accounted for by changes in reward magnitude discrimination, reward location per se, or sex of the animal. These findings, combined with other recent publications confirms and extends our working hypothesis that the LHb enables adaptive and flexible responding, particularly when established rules must be flexibly applied on a trial by trial basis. Since there are no known direct anatomical connections between LHb and HPC, future research is needed to understand how these structures communicate to enable flexible and rapid responding.

Keywords: baclofen; delay discounting; muscimol; rats; spatial delayed alternation.

Figures

Figure 1
Figure 1
Schematic of selected afferent, bidirectional and efferent pathways of LHb (red circle) and the broader HPC-mPFC-LHb circuit. Afferent structures/pathways are shown in blue, bidirectional structures/pathways are shown in red, efferent structures/pathways are shown in yellow. The mPFC-LHb-HPC circuit structures/pathways are shown in green. LHb, lateral habenula; HPC, hippocampus; mPFC, medial prefrontal cortex; vDBB, vertical diagonal band of Broca; VP, ventral pallidum; LPO, lateral preoptic area; SCN, suprachiasmatic nucleus; EPN, entopeduncular nucleus; LH, lateral hypothalamus; VTA, ventral tegmental area; RMTg, rostromedial tegmental nucleus; IPN, interpeduncular nucleus; MnR, median raphe; DR, dorsal raphe; PAG, periaqueductal gray. Rat brain background credited to Brown (2018).
Figure 2
Figure 2
Schematic of Delayed-Alternation Task. (1) Animals waited on the start arm for 10 s during the start/delay epoch. The arm segment closest to the center platform was lowered to confine the animal to the start arm region for the delay period. (2) The choice epoch started when the animal entered the center platform of the maze and made a choice to the left/right reward arm. The blue circle at the end of each reward arm signifies the metal cup where sugar pellets were delivered through food dispensers upon the animal’s arrival. (3) The reward epoch, which was analyzed only for the rewarded choice, while the other reward or start arm was lowered. Animals consumed the sugar pellet once they obtained the reward at the metal cup. (4) The return epoch began when the animal finished reward consumption and returned to the recently raised start arm (start arm location was pseudorandomized over the 60 choice trials). (5) Once the animal finishes the delay period, the next trial began. Only alternating choices were rewarded, while repeated entrance to the same reward arm resulted in no reward. After the animal returned to the start arm and waited for the 10 s delay period, a new trial began.
Figure 3
Figure 3
Experiment 1: LHb inactivation impairs learned spatial delayed-alternation but does not alter initial learning. (A) Drug treated animals are more accurate than would be expected by chance (t(2) = 7.94, p < 0.05). (B) LHb inactivation altered flexible performance of a learned behavior performed at asymptotic levels. (C) No effects of treatment group (F(2,16) = 0.80, ns.), learning phase (F(1,16) = 1.01, ns.), or interaction (F(2,16) = 1.55, ns.) were observed according to a two-way ANOVA on the ratio of preferred place location over the total number of choices during a session. (D) Examination of egocentric turn bias using a two-way ANOVA revealed a significant effect of treatment group (F(2,16) = 5.11, p < 0.05), learning phase (F(1,16) = 38.26, p < 0.01), and a significant interaction (F(2,16) = 4.23, p < 0.05) on the ratio of preferred egocentric turn direction over the total number of choices during a session. Bonferroni post hoc tests showed that the turn bias ratio was significantly lower during asymptotic performance for the Sal-Sal and Drug-Sal groups but not for the Sal-Drug group (p < 0.01). (E) A one-way ANOVA revealed a similar level of performance accuracy following the final day of asymptotic testing, F(2,16) = 3.03, ns. ranging from (80.79 ± 3.32) in the Drug-Sal group to (90.28 ± 1.69) in the Sal-Sal group. (F) Cannula placements in the LHb. *p < 0.05. **p < 0.01. ns = not significant.
Figure 4
Figure 4
Experiment 2: LHb and HPC disconnection decreases accuracy in a delayed-alternation task. (A) Schematic of ipsilateral, contralateral and unilateral infusion cannula placements within HPC and LHb. (B) Histological example of cannula placements in the HPC and LHb. (C) Any treatment group that impaired both the LHb and HPC showed significant impairments in accuracy (F(3,32) = 5.83, p < 0.05). (D) Results revealed no effect of day of injection on overall accuracy on the delayed alternation task (F(7,64) = 0.89, ns.). (E) Egocentric turn bias significantly decreased from the Learning phase (0.81 ± 0.04) to Asymptote phase (0.60 ± 0.02; t(8) = 5.12, p < 0.01). (F) The place bias remained at a relatively low level in both Learning and Asymptote phases (Learning = 0.57 ± 0.02, Asymptote = 0.52 ± 0.01) and did not significantly differ from each other (t(8) = 1.78, ns.). (G) A two-way ANOVA of turn bias scores during asymptotic performance revealed main effects of both drug treatment (F(1,32) = 17.91, p < 0.01), and injection condition (F(3,32) = 13.72, p < 0.01), with no interaction effect (F(3,32) = 4.39, ns.). Bonferroni post hoc tests indicated that there was a significant difference between saline and drug treatment in the contralateral disconnection group (t(8) = 3.03, p < 0.05). (H) A two-way ANOVA of place bias scores did not show main effects for either drug treatment (F(1,32) = 2.86, ns.) or treatment area (F(3,32) = 1.06, ns.) but did reveal a significant interaction (F(3,32) = 3.03, p < 0.05). Bonferroni post hoc tests indicated that the ipsilateral disconnection group significantly increased place bias scores under drug vs. saline treatment, (t(8) = 3.38, p < 0.01). *p < 0.05. **p < 0.01. ns = not significant.
Figure 5
Figure 5
Schematic of Delay-Discounting Task. (A) Reward arm (marked in yellow) and two start arms (marked in red) on the elevated T-maze. (B) Sooner smaller choice: the animal waited for 3 s to obtain one sugar pellet. Later larger choice: the animal waited for 10 s and obtained four sugar pellets. Once the animal entered one reward arm, the other reward and start arms were blocked by wooden barriers to confine the animal to the chosen arm for the entire delay period. (C) For each day of testing, three blocks of trials were offered, each associated with a different length of LL delay (10 s, 20 s, or 40 s). The SS delay remained constant (3 s). The blocks were separated by an inter-block interval of 5 min. Drug infusion sequences follow an ABBA pattern. On day 1 and day 4, a certain animal would receive one type of infusion (0.9% saline or muscimol), and day 2 and day 3 of the alternative type of infusion.
Figure 6
Figure 6
Experiment 3: Bilateral LHb inactivation resulted in impaired performance in delay-discounting task. (A) Bilateral HPC inactivation resulted in significantly impaired performance during the delay-discounting task. The figure shown was adapted from Davis et al. (2016). (B) Histological illustration of cannula placements in the LHb. (C) Bilateral LHb inactivation resulted in a significant impairment in delay-discounting task. There was a significant interaction effect (F(2,16) = 34.725, P < 0.01), no drug effect (F(1,8) = 3.510, p = 0.098, ns.), and a significant delay effect (F(2,16) = 15.561, p < 0.01). (D) Impairments were not due to sex differences. No interaction effect was observed after muscimol infusion (F(2,14) = 0.206, p = 0.816, ns.), and no significant effect of sex was observed (F(1,7) = 0.012, p = 0.916, ns.). No interaction effects were observed under saline infusion (F(2,14) = 0.371, p = 0.697, ns.) nor was there a significant sex effect (F(1,7) = 0.177, p = 0.687, ns.). (E) Impairments were not due to a spatial bias toward one or the other reward locatinos. No interaction effect was found after muscimol infusion (F(2,14) = 0.127, p = 0.882, ns.) and and there was no effect of reward location (F(1,7) = 0.057, p = 0.819, ns.). No interaction effects were seem after saline infusion (F(2,14) = 2.268, p = 0.140, ns.) and there was no effect of reward location (F(1,7) = 2.356, p = 0.169, ns.). (F) Intact reward magnitude discrimination was observed after LHb inactivation. Given a consistent delay for either small or large reward choices, animals preferred locations associated with a larger reward, paired sample t-test (t = 10.513, p < 0.01). **p < 0.01. ns = not significant.

Similar articles

See all similar articles

References

    1. Admon R., Pizzagalli D. A. (2015). Dysfunctional reward processing in depression. Curr. Opin. Psychol. 4, 114–118. 10.1016/j.copsyc.2014.12.011 - DOI - PMC - PubMed
    1. Agetsuma M., Aizawa H., Aoki T., Nakayama R., Takahoko M., Goto M., et al. . (2010). The habenula is crucial for experience-dependent modification of fear responses in zebrafish. Nat. Neurosci. 13, 1354–1356. 10.1038/nn.2654 - DOI - PubMed
    1. Aghajanian G., Wang R. Y. (1977). Habenular and other midbrain raphe afferents demonstrated by a modified retrograde tracing technique. Brain Res. 122, 229–242. 10.1016/0006-8993(77)90291-8 - DOI - PubMed
    1. Araki M., McGeer P. L., Kimura H. (1988). The efferent projections of the rat lateral habenular nucleus revealed by the PHA-L anterograde tracing method. Brain Res. 441, 319–330. 10.1016/0006-8993(88)91410-2 - DOI - PubMed
    1. Azmitia E. C., Segal M. (1978). An autoradiographic analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat. J. Comp. Neurol. 179, 641–667. 10.1002/cne.901790311 - DOI - PubMed

LinkOut - more resources

Feedback