- Research article
- Open Access
Characterization of a thalamic nucleus mediating habenula responses to changes in ambient illumination
- Ruey-Kuang Cheng†1,
- Seetha Krishnan†2,
- Qian Lin2,
- Caroline Kibat3 and
- Suresh Jesuthasan1, 3, 4, 5Email authorView ORCID ID profile
© Jesuthasan et al. 2017
Received: 17 July 2017
Accepted: 25 September 2017
Published: 3 November 2017
Neural activity in the vertebrate habenula is affected by ambient illumination. The nucleus that links photoreceptor activity with the habenula is not well characterized. Here, we describe the location, inputs and potential function of this nucleus in larval zebrafish.
High-speed calcium imaging shows that light ON and OFF both evoke a rapid response in the dorsal left neuropil of the habenula, indicating preferential targeting of this neuropil by afferents conveying information about ambient illumination. Injection of a lipophilic dye into this neuropil led to bilateral labeling of a nucleus in the anterior thalamus that responds to light ON and OFF, and that receives innervation from the retina and pineal organ. Lesioning the neuropil of this thalamic nucleus reduced the habenula response to light ON and OFF. Optogenetic stimulation of the thalamus with channelrhodopsin-2 caused depolarization in the habenula, while manipulation with anion channelrhodopsins inhibited habenula response to light and disrupted climbing and diving evoked by illumination change.
A nucleus in the anterior thalamus of larval zebrafish innervates the dorsal left habenula. This nucleus receives input from the retina and pineal, responds to increase and decrease in ambient illumination, enables habenula responses to change in irradiance, and may function in light-evoked vertical migration.
The habenula is an evolutionarily conserved structure [1, 2] that influences multiple behaviors, ranging from fear [3–5], to learning [6, 7], addiction , sleep , aggression [10, 11], and performance under stress . One function of the habenula is to regulate the release of broadly acting neuromodulators such as serotonin, dopamine, epinephrine and histamine [12–15]. To precisely control these neuromodulators, the habenula integrates diverse variables, including internal state, reward value, and sensory stimuli. This information reaches the habenula from distinct sources. For example, circadian time is transmitted to the habenula by hypocretin-secreting neurons located in the hypothalamus . Negative reward or punishment is conveyed by neurons of the entopeduncular nucleus (internal segment of the globus pallidus) . Olfactory stimuli evoke activity in the habenula [18, 19] via a direct innervation of mitral cells from the olfactory bulb . Light, as well as loss of light, can also cause activity in the habenula, as has been demonstrated in rat , mouse , pigeon , and zebrafish [18, 24], but the neurons regulating habenula responses to changes in ambient illumination are not yet well defined.
The habenula is divided into two major regions based on the pattern of connectivity. In mammals, these are called the medial and lateral habenula, while in fish, these are the dorsal and ventral habenula . In larval zebrafish, short pulses of red light cause asymmetric depolarization of the dorsal habenula, with a stronger response on the left side . This response is lost in fish lacking eyes . However, no direct pathway from the retina to the habenula has been documented [26, 27]. By retrograde tracing in adult zebrafish, Turner et al.  proposed that the habenula receives input from the nucleus rostrolateralis, a diencephalic structure with retinal input that has been described in several ray-finned fish [29, 30]. Because injections into both left and right habenula led to labeling of this structure, as well as due to potential artifacts in labeling, the authors concluded that the source of light-evoked activity in the habenula could not be determined . Here, we set out to characterize the nucleus by which ambient illumination affects activity in the habenula and to explore the function of this nucleus in innate responses to change in illumination.
Habenula afferents mediating response to light ON and OFF target the dorsal left neuropil
Two-photon microscopy was used next, as this allows higher spatial resolution imaging before, during, and after delivery of a more precisely timed light stimulus. Imaging was carried out using GAL4s1011t, UAS:GCaMP6s fish, in which expression of the calcium reporter is restricted mainly to habenula neurons . In agreement with wide-field microscopy, onset of light was found to trigger a response in the dorsal left neuropil of the habenula (Fig. 1d–h, blue pixels). Responses in the neuropil, which contains dendrites of habenula neurons, correlated with but preceded the response of the cell body of habenula neurons (Fig. 1i–k). These observations suggest that neurons mediating the habenula response to light reside outside the habenula and target the dorsal left neuropil.
A thalamic nucleus innervates the habenula
Additional file 1: Movie 1. Z-stack through a brain following DiD injection into the dorsal left habenula neuropil. DiD label (cyan) is seen bilaterally in two clusters of neurons in the anterior thalamus, starting from a depth of 65 μm from the first plane. Sparse labeling can also be seen in the ipsilateral entopeduncular nucleus (EN), at a depth of approximately 100–110 μm. Glutamatergic neurons are labeled by vGlut2a:GAL4,UAS:eGFP (yellow), while GABAergic neurons are labeled by gad1b:DsRed (magenta). The left fasciculus retroflexus is labeled by axons from the habenula. This is a dorsal view, with anterior to the left. Gamma = 0.45. (MP4 14880 kb)
Additional file 2: Movie 2. Z-stack through the brain following DiD injection into the dorsal right habenula neuropil. Retrogradely labeled cells are seen primarily in the ipsilateral entopeduncular nucleus (arrow). Labeled axons are also visible in the neuropils of the left habenula. These may arise from neurons that innervate the anterior right thalamus and/or from the right entopeduncular nucleus (arrow). This is a dorsal view, with anterior to the left. (MP4 11261 kb)
Additional file 3: Movie 3. Z-stack of 6-day-old gad1b:DsRed, vglut2a:GAL4, UAS:eGFP fish. GABAergic neurons (magenta) are visible in the thalamus, below the habenula. Arrows indicate the anterior thalamic neuropil, which contains DsRed-labeled fibers (~50 μm below the first plane). The entopeduncular nucleus does not contain DsRed-labeled fibers. In the first frame, DsRed-labeled neurites are visible in the optic tectum, but not in the habenula neuropil. Anterior is to the left. The stack goes from dorsal to ventral. rHb right habenula, lHb left habenula, OT optic tectum, EN entopeduncular nucleus. (MP4 18586 kb)
Thalamic nuclei usually contain glutamatergic projection neurons , but may, in rare cases, extend GABAergic projections . When DiD was injected into the dorsal left habenula, approximately 45% of retrogradely labeled thalamic neurons expressed eGFP under the control of the vGlut2 GAL4 driver  (Fig. 3b, arrowheads; Additional file 4: Movie 4; n = 3 fish), consistent with projection neurons being glutamatergic. We asked whether any of the afferent neurons might be GABAergic, as this is one possible mechanism for the suppression of activity seen in Inh neurons. However, no retrograde DiD label was seen in thalamic cells labeled by gad1b:DsRed (Additional file 1: Movie 1 and Additional file 4: Movie 4), nor were there DsRed-labeled neurites in the dorsal left habenula neuropil (Additional file 3: Movie 3), as would be expected if there was innervation by GABAergic neurons. Moreover, when the fixable dye CM-DiI was injected into the dorsal left habenula, followed by immunofluorescence with the antibody to GAD65/67, no double-labeled cells were seen (Fig. 3f, n = 6 fish). It is unclear whether this is due to the low probability of labeling the relevant GABAergic cells with lipophilic tracing or with the transgene or, more simply, if there are no GABAergic projections from the thalamus to the habenula. It is evident, however, that the anterior thalamus sends glutamatergic projections to the habenula.
Additional file 4: Movie 4. Lateral view of a fish following DiD injection in the dorsal left neuropil of the habenula. The right side of the fish shown in Movie 1. Thalamic neurons that have been retrogradely labeled are shown in cyan. Glutamatergic neurons are labeled by vGlut2a:GAL4,UAS:eGFP (yellow), while GABAergic neurons are labeled by gad1b:DsRed (magenta). DiD labeled cells extend neurites into neuropil of the anterior thalamus. A number are labeled by eGFP (arrows), but none are labeled by DsRed. The optic tract is visible in the DIC image, and contains eGFP labeled axons. Note that anterior is to the right in this stack. (MP4 7315 kb)
Light ON and OFF evokes activity in the anterior thalamus
Additional file 5: Movie 5. Lateral view of the left anterior thalamus following injection of DiD (cyan) into the dorsal left habenula neuropil and DiI (yellow) into the right eye. The arrow shows intermingling of retinal and habenula afferent fibers in the thalamic neuropil. The stack runs from lateral to medial, and habenula afferents and retinal ganglion cell terminals meet anteriorly and medially to the optic tract. fr fasciculus retroflexus. This is a 6-day-old fish, with anterior to the left. (MP4 2837 kb)
Additional file 6: Movie 6. Z-stack of a 6-day-old fish following CM-DiI injection into the pineal and DiD into the dorsal left habenula. Pineal axons (red) project laterally and then posteriorly. Arrows indicate axons that enter the anterior thalamic neuropil, where retrogradely labeled fibers from the habenula (cyan) are visible. This is a dorsal view, with anterior to the left. (MP4 8313 kb)
Lesion of the anterior thalamic neuropil inhibits habenula response to light ON and OFF
Optogenetic manipulation of the thalamus affects habenula response to irradiance change
Optogenetic manipulation of the thalamus disrupts an innate behavioral response to irradiance change
Finally, we asked whether the anterior thalamic nucleus might be involved in an innate behavior that is responsive to changes in light. We hypothesized that one such behavior may be light-evoked vertical migration . Larval zebrafish normally move upwards to the surface of a water column in the presence of blue or green light, but move downwards when the lights are switched off . We tested the effect of optically manipulating the thalamus using the anion channelrhodopsins, reasoning that the presence of these channels would disrupt normal light-controlled responses; if no difference was seen, then the hypothesis should be rejected.
Imaging with wide-field and two-photon microscopy demonstrates that the dorsal left neuropil of the zebrafish habenula is stimulated by change in ambient illumination, consistent with previous reports of an asymmetric response in habenula neurons to a flash of light . Lipophilic tracing demonstrates that this neuropil is asymmetrically innervated by a nucleus in the anterior region of both left and right thalamus. The anterior thalamic nucleus receives input from the retina and pineal, and responds to light ON and OFF. Lesion of the anterior thalamic neuropil or optogenetic silencing of the thalamus inhibited light-evoked activity in the habenula, while optogenetic stimulation of the thalamus drove activity in the habenula. Thus, by optical recording, anatomical tracing, optical manipulation, and lesion, our data suggests that an anterior thalamic nucleus mediates the habenula responses to irradiance change in larval zebrafish.
The thalamic nucleus that projects to the habenula can be functionally separated into two domains, based on the response to light – excitation to light OFF in the anterior-ventral regions and excitation to light ON more dorso-posteriorly. This neuropil contains two previously defined targets of retinal ganglion cells, AF2 and AF4 , that have this location. AF4 is innervated predominantly by M3 and M4 retinal ganglion cells, which extend their dendritic tree into the proximal layer of the inner plexiform layer and are considered ON neurons . AF2 is innervated by B1 retinal ganglion cells that have dendrites in the distal layer , and these may account for the OFF responses in the thalamus and habenula. The pineal may also be responsible for a component of OFF responses, namely pineal cells appear to depolarize in darkness, and pineal fibers innervate the thalamic neuropil of larval zebrafish, as has been reported for adult zebrafish .
As in the anterior thalamus, a response to the loss of light was seen in the habenula. This has a number of implications. Firstly, it suggests that darkness itself may be a stimulus, in which case the level of activity in habenula neurons during darkness prior to a light stimulus cannot be taken to be a ‘ground’ state. Such activity may include what has been termed spontaneous activity , which may reflect the current state of the animal (i.e., the effects of being in the dark, which is aversive to larval zebrafish ). Secondly, the fact that there is more than one class of habenula response to darkness implies that there may be more than one mechanism involved. In particular, the suppression of activity in the presence of light in Inh neurons implies that a part of the OFF response could involve active inhibition. As yet, there is no evidence that there is direct hyperpolarization of habenula neurons during light ON. However, inhibition need not occur in the habenula, but could occur in the thalamus, where there are GABAergic neurons that extend neurites into the thalamic neuropil. Inhibition of thalamic OFF neurons by thalamic ON neurons, for example, could lead to the observed pattern in habenula Inh neurons.
The thalamic nucleus mediating activity in the habenula may represent the nucleus rostrolateralis, as proposed by Turner et al. . The nucleus rostrolateralis was initially described as a dorsal thalamic nucleus that receives retinal input . However, it has recently been suggested that this nucleus is an extension of the habenula, due to apparent innervation of the interpeduncular nucleus (IPN) . We found no evidence that the nucleus identified here has a direct connection to the IPN. The GAL4s1020t, UAS:GCaMP6s line, which was used for calcium imaging of the thalamic response, for example, does not label axons extending to the IPN. Moreover, the GAL4s1011t driver, which labels the habenula neurons and axons that extend to the IPN, does not label the nucleus with retinal input. It is thus unclear whether the nucleus identified here is different from the nucleus rostrolateralis described in the butterfly fish, or if there was a labeling artifact in the tracing experiment .
While this manuscript was in review, it was suggested that light-evoked activity in the habenula is driven by input from the thalamic eminence , an “ambiguous thalamic structure”  that has been proposed to give rise to the glutamatergic bed nucleus stria medullaris [34, 54] or the ventral entopeduncular nucleus, a homolog of the globus pallidus . However, the nucleus characterized here is distinct from the ventral entopeduncular nucleus, which is located more anteriorly and ventrally . It also contains GABAergic neurons, and is thus unlikely to be the bed nucleus stria medullaris. It is possible that the nucleus here is an additional derivative of the thalamic eminence, although this remains to be demonstrated with lineage tracing. Intriguingly, Zhang et al.  showed that the retinal inputs to AF4 express the melanopsin-related gene opn4xa, consistent with our finding that the thalamic response to light is stronger for blue light relative to red light, and another report that the habenula response is stronger for blue light . In mammals, melanopsin-expressing retinal ganglion cells target a number of thalamic structures, including the intergeniculate leaflet and the margin of the lateral habenula . The latter region may correspond to the para-habenular termination zone, which is located in the anterodorsal thalamic nucleus . Whether either of these regions is homologous to the zebrafish nucleus described here remains to be determined.
Neurons in the anterior thalamus have a prominent sustained response to blue light (Fig. 4a–e; ), and may be involved in a behavior that is evoked by blue light, which is vertical migration. This response is disrupted by expression of anion channelrhodopsins in the anterior thalamus, suggesting that the behavior is not independent of the thalamus. A limitation of this experiment, however, is that the driver line used also causes expression of the channel in spinal motor neurons . Silencing of these neurons may contribute to reduced ability of GrACR1 and GtACR2 fish to move upwards in the light. However, the offset of light, which causes activity in networks containing light-gated chloride channels (Fig. 8) [41, 43], led to upward movement. This is unlikely to be due only to rebound activation of motor neurons, as there is a choice of which direction to move. Instead, the upward movement at light offset is consistent with the hypothesis that activation of the thalamus may drive vertical migration.
A projection from the thalamus to the habenula may be evolutionarily conserved in vertebrates. Using retrograde tracing with horseradish peroxidase, a projection from the dorsal thalamus to the habenula has been reported in a lizard  and in a frog . In humans and rabbits, a thalamo-habenula projection was proposed many years ago based on degeneration experiments [59, 60], but evidence with modern tracing techniques is lacking. Hints of a projection can be seen in a tracing experiment performed in rats , but this remains to be confirmed. The mesoscale mouse connectome project  also suggests that such a projection may exist, but the large label volumes do not allow the possibility of labels from neighboring regions to be excluded. In humans, resting state functional magnetic resonance imaging indicates that the habenula and thalamus are functionally connected [63, 64]. However, it remains to be determined whether this connection is direct. The findings in lower vertebrates suggest that it may be worthwhile revisiting efferent connectivity of the anterior thalamus in mammals and investigating if and how this mediates non-visual responses to light.
A nucleus in the anterior thalamus of zebrafish enables habenula responses to increase and decrease in ambient illumination. This nucleus is innervated by the retina and pineal organ. It may function in vertical migration triggered by light.
Experiments were performed in accordance with guidelines issued by the Institutional Animal Care and Use Committee of the Biological Resource Centre at Biopolis, Singapore.
Zebrafish (Danio rerio) lines used for this study were UAS:GCaMP6s sq205 , SqKR11Et , sqKR4Et , Et(-0.6hsp70l:Gal4-VP16)s1011t , Et(-0.6hsp70l:Gal4-VP16)s1020t , UAS:GCaMP3 sq200 , elavl3:GCaMP6f a12200 , UAS:ChR2-eYFP , gad1b:DsRed , vGlut2a:GAL4 , UAS:eGFP, UAS:GtACR1 sq211 , UAS:GtACR2 sq212 , and AB wildtype. For brevity, the enhancer trap GAL4 lines are referred to as GAL4s1011t and GAL4s1020t.
Zebrafish larvae were anaesthetized in mivacurium and embedded in low-melting temperature agarose (1.2–2.0% in E3; egg water: 5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4) in a glass-bottom dish (Mat Tek). They were imaged on a Nikon two-photon microscope (A1RMP), attached to a fixed stage upright microscope, using a 25× water immersion objective (NA = 1.1). The femtosecond laser (Coherent Vision II) was tuned to 920 nm for GCaMP imaging. Stacks were collected in resonant-scanning mode with a 525/50 nm bandpass emission filter and with 8× pixel averaging; single-plane images were collected in galvano-scanning mode with 2× pixel averaging.
Light stimuli were generated by 5 mm blue LEDs (458 nm peak emission) powered by a 5 V TTL signal from a control computer and synchronized with image capture using a National Instruments DAQ board, controlled by the Nikon Elements software. Light intensity at the sample was 0.13 mW/cm2.
For wide-field microscopy, excitation was provided by LEDs (Cairn OptoLED) at 470 nm. Images were captured on a Zeiss Axio Examiner with a 20× water immersion objective, using a Flash4 camera (Hamamatsu) controlled by MetaMorph. After background subtraction, change in fluorescence was measured using MetaMorph.
Image data analysis
Initial data preprocessing
Data was analyzed using custom written codes in Python. Raw images obtained were first registered using cross-correlation to correct for any vertical/horizontal movement artifacts. Then, a median spatial filter of size 3 was applied to remove spatial noise. A darker region outside the fish was chosen as the background and subtracted from the image to remove any signal that did not arise from GCaMP fluorescence. Non-linear trends in the data were de-trended using polynomials of order 2–3.
Pixel-based analysis in single fish
In order to look at the overall spatial distribution of responses, which included both neuropils and cells, we performed clustering via K-means using the Thunder platform . Data here were normalized into Z-scores by subtracting the overall mean and dividing by the standard deviation of each pixel over time and smoothed with a rolling window. Since pixel-based analyses are sensitive to noise, and neighboring pixels with the same response could have varying standard deviation (in case of cell segmentation, pixels forming a region of interest are averaged to obtain its intensity value), z-scores that account for both mean and standard deviation were used. Clusters obtained using pixel-based K-means analysis also provided the basis for the type of responses we looked for in segmented neurons.
K-means clustering was performed to identify pixels with similar response profiles. This algorithm classifies the pixels into clusters, where the number of clusters, K, is chosen by the user. The end results are K cluster centers and labeling of pixels that belong to each cluster. Given the uncertainty of the optimal cluster number, an iterative approach was used to separate pixels that relate to evoked responses versus pixels that do not (here referred to as independent clusters). The number of clusters were chosen to reveal as many stimulus-related clusters as possible, until there was little change in the number and types of stimulus-related clusters and an increase in independent clusters. In normal fish, clusters related to evoked activity were easy to obtain. Clusters that are stimulus-independent were removed from the spatial and temporal plots for clarity. Examples of such clusters are shown in Additional file 8: Figure S2. In all cases, K-means cluster centers showing evoked responses to light ON were colored in shades of blue and those showing responses to light OFF were colored in shades of red. Pixels belonging to the cluster were colored similarly and superimposed on an average image of the plane analyzed. In different datasets (Figs. 1e–f, 4a–f, n–o, and 5c–e), this analysis provided an optimal k of 6–10. The 2–4 clusters that did not correspond to evoked activity were not included while plotting.
Cells were manually segmented in ImageJ. The average intensity of pixels within a region of interest across time were saved for further analysis. ∆F/F0 of the temporal traces were calculated by subtracting and then dividing by the mean of the total fluorescence during a baseline period (usually 10 seconds before first stimulus). A rolling window average was performed to smooth traces.
Pixel-based K-means analysis revealed many categories of responses to changes in irradiance. Using that as a basis, temporal traces from the cells were first broadly classified as those responding to light ON or OFF. This was done by calculating their correlation coefficient to a square wave that is 1 when the light is ON or OFF and 0 during other time periods. High correlation to these traces indicated that the pixel or cell is responding to light ON or OFF, respectively (from multiple runs, a correlation coefficient of 0.4 and above seemed to provide accurate classification). Inspecting the cell traces in the ON and OFF categories revealed further classifications that could be made based on time of response (transient or sustained) and direction of response (excitatory or inhibitory). Cells responding transiently to both ON and OFF were also found. The temporal traces from the many categories in individual fish are plotted as heatmaps (e.g., Figs. 2a and 4g). In experiments looking for the presence or absence of activity (effects of anterior thalamus neuropil ablation, parapineal ablation, enucleation, red vs. blue response), the broad categories of ON and OFF were used. Spatial distribution of these categories are also plotted (e.g., Figs. 2b and 4h).
Similar to the cell responses, pixels from habenula, thalamic neuropil, and the pineal were similarly classified. Pixels from multiple fish were overlaid on each other and image transparency was adjusted to view the compiled response. Since locations of responses were largely similar and different classes spatially distinct in the neuropil of individual fish, the overlay did not mask any response.
Where possible, boxplots were plotted to show the full distribution of the data. The box in the boxplot ranges from the first quartile to the third quartile, and the box shows the interquartile range (IQR). The line across the box is the median of the data. The whiskers extend to 1.5*IQR on either side of the box. Anything above this range is defined as an outlier and plotted as a black diamond in the plots.
Percentage of active cells/pixels vs. percentage of cells/pixels
The percentage of active cell/pixels was calculated by dividing the number of active cells/neuropil pixels by the total number of cells/neuropil pixels. This provides an indication of the response across individual animals and has been shown as boxplots or individual data points. Histograms, on the other hand, display the percentage of cells/pixels obtained by dividing the number of cells/neuropil pixels with a particular ∆F/F0 by the total number of cells/neuropil pixels.
The Kolmogorov–Smirnov test was used to calculate the differences in distribution of amplitude or response duration. Histograms are shown in all cases. For non-parametric paired distributions of number of cells, a Wilcoxon signed rank test was used and a Mann–Whitney U test was used for independent data. Test statistic and P values are reported.
DiD (Thermo Fisher Scientific) was dissolved in 50 μL ethanol to make a saturated solution. This was heated to 55 °C for 5 minutes prior to injection into the fish that had been fixed in 4% paraformaldehyde. Fish were mounted in 2% low melting temperature agarose dissolved in PBS. The dye was pressure injected into the habenula under a compound microscope (Leica DM LFS), using a 20× water immersion objective. For labeling the retina, a saturated solution of DiI (Thermo Fisher Scientific) in chloroform was used. Injections were carried out under a stereomicroscope (Zeiss Stemi 2000). After injections, fish were stored at 4 °C overnight to allow tracing, and then imaged with a 40× water immersion objective on a Zeiss LSM 710 confocal microscope.
CM-DiI (Thermo Fisher Scientific) was dissolved in ethanol (1 μg/μL). Fish were mounted in 2% agarose in E3, injected on a compound microscope, then allowed to recover in E3 at 28 °C for 4 hours.
Larvae were fixed in 4% para-formaldehyde/PBS overnight at 4 °C. They were then rinsed in PBS. The brains were dissected out, and permeabilized using 1% BSA (fraction V; Sigma), 0.1% DMSO and 0.1% Triton X-100. The anti-GAD65/67 (Abcam ab11070, RRID:AB_297722; 1:500) has been previously used in zebrafish . The brains were incubated in the primary antibody overnight, rinsed several times in PBS, then incubated in secondary antibody (Alexa 488 goat anti-rabbit; 1:1000). After washing, these were mounted in 1.2% agarose/PBS. Imaging was carried out using a Zeiss LSM 800 laser scanning confocal microscope, with a 40× water immersion objective.
Five-day-old fish were anaesthetized in Ringer’s saline containing buffered tricaine. The eyes were removed using electrolytically sharpened tungsten needles. Fish were allowed to recover for several hours in anesthetic-free saline. Activity was recorded 2–4 hours after eye removal. To enable lateral imaging of the thalamus (Fig. 5c, d), one eye was removed using this method.
Five dpf GAL4s1020t, UAS:ChR2-eYFP, elavl3:GCaMP6f larvae were used. All experiments were performed on fish lacking eyes. Fish were mounted in 1.2% agarose in Ringer’s saline, and imaged using two-photon microscopy as described above, at 1 Hz. Optical stimulation was carried out using a 50 μm fiber optic probe (Doric Lenses). The probe was held with a pipette holder (UT-2, Narishige), and the tip was positioned approximately 20 μm from fish, at the level of the thalamus, using a hanging drop micromanipulator (MO-202U, Narishige). The 465 nm LED (Doric) was driven with a current of 900 mA, 30 seconds after the start of imaging; 10 pulses were provided, with a pulse duration of 25 milliseconds and a frequency between 1 and 8 Hz. Each fish was exposed to at least three pulse trains. For Fig. 9b, c, the average of the first 29 frames was used as a reference. The ratio of all frames relative to this reference was obtained using FIJI (RRID:SCR_002285). The analysis to generate Fig. 9g was blind to the genotype.
elavl3:GCaMP6f larvae were anaesthetized and then mounted in 2% low-melting temperature agarose. First, the response of dorsal habenula neurons to light pulses was recorded. Lesions were then created with the femto-second laser tuned to 960 nm and fixed on a single point. Several pulses, each lasting 100–500 msec, were used. Lesioning was monitored by time-lapse imaging GCaMP6f fluorescence before and after each pulse, and was terminated when a cavitation bubble was seen; this was visible by simultaneously collecting light at 595 nm. Animals with bleeding in the brain after lesioning due to bursting of a blood vessel in the thalamus were discarded. The dorsal habenula was then re-imaged at the focal plane that was initially recorded, as determined by the focus motor, with care taken to ensure that cell shapes matched.
As described elsewhere , six naive larvae were tested simultaneously. Fish were placed individually in a chamber (3 cm L × 1 cm W × 5 cm H). After 3 min of adaptation to light and habituation to the chamber, six cycles of alternating light/dark were delivered, each consisting of 1 min light ON and 1 min light OFF. A green (24 V, 525 nm peak, TMS-lite) or blue (470 nm peak, TMS-lite) LED backlight was the only visible light source in the incubator. The irradiance of the green light was 3.8 mW/cm2, while the irradiance of blue light was 6.0 mW/cm2, as measured using a Thorlabs light meter (PM100A and S120VC). Videos were taken at 17 fps, 1096 × 1096 pixel resolutions, using custom-written Python codes for real-time tracking of the fish position in the tank. The codes also control a USB3.0 Basler camera (acA2040-90umNIR) attached with a 1:1.8/4 mm lens (Basler) and a 830 nm longpass filter (MIDOPT, LP830) for capturing images at the IR range. Four infrared LED bars (850 nm peak TMS-lite) were used for illumination. The LED backlight was controlled by Python codes driving a microcontroller board (Arduino Uno) connected to a power supply switch (TMS-lite). The entire experiment for one transgenic line was carried out in one afternoon (3–6 pm). A total of 57 fish were tested (18 GtACR1, 17 control siblings, tested at 8 dpf; 10 GtACR2, 12 control siblings, tested at 11 dpf).
Expression of GtACR1 or GtACR2 in each fish was determined after the experiment using a fluorescent stereomicroscope. No fish were excluded from analysis. The x-y coordinate data were analyzed using custom-written macros in Excel (Microsoft). The correlation coefficient of each fish (Fig. 10c, e) was calculated using the correl function in Excel to correlate the vertical position of the fish in the tank, normalized from 0 (bottom) to 1 (top), with the LED backlight status (0 OFF and 1 ON). To determine the initial movement of the fish upon each light offset, especially when the fish was in the middle of the tank (defined as between 0.25 and 0.75 in the y-axis), we calculated the position of the fish at the sixth second after light offset (i.e., after the first 10% of darkness). Upward movement was defined as vertical position at t6 > t0 (red dots in Fig. 10d, f) and downward movement was defined as vertical position at t6 < t0 (blue dots in Fig. 10d, f). Because each fish has more than one data point in the six ON/OFF cycles in Fig. 10d, a multilevel analysis was conducted to rule out the nested cluster (fish). Locomotion was calculated as distance moved by each fish under light ON and OFF and averaged across six cycles.
We thank David Hildebrand and Isaac Bianco for providing the elavl3:GCaMP6f line, and Claire Wyart for providing the UAS:ChR2-eYFP line. The drawing in Fig. 5b was obtained from www.uoneuro.uoregon.edu.
This work was supported by core funding from the Institute of Molecular and Cell Biology and from a Lee Kong Chian School of Medicine, Nanyang Technological University Start-up grant to SJ. SK and QL were supported by NGS fellowships from the National University of Singapore.
Availability of data and materials
The code and data used are available in the Figshare repository, https://figshare.com/projects/Characterization_of_a_thalamic_nucleus_mediating_habenula_responses_to_changes_in_ambient_illumination/18790.
Experiments were designed by CRK, QL and SJ. CRK carried out two-photon imaging and optogenetic experiments with anion channelrhodopsins. SK developed software and analyzed imaging data. QL performed parapineal lesioning. CK generated the UAS:GCaMP6s line. SJ performed wide-field imaging and analysis, dye tracing, antibody label, ChR2 manipulation and wrote the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Stephenson-Jones M, Floros O, Robertson B, Grillner S. Evolutionary conservation of the habenular nuclei and their circuitry controlling the dopamine and 5-hydroxytryptophan (5-HT) systems. Proc Natl Acad Sci U S A. 2012;109:E164–73.View ArticlePubMedGoogle Scholar
- Bianco IH, Wilson SW. The habenular nuclei: a conserved asymmetric relay station in the vertebrate brain. Philos. Trans R Soc Lond B Biol Sci. 2009;364:1005–20.View ArticleGoogle Scholar
- Lee A, Mathuru AS, Teh C, Kibat C, Korzh V, Penney TB, et al. The habenula prevents helpless behavior in larval zebrafish. Curr Biol. 2010;20:2211–6.View ArticlePubMedGoogle Scholar
- Agetsuma M, Aizawa H, Aoki T, Nakayama R, Takahoko M, Goto M, et al. The habenula is crucial for experience-dependent modification of fear responses in zebrafish. Nat Neurosci. 2010;13:1354–6.View ArticlePubMedGoogle Scholar
- Zhang J, Tan L, Ren Y, Liang J, Lin R, Feng Q, et al. Presynaptic excitation via GABAB receptors in habenula cholinergic neurons regulates fear memory expression. Cell. 2016;166:716–28.View ArticlePubMedGoogle Scholar
- Matsumoto M, Hikosaka O. Representation of negative motivational value in the primate lateral habenula. Nat Neurosci. 2009;12:77–84.View ArticlePubMedGoogle Scholar
- Amo R, Fredes F, Kinoshita M, Aoki R, Aizawa H, Agetsuma M, et al. The habenulo-raphe serotonergic circuit encodes an aversive expectation value essential for adaptive active avoidance of danger. Neuron. 2014;84:1034–48.View ArticlePubMedGoogle Scholar
- Fowler CD, Lu Q, Johnson PM, Marks MJ, Kenny PJ. Habenular α5 nicotinic receptor subunit signalling controls nicotine intake. Nature. 2011;471:597–601.View ArticlePubMedPubMed CentralGoogle Scholar
- Aizawa H, Cui W, Tanaka K, Okamoto H. Hyperactivation of the habenula as a link between depression and sleep disturbance. Front Hum Neurosci. 2013;7:826.View ArticlePubMedPubMed CentralGoogle Scholar
- Chou M-Y, Amo R, Kinoshita M, Cherng B-W, Shimazaki H, Agetsuma M, et al. Social conflict resolution regulated by two dorsal habenular subregions in zebrafish. Science. 2016;352:87–90.View ArticlePubMedGoogle Scholar
- Golden SA, Heshmati M, Flanigan M, Christoffel DJ, Guise K, Pfau ML, et al. Basal forebrain projections to the lateral habenula modulate aggression reward. Nature. 2016;534:688–92.View ArticlePubMedPubMed CentralGoogle Scholar
- Jhou TC, Fields HL, Baxter MG, Saper CB, Holland PC. The rostromedial tegmental nucleus (RMTg), a GABAergic afferent to midbrain dopamine neurons, encodes aversive stimuli and inhibits motor responses. Neuron. 2009;61:786–800.View ArticlePubMedPubMed CentralGoogle Scholar
- Quina LA, Tempest L, Ng L, Harris JA, Ferguson S, Jhou TC, et al. Efferent pathways of the mouse lateral habenula. J Comp Neurol. 2015;523:32–60.View ArticlePubMedGoogle Scholar
- Morley BJ, Spangler KM, Javel E. The development of somatostatin immunoreactivity in the interpeduncular nucleus of the cat. Brain Res. 1985;352:241–8.View ArticlePubMedGoogle Scholar
- Wang RY, Aghajanian GK. Physiological evidence for habenula as major link between forebrain and midbrain raphe. Science. 1977;197:89–91.View ArticlePubMedGoogle Scholar
- Appelbaum L, Wang GX, Maro GS, Mori R, Tovin A, Marin W, et al. Sleep-wake regulation and hypocretin-melatonin interaction in zebrafish. Proc Natl Acad Sci U S A. 2009;106:21942–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Hong S, Hikosaka O. The globus pallidus sends reward-related signals to the lateral habenula. Neuron. 2008;60:720–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Dreosti E, Vendrell Llopis N, Carl M, Yaksi E, Wilson SW. Left-right asymmetry is required for the habenulae to respond to both visual and olfactory stimuli. Curr Biol. 2014;24:440–5.View ArticlePubMedPubMed CentralGoogle Scholar
- Krishnan S, Mathuru AS, Kibat C, Rahman M, Lupton CE, Stewart J, et al. The right dorsal habenula limits attraction to an odor in zebrafish. Curr Biol. 2014;24:1167–75.View ArticlePubMedGoogle Scholar
- Miyasaka N, Morimoto K, Tsubokawa T, Higashijima S-I, Okamoto H, Yoshihara Y. From the olfactory bulb to higher brain centers: genetic visualization of secondary olfactory pathways in zebrafish. J Neurosci. 2009;29:4756–67.View ArticlePubMedGoogle Scholar
- Zhao H, Rusak B. Circadian firing-rate rhythms and light responses of rat habenular nucleus neurons in vivo and in vitro. Neuroscience. 2005;132:519–28.View ArticlePubMedGoogle Scholar
- Sakhi K, Wegner S, Belle MDC, Howarth M, Delagrange P, Brown TM, et al. Intrinsic and extrinsic cues regulate the daily profile of mouse lateral habenula neuronal activity. J Physiol. 2014;592:5025–45.View ArticlePubMedPubMed CentralGoogle Scholar
- Semm P, Demaine C. Electrophysiology of the pigeon’s habenular nuclei: evidence for pineal connections and input from the visual system. Brain Res Bull. 1984;12:115–21.View ArticlePubMedGoogle Scholar
- Bianco IH, Engert F. Visuomotor transformations underlying hunting behavior in zebrafish. Curr Biol. 2015;25:831–46.View ArticlePubMedPubMed CentralGoogle Scholar
- Amo R, Aizawa H, Takahoko M, Kobayashi M, Takahashi R, Aoki T, et al. Identification of the zebrafish ventral habenula as a homolog of the mammalian lateral habenula. J Neurosci. 2010;30:1566–74.View ArticlePubMedGoogle Scholar
- Burrill JD, Easter Jr SS. Development of the retinofugal projections in the embryonic and larval zebrafish (Brachydanio rerio). J Comp Neurol. 1994;346:583–600.View ArticlePubMedGoogle Scholar
- Robles E, Laurell E, Baier H. The retinal projectome reveals brain-area-specific visual representations generated by ganglion cell diversity. Curr Biol. 2014;24:2085–96.View ArticlePubMedGoogle Scholar
- Turner KJ, Hawkins TA, Yáñez J, Anadón R, Wilson SW, Folgueira M. Afferent connectivity of the zebrafish habenulae. Front Neural Circuits. 2016;10:30.View ArticlePubMedPubMed CentralGoogle Scholar
- Butler AB, Saidel WM. Clustered phylogenetic distribution of nucleus rostrolateralis among ray-finned fishes. Brain Behav Evol. 2003;62:152–67.View ArticlePubMedGoogle Scholar
- Saidel WM. Nucleus rostrolateralis: an expansion of the epithalamus in some actinopterygii. Anat Rec. 2013;296:1594–602.View ArticleGoogle Scholar
- Hendricks M, Jesuthasan S. Asymmetric innervation of the habenula in zebrafish. J Comp Neurol. 2007;502:611–9.View ArticlePubMedGoogle Scholar
- Lin Q, Jesuthasan SJ. Masking of a circadian behavior in larval zebrafish involves the thalamo-habenula pathway. Sci Rep. 2017;7:4104.View ArticlePubMedPubMed CentralGoogle Scholar
- Chen T-W, Wardill TJ, Sun Y, Pulver SR, Renninger SL, Baohan A, et al. Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature. 2013;499:295–300.View ArticlePubMedPubMed CentralGoogle Scholar
- Mueller T. What is the thalamus in Zebrafish? Front Neurosci. 2012;6:64.View ArticlePubMedPubMed CentralGoogle Scholar
- Sherman SM, Guillery RW. The role of the thalamus in the flow of information to the cortex. Philos Trans R Soc Lond B Biol Sci. 2002;357:1695–708.View ArticlePubMedPubMed CentralGoogle Scholar
- Satou C, Kimura Y, Hirata H, Suster ML, Kawakami K, Higashijima S-I. Transgenic tools to characterize neuronal properties of discrete populations of zebrafish neurons. Development. 2013;140:3927–31.View ArticlePubMedGoogle Scholar
- Moore RY, Speh JC. GABA is the principal neurotransmitter of the circadian system. Neurosci Lett. 1993;150:112–6.View ArticlePubMedGoogle Scholar
- Venugopalan V, Guerra 3rd A, Nahen K, Vogel A. Role of laser-induced plasma formation in pulsed cellular microsurgery and micromanipulation. Phys Rev Lett. 2002;88, 078103.View ArticlePubMedGoogle Scholar
- Vogel A, Noack J, Hüttman G, Paltauf G. Mechanisms of femtosecond laser nanosurgery of cells and tissues. Appl Phys B. 2005;81:1015–47.View ArticleGoogle Scholar
- Govorunova EG, Sineshchekov OA, Janz R, Liu X, Spudich JL. Natural light-gated anion channels: A family of microbial rhodopsins for advanced optogenetics. Science. 2015;349:647–50.View ArticlePubMedPubMed CentralGoogle Scholar
- Mohamed GA, Cheng R-K, Ho J, Krishnan S, Mohammad F, Claridge-Chang A, et al. Optical inhibition of zebrafish behavior with anion channelrhodopsins. BMC Biology. 2017;15:103.Google Scholar
- Takeuchi T, Duszkiewicz AJ, Sonneborn A, Spooner PA, Yamasaki M, Watanabe M, et al. Locus coeruleus and dopaminergic consolidation of everyday memory. Nature. 2016;537:357–62.View ArticlePubMedPubMed CentralGoogle Scholar
- Raimondo JV, Kay L, Ellender TJ, Akerman CJ. Optogenetic silencing strategies differ in their effects on inhibitory synaptic transmission. Nat Neurosci. 2012;15:1102–4.View ArticlePubMedPubMed CentralGoogle Scholar
- Fernandes AM, Fero K, Driever W, Burgess HA. Enlightening the brain: linking deep brain photoreception with behavior and physiology. Bioessays. 2013;35:775–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Davies WIL, Tamai TK, Zheng L, Fu JK, Rihel J, Foster RG, et al. An extended family of novel vertebrate photopigments is widely expressed and displays a diversity of function. Genome Res. 2015;25:1666–79.View ArticlePubMedPubMed CentralGoogle Scholar
- Fernandes AM, Fero K, Arrenberg AB, Bergeron SA, Driever W, Burgess HA. Deep brain photoreceptors control light-seeking behavior in zebrafish larvae. Curr Biol. 2012;22:2042–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Aarts E, Verhage M, Veenvliet JV, Dolan CV, van der Sluis S. A solution to dependency: using multilevel analysis to accommodate nested data. Nat Neurosci. 2014;17:491–6.View ArticlePubMedGoogle Scholar
- Warp E, Agarwal G, Wyart C, Friedmann D, Oldfield CS, Conner A, et al. Emergence of patterned activity in the developing zebrafish spinal cord. Curr Biol. 2012;22:93–102.View ArticlePubMedGoogle Scholar
- Yáñez J, Busch J, Anadón R, Meissl H. Pineal projections in the zebrafish (Danio rerio): overlap with retinal and cerebellar projections. Neuroscience. 2009;164:1712–20.View ArticlePubMedGoogle Scholar
- Jetti SK, Vendrell-Llopis N, Yaksi E. Spontaneous activity governs olfactory representations in spatially organized habenular microcircuits. Curr Biol. 2014;24:434–9.View ArticlePubMedGoogle Scholar
- Steenbergen PJ, Richardson MK, Champagne DL. Patterns of avoidance behaviours in the light/dark preference test in young juvenile zebrafish: a pharmacological study. Behav Brain Res. 2011;222:15–25.View ArticlePubMedGoogle Scholar
- Saidel WM, Butler AB. Retinal projections in the freshwater butterfly fish, Pantodon buchholzi (Osteoglossoidei). II. Differential projections of the dorsal and ventral hemiretinas. Brain Behav Evol. 1991;38:154–68.View ArticlePubMedGoogle Scholar
- Zhang B-B, Yao Y-Y, Zhang H-F, Kawakami K, Du J-L. Left habenula mediates light-preference behavior in zebrafish via an asymmetrical visual pathway. Neuron. 2017;93:914–28.e4.View ArticlePubMedGoogle Scholar
- Mueller T, Guo S. The distribution of GAD67-mRNA in the adult zebrafish (teleost) forebrain reveals a prosomeric pattern and suggests previously unidentified homologies to tetrapods. J Comp Neurol. 2009;516:553–68.View ArticlePubMedPubMed CentralGoogle Scholar
- Hattar S, Kumar M, Park A, Tong P, Tung J, Yau K-W, et al. Central projections of melanopsin-expressing retinal ganglion cells in the mouse. J Comp Neurol. 2006;497:326–49.View ArticlePubMedPubMed CentralGoogle Scholar
- Morin LP, Studholme KM. Retinofugal projections in the mouse. J Comp Neurol. 2014;522:3733–53.View ArticlePubMedPubMed CentralGoogle Scholar
- Díaz C, Puelles L. Afferent connections of the habenular complex in the lizard Gallotia galloti. Brain Behav Evol. 1992;39:312–24.View ArticlePubMedGoogle Scholar
- Kemali M, Guglielmotti V, Gioffré D. Neuroanatomical identification of the frog habenular connections using peroxidase (HRP). Exp Brain Res. 1980;38:341–7.View ArticlePubMedGoogle Scholar
- Marburg O. The structure and fiber connections of the human habenula. J Comp Neurol. 1944;80:211–33.View ArticleGoogle Scholar
- Cragg BG. The connections of the habenula in the rabbit. Exp Neurol. 1961;3:388–409.View ArticlePubMedGoogle Scholar
- Moore RY, Weis R, Moga MM. Efferent projections of the intergeniculate leaflet and the ventral lateral geniculate nucleus in the rat. J Comp Neurol. 2000;420:398–418.View ArticlePubMedGoogle Scholar
- Oh SW, Harris JA, Ng L, Winslow B, Cain N, Mihalas S, et al. A mesoscale connectome of the mouse brain. Nature. 2014;508:207–14.View ArticlePubMedPubMed CentralGoogle Scholar
- Ely BA, Xu J, Goodman WK, Lapidus KA, Gabbay V, Stern ER. Resting-state functional connectivity of the human habenula in healthy individuals: associations with subclinical depression. Hum Brain Mapp. 2016;37:2369–84.View ArticlePubMedPubMed CentralGoogle Scholar
- Torrisi S, Nord CL, Balderston NL, Roiser JP, Grillon C, Ernst M. Resting state connectivity of the human habenula at ultra-high field. Neuroimage. 2017;147:872–9.View ArticlePubMedGoogle Scholar
- Teh C, Chudakov DM, Poon K-L, Mamedov IZ, Sek J-Y, Shidlovsky K, et al. Optogenetic in vivo cell manipulation in KillerRed-expressing zebrafish transgenics. BMC Dev Biol. 2010;10:110.View ArticlePubMedPubMed CentralGoogle Scholar
- Baier H, Scott EK. Genetic and optical targeting of neural circuits and behavior--zebrafish in the spotlight. Curr Opin Neurobiol. 2009;19:553–60.View ArticlePubMedPubMed CentralGoogle Scholar
- DA Wolf S, Bertoni T, Böhm UL, Bormuth V, Candelier R, Karpenko S, et al. Sensorimotor computation underlying phototaxis in zebrafish. Nat Commun. 2017;8:651. Google Scholar
- Arrenberg AB, Del Bene F, Baier H. Optical control of zebrafish behavior with halorhodopsin. Proc Natl Acad Sci U S A. 2009;106:17968–73.View ArticlePubMedPubMed CentralGoogle Scholar
- Freeman J, Vladimirov N, Kawashima T, Mu Y, Sofroniew NJ, Bennett DV, et al. Mapping brain activity at scale with cluster computing. Nat Methods. 2014;11:941–50.View ArticlePubMedGoogle Scholar
- Wyart C, Del Bene F, Warp E, Scott EK, Trauner D, Baier H, et al. Optogenetic dissection of a behavioural module in the vertebrate spinal cord. Nature. 2009;461:407–10.View ArticlePubMedPubMed CentralGoogle Scholar