Mediodorsal thalamus and ventral pallidum contribute to subcortical regulation of the default mode network
Data files
Mar 01, 2024 version files 505.21 KB
Abstract
Humans and other animals readily transition from externally to internally focused attention, and these transitions are accompanied by coactivation of a group of brain regions collectively known as the default mode network (DMN). While the DMN was considered a cortical network, recent evidence suggests subcortical structures are part of the DMN. Here we investigated the role of ventral pallidum (VP) and mediodorsal thalamus (MD) in DMN regulation in the tree shrew, a close relative of primates. We combine electrophysiology and deep learning-based motion tracking to perform unsupervised classification of behavioral states. We found gamma oscillations in VP and MD coordinated with gamma in the anterior cingulate (AC) cortex specifically during DMN states. Similar enhancements were found for high gamma, but only at subcortical sites. Cross-frequency coupling between gamma and delta oscillations were higher during DMN than other behaviors, underscoring the engagement of MD, VP, and AC circuits. Our findings highlight the importance of VP in DMN regulation in the tree shrew, consistent with rodent studies, and demonstrate a role for MD thalamus in DMN regulation. Our results extend homologies in DMN regulation among mammals, and underline the importance of thalamus and basal forebrain to the regulation of DMN brain states.
README
Variables
Fig1B
Deep Lab Cut likelihood before and after relabeling of misclassified frames.
This data set contains 4 variables. Likelihood from two body parts(neck and nose) before and after relabelling misclassified frames, 12 video segments, 2400 frames each.
nk1: Neck position likelihood before relabelling misclassified frames.
nk2: Neck position likelihood after relabelling misclassified frames.
ns1: Nose position likelihood before relabelling misclassified frames.
ns2: Nose position likelihood after relabelling misclassified frames.
Fig1C
Scatter plot shows the likelihood before and after relabeling from 12 segments of a 4-hour session.
This data set contains 4 variables.
NK_bf: Average likelihood of neck position from each video segment before relabelling misclassified frames.
NK_af: Average likelihood of neck position from each video segment after relabelling misclassified frames.
NS_bf: Average likelihood of nose position from each video segment before relabelling misclassified frames.
NS_af: Average likelihood of nose position from each video segment after relabelling misclassified frames.
Fig1D1E1G
D, a 3-hour segment showing both accelerometer (ACL) and nose movement speed signals from DeepLabCut (DLC). E, a 10-minute segment taken from D with a detailed view of the complementary movement sensor signals ACL and DLC. G. Three states assigned by Hidden Markov Model based on ACL/DLC input.
This data set contains 6 variables.
ACL3h: 3 hours of acceleration (m/s2) data in 2fps.
DLC3h: 3 hours of DeepLabCut speed (a.u.) in 2fps aligned with ACL3h.
ACL10m: 10 minutes of acceleration (m/s2) data in 2fps.
DLC10m: 10 minutes of DeepLabCut speed (a.u.) in 2fps aligned with ACL10m.
st/en each has 3 cells. Each cell contains the starting frame(st) or ending frame(en) of HMM state 1, state 2 or state 3.
Fig2A2B
A. Group analysis of the transition probability for the Hidden Markov Model. B. Group analysis of the output probability distribution of the Hidden Markov Model. Here, Lo/Hi refers to low ACL and high DLC activation.
This data set contains 6 variables.
Tr_From_state1, Tr_From_state2, Tr_From_state3: The transition probability.
Tr_From_state1: Contains two dimensions. First dimension refers to the state that transfer from state 1 to (state 1, state 2, state 3). Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Tr_From_state2: Contains two dimensions. First dimension refers to the state that transfer from state 2 to (state 1, state 2, state 3). Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Tr_From_state3: Contains two dimensions. First dimension refers to the state that transfer from state 3 to (state 1, state 2, state 3). Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Output_state1, Output_state2, Output_state3: The observation probability distribution matrix.
Output_state1: Contains two dimensions. First dimension refers to the output symbols (Low/Low, High/High, Low/High and High/Low combinations of ACL and DLC signal) distribution in state 1. Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Output_state2: Contains two dimensions. First dimension refers to the output symbols (Low/Low, High/High, Low/High and High/Low combinations of ACL and DLC signal) distribution in state 2. Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Output_state3: Contains two dimensions. First dimension refers to the output symbols (Low/Low, High/High, Low/High and High/Low combinations of ACL and DLC signal) distribution in state 3. Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Fig3A
Shows an example of the typically high correspondence between the states output by the HMM, and the behavioral sets derived from the manually labelled data.
This data set contains 3 variables.
sort_ann: Including starting frame, end frame and the manual behavior tag.
on: Including three cells, onset frame of HMM state 1, state 2 and state 3.
off: Including three cells, offset frame of HMM state 1, state 2 and state 3.
Fig3B-3D
B-D, pie charts showing the composition of different behaviors observed in the 3 HMM states.
This data set contains 3 variables.
SLP reflects different manual labels contribution in HMM State 1. The order is sleep posture and other.
DMN reflects different manual labels contribution in HMM State 2. The order is quiet awake, eating/drinking, grooming, in or near nest box, unclassified and other.
EXP reflects different manual labels contribution in HMM State 3. The order is locomotion, exploration, unclassified and other.
Fig3E
Mean overlap of the HMM output and manually scored behavioral sets.
This data set contains 1 variable.
ovl_all has two dimensions. First dimension reflects 5 datasets. Second dimension reflects the overlap percentage of the HMM state (HMM state 1, HMM state 2, HMM state 3) output and manually scored behavior sets.
Fig3F
Overall distribution of time spent in the different states revealed by the HMM for all animals and all sessions.
Timespent: Contains two dimensions. First dimension refers to the distribution of time spent in each HMM state (HMM state 1, HMM state 2, HMM state 3). Second dimension refers to different animals (1777, 1795, 2172), all the repeats of probability value are in each cell.
Fig4A
Example PSD of LFPs recorded from a single animal in the VP during the three HMM derived states.
This data set contains 4 variables showing PSD of an example site during three HMM-defined states.
Freq reflects LFP frequency. Sleep_PSD, DMN_PSD, Exp_PSD reflect Power Spectrum Density of HMM-defined sleeping posture state, DMN activation state, and locomotion/exploration state respectively.
Fig4B
Scatter plots showing the spectral power at gamma frequencies (40-60 Hz) during DMN related behavioral activity vs exploratory behaviors in the four brain areas.
This data set contains 8 variables. Each has three cells, represents three animals (1795, 2172, 1777), V1 only has data from 2172 and 1777.
V1_DMNG, MD_DMNG, AC_DMNG, VP_DMNG show average Gamma band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined DMN activation state (DMN).
V1_ExpG, MD_ExpG, AC_ExpG, VP_ExpG show average Gamma band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined locomotion/exploration state (Exp).
Fig4C
Average magnitude of the difference in gamma power between DMN and exploration behaviors in the four brain areas.
"G" contains 4 cells show the magnitude of PSD difference in Gamma band between DMN activation state and locomotion/exploration state, each cell represents one brain region: V1, MD, AC and VP respectively.
Fig4D
Same as (B,C) but for the high gamma band (60-150 Hz).
This data set contains 8 variables. Each has three cells, represents three animals (1795, 2172, 1777), V1 only has data from 2172 and 1777.
V1_DMNH, MD_DMNH, AC_DMNH, VP_DMNH show average High Gamma band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined DMN activation state (DMN).
V1_ExpH, MD_ExpH, AC_ExpH, VP_ExpH show average High Gamma band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined locomotion/exploration state (Exp).
Fig4E
Same as (B,C) but for the high gamma band (60-150 Hz).
"HG" contains 4 cells show the magnitude of PSD difference in High Gamma band between DMN activation state and locomotion/exploration state, each cell represents one brain region: V1, MD, AC and VP respectively.
Fig4F
Average directional interaction strength (Granger causality) calculated for the gamma band (40-60 Hz) between VP, AC, MD, V1 during DMN-related states. Line thickness proportional to the magnitude of the Granger Causality.
This data set contains 1 variable.
c1 includes the average gamma band directional interaction strength of MD>V1; AC>V1; VP>V1; V1>MD; AC>MD; VP>MD; V1>AC; MD>AC; VP>AC; V1>VP; MD>VP; AC>VP respectively.
Fig4G
Directional interactions among DMN-related sites (VP, AC, MD) during DMN-related states for gamma, left, and high-gamma band, right.
This data set contains 2 variables.
G_Granger has two dimensions. First dimension is the 15 sessions. And second dimension corresponds to the gamma band directional interaction strength during DMN state, of MD>AC; MD>VP; AC>MD; AC>VP; VP>MD; VP>AC respectively.
HG_Granger also has two dimensions. First dimension is the 15 sessions. And second dimension corresponds to the high gamma band directional interaction strength during DMN state, of MD>AC; MD>VP; AC>MD; AC>VP; VP>MD; VP>AC respectively.
Fig4H
Scatter plots comparing overall Granger causality values within VP, AC and MD between DMN and exploratory behavioral states.
This data set contains 4 variables.
G_Exp is the gamma band directional interaction strength of six directions between MD, AC and VP during locomotion/exploration state.
G_DMN is the gamma band directional interaction strength of six directions between MD, AC and VP during DMN state.
HG_Exp is the high gamma band directional interaction strength of six directions between MD, AC and VP during locomotion/exploration state.
HG_DMN is the high gamma band directional interaction strength of six directions between MD, AC and VP during DMN state.
Fig5A5B
A, example of average gamma power (40-60 Hz) for behavioral epochs in chronological order throughout a single session. B, average gamma amplitude of all epochs for the different behaviors in the session shown in A.
This data set contains 4 variables.
Gamma_example: the average gamma power in VP of all the behavior epochs from one example session.
Behavior_example: corresponding behavior labels of all the epochs.
mean_Gamma: mean gamma power of each behavior label (sleep, quiet awake, eating/drinking, grooming, nest box, locomotion, exploring).
epo_n: number of epoch belongs to each behavior label.
Fig5C
Scatter plots showing gamma amplitude during each of the DMN related behaviors vs gamma power during active states (Locomotion and Exploration) in all five manual labelled datasets.
This data set contains 2 variables.
act_amp represents the average gamma amplitude during active states (locomotion and exploration).
DMN_amp has two dimention, shows the average gamma amplitude during each DMN state. First dimension is five datasets, second dimension represents different DMN states (Quiet awake, Eating&drinking, Grooming, Nest box).
Fig6A
Example of LFP power spectral density (PSD) in AC for three behavioral states.
This data set contains 4 variables showing PSD of an example site during three HMM-defined states.
Freq reflects LFP frequency. Sleep_PSD, DMN_PSD, Exp_PSD reflect Power Spectrum Density of HMM-defined sleeping posture state, DMN activation state, and locomotion/exploration state respectively.
Fig6B
Scatter plots showing delta band (0.5-4 Hz) PSD during sleep posture vs active states (including DMN and Exploratory states) in the four brain areas.
This data set contains 12 variables. Each has three cells, represents three animals (1795, 2172, 1777), V1 only has data from 2172 and 1777.
V1_DMND, MD_DMND, AC_DMND, VP_DMND show average Delta band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined DMN activation state (DMN). In Fig6B, they are shown as active state.
V1_ExpD, MD_ExpD, AC_ExpD, VP_ExpD show average Delta band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined locomotion/exploration state (Exp). In Fig6B, they are shown as active state.
V1_SLPD, MD_SLPD, AC_SLPD, VP_SLPD show average Delta band PSD from different brain regions (V1, MD, AC, VP) during HMM-defined sleeping posture state (SLP).
Fig6C
Average magnitude of the difference between delta band PSD during sleep posture and two active states (DMN and Exploration).
This data set contains 2 variables.
" DMNd " contains 4 cells show the magnitude of PSD difference in Delta band between sleep posture state and DMN activation state, each cell represents one brain region: V1, MD, AC and VP respectively.
" EXPd " contains 4 cells show the magnitude of PSD difference in Delta band between sleep posture state and locomotion/exploration state, each cell represents one brain region: V1, MD, AC and VP respectively.
Fig6D
Average delta amplitude of each epoch in chronological order throughout one session.
This data set contains 2 variables.
Delta_example: the average delta power in AC of all the behavior epochs from one example session.
Behavior_example: corresponding behavior labels of all the epochs.
Fig6E
Warren-Sarle test of bimodality showing the delta amplitude distribution of sleep posture epochs was bimodal for the example session.
This data set contains 4 variables.
delta1: the delta amplitude range.
occurences1: the occurrences in each delta amplitude range during sleep posture state.
delta2: the delta amplitude range.
occurences2: the occurrences in each delta amplitude range during other states (DMN state and locomotion/exploration state).
Fig6F
Across the datasets, the percentage of sleeping posture in epochs with high delta activity occurred.
This data set contains 1 variable.
perc_sws: The percentage of sleep posture state occurrence in epochs with high delta activities from 5 datasets.
Fig7A
A raw segment of the LFP taken from MD thalamus with prominent delta and gamma activity is shown at top, at the bottom the amplitude of delta and gamma power for the same segment.
This data set contains 4 variables.
RawData reflects a LFP raw data segment from MD.
Time reflects the time point of the data points.
Delta shows the delta band (0.5-4 Hz) filtered from the example segments.
Gamma_env shows the gamma band (40-60 Hz) envelope from the example segments.
Fig7B
The delta-gamma cross-frequency coupling in polar format. The cfc vector is computed to the centroid of the angular cfc distribution
This data set contains 4 variables.
px, py: coordinates of cfc distribution in polar format of an example dataset.
centroidx, centroidy: centroid coordinate of the angular cfc distribution.
Fig7CDE
The cfc vector distribution for sleeping posture C, DMN behaviors D, locomotion/exploration E, in MD, AC and VP.
This data set contains 3 variables.
xycentroidSLP shows the centroid coordinates of the angular cfc distribution of each dataset in HMM-defined sleep posture state. It has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different datasets.
xycentroidDMN shows the centroid coordinates of the angular cfc distribution of each dataset in HMM-defined DMN state. It has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different datasets.
xycentroidEXP shows the centroid coordinates of the angular cfc distribution of each dataset in HMM-defined locomotion/exploration state. It has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different datasets.
Fig7FG
F, Cfc coupling strength in MD, AC and VP. G, cfc preferred angle in MD, AC and VP.
This data set contains 4 variables.
CouplStrength shows the mean cfc coupling strength. Data has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different HMM behavioral states (HMM-defined sleep posture state, HMM-defined DMN state, HMM-defined locomotion/exploration state).
CouplStrengthSD shows the standard deviation of cfc coupling strength. Data has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different HMM behavioral states (HMM-defined sleep posture state, HMM-defined DMN state, HMM-defined locomotion/exploration state).
PrefAngle shows the mean cfc preferred angle. Data has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different HMM behavioral states (HMM-defined sleep posture state, HMM-defined DMN state, HMM-defined locomotion/exploration state).
PrefAngleSD shows the standard deviation of cfc preferred angle. Data has two dimensions, first dimension is the brain regions (MD, AC, VP) and second dimension is the different HMM behavioral states (HMM-defined sleep posture state, HMM-defined DMN state, HMM-defined locomotion/exploration state).
Methods
Materials and methods
The local ethical committee on animal experimentation (canton of Fribourg), approved all experimental procedures.
Animals. Three adult tree shrews, T. belangeri, of either sex were housed under a 13/11 LD cycle in a 3 m3 cage with branches, some enrichment elements, and ad libitum access of food and water. The cage was connected to a nest box with a tube.
Surgical Procedures. Animals first received i.m injections of Alfaxan (40 mg/kg) to induce anaesthesia and Atropine (0.08 mg/kg) to prevent secretions. Animals were then intubated using a modified otoscope (Bebird, Alhambra, CA), ventilated at 100 bpm (Small Animal Ventilator, Harvard Apparatus, Cambridge, MA), and placed in a stereotactic frame (David Kopf Instruments Tujunga, CA). Anaesthesia was maintained with isoflurane (1-3%) in pure oxygen, and end tidal CO2 was monitored (Physiosuite, Kent Scientific Torrington CT) and maintained at ~4%. Lidocaine (0.5 ml 1%) was injected near the incision site, a midline incision was made and the skull was exposed. Three 1.5 mm stainless-steel bone screws (WPI Hertfordshire, UK) were implanted with two located above the cerebellum as a reference and ground. Burr holes were drilled, and epoxy coated tungsten electrodes (FHC Bowdoin ME) with a tip resistance ~ 150 kW were lowered to the recording sites: Ventral pallidum (AP 7.6 mm, ML 3.0 mm, DV -7.8 mm), anterior cingulate cortex ( AP 11.1 mm, ML 0.8 mm, DV -1.5 mm) mediodorsal thalamus (AP 4.9 mm, ML 1.0 mm, DV -5.4 mm) and primary visual cortex (AP 2.0 mm, ML 1.4 mm, DV -1.0 mm). All coordinates are from the interaural line. Electrodes were fixed to the skull with super glue (LOCTITE, Westlake OH) and Paladur dental cement (Kulzur Inc. Hanau Germany). Electrodes were wired to a socket connector, and the connector was attached to the skull with dental acrylic. The incision was closed about the connector with sutures, and the animal was allowed to recover for at least one week prior to testing.
Data Acquisition. LFP and accelerometer data were collected using a wireless battery-powered data logger (Neurologger 2A, Zürich Switzerland), additionally, an infrared receiver on the neurologger was used for aligning the LFP and accelerometer data with video recordings. All channels of neural signals and accelerometer data were digitized at 400 Hz and no further filtering was performed on the LFP data.
Home cage Recording. Video recordings of the animals in their home cage were made using a wide field, 103⁰x58⁰, CMOS camera (DS-2CD2143G0-IS, HIKVISION, Hangzhou China) mounted on top of the cage. After connecting the Neurologger, tree shrews were initially kept in their nest box for 10 min in order to acclimate to the Neurologger device. Home cage recordings typically lasted for 5-6 hours, between 9:00-18:00 during the animals’ perspective daytime.
DeepLabCut Tracking. DeepLabCut (DLC) (Mathis et al., 2018; Nath et al., 2019) was used to track the animal’s location in the home cage. First, videos were pre-processed by cropping appropriately and down sampling to 2 fps, and then cut into 20 minute segments. The model was trained for 500,000 iterations after manually labelling 50 frames from each video segment. For each frame, we labelled the nose and neck. The output consisted of body part coordinates in x,y coordinates and the corresponding likelihood estimate, using the same model to analyse all the video segments from the same animal. In most circumstances, DLC can accurately track the tree shrews in most of the frames. For those videos with obvious missing frames, an additional 50 frames were hand labelled, and the network was retrained with additional 200,000 iterations.
Preprocessing and Spectral Analysis. We partitioned the LFP data into 0.5-s epochs for further analysis. Power spectra were calculated by fast Fourier transforms (FFT). We calculated the band power by calculating the mean value of the power spectrum between 40 - 60 Hz (gamma band) and 60-150 Hz (high gamma band).
Hidden Markov Model.
We designed an HMM with three states to capture the three groups of behavioural states detailed in the results, and four output symbols corresponding to High/High, High/Low, Low/High, and Low/Low combinations of ACL and DLC signal values based on preliminary observations of our data. The thresholds between High and Low sensor values were determined based on the median value of the signal across the recording session. We used maximum likelihood estimation to find HMM state transition and output symbol emission probabilities and the Viterbi algorithm to compute the most probable state sequence given the estimated parameters.
Granger Causality.
To test the information transfer between VP, AC, MD, and V1 brain regions, we used LFPs and a multivariate linear vector autoregressive (VAR) model from Matlab Multivariate Granger Causality (MVGC) toolbox (Barnett and Seth, 2014) for granger causality analyses. The maximum model order for model order estimation was 20 ms, and Akaike information criteria (AIC) was used. The model parameter for the VAR model estimation was the locally weighted linear regression (LWR). We used F-testing with a false discovery rate (Q < 0.05) for the pairwise conditional Granger causality estimation. Kruskal-Wallis test was applied to determine the significance between the information transfer directions. T-test was applied to compare the significance of granger causality between the DMN state and the exploration state.
Cross-frequency Coupling
We first filtered our raw data with butterworth band pass filter into delta (0.5-4 Hz) and gamma (40-60 Hz) bands. Then we applied the Hilbert transform to extract the instantaneous phase from the delta band and the instantaneous amplitude from the gamma band. We display the cross-frequency coupling in polar format (Fig. 7B).
Statistical analyses reproducibility.
Experiment details are provided in the text and Materials and Methods section. All the statistical analyses were performed in Matlab. In comparison of likelihood in Fig.1, we applied the non-parametric Wilcoxon signed rank test as the data did not follow a normal distribution. For the same reason, to compare granger causality values within VP, AC, and MD, we used Kruskal-Wallis test (Fig.4G). For normally distributed data with equal variance, t-test or ANOVA tests were applied according to group number. In Fig.6E, we applied the Warren-Sarle test in order to assess the bimodality of the delta band amplitude distribution for sleep posture epochs. In Fig.7G, we used the Circular Median test (Matlab CircStat Toolbox) for a non-parametric multi-sample test of equal medians for circular data.