Medium spiny neurons activity reveals the discrete segregation of mouse dorsal striatum
- Instituto de Neurociencias CSIC-UMH, Spain
Figures
Figure 1 with 2 supplements
Analysis and classification of SWO in dorsal striatum.
(A) Schematic representation of the in vivo recording setup. (B) Morphological reconstruction of DLS-MSN (left) and DMS-MSN (right). Different scales show neuron magnitude and its dendritic spines, confirming that the recorded neuron is a MSN. (C) Representative LFP (top) and whole-cell patch-clamp recording of a MSN (bottom). (D) Featurization of the SWO of DLS- and DMS-MSNs (see Materials and methods). (E) RFE of the computed features (feature number) to determine their order of relevance classifying DLS- and DMS-MSNs. A SVM with a linear kernel was selected as classification algorithm. (F) Example ROC curve of the DLS-DMS classification for one of the crossvalidations using the three most relevant features, illustrated in D. (G) Schematic representation of the data distribution of DLS and DMS groups for the most relevant features (numbers 11, 5, and 9, respectively). White number represent the mean; the radius of the circle represents the variance. (H) Subspace of classification based on the RFE results. Dots represent individual MSNs recorded in DLS (red) or DMS (blue). Classification hyperplane obtained after training the SVM with linear kernel in black. p Values obtained using the Wilcoxon Rank-sum test.
Figure 1—figure supplement 1
Classification of DLS- and DMS-MSNs at animal level.
(A) One-animal-out classification, ROC curve of the DLS/DMS. (B) Subspace of classification based on the RFE results. Each dot represents the average of all neurons recorded in single mouse, in DLS (red) or DMS (blue).
Figure 1—figure supplement 2
Correlation between the SWO features and electrophysiological properties of MSNs.
Correlation coefficient of the 13 SWO features, showed in Figure 1 and the intrinsic properties of MSNs. Note that all coefficients are < 0.03.
Figure 2 with 2 supplements
Beta band in membrane voltage of dorsal striatal MSNs.
(A) Example of energy of the beta band (red trace) during the MSN SWO recorded in DMS. (B) Energy of the beta band of DLS- and DMS-MSNs. (C) Average of phase alignment of beta band to the SWO in DLS- (left) and DMS-MSNs (right) (Raylegh test, p<0.001 in both cases). Asterisk indicates the radial position of the beta peak. Beta peak occurs first in DMS-MSNs (p<0.001). (D) Representative examples of beta phase locking in DLS- (left) and DMS- MSNs (right). White line represents the beginning of each Up state. p Values in B obtained using the Wilcoxon Rank-sum test. p Values of the phase locking in C were computed using Rayleigh test.
Figure 2—figure supplement 1
Theta and Gamma band in membrane potential of dorsal striatum.
(A) Representative example of the energy of the theta band (orange trace) of a DLS-MSN during SWO regime. (B) Representative example of the energy of the Gamma band (red trace) of a DLS-MSN during SWO regime. (C) Energy of the Theta band of DLS- and DMS-MSNs. (D) Energy of the Gamma band of DLS- and DMS-MSNs. Same neuron in A and B. p Values obtained using the Wilcoxon Rank-sum test.
Figure 2—figure supplement 2
Example of decomposition of different traces by the NA-MEMD.
(A) Example traces of a S1 LFP (left), M1 LFP (center), and a whole-cell recording in DLS (right). (B) Consecutive IMFs obtained by applying NA-MEMD to the signals in a, starting from the seventh IMF. The IMF classifies the oscillatory activity from the fastest to the slowest.
Figure 3 with 1 supplement
Study of the SWO in the DCS striatum and determination of the functional boundary between DLS and DMS.
(A) Schematic representation of the ‘boundary question’. (B) Three possible hypotheses about DCS-MSNs SWO. Hyp. 1: There is an intermediate distribution between DLS and DMS SWO (top); Hyp. 2: There is a different type of SWO distribution (middle); Hyp. 3: There is no specific DCS distribution of SWO and the MSNs recorded at DCS are a combination of DLS- and DMS-MSNs (bottom). The black line displays the distribution of the SWO combining both DLS and DMS. (C) Distribution of DLS-, DMS-, and DCS-MSNs in the subspace of classification determined by the RFE. Classification plane in black. Orthogonal hypervector to the classification plane in gray. (D) Distribution of dorsolateral, dorsomedial and dorsocentral MSNs and the combining DLS-DMS function along the orthogonal hypervector to the hyperplane classification. All comparison between distributions are significant (p<0.01) except for DCS with DMS-DLS (p=0.681). (E) Percentage of neurons recorded in the DCS coordinate, responding or not to visual stimulation and classified as DLS or DMS. (F) Waveform average of visual responses recorded in an MSN from DCS (green) and DMS (blue). (G–I) Averages of onset (G), amplitude (H), and slope (I) of the visual responses recorded in MSNs from DCS coordinate and DMS. p Values obtained using the Wilcoxon Rank-sum test.
Figure 3—figure supplement 1
Z-scoring of the SWO features.
Figure 4
Number of peaks and depolarized/hyperpolarized ratio of the SWO in dorsal striatum.
(A) Detection of peaks (black arrows) in the Up state in a DLS-MSN (red) and a DLS-MSN (blue) (see Materials and methods). (B) Representative examples of SWO in different cortical regions recorded. (C) Number of peaks per Up state in the different cortical regions (left). Non-labeled comparisons are not significant. (D) Extraction of depolarized/hyperpolarized events by computational approach (see Materials and methods). Top: Detection of the events using a threshold in the first derivative of the Vm. Scale bar represents 0.1 dmV/dt. Bottom: Representation of the detected depolarized (green) and hyperpolarized (orange) events from whole-cell recording of a DLS-MSN (middle). (E) Average of the depolarized/hyperpolarized ratio aligned to the SWO cycle of DMS- and DLS-MSNs. Comparison of positive (p<0.0072) and negative (p<0.0398) values at peaks. (F) Grand average of the transitions from Down to Up (upper) and Up to Down (bottom). Shaded bars in E and F represent SEM. p Values obtained using the Wilcoxon Rank-sum test. In C, alpha values for multiple comparisons were corrected using Holm-Bonferroni correction.
Figure 5
Integration of the cortical SWO in dorsal striatum.
(A) Example of simultaneous recordings from a DLS-MSN (red) and double LFPs from FrA (purple) and S1 (black) together with SWO extraction by NA-MEMD (orange line). (B) Example of simultaneous recordings from a DMS-MSN (blue) and double LFP from M1 (green) and V1 (ochre) together with SWO extraction by NA-MEMD (orange line). (C) Histogram of the membrane potential values of the MSNs in A and B, the black arrow indicates the dissimilar shape of the bimodal distributions. (D) Correlation coefficient of different cortical regions to DLS- and DMS-MSNs. (E) Raster plot showing the correlation between all MSNs and the LFP from FrA and V1. (F) Probability of propagation of the Up state from different cortical regions to DLS- and DMS-MSNs. p Values obtained using the Wilcoxon Rank-sum test. In D and F, alpha values for multiple comparisons were corrected using Holm-Bonferroni correction.
Figure 6
Sequential activation of the dorsal striatum during the SWO.
(A) Representative example of simultaneous double in vivo whole-cell recordings in DLS (top, red) and DMS (bottom, blue). (B) Inset of the shaded part in A, showing two aligned membrane potential traces to their Down states. Notice that DLS-MSN onset is preceding the DMS-MSN. (C) Raster plot of the same example of a paired recording in DLS (top) and DMS (bottom) MSNs. Each line represents a 1 s time window aligned to the onset of each of the Up states of the DLS-MSN (white line). (D) Example of the distribution of delays between the onset of the Up state in the DMS- relative to the DLS-MSN. Positive values indicate that the Up state arrive later to the DMS-MSN. Same neuron in A, B, C, and D. (E) Directional probability of DS-MSNs, obtained from pairs of whole-cell recordings (N = 6). (F) Directional probability of V1 and FrA cortical areas, obtained from pairs of LFPs (N = 54). p Values obtained using the Wilcoxon Rank-sum test.
Figure 7 with 1 supplement
Differences between direct and indirect pathway MSNs in dorsal striatum.
(A) Example showing an in vivo identification of an MSN using the optopatcher. Responses in D2-ChR2-YFP mice (top trace, ChR2+, green) to light pulses, inducing depolarization in the MSN. Negative cells (bottom trace, ChR2-, black) did not respond to light pulses. Blue squares indicate the intensity of the light pulse stimulation, from 20 to 100%. (B) RFE to optimize the classification of dMSNs and iMSNs in DLS (red) or DMS (blue). A SVM with a linear kernel was selected as classification algorithm. (C) Significant differences in three SWO features were found in DLS (red) between direct (dark) and indirect (light) pathways, but not in DMS (blue). p Values obtained using the Wilcoxon Rank-sum test.
Figure 7—figure supplement 1
Beta, Theta, and Gamma bands in membrane voltage of direct and indirect pathways in the DLS and DMS.
Energy of the Beta, Theta, and Gamma bands of direct and indirect DLS- and DMS-MSNs. p Values obtained using the Wilcoxon Rank-sum test.
Tables
Table 1
Intrinsic properties of DLS and DMS of direct and indirect MSNs.
Comparisons between DLS- and DMS-MSNs (p<0.05, *symbol). Comparisons between indirect DLS-MSNs and indirect DMS-MSNs (p<0.01, # symbol). All values are means ± SEM. p Values obtained using the Wilcoxon Rank-sum test.
| Input Resistance (MΩ) | Resistance Down state hyp. (MΩ) | Resistance Down state dep. (MΩ) | Resistance Up state hyp. (MΩ) | Resistance Up state dep. (MΩ) | Capacitance (pF) | Tau (ms) | |
|---|---|---|---|---|---|---|---|
| DLS | 312± 25 * | 292± 24 * | 313± 23 * | 304± 23 | 315± 27 | 19.45± 1.52 | 4.57± 0.23 |
| dMSNs | 297± 33 | 279± 31 | 284± 28 | 289± 30 | 296± 35 | 20.72± 1.98 | 4.52± 0.29 |
| iMSNs | 338± 41 # | 312± 36 # | 355± 40 # | 324± 37 | 342± 41 | 17.58± 2.35 | 4.63± 0.40 |
| DMS | 255± 16 * | 241± 20 * | 259± 17 * | 262± 20 | 262± 17 | 21.68± 1.55 | 4.61± 0.15 |
| dMSNs | 260± 22 | 249± 28 | 265± 23 | 269± 28 | 267± 23 | 23.20± 2.21 | 4.77± 0.19 |
| iMSNs | 241± 18 # | 223± 18 # | 248± 16 # | 248± 20 | 252± 20 | 18.43± 1.04 | 4.26± 0.24 |
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Genetic reagent (M. musculus) | BAC-Cre Drd2-44 or STOCK Tg(Drd2-cre)ER44Gsat/Mmcd | GENSAT | RRID:MMRRC_017263-UCD | Males and females used |
| Genetic reagent (M. musculus) | Ai32 or Ai32(RCL-ChR2(H134R)/EYFP) or B6;129S-Gt(ROSA)26Sortm32(CAG-COP4∗H134R/EYFP)Hze/J | The Jackson Laboratory | Stock No: 012569 | Males and females used |
| Genetic reagent (M. musculus) | C57BL/6J or C57BL/6NCrl | Charles River Laboratories | Strain Code: 027 | Males and females used |
| Other | Cy3 conjugated streptavidin | Jackson ImmunoResearch Laboratories | Cat#: 016-160-08 Lot. #125000 | 1:1000 |
| Chemical compound, drug | Ketamine, Ketamidor | Alvet Escartí S.L. | Ref. # 078100377 | 100 mg/ml |
| Chemical compound, drug | Medetomidine, Sedine | Alvet Escartí S.L. | Ref. # 005100740 | 10 ml |
| Chemical compound, drug | Sodium Pentobarbital, Dolethal | Alvet Escartí S.L. | Ref. # 015P5502 | 200 mg/ml, 100 ml |
| Software, agorithm | Spike2 | Cambridge Electronic Design Limited (CED) | n/a | Version 9 |
| Software, algorithm | Matlab | Mathworks | n/a | Version 2018 |
| Software, algorithm | Support Vector Machine (SVM) | Cortes and Vapnik, 1995 doi: https://doi.org/10.1007/BF00994018 | n/a | https://www.scipy.org/ |
| Software, algorithm | NA-MEMD | Rehman and Mandic, 2010 doi: https://doi.org/10.1098/rspa.2009.0502 | n/a | http://www.commsp.ee.ic.ac.uk/~mandic/research/emd.htm |
Additional files
-
Supplementary file 1
Additional data.
(A) Featurization of the SWO of DCS-MSNs. Note that the number preceding the feature labels are the same as the ones showed in Figure 3. All values display means ± standard deviation. (B) Data set description. The three first rows are total values; the rest display the mean ± standard deviation.
- https://cdn.elifesciences.org/articles/60580/elife-60580-supp1-v2.docx
-
Transparent reporting form
- https://cdn.elifesciences.org/articles/60580/elife-60580-transrepform-v2.pdf
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Medium spiny neurons activity reveals the discrete segregation of mouse dorsal striatum
eLife 10:e60580.
https://doi.org/10.7554/eLife.60580
