Skip to main content
NCBI home page
Neuropsychopharmacology logoLink to Neuropsychopharmacology
. 2024 Jan 5;49(5):796–805. doi: 10.1038/s41386-023-01788-w

Comparative rhythmic transcriptome profiling of human and mouse striatal subregions

Kaitlyn A Petersen 1, Wei Zong 2, Lauren M Depoy 1, Madeline R Scott 1, Vaishnavi G Shankar 1, Jennifer N Burns 1, Allison J Cerwensky 1, Sam-Moon Kim 1, Kyle D Ketchesin 1, George C Tseng 2, Colleen A McClung 1,✉
PMCID: PMC10948754  PMID: 38182777

Abstract

The human striatum can be subdivided into the caudate, putamen, and nucleus accumbens (NAc). In mice, this roughly corresponds to the dorsal medial striatum (DMS), dorsal lateral striatum (DLS), and ventral striatum (NAc). Each of these structures have some overlapping and distinct functions related to motor control, cognitive processing, motivation, and reward. Previously, we used a “time-of-death” approach to identify diurnal rhythms in RNA transcripts in these three human striatal subregions. Here, we identify molecular rhythms across similar striatal subregions collected from C57BL/6J mice across 6 times of day and compare results to the human striatum. Pathway analysis indicates a large degree of overlap between species in rhythmic transcripts involved in processes like cellular stress, energy metabolism, and translation. Notably, a striking finding in humans is that small nucleolar RNAs (snoRNAs) and long non-coding RNAs (lncRNAs) are among the most highly rhythmic transcripts in the NAc and this is not conserved in mice, suggesting the rhythmicity of RNA processing in this region could be uniquely human. Furthermore, the peak timing of overlapping rhythmic genes is altered between species, but not consistently in one direction. Taken together, these studies reveal conserved as well as distinct transcriptome rhythms across the human and mouse striatum and are an important step in understanding the normal function of diurnal rhythms in humans and model organisms in these regions and how disruption could lead to pathology.

Subject terms: Molecular biology, Circadian rhythms and sleep

Introduction

The striatum is a part of the brain composed of three subregions responsible for a variety of cognitive and motor functions. In general, the ventral striatum (primarily the nucleus accumbens (NAc)) is involved in reward and motivation, the caudate nucleus is associated with cognition and habitual behavior, and the putamen is largely associated with motor learning and control [1]. Importantly, dysfunction of the striatum is linked to neurological disorders and psychiatric illnesses, including substance use disorders, depression, Parkinson’s disease, obsessive-compulsive disorder, and schizophrenia [2–6].

Circadian clocks are endogenous timekeeping processes that cycle with a period of ~24 h and drive rhythms in most biological processes, including brain functioning [7, 8]. For example, in humans, many neurons are active during the day to facilitate learning, attentiveness, movement, and other activities. At night, other neuronal processes take over to promote sleep, memory consolidation, and provide restorative actions [9]. A synchronizing principal pacemaker resides in the suprachiasmatic nucleus (SCN) of the hypothalamus; however, the core molecular clock is present in nearly every cell in mammals [10]. All cellular clocks share the same clock machinery as those in the principal clock of the SCN, but locally regulate tissue or brain region-specific genes and functions [11–13]. The core molecular clock is a transcriptional/translational feedback loop, consisting of the CLOCK or NPAS2 proteins, which dimerize with ARNTL (also known as BMAL1), leading to the transcription of multiple genes, including the Period (PER) and Cryptochrome (CRY) genes [9]. The PER and CRY proteins then form a complex with other factors, enter the nucleus, and inhibit the activity of CLOCK/NPAS2 and ARNTL. Our lab and others have found that there are many transcripts whose expression levels follow a diurnal rhythm, likely regulated by the core molecular clock [14–17]. These studies have found that many of the same transcripts are rhythmic across various brain regions with very similar phase (i.e., rhythms peaking at roughly the same time of day) [14, 16, 17]. These genes not only include core circadian genes directly involved in molecular clock function, but thousands of other “clock-controlled genes”. In addition to transcripts that are rhythmic across cell types and brain regions, there are also a number of rhythmic transcripts that can be highly specific depending on the cell type or region [18].

Recently, our group identified molecular rhythms across the three striatal subregions collected from postmortem human brain tissue in subjects without psychiatric or neurological disorders [15]. In the current study, we aimed to perform a similar analysis of mouse striatal tissue to determine the level of evolutionary conservation of these rhythmic patterns. First, to directly compare to human subjects who were exposed to a day/night cycle, we utilized mice kept under a 12:12 light/dark cycle rather than those kept under constant conditions. While this did not allow us to analyze transcripts with free-running circadian rhythms, it allowed for a direct comparison to human subjects. Second, the human dorsal striatum is anatomically separated into caudate nucleus and putamen while the mouse dorsal striatum is a solid structure. However, the striatum of both primates and rodents contains the NAc [19]. Behavioral studies differentiate the rodent dorsal striatum into lateral (DLS) and medial (DMS) regions that have comparable functions to the putamen and caudate, respectively [20, 21]. Electrophysiological studies have also demonstrated unique properties associated with DMS and DLS [22]. For these reasons, we chose to collect DMS and DLS tissue as well as NAc tissue to compare to human caudate, putamen and NAc. Such a characterization in mouse brain could orient future brain-related studies, by providing knowledge regarding the similarities and/or differences between humans and mice.

Materials and methods

A brief summary of the “Materials and Methods” follows. See Supplementary Information for details.

Animals and housing conditions

Adult male and female C57BL/6J mice (Jax stock no: 000664) (~10 wks old) were maintained on a 12:12 light/dark schedule (lights on at 7 AM (zeitgeber time 0 (ZT0)) and off at 7 PM) and were provided with food and water ad libitum. All animal use was conducted in accordance with the National Institute of Health guidelines and approved by the Institutional Animal Care and Use Committees of the University of Pittsburgh.

Mouse RNA extraction and data pre-processing

Mice were sacrificed across 6 times of day, 4 h apart (ZT 2, 6, 10, 14, 18, and 22; n = 5–6 mice/sex/timepoint). Bilateral 1-mm punches were taken centered over the NAc, DMS, and DLS. Total RNA was extracted using the RNeasy Plus Micro Kit (Qiagen). RNA quantity and quality were assessed using fluorometry (Qubit RNA Broad Range Assay Kit and Fluorometer; Invitrogen) and chromatography (Bioanalyzer and RNA 6000 Nano Kit; Agilent), respectively. Libraries were prepared using the Smartseq stranded total RNA ultra-low input sample preparation kits (Illumina). Paired-end dual-indexed sequencing (75 bp) was performed using NextSeq 500 platform (Illumina) at the University of Pittsburgh Health Sciences Sequencing Core at UPMC Children’s Hospital of Pittsburgh. A total of 40 million reads per sample was targeted.

After quality control, HISAT2 (HISAT2v2.1.0) was used to align reads to the reference (Mus musculus Ensembl GRCm38) using default parameters. The resulting bam files were converted to expression count data using HTSeq (HTSeq v0.10.0) with default union mode. Filtering and normalization were performed separately for NAc, DMS, and DLS samples. After filtering, 11,867, 12,101, and 11,395 genes remained for NAc, DMS, and DLS samples, respectively. DESeq2 was then used to normalize for sequencing depth and RNA composition followed by a log2 transformation [23].

Mouse rhythmicity analysis

A parametric cosinor model assuming a sinusoidal relationship between the gene expression level and the zeitgeber time was used to detect rhythmicity, where the rhythmicity p-value can be calculated from the F test [24]. Scatter plots were generated for the core circadian clock genes in each region. Phase concordance plots were also generated to investigate differences in phase between NAc and DMS, NAc and DLS, and DMS and DLS. Rank rank hypergeometric overlap (RRHO) was used as a threshold-free approach to evaluate the overlap in rhythmic transcripts between striatal regions. To determine whether transcripts were significantly more rhythmic between regions, a gain-loss analysis was used [25].

Human postmortem brain samples

Human postmortem RNA-sequencing data from the striatum was obtained from Ketchesin et al. [15] (NCBI GEO accession no. GSE202537). Briefly, NAc, caudate, and putamen tissue samples were obtained through the University of Pittsburgh Brain Tissue Donation Program and the NIH NeuroBioBank. Total RNA from the NAc, caudate, and putamen of 59 subjects was extracted, processed, and analyzed as described previously [15].

Pathway and biological process enrichment analysis

Ingenuity Pathway Analysis (IPA) software (Qiagen) was used to identify enriched pathways in the top rhythmic transcripts. As an exploratory analysis, a significance threshold of p < 0.05 without multiple comparison correction was used for the rhythmic transcript input list to compare between regions. p < 0.05 (-log10(p-value) >1.3) was used as a significance threshold to identify significantly enriched pathways. The web-based portal Metascape was used to perform biological process enrichment (March 2023) [26]. Process enrichment was accomplished using GO biological processes as the ontology source. Transcript lists were based on the p-value (p < 0.01) of the respective rhythmicity analysis.

Results

Rhythmic transcript expression in the striatal subregions of mouse

To identify diurnal rhythmic gene expression patterns in mouse NAc, DLS and DMS we isolated RNA and performed sequencing and rhythm analysis. Timepoints of mouse tissue collection are depicted in Fig. 1A. We detected ~12–15,000 transcripts in mouse striatal regions. The absolute number of significantly rhythmic transcripts varied according to the stringency of statistical cut off. When we select a moderately stringent cut-off of p < 0.01 we identified many transcripts with a diurnal rhythm in mouse striatal regions (Table S1). Venn diagrams were also constructed for rhythmic transcripts in mouse at p < 0.05 to capture transcripts which might just miss a more stringent cut off (Fig. 1B). There were 846 rhythmic transcripts (p < 0.05) shared between all three regions in mouse (Fig. 1B). By performing ingenuity pathway analysis (IPA) on overlapping rhythmic transcripts between mouse striatal subregions, we were able to detect patterns of expression related mainly to the unfolded protein response, stress responses, and circadian rhythm signaling (Fig. 1C). Because we used an equal number of male and female animals, we were able to investigate differences in sex in the mouse (Fig. S1). While there were differences in specific rhythmic genes expressed between sexes, the pathways these genes represented were largely the same as what was found in the combined dataset.

Fig. 1. Overlap in rhythmic transcripts between striatal regions in mice.

Fig. 1

Open in a new tab

A. Human TOD values for subjects plotted around a 24 h circle plot. Suns and moons indicate time of mouse tissue collection (ZT2, 6, 10, 14, 18, 22) B Venn diagram showing overlap of rhythmic transcripts in the mouse striatum at a significance threshold of p < 0.05. Transcripts in common between only the NAc and DMS (452 transcripts) and DMS and DLS (2048 transcripts) show a higher degree of overlap than the NAc and DLS (130 transcripts). There were 846 transcripts in common between all three striatal regions. C Top five pathways enriched for rhythmic transcripts in common between all three regions, NAc and DMS, NAc and DLS, and DMS and DLS. Overlapping rhythmic transcripts were from the Venn diagram in B. D RRHO plots indicating a high degree of overlap in rhythmic transcripts between the mouse NAc and DMS, NAc and DLS, and DLS and DMS (left). In contrast, there was much less overlap between the human striatal regions (right).

Overlapping and unique rhythmic transcript expression in human and mouse striatum

We then compared rhythmic transcript expression in mouse to those we previously identified in the human striatum [15]. Timepoints of mouse tissue collection and human TOD are depicted in Fig. 1A. The top 10 transcripts with significant rhythms in expression are listed in Table S2. Many of the top rhythmic transcripts in the mouse and human striatum are core circadian genes including DBP, CIART, NR1D1, NR1D2, and ARNTL. In the DMS and mouse NAc, the remainder of the top rhythmic transcripts consist of protein-coding genes. Previously we found that the transcripts with the greatest diurnal rhythmicity in human NAc were noncoding RNAs including most prominently snoRNAs [15]. Many of these snoRNAs are expressed in mouse, but they did not have significant diurnal rhythmicity, suggesting their rhythmicity is possibly uniquely human. To further investigate rhythmic overlap between mouse and human striatal regions, we employed rank–rank hypergeometric overlap (RRHO) analysis, a threshold-free approach that examines the degree of overlap of identified genes across two datasets (Fig. 1D). We found a high degree of overlap between all three regions in the mouse striatum. This overlap between mouse striatal subregions was more pronounced than the overlap between human striatal subregions.

Pathway analysis for evolutionarily conserved and unique rhythmic transcripts in each region between species

To gain additional insight into potential functional roles of rhythms in the striatum, we identified pathways enriched for rhythmic transcripts in each striatal region between human and mouse. Figure 2A shows the number of overlapping and unique genes between species within comparable regions at p < 0.05. Since mice are both genetically identical and live in a very structured 12/12 light/dark (LD) environment, an experimental consequence is that more rhythmic transcripts will inevitably be identified in mouse than human due to the lower level of environmental and genetic variability. Therefore, any transcripts that have greater levels of rhythmicity in humans compared to mice are of interest as it suggests that this rhythm is highly robust and perhaps uniquely human. A gain/loss analysis was performed to detect transcripts that were uniquely rhythmic in each region between human subjects and mice (Fig. 2B). In the human NAc, there were uniquely rhythmic pathways associated with the immune response such as natural killer cell signaling and B cell receptor signaling, whereas the mouse NAc had pathways uniquely involved in the unfolded protein response and microRNA biogenesis. In the caudate, pathways were associated with nucleotide excision and repair as well as autophagy. Interestingly, in the DMS, there were many uniquely rhythmic pathways associated with neuronal signaling such as synaptic long-term depression and axonal guidance signaling. Lastly, in the putamen, protein ubiquitination and DNA methylation were uniquely rhythmic whereas in the DLS uniquely rhythmic pathways were involved with cell differentiation and G protein receptor signaling.

Fig. 2. Overlapping and distinct rhythmic transcripts and pathways in human and mouse striatum.

Fig. 2

Open in a new tab

A Venn diagrams showing overlap of rhythmic transcripts between the mouse and human striatum at a significance threshold of p < 0.05. The highest degree of overlap occurs between the mouse DLS and human putamen with 1411 transcripts. B Top five pathways enriched for rhythmic transcripts either unique to mouse or unique to human in the striatal regions. Unique rhythmic transcripts were from the Venn diagram in A. C Top five pathways enriched for rhythmic transcripts that overlap between mouse and human within each striatal subregion. D Metascape-derived heatmaps of biological processes enriched for rhythmic transcripts (p < 0.01) in each striatal subregion of mouse and human samples.

Figure 2C shows the pathways associated with overlapping transcripts within each striatal subregion between species. As expected, many of the overlapping transcripts are involved in the circadian rhythm signaling pathway which contains transcripts such as PER2 and ARNTL. Furthermore, all three region comparisons have the unfolded protein response as a highly rhythmic pathway. In the human and mouse NAc, the 290 overlapping transcripts coalesce into 3 main pathways: circadian rhythm signaling, RNA processing/splicing, and protein folding. In the caudate and DMS, EIF2-signaling, related to translation initiation and NRF2-Mediated Oxidative Stress Response are highly rhythmic across species. In the putamen and DLS, the unfolded protein response is the top shared rhythmic pathway. We also used Metascape to determine the biological processes implicated in the top rhythmic transcripts (p < 0.01) between all 6 comparable regions of the mouse and human striatum (Fig. 2D). While each group had unique aspects of biological process enrichment, the most amount of overlap between regions was associated with the unfolded protein response and regulation of RNA splicing. A previous study by Mure et al. sequenced several regions of the baboon brain including the putamen [18]. Even though different methodology was used to perform circadian rhythm analysis, we performed an exploratory investigation between the mouse, human and baboon putamen. Similarly, basic comparisons between the human, mouse, and baboon datasets revealed a few hundred overlapping rhythmic transcripts largely involved in mRNA processing, translation, and ER stress (Fig. S2).

Phase relationships within and between species and striatal subregions in core clock genes

We next wanted to investigate phase relationships in rhythmic transcripts within and between striatal regions. Most of the core circadian clock genes were identified as highly rhythmic across the striatum. For example, the known circadian genes ARNTL, CRY1, PER2, and REV-ERBα all have strong expression rhythms in each region in both human tissue and mice (Fig. 3). We compared the timing of peak expression of each circadian clock gene between species within each striatal region by plotting the phase of the human peak (blue line) onto the graph of mouse transcript expression over 24-h (Fig. 3). Humans are a diurnal species while mice are nocturnal, therefore, one might predict a 12-h phase difference in rhythmic gene expression, particularly in regions related to motor control. Interestingly, we found instead that the timing of expression of core circadian transcripts were shifted between human and mouse by ~6–8 h.

Fig. 3. Shift in timing of core clock gene expression between the human and mouse striatum.

Fig. 3

Open in a new tab

Circadian gene expression patterns of four canonical circadian clock genes in the mouse NAc (Top), DMS (Middle), and DLS (Bottom). In the scatterplots, each dot represents an animal, with the x axis indicating Time of Death (TOD) on a ZT scale (0 to 24) and the y axis indicating level of transcript expression. The red line is the fitted sinusoidal curve. The blue line represents the peak time of the clock gene in human subjects. All four clock genes show rhythms in each striatal region, with consistent peak times across regions. There is a distinct shift in peak timing between human and mouse of ~6–8 h. Peak time and p-values are located above each scatterplot.

Phase relationships between species in all rhythmic transcripts

We then investigated phase relationships in all significantly rhythmic transcripts in each striatal region. Within species, the peak time of expression for each transcript was consistent across regions, indicating phase concordance for these rhythmic genes (Fig. 4). For a given transcript, phases were plotted between the two regions, and transcripts were considered concordant if their phase differences fell within a window of ±4 h [15]. As in RRHO analysis, we saw a high degree of phase concordance between the 3 regions in mouse with upwards of 94% with all comparisons. In humans, however, there was a much lower degree of overlap and clustering of rhythmic transcripts was observed, suggesting unique phase differences exist between striatal subregions. Notably, this clustering was not present in mice, suggesting more consistent timing of individual transcript rhythms in mice across striatal regions compared to humans.

Fig. 4. Phase relationships within striatal subregions of mouse and human.

Fig. 4

Open in a new tab

Phase concordance plots showing the phase relationship in rhythmic transcripts (p < 0.01) between the NAc and caudate/DMS A, NAc and putamen/DLS B, and caudate/DMS and putamen/DLS C. For a given transcript, phases were plotted between the two regions, and transcripts were considered concordant if their phase differences fell within a window of ±4 h. Both axes indicate Time of Death (TOD) on a ZT scale (0 to 24). (A, left) There was a very high phase concordance between the NAc and the DMS in mice (94%) and (A, right) a comparably low phase concordance between the NAc and caudate in humans (46%). (B, left) There was a very high phase concordance between the NAc and DLS (96%) and (B, right) a high phase concordance between the NAc and putamen (89%). (C, left) There was a very high phase concordance between the DMS and DLS (98%) and (C, right) a high phase concordance between the caudate and putamen (78%).

Human and mouse day/night differences

We created radar plots to further analyze the timing of gene expression rhythms (Fig. 5A–C). Overlapping significantly rhythmic transcripts between species were plotted at their peak time for each region. There was an approximate 3-h difference between the average peak time of all overlapping transcripts between mouse and human, although not always in the same direction depending on region.

Fig. 5. Peak timing differences between human and mouse striatum.

Fig. 5

Open in a new tab

Radar plots showing peak expression times (ZT) of transcripts with diurnal rhythms in the A NAc, B DMS/Caudate, and C DLS/Putamen at p < 0.05. The radius indicates the percentage of transcripts peaking at that ZT. The red line indicates the time chosen to split the day into either dawn/dusk or day/night depending on the natural peak patterns of the data. D–F Top pathways enriched for rhythmic transcripts peaking in either the day/dawn or night/dusk between the striatal regions in human and mouse. G Upstream regulators enriched for rhythmic transcripts peaking in either the day/dawn or night/dusk between the striatal regions in human and mouse.

We further analyzed the timing of rhythmic transcripts between species by performing pathway analyses on transcripts separated by average peak timing. Transcripts peaking surrounding dawn (6 h before/after sunrise/lights on) versus dusk (6 h before/after sunset/lights off) were analyzed in the NAc and Caudate/DMS (Fig. 5D–F). In the DLS/putamen, however, the peak times of transcripts naturally split into day and night, where day was considered ZT0-ZT12 (lights on) and night ZT12-ZT24 (lights off). We also investigated the top predicted upstream regulators involved in these time-specific processes (Fig. 5G). Interestingly, in the NAc, many pathways that were significantly rhythmic in humans taken from dawn peaking transcripts were also rhythmic in mice surrounding dusk and vice versa. These results were perhaps expected due to humans being diurnal and mice being nocturnal. However, this pattern did not hold true in the caudate/DMS and putamen/DLS where similar pathways were peaking at the same times of day between species. Some interesting pathways that fell into this category were involved in mitochondrial dysfunction, oxidative phosphorylation, and calcium signaling.

Lastly, there were species- and time-specific predicted upstream regulators in each striatal region (Fig. 5G). During the night, predicted top upstream regulators represented by rhythmic transcripts included the glucose-dependent transcription factor MLXIPL (caudate/DMS), the Wnt signaling transcription factor TCF7L2 (Human NAc), and HNF4a, a transcription factor that may exert direct effects on various genes specifically associated with amine synthesis, such as genes involved in serotonin metabolism and related immunological functions [27]. During the day, HTT (NAc, DMS), PRDM5 (putamen) and miR-199a-5p (DLS) were amongst top predicted upstream regulators. The circadian protein CLOCK was also found to be a predicted regulator in the caudate. Interestingly, XBP1, a transcription factor that was recently found to regulate 12-h rhythms in mouse liver, was also a predicted upstream regulator in both species in the NAc (opposing times of day) and the putamen/DLS (corresponding times of day) [28].

Discussion

In this study, we utilized a TOD analysis to identify transcripts in the mouse striatum that have 24-h rhythms and compared those results to transcripts previously identified in humans [15]. This work is among the first to compare gene expression rhythms in the brains of humans and mice outside of the SCN with the goal of examining evolutionary conservation of molecular rhythms in the brain across species and providing valuable information to investigators using mice as a model system to understand human biology. Mice are one of the most widely used species in preclinical models and we demonstrate that there are key similarities and differences in circadian rhythms in the transcriptome of the striatum between species that may need to be considered when designing experiments.

In the mouse, our findings reveal many transcripts in each region of the striatum that display a significant 24-h rhythm in expression, ranging from 15 to 50% of all transcripts depending on subregion. In addition to core clock genes, many of the rhythmic transcripts identified in all three regions are involved in the unfolded protein response (UPR) and the oxidative stress response. Studies have demonstrated the importance of the circadian clock for counteracting ER stress and coordinating UPR activity during periods of high secretory demand to control energy expenditure [29]. Therefore, it is not surprising to see diurnal rhythms in basic cellular function as life evolved to adapt to daily, predictable changes in light, temperature, and resource availability [30].

When we compared results between mice and humans, we found many overlapping transcripts that exhibited diurnal rhythms. Most of these transcripts appeared to regulate basic biological functions that are evolutionarily conserved such as RNA splicing and protein folding. One striking finding was the lack of rhythmic snoRNAs in the mouse that we previously found were among the top rhythmic transcripts in the human NAc. Many of the rhythmic snoRNAs target rRNAs and other snoRNAs to facilitate posttranscriptional modifications and RNA splicing [31]. RNA splicing was a significant pathway in the mouse NAc, suggesting this process is likely still rhythmic in mouse, but does not involve these specialized snoRNAs. The study of snoRNA biogenesis and their functional impact on transcription is still a novel field and as research moves forward, we posit that species differences, and potential rhythmicity differences, should be considered.

We then performed a gain/loss analysis to investigate rhythmic transcripts unique to each species within region. Interestingly, in the DMS, unique rhythmic transcripts were involved in synaptic functioning and axonal guidance. Genes in these pathways largely included voltage-gated calcium channel subunits, AMPA receptor subunits, and G protein subunits. Studies have shown that AMPA receptors in the striatum are important for learning and memory and that there are activity dependent changes in receptor subunits [32]. These transcripts showed peak times in the late dark phase when activity is high. One potential explanation for why these genes could be uniquely rhythmic in mice and not humans is that humans may need more cognitive flexibility across all times of day. Another possibility is that our modern lifestyle has caused an increase in wakefulness and activity at night, disrupting molecular rhythms in genes which perhaps had weaker rhythmic regulation. Alternatively, we may have identified more rhythmicity in neuronal signaling in mice due to a difference in the proportion of cell types in mice versus humans. In the cortex, there is a much higher ratio of astrocytes and glia to neurons in humans compared even to other non-human primates [33]. This high ratio of glia to neurons has also been documented in the human striatum [34]. Since we analyzed homogenized tissue, differences in cell type ratios could be causing apparent changes in significant rhythmic processes between species and this will have to be determined in future single cell RNA-sequencing experiments.

In contrast to the mouse, the putamen had unique rhythmic transcripts involved in ubiquitination, a process involved in the removal of proteins and cell proliferation. The protein ubiquitination pathway was also highly rhythmic in the DLS, however we found many more unique rhythmic transcripts involved in this process expressed in humans. One explanation for this is the potential greater need for timed protein degradation based on higher metabolic demand in humans compared to mice.

Lastly, in the human NAc, there were many uniquely rhythmic transcripts related to immune functioning, such as natural killer cell signaling and B cell receptor signaling. Once again, while some immune-related pathways such as PI3K signaling in B lymphocytes were also rhythmic in mice, there were more immune-related rhythmic genes found in human subjects. This could suggest that humans need a more optimally timed immune system to promote neuronal health and homeostasis. Interestingly, current evidence suggests that dopaminergic signaling in the NAc modulates the immune system [35–37]. The rhythmic expression of these pathways in the NAc could indicate that rhythmicity in dopamine signaling is contributing to rhythmicity in immune functioning in this region.

Next, we looked at the overall timing of rhythmic transcripts. We found that transcripts were typically in phase between regions within species and that, in general, mice show a higher phase concordance than human subjects, suggesting more alignment between regions. This may be due to less functional and anatomical distinction between mouse striatal subregions. Studies have shown that there are functional differences between DMS and DLS in spatial memory and learning as well as habit and goal directed behaviors [38, 39]. However, in humans and nonhuman primates there is a large connection between the dorsolateral prefrontal cortex and the striatum that rodents do not have [40]. As such, the human striatum gains much more complexity with evolution compared to the mouse striatum, which could lead to more nuanced rhythmicity and function between the caudate and putamen compared to the DMS and DLS.

Lastly, we investigated day/night differences between species. Surprisingly, many of the overlapping pathways peaked at the same time of day, suggesting a conservation of rhythmicity independent of nocturnal/diurnal differences. Interestingly, many of these pathways peaked at night across species. These were genes involved in mitochondria function, transcription, protein folding, and ubiquitination. Pathways that peaked during the day were mostly involved in calcium signaling and cell signaling. Similar results were found in pathway analysis in another study that investigated striatal diurnal gene networks in mice [41]. This could suggest that, regardless of species and sleep timing, many basic cellular functioning and homeostatic mechanisms are occurring at night. Most pathways that had opposing effects (i.e., peak during the day in humans and during the night in mice or vice versa) were found in the NAc and were mostly involved in metabolism. Perhaps energy efficiency in the NAc is important for reward-related behaviors, specifically during times when activity is high.

Importantly, differences in rhythmicity between human subjects and mice may also be due to the rigor of analysis we can perform in each species. As mice were evenly sacrificed across 6 times of day and live in a controlled environment, we had improved statistical power and less variability, allowing clear identification of rhythmic gene expression. The nature of TOD analysis in humans inevitably leads to higher variability in samples. For this reason, we acknowledge that there are likely many transcripts that have rhythmicity in mice that may be rhythmic in human subjects but were unable to be captured in this study. In the future, adding more subjects to increase the sample size across all times of day would improve our statistical power to detect additional transcripts that overlap with mice. In future studies, it would also be interesting to further compare mouse, human, and non-human primate circadian data.

Another limitation is the imbalance between the sexes within the human subjects with most subjects being male. This has given us limited statistical power to reliably measure potential sex differences in rhythmicity, especially between species. Future studies will focus on utilizing larger sample sizes in male and female human subjects to specifically test whether any of the rhythmicity patterns are sex- and species-specific. Additionally, mice were sacrificed in early adulthood (~10 wks old) whereas the human postmortem samples had a mean age of 47 years. In the future it would be informative to investigate how circadian rhythms in the mouse brain change throughout aging to compare to human tissue taken from older adults.

Taken together, this work compares the relationship of rhythmic transcription between the human and mouse striatum. These results are critical for understanding the similarities and differences between the transcriptome of humans and mice to better inform future studies. Understanding rhythmic transcription in control subjects is the first step in understanding molecular rhythm disruptions associated with various pathologies. This study demonstrates the continued importance of considering the time of day of experimentation as well as species differences when performing preclinical work.

Supplementary information

Supplemental material (1.3MB, pdf)
Data Set 1 (5.1MB, xlsx)

Author contributions

Conceptualization: KAP, CAM; Methodology: GT, CAM; Data acquisition: KAP, LMD, MRS, VGS, JNB, AJC, SMK, KDK; Data analysis: KAP, WZ; Writing: KAP, LMD, MRS, KDK, CAM.

Funding

This work was funded by the National Institutes of Health DA039865, DA046346, MH111601, MH106460, NS127064 (PI: CAM), and MH128763 (PI: KDK), the Brain and Behavior Research Foundation (30823 (P&S Fund), PI: KDK), and the Wood Next Foundation (PI: CAM) (https://www.woodnext.org/).

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

The online version contains supplementary material available at 10.1038/s41386-023-01788-w.

References

  • 1.Haber SN. Corticostriatal circuitry. Dialogues Clin Neurosci. 2016;18:7–21. doi: 10.31887/DCNS.2016.18.1/shaber. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Casale AD, Kotzalidis GD, Rapinesi C, Serata D, Ambrosi E, Simonetti A, et al. Functional neuroimaging in obsessive-compulsive disorder. Neuropsychobiology. 2011;64:61–85. doi: 10.1159/000325223. [DOI] [PubMed] [Google Scholar]
  • 3.Cooper S, Robison AJ, Mazei-Robison MS. Reward circuitry in addiction. Neurotherapeutics. 2017;14:687–97. doi: 10.1007/s13311-017-0525-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Dickson DW. Neuropathology of Parkinson disease. Parkinsonism Relat Disord. 2018;46:S30–S33. doi: 10.1016/j.parkreldis.2017.07.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Middleton FA, Strick PL. Basal ganglia output and cognition: evidence from anatomical, behavioral, and clinical studies. Brain Cogn. 2000;42:183–200. doi: 10.1006/brcg.1999.1099. [DOI] [PubMed] [Google Scholar]
  • 6.Weinstein JJ, Chohan MO, Slifstein M, Kegeles LS, Moore H, Abi-Dargham A. Pathway-specific dopamine abnormalities in schizophrenia. Biol Psychiatry. 2017;81:31–42. doi: 10.1016/j.biopsych.2016.03.2104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Logan RW, McClung CA. Rhythms of life: circadian disruption and brain disorders across the lifespan. Nat Rev Neurosci. 2019;20:49–65. doi: 10.1038/s41583-018-0088-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Roenneberg T, Merrow M. The circadian clock and human health. Curr Biol. 2016;26:R432–R443. doi: 10.1016/j.cub.2016.04.011. [DOI] [PubMed] [Google Scholar]
  • 9.Takahashi JS. Molecular neurobiology and genetics of circadian rhythms in mammals. Annu Rev Neurosci. 1995;18:531–53. doi: 10.1146/annurev.ne.18.030195.002531. [DOI] [PubMed] [Google Scholar]
  • 10.Ko CH, Takahashi JS. Molecular components of the mammalian circadian clock. Hum Mol Genet. 2006;15:R271–R277. doi: 10.1093/hmg/ddl207. [DOI] [PubMed] [Google Scholar]
  • 11.Panda S, Antoch MP, Miller BH, Su AI, Schook AB, Straume M, et al. Coordinated transcription of key pathways in the mouse by the circadian clock. Cell. 2002;109:307–20. doi: 10.1016/S0092-8674(02)00722-5. [DOI] [PubMed] [Google Scholar]
  • 12.Yan J, Wang H, Liu Y, Shao C. Analysis of gene regulatory networks in the mammalian circadian rhythm. PLoS Comput Biol. 2008;4:e1000193. doi: 10.1371/journal.pcbi.1000193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Yoo SH, Yamazaki S, Lowrey PL, Shimomura K, Ko CH, Buhr ED, et al. PERIOD2::LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. Proc Natl Acad Sci USA. 2004;101:5339–46. doi: 10.1073/pnas.0308709101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Chen C-Y, Logan RW, Ma T, Lewis DA, Tseng GC, Sibille E, et al. Effects of aging on circadian patterns of gene expression in the human prefrontal cortex. Proc Natl Acad Sci USA. 2016;113:206–11. doi: 10.1073/pnas.1508249112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ketchesin KD, Zong W, Hildebrand MA, Seney ML, Cahill KM, Scott MR, et al. Diurnal rhythms across the human dorsal and ventral striatum. Proc Natl Acad Sci USA. 2021;118:e2016150118. doi: 10.1073/pnas.2016150118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Li JZ, Bunney BG, Meng F, Hagenauer MH, Walsh DM, Vawter MP, et al. Circadian patterns of gene expression in the human brain and disruption in major depressive disorder. Proc Natl Acad Sci USA. 2013;110:9950–5. doi: 10.1073/pnas.1305814110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Seney ML, Cahill K, Enwright JF, Logan RW, Huo Z, Zong W, et al. Diurnal rhythms in gene expression in the prefrontal cortex in schizophrenia. Nat Commun. 2019;10:3355. doi: 10.1038/s41467-019-11335-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Mure LS, Le HD, Benegiamo G, Chang MW, Rios L, Jillani N, et al. Diurnal transcriptome atlas of a primate across major neural and peripheral tissues. Science. 2018;359:eaao0318. doi: 10.1126/science.aao0318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Mitchell IJ, Cooper AJ, Griffiths MR. The selective vulnerability of striatopallidal neurons. Prog Neurobiol. 1999;59:691–719. doi: 10.1016/S0301-0082(99)00019-2. [DOI] [PubMed] [Google Scholar]
  • 20.Graybiel AM. Habits, rituals, and the evaluative brain. Annu Rev Neurosci. 2008;31:359–87. doi: 10.1146/annurev.neuro.29.051605.112851. [DOI] [PubMed] [Google Scholar]
  • 21.Hauber W, Schmidt WJ. Differential effects of lesions of the dorsomedial and dorsolateral caudate-putamen on reaction time performance in rats. Behav Brain Res. 1994;60:211–5. doi: 10.1016/0166-4328(94)90149-X. [DOI] [PubMed] [Google Scholar]
  • 22.Alegre-Cortés J, Sáez M, Montanari R, Reig R. Medium spiny neurons activity reveals the discrete segregation of mouse dorsal striatum. eLife. 2021;10:e60580. doi: 10.7554/eLife.60580. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550. doi: 10.1186/s13059-014-0550-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Cornelissen G. Cosinor-based rhythmometry. Theor Biol Med Model. 2014;11:16. doi: 10.1186/1742-4682-11-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ketchesin KD, Zong W, Hildebrand MA, Scott MR, Seney ML, Cahill KM, et al. Diurnal alterations in gene expression across striatal subregions in psychosis. Biol Psychiatry. 2022. 10.1016/j.biopsych.2022.08.013. [DOI] [PMC free article] [PubMed]
  • 26.Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10:1523. doi: 10.1038/s41467-019-09234-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Yamanishi K, Doe N, Sumida M, Watanabe Y, Yoshida M, Yamamoto H, et al. Hepatocyte nuclear factor 4 alpha is a key factor related to depression and physiological homeostasis in the mouse brain. PLoS ONE. 2015;10:e0119021. doi: 10.1371/journal.pone.0119021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Pan Y, Ballance H, Meng H, Gonzalez N, Kim SM, Abdurehman L, et al. 12-h clock regulation of genetic information flow by XBP1s. PLoS Biol. 2020;18:e3000580. doi: 10.1371/journal.pbio.3000580. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Díaz-Hung ML, Martínez G, Hetz C. Chapter Two - Emerging roles of the unfolded protein response (UPR) in the nervous system: a link with adaptive behavior to environmental stress? In: Kepp O, Galluzzi L, editors. International review of cell and molecular biology, Academic Press; 2020, vol. 350 p. 29–61. [DOI] [PubMed]
  • 30.Chaix A, Zarrinpar A, Panda S. The circadian coordination of cell biology. J Cell Biol. 2016;215:15–25. doi: 10.1083/jcb.201603076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hombach S, Kretz M. Non-coding RNAs: Classification, biology and functioning. In: Slaby O, Calin GA, editors. Non-coding RNAs in colorectal cancer, Cham: Springer International Publishing; 2016. p. 3–17. [DOI] [PubMed]
  • 32.Ferretti V, Perri V, Cristofoli A, Vetere G, Fragapane P, Oliverio A, et al. Phosphorylation of S845 GluA1 AMPA receptors modulates spatial memory and structural plasticity in the ventral striatum. Brain Struct Funct. 2015;220:2653–61. doi: 10.1007/s00429-014-0816-7. [DOI] [PubMed] [Google Scholar]
  • 33.Sherwood CC, Stimpson CD, Raghanti MA, Wildman DE, Uddin M, Grossman LI, et al. Evolution of increased glia–neuron ratios in the human frontal cortex. Proc Natl Acad Sci USA. 2006;103:13606–11. doi: 10.1073/pnas.0605843103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.von Bartheld CS, Bahney J, Herculano-Houzel S. The search for true numbers of neurons and glial cells in the human brain: a review of 150 years of cell counting. J Comp Neurol. 2016;524:3865–95. doi: 10.1002/cne.24040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ben-Shaanan TL, Azulay-Debby H, Dubovik T, Starosvetsky E, Korin B, Schiller M, et al. Activation of the reward system boosts innate and adaptive immunity. Nat Med. 2016;22:940–4. doi: 10.1038/nm.4133. [DOI] [PubMed] [Google Scholar]
  • 36.Ben-Shaanan TL, Schiller M, Azulay-Debby H, Korin B, Boshnak N, Koren T, et al. Modulation of anti-tumor immunity by the brain’s reward system. Nat Commun. 2018;9:2723. doi: 10.1038/s41467-018-05283-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Brod S, Rattazzi L, Piras G, D’Acquisto F. ‘As above, so below’ examining the interplay between emotion and the immune system. Immunology. 2014;143:311–8. doi: 10.1111/imm.12341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Gremel CM, Costa RM. Orbitofrontal and striatal circuits dynamically encode the shift between goal-directed and habitual actions. Nat Commun. 2013;4:2264. doi: 10.1038/ncomms3264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Pooters T, Gantois I, Vermaercke B, D’Hooge R. Inability to acquire spatial information and deploy spatial search strategies in mice with lesions in dorsomedial striatum. Behav Brain Res. 2016;298:134–41. doi: 10.1016/j.bbr.2015.11.001. [DOI] [PubMed] [Google Scholar]
  • 40.Carlén M. What constitutes the prefrontal cortex? Science. 2017;358:478–82. doi: 10.1126/science.aan8868. [DOI] [PubMed] [Google Scholar]
  • 41.Wang N, Langfelder P, Stricos M, Ramanathan L, Richman JB, Vaca R, et al. Mapping brain gene coexpression in daytime transcriptomes unveils diurnal molecular networks and deciphers perturbation gene signatures. Neuron. 2022;110:3318–3338.e9. doi: 10.1016/j.neuron.2022.09.028. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental material (1.3MB, pdf)
Data Set 1 (5.1MB, xlsx)

Articles from Neuropsychopharmacology are provided here courtesy of Nature Publishing Group

RESOURCES