Spatial transcriptomics reveal neuron–astrocyte synergy in long-term reminiscence


All animal experiments have been performed following protocols authorized by the Administrative Panel on Laboratory Animal Care at Stanford College. The TRAP2:Ai14 mouse line was a present from the Luo laboratory at Stanford. TRAP210 mice have been heterozygous for the Fos2A-icreER allele, and homozygous for Ai14 within the C57BL/6 background. Gt(ROSA)26Sortm1.1(CAG-cas9*,-EGFP)Fezh/J mice have been acquired from Jackson Laboratory. Mice have been group-housed (most 5 mice per cage) on a 12 h mild:darkish cycle (07:00 to 19:00, mild) with meals and water freely obtainable. Mice have been stored with ambient temperature at 21.1 ± 1.1 °C and humidity at 55 ± 5%. Male mice 49–56 days of age have been used for all of the experiments. Mice have been dealt with day by day for 3 days earlier than their first behavioural experiment. The animal protocol no. 20787 was authorized by Stanford College APLAC and IACUC. All surgical procedures have been carried out underneath avertin anaesthesia and carprofen analgesia, and each effort was made to reduce struggling, ache and misery.


The next primers: TCCTGGGCATTGCCTACAAC (ahead), CTTCACTCTGATTCTGGCAATTTCG (reverse) and ACCCTGCTGCGCATTG (reporter) have been used for genotyping of the Fos2A-icreER allele; CTGAGCTCACCCACGCT (ahead), GGCTGCCTTGCCTTCTCT (reverse), ACTGCTCACAGGGCCAG (reporter) for wild-type allele; CGGCATGGACGAGCTGTA (ahead), CAGGGCCGGCCTTGTA (reverse) and AATTGTGTTGCACTTAACG (reporter) have been used for genotyping of the Rosa-Ai14 allele; TTCCCTCGTGATCTGCAACTC (ahead), CTTTAAGCCTGCCCAGAAGACT (reverse) and CCGCCCATCTTCTAGAAAG (reporter) for Rosa wild-type allele.

Concern conditioning

The concern conditioning coaching was performed in accordance with beforehand described strategies9. Every mouse was positioned individually within the concern conditioning chamber (Coulbourn Devices), which was positioned on the centre of a sound-attenuating cubicle. Prior to every session, the chamber was cleaned with 10% ethanol to offer a background odour, whereas a air flow fan generated background noise at round 55 dB. The coaching started with a 2-min exploration interval, after which the mice acquired three tone-foot shock pairings separated by 1-min intervals. Every tone, an 85 dB 2-kHz sound, lasted for 30 s, and was adopted by a 2-s foot shock of 0.75 mA, with each ending concurrently. Following every pairing, the mice remained within the chamber for an extra 60 s earlier than being returned to their residence cages. For context recall, the mice have been reintroduced to the unique conditioning chamber for five min, 16 days after the coaching. Injections of 4-hydroxytamoxifen injections have been administered instantly previous to the recall experiments, inside 30 min. Within the HC and NR teams, 4-hydroxytamoxifen was injected at an identical time to the opposite two teams throughout the recall. The behaviour of the mice was recorded and analysed utilizing FreezeFrame software program (model 4; Coulbourn Devices), with immobile bouts lasting over 1 s being thought of as freezing.

Mind tissue dissociation and circulate cytometry

Basolateral amygdala was microdissected from stay sections reduce by a vibratome (300 μm thick). Tissue items have been enzymatically dissociated utilizing a papain-based digestion system (LK003150, Worthington). In short, tissue chunks have been incubated with papain (containing L-cysteine), DNase I, and kynurenic acid for 1 h at 37 °C and 5% CO2. After incubation, tissues have been triturated with 300 µm glass pipette ideas, then 200 µm glass pipette ideas, and 100 µm glass pipette ideas. Cell suspensions have been then centrifuged at 350g for 10 min at room temperature, resuspended in 1 ml EBSS with 10% v/v ovomucoid inhibitor, 4.5% v/v Dnase I, and 0.1% v/v kynurenic acid, and centrifuged once more. The supernatant was eliminated, and 1 ml synthetic cerebrospinal fluid (ACSF) was added to the cells. ACSF contained 2.5 mM KCl, 7 mM MgCl2, 0.5 mM CaCl2, 1.3 mM NaH2PO4, 110 mM choline chloride, 24 mM NaHCO3, 1.3 mM sodium ascorbate, 20 mM glucose, and three mM sodium pyruvate, 2 mM thiourea, and 13.2 mM trehalose. Cells have been then handed via a 70 μm cell strainer to take away particles. Hoechst stain (1:2,000; H3570, Life Applied sciences) was added and incubated in the dead of night at 4 °C for 10 min. Samples have been centrifuged (350g for 8 min at 4 °C) and resuspended in 0.5 ml of ACSF and stored on ice for circulate cytometry. Stay cells have been sorted utilizing the BD Vulcan into 384-well plates (Bio-Rad) instantly into lysis buffer, oligodT, and layered with mineral oil. After sorting, the plates have been instantly snap frozen till reverse transcription.


The Smartseq3 protocol was used for whole-cell lysis, first-strand synthesis and cDNA synthesis, as beforehand described with modifications. Following cDNA amplification (23 cycles), the focus of cDNA was decided by way of Pico Inexperienced quantitation assay (384-well format) and normalized to 0.4 ng µl−1 utilizing the TPP Labtech Mosquito HTS and Mantis (Formulatrix) robotic platforms. In-house Tn5 have been used for cDNA tagmentation. Libraries have been amplified utilizing Kapa HiFi. The libraries have been then sequenced on a Novaseq (illumina), utilizing 2 × 100-bp paired-end reads and a pair of × 12-bp index reads, with a mean of two million reads per cell.

Bioinformatics and information evaluation for scRNA-seq

Sequences from Nextseq or Novaseq have been demultiplexed utilizing bcl2fastq, and reads have been aligned to the mm10 genome augmented with ERCC (Exterior RNA Controls Consortium) sequences, utilizing STARsolo 2.7.9a. We utilized customary algorithms for cell filtration, function choice and dimensionality discount. In short, genes that appeared in fewer than 5 cells, samples with fewer than 2,000 genes and samples with lower than 50,000 reads have been excluded from the evaluation. Out of those cells, these with greater than 10% of reads as ERCC or greater than 20% mitochondrial have been additionally excluded from evaluation. Counts have been log-normalized after which scaled the place applicable. Canonical correlation evaluation (CCA) perform from the Seurat70 bundle was used to align uncooked information from a number of experiments. The highest 20 canonical parts have been used. After alignment, related options have been chosen by filtering expressed genes to a set of two,000 with the very best constructive and detrimental pairwise correlations. Genes have been then projected into principal part area utilizing the strong principal part evaluation. DEG evaluation was achieved by making use of the Mann–Whitney–Wilcoxon take a look at on numerous cell populations.

To search out memory-induced genes in every sort of neurons, sequence of strict standards have been utilized. First, we eliminated the background activation by excluding the DEGs resulted from FR versus NF amongst tdT detrimental neurons. This ensures their specificity that DEGs are activity-dependent, fairly than a basic enhance in all cells brought on by expertise. Second, DEGs have to be differentially expressed when FR TRAPed cells are in comparison with NR and HC controls, making certain that the DEGs have been distinctive to neuronal ensembles related to reminiscence recall, and never a results of baseline exercise (HC) or exercise remaining from the preliminary concern studying (NR). Lastly, every DEG needed to meet the standards of being expressed in a quater of cells and exhibiting no less than a 1.75-fold change. By adhering to those requirements, a complete of 107 DEGs have been acknowledged as ‘remote-memory-associated DEGs’ throughout 6 distinct neuron varieties, BLA.Int.Pvalb was not included within the evaluation as a result of inadequate numbers of cells. EnrichR was used for GO, KEGG and REACTOME pathway evaluation and classification of enriched genes (log2FC > 0.5 and P < 0.05) in every subpopulation.

scRNA-seq information from mPFC cells have been mapped to mm10 genome with full-length tdTomato assemble (together with Woodchuck Hepatitis Virus Posttranscriptional sequence included in Ai14 line71), which improved the sensitivity in calling tdT+ cells. Knowledge from BLA and mPFC cells have been built-in utilizing CCA. TRAPed neurons from the every built-in inhabitants have been analysed, besides B-P.Int.Pvalb and B-P.Int.Gpr88 neurons, as a result of restricted cell quantity. DEGs with P < 0.05 (Mann–Whitney–Wilcoxon take a look at) have been thought of as vital DEGs (highlighted in orange in Fig. 5d and Prolonged Knowledge Fig. 12f).

After unbiased clustering astrocytes, RNA velocyto40 and Monocle341 have been utilized to deduce astrocytic trajectory. DEGs between FR and NF situations have been estimated utilizing Mann–Whitney–Wilcoxon take a look at on every astrocyte inhabitants. R, RStudio, Python have been used for information evaluation.


The RNAscope multiplex fluorescent reagent package v2 (323100, ACD) and RNAscope 4-Plex probes have been used to conduct the RNAscope experiment in accordance with the producer’s pointers. The probes employed have been both obtained from obtainable shares or specifically created by ACD.

Gene choice for MERFISH measurements

We used a mix of single-cell RNA sequencing information and literature to pick out genes for MERFISH. Our choice standards concerned figuring out cell-type-marker genes for a selected cell inhabitants utilizing a one-vs-all method. To do that, we carried out a Mann–Whitney–Wilcoxon take a look at for every gene between the cells inside the cell inhabitants and all different cells not in that inhabitants, and corrected the ensuing P values for a number of speculation testing to acquire false discovery rate-adjusted P values. A gene was thought of a cell-type marker for a selected cell inhabitants if it met the next standards: (1) it was expressed in no less than 30% of cells inside the specified inhabitants; (2) the false discovery rate-adjusted P worth < 0.001; (3) gene expression within the specified inhabitants was no less than fourfold increased than the common expression in all cells not in that inhabitants; and (4) expressed in a fraction of cells inside the specified inhabitants that was no less than 2 instances increased than every other inhabitants of cells. We then sorted the marker genes for every inhabitants by fold change in expression relative to cells outdoors the inhabitants, and saved the highest 5 marker genes for every inhabitants to make use of for marker choice. Along with these markers, recognized genes associated to microglia, astrocytes and OPCs from the literature and included. Lastly, DEGs from distant memory-associated genes have been added to the panel with a complete variety of 158 genes.

Tissue processing for MERFISH and RNAscope

Mind tissue samples have been processed utilizing a fixed-frozen protocol for each MERFISH and RNAscope. In short, mice have been euthanized utilizing CO2 and perfused with chilly 4% paraformaldehyde. Mind tissue was dissected and adopted by incubation at 4 °C in 4% paraformaldehyde in a single day, 15% sucrose for 12 h, and 30% sucrose till sink. Mind tissue was frozen in OCT utilizing dry ice and saved at −80 °C till sectioning. Sectioning was carried out on a cryostat at −18 °C. Slices have been eliminated and discarded till BLA area was reached.

Slices with 10 μm in thickness have been captured onto Superfrost slides for RNAscope and MERSCOPE slides for MERFISH. The identical anatomical area was recognized for imaging publish hoc after pattern preparation, earlier than the beginning of RNAscope or MERFISH imaging.

Pattern preparation and MERFISH imaging

Slides with tissue sections have been processed in accordance with MERSCOPE protocol (Vizgen). In short, slides with tissue sections have been washed 3 times in PBS, after which saved in 70% ethanol at 4 °C for 18 h to permeabilize the tissue. Tissue slices from the identical mouse have been reduce on the identical time and distributed onto 4 coverslips. After permeabilization, the samples have been faraway from 70% ethanol and washed with Pattern Prep Wash Buffer (PN 20300001), then incubated with Formamide Wash Buffer (PN 20300002) at 37 °C for 30 min. Gene Panel Combine (RNA probes) was incubated for 48 h at 37 °C. After hybridization, the samples have been washed in Formamide Wash Buffer for 30 min at 47 °C for a complete of two instances to take away extra encoding probes and polyA-anchor probes. Tissue samples have been then cleared to take away lipids and proteins that contribute fluorescence background. In short, the samples have been embedded in a skinny 4% polyacrylamide gel and have been then handled with Clearing Premix (PN 20300003) for 36 h at 37 °C. After digestion, the coverslips have been washed in Pattern Prep Wash Buffer 2 instances and stain with DAPI/PolyT combine for 15 min. Slides have been washed with Formamide Wash Buffer adopted by Pattern Prep Wash Buffer earlier than imaging. Lastly, slides have been loaded to MERSCOPE Move Chamber and imaged at each 20× and 63× magnification.

MERFISH information processing

MERFISH imaging information have been processed with MERlin72 pipeline with cell segmentation utilizing CellPose73, a deep learning-based cell segmentation algorithm based mostly on DAPI staining. Decoding molecules have been then assigned to the segmented nuclei to supply a cell-by-gene matrix that checklist the variety of molecules decoded for every gene in every cell. The MERFISH expression matrix for every pattern was concatenated with the normalized, log-transformed with Scanpy74 and built-in utilizing Concord75 and Leiden76 clustering was utilized to separate the cells into distinct clusters. TRAPed neurons have been assigned based mostly on tdTomato expression. DEGs from a comparability of FR-TRAPed and NF-TRAPed situations have been estimated utilizing Mann–Whitney–Wilcoxon take a look at. Peri-engram cells have been computed as follows: for every engram cell (tdT+), its peri-engram cells have been counted inside a radius of 30 μm.

CalEx injection and behavioural experiments

AAVs carrying CalEx51 or tdTomato have been generated by Addgene based mostly on the vector pZac2.1-GfaABC1D-mCherry-hPMCA2w/b (AAV5, Addgene 111568) or pZac2.1 gfaABC1D-tdTomato (AAV5, Addgene 44332). Stereotaxic process of viral microinjection has been described beforehand. In short, mice with concern coaching (inside 12 h or after 24 h) have been anaesthetized and positioned onto a stereotaxic body (mannequin 1900, KOPF). Mice have been injected with Carprofen (5 mg kg−1) subcutaneously earlier than and after surgical procedure. AAVs carrying hPMCA2w/b (CalEx) or management (tdTomato) viruses have been loaded by way of a glass pipette related with a ten μl Hamilton syringe (Hamilton, 80308) on a syringe injection pump (WPI, SP101I) Bevelled glass pipettes (1B100–4; World Precision Devices) stuffed with viruses have been positioned into the BLA (1.3 mm posterior to the bregma, 3.4 mm lateral and to the midline, and 4.6 mm from the pial floor). Both 0.3 μl of AAV5 GfaABC1D mCherry-hPMCA2w/b (7 × 1012 viral genomes (vg) per ml) or 0.3 μl AAV5 GfaABC1D tdTomato (7 × 1012 vg ml−1) have been injected at 100 nl min−1. Glass pipettes have been withdrawn after 10 min and scalps have been cleaned and sutured with sterile surgical sutures. Mice have been allowed to recuperate in clear cages for 7 days. behavioural experiments (recall) have been carried out three weeks after surgical procedures. Schematic illustrations (Figs. 1a and 4a,f and Prolonged Knowledge Fig. 7h,o) created with

Open discipline

Mice have been positioned within the centre of 40 × 40 cm white field and allowed to freely probe for 15 min. Movies have been recorded and analysed by BIOBSERVE III software program. The 20 × 20 cm area within the centre was outlined because the central zone. The overall distance travelled and the exercise exploring the centre space have been analysed to guage the topic’s locomotor skill and anxiousness ranges.

Oligos and antibodies

For quantitative PCR evaluation, particular primers have been designed to amplify the Igfbp2 gene: Igfbp2 FW (GTCTACATCCCGCGCTG) and Igfbp2 RV (GTCTCTTTTCACAGGTACCCG). Moreover, for CRISPR–Cas9 gene enhancing, six gRNAs (Igfbp2 guides 1–6) have been chosen to focus on distinct areas of the Igfbp2 gene. These gRNAs have been designed based mostly on predicted specificity and effectivity: Igfbp2 information 1 (CTACGCTGCTATCCCAACCC), Igfbp2 information 2 (GCCAGACGCTCGGGCGTGCA), Igfbp2 information 3 (AGAAGGTCAATGAACAGCAC), Igfbp2 information 4 (GCCCTCCTGCCGTGCGCACA), Igfbp2 information 5 (CTCTCGCACCAGCTCGGCGC), and Igfbp2 information 6 (CGTAGCGTCTGGGCGCAGCG).

Antibodies focusing on mCherry (Thermo Fisher M11217) and cFOS (Synaptic Programs 226308) have been utilized for immunostaining following producers’ manuals.

Inclusion and ethics assertion

We, the authors of this manuscript, acknowledge the significance of inclusion and moral concerns in scientific analysis. Our work is guided by the ideas of equity, transparency, and respect for human dignity.

We affirm our dedication to selling variety and inclusivity in science, recognizing that various views, backgrounds, and experiences enrich analysis and improve scientific discovery. Now we have made efforts to make sure that our research is performed in a way that respects and consists of people of all races, ethnicities, genders, sexual orientations, skills, and different elements of human variety.

Now we have obtained all obligatory moral approvals and have adopted applicable pointers and rules for the analysis performed. Now we have taken measures to guard the privateness and confidentiality of analysis contributors, together with acquiring knowledgeable consent and making certain information safety.

We acknowledge the potential for hurt in scientific analysis and have taken steps to reduce any potential hurt to analysis contributors or others affected by our work. Now we have fastidiously thought of the potential implications of our analysis and have taken duty for making certain that our work is performed in a way that upholds moral and ethical requirements.

We acknowledge that scientific analysis has the potential to affect society in profound methods and we’re dedicated to partaking in accountable analysis practices that promote the well-being of people and society as an entire.

In abstract, we affirm our dedication to inclusive and moral analysis practices and acknowledge our duty to conduct analysis that’s performed with integrity, respect, and social duty.

Reporting abstract

Additional data on analysis design is obtainable within the Nature Portfolio Reporting Abstract linked to this text.

Leave a Reply

Your email address will not be published. Required fields are marked *