Glycyrrhizic acid inhibits myeloid differentiation of hematopoietic stem cells by binding S100 calcium binding protein A8 to improve cognition in aged mice

Animals

C57BL/6 mice (8W) were purchased from Shanghai SLAC Laboratory Animal (Shanghai, China). Aged C57BL/6 mice (16 months) were from laboratory-reared reserves. B-NDG severely immunodeficient mice (8 weeks) were purchased from Beijing Biocytogen (Beijing, China). All mice used in experiments were female and housed in the specific-pathogen-free-grade laboratory animal center of Tongji University School of Medicine. Mice were housed five per cage, maintained on a 12-h light/dark cycle, at an appropriate temperature, with free access to water and food. Animal care and procedures for this study were in accordance with institutional guidelines and the Animal Welfare Act, and all protocols involving experimental animals were approved by Experimental Animal Ethics Committee of Tongji University School of Medicine. Mice of equal weights in each age group were randomly grouped. GA was administered every other day by tail vein injection, and the same dose of phosphate-buffed saline (PBS) was injected into the tail vein of the control group.

10 × Single-cell sequencing

PBMCs were isolated from the blood of mice. Cell viability was examined under a microscope with 0.4% trypan blue staining. When the survival rate of cells reached 80% or more, library construction experiments were performed. Single-cell libraries were constructed using Chromium™ Controller and Chromium™ Single Cell 3ʹ Reagent Version 3 Kit (10 × Genomics, Pleasanton, CA). Single cells, reagents, and gel beads were enclosed in “gel beads in emulsion” (GEMs). Lysis and barcoded reverse transcription of single-cell polyadenylate mRNA was performed within each GEM. RT-GEMs were cleaned up to amplify cDNA, which was subsequently fragmented. Fragmented ends were repaired and A-tailing was added at the 3' end. Aptamers were ligated to fragments that were screened by two-sided solid-phase reversible immobilization (SPRI). After sample index PCR, another two-sided SPRI screen was performed. The final library was checked for fragment size distribution using an Agilent 2100 Bioanalyzer (Santa Clara, CA) and the library was quantified using real-time quantitative PCR with TaqMan probes. Finally, sequencing was performed using the DNBseq platform (BGI Group, Shenzhen, China).

Analysis of single-cell transcriptomics data

Single-cell FASTQ sequencing reads from each sample were processed and converted to digital gene expression matrices after mapping to the reference genome using the Cell Ranger Single Cell Software Suite (v3.1.0) [35] provided on the 10 × genomics website. Individual datasets were aggregated using the CellRanger aggr command without subsampling normalization. The aggregated dataset was then analyzed using the R package Seurat (v 3.1.0) [36]. First, the dataset was trimmed of cells expressing fewer than 200 genes. Next, the number of genes, UMI counts, and percentage of mitochondrial genes were examined to identify outliers. Because an unusually high number of genes can result from a ‘doublet’ event, in which two different cell types are captured together within the same barcoded bead, cells with > 90% of maximum genes were discarded. Cells containing > 7.5% mitochondrial genes were presumed to be of poor quality and also discarded. Next, gene expression values underwent library-size normalization, in which raw gene counts from each cell were normalized relative to the total number of read counts present in that cell. The resulting expression values were then multiplied and log-transformed. Subsequent analyses were conducted using only the most highly variable genes in the dataset. Principal component analysis was used for dimensionality reduction, followed by clustering in principal component analysis space using a graph-based clustering approach. Uniform Manifold Approximation and Projection (UMAP) was then used for two-dimensional visualization of the resulting clusters. For each clusters marker genes were identified using the FindConservedMarkers function as implemented in the Seurat package (logfc.threshold > 0.25 and minPct > 0.25). Next, clusters were annotated with known cell types according to the Cell Marker database [37]. Differently expressed genes across samples were identified using the FindConservedMarkers function in Seurat (logfc.threshold > 0.25, minPct > 0.25, and Padj ≤ 0.05). Finally, pseudotime trajectory analysis was conducted with the R package Monocle2 [38].

Isolation and purification of Lin−CD117.+ hematopoietic stem cells (HSCs)

C57BL/6 mice (aged 8 weeks) were euthanized by cervical and sterilized in 75% alcohol for 30 s, preferably allowing the fur of mice to be completely wetted. Next, the femur and tibia were removed and placed in a Petri dish containing pre-chilled PBS for cleaning. The attached muscle tissue was removed as cleanly as possible, and then the clean bone with muscle tissue removed was placed in pre-chilled PBS. Next, both ends of the tibia were cut and the bone marrow was flushed out with a 1-mL syringe until the bone turned white. Bone marrow cells were gently dispersed, and the resulting bone marrow monocyte PBS suspension was collected and filtered through a 40-µm cell sieve. Subsequently, the bone marrow was centrifuged at 440 × g for 5 min and the supernatant was discarded. Single bone marrow cells were resuspended in 5 mL of red blood cell lysate, and lysis was terminated with 5 mL of pre-cooled PBS after 1–2 min. Next, the cells were centrifuged at 440 × g for 5 min, the supernatant was discarded, and 5 mL of pre-cooled PBS was used to wash the cells. An aliquot was taken for counting. After negative selection of lineage cells using a mouse Lineage Cell Depletion Kit (Miltenyi Biotech), the cells were subjected to positive selection of CD117+ cells to obtain the target Lin−CD117+ HSCs.

OP9/OP9-DL1 co-culture system

OP9/OP9-DL1 cells were co-cultured as previously described [20,21,22]. OP9 was co-cultured with Lin−CD117+ HSCs for directed differentiation into B cells, while OP9-DL1 was co-cultured with Lin−CD117+ HSCs for directed differentiation into T cells. Before preparing for differentiation, OP9/OP9-DL1 cells grown in log phase were seeded at a density of 3 × 104 cells/well in 12-well plates, and 105 Lin−CD117+ HSCs were added to each well after apposition. Lin−CD117+ HSCs were co-cultured with OP9/OP9-DL1 cells in 20% fetal bovine serum (FBS), 5% penicillin/streptomycin (P/S), 5 ng/mL recombinant mouse IL-7, and 5 ng/mL rmFit3-L in α-Minimum Essential Medium. After 12 d of differentiation into B cells, or 12 and 28 d of differentiation into T cells, control (Ctrl) and GA groups with a glycyrrhizic acid concentration of 3.125 µM were evaluated (n = 3 replicates per group).

Cell preparation and flow cytometry analysis

Briefly, harvested mouse spleens were infiltrated with phosphate buffer, washed, lysed with lysis buffer, and filtered with a 70 µm cell strainer. Fresh blood samples were collected in heparin sodium tubes. Flow cytometry was used to directly quantify numbers of T cells (CD3 +) and B cells (CD45R/B220 +) in the spleen.

According to Roxanne Holmes' research [20], OP9-DL-1 cells are bone marrow-derived stromal cells that peripherally express the Notch ligand Delta-like 1 to support in vitro differentiation and development of T cells. T cell development proceeds through CD44 and CD25 yin-yang selection (DN phase changes) to yield CD44-CD25- (double-negative) cells, followed by CD4 + CD8 + double-positive cells, which ultimately mature into CD4 + /CD8 + single-positive cells. OP9 cells support in vitro differentiation and development of B cells [21]. At the end of co-culture, the upper cells were collected, rinsed 2–3 times with PBS, processed immunocytochemically, and quantified by flow cytometry (Fig. S9). Numbers of DN1 (CD44 + CD25-), DN2 (CD44 + CD25 +), DN3 (CD44-CD25-), DN4 (CD44-CD25-), T cells (CD4 + /CD8 +), B cells (CD45R/B220 +) and CD11b + cells were assessed. The optimal working concentration for all antibodies was 1 μg/mL. Data were analyzed with FlowJo software (Ashland, OR).

Label-free quantitative (LFQ) proteomics

Following extraction of total protein from PBMCs, a bicinchoninic acid assay was used to quantify protein and divide it into six 100-µg aliquots. For each group of three aliquots, methanol was added to the control group and the small molecule GA was added to the treatment group. After incubating samples at 25 °C for 10 min, PK (1:100) was added for incubation at 25 °C for 5 min. Subsequently, samples were transferred to a water bath with a temperature greater than 95 °C to completely inactivate PK. Next, samples were cooled and placed at room temperature for 5 min. After adding 2% Sodium deoxycholate (DOC) and 200 mM Ammonium bicarbonate (ABC), DL-Dithiothreitol (DTT) was added to samples and incubated at 37 °C for 30 min. Next, Iodoacetamide (IAA) was added and incubated at room temperature (protected from light) for 45 min. Subsequently, trypsin was added according to the ratio (1:50) and digested overnight in a 37 °C water bath. The following day, 50% Trifluoroacetic acid (TFA) was added to terminate the enzymatic termination and precipitation, and the enzymatically cleaved peptides were eluted. After elution, peptides were digested and resuspended in 100 µL of 0.1% Formic acid (FA) to yield a final concentration greater than 2 µg/µL. Finally, mass spectrometry was performed.

nanoLC-MS/MS analysis

Two microliters of total peptide were taken from each sample and separated with an EASY-nLC1200 nano-UPLC Liquid Phase System (Thermo Fisher Scientific). Data were collected using a mass spectrometer (Q-Exactive HFX; Thermo Fisher Scientific) equipped with a nano-electrospray ion source. Separation was performed with a reversed-phase column (100 μm ID × 15 cm, Reprosil-Pur 120 C18-AQ, 1.9 μm). The mobile phase adopted the acetonitrile–water-formic acid system, in which mobile phase A was 0.1% formic acid-98% aqueous solution (2% in acetonitrile) and phase B was 0.1% formic acid-80% acetonitrile (20% in water). After the column was equilibrated with a 100% A phase, the sample was directly loaded onto the column and passed through the column ladder degree separation with a flow rate of 300 nL/min and gradient length of 120 min. Mobile phase B ratios were applied as follows: 5% for 2 min, 5%–22% for 98 min, 22%–45% for 16 min, 45%–95% for 2 min, and 95% for 2 min. Mass spectrometry analysis used a data-dependent acquisition mode with a total analysis time of 120 min and positive ion detection mode.

Molecular docking

The binding of GA and S100A8 protein was simulated with CB-Dock (http://clab.labshare.cn/cb-dock/php/blinddock.php) and AutoDock software. Homology modeling was performed in SWISS-MODEL to predict the S100A8 protein structure with and without GA secondary structures. The CB-Dock computing program for AutoDock Vina was used to select the structure with highest score, as predicted by a total of 50 simulation positions according to the binding affinity sorting. The highest affinity group from the predicted results was put into PyMOL for visualization. The hydrogen bond formed between GA and S100A8 protein is the proposed binding site.

Surface plasmon resonance (SPR) analysis of GA with S100A8

The binding ability of GA to S100A8 protein was measured by SPR on a BIAcore3000 system (GE Healthcare, Chicago, IL). The S100A8 protein was immobilized on a CM5 chip by its amine group. A 10-µL aliquot of 0.1 mg/mL S100A8 was injected into the flow cell at a rate of 5 µL/min. Successful immobilization of S100A8 was verified by adding approximately 5000 RU to the sensor chip. No S100A8 was injected into the first flow cell as a control. GA was diluted in buffer containing 20 mM Tris–HCl, 150 mM NaCl, and 1 mM TCEP (pH 7.2). After fixation, varying dilutions of GA were injected at 30 µL/min for 3 min. After sample injection, the flow buffer was allowed to dissociate by passing over the sensor for 3 min. The sensor surface was regenerated by injecting 20 µL of 10 mM glycine–HCl solution (pH 2.25). Generated data were analyzed at 25 °C using Bioassessment Software 4.1.1 (GE Healthcare).

RNA sequencing (RNA-seq)

Total RNA from samples was extracted to create an RNA-seq library, which was analyzed by BGI USA (Cambridge, MA) using a BGISEQ-500 sequencer. Raw sequencing reads were cleaned by removing reads containing aptamer or poly-N sequences, and reads with low-quality base ratios. Afterwards, clean reads were obtained and stored in FASTQ format. Clean reads were mapped to the reference genome using the HISAT2 (v2.0.4) [39]/Bowtie2 (v2.2.5) [40] tool. Gene expression levels were calculated using RESM software [41] and normalized to determine gene expression. Heat maps were plotted using Graphpad Prism 8 (GraphPad, San Diego, CA). To gain insight into phenotypic variation, Kyoto Encyclopedia of Genes and Genomes (KEGG, https://www.kegg.jp/) enrichment analysis of annotated differentially expressed genes was performed by Phyper (https://en.wikipedia.org/wiki/Hypergeometric_distribution) based on hypergeometric tests. Terms and pathways were corrected for significance levels by Bonferroni's strict threshold (Q value ≤ 0.05).

qRT-PCR

Total RNAs were extracted from cells using TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. cDNA was prepared using reverse transcriptase (Takara, Kusatsu, Japan) according to the manufacturer's protocol. qRT-PCR reactions were performed using SYBR green-fluorescent dye and the primer sequences below.

sgRNA and CRISPR/Cas9 vector construction, virus generation, and transduction

sgRNA of S100A8 (sgRNA1F, 5'-caccgGACATCAATGAGGTTGCTCA-3'; sgRNA1R, 5'-aaacTGAGCAACCTCATTGATGTCc-3'; sgRNA2F, 5'-caccgGTCCTCAGTTTGTGCAGGTG-3'; and sgRNA2R, 5'-aaacCACCTGCACAAACTGAGGACc-3'; http://www.e-crisp.org/E-CRISPR) were cloned into the Lenti-CRISPR v2 vector following linearization by BsmB1 enzyme. Lentiviruses were generated by transiently transfecting 293 T cells with lentiviral plasmids VSVG and REV, and Pmdl packaging plasmids using Lentifit™ (HanBio Technology, Shanghai, China). Lentiviral particles were collected using ultracentrifugation. Lin−CD117+ HSCs were cultured in Ham's F-12 Nutrient Mixture with 1 × Penicillin–streptomycin–glutamine (PSG), 10 mM (4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid) (HEPES), 1 mg/mL Polyvinyl alcohol (PVA), 1 × Insulin–transferrin–selenium–ethanolamine (ITSX), 100 ng/mL thrombopoietin (TPO), and 10 ng/mL stem cell factor for 24 h before adding lentiviral particles. Particles were incubated at room temperature for 15–30 min and then added to cells in culture dishes for incubation at 37 °C overnight. Mouse embryonic fibroblasts (MEF) cultured in Dulbecco’s Modified Eagle Medium with 10% FBS and 5% P/S were infected with virus by the 1/2 small-volume infection method. After 4 h of infection, the medium was replenished with complete culture volume. After 24 h of infection, the culture medium was replaced with fresh complete medium for continued culture. Transfected cells were positively screened with 2 μg/mL of puromycin.

Immune Reconstruction (IR)

B-NDG mice are severely immunodeficient with severe deficiencies of T, B, and NK cells, which could be reconstituted with Lin−CD117+ HSCs modified with lentiviral particles in vitro. Lin−CD117+ HSCs were infected with LV201 and OEa8 lentiviral particles, and 105 particles/mouse were intravenously injected into the tail of B-NDG mice. One week after IR, mice in good health were administered GA (5 mg/kg every 2 d, lasts 30 days) by tail vein injection. Experiments were divided into five groups: non-IR, IR + PBS, IR + GA, IR-OEa8 + PBS, and IR-OEa8 (PM69) + GA.

Open field test (OFT)

A square box (50 × 50 × 30 cm) was used to evaluate the anxiety of mice. Anxious mice prefer to stay in the corner of the box and make stereotyped movements along the sides. At the beginning of the experiment, mice were placed in the same corner of the box, and the trajectory of mice in the box was tracked and recorded for 15 min. The anxiety of mice was judged by observing the time mice spent in the central area.

New object recognition (NOR)

Novel object recognition experiments use the rodent's natural instinct to approach and explore novel objects to test the behavioral sensitivity of the animal's recognition memory. The NOR experiment was divided into three phases: the adaptation period, familiarization period, and constant-temperature test period. Adaptation period: The new object was a square black box, and mice were put into one corner of the black box and allowed to explore the box freely for 10 min to adapt and reduce the stress of mice to the unfamiliar environment. Familiarization period: Two identical objects (A1 and A2) were placed diagonally across the bottom of the black box at a certain distance from the side, and mice were put into one corner of the black box and allowed to freely explore and familiarize with these two objects for 10 min. After familiarization, mice were put back into the cage for familiarization and tested after a certain period of time. Test period: A1 of the two identical objects was replaced in the black box with new object B. Mice were placed into the same corner of the black box they were put into during adaptation and familiarization, and allowed to explore the new object freely for 5 min. Video recordings were used to track the exploration times of mice for A2 and B. Preference for the new object was quantified by "discrimination index", expressed as D2 and calculated according to the formula: D2 = N/(FN + F), where "N" is the exploration time of mice for B, "FN" is the exploration time of mice for B, and "F" is the exploration time of mice for A2.

Statistical analysis

Data were statistically analyzed using Graphpad Prism 8.0 and expressed as mean ± standard error of the mean (SEM). Statistical comparisons between two groups were performed using an unpaired t-test. Probability values less than 0.05 were considered statistically significant.

留言 (0)

沒有登入
gif