High glucose

Cancer cells. The triple-negative breast cancer cell line 4T1 (ATCC) and its clonal variants, derived from a mouse mammary cancer, were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) with high glucose (GIBCO, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; GIBCO) and penicillin-streptomycin (GIBCO).
Whole-exome sequencing. Whole-exome sequencing (WES) was performed using 200 ng of genomic DNA isolated from frozen cells using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). Exome sequencing and bioinformatics analysis were performed at the DNA Chip Laboratory (Tokyo, Japan). Exome capture was performed using a SureSelect Mouse All Exon kit V1 (Agilent Technologies Japan, Tokyo, Japan) according to the manufacturer’s instructions. Exome sequencing was performed on an Illumina NovaSeq 6000 platform using 2×150-bp paired-end reads (Illumina, Tokyo, Japan). Reads were aligned against the UCSC mm9 reference mouse genome using WA-mem (v0.7.17) (14 (link)). After excluding chimeric reads, duplicate reads were eliminated using Picard (Broad Institute, Cambridge MA, USA). Basic alignment and sequence quality control were performed using the GATK4 best-practice pipeline (15 (link),16 (link)). Somatic single-nucleotide variants (SNVs) and insertion/deletion (indel) variants were analyzed with Matched Normal Variant Call by A1-R using the Strelka2 program (ver 2.9.10) (GitHub, Inc., San Francisco, CA, USA) (17 (link)). All variants were annotated with dbSNP128 and Snpeff ver 3.6 using the GRCm37.67 database. Frameshift, stop-gain, stop-loss, start-loss, splicing, non-synonymous mutations, among others were included as high- and moderate-impact variants, while other variants were excluded.
Preparation of RNA-seq library. Total RNA was purified using an miRNeasy Mini Kit (Qiagen) and treated with DNase. RNA integrity and quantity were measured on an Agilent 2100 Bioanalyzer (RNA Nano kit, Agilent Technologies). A portion (100 ng) of the purified total RNA was used to create an RNA-sequencing library with the TruSeq stranded mRNA library prep kit (Illumina).
RNA-sequencing. The resulting library was sequenced using the NovaSeq 6000 System (Illumina) for 100-base single reads and 100-base paired-end read sequencing. The STAR-2.7.5c alignment tool (Agilent Technologies) was used with default parameters to align the raw reads to the reference mouse genome (UCSC mm10). Mapped reads were then assigned to genes using HTSeq-count (GitHub, Inc.) with gene coordinates based on the RefSeq.gtf file. Data were analyzed using the Subio platform (Subio, Inc., Aichi, Japan). A gene was considered over-expressed if it showed at least a 2-fold change and the false discovery rate (FDR) was 0.05. Gene-set enrichment analysis (GSEA v4.1.0, Broad Institute) was performed on all genes ranked according to fold-change using the Gene Ontology gene set v7.4 (MSigDB, Broad Institute).
Neoantigens. Tumor mutation burden (TMB) was calculated for each category of 4T1-HA cell as the number of somatic single-nucleotide and indel mutations per mega base (Mb) of sequenced genome (18 (link)). Somatic single-nucleotide mutations included nonsynonymous mutations as well as synonymous mutations. Non-coding changes were excluded. All experiments used BALB/c mice. The major histocompatibility (MHC) haplotype antigens for BALB/c mice are H2-Kd, H2-Dd, and H2-Ld. vCF output files from the Strelka2 pipeline and RNA-sequence- expression values from Kallisto were obtained using the neo-peptide prediction tool MuPeXI version 1.2.0. For neoantigen prediction, SnpSift was used to filter only nonsynonymous mutations (missense and nonsense) from each VCF file (19 (link)). Non-synonymous mutations, gene expression counts, MHC type of each sample, and peptide length (8-11mer) were entered into MuPeXI. Affinity of WT and mutant peptides for BALB/c MHC class I alleles were calculated. For NetMHCpan-4.0 analysis, the predicted binding value, IC₅₀ ≤500, for MHC class I alleles was set as the threshold for neoantigen selection (20 (link)).

We obtained NRK-52E cells from the Department of Cardiology, Harbin Medical University, and cultured them in Dulbecco’s modified Eagle medium containing high-glucose (Gibco/Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum, (Gibco/ Life Technologies), then placed at 37 °C in a humidified atmosphere containing 5% CO₂. Afterward, cells reached 70% confluent were used for the next experiments. In order to choose the appropriate inducer, the NRK-52E cells were divided into two groups: (1) TGF-β1group(0,5,10ng/ml), (2) Ang II group (0,100,1000nmol/L). The TGF-β1 group was treated with recombinant human TGF-β1 (Pepro Tech, Rocky Hill, NJ, USA) for 24,48,72 h, (Figure S2). The cells were cultured and treated in a 35 mm dish, and the volume of DMEM 2 ml in each dish.
The TGF-β1-induced cells were divided into 8 groups, control group(only induced by TGF-β1), DVDMS group, Ultrasound group; NL-SDT group (DVDMS + Ultrasound), NL-SDT + CQ(10µmol/L) group¹⁸ , NL-SDT + 3-MA (10 mmol/L) group⁴³ , NL-SDT + NAC (5 mmol/L) group⁴⁴ . NL-SDT + LY294002 (50µmol/L) group^{45 (link)}, NL-SDT + rapamycin(500nmol/L) group^{46 (link)}. 3-MA is a PI3K selective inhibitor that blocks the formation of autophagosomes. LY294002 can suppress the PI3K/Akt signaling pathway, including inhibition of Akt phosphorylation. Rapamycin is a specific mTOR inhibitor that promotes autophagy. We have also validated the above drug concentrations at different concentrations (data not shown). After 24 h of treatment, the cells were harvested to determine the related parameter.