Clinical sample collection

The study was approved by the Clinical Research Ethics Committee of Shenzhen University Health Science Center (Approved No. 2,019,015). Written informed consent was obtained from all study patients. Tumor tissues of BC-DM and BC-no-DM patients were obtained from subjects who underwent surgical resection at Cancer Hospital of Harbin Medical University and Peking University Shenzhen Hospital. After surgical resection, tumor tissues were quickly frozen in liquid nitrogen and then stored at − 80 °C until further processing. All patients had a definite pathological diagnosis of breast cancer. No patients received radiotherapy or chemotherapy prior to surgery. The histological type, grade, and TNM-stage were classified according to the American Joint Committee on Cancer(AJCC)TNM staging system (7th edition) [24] Diabetes was defined as having fasting plasma glucose (FPG) level ≧ 7.0 mmol/l, according to the criteria of American Diabetes Association (ADA) [25]. Clinicopathological characteristics of patients were summarized in Table S1–2.

tRF & tiRNA sequencing

Total RNA was extracted from tumor tissues using the TRIzol® reagent (Invitrogen, MA, USA), according to the manufacturer’s protocol. To remove RNA modifications that affect small RNA-seq library construction, the RNA samples were first pretreated with the following reagents: 3’-aminoacyl (charged) deacylation to 3’-OH for 3’adaptor ligation, 3’-cP (2’,3’-cyclic phosphate) removal to 3’-OH for 3’adaptor ligation, 5’-OH (hydroxyl group) phosphorylation to 5’-P for 5’-adaptor ligation, m1A and m3C demethylation for efficient reverse transcription. Pretreated RNA sample was then subjected to tRF&tiRNA-seq library preparation. Library preparation procedures included: 3’ and 5-adapter ligation, cDNA synthesis, PCR amplification, and size selection of 134–160 bp PCR amplified fragments (corresponding to 14-40nt small RNAs). The completed libraries were sequenced on NextSeq system using NextSeq 500/550 V2 kit (Illumina) at Aksomics Inc. (Shanghai, China). Image analysis and base calling were performed using Solexa pipeline v1.8 (Off-Line Base Caller software, v1.8). Sequencing qualities were examined by FastQC and trimmed reads were aligned to the precursor and mature tRNA sequences in GtRNAdb database by bowtie software. The remaining reads were aligned to miRNA reference sequences with miRDeep2 database. The tRF & tiRNA expression levels were measured and normalized to the number of transcripts per million of total aligned tRNA reads (TPM). The differential expression of tRFs& tiRNAs between groups were compared by calculating the fold change for each tRF. Paired P-value < 0.05 was considered statistically significant.

Cell culture and transfection

The human BC cell lines (MDA-MB-231, MCF-7, BT-549, and T-47D) were obtained from the Cell Bank of Chinese Academy of Biological Sciences (Shanghai, China). All cell lines were authenticated through short-tandem repeat (STR) DNA profiling. No contamination of mycoplasma was found in these cell lines. Among these cell lines, MDA-MB-231, MCF-7, and BT-549 cells were cultured in Dulbecco’s Modified Eagle’s medium (DMEM) (Gibco, Shanghai, China) supplemented with 10% (v/v) fetal bovine serum (FBS) (Gibco, Shanghai, China, Table S3). T-47D cells were incubated in RPMI 1640 (Gibco, Shanghai, China) containing 10% FBS. All cells were incubated in a humidified atmosphere of 5% CO2 at 37oC.

The tRF-Cys-GCA-029 mimic, tRF-Cys-GCA-029-inhibitor, and their corresponding negative controls (NC) were synthesized by Ribobio Co. (Guangzhou, China). Cells were cultured on 6-well plates to confluence for 24 h. The tRF mimic, inhibitor, and NCs were transfected into cells, respectively, using Lipofectamine 3000 reagent (Table S4) according to the manufacturer’s instructions.

Quantitative real-time PCR (qRT-PCR)

Total RNA was isolated using the TRIzol reagents (Life Technologies, Shanghai China) according to the manufacturer’s protocol. 1 mg of total RNA was reverse transcribed into cDNA using the Takara cDNA kit (Takara Bio, Beijing, China). The transcript level of specific gene or tRF was amplified using the Takara qPCR kit and was normalized to U6. The primers were synthesized by Ribobio Co. (Guangzhou, China).

Cell proliferation assay

Cell Counting Kit-8 (CCK-8, Dojindo, Shanghai, China) kit was used to examine cell proliferation rate according to the manufacturer’s instructions. Briefly, the cells were planted on 96-well plate with a density of 5 × 103 cells/well. After incubation for 24, 48, 72, 96, 120 h,10 µl CCK-8 reagent and 100 µl fresh DMEM with 10% FBS were added, then cells were incubated at 37 ℃ for 1 h. The absorbance at 450 nm was measured by a microplate reader.

Cell migration assay

Cell migration was determined by Falcon Transwell chambers (Life Sciences, Shanghai, China) following the manufacturer’s instructions. Briefly, transfected cells were placed on the upper surface of the transwell insert. After 16 h, the invasive cells in the lower chamber were collected and washed with PBS, and then fixed with methanol for 60 mins. The fixed samples were washed and stained with 1.0% crystal violet staining solution (Beyotime, Shanghai, China). The number of invasive cells were counted in five randomly selected microscope visions and photographed.

Apoptosis assay

FITC Annexin V Apoptosis Detection Kit (BD Biosciences, CA, USA) was used to measure cell apoptosis, according to the manufactures’ protocols. In brief, BC cells transfected with tRF-Cys-GCA-029 mimic, inhibitor, or NC were stained with Annexin V-FITC and propidium iodide. The apoptotic rates of cells were analyzed using flow cytometer (Beckman Coulter, Inc., IN, USA) equipped with CytExpert software.

Diabetes-breast cancer animal model

BalB/C mice were purchased from the Biocytogen Animal Center (Zhuhai, China). Mice were firstly acclimated to the housing facilities for 7 days and then fed with high fat diet (HFD, 1135DM-5, Boaigang, Beijing, China). After 12 days of HFD feeding, streptozotocin (STZ, once 50 mg/kg body weight/day for 3 days) was injected intraperitoneally to induce diabetes. The corresponding diet lasted for the entire experimental period (50 days). Glycemic levels were measured on whole blood obtained from the snipped mouse tail using a glucometer (Yuwell 550, Shanghai, China). Diabetes was confirmed by the presence of hyperglycemia (> 11.1 mmol). 4T1 cells (2.5 × 105) were subcutaneously injected into the right flank of mice. When the tumors attained a size of 50 mm3, mice were randomly divided into 2 groups: control group (injected with5 nmol/10µl tRF-Cys-GCA-029 NC), tRF treatment group (injected with 5 nmol/10µl tRF-Cys-GCA-029 mimic). NC or tRF mimic were locally injected into the tumor mass once every 3 days, respectively (Fig. 4A). After implantation, mice were followed with repeated measurements of blood glucose levels and body weight. A caliper was used to measure the length and width of each tumor. The equation of volume = length × width2 was used to calculate tumor volumes. After two weeks treatment (a total of 6 injections), the animals were sacrificed, and the xenograft tumors were excised for tumor weight analyses. All animal study procedures were performed in accordance with the guidelines approved by the Institutional Animal Care and use Committee of Shenzhen University Medical School (Approved No. IACUC-202,300,060).

RNA-seq

RNA sequencing was carried out to identify the target genes regulated by tRF-Cys-GCA-029. MDA-MB-231 cells were transfected with tRF-Cys-GCA-029 mimic or scrambled control for 48 h. Three biological replicates for each sample were included in this experiment. Thereafter, RNA was isolated from the treated MDA-MB-231 cells using the TRIzol (Invitrogen, Shanghai, China) reagents. The total RNA quantity and purity were analyzed with Bioanalyzer 2100 and RNA 6000 Nano LabChip Kit (Agilent, CA, USA). High-quality RNA samples with RIN number > 7.0 were used to construct sequencing library, and then performed the 2 × 150 bp paired-end sequencing (PE150) on an Illumina Novaseq™ 6000 (at LC-Bio Technology CO., Ltd., Hangzhou, China) following the vendor’s recommended protocol. We aligned reads of all samples to the reference genome database using the HISAT2 package (version 2.1.0), which initially removed a portion of the reads based on quality information and then mapped the reads to the reference genome. The mapped reads of each sample were assembled using StringTie software (version 1.3.3) with default parameters. Genes differentially expressed between sample groups were calculated using the DESeq2 software (Table S5). The genes with the parameter of false discovery rate (FDR) below 0.05 and absolute fold change ≥ 2 were considered differentially expressed genes.

Quantification of L-lactate and pyruvate

For measurement of L-lactate and pyruvate levels, 1 × 106 cells were seeded into each well of the six-well plate. Cells were transfected with tRF-Cys-GCA-029 mimic, tRF-Cys-GCA-029-inhibitor, and NCs, respectively. L-lactate and pyruvate levels in the supernatants of cell culture media were quantified using the L-lactate assay kit and the pyruvate assay kit (Jiancheng Bioengineering Institute, Nanjing, China) according to the manufacturer’s instructions. The sample absorbance (530 nm for L-lactate and 505 nm for pyruvate) was detected using the Synergy HTX Multi-Mode Microplate Reader (BioTek, Guangzhou, China). All experiments were performed in triplicate.

Extracellular Acidification Rate (ECAR)

The extracellular acidification rate (ECAR) of cells were assessed using the Seahorse XFe24 Flux Analyzer (Seahorse Bioscience, Agilent). The glycolytic stress test kit (Seahorse) was used for ECAR detection. Briefly, the transfected MDA-MB-231 cells (1 × 105 cells/well) were seeded in the 24-well XF Seahorse incubation microplate as the protocol indicated. Cells were cultured at 37 °C overnight in XF base medium (pH 7.4) for adhesion. After baseline measurements, glucose (10 mM), glutamine (1 mM), 2-DG (50 mM) and oligomycin (1 µM) were added sequentially into each well at indicated time points. Data of ECAR were analyzed by Seahorse XF Glycolysis Stress Test Report Generator package.

Western blot analysis

Cellular proteins were extracted with cell lysis buffer (Promega, Madison, WI, USA). Thirty microgram of protein extraction was separated on an 10% sodium dodecyl sulfate-polyacrylamide Gel electrophoresis (SDS-PAGE) gel and then transferred to polyvinylidene fluoride (PVDF) membranes. After blockage by 5% Bovine serum albumin (BSA) at room temperature for 1 h, the membranes were incubated with primary antibodies (anti-PRKCG, 1:2000 dilution, Proteintech, anti-β-actin, 1:1000 dilution, Abcam) at 4 °C overnight. Then, blots were incubated at room temperature for 90 min with horseradish peroxidase (HRP) conjugated beta-actin secondary antibodies (diluted 1:5000, ZSGB-BIOZS). The bands were scanned and visualized by the GS700 imaging densitometer (Bio-Rad Laboratories) and analyzed by Image Studio software.

Ribosome profiling

Polysome profiling was performed as previously reported [26]. In brief, BC cells were pretreated with cycloheximide (CHX,100 µg/ml) at 37 °C for 15 min. Cells were collected and lysed on ice with lysis buffer (20mM Tris-HCl (pH 7.5),50mM KCl,10mM MgCl2, 1mM DTT,100mM CHX,200 µg/ml Heparin,1% Triton X-100) for 10 min. After centrifuging at 13,000 g for 10 min at 4 °C, the lysate was loaded onto a 10-50% sucrose gradient and further ultracentrifuged at 4 °C for 4 h at 36,000 rpm. Samples were subjected to fractionate using the Gradient Station (BioComp Instruments, Fredericton, Canada) and monitored at 254 nm. Thirteen fractions of equal volume were collected and their absorbances were measured. The 1–6 fractions were pooled together as light fraction (monosome) and the fractions 7–13 were pooled as the heavy weight fraction (polysome). The fractions were lysed in TRIzol (Invitrogen™) and isolated RNA was used for qRT-qPCR analysis.

Luciferase reporter assays

The PRKCG wild-type (WT) and mutant-type (MT) sequences were cloned into pmirGLO-basic vector. HEK-293T cells were cotransfected with luciferase reporters, either WT or MT, in combination with tRF-Cys-GCA-029 mimic or NC, using Lipofectamine 2000 reagents (Invitrogen, 11668-027). The luciferase activity was measured by Dual-Luciferase Reporter Assay System (Promega, E1910). And then the firefly luciferase activity values were normalized to the Renilla luciferase activity values that reflect expression efficiency.

Bioinformatics and statistical analyses

PRKCG mRNA sequence was downloaded from the NCBI RefSeq database (https://www.ncbi.nlm.nih.gov/gene/; NCBI transcription ID: NM_0013163.29.2). The putative tRF-Cys-GCA-029 binding sites in the PRKCG genome sequence were predicted on the basis of 20 kcal/mol of minimal free energy threshold using two independent bioinformatics tools, Targetscan and miRanda (Table S5).

Data are expressed as the means ± SD from at least three independent experiments. Statistical analyses were performed using Graph Pad Prism 8 and (Graph Pad, USA) and R program. Comparison between groups was conducted using Student’s t-test (for parametric data) or the Mann–Whitney test (for non-parametric data). All tests were two tails and P-value < 0.05 was considered statistically significant.