Fig 1.
Isolation and characterisation of diabetic PMVs.
(A) Analysis of Tyrode’s buffer, healthy control PMVs (NPMV) and diabetic-patient derived PMVs (DPMV) by NTA. Left panels display particle concentrations (1x108) and size, right panels display individual particle size verses intensity (a.u.), n = 3. (B) PMV particle concentration for Tyrode’s buffer, NPMV and DPMV, n = 3. (C) PMVs were labelled with Annexin V-FITC and/or CD41a-PE, gated around singlets and 10,000 events collected. Example dot plot for single and dual labelled PMVs with quadrant analysis, n = 3. Data are representative of all PMV samples, *** = p<0.001.
Fig 2.
PMVs isolated from poorly controlled diabetic patients harbour an oncogene-polarised miRNA profile compared with PMVs derived from healthy volunteers.
(A) Multi-dimensional scaling (MDS) plot of miRNA profiles in diabetic PMV samples (D; n = 4) and healthy controls (C; n = 8). (B) Heatmap of expression profiles of all differentially expressed miRNAs in diabetic PMVs compared with normal PMVs. Within the plot, red colour shows diabetic PMVs and green normal. (C) Cluster plot of enriched biological processes associated with mRNA targets of the down regulated (Down) or up regulated (Up) miRNAs. Gene ratio—genes in ontology/total gene set and false discovery rate (FDR)-. mRNA transcripts represented in the plot are experimentally validated miRNA-mRNA interactions from the miRTarBase database. (D) Distribution of known tumour suppressor and oncogenic miRNAs in the set of differentially expressed miRNAs in diabetic PMVs compared with normal PMVs.
Fig 3.
PMVs isolated from poorly controlled diabetic volunteers are internalised by TNBC cells, in vitro.
PMVs were stained with 1mM CellTracker™ Red CMTPX, free-dye removed and labelled PMVs cocultured with MCF-7 or MDA-MB-231 cells for 4h prior to analysis by confocal microscope.
Fig 4.
PMVs isolated from poorly controlled diabetic volunteers potentiate TNBC invasion, in vitro.
(A) MDA-MB-231 cells were cultured for 48h in complete media plus vehicle (PBS), serum-free (SF) media plus vehicle (PBS), SF media with NPMV, or SF media with DPMV and total viable cells assessed via CellTiter-Glo®, n = 3. (B and C) MDA-MB-231 cells were cultured in SF media or co-incubated with either healthy control PMVs (NPMV), or diabetic-patient derived PMVs (DPMV) in either uncoated or extra-cellular matrix (ECM)-coated transwell inserts to determine migration (B) and invasion (C), respectively. Migration and invasion rates are relative to cells cultured in SF media, n = 6. (D) MDA-MB-231 cells were cultured in SF media or co-incubated in SF media with either healthy control PMVs (NPMV), diabetic-patient derived PMVs (DPMV), or TGFβ (10ng/ml) for 24h prior to isolation of total RNA and analysis of target gene expression by RT-qPCR, n = 3. * = p>0.05; ** = p<0.01; *** = p<0.001.
Fig 5.
PMVs isolated from poorly controlled diabetic volunteers deliver oncogenic miRNA to TNBC cells, in vitro.
(A) MDA-MB-231 cells were cultured in the presence of actinomycin D (1 μg/ml) for 24h cultured prior to culturing for a further 24h in serum-free (SF) media, SF media with healthy control PMVs (NPMV), or SF media with diabetic-patient derived PMVs (DPMV). Total RNA was isolated from cells and RT-qPCR performed against target miRNAs, n = 3. (B) Total RNA from (A) was utilised to assess via RT-qPCR the transcript levels of validated cognate targets of each of the miRNA shown in panel A, n = 3. * = p<0.05.
Fig 6.
Model for how DPMV might impact on TNBC progression.
Circulating PMVs harbour an altered and pro-oncogenic miRNA profile in T2DM patients (indicated by HbA1c>50mmol/mol) that impact on TNBC invasion, in vitro, and might similarly drive tumourigenesis in diabetic patients with breast cancer.