Archives

  • 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • br SRC and SRC depletions alter luminal

    2019-10-29


    3.4. SRC-2 and SRC-3 depletions alter luminal, basal and EMT gene programs
    Since prominent changes in the morphology of 3D cultures pointed to the development of the luminal cellular phenotype, we decided to 
    investigate more closely the PANTHER GO term Developmental process. Its subordinate category, Mesoderm development is tightly associated with EMT, as well as stemness in cancer biology. Therefore, we com-pared shSRC-2 and shSRC-3 DE genes to a set of 116 EMT-stemness associated genes. The 116 gene set was derived from literature by Lawson et al. and is defined as clinically relevant for EMT, stemness and mammary lineage specification in normal development of mammary ducts [16]. Therefore, we performed Gene Set Enrichment Analysis (GSEA) of this EMT category within the shSRC-2 and shSRC-3 micro-array datasets (Table 1) and found EMT to generally be depleted in both cell lines compared to the shKTR cells. More detailed analysis showed that individual SRC-2 and SRC-3 depletions mostly increased the ex-pression of genes associated with luminal phenotype and, vice versa, decrease expression of genes associated with basal and stem-like phe-notype. The EMT subcategory Luminal was enriched in SRC-2 and SRC-3 knockdown cells, while the subcategory Basal/Stem was depleted. Ad-ditionally, the heterogeneous phenotype group Mixed, composed of genes known to promote pluripotency in POM 1 and angio- and lym-phangiogenesis was enriched in shSRC-2 cells, while the TNBC gene set, related to the highly aggressive basal-like triple-negative breast cancer, was depleted.
    Leading edge analysis of statistically significant enriched and de-pleted groups identified four out of five genes, responsible for specifi-cation into a particular cell type (lineage-specific genes) significantly changed by SRC-2 and SRC-3 depletion (Supplementary tables 3 and 4). More specifically, luminal lineage-specific genes CD24 and MUC-1 were upregulated, while basal lineage-specific genes KRT5 and TP63 were
    Fig. 4. SRC-2 and SRC-3 knock-down induces morphological changes in three-dimensional MCF-7 cell cultures. Morphology of shKTR (a), shSRC-2(b) and shSRC-3
    (c) as observed by light microscopy. Yellow asterisk indicates the hollow acinar structures. Scalebar a-c: 100 μm. (d) Quantification of the number of acinar structures/spheroids (**** p ≤.0001). (e) Quantification of the size of the biggest acini/spheroid (**** p ≤.0001). Haematoxylin/Eosin staining of FFPE sections of shKTR (f), shSRC-2 (g), and shSRC-3 (h) spheroids reveal the morphology and the prominent lumen in shSRC-2 and shSRC-3 spheroids. Scalebar f-h: 100 μm. Immunofluorescence shows prominent staining of the luminal cytokeratin (K8, green) in shSRC-2 (j), and shSRC-3 (k), compared to shKTR (i). Counterstain DAPI (blue). Scalebar i-k: 50 μm. Chromogenic immunostaining for Ki67 (DAB; brown) reveals reduced proliferation in the differentiated acinar structures in shSRC-2 (m), and shSRC-3 (n), compared to shKTR (l). Counterstain hematoxylin (blue). Scalebar l-n: 100 μm.
    O. Bozickovic et al.
    Table 1
    Gene Set Enrichment Analysis of epithelial-to-mesenchymal transition (EMT) gene set and its subcategories Luminal, Basal/Stem, TNBC and Mixed.
    Rank Gene set Size ES NES Nominal P- FDR (%)
    value
    Stem
    overall
    overall
    ES, Enrichment Score; NES Normalized Enrichment Score; FDR, false-discovery rate; TNBC, triple-negative breast cancer. a Mixed: Pluripotency, (Lymph)angiogenesis.
    downregulated. The expression of CD24, KRT5 and TP63 was verified by RT-qPCR analysis showing the same direction of changes as in the microarray data set (Fig. 5). Additionally, we analyzed gene expression of the initiators and sustainers of the EMT (SNAI1, SNAI2, HEY2 and TGFBR2) and the structural cellular components affected by EMT (CDH1, CDH3) (Fig. 5). Moreover, SNAI1, CDH1 (E-cadherin) and ERBB1 were not identified in the microarray analysis applying the mentioned fold change cut-off, but were also chosen for the RT-qPCR analysis as they are considered to be biomarkers of EMT. Interestingly, we observed a strong downregulation of the EMT-inducing genes SNAI2 and TGFBR2 (encoding the TGFβ receptor II) in both shSRC-2 and shSRC-3 cells, alongside an opposing decrease in the major epithelial biomarker CDH1 and a mild increase in SNAI1, specific for shSRC-2 cells only.