Skip to main content
. 2019 Mar 26;8:e43627. doi: 10.7554/eLife.43627

Figure 5. Mitochondrial Complex I expression and oxidative stress in lung micrometastases.

(A) Gene Ontology (GO) analysis of biological processes (BP) of genes that were upregulated in MDA231 lung micrometastases compared to brain micrometastases or mammary tumors. The top functional groups and their corresponding pvalues are shown (n = 3); (B) Gene Set Enrichment Analysis (GSEA) analysis of nuclear Complex I genes was performed for the genes identified by Flura-seq in the indicated pairs of MDA231 lung and brain micrometastases and mammary tumors. p-Values were calculated by random permutations; (C) Oxidative stress in lung and brain tissue sections containing micrometastases were examined by IHC using anti-4-HNE antibody. Scale bars, 100 μm (top) and 20 μm (bottom); (D) Heatmap representation of the expression of genes encoding known antioxidant factors in MDA231 tumors from the indicated organs. The highlighted genes were also upregulated in clinical samples of lung metastasis from breast cancer patients (Figure 6D); (E) IHC analysis of GPX1, an antioxidant gene product identified by Flura-seq to be selectively upregulated in lung micrometastases. Scale bars, 100 μm (top) and 20 μm (bottom); (F) IHC analysis of NRF2 in lung and brain micrometastases. Scale bars, 100 μm (top) and 20 μm (bottom); (G) GSEA analysis of the NRF2 response gene signature applied to Flura-seq data from the indicated pairs of MDA231 lung and brain micrometastases and mammary tumors (n = 3). p-Values were calculated by random permutations.

Figure 5.

Figure 5—figure supplement 1. Differential gene expression in brain and lung micrometastatic cells.

Figure 5—figure supplement 1.

(A) Genes differentially expressed in lung and brain micrometastases relative to mammary fat pad tumors were identified by Flura-seq, and the overlap of the upregulated genes (left) and downregulated genes (right) in the organs are shown. (B, C) Upregulation of Complex I genes and antioxidant response in lung micrometastases is dependent on the tissue microenvironment. MDA231 cells from lung and brain micrometastases, and from mammary tumors were isolated and cultured in vitro under identical conditions, and the expression of Complex I genes (B) and the NRF2 gene signature (C) were subjected to GSEA for the indicated tissue pairs. p-Values were calculated by random permutations.
Figure 5—figure supplement 2. Oxidative stress and antioxidant programs are elevated in lung micrometastases relative to brain micrometastases in HCC1954 xenograft metastasis model.

Figure 5—figure supplement 2.

(A–C) Oxidative stress (A), antioxidant GPX1 (B) and NRF2 (C) in lung and brain tissue sections containing HCC1954 micrometastases were examined by IHC using anti-4-HNE, anti-GPX1 and anti-NRF2 antibodies respectively. Scale bars, 100 μm (top) and 20 μm (bottom) (n = 3). Shown IHCs are from serial sections of organs derived from the same mouse.