Fig 1.
HP1β message abundance is associated with survival of breast cancer patients.
A microarray database of 295 breast cancer patients (NKI-295 dataset) was analyzed and HP1β message signals were investigated. A. Kaplan-Meier analyses indicated that HP1β mRNA abundance is inversely correlated with both disease-free survival (DFS) and overall survival (OS) for breast cancer patients. B. Group of high HP1β expressors is associated with aggressive and poorly differentiated breast cancers. Low or high HP1β message abundance are denoted from microarray database from the public domain [23].
Fig 2.
Levels of HP1 protein in breast cancer tissues are heterogeneous.
A. Normal skin and breast tissues were stained with an anti-HP1β antibody (upper panel). Representative examples of breast cancer sample staining by an anti-HP1β antibody (lower panel). Depending on expression level and subcellular localization of HP1β, cancer samples were classified to three groups. B. Comparison of expression HP1 subtypes in breast cancer samples. Three sets of 190 breast cancer tumors were stained with individual subtypes anti-HP1 antibodies, anti-HP1α, HP1β or HP1γ (Abcam antibodies: ab77256, ab10478, ab10480). IHC staining patterns were compared and IHC staining scores were determined as shown in S1 Fig. Scale bars: 100 μm. HP1-High indicates the group of tumors with abundant expression of all three HP1 subtypes. HP1-Low is the group with no or low expression of all three HP1 subtypes. HP1-Mixed group of cancer samples are high level expression with one or two HP1 subtypes. Representative images are shown.
Table 1.
Correlation analyses of HP1β expression level with several molecular and pathological cancer markers.
Fig 3.
Positive correlation of HP1 and Ki-67 levels in breast cancer.
Levels of HP1α, HP1β and HP1γ IHC signal were positively correlated with Ki-67 levels in breast cancer patients. 0 indicates no or low expression and 1 denotes high expression of respective HP1 subtype, as shown in Fig. 2, and Ki-67 level, respectively.
Fig 4.
PARP inhibitor, ABT-888, induces apoptosis in HP1-depleted breast cancer cells.
ABT-888 (20 μM or 200 μM) was used to treat MCF7 and individually HP1-depleted MCF7 cells for 72 hours. Cells were stained with Annexin V and propidium iodide staining and cell population was analyzed by flow cytometry. A. Cell death percentage on Y-axis is for the double stained population using Annexin V (FITC-conjugated) and propidium iodide. B. Annexin V stained cell population was analyzed by flow cytometry. Y axis: cell numbers, X axis: FL1 values (Annexin V). C. HP1β-depleted MCF7 breast cancer cells are more prone to apoptosis by ABT-888 and carboplatin. Apoptosis of MCF7 and individually HP1-depleted MCF7 cells was determined after treatment of ABT-888 (20 μM), carboplatin (20 μM) or ABT-888/carboplatin combination (20 μM/20 μM) for 72 hours. A representative Annexin V staining of MCF7 and HP1β depleted MCF7 cells. D. Percentage of apoptotic MCF7 and HP1-depleted MCF7 cells were determined by double staining with Annexin V and propidium iodide followed by analysis using flow cytometry. *: p < 0.05 from analysis of three independent assays, Student's t-test.
Fig 5.
HP1β is a biomarker for breast cancer prognosis and PARP inhibitor therapy.
Respective HP1 expression level is frequently altered in breast cancer cells, suggesting the diverse role of each HP1 subtype in breast cancers. This model shows that the expression level of HP1 subtype in breast cancer cells may be either decreased or increased according to cancer stage, grade, cancer cell proliferation (Ki-67) and aggressiveness. PARP inhibitor therapy may be an effective therapy for patients with no/low HP1β expression. Combination therapy with epigenetic drugs (including H3K9 methylation inhibitors) or alternative therapy is necessary for patients with breast cancers of high HP1 abundance.