Curcumin induces BRCA1 expression and phosphorylation. (A) MDA468, HCC1806, and MCF12A cells were treated with curcumin (10 μM) for 6 or 24 h. As a control, cells were treated for 24 h with the volume of ethanol, E, equivalent to that present in 10 μM curcumin. Total cell lysates were immunoblotted (50 μg) for total BRCA1 and BRCA1 phosphorylated at serine 988 (p-S988), a chk2-specific phosphorylation site. Curcumin induced S988 phosphorylation and increased total levels of BRCA1 within 6 h in MDA468 and HCC1806 triple negative cells, but not in MCF12A non-transformed cells. (B) Immunoblots for phosphorylated serine 1189 and serine 1280 (ATM phosphorylation sites) were performed on lysates (50 μg) from MDA468 cells treated with curcumin (10 μM) for 6 or 24 h or ethanol, E, for 24 h. Curcumin induced transient phosphorylation of both residues by 6 h, suggesting short-term activation of ATM by curcumin.

Curcumin induces DNA damage and cytoplasmic localization of BRCA1. (A) MDA468 cells were treated with ethanol, E, or curcumin, C (10 μM), for 24 h. Cells were fixed, stained with phospho-gamma H2Ax antibody followed by secondary antibody conjugated to green fluorescent protein (GFP). Phospho-gamma H2Ax foci (arrow) and DAPI nuclear staining were detected by immunofluorescent microscope. Representative DAPI, GFP (phospho-gamma H2Ax), and merged DAPI plus GFP photographs are shown using 100X objective lens. (B) HCC1806 and MDA468, and (C) HCC1937 and MCF12A cells were treated with ethanol or curcumin (10 μM) for 24 h. Cells were fixed, stained with anti-BRCA1 mouse antibody, followed by secondary GFP-conjugated anti-mouse antibody. BRCA1 localization was detected by immunofluorescent microscope. Representative photographs are shown using 4X objective lens. (D) Cells from (B) and (C) above that were in the ethanol, E, and curcumin, C, treatment groups were counted in five random non-overlapping fields for nuclear + cytoplasmic, N + C, versus cytoplasmic, C, staining only. Error bars represent standard deviation between the five fields per treatment group per line. Curcumin increased BRCA1 localization in the cytoplasm in HCC1806 and MDA468 cells (p ≤ 0.001), but did not show statistically significant changes in localization in HCC1937 and MCF12A cells.

Curcumin inhibits anchorage-independent growth and migration of triple negative breast cancer cells. Cells plated in matrigel were treated with ethanol, E, corresponding to highest dose of curcumin, 5 μM curcumin, or 15 μM curcumin. Media plus drug (or ethanol) was changed twice a week for two weeks. Experiments were done in duplicate and performed twice. (A) Representative photographs are shown using 4X objective lens with a 202-μm magnification bar shown at the right of photo. (B) For HCC1806 and MDA468 cells, matrigel was dissolved using dispase and cells were counted by trypan blue exclusion. Because of clumping, HCC1937 colonies were counted by microscopic examination. Number of viable colonies is shown as a percentage of the control per line. Error bars represent standard deviation between replicates. Curcumin inhibited anchorage-independent colony growth in a dose-dependent manner in HCC1806, MDA468, and HCC1937 cells, with statistically significant (^*p ≤ 0.05) inhibition at 15 μM curcumin in all three lines. (C) HCC1806 cells were plated at confluence. The next day cells were scratched down the middle and then treated with ethanol or 15 μM curcumin for 24 h. Representative photos taken with 4X objective lens are shown for 0 h (no curcumin or ethanol treatment) and 24 h. Arrow shows closed wound in control cells, indicating migration of HCC1806 cells within 24 h in presence of ethanol, whereas curcumin prevented migration of HCC1806 TNBC cells.

Estrogen receptor alpha and HER2 expression status of breast cancer cell lines. SKBR3, MCF7, HCC1937, HCC1806, and MDA468 cell lysates (50 μg) were immunoblotted for ER alpha, HER2, and actin. In comparison to ER+ MCF7 cells and HER2-over-expressing SKBR3 cells, HCC1937, HCC1806, and MDA468 are ER-negative and do not over-express HER2.

Modulation of BRCA1 in response to curcumin appears to be independent of NF-kB inhibition. (A) HCC1806 cells were untreated, treated with 10 μM curcumin or corresponding volume of ethanol for 6 h (6) or 24 h (24), and then lysed for nuclear protein. DNA binding activity of NF-kB was determined for nuclear extracts using NF-kB (p65) Transcription Factor Assay colorimetric kit (Cayman Chemical). Samples were run in duplicate. Fold change of the average of each sample relative to the average of untreated samples was determined; error bars reflect standard deviation between duplicates. Curcumin inhibited NF-kB transcription factor activity in HCC1806 cells. (B) HCC1806 cells were treated with 10 μM curcumin (Curc) or corresponding volume of ethanol (E) for 6 h or 24 h. Total protein lysates (50 μg) were immunoblotted for phosphorylated and total p65 NF-kB. Curcumin inhibited phosphorylation of p65 NF-kB consistent with inhibition of NF-kB function. (C) HCC1806 cells were treated with 10 μM of IKK inhibitor wedelolactone (Wedelo) or corresponding volume of solvent DMSO for 6 h or 24 h. Total protein lysates (50 μg) were immunoblotted for phosphorylated and total BRCA1. In contrast to curcumin, wedelolactone did not induce phosphorylation of S988 on BRCA1 or expression of total BRCA1, suggesting that modulation of BRCA1 may occur independently of IKK inhibition.

Curcumin promotes cell death of triple negative breast cancers. (A) Triple negative MDA468, HCC1806, and HCC1937, and ER+/HER2-over-expressing BT474, HER2-over-expressing SKBR3, ER+ MCF7, and non-transformed MCF12A cells were treated with 5, 10, or 20 μM curcumin for 72 h. Control cells were treated with ethanol corresponding to the highest dose of curcumin, since curcumin is dissolved in ethanol. Surviving cells were counted by trypan blue exclusion. For each treatment group, cell viability is shown as a percentage of the ethanol control group per line. Experiments were done in duplicate or triplicate at least twice. Error bars represent standard deviation between replicates. In comparison to MCF12A non-transformed mammary epithelial cells, curcumin inhibited viability of all cancer lines (^**p < 0.005 at 10 μM and 20 μM). (B) Results from the experiment in (A) are shown for the 10 μM curcumin dose for comparison of triple negative breast cancer (TNBC) and non-TNBC cells. TNBC cells HCC1806 and HCC1937 showed statistically significant (^**p < 0.007) higher sensitivity to curcumin versus non-TNBC lines. MDA468 cells showed a trend of being slightly more sensitive to curcumin than non-TNBC cells, although this difference was not statistically significant (p = 0.16, p = 0.14, p = 0.06, respectively for MDA468 versus MCF7, SKBR3, BT474). (C) Cells were treated with ethanol, E, corresponding to highest dose of curcumin, 5 μM curcumin, or 15 μM curcumin for 24 h. Total lysates (50 μg) were immunoblotted for PARP and actin. HCC1806 and MDA468 showed significant cleavage of PARP consistent with induction of apoptosis within 24 h of curcumin treatment. MCF12A cells did not show evidence of PARP cleavage in response to curcumin, consistent with trypan blue results in (A) which demonstrate that MCF12A non-transformed mammary epithelial cells are not sensitive to curcumin at these doses and time points. (D) HCC1806 and MDA468 cells were treated with ethanol, E, as a control or 10 μM curcumin for 6 or 24 h. Total lysates (50 μg) were immunoblotted for survivin and actin. Curcumin suppressed expression of the anti-apoptotic protein survivin within 24 h, consistent with induction of apoptosis.