HMGB1 modulates the balance between senescence and apoptosis in response to genotoxic stress

ABSTRACT High mobility group box‐1 (HMGB1) is involved in various diseases and is associated with the resistance of many types of human cancers to chemotherapy; however, its role in cancer metastasis remains unexplored. This study examined the HMGB1 status of both highly and poorly metastatic cancer...

Full description

Saved in:
Bibliographic Details
Published in:The FASEB journal 2019-10, Vol.33 (10), p.10942-10953
Main Authors: Lee, Je-Jung, Park, In Ho, Rhee, Woo Joong, Kim, Hee Sue, Shin, Jeon-Soo
Format: Article
Language:English
Subjects:
camptothecin
cancer
cell fate
doxorubicin
metastasis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:ABSTRACT High mobility group box‐1 (HMGB1) is involved in various diseases and is associated with the resistance of many types of human cancers to chemotherapy; however, its role in cancer metastasis remains unexplored. This study examined the HMGB1 status of both highly and poorly metastatic cancer cells in response to genotoxic stress. The weakly and highly metastatic mouse melanoma cell lines (B16 vs. B16‐F10), human melanoma cell lines (SK‐MEL‐28 vs. SK‐MEL‐24), colon cancer cell lines (DLD‐1 vs. LS174T), and wild‐type (WT) vs. HMGB1 knockout (KO) mouse embryonic fibroblasts (MEFs) were treated with doxorubicin (Dox) and camptothecin (CPT), and then cellular morphology, senescence‐associated β‐galactosidase staining, lactate dehydrogenase release, and caspase‐3 activation were used to assess cell fate. To investigate the role of HMGB1 in p21 expression, HMGB1 and p21 expressions were examined by Western blotting, and the HMGB1‐mediated p21 promoter luciferase assay was performed after small interfering RNA or overexpression of HMGB1 prior to Dox treatment. Although highly metastatic mouse melanoma B16‐F10 cells preferred senescence, with persistent HMGB1 expression, poorly metastatic B16 cells entered apoptosis, with decreasing HMGB1 levels via cleavage under Dox treatment. Similarly, more metastatic human melanoma SK‐MEL‐24 and human colon cancer LS174T cells underwent senescence, whereas fewer metastatic melanoma SK‐MEL‐28 and DLD‐1 cells exhibited apoptosis under Dox stimulation. In senescent B16‐F10, SK‐MEL‐24, and LS174T cells treated with Dox, p21 levels were increased by persistent HMGB1 expression. Furthermore, HMGB1 depletion caused a senescence‐apoptosis shift with p21 down‐regulation in B16‐F10 cells, and HMGB1 overexpression switched from apoptosis to senescence concomitantly with increased p21 expression in B16 cells after Dox treatment. The same effects were observed in both cell pairs of mouse melanoma and human colon cancer cells treated with CPT, another genotoxic stressor. Indeed, although WT MEF entered senescence accompanied by p21 increase, HMGB1 KO underwent apoptosis with p21 decrease by Dox treatment. In our cell model system, we demonstrated that highly metastatic cancer cells preferentially enter senescence, whereas apoptosis predominates in weakly metastatic cancer cells under genotoxic stress, which depends on the presence or absence of HMGB1, suggesting that the HMGB1‐p21 axis is required for genotoxic stress–induced senesc
ISSN:0892-6638
1530-6860
DOI:10.1096/fj.201900288R