Development of biomarkers for predicting recurrence by determining the metastatic ability of cancer cells

Adjuvant chemotherapy has been carried out for patients with cancer who underwent curative resection, but it is basically not needed for patients without micro-metastatic lesions who undergo a perfectly curative surgical operation. The patients who need adjuvant chemotherapy are defined as those whose micro-metastases cannot be detected by imaging modalities in the other sites of the resective areas, despite curative resection for the primary sites. If biomarkers to efficiently evaluate the metastatic potential of each patient could be developed, we may be able to provide personalized adjuvant chemotherapy in the clinical setting. Actinin-4 (ACTN4, gene name ACTN4) is an actin-bundling protein that we identified in 1998 as a novel molecule involved in cancer invasion and metastasis. Protein overexpression of actinin-4 in cancer cells leads to the invasive phenotype, and patients with gene amplification of ACTN4 have a worse prognosis than patients with a normal copy number in some cancers, including pancreas, lung, and salivary gland cancers. In this review, the biological roles of actinin-4 for cancer invasion and metastasis are summarized, and the potential usefulness of actinin-4 as a biomarker for evaluation of metastatic ability is examined.

Structural and Biochemical Characterization of EFhd1/Swiprosin-2, an Actin-Binding Protein in Mitochondria

Ca²⁺ regulates several cellular functions, including signaling events, energy production, and cell survival. These cellular processes are mediated by Ca²⁺-binding proteins, such as EF-hand superfamily proteins. Among the EF-hand superfamily proteins, allograft inflammatory factor-1 (AIF-1) and swiprosin-1/EF-hand domain-containing protein 2 (EFhd2) are cytosolic actin-binding proteins. AIF-1 modulates the cytoskeleton and increases the migration of immune cells. EFhd2 is also a cytoskeletal protein implicated in immune cell activation and brain cell functions. EFhd1, a mitochondrial fraternal twin of EFhd2, mediates neuronal and pro-/pre-B cell differentiation and mitoflash activation. Although EFhd1 is important for maintaining mitochondrial morphology and energy synthesis, its mechanism of action remains unclear. Here, we report the crystal structure of the EFhd1 core domain comprising a C-terminus of a proline-rich region, two EF-hand domains, and a ligand mimic helix. Structural comparisons of EFhd1, EFhd2, and AIF-1 revealed similarities in their overall structures. In the structure of the EFhd1 core domain, two Zn²⁺ ions were observed at the interface of the crystal contact, suggesting the possibility of Zn²⁺-mediated multimerization. In addition, we found that EFhd1 has Ca²⁺-independent β-actin-binding and Ca²⁺-dependent β-actin-bundling activities. These findings suggest that EFhd1, an actin-binding and -bundling protein in the mitochondria, may contribute to the Ca²⁺-dependent regulation of mitochondrial morphology and energy synthesis.

Molecular dissection of the actin-binding ability of the fission yeast α-actinin, Ain1, in vitro and in vivo

A contractile ring (CR) is involved in cytokinesis in animal and yeast cells. Although several types of actin-bundling proteins associate with F-actin in the CR, their individual roles in the CR have not yet been elucidated in detail. Ain1 is the sole α-actinin homologue in the fission yeast Schizosaccharomyces pombe and specifically localizes to the CR with a high turnover rate. S. pombe cells lacking the ain1+ gene show defects in cytokinesis under stress conditions. We herein investigated the biochemical activity and cellular localization mechanisms of Ain1. Ain1 showed weaker affinity to F-actin in vitro than other actin-bundling proteins in S. pombe. We identified a mutation that presumably loosened the interaction between two calponin-homology domains constituting the single actin-binding domain (ABD) of Ain1, which strengthened the actin-binding activity of Ain1. This mutant protein induced a deformation in the ring shape of the CR. Neither a truncated protein consisting only of an N-terminal ABD nor a truncated protein lacking a C-terminal region containing an EF-hand motif localized to the CR, whereas the latter was involved in the bundling of F-actin in vitro. We herein propose detailed mechanisms for how each part of the molecule is involved in the proper cellular localization and function of Ain1.