**1. Introduction**

332 Advances in Cancer Therapy

Turke, A.B., Zejnullahu, K., Wu, Y.L., Song, Y., Dias-Santagata, D., Lifshits, E., Toschi, L.,

Uehara, Y., Minowa, O., Mori, C., Shiota, K., Kuno, J., Noda, T., Kitamura, N. (1995)

Van der Horst, E.H., Chinn, L., Wang, M., Velilla, T., Tran, H., Madrona, Y., Lam, A., Ji, M.,

Wang, W., Marimuthu, A., Tsai, J., Kumar, A., Krupka, H.I., Zhang, C., Powell, B., Suzuki,

Wang, W., Li, Q., Yamada, T., Matsumoto, K., Matsumoto, I., Oda, M., Watanabe, G.,

Watanabe, K., Chirgadze, D.Y., Lietha, D., de Jonge, H., Blundell, T.L. & Gherardi, E. (2002)

Weidner, K.M., Behrens, J., Vanderkerckhove, J. & Birchmeier, W. (1990) Scatter factor:

Yano, S., Wang, W., Li, Q., Matsumoto, K., Sakurama, H., Nakamura, T., Ogino, H.,

Yu, Y. & Merlino, G. (2002) Constitutive c-Met signaling through a nonautocrine mechanism

amplification in EGFR mutant NSCLC. *Cancer Cell*, 17: 77-88

factor/scatter factor. *Nature,* 373: 702-705

*Clinical Cancer Research*, 15; 6630-6638

*Journal of Molecular Biology*, 319: 283-288

*Cell Biology,* 111: 2097-2108

*Research*, 68: 6479-6487

2951-2956

14592-14597

*Neoplasia*, 11: 355-364

3563-3568

agonist to an antagonist. *Proceeding National Academy of Science USA,* 104:

Rogers, A., Mok, T., Sequist, L., Lindeman, N.I., Murphy, C., Akhavanfard, S., Yeap, B.Y., Xiao, Y., Capelletti, M., Iafrate, A.J., Lee, C., Christensen, J.G., Engelman, J.A. & Jänne, P.A. (2010) Preexistence and clonal selection of MET

Placental defect and embryonic lethality in mice lacking hepatocyte growth

Hoey, T.C. & Sato, A.K. (2009) Discovery of fully human anti-MET monoclonal antibodies with antitumor activity against colon cancer tumor models in vivo.

Y., Nguyen, H., Tabrizizad, M., Luu, C. & West, B.L. (2006) Structural characterization of autoinhibited c-Met kinase produced by coexpression in bacteria with phosphatase. *Proceeding National Academy of Science USA* 103:

Nakamura, T., Nishioka, Y., Sone, S. & Yano, S. (2009) Crosstalk to stromal fibroblasts induces resistance of lung cancer to EGFR tyrosine kinase inhibitors.

A new crystal form of the NK1 splice variant of HGF/SF demonstrates extensive hinge movement and suggests that the NK1 dimer originates by domain swapping.

molecular characteristics and effect on the invasiveness of epithelial cells. *Journal of* 

Kakiuchi, S., Hanibuchi, M., Nishioka, Y., Uehara, H., Mitsudomi, T., Yatabe, Y., Nakamura, T. & Sone, S. (2008) Hepatocyte growth factor induces gefitinib resistance of lung adenocarcinoma with EGF Receptor activating mutations. *Cancer* 

promotes metastasis in a transgenic transplantation model. *Cancer Research*, 62:

Survivin is a cancer associated protein that is present in all embryonic cells, but whose expression is normally limited to actively proliferating cells in adults. During mitosis it is part of a conserved complex of chromosomal passenger proteins (CPPs), which is essential for chromosome biorientation and cell division (Ruchaud et al., 2007). Defects in CPP function cause errors in mitosis and cytokinesis and can lead to genomic instability, or "aneuploidy", a status indicative of tumorigenesis. First discovered by Altieri and coworkers (Ambrosini et al., 1998), survivin is the fourth most upregulated mRNA in the cancer transcriptome (Velculescu et al., 1999), and in many tumours its expression is detected throughout the cell cycle. Importantly increased survivin abundance correlates with tumour resistance to conventional radiation and chemotherapies, a correlation recapitulated in cultured cells in the laboratory (Chakravarti et al., 2004; Colnaghi et al., 2006). While involvement in mitosis alone is a valid reason for considering a protein as a biomarker, or a target for cancer therapy, the oncotherapeutic potential of survivin is compounded by the fact that it is also an inhibitor of cell death (for review see Altieri, 2008). The principle aim of this article is to highlight the manifestations of survivin expression in the nucleus and to discuss how these might be exploited for oncotherapeutic gain.

#### **2. Subcellular localisation of survivin as a prognostic marker in cancer**

Numerous clinical studies have reported differential localisation of survivin in the cytoplasm, the nucleus, or both in tumours and have correlated this with disease severity and patient survival, with the aim to determine the prognostic value of its localisation retrospectively. However, despite the plethora of data collected, the prognostic significance of the subcellular distribution of survivin in cancer remains unclear. In 2005 a review of the subject returned a hung jury, revealing more opposing conclusions and contradictions than commonalities (Li et al., 2005); see also (Xie et al., 2006). As most of these clinical-based studies used histopathological samples, the lack of uniformity in sample collection and preservation between labs, together with differences in antibody protocols and the subjective nature of the analysis are likely to have contributed to the variation within these data (Li et al., 2005). Differences in data acquisition and interpretation aside, survivin localisation does vary between tumour types, and if one considers more detailed factors within specific types of cancer, such as its grade, stage of differentiation, and whether it is

Nuclear Survivin: Cellular Consequences and Therapeutic Implications 335

proposed by (Knauer et al., 2007), more efficient at binding CRM1. Given the variance in the above studies and the proximity of the proposed NES to its homodimerisation interface (Verdecia et al., 2000), Engelsma et al., (2007) investigated this region in more detail. They proposed that residues 84-109, shown in red on the crystal structures in Figure 2, constitute a more complete sequence for the central NES. Furthermore, by mutating F101 and L102 to inhibit homodimerisation, they found that survivin was more efficiently removed from the nucleus than the wild type form, suggesting that monomeric survivin is more efficiently exported than homodimeric survivin. These authors further reported a second, unconventional NES present in the C-terminus of survivin, depicted in orange in Figure 2, and spanning the alpha-helix from residue 119 to its end (Engelsma et al., 2007; Rodriguez et al., 2002). Presumably this second NES would be operative in both monomers and dimers,

Trafficking of survivin between the nucleus and cytoplasm is also susceptible to regulation by post-translational modifications, including acetylation and phosphorylation. Survivin acetylation, which occurs at K129, and is facilitated by the acetyl transferase CREB-binding protein (CBP), has been implicated in aiding survivin transit out of the nucleus (Wang et al., 2010), presumably by affecting the activity of the C-terminal NES (Engelsma et al., 2007). Glucogen synthase kinase 3β (GSK3β) facilitates translocation of survivin into the nucleus, but whether phosphorylation of survivin by GSK3β is involved is not yet known (Li et al., 2008). Work from my lab has recently provided two lines of evidence that suggest that casein kinase 2 (CK2) phosphorylation helps to maintain survivin within the cytoplasm, as treatment with the CK2 specific inhibitor, TBB (4,5,6,7 tetrabromobenzotriazole), or mutation of threonine 48 (T48), the unique site targeted by CK2 in survivin, causes its accumulation in the nucleus (Barrett et al., 2011). This latter observation is curious because T48 lies in the BIR domain of survivin, which is from distinct from either NES, and would not affect binding to CRM1. Perhaps CK2 affects

The molecular weight cut off for passive import across the nuclear membrane is approximately 45kD, thus at 16.5 kD survivin is sufficiently small to access the nucleus by passive diffusion even as a homodimer, or as a monomer tagged with a large fluorescent moiety such as green fluorescent protein, GFP (27kD). Whether an active component contributes to its import is unclear, but is supported by the observation that transit from the cytoplasm to the nucleus is abrogated when cells are chilled (Rodriguez et al., 2002), and would have the potential for its regulation, as suggested above for CK2, which could be exploited under challenging or stressful conditions. As survivin itself has no recognisable nuclear localisation signal (NLS) of its own, to facilitate nuclear entry survivin would presumably require a chaperone. While its mitotic partners, borealin, inner centromeric protein (INCENP) and aurora-B all have one or more NLS, co-expression analysis demonstrated that none of them can coerce survivin into the nucleus, instead, when coexpressed with borealin, borealin no longer localises to the nucleoli but is forced to relocate to the cytoplasm (Rodriguez et al., 2006). This is somewhat surprising given that borealin interacts with survivin at its homodimerisation interface and could potentially mask part of the central NES (Figure 2B), nevertheless, it points to the subtleties lying within the structure

and would act to further ensure survivin removal from the nucleus.

survivin import or its stability?

**4. Getting into the nucleus** 

benign or malignant, this approach could yield fruitful predictions. Indeed, Gallagher, Duffy and co-workers have recently validated this approach in tumour microarrays of breast cancers, in which they have recorded not just the cytoplasmic and nuclear placement of survivin in tumours, but the ratio between their distributions. Their automated analysis clearly removes the subjectivity of the interpreter, and also permits its application to high through-put screening (Brennan et al., 2008; Rexhepaj et al., 2010). In summary, while survivin's subcellular localisation may have prognostic value across a wide spectrum of tumours, correlating its distribution, or the ratio of its distribution with disease outcome within specific tumour types may be more effective.
