Survivin, a 16 KDa protein as a monomer and 32 KDa as a dimer, is the smallest member in the Inhibitor of Apoptosis Protein (IAP) family, and contains a BIR domain, which is characteristic of this family of proteins. Unlike other members of the IAP family, Survivin does not have ubiquitin ligase activity (E3) and is the only member protein that forms a homodimer in solution (Chantalat et al., 2000; Muchmore et al., 2000; Verdecia et al., 2000). Interestingly, it is also a component of the chromosome passenger complex that associates with Aurora-B, and it follows a similar pattern of expression and localization during mitosis. Its expression has been found to peak at G2/M and its degradation occurs in a cell-cycle dependent manner (Zhao et al., 2000). In differentiated tissue, Survivin expression is virtually absent, in contrast to its high expression in actively proliferating lineages, including CD34+ hematopoietic stem and progenitor cells (when stimulated by the combination of thrombopoietin (TPO), stem cell factor (SCF), and Flt3 ligand (FL)) (Fukuda and Pelus, 2002), vascular endothelial cells (Tran et al., 1999), vascular smooth muscle cells (Blanc-Brude et al., 2002), thymus T- and B-cells (Kobayashi et al., 1999), and particularly in tumor cells (reviewed in Li (2005)). Survivin can be found in three splice variants that differ in size (Survivin 2B, Delta Ex3, and 3B) as a result from translation of an alternate exon 2B, skipping of exon 3 and/or a frameshift with premature stop codon (Conway et al., 2000). However, these splice variants still retain two features in common: the dimer interphase and the BIR domain at the N-terminus (Li, 2005). Published studies have suggested that survivin can form homodimers or heterodimers with its splice variants (Song et al., 2004; Li, 2005). These homodimers/heterodimers are hypothesized to have distinct functions in regulating apoptosis or cellular proliferation, depending on the type of dimer and its subcellular localization (Song et al., 2004). The Survivin 2B variant is cytosolic, while the Delta Ex3 variant is localized mainly in the nucleus. The Delta Ex3 variant contains a nuclear localization sequence (NLS, R/K-rich region 81RRKNLRKLRRK91) (Li, 2005). Survivin 2B expression is lost at later stages of malignancy, while normal Survivin and its Delta Ex3 variant maintain a high expression profile, suggesting a differential role in tumor development (Fortugno et al., 2002; Krieg et al., 2002). In vitro studies have demonstrated that Survivin's localization to the central spindle and midbody at telophase is dependent on phosphorylation at Thr117 by Aurora-B, and mutation of this site leads to disruption of its association with INCENP (Wheatley et al., 2004), suggesting that phosphorylation of Thr 117 is important for Survivin's role as a chromosome passenger protein. Homozygous deletion of Survivin in mice results in embryonic lethality at day 4.5, characterized by the presence of catastrophic mitosis (cell death during mitosis), giant multinucleate cells, in addition to a large population of polyploid cells (Uren et al., 2000). Forced overexpression of Survivin has been shown to inhibit IL-3-induced apoptosis in B-lymphocytes (Ambrosini et al., 1997) and in UV-induced apoptosis in primary keratinocytes (Grossman et al., 2001). In addition, published studies have suggested that overexpression of Survivin shortens G1 phase arrest and accelerates S phase, potentially through activation of Cdk2/Cyclin-E complex (Suzuki et al., 2000a,b). The important role of Survivin in regulating endomitosis in polyploidizing megakaryocytes and vascular smooth muscle cells has also been described (Zhang et al., 2004; Gurbuxani et al., 2005; Nagata et al., 2005). It has been shown that during these endomitotic cell cycles, Survivin does not colocalize with Aurora-B or INCENP, as typically observed at the centromere and at the central spindle/midbody during cytokinesis of normally dividing cells. Overexpression of Survivin was shown to reduce polyploidization in cultured primary vascular smooth cells (Nagata et al., 2005) as well as in megakaryocytes (Gurbuxani et al., 2005).
With regard to its antiapoptotic properties, Survivin has been shown to bind to Smac/Diablo, a caspase activator and/or to procaspase 9 via the hepatitis B X-interacting protein complex to mediate this effect (Marusawa et al., 2003; Song et al., 2003). A study by Song et al. (2004) demonstrated that a single amino acid change (Asp53_Ala53) converts Survivin from an antiapoptotic to proapoptotic regulator, suggesting that it has a dual role in controlling cell death at mitosis. Studies of Survivin function as both a chromosome passenger protein and as an anti/pro apoptotic factor has been a subject of much interest. Recent work has described a new type of cell death, termed “mitosis catastrophe,” often observed in cells with defective mitosis spindle assembly checkpoint, chromosome mis-segregation and abortive cytokinesis (reviewed in Castedo et al. (2004)). Although “mitosis catastrophe” is believed to be triggered by aberrant events during mitosis and not signals originating in G1 or S-phase, this type of programmed cell death still converges on the action of caspases, as suggested by several studies (Li et al., 1999; Grossman et al., 2001; Carter et al., 2003). Survivin may ensure the survival of cells with correct chromosome segregation by directly inhibiting caspases through its anti-apoptosis and/or chromosome checkpoint properties. On the other hand, Survivin, through its pro-apoptotic properties, may also ensure that cells undergo apoptosis if mitotic events are defective.