Role of Smac/DIABLO in cancer progression

Several studies have shown that overexpression of Smac/DIABLO sensitizes neoplastic cells to apoptotic death [18, 19]. These findings prompted the development of peptides derived from NH₂-terminal of smac/DIABLO and small molecules that mimic Smac/DIABLO functions as therapeutic agents in order to induce death or to increase the apoptotic effect of chemotherapeutic agents. Permeable NH₂-terminal peptides of Smac/DIABLO sensitize Hodgkin lymphoma cells to apoptosis mediated by B granzyme [18] and induce caspase-3 activation mediated by cytochrome- c [19]. Moreover, NH₂-terminal peptides of Smac/DIABLO fused to Drosophila antennapaedia penetratin sequences enhance apoptosis mediated by different antineoplastic agents in breast cancer [20] and glioblastoma cell lines [21]. Similarly, a small Smac-mimic compound is able to increase the apoptotic effects of death factors such as TRAIL and TNF-α [22]. It is interesting to note that this small molecule induces apoptosis by itself in MDA-MB-231 breast cancer cells, which have high expression levels of XIAP and c-IAP1. In contrast, it only sensitizes MDA-MB-452 and T47D cells, which have low IAPs expression, to TRAIL or etoposide. [23]. Recently, it has been described that compounds that mimic Smac/DIABLO induce the activation of the NF-kB pathway eliciting TNF-α-dependent apoptosis via caspase-8 activation [24]. This activation depends on c-IAP1 and -2 degradation [25]. This was replicated in vivo using a malignant glioma xenograft mice model, in which co-administration of Smac/DIABLO peptides and TRAIL sensitized glioma cells to apoptotic death and induced tumor regression [26]. In the same line, tumor regression was observed when nude mice with xenografted hepatocarcinoma tumors where locally treated with an adenovirus expressing Smac/DIABLO and 5-Fluorouracyl [27]. Taken together, these results show that the NH₂-terminal Smac/DIABLO derivatives and small molecules that mimic its function could be useful as adyuvant therapy in tumors with low levels of IAPs and as stand-alone therapy in tumors with high expression of IAPs. Further work is needed to address the potential use of these drugs in humans. Nevertheless, these studies show that particular expression levels of IAPs are key to the cellular response to these peptides or small molecules and underline the posible contribution of Smac/DIABLO levels to the intrinsic resistance to antineoplastic drugs.

Due to the importance of Smac/DIABLO in determining the sensitivity of cancer cells to apoptotic death induced by diverse stimuli, it is important to investigate if its expression levels could be negatively regulated during the initiation or progression of cancer. This information could be useful as a prognostic or therapeutic marker or to provide new targets for drug design. As mentioned before, it could be expected that cells with lower Smac/DIABLO and therefore, higher apoptosis resistance, should be selected during cancer progression. This selection would contribute to the higher intrinsic resistance in more advanced cancer stages. All the studies have been summarized in Table 1.

Two recent studies of Renal Cell Carcinoma (RCC) patients showed that Smac/DIABLO expression at both mRNA and protein level were not associated with stage or grade of tumor [28, 29]. However, another study demonstrated a significative inverse correlation between Smac/DIABLO protein expression level and both stage and histologic grade of RCC [30]. Regardless of this discrepancy, a clear correlation with progression was made in the three studies: patients with metastasic disease had lower Smac/DIABLO levels than those with localized disease [28‐30]. In addition, it was shown that patients that undergo radical nephrectomy for RCC with positive Smac/DIABLO expression had a longer disease-specific survival when compared with those with negative expression [30]. Moreover, RCC patients with low Smac/DIABLO expression had a four times higher risk to die than those with high expression [29].

Similarly, Sekimura et al have shown that Smac/DIABLO mRNA levels decreased significantly during lung cancer progression [31]. Since it has been shown that IAPs levels are upregulated in non small cell lung cancer [32], this reduction should increase further the apoptotic threshold of these cells. As expected from these results, decreasing XIAP expression in vitro sensitized lung cancer cells to apoptosis induced by diverse stimuli [33].

It has also been shown that testicular germ tumors with more advanced malignant phenotype presented lower Smac/DIABLO expression levels [34]. Moreover, the XIAP-Smac/DIABLO ratio increased significatively in clinical stage III tumors when compared with stages I or II. In this study an inverse correlation between Smac/DIABLO expression levels and cancer progression was also demonstrated [34].

Finally, it has also been shown that Smac/DIABLO mRNA and protein expression is reduced in hepatocellular carcinoma (HCC) when compared with normal hepatic tissue [35]. This reduction directly correlated with progression. Moreover, in this study, the authors showed that the IAP survivin, increased in parallel with cancer progression [35]. All these studies showed that Smac/DIABLO expression inversely correlated with cancer progression, aggressive behavior, as shown by metastasic disease, and bad prognosis. These correlations could be due to an increased apoptotic threshold, which should allow cancer cells to withstand not only invasion and metastasis, but to provide a relative increased intrinsic resistance to chemo- or radiotherapy. Further in vivo experiments specific for each tumor should help to solve this question.

Recently, our group analyzed the participation of Smac/DIABLO in cervical cancer patients. Surprisingly, we found that Smac/DIABLO expression did not correlate inversely with progression. As a matter of fact, normal cervix barely expressed the Smac/DIABLO RNA messenger. During cancer progression, a subset of cervical cancer samples expressed de novo this protein [36]. We could not find any correlation between Smac/DIABLO expression and stage or grade for both squamous cell carcinoma and adenocarcinoma [37]. However, higher immunostaining was directly associated with local recurrence in squamous cell carcinoma. We also established a positive correlation between Smac/DIABLO immunoreactivity and high microvascular density [37], which could be accounting for the recurrence results. More investigation is needed to clarify the relation between Smac/DIABLO and angiogenesis in cervical cancer.

A similar finding in gastric cancer patients supports these results [38]. Gastric adenocarcinomas showed a significant higher expression of Smac/DIABLO than non-neoplastic gastric mucosa [38]. However, no difference in the ratio XIAP-Smac/DIABLO between non-neoplasic gastric mucosa and gastric adenocarcinoma was found [38]. This discrepancy could be due to decrease in the expression of XAF1 (XIAP-associated factor-1) found in these samples [38]. XAF1 is a direct inhibitor of XIAP, so this decrease could be also be increasing the apoptotic threshold of these tumors [39]. Supporting these results, a recent report showed that a decrease of XIAP in gastric cancer cells increase the apoptotic response induced by cisplatin and mitomycin-c [40].

Analyzing a single protein such as Smac/DIABLO or a particular IAP in cancer samples, will only give a partial vision of the mitochondrial apoptotic threshold, since it is known that these proteins are regulated by themselves and, in a tissue-specific pattern, specific IAPs functions overlap. In the case of IAPs (Figure 1), although Smac/DIABLO is not able to degradate XIAP, it induces the ubiquitination and degradation of other members of the family, such as c-IAP1 and c-IAP2 [41]. In addition, a recently cloned Smac/DIABLO isoform, Smac3, induces the auto-ubiquitination and degradation of XIAP [42]. As mentioned before, another negative regulator of XIAP is XAF1. This protein antagonizes the inhibitory activity of XIAP toward caspase-3 [39]. Similarly to IAPs, Smac/DIABLO is also regulated by IAPs family members (Figure 2). Livin is able to ubiquitinate and degrade Smac/DIABLO [43]. Also, in a reciprocal manner, XIAP, c-IAP1, c-IAP2 and Apollon induce the ubiquitination and degradation of Smac/DIABLO [11, 44, 45]. Although in vivo relevance has not yet been demonstrated for these mechanisms, a complicated scenario emerges from these studies, underlying the need for more research focusing not only in Smac/DIABLO protein and isoforms, but also in their interaction partners in a tissue-specific manner.