**4. Enhancing the antitumor efficacy of soluble TRAIL**

Despite very promising preclinical in vitro and in vivo antitumor evidences, phase I/II clinical trials have demonstrated limited antitumor activity of soluble TRAIL likely due to

1997a), TRAIL-R4 (Degli-Esposti et al., 1997), and osteoprotegerin (OPG) (Emery et al., 1998). Both TRAIL-R1 and TRAIL-R2 are type I transmembrane proteins containing a cytoplasmic death domain (DD) motif that engage apoptotic machinery upon ligand binding (Almasan & Ashkenazi, 2003), whereas the other three receptors either act as decoys or transduce antiapoptotic signals (Wang & El-Deiry, 2003). TRAIL-R3 and TRAIL-R4 have close homology to the extracellular domains of agonistic receptors. TRAIL-R4 has a truncated, nonfunctional cytoplasmic DD, while TRAIL-R3 exists on the plasma membrane as a glycophospholipid-anchored protein lacking the cytosolic tail. The physiological relevance of OPG as a soluble receptor for TRAIL is unclear, but a recent study suggests that cancer-derived OPG may be an important survival factor in hormone-resistant prostate

Soluble TRAIL forms homotrimers that bind three receptor molecules, each at the interface between two of its subunits. A Zn atom bound to cysteine residues in the trimeric ligand is essential for trimer stability and optimal biologic activity. Binding of TRAIL to the extracellular domain of agonistic receptors results in trimerization of the receptors and clustering of the intracellular DDs, which leads to the recruitment of the adaptor molecule Fas-associated protein with death domain (FADD). Subsequently, FADD recruits initiator caspase-8 and -10, leading to the formation of the death-inducing signaling complex (DISC), where initiator caspases are autoactivated by proteolysis. Once they become enzymatically active, caspase-8 and/or -10 are released from the DISC and signal through two different proteolytic pathways that converge on caspase-3 and lead to cellular disassembly (Kaufmann & Steensma, 2005). In type I cells, activation of initiator caspases upon death receptors ligation is sufficient to directly activate downstream effector caspases, such as caspase-3 and/or -7 (Scaffidi et al., 1998). This extrinsic pathway is independent of the mitochondria and is not blocked by overexpression of Bcl-2. In type II cells, the commitment from death receptor ligation to apoptosis is less direct (Scaffidi et al., 1998). The amount of initially cleaved caspase-8 and/or -10 is not enough to directly trigger effector caspases activation. Consequently, apoptotic signaling requires an amplification loop by mithocondrial pathway engagement through caspase 8-mediated cleavage of Bid (BH3 interacting death domain agonist), which, in turn, induces the cytosolic Bcl-2 family member Bax (Bcl-2-associated X protein) and/or the loosely bound mitochondrial homolog Bak (Bcl-2 antagonist/killer) to insert into the mitochondrial membrane, where they contribute to the mitochondrial release of cytochrome c (Lucken-Ardjomande & Martinou, 2005). In the cytosol, cytochrome c binds the adaptor protein Apaf-1 (apoptotic protease activating factor 1) to form an apoptosome with recruitment and activation of the apoptosis-initiating caspase-9, which proteolytically activates additional caspase-3. These events are further amplified by apoptogenic factors released from the mitochondrial space, including Smac/DIABLO (second mitochondrial activator of caspases/direct IAP-binding protein

cancer cells (Holen et al., 2002).

**3. TRAIL-induced apoptosis signaling** 

with low pI) (Verhagen & Vaux, 2002).

**4. Enhancing the antitumor efficacy of soluble TRAIL** 

Despite very promising preclinical in vitro and in vivo antitumor evidences, phase I/II clinical trials have demonstrated limited antitumor activity of soluble TRAIL likely due to its short half-life and the consequent short exposure of tumor cells to the molecule (Ashkenazi et al., 2008). Because of soluble TRAIL's short half-life (Ashkenazi et al., 1999; Kelley et al., 2001; Walczak et al., 1999), it seems unlikely that the recommended soluble TRAIL dose of 8 mg/kg body weight will allow adequate exposure of tumor cells at high drug concentrations (Ashkenazi et al., 2008). Strategies to enhance the therapeutic activity of soluble TRAIL include combining it with conventional chemotherapy (Ballestrero et al., 2004) or with new agents such as histone deacetylase inhibitors that upregulate TRAIL-R1 and/or TRAIL-R2 (Inoue et al., 2004).

Gene therapy approaches have also been proposed to enhance TRAIL-mediated tumor cell targeting. Recently, a TRAIL-expressing adenoviral vector (Ad-TRAIL) has been shown to cause direct tumor cell killing, as well as a potent bystander effect through presentation of TRAIL by transduced normal cells (Lee et al., 2002). Thus, using Ad-TRAIL might be an alternative to systemic delivery of soluble TRAIL possibly resulting in better tumor cell targeting and increased tumoricidal activity (Armeanu et al., 2003; Griffith et al., 2000; Griffith & Broghammer, 2001; Kagawa et al., 2001; Lee et al., 2002). However, systemic Ad-TRAIL-based gene therapy requires efficient infection of target tumor cells as well as avoidance of immune clearance, and is limited by several safety and toxicity issues related to intravenous adenovector administration (Harrington et al., 2002). Intratumoral injection of TRAIL-encoding adenovectors has been successfully explored in a number of experimental models; however, this approach results in local antitumor activity and has little, if any, value in the treatment of disseminated tumors.

Alternatively, cell-based vehiculation of the full-length, membrane-bound (m)TRAIL (Griffith et al., 2009) has been proposed to achieve an optimal systemic delivery. Indeed, genetically modified stem/progenitor cells represent an innovative approach for delivery of anticancer molecules (Harrington et al., 2002; Introna et al., 2004). Due to their homing properties, systemically injected stem/progenitor cells could infiltrate both primary and metastatic tumor sites, thus allowing tumor-specific targeting (Burger & Kipps, 2006; Jin et al., 2006; Kaplan et al., 2007; Kucia et al., 2005; Loebinger et al., 2009; Najbauer et al., 2007; Rafii et al., 2002), and potentially overcoming limitations inherent to the pharmacokinetic profile of soluble drugs (Aboody et al., 2008; Griffith et al., 2009; Sasportas et al., 2009). Neural or mesenchymal stem cell-mediated mTRAIL delivery has been investigated in solid tumors (Grisendi et al., 2010; Kim et al., 2008; Loebinger et al., 2009; Menon et al., 2009; Mohr et al., 2008; Uzzaman et al., 2009).

In order to optimize the use of TRAIL-encoding adenovectors for the treatment of systemic tumors, we have recently investigated a cell-based approach using mobilized CD34+ hematopoietic cells transduced with a replication-deficient Ad-TRAIL (CD34-TRAIL+) encoding a full-length mTRAIL under the control of the CMV promoter (Carlo-Stella et al., 2006; Griffith et al., 2000). Several lines of evidence support the use of gene-modified CD34+ cells as optimal vehicles of antitumor molecules. In fact, CD34+ cells are already widely used in the clinical setting. Additionally, they can migrate from the bloodstream into tumor tissues due to the expression of adhesion receptors that specifically interact with counterreceptors on endothelial cells in the tumor microenvironment (Burger & Kipps, 2006; Kaplan et al., 2005; Verfaillie, 1998). Moreover, up-regulation of inflammatory chemo-attractants in the tumor microenvironment provides with a permissive milieu that potentially allows for homing of systemically delivered CD34-TRAIL+ cells and efficient tumor targeting (Jin et al., 2006). Using a multiplicity of infection (MOI) of 500, the transduction protocol optimized for the transduction of CD34+ cells consistently results in a transduction efficiency higher

Targeting TRAIL Receptors with Genetically-Engineered CD34+ Hematopoietic Stem Cells 665

intravenous injection of CD34-TRAIL+ cells (3 × 106 cells/mouse) (Lavazza et al., 2010). Tumor and healthy tissue sections were immunostained with an anti-human CD45 antibody and digitally recorded to count transduced cells on entire tissue sections.3 Early following injection, transduced cells were detected at high frequencies in the lung, liver and spleen (**Figure 1**). CD34-TRAIL+ cells progressively decreased and were no longer detectable in these tissues 24 hours after injection. Bone marrow CD34-TRAIL+ cells peaked 5 hours after injection and were detectable up to 24 hours. Low frequencies of transduced cells were detected within tumors as early as 30 minutes following injection. They progressively increased and peaked 48 hours post-injection when on average 188 ± 25 CD45+ cells per 105 tumor cells (i.e., 0.2 ± 0.03%) were recorded (**Figure 1**). Overall, kinetics data suggest that transduced cells transiently circulate through healthy tissues, whereas they are preferentially recruited within tumor nodules, allowing to hypothesize that homing signals

by tumor endothelial cells actively promote intratumor homing of transduced cells.

Fig. 1. Tissue kinetics of CD34-TRAIL+ cells. Lung, liver, spleen, femur, and tumor nodules were harvested from tumor-bearing NOD/SCID mice 0.5 (), 5 (), 24 (), and 48 () hours after a single intravenous injection of CD34-TRAIL+ cells (3 × 106 cells/mouse). Shown is the quantification of CD34-TRAIL+ cells on digitally acquired tissue sections stained with anti-CD45. Frequency of CD34-TRAIL+ cells is expressed as the mean (± SD)

3 Images of tissue sections were acquired at 20× magnification with an automatic high-resolution scanner (dotSlide System, Olympus, Tokyo, Japan) and subdivided into a collection of non-overlapping red, green, and blue (RGB) images in TIFF format (final resolution 3.125 pixels/µm). Image analysis was carried out using the open-source ImageJ software (http://rsb.info.nih.gov/ij/). Routines for image analysis were coded in ImageJ macro language and executed on RGB images without further treatment. Per each experimental condition, at least three sections from different tumor nodules or healthy tissues were analyzed. Intratumor frequency of CD34-TRAIL+ cells was expressed as the number of CD45+ cells per total cells per tisue section. Total cells were counted by the ImageJ internal function for particle analysis, whereas CD45+ cells were manually counted in all images from whole scanning of

number of CD45+ cells per tissue section.

histochemically stained tissue sections.

than 80% (range 70% - 96%), a high level expression of mTRAIL, and a cell viability ≥85%. Flow cytometry analysis of CD34-TRAIL+ cells shows significant levels of transgene expression for at least 96 hours after transduction, and Western blot analysis reveals the presence of 32- and 55-kDa proteins, which are the expected products for full-length monomer and dimer TRAIL, respectively (Carlo-Stella et al., 2006).
