AKT as locus of cancer angiogenic robustness and fragility

A cancer is a complex and a robust system (Radisavljevic, 2004a, 2008). Complexity of signaling in cancer and interconnections of signaling genes make complex organization of the signaling networks. Metastatic cancer, relapsed leukemia, relapsed lymphoma or progressed myelodysplastic syndromes (MDS) to acute myeloid leukemia (AML; Takahashi et al., 1995; Salven et al., 1997; Podar and Anderson, 2005; Ayala et al., 2009) have hyper complex properties because of the positive feedback loops of macromolecular signaling networks (Radisavljevic, 2008). Integrated signals from positive feedback loop systems create an extreme robustness in the cancer complex system (Radisavljevic, 2008). Integrated signaling modules with their interacting macromolecules, their complexity and robustness of nodal network can change cancer phenotype module from robust to the extreme robust (Radisavljevic, 2008). The cancer robustness is generated from simple signaling pathways such as the NOS/NO/Rb, the PI3K/AKT/mTOR/RAN proliferative pathways, and pure angiogenic pathway the VEGF/PI3K/AKT/NOS/NO/ICAM-1 (Radisavljevic et al., 2000; Radisavljevic, 2003, 2004a, b, 2008; Radisavljevic and Gonzalez-Flecha, 2004). A cancer complexity can amplifies small perturbation in the system causing failure of the complex robust system. The activation of apoptotic signaling pathway initiated by the inhibition of the NOS molecule through the NOS/NO/ROCK/FOXO3a signaling pathway causes complete failure of all cancer system (Radisavljevic, 2003). On the other hand, inhibition of the AKT by the trivial oxido-reductive perturbation, an inactivation of the H₂O₂ will dephosphorylates AKT and causes failure of all cancer robust system (Radisavljevic, 2008). A complex robust cancer system functions with multiple parameter elements. A macromolecules as a signaling elements have own fluctuations, variability and errors, and robust system is a complex system which tolerates that errors. Regulatory interactions between genes are made by negative feedback loops in normal cells, and in cancer cells with positive feedback loop systems of the interactive macromolecules in the cancer interactome. A cancer complexity is built from many macromolecular elements participating as functional elements of the signaling complex network system. Thus, protein interactome is a part of signaling interactome network, but they are two separate complex systems (Radisavljevic, 2008). Protein–protein interaction during signaling is followed by the conformational change of the protein and entropy/enthalpy of free energy change (Kortemme and Baker, 2002). The interacting synchronization of these two complex systems creates robustness, but extreme cancer robustness is generated by the positive feedback loops of amplified signals from the hypoxic cancer microenvironment in the metastatic cancer, relapsed leukemia or lymphoma. Signaling interactome has own complex nodal organization of the signaling genes, the structure composed of signaling nodes and signaling loci such as AKT locus in robust cancer system (Radisavljevic, 2008).

Robust system is a dynamic and stable system described by Lyapunov stability theory and mathematical function of dynamic systems. A complexity of the robust system is well described by the mathematical model of the high optimized tolerance (HOT) theory from statistical physics (Carlson and Doyle, 1999, 2002). The HOT system has high performance, high structural internal complexity with high densities of the interaction, simple robust external behavior and reliability, with the risk to potentially cascading failure initiated by possibly quite small trivial perturbations (Carlson and Doyle, 1999, 2002; Zhou et al., 2002). HOT theory postulated that system which acquires robustness against a common perturbation tends to be extremely fragile to some unexpected, tiny (trivial) perturbation (Carlson and Doyle, 1999, 2002; Csete and Doyle, 2002). HOT power law presents complex system as robust and fragile (Carlson and Doyle, 1999, 2002). Complex cancer system is the best example of the robust system, which can achieves extreme robustness when signaling becomes integrative of the positive loops and such a system is fragile at the same time if the key AKT locus is targeted (Radisavljevic, 2004a, 2008).

Cancer Angiogenic Locus

A cancer requires persistent angiogenesis (neovascularization) for its survival and growth (Folkman, 1971, 1995; Folkman and Shing, 1992; Tzankov et al., 2007). Angiogenesis is crucial in solid cancers, metastasis, hematological malignancies such as lymphoma, acute and chronic leukemia of myeloid and lymphoid lineages, multiple myeloma, and myelodysplastic syndromes (Takahashi et al., 1995; Salven et al., 1997; Podar and Anderson, 2005; Ayala et al., 2009). Angiogenesis is the formation of new blood vessels involving degradation of extracellular matrix and migration of endothelial cells and pericytes (Folkman, 1971, 1995; Folkman and Shing, 1992). New blood vessels can occur from pre-existing vessels by branching of new capillaries (sprouting angiogenesis) or from enlargement, splitting and fusion of pre-existing vessels (non-sprouting angiogenesis; Fidler and Ellis, 2004). Angiogenesis involves degradation of the parent venule basement membrane, endothelial cell proliferation, migration and development of sprouts, and generation of new basement membrane (Folkman, 1995). The most potent regulator of angiogenesis is the vascular endothelial growth factor (VEGF), also known as a vascular permeability factor and its increased level have been correlated with poor prognosis in cancer, lymphoma and leukemia (Bussolino et al., 1996; Ferrara, 1996). Also, VEGF has role in organ development, wound healing, tissue regeneration, rheumatic artitis, and retinopathies (Folkman and Klagsbrun, 1987).

Angiogenesis is necessary for the development of a malignant phenotype (Hanahan and Folkman, 1996). Higher levels of VEGF, VEGFR1, VEGFR2, and plasma endostatin have been found in AML, MDS and have been correlated with worse survival (Aguayo et al., 1999, 2000). Angiogenesis is increased in acute promyelocytic leukemia (APL) and was mediated by VEGF. The all-trans retinoic acid (ATRA) inhibits VEGF production and decreases microvessel density in acute promyelocytic leukemia (Kini et al., 2001).

VEGFR-2 (KDR/Flk1 receptor) is a high-affinity VEGF receptor (Terman et al., 1992) and a member of the tyrosine kinase receptor superfamily (de Vries et al., 1992) which plays a crucial role in hematopoiesis (Klagsbrun and D'Amore, 1996). VEGF-1 receptor (Flt-1) also is an important for hematopoietic stem cell survival and proliferation (Gerber et al., 2002). Knockout of VEGFR-2 gene in mice fails to develop hematopoietic cells (Shalaby et al., 1995). Pluripotent hematopoietic cells express VEGFR-2 (Ziegler et al., 1999). A CD34 positive stem cells are precursors of both human hematopoietic and endothelial cells which express VEGFR-2 (Terman et al., 1992). VEGFR-2 is overexpressed in chronic lymphocytic leukemia (CLL) and patients with high VEGFR-2 levels have shortened survival (Ferrajoli et al., 2001). During tumor angiogenesis endothelial precursor cells are mobilized from the bone marrow, transported trough bloodstream and incorporated into the walls of growing new blood vessels. Thus, same mechanism was seen in early embryogenesis (Rafii, 2000). That funding was confirmed by the bcr-abl fusion protein found in the bone marrow endothelial cells of chronic myeloid leukemia (CML) and in B cell lymphomas (Gunsilius et al., 2000; Streubel et al., 2004).

Hypoxia-inducible factor-1 (HIF-1α), a transcription factor is overexpressed in human cancers and associated with increased cancer vascularity, cancer progression and invasion (Semenza, 2003). Hypoxia develops when the cancer growth exceeds the rate of new blood vessel growth (Giatromanolaki and Harris, 2001). In cancer overgrowth exist hypoxia which generates HIF-1α. Its degradation under hypoxic condition is inhibited allowing HIF-1α to accumulate, dimerize with HIF-1β (aryl hydrocarbon nuclear translocator, ARNT) forming HIF which translocates to the nucleus. This hetrodimer is a basic-helix-loop-helix (bHLH) transcription factor, a protein structural motif that characterizes a family of transcription factors that activates transcription of HIF-1α, erythropoietin, transferrin, endothelin-1, inducible nitric oxide synthase, heme oxygenase-1, vascular endothelial growth factor, insulin-like growth factor-2 (IGF-2), insulin-like growth factor binding protein-1, -2, and -3 (IGFBP-1, -2, and -3), glucose transporters glycolytic enzymes (Tennant et al., 1996; Mazzucchelli et al., 2000), hepatocyte growth factor (HGF; Trusolino and Comoglio, 2002) and upregulating MET protooncogene (Pennacchietti et al., 2003), causing increase in invasive growth and metastasis (Harris, 2002; Trusolino and Comoglio, 2002; Pennacchietti et al., 2003; Semenza, 2003). In normoxia, degradation of HIF-1α is initiated by the hydroxylation on the two proline residues by prolyl hydroxylases (Ivan et al., 2001; Jaakkola et al., 2001). Thus, HIF-1α is bound to the tumor suppressor Von Hippel-Lindau (pVHL) during proline hydroxylation then ubiquitylated and degradated in the proteasome (Metzen et al., 2003). Hypoxia developed in cancer, generates resistance to radiation (Gray et al., 1953), chemotherapy (Brown and Giaccia, 1998) and creates a more aggressive phenotype (Graeber et al., 1996; Koong et al., 2000).

Increased bone marrow microvessel density was found in acute limphoblastic leukemia (ALL; Perez-Atayde et al., 1997). Increased vascularity also was found in bone marrow in patients with AML (Hussong et al., 2000) as well as VEGF expression on their blasts (Fiedler et al., 1997). VEGF expression were found in aggressive non-Hodgkin's lymphomas including diffuse large B cell lymphoma (DLBCL), mantle cell lymphoma (MCL), indolent lymphoma, and chronic lymphocytic leukemia/small lymphocitic lymphoma (CLL/SLL; Ruan and Leonard, 2009). Lymphoma growth and progression depends of the VEGF autocrine and paracrine stimulation where lymphoma cells express VEGF and VEGFR (Ruan et al., 2009). DLBCL cells express VEGF, VEGFR1, and VEGFR2 suggesting autocrine and paracrine effects (Gratzinger et al., 2010).

VEGFA is produced by cancer cells and binds to the VEGFR-1 and VEGFR-2 (Jain, 2005). VEGFR-1 is expressed on hematopoietic cells and endothelial progenitor (Hattori et al., 2002). VEGFA cause transient loss of endothelial cells connections (Suarez and Ballmer-Hofer, 2001). Anti-angiogenic therapy that inhibits VEGFA through VEGFR signaling may restore endothelial connections and reduces functional shunting in cancer (Jain, 2005, 2008; Zhong and Bowen, 2006; Fukumura and Jain, 2007; Jain et al., 2007; Bergers and Hanahan, 2008). VEGFR-2 is the primary receptor for the transmission of VEGF signals in endothelial cells (Gerber et al., 1998) through the AKT signaling (Radisavljevic et al., 2000; Radisavljevic, 2003, 2004a, 2008). VEGF-C and VEGF-D are involved in lymphangiogenesis and tumor angiogenesis via VEGFR-3 (Tammela et al., 2008). VEGF gene expression is regulated by HIF-1α (Kaelin, 2007). Chemokine SDF-1α promotes cancer angiogenesis by the CXCR4 on endothelial progenitor cells from bone marrow (Jin et al., 2006).

A cancer is hypoxic microenvironment with increased levels of reactive oxygen species (ROS) including H₂O₂ and nitrosative radicals (Upham et al., 1997; Radisavljevic, 2003, 2004a, b, 2008; Radisavljevic and Gonzalez-Flecha, 2003; Chen et al., 2006) affecting gap junctions. Hypoxia causes dephosphorylation of Cx43 at Ser365, resulting in the loss of Cx43 function in gap junctions (Jain, 2005; Solan et al., 2007) and H₂O₂ cause hyperphosphorylation of the Cx43 leading to the loss of gap junction function (Upham et al., 1997) initiating formation of functional shunts (Pries et al., 2010). Activity of the AKT controls phosphorylation of Cx43, but Cx43 ubiquitination is not necessary for its regulation (Dunn et al., 2012). Targeting angiogenic AKT locus rather then targeting peripheral signaling elements VEGF or HIF-1α which control just one small part of signaling network will restore normal circulation and eliminates functional shunts. Thus, targeting angiogenic integrative AKT locus can block all signals including the VEGF, HIF-1α, H₂O₂, HGF, and other growth factor signals (Fig. 1). This new principle for new locus chemotherapy, targeting AKT angiogenic locus, can solve the problems of the cancer drug resistance and radiation therapy resistance. AKT activity is regulated by the balance between oxidant H₂O₂ and antioxidant catalase (Radisavljevic and Gonzalez-Flecha, 2003; Radisavljevic, 2004a, 2008). We have shown in previous experiments that the AKT is upstream of the mTOR (Radisavljevic and Gonzalez-Flecha, 2004) in signaling pathways activated by reactive oxygen species such as H₂O₂ inducing phosphorylation and activation of AKT, transmitting signals toward mTOR and then to the RAN protein. AKT activation by the H₂O₂ induces proliferation trough the H₂O₂/PI3K/AKT/mTOR/RAN pathway (Radisavljevic and Gonzalez-Flecha, 2003), but hyperstimulation by the H₂O₂ caused apoptosis (Radisavljevic and Gonzalez-Flecha, 2003). Thus, balance between oxidants/antioxidants is the crucial element in the signaling interactome determining final response of apoptosis or proliferation (Radisavljevic and Gonzalez-Flecha, 2003; Radisavljevic, 2004a, 2008). Other authors concluded that AKT sensitizes cells to oxidative apoptosis (Nogueira et al., 2008), instead that AKT does not determines oxidative response nor sensitizes cells to oxidative apoptosis, but rather balance of the oxidants/antioxidants determines response of the AKT in cancer system (Radisavljevic et al., 2000; Radisavljevic, 2003, 2004a, b, 2008; Radisavljevic and Gonzalez-Flecha, 2003, 2004).

It was reported (Semenza, 2003) that HIF-1α is a master regulator of clusters of genes in the cancer. The HIF-1α signaling is regulated through the AKT (Lee et al., 2008). Thus, it is rather that AKT is master locus of cancer robustness then HIF-1α, because stimulating signals coming from the HIF-1α are regulated by AKT locus, and AKT locus is regulated by the balance of the oxidants/antioxidants (Radisavljevic, 2008).

Phosphorus incorporated in the DNA gives robustness to all spiral structure. Also, incorporation of the phosphorus into AKT protein, which becomes phosphorylated protein and active element in angiogenic signaling, gives robustness to all angiogenic signaling network. Phosphorylated AKT is crucial element of the agiogenic robustness giving all system extreme robustness by positive feedback loops in metastatic cancers (Radisavljevic, 2008), relapsed leukemia or relapsed lymphoma (Uddin et al., 2006), but AKT is also a locus of angiogenic fragility because its dephosphorylation causes failure of the whole signaling robust system (Radisavljevic, 2008). The balance between phosphorylated and dephosphorylatied AKT is a crucial element controlling cancer angiogenic robustness. The AKT phosphorylation is controlled by balance between H₂O₂/catalase (oxidant/antioxidant) which switching AKT activity (Radisavljevic, 2008).

Protein interactome is a part of signaling interactome system and synchronization of these two systems creating robustness. Extreme cancer robustness is generated by the positive feedback loops by the amplification of the signals from the hypoxic microenvironment in the metastatic cancers, relapsed leukemia or relapsed lymphoma through the phosphorylated AKT which is regulated by the balance of the oxidant/antioxidant. Upstream angiogenic locus is the AKT and downstream angiogenic node is NOS, which directly controls angiogenic pathway (Radisavljevic, 2004a, 2008). Upstream AKT locus integrates NOS pathway and mTOR signaling pathway through the NOS/NO/Rb cell proliferative pathway, the VEGF/PI3K/AKT/NOS/NO/ICAM-1 angiogenic and cell migratory pathway, the PI3K/AKT/mTOR/RAN, proliferative pathway, and the NOS/NO/ROCK/FOXO3A as apoptotic pathway, in the nodal organization of signaling proteins (Fig. 1). The AKT locus integrates amplified signals from the positive feedback loops of the HIF-1α stimulation, the VEGF stimulation, the HGF stimulation, as well as H₂O₂ signaling, generated in the hypoxic cancer microenvironment, in the one complex and robust amplified angiogenic signal forming functional shunts (Fig. 1). Recently was reported that RAN is a potential therapeutic target for cancers (Yuen et al., 2012). However, RAN is only effector molecule in the PI3K/AKT/mTOR/RAN pathway (Radisavljevic, 2004a, 2008). Thus, AKT is a crucial locus in the nodal signaling network which can be targeted as angiogenic locus in cancer. Targeting the AKT upstream angiogenic locus in the signaling networks is the best way to target cancer angiogenic robustness.

Conclusions

AKT is an upstream locus of the integrated angiogenic robustness and fragility. Integrated signaling from hypoxic microenvironment induces creation of positive loops which form functional shunts and extreme robustness in metastatic cancers, relapsed leukemia and lymphoma. A complex angiogenic singling network becomes integrated through the upstream AKT angiogenic locus. Phosphorylated AKT locus is regulated by the balance of the oxidant/antioxidant. The AKT locus regulates two downstream nodes the NOS and mTOR. The downstream NOS node is rather integrated angiogenic node then single angiogenic node. It integrates NOS/NO/Rb proliferative pathway, the VEGF/PI3K/AKT/NOS/ICAM-1 angiogenic and cell migratory pathway and the NOS/ROCK/FOXO3a apoptotic pathway. The mTOR node controls the PI3K/AKT/mTOR/RAN cell proliferative pathway. Targeting upstream angiogenic AKT locus represents new principle to target cancer angiogenic robustness by new locus chemotherapy.