Distinct Roles For Two Omnipresent Partners Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Lung cancer is the most common malignancy in the world with 1.61 million new cases diagnosed every year. Moreover, according to CDC and WHO cancer statistics (the GLOBOCAN project), mortality rates render lung cancer one of the commonest causes of death from malignancies, for both men and women [1, 2].

Advanced glycation end-products (AGEs) are a complex and heterogeneous group of compounds formed via a non-enzymatic reaction between reducing sugars and amine residues on proteins, lipids, and nucleic acids [3-7]. AGEs are prevalent in pathological conditions marked by hyperglycemia, inflammation, and oxidative stress, and also in renal failure and aging. The formation of AGEs is implicated in the development of diabetes mellitus and its complications [6, 8-10], as well as in a wide and disparate range of pathologies, such as rheumatoid arthritis [11], polycystic ovary syndrome [12], renal and cardiovascular diseases [6, 8, 10], malignancies [13, 14], neurodegeneration and Alzheimer's disease [15, 16].

The receptor for advanced glycation end-products (RAGE) is a transmembrane pattern recognition receptor of the immunoglobulin superfamily [17-19]. RAGE was initially identified and characterized for its ability to bind AGEs [17, 18]. The current body of research underscores the important role of RAGE in pulmonary physiology and disease, especially in lung cancer [20, 21]. However, many aspects of the implication of RAGE in lung physiology and pathophysiology still remain ambiguous, whereas even less has been reported on the possible correlation between AGEs and lung cancer.

This review aims: a) to identify and interpret evidence relating AGEs and lung cancer and b) to clearly define and illustrate important, newly disclosed aspects of RAGE function in lung physiology and lung cancer.

Keywords: advanced glycation end-products, receptor for advanced glycation end-products, lung cancer, ROS, ligands, charge


The study of AGEs represents one of the most promising areas of research. Although the initial chemistry behind their formation has been known since the early 1900's, it is only during the last few decades that significant work has been done to expand upon this.

Non-enzymatic glycosylation, which eventually leads to the formation of AGE-modified proteins and other molecules, is enhanced at sites of sustained inflammation, in the presence of hyperglycemia, in aging, in end-stage renal disease, and under other conditions associated with oxidative stress [6, 8, 9].

AGEs accumulation in diabetes may result from both chronic hyperglycemia promoting the generation of AGEs and concomitant impaired renal function, since the kidney is the major site of AGEs clearance [8, 9]. AGE-modified proteins are more resistant to enzymatic degradation and it is likely that this further promotes local tissue AGE accumulation. The role of AGEs in the development of diabetic complications is well-known. Diabetic complications are multifactorial in origin; however the biochemical process of advanced glycation, which is accelerated in diabetes as a result of chronic hyperglycemia and increased oxidative stress, has been postulated to play a central role in the development of microvascular and macrovascular disease [6, 9, 10]. AGEs accumulate in most sites of diabetic complications, including kidney, retina, peripheral nerves and atherosclerotic plaque [8, 10, 22, 23].

Apart from diabetes-related complications, AGEs have also been implicated in a wide and disparate range of diseases such as rheumatoid arthritis, polycystic ovary syndrome and Alzheimer's disease [11, 12, 16]. Histopathological studies show AGEs accumulation in a variety of tissue types, including glomerular basement membrane and mesangial and endothelial cells of the kidneys [6, 8], skin [24, 25], amyloid plaques in Alzheimer's disease [16], synovial tissue in rheumatoid arthritis [11, 26], myocardium [27, 28] and liver [29].

AGEs may also be derived from exogenous sources, such as tobacco smoke and certain foods, particularly those that are heated [30, 31]. Tissue and circulating AGEs levels are higher in smokers and in patients on high AGEs diets, with concurrent increases in inflammatory markers. Tobacco-derived AGEs have been observed in the lens and coronary artery vascular wall of cigarette smokers [32]. The western diet is today typically full of AGEs. Food processing, especially prolonged heating, has an accelerating effect in the generation of glyco-oxidation and lipo-oxidation products, and a significant proportion of ingested AGEs is absorbed [31, 33-35].

Despite their complexity and widespread pathological distribution, AGEs mediate their effects through a certain repertoire of mechanisms that include structural and functional modification of proteins, lipids and nucleic acids, and formation of covalent cross-links between proteins or other adducts [3-7, 36].

RAGE is a transmembrane pattern recognition receptor of the immunoglobulin superfamily and is encoded on chromosome 6, in the Class III major histocompatibility locus, which encodes members of the innate immune system. According to the Human Gene Nomenclature Committee, it is now known as AGER (advanced glycosylation end-product-specific receptor), while the symbol RAGE is considered to be a synonym. RAGE structure consists of five domains: three extracellular immunoglobulin-like domains (V, C1, C2), a single transmembrane domain, and a short intracellular negatively charged C-terminal tail. The V-domain is critical for ligand binding, whereas the cytosolic tail is essential for downstream intracellular RAGE signaling. Interaction of the RAGE cytoplasmic domain with diaphanous-1 is required to transduce signals coming from RAGE-ligand engagement and stimulate fundamental signaling networks. [19, 37, 38].

RAGE was initially identified and characterized for its ability to bind AGEs [17, 18] and, indeed, AGEs were thought to be its main activating ligands. However, many more ligands have been identified since then, such as the High Mobility Group family proteins, including HMGB1 (high mobility group box 1: amphoterin), S100 proteins (members of the calgranulin family), integrin Mac-1, matrix proteins such as Collagen IV and I, amyloid-β peptide (relevant to the pathophysiology of Alzheimer's disease), and amyloid A (systemic amyloidosis). Nowadays, RAGE is considered as a multiligand pattern recognition receptor [19, 37], which recognizes a variety of structurally different molecules. It is believed that RAGE identifies the three-dimensional structure of ligands, rather than specific amino acid sequences. RAGE-ligand interaction leads to the activation of diverse cascades, depending on the ligand, the particular environment, and the cell-tissue type, and stimulates a considerable variety of distinct biological pathways related to cell differentiation, proliferation, adhesion and migration [38-43]. These pathways are related to the development of common and severe disease states, such as neurodegenerative diseases [15, 16], atherosclerosis, diabetes [15, 44, 45], chronic inflammation and cancer [46, 47].

RAGE presents a particular challenge to the well-established concept of molecular recognition stating that every receptor has a ligand that binds to a specific area on its surface [19]. For example, any intracellular protein could be AGE-modified leading to numerous possible AGE-modified products. In addition, non-enzymatic glycation can modify more than one site on a single molecule, creating many potential ligands for RAGE. Furthermore, apart from AGEs, the catalogue of RAGE-ligands contains a long list of structurally diverse molecules. Indeed, not only is it still unknown how one receptor is able to recognize a vast number of seemingly unrelated structures, but, moreover, how the same receptor stimulates diverse intracellular events is also not known. In the same context, it is not well understood why some ligands, such as HMGB1 and carboxymethyllysine (CML-AGE) demonstrate strong pro-inflammatory signaling through RAGE, while similar molecules, such as pentosidine-AGE and pyrraline-AGE seem to have much less or no signaling potential. Additionally, high RAGE-ligand binding affinity does not necessarily correlate with signaling generation and cellular activation [48-50].

RAGE, AGEs and cancer biology:

RAGE, inflammation and cancer

The correlation of RAGE with cancer pathophysiology has been extensively studied and is well documented. Several detailed reviews and experimental reports clearly show that the multiligand-RAGE axis is implicated in chronic inflammation, together with regional and systemic immune response deregulation and intercellular communication derangement [46, 51-53].

The link between inflammation and cancer was proposed more than a century ago by Rudolf Virchow, who noticed the infiltration of leukocytes into malignant tissues. Since then, a broad range of experimental and clinical evidence has shed considerable light on our understanding of the central role of chronic inflammation in tumor pathophysiology. Today, it is widely accepted that many neoplastic diseases are driven, at least in part, by chronic and often subclinical inflammation. Detailed reviews and reports on the role of the multiligand receptor RAGE and its ligands in the development of inflammation, tumor microenvironment and tumor promotion have been published and the interested reader is referred to these for further study [46, 47, 49, 51]. Christoffer Gebhardt et al. in their report [51] provide evidence that RAGE signaling drives the strength and maintenance of inflammatory reaction during tumor promotion. Their study was based on an experimental in vivo model of chemically induced carcinogenesis, using both wild-type and RAGE-deficient female mice. They found that RAGE is associated with the recruitment of the CD11b+ Gr-1+ subset of myeloid cells, known to induce T lymphocyte dysfunction and inhibit antitumor adaptive immunity. In addition, RAGE expression was associated with significant changes in the macrophage inflammatory protein (MIP) family members, which are chemokines responsible for the recruitment of polymorphonuclear cells at sites of inflammation. As concerns the relation between RAGE and the upregulation of RAGE-ligands, it was demonstrated that RAGE expression led to upregulation of the calcium-binding S100A8/A9 complex, which further engages with RAGE in order to induce intracellular signaling. These proteins act as proinflammatory mediators in acute and chronic inflammation and are strongly upregulated in chemically induced skin carcinogenesis models. In addition, expression of RAGE was involved in the upregulation of prostaglandin-endoperoxide synthase 2 enzyme, which has an important role as proinflammatory mediator, thus representing a target for cancer chemoprevention by NSAIDs.

The targets of RAGE signaling identified in the study of Christoffer Gebhardt et al. are well known target genes of NF-kB, an essential player in the link between inflammation and cancer development and progression [54, 55]. RAGE has the capacity to induce sustained NF-kB activation and is a potent upregulator of NF-kB signaling [56, 57]. The generation of an inflammatory microenvironment supports tumorigenesis by promoting cancer cell survival, proliferation, migration and invasion, and is strongly dependent on the activation of NF-kB and other transcription factors; these in turn regulate the expression of cytokines, such as TNFa, IL-1, IL-6, that are critically involved in the crosstalk between cancer cells and cells of the tumor stroma [47, 49, 58].

Numerous findings have advanced our knowledge about the contribution of RAGE to the regulation of innate and adaptive immune responses. RAGE expression has been functionally linked to most key cell types involved integrally in an immune response, i.e. monocytes/macrophages, neutrophils, dendritic cells, as well T and B-lymphocytes [42, 59-62]. RAGE is highly expressed on endothelial cells and acts as an adhesion receptor for leucocytes by direct interaction with β2 integrin Mac-1 [63]. In the context of inflammation-associated cancer, RAGE and RAGE-ligands are overexpressed in most types of solid tumors [51, 64]. S100A9 overexpression is critically important for accumulation of myeloid-derived suppressor cells (MDSCs). Tumor cells secrete S100A8/S100A9 that binds to RAGE on MDSCs and promotes their migration and accumulation [65]. RAGE/S100-dependent recruitment of MDSCs is a major immunological abnormality in cancer, resulting in T-cell tolerance and suppression of antitumor immune response.

Apart from the role of S100A8/A9 and RAGE in inflammation and cancer development [66], another important RAGE ligand, amphoterin (HMGB1) is commonly implicated in tumor biology. Amphoterin is a high-mobility group I non-histone chromosomal DNA-binding protein; however it can be found in the extracellular space as well. Both amphoterin and RAGE are upregulated in inflammation and are expressed in many tumor types, including pancreatic, colon, gastric, breast and prostate cancers and leukemia; cancer cells with high metastatic ability display strong expression of amphoterin, usually with high RAGE co-expression [67-73].

Armando Rojas et al. discuss the role of the multiligand/RAGE axis and its contribution to cancer biology, particularly in the multicellular crosstalk established in the inflammatory tumor microenvironment [47]. Development of inflammation within the microenvironment of neoplastic tissue is now generally thought to promote tumor growth and metastasis. Evidence derived from epidemiological studies and basic research have shown that organ-specific carcinogenesis is linked to the development of chronic local inflammation, as has been reported with regard to the relationship between a) Helicobacter pylori-induced gastric inflammation and gastric cancer/lymphoma, b) prostatitis and prostate cancer c) bowel disease and colon cancer, d) chronic cholecystitis and gall bladder cancer etc.

A key consequence of RAGE engagement is the activation of multiple signaling pathways that incorporate or are related to reactive oxygen species (ROS) production and enhanced oxidative stress, ERK1/2, p38 and SAPK/JNK MAP mitogen activated protein kinases (MAPKs), p21ras and Rho GTPases (Rac1 ,Cdc42), phosphoinositol-J kinase (PIJK) and the JAK/STAT pathway. These effects have important downstream inflammatory consequences , such as the activation of transcription factors that include NF-kB, activator protein-1 (AP-1), and signal transducer and activator of transcription 3 (STAT-3). Through RAGE engagement and activation of these complex multiple signaling pathways, RAGE and its ligands promote tumor growth and invasion, hypoxia resistance, angiogenesis of tumor vasculature, and modulation of the host immune response, contributing to the immunosuppressive state seen in cancer patients [46, 47, 49].

It is thus not surprising that most cancer cells upregulate both the expression of RAGE and its ligands. RAGE-ligands act in an autocrine fashion and induce direct activation of cancer cells by stimulating proliferation, invasion and metastasis. Moreover, they act in a paracrine manner over RAGE-positive cells within the tumor microenvironment, including fibroblasts, leukocytes and vascular cells. AGEs, S100/calgranulins, and HMGB1 are RAGE-ligands, which are particularly relevant in tumor biology. Several clinical studies have demonstrated a strong association between RAGE expression and malignant potential of various cancer types, such as gastric [71], colorectal [74, 75], pancreatic [76] and prostate cancer [69], as well as oral squamous cell carcinoma [77]. However, some types of cancer constitute a conspicuous exception to this almost universal rule. Examples of these exceptions are esophageal cancer [78] and, especially, lung cancer [20].

AGEs and inflammatory responses

The biology of AGEs extends beyond settings such as diabetes, kidney failure and atherosclerosis. The observation that AGEs are formed in inflammatory lesions, such as in the joints of patients with dialysis-related amyloidosis, or in highly oxidative states, suggested that the potential biological effects of these modified complexes are more extensive [79]. Formation of CML-AGEs may be driven by activated myeloperoxidase, indicating that recruitment and accumulation of neutrophils in areas of inflammation may further stimulate generation of CML-modified proteins and lipids. Indeed, activated myeloperoxidase is present in human atherosclerotic lesions, a fact that favors the concept that in inflammatory foci there is sustained formation of AGEs [79].

RAGE engagement by AGE elicit proinflammatory responses in many cell types, including endothelial cells, smooth muscle cells and mononuclear phagocytes, mediated by NF-kB and consequent cytokine expression. Engagement of RAGE by AGEs leads to activation of signaling cascades, finally resulting in activation of nuclear transcription factors. Transcription factors, including NF-kB, translocate into the nucleus and enhance the transcription of numerous proinflammatory cytokines, adhesion molecules and other proteins [10, 79-83]. These include endothelin-1, intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), E-selectin, vascular endothelial growth factor (VEGF), tissue factor, interleukines, TNF-a, and RAGE.

AGEs furthermore modulate immune responses by interfering with monocyte migration and maturation. In one of the first reports on RAGE-AGE interactions by Ann Marie Schmidt, it was experimentally shown that RAGE, in response to soluble AGEs, activates expression of cytokines and growth factors and induces migration of mononuclear phagocytes [62]. A more recent study showed that AGEs modulate the maturation and function of dendritic cells in peripheral blood [84]. In particular, exposure of dendritic cells to AGEs resulted in inhibition of expression of maturation markers and a dose-dependent loss in their capacity to stimulate proliferation of T-cells. In addition, in contrast to soluble AGEs, AGEs located in basement membranes inhibited monocyte migration. However, other studies show that subendothelial AGEs can selectively enhance monocyte migration across an intact endothelial cell monolayer and that AGEs can induce monocyte CD147 (extracellular matrix metalloproteinase inducer - EMMPRIN) expression, an effect mediated by inflammatory pathways and RAGE [85, 86]. Furthermore, in non-diabetic patients with renal failure, AGE pentosidine is found to be associated with monocyte activation and, moreover, plasma levels of pentosidine and neopterin (a monocyte activation product) increased in parallel with the progression of renal failure [87].

AGEs are also involved in polymorphonuclear leukocyte dysfunction. Advanced glycation end-products depress superoxide production by stimulated polymorphonuclear leukocytes. As superoxide plays an essential role in bactericidal activity, polymorphonuclear leukocyte dysfunction may be a contributory factor to the increased prevalence and severity of bacterial infection seen in diabetic patients [88]. In addition, RAGE mediated neutrophil dysfunction is evoked by advanced glycation end-products [60]. In particular, AGE-modified albumin inhibits transendothelial migration and compromises the ability of neutrophils to kill ingested bacteria. This is the result of aberrant signal processing and altered neutrophil responses, when RAGE, which is expressed on the cell membrane of neutrophils, is engaged by AGEs.

Interestingly, activated phagocytes (neutrophils) are capable of generating CML-AGEs at sites of inflammation, through the heme-enzyme myeloperoxidase. This pathway is independent of hyperglycemia and might explain the presence of AGEs in atherosclerotic and inflammatory lesions found in non-diabetic patients [89].

AGEs in tumor biology:

Engagement of RAGE by AGEs activates signaling pathways that include mitogen activated protein kinases (MAPKs). MAPKs pathways are related to cell growth, proliferation and differentiation, mediate stress and apoptosis and play a central role in cancer biology [90]. Thus, depending on the extracellular signal and cell type, AGE-dependent activation of MAPKs pathways may promote or inhibit cell growth. This is effectively documented in a recent study of Ju Young Kim et al. [91]. This study investigated whether advanced glycation end products and RAGE are involved in the proliferation of leukemia cells (HEL cells). It was demonstrated that AGEs directly induced the proliferation of human acute myeloid leukemia cells and cell lines via the MAPKs, PI3K and JAK/STAT (janus kinase / signal transducer and activator of transcription) pathways. In contrast, AGEs did not affect the proliferation of normal mononuclear cells. This may suggest that AGEs act on leukemia cells selectively. Indeed, RAGE expression, both at the mRNA and protein level was identified in all leukemia cell lines. Additionally, it was observed that HEL cells exhibited a significant dose-dependent increase in the number of cells in the S phase of the cell cycle after exposure to AGEs and AGEs had modulatory effects on cell-cycle progression regulatory proteins CDKs (cyclin-dependent kinases). These results suggest that RAGE ligation by AGEs may induce the proliferation of HEL cells via the MAPKs, PI3K and JAK/STAT pathways, eventually leading to the activation of nuclear transcription factors, such as NF-kB and c-myc, and ultimately triggering the cell-cycle machinery.

Most aggressive tumors induce the secretion of angiogenesis factors. The angiogenic process incorporates endothelial cell activation, proliferation, migration, tube formation and capillary growth [92]. Among various angiogenic factors, the most notable is the VEGF, which exerts its mitogenic activity especially on endothelial cells. The genes of this potent angiogenic factor have a kB binding site and are regulated by activated NF-kB. Several studies show the correlation of advanced glycation end-products and angiogenesis, both in vitro and in vivo, and the AGE/NF-kB/VEGF axis seems to play an important role both in the pathogenesis of diabetic microangiopathy and the tumor-related angiogenesis [93-96].

DNA is susceptible to glycation by glyoxal (G) and methylglyoxal (MG), a process that leads to the formation of nucleotide advanced glycation end-products (nucleotide AGEs). The nucleotide most reactive under physiological conditions is deoxyguanosine (dG). Presence of nucleotide dG-G and dG-MG in DNA is associated with increased mutation frequency, DNA strand breaks and cytotoxicity. In contrast to the extensive literature dealing with AGE-modified proteins, little has been written on the formation and impact of DNA glycation on cell physiology and disease. Yet, the formation of AGE-DNA adducts may contribute to genetic instability, mutagenesis and increased risk for cancer development in patients with metabolic diseases [97].

In order to assess the potential contribution of DNA glycation to genetic instability, Tamae D et al. prepared shuttle vectors containing defined levels of the DNA glycation adduct N(2)-(1-carboxyethyl)-2'-deoxyguanosine (CEdG) and transfected them into human fibroblasts, which differed only in their capacity to conduct nucleotide excision repair (NER) [98]. In the NER-compromised fibroblasts, the frequency of induced mutation increased up to 18-fold relative to background, compared to the 5-fold increase observed at the highest adduct density in NER-competent cells. The authors underscored the fact that NER was the primary-if not exclusive mechanism for repair of this adduct in human fibroblasts. Consistent with predictions from biochemical studies using CEdG-substituted oligonucleotides, guanine transversions were the predominant mutation resulting from replication of MG-modified plasmids.

AGEs are markedly elevated in patients with chronic renal insufficiency, a disease which is related to both increased cardiovascular complications and enhanced cancer incidence. Stopper H. et al. investigated the potential effects of AGEs on DNA integrity in porcine renal tubular cells. Incubation of the porcine kidney cells with AGE- modified bovine serum albumin (AGE-BSA) - carboxymethyllysine-BSA as well as methylglyoxal-BSA resulted in a significant increase in DNA damage. Therefore, AGE-induced genotoxicity may be related to the enhanced cancer incidence observed in patients with chronic advanced renal failure [99].

Several defense mechanisms are recruited for the protection and recovery of DNA from glycation. These mechanisms include the suppression of nucleotide glycation by glyoxalase I, aldehyde reductases and dehydrogenases, as well as base excision repair. Overexpression of glyoxalase I has been observed in drug-resistant malignant cells and this may be an example of an undesirable effect of the enzymatic protection against DNA glycation. Glyoxalase I-mediated drug resistance was found in human leukemia and lung carcinoma cells. In addition, glyoxalase I overexpression has also been demonstrated in invasive ovarian cancer and breast cancer [100]. Consequently, both normal and malignant cells are vulnerable to DNA damage from AGEs; malignant cells are "aware" of this threat and utilize defense enzymatic mechanisms in order to protect their genome from glycation.

Glucose metabolism and cancer: a bioenergetic viewpoint in tumor biology

The theoretical basis of the correlation between glucose metabolism and tumor biology has its roots far back in the third decade of the 20th century. According to Warburg's original hypothesis tumors rely primarily on anaerobic glycolysis for energy production and exhibit higher glucose uptake (Warburg et al. 1924, Warburg 1956). The role of glucose metabolism in cancer was reviewed in detail by Rainer Wittig and Johannes F. Coy [101]. Glucose fermentation is linked to aggressiveness in cancers regardless of the origin and is diagnostically exploited by the utilization of [18F] fluoro-2-deoxyglycose-PET. Indeed, high glucose uptake by certain histological subtypes of non-small cell lung cancer (NSCLC) positively correlates with high Ki67 proliferation index scores and poor differentiation [102]. The metabolism of aggressive cancer cells is frequently dominated by consumption of large amounts of glucose that exceeds by 20 to 30 times the needs of normal cells. It is well documented that elevated fermentation of glucose and the consequent lactate overproduction confer selective advantages on cancer cells, i.e. they a) modulate the tumor microenvironment, b) promote the activation of matrix metalloproteases, c) suppress the proliferation and function of cytotoxic T-lymphocytes and d) correlate with the likelihood of distant metastases and poor survival. In addition, glucose fermentation and loss of oxidative ATP production have been linked to the activation of the AKT pathway, which is related to tumor resistance to common anticancer drugs, irrespective of their intracellular targets. The authors emphasize that several in vitro and in vivo studies have revealed the induction of resistance to therapy after the administration of glucocorticoids and express concerns about the frequently applied practice of glucocorticoid administration for the suppression of cancer cachexia [101].

AGEs are typically related to enhanced generation of ROS. Increased ROS generation reflects mitochondrial dysfunction and it is commonly related to decline in energy production (ATP). However, it is not yet clear whether increased ROS production is overall advantageous or detrimental to cancer cells. Intriguingly, ROS overproduction is paradoxically "normal" for some types of cells, such as activated macrophages [103], certain stem cells [104] and malignant cells, including lung cancer cells [105-109]. These cells are capable of utilizing ROS for their metabolic and functional needs, without being adversely affected by the toxic effects of ROS. In these cases, there is evidence that the cells share a common protective mechanism against ROS, which is the upregulation of the activity of endogenous antioxidant and redox buffering systems, such as superoxide dismutases (SODs), catalases and glutathione peroxidases.

H. Pelicano et al. in their review [110] not only provide data on the mechanisms of increased glycolysis in cancer cell metabolism, but also provide supporting evidence that inhibition of glycolysis may be used for anticancer treatment. These authors point out that one of the most prominent metabolic alterations in cancer cells is the increased rate of glycolysis and their dependency on glycolytic pathway (Embden-Meyerhof pathway) for ATP generation (Warburg effect). The production of ATP is much more efficiently achieved through oxidative phosphorylation, which yields 36 ATP from 1 molecule of glucose, than through glycolysis, which yields only 2 ATP per glucose. Consequently, even a minimal decline in mitochondrial respiration would require a substantial increase of glycolytic activity to compensate for cell energy needs. However, the mechanisms by which this metabolic alteration evolves during cancer development are more complex. These mechanisms include a) mitochondrial defects and malfunction in respiration and oxidative phosphorylation, b) the hypoxic environment in cancer tissues, which works as a strong modulator of energy metabolism that forces cancer cells to use glycolytic pathway for ATP synthesis, c) oncogenic signals that render cancer cells addictive to glycolysis for ATP production, such as the PI3k/AKT signaling pathway and Ras, Src and Bcr-Abl oncogenes and d) alterations of enzyme expression, such as increase of hexokinase II, which in turn promotes glycolysis.

The increased dependence of cancer cells on the glycolytic pathway for ATP synthesis provides a biochemical basis for the development of novel glycolytic inhibitors as a new class of anticancer agents. A well-known paradigm of a compound that is used in clinical praxis is the tyrosine kinase inhibitor Imatinib. This drug targets the Bcr-Abl oncogene that is responsible for the development of chronic myeloid leukemia. Imatinib decreases the activity of hexokinase and glucose-6-phosphate dehydrogenase in leukemia cells, leading to suppression of glycolysis and ATP production.

AGEs are involved in glucose metabolism and energy production derangement. The addition of AGEs to the cell culture in a neuroblastoma cell line experimental study [111] resulted in decreased cellular ATP levels, increased glucose consumption and increased lactate production. All of the AGE-induced metabolic derangements could be attenuated by antioxidants such as a-lipoic acid, 17β-estradiol, aminoguanidine, and pyruvate, which is an energy substrate and an antioxidant as well. In particular, when administered 1 hour before the addition of AGEs in the cell culture, all the antioxidants were able to restore, at least partially, the AGE-induced ATP depletion, suggesting that the reduction of ATP synthesis by AGEs involves the production of ROS, such as superoxide or hydrogen peroxide. AGE-related decrease in ATP production could be caused by impaired glucose flux due to inhibition of glycolytic enzymes, Krebs cycle, or mitochondrial function. Investigators of the same cell line model [111] measured the AGEs effects on glucose consumption and lactate production. At 100μM bovine serum albumin AGE (BSA-AGE), glucose consumption increased up to 5-fold, whereas lactate production was already increased at a concentration of 50μM BSA-AGE. Similarly, glucose consumption and lactate production were normalized by antioxidants. These data suggest that AGEs impair glucose flux through the Krebs cycle and oxidative phosphorylation, resulting in the conversion of pyruvate to lactate rather than to acetyl-CoA. They also indicate that the metabolic disturbances in glucose metabolism and the resulting shift to anaerobic glycolysis are mediated by oxidative free radicals.

In summary, the authors of this neuroblastoma cell line experimental study present a model of AGE-induced radical synthesis and impairment of energy production: binding of AGEs to RAGE results in increased production of ROS that inactivates ROS-sensitive enzymes (including those of the tricarboxylic acid cycle) and mitochondrial respiratory chain complexes. This leads to the impairment of glucose flux through the pyruvate dehydrogenase and mitochondrial respiratory chain, which results in ATP depletion and enhanced lactate and ROS overproduction.

Many of the AGE-induced changes in glucose metabolism have also been reported in diabetes, including increased plasma lactate and low intracellular ATP levels [112, 113]. In addition, AGE-induced metabolic changes such, as increased levels of lactic acid, have been reported in patients with Alzheimer's disease [114]. Numerous other in vitro and in vivo studies, most of them related to diabetes and its complications, link AGEs, usually through RAGE signaling, with major intracellular metabolic alterations, including ROS production, mitochondrial respiratory chain and electron transfer dysfunction and decrease in ATP production, thus creating an energy deficit for the cell [115-117]. AGEs and ROS major effects on cellular level are briefly illustrated in figure 1.

AGEs in particular cancer cells/tissues

Tumor metabolism is characterized by higher glucose uptake and increased anaerobic glycolysis in order to meet the tumor's energy needs. A consequence of increased glycolysis is the non-enzymatic glycation of proteins, lipids and nucleotides, which ultimately leads to the formation of AGE adducts. Seen in this light, tumors provide an environment that favors generation of AGEs. Van Heijst et al. studied the presence of AGEs in human cancer tissues [14] using two AGE-specific antibodies. The first one was directed against the glucose-derived AGE CML [Nε-(carboxymethyl)lysine], which is a major AGE-adduct derived from lysine glycation. The second one was directed against argpyrimidine, which is a methylglyoxal-arginine modification. Methylglyoxal is a dicarbonyl compound that is produced as a side product during glycolysis and is a precursor for particular forms of AGEs. Investigators examined the presence of CML-AGEs and argpyrimidine in four different types of human cancer tissues: squamous cell carcinoma of the larynx, breast and colon adenocarcinomas, and leiomyosarcoma. They observed the presence of both CML and argpyrimidine in all four types of tumors, although at different levels of expression. The highest expression of CML was found in colon adenocarcinomas and leiomyosarcomas, whereas the highest expression of argpyrimidine was found in squamous cell carcinomas of the larynx and adenocarcinomas of the breast. CML was present both in tumor cells and in tumor stroma, i.e. fibroblasts, macrophages and capillaries. The authors concluded that AGE-modified proteins may be involved in the biology of tumors. However, in a recent prospective case-cohort study, high serum levels of CML-AGEs were not related to increased risk of colorectal cancer among Finnish male smokers [118].

Melanoma is one of the most invasive and metastatic cancers with high mortality rates. AGEs, formed at an accelerated rate under oxidative stress might be involved in the growth and invasion of melanoma through interaction with RAGE. CML formation in sun-exposed areas of human skin with actinic elastosis is accelerated by UV-induced oxidation [24] and AGEs generate active oxygen species into the skin under UVA irradiation [119], suggesting the creation of a vicious cycle of AGEs formation. RAGE expression and the effects of AGEs on melanoma growth and migration were investigated in a both in vitro and in vivo experimental study with human melanoma cell lines and tumor-bearing mice [13]. RAGE expression, both at the protein and mRNA level was found to be weak in normal melanocytes, whereas it was higher both in the membrane and cytoplasm of melanoma cells and melanoma cell extractions (immunohistochemistry and western blot). RAGE blockade with anti-RAGE antibodies suppressed growth of implanted melanoma in immunocompetent mice. In tumor-bearing mice treated with anti-RAGE antibodies, survival rates were prolonged and spontaneous lung metastases were inhibited. Both glyceraldehyde and glycolaldehyde-derived AGEs significantly stimulated cell proliferation, migration and invasion of the melanoma cells in vitro. Anti-RAGE antibody inhibited proliferation of the melanoma cells induced by AGEs. Interestingly, other AGEs classes (glucose-derived, methylglyoxal-derived, glyoxal-derived and CML) failed to enhance melanoma cell proliferation, migration and invasion. Furthermore, investigators studied the expression of AGE and RAGE in human melanoma tissue specimens in vivo. RAGE was detected in the cytoplasm of melanoma cells, whereas it was scarcely detected in normal melanocytes. CML-AGEs and all five distinct AGEs reported in the study were present in the melanoma tumor specimens, both within the cytoplasm and the extracellular matrix, whereas they were barely detected in normal skin. The presence of AGEs in the tumor extracellular matrix suggests that not only tumor cells but also stromal cells located in the tumor microenvironment synthesize AGEs.

Similarly, in another experimental mice model, high doses of AGE-BSA (bovine serum albumin) were found to induce skin tumors through a mechanism which may be associated with oxidative stress and mutagenesis [120]. It is known that nucleotide bases also participate in advanced glycosylation reactions, producing DNA-linked AGEs that cause mutations and DNA transposition. CML (carboxymethyllysine) may enhance cancer progression by inducing DNA damage and glyoxal and methylglyoxal can cause covalent modifications of nucleic acids and histones, changes that are genotoxic and may induce carcinogenesis [25, 121].

Two other in vitro experimental studies show the correlation of AGEs and cancer cell biology in human acute myeloid leukemia cells [91] and in lung cancer cells [122]. The former has already been presented above, and the latter will be addressed immediately below.

AGEs and lung cancer

AGE-derived adducts derange cellular and extracellular structures and perturb cell energy production. Inherently connected to increased oxidative stress, AGEs are related to the pathogenesis of numerous and diverse pathological conditions, including cancer. The receptor for AGEs, RAGE, has been extensively studied for its implication in tumor biology, including lung cancer. The assumption that AGEs may be associated with the biology of lung cancer as well, would appear justified. This is also supported by the fact that tumors generally exhibit an increased glycolytic rate, which enhances the formation of AGEs. However, data regarding the potential effects of endogenous, food- and tobacco-derived AGEs on the development of lung cancer are fairly limited and conflicting.

The effects of AGEs on cell viability, migration and invasion of lung cancer A549 cells were examined in vitro [122]. Human lung adenocarcinoma A549 cells that express RAGE were cultured and supplemented with both unglycated BSA and glyceraldehyde-AGE-BSA. Glycer-AGEs were found to attenuate proliferation of A549 cells and decrease cell viability. However, glycer-AGEs-treated cells showed enhanced migration and greater invasion through the Matrigel matrix, as compared to unglycated control BSA-treated cells. Furthermore, activity of Rac1 was induced by glycer-AGEs. Activation of the small GTPases Rac1 and Cdc42 is known to induce actin cytoskeleton reorganization - an effect of major importance for RAGE-dependent cell migration [38, 123]. Thus, the first conclusion of the study was that glycer-AGEs enhanced the migration and invasion capacity of A549 lung cancer cells, rather than their proliferation. In addition to cell migration and invasion assays, investigators studied the effects of glycer-AGEs on matrix metalloproteinase 2 (MMP-2) activity. MMPs are commonly upregulated in tumor stroma and are considered as promoters of tumor invasion and metastasis [124]. MMP-2 in particular acts as a key enzyme for the degradation of type IV collagen, which is both a major component of the normal lung tissue and an important target for RAGE-mediated adhesion of alveolar cells (AT I cells) on basal lamina. Several studies show that AGEs induce the expression of CD147, which is involved in the regulation of matrix metalloproteinases (MMP) expression, including MMP-2 [85, 125] and is correlated with tumor progression in numerous malignant tumors, including lung cancer [126-128]. In this study, no significant increase in mRNA expression or the activated form of MMP-2 was observed after addition of glycer-AGEs. However, it must be noted that different types and molecular concentrations of AGEs exhibit different biological effects; thus conclusions from individual experimental studies cannot be generalized.

In contrast, Bartling et al. demonstrated that glyoxal, which is a dicarbonyl compound and highly reactive precursor of AGEs, actually decreased the activity of MMP-2 released from lung fibroblasts, in a dose-dependent manner [129]. Since the increased serum and tissue level of AGEs is a characteristic finding of older individuals, this may be linked with the less invasive behavior of tumors at very advanced age. This suggestion is in agreement with the results of another study on the effects of age-associated changes of extracellular matrix collagen on lung cancer cell migration [130]. AGEs load is commonly increased in aged tissues, including lung tissue, because of lifelong protein glycation [15, 130-132]. Although advanced age is related to increased incidence of lung cancer [1] elderly patients with lung cancer demonstrate slower tumor growth progression and less metastatic disease [133, 134]. Since extracellular matrix alterations contribute to cancer development, investigators studied the effects both of aged-fibrillar rat tail collagen and AGE-modified collagen on the invasiveness of two human lung cancer cells (H322 and H358) in vitro. The study revealed that migration of lung cancer cells through a matrix of collagen was reduced in old collagen assays compared with young ones. Similarly to old collagen, AGE-modified collagen reduced the invasiveness of lung cancer cells. This inhibitory effect was more prominent in H322 cells, which demonstrate a higher basic migration potential compared to H358 cells. Cancer cell proliferation was not affected in either aged-collagen or AGE-modified collagen. Further analyses showed that old and AGE-modified collagen adversely affect certain mechanisms of cell migration, including cell adhesion and proteolytic degradation of collagen by membrane-type matrix metalloproteinases (MT-MMPs). Adhesion of lung cancer cells to old-rat collagen, as well as to several types of AGE-modified collagen was less efficient. In addition, old and AGE-modified collagen proteolysis from MMPs was reduced and the degree of reduction was inversely correlated with AGEs level. Taken together, these data indicate that age-related alterations of extracellular matrix are related to a decreased invasive behavior of lung carcinoma cells, in which AGE-modified collagen may play a significant role.

AGE-related plasma fluorescence has been studied for its prognostic value on the outcome of NSCLC patients [135]. Seventy NSCLC patients after curative resection surgery were tested for plasma AGE-fluorescence. The study was retrospective and the prognostic value of AGE-related plasma fluorescence on tumor reoccurrence and five-year survival rate were investigated. The worst prognosis was associated with patients having a low level of AGE-related fluorescence, with a five-year survival of only 10%. The best five-year survival rates were demonstrated in patients with moderately high AGE-fluorescence, reaching 60%, who also exhibit a later tumor reoccurrence. However, this positive correlation was partially abolished in patients with very high AGE-fluorescence, with survival rates dropping to 35% (still higher than in the very low AGE-fluorescence group). Additional in vitro studies demonstrated that treating NSCLC cells with patient's plasma revealed an inverse correlation between the growth of the cell spheroids and the levels of AGEs plasma fluorescence. The impact of circulating levels of AGEs on NSCLC growth was confirmed in AGE-rich diet fed mice in vivo. Mice with higher levels of circulating AGEs developed smaller tumors in comparison with mice with normal AGE levels. This study demonstrated both the tumor growth modifying effect of particular levels of circulating AGEs and their possible role as prognostic markers in the outcome of NSCLC patients after curative surgery. However, the authors underline the pathophysiologic role of highly increased AGE levels in a) induction of oxidative stress, b) exhaustion of antioxidant systems and c) support of tumor progression. In addition, it is mentioned that in breast cancer, elevated plasma levels of protein carbonyls are associated with induced cancer progression. Factors that determine the final biological effects of AGEs in a particular context include a) the quantity and the particular forms of AGE variants, b) the metabolic profile of the patient, including glucose and lipid metabolism and insulin sensitivity c) the dietary habits, d) the balance between energy production, oxidative stress and antioxidant activity, e) the capacity of degradation and excretion of AGE compounds and f) the individual cells/tissues that are exposed to AGEs.

Lung cancer cell evasion of apoptosis may be triggered by AGE-modification of heat shock protein 27 (HSP27) [136]. The expression of two major AGEs, i.e. Nε-(carboxymethyl)lysine (CML) and argpyrimidine (a methylglyoxal-arginine adduct) was studied by immunohistochemistry on sections of squamous cell and adenocarcinoma NSCLC tissues. In addition, expression profile was investigated in both human lung squamous carcinoma cell line SW1573 and adenocarcinoma cell line H460. The study revealed a moderate to strong cytoplasmic CML staining of cancer cells in the squamous cell carcinoma and adenocarcinoma tissues. Tumor stroma, i.e. fibroblasts, macrophages and capillaries, also demonstrated a strong CML staining. In contrast, argpyrimidine staining was found to be strong only in the cytoplasm of the squamous cell carcinoma tissues. Argpyrimidine staining was weakly positive in adenocarcinomas and virtually absent in tumor stroma of both types of lung cancer tissues. Additional cell line experimental studies showed that argpyrimidine was predominantly found in the human lung squamous carcinoma cell line SW1573, whereas it was hardly found in the adenocarcinoma cell line H460. This weak argpyrimidine expression in lung adenocarcinoma was tumor specific. Furthermore, it may be lung specific as well, since other adenocarcinomas, mainly breast, strongly express argpyrimidine.

Heat shock proteins play an essential role in cell biology through their implication in multiple cellular functions, including immune responses, cell differentiation and cell apoptosis [137-139]. In this study it was demonstrated that human lung squamous cell carcinoma tissues strongly express heat shock protein 27 (Hsp27) which, in addition, is co-localized with argpyrimidine. The same strong Hsp27 staining was also seen in the adenocarcinoma tissues; although argpyrimidine staining was found to be weak, co-localization with argpyrimidine was still present. Consequently, co-localization of Hsp27 and argpyrimidine is characteristic for NSCLC tissues, independently of the staining intensity. This pattern was confirmed in the cell line immunostaining both of squamous carcinoma cell line SW1573 and adenocarcinoma cell line H460. Immunoprecipitation assays with a lysate of SW1573 cells showed that argpyrimidine completely co-immunoprecipitated with Hsp27, proving that Hsp27 is the major argpyrimidine-containing protein.

Investigators sought to determine whether the differences in argpyrimidine-modified Hsp27 levels between human lung squamous cell carcinomas and adenocarcinomas have any functional effect on the biology of these tumors. The correlation of expression of argpyrimidine-modified Hsp27 and active caspase-3 was examined. Caspase-3 belongs to a family of intracellular cysteine proteases involved in programmed cell death, proliferation and inflammation [140]. Activated caspase-3 downstreams cell apoptosis. A statistically significant difference of expression of active caspase-3 was detected between adenocarcinoma and squamous cell carcinoma tissues. Adenocacinoma tissues, which demonstrate low expression of argpyrimidine-modified Hsp27, showed a higher positivity for active caspase-3, whereas squamous cell tissues, with moderate to high expression of argpyrimidine-modified Hsp27, showed low positivity for active caspase-3. To establish a difference in caspase-3 activation between human lung squamous carcinoma SW1573 cells and adenocarcinoma H460 cells, investigators examined the response of these cells to cisplatin-induced apoptosis. The mechanism of cytotoxicity and cell death induced by cisplatin involves cytochrome-c dependent caspase activation. Cells were incubated with increasing concentrations of cisplatin. A strong activation of caspase-3 was detected in H460 cells, which are cells with low expression of argpyrimidine-modified Hsp27, whereas the opposite was found for SW1573 cells, in which activation of caspase-3 was very low. A further experimental assay confirmed this correlation: H460 cells were pre-treated with a glyoxalase I inhibitor, which increases intracellular methylglyoxal levels and consequent argpyrimidine modification. This time, cisplatin-induced caspase-3 activation was significantly reduced.

These results indicate that high expression of argpyrimidine-modified Hsp27 may be related to cancer cell evasion of apoptosis and development of resistance against cisplatin-induced caspase-3 activation.

The presence of CML-AGEs both in squamous cell and adenocarcinoma cancer cells and, furthermore, in tumor stroma of lung cancer tissues may influence cancer development through a number of diverse mechanisms. CML-modified proteins are capable of activating multiple signaling pathways related to upregulation of growth factors, activation of NF-kB and induction of ROS production. Moreover, AGEs adducts induce DNA damage and mutagenesis. In addition, AGEs may be implicated in the development of mechanisms by which cancer cells evade apoptosis and exhibit chemotherapy resistance. However, many in vitro and in vivo studies show that AGEs do not induce lung cancer cell proliferation. In addition, the presence of AGEs in the extracellular matrix may reduce cancer cell adhesion and invasion and may be related to a better outcome in patients already suffering from lung cancer.

RAGE in lung physiology and lung cancer

Ann Marie Schmidt et al. and Michael Neeper et al. were the first to identify the receptor for AGEs [17, 18]. They simultaneously published the results of their experimental studies that led to the identification of RAGE (July 25 1992). RAGE was isolated from bovine lung and found to be present on the endothelial cell surface.

RAGE is expressed only at low levels in the majority of tissues under normal conditions. In contrast, the expression of RAGE is highly enhanced in many pathological conditions, such as acute and chronic inflammatory diseases, diabetes, atherosclerosis, advanced kidney disease, heart failure, stroke, neurodegenerative diseases, and cancer. The multiligand/RAGE axis-dependent development of sustained inflammation seems to play a key role in the pathogenesis of these disparate diseases and it may represent the common causative mechanism that links them together. The reader is referred to analytical review articles for a more detailed reference to RAGE implication in the pathophysiology of numerous and heterogeneous diseases, including cancer [21, 37, 46, 47, 49].

Unlike other tissues, healthy lung expresses RAGE strongly. This "paradox" is suggestive of distinct roles for this receptor in lung tissue homeostasis. Studies have showed that RAGE possesses a number of important roles in lung physiology that include enhancement of cell spreading and adhesion to extracellular matrix, and diminished proliferative capacity of RAGE-expressing cells [21]. RAGE is primarily expressed and localized on alveolar epithelial type I cells (ATI). In fact, it is considered as a highly specific marker for human ATI cells, specifically located at the basolateral membrane [41, 141, 142]. The major contribution of RAGE in lung physiology is comprehensively described in the publication of Nina Demling et al. [41]. The lung alveolar epithelium comprises two cell types: ATI and ATII cells. ATII cells are cuboid in shape and, although more abundant in number, cover much less area of the alveolar surface. They are the primary site of surfactant synthesis and they have the capacity to proliferate and differentiate into ATI cells. In contrast, ATI cells are thin and squamous and cover more than 95% of the alveolar surface of the lung. They have the capacity neither for further differentiation nor for self-renewal. However, their thin and flat morphology makes them ideal for the formation of the functional surface that is involved in bidirectional gas exchange [143]. In their study Demling et al. demonstrated that RAGE enhances the adherence of alveolar epithelial cells to collagen-coated surfaces and strongly induces alveolar cell spreading. Their data suggested that RAGE enforces the adherence of ATI cells to the basal lamina and specifically to collagen IV, which is a major component of the alveolar basal lamina. The localization of RAGE at the basolateral membrane of ATI cells is in accordance with this suggestion. During transdifferentiation of ATII cells to ATI in vitro, RAGE protein expression and transcript levels become increased. Expression of RAGE promotes spreading of adherent cells seeded on collagen IV coated glass to a strikingly marked degree. The authors underline that the cells became so thin that they could only be detected by differential interference microscopy. Consequently, RAGE appears both to assist ATI cells to obtain a flat and extended morphology, in order to ensure effective gas exchange, and to induce alveolar stability, by strengthening the adherence of the cells to the alveolar basal membrane. Induction of cell spreading was entirely dependent on a coating with collagen which forms three dimensional fibril meshworks; however, the epitope in collagen to which RAGE binds is not known.

In the context of the developing lung, it has been shown that RAGE expression increases gradually from fetal to birth and to adult rat lungs, both at protein and mRNA level [144]. In particular, fetal, term, 4-day, 8-day and adult rat lungs were studied both with western blot and immunohistochemistry. A steady increase from fetal to adult lungs was demonstrated both in membrane and sRAGE protein expression. Receptor positive staining was predominant in type I pneumocytes (AT I cells). The study illustrated that the neonatal rat lung, which is not fully alveolarized, demonstrates low RAGE expression, whereas the post-natal upregulation of RAGE expression reflects ongoing alveolarization characterized by an expansion of type I epithelial cell population. These data and their interpretation are in complete agreement with the results of the study of Nina Demling et al. [41] according to which RAGE promotes cell adherence and spreading of alveolar epithelial cells and, furthermore, is a highly selective differentiation marker of human alveolar epithelial type I cells. However, whereas RAGE expression is required for normal alveolar formation, it is the accurate regulation of this expression which is a prerequisite condition for normal lung development. Indeed, mice overexpressing RAGE during embryogenesis, not only showed significant lung hypoplasia, but also exhibited 100% mortality rates [145].

The above findings indicate that RAGE possesses a central role in lung development and physiology, through its implication in lung alveolarization and alveolar epithelial cell differentiation and adhesion. Thus, a potential role of deregulation of RAGE expression in lung cancer appears likely. RAGE and RAGE-ligands are involved in cancer development and metastasis. Ligands such as S100/calgranulins and HMBG1 are expressed and secreted by cancer cells and lead to enhanced expression of cytokines and growth factors, activation of NF-kB, and increased cell migration [46, 49]. RAGE expression parallels tumor invasiveness and metastasis potential, whereas blockade of RAGE signaling attenuates these effects. While this is the case for many cancers, such as gastric, colon, pancreatic and prostate cancer, it is not for lung carcinomas. Buckley and Ehrhardt, in their review [21] highlight what is already known from previous studies, i.e. that, strangely, lung cancers, among the most invasive malignancies, express low levels of RAGE. In point of fact, in lung cancer, lower RAGE levels are related to increased tumor growth and invasiveness, whereas RAGE expression in cancer cells may be related to diminished tumor growth [20, 146]. Furthermore, RAGE genes may have a role in the diagnosis of lung neoplasms, via the discrimination of normal from malignant lung tissue [147, 148], so that soluble and endogenous secretory (sRAGE - esRAGE) isoform levels may serve as diagnostic and prognostic biomarkers for lung cancer [149, 150].

In 1997, Peter Schraml et al. investigated the differences of RAGE expression between nine paired normal lung and NSCLC tissues [151]. Their study demonstrated a strongly reduced or absent expression in NSCLC tissues, both at the transcriptional and protein level. Two more recent studies on RAGE implication in lung cancer development and progression come from Bartling et al. [20, 146]. In the first study [20] the aim was the quantification of expression of RAGE in normal human lung and NSCLC tissues. Lung specimens from tumor and paired non-tumor tissues of 34 NSCLC patients, who underwent pulmonary resection surgery, were included. Comparisons with different histological subtypes, tumor stage (TNM) and degree of differentiation were to be made. The study revealed that RAGE is markedly reduced at the mRNA level in NSCLC samples compared to normal lung. The reduction was similar in squamous cell lung carcinomas and adenocarcinomas. These differences were even more prominent at the protein level, suggesting further post-transcriptional regulation. Additionally, it was demonstrated that downregulation of RAGE correlates with higher tumor stages. In contrast, overexpression of full-length human RAGE in lung cancer cells (NCI-H358) was related to diminished tumor growth.

However, benign lung neoplasms (hamartomas) were also characterized by reduced RAGE levels. In addition, metastatic lesions originating from distant tumors (kidney and colon) clearly showed reduction of RAGE expression compared to individual lung specimens.

Another part of the study was the investigation of the impact of RAGE on the proliferation of lung cancer cells. In vitro experiments were performed with different cell lines, all of which showed low mRNA and protein RAGE expression. In addition, NCI-H358 cells were chosen for overexpression of RAGE. Overexpression of RAGE resulted in diminished proliferation of NCI-H358 cells in monolayer cultures in vitro. Initial indications were therefore that RAGE suppresses lung tumor growth. However, in three-dimensional spheroid cell cultures, the differences between RAGE-transfected cells and mock controls (control NCI-358 cells) were not significant. Interestingly, ΔcytoRAGE-transfected NCI-H358 cells, i.e. cells lacking the cytoplasmic domain of RAGE and, therefore, blockaded RAGE signaling, demonstrated increased tumor growth in comparison to RAGE-transfected cells. These data suggest that RAGE re-expression in cancer cells does not necessary impair tumor growth, and that RAGE signaling encompasses factors which are present in normal NCI-H358 cancer cells, but are neutralized in mutant ΔcytoRAGE cells that lack the cytoplasmic domain of RAGE. Furthermore, it is clearly demonstrated that in vitro monolayer cell studies may be not be representative of biological processes occurring in vivo and, in fact, may be misleading.

In the same experimental study NCI-H358 cells were cultured on collagen layers. The morphology of the RAGE-expressing cells and localization of RAGE were observed. A spreading, epithelial-like growth of both RAGE and ΔcytoRAGE-expressing cells was observed, which was not seen with regard to NCI-H358 control cells. In particular, RAGE-deficient lung cancer cells exhibited a multicellular-complexed proliferation pattern. Localization of RAGE was determined by immunocytochemistry, which revealed strong membrane localization in single cells, and redistribution of RAGE towards intercellular contact sites at higher cell densities. Interestingly, ΔcytoRAGE-expressing cells, although lacking the cytosolic domain, exhibit preferential localization of the truncated receptor at cell-cell contact sites as well. Thus, formation of intercellular contacts requires either the full-length or the truncated receptor and does not require RAGE signaling. Localization of RAGE at cell-cell contacts, as well as the spreading acquired morphology and growth of cells expressing the receptor, imply that downregulation of RAGE may be considered a critical step in tissue reorganization during tumor development. As concerns cell migration, overexpression of RAGE did not induce cell migration in vitro, whereas ΔcytoRAGE transfection did. This observation is consistent with the finding that both metastatic tissue (from other primary cancers) and benign tumors demonstrate reduced RAGE expression. Thus, activation of RAGE per se does not contribute to tumor cell migration and metastases.

The authors' basic conclusion is that down-regulation of RAGE in human NSCLC may contribute to a loss of normal epithelial structure organization and differentiation and concomitant oncogenic transformation without restriction of cancer cell proliferation. Furthermore, in patients with NSCLC downregulation of RAGE correlates well with higher tumor stages. Another interesting point is that cell localization of RAGE is independent of the intracellular domain and there may be RAGE-dependent effects exclusively mediated by the extracellular region.

The second study of Bartling et al. was based on the well-established knowledge that cancer formation and progression is regulated by the tumor microenvironment [146]. Tumor stroma consists predominantly of fibroblasts, extracellular matrix, immune-inflammatory cells, and blood vessels. In this context, fibroblasts are involved both as tumor stroma factors and as mesenchymal cells capable of producing and releasing growth factors and cytokines. The main aim of the study was to investigate the RAGE-dependent impact of fibroblasts on tumor cell growth. They observed that cultivation of human lung cancer cells H358 (a cell line that corresponds to NSCLC) with lung fibroblasts WI38 augmented the proliferation of cancer cells, both in monolayer and spheroid culture models. By contrast, when lung cancer cells overexpressed full-length RAGE, the proliferative stimulus of fibroblasts was reduced. Fibroblast-induced cancer cell proliferation was associated with increased activation of particular MAP kinases (p42/44) in lung cancer cells. Other consequences of the presence of fibroblasts were reduced spontaneous cell death and mitochondrial depolarization, independently of the expression of RAGE.

In this study, the behavior of ΔcytoRAGE-expressing cells was investigated as well. Curiously, these cells often tended to a higher stimulation of p44/42 MAPKs (ERK1/2), and, moreover, in three-dimensional cultures they formed the largest spheroids in response to fibroblasts impact, thus behaving like mock-transfected cells.

The authors state that stromal fibroblasts of malignant tumors exhibit a divergent phenotype, which supports tumor growth and propagation. In fact, from the early s