Value of 18F-FDG PET/CT in discrimination between indolent and aggressive non-Hodgkin's lymphoma: A study of 328 patients.

OBJECTIVE: Non-Hodgkin's lymphoma (NHL) cases with inconclusive biopsy findings are not infrequently referred for fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT). We searched for maximum standardized uptake value (SUVmax) cut-off values that could discriminate between indolent and aggressive NHL in conventional non-time of flight (non-TOF) 18F-FDG PET/CT and TOF 18F-FDG PET/CT. SUBJECTS AND METHODS: Retrospectively, 328 patients were selected by the following inclusion criteria: biopsy-proven NHL with no more than one histopathological type; new cases with less than 90 days between obtaining biopsy and 18F-FDG PET/CT scanning; recurrent cases with time interval more than six months since the last therapy with no history of transformation; and blood glucose less than 150mg/dL. Two hundred forty six (246) selected patients were scanned with non-TOF PET/CT, and 82 patients were scanned with TOF 18F-FDG PET/CT. RESULTS: The SUVmax of NHL tends to be higher in TOF 18F-FDG PET/CT than non-TOF 18F-FDG PET/CT. New aggressive NHL had significantly higher SUVmax than new indolent NHL in both, non-TOF 18F-FDG PET/CT (13.6±7.7g/mL vs. 5.3±3.4g/mL, P<0.0001) and TOF 18F-FDG PET/CT (20.5±9.8g/mL vs. 6.6±4.7g/mL, P<0.0001). A receiver operating characteristic curve analysis for new cases in non-TOF 18F-FDG PET/CT (n=204), demonstrated SUVmax of 10g/mL as the most balanced cut-off between aggressive and indolent NHL, with the area under the curve (AUC) of 86%, specificity of 94%, and sensitivity of 71%. While SUVmax of 13g/mL was the most balanced cut-off for new cases in TOF 18F-FDG PET/CT (n=57), with AUC of 91%, specificity of 95.5%, and sensitivity of 77%. CONCLUSION: Both SUVmax>10g/mL in non-TOF 18F-FDG PET/CT and >13g/mL in TOF 18F-FDG PET/CT were highly suggestive of an aggressive nature of NHL, while there was an overlap between indolent and aggressive NHL in the lower SUVmax levels.

Metabolic alterations in bladder cancer: applications for cancer imaging.

Treatment planning, outcome and prognosis are strongly related to the adequate tumor staging for bladder cancer (BC). Unfortunately, a large discrepancy exists between the preoperative clinical and final pathologic staging. Therefore, an advanced imaging-based technique is crucial for adequate staging. Although Magnetic Resonance Imaging (MRI) is currently the best in vivo imaging technique for BC staging because of its excellent soft-tissue contrast and absence of ionizing radiation it lacks cancer-specificity. Tumor-specific positron emission tomography (PET), which is based on the Warburg effect (preferential uptake of glucose by cancer cells), exploits the radioactively-labeled glucose analogs, i.e., FDG. Although FDG-PET is highly cancer specific, it lacks resolution and contrast quality comparable with MRI. Chemical Exchange Saturation Transfer (CEST) MRI enables the detection of low concentrations of metabolites containing protons. BC is an attractive target for glucose CEST MRI because, in addition to the typical systemic administration, glucose might also be directly applied into the bladder to reduce toxicity-related complications. As a first stage of the development of a contrast-specific BC imaging technique we have studied glucose uptake by bladder epithelial cells and have observed that glucose is, indeed, consumed by BC cells with higher intensity than by non-transformed urothelial cells. This effect might be partly explained by increased expression of glucose transporters GLUT1 and GLUT3 in transformed cells as compared to normal urothelium. We also detected higher lactate production by BC cells which is another cancer-specific manifestation of the Warburg effect. In addition, we have observed other metabolic alterations in BC cells as compared to non-transformed cells: in particular, increased pyruvate synthesis. When glucose was substituted by glutamine in culture media, preferential uptake of glutamine by BC cells was observed. The preferential uptake of glucose by BC cells gives an opportunity to develop NMR based imaging procedures where glucose or its derivatives can serve as a contrasting agent. In addition, metabolic alterations observed in BC cells could provide the basis for development of new anti-cancer therapeutics.

Genotoxic and cytotoxic effects of the environmental pollutant 3-nitrobenzanthrone on bladder cancer cells.

3-Nitrobenzanthrone (3-NBA), a potential human carcinogen, is present in diesel exhaust. The main metabolite of 3-NBA, 3-aminobenzanthrone, was detected in urine of miners occupationally exposed to diesel emissions. Environmental and occupational factors play an important role in development of bladder cancer (BC), one of the most frequent malignancies. It is expected that exposure of urothelium to 3-NBA and its metabolites may induce BC initiation and/or progression. To test this hypothesis, we studied geno- and cytotoxicity of 3-NBA using an in vitro BC model. 3-NBA induced higher levels of DNA adducts, reactive oxygen species and DNA breaks in aggressive T24 cells than in more differentiated RT4 cells. To understand the nature of this difference we examined the role of several enzymes that were identified as 3-NBA bio activators. However, the difference in DNA adduct formation cannot be directly linked to the different activity of any of the examined enzymes. Conversely, the difference of tested cell lines in p53 status can partly explain the distinct levels of 3-NBA-DNA adducts and DNA damage induced by 3-NBA. Therefore, we assume that more aggressive T24 cells are more predisposed for DNA adduct formation, DNA damage and, possibly, mutations and as a result further tumorigenesis.

Clostridium perfringens enterotoxin as a potential drug for intravesical treatment of bladder cancer.

The current intravesical treatment of bladder cancer (BC) is limited to a few chemotherapeutics that show imperfect effectiveness and are associated with some serious complications. Thus, there is an urgent need for alternative therapies, especially for patients with high-risk non-muscle invasive (NMIBC). Clostridium perfringens enterotoxin (CPE), cytolytic protein binds to its receptors: claudin 3 and 4 that are expressed in epithelial cells. This binding is followed by rapid cell death. Claudin 4 is present in several epithelial tissue including bladder urothelium and its expression is elevated in some forms of BC. In addition to directly targeting BC cells, binding of CPE to claudins increases urothelium permeability that creates conditions for better accession of the tumor. Therefore, we evaluated CPE as a candidate for intravesical treatment of BC using a cellular model. We examined cytotoxicity of CPE against BC cells lines and 3D cultures of cells derived from surgical samples. To better elucidate cellular mechanisms, activated by CPE and to consider the use of CPE non-toxic fragment (C-CPE) for combination treatment with other drugs we synthesized C-CPE, compared its cytotoxic activity with CPE and examined claudin 4 expression and intracellular localization after C-CPE treatment. CPE induced cell death after 1 h in low aggressive RT4 cells, in moderately aggressive 5637 cells and in the primary 3D cultures of BC cells derived from NMIBC. Conversely, non-transformed urothelial cells and cells derived from highly aggressive tumor (T24) survived this treatment. The reason for this resistance to CPE might be the lower expression of CLDNs or their inaccessibility for CPE in these cells. C-CPE treatment for 48 h did not affect cell viability in tested cells, but declined expression of CLDN4 in RT4 cells. C-CPE increased sensitivity of RT4 cells to Mitommycin C and Dasatinib. To better understand mechanisms of this effect we examined expression and phosphorylation status of EphA2 and Src after C-CPE treatment and found changes in expression and phosphorylated status of these regulatory molecules. These observations show that after additional preclinical studies CPE and C-CPE in combinations with other drugs can be considered as a potential modalities for intravesical treatment of BC because of its ability to effectively destroy BC cells expressing claudin 4 and low toxicity against normal urothelium.

Aristolochic acid-induced apoptosis and G2 cell cycle arrest depends on ROS generation and MAP kinases activation.

Ingestion of aristolochic acids (AAs) contained in herbal remedies results in a renal disease and, frequently, urothelial malignancy. The genotoxicity of AA in renal cells, including mutagenic DNA adducts formation, is well documented. However, the mechanisms of AA-induced tubular atrophy and renal fibrosis are largely unknown. To better elucidate some aspects of this process, we studied cell cycle distribution and cell survival of renal epithelial cells treated with AAI at low and high doses. A low dose of AA induces cell cycle arrest in G2/M phase via activation of DNA damage checkpoint pathway ATM-Chk2-p53-p21. DNA damage signaling pathway is activated more likely via increased production of reactive oxygen species (ROS) caused by AA treatment then via DNA damage induced directly by AA. Higher AA concentration induced cell death partly via apoptosis. Since mitogen-activated protein kinases play an important role in cell survival, death and cell cycle progression, we assayed their function in AA-treated renal tubular epithelial cells. ERK1/2 and p38 but not JNK were activated in cells treated with AA. In addition, pharmacological inhibition of ERK1/2 and p38 as well as suppression of ROS generation with N-acetyl-L-cysteine resulted in the partial relief of cells from G2/M checkpoint and a decline of apoptosis level. Cell cycle arrest may be a mechanism for DNA repair, cell survival and reprogramming of epithelial cells to the fibroblast type. An apoptosis of renal epithelial cells at higher AA dose might be necessary to provide space for newly reprogrammed fibrotic cells.

Circular STAG2 RNA Modulates Bladder Cancer Progression via miR-145-5p/TAGLN2 and Is Considered as a Biomarker for Recurrence.

The current study aimed to elucidate the regulatory mechanisms of the circRNA hsa_circ_0139697 (circSTAG2(16-25)) in BCa and to consider the opportunity of using circSTAG2(16-25) isolated from BCa patient urine as a marker for disease development prediction. The selection of this circRNA was determined by the special role of its parental gene STAG2 in BCa biology. The circRNA hsa_circ_0139697 was chosen from 25 STAG2 circRNAs due to its differential expression in the urine of BCa patients and healthy volunteers. Higher levels of circSTAG2(16-25) were detected in urine samples obtained from patients with recurrent tumors. A higher expression of circSTAG2(16-25) was also detected in more tumorigenic BCa cell lines. The overexpression of circSTAG2(16-25) in BCa cells induced the elevation of proliferation, motility, and invasion. To study the mechanisms of circSTAG2(16-25) activity, we confirmed that circSTAG2(16-25) can bind miR-145-5p in vitro as was predicted by bioinformatic search. miR-145-5p was shown to suppress some genes that promoted BCa progression. One of these genes, TAGLN2, encodes the protein Transgelin 2, which plays a role in BCa cell motility and invasion. Therefore, the possible mechanism of action of circSTAG2(16-25) could be sponging the tumor suppressor miR-145-5p, which results in activation of TAGLN2. In addition, circSTAG2(16-25) might be considered as a potential biomarker for recurrence prediction.

A fluorescence-based analysis of aristolochic acid-derived DNA adducts.

Aristolochic acids (AAs), major components of plant extracts from Aristolochia species, form (after metabolic activation) pro-mutagenic DNA adducts in renal tissue. The DNA adducts can be used as biomarkers for studies of AA toxicity. Identification of these adducts is a complicated and time-consuming procedure. We present here a fast, nonisotopic, fluorescence-based assay for the detection of AA-DNA adducts in multiple samples. This approach allows analysis of AA adducts in synthetic DNA with known nucleotide composition and analysis of DNA adducts formed from chemically diverse AAs in vitro. The method can be applied to compare AA-DNA adduct formation in cells and tissues.

Glutamate dehydrogenase requirement for apoptosis induced by aristolochic acid in renal tubular epithelial cells.

Ingestion of aristolochic acids (AA) contained in herbal remedies results in a renal disease and, frequently, urothelial malignancy. The genotoxicity of AA in renal cells, including mutagenic DNA adduct formation, is well-documented. However, the mechanisms of AA-induced tubular atrophy and renal fibrosis are largely unknown. Epithelial cell death is a critical characteristic of these pathological conditions. To elucidate the mechanisms of AA-induced cytotoxicity, we explored AA-interacting proteins in tubular epithelial cells (TEC). We found that AA interacts with a mitochondrial enzyme glutamate dehydrogenase (GDH) and moderately inhibits its activity. We report that AA induces cell death in GDH-knockdown TEC preferentially via non-apoptotic means, whereas in GDH-positive cells, death was executed by both the non-apoptotic and apoptotic mechanisms. Apoptosis is an energy-reliant process and demands higher adenosine 5'-triphosphate (ATP) consumption than does the non-apoptotic cell death. We found that, after AAI treatment, the ATP depletion is more pronounced in GDH-knockdown cells. When we reduced ATP in TEC cells by inhibition of glycolysis and mitochondrial respiration, cell death mode switched from apoptosis and necrosis to necrosis only. In addition, in cells incubated at low glucose and no glutamine conditions, oxaloacetate and pyruvate reduced AAI-induced apoptosis our data suggest that AAI-GDH interactions in TEC are critical for the induction of apoptosis by direct inhibition of GDH activity. AA binding may also induce changes in GDH conformation and promote interactions with other molecules or impair signaling by GDH metabolic products, leading to apoptosis.

Organoid model of urothelial cancer: establishment and applications for bladder cancer research.

3D cancer cell models are suitable for drug evaluation because they more precisely mimic tissue architecture than 2D cultures. To study cytotoxicity of anticancer agents, the most sensitive CellTiter-Glo 3D assay is used. However, this is an end point assay, so it is not possible to consider the variance of the starting material amount in the final reading. It is difficult to maintain an even plating density of 3D organoids for cytotoxicity analysis. We present a simple, 3D bladder cancer culture that can be maintained, cryopreserved and used for molecular and drug response studies. We applied a simple modification of the drug response assay for 3D cultures by measuring the background signal with the CellTiter Blue assay before drug application.

Evaluation of the total distribution volume of 18F-FBPA in normal tissues of healthy volunteers by non-compartmental kinetic modeling.

OBJECTIVE: Boron neutron capture therapy (BNCT) is a noninvasive radiation therapy method for cancer treatment. In BNCT, 4-borono-2-[18F]-fluoro-L-phenylalanine (18F-FBPA) PET has been employed to estimate 10B accumulation in target tumors and normal tissues if 10B borono-L-phenylalanine (10B-BPA) is used as a boron carrier. The purpose of the current study was to evaluate the total distribution volume (Vt) of 18F-FBPA in normal organs of healthy volunteers by kinetic analysis and to estimate boron concentration in normal organs for the therapeutic dose of 10B-BPA using obtained Vt values. METHODS: Six healthy volunteers were injected with 18F-FBPA (3-5 MBq/kg), and 7 PET-CT scans were performed subsequently. 18F-FBPA radioactivity in whole blood and plasma was measured before, and eight times after the injection. PET images were analyzed by PMOD software. Twelve volumetric regions of interest including the brain, heart, right lung, spleen, liver, parotid salivary glands, esophagus, stomach, pancreas, intestines, and bone marrow were drawn manually for each subject and analyzed with the Logan plot and two Ichise multilinear analyses (MA1 and MA2). The better model was defined by several goodness-of-fit parameters and residual distribution. After Vt values had been derived, boron concentration was estimated in ppm for the 10B-BPA-fructose (10B-BPA-fr) dose 30 g 1 and 2 h post-injection using Vt and interpolated plasma activity data. RESULTS: The Ichise MA2 model showed the best fit among all models. Akaike Information Criterion (AIC) was the lowest for the Ichise's MA2 in all regions (mean AIC value - 14.0) comparing to the other models (Logan plot mean AIC 31.4; Ichise MA1 model mean AIC - 4.2). Mean Vt values of the Ichise MA2 model ranged from 0.94 ± 0.14 ml/ml in the pancreas to 0.16 ± 0.02 ml/ml in the right lung. Estimated boron concentration for 10B-BPA-fr had the highest value in the pancreas (14.0 ± 1.9 ppm 1 h after, and 5.7 ± 1.7 ppm 2 h after the 18F-FBPA administration) and the lowest value in the right lung (2.4 ± 0.3 ppm 1 h, and 1.0 ± 0.3 ppm 2 h post-injection). CONCLUSION: The 10B concentration in normal tissues was best estimated using Vt values of 18F-FBPA with the Ichise multilinear analysis 2 (MA2). TRAIL REGISTRY: The UMIN clinical trial number: UMIN000022850.

Evaluation of D-isomer of 18F-FBPA for oncology PET focusing on the differentiation of glioma and inflammation.

OBJECTIVES: L-4-borono-2-18F-fluoro-phenylalanine (L-[18F]FBPA), a substrate of L-type amino acid transporter 1 (LAT1), is a tumor-specific probe used in positron emission tomography (PET). On the other hand, it has not been examined whether another isomer D-[18F]FBPA accumulates specifically in the tumor. Here, we compared the accumulation of D-[18F]FBPA in C6 glioma and inflammation to evaluate the performance of D-[18F]FBPA as a tumor-specific probe. METHODS: HEK293-LAT1 and HEK293-LAT2 cells were tested for [14C]-leucine or [14C]-alanine transport, and IC50 values of L- and D-FBPA were evaluated in both cell types. PET was conducted in rat xenograft model of C6 glioma with LAT1 expression and model of turpentine oil-induced subcutaneous inflammation (n=10 for both models). The concentrations of D-[18F]FBPA were compared between glioma and inflammatory lesion using standardized uptake value (SUV). RESULTS: In contrast to L-FBPA, which inhibited substrate uptake in both HEK293-LAT1 and -LAT2 cells, D-FBPA showed no inhibitory effect on both cells, suggesting low transporter selectivity of D-[18F]FBPA against LAT1 and LAT2. Static PET analysis showed low accumulation of D-[18F]FBPA in C6 glioma and inflammatory lesion (SUVmax=0.80±0.16, 0.56±0.09, respectively). Although there was a statistical difference in SUVmax between these tissues, it was difficult to distinguish glioma from inflammation on the PET image due to its low uptake level. Therefore, it was suggested that D-[18F]FBPA is not a suitable tumor-specific probe for oncology PET in contrast to L-[18F]FBPA. CONCLUSION: This study demonstrated that D-[18F]FBPA is not a LAT1-specific PET probe and shows low uptake in C6 glioma, indicating its unsuitability as a tumor diagnosis PET probe.

Preserved Cerebral Oxygen Metabolism in Astrocytic Dysfunction: A Combination Study of 15O-Gas PET with 14C-Acetate Autoradiography.

Fluorocitrate (FC) is a specific metabolic inhibitor of the tricarboxylic acid (TCA) cycle in astrocytes. The purpose of this study was to evaluate whether inhibition of the astrocyte TCA cycle by FC would affect the oxygen metabolism in the rat brain. At 4 h after the intracranial FC injection, the rats (n = 9) were investigated by 15O-labeled gas PET to measure the cerebral blood flow (CBF), the cerebral metabolic rate of oxygen (CMRO2), oxygen extraction fraction (OEF), and cerebral blood volume (CBV). After the 15O-gas PET, the rats were given an intravenous injection of 14C-acetate for autoradiography. 15O-gas PET showed no significant differences in any of the measured parameters between the ipsilateral and contralateral striatum (high dose group: CBF (54.4 ± 8.8 and 55.3 ± 11.6 mL/100mL/min), CMRO2 (7.0 ± 0.9 and 7.1 ± 1.2 mL/100mL/min), OEF (72.0 ± 8.9 and 70.8 ± 8.2%), and CBV (4.1 ± 0.8 and 4.2 ± 0.9 mL/100mL), respectively). In contrast, the 14C-acetate autoradiography revealed a significant inhibition of the astrocyte metabolism in the ipsilateral striatum. The regional cerebral oxygen consumption as well as the hemodynamic parameters were maintained even in the face of inhibition of the astrocyte TCA cycle metabolism in the rat brain.

Protein misfolding and aggregation in neurodegenerative diseases: a review of pathogeneses, novel detection strategies, and potential therapeutics.

Protein folding is a complex, multisystem process characterized by heavy molecular and cellular footprints. Chaperone machinery enables proper protein folding and stable conformation. Other pathways concomitant with the protein folding process include transcription, translation, post-translational modifications, degradation through the ubiquitin-proteasome system, and autophagy. As such, the folding process can go awry in several different ways. The pathogenic basis behind most neurodegenerative diseases is that the disruption of protein homeostasis (i.e. proteostasis) at any level will eventually lead to protein misfolding. Misfolded proteins often aggregate and accumulate to trigger neurotoxicity through cellular stress pathways and consequently cause neurodegenerative diseases. The manifestation of a disease is usually dependent on the specific brain region that the neurotoxicity affects. Neurodegenerative diseases are age-associated, and their incidence is expected to rise as humans continue to live longer and pursue a greater life expectancy. We presently review the sequelae of protein misfolding and aggregation, as well as the role of these phenomena in several neurodegenerative diseases including Alzheimer's disease, Huntington's disease, amyotrophic lateral sclerosis, Parkinson's disease, transmissible spongiform encephalopathies, and spinocerebellar ataxia. Strategies for treatment and therapy are also conferred with respect to impairing, inhibiting, or reversing protein misfolding.

Preliminary feasibility study on differential diagnosis between radiation-induced cerebral necrosis and recurrent brain tumor by means of [18F]fluoro-borono-phenylalanine PET/CT.

OBJECTIVES: A previous study reported that a differential diagnosis between glioblastoma progression and radiation necrosis by 4-borono-2-[18F]-fluoro-phenylalanine ([18F]FBPA) PET can be made based on lesion-to-normal ratio of [18F]FBPA accumulation. Two-dimensional data acquisition mode PET alone system, with in-plane resolution of 7.9 mm and axial resolution of 13.9 mm, was used. In the current study, we aimed to confirm the differential diagnostic capability of [18F]FBPA PET/CT with higher PET spatial resolution by three-dimensional visual inspection and by measuring mean standardized uptake value (SUVmean), maximum SUV (SUVmax), metabolic tumor volume (MTV), and total lesion (TL) [18F]FBPA uptake. METHODS: Twelve patients of glioma (9), malignant meningioma (1), hemangiopericytoma (1), and metastatic brain tumor (1) were enrolled. All had preceding radiotherapy. High-resolution three-dimensional data acquisition mode PET/CT with in-plane resolution of 4.07 mm and axial resolution of 5.41 mm was employed for imaging. Images were three-dimensionally analyzed using the PMOD software. SUVmean and SUVmax of lesion and normal brain were measured. Lesion MTV and TL FBPA uptake were calculated. The diagnostic accuracy of [18F]FBPA PET/CT in detecting recurrence (n = 6) or necrosis (n = 6) was verified by clinical follow-up. RESULTS: All parameters showed significantly higher values for tumor recurrence than for necrosis. SUVmean in recurrence was 2.95 ± 0.84 vs 1.18 ± 0.24 in necrosis (P = 0.014); SUVmax in recurrence was 4.63 ± 1.23 vs 1.93 ± 0.44 in necrosis (P = 0.014); MTV in recurrence was 44.92 ± 28.93 mL vs 10.66 ± 8.46 mL in necrosis (P = 0.032); and mean TL FBPA uptake in recurrence was 121.01 ± 50.48 g vs 12.36 ± 9.70 g in necrosis (P = 0.0029). CONCLUSION: In this preliminary feasibility study, we confirmed the possibility of differentiating tumor recurrence from radiation necrosis in patients with irradiated brain tumors by [18F]FBPA PET/CT using indices of SUVmean, SUVmax, MTV, and TL 18FBPA uptake.

Prostate cancer cell adhesion to bone marrow endothelium: the role of prostate-specific antigen.

Bone metastasis is the most frequent complication of prostate cancer (PC). Elucidation of the biological basis of this specificity is required for the development of approaches for metastatic inhibition. We investigated the possibility that the preferential attachment of PC cells to bone marrow endothelium (as opposed to endothelium from other organs) affects this specificity. We selected, from peptide phage-displayed libraries, peptide ligands to surfaces of PC cells (C4-2B) attenuated (30-40%) binding of C4-2B cells to bone marrow endothelial cells (BMECs). We then determined the molecules on the surface of C4-2B cells interacted with the selected peptides using column affinity chromatography and a cDNA expression phage-displayed library generated from C4-2B cells in T7 phage. We identified a phage from the cDNA library that specifically bound to one of the selected peptides-L11. This phage displayed the amino acid sequence homologous for the COOH-terminal portion of prostate-specific antigen (PSA). To examine the possible direct involvement of PSA in the interactions between PC and BMECs, we performed a cell-cell adhesion assay. Antibodies to PSA attenuated PC cells adhesion to BMECs. In addition, exogenous proteolytically active PSA modulated this adhesion. Finally, inactivation of mRNA coding PSA by a small interfering RNA (siRNA) diminished C4-2B cell adhesion to BMECs. These results indicate that PSA expressed as secreted and surface-associated molecules in C4-2B cells is involved in cell-cell interactions and/or digests components of bone marrow endothelium for preferential adhesion and penetration of PC cells. The suggested experimental approach is a promising strategy for identification of cell surface molecules involved in intercellular interactions.

Phage display selection and evaluation of cancer drug targets.

Techniques for the construction of phage display libraries of combinatorial proteins have dramatically improved. This has allowed researchers to expand the applications to the field of cancer biology. The most direct use of protein phage-displayed libraries is the selection of ligands for individual proteins. This includes identification of peptide ligands for receptor signaling molecules: integrins, cytokine and growth factor receptors. Selected peptides may be used as competitors for natural ligands and for the mapping of binding epitopes. This approach has been exploited for delineation of intracellular signal transduction pathways and for the selection of enzyme substrates and inhibitors. Recently, more complicated biological systems were used as targets for biopanning. This includes combination of soluble proteins, cellular surfaces and even the vasculature of whole organs. cDNA expression libraries in phage-based vectors have been recently introduced. The use of phage as a vector for targeted gene therapy is also considered. These and other applications of phage display for cancer research will be reviewed.

Phage matrix for isolation of glioma cell membrane proteins.

Cell-binding ligands for RG2 rat glioma were identified in our recent study from a library of peptides that are displayed as fusion molecules on phage particles. Here, one of the phage clones was used to affinity purify those cell membrane components to which the displayed peptides bind. This phage clone, displaying the ELRGDSLP peptide, was shown to recognize glioma cells specifically in comparison to control phage-expressing peptides of either similar or irrelevant sequences. Blocking experiments with synthetic RGDS peptide demonstrated that the phage-glioma cell recognition occurs via the RGD motif known to be present in many integrin-binding proteins. To form an affinity matrix that would bind to glioma cell membrane molecules, ELRGDSLP phage particles were cross-linked using dextran polymer. Whole cell lysate from RG2 rat glioma cells was passed through the matrix, resulting in the isolation of cell membrane components having strong affinity to the peptides on phage and molecules associated with those components. One of the isolated proteins was found to be CD44s, a cell surface adhesion molecule involved in glioma cell invasion and migration, which likely formed a complex with an RGD-binding integrin. Cell membrane proteins isolated with this innovative approach could be used for the design of cell-specific anticancer treatments.

Osteopontin traffic in hypoxic renal epithelial cells.

Osteopontin (OPN), a secretory RGD-containing phosphoprotein, is induced in acute renal injury where it plays a renoprotective role. To investigate in depth the mode of OPN secretion under stress conditions, we analyzed OPN traffic in human renal proximal tubular epithelial cells (RPTEC). Western blot analysis and fluorescence microscopy revealed trace amounts of OPN in intact cells, whereas cytoplasmic OPN levels were significantly increased after 24-48 h hypoxia. Immunoelectron microscopy of RPTEC showed predominantly apical localization of gold-labeled OPN under normal conditions. Hypoxia (24 h) increased 2.5-fold immunodetectable gold-labeled OPN at the apical plasma membrane; further reoxygenation (2 h) augmented apical and basolateral labeling 2- and 10-fold, respectively. Analysis of apical and basolateral medium conditioned by RPTEC grown on semipermeable membranes using a specially developed ELISA showed a global decrease in secreted OPN after hypoxia, which recovered following 2 h reoxygenation. Agents known to disrupt the function of the Golgi apparatus (brefeldin A, monensin) or actin cytoskeleton (cytochalasin B) significantly inhibited OPN-GFP secretion in normoxic cells. In cells recovering from hypoxia, however, OPN secretion required functional Golgi apparatus, but was not affected by cytochalasin B. These findings demonstrate that stress inhibits OPN secretion by the process dependent on the functional Golgi apparatus and actin cytoskeleton; recovery restores OPN secretion, although its polarity may become perturbed.

A claudin 3 and claudin 4-targeted Clostridium perfringens protoxin is selectively cytotoxic to PSA-producing prostate cancer cells.

Prostate cancer is the second leading cause of non-cutaneous cancer-related death in males, and effective strategies for treatment of metastatic disease are currently limited. The tight junction proteins, claudin 3 and claudin 4, serve as cell-surface receptors for the pore-forming Clostridium perfringens enterotoxin [CPE]. Most prostate cancer cells overexpress claudin 3 and claudin 4, and claudins are aberrantly distributed over the plasma membrane, making these cells particularly sensitive to cytolysis by CPE. Prostate cancer cells secrete PSA locally that is proteolytically active; however, circulating PSA is inactivated via binding to protease inhibitors. To overcome systemic toxicity of CPE, a modified protoxin was constructed with a tethered ligand attached to the C-terminus connected by a flexible linker containing a PSA-specific protease cleavage site. This engineered protoxin selectively and efficiently lyses PSA-producing prostate cancer cells whereas CLDN3 and CLDN4 positive cells that do not express PSA are resistant to cytolysis.

VEGF-induced mobilization of caveolae and increase in permeability of endothelial cells.

Glomerular epithelial cells (GEC) are a known site of vascular endothelial growth factor (VEGF) production. We established immortalized rat GEC, which retained the ability to produce VEGF. The isoforms expressed by GEC were defined as VEGF-205, -188, -120, and -164. The electrical resistance of endothelial cells cultured on GEC-conditioned matrix, an indicator of the permeability of monolayers to solutes, was significantly increased by the treatment with the neutralizing polyclonal antibodies to VEGF and decreased by VEGF-165. Transfection of endothelial cells with green fluorescence protein-caveolin construct and intravital confocal microscopy showed that VEGF results in a rapid appearance of transcellular elongated structures decorated with caveolin. Transmission electron microscopy of endothelial cells showed that caveolae undergo rapid internalization and fusion 30 min after application of VEGF-165. Later (36 h), endothelial cells pretreated with VEGF developed fenestrae and showed a decrease in electrical resistance. Immunoelectron microscopy of glomeruli confirmed VEGF localization to podocytes and in the basement membrane. In summary, immortalized GEC retain the ability to synthesize VEGF. Matrix-deposited and soluble VEGF leads to the enhancement of caveolae expression, their fission and fusion, formation of elongated caveolin-decorated structures, and eventual formation of fenestrae, both responsible for the increase in endothelial permeability.