DLDH, DERIVATIVES THEREOF AND FORMULATIONS COMPRISING SAME FOR USE IN MEDICINE

Abstract

The invention provides methods for treating a proliferative disease or disorder such as cancer by administering to patients suffering from the disease or disorder DLDH or a derivative thereof.

Claims

1.-25. (canceled)

26. A pharmaceutical composition comprising dihydrolipoamide dehydrogenase (DLDH), or a DLDH-based material, for use in medicine, wherein the DLDH-based material is selected from the group consisting of a. DLDH associated with at least one peptide, b. DLDH complex with at least one metal or metal oxide, and c. DLDH associated with at least one peptide as a complex with at least oen metal or metal oxide.

27. (canceled)

28. The pharmaceutical composition according to claim 26, for the treatment of a proliferative disease or disorder.

29. A peptide-modified material comprising DLDH and at least one integrin binding domain.

30. The peptide-modified material according to claim 29, wherein the integrin binding domain is arginine-glycine-aspartic acid (RGD).

31. The peptide-modified material according to claim 29, being associated with at least one metal or metal oxide.

32. The peptide-modified material according to claim 30, wherein the at least one RGD is associated with DLDH through at least one of the N terminus of the DLDH and the C terminus of the DLDH.

33. The peptide-modified material according to claim 30, being in the form of DLDH-RGD.sub.2.

34. A pharmaceutical composition comprising a peptide-modified material according to claim 29.

35. A method for treatment or prevention of a disease or disorder, the method comprising administering to a subject in need thereof an effective amount of DLDH or a DLDH-based material, wherein the DLDH-based material is selected from a. DLDH-associated with at least one peptide, b. DLDH complex with at least one metal or metal oxide, and c. DLDH associated with at least one peptide as a complex with at least one metal or metal oxide.

36. A photodynamic therapeutic method comprising administering to a subject in need thereof an effective amount of a photo-reactive DLDH-based material selected from DLDH-RGD, DLDH-TiO.sub.2 and DLDH-RGD.sub.2-TiO.sub.2, and irradiating said subject or a region of the subject's body in order to render active said photo-reactive material.

37. A composition comprising a peptide comprising at least one CHED motif and a pharmaceutically acceptable carrier.

38. The composition according to claim 37, wherein the CHED motif is associated with a metal or a metal oxide.

39. The method of claim 35, wherein the DLDA-based material comprises DLDH complex with TiO.sub.2.

40. The method of claim 35, wherein the DLDA-based material comprises DLDH associated with RGD.sub.2 and complexed with TiO.sub.2.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0093] In order to better understand the subject matter that is disclosed herein and to exemplify how it may be carried out in practice, embodiments will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:

[0094] FIGS. 1A-Dpresent ESEM micrograms of: FIG. 1AAnatase on metal plates; FIG. 1BAnatase on glass covered TitanShield nano-tubes (NT); FIG. 1CAnatase nano-particles (NP); FIG. 1D- Rutile on metal plates; FIG. 1ERutile micro-particles (P); and FIG. 1FP25-NP.

[0095] FIGS. 2A-Bpresent ROS production by TiO.sub.2 anatse NP: FIG. 2Apresents ROS production by different amounts of the oxide after 30 min of UVA illumination. FIG. 2Bpresents ROS production at different time intervals of UVA illumination by a constant amount of TiO.sub.2 anatse (0.4 mg/ml).

[0096] FIGS. 3A-Bpresent ROS production by DLDH-RGD.sub.2. FIG. 3A presents ROS production by 50 nM DLDH-RGD.sub.2 after 30 min at UVA illumination or in the dark, in presence or absence of the substrates dl-dihydrolipoamide (DHL, (0.5 mM) and NAD.sup.+ analog acetylpyridine adenine dinucleotide (AcPyAD, 0.9 mM). FIG. 3B presents the time-dependent ROS production by increasing concentrations of DLDH-RGD.sub.2.

[0097] FIG. 4presents a binding curve for DLDH-RGD.sub.2 to TiO.sub.2 anatase nano particles (NP). Dotted line represents a non-linear regression curve (based on saturated bindind model, Prism-GraphPad).

[0098] FIGS. 5A-Bshow the binding of DLDH-RGD.sub.2 to TiO.sub.2. TiO.sub.2 rutile discs were placed in polystyrene plates and coated with DLDH-RGD.sub.2 by 1 h incubation in 0.1M sodium bicarbonate buffer, pH 7.5, at room temperature. After washing, bound DLDH was detected by reaction with specific anti DLDH-antibodies followed by reaction with secondary gold labeled (12 nm) anti-IgG antibodies (FIG. 5A). FIG. 5B shows a control not containing DLDH.

[0099] FIGS. 6A-C present the effect of L-carboxylic acids on DLDH binding to TiO.sub.2. FIG. 6A shows the binding of glutamate (black) and aspartate (grey) to TiO.sub.2 NP (solid) and to DLDH-RGD.sub.2-coated TiO.sub.2 NP (striped).A sample of each amino acid (1.0 mg-ml.sup.1 in 1.0 ml of sodium bicarbonate buffer, pH7.5, containing 1.0M NaCl) was incubated for 1 h at room temperature with 30 mg of intact or DLDH-RGD.sub.2-coated TiO.sub.2 nanoparticles. Binding of the amino acid to TiO.sub.2 was determined by a fluorescamine assay of samples withdrawn from the supernatants at various time intervals. FIG. 6B shows DLDH-RGD.sub.2 binding to Glutamate-coated particles. DLDH (300 g in the same buffer) was incubated with glutamate-coated TiO.sub.2 nanoparticles (30 mg). FIG. 6C shows the inhibiting effect of glutamate (black) or aspartate (grey) on DLDH-RGD.sub.2 binding to TiO.sub.2 NP.

[0100] FIG. 7presents enzyme activities of soluble (DLDH-RGD.sub.2, squares) and TiO.sub.2-adsorbed (TiO.sub.2-DLDH-RGD.sub.2, circles). Initial velocities of substrate reduction by the enzyme in solution were monitored by OD.sub.363.

[0101] FIGS. 8A-Cshow the time-dependent release of DLDH-RGD.sub.2 from the complex with TiO.sub.2 upon UVA illumination in vitro experiments. FIG. 8A shows in vitro results. FIG. 8B shows a similar effect in B 16F10 cells. The cells (10.sup.5 cells per well) were incubated with the complex and illuminated with UVA for up to 60 min The protein was prestained with fluorescamine and was monitored in the cells using confocal microscopy with Leica SPS. A & B are at Dark conditions, C & D are after UVA illumination; A & CNo protein added; B & D8 M represent the complex prepared with 80 g/ml protein+11.6 mg/ml TiO.sub.2. FIG. 8C shows quantitation by mean fluorescence intensity of wells B/60 (black, control) and D/60 (grey) of. FIG. 8B.

[0102] FIG. 9 shows the binding of DLDH-RGD.sub.2 to various metal oxides in 0.1M ammonium bicarbonate buffer, pH 8.0, in the presence (solid lines) of 3.0M NaCl or in its absence (dashed lines). PZC values are shown.

[0103] FIGS. 10A-B show the effect of fluorescamine-labelling on DLDH activity. FIG. 10A shows the enzyme activity of DLDH-RGD.sub.2 at different substrate concentrations. FIG. 10B shows the enzyme activity of the fluorescently-labeled protein at different substrate concentrations.

[0104] FIGS. 11A-B show TiO.sub.2DLDH-RGD.sub.2 effect on HeLa morphology before/after UVC, 10 min FIG. 11A and FIG. 11B show results obtained in the dark. FIG. 11A and FIG. 11B show results obtained under UV illumination. FIG. 11B and FIG. 11D show results obtained in the presence of RGD.sub.2-DLDH. FIG. 11A and FIG. 11C in the absence of RGD.sub.2-DLDH.

[0105] FIG. 12 TiO.sub.2DLDH-RGD.sub.2 effects on HeLa survival before/after UVC, 10 min. FIG. 12A shows % survival of the cells after UVC illumination or in the dark, in the presence or absence of RGD.sub.2-DLDH. FIG. 12B and FIG. 12C shows FACS analysis of the cells described in FIG. 12A under UVC illumination (FIG. 12C) or in the dark (FIG. 12B).

[0106] FIGS. 13A-C show the cell survival after UVA illumination. FIG. 13A shows a diagram of the setup for the ROS assay with anatase NT-covered net FIG. 13B shows the cell survival of cells on anatase NT -covered netDLDH-RGD.sub.2 and Hela survivalUVA, 1 h FIG. 13C shows FACS analysis.

[0107] FIGS. 14A-B: FIG. 14A shows cell survival (HEK293normal, cervicalHela) with/without UVA illumination (1 h, 365 nm) after 48 h incubation. FIG. 14B shows cell membrane v.sub.3 integrin expirations by FACS analysis.

[0108] FIG. 15 shows dose dependence of B 16F10 death in presence or absence of TiO.sub.2 tested by Confocal Microscopy (Leica SP5). The cell nuclei are stained by Draq5 (Far red).

[0109] FIG. 16 shows the cytotoxic effect on B16F10 cells of the TiO.sub.2/protein complex at different ratios after 24 h, with/without 1 h of UVA illumination, by Leica SP5.

[0110] FIGS. 17A-C show B 16F10 cells incubated with different concentrations of DLDH-RGD.sub.2 after 48 h of incubation. FIG. 17A show microscopic images of CM FIG. 17B shows FACS analyses of early and late apoptosis markers and FIG. 17C shows FACS analysis of cell Cycle.

[0111] FIGS. 18A-E show FACS analysis of 3 different human melanoma cell lines (A375, WM368, WM3314) and mouse melanoma (B16F10) treated with different concentrations of DLDH-RGD.sub.2. FIG. 18A shows % surviving cell. FIG. 18B shows Early apoptosis. FIG. 18C shows Necrosis. FIG. 18D shows Late apoptosis. FIG. 18E shows shows cell membrane v3 integrin expirations by FACS analysis.

[0112] FIG. 19A shows the cytotoxic effect on B 1 6F10 cells of the DLDH-RDG.sub.2 during time (min), obtained by confocal microscopy by Leica SP5. White arrows point to the cytoplasm, yellow to the nuclei. FIGS. 19B-E show the cytotoxic effect of DLDH-RDG.sub.2 during time (min) on B16F10 cells by following stained internal markersNuclei, DLDH, PI and PSIVA.

[0113] FIG. 20A shows the cytotoxic effect of DLDH-RGD.sub.2 on cancer and normal cells by LSM510. FIG. 20B shows cell membrane v.sub.3 integrin expirations by FACS analysis.

[0114] FIG. 21 shows the dependence of DLDH-RGD.sub.2 cytotoxicity on RGD. B16F10 cells were incubated with 1 M of either fluorescamine-labelled DLDH-RGD.sub.2 (FIG. 21A) or DLDH (FIG. 21B). FIG. 21 C shows solvent control. The confocal images were taken with Leica SP5.

[0115] FIGS. 22A-B show the inhibitory effect of free RGD on fluorescamine-labeled DLDH-RGD.sub.2 penetration to HEK293B.sub.3 cells. FIG. 22A shows fluorescamine-labelled DLDH-RGD.sub.2 and FIG. 21B shows fluorescamine-labelled DLDH-RGD.sub.2+free RGD.

[0116] FIGS. 23A-E show the incubation of OVCAR 3 (100,000 per well) with DLDH-RGD.sub.2 (FIG. 23A), DLDH (FIG. 23B), ovalbumin (FIG. 23C), glycine (FIG. 23D) and fluorescamine alone (4 ug/0.25 ml of DMEM without FCS, FIG. 23E). The time (in sec) since addition of the protein to the cells is indicated in each figure. FIG. 23F shows cell membrane v3 integrin expirations by FACS analysis.

[0117] FIGS. 24A-B show the degradation of ds-DNA phage by DLDH-RGD.sub.2. Composition of each lane is shown in the table below. FIG. 24Ashows fragmented DNA and FIG. 24B shows intact DNA (0.15 ug-ml.sup.1 each). Substrates are DHL (0.5 mM) and NAD analog AcPyAD (0.9 mM). The buffer used was 0.1M sodium bicarbonate, pH7.4.

TABLE-US-00001 TABLE 1 Lane(s) DNA DLDH-RGD.sub.2 Substrates 1 + 2, 7 + + 3, 10 + 1 M + 4 1 M 5 10 M 6, 11 + 10 M + 8 + 10 M 9 + 1 M 12

DETAILED DESCRIPTION OF EMBODIMENTS

[0118] Dihydrolipoamide dehydrogenase (DLDH) is a homodimeric mitochondrial flavin-dependent oxidoreductase enzyme. It comprises an essential constituent of the 2-oxo acid dehydrogenase cycles which convert 2-oxo acids to the corresponding acyl-CoA derivatives. It catalyzes the NAD.sup.+-dependent oxidation of dihydrolipoic acid (or amide) into lipoic acid (or amide). Interestingly, NAD.sup.+ dependent metabolic and signaling pathways are highly altered in cancer cells. The enzyme's activity is critical for energy and redox balance in the cell and is often associated with elevated levels of Reactive Oxygen Species (ROS) production. Additionally, bioinformatics analysis indicates that the protein possess sequence and structural homology with the apoptosis-inducing factor (AIF), a central player in apoptotic death. These characteristics make DLDH a relevant potential anti-cancer molecule for the selective cytotoxicity of cancer cells. However, involvement of DLDH in cytotoxicity has never been reported.

[0119] Integrins are a family of cell surface receptors which are over-expressed on all tumor vascular cells and an array of cancer types. Twenty-four integrin heterodimers are currently identified and formed by the combination of at least 18 -subunits and 8 -subunits. These integrin receptors play a key role in the cross-talk between the cell and its surrounding stroma, binding to ECM ligands, cell surface ligands and soluble ligands. A subset of these 24 integrins, including v.sub.3 integrin, interact with proteins of the extra cellular matrix through an RGD recognition site and offer a docking site for endothelial cells, inflammatory cells and cancer cells. v.sub.3 plays a pivotal role in cancer pathogenesis and is intensively studied. Besides its mechanical function, this integrin is a true signaling molecule which participates in activation of cell migration, survival and angiogenesis and communicates with an array of growth factor receptors in cancer, such as tyrosine kinas to trigger tumorigenesis. The high expression of this integrin, to such an extent that new imaging approaches based on its expression are now under development is of clinical significance as correlation with tumor progression in several cancer types has been documented. Therefore utilizing RGD-recognition integrins, such as v.sub.3, are attractive and rational targets for cancer treatment.

[0120] The inventors have bio-engineered a recombinant human DLDH with tails of the integrin binding domain, arginine-glycine-aspartic acid (RGD), on both ends of the molecule (DLDH-RGD.sub.2) and generated a protein capable of specifically targeting integrin expressing cancer cells.

[0121] Titanium (Ti) is a powerful biocompatible material which is extensively being used in medical biomaterials applications e.g. in implantation. The oxide layer formed on the surface upon exposure to air, is important for protein adsorption which is considered to be an initial step in induction of differentiation of bone cells. Of the three naturally occurring crystallographic forms of titanium dioxide (TiO.sub.2), anatase, rutile and brookite the former possesses highest photocatalytically activity which results from its higher hydrophilicity. Its high photoreactivity, physical stability and commercial availability as well as its low toxicity make TiO.sub.2 the material of choice as a detoxifier which destroys cells, bacteria and organic toxic materials which is widely used for biomedical treatments including the destruction of cancer cells. The cytotoxic effect derived from ROS production of upon photo-excitation of TiO.sub.2 by far UV, has been examined in several cancer models in vitro. However, serious damage to the surrounding healthy cells limits the applicability of this method. Thus, developing a technique that will achieve TiO.sub.2 photo-oxidative effect at the visible or near UV (>351 nm) range as well as delivering the TiO.sub.2 selectively to the cancer cells are desired. When TiO.sub.2 is encountered with a human tissue it is rapidly covered with plasma and extra cellular matrix (ECM) proteins which strongly affect the biorecognition process. The adherence of TiO.sub.2 to most proteins is dominated by weak, reversible, electrostatic or hydrophobic bonds which often results in failure in achieving strong attachment of cells and tissues.

[0122] Previously, a cell wall protein capable of strong adherence to coal fly ash (CFA) and TiO.sub.2 particles, expressed solely during the late logarithmic phase of growth, has been isolated in our laboratory from the marine actinobacterium Rhodococcus rubber GIN1 and designated TiBP. The protein/oxide interaction occurred at high salt concentrations and its release from the oxide required high concentrations of SDS/urea, indicating non-electrostatic mechanism of binding, presumably via coordinative bonds. This is in contrast to most proteins that bind TiO.sub.2 via relatively weak charge related interactions. Peptide mapping and sequencing revealed that TiBP is an exocellular form of DLDH. Docking analysis experiments performed by our group have led to identification of a putative TiO.sub.2 binding site (CHED motif) on the protein molecule.

[0123] Conventional therapeutic strategy in cancer is based on drugs that increase ROS generation and induce apoptotic cell death. These ROS moieties have been shown to selectively affect cancer cells but protect normal cells from ischemic damage. The novel approach disclosed herein for cancer therapy is based on combining the independent photoreactive ROS production capability of TiO.sub.2 with that of DLDH and, together with the targeting effect of the RGD tails, to produce a potent and selective anti-cancer effect. For the proof of concept the inventors have studied each of the components of the complex, individually or combined, in three cancer cell models: ovarian cancer, cervical cancer and cutanous melanoma, all over-expressing RGD-recognizing integrins, such as v.sub.3. DLDH itself is proposed to serve as a novel therapy and that illumination of the complex (DLDH-RGD.sub.2-TiO.sub.2) will produce high synergistic ROS activity and cell death and may serve as a neo-radiation targeted treatment in cancer.

[0124] It is further suggested that short metal-oxide motifs identified within the DLDH sequence, may be used to achieve a short peptide that can maintain the titanium oxide binding capabilities with high affinity and may also be effective in producing ROS and cancer cell death upon illumination.

[0125] Experimental

[0126] Materials

[0127] All reagents used in this study were of analytical grade and, unless otherwise specified, were purchased from Merck (Darmstadt, Germany) or Sigma-Aldrich (St. Louise, Mich.). Several TiO.sub.2 forms were used as carriers in the present study: (1) anatase nano- and microparticles (Sigma, 23 and >100 nm, respectively) (2) TitanShield Colloid solution which form 8 nm particle net on glass disks and (3) oxidizedrutile, prepared by 5 h heating at 850 C. and anatase, prepared by 2 h heating at 450 C. (FIG. 1). The crystallographic forms and the metal composition of these preparations were confirmed by XRD, EDS and XRF (data not shown).

[0128] DLDH Cloning

[0129] Calcium chloride treated competent Escherichia coli BL21 (DE3) cells harboring the expression vector were aerated at 37 C. in Terrific Broth media, supplemented with 25 mg-ml.sup.1 kanamycin for 16 h. The cells were harvested, resuspended in 50 mM sodium phosphate buffer, pH7.5 and sonicated in the presence of 10 mg-ml.sup.l DNase E. and protease inhibitor cocktail (Sigma-Aldrich, St. Louise, Miss.). After sonication it was centrifuged (20,000 g for 30 min at 4 C.) and the supernatant fluid was collected and used for further purification. Recovery of the native form of the enzyme in the soluble fraction was indicated by its yellow color originating from the FAD prostatic group, as well as its enzymatic activity.

[0130] Lysozyme and diaphorase were purchased from Sigma-Aldrich. A pET28b-WT-DLD plasmid carrying the human dldh gene encoding DLDH (UniProt P09623), excluding the N-terminal 1-35 signal peptide region and containing an N terminal His.sub.6 tag (kindly provided by Prof. Grazia Isaya from the Mayo clinic college, Rochester, Minn.) was transformed into competent E. coli BL21 cells.

[0131] Protein Purification

[0132] The expressed His-tagged protein was isolated by immobilized metal affinity chromatography (IMAC). The supernatant of the cell extract was loaded onto a Fast-Ni Column (5 ml, GE Healthcare, Upsalla), connected to Akta Chromatographic System (GE Healthcare, Upsalla). The column was washed with the washing buffer (50 mM potassium phosphate buffer, pH 6.0, containing 300 mM NaCl, 10% glycerol and 20 mM imidazole) at a flow rate of 2 ml-min.sup.1, until no protein was detected by OD.sub.280. The His-tagged protein was then eluted with the elution buffer (0.5M imidazole in washing buffer, pH 6.0). The fractions were pooled and dialyzed against 0.1M sodium bicarbonate buffer, pH 7.5, at 4 C. for 16 h. The pooled protein fraction was further purified by gel filtration chromatography on a Superdex 200 column (30010 mm, Akta Chromatographic System, GE Healthcare, Upsalla) collecting 1 ml fractions at a flow rate of 1 ml-min.sup.1 and analyzed by SDS-PAGE (12%) as commonly used. The gels were stained with Coomassie Blue R250. A Precision Plus Protein Standards Dual Color (Bio-Rad) markers mixture was used.

[0133] Determination of Protein and Peptide Concentrations

[0134] Protein concentrations were determined by absorbance at 280 nm using extinction coefficients of 0.479 calculated from the amino acid compositions of DLDH-RGD.sub.2 (by the Expasy ProtParam application http://web.expasy.org/protparam/) or by the Bradford assay as commonly used

[0135] Oxide-Binding Assays

[0136] TiO.sub.2 binding activity was determined by incubation of DLDH-RGD.sub.2 or lysozyme with TiO.sub.2 (Anatase) nanoparticles at a ratio of 10-15 g protein per mg of beads in the indicated buffer. After agitation for 1 h at room temperature, the beads were sedimented by centrifugation in an Eppendorf (Hamburg, Germany) centrifuge for 15 min at 11000 g. The concentration of the non-adsorbed protein in the supernatant was determined as mentioned above.

[0137] Binding of aspartate and glutamate to native TiO.sub.2 or to DLDH-RGD2-coated TiO.sub.2 was determined by incubation for 1 h at room temperature of the particular amino acid (1.0 mg-mL.sup.1) with 30 mg of TiO.sub.2 or TiO.sub.2-DLDH-RGD2 (Anatase) nanoparticles in 1.0 mL of sodium bicarbonate buffer, pH 7.5, containing 1.0M NaCl. Binding of amino acids to the TiO.sub.2 particles was determined by fluorescamine assay of samples withdrawn from the supernatants at various time intervals.

[0138] Enzyme Activity Assays

[0139] The substrate dl-dihydrolipoamide (DHL) was prepared by reduction of dl-lipoamide with sodium borohydride. Briefly, a suspension of 200 mg dl-lipoamide in 4 ml methanol and 1.0 ml of 2 distilled water, was cooled to 0 C. and stirred while dripping a cold solution (1 ml) of sodium borohydride (200 mg-ml.sup.1 in 2 distilled water) until the solution became clear and colorless. The solution was then acidified with dilute hydrochloric acid to pH2 and extracted with 5 ml chloroform. The chloroform extract was dried and evaporated in a desiccator. The residual material was crystallized from hexane/benzene (1:2.5). The product was recovered by centrifugation for 5 minutes at 4 C. (3800 g) using a Heraeus Megafuge 1.0 centrifuge (ThermoFisher Scientific Inc., Waltham, Mass.). The precipitate was air-dried, and stored at 20 C. Before used, the substrate was dissolved in acetone at a concentration of 120 mM.

[0140] DLDH-RGD.sub.2

[0141] The reaction buffer contained 0.9 mM of the NAD.sup.+ analog acetylpyridine adenine dinucleotide (AcPyAD) in 2.5 mM sodium phosphate buffer, pH 7.6 containing 1.0 mM EDTA and the substrate DLH (0.1-0.5 mM). The reaction was initiated by addition of DLDH-RGD.sub.2 (10 l of 2 mg-ml.sup.1) into 1 ml of reaction buffer. Reduction rate of AcPyAD was continuously monitored by the absorbance at 363 nm. Activity was expressed as product produced (mM-min.sup.1), based on an extinction coefficient of a 9.110.sup.3M.sup.1 cm.sup.1 of AcPyAD.

[0142] TiO.sub.2-DLDH-RGD.sub.2

[0143] Bound DLDH-RGD.sub.2 was prepared by binding samples of 20 g of DLDH-RGD.sub.2 to 10 mg of TiO.sub.2 (Anatase) nanoparticles (100% bound) as described above. Prior to the experiment the particles were thoroughly washed with the assay buffer, excluding the substrate DHL, to remove any unbound enzyme. 10 mg of beads were mixed with 1.0 ml assay buffer, excluding the DHL substrate. Activity of the oxide-bound enzyme was determined under the above described conditions for the enzyme in solution. Since continuous, on-line monitoring was not possible due to the suspension turbidity, a discontinuous assay was applied as follows: The beads were agitated with 1.0 ml of the reaction buffer (excluding DHL) for 1 min and then sedimented by short centrifugation. The absorbance of the supernatant at 363 nm was monitored and used as reference. The beads were then re-suspended in the same buffer and the enzymatic reaction was initiated by addition of DHL (0.05-0.5 mM). After 1 min agitation the beads were sedimented and the absorbance at 363 nm of the supernatant was measured again. This process was repeated 5 times. Only the incubation times of the beads with the reaction mixture were taken into account in the activity assay. Incubation of the DLDH-RGD2 carrying beads in the reaction mixture without DHL served as a reference. Activity was defined as the OD.sub.363-min.sup.1 between the absorbance measurement and the former one.

[0144] ROS Generation Assay

[0145] The detection of ROS generated by A-NP activity is based on the reduction of Fe.sup.+3 to Fe.sup.+2-cytochrome C. This test was performed under UVA illumination for 30 min After incubation, the absorption of the liquid measured by spectrophotometer at a wavelength of 550 nm (E.sub.M 550 nm=2.110.sup.4 M.sup.1 cm.sup.1).

[0146] ROS generation assay in vitro was carried out by cytochrome C reduction.

[0147] Photocatalysis Assay

[0148] Determination of the photocatalytic effect was made by photo-degradation of Methylene Blue and Degradation of Methyelene Blue by spectrum colorimetric assay.

[0149] The Cytotoxicity Effect

[0150] The different complex components, separately and combined, examined in vitro at various concentrations in three cancer cell models (ovarian cancer, cervical cancer and cutaneous melanoma) in cultures (24 wells) as well as in control cells (integrin positive; CV-1 cells and integrin negative; HEK293 cells) in the presence/absence of a selected optimal protocol of illumination and will be assessed for: Cell viability (WST-1, ELISA), Absolute cell number (FACS), Cell cycle (PI, FACS) and Cell death (Annexin-PI, FACS). Confocal microscopy (LSM 510, Leica SP5) used to analyze the interaction of the biocomplex with the cells and its internalization process.

[0151] DyesDraq5, Heuaecst, Fluorescamin, Annexin, PI, Psiva

[0152] Fluorescamine Labelling of Proteins

[0153] Protein, (typically 1-20 mg in 0.1M sodium carbonate buffer, pH 7.5-8.0, 1.0 ml) was mixed with 0.2 ml of 1M sodium borate buffer, pH9.5. Then 0.1 ml of fluorescamine (0.1 mg-ml.sup.1 of acetone) was added, and thoroughly Vortexed for about 15 sec.

[0154] Cell lines: Human ovarian adenocarcinoma cells were OVCAR-3 (ATCC HTB-161). Human cervical cancer (HeLA) and human melanoma cells (WM3314, A375, WM3682, WM3526) and mice melanoma (B16F10). Human normal embryonic kidney cells, HEK 293 (ATCC CRL1573) serve as healthy controls. The cells were cultured in RPMI1640 supplemented with 10% heat-inactivated FBS and antibiotics.

[0155] Flow cytometry: For absolute cell number, the cells were harvested in a fixed volume and counted. For Annexin-PI assay, cells were harvested and incubated with Annexin v-FITC and PI (BioVision) according to manufacture instructions and analyzed by flow cytometry. Annexin-/PI-, surviving cell fraction; Annexin+/PI-, early apoptosis; and Annexin+/PI+, late apoptotis. For cell cycle, the cells were permeablized by 70% ethanol and PI was added and the cells were analyzed by FACS.

[0156] Results

[0157] 1. Characterization of the Individual Component of the TiO.sub.2-DLDH-RGD.sub.2 Compex

[0158] 1.1 TiO.sub.2 Preparations

[0159] Photoreactivity Upon UVA and UVC Illumination

[0160] The various TiO.sub.2 preparations described in FIG. 1, were tested for MB degradation capability under both UVC and UVA illumination. As summerize in Table 2, highest activities were obtained with the anatase nanoparticles of Sigma.

[0161] Production of Reactive Oxygen Species (ROS) which is the main cause for photodegradation was then analyzed by a specific in vitro assay based of Cyt C reduction for most efficient TiO.sub.2 form in Table 2, Anatase NP preparation as shown in FIG. 2A and FIG. 2B.

[0162] 1.2 DLDH-RGD.sub.2

[0163] DLDH-RGD.sub.2-recombinant DLDH with RGD tails on both termini prepared in E.coli and obtained as an active, 255 kDa holo(FAD) dimeric enzyme, as shown by spectrophotometry and FPLC analysis. Neither the TiO.sub.2-binding properties nor its enzymatic activity DLDH were affected by the addition of the RGD tails (data not shown). It is pertinent to note that DLDH is enzymatically active only as a dimer and only when FAD is bound to the molecule.

[0164] ROS Production By DLDH-RGD.sub.2

[0165] It was anticipated that the ROS generating ability of DLDH-RGD.sub.2 should be associated with redox enzyme activity, an effect which was reported before to occur in the mitochondria of the living cells. The ROS generating activity was measured in the presence of the two co-substrates, DHL and the NAD analog (AcPyAD) or their absence, under UVA illumination or in the dark (FIG. 3A). The enzyme was found to be stable under UVA illumination and ROS production conditions (data not shown). The Figure indicates that DLDH-RGD.sub.2 produces a similarly high ROS activity under dark and UVA conditions provided that the substrates are present. Next, ROS activity was measured by increasing concentrations of DLDH-RGD.sub.2 at different incubation times (FIG. 3B). As shown in this Figure, ROS generation increases by DLDH-RGD.sub.2 in a time dependent manner. Comparable ROS production was observed by the different DLDH-RGD.sub.2 concentration (0.1-1 M). This led us to further study the activity of DLDH-RGD.sub.2 in vitro.

[0166] 1.3 TiO.sub.2-DLDH-RGD.sub.2

[0167] Binding of DLDH-RGD.sub.2 to TiO.sub.2

[0168] The binding isotherm of DLDH-RGD.sub.2 to TiO.sub.2 was analyzed next. A constant amount of the protein was incubated with increasing amounts of TiO.sub.2 particles and the amounts of bound protein were determined. As shown in FIG. 4 a saturation curve was obtained with an apparent Kd of 9.431.38 mg DLDH-RGD.sub.2 per g TiO.sub.2 and a Bmax of 17.150.67 mg DLDH-RGD.sub.2 bound per g of TiO.sub.2.

[0169] In order to visualize the DLDH-RGD.sub.2 binding to TiO.sub.2 we incubate DLDH-RGD.sub.2 to Rutile plates, and used gold-NP-DLDH-antibody. FIG. 5A shows the gold NP on Rutile plates. A control without DLDH is shown in FIG. 5B.

Inhibitory Effect of Carboxylic Acids on DLDH-RGD.sub.2 Binding to TiO.sub.2

[0170] Carboxylic acids, have been shown in the literature to associate with TiO.sub.2 While at low pH their interaction with the oxide is mainly electrostatic, at neutral pH the binding of glutamate is mainly via coordinative bonds while that of aspartate is dramatically reduced. Therefore, we set to study the effect on DLDH-RGD2 binding to TiO.sub.2 of carboxylic amino acids addition to the reaction mixture. As shown in FIG. 6A, glutamate readily binds to TiO.sub.2 (Anatase) at pH 7.5 while aspartate binds much less. This binding was abolished at pH 9.5 (data not shown). The binding capacities obtained for glutamate and aspartate (27.0 and 8.0 mg amino acid per gram of TiO.sub.2, respectively) indicate coverage of 2.23 and 0.66 molecules per square nm for the two amino acids, respectively. Assuming about 10 Ti atoms per nm and binding of one carboxylic acid to two Ti atoms this figure represents close to full coverage of the Ti surface by glutamate and 30% of that for aspartate.

[0171] As shown in FIG. 6B, pre-coating of the TiO.sub.2 nanoparticle with DLDH hardly affected glutamate and aspartate binding by the oxide. In contrast, binding of DLDH to glutamate-pre-coated nanoparticles was much reduced which might be expected due to the high coverage of the TiO.sub.2 surface by the amino acid. The ability of glutamate and aspartate to compete with DLDH binding was exemplified next. When each of the amino acids was included in the reaction mixture, concomitantly with DLDH, inhibition of the protein binding to the oxide was observed which was higher for glutamate than for aspartate (FIG. 6C). It is pertinent to note that chemical modification of DLDH by blocking carboxylic groups with carbodiimide resulted in complete abolishment of DLDH binding to TiO.sub.2 (data not shown).

The Enzymatic Activity of DLDH-RGD.sub.2 and TiO.sub.2-DLDH-RGD.sub.2

[0172] To determine whether DLDH-RGD.sub.2 binding to TiO.sub.2 affects its activity, the enzymatic activities of the TiO.sub.2-adsorbed enzyme with that of the enzyme in solution were compared. Michaelis-Menten curves for dl-dihydrolipoamide (DLH) oxidation by the soluble and the TiO.sub.2-bound DLDH-RGD.sub.2 depicted in FIG. 7 clearly show that the enzyme retains its activity upon binding to TiO.sub.2 with some increase in the apparent Km value (0.215 to 0.327 mM upon TiO.sub.2-binding) and practically no change in kcat (3.55 vs 3.25 sec.sup.1, respectively).

[0173] The Effect of UVA Illumination on DLDH-RGD.sub.2 Release from the Complex with TiO.sub.2

[0174] The TiO.sub.2-DLDH-RGD.sub.2 complex is fully stable in vitro in the dark (FIG. 8A). However, when illuminated with UVA at 365 nm it is released, at least in part, from the TiO.sub.2 surface, in a form that retains the enzyme activity and cytotoxicity of the protein. A similar effect was observed when the complex was added to B 16F10 cells. When the cells were kept in the dark the complex remained intact and the protein failed to enter the cell. Upon UVA illumination the protein was released from the complex and penetrated the cells (FIG. 8B). The mean fluorescence intensity in the blue laser channel, representing labeled DLDH was quantified. As shown in FIG. 8C, a marked increase in the insertion of labelled DLDH-RGD.sub.2 into the cells over time under UVA illumination (min) was observed upon UVA illumination, but not in the dark.

[0175] Oxide-specificity of DLDH-RGD.sub.2

[0176] The capability of DLDH-RGD.sub.2 to form complexes with metal oxides other than TiO.sub.2 at pH 8.0 was tested with the acidic SiO.sub.2 and MnO, the amphoteric Al.sub.2O.sub.3 and Fe.sub.2O.sub.3 (magnetite) and the basic ZnO and MgO. Of these, DLDH-RGD.sub.2 was found to also bind magnetite and ZnO. Similarly to TiO.sub.2, the binding was not affected by NaCl presence (FIG. 9).

2. Cytotoxic Effect of DLDH-RGD.SUB.2 .on Cells

[0177] As mentioned in the Experimental section, DLDH-RGD.sub.2 was fluorescently labeled to enable its monitoring in biochemical and confocal experiments. Neither the enzyme activity (FIG. 10) nor the TiO.sub.2 binding capability (data not shown) were affected by the protein labelling.

2.1 The Cytotoxic Effect of TiO.sub.2-DLDH-RGD.sub.2 Under UVA Illumination or In the Dark.
Effect of Different TiO.sub.2 Disks PreparationsDLDH-RDG.sub.2 on the Morphology of HeLa Cells Before/After UVC (254 nm) Illumination

[0178] Since in this work we aim at using the combined cytotoxic ROS production effect of TiO.sub.2 and that of DLDH on cancer cells, we next examined the components of the biocomplex (TiO.sub.2, DLDH, RGD) on a selected cancer cell-line (HeLa).

[0179] FIGS. 11A-D depicts TiO.sub.2-coated disks (amorphous, rutile and anatase), prepared by thermal treatment of TiO.sub.2 plates (see above). The disks were further coated with DLDH-RGD.sub.2 protein. Uncoated disks served as controls. The disks seeded with HeLa cells (150,000/well) which over-express v.sub.3 integrin. Before illumination, the control cells in the absence (FIG. 11A) or presence (FIG. 11B) of DLDH-RGD.sub.2 protein, maintained classical HeLa morphology and density, the cell phenotype appeared to be more round and condensed. Following UVC illumination, 10 min, blebbing of the cell membrane, as well as a reduction in cell density, a feature of apoptosis, was evident (FIG. 11C). Interestingly, this effect was more significant in the presence of the DLDH-RGD.sub.2 protein (FIG. 11D). After an overnight incubation, nuclear dye (Hoechst) was added and the cells were visualized by fluorescent microscopy.

[0180] The cells from the same experiment were collected and examined for survival by Annexin-PI (An-/PI-) Fluorescence-activated cell sorting (FACS) analysis. Before Illumination (FIG. 12A, black bars), the cells survival remained high and was similar for the different conditions. Illumination by UVC for 10 minutes (grey bars) induced a differential decrease in survival in comparison to control non-illuminated cells. In the absence of DLDH-RGD.sub.2 protein, two titanium conformations, rutile and amorphous, were least potent (58 and 62% average survival rate, respectively), while control (HeLA) and anatase treated cells exhibited a comparable significant reduction in cell survival (39 and 41%, respectively).

[0181] Illumination of DLDH-RGD.sub.2 protein in amorphous preparation had no additional effect on survival. However, the addition of the protein induced a significant effect in rutile disks (39% survival) and more potently in control (26% survival) and cells grown in the presence of anatase disks (25% survival). Representative results before (FIG. 12B) and after UVC illumination (FIG. 12C) are depicted, with the percentage of surviving cells (Annexin-/PI-) in each of the treatments shown in a text box within each graph (*, p<0.05; **, p<0.005). Taken together, these results imply that the titanium binding protein with its integrin-binding-RGD tails autonomously, as well as in combination with anatase TiO.sub.2, sensitizes cell's response to UVC, probably due to its ROS producing ability.

Effects of Anatase Nanotubes (NT) Glass Matrix and Its EffectDLDH-RDG.SUB.2 .Protein on the Survival of HeLa Cells Before/After UVA (365 nm) Illumination

[0182] To improve the illumination methods, radiation was switched to UVA light, which is of a longer wavelength (365 nm) in the near UV range. This radiation flux is far below the radiation flux considered safe by International standards and is in addition less mutagenic and more penetrable than UVC. Similar UVA irradiation are currently under use in various skin diseases (Malinowska et al. 2011). The next step was to examine additional matrixes that might be more relevant for our experimental model. Glass cover slips were chosen for further validation. In order to increase surface area and ROS activity of the TiO.sub.2, we have developed 8 nm anatase-NT (see above). The glass plates were prepared anatase-NTDLDH-RGD.sub.2 protein. Uncoated glass cover slips served as controls. HeLa cells (150,000 cells/24 wells) were seeded in the presence of the different preparations (FIG. 13A) and were examined by FACS before and after an hour of UVA illumination for apoptotsis/survival. Before UVA illumination (black bars), no significant apoptosis was observed under the different conditions. However, following an hour of illumination under UVA (grey bars), the titanium produced apoptosis in 18% of cells and a significant synergistic effect was further observed in the presence of the DLDH-RGD.sub.2 protein (44% apoptosis). Representative results (FIG. 13B) before and after UVA are depicted, with the percentage of surviving cells (Annexin-/PI-) in each of the treatments shown in a text box within each graph (*, p<0.05; **, p<0.005). The nets were covered with/without the anatase-NT with/without DLDH-RGD.sub.2 protein in the presence/absence of an hour of UVA illumination (FIG. 13B). No significant apoptosis was observed in control and anatase-NT treated cells in the absence or presence of UVA illumination without DLDH-RGD.sub.2. However, nets covered with anatase-NT and the DLDH-RGD.sub.2 produced a similar induction of apoptosis (35%) without illumination and after UVA illumination. Representative results (FIG. 13C) before and after UVA are depicted, with the percentage of surviving cells (Annexin-/PI-) in each of the treatments shown in a text box within each graph (*, p<0.05).

Exploring the Cytotoxic Effect of TiO.sub.2-DLDH-RGD.sub.2 (NP) in Cancer Cells

[0183] To examine the cytotoxic effect of the nanobiocomplex (TiO.sub.2DLDH-RGD.sub.2) under UVA illumination on cervical cancer cells (Hela) and normal cells (HEK293), the cells incubated (100,000 cells/well) for 1 h under UVA illumination or without, in the presence of TiO.sub.2-DLDH-RGD.sub.2 (NP)-(1 m). After 48 h incubation (for recovery), cell survival was tested by confocal microscopy and FACS analysis (FIG. 14A). UVA illumination activates the TiO.sub.2 and induces cell death in both cancer (Hela) and normal (HEK293) cells (FIG. 14B). In Hela this effect is enhanced upon TiO.sub.2-DLDH-RGD.sub.2 addition (in B16F10 only). No such effect was observed without UVA illumination.

[0184] FIG. 15 shows the cytotoxic effect is observed after 48 h incubation in the presence of increasing concentrations of A-NP (0.05-5 mg/ml) following UVA illumination (1 h). The cytotoxic effect is evident by a reduction in cell densitymouse melanoma cell line (B16F10).

[0185] This experiment was repeated with the addition of 0.5 mg of fluorescamine-labeled DLDH-RGD.sub.2 and different amounts of A-NP (0.05-5 mg/ml). The cell nuclei, stained by Draq5 are indicated by a red color, whereas the fluorescamine-labelled protein is in blue. The cells were incubated for 1 h either in the dark or under UVA illumination. While the cells remained intact in the dark a substantial amount of nuclear shredding, indicative of apoptosis, was shown with increasing TiO.sub.2 concentrations (FIG. 16). These results indicate the requirement of UVA illumination for the nanobiocomplex activation.

[0186] As shown in FIG. 17A, incubation for 48 h at 37 C. of DLDH-RGD.sub.2 at increasing amounts concentration (0.5-10 M) with 100,000 mouse melanoma cell line (B16F10) leads to incorporation of the protein into the cells and initiation of an apoptotic effect. FACS analysis (FIG. 17B) showed increased early and late apoptosis which occur with increasing DLDH concentration. The percentage of cells in SubG1 (FIG. 17C) increases from 20% to 83%. Incubation of the cells with DLDH without RGD tails resulted in a similar but slower effect. In contrast, ovalbumin, served as a control, failed to enter the cells and cause cell death. FACS analysis was repeated with four different human melanoma cell lines (FIG. 18A and B).

2.2 The Cytotoxic Effect of DLDH-RGD.sub.2 (in the Absence of TiO.sub.2) in B16F10 Melanoma cells, Under UVA Illumination or in the Dark.

[0187] It was assumed that this phenomenon was due to an excess of unbound protein (DLDH-RDG.sub.2).

[0188] In FIG. 19A, confocal images (time-laps video) of an in-situ apoptosis assay in B 16F10 melanoma cells (100,000 cells/well) that were incubated with 5 m DLDH-RGD.sub.2 during time (min), without illumination, are shown. The assay follows in multiplex the cell nuclei, which is stained by Draq5 (far red), Fluorescamine-labeled DLDH-RGD.sub.2 (blue), Psiva (green), indicative for early apoptosis and PI (pink) is indicative for late apoptosis. It is noticeable that DLDH-RGD.sub.2 enters the cytoplasm within 5 min leading quickly to the destruction of the cell nucleus.

[0189] The mean fluorescence intensity of the each laser channel was measured during time. FIG. 19B shows the disappearance of the nuclei that points to cell death, FIG. 19D shows the insertion of fluorescamine-label DLDH-RGD.sub.2 into the cells, in FIG. 19C the increase of the PI dye that detects late apoptotic death is shown and in FIG. 19E the increase of the Psiva dye that detects early apoptotic death is shown.

[0190] By comparing the cytotoxic effect on mouse melanoma cell line (B16F10) of TiO.sub.2-DLDH-RDG.sub.2 with that of DLDH-RGD.sub.2 it is obvious that the complex requires UVA illumination to be effective, while the protein alone acts on the cells also in the dark. Considering the data in FIG. 3 showing that ROS production by DLDH-RGD.sub.2 is independent of UVA illumination, led us to hypothesize that UVA activation is needed to release the protein from the complex in situ. To examine this possibility we subjected the release of DLDH-RGD.sub.2 from the complex TiO2-DLDH-RDG.sub.2 upon UVA illumination in cell free conditions (FIG. 8).

[0191] The cytotoxic effect of DLDH-RGD.sub.2 on melanoma cells compared to normal cells was investigated next. FIG. 20A depicts the cytotoxic effect of Fluorescamine-labelled DLDH-RGD.sub.2 (5 M) upon incubation with HEK293 (normal kidney cells) and B16F10 cells (100,000 cells/well), for 6-48 hrs. The cell nuclei were stained by Draq5 (red). As shown in the Figure, no cytotoxic effect was observed with the HEK293 cells for at least 24 h, apparently due to its low integrin expression. In contrast, B16F10 cells which highly express the integrin, as expected, were susceptible to DLDH-RGD.sub.2 showing high cytotoxicity within 6 h of application. FIG. 20B shows cell membrane v3 integrin expirations by FACS analysis.

2.3 The Effect of the RGD Tails on the Cytotoxicity of DLDH

[0192] Next, the inventors compared the contribution of the RGD tails to the DLDH protein cell penetration. B16F10 (100,000 cells/well) were seeded overnight. Fluorescamine-labelled DLDH (FIG. 21A) or DLDH-RGD.sub.2 (FIG. 21B) were added to each well and the cytotoxicity was visualized under Leica SP8. The solvent served as negative control (FIG. 21C). The cells nucleauses were stained by Draq5 (Far red), and protein stained by Fluorescamine (blue). Results indicated that protein penetration to the cells was enhanced in the presence of the RGD tails (FIG. 21B), leading to nuclear destruction and cell death.

[0193] Next, it was examined whether RGD tri-peptides inhibit DLDH penetration to the cells. HEK 293 cells (normal kidney cells) were transfected to express the integrin v.sub.3 (HEK293B3). 100,000 cells/well were seeded overnight. Fluorescamine-labelled DLDHRGD (1 M) was add to each well in presence (1 mM) or absence of RGD (as a competitive inhibitor) and the penetration rate as well as the cytotoxicity were determined by confocal Microscopy (Leica SP8). The cells nuclei was stained by Draq5 (red). Blue marks the Fluorescamine labelled DLDH-RGD.sub.2. As shown in FIG. 22, while in the absence of RGD addition, DLDH quickly incorporated into the cells, inducing cell death as previously shown (FIG. 22A), while co-incubation of the protein in the presence of RGD (1 mM), which act as v.sub.3 antagonist, potently prevented DLDH-RGD.sub.2 from entering the cells (FIG. 22B). These results further indicate the integrin mediated nature of DLDH-RGD.sub.2 take in by the cancer cells.

[0194] In FIG. 23 the intake of fluorescamine(Fluram)-labelled DLDH-RGD.sub.2, DLDH (devoid of RGD tails) and Ovalbumin and glycine, as a negative controls, are shown. The nuclei of the selected cancer cell-line (OVCAR3, ovarian) were stained bleu using DraQ5. Results show that DLDH-RGD.sub.2 is the first to enter the cells (37 sec, probably by endocytosis) and reach out-side of the nucleus (60 sec), leading to apoptotic cell blebbing (83 sec) and collapse (128sec), which not occur at the controls. The experiment was videoed during 370 sec at LSM 510 META (X63).

[0195] DLDH-RGD.sub.2 was shown to incorporate the cancer cells significantly faster than DLDH. However, the time interval from cell entry till cell death (23 sec to apoptosis and additional 45 sec the cell collapse) was comparable. The controls, ovalbumin and glycine entered the cells but did not initiate apoptosis process (Table 3).

TABLE-US-00002 TABLE 3 comparison between the complex components during incubation time (sec). Function DLDH-RGD.sub.2 DLDH OV Incorporation 37 68 83.5 into the cell Apoptosis 83 136.5 X Cell destruction 128 182 X
2.4 DLDH as a DNA degrading enzyme

[0196] The nucleus destruction by DLDH led us to examine the possibility that the enzyme possesses an activity of DNase. Such an activity is shown in FIG. 24. When DLDH (1 or 10 g) was incubated with X phage (0.15 g), degraded (FIG. 24A) or intact (FIG. 24B). Degradation of the phage by DLDH was observed only in the presence of its substrates and was more pronounced at the higher enzyme concentration (lanes 6 and 11).