The prevention of neovessel formation or angiogenesis is a recent popular strategy for limiting and curing cancer. Diaminothiazoles are a class of compounds that have been reported to show promise in the treatment of cancer by inhibiting cancer cell proliferation and inducing apoptosis, because of their effects on microtubules and as inhibitors of cyclin-dependent kinases. Many microtubule-targeting agents are being studied for their antiangiogenic activity, and a few have shown promising activity in the treatment of cancer. Here, we report that diaminothiazoles can be highly effective as antiangiogenic agents, as observed in the chick membrane assay. The lead compound, 4-amino-5-benzoyl-2-(4-methoxyphenylamino)thiazole (DAT1), inhibits endothelial cell processes such as invasion, migration, and tubule formation, which require a functional cytoskeleton. DAT1 also decreases the expression of cell adhesion markers. The antiangiogenic activities of DAT1 occur at concentrations that are not cytotoxic to the normal endothelium. Analysis of intracellular signaling pathways shows that DAT1 inhibits Akt phosphorylation, which is actively involved in the angiogenic process. The antiangiogenic properties of diaminothiazoles, in addition to their promising antimitotic and cytotoxic properties in cancer cell lines, give them an extra advantage in the treatment of cancer.
Angiogenesis, which is a crucial requirement for tumor progression and metastasis, is a potential target in cancer chemotherapy. Because tumors require the development of a vascular supply to grow beyond 1 to 2 mm in diameter, constant vascular growth and remodeling are necessary. Compromising this vascular supply leads to indirect killing of the tumor. Because of the genetic stability of the endothelial cells that line the blood vessels and the accessibility of the tumor vasculature, this approach promises less resistance (Hinnen and Eskens, 2007).
Activation of the angiogenic switch in tumors results in degradation of the basement membrane and invasion of the extracellular matrix, followed by endothelial cell migration, proliferation, and capillary network formation (Denekamp, 1982). Each stage can be a target for therapeutic strategies. A number of cytotoxic chemotherapeutic agents exhibit antiangiogenic activity. In addition, novel antiangiogenic compounds have been developed, and many that work at different stages of angiogenesis are undergoing clinical trials (Kerbel and Folkman, 2002; Eichhorn et al., 2004). However, agents that inhibited only endothelial cell proliferation in vitro did not have any effect on angiogenesis in vivo (Belotti et al., 1996; Klauber et al., 1997), which indicated that cytotoxicity was not sufficient to inhibit angiogenesis. Colchicine was among the first agents shown to possess antiangiogenic activity. Subsequently, a number of microtubule-targeting agents were shown to inhibit angiogenesis efficiently (Belotti et al., 1996; Klauber et al., 1997; Vacca et al., 1999; Hotchkiss et al., 2002). At low concentrations, tubulin-binding agents disrupt the dynamic instability of microtubules, thereby causing changes in the endothelial cytoskeleton (Vacca et al., 1999; Bocci et al., 2002; Grant et al., 2003; Wang et al., 2003). Tubulin-depolymerizing agents have been shown to increase vascular permeability as a result of their role in the maintenance of the endothelial barrier (Watts et al., 1997). Combretastatin A4 phosphate (CA4P) is a highly effective antiangiogenic agent that causes vascular shutdown, leading to tumor death (Iyer et al., 1998). CA4P and a few other tubulin-binding agents that show antiangiogenic activity are undergoing clinical trials (Mauer et al., 2008; Mita et al., 2008).
Diaminothiazole analogs were found to be cytotoxic in different types of cancer cells, with a few of the series being effective even at nanomolar concentrations (Romagnoli et al., 2009) (N. E. Smith, S. L. Smitha, S. U. Lekshmi, R. J. Komalam, K. V. Sreerekha, A. Bharathan, K. N. Rajasekharan, and S. Sengupta, unpublished results). The lead compound, 4-amino-5-benzoyl-2-(4-methoxyphenylamino)thiazole (DAT1), was shown to bind tubulin at the colchicine binding site, causing mitotic arrest and cell death (Sengupta et al., 2005). The promising activity of CA4P and a few other tubulin-binding agents in clinical trials prompted us to study the antiangiogenic activity of diaminothiazoles. Although there are reports of probable antiangiogenic activity of diaminothiazoles attributable to their inhibition of the protein tyrosine kinase Axl and cyclin-dependent kinases (Chu et al., 2000; Goff et al., 2011), our studies in angiogenic models prove their antiangiogenic activity. We also studied the probable signaling pathway underlying the antiangiogenic activity of the lead compound DAT1.
Materials and Methods
DAT1 and its analogs were synthesized through a green chemistry route that involved some process simplification of a procedure reported earlier (Sreejalekshmi et al., 2006). The details of the synthesis and characterization of the analogs will be published separately.
MCDB131 medium was from Invitrogen (Carlsbad, CA), and FBS was from PAN-Biotech GmbH (Aidenbach, Germany). Matrigel and the invasion plates were from BD Biosciences (San Jose, CA). Antibodies specific for CD31, Akt, and phospho-Akt (Thr308) were from Cell Signaling Technology (Danvers, MA). β-Actin-specific antibody and VEGF were from Santa Cruz Biotechnology (Santa Cruz, CA). PI3K inhibitor, E-selectin antibody, anti-mouse horseradish peroxidase antibody, and anti-rabbit horseradish peroxidase antibody were from Sigma-Aldrich (St. Louis, MO). The ECL Plus Western blotting detection system was from GE Healthcare (Chalfont St. Giles, Buckinghamshire, UK). All other fine chemicals were from Sigma-Aldrich.
Isolation and Culture of HUVECs.
Endothelial cells were isolated through controlled collagenase perfusion of the umbilical vein (Jaffe et al., 1973). The endothelial cells were selected with two markers, i.e., the presence of von Willebrand factor (VWF) and the uptake of acetylated low-density lipoprotein (Ac-LDL). For characterization, the cells were grown on 12-mm glass coverslips, in 24-well plates, for 48 h. The cells were then washed with PBS and fixed with 4% paraformaldehyde. After blocking of the monolayer with PBS containing 3% bovine serum albumin, the cells were permeabilized with 0.2% Triton X-100 in PBS. They were subsequently incubated with primary antibody against VWF (Sigma-Aldrich), at a concentration of 5 μg/ml, for 4 h at 37°C, followed by incubation with Alexa 546-conjugated secondary antibody for 45 min. The monolayer was washed extensively, mounted with glycerol, and observed with fluorescent microscope filter sets (Nikon TE2000E; Nikon Tokyo, Japan). Images were captured with a Retiga Exi camera (Photometrics, Tucson, AZ) and analyzed with NIS-Elements software (Nikon). For analysis of Ac-LDL uptake, cells grown on glass coverslips were incubated for 4 h at 37°C in 5% CO2 with 5 μg/ml acetylated low-density lipoprotein labeled with 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (Dil) (Biomedical Technologies, Cambridge, MA). The cells were washed and imaged with rhodamine filter sets for the fluorescent microscope, by using a 40× objective. Only cultures containing 100% endothelial cells were used for experiments. For experiments, the cells were resuspended in 5 ml of fresh MCDB131 medium containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin and were incubated at 37°C in 5% CO2; cells were used within the third passage.
HUVECs were seeded onto 96-well microtiter plates at a density of 3 × 103 cells per well and were grown for 24 h in MCDB131 medium supplemented with 10% fetal bovine serum. The effect of DAT1 on the growth of HUVECs was measured after 48 h by using the MTT dye-reduction assay (Mosmann, 1983).
HUVEC Invasion Assay.
Invasion of HUVECs was determined in an invasion chamber (BD Biosciences), which consisted of a multiwell insert plate containing 8.0-μm pores occluded with Matrigel. The cells were seeded in the upper compartment of the chamber, in 0.2 ml of MCDB131 without serum. The lower compartment contained MCDB131 supplemented with 5% FBS. After incubation for 2 h at 37°C, MCBD131 containing different concentrations of the drug was added to the upper compartment and the cells were incubated for an additional 22 h. The cells that had migrated (cells on the lower surface of the filter) were stained with propidium iodide and quantitated by using a bottom-reading spectrofluorimeter (Tecan Infinite M200; Tecan, Grödig, Austria), with excitation and emission wavelengths of 529 and 550 nm, respectively. The inhibition of migration was calculated as [1 − (migrated cells, treated/migrated cells, control)] × 100%.
Migration Assay (Scratch Assay).
Endothelial cells were seeded in MCDB131 medium containing 10% FBS, in 35-mm graduated dishes, at a density of 2 × 105 cells per plate and were incubated at 37°C. HCT116 cells were seeded in Dulbecco's modified Eagle's medium containing 10% FBS, in 60-mm graduated dishes, at a density of 5 × 105 cells per plate and were incubated at 37°C. When the cells were 80 to 90% confluent, wounds were made through the cell layer by using a sterile, 200-μl, pipette tip. The cells were then rinsed carefully with PBS, without disturbances to the monolayer. The cells were treated with fresh medium or DAT1-containing medium and were observed at 0, 6, 12, 24, and 48 h with a phase-contrast microscope with a 4× objective.
HUVEC Tube Formation Assay.
Matrigel (0.05 ml per well) was coated on a 96-well plate and was allowed to solidify for 1 h at 37°C. HUVECs were then seeded at a density of 2 × 105 cells per well in MCDB131 containing 5% FBS, VEGF (5 ng/ml), and different concentrations of DAT1. After incubation for 24 h, images of the cords formed were taken by using an inverted microscope (Nikon Eclipse TE300). The effect of DAT1 on preformed cords was studied by adding DAT1 to the cords after formation and monitoring the disruption of the cords with time.
Chicken Chorioallantoic Membrane Assay.
Fertilized chicken (white leghorn) eggs were incubated at 37°C and 50% humidity and were allowed to grow for 4 days. A 1-cm2 window was made perpendicular to the broad end of the egg. The eggs were incubated for 4 more days, and different concentrations of the drug and VEGF (5 ng/ml) were added within sterile plastic O-rings between preexisting vessels. PBS was used as the vehicle for delivery of the drug and VEGF. After incubation for 24 h, the eggs were viewed and photographed with a stereomicroscope. Angiogenesis was quantified by counting the number of blood vessel branch points in the images, by using Image-Pro Plus software (MediaCybernetics, Bethesda, MD). The percentage inhibition of neovessel formation was determined through comparison with untreated control preparations.
The cells to be analyzed were harvested and lysed with phospholysis buffer (1% Nonidet P-40, 10% glycerol, 137 mM NaCl, 20 mM Tris-HCl, pH 7.4, 20 mM NaF, 1 mM sodium pyrophosphate, 1 mM sodium orthovanadate, 1% Triton X-100, 5 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin). The whole-cell lysates (40–50 μg) were immunoblotted with appropriate dilutions of primary antibody, followed by incubation with horseradish peroxidase- or alkaline phosphatase-conjugated secondary antibody. The proteins were observed by using the enhanced chemiluminescence detection system or nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate reagent. β-Actin was used as a loading control.
Analysis of Intracellular Signaling Pathways.
HUVECs in culture were treated for 48 h with an inhibitor of PI3K/Akt [2-(4-morpholinyl)-8-phenyl-chromone (LY294002)] at a concentration of 20 μM, and expression of the angiogenic markers E-selectin and CD31 was analyzed through immunoblotting of equivalent amounts of cell lysates with specific antibodies, as described above. The effect of DAT1 on the activation of Akt through phosphorylation was studied with Western blotting.
To establish diaminothiazoles as highly promising anticancer agents, we studied their effects on tumor angiogenesis. Blocking of tumor angiogenesis is an essential strategy for preventing cancer progression and metastasis. Some proangiogenic growth factors are actively involved in tumor growth. VEGF is one of the most-potent proangiogenic factors required for the growth of tumors. To simulate the diseased condition in cancer, in which there is excess blood vessel formation, VEGF was used in all assays.
Antiangiogenic Activity of DAT1 and Its Analogs.
The chicken chorioallantoic membrane is used as a favorable model to study the antiangiogenic activity of compounds because it provides a natural medium of growing blood vessels with all of the components for the complex host interactions (Auerbach et al., 2003). We analyzed the antiangiogenic activity of DAT1 and its analogs (Supplemental Fig. 1) in this model after 24 or 48 h. The effects of five of the compounds tested are shown in Table 1. The concentrations of the compounds that caused 50% inhibition of neovessel formation and the numbers of vessel branches within the ring area are indicated. The percentage inhibition was calculated as the percentage decrease in comparison with untreated control samples. Representative images of the effects of the drugs are shown in Fig. 1. The compounds showed similar effects after 24 or 48 h.
Isolation and Characterization of HUVECs.
Endothelial cells isolated from the umbilical vein with the collagenase perfusion method were characterized by the expression of two markers. The VWF glycoprotein is a prominent marker of endothelial cells. It participates in platelet-vessel wall interactions (Zanetta et al., 2000). Another characteristic of endothelial cells is the uptake of Ac-LDL. We stained the cells with an antibody against VWF and viewed them by using a secondary antibody conjugated with Alexa 546. Endothelial cells also take up Ac-LDL through the scavenger cell pathway for low-density lipoprotein (Voyta et al., 1984). We monitored the uptake of Ac-LDL by the cells by using a fluorescently labeled analog, Dil-Ac-LDL. Only preparations that contained more than 99% endothelial cells (Supplemental Fig. 2) were used for the experiments. In addition, the cells expressed CD31 (see Fig. 5).
DAT1 Induction of Cytotoxicity in HUVECs at Relatively High Concentrations.
Diaminothiazoles have been shown to inhibit the viability of various cancer cells as a result of their actions on microtubules. To determine whether the observed antiangiogenic effects of DAT1 and its analogs were attributable to similar actions on endothelial cells, the cytotoxic activity of the lead compound DAT1 against HUVECs was evaluated with the MTT assay. A dose-response curve for the effects of DAT1 on endothelial cell viability is indicated in Fig. 2. DAT1 reduced cell viability by 50% after treatment for 48 h at a concentration of 4.47 ± 0.99 μM. This value is ∼14 times higher than the cytotoxicity of DAT1 in a variety of human tumor cell lines (average IC50, 0.35 μM) (Sengupta et al., 2005).
DAT1 Inhibition of Invasion, Migration, and Cord Formation of HUVECs.
The effects of DAT1 on HUVEC invasion and migration were studied by using the BD Biosciences Biocoat angiogenesis system. This system consists of a multiwell insert plate containing 3.0-μm pores occluded with Matrigel. It replicates the process of new blood vessel formation, in which endothelial cells escape from the vessels they line by secreting matrix metalloproteases and then migrate toward the angiogenic stimulus. A concentration of 0.5 μM DAT1 inhibited the invasion of HUVECs to the bottom compartment by 81.45 ± 26.23%, whereas 0.05 μM DAT1 inhibited invasion by 27.8 ± 3.75%.
Migration of endothelial cells was also monitored with the scratch assay in the presence of DAT1. In the case of untreated control preparations, cells began migrating after 6 h and the entire area of the scratch was closed with cells in 24 h. Migration of HUVECs was inhibited in the presence of DAT1 at a concentration as low as 0.04 μM even up to 24 h (Fig. 3a). Because cancer cells migrate in the case of tumor metastasis, the effect of DAT1 on the migration of HCT116 colon cancer cells also was monitored with the scratch assay. In the case of untreated control preparations, cells began migrating after 6 h and the entire area of the scratch was closed with cells in 48 h. In the presence of 0.04 μM DAT1, although the migration started at 24 h to a lesser extent, there seemed to be no further migration at 48 h. In the cells treated with 0.2 μM DAT1, migration started only at 48 h (Fig. 3b). These results indicated that DAT1 inhibited the migration of cells in a dose-dependent manner.
After the new outgrowth of endothelial cells, there is the need for their reorganization into a tubular structure. This can be replicated in vitro, because endothelial cells form cords spontaneously on extracellular matrix components. Visible inhibition of cord formation (observed as sparse and disrupted cords) was observed with concentrations of DAT1 as low as 16 nM, after treatment for 24 h (Fig. 4a). The inhibition was quantitated by counting the number of three-point branches in different fields and in three repeated experiments (Fig. 4b). When DAT1 was added to preformed cords, it resulted in their disruption by as early as 2 h at a concentration of 40 nM (Fig. 4c). This finding proves that DAT1 is able to inhibit and to disrupt the tube formation of endothelial cells.
Decreased Expression of Cell Adhesion Markers with DAT1.
E-selectin is a cell adhesion molecule that is involved in cell-cell contacts. Cell-cell interactions between endothelial cells are required for the formation of capillary tubes, which suggests a functional role for E-selectin in angiogenesis (Brooks, 1996; Yasuda et al., 2002). Levels of CD31, which is essential for migration and tubular structure formation, increase before endothelial cell network formation (Yang et al., 1999; Cao et al., 2009). We observed decreases in the cell-associated levels of both of these markers in the immunoblots of HUVEC samples after treatment with DAT1, at both concentrations (Fig. 5a).
DAT1 Targeting of Akt Pathway.
Akt is one of the most important proteins for activation of the angiogenic process in endothelial cells (Jiang et al., 2000). To study the signaling pathways inhibited by DAT1, we analyzed the expression of E-selectin after the addition of LY294002, a potent inhibitor of PI3K activity that has been shown to exhibit antitumor effects both in vitro and in vivo (Brognard et al., 2001).
Treatment of HUVECs with DAT1 in the presence of the PI3K/Akt inhibitor LY294002 resulted in reduction in the expression of E-selectin, compared with cells treated with inhibitor alone (Fig. 5b). DAT1 also inhibited the phosphorylation of Akt at Thr308 at concentrations as low as 100 nM, as revealed by the decreased levels of phospho-Akt after probing with antibody that detected phosphorylation at the Thr308 site (Fig. 5c).
With the emergence of antiangiogenic strategies, the scope of cancer treatment has expanded widely. The development of antiangiogenic compound-based therapy may improve the outcomes of cancer treatments, with the promise of reduced side effects and increased treatment efficacy. DAT1 inhibited the proliferation of HUVECs at a higher concentration (IC50 = 4.47 μM), compared with its IC50 in cancer cells (0.35 μM), which indicates that it would be less toxic to normal endothelial cells. Some microtubule-targeting agents show maximal antiangiogenic effects when administered at low doses over a longer time span, i.e., metronomic therapy (Belotti et al., 1996). Metronomic therapy was shown to be effective for the treatment of human cancers, with low levels of side effects, in several preclinical and clinical studies (Kerbel and Kamen, 2004). The antiangiogenic activity that DAT1 showed at a much lower effective concentration, compared with the concentration required for its antiproliferative activity and apoptosis induction in cancer cells (S. A. Thomas, S. Vasudevan, S. U. Lekshmi, T. R. Santhoshkumar, K. N. Rajasekharan, and S. Sengupta, unpublished result), could indicate its promising activity in this mode of treatment. This, in turn, could be attributable to its effect on the microtubule dynamics of endothelial cells. Earlier studies showed that low concentrations of microtubule-targeting agents induced an unexpected increase in microtubule dynamic instability in human endothelial cells, which was associated with inhibition of cell motility (Pasquier et al., 2005; Pourroy et al., 2006). This could be attributable to impairment of cell polarization or of microtubule-induced focal adhesion turnover (Hotchkiss et al., 2002).
In all of the assays, the effect of DAT1 was observed at concentrations well below the concentration at which cytotoxicity was observed. In the cord formation assay, the disruption of preformed cords was observed at a very low concentration and after a short treatment time, which rules out the cytotoxic effect of DAT1. This disruption may be attributable to the effects of DAT1 on the endothelial cytoskeleton and the adhesion molecules.
Increased levels of phosphorylated Akt have been observed in various cancers that are resistant to radiation (Gupta et al., 2002; Choe et al., 2003; Schmitz et al., 2004). On activation, Akt has been reported to activate nuclear factor κB and to up-regulate the antiapoptotic protein Bcl2, thus promoting cell survival (Béraud et al., 1999; Kane et al., 1999; Pugazhenthi et al., 2000). The PI3K/Akt pathway is involved in regulation of the angiogenic processes of survival, migration, and tube formation (Gupta et al., 1999; Yu and Sato, 1999). Many reports have shown that the expression of adhesion molecules in endothelial cells is regulated through the PI3K/Akt pathway. It was found that the expression of vascular adhesion molecule 1 in human intestinal microvascular endothelial cells is regulated by the PI3K/Akt/mitogen-activated protein kinase/nuclear factor κB pathway (Binion et al., 2009). Blockade of the PI3K/Akt pathway abrogated the suppression of tumor necrosis factor α–induced cell surface expression of E-selectin in HUVECs (Gong et al., 2006). Therefore, endothelial cells require activation of this pathway during angiogenesis (Zhu et al., 2002; Bullard et al., 2003). We observed down-regulation of phosphorylated Akt after treatment with DAT1, which indicates that this is one of the signaling pathways through which DAT1 reduces the expression of E-selectin.
Inhibition of angiogenesis is desirable for the treatment of a number of conditions, such as rheumatoid arthritis, macular degeneration, and cancer. Our experiments with colon cancer cells indicate the antiangiogenic activity that diaminothiazoles might possess in a cancer setting. The antiangiogenic properties of diaminothiazoles observed in both in vitro and in vivo models at concentrations well below the cytotoxic doses, along with their promising cytotoxicity against cancer cells, make them more promising clinically.
Participated in research design: Thomas and Sengupta.
Conducted experiments: Thomas, Thamkachy, and Ashokan.
Contributed new reagents or analytic tools: Komalam, Sreerekha, Bharathan, Santhoshkumar, and Rajasekharan.
Performed data analysis: Thomas, Thamkachy, and Sengupta.
Wrote or contributed to the writing of the manuscript: Thomas, Santhoshkumar, and Sengupta.
We thank the director of Dr. Govindan's Hospital for the samples for HUVEC isolation. We also thank Swathilekshmi and Ronodip Kar for their help in some experiments.
This work was supported by the Department of Science and Technology, Government of India; Rajiv Gandhi Centre for Biotechnology core funding from the Department of Biotechnology and the Council of Scientific and Industrial Research, University Grants Commission, Government of India; and the Kerala State Council for Science, Technology, and Environment, Government of Kerala (fellowships to S.A.T., R.T., R.J.K., and K.V.S.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- vascular endothelial growth factor
- human umbilical vein endothelial cell
- 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
- combretastatin A4 phosphate
- von Willebrand factor
- acetylated low-density lipoprotein
- 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate
- phosphatidylinositol 3-kinase
- phosphate-buffered saline
- fetal bovine serum.
- Received January 27, 2012.
- Accepted March 12, 2012.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics