UC Berkeley

Cancer unforgivingly impinges upon millions of lives daily and is a predominant cause of death worldwide. Conventional cancer treatment is comprised of surgery, radiation, chemotherapy, hormone, immune, and targeted therapy, however, with the deleterious side effects and eventual tumor resistance that invariably ensue, there is an urgency to develop safer, less invasive therapies. Within the past decade, a significant transition in cancer therapeutics has unfurled as molecular targeted therapies have emanated as an alternative treatment for an array of cancers which include breast, colorectal, lung, pancreatic, lymphoma leukemia, and multiple myeloma. Molecular targeted therapies suppress critical biochemical pathways or mutant proteins that are required for tumor cell growth and survival. Within a molecularly defined cluster of patients, these drugs have the capacity to arrest tumor progression and can impel dramatic regressions. Identifying novel classes of highly potent therapeutic agents that specifically act on molecular targets with diminished side effects after continued treatment has been a challenging obstacle to overcome in the treatment of cancer. Among the array of molecular targeted agents employed, indole-3-carbinol (I3C), has emerged as a viable anti-cancer agent. I3C is a bioactive component in cruciferous vegetables that is derived by hydrolysis from glycobrassicin in Brassica and exhibits multiple anticarcinogenic and antitumorigenic properties as well as chemo-preventative and strong anti-tumor properties in vivo in rodent model systems and in human cancer cell xenograft tumors. A major headway in understanding the I3C anti-proliferative mechanism is our discovery that I3C triggers distinct and overlapping sets of anti-proliferative signaling events by direct interactions with specific target proteins. Through a multi-faceted series of cellular and biochemical experiments, our lab has cemented three I3C target proteins which include human neutrophil elastase, E3 ubiquitin ligase NEDD4-1, and oncogenic B-RAF V600E serine/threonine kinase.

Because of I3C’s off-target and nonspecific cytotoxic effects and its propensity to dimerize into its natural condensation product DIM, which triggers distinctly different antiproliferative cascades, there is a need to generate more potent, target-specific compounds this ultimately lead to a promising new compound, 1-benzyl-I3C, which is substantially more effective in suppressing enzymatic activity, inducing anti-proliferative effects in melanoma and breast cancer cells, and diminishing tumor size in mouse xenografts . Here, we demonstrate that both I3C and 1-benzyl-I3C serve as molecular scaffold for creating a novel, robust enzymatic small molecule inhibitors aimed at disrupting specific target proteins in melanoma and breast cancer cells, particularly NEDD4-1 and elastase respectively. By executing an in silico approach, we were able to make predictions about the mechanistic and structural nature of a set of five synthetic I3C and 1-benzyl-I3C derived analogs and employed a combination of in vitro protein thermostability assays and enzymatic assays to substantiate our predictions. Notably, compounds 2242 and 2243, the two indolecarbinol analogues with added methyl groups that result in a more nucleophilic benzene ring π system, further inhibited NEDD4-1 enzymatic activity more dramatically with IC50s of 2.7 µM and 7.6 µM, respectively. Interestingly, compounds 2242, 2160, and 2243 inhibited elastase activity much more significantly as well with and an IC50 of 30.4 μM, 25.1 μM, and 16.9 μM respectively. In both NEDD4-1 and elastase enzymatic studies, the potency of compound 2242, 2160, and 2243 appeared to be sensitive to structural changes on the phenyl ring, more notably when methyl substituents were added to the para and ortho positions. The activity was abolished with the addition of bulky chemical groups like the thiophene substituent attached to the indole of 2163 and the electron donating methoxy group attached to the phenyl moiety of 2244 in the meta and ortho positions.

The quest for additional I3C target proteins has been facilitated through our understanding of homology modeling and identification of patterns in protein folds and ligand binding sites. In order to make predictions about additional indolecarbinol target proteins, we obtained the crystal structures of the four I3C target proteins to date (human neutrophil elastase, ubiquitin E3 ligase NEDD4-1, oncogenic BRAF V600E serine/threonine kinase, WNT) and used them as the starting template structure for creating homologous protein models. Homology models for various homologues of the target proteins were generated to ascertain whether or not similarities in potential indolecarbinol binding sites could be visibly discerned after performing a computational docking analysis of indolecarbinol compounds into their most feasible predicted binding sites. After examining the indolecarbinol binding sites in the homology models, it is evident that the homologous proteins that have greater than 50% sequence identity share a distinct structural architecture that confers binding to the indolecarbinol compounds, but generally, homologous proteins that share 30% sequence identity or below tend to lose essential I3C contact residues.

Examining the binding modes of I3C, 1-benzyl-I3C, and their corresponding synthetic derivatives in complex with their target proteins, namely, elastase, NEDD4-1, oncogenic BRAF V600E serine/threonine kinase, and wnt, can illuminate various patterns in binding sites and can hence provide valuable information for locating additional indolecarbinol target proteins. Conceivably, information garnered from these studies will be useful in making predictions allowing identification of additional indolecarbinol compound target proteins.

The results presented here provide the fundamental framework for understanding the mode of inhibition of natural and synthetic indolecarbinol compounds and the proteins they effectively target.