Carbonic anhydrase inhibitors: E7070, a sulfonamide anticancer agent, potently inhibits cytosolic isozymes I and II, and transmembrane, tumor-associated isozyme IX

https://doi.org/10.1016/j.bmcl.2003.09.062Get rights and content

Abstract

E7070 [N-(3-chloro-7-indolyl)-1,4-benzenedisulfonamide] is an anticancer drug candidate under clinical development for the treatment of several types of cancers. We prove here that this compound also acts as a potent carbonic anhydrase (CA) inhibitor. Similarly to the clinically used drugs acetazolamide, methazolamide and topiramate, E7070 showed inhibition constants in the range of 15–31 nM against isozymes I, II and IX, being slightly less effective as a CA IV inhibitor (Ki of 65 nM). The X-ray crystal structure of the adduct of hCA II with E7070 revealed unprecedented interactions between the inhibitor and the active site, with three different conformations of the chloroindole fragment of the inhibitor interacting with different amino acid residues/water molecules of the enzyme. A superimposition of these conformations with those of other sulfonamide/sulfamate CA inhibitors indicated that similar regions of the hCA II active site could be involved in the interaction with inhibitors.

Introduction

Sulfonamides possess many types of biological activities, and representatives of this class of pharmacological agents are widely used in clinic as antibacterial, hypoglycemic, diuretic, anti-hypertensive and antiviral drugs among others.1, 2 Recently, a host of structurally novel sulfonamide derivatives have been reported to show substantial antitumor activity in vitro and/or in vivo.1, 2, 3, 4, 5, 6, 7, 8, 9 All these derivatives incorporate in their molecules a common chemical motif of aromatic/heterocyclic sulfonamide, but there are a variety of mechanisms of their antitumor action, such as inhibition of tubulin polymerization, cell cycle perturbation in the G1 phase, carbonic anhydrase (CA) inhibition, functional suppression of the transcriptional activator NF-Y, angiogenesis (matrix metalloproteinase or integrin α2) inhibition, and so on.1, 2, 3, 4, 5, 6, 7, 8, 9 Among such compounds selected via elaborate screenings of compound libraries or obtained through computer-based drug design, E7070 (1) is one of the most potent anticancer sulfonamides ever reported.8, 10, 11, 12 This statement is based on the significant in vivo efficacy of 1 against various human tumor xenografts in nude mice, for example colon carcinomas HCT116, LS174T, SW620 and HCT15 and lung carcinomas LX-1 and PC-9.8, 10 In particular, this compound caused complete tumor regression in the xenografts of HCT116 and LX-1 following its intravenous administration at 25 mg/kg daily for 8 days.10 Currently, 1 is under clinical evaluation (phase I trials in Japan and phase II trials in Europe and the United States), thus far demonstrating some objective antitumor responses in patients with colorectal cancer, non-small cell lung cancer, and so on.11, 12, 13, 14

The precise mechanism of action has not been determined, but 1 is considerably different from conventional anticancer drugs in clinical use with respect to its effects on the cell cycle and its tumor type selectivity. In HCT116, one of the most sensitive cancer cell lines, 1 was found to suppress the expression of cyclin E and the phosphorylation of CDK2, both of which are essential for the G1 to S transition.8 A more detailed mechanistic study by Fukuoka et al.15 clarified that 1 disrupted cell-cycle progression at multiple points, including both G1/S and G2/M transitions, in a human non-small cell lung cancer cell line A549. In their experiments using A549 and its 1-resistant subline A549/ER, the compound was shown to inhibit pRb phosphorylation, to reduce the protein expression of cyclin A, cyclin B1, CDK2 and CDC2, and to suppress CDK2 catalytic activity with the induction of p53 and p21 proteins only in parental (drug sensitive) A549 cells. All these observations suggest that 1 belongs to a new class of cell cycle inhibitors affecting multiple cell-cycle checkpoints. More recently, DNA microarray analysis has further revealed that the drug treatment at 8 μM for 12 h altered at least 3-fold the expression levels of 60 transcripts; 58 of these were down-regulated, and the remaining two were up-regulated.16, 17 An observation of particular interest is profound transcriptional repression of subsets of energy metabolism and cell-cycle genes, including mitochondrial NAD(P)+-dependent malic enzyme, DNA polymerase α and cyclin H, all of which are also significantly down-regulated by 1 in common in several other cancer cell lines (T. Owa, unpublished data).

Although the gene expression analysis has actually provided an important insight into the mode of action of 1, it is also critical to identify a cellular protein target(s) of 1 for understanding accurately its anticancer mechanism(s). A most recent report by Oda et al. has described that cytosolic malate dehydrogenase (cMDH) is a specific binding protein and apparently a primary cellular target of 1 on the basis of quantitative proteomic analysis.18 Independent of this study, we have investigated whether 1 shows inhibitory activity against CAs as this compound is a primary sulfonamide fitting for a general formula of CA inhibitors.2, 5 These enzymes play a key role in a multitude of physiological and physiopathological processes, including tumorigenesis.2, 5, 6, 19 Recently, at least two tumor-associated CA isozymes have been identified (CA IX and CA XII)19 among the fourteen isozymes isolated in humans.2, 5, 6 Of these, CA IX shows restricted expression in normal tissues but is tightly associated with different types of tumors, mostly due to its strong induction by tumor hypoxia that involves HIF-1 binding to a hypoxia response element in the CA9 gene promoter.19 CA XII is present in various normal tissues and overexpressed in some tumors. It is also induced by hypoxia, but the underlying molecular mechanism remains undetermined. Both CA IX and CA XII are negatively regulated by von Hippel Lindau tumor suppressor protein and their expression in renal cell carcinomas is related to inactivating mutation of VHL gene.19 The high catalytic activity of these two isozymes supports their role in acidification of tumor microenvironment that facilitates acquisition of metastatic phenotype.19 Therefore, modulation of extracellular tumor pH via inhibition of CA activity represents a promising approach to anticancer therapy. Indeed, novel sulfonamide/sulfamate CA inhibitors with nanomolar activity against the tumor-associated isozyme IX have been reported by this group,20, 21, 22 together with classical sulfonamide drugs such as acetazolamide 2 and methazolamide 3. On the other hand, recently developed topically acting antiglaucoma sulfonamides such as the bis-sulfonamide 423 and the perfluorophenyl derivative of methazolamide 524 or the sulfamate antiepileptic drug topiramate 625 have not been investigated for their interactions with the tumor-associated isozyme CA IX.

Here, we report a detailed CA inhibition study of 1 against four isozymes, the cytosolic CA I and II, the membrane-bound isozyme IV, as well as the transmembrane, tumor-associated isozyme CA IX. An X-ray crystallographic study for the hCA II–1 adduct has also been performed, and is reported here. The structure of this adduct evidenced unprecedented interactions between the inhibitor and the enzyme active site and may be useful for the drug design of more active agents belonging to this class of pharmacological agents.

Section snippets

Chemistry and CA inhibition

1 was prepared as previously reported.8 The other sulfonamides/sulfamates (26) used as reference compounds were either commercially available (2, 3 and 6) or prepared as previously reported (4 and 5).23, 24 Inhibition data against four CA isozymes with the six compounds 16 are shown in Table 1.

These data clearly indicate that 1 acts as a strong CA inhibitor against the four investigated isozymes CA I, II, IV and IX, with potencies comparable to those of the reference sulfonamide/sulfamate

Crystallography

The hCA II–1 adduct obtained by co-crystallization, was subjected to detailed X-ray crystallography. The programs SHELX9731 and O32 were used to build the model and to compute the Fourier maps. The last refinement cycle yielded a final R factor of 0.21 (Rfree=0.25) with a final temperature factor of the inhibitor atoms ranging between 8.5 and 24.0 Å2. The final number of water molecules was 195 and the final rmsd's from ideal geometry for bond lengths and angles were 0.02 Å and 1.8°,

References (39)

  • Y Ozawa et al.

    Eur. J. Cancer

    (2001)
  • C.J Punt et al.

    Ann. Oncol.

    (2001)
  • D Vullo et al.

    Bioorg. Med. Chem. Lett.

    (2003)
  • F Mincione et al.

    Bioorg. Med. Chem. Lett.

    (2001)
  • A Casini et al.

    Bioorg. Med. Chem. Lett.

    (2003)
  • R.G Khalifah

    J. Biol. Chem.

    (1971)
  • K Hakansson et al.

    J. Mol. Biol.

    (1992)
  • A Casini et al.

    Bioorg. Med. Chem. Lett.

    (2003)
  • A Scozzafava et al.

    Curr. Med. Chem.

    (2003)
  • C.T Supuran et al.

    Med. Res. Rev.

    (2003)
  • A Casini et al.

    Curr. Cancer Drug Targets

    (2002)
  • T Owa et al.

    Exp. Opin. Ther. Pat.

    (2000)
  • C.T Supuran et al.

    Exp. Opin. Ther. Pat.

    (2002)
  • C.T Supuran et al.

    Curr. Med. Chem. Imm., Endoc. Metab. Agents

    (2001)
  • C.T Supuran et al.

    Exp. Opin. Ther. Pat.

    (2000)
  • T Owa et al.

    J. Med. Chem.

    (1999)
  • Y Funahashi et al.

    Cancer Res.

    (2002)
  • C.T Supuran

    Exp. Opin. Investig. Drugs

    (2003)
  • C van Kesteren et al.

    Anticancer Drugs

    (2002)
  • Cited by (273)

    View all citing articles on Scopus
    View full text