|Home | About | Journals | Submit | Contact Us | Français|
An efficient synthesis of apricoxib (CS-706), a selective cyclooxygenase inhibitor, was developed using copper catalysed homoallylic ketone formation from methyl 4-ethoxybenzoate followed by ozonolysis to an aldehyde, and condensation with sulphanilamide. This method provided multi-gram access of aprocoxib in good yield. Apricoxib exhibited potency equal to celecoxib at inhibition of prostaglandin E2 synthesis in two inflammatory breast cancer cell lines.
Cyclooxygenase (Cox) enzymes, also known as prostaglandin H synthases (PGHS), catalyze the rate limiting step in the formation of inflammatory prostaglandins, most notably the inflammatory mediator, prostaglandin E2 (PGE2).1, 2 Cox-1 is constitutively expressed and PGE2 derived from this isoform is associated with survival of specific populations of epithelial cells such as crypt stem cells in the gastrointestinal tract.3 Cox-2 is transcribed from an inducible immediate early gene primarily responsible for the production of PGE2. Since this isoform was first cloned and sequenced in 1992,4 numerous studies have documented the association between elevated gene expression and proliferation, invasion, angiogenesis and metastasis in human tumors from different organ sites, thus validating Cox-2 as a useful therapeutic target.5, 6 In invasive breast cancer, Cox-2 mRNA and protein are elevated regardless of hormone receptor or Her-2 neu status.7, 8, 9
Selective Cox-2 inhibitors, such as celecoxib (Celebrex®, Pfizer, Inc), in addition to treating pain and inflammation, showed great promise as inhibitors of tumor growth and angiogenesis.10, 11 Unfortunately, due to the observation of unacceptably high rate of cardiovascular side effects,12, 13 the Cox-2 selective inhibitor rofexocib (Vioxx®, Merck, Inc) was removed from the market on September 30, 2004, with a black box warning issued for Celebrex®. Consequently, efforts to develop next generation Cox2 inhibitors with different toxicity profiles that can be used clinically as anti-tumor, anti-angiogenesis and chemopreventive agents10, 11 continue.
Apricoxib, (CS-706, 1) 2-(4-ethoxyphenyl)-4-methyl-1-(4-sulfamoylphenyl)-pyrrole, a small-molecule, orally active, selective COX-2 inhibitor was discovered by investigators at Daiichi Sankyo in 1996.14, 15 Clinical studies demonstrated potent analgesic activity16, 17 and preclinical studies demonstrated good pharmacokinetics, pharmacodynamics and gastrointestinal tolerability.18 As an anticancer agent, preclinical studies demonstrated efficacy in biliary tract cancer models19 and colorectal carcinoma,20 and Recamp et al.21 recently reported a Phase I trial of 1 in combination with erlotinib in lung cancer.
Our goal is to evaluate the clinical potential of apricoxib in inflammatory breast cancer models, which requires multi-gram quantities. Daichi Sankyo reported the synthesis of 1 in a patent14 and appears to have provided the compound for the clinical and pre-clinical studies mentioned above. The original synthetic route is outlined in Scheme 1. Though the initial two steps were accomplished with decent yields, the final step of pyrrolidine formation followed by dehydration and dehydrocyanation produced only 3% of 1 as a brown powder. The yield in the last step of the synthesis of the 2-(4-methoxyphenyl) analog, 2-(4-methoxyphenyl)-4-methyl-1-(4-sulfamoylphenyl)-pyrrole, was 6%,14 indicating that this synthesis route is problematic.
The route used to prepare the 4-ethyl analog, 2-(4-ethoxyphenyl)-4-ethyl-1-(4-sulfamoylphenyl)-pyrrole, in the same patent employed the well known Paal-Knorr condensation of the intermediate γ-ketoaldehyde 7 and sulfanilamide (Scheme 2).14 The process involved conventional enamine alkylation with the appropriate phenacyl bromide in an inert solvent followed by acid hydrolysis leading to the desired dicarbonyl compound. In our hands, coupling of 4-ethoxyphenacyl bromide (5)22 and 1-(N,N-diisopropylamino)-1-propene (6) in solvents such as toluene, CH2Cl2, dry THF at 0°C, room temperature, or reflux produced at best 18% yield of 7. Purification of 6 through distillation was fruitless so crude enamine was used, which likely contributed to the low yields. The corresponding 1-piperidino-1-propene also did not give high yields of 7. In a later patent researchers at Daichi Sankyo used γ-ketoaldehyde protected as a ketal23 and condensed this with sulfanilamide under acid conditions.22 Note that 5 and 6 also had to be synthesized which add further steps to the preparation of 1.
We envisioned that 7 could be prepared by ozonolysis of homoallylic ketone (8) (Route B). A recent development in the synthesis of homoallylic ketones by Dorr et al. via copper-catalyzed cascade addition of alkenylmagnesium bromide to an ester24 was examined. Treatment of commercially available methyl 4-ethoxybenzoate with 1-propenylmagnesium bromide (4.0 equiv) in presence of CuCN (0.6 equiv) resulted in 95% yield of desired ketone 8 after silica gel chromatography, along with a minor amount of unreacted ester (Scheme 3).25
The product was a mixture of cis/trans R/S stereoisomers, as detected in the 1H NMR spectrum, and was used directly in the next step without separation. Ozone was bubbled through a solution of 8 in MeOH/CH2Cl2 at −78°C, until all starting materials were consumed. The ozonide was then reduced to aldehyde 7 by treatment with Me2S overnight. Removal of volatiles and subsequent addition and evaporation of toluene gave the crude 1,4-dicarbonyl compound 7 which was sufficiently pure for the following condensation step. The 1H NMR signal at 9.78 ppm of the crude product confirmed the formation of the aldehyde. No attempt was made to characterize the enantiomeric ratio of 7 since the dehydration/aromatization reaction of the next step removes the chirality of the product. Treatment of 7 with sulfanilamide in 40% acetic acid-acetonitrile at 70°C for three hours resulted in a brown product. Purification by silica gel flash chromatography yielded 71% of pure 1 as a white solid.26
Our lots of apricoxib were assayed for the ability to inhibit production of PGE2 in two inflammatory breast cancer lines, SUM149 and SUM190.27 As shown in Table 1, after stimulation with arachidonic acid, SUM149 cells produced 3 pg/mL/1000 cells of PGE2. Apricoxib at 100 nM resulted in 70% inhibition (1 pg/mL/1000 cells). Concentrations of 1 and 10 μM resulted in 91% and 97% inhibition, respectively. Although SUM190 cells produced double the amount of PGE2, the degree of inhibition with apricoxib was the same. The well known Cox-2 inhibitor, celecoxib was equally potent as 1.
In summary, we present a highly efficient synthesis of the promising COX-2 inhibitor, apricoxib, in only three steps from commercially available starting materials. The key is the improved synthesis of the γ-ketoaldehyde intermediate, 7. Multigram quantities have been prepared for use in preclinical studies. Our lots of apricoxib potently inhibit COX-2 activity in inflammatory breast cancer cells. Details of further biological evaluation will be published under separate cover.
We are grateful to the National Cancer Institute for support (PKM, JSM) CA096652, the Cancer Center Support Grant CA016672 for support of both our NMR facility and the Translational Chemistry Core facility which provided HRMS, the American Airlines-Komen For the Cure Foundation Promise Grant KGO81287 (FMR, EMF, and ALB) and The State of Texas Fund for Rare and Aggressive Breast Tumors (FMR, EMF, and ALB).
1H, 13C, and COSY NMR spectra of compounds 1 and 8.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.