Deidre M. Johns, Ph.D.

American Cancer Soc. New England Div. – Broadway on Beachside Postdoctoral Fellow

 

Ph.D. University of Colorado, Boulder

(advisor: Prof. Tarek Sammakia)

 

B. S. Chemistry, UC San Diego

Send me e-mail at djohns@lamar.colostate.edu

 

Here’s a quick description of my postdoctoral research . . .

 

Design and Synthesis of Anti-cancer Analogues of Antimycin A

 

Most anti-cancer drugs act on DNA to induce cell death in cancer cells.  One mode of multidrug resistance and gamma-radiation resistance is over-expression of anti-apoptotic members of the Bcl-2 family of proteins (Bcl-2 and Bcl-XL), which occurs in 70% of breast cancer, 90% of colorectal adenocarcinomas, 30-60% of prostate cancer, 80% of B-cell lymphomas, and many other cancer forms.  Bcl-2 proteins serve to regulate apoptosis, and over-expression of the anti-apoptotic members (Bcl-2 and Bcl-XL) protects the cells against apoptosis.  Drugs that act as alkylating agents, antimetabolites, microtubule inhibitors, and anti-tumor antibiotics are ineffective against cells that over-express anti-apoptotic Bcl-2 proteins.  Antimycin A3 (Figure 1) was recently found to selectively induce apoptosis in cells over-expressing Bcl-2 and Bcl-XL (IC50 = 3.4 µM) and is substantially less toxic to other cells.  It is effective against multidrug resistant cell lines, exhibiting a cell viability profile complimentary to doxorubicin and cisplatin.  This potentially offers a new mode for targeting a broad spectrum of resistant forms of cancer.

 

(+)-Antimycin A and 2-Methoxy Antimycin A

 

 

The antifungal antibiotic antimycin A3 was isolated from streptomyces sp. along with seven closely related compounds, antimycin A1-A8, that differ only in the alkyl substituent at C7.  They inhibit electron transfer activity of ubiquinol-cytochrome c oxidoreductase (cytochrome bc1 complex), a membrane-bound complex comprised of multiple subunits, generally inhibiting mitochondrial respiration.  Despite the activity of antimycin A3 against resistant cancer cell lines, the inhibition of mitochondrial respiration renders it toxic (LC50 of 0.9 mg/kg in mice).  Structure-activity relationships of antimycin suggest that the mitochondrial inhibitory activity can be minimized independently from the apoptotic activity.  A methylated derivative of antimycin A, 2-methoxy antimycin A (2), is 1000-fold less potent of a mitochondrial respiration inhibitor than antimycin A and retains anti-cancer activity.

 

We designed analogues of antimycin A using available structure-activity relationships to potentially minimize toxicity and increase anti-cancer activity (7). The bislactone core of antimycin A will be replaced by a bislactam core in our analogues.  It is well known that lactones and esters are less hydrolytically stable than peptides.  The bislactone core is expected to be significantly more hydrolytically stable, which confers a longer duration of drug action and lower effective doses.

 

Retrosynthetic Disconnection of Antimycin Bislactam Analogues

 

 

My target library can be retrosynthetically disconnected to a core structure (4), which enables late-stage introduction of C8 esters, C3 amides, and various residues at C7 via an olefinic synthetic handle. The synthesis of 9-membered lactams is a significant challenge and few syntheses have been reported.  Our synthesis will constitute the first 9-membered bislactam synthesized and characterized using modern methods.  The proposed cyclization will be performed on a significantly more complex substrate than previous syntheses of 9-membered aminolactams. The total synthesis of these compounds is currently being developed.  Once the synthesis is completed, the compounds will be evaluated by Dr. Hockenbery at the Fred Hutchinson Cancer Research Center.

 

 

 

 

Synthetic Studies on the Cinchona Alkaloid, Quinine:  Development of a Ketone-enolate Pd-mediated Allylic Alkylation

 

         Malaria continues to be one of the world’s most deadly diseases, killing more than one million people each year.  The spread of drug-resistant strains of malaria has rendered the available treatments ineffective on their own in many parts of the world.  Quinine, the first known treatment for malaria, inhibits the growth of malaria parasites, including Plasmodium falciparum, which causes the most severe form of the disease.  Some regions have returned to using quinine. Despite synthetic interest in quinine for over 100 years, the first stereocontrolled synthesis was reported in 2001 by Stork and co-workers.

 

         We are interested in developing a versatile synthesis of quinine.  Makoto Mori, a former researcher in our laboratory, who was visiting from Sankyo CO, Ltd., and I are investigating a synthetic strategy that involves a novel disconnection of the quinuclidine core that differs from the classic disconnection accomplished asymmetrically independently by Jacobsen and Kobayashi in 2004.

 

Retrosynthetic Disconnection of Quinine

 

 

 

         We developed the first synthetic approach to quinine via a stereocontrolled C3-C4 ring closure reaction to construct the quinuclidine azabicyclo[2.2.2]octane ring system. Another novel aspect of our approach is the establishment of the C8 and C9 stereogenic centers early in the synthesis through the use of an asymmetric aldol reaction.

 

Synthesis of (R)-7-hydroxy-quinine

 

 

         Our modified Pd-mediated allylic alkylation enables a TMS-enol ether (pre-formed ketone enolate equivalent) to participate as the nucleophile in an intramolecular cyclization. Surprisingly, none of the undesired C3-vinyl stereoisomer was observed from the cyclization. Immediate reduction of the quinuclidinone ketone product to avoid equilibration at C8 and β-elimination, yields an alcohol product that exists predominantly as two rotamers about C9 as determined by rOesy NMR and confirmed by variable temperature NMR. The standard π-allyl Pd-based allylic alkylation mechanism fails to explain the observed stereoselectivity. We have envisioned an alternative mechanistic explanation that can only result in a single, observed and desired stereoisomer.  It involves Pd-mediated etherification, followed by a Claisen rearrangement. Finally, silyl deprotection afforded (R)-7-hydroxy-quinine. This diol and 8-dehydro-7-keto-quinine were both found to be inactive against Plasmodium falciparum HB3 and Dd2 strains.

 

“Synthetic Studies on Quinine: Quinuclidine Construction via a Ketone Enolate Regio- and Diastereoselective Pd-mediated Allylic Alkylation.” Johns, D. M.; Mori, M.; Williams, R. M.  Org. Lett. Accepted for publication. No. ol0615424s.

 

 

 

Total Synthesis of HDAC Inhibitor FK228

 

The depsipeptide, FK228 (1) (previously named FR-901228) exhibits potent antitumor activity against a range of murine and human solid tumor cells (IC50 = 0.02-0.2 nM, synovial carcinoma cells). It is a potent Zn-dependent histone deacetylase (HDAC) inhibitor.

 

Structures of Depsipeptide FK228 and a Related Depsipeptide FR-901375

 

 

The structure of FK228 consists of a 16-membered cyclic depsipeptide and a 15-membered macrocyclic dithiane.  Retrosynthetic disconnection of FK228 reveals a tetrapeptide portion and a β-hydroxy mercapto heptanoic acid, which I prepared in 10 steps (10% overall yield) from commercially available materials.

 

Retrosynthetic Disconnection

 

The β-hydroxy mercapto acid was coupled with the tetrapeptide portion (prepared by Dr. Yasuo Noguchi a former visiting scientist to our group from Sankyo Co., LTD) after N-Fmoc deprotection.  Macrolactonization using Mitsunobu conditions described by Simon and co-workers provided the 16-membered cyclic depsipeptide.  The total synthesis of FK228 was completed in 22 total steps (2.2% yield) from commercially available materials.

 

Completion of FK228 Synthesis