Noteworthy Chemistry

April 14, 2014


This prodrug releases cisplatin and aspirin. Inflammation plays an important role in the progression of tumors, so a combined chemotherapy and anti-inflammatory strategy against cancer is desirable. Administering formulations that contain two free drugs, however, would likely result in pharmacokinetic and biodistribution problems. S. Dhar and co-workers at the University of Georgia (Athens) developed platin-A, a prodrug that simultaneously releases cisplatin and aspirin.

Platin-A (1), a Pt(IV) compound, is prepared by the reaction of cisplatin with “aspirin anhydride” (2-acetoxybenzoic anhydride). This compound is kinetically more stable than the labile cisplatin. 

Biodegradation of platin-A

The authors tested 1 against androgen-responsive LNCaP and androgen-unresponsive DU145 prostate cancer cells. The compound was reduced in the tumor microenvironment to release cisplatin (2) and aspirin (3). In cancer cells, with an intracellular pH value of >7.4, the aspirin hydrolyzed to salicylate (4).

Cisplatin induces apoptosis in the nuclear DNA; aspirin inhibits the cyclooxygenase (COX) enzyme; and salicylate controls the level of inflammatory responses in tumor-associated macrophages. The cytotoxicity of compound 1 is comparable with cisplatin.

This platin-A study highlights the importance of combination anticancer and anti-inflammatory drugs to increase potency and reduce side effects. (Angew. Chem., Int. Ed. 2014, 53, 1963–1967; José C. Barros)

Back to top

Optimize Migita and Heck couplings in the synthesis of axitinib. Axitinib is a vascular endothelial growth factor (VEGF) antagonist that is being developed as a cancer drug. A Migita coupling of 6-iodoindazole and 2-thio-N-methylbenzamide, the first step in the synthesis of Axitinib, is catalyzed by Pd-XantPhos. [The ligand XantPhos is 4,5-bis(diphenylphosphino)-9,9-dimethylxanthene.]

R. A. Singer and co-workers at Pfizer (Groton, CT) found that the order of reagent addition and a completely inert reactor environment are critical to the success of this reaction. If the methylbenzamide is added too early, it forms a catalytically inactive complex with palladium. Similarly, if the reactor space is not completely inert, dissolved oxygen oxidizes the thiol group to the disulfide, which undergoes oxidative addition to palladium and inactivates it. Slow addition of the benzamide to the other reactants gives the best results.

In the subsequent Heck coupling (carried out after the Migita coupling product is iodinated), Pd-XantPhos is again the catalyst of choice. Complete conversion is achieved over 24 h, compared with 10–20% conversion in the same time in the absence of the ligand. The choice of an amine base is also critical: i-Pr2NH gives full conversion over 24 h. If bases such as n-Bu3N are used, 24-h conversion drops to 75%; with N-ethylmorpholine or 4-dimethylaminopyridine, conversion drops to as little as 34%. (Org. Process Res. Dev. 2014, 18, 266–274; Will Watson)

Back to top

Why do some crystals have nongrowing faces? Chemists have long known that the growth rates of different crystals with the same composition can differ, a phenomenon called growth rate dispersion. Although many factors contribute to this effect, the detailed mechanism for growth rate dispersion remains elusive. One extreme case of growth rate dispersion is the formation of nongrowing faces.

Examples of crystals with different growth rates

M. M. Mitrović and coauthors at the University of Belgrade (Serbia) and the University of Banja Luka (Bosnia and Herzegovina) studied the nongrowing {100} faces of NaClO3 crystals in supersaturated aqueous solutions. Because nongrowing faces are very stable, the probability of observing them is not affected by dissolution or refaceting. In highly supersaturated solutions, however, otherwise nongrowing faces began to grow randomly.

Based on this observation, the researchers excluded microscopic conditions around the crystals as causes of nongrowing faces. Instead, they identified the microstructure of crystal faces as an important factor because growth did not occur on faces that were free of dislocation sources or had sources of opposite signs. Immobile impurities also may be responsible for nongrowing faces because they may block step spreading and incoming growth units. (Cryst. Growth Des. 2014, 14, 972–978; Xin Su)

Back to top

A single amino acid change makes avian H7N9 influenza more severe in humans. The avian influenza A virus strain H7N9 that caused serious illness and death in eastern China was identified in March 2013. By October, 137 human cases had been identified; 45 of these ended in death. Additional cases continue to be reported. Infected patients typically develop a rapidly progressing pneumonia, which leads to respiratory failure and acute respiratory distress syndrome.

The H7N9 strain crosses easily from birds to humans because mutations in the viral hemagglutinin protein allow the virus to bind avianlike and humanlike receptors in the human respiratory tract. Analysis of the H7N9 genome showed that all eight of the flu gene segments are of avian origin. Six come from H9N2 viruses; the two surface proteins are derived from other avian subtypes.

Amino acid changes in polymerase basic protein 2 (PB2) occur in H7N9 viruses isolated from humans and are believed to be a result of adaptive changes to the virus’s new mammalian host. Avian viruses have glutamic acid (E) at PB2 residue 627, but 13 out of 16 H7N9 mammalian viral isolates from the recent outbreak had lysine (K) at that position. This lysine substitution also is found in human isolates of highly pathogenic avian viruses such as H5N1 and H7N7.

Mutations such as this one can play a role in enhancing avian viral replication efficiency and virulence in mammals. To understand the importance of the E627K mutation in PB2 to the pathogenicity of H7N9 mammalian infection, J. S. M. Peiris and colleagues at the University of Hong Kong generated recombinant viruses with individual point mutations in the PB2 gene. They then studied the effects of these mutations on polymerase activity, viral replication, and pathogenicity.

The authors determined that the human H7N9 virus with the E627K mutation is more pathogenic than avian isolates without the mutation. The mutated viruses exhibit higher polymerase activity, replication competence, pro-inflammatory cytokine expression, and disease severity than their nonmutated counterparts. (J. Virol. 2014, 88, 3568–3576; Abigail Druck Shudofsky)

Back to top

Vary the number of fused benzene rings to tune light emission. Luminescent molecules usually consist of rigid polynuclear aromatic rings; luminophores with flexible alicyclic rings are rare. S. Irle, S. Yamaguchi, and coauthors at Nagoya University (Japan), the Japan Science and Technology Agency (Nagoya), the University of Costa Rica (San Pedro Montes de Oca), and the Institute of Transformative Bio-Molecules/JST-CREST (Nagoya) synthesized a series of compounds (13) that have flexible cyclooctatetraene cores and rigid aceneimide wings. They found that the molecules’ luminescence properties are strongly influenced by the length of their aceneimide “wings”.

Compound 1 contains single benzene rings in its two wings. Neither its solution nor its solid emit light upon photoexcitation. When the wings contain two fused benzene rings as in 2, the solid powder is luminescent, but the dilute solution is nonemissive.

When the number of fused benzene rings is increased to three (compound 3), the solution and solid are luminescent. This remarkable tunability of light-emitting behavior may have uses in the design of multifunctional organic electronic materials. (Chem.—Eur. J. 2014, 20, 2193–2200; Ben Zhong Tang)

Back to top

“Separate” enantiomers with NMR spectroscopy. When NMR spectroscopy is used to characterize enantiomers, chiral auxiliary reagents are required to isolate the signals from each enantiomer. But the use of chiral auxiliaries often leads to unsatisfactory signal resolution of because of overlapping peaks. This drawback greatly limits the scope of NMR in chiral analysis.

To circumvent such problems, Lokesh, S. R. Chaudhari, and N. Suryaprakash at the Indian Institute of Science (Bangalore) developed a 2-D 1H NMR experiment that allows overlapped signals to be resolved and enantiomeric signals in a coupled-spin system to be separated. They call their technique resolved-total correlated spectroscopy, or RES-TOCSY.

The authors’ RES-TOCSY experiments start with

a semi-selective π/2 pulse, which excites selective protons (Hi) in each enantiomer/diastereomer followed by t1 evolution. During t1 evolution the combined use of a selective π pulse and a hard π pulse causes decoupling of Hi with the remaining protons, which allows the evolution of a chemical shift of Hi only. The consequence of this is the separation of enantiomer/diastereomer peaks (ΔδiR,S) of Hi in the indirect dimension. During isotropic mixing the separated proton magnetization is transferred to all the other coupled protons of the respective enantiomer/diastereomer, resulting in evolution of both couplings and chemical shifts in the direct dimension.

A 2-D RES-TOCSY spectrum yields a separate set of chemical shift signals for each isomer.

The authors tested their RES-TOCSY experiment for several common uses, including resolving enantiomers in the presence of starting materials and resolving spectra with significant broadening and overlapping. They also used RES-TOCSY quantitatively for accurate enantiomeric excess measurements.

This technique has better inherent resolution than the J-resolved method. It not only produces unambiguous assignments of peaks in enantiomers or diastereomers, but it also extends the scope of NMR for qualitative and quantitative characterization of chiral compounds. (Org. Biomol. Chem. 2014, 12, 993–997; Xin Su)

Back to top

Mediate bromodifluoroallylation with palladium. Fluorinated compounds are found in pharmaceuticals, agrochemicals, and modern materials; but introducing the difluoromethylene group (CF2) into organic molecules has not been fully explored. The CF2 group can be seen as a bioisostere of oxygen or carbonyl in biological environments. Current difluorination methods require the use of toxic Et2NSF3 or deoxofluor.

X. Zhang and co-workers at the Chinese Academy of Sciences (Shanghai) developed a method for difluoroallylating organoborons that uses readily available 3-bromo-3,3-difluoropropene (BDFP, 2) and palladium catalysts. The authors used 4-tert-butylphenylboronic acid (1) as their model substrate and found that its reaction with Pd2(dba)3 (0.4 mol%) and K2CO3 (≈3 equiv) in water–dioxane at 80ºC gave a 93% yield of a mixture of α- and γ-substituted products with high selectivity. (The ligand dba is dibenzylideneacetone.) The α/γ ratio ranged from 3.2:1 to >20:1. Product 3 in the figure is the α-isomer.

Nonemissive to fully luminescent fused ring systems
Bromodifluoroallylation of an areneboronic acid

The authors expanded the method to several functionalized areneboronic acids, an organoborate, and a potassium trifluoroborate salt. They then scaled the reaction up to 10 g and obtained an 80% yield; in this case, Pd(PPh3)4 (0.01 mol%) was the best catalyst.

Substituted BDFP reagents were used to obtain compounds that are difficult to prepare otherwise. This method of gem-difluoroallylation requires low catalysts loadings, is regioselective, and has excellent functional group tolerance. The authors believe that this method will be useful in drug discovery and development. Further studies are under way to discover the reaction mechanism and to develop additional derivative reactions. (J. Am. Chem. Soc. 2014, 136, 1230–1233; José C. Barros)

Back to top

What do you think of Noteworthy Chemistry? Let us know.