Current Topics in Medicinal Chemistry - Volume 7, Issue 7, 2007

It gives me great pleasure to forward this special theme issue devoted to the topic, ‘Modified Nucleic Acids as Substructures in Medicinal Chemistry and Drug Development’. With the discovery of siRNA being recognized by Nobel Prize in medicine in 2006, the topic warrants such attention. The collection of papers, contributed by several leading researchers, illustrates the generation of therapeutically driven chemical entities as mimics of nucleic acids. The common objective in each case is the evolution of nucleic acids into therapeutically viable oligonucleotide (ON) structures and to find some straightforward answers to several technical challenges. Each contributor has surpassed milestones on diverse roadmaps that lead to a common destination i.e application of antisnese (AS) principle for the therapeutic solution to several untreatable diseases. The authoritative articles by the contributors summarize their own research work and also aptly take note of the current relevant literature. Prakash and Bhat focus on the progress made in RNase H mediated antisense approach, Summerton comprises the morpholinos, their application and comparison with S-oligos, Koizumi details the ONs with restricted geometry in N-sugar conformation. Peptide nucleic acid, PNA, the highly competent DNA mimic is further exploited in application perspective. We have two distinct articles dedicated to the nucleobase and backbone PNA modifications by Hudson and Corradini, respectively. The P-chirality is a major issue in the phosphodiester modifications in ONs and Guga has extensively reviewed this topic for both chemistry and biology. Ganesh and myself divulge upon the literature that highlights nucleic acid structure-editing for RNA selectivity and the logic behind such selection. In addition to the knockdown/down-regulating AS therapeutics, the corrective AS principle needs such selection where small molecular therapies are not common. The important bottleneck for application of modified ONs as medicines is their cellular delivery. Lebleu and co-authors have given sufficient attention to the various ways of realizing this in their article. Considering the nature of the special issue, there are certain quite understandable overlaps, but authors have treated the topics in different perspectives. The issue thus covers concise efforts towards a common goal. It is interesting to see that we have articles from Canada, France, India, Italy, Japan, Russia, U.K. and U.S.A. that also incidentally represent all the continents. I hope this gives a comprehensive picture of today's modified oligonucleotide substructures and to some extent their biological evaluation. I trust that the readers will enjoy the special issue and it will be beneficial for the future research initiatives. I owe all the contributors special thanks for their efforts. I am personally grateful to Bentham Science Publications for giving me this opportunity to bring out this issue.

Chemically modified antisense oligonucleotides are currently progressing in multiple clinical trials. Among several chemical modifications made, modification of the 2'-position has proved most successful. Second generation antisense oligonucleotides incorporating these 2'-modifications exhibit high binding affinity to target RNA, enhanced metabolic stability, and improved pharmacokinetic and toxicity profiles. This is, in part, due to the enhanced biophysical properties of second generation antisense oligonucleotides. 2'-Modifications that influence the sugar to adopt a 3'-endo sugar pucker can improve properties such as affinity. 2'- Modifications that provide a gauche effect and/or a charge effect can play a significant role in the level of nuclease resistance. The heterocyclic base modifications such as 2-thiothymine provides additive effect on the affinity of 2'-F and 2'-O-MOE modifications. This review summarizes the structural and biophysical properties of selected 2'-modified nucleosides which are candidates for use in oligonucleotide theraputics.

Generally a gene knockdown agent should achieve high sequence specificity and should lack off-target effects (non-antisense effects due to interactions with structures other than gene transcripts). Three major gene knockdown types are compared with respect to offtarget effects and sequence specificities: 1) phosphorothioate-linked DNA (S-DNA); 2) short interfering RNA (siRNA); and, 3) Morpholino. S-DNAs cause multiple off-target effects, largely because their backbone sulfurs bind to many different proteins. S-DNAs also achieve poor sequence specificity because S-DNA/RNA duplexes as short as 7 base-pairs are cleaved by RNase H. siRNAs cause several off-target effects, but improved designs may soon avoid such effects. siRNAs also provide only limited sequence specificity because their short guide sequences largely determine which gene transcripts will be blocked or cleaved, and those guide sequences appear to recognize insufficient sequence information to uniquely target a selected gene transcript. This specificity limitation is inherent in their mechanism of action and so probably cannot be greatly improved. Morpholinos are virtually free of off-target effects - probably because they cannot interact electrostatically with proteins. Morpholinos also achieve exquisite sequence specificity - in large part because they must bind at least about 14 to 15 contiguous bases to block a gene transcript, and this constitutes sufficient sequence information to uniquely target a selected gene transcript. Because of their freedom from off-target effects, exquisite sequence specificity, complete stability in biological systems, and highly predictable targeting, Morpholinos dominate the most demanding of all gene knockdown applications, studies in developing embryos.

As first-generation antisense oligonucleotides, more than a dozen phosphorothioate oligodeoxynucleotides (PS ODNs) have been clinically developed, but only one has reached the market. To improve the drawbacks of PS ODNs, such as low affinity to target mRNA and non-specific binding to proteins, modified oligonucleotides with 2'-modified sugars such as 2'-O-(2-methoxy)ethyl and 2'-F modification or with bridged sugars such as oxyalkylene linkages between 2'-oxygen and 4'-carbon, have been synthesized as 2'-MOE, 2'-F RNA, 2',4'-BNA/LNA and ENA oligonucleotides. They have shown properties of higher affinity to complementary single-stranded RNA and DNA than those of PS ODNs due to their preorganized N-conformation. On the basis of the properties of these newly designed oligonucleotides, their in vitro and in vivo applications for gene silencing as true antisense oligonucleotides have been reported. In this review, antisense applications with these modified oligonucleotides are focused on.

Peptide nucleic acid (PNA) is an oligonucleotide mimic originally designed upon a repeating N-(2-aminoethyl)glycine polyamide backbone to which nucleobase heterocycles are attached through a methylene carbonyl linkage to the α-amino group. These molecules possess remarkable hybridization properties with DNA or RNA forming complexes with high stability and with excellent sequence discrimination despite the substantial structural divergence from natural nucleic acids. Since the disclosure of PNA, a vibrant research community with interest in the chemistry and applications of polyamide-based nucleic acid analogs has developed. This has led to the synthesis and evaluation of a wide variety of modified polyamide nucleic acids. The focus of this report is a comprehensive review of nucleobase modifications in aminoethylglycine (aeg) PNA with reference, where appropriate, to the same modification in DNA or RNA.

Peptide nucleic acids (PNAs) are polyamidic oligonucleotide analogs which have been described for the first time fifteen years ago and were immediately found to be excellent tools in binding DNA and RNA for diagnostics and gene regulation. Their use as therapeutic agents have been proposed since early studies and recent advancements in cellular delivery systems, and in the so called antigene strategy, makes them good candidates for drug development. The search for new chemical modification of PNAs is a very active field of research and new structures are continuously proposed. This review focuses on the recent advancements obtained by the modification of the PNA backbone, and their possible use in medicinal chemistry. In particular two classes of structurally biased PNAs are described in details: i) PNAs with acyclic structures and their helical preference, which is regulated by stereochemistry and ii) cyclic PNAs with preorganized structures, whose performances depend both on stereochemistry and on conformational constraints. The properties of these compounds are discussed in terms of affinity for nucleic acids, and several recent examples of their use in cellular or animal systems are presented.

Internucleotide phosphodiester linkages in non-modified oligonucleotides are quickly degraded by nucleolytic enzymes present in the cells and this feature practically eliminates natural DNA and RNA molecules from medical applications and from many structural and mechanistic studies. P-chiral oligonucleotide analogs, in which one of the non-bridging phosphate oxygen atoms is substituted with another heteroatom (e.g. S, Se) or a chemical group (e.g. CH3, BH3 -), have significantly greater nuclease resistance and also offer important possibilities for detailed studies of interactions with other biomolecules at the molecular level. Notably, these substitutions do not disrupt hydrogen bonding between nucleobases and affect the overall geometry of the oligomers to only low or moderate extent, although important changes of hydration patterns and changes of interactions with metal ions are observed. Such the probes, including isotopomeric species labeled with a heavy oxygen isotope, possessing phosphorus atoms of selected absolute configurations, have been used for elucidation of the mode of action of many enzymes (nucleases, transferases, kinases), ribozymes and DNA-zymes, as well as for investigations on thermodynamic stability of nucleic acids complexes (duplexes, triplexes, i-motif) and for studies on a mechanism of conformational changes of B-Z type. They are also useful tools for analysis of interactions of the phosphoryl oxygen atoms in natural precursors with functional groups of proteins. The synthetic routes to stereodefined forms of selected types of Pchiral oligonucleotides are presented, as well as recently developed methods for their configurational analysis at micromolar concentration. Selected examples of application of diastereomerically pure P-chiral oligonucleotides for structural, biochemical and biological experiments are discussed.

The synthesis of backbone-modified nucleic acids has been an area of very intense research over the last two decades. The main reason for this research activity is the instability of nucleic acid based drugs in the intracellular conditions. Changes in the sugarphosphate backbone invariably bring about the changes in the complementation properties of the nucleic acids. The naturally occurring deoxyribose- (DNA) and ribose (RNA) sugar-phosphate backbones are endowed with considerable differences in their binding affinities towards themselves. This occurs because of the different sugar conformations prevalent in DNA and RNA and the subtle structural changes accruing from these in hydrogen bonding, base-stacking interactions and hydration of major/minor grooves. The six-atom phosphodiester linkages and pentose-sugars give immense opportunities for chemical modifications that lead to several backbonemodified nucleic acid structures. This article is focused on such modifications that impart RNA-selective binding properties to the modified nucleic acid mimics and the rationale behind the said selectivity. It is found that the six-atom sugar-phosphate backbone could be replaced by either one-atom extended or one-atom edited repeating units, leading to the folded or extended geometries to maintain the internucleoside distance-complementarity. Other important contributions come from electronegativity of the substituent groups, hydration in the major/minor groove, base stacking etc.

Specific control of gene expression by synthetic oligonucleotides (ON) is now widely used for target validation but clinical applications are limited by ON bioavailability. Moreover, most currently used strategies for physical and chemical delivery cannot be easily implemented in vivo. This article reviews new strategies which appear promising for ON delivery. The first part deals with ON chemical modifications aiming at improving cellular uptake as for instance the grafting of cationic groups on the ON backbone. The second part concerns ON conjugation to cell penetrating peptides.

Emerging from the shadow of Vioxx™. Nonsteroidal antiimflammatory drugs (NSAIDs) are a class of medicines prescribed to treat acute and chronic pain [1]. NSAIDs reduce prostaglandin production by inhibition of cyclooxygenase (COX) to afford analgesia. However, while very effective for the treatment of pain and inflammation, this mechanism elicits adverse gastrointestinal (GI) and renal effects. The discovery of COX isoforms (COX-1 and COX- 2) and differential functions led to the development of selective COX-2 inhibitors for the treatment of pain that minimized the adverse GI effects [1]. In 1999 Pfizer launched Celebrex™ (30-fold selective COX-2 inhibitor) and Merck launched Vioxx ™ (272-fold selective COX-2 inhibitor) [2, 3]. It has been 28 months since Merck & Co. announced the results of the APPROVe trial and withdrew Vioxx™ from the market due to adverse cardiovascular events such as heart attack and stroke; as a result, over 27,000 lawsuits have been filed against the pharmaceutical giant [3]. The cardiovascular risks stems from the selective inhibition of COX-2 which reduces COX-2-mediated vasodilation and inhibition of platelet aggregation. Unlike the nonselective NSAIDs, which balance COX-1-mediated platelet activation with COX-2 vasodilation and platelet inhibition, selective inhibition of COX-2 with drugs like Vioxx may generate an imbalance that predisposes platelet aggregation and ischemia [1]. Despite the potential risks, arthritis patients clamored for the drug, as for many patients, only Vioxx™ relieved their arthritis pain. In November of 2006, Merck announced that it received an approvable letter from the FDA for Arcoxia™, a second generation COX-2 inhibitor with even greater COX-2 selectivity (344-fold) [1, 3]. Also in November of 2006, Merck announced the results of a large outcomes study, coined the MEDAL (Mulitnational Etoricoxib and Diclofenac Arthritis Longterm) program, which demonstrated that Arcoxia™ had similar rates of cardiovascular thrombotic events as Diclofenac™, the most prescribed traditional NSAID in the world [3]. In addition, the MEDAL study showed that Arcoxia™ had a lower rate of confirmed upper GI effects (including ulcers, bleeding and obstructions) than Diclofenac™. However, a recent article in the Wall Street Journal raised issues with the MEDAL study [4]. In the article, some doctors were quoted as saying that the study was of limited scope and questioned the comparison of Arcoxia™ to DiclofenacTM, which ‘...acts like a COX-2 inhibitor in the first place.’ They advocated comparing Arcoxia™ to a less-similar painkiller such as naproxen. Despite these issues, Arcoxia™ is currently available in 62 countries around the world, but sales in the US are pending an FDA decision that is expected in April of 2007 [3]. REFERENCES [1] Martina, S.D.; Vesta, K.S.; Ripley, T.L. Etoicoxib: A highly selective COX-2 inhibitor. Ann. Pharmacother. 2005, 39, 854-862. [2] For detailed information and press releases describing Celebrex™, see www.pfizer.com [3] For detailed information and press releases describing Vioxx™, Arcoxia™, the APPROVe and MEDAL studies see www.merck.com [4] Whalen, J. ‘Drug makers try to bring back Cox-2 inhibitors’ in the Wall Street Journal, Eastern edition, Jan. 19, 2007, B.1.