1. Introduction
Antisense technology is one of the therapeutic platforms to regulate gene expression in vivo1. The synthetic antisense oligonucleotides would influence gene expression and inhibit protein synthesis. Proteins have a crucial role in all the cellular processes of human metabolism. The majority of human diseases are the result of inappropriate protein production or disordered protein performance2. To inhibit the production of disease-causing proteins, antisense technology based drugs could be designed to bind sequence specifically to concerned mRNA. A wide spectrum of diseases including infectious, inflammatory, cancer, and cardiovascular diseases can be treated by specially designed and synthesized antisense drugs. Antisense oligonucleotides (ASOs) are short chemically modified oligonucleotides that bind to their complementary mRNA by Watson-Crick base-pairing and modulate its function3.  The ASOs can be more effective and less toxic in targeting any disease because they can have the capacity to bind sequence specifically to the target mRNA. Single-stranded DNA based oligonucleotides activate the RNase H antisense mechanism4. The RNase H1 enzyme is ubiquitously expressed which cleaves the RNA strand selectively from the RNA/DNA hetero duplex5. The investigation of therapeutic applications of antisense oligonucleotides against various diseases is moving at a fast pace6. There are six antisense technology based drugs namely Fomivirsen7,8, Pegaptanib6, Mipomersen9, Eteplirsen9, Defibrotide9 and Nusinersen9 that have been approved by the FDA from 1998 to 2016. It has been observed in recent years that there is a rapid increase in the number of antisense molecules entering into phase III clinical trials8. In 1970, first time Zamecnik and Stephenson proposed ASOs as therapeutic agents10. The standard unmodified nucleic acids have confined stability in biological media and undergo rapid degradation by nucleases11,12. Chemical modifications would be required to protect the oligonucleotides from the cellular nucleases, enhance the stability, improve binding affinity for the target RNA, and improve pharmacokinetic properties in animals to elicit a functional antisense response. The protective modifications could be introduced at three different sites on the nucleotide13. The nitrogen base can be altered or changes in phosphate backbone can be made for DNA and RNA nucleotides. Apart from these two, in RNA nucleotides, the 2’ hydroxyl group also can be modified. These modifications are categorized into three generations.
First generation antisense modifications are majorly backbone based modifications like phosphorothioates, methyl phosphonates, and phosphoramidates modifications. One of the non-bridged oxygen attached to the phosphate is replaced by sulfur in the phosphodiester backbone of the nucleotide (Phosphorothioates), a methyl group (methylphophonates) and amines (phosphoramidates). In all these backbone modifications, phosphorothioates (PS) are successful and are widely used for gene-silencing because of their resistance against nucleases and the ability to induce the RNase H functions14. However, the binding affinity to the target sequences, specificity and cellular uptake profiles of phosphorothioates are less satisfactory15. The issues raised with first generation antisense modifications are solved up to some extent with second generation modifications that are majorly based on sugar based modifications. The 2’-O-methyl (OMe) and 2’-O-methoxyethyl (MOE) are well explored and important members of second generation modifications. These sugar based modifications OMe and MOE can be further combined with the phosphorothioate backbone linkage16. It was reported that the antisense oligonucleotides having the 2’-O-methoxyethyl modification are less toxic than phosphorothioate antisense oligonucleotides and also shows enhanced affinity towards their complementary RNAs and improved pharmacokinetic properties17,18. To improve the thermal stabilities of the antisense oligomer bound to either complementary DNA or RNA, several nucleic acid analogs have been studied and developed. The third generation modifications are peptide nucleic acids (PNAs), locked nucleic acids (LNAs) or bridged nucleic acids, hexitol nucleic acids, and morpholino oligonucleotides to name a few19-22. PNAs are oligonucleotide analogues in which the sugar-phosphate backbone has been completely replaced by pseudo peptide linkages. This modification offers increased stability and favourable hybridization kinetics. However, these constructs have problems of solubility and delivery difficulties and they can’t activate the RNase H cleavage mechanism23. LNAs are the most promising third generation modifications. LNA nucleotides are a class of nucleic acid analogues in which the ribose ring is locked by a methylene bridge connecting the 2’-O atom and the 4’-C atom. The LNAs show increased thermodynamic stability and improved nucleic acid recognition24,25. Apart from these generations of modifications, several nucleobase based modifications, backbone based modifications, furanose sugar based modifications, six-membered ring analogues, bicyclo and tricyclo modifications, constrained nucleic acids, etc. have been designed by medicinal chemists26. The novel “chimera” modifications in which more than one modification has been made to improve the nuclease resistance and target binding affinity as well. For example, single nucleotide modification can have phosphorothioate backbone modification and MOE sugar based modification. Another advanced strategy being explored recently is the concept of gapmer design. Gapmer is a designed antisense oligomer strand in which both the ends (2-5nt) have particular modification for example MOE or LNA to increase binding affinity, improve pharmacokinetic properties and in middle region phosphorothioate modification to increase nuclease resistance and activate RNase H activity27-31.
The importance of antisense technology and other oligonucleotide based therapeutics is continuously increasing. Several studies are being carried out by various labs to design novel drugs against some critical diseases. There is a strong need to identify or design novel modifications to improve the performance of existing antisense oligonucleotides32. Various modified antisense oligonucleotides are used in targeting different regions of the mRNA of disease-causing genes. However, the structural information about these antisense modifications is limited. The structural parameters calculated through computational methods at the quantum mechanical level or molecular dynamics level of the modifications are very useful for understanding the mechanism of action of these modifications and also help in designing novel modifications. The quantum chemical and molecular dynamics simulations studies of some of the antisense modifications have been reported in the literature33-36. Many antisense molecules are available in the market as FDA approved drugs, where PS, MOE, LNA modified molecules are being used. There is a tremendous scope to develop new modifications that can have strong binding affinity with DNA/RNA and less toxicity. In the present study, five new antisense modifications have been proposed. All the modifications have been introduced in the cytidine nucleotide for the monomer level study. This has been done to understand the effect of particular modification considering any one nucleobase (here cytosine) at monomer level by comparing it with its standard nucleotide with the same nitrogen base. The efforts of quantum chemical studies at the monomer and base-pair level would be helpful to understand the various properties of the modifications in order to identify better modifications. Modification of nucleic acids can affect chemical stability, pairing, and conformation. The structures of these proposed modifications are designed based on the LNA structure. LNA has strong binding properties but higher toxicity compared to MOE37-40. The proposed modifications are designed by changing various atoms or by adding different oxy or nitro groups to the LNA. The backbone and sugar based chemical modifications are known to improve the pairing affinity, metabolic stability, and cellular uptake of RNA. These parameters are absolutely essential in the discovery and development of highly active oligonucleotide / RNA based therapeutic candidates. To enhance the potency and reduce the potential toxicity of antisense oligomers, numerous chemical modification geometries have been established and tested. The basic idea behind proposing these modifications is to increase the binding affinity as well as to reduce the toxic effects and improve the pharmacokinetic properties. The proposed modifications are designed in such a way that they share the structural components from LNA, MOE and other electronegative groups / bulky groups attached to methylene bridge carbon of LNA. In the proposed modifications, the methylene bridge carbon is replaced with nitrogen in A1 and A2 modifications. In A2, an extra methoxy group is added to the bridged nitrogen. For A3, A4 and A5 modifications, dimethyl amine group, amine group and dihydroxy amine group are added to the bridged carbon respectively. For proposed modifications, the methoxy group and other amine groups are added because they are already well established as 2’-O sugar modifications and incorporation of these proved to show good pharmacokinetic properties17,18,41. In all proposed modifications, the electronegative groups and other small chemical groups were added to the basic LNA structure to improve the pharmacokinetics and reduce the toxicity without disturbing the strong binding nature of LNA. All these proposed modifications are studied thoroughly through quantum calculations. The 2D structural representation of all the modifications used in the current study has been given in Figure 1 and the modification details have been given in Table 1. The study also aims to identify the stable conformations of the antisense molecules selected from the literature and the proposed novel modifications. Comparison has been made in terms of quantum chemical descriptors which helps to evaluate the different standard antisense modifications and the proposed new modifications.
The conformational search and optimization of all modifications at the monomer level was carried out and identified the most stable conformation for each modification by using in-house developed conformation generation and optimization tool TANGO42. The quantum chemical calculations have been carried out for the most stable conformations of all the modifications and derived various quantum chemical descriptors. This study may help in understanding the structural and functional significance of these novel antisense modifications in exhibiting lower toxicity with higher binding affinity and increased potency. Structural insight and information regarding various quantum chemical properties could be useful in their functional understanding, which may guide in the design of better modifications.