Designing Antisense-oligonucleotides

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Designing Antisense-oligonucleotides

ASOs are generally made up of 15 to 20 nucleotides which are complementary to the mRNA against which they are targeted. Long RNA molecules tend to form secondary and tertiary structures. So it is imperative to first identify target sites of the mRNA that can be accessed by the ASO. It is thought that on average only one in eight ASOs binds effectively and specifically to a target mRNA. However, the percentage of active ASOs varies from one target to another. We can test many oligonucleotides for their antisense efficiency. But less tedious and more sophisticated methods for systematic optimization of the antisense effect are available (Stein, 2001).

Computer generated structural models of long RNA molecules cannot accuratelt represent the RNA structure found inside a living cell and hence has been of only limited use in designing efficient ASOs. As a result many strategies have been developed for achieving this purpose. Valuable information regarding the accessible sites can be obtained by using random or semi-random oligonucleotide libraries and RNase H which is followed by primer extension (Ho et al., 1998). Non-random oligonucleotide libraries where template DNA is digested to generate target specific ASOs have also been developed. Screening many specific oligonucleotides against the transcript when done in the presence of RNase H is a simple and direct method which provides comparable information regarding the structure of the target RNA and helps in evaluating the extent of cleavage that individual oligonucleotides induce (Kurreck et al., 2002). A DNA array designed to map hybridization sites of ligonucleotides represents the most sophisticated approach employed thus far. In vitro transcribed RNA molecules tend to differ in structure from those inside biological systems. Also certain target sites are shielded by RNA binding proteins. So it is advantageous to screen the efficiency of oligonucleotides in cell culture or cell extracts (Tu et al., 1998).

When designing ASOs some precautions have to be taken. ASOS that contain four contiguous guanosine residues is to be avoided since they may form G-quartets via formation of Hoogsteen base pair that decreases the available oligonucleotide concentration and produce undesirable side-effects. This problem may be oversome by using modified guanosines like 7-deazaguanosine which are incapable of forming Hoogsteen base pairs (Stein, 2001).

Oligonucleotides that contain CpG motifs are not to be used in in vivo experiments as this particular motif stimulates immune responses in mammalian systems. Bacterial and viral DNA exhibit CG dinucleotide more frequently than human genome and hence the immune system uses it as a marker to signify infection (Dove, 2002).

In order to prevent significant homology with non-target mRNAs, performing a database search for oligonucleotide sequence is necessary.

Structural Modifications of Oligonucleotides:

The affinity of antisense oligonucleotides to the target mRNA is directly proportional to the stability of the nucleic acid hybrid. Higher the affinity more is the gene repressing activity. Unmodified DNA and RNA are rapidly degraded by endogenous nucleases and hence have very short in vivo half life. Also they have poor pharmacokinetics properties and distribute poorly in vivo which makes them undesirable and unacceptable therapeutic agents for systemic administration. Many efforts have been made to chemically modify DNA or RNA to make them resistant to degradation by nucleases and also to increase their intrinsic affinity to complementary target mRNAs. These modifications have focused on antisense oligonucleotide backbone, phosphodiester bond and sugar moiety. These efforts have produced three generations of nucleic acid analogs (Gleave et al., 2005).

Figure : Oligonucleotide modification targets. (P. Toth, 2011)

First-generation ASOs:

Deoxyribooligonucleotides with sulphur atom in place of the non-bridiging oxygen in the nucleotide backbone of the oligonucleotide was produced. This modified the backbone of the oligonucleotide from phosphodiester to phosphorothioate. Phosphorothioates are the main representatives of first generation ASOs and are most widely used even today. They were first synthesized by Eckstein et al. in 1960s and were used as ASOs for the first time by Matsukura et al. to inhibit HIV replication. Phosphorothioates are more stable than unmodified oligonucleotide to the action of nucleases. So their half-life is extended. The half-life of unmodified oligonucleotides is ~1 h while that of phosphorothioates is ~9 – 10 h.

In addition to this, phosphorothioates have advantageous features like the ability to form regular Watson-Crick base pairs and activate RNase H. They can be easily synthesizes and as they carry a negative charge they can be easily delivered inside the cell. They also show favourable pharmacokinetic properties (Kurreck, 2003).

However phosphorothioate backbone possesses certain non-specific effects like interacting with polyanions like heparin binding proteins such as vascular endothelial growth factor and platelet derived growth factor which cause cellular toxicity at high doses. This has even led to cardiovascular collapse and death in experimental animals. They have also been found to cause prolongation of clotting times by temporarily inhibiting the clotting cascade (Toth, 2011). Also the affinity of phosphorothioates towards complementary RNA molecules is slightly reduced as compared to their corresponding phosphodiester oligonucleotides. There is decrease in melting temperature of a heteroduplex by ~0.5 ° C. But this shortcoming is at least in part compensated by greater specificity of hybridization of phosphorothioates.

Figure : First Generation ASO (Toth, 2011)

Second-generation ASOs:

Second generation ASOs succeed in solving some of the problems associated with phosphorothioates. In these the oligonucleotide backbone is modified by adding alkyl groups at the 2’ position of ribose. The alkyl groups are 2’- O-methoxyethyl (MOE) and 2’-O-methyl.

Figure : Second Generation ASOs (Toth, 2011)

They have increased potency, nuclease activity and in vivo half-life and are less toxic as compared to first generation ASOs. However the methy groups render the ASO incapable of inducing RNase H cleavage of the target mRNA and so their antisense effect is only due to steric block of translation. This feature can be advantageously used in alteration of splicing. In this case, as against the typical role of ASOs in which they suppress protein expression, ASOs are used to block a splice site and magnify the expression of a protein variant that is alternatively spliced (Sierakowska et al., 1996).

However, in many cases cleavage of target mRNA by RNase H is desired for increasing antisense potency. For this purpose ‘Gapmer technology’ was developed. ‘Gapmers’, basically a further improvement of second generation ASOs, have their ends modified with 2’MOE while the nucleotide backbone is phosphorothioate, just like first generation ASOs. The 2’MOE ends prevent degradation and improve half-life and binding affinity while the phosphorothioate center allows degradation of the mRNA-antisense complex. Contiguous stretch of a minimum of four to five deoxy residues between the flanking methyl nucleotides was found to be sufficient for activating RNase H in Escherichia coli and humans (Monia et al., 1993).

Figure : Gapmer

Gapmers also help in solving the problem of ‘irrelevant cleavage’. The specificity of ASO is reduced by the many shorter sequences it nests. For example, a 15-mer can be seen as eight overlapping 8-mers which can activate RNase H. Each of them will occur multiple times in the genome and could bind to non-targeted mRNAs inducing their cleavage by RNase H. Gapmers which have a central core of 6-8 oligonucleotides and methyl nucleotides at both ends which are unable to recruit RNase H can be used to eliminate irrelevant cleavage since they will bring about RNase cleavage of only one target sequence (Kurreck, 2003).

Third-generation ASOs:

Chemical modification of the furanose ring of oligonucleotides and also the modifications of riboses and phosphate linkages led to the development of third generation ASOs. They are more resistant to degradation by nucleases & peptidases, have better target affinity as well as improved pharmacokinetic profiles as compared to the previous generations of ASOs.

Third generation ASOs do not activate RNase H and produce their antisense effects by achieving translational arrest through steric hindrance of ribosomal machinery. As they are uncharged molecules, serum proteins which generally bind poly-anions do not bind to them. While this reduces the chance of non-specific interactions, it also leads to its hastened clearance from the body. Furthermore, having electrostatically neural backbones reduces their solubility & makes their uptake difficult (Kurreck, 2003).

Commonly used third generation ASOs are as follows.

  1. Peptide Nucleic Acid (PNA)

In Peptide nucleic acids (PNAs), polyamide linkages replace the deoxyribose phosphate backbone. Nielsen et al. were the first to introduce PNA. They are commercially produced by Applied Biosystems which is based in USA.

PNA-DNA and PNA-RNA duplexes have higher thermal stability and fewer mismatched base pairs. They are also resistant to nucleases and proteases. PNAs can also be used for correcting aberrant splicing. They are non-toxic and have low affinity for nucleic acid binding proteins. However, being electrostatically neutral, they present the problems of inefficient solubility and cellular uptake.

On account of their ability to recognize dsDNA, PNAs can be used to modulate gene expression or induce mutation due to strand invasion of chromosomal duplex DNA.

  1. Locked Nucleic Acid (LNA)

LNA is one of the most promising candidates of the third generation ASOs. In Locked nucleic acid (LNA) a methylene bridge links the 4’ carbon with the 2’ oxygen of the ribose. Wengel and Imanishi laboratories were the first to synthesize LNAs. They are commercially manufactured by Progligo.

Conformational change in the DNA/RNA duplex induced by the introduction of LNA into a DNA is towards the A type helix. As a result the cleavage of target mRNA by RNase H is prevented. Degradation of mRNA can be brought about by chimeric DNA/LNA gapmer containing a contiguous stretch of 7-8 DNA monomers in the center which will induce RNase H activity.

LNA and chimeric DNA/LNA oligonucleotides have increased stability against nucleolytic degradation, potent biological activity, high target affinity and are non-toxic.

  1. Morpholino oligonucleotides (MF)

Morpholino oligomers are non-ionic DNA analogs. In them a morpholino moiety and phosphoramidite linkage replaces the ribose phosphodiester bond. They are commercially produced by Gene Tools LLC, USA.

Inhibition of gene expression can be achieved by targeting the 5’ untranslated region. Alternatively we can also target the first twenty-five bases downstream of the start codon which would prevent ribosomes from binding. They are resistant to nuclease and less likely to interact non-selectively with cellular proteins due to the absence of a negative charge.

  1. Cyclohexene Nucleic Acid (CeNA)

In CeNA a six membered ring replaces the five membered furanose ring. They have a high degree of conformational rigidity and stability against nucleases. Also CeNA/RNA hybrids are known to activate RNase. However, kcat is 600 fold lower as compared to a DNA/RNA duplex.

  1. N3’- P5’ Phosphoroamidates (NPs)

In NPs a 3‘- amino group replaces the 3’- hydroxyl groupof the 2’- deoxyribose ring. NPs are resistant to nucleases and show high affinity towards complementary mRNA. As NPs do not induce RNase cleavage they can be used for modulation of splicing.

Figure : Third generation ASOs (Kurreck, 2003)


A nuclease resistant ribozyme was designed such that it contained 5 unaltered ribonucleotides. At position 4 was a 2’ C-allyl uridine. Rest of the positions were having 2’ O-methyl RNA. An inverted thymidine group protected the 3’ end. The serum half-life of this modified ribozyme was <10 days as against >1 minute of unmodified ribozyme. An improved version has 4 phosphorothioate bonds in one of the substrate recognition arm and an inverted 3’ – 3’ deoxybasic sugar (Beigelman et al., 1995).

Figure : Modified nucleotides used in stabilizing ribozymes (Kurreck, 2003).


Introducing modified nucleotide at the ends of both the strands helped in enhancing the stability of siRNA. A siRNA with 4 methylated monomers at the 3‘ end and two 2’ O-methyl RNA nucleotides at the 5’ end produced prolonged gene silencing effect in cell culture. However, there was decreased siRNA activity when the methylated stretch of nucleotides was extended an bullku 2’ allyl substituent containing nucleotides were introduced.