Humor me while I brainstorm about my current project. I know that publishing this places my thoughts in the public domain, but since I haven’t a hope of getting something like this into a lab setting for years, and it’s just an untested intellectual exercise without an actual sequence attached, I don’t mind.

The aim is to design an antisense oligonucleotide (AS-ON) for HIV reverse transcriptase to be used as chemotherapy for one strain of the disease. Reverse transcriptase is the enzyme that transcribes the single-stranded RNA viral genome into double-stranded DNA, like the host’s. Blocking the production of reverse transcriptase, then, would block the reproduction of the virus. The AS-ON binds to the mRNA of reverse transcriptase to block translation in one of two ways: a. it hitches onto the 5′ end to block the attachment of ribozymes, or, b. it binds and activates RNAse H, which degrades the mRNA.

In designing an AS-ON treatment, there’s a lot to take into account. First of all, you need to get it into the cells for it to be effective, but straight up DNA or RNA will be degraded before it gets there. One gets around this by using either a lipid vector or using one of the various generations of modified DNA and RNA types that have been developed to cope with this. Then you have to worry about binding specificity, immune response, toxicity, efficiency, and target availability.

Immune response becomes troubling when the AS-ON contains a high percentage of guanine and cytosine residues. GC pairs are found most commonly in bacteria and viruses, and our immune system knows it. Four or more consecutive guanines can bind back on themselves, too, removing the AS-ON from circulation. 

Targeting is complicated by the sometimes complex secondary structures of mRNA molecules, which blocks much of the molecule’s sequence. One either has to run a series of tests to determine available binding sites or just target the 5′ end and hope for the best. Unfortunately, I have to do the latter. Furthermore, to avoid side effects from misbinding, you have to blast your designed sequence against the patient species’ mRNA to make sure there are no overlaps with native proteins.

What I’ve been looking into involves Locked DNA (LNA), which is basically DNA with a methyl bridge. This confers a number of great properties: increased affinity or the mRNA substrate, high potency, resistance to degradation, and apparent lack of toxicity and enhanced cellular substrate. However, it can not activate RNAse H. We also don’t really know how much increased affinity detracts from target specificity.

To remedy that problem, I’ve decided to use a chimeric strand of LNA-DNA. This will activate RNAse H, and the LNA confers all of its good qualities while the DNA center may increase specificity (again, I can’t run in vivo trials to find out). Now, I could have used phosphorothioate (PS) DNA or 2′-O-methoxy-ethyl (MOE) RNA for the internal stretch. More trials have been run on those sorts of hybrids. PS DNA would increase the rate of cellular uptake to fantastic levels; however, it also gives increased toxicity from (it is thought) binding nuclear proteins. LNA-MOE RNA chimeric AS-ONs have been tested, and also have increased cellular uptake and effectiveness, but it won’t activate RNAse H, either, and that’s a tool I really need to exploit. There are a bunch of other 3rd generation monomers that have been developed for AS-ONs, but none of them quite offers the range of benefits that follows an LNA-DNA chimera strand.

So, LNA-DNA it is.

In other news, Fulham handily won today’s match with Portsmouth. Yeah!