combatting HIV with infected res/mod system

[Dak]

I did my dissertation on this, and i think it's a relitively interesting idea.

 

the basic jist is as follows:

 

find a restrictase and methylase that are both small enough to fit through the nuclear membrane (<50kDa) so that they can get from the cytoplasm to the genome.

 

find a methylase that methylates the sequence HCH (H = notG) at the 4 position (eg, H 4-methyl-C H). this is so that the methylation will not interfere with the eukaryotic gene silensing method (of methylating CG pairs), and because, afaict, 4-methyl-cytosine glycolase does not exist (so the BER mechanism will not reverse the methylations).

 

Find a restrictase that recognises a sequence containing HCH, that will not restrict the sequence if the C is 4-methylated (so that it will actually work properly with the methylase), and is a 4-cutter (for reasons that will be explained later).

 

both res and mod enzyme need to function at 36C, and be expressable by humans, and the genes need to be small enough to fit in a vector.

 

Citrobacter feundii has the closest res/mod system i could find:

 

 

             size          seq                gene length   temp
Cfr9I        36.8kDa   CCCGGG                 <1KBp           37C
M.Cfr9I     4.7kDa     C(4-methyl-C)CGGG     <1KBp           37C

 

dont know if they'll express in humans, and its a 6-cutter cos i couldn't find a 4-cutter with the correct qualities (tbh, i'm surprised i found this one).

 

anyhoo, the plan is:

 

whack the methylase gene into CD4+ cells' (or hematopoietic stem cells') genomes, and allow the cells to make methylase and the genome to become methylated

 

either whack the res gene in later, or at the same time but make sure it'll become active after the methylase has had a chance to protect the host's genome.

 

now, the CD4+ cells have a res/mod system (ta-da). cell division will result in hemi-methylated DNA, which will be protected from restriction and will become fully methylated as is normal for res/mod systems.

 

any invading HIV will reverse-transcribe in the cytoplasm, and will be restricted.

 

if any HIV inserts into the genome it'll create an unmethylated patch and the CD4+ cells genome will be restricted, essentially creating a suicide system so that the infected CD4+ cells kill themselves rather than pass the disease on.

 

theres a slight chance (as with any res/mod system) that the invading DNA will pick up the modification before being restricted, in which case the fact that HIV is spread in RNA form will prevent it from carrying the modification about with it.

 

so, yeah, it should be increadably hard for HIV to spread in the individual if their CD4+ cells have a res/mod system.

 

an obvious problem is that any HIV already present will be methylated along with the host genome, so will be completely immune to the res/mod system. but it might be useful as a vaccine for those at high risk, or as a treatment to keep CD4+ count up (or, i suppose, as a way of making a population of clean CD4+ cells in advance of a 'cull' of HIV+CD4+ cells, using a suicide gene or something).

 

that is, of course, assuming that it'd work, and whatnot. there were loads of potential pitfalls that i couldn't figure out, like wether 4-methyl-C is actually invisable to the BER system, and what exactly would happen with an invading HIV that was restricted in the genome (the pre-integration-complex would have it's DNA cut up, but the end bits of DNA, wrapped in integrase, would be intact; would this insert dijested DNA into the genome?) and so on and so forth.

 

but, one potential huge benifit (and the reason it needs to be a 4-cutter) is that it's incredibly unlikely that HIV could evolve out all instanses of a 4-bp sequence from it's genome, so it could be the one treatment that HIV can't evolve around.

 

So, yeah, whatcher recon? feasable?