errata: new anti-HIV strategy, no HIV resistance thinkable
marcos at uchicago.edu
Sat Dec 2 21:08:46 EST 2000
Dear bionet virology Members
I removes a few errors and unclarities from my previous posting. Please
disregard it and read only this one.
Dr, Marcos Antezana
////////////// anti-HIV strategy, HIV-resistance unlikely
HIV relies partially on nucleus-encoded proteins and even packages some
of them in its viroids. If one were to neutralize the latter proteins
possibly already at the viroid-stage, the virus would not be able to
evolve resistance because nucleus-encoded proteins do not mutate at high
rates and their evolution is tied to the fate of the host.
One of the latter proteins is the DNA mismatch-repair enzyme called
uracyl-DNA-glycosylase (UDN) that removes uracyl from UG mismatches in
DNA (I suspect it may remove even other mismatches from the DNA part of
DNA/RNA hybrids created by reverse transcription)
It is very unlikely that the virus packages the glycosylase in its
viroids for something other than mismatch repair. Also the virus must
be at a great disadvantage if it does not find the glycosylase in the
cytoplasm upon infection. The glycosylase should be found in
appreciable quantities only in the nucleus. The glycosylase packaged in
the viroid must therefore be used to create a glycosylase-rich
micro-environment in the cytoplasmic region where reverse transcription
will take place.
My idea is to increase the mutation rate and inhibit the glycosylase
activity both inside viroids and in the cytoplasm (but not in the
nucleus). This should mutate the virus to death over evolutionary time
(Eigens hypercycle theory) and/or in just one generation (useless
reverse-transcripts, non-viable viroids) so it is very likely that the
virus will fare too well if one mutagenizes it beyond a critical
More specifically, the treatment consists in using, separately if
necessary, but concomitantly for maximum effectivity:
1) A mutagen with low-nuclear-penetrance, hopefully RNA-specific and
inactive on DNA, to increase the mutation rate in the RNA based
HIV-genomes packaged in viroids,
2) A glycosylase inhibitor with low-nuclear-penetrance
1. will mutagenize the virus before infection
2. will make mismatch repair after reverse transcription very
error-prone (this has been proven already)
2 will also impede that repair take place between the two RNA genomes
both within a viroid and at infection time when the two HIV genomes are
injected in the host (since the two HIV genomes are not complementary,
such repair would require that HIV-genome regions with C have
complementary sequences elsewhere in the HIV genome to serve as
templates, and also that U-containing regions don't have such).
and 2 will also impede that repair excises anomalous U in the
reverse-transcript/HIV-genome hybrid (DNA/RNA hybrid).
The RNA-specific mutagenicity of 1 and the low nuclear-penetrances of
both compounds are only necessary to avoid increasing the mutation rate
of nuclear DNA. But RNA-specific mutagenicity is not necessary if low
nuclear-penetrance is guaranteed. Moreover attaining low
nuclear-penetrance would allow using a mutagen that also damages DNA
i.e. one that would also mutagenize the HIV reverse-transcript.
Non-RNA-specific mutagens that reach the nucleus and produce UG
mismatches are known and could be used, but should create problems if
they overwhelm the nuclear glycosylase by producing too many mismatches
to repair, which ultimately should cause cancer.
It should be easy to create low nuclear-penetrance constructs by binding
the mutagen or the inhibitor at the tip of elongated lipophilic
molecules that would penetrate the viroids but that would have a bulky
plug at the end so the construct cannot be fully absorbed through
A presence restricted to the proximity of the interior surface of cells
and viroids should be enough for the mutagens/inhibitors to enter into
contact and damage the viroids contents while guaranteeing that the
constructs don't reach the nucleus. It is important not to use peptides
or other biomolecules for the constructs so that the cell cannot cleave
them thereby allowing the mutagen and the inhibitor to reach the
nucleus, ultimately poisoning the cell.
Ideally the mutagen could have an antidote so that one can neutralize it
when it is not needed anymore, to give relief to cytoplasmic mRNA and
rRNA populations. One possibility is choosing one that becomes inactive
after it mutagenizes a base (nitrous acid?). Also the mutagen should be
covered by a layer of molecules that are shaved off spontaneously in the
cytoplasm but not in the blood stream/etc.
The mutagen would affect the normal RNA populations of cells but it is
unlikely that such effects will be dire: Some aging maybe but one that
would be transient (normal turn-over should take care ultimately of
anomalous mRNAs and rRNAs, and anomalous proteins from mutated mRNAs).
Similarly one could raise the mutation rate up to, but not beyond, the
saturation level of the glycosylase, but such dosages maybe be hard to
In any case the treatment with only the glycosylase-inhibitor construct
should not have any side effects since the enzyme main activity is
repairing double-stranded DNA in the nucleus. DNA, and native DNA
repair even less, should not be found in the cytoplasm. Mitochondrial
DNA and RNAs may be a problem, however.
best to you all
P.S. Has anybody tried to deliver a non-specific one-time
ribonuclease/protease that would cleave only once and then become
inactive. This would kill viroids but do hardly any lasting damage to
cells (especially the ribonuclease).
More information about the Virology