Re: Re: HIV-1 does not encode glutathione peroxidase. 14 January 2005
Previous Rapid Response Next Rapid Response Top
Nicholas Bennett,
Infectious Disease Postdoc/Clinician
Department of Pediatrics, University Hospital, Syracuse NY

Send response to journal:
Re: Re: Re: HIV-1 does not encode glutathione peroxidase.

I think Mr Whitehead has rather proved Foster's theory wrong himself! We all seem to agree that there is evidence for a general lowering of GP (glutathione peroxidase) activity in AIDS patients and in those with ARC. But the investigations into the theoretical HIV-1 encoded gene show that:

"To validate the theoretical predictions, we cloned the hypothetical HIV-1 gene and found it to encode functional GPx activity when expressed as a selenoprotein in mammalian cells. In transfected canine kidney cells, the increase in GPx activity ranged from 21% to 43% relative to controls (average 30%, n = 9, P < 0.0001), whereas, in transfected MCF7 cells, which have low endogenous GPx activity, a near 100% increase was observed (average 99%, n = 3, P < 0.05). " [1]

Furthermore Brian Foley's observations on the limited probability of this putative sequence being an active gene argue strongly against it being expressed as a functional protein in vivo.

If, and it appears so, HIV does affect GP activity, it seems highly unlikely it can be through the expression of additional GP. If anything, this enzyme would be protective for oxidative stress by catalysing the reduction of oxidative intermediates. I suppose one caveat is that an overactive enzyme may act as a selenium sink, but we're back to the problem that there seems little evidence that this enzyme is even expressed in vivo, never mind actually active. I'm also surprised at the claims about Hep B virus, since it too has a highly restricted genome and appears unlikely in my mind to want to encode another protein. One wonders if this group are seeing what they want to see..?

In response to the specific questions asked though I would have to answer:

I disagree that oxidative stress in involved in the causation of AIDS, but I do think it may be implicated in the rate of progression. This is similar to other situations, such as the Hepatitis delta agent which is harmless alone but greatly augments the toxicity of the hepatitis B virus when present as a co-infection.

I have seen evidence that antioxidants provide short-term clinical benefits, but then so did monotherapy AZT! I don't know if there is good evidence on long-term survival benefits. I predict however they would be rather better than monotherapy AZT... One interesting idea is whether, in badly antioxidant deficient HIV+ people, supplementation is as good as, additive to, or synergistic with HAART. Similarly to HAART, are the benefits less in those who aren't actually badly affected? These are questions that need answering.

I'm unsure as to whether it can be concluded that Tat per se can be blamed solely for any reduction in antioxidant protective measures. Tat trans-activates the HIV genome so in vivo any of the other proteins may be to blame. There is some interesting work in transgenic mice that may help here [2]. My caution is perhaps because viral trans-activators have been investigated for host cell interactions in the past (e.g. HTLV, Hep B) with largely no results! If Tat does have a direct, clear effect on a host cell gene or genes it may be an interesting precedent.

I note that in the UV paper cited they find no decrease in available Selenium but a decrease in GP activity. This perhaps argues that selenium deficiencies are contributary but not causal to the GP activity decrease (and hence this also argues against the theory of HIV encoding an active enzyme!). It also seems counter-intuitive that a virus would shut down a host gene only to replace with the same function - an altered function would make sense (e.g. this selenium sink idea) but then why would it show activity as a cloned gene product in vitro?

It is interesting that abberation of one DNA-damaging pathway (oxidative radicals) results in augmentation of another (UV attack): but not too surprising perhaps if similar end-pathways of DNA repair are utilised. One wonders if this has implications for the various AIDS- related cancers.

Of interest also is another "throw away" line in the web link provided by Mr Whitehead:

"Studies in knockout mice indicate that mRNA expression levels respond to luminal microflora, suggesting a role of the ileal glutathione peroxidases in preventing inflammation in the GI tract."

In the light of the fact that HIV infection is associated with malabsorption, so much so that in some areas it is simply called "slim disease", perhaps this sheds light on a possible mechanism.

Nick Bennett njb35@cantab.net

1. Zhao et al: .PNAS June 6, 2000 vol. 97 no. 12 pp6356-6361 "Biochemistry Molecular modeling and in vitro activity of an HIV-1-encoded glutathione peroxidase."

2. Choi et al. J Biol Chem. 2000 Feb 4;275(5):3693-8. "Molecular mechanism of decreased glutathione content in human immunodeficiency virus type 1 Tat-transgenic mice."

Competing interests: None declared