Re: Re: Clinical trials of antioxidants 15 February 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: Clinical trials of antioxidants

I would like to echo Mr Whitehead, as I have mentioned previously, in being in support of larger-scale investigations of the benefits of treating antioxidant deficiencies. At some point it would be nice to think I would be in a position to make a more practical contribution to the idea!

As regards HHV8 and carcinogenesis, Mr Whitehead may have missed the several posts I made previously highlighting the fact that HHV8 contains various genes that are related to or can function similarly to human cell- cycle regulators. HHV8 is not only epidemiologically associated with KS, but can be detected in all forms of KS, from almost all samples, but not from surrounding tissues. The fact that it contains genes that are demonstrably and theoretically capable of causing cancer is enough for most people.

The phrase "cutaneous lesions resembling Kaposi's Sarcoma" is about as proof-positive as you'll ever get in a journal. It's tantamount to saying "we created KS"! They can't of course say that because KS is a naturally occuring disease, whereas this was "only" created by adding the viral gene(s) artificially. As Yang et al [1] conclude for example:

"Here, we report that a single gene of HHV8 is sufficient to induce in mice an angioproliferative disease that bears striking similarity to human KS. Clinically, the lesions start as erythematous maculae or plaques and progress to violaceous, purplish nodules and tumors that develop predominantly in the skin and to a lesser extent in intestine, heart, and skeletal muscle. Histologically and ultrastructurally, these lesions show characteristics similar to those of human KS and Kaposiform hemangioendotheliomas. Lesions within the skin developed and spread within the dermis while sparing the overlying epidermis. Like KS, the nodular angioproliferative tumors formed slit-like vascular spaces containing erythrocytes and were surrounded by spindle cells and infiltrating inflammatory cells. Finally, vGPCR, VEGF, and CD34, markers that have been detected in KS, are also detected within these lesions."

One wonders what description of a lesion Mr Whitehead would consider better than this, as evidence supporting HHV8 in causing Kaposi's Sarcoma? He uses the phrase "at best" as if the findings somehow fall short of meeting his expectations, whereas for me as a virologist they far exceed my wildest expectations! For example, we cannot (yet?) induce an AIDS- like illness by adding a single gene from HIV into an animal model. To do this for HHV8 and KS is a holy grail of viral pathogenesis.

HHV8 alone is clearly not sufficient, due to the fact that most of those infected with the virus do not get KS, but immune deficiency increases the risk of KS. Also the model of Yang et al is obviously full of caveats because it is a transgenic animal rather than a viral infection. But still, it's striking enough.

The report of a possible mouse model of AIDS mentioned by Mr Whitehead is interesting because it highlights the problem I previously mentioned to Alex Russell some time ago, regarding the prolific number of mouse retroviruses: both exogenous and endogenous. The followup papers to that report clearly demonstrate the fact that this was the result of a previously unknown retrovirus infection, which is sexually and horizontally transmissible. Two endogenous retroviruses were activated and the co-infection results in the disease process mentioned (a lymphoproliferative condition and an autoimmune CD4 T cell depletion) [2].

This of course has several implications - the use of transgenic porcine organs in human transplantations has had the worry of porcine endogenous retrovirus activation for some time. An interaction between one such virus and a human endogenous RV cannot be easily ruled out, nor the outcome predicted. Additionally, while HIV sequences are not detected in the human genome, HIV-like sequences have been detected in certain cancers, which may implicate a new exogenous retrovirus. Recombination between an ancestral SIV and another human RV may have resulted in the formation of HIV. Such a theory is supported by the observation that in this mouse model of co-infections, possible co-packaging of both virus genomes within virions occured [2].

Mostly speculation, but this result highlights the fact that multiple viral co-infections may have unpredictable outcomes.

Nick Bennett njb35@cantab.net

1. Yang et al J Exp Med. 2000 Feb 7;191(3):445-54. "Transgenic expression of the chemokine receptor encoded by human herpesvirus 8 induces an angioproliferative disease resembling Kaposi's sarcoma."

2. Hook J Virol. 2002 Dec;76(23):12112-22. "Characterization of a novel murine retrovirus mixture that facilitates hematopoiesis."

Competing interests: None declared