Re: HIV EM and purification 2 July 2004
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: HIV EM and purification

I thank Robert Da Prato for his cogent reply, and agree that with removal of cells and subsequent centrifugation/fractionation then I see no reason why viral particles shouldn't be prepared in some manner for EM. The yields would still be remarkably small however in the case of HIV, as I pointed out before (200 patients of blood required for one salt-grain of virus, although I imagine less might suffice for an EM section). I wouldn't like to handle it at any rate!

As a comparison we can compare levels of virus in culture from that in the periphery. It's well known that the vast majority of viruses in the host are defective in some way, and that while viral load measures genomes only a small proportion of those correspond to infectious virus. The estimated proportion of defective viruses for HIV is greater than that for, say Herpes viruses, at between 1 in 10,000 and 1 in 100,000 depending on which report you read (but bear in mind these are numbers from peripheral blood!). Herpes simplex runs at around 1 in 100 as I recall, probably because it's a more stable DNA virus. HIV in culture might, in good hands, produce a million infectious units per ml. Some more recent work raises this to 10 million infectous units per ml. So, a vector titre of 1,000,000 would in theory equate to a "viral load" of at least 10 billion per ml. And this still requires concentration...unless you look at cell surfaces for budding virus.

Other viruses do indeed grow to higher titres, in culture at least (adenovirus vectors routinely reaching 10^9 to 10^12 pfu/ml). I am unable to find much evidence for peripheral measures except that Hepatitis C seems to routinely exceed viral load titers seen in HIV (a low value being 200,000 with the scale rising to 25 million as high - treatment is recommended for HIV at a level of 50,000 per ml, roughly equating to a 50% risk of progression to AIDS in 2 years as I recall). It should be remembered that Hep C was the first virus to be entirely isolated using molecular methods, because no culture system existed at the time and it wasn't possible to harvest virions directly from the patient. This despite the high peripheral virus titers!

In contrast, EM pictures of HIV have been readily obtained from lymph node biopsies of infected patients: a 5-minute search on PubMed brings up this paper from 1986.

Le Tourneau et al. Hum Pathol. 1986 Oct;17(10):1047-53 "LAV-like viral particles in lymph node germinal centers in patients with the persistent lymphadenopathy syndrome and the acquired immunodeficiency syndrome-related complex: an ultrastructural study of 30 cases."

And there are many others from the "related articles" link. Interestingly they consider this process important in making the diagnosis of pre-AIDS (ie before significant immune deficiency) - a sentiment echoed by people such as the Perth Group perhaps! In 1986 of course one must remember that the diagnostic tests of serology and PCR were far less developed than they are now. Of course LN sections are hardly pure by any stretch of the imagination, but these findings neatly counter the criticisms of "viral load insufficient to cause disease", especially considering my earlier comments on peripheral virus being the tiny minority of the total.

Competing interests: None declared