Re: Retrovirologists, retroviruses 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: Retrovirologists, retroviruses and purification

The statements on retroviral isolation and purification (for various personal values of "pure") repeatedly made by the Perth Group seem ironic considering they have no first-hand experience with such matters.

However, it seems "common sense" that such stringent standards as 100% pure virus preparations are hardly a prerequisite for discovering new viral proteins or RNA to a degree sufficient to satisfy most scientists of their existence. At the very least, I ask the Perth Group where exactly the sequences deposited in the Genbank database are supposed to come from? They have the genetic appearance of an exogenous complex retrovirus and produce virus when innoculated as plasmids into culture systems. The sequences do not appear in the human genome. The fact that this is a post -hoc discovery to the original isolation of the clones should be seen as confirmation, rather than dismissed because the original findings do not meet artificial and impractical criteria for isolation.

As an example: Montagnier's original discovery of HIV describes how proteins from a culture of the putative new virus reacted with sera from patients with AIDS. In a Western Blot, this not only gives a measure of identity (self versus non-self) but also a degree of analysis (size). Such methods are frequently used to show the prescence of new proteins in a mixture and do not detract from the question of whether the protein exists or not. The inclusion of appropriate controls (non-infected samples) neatly sidesteps the problem of isolation, since proteins and RNA found only in infected cultures can most likely only come from an exogenous source such as (for example) a virus in a virus culture. For the record I do not personally agree with the use of the word "isolation" in that context, but "detection" seems entirely appropriate and does nothing to detract from the conclusions of the paper.

One must however, grant that inducible expression of self proteins or RNA is not a more likely explanation. Previous arguments I have seen use the appalling explanation of "culture stimulation" to explain these findings, which I hope will not surface here. One only has to ask "why does the control culture not show the same 'stimulation'?", and assume that critics who use such an argument (such as the Perth Group in their comments on "confounding molecules" of culture supernatants) are entirely unfamiliar with the practise of using controls in their experiments. Subsequent work, as I have mentioned, has shown that the genetic sequences of HIV are not present in the human genome, so they must be derived from exogenous nucleic acid, which in the case of HIV is in fact integrated into the host cell chromosomes. The fact that individual cells within an infected host display variation in detection of HIV further highlights the fact that these are not "susceptibility" markers of some other future disease.

As regards the appearance of new proteins, the coincidence of seropositivity argues against these being inducible host proteins, and the size can be used as a crude by effective means of identification. Certainly it is possible to judge with the naked eye whether a particular Gag protein on an SDS-PAGE gel comes from HIV-1 or HIV-2, and a pattern of sizes showing approximately 120kD, 55kD, 41kD, 24kD is suggestive to any retrovirologist that it may show the envelope and capsid proteins. In addition, the slight cross-reactivity to HTLV sera described in the early papers increases the likelihood that these proteins are from a related retrovirus rather than host-derived or indeed from any other type of pathogen.

And in the end of the day, the presumed HIV RNA actually encodes the presumed HIV proteins, which in and of itself is a massive clue as to their mutual identity. One must agree that the situation is beyond reasonable doubt. Being able to extract a single species of RNA from the band at 1.16mg/L was always possible, since it was why retroviruses were originally presumed to have diploid RNA genomes and how the early work on mapping was performed. It is entirely possible to reverse-transcribe an entire HIV genome from such a prep, since I myself have done so albeit using a more crude sample (pellet rather than density band). The trouble of course is that I wasn't aiming to answer the Perth Group's questions on isolation, merely perform site-directed mutagenesis - the point being that the field has moved on beyond trying to address such questions.

The Perth Group appear fixated with wanting a quick-fix (i.e. one paper) comprehensive answer to their questions, which as far as I am aware is not possible using historically relevant data. Much of the data directly answering their questions was published after the acceptance of the existance of HIV and its causal link to AIDS - in particular I refer them to my earlier reference using anion-exchange chromatography, since as well as providing an excellent EM the protein content of the virions was also extensively analysed. Since they seem well practised in pulling together disparate facts from papers, I urge them to undertake a systematic review of the literature with a view to understanding why the conclusions about HIV's existence and pathogenesis have been made, rather than trying to support a tired and well-refuted argument. If they desire a comprehensive review then I suggest they indulge themselves with a little light reading of Field's Virology.

My point is that within the limitations of the science at the time, the approach used was entirely reasonable. One must bear in mind that many of the quotes used by the Perth Group refer not to complex cytotoxic lentiviruses but to comments and suggestions made with reference to simple non-cytotoxic oncoviruses such as RSV, ALV etc. These viruses are well known to produce far better quality preps than HIV due to the nature of HIV to kill cells in culture and produce a soup of cellular debris. This toxic effect has at least been acknowledged by the Perth Group, which makes their insistence of requiring such soup to have been removed from the very first analyses (of these unknown, at the time uncharacterised virus preps) all the more surprising. Anyone familiar with the practical problems of benchwork would acknowledge the difficulty of attaining perfection in the very early days of any line of investigation.

In addition, the later quotes will include assumptions of using refinements to the techniques of isolation and purification far and away superior to the mere density banding at 1.16g/ml in a sucrose gradient. It is entirely possible to directly address the problems of viral versus host protein and RNA in a prep using modern methods, and the fact that these findings are retrospective does nothing to detract from the original work. Science works by progression, adaptation, correction and augmentation of knowledge. Many of the original assumptions and conclusions about HIV (and many other pathogens) have been found to be wrong, but the actual existence of the virus and its ability to replicate has, if anything, been more firmly established than ever before.

Additionally, I should point out that having worked with HIV-1 and HIV-2, and with other researchers with the same and far greater experience, I believe that my opinions are in line with the vast majority of those working in the field, including those of Brian Foley and Christopher Noble. That's not to say that purification and isolation are not goals to be attained, but a huge amount of work can be done in their abscence including, but not limited to, detection, proof of exogenous existence, and causal links to disease processes.

Competing interests: None declared