Infectious Disease Postdoc/Clinician
Department of Pediatrics, University Hospital, Syracuse NY
Send response to journal:
I am extremely encouraged that the Perth Group are finally addressing the important point of Montagnier's original RT findings.
What the Perth Group say is the entire crux of their problem with science in general. I'm sure if they directed their energies towards any field they would find just as many "problems" as with HIV.
As Montagnier is reported to have said: "Taken in isolation, each of the properties [including RT activity] isn't truly specific".
But that is exactly what the Perth Group do. They take things is isolation: which is why my pithy response to their criticisms of the Richieri paper was to READ THE REST OF IT, which I repeat in the hope the message gets home. Facts must NEVER be taken in isolation, ultimately the entirety of the rest of the literature must be considered as well. What I see repeatedly is facts taken in isolation, out of context sometimes, and THEN put together into a haphazard argument.
For example, the Perth Group ask me how I would distinguish RT activity as coming from a retrovirus or something else. I would perform other experiments: firstly using uninfected cultures in parallel to compare endogenous levels. Secondly I could characterise the enzyme in terms of buffer specificity, so I wouldn't just be looking at pH 7.8 and Mg 5mM, but at a range of magnesium concentrations and acidities to distinguish it from other enzymes. I could also look for proteins (typical RV sizes) or RNA (40S, the antiquated method of showing RVs in Duesberg's day) in the preps, since I would expect these to be far more easily detected that viral particles under EM. I could also use Actinomycin D as described previously. With more modern reagents the possibilities are almost endless...culminating with sequencing of the RNA.
To be honest, one would generally cover most bases by simply using uninfected cultures and supernatants as a control. For a real-life example, in my experiments a truly negative sample (buffer only) would give approx 20 RT units (a measure of incomplete washing of radionucleotide), uninfected cultures have counts of around 30-50 (endogenous RT activity, cellular polymerases) and viral cultures and supernatants run between 200 and 2000 depending on the efficiency of the experiment. I typically didn't consider RT to be present until the sample was at least twice that of the uninfected culture (over 100 units), and that correlated with later timepoints rising to high values. By expressing RT values relative to the uninfected cultures you neatly sidestep all the DNA polymerase and endogenous RT!
Note that my RT was purely to monitor viral replication or production levels: later confirmatory work included western blots, RT-PCR and sequencing, the real experiments.
Montagnier also performed extra experiments, he did not simply look for "RT activity" in a single buffer without extra confirmation. Had he done so I would say the Perth Group had an excellent case! Sadly, ignoring the minor fact that he didn't somewhat detracts from the argument...i.e. renders it useless. He ensured the enzyme was not DNA polymerase by seeing what the optimal buffer was and the degree of inhibition by actinomycin D. That clinched it for me: to date I am unaware of an enzyme that would fit the findings other than RT.
I am sure that stimulation of primary cells is necessary, but I take issue with the Perth Group's persistence in saying "stimulation (oxidation)". The two terms are NOT synonymous, and linking them in this way is extremely misleading. Many mitogens act not through "oxidation" directly but through specific receptors and cellular signalling pathways. In fact, T cell responses and IL-2 production are reduced by oxidative reagents  despite oxidative agents increasing the effects of NF-kB. As well as showing how a sledgehammer approach can affect multiple pathways simultaneously, it suggests that an activating cytokine (one kind of mitogen) like IL-2 is far more subtle.
Primary T cells require IL-2 stimulation in order to survive ex vivo (they normally get this stimulation in the body). I would not call IL-2 an oxidising agent. HIV may infected non-dividing cells (a property quite distinct from simple retroviruses like RSV) but it requires cellular activation in order to replicate. This is because it has NF-kB binding sites in its leader region: in the same way as we wouldn't expect a T cell to produce cytokines or replicate without NF-kB we cannot expect HIV to replicate without NF-kB. It is no coincidence that a T-cell trophic virus utilises the major transcription factor of T cell activation.
As such, it makes perfect sense that all the subsequent events of viral activation (particles, antigenic expression, cytopathic effect - is this the Perth Group admitting that particles and cytopathic effects exist...?) depend on activation.
I also noted that the Jurkat T cell line produces HIV without exogenous stimulation (it makes its own IL-2). How does this square with the Perth Group's insistence that the activation of HIV depends on oxidizing agents?
It is well known that producing permanent cell lines from human cells requires considerable effort. Murine cells conversely almost always immortalise (this may be due to differences in telomere length or other mechanisms of cellular genetic stability). As such without reading the references provided I know from my education that it's obvious that a great many cell lines DO require stimulation in order to become established, and agree with the Perth Group on this. In this situation, some of the agents used are quite harsh, and I'm prepared to accept the use of the word "oxidising" in this context for certain cell lines, although I can't comment on the specifics of H9 and the parent HUT-78.
However, stimulation of cells to activate HIV is an entirely different situation, and shouldn't be included haphazardly. If immortalisation of cells is equivalent to conscription of an army, activation is equivalent to giving a General orders for his or her battalion (petri dish).
 Morel and Barouki. Biochem. J. (1999) 342, 481–496 Repression of gene expression by oxidative stress.
Competing interests: None declared