Authenticity of XMRV integration sites

retroviral integrationIntegration of retroviral DNA into the cellular genome is essential for the production of new infectious particles. A strong argument that the novel human retrovirus XMRV is not a laboratory contaminant is the finding that viral DNA is integrated in chromosomal DNA of prostate tumors. Nucleotide sequence analyses of 14 integration sites in prostate tumor DNAs from 9 different patients previously revealed the expected viral sequences linked to human DNA. But two of these integration sites are identical to those found in a prostate tumor cell line infected with XMRV.

A search of the nucleotide sequence database with the previously identified XMRV integration site sequences revealed that 2 of the 14 sequences (from 2 patients) were identical to two XMRV integration sites in DU145 cells. This cell line was established in 1978 from the brain metastasis of a human prostate tumor. In early 2010 2007 DU145 cells were infected with XMRV, and sequences of two integration sites were determined (the database entries can be found here and here).

Identical retroviral integration sites have never been reported in independently infected cells. Furthermore, XMRV infection of DU145 cells was done in the same laboratory in which the XMRV integration sites were identified in prostate tumor DNA. The conclusion is that two of the 14 XMRV integration sites in prostate tumor DNA are likely to be the result of contamination. These prostate tumor DNA samples were probably contaminated with DNA from XMRV-infected DU145 cells.

These observations do not directly impugn the veracity of the other 12 XMRV integration sites identified in prostate tumor DNA. However, when DNA contamination occurs it is often ubiquitous. Hence the authors write:

Whilst it is conceivable that the other 12 integration sites apparently derived from prostatic tumor tissues are genuine patient-derived sequences, we suspect that some or all of them may also be the result of contamination with DNA from experimentally infected DU145 cells.

This possibility can and must be addressed experimentally.

Update: While writing this post I received an abstract from the 2011 Conference on Retroviruses and Other Opportunistic Infections (CROI) entitled “XMRV probably originated through recombination between two endogenous murine retroviruses during passage of a human prostate tumor in nude mice”. As usual I will await publication of this story in a peer-reviewed journal before discussing it further.

Garson JA, Kellam P, & Towers GJ (2011). Analysis of XMRV integration sites from human prostate cancer tissues suggests PCR contamination rather than genuine human infection. Retrovirology, 8 (1) PMID: 21352548

Stone, K., Mickey, D., Wunderli, H., Mickey, G., & Paulson, D. (1978). Isolation of a human prostate carcinoma cell line (DU 145) International Journal of Cancer, 21 (3), 274-281 DOI: 10.1002/ijc.2910210305

Dong B, Kim S, Hong S, Das Gupta J, Malathi K, Klein EA, Ganem D, Derisi JL, Chow SA, & Silverman RH (2007). An infectious retrovirus susceptible to an IFN antiviral pathway from human prostate tumors. Proceedings of the National Academy of Sciences of the United States of America, 104 (5), 1655-60 PMID: 17234809

158 thoughts on “Authenticity of XMRV integration sites”

  1. The statement “Is it contamination, or is it infectious?” really is a logical fallacy (false dilemma) the way you phrased it, as both possibilities are not mutually exclusive.

    In fact, I think this has confused patients the most: the fact that some virus is a contaminant does not rule out the possibility that this virus is infectious or is harmful for a human. After all, if this is a contaminant, this contaminant has still been “created” in a human cell line.

  2. Of course there are other variables that could explain all this. By far the most likely explanation at this point is the c-word however.

    BTW, Danielson et al. have not “shown” that the CDC method is incapable of detecting pol or gag. It seems that you are implying that. Moreover, there are currently no people on this planet that are conclusively “known to be infected” with XMRV.

  3. I said: “Just like patients tend to assume that the original authors cannot get any new findings published due to a hostile scientific environment”. How on earth should I logically have to assume that your reply to that was only aiming at the “hostile environment” bit (which is also unsubstantiated by any evidence, by the way)?

    For your logical understanding: the statement “if a paper was rejected, then it was probably not good” does not exclude the possibility that no paper was submitted. But yeah, propositional logic is a bitch….

  4. You can say this ad nauseum, but there simply are no positive clinical samples at this moment.

    Also, does any attempt of replication of the original study not rule out the validation of an assay through a test of a positive clinical sample? After all, the original study didn’t calibrate to a known positive sample. Thus, when any independent confirmation attempt does calibrate its assay to a known positive sample, it essentially stops being a replication of the original study.

  5. You can say that ad nauseum, but there are. Replication of a methodology thorough use of known clinical positive is scientific. Validation is separate again. Which original study are you now referring to, it is not clear? I will guess that you mean Lombardi et al. The authors of that paper applied scientific knowledge of how retroviruses work to find the virus. They also had some luck, as their assay was able to identify the diversity of the virus in their cohort. Others are not calibrating their assay to identify that range, nor do they know what range is in their cohorts, if indeed it is there. No other study has used Fukuda and Canadian criteria.

  6. That MRV’s are easy to find – did you not re-read your post?

    Please name one piece of evidence that supports the idea that MRV’s are a contaminant? I do mean evidence though, and not hypothesis which fail to explain all the evidence that shows this to be a human retrovirus.

  7. Please name one piece of evidence that supports the idea that MRV’s are a contaminant? I do mean evidence though, and not hypothesis which fail to explain all the evidence that shows this to be a human retrovirus.

    The CDC in Satterfiled et al. used the same assay that Danielson et al. showed to be incapable of detecting XMRV.

    In the BWG in phase IIa the CDC found MRV’s in plasma using Nested PCR for XMRV gag, qRT-PCR (pro), and qRT-PCR (int). In phase IIb the CDC found no MRV’s, but this time used Nested RT-PCR for XMRV gag and envelope, and quantitative RT-PCR for MLV gag and integrase.

    There is no conclusive by the way, only hypothesis which currently have not been challenge by any other hypothesis that accounts for all the published evidence.

  8. Then why assume, you can ask. So are you saying that there was no HIV hostile environment or MRV one now? If the latter, what has changed in the world to make you believe that is the case?

    I of course have not claimed that any paper has been submitted, that is your addition.

  9. I introduced Ebola, because you don’t seem to be convinced with the current biosafety level for handling XMRV (which is BSL-2 and for some people BSL-3). The only level higher than that is BSL-4 which is for Ebola and similarly deadly viruses with no treatment.

    Medics are at risk of infecting themselves with a whole lot of other more deadly viruses too. Do you expect people to carry “XMRV positive” badges on themselves? Patient samples are always handled with caution in hospitals or in labs. I’m pretty sure you’re already aware of this and I don’t understand why you’re still trying to push the panic button.

  10. I am beginning to think that you don’t really understand science. How can you claim that an animal model study is not scientific? Scientific method doesn’t have to be repetition of other studies to confirm or reject the observations.

  11. For all the screams of a “proper replication study”, instead of arguing, I assume it would be very easy to point me to a few of these “precise replication studies” in the field of retrovirology? Can you please direct me to the replication attempts that any retrovirologist (preferably Mikovits or Ruscetti) has made, using the exact methodology that the original study used. I guess that this would be very easy, not?

    And you would be correct in saying that “replication of a methodology thorough use of known clinical positive is scientific”. However, that is not what I argued. Can you thus point me to a valid replication of a “breakthrough” study that used the reported positives of said study as a way to calibrate its assays?

  12. Only you have suggested anything about biosafety. If it were perfectly safe to handle MRV’s, then there would be no need to test lab tech’s for the virus, but that is happening.

    I also wouldn’t suggest a virus associated with prostate cancer and ME is not deadly, when investigation into that is only in it’s infancy. Obviously there are always pathogens which take hold faster and have high death rates, but its still a serious concern for any tech or medic.

    Really the issue is that certain groups are using assay which have no demonstrated diagnostic sensitivity or specificity, and there is a risk. That won’t go away.

  13. I did re-read my post. I would suggest that you should try to properly address which part of a post you reply to with a single word.

    Furthermore, I asked you which part of my post was incorrect. There is no part of my post that states that MRV’s are easy to find. I suggest you re-read my post and re-consider your reply.

    I will reiterate: in science, evidence is data that supports (or rejects) a given hypothesis. For instance, consider the two (mutually exclusive) possibilities: that XMRV is a contaminant and that XMRV is not a contaminant. Suppose you have four commercially available PCR kits and decide to test them for mouse contamination. A prioro, you would expect (some of) the kits to be contaminated if XMRV is a contaminant and none of the kits to be contaminated if XMRV was not a contaminant. (Of course, neither data could conclusively proof either hypothesis). Therefore, in a scientific sense, finding one of the commercially available kits to be contaminated with XMRV is evidence that XMRV is a contaminant. Please note again that evidence does not mean that somehing is conclusively proven.

    Also, please note that I have used the very weakest of all “contamination studies” to explain how the process of gathering scientific evidence works. The other findings present of course much stronger lines of evidence. Together they provide a convincing and compelling argument.

  14. Oops. Incorrect that I said they were. Currently the evidence does not support the contamination belief. Why would you expects kits to be contaminated? Oakes and Robinson et al. showed there are different ways to contaminate samples. That in itself is not evidence. No study has shown kits to be contaminated with XMRV though.

  15. I take this as a “no, I haven’t got any examples of any scientists performing a true replication study in the field of retrovirology”. : )

  16. Could you propose one single experiment that an independent lab could perform that could provide us with “evidence” (as you define it) that XMRV is a contaminant?

  17. Now you are getting it. The evidence that this is a human retrovirus is too strong. If they cannot account for all the results of all the positive studies and more, then it cannot be a contaminant.

  18. This is the reference showing how an increase in the level of NF-kappa B increases the replication rate of XMRV, published in the peer reviewed journal of virology.

    How does an increase in the level of NF-kappa B increase the replication rate of a contaminant?

    This is the reference showing the elevated rate of NF-kappa B in DU145 cells.

    If DU145 cells have XMRV integrated into their DNA, can anyone explain why the XMRV does is not transcribed and the fact that DU145 cells do not express XMRV, given the high levels of nf-kappa b and oxidative stress known to exist within these cells?

    Mr Racaniello I would be interested to hear your comments and observations on this matter?

  19. This is the reference showing how an increase in the level of NF-kappa B increases the replication rate of XMRV, published in the peer reviewed journal of virology.

    How does an increase in the level of NF-kappa B increase the replication rate of a contaminant?

    This is the reference showing the elevated rate of NF-kappa B in DU145 cells.

    If DU145 cells have XMRV integrated into their DNA, can anyone explain why the XMRV does not transcribe and the fact that DU145 cells do not express XMRV, given the high levels of nf-kappa b and oxidative stress known to exist within these cells?

    Mr Racaniello I would be interested to hear your comments and observations on this matter?

  20. I merely asked for a link to a (in your view) *true* replication study in the field of retrovirology. If this is the normal and even only way that science progressess, this shouldn’t be exactly hard, but apparently it is.

    Now, can you point me to this *true* replication study in the field of retrovirology, preferably by Judy Mikovits?

  21. It’s not hard, people only have to display a will to undertake a replication study. Why single out retrovirology and Mikovits? This is the scientific method.

  22. No, I am not “getiing it”. I asked you for a possible experiment that could provide us with (your definition of) evidence that XMRV is a contaminant.

    To illustrate, while be both know the law of gravity to be true because “the evidence is too strong”, I can still propose an experiment (or a milllion) that would provide us with evidence that the hypothesis is false, if it were false. This (falsifiability) lies at the root of the scientific method and seperates science from pseudoscience.

    When you are unable to think of an experiment (by independent scientists) that can provide evidence against your hypothesis, your hypothesis is unscientific. So please rationally reconsider your previous answer and provide us with a possible independent experiment or accept the fact that your hypothesis is pseudoscientific.

  23. You implicitly accuse scientists of being lazy for not doing *true* replication studies, but are unable to even find a single example within the field for us to behold?

  24. If it’s not a contaminant there won’t be an experiment that can show it to be so, so yes you are now getting it.

  25. XMRV was recently transmitted from mice to humans, either from a single source, or at least from a single (sub) species of mice, and that all XMRV-positive individuals known today were infected with this newly emerged virus only recently, as a very high sequence identity is normally only seen after a direct retrovirus transmission.

  26. Please read the following for your understanding. It’s short, easy to read:

    Please come back and reply when you understand this very basic scientific principle. If you cannot propose any such experiment, you are essentially agreeing that your views on the topic are unscientific (more specifically, pseudoscientific). You wouldn’t want that, would you?

  27. It is down to other researchers to prove whether the evidence presented for MRV’s is incorrect. Currently nothing has been presented that is capable of explaining all the observations, and in fact has been shown to be unsupportable in the face of that evidence. Therefore contamination is not a hypothesis.

  28. “Thus the observation is not scientific.”

    These are your words. I didn’t suggest it, you said it! For that observation to be not scientific, the authors would have had to claim that the animal model reflects exactly what happens in humans, but they never say that.

  29. The only thing you have showed, is that you haven’t read the link I provided. I could think of plenty experiments that would provide evidence against the HIV hypothesis if it were actually incorrect.

    If “contamination is not a hypothesis”, you would agree that Silverman and F. Ruscetti are lousy scientists, as they have both publicly acknowledged that contamination of samples could be an explanation for different findings?

    Our argument is as circular as your logic and my time is limited, so I think I will rest my case. We may meet again if/when another sound negative study is published and you feel the need to “debunk” and “refute” it, so perhaps till next time…

  30. Your link would apply to HIV as equally as MRV’s, not exactly sticking to the issues at hand. Incorrect about those who entertain the idea of contamination, but know it to not be a contaminant, as no hypothesis has as yet explained all the evidence that shows it to be a human retrovirus. I notice that you are unable to challenge the information I have provided on multiple posts, and are now bowing out. Hope you are able to deal with this once further positive studies are published. I am going nowhere.

  31. No, you said I said it was not scientific, I said it didn’t adhere to the scientific method. It is Dove who has stated that they have used the same assay and been unable to find it in clinical positives. If it is not an assay that has been shown capable of detecting a known positive, then they have no adhered to the scientific method.

  32. are you blind or something!?

    “The titres in the infected monkeys would be massively higher than in human PMBC’s, with a natural infection pattern. That is artificial inoculation versus natural infection, and can be used to determine tissue tropism. Thus the observation is not scientific.”

    You wrote this paragraph! period. Look up on this thread.

  33. I don’t think there was anything about the comments from Gob that are reflexive. In my humble opinion they make more sense, and I have to agree they do not say that the reserach should be ignored.

  34. If you were at CROI last week, then you would have seen how worried many of the scientists were about this study.

    If XMRV is a contaminent, why is the CDC testing samples. Why did the questions deal with the safety of lab workers? Why did one questioner say he was worried now because they have been finding intermittent positive findings in lab workers. Since you were encouraged by “how quickly, efficiently and expertly different scientific teams have addressed the XMRV issue” why would anyone worry at all? If I was a lab worker at the CDC, who worked with these items over the years, youwouldn’t get me believing CDC lab test results. I’d want the WPI to check me, but that’s just me.

  35. Based on the above, it would seem that ANY lab that finds positive XMRV must have contamination. No? After all it’s a contaminent, so anyone who finds it must be contaminated, so it would seem.

    At CROI they said XMRV acts differently than other retroviruses. If that’s the case, looking for it in the same old way might not find it –doing the same thing over and over and expecting a different result is, well is you get the idea.

  36. I find Gob’s comments quite bizarre. There are clearly people on here who have an agenda which is clouding their judgement. And it’s not the virologists. By the way, I’ve had CFS for many years and want an answer. That answer must come from hardcore science.
    Gob : try actually listening to what people are saying to you. It would save everyone – you included – a lot of time.

  37. I agree with your comment, but I would replace “bizarre” with “misguided” (I’m also a long-term ME/CFS sufferer, pre 1978).

    I appreciate comments expressed here by some who strongly suspect XMRV is a lab contaminant but who also acknowledge that ME/CFS needs far more attention from scientists who are not connected with the psychobabble brigade.

    The Clinical Director of WPI (Dr Deckoff-Jones) addresses the CROI abstract in her post “Cover-up and contamination theories” on her blog. A lot of her information and that presented here is way over my head; I think I follow the basic concepts correctly.

    Reading through the commentary here, I come down on the side of those who think it looks very likely that XMRV is a lab contaminant.

    There is comment here about how antibodies with cross-reactivity could explain why antibodies to XMRV were found in some studies (ancillary question for Snail: Why don’t men with RA develop cross-reactive antibodies to HIV-1?). However, Dr Deckoff-Jones doesn’t allude to the possibility of cross-reactivity but seems to consider the finding of an immune response evidence of infection (do I confuse two separate issues?).*

    She also associates XMRV with autism (a point raised here by Snail). Are there any studies that have found XMRV in people with autism?

    I’m unsure what to think of WPI (friend or foe?). By having their study published in Science WPI moved ME/CFS to the centre of scientific attention. If a few good scientists became aware of the plight of ME/CFS sufferers as a result of the subsequent brouhaha then WPI achieved a magnificent feat. However, if WPI dismantles that achievement by ostracising good scientists (via too much eccentricity) ME/CFS will sink into neglect yet again. WPI seems in danger of doing just that at present. Is WPI marketing an unproven test (for XMRV) to desperate people? So many questions, so few answers ….

    For years the psychobabble brigade has advised that we are exhausted because we have a sickness mindset. At present a subset of ME/CFSers seems unhelpfully fixated on an XMRV mindset whereby if they believe strongly enough (and acrimoniously enough) it will become reality.

    * Some readers might not have seen these comments by ERV in Derek Lowe’s post ‘XMRV: It’s Ugly, But That’s Science’

    Quoting ERV:
    “The deal with the ‘anti-XMRV antibodies’ is something we have been dealing with in HIV-testing for decades, and has been explored in previous papers.
    Everyone makes antibodies to everything. That is how your immune system works. If you never ‘see’ an invader, you can never make a *potent* immune response to said invader. Thus, even a virgin who has never been exposed to any needle, ever could walk into a clinic one day, take a fast-HIV test (detects anti-antibodies in saliva), and come up ‘HIV positive’.
    That is why antibody tests are not the be-all-end-all-final-diagnosis tool for HIV-1 infection.
    Nor is 1950-esque ‘viral culture’.
    Real-Time quantitative PCR is.
    Because this is 2011.
    Furthermore, that virgin who tests ‘positive’ one day might next negative with the very same test a week later. Maybe they were just getting over a cold. Maybe their allergies were acting up. An immune disregulation could knock things off balance. Or maybe it was just a transient occurrence.
    In any case, it seems clear that the patients chosen for the initial paper had immune system abnormalities, while other papers chose people with a CFS diagnosis. Immune disregulation could lead to ‘anti-XMRV-antibodies’, or ‘anti-HIV-antibodies’, or ‘anti-anything-we-are-looking-for-antibodies’ that are not real. They are not really in response to infection, but a sign of immune annoyances or simply chance.
    The anti-XMRV antibodies isolated *WERE NOT REAL*.
    We know this, because another lab *also* found ‘anti-XMRV-antibodies’ (more in their controls than their CFS patients). They then investigated whether those antibodies were able to neutralize XMRV.
    They were not.
    Which means they were antibody-noise, not a real antibody response in response to a real infection.
    Now, I do not expect any person outside HIV-world/immunology to know this. Certainly not CFS patients, certainly not Average Joes/Janes. Makes one wonder why Judy Mikovits, who is not an Average Jane, does not know this information. Or at least pretends not to know this information when addressing her employers and CFS patients.”

    ” … The antibody ‘test’ the WPI used is a Western Blot. You take viruses, blow them up, and you look for antibodies that ‘see’ the guts. If you look at figure 4 in the Science paper, there are lots and lots and lots of bands in their gels that were supposed to be for one thing– XMRV env or XMRV gag. But… there were lots and lots and lots of bands? Because of non-specific crap… so how do you know what is crap and what is real?
    When Abbot framed an XMRV Western test off the HIV-1 Western test, they did not count someone as ‘positive’ until their sera had antibodies to three different proteins (it ended up being 3 people out of ~3,000-ish?).
    What is better, if you have a gun to your head and you MUST look for antibodies and nothing else, is to see whether those antibodies actually recognize and neutralize real virus. Not viral guts. Real virus. Every HIV-1 patient on Planet Earth makes neutralizing HIV-1 antibodies… its just that its not enough, and they are too late to prevent disease. Its a standard test I do all the time.
    Anyway, when scientists mixed their ‘XMRV positive’ sera samples with real, live, XMRV (not guts), it had no effect on the virus. AAAAAAALL those ‘anti-XMRV-env’ antibodies, and not one virus was neutralized? Defies logic.
    Unless the antibodies are, in fact, non-specific.”

  38. Sorry, but I really do disagree. It is evident that the comments are born out of science and not an agenda. I think perhaps the danger is that some like yourself are stuggling with this.

  39. A retrovirus being a recombinate is not mutually exclusive of it being a contaminant as well. For example HIV is a pathogenic retrovirus, but if you had a big vat of HIV and spilled it all over your lab then the HIV would be labelled a contaminant for whatever experiments you were trying to do. Take that example except substitute less-than-microscopic particles of the retrovirus in question, mouse DNA, etc. in place of a vat and that is how a recombinate could also be a contaminant.

  40. The recombination paper presented at CROI is proposing that a lab procedure is likely to have produced an identical XMRV as that reported in the disease association studies sometime between 1993-1996 and thus the studies which have reported an association between XMRV and disease were most likely contaminated by the retrovirus in question and thus are not genuine findings.

  41. The quote from ERV below isn’t correct. She writes “> The anti-XMRV
    antibodies isolated *WERE NOT REAL*. No one isolated any antibodies in
    the original Science paper. What was done was to show that sera from 9
    of 18 CFS patients reacted, by flow cytometry, with cultured cells
    that were engineered to produce a surface glycoprotein (env) from
    spleen focus forming virus, which the authors say has close homology
    with the env protein of XMRV. The positive signals could have been
    from cross-reactive antibodies, directed against another unrelated
    protein; or they might have resulted from infection by a related
    murine retrovirus. The results do not show that the antibodies in the
    patients were directed against XMRV.

  42. Monoclonal antibody to SFFV env will only react with the SU protein of a MLV virus. There were no other MLV viruses in the experimental environment with the same genomic sequence as XMRV. Would you care to explain what a monoclonal antibody to SFFV env could cross react with and do you have a reference please?

  43. I will post this here too.

    Monoclonal antibody to SFFV env will only react with the SU protein of a MLV virus. There were no other MLV viruses in the experimental environment with the same genomic sequence as XMRV. Would you care to explain what a monoclonal antibody to SFFV env could cross react with and do you have a reference please?

  44. Patrick S. Moore and Masahiro Shuda voiced some criticism regarding the antibody part of the Lombardi paper. I have included the following part in case you (or anybody else) hasn’t read this yet and is interested:

    “Flow cytometry and immunostaining with murine leukemia virus (MLV) antibodies were used to directly detect viral proteins in patient cells (see Figure 2A of the paper). The CFS peripheral blood cells have robust monotonic staining rather than the bimodal peaks that are expected from a mixture of infected and uninfected populations of peripheral blood cells. It is not certain whether this level of viremia for an exogenous retrovirus is medically possible. It may, perhaps, be possible but it seems improbable and is a pattern more consistent with a cross-reactive endogenous retroviral antigen.

    To confirm this finding, CFS peripheral blood cells (without negative controls in Figure 2B) were immunoblotted using cross-reactive spleen focus-forming virus (SFFV) and MLV antibodies. XMRV gp70 and p30 proteins are found at higher levels in 2 out of 5 CFS peripheral blood samples (1150 and 1221) than in the positive control — HCD-57 cells directly infected with SFFV — a very remarkable result. Repetition with negative control samples (see Figure 2C of the paper) has the higher molecular weight bands cut from the photograph, thus we cannot interpret potential positivity for p30 gag precursor proteins among the control samples (see CFS samples 1199 and 1220 in Figure 2B).

    Finally, the positive control HCD-57 cell lane in Figure 2C lane 8 has a completely different banding pattern from the very same control in Figure 2B lane 7 for the p30 gag protein. The elementary issue of whether the authors are measuring XMRV has to be clarified.”

    Full review:

Comments are closed.

Scroll to Top