A virus that originates in mice, which last year was hailed as a possible cause of chronic fatigue syndrome, or ME, is not the cause of the disease, say scientists.
Four papers published by the journal Retrovirology all come to the conclusion that the finding of the mouse virus XMRV in human cell samples was not the breakthrough that researchers and doctors had hoped for. Further research suggests that the samples were contaminated with mouse DNA.
"Our conclusion is quite simple: XMRV is not the cause of chronic fatigue syndrome," says Professor Greg Towers, a Wellcome Trust Senior Research Fellow at University College London (UCL) and an author of one of the papers.
"All our evidence shows that the sequences from the virus genome in cell culture have contaminated human chronic fatigue syndrome and prostate cancer samples.
"It is vital to understand that we are not saying chronic fatigue syndrome does not have a virus cause – we cannot answer that yet – but we know it is not this virus causing it."
The discovery of xenotropic murine leukemia-related virus (XMRV) in patients with CFS by the Whittemore Peterson Institute in Reno, Nevada, published in the highly reputable journal Science, set the CFS/ME community alight. Many people who had been desperately ill for years without any idea of the cause, believed that this could be the answer. With a viral cause, CFS should be treatable and preventable.
But attempts to replicate the institute's findings have largely failed and the four papers published today leave little scope for further work on XMRV.
"Studies conducted by four completely independent research groups working around the globe have all reached the same conclusion: it is likely that the evidence for mouse virus found in human samples was due to contamination by mouse DNA," says Mark Wainberg, editor of the open-access journal Retrovirology.
The research includes evidence from a team led by Professor Myra McClure from Imperial College, London who looked at samples from prostate cancer patients, where XMRV had also been identified. They noted that XMRV was routinely found in around 5% of US citizens with prostate cancer, but was only very rarely found in European patients. Analysing British, Korean and Thai samples with a highly sensitive assay technique, however, revealed markers for mouse DNA other than XMRV. Professor Brigitte Huber and her team from Tufts University, near Boston, found similar results in samples from CFS patients.
In another paper, Dr Takayuki Miyazawa from Kyoto University offered evidence that the mouse DNA contamination may come from a particular manufacturer of testing kits commonly used to identify XMRV in tissue samples. When analysing the reagents in the kits without any human tissue present, the team found markers for mouse DNA.
The fourth paper came from Greg Towers and colleagues at UCL. They carried out retrospective analysis of previous research that seemed to support the claim and found that mouse DNA contamination was very likely in most of these studies. They concluded that researchers have been trying to identify XMRV using a DNA marker that may not be unique to XMRV after all. "Collectively, these results cast serious doubts on the PCR evidence used to support claims of MLV- related viruses in prostate cancer and CFS patients," writes Prof Robert Smith, Department of Pathology, University of Washington, Seattle in the USA in an editorial. "Future assessments of the prevalence of XMRV should include more rigorous PCR and phylogenetic tests to exclude the possibility of contamination."
Comments in chronological order (Total 9 comments)
20 December 2010 7:09PM
Urgh, I know that mice can get anywhere, but one would have hoped that sophisticated medical testing equipment would be sterile??
Yuk.
20 December 2010 7:31PM
f this is contamination why aren’t controls testing positive at the same rate as patient samples?
And why are patients testing antibody positive using a test that has been available since August? Why wasn't the news of a serological test reported on? Mounting an antibody response to a lab contaminant is biologically impossible.
This is basic bio-science 101.
Why are you quoting as FACT papers that have yet to be properly peer reviewed , seeing as “Retrovirology” is an open access, fee for publication journal?
And last time I checked op-ed pieces like the on quoted from Robert Smith do NOT equal verified scientific fact.
Oh and by the way the X in XMRV stands for XENOTROPIC. This means that this is NOT a mouse virus any more than HIV is the exact same virus as SIV.
This is one of the worst cases of yellow journalism I’ve ever seen.
20 December 2010 7:57PM
I am extremely disappointed with the Guardian for reporting this story in this way. Were you paid to print this story without any independent investigation? There has been a serious failure of journalistic inquiry here.
The papers that have found Human Mulv-Related gamma retro-Viruses have disproven contamination by several means already. The papers published today have done nothing to refute that.
1. Whole virus was isolated in the original Lombardi paper. This would not be possible with a contaminant.
2. The labs that can find the virus are CONSISTENTLY finding it, test after test, in the same patients, and in FAR greater numbers in patients than in controls. If this was a contaminant, the numbers of positives in patients and controls would be more similar.
3. It is impossible for the body to generate an immune response to a contaminant. Several labs have proven that there is an immune response.
4. A contaminant does not have sequence variation in the same lab in different samples, and does not mutate over time. Alter-Lo found the expected level of viral mutation in fresh samples taken 15 years after the stored samples they had tested positive.
5. 4 labs have independently found HMRVs and have tested rigorously for contamination, following all recommended procedures.
6. NONE of the labs that have found HMRV in patient samples have ever handled mice.
20 December 2010 8:22PM
A quote from Dr Harvey Alter, (distinguished NIH investigator and discoverer of the Hep C virus), re: the issue of contamination. It was made last week, 14/12, at the latest meeting of the FDA (US) Blood Study Advisory meeting.
"But I still want to counter by saying that the current evidence for disease association is very strong, that XMRV or MLV is strongly associated with CFS.
In those labs who do find the agent, it’s very reproducible. Year after year, same patients test positive.
Confirmed by sequencing, reproducible over time.
Dr. Hanson has demonstrated how critical the assays are. When they tweaked the assays, they are getting findings identical to the Lo lab.
Diversity of XMRV/MLV being confirmed in WPI lab, so XMRV is not the only agent being confirmed there now.
Using 100s of neg controls in same lab (Lo’s), all results for contamination were negative. Lo has done what Coffin recommended to test for contamination, and results were also negative. Always negative for contam. It isn’t logical to suggest otherwise.
Stoye used single-case, anecdotal information to try to make a case. Simply because it has happened in the past isn’t a valid basis to negate reproducible data from 4 diff laboratories.”
20 December 2010 8:38PM
WOW. I thought my UK friends might be overexagerating the biased reporting on ME across the pond. I guess not.
We in America get a more balanced news coverage of XMRV in ME/CFS and Prostate cancer.
Did you realize the Whittemore Peterson Institute has an electron microscope picture of this retrovirus and they have pictures of it infecting human cells. That is not contamination.
Please do your country justice and give media attention to both sides of the argumant and report the facts that show this is probably not contamination from Doctors Alter and Lo from the NIH and FDA who clearly feel what they have found is not contamination from there comments at the Blood Working Group Meeting in the US last week.
20 December 2010 8:57PM
"Scientists conclude mouse virus does not cause ME." Oh, great! Now I can use my mouse with peace of mind - without suffering chronic fatigue syndrome when surfing the net. Phew!
20 December 2010 9:57PM
Ok, I've read the XMRV blogs with continuing exasperation over the last few months, and here are a few points i'd like to get straight.
1: No-one (and I mean NO-ONE!) is disputing that the XMRV virus exists. It is a mouse virus that has been transferred to at least one human cell line that was isolated by growing them in lab-mice many years ago. The issue is: ARE PEOPLE WITH CFS INFECTED WITH IT? The problem is that retroviruses similar to XMRV have been infecting mice for millions of years, to the extent that the mouse genome itself is littered with their fossils. What has been shown today is that many samples that "test positive for XMRV or X-MLV" also test positive for mouse DNA (by a far more sensitive test than Lo and Alter used to discount this possibility). One might therefore legitimately ask why is there less detection in the controls? First answer to that is there isn't in the majority of studies because the level of detection in those studies was low - most people can't detect it in clinical samples very often. This has nothing to do with those studies being done by poor and slap-dash scientists - as anyone who does these procedures knows it is very easy to get false positives. It also has nothing to do with european molecular scientists being inferior to their US peers - need i remind you where HIV was first isolated (ok then, France!), only to be found a year later in a US lab which claimed credit for its discovery, but was subsequently shown to have contaminated their cell cultures with the French virus! The second point on this is that one must remember that many reagents used in labs are of mouse origin and that the more a sample is handled on the bench (as the CFS samples naturally naturally have been compared to the controls) the likelihood of cross-contamination increases exponentially - you don't need much, the techniques we are talking about amplify signal trillions upon trillions of times to detect it, one extraneous molecule will do it. Therefore you need exquisitely clean isolation conditions under which to process the samples - its intriguing that according to the Wall Street J blog yesterday, when samples form previously "XMRV+" patients were processed by the blood transfusion service before being sent out, no lab could detect XMRV, even the WPI!
2: No virologist involved in the XMRV story over the last year is disputing that CFS/ME is a debilitating disease whose cause needs to be understood, or that an infectious agent might be the cause. The reason so many retrovirologists (people like me who work on viruses like HIV and XMRV) have been trying to figure this out is not because any of them are involved in X, Y or Z conspiracy to suppress the truth (to what end one wonders?), but simply because this is their job. Put bluntly, in these times of strapped scientific funding, none of us would want anything more than a mysterious human disease to be associated with a Retrovirus so that we could get grants to work on it, understand its biology, and try to cure it. Unfortunately doing science is about finding the right answer, not the most popular one. I'm afraid XMRV's association with CFS will be another footnote in the long history of trying to associate human immune disorders with a direct viral cause - it's extremely difficult to prove causation, and one can only do it if the detection methods are reliable. The studies published today show that the tests used at the moment cannot discriminate between XMRV virus infection and contamination of samples (at an extremely low-level sometimes) with DNA of mouse origin.
3: To those above who say that there are EM images of XMRV. Yes, D'uh. They're of the bloody cell line that we know is infected with the damn virus! Its not difficult. I could train a teenager to take that picture in a day or two.
4: For the ill-informed "talkingFox", as someone who does this on a regular basis, the journal Retrovirology is a rigorously anonymously peer-reviewed scientific journal. Public-Access journals ask the authors to pay a contribution to the publication costs ONLY after this process is complete, and they do it so that scientists in poorer countries can access the information without paying exorbitant subscription charges. If one wants to point the finger at vested interests, perhaps one should consider that people associated with the original paper in Science are marketing an "XMRV-test" (that the current papers convincingly demonstrate cannot be and is not XMRV-specific) at several hundred dollars a pop. Sick and desperate people are being told that this provides hope for their cure and recovery, despite the lack of convincing follow-up studies that concur with the XMRV/CFS hypothesis. Some of these people are trying HIV drugs off-label with potentially harmful side effects while this argument rattles on. This to my mind is not only unethical, it is immoral.
20 December 2010 10:03PM
This is absurd. If ME/CFS is not caused by a retrovirus, then why are people getting better on antiretrovirals?
Please google: Dr. Jamie Deckoff Jones to find her blog, X Rx, which describes the effect antiretrovirals are having on her health. Click on the link HAART x 5 1/2 months.
Perhaps some large scale studies in which people are given antiretroviral medication would more clearly determine if a retrovirus is or is not causing ME/CFS.
Antiretroviral treatment is costly. Are these conclusions being reached to protect the insurance industry?
Shame on any scientist who spins science to support industry rather than the truth. Shame on any journalist and newspaper that does not dig deeper to find the whole story. Millions are seriously ill. You are enabling a silent holocaust.
20 December 2010 10:43PM
There is more to this than meets the eye.
Just because contamination has been the cause in other studies does not prove that it was in the studies done by Whittemore Peterson Institute or the Lo Alter NIH/FDA study.
It is significant that the UK media rush to publish this but haven't bothered to keep the public informed ...on the earlier significant research.
Before they continue publishing they should do their homework and I am sure the Whittemore Peterson Institute would be more than happy to share their findings especially on UK patients who have now tested positive to XMRV.
It's time for good investagative Science Journalism and not just parrotting such a weak study touting it as conclusive without looking at alternate science.
Shame on our UK media, they do not have a good track record with this illness ME/CFS if they did they would realise the huge controversy that has left patients struggling years without researching the cause.