XMRV and human PBMCs: DO NOT WANT!!!!!!!!

Though I am definitely (and I hope clearly) a supporter of animal research in science (plug for Speaking of Research), it is also definitely something I do not like to do. Contrary to how animal lib freaks frame us, while I fully and absolutely support and appreciate my fellow scientists who are animal researchers, I personally have psychological issues with performing animal research myself*.

I feel it is a big part of my responsibility as a lab-bench researcher to do the best damn in vitro work I can do, to make animal researchers jobs as painless as possible.

Everything I do in the lab is in vitro– whether its artificial quasispecies generated in the lab and ‘lab guided’ evolution, or playing with viruses evolved in and isolated from nature. I try to do my part for animal research by doing basic research the best I can. Thus, I can work with animal researchers to design experiments with the best data available, thus we can get the most out of our experiments by using as few animals as possible. Then I get great data/reagents back from them, design things better in the lab, and so on and so forth. The fewest animals are used when everyone is at their best and working together.

Great in vitro data from logically designed experiments save wasting the lives of research animals. So despite my XMRV burn-out, I do want to cover this extremely data-rich in vitro paper:

Severe Restriction of Xenotropic Murine Leukemia Virus-Related Virus Replication and Spread in Cultured Human Peripheral Blood Mononuclear Cells

Long story super-short: Human PBMCs to XMRV: DO NOT WANT!!!!!!

Long story short: Human PBMC (white blood cells) dont want or like XMRV. They are very inefficiently infected, and when they are infected, they produce very little virus. The little virus they do produce is hypermutated, and almost certainly abortive.

Long story not short at all:

The focus of this paper is a group of proteins every cell in your body has to defend themselves against retroviruses (want me to make an evolutionary guess, they protect us from wayward ERVs)– APOBEC (Wikipedia page for a specific one, A3G). APOBEC is packaged in babby viruses, and then when the babby goes on to infect a new cell, APOBEC screws up their reverse transcription process and makes a ton of mutations in the new proviral DNA. You might say “OY! ERV! Retroviruses mutate all the time anyway, rite? So why do they give a crap if ABOBEC makes them mutate more? Wouldnt that be better for a retrovirus?”

While Creationists (and other anti-/non-science groups) might present viruses as these mindless mutating machines, the number of ‘mistakes’ that happen during reverse transcription is a evolved number. It makes those mistakes because they provide the ability to explore as much sequence space as possible… without making so many mistakes that it just wont work as a virus anymore (error threshold). APOBEC pushes retroviruses over that error threshold (unless the virus has evolved a protein to counter the APOBEC).

So what happens is– A retrovirus productively infects a cell. The infected cell produces lots of babby viruses. APOBEC stows away in those babby viruses, which go on to infect new cells… BUT! During reverse transcription, APOBEC introduces lots of unwelcome mutations. Which means even if that new cell is ‘infected’, the virus that inserts into its genome is crap. It cant make any more babby viruses.

APOBEC is like the retroviral version of Cockroach Hotel: Retroviruses can check in, but they dont check out.

An obvious observation is that XMRV has no apparent counter to human APOBEC, but it might. But previous experiments didnt seem to indicate there was one. ?

So… if XMRV is infecting humans… how is it getting around our APOBECs? If XMRV is not getting around our APOBECs… is XMRV infecting humans?

These folks started in cell lines. As Fraulein Maria would say ‘A very good place, to start!’ They infected 10 billion CEM cells (a T-cell line that expresses a lot of APOBEC3G and 3F) and 10 billion CEM-SS cells (a related T-cell line that expresses little/no A3G or A3F) with either 10^4, 10^5, 10^6, 10^7, 10^8, or 10^10 XMRV viruses (based off RNA copies/volume).

At 10^4, not a damn cell was infected, either line. But when they used more virus, the CEM-SS cells were infected and made lots and lots of baby viruses, while the CEM cells exposed to the same quantity of virus, were like ‘Meh.’ They got infected, they made *some* babby viruses… but those babby viruses didnt really go on to infect more cells to produce MORE babby viruses.

This result is consistent with the expectation that most of the proviruses in the CEM cells will be hypermutated and will be unable to produce viral particles.

CEMs, CEMs, a human T-cell line, Cockroach XMRV hotel.

But cell lines are not ‘human’. So the next step was to isolate white blood cells from a few different people (one of the putative targets of XMRV, which express lots of A3G and A3F), made 10 billion cell aliquots of those cells, and did experiments with them.

The cells pretty much wanted nothing to do with XMRV. Theres not much else to say. They did the same thing as the CEMs– some of the cells were infected, but they didnt *produce* babby viruses:

Importantly, we did not observe any increase in viral RNA copies in the infected PBMCs during the next 15 days, indicating that there was little or no virus replication and spread.

If the cells are good targets, you expect to put on ‘some’ virus, and get out A TON OF VIRUS. Didnt happen here.

Furthermore, the few babby viruses that were made were mutated by APOBEC3G and 3F. They sequenced the babby viruses and saw A3G/A3Fs mutational signatures. Actually, they only sequenced a ~1200 nt region of the virus, and they saw more mutations in viruses we *know* came from a clonal stock than I see in the ‘XMRV sequences’ uploaded to Genbank, which theoretically come from ‘wild type totally in nature’ variants.


They also made a neat indicator cell-based ‘test’. Mix 100,000 PBMC from someone with their indicator cells. If at least 7 of those cells are infected with XMRV or any other MLV-like-creature, their test will see it. No PCR with magic primers. No flow that could be non-specific. No antibodies that could be non-specific. Just: Is there a gammavirus up in here or not? Are the indicator cells green or not?

While that is neat, there is one more thing I would have wanted in this paper– XMRV RNA is not increasing in infected PBMC cultures (not getting multiple rounds of infection). I would have liked to have seen PBMC supernatant filtered, and put on, say, the CEM-SS. I dont care if the CEM-SS are infected. They should be. I want to know if XMRV RNA increased in *THAT* culture. How much of the crap virus coming off the PBMC is still functional? I dont think they have the data to say this:

Although XMRV replication and spread in PHA-activated PBMCs was severely restricted, PBMCs could serve as a reservoir of replication-competent XMRV and facilitate infection of A3G/A3F-deficient cells.

Yes, they ‘could’, but it would have been nice to have some data to see how ‘couldliest’ that possibility is. But maybe they did and I just dont totally ‘get’ their indicator cells.

Thats the worst thing I can say about this paper. Its straight forward. The data moves the field forward. HUZZAH!

So why the hell was I ranting about animal research in the beginning?

Because the data in this paper could have helped this paper. It could have made the deaths of those eight rhesus macaques most likely to yield the largest quantity of data/life. As is, the take-away message from that paper is “If you pump 10^6 XMRV IV into macaques (a dose that would leave you with a pile of dead monkey with SHIVs), pretty much nothing happens. No clinical symptoms. The XMRV gets in, but there isnt much (any?) productive infection.” Where did the >10^6 number come from? What was the logic? Why not 10? Why not 10^10? 10^6 is a dead-monkey-dose with SHIV, but 10^6 XMRV on human isolated PBMC did jack shit. So why would you put 10^6 in an animal? Think of that this way– lets say you put a few cells of algae in a fish tank, and they didnt take hold to make your fish tank gross. Why would you take that same ‘dose’ of algae and expect it to take hold in the Colorado River? And when the hell is any human exposed to >10^6 XMRV IV in real life? Wat? And Im looking at previous XMRV publications that indicated APOBEC is a Big Deal with XMRV restriction. At least with HIV-1, human APOBEC, human A3G is, what, 100-fold more potent than macaque A3G? Wheres the macaque in vitro data, like the data that is the topic of this post? They talk about APOBEC all over the place, but didnt feel it was necessary to investigate human vs macaque in vitro before they decided they made an animal model?? Wtf?

That paper is a rush job that gave us very little information in exchange for 8 dead monkeys, before basic bench research was done.

We can do better.

* I want to make it clear that I am not making any ethical judgments here. Animal researchers are filed under the jobs that “Thank god someone else can do this, because I cant”, like a police officer, or daycare teacher.


  1. #1 hematophage
    February 21, 2011

    I trap wild rodents for my work, and I HATE killing them. I’ve lost like five or six to trap death, accidental anesthetic overdose or my dog, and I always feel terrible about it.

    The dead parasites, not so much.

    To continue adding nothing of substance to this (excellent) post, I will forever think of diminutive viruses as “babbies”.


  2. #2 theshortearedowl
    February 21, 2011

    Wow. Too bad the XMRV genie is already out of the bottle. This is the problem with so many things – it takes 10 seconds to say something stupid, and 2 years to prove it wrong.

    I know exactly what you mean about animal research. And rather than praised for dealing with things other people don’t want to deal with, they get their cars firebombed.

  3. #3 Poodle Stomper
    February 21, 2011

    the number of ‘mistakes’ that happen during reverse transcription is a evolved number.

    An outstanding but often overlooked point!

    I do animal research myself (mice) and while killing them is inevitable in our research we do our best to make it as humane and painless as possible. I don’t like doing it but I understand the need and do respect and appreciate the sacrifice they (unwillingly) make.

  4. #4 Kemanorel
    February 22, 2011

    I don’t think I could kill mice. They would remind me too much of my sugar gliders.

  5. #5 Paul
    February 22, 2011

    Interesting post Erv. You’re right about the need to get as much information from in vitro studies as possible before proceeding to animal studies.

    In a way I think the problem here is not to dissimilar to that with the preclinical evaluation of neuroprotective stroke treatments in the past, where the design of many preclinical studies differed so much from that of subsequent clinical trials (in time to treatment, dosage etc) that the real surprise would have been if they had yielded similar outcomes. These were problems that could and should have been avoided.

    The stroke research community does seem to have learned the lessons of past mistakes and got its house in order, and I think the rest of the biomedical research community could learn from their example. There is little room for short cuts in science; important as animal studies are to medical researchers they are not magic.

  6. #6 Poodle Stomper
    February 22, 2011


    I just googled pictures o f sugar gliders and must have one now. =) I don’t think I could sacrifice one of those either.

  7. #7 In Vitro Infidelium
    February 22, 2011

    Apart from a huge level of endemic redundancy in the XMRV ‘we’ve got an organism, now let’s find a disease’ research merry go round, which is largely what underlies the wasteful Macaque study, there’s a sick irony in that the CFS/XMRV lunatic fringe are ever more resembling the hard line antivivisectionists in their harassment of ‘unapproved of’ academics: http://www.mecfsforums.com/index.php/topic,5722.0.html

    No surprise that WPI promoter ADM, spun the Macaque study as revealing yet another miracle http://blogs.wsj.com/health/2011/02/17/xmrv-study-shows-virus-can-cause-persistent-infection-in-monkeys/tab/comments/:

    “This is a tantalizing finding because it raises the prospect that someone could be infected with XMRV but show no clinical symptoms of disease until years, possibly decades, later.”

    ADM’s fans are unfortunately unwilling to acknowledge how dumb this statement is and given ADM is now a messiah of XMRV+ (ya know like HIV+) disease, her half witted diatribes are treated as gospel. ADM has made no comment on ‘Severe Restriction of Xenotropic Murine Leukemia Virus-Related Virus Replication and Spread in Cultured Human Peripheral Blood Mononuclear Cells’, though one of the CFS forums has this gem:

    “Does this tie in with the statements Judy [Mikovits] has made about the difficulties they have found finding a cell line that XMRV will grow in. They picked LnCaP and that works fine but are having trouble finding any other cell lines. I seem to remember her saying that with a lot of other cell lines it just sits there and doesn’t increase.”

  8. #8 Kemanorel
    February 22, 2011

    I just googled pictures o f sugar gliders and must have one now. =) I don’t think I could sacrifice one of those either.

    If you thought they just looked cute, you don’t even know the half of it. They *bond* with you, and it is cute as hell when something scares them and, for protection, they run down your shirt. It’s even cuter when they jump and glide to you from somewhere to do that.

    There’s quite a few rules to learn about them if you want them, but *they are totally worth it*.

    The biggest thing to consider before getting one (I HIGHLY recommend at least two) they live 12-15 years so be ready for a commitment. Because of the way they bond you can’t just give them up, unless you have a heart that’s really just a cold, black lump of obsidian forged from the fires of Mt. Doom.

    Anyways… yeah, definitely have an adversion to killing cute little animals now. Can’t do it.

  9. #9 Douche
    February 22, 2011

    n. A turning towards; attention.

  10. #10 Kemanorel
    February 22, 2011

    n. A turning towards; attention.

    By bad, douche. I meant to say “aversion.”

  11. #11 RRM
    February 22, 2011

    @In Vitro Infidelium

    Not only WPI promotors, but WPI employees themselves are saying the most bizarre things. The clinical director (Jamie Deckoff-Jones) about the PACE study:

    “It’s unbelievable that anyone would be willing to put their name on it. While new people are being infected [with XMRV], new babies being born with it every day,…”

    About the rhesus macaques study:

    “I don’t see how anyone can read the following [rhesus macaque] paper and still be in doubt. It is a public health emergency. To spend years arguing about what isn’t, instead of coming together to discover what is, no matter how devastating, compounds the crime against humanity that has already occurred: [link to rhesus macaque study]”

    Yeah Jamie, how can we still be in doubt when Mikovits was just as good as you expect a freaking rhesus macaque to be, in differentiating between (self chosen) pedigreed positive samples and negative samples in the blinded setting of the Blood Working Group, Oh wait, Mikovits did slightly worse than mere chance would suggest…

  12. #12 Alan Dove
    February 22, 2011

    Great post, and an excellent explanation of the selective pressures that drive viral mutation rates.

    Unfortunately, now I have a voice in my head chanting “How is babby virus formed? How is babby virus formed?” For anyone who missed the meme the first time around: http://www.youtube.com/watch?v=Ll-lia-FEIY .

  13. #13 D. C. Sessions
    February 22, 2011

    it takes 10 seconds to say something stupid, and 2 years to prove it wrong.

    And forever to get the believers to quit insisting that it’s right anyway.

  14. #14 RRM
    February 22, 2011

    Another negative CDC study:


    It seems they did everything you could reasonably expect (e.g. change the cohort to more severe cases, check the serology assay on the infected rhesus macaque mentioned in this blog), and again Switzer et al. found absolutely nothing.

    Bust just check the wacko comments to see the XMRV story just evolving in another “lost” Wakefield style case:

  15. #15 Charl
    February 23, 2011

    Thanks for the link, RRM! That’s a pretty nice paper. Not a massive sample, but combining the “classic” Lombardi tests with more powerful assays too. The CDC do seem to have taken on board a lot of the criticism they received before.

  16. #16 RobertC
    February 27, 2011

    My son is a cardiac patient. First OHS at age 5 days. Many more sense.

    I know that his care would not be possible without animal testing.

    So, to all those who struggle with their work. I know it can be hard. But, from the bottom of my heart.

    Thank You.