Oops: Contaminated cell-lines from the NIH

One of more genuinely stupid comments I hear about HIV-1 research goes along these lines:

Why we spenden so much moneys on HIV when we aint got no vaccine and we could be spendin it on cancer/heart disease/diabeetus/whatever disease I personally am at risk for?

No, we dont have a vaccine for HIV-1 yet. But that doesnt mean HIV-1 research for the past 25 years has been ‘worthless’. HIV-1 research has wildly expanded our understanding of the human immune system, ie helps all of us.

*rolleyes*

The field has also developed an army of assays, cell lines, reagents, a million tools for characterizing HIV that did not exist before 1984. But the development of all of these new tools created a new problem: standardization. If every lab is using a different ‘home-made’ assay to study HIV, how do you compare their results?


So in 1988 the National Institute of Health created a ‘bank’ for all the tools labs had developed to help keep things standardized, the AIDS Reagent and Reference and Reagent Program. You dont have to develop and characterize your own antibodies to gp120 anymore– you just order #288 and you know exactly how much to use for a Western Blot or ELISA or whatever, and everyone ‘believes’ your results.

One of the reagents I use every day is a cell line called TZM-bl. TZM-bls are cells that used to be cervical cancer, but HIV-1 researchers got it to express all the receptors HIV-1 needs (CD4, CXCR4, CCR5) so TZMs are suceptable to HIV infection. TZMs also have a couple of genes inserted into their genome that are under the control of HIV Tat, β-galactosidase and luciferase. So, when TZMs are infected with HIV-1, they will stain blue or glow (depending on what assay you want to use).

These assays are really useful when youre trying to figure out how many infectious viruses you have in a stock solution (with HIV, you have a lot more dead viruses than you have ‘live’– you dont want to count the dead ones). You take your stock, do a series of dilutions, and see how far out you have to dilute the stock to get one blue TZM cell. Say you have to dilute it 1,000 times– that means you have 1,000 infectious viruses per X volume. So if you need 5,000 viruses for your experiment, you know exactly how much of your stock solution you need. Yay!

Well, heres where the ‘oops’ comes in. Every HIV-1 lab in the country uses TZM-bl cells for something. All from the same stock from the NIH reagent bank. Turns out this stock has been contaminated with another retrovirus– murine leukemia virus.

Identification of Gammaretroviruses Constitutively Released from Cell Lines Used for Human Immunodeficiency Virus Research

This doesnt effect me at all, personally. MLV is a gamma retrovirus, so it has no effect on the blue/glow assay I do every day (MLV doesnt have the right proteins to interfere with that assay, which is HIV specific), BUT it could be screwing over other labs if they are doing something different (like measuring reverse transcriptase activity levels).

This paper brings up a very, very important lesson: ALWAYS DO ALL THE RIGHT CONTROLS IN YOUR EXPERIMENTS, INCLUDING MOCK/NEGATIVE CONTROLS!!!

This lab discovered that something was ‘wrong’ with their cells (which turned out to be ALL of our cells) by doing all of the appropriate controls… and getting unexpected results. If they had skipped their negative controls, they wouldnt have noticed that they were getting reverse transcriptase activity off of their uninfected cells.

Can never have too many controls.

Comments

  1. #1 Becca
    January 5, 2009

    obnoxiously technical point- Doesn’t our good old LacZ (I presume the source of that lovely classical blue color) produce the pigment if you put it on the right substrate? And isn’t this, strictly speaking, an enzymatic process of the organism and not a “dye”?

  2. #2 JustaTech
    January 5, 2009

    Ah, ERV, you’ve hit on my boss’ favorite complaint about immunologists! “More controls! I want more controls than experimental samples!”

    Of course, our work is to build a standardized HIV-vaccine non-primate model for quantitative testing.

    How many controls do you usually do? 1 positive and 2 negative?

  3. #3 Peggy
    January 5, 2009

    Are there really that many labs where people don’t do negative controls? I felt like that was really pounded into my head when I was a grad student.

    Becca: I think ERV explained it correctly – the assay she uses causes the cells to stain blue (or glow) when HIV is present in the cells that have the expression of a reporter enzyme (e.g., beta-galactosidase/lacZ) under control of HIV-tat. She just didn’t include all the steps of the assay, which, like you say, involve adding a substrate which the enzyme converts into a detectable product.

  4. #4 Chris Noble
    January 6, 2009

    Abbie, your real problem is that the AIDS Reagent Program doesn’t exist. It’s a figment of your imagination.

    HIV/AIDS and parapsychology: science or pseudo-science?

    To establish such claims, one might ask at the minimum to be given samples of actual particles of this miraculous agent; but they are as unavailable as the poltergeists and apparitions that constitute some of the more exotic claims made by some parapsychologists.

    I don’t know what sort of cellular debris you’re playing with but it isn’t HIV.

    PS. The Loch Ness monster is real.

  5. #5 Brian
    January 6, 2009

    the AIDS Reagent and Reference and Reagent Program.

    Unless John Cleese is in charge, I’m pretty sure there’s one too many Reagents in that name.

  6. #6 Rrr
    January 6, 2009
    the AIDS Reagent and Reference and Reagent Program.

    Unless John Cleese is in charge, I’m pretty sure there’s one too many Reagents in that name.

    Posted by: Brian | January 6, 2009 6:06 AM

    That’s most probably likely due to a double negative control, as ordered by the chief head boss of his nest egg, Department of Redundancy Department.

  7. #7 LHD
    January 6, 2009

    I think it’s interesting that you put an objection that you find insipid, ignorant and stupid in a Southern accent. I think someone needs to get over their inane prejudice. Just saying.

  8. #8 Lab Rat
    January 6, 2009

    heh, on a similar note, my supervisor had some stuff in the fridge that was meant to be campylobacter, we’ve a strong suspicion that it’s actually artobacter and they’ve sent us the wrong stuff. Crossed lines on the order form I expect…

  9. #9 MattK
    January 6, 2009

    I think it’s interesting that you put an objection that you find insipid, ignorant and stupid in a Southern accent. I think someone needs to get over their inane prejudice. Just saying.

    I think it’s grammatically incorrect rather than southern specifically(under educated Canadians also say “ain’t”) In fact, one could argue that it is more lolspeak rather than southern.

    On another note, Abbey, did you see that Oklahoma is trying to pass legislation to ‘teach the controversy’? Stranger fruit has a post about it.

  10. #10 MattK
    January 6, 2009

    oops, I meant Abbie

  11. #11 Paper Hand
    January 7, 2009

    This doesnt effect me at all,

    Affect, not effect

  12. #12 a lurker
    January 8, 2009

    Misinformed ? stupid

  13. #13 TomJoe
    January 8, 2009

    ALWAYS DO ALL THE RIGHT CONTROLS IN YOUR EXPERIMENTS, INCLUDING MOCK/NEGATIVE CONTROLS!!!

    Damn skippy. To which, if you’re not using controls, you’re not doing science. You’re wasting time and money.

    On a related note, when you request items from a culture collection, it’s always wise to test it once you receive it. When I request bacterial cultures, I always amplify the 16S gene and send it off for sequencing to confirm that the isolate I have is the actual organism I’m attempting to study. No point in performing a battery of tests only to find out that your organism in question is something entirely different … and utterly useless.

  14. #14 Roland N
    February 19, 2009

    Several good comments precede this one. They point to the lack of adequate controls and the false assumption that all cell lines received from major repositories need no further scrutiny. Much cross-contamination and misidentification of cell lines occurs soon after the cultures are purchased. The problem is rampant and warrants our careful attention and action. See Curbing Rampant Cross-Contamination and Misidentification, 2008 BioTechniques, 45:221-227. This is a major problem regarding scientific integrity and good science.

  15. #15 Roland Nardone
    February 19, 2009

    The problem of cross-contamination of cell lines has been around for 45 years and is continuing unabated. Editors of science journals, Boards of professional societies, and many of those setting policy for the award of research grants have not responded to the challenge with imagination and courage. Where is the leadership the public deserves?

The site is undergoing maintenance presently. Commenting has been disabled. Please check back later!