Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Featured Medical Using CRISPR to cut HIV out of infected cells

  1. May 2, 2017 #1

    Ygggdrasil

    User Avatar
    Science Advisor

    A major impediment to curing patients with HIV is the fact that the virus integrates its DNA into the genome of the cells it infects. While antiretroviral drugs and therapies can prevent infection of new cells and kill cells that are actively producing virus, a "latent reservoir" remains that can cause an active infection to re-emerge after drug treatment has stopped. Therefore, current treatments for HIV-infected individuals requires them to take antiretroviral drugs for their entire lives.

    One approach that various groups are exploring to attempt to rid the body of these latent reservoirs is to use CRISPR gene editing technology to cut the virus out of infected cells. Recently, researchers published a study in which they performed a proof-of-principle experiment showing that they could use CRISPR to cut HIV out of infected mice (mice cannot normally be infected with HIV, a human virus, so the researchers used mice that had been engineered to carry humanized cells that are susceptible to HIV infection):

    Yin et al. In Vivo Excision of HIV-1 Provirus by saCas9 and Multiplex Single-Guide RNAs in Animal Models. Mol Ther. 25 (2017). Published online 30 March 2017. doi:10.1016/j.ymthe.2017.03.012

    See also the accompanying commentary article: http://www.sciencedirect.com/science/article/pii/S1525001617301703

    The work builds off of previous studies by the same group showing that the virus could be cut out of human cells in the laboratory. The researchers also published a smaller-scale demonstration in 2016. Their study engineers a virus (the normally non-pathogenic adeno-associated virus) to deliver CRISPR throughout the body. They are able to detect successful cutting of the virus at various sites in the infected mice's bodies.

    Still, much more work needs to be done before this work could be applied to AIDS patients. Curing patients of HIV infection is a very difficult task as it requires eliminating all latent reservoirs. While the researchers show that they can detect successful cutting of the latent provirus in various cell types, they do not show evidence that their therapy removes all proviruses or that their delivery system can target all cells that may be harboring the latent HIV reservoirs. It will likely be very difficult for the researchers' method to eliminate 100% of the virus, and testing the ability of their delivery method to target all infected cells will be particularly difficult as mouse models of HIV may not harbor the same types of reservoirs present in humans. Still, the research represents an important step forward in developing CRISPR technology into a potential cure for HIV.

    (Thanks to @mfb to pointing the study out to me)
     
    Last edited: May 10, 2017
  2. jcsd
  3. May 3, 2017 #1

    Drakkith

    User Avatar

    Staff: Mentor

    You can say that again. With a mouthful of peanut butter. "Very difficult" is likely a severe understatement.
     
  4. jcsd
  5. May 3, 2017 #2

    BillTre

    User Avatar
    Science Advisor

    Pretty cool in many ways!

    I'm not sure about how HIV works;
    1) If you cut out all the genome inserted copies, will the cytoplasmic materials be able to maintain its infection in the cells?
    2) Or will it get used up and the cellular infection reduce to nothing?
    So, do you have to use both the drugs and the Crispr or can you just use the Crispr (over longer term)?

    Does HIV really infect all the cells (or maybe cell types) in the body?

    How many cell types does the virus infect?
     
  6. May 3, 2017 #3

    Drakkith

    User Avatar

    Staff: Mentor

    I believe HIV only infects certain types of immune cells.

    From this wiki article:
    So it shouldn't be able to infect liver cells or skin cells and such.
     
  7. May 3, 2017 #4

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor
    2016 Award

  8. May 3, 2017 #5

    Ygggdrasil

    User Avatar
    Science Advisor

    Yes, that's the reason many are skeptical that such an approach could actually cure an individual of HIV.

    As you suggest, the CRISPR approach would likely have to be combined with antiretroviral drugs. The drugs would deal with cells actively producing virus and prevent infection of new cells while the CRISPR approach would hopefully deal with latently infected cells.

    As @Drakkith mentioned, HIV mainly infects immune cells (specifically, HIV latches onto the CD4 protein to gain entry into the cell, so it only infects cells expressing CD4 and other required co-receptor proteins). However, immune cells reside in many locations in the body, so the CRISPR delivery method would need to access these cells all throughout the body.
     
  9. May 3, 2017 #6

    BillTre

    User Avatar
    Science Advisor

    That's what I thought.
    I guess I was confused by one of the links saying:
    "Khalili and his team engineered the animals to incorporate specific HIV genes into nearly every cell in their body, from the brain, heart, liver, kidney lungs and spleen to their blood cells.".
    But that just the test animals.
    HIV in many cell types would present greater problems for virus targeting and removal.
     
  10. May 3, 2017 #7

    Ygggdrasil

    User Avatar
    Science Advisor

    Yes, the mouse experiments use fairly artificial systems for studying HIV (because mice don't normally get infected by HIV and one would not want to do initial studies in humans or related primates). Most other studies employ lentiviruses to deliver the CRISPR system as HIV is also a lentivirus and would be expected to show similar tropism. Whether the AAV vectors they developed can recapitulate HIV's tropism is still unclear and would need to be addressed before this approach could be applied in more practical studies.
     
  11. May 3, 2017 #8

    BillTre

    User Avatar
    Science Advisor

    They seemed vague about their success percentage-wise of removing HIV (guess that's the point of @Drakkith's comment).

    Are those numbers known?
     
  12. May 3, 2017 #9

    Ygggdrasil

    User Avatar
    Science Advisor

    Not really. The authors mostly use qualitative assays to detect provirus cleavage, but don't do much to quantify the fraction of intact provirus remaining. Probably one of the reasons why this study was not published in a higher impact journal.
     
  13. May 4, 2017 #10

    DrDu

    User Avatar
    Science Advisor

    HIV is level 2 and relatively harmless, as skin contact or inhalation won't usually lead to infection. I once worked in the development for test for HIV and Hepatitis and also had access to the labs. Working with Hepatitis B is much more risky.
    Somewhere at home I have a book on how to set up a bacteriology lab for medical students from the beginning of last century.
    It is recommended that the experiments with antrax are only to be performed by experienced students and that the wooden board on which experiments are to be performed are burnt afterwards :-)
     
  14. May 4, 2017 #11

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor
    2016 Award

    Yeah, we also stopped mouth pipetting a while ago :)

    One thing my students learn during their training in cell culture- the air we breathe is *filthy*. Full of bacteria, yeast, mycoplasma, mold spores... you name it, you breathe it in with every breath. One gains considerable experience in microbiology while doing cell culture.
     
  15. May 10, 2017 #12
    On the one hand, you want to be certain that HIV sequences are excised from 100% of genomes containing them. On the other hand, you need to prevent expressable DNA from even one cell from being excised if it's not an essential HIV gene. If one integrated HIV genome is ever expressed, a multitude of virions will be produced and infect other cells and a new infection will result. If one cell suffers a genetic disruption caused by mis-targeted CRISPR, mightn't a cancer be the result? That's the problem with biology. A single elementary event can have compounded consequences.
     
  16. May 10, 2017 #13

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    A bit of collateral damage can be tolerable. It can lead to a slightly higher risk of cancer, but if you can cure HIV that way that is perfectly fine. An airplane flight gives you a slightly higher risk of cancer as well, and people use it just to get from A to B.
     
  17. May 11, 2017 #14
    Quite true. A quantitative risk-benefit analysis must be done. That requires data which we don't have yet.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Using CRISPR to cut HIV out of infected cells
Loading...