CRISPR, about to be crisped

 a recent quite shocking article found that disruptions of the genome from double-stranded DNA breaks, as happens with CRISPR, seem to lead to long-term genetic dysfunction with profound effects both locally and more remotely in the genome; these adverse changes persist through several rounds of successive cell divisions and are likely inherited by offspring: see gene editing leaves heritable impairment Science2025 in dropbox or DOI: 10.1126/science. adk6662.  There is lots of complex biology in the article. i will summarize major points below

A few background clarifications:

-- Cas9 (CRISPR-associated protein 9) is effectively a molecular scissors that cuts double stranded DNA at precise locations in the genome and is the methodology allowing CRISPR to then alter the genetic material. this entire process is commonly referred to as CRISPR/Cas9: https://pmc.ncbi.nlm.nih.gov/articles/PMC4975809/

-- chromatin is the complex, compact, coiled 3-dimensional structure involving DNA wrapped around a histone scaffold. packages of this massive amount of DNA being in this very compact form allows the chromatin to fit in the cell nucleus and also allows the coiled chromatin to interact with even distant areas of the genome. in these more remote connections, this contact of chromatin with the genome facilitates interactions that influence the gene expression through methylation, where the added methyl groups are added to the DNA and typically silence the genes by preventing transcription without altering the DNA sequence

        -- in the event that the chromatin did not reconfigure as it was before the Cas9-induced double stranded DNA break, these connections between the chromatin and the more distant parts of the genome would effectively be broken, thereby profoundly affecting the gene function by silencing those important baseline chromatin interactions

        -- and there is the potential that these epigenetic effects changes can also pass through to offspring (https://pmc.ncbi.nlm.nih.gov/articles/PMC4264925/ or https://gmodestmedblogs.blogspot.com/2015/07/marijuana-passing-through-generations.html)

            -- another article found that reductions in gene expression in chromosomes can be heritable: https://www.nature.com/articles/s41586-023-06157-7

-- the outcome tested by this study is whether after the double stranded breaks in the gene by Cas9 is done and the CRISPR patches the break with new DNA material, does the chromatin revert to its natural folding and interactions with the genome or not??

Details:

-- this study employed cells of the human HeLa Kyoto cervical cancer cell line that were grown under standard cell culture conditions (5% CO2, humidified atmosphere) in Dulbecco’s modified Eagle’s medium containing 10% heat-inactivated FBS and penicillin–streptomycin antibiotics

-- these HeLa cell lines had genetic modifications of the MYC-EGFP-mAID clones on the genome

    -- the c-MYC gene is transcribed in a topology-dependent manner, making the c-MYC TAD (topologically-associated domain, that has conformation-sensitive genes that could affect long-lasting chromatin alterations), an "ideal environment to quantitatively interrogate the dynamics, magnitude, and temporal fluctuation of gene expression after disrupting the 3D chromatin arrangement after DNA breakage"

    -- the researchers established 12 Cas9 cut sites spanning the entire c-MYC topologically associate domains in order to see if there were specific sites on c-MYC causing aberrations in the effect of the double-stranded cuts along the c-MYC TAD

-- the researchers performed the following:

    -- Cas9-inducted double stranded breaks (DSBs) in genomic loci harboring topologically sensitive protein-coding genes as well as regulatory RNA species in order to assess the long-term consequences of the DNA breakage on chromatin topology and gene activity

    -- quantitative imaging of large cell populations, DNA and RNA fluorescence in situ hybridization (FISH), and Region Capture Micro-C as readouts

Results:

-- A single Cas9 DSB anywhere in the c-MYC TAD has a protracted effect on c-MYC protein expression

    -- the DSB-induced chromatin alterations created by the Cas-9 scissors do not recover to pre-damage level; these alterations persist as lasting changes in the 3D arrangement; a single cut in the DNA (vs a double-stranded cut) did not have this effect

    -- as a result, there are impaired gene expression throughout large chromatin neighborhoods, affecting the genome across the entire c-MYC TAD (topologically-associated domain) and at sites far from the DSB

    -- as a further test of whether the c-MYC TAD results are generalizable, they assessed the mRNA production of several genes within or in the immediate neighborhood of the c-MYC TAD and found mRNA production was partially decreased after 6 hours of the Cas9 cuts, confirming a wide reach of acute transcriptional repression during ongoing DSB repair 

-- Alteration of c-MYC expression after a single Cas9 DSB is accompanied by changes in the 3-D chromatin structure and is heritable to daughter cells over several generations

    -- they performed six Cas9 cuts across the c-MYC TAD, and they assayed c-MYC protein production after 24, 48 and up to 96 hours after the Cas9 RNP transfection, finding that the c-MYC repression was maintained the subsequent 6-72 hours

        -- there was no difference in the reduction of c-MYC protein levels vs control cells transfected with non-targeting Cas9 cuts

        -- and, there was a loss of chromatin interactions to upstream sites at all time points measured

    -- a single DSB led to an intrinsic memory of the DNA damage

    -- and this was propagated after multiple cell divisions to the daughter cells, finding inheritability of these initial changes

-- The DSB-recovered c-MYC locus remains less responsive to physiological stimulation

    -- the researchers then quantified the c-MYC TAD susceptibility to respond to epidermal growth factor (EGF) stimulation, a canonical driver of c-MYC expression

    -- there was a 60-80% drop in the ability of gene regulatory elements to sense and respond to physiological stimuli

Conclusions:

-- their findings in this study may well apply to all types of gene editing, whether SDN1 (gene disruption), SDN2 (gene modification via insertion of a repair template), or SDN3 (gene insertion), since these all require a double-stranded DNA break (DSB)

-- one implication of this study is that patients who already had CRISPR/Cas9-based gene therapy (eg sickle-cell, beta-thalassemia) should be examined for altered gene expression

-- there are other factors that lead to double-stranded DNA breaks, such as ionizing radiation, mutagenic chemicals, nuclear accidents, exposure to radioactive chemicals, etc; these can also possibly lead to chromosomal dysfunction

    -- as with all genetic mutations, there may be some that are actually beneficial (a documented minority of them), and over evolutionary time (which won't be very consoling to the individuals involved) many well evolve to be beneficial

-- the researchers called this phenomenon of chromatin distortion from CRISPR "chromatin fatigue" 

-- CRISPR/Cas9 can therefore lead to major genetic changes, called chromothripsis, "an extremely damaging form of genomic rearrangement that results from the shattering of individual chromosomes and the subsequent rejoining of the pieces in a haphazard order" https://www.gmwatch.org/en/106-news/latest-news/19885 , notable because:

    -- this seems to be an irreversible process with this "on-target" genetic manipulation of CRISPR/Cas-9 that then leads to upstream genetic dysfunction

    -- it is fundamental to the targeted change from CRISPR/Cas9; so even advances in the technology of CRISPR in the future to make it more precisely targeted will not help, since CRISPR itself requires the initial double-strand breaks

    -- the upstream results in the genome could potentially be carcinogenic

        -- and the effects could be an inherited disease in the affected patients children

    -- there is therefore a push for more CRISPR regulation, given the unintended outcomes of gene editing

-- Plants are routinely exposed to mutagenesis through similar mechanisms (eg CRISPR/Cas), which in many instances leads to deformed, infertile, and non-viable plants (https://www.gmwatch.org/en/106-news/latest-news/20621-gene-editing-disrupts-multiple-gene-functions-through-large-scale-epigenetic-changes-in-a-way-that-persists-through-successive-cell-generations?utm_medium=email&utm_source=sendpress&utm_campaign)

    -- these types of unintended consequences are a concern in conventional plant breeding, but gene editing is 1000-10,000 times more powerful a mutagen than chemical and radiation-based mutagenesis, which in turn is far more mutagenic than natural reproduction

        -- genetic modifications seem to be much worse than the past non-targeted mutagenesis that has been used for decades by conventional plant breeders. chromothripsis (as described above) has been found: https://www.gmwatch.org/en/news/archive/2019/19223

        -- https://www.mdpi.com/2673-6284/10/3/10: this review surveys the literature for unintended outcomes from old and new genetic modification techniques (including gene editing) in plants. It reveals a wide variety of unintended consequences, from small to large genomic alterations, giving rise to "legitimate safety concerns". 

        -- so, with CRISPR gene editing, there seems to be quite similar results in gene-edited plants as in humans

              -- The authors also found that only five articles used whole-genome sequencing, which could identify adverse effects in plants. they speculate that some larger DNA damages could only be found using long-read sequencing, which is generally not performed. As a result, the occurrence of unintended DNA damages is underestimated in the literature: https://www.mdpi.com/2223-7747/11/21/2997 and https://www.mdpi.com/2673-6284/10/3/10

    -- there is ensuing chromatin fatigue in plants as in humans

    -- another study found that CRISPR-Cas9 gene editing with short insertion or deletion mutations is associated with a range of different gene sequence rearrangements as unintended outcomes: https://www.mdpi.com/2223-7747/11/21/2997 

    -- these concerns have led to strict anti-GMO restrictions in many countries (https://www.gmwatch.org/en/106-news/latest-news/20504 and https://www.mdpi.com/2673-6284/10/3/10 .

-- as tangential comment on unintended consequences which can happen with any new therapy:

    -- i have not prescribed DPP-4 inhibitors for patients with diabetes, since:

        -- DPP-4 is serine protease located throughout the body that targets a very wide array of proteins, including the incretins GLP1 and GIP (the target for diabetes), but also B-natriuretic peptide, stromal cell-derived Factor-1, CCL5 (a chemokine that attracts immune cells to sites of inflammation), CXCL10 and CXCL11 (chemokines that recruit T-cells to inflammatory sites, CCL11 (involved in allergic responses and eosinophil recruitment), growth-hormone releasing factor, GLP-2 (involved in intestinal growth and function), substance P and others; many of these could affect the immune system

        -- and DPP-4s are really minimally effective for lowering A1c levels, only in the 0.5-0.8% range

        -- ie, not a great benefit, and having the potential for distorting many important systems in our bodies

    -- this DPP-4 inhibitor example is more extreme than others (we did know in advance that DPP-4 is an important enzyme with multiple effects in the body), it does reinforce to me that we should always be wary of new meds that have become available recently and do not have enough time to detect longer-term adverse outcomes (does cancer develop more often 5 years later because of undetected effects on one of the many aspects of our immune system??). i am particularly concerned about all of the potentially severe long-term problems with the monoclonal antibodies that target the immune system

        -- so, my summary is that unless there is a clear, profound benefit (eg, GLP-1's for diabetes, or a new and dramatically better meds for HIV such as INSTIs), why start a new med which has potentially profound unintended consequences for a disease that has an array of other meds that work well????  so the caveat to me is that if older tried-and-true meds work well and the new ones do not offer a dramatic improvement, why not wait for more and longer-term studies on the new ones??

Limitations/further questions:

-- it is important to know if these chromatin derangements from double-stranded breaks in the DNA that can pass on to daughter cells over several generations are found in other areas of the genome (this study focused on the MYC-EGFP-mAID clones, though the authors do argue that this should be representative of other areas)

    -- clearly there should be other studies done confirming the above conclusions

    -- as mentioned, it would be useful to assess patients who have already had Cas9/CRISPR done, such as those with sickle cell disease or thalassemia, to see if the above chromatin findings are evident. and, given that CRISPR has been around for several years (the first human CRISPR for sickle cell was in 2019, and regulatory approval was in late 2023), we would have longer-term information on the status of the chromatin and resulting gene function

so,

-- this study raises profound questions about the utility/potentially very serious unintended adverse effects of CRISPR for patients and their offspring

-- and it brings up the general issue of unintended adverse outcomes, which really does need to be studied for all new procedures and therapies over time

-- in the case of CRISPR, there really needs to be more information, including assessing other targeted genetic locations and their specific effects on chromatin and gene function. one useful study that could be done now would be to test the chromatin effects in patients who have had sickle cell or thalassemia CRISPR done. this would also help elucidate the very long term effects of CRISPR on chromatin

    -- and, it would be very useful to assess the offspring of these individuals. do they have even subtle derqangements of their immune system? are they more prone to infections? cancers? other diseases?

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them