Every Cell in Your Body Has the Same DNA. Except It Doesn’t.

May 21, 2018 · 63 comments
Dr. Kim Meyers (Glenview, Illinois)
The case of the infant with the heart arrhythmia illustrated, is an example of a germline mutation. That is an example of genetic chimerism. When one thinks of a tumor in one of our organs, such as a kidney cancer, one sees an example of genetic mosaicism. In that situation, the vast majority of the kidney cells have retained the genes necessary to suppress tumor growth. In some of the cells, the mutations have overcome the normal controls at the cellular level, and cancer occurs. All of us have genetic mosaicism that we can see with our own eyes. All you have to do, is take off your shirt and look in the mirror. You will see various growths and pigmentations that represent small mutations leading to benign tumors on the surface of your skin. While the overwhelming majority of your skin cells have the same basic genetic code, the cells that have mutations, lead to the various lesions we see on our skin. Many of these growth’s occur as we get older. For instance, seborrheic keratoses, are an example of how multiple mutations over decades will accumulate. This leads to DNA that creates small benign growths due to multiple mutations after millions of cell divisions. As a physician, it took me quite a long time to realize why tumors appear in otherwise normal organs. Genetic mosaicism is the cause of that. This explains why a small number of our trillions of cells lead to a small number of “cellular criminals” that do us harm.
Philly Spartan (Philadelphia, PA)
How is Astrea doing? Well, I hope.
Rosy (Dallas)
I'm a bit confused. What's the difference between mosaicism and masochism?
grodh2 (Charlotte, NC)
Very interesting, I was wondering about Beckwith-Wiedemann Syndrome, with hemihypertrophy, one side of the body is larger than the other, with increased risk for tumors such as hepatoblastoma and others, perhaps they have mosaic cells, like the person with different brain "halves". Also Lyonization, where women randomly inactivate one X in each cell, sometimes it winds up with much more of one X inactivated compared to the other, and so may have symptoms of an X with a mutation. But that would be very unlikely if it were random. Maybe what is happening is different mosacisms of the cells.
Pontifikate (san francisco)
Fascinating article. I'm a lay person who wonders if Meckle's diverticulum may be related to mosaicism. My brother had a Meckle's and the way it was explained to me was that a stomach cell had migrated to his intestine when he was a zygote or fetus. Makes one think about your gut having a brain (if brain cells migrated there). Having a gut instinct may have some truth.
Marcie Strauss (White Plains, NY)
What are the implications of mosaicism for identical twins? Does it provide an explanation for why even though these type of twins start out with the same exact DNA/genome, they may end up being different in more ways than expected?
IN (NYC)
The writer replies to many of the comments, and clarifies some of the concerns. This is a wonderful way to enhance the value of any article that introduces new and important ideas and findings.
Henry (Woodstock, NY)
Does this have any impact on forensics?
Kaikopere (Ohakune)
I was wondering about forensics too. Presumably some of the 9/11 victims had mosaic DNA. Did they also have enough unique DNA for all their body parts to be matched up? And if police take bits of a mosaic attacker's skin from under an assault victim's fingernails, and there is only a record of his blood DNA on the police data base, will there still be enough similarities in the skin DNA for an arrest?
Don (NH)
@Kaikopere There is no DNA in an erythrocyte (red blood cell) because they're not nucleated cells. Birds and dinosaurs on the other hand do have nucleated red blood cells hence the premise in Jurassic Park. Not unlike you, I was curious about DNA evidence but then realized that regardless of where a specimen is taken or evidence recovered from there will never be 100% mosaicism or vice versa, hence it's incredible accuracy.
Sarah Townsend (Vancouver BC)
Our Dear Mr. Zimmer, You have done it again - this is marvellous ! I was especially taken with your witches broom analogy; just a brilliant way to illustrate a complex phenomenon. Thank you for your insightful and imaginative work. You, sir, are a treasure.
Tom Hayden (Minneapolis)
I'm guessing natural selection normally weeds out dangerous imperfections/mosaics, and the beat/species carries on. We may think of our genome as stable but evolution tells a much different and ephemeral story.
Greeley Miklashek, MD (Spring Green, WI)
Interesting collection of anecdotal findings. However, my understanding of human population genetics derived from micro-DNA analyses is that modern humans have a gross 99.9% overall genetic identity and that the only other mammal with this degree of genetic inter-member identity is the blind mole rat. I read somewhere (can't find the reference now) that only 18 species are known to have our degree of genetic identity, and that the 16 other than us and the mole rat are all social insects. This was long thought to be from the nuclear winter of the Sumatran super-volcano Mt. Toba 73.5kya, but the genetic bottleneck responsible is now thought to have been about 50kya. In any case, this amazing genetic similarity between all humans on earth today strongly suggests that our "diseases of civilization" are caused by environmental effects. Our hunter-gatherer contemporaries have very few of our diseases. In Donnison's 1937 book about his experiences treating mostly traditional living Kenyan hunter-gatherers and pastoralists, combined with the similar medical practices of his fellow Brits in Africa, on pages 10-11, records 238,851 patients examined, but not one case of heart disease found! We're looking in the wrong place to explain our increasing load of common "diseases of civilization". Our modern Western "built" environment is the real culprit and I just published a book laying out over 100 supporting references: "Stress R Us". It's on Amazon Books. Stress R Us
Julie (Denver)
How could it not matter that what was commonly believed to be our genetic signiture to the degree that some religious folks have used “our unique genetic code is proof of a soul” and DNA evidence is being used in the courts to both convict and release crime suspects? The implications of such findings seem more than just anecdotal.
Jippo (Boston)
I posit that there was no heart disease found in the populations you referenced because individuals with those conditions died out.
Greeley Miklashek, MD (Spring Green, WI)
Actually they would have even longer lifespans than us, were it not for our constant heroic medical care, which is now costing us nearly 25% of the GDP. We are terribly sick and dying; they are not. Read "Stress R Us" for documentation or just savor your own unfounded bias. Stress R Us
James Gaston (Vancouver island)
Fascinating! Thank you.
Daniel B (Granger, In)
These findings may confirm the concept of fluid sexuality in humans.
Caroline (San Jose)
“The skin cells in her saliva, the bladder cells in her urine and her blood cells each originated from a different layer of cells in two-week-old embryos.” The idea of mixed cells in organs reminds me of endometriosis where the endo cells show up, cramp, and bleed in other areas of the body besides the uterus. I have no conclusion or question other than sometimes patterns emerge from disparate information and help solve a problem. Something that I learned this week (from an interview on NPR’s Fresh Air) was that the placenta is a precursor to all of the baby’s organs. Before those organs develop, the placenta serves as the kidneys, the lungs, the brain etc. for the growing fetus. It make me wonder what role the placenta has in this process of mosaicism. I am obviously not a medical person, but love trying to make linkages across ideas.
Dexter Ford (Manhattan Beach, CA)
There are a few compelling, obvious questions that were not addressed, but which should have been. How identical are apparently identical twins? How identical can they be? Corollary: how accurate can popular DNA sequencing tests be, if different samples of tissue and/or saliva can be profoundly genetically distinct? And doesn't this put the entire field of DNA testing, as used in solving crimes, in grave doubt? The science is based on the presumption that the DNA found in a small voluntary sample, or even on a discarded coffee cup, is representative of every cell in a suspect's body. Samples from relatives have been used recently to track down suspected serial killers. If it's not true that an overwhelming majority of a person's cells are genetically identical, science, society and the law have placed their faith in an entirely specious form of identification by genetic-testing.
gratis (Colorado)
Thanks for the article. It is news to me, but i am not surprised. Nature loves diversity. There is no "normal" because no two anythings are the same. Everything is different. Nature keeps telling us this, but people keep wanting uniformity.
joymars (Provence)
This fills in needed explanation for biological evolution. Mutations happen constantly and are everywhere, not just in one individual and only at its birth. I’m tempted to think that how non-organ parts of the body age has something to do with this sort of mutating. Much aging is not caused by ordinary wear and tear, like it is in the case of skin, heart or liver. For instance, eyebrows age. What signals that? It could be these ordinary mutations. There is no genetic marker for that kind of aging because it happens after fertility. But given enough time, these random mutations would happen a lot. So maybe what looks like an aging part of the body is actually just accumulated random mutations. ?
morphd (midwest)
So it appears 'natural variation' often occurs within an individual organism and not just among members of a species. As most biologists understand, the vast majority of genetic mutations that underlie natural variation are either detrimental or neutral. It's only the rare exception that leads to something positive for the organism. And the phenomenon was observed in some fashion well before all the supposed ills of modern technology arrived on the scene. One might hope information like this would undermine the idealized but false notion of Nature's perfection that seems to drive many anti-GMO and anti-vaccine activists. But asking for rational thinking from them is probably asking for too much.
marek pyka (USA)
You make a vast, erroneous assumption that mutations are spontaneous and not caused by some environmental object, such as tetraflourocarbons or some other tissue-ripping molecule.
Lou (Oregon)
I read this with interest. Some of my ow research is in tracking chimeric combinations in humans. Has anyone tracking these mutations, cross checked to see if there was a chimeric component. Many examples now exist, showing the fusion of two or more embryos, into a chimera. A famous example showed the children of one mother, were actually the product of her "twin" whose genes built the patients Reproductive elements.
Mykeljon (Canada)
Does this mean that DNA evidence in court is not as reliable as we had assumed?
Danie (Martin)
I'm awaiting testing for a form of mosaicism. I have 2 unknown versions of the gene for a rare genetic condition called Fanconi Anemia. A blood test came back normal but in some cases the rest of the body can have a different version of the gene, so I need the test repeated with a skin biopsy to be sure. That's because it affects DNA repair and random changes in the bone marrow can sometimes lead to a HEALTHY version appearing and crowding out the unhealthy marrow cells; sort of a natural gene therapy. But the rest of the body stays with the unhealthy version.
Cookie Monster (San Diego)
I did not know that. Thanks.
JulieN (Southern CA)
In my clinical lab, we do an assay for mosaicism found in people with overgrowth of a portion of their bodies and other related symptoms. The only way to successfully find these kinds of genomic variants is by very deep sequencing, meaning they kind of sequencing that reads through more than 1000 copies of the gene to detect the low level of mosaic cells. This only works when we test the tissue that is affected in most cases. If you are concerned that you have one of these syndromes, you would need to see a clinical geneticist. Note that the direct-to-consumer tests will not likely find mosiacism at this level because they don't have the ability to sequence like this. I think that in some ways we are all mosaics, and that makes sense if you think about the numbers of cell divisions that have to happen successfully to build a human. If a single cell developes an error in the DNA in one division of one cell, that cell may go on to replicate many thousands of times, and thus the error spreads. We see this in people who have "pre-cancerous" results from biopsies, and we see this in children with birth defects or overgrowth. Mosaicism is not the same as chimerism, and has nothing to do with disorders such as Sickle Cell, which is inherited from both parents who are carriers of this disease, in most cases. Please see a geneticist or genetic counselor if you have questions about your health or the health of your child in relationship to this issue.
Maya Posch (Germany)
The implications of this research on healthcare is likely to be massive. We're so used to taking a bit of blood, running tests on its white blood cells and determine the genetic makeup of an individual from that. This throws all of that out of the window. As a chimaera myself (46, XX/46, XY), I'm acutely aware of the implications of having dissimilar DNA and even complete cell lines within one's body's structures. Though one can distinguish chimaerism and mosaicism, I would postulate that the effective difference is far less significant than one might assume at first glance. My body is undergoing changes which would not happen to a regular, non-chimaeric person, such as going through a second puberty which completely shifts fundamental responses to hormones around. Starting off with two cell lines as a zygote from merged zygotes is very similar in that regard to having diverging cell lines during development.
FunkyIrishman (member of the resistance)
This is exciting research and discovery, however... What I am concerned about ( and many a scientist/person as well ) is where the powers that be/corporations are trying to patent every step or discovery along the way. This is preventing a rush to discovery from as many people and for as many people as possible, and putting us on paths that may not be the best for the whole of humanity. We must remain vigil.
Paul Knoepfler (Davis, CA)
Great piece. In a general sense this has been known for a long time, but the more we learn, the more interesting and complicated it gets. I wonder about connections to autoimmunity. Also, this likely could impact effectiveness of CRISPR gene editing clinically in the future.
Bears (Kansas City, MO)
Hewson Swift was the first to measure the amount of DNA in various types of cells and in organelles such as mitochondria and chloroplasts, work that helped convince biologists that they had their own genomes. His finding that nearly all cells from an animal have the same amount of DNA, and that germ cells have half as much, confirmed the principal of DNA constancy and helped bring to an end any lingering skepticism about the genetic role of DNA. The exceptions cited prove the rule.
Nicole Iovine (Gainesville Florida)
This article focuses on the negative impact of mosaicism, but we should remember that the presence of whatever percentage of normal cells can mean the difference btwn fetal/embryonic non-viability, shortened life-span or severe morbidity versus survival and hopefully a good life.
Kodali (VA)
This adds complexity to precision medicine. Different parts of the body has a different response for a given cancer treatment. It is even more important for cancer patients donate their tissues to national tissue bank so that future drug developments can not only be patient specific but also organ specific.
Stefan (PA)
A lot of the commentators here are mixing up chimerism with mosaicism. In chimerism, a person's cells may have 2 distinct genotypes due to a fusion of more than one fertilized zygote. For these individuals, DNA tests may identify 2 distinct individual profiles from the same sample. In mosaicism, some cells may carry mutations while others don't but the cells are both from the same genetic individual. Furthermore, witches broom is more like cancer than it is typical mosaicism described in this article. In witches broom the growth is typically caused by pathogens such as viruses
Julie stein (Boca)
I had an allo stem cell transplant. I am a chimera!
S (C)
Can sex-determining X and Y chromosomes also vary across cells - mosaic - in some individuals?
Kara Ben Nemsi (On the Orient Express)
Non-disjunction can occur at any cell division, so, yes. https://www.sciencedirect.com/topics/biochemistry-genetics-and-molecular...
A Feldman (PA)
Yes! Mutations of the the X chromosome are especially common. Changes in X and Y come in many flavors, many compatible with life, though some health issues, and many terrific people walking around with all manner of variant X0/XX (search Turner’s Syndrome) 45,X/46,XY mosaicism, or X0/XY mosaicism and mixed gonadal dysgenesis, Klinefelter syndrome 47,XXY many more. Folk with mutations of one of the sex chromosomes are more likely to carry mosaic cell lines than not, and there are lots of combinations. Many woman have some XY cell line, many men an extra X.
TrenchCoat (New York)
Wondering how this research is aligned with chimerism -- where an embryonic twin identical or fraternal may be absorbed into a surviving fetus that carries cells with separate DNA (in the case of a fraternal twin.
kenw (palo alto, ca)
I have read The Origin of Species and many commentaries on Darwin's work, but I have never seen any quote of Darwin's that mentions the inheritance of "particles" that determine physical traits. If anyone knows of such a statement I would really like to know the source.
Kaleberg (Port Angeles, WA)
Darwin did not know the mechanism of inheritance. He knew that traits were passed down to offspring, but he didn't know how. He published Origin of the Species in 1859, several years before Gregor Mendel published his work on pea plants. At any rate, Mendel's work remained virtually unknown until 1900, at which point it became a basis of what is now the science of genetics. It is not necessary to understand a mechanism, in this case transmission of genes, to recognize its effects, in this case, evolution.
rgarcia (Maryland)
https://embryo.asu.edu/pages/charles-darwins-theory-pangenesis
Stefan (PA)
See: Darwin C. The Variation of Animals and Plants under Domestication 1868 John Murray This work by Darwin discusses his concept on Pangenesis and it came after Origin of Species. "...postulating the existence of numerous and minute hereditary molecules, termed gemmules, which were supposed to be thrown off by all cells of the body, to multiply by self-replication, to be dispersed through tissues or from cell to cell, to be incorporated into the reproductive cells, and to be transmitted from parents to offspring." https://academic.oup.com/bioscience/article/64/11/1037/2754242
Dan Styer (Wakeman, OH)
"Those seeds have produced pink grapefruit trees ever since. Grapefruits are propagated by budding or by grafting, not through seeds.
Jami Flicker (Tualatin, OR)
My daughter has tuberous sclerosis complex most likely connected to a mutation on her TSC2. This genetic multisystem disorder has a high incidence of epilepsy/autism. When she was diagnosed at 4 months, we met with a geneticist who explained the possibilities of mosaicism. This was very interesting to me, Thank you. Jami
Agent GG (Austin, TX)
What are the legal consequences for identifying someone using a DNA sample, if the DNA sample is not unique or uniform?
morphd (midwest)
Assuming there are sufficient genetic markers I'd think false positives would be exceedingly rare though perhaps mosaicism could lead to false negatives at some low frequency (e.g. one or more genetic markers vary between say blood and semen samples). Just my speculation as a plant molecular geneticist.
Keely (NJ)
I was born with Sickle Cell, a blood disorder caused by a hemoglobin mutation and I always wondered if my body produced all misshapen cells or were some normal and some not? I still don't know but this article makes me think its the latter.
Healthy Nurse (Chicago)
This makes me interested in how this would affect the very popular heredity tests everyone is doing (like 23 and Me), if all cells aren't of similar composition. So do these tests sample a large number of cells, or only one? If DNA of only one cell is tested, then the results are extremely fuzzy.
Rajiv McCoy (Baltimore, MD)
@Healthy Nurse Tests like 23andMe sample a large number of cells, so they are essentially looking at the average genotype in the sample. Somatic mutations (those arising after fertilization and resulting in mosaicism) would have to achieve high frequency in the sample of cells to be detected in such a test. This could occur if the mutation arises very early in development or if the mutation confers a fitness advantage to the cells (e.g. cancer-causing mutations). However, in order to be detected, these mutations would have to occur in the less than 1% of the genome that is actually examined by the genetic tests, which focus on genomic locations that are already known to vary in the human population. So while mosaicism is common, it is mostly undetected and unlikely to have much impact on genetic tests like 23andMe.
Richard (Albany, New York)
Great article, thanks. It would be neat if there were a way to append a limited set of references for more in depth reading.
Sewgirl (NYC)
Fascinating. How will this affect drug testing in performance athletes? I think I recall a cyclist who was thought to be doping with EPO but then was a mosaic. They could not tell what cells were his and what were not. Interesting implications for many areas.
Paul Davis (Philadelphia, PA)
I don't know of any performance drug testing that is based on genetics. It would make no sense, since the whole point of PEDs is to alter performance in ways somewhat independent of genetics. The EPO tests in particular look for the effect of EPO on blood cells, not specific genetics of blood cells. Introducing bloods cells from another individual would likely show up as vastly more divergent than mosaicism.
lou (Georgia)
What does this say about having your genome checked for mutations for various health conditions/diseases, or in fact anything? If you take one sample, the result is different from another sample.
Danie (Martin)
Some conditions do indeed need more than one test to be sure.
Robby (Utah)
I'm a layman on the topic of DNA. Does this mean DNA evidence will become questionable?
Eric (Texas)
In the instances discussed there were a few genes affected. In my opinion there would still be a distinct signature for each person even if that signature varied slightly within that person. Fruit flies share 60% of the genes with humans but there is no mistaking fruit fly DNA from humans.
anae (NY)
@Robby - Questionable? Not in the way you're thinking. But the blood test from a guilty person might not be a genetic match for the guilty person's skin cells left at a crime scene. And check this out - the State treated this woman as a criminal and almost took her children away because their DNA samples didn't match. The mom was a mosaic - different types of cells in her body actually had different DNA - as different as two separate people might have. https://en.wikipedia.org/wiki/Lydia_Fairchild
Robin (Canada)
Great question! Hey NYT, on an otherwise quiet comment section, this question has raised some interest. Can you guys follow up? Thanks.
drhagus (Brooklyn, NY)
Another excellent article by Carl Zimmer. As a neuroscientist working in basic research with a particular interest in developmental disorders such as autism (ASD), the emerging appreciation of genetic mosaicism arising through post-zygotic mutation has been a revelation. In Rett syndrome (a rare syndromic ASD related disorders), mosaicism arising from random X-inactivation (females have two X chromosomes, with Rett patients carrying one mutated version that causes the disease) throughout the brain explains much of the phenotypic variation observed in the clinic. For non-syndromic ASD (the vast majority of ASD cases), the identification of causative genetic variation remains elusive despite the remarkable depth of recent whole genome sequencing efforts. Post-zygotic mutations that arise during early stages of brain development within neuronal lineages are effectively invisible to whole genome analyses that analyze genomic DNA from blood or skin, and thus may represent the 'dark matter' of ASD genetic etiology (cheers to Dr. Chris Walsh and colleagues for exploring this). For this scientist, it is humbling to think about the complex phenotypic variation that could arise from post-zygotic genetic mosaicism within the hundreds of potentially ASD-associated genes, but this seems likely to be the reality.
W in the Middle (NY State)
On autism, long wondered if it’s delayed-onset – or even absence - of successively more-abstract semantic “Helen Keller moments”…Perhaps: > As many adaptive approximately-matched filters as visual cognition, but self-organizable in temporal space rather than physical space – with many of the trouble spots at less-accessible hidden middle levels of cognition > Partly hardwired for different “Q’s” and coherence lengths and levels…Some things/folks process eidetically and some a bit more abstractly…Most familiar example – hearing the chromatic scale, but not in the pitch-perfect way that musicians can tell they’re in one key vs another…In listening to other scales, e.g. EDO-19 or 53, perceive a calliope’s wheezing rather than some slightly different ratio Not at all expert – but if something between “dog whispering” and full “talk therapy” that could correctly reboot the system…Might be helpful... At coherence length of paragraphs – conversation being more or less crowd-based synthesis from sole-based sentences, with borderline affirmative/ adversarial gaming…Visual re-enters as eye and body micro-movements – even before more macro-level “body language” Again, not expert – but for a while a useful mercenary vs one of the more insidious neurological diseases - one with a simplest genomic tell... At a structural level, the difference between right and left index fingerprints…Very similar – but one with double loop whorl... Answers maddeningly invisible in plain sight…