I thought this article in the current New Yorker might generate some interesting discussion here.
http://www.newyorker.com/magazine/2016/05/02/breakthroughs-in-epigenetics
161 thoughts on “Twins and Epigenetics”
Leave a Reply
You must be logged in to post a comment.
BruceS,
Thanks for you contributions and comments. I’m at the NIH almost all day today, but maybe tomorrow and the days after, I’ll try to give point by point responses. If there is something you feel I’m not communicating well, speak up and I’ll try to clarify. The stuff isn’t as hard as it seems, there is just a lot of jargon. Thankfully, because of the way TSZ is set up, we can show pictures and diagrams. Here are the sort of diagrams presented in my classes, this one is the most central to the discussion.
It shows the histones being wrapped around by DNA. This is right from the NIH RoadmapEpigenomics project. It is central to the discussion at hand. Does it make sense to you?
BruceS,
Veeeery eeenteresting! In particular, it shows that I have been taking rather too much on trust. I had assumed, when people claimed epigenetic inheritance, that it might conceivably occur cell-to-cell within an organism (I was mainly dubious about multi-generational persistence). I thought there were some grounds for supposing it happened (outside of conservative replication of methylated cytosine, which does happen). Likewise with histone control of expression. But, according to Ptashne and Greally, even those claims lack experimental support.
“there is no evidence that coiling and uncoiling of DNA has a causal effect on gene activity.”
“There is no evidence, despite years of research, that nucleosome states can be “copied” for transmission to daughter cells. ”
“It is true that enzymes that modify histones have been found—lots of them. A striking problem is that, after all this time, it is not at all clear what the vast majority of these modifications do.”
Yet, according to Sal, it is to implement RAM.
“By ‘epigenetic code’ the author seems to mean specific arrays of nucleosome modifications, imparted over time and cell divisions, marking genes for expression. This idea has been tested in many experiments and has been found not to hold.”
It all rather confirms my prejudices.
I asked a while back where in the linked PDF there is any mention of epigenetic changes being transmitted to future generations, as in a permanent modification to the lineage.
And perhaps you’d clarify how this is conceptually different from evo/devo. Or conceptually different from stuff that been the subject of research for the last 50 years.
Just to plug one rabbit-hole we’ve already been down with Sal: when I talk of epigenetic inheritance, I mean inheritance of the epigenetic state, not genetic inheritance of the capacity to establish a particular epigenetic state.
Here’s the money quote from the New Yorker, boldface added:
I’m curious what Sal makes of this, and I thank him for inviting questions.
Let me clarify. If Sal is simply reciting some new discoveries in the details of how development works, what is his point?
It’s really complicated; therefore YEC?
It’s really complicated; therefore no junk DNA?
I don’t see his point.Or if I see his point, I don’t see the evidence or the connections.
As far as I can see, the appeal comes from the fact that genetics has grubby associations, due to neo-Darwinism, Dawkins and Determinism. And other stuff that doesn’t begin with D. All of these associations are misplaced, but basically any paradigm will do, as long as it’s not the current one.
Okay, but what is Sal’s new paradigm? Does he have any evidence that new discoveries in epigenetics are actually invalidating 40 0r 50 years work on the subject? Or is knowledge just being extended?
Sal seems to relate every new discovery to evolution is dead, or Larry Moran doesn’t know what he’s talking about.
Since Sal has called for questions, my question is how Moran and Coyne are wrong.
Specifically, with links.
Ptashne in comments at whyevolutionistrue:
Ptashne’s response was masterful.
In a couple of places, my initial reaction was “Ooh, I don’t think that is universally true”.
For instance, I found “And there is no evidence that coiling and uncoiling of DNA has a causal effect on gene activity.” somewhat surprising. So I went searching for a counter-example in my old stomping grounds.
In Zhang et al 2014 I find this gem:
which Sal is welcome to quote
xxxx, so long as he includes the next sentence:which result Mark was fully aware of… he being the editor for the article in question, ‘n’all.
It really is the case that transcription factors RULE, Sal’s efforts to find random quotations that suggest otherwise to him notwithstanding.
If epigenetic modifications are inherited, they then form another inheritance system that can respond to natural or artificial selection. There is modest evidence that there is such inheritance. But it rapidly reverts to its original state. Thus to make a long-term change one needs either to keep fighting this reversion by continued strong natural selection, or to stabilize the change by changes in the DNA.
What there is not, is any evidence that inherited epigenetic changes are preferentially “mutating” in an adaptive direction. If people experience a famine and have epigenetic changes that are inherited, they are not changes that preferentially help the organism cope with famine. Instead they are just changes that increase rates of cancer and heart disease.
This of course will not stop all the woo peddlers who are just sure that you inherit, epigenetically, the mystic wisdom of your ancestors. Or who peddle some cream that just must make both you and your offspring thinner. Nor will it stop the neo-Lysenkoists who are just sure that DNA is not the mechanism of inheritance. Or the creationists who are sure that Mendelian changes are not the basis of evolutionary change and that epigenetics shows that many decades of the Modern Synthesis are all wrong.
Joe Felsenstein,
Even there, the distinction needs to be made between inheritance passing along a cell line from grandmother to grandchild, and a mark that happened to be set in an egg in a foetus in the stress-exposed grandmother.
DNA_Jock,
Transcription factors … aren’t they like little logic gates … ? 😉
A repeat post:
This is the part that Sal obtusely still cannot wrap his head around:
Is there sometimes (not always) an aspect of inter-generational memory (metaphorically speaking) to epigenetics?
Without belaboring the finer details – short answer = yes.
Next question:
Would nucleosome modification (including Histone acetylation) play some role in the expression of this so-called “memory” effect?
Again, the short answer is “yes”.
But herein lies the rub – nucleosome modification is a result and NOT a cause of epigenetics
… a point that Sal has repeatedly been attempting to deny in a confused and often incoherent series of contradictory posts
BUT (drum roll please) happens to be the very point made in the citation that Sal has just tossed our way!
…for further elucidation of Sal’s lack of lucidity please refer to
BASTA!
Joe Felsenstein,
Hi Joe
I asked you a question about epigenetics & evolution on an earlier occasion
Is it not possible that the very possession of some version of multigenerational epigenetic inheritance (in and of itself) as some ‘toggle-switch’ alternating between default settings (fasting mode vs famine mode for example) could be construed as adaptive because children typically experience the same conditions as their parents but at the same time allows quick resets along Ptashne’s version of Lambda autogenous regulation of lysogeny because ancestral responses would be detrimental if the environments of the progeny and the ancestors were different?
Whew so sorry for the long run-on question, but I hope you get my drift.
BruceS,
Ok, as I promised I’m trying to being responding point by point by some of the quotations on Coyne’s blog. I have to provide a little more preliminary info before addressing things point by point. I will refer to these preliminaries in my reponses.
The diagram from the Roadmap Epigenomics project shows the DNA uncoiling from a chromosome and revealing the nucleosome/histone complexes. The NIH Roadmap Epigenomic project can be viewed somewhat as ENCODE 2.0 as it performs and extends many of the experiments and measurements that ENCODE did but on a different set of cells in various tissue types. These projects gathered data on the histone states of histones in various locations in the genome and in various cells. ENCODE sampled 150 cell lines and RoadmapEpigenomics 111 — but that his only the tip of the iceberg since an adult human has about 100 trillions cells!
If most of the genome is junk as some evolutionary biologists claim, then it might stand to reason what is going on in the histones attached to this “junk” is also junk information. This probably was one of the reasons a critic of the half-billion dollar ENCODE/RoadmapEpigenomics project said this:
So that gives a backdrop of some of the prejudices in some scientific quarters against the study and interpretation of the epigenome.
That said, here is a diagram of a single nucleosome and the histone tails. So if you look at the above diagram from RoadmapEpigenomics, focus on the part that says “histone modifications”, you can see a zoomed in version of the amino-acid positions on the histone tails that are actually modified. This shows that even individual amino acid positions can record memory marks in the form of chemical enhancements (like Acetylation, Methylation, Phosphorylation, Ubiquitylation, Sumoylation, etc.)
The listing of possible histone modifications is growing and it seems to have meaning as attested to in this table:
http://www.cellsignal.com/contents/resources-reference-tables/histone-modification-table/science-tables-histone
The above information is important to understanding my objections to the stuff said in Coyne’s blog. I will refer to this data again.
OK, now I’ll contest some of the more prominent criticism on Coyne’s blog. Starting with the Nobel Laureate Sidney Altman:
The epigenetic code is a hypothesis that is based on the observation that the patterns of histone modifications are different between cell types. It suggests the histones provide meaning for the identity and functioning of a cell, hence it is fair to say histones along with other epigenetic marks form some sort of code or instructions or control mechanism for the cell. For Sidney Alman to say “I’m not aware that there is such a thing as an epigenetic code” is a bit ironic given it seems it is common knowledge as a working hypothesis as attested to by Wikipedia:
Altman complains the New Yorker should be peer reviewed? Is he saying the term “epigenetic code” somehow won’t pass peer review? From the prestigious scientific journal Nature
http://www.nature.com/ncb/journal/v9/n1/full/ncb0107-2.html
So what’s Altman’s excuse for saying he’s not aware of such a code. Granted, he may complain it’s not fully elucideated or the hypothesis completely worked out, but it exists at least as a belief and possible reality in the mind of some scientists, certainly enough to appear in a peer-reviewed scientific journal.
Like I said, something about this barrage of criticism doesn’t seem completely wholesome — just my opinion.
“or I the Lord your God am a jealous God, visiting the iniquity of the fathers on the children to the third and the fourth generation of those who hate me”
stcordova,
Something about the all-over-the-internet phenomenon of Creationists ‘bigging-up’ ENCODE and epigenetics against all expert criticisms of over-interpretation seems decidedly unsavoury. It’s like there’s a seminary – I dunno, a conspiracy or something – where the Faithful learn the latest fad, then go forth and multiply.
Why do Creationists suddenly give a shit about genome mapping and the possibility of non-genetic inheritance? Why are the critics – every one of ’em – wrong?
[Insert pretty picture here].
stcordova,
He’s unaware that it’s a real thing, not unaware that it’s a concept.
Exactly right, Allan. Sal tries a little map/territory legerdemain there.
ENCODE is expert science, despite Gruar calling ENCODers ignoramuses.
http://www.sciencemag.org/news/2015/02/massive-project-maps-dna-tags-define-each-cells-identity
ENCODERs have the attention of medical researchers for good reason. At the ENCODE 2015 conference the MD’s there didn’t even mention Graur, Coyne, or Moran. That’s about the level of regard that their “expert” opinion commands.
The Encode project involved 442 researchers, based at 32 institutes around the world, and required 300 years of computer time and five years in the lab to get their results. — Forbes
Researchers at MIT, Yale, Harvard, Stanford, …and even Larry Moran’s University. Ewan Birney, a chief architect of ENCODE was called “the scientific equivalent of Saddam Hussein” by Dan Graur. Birney got elected to fellowship in the Royal Society. Graur? Reduced to name calling attacks.
What is represented at Coyne’s blog is the reaction of Old School scientists whose ideas are getting dissed by the new generation with new viewpoints.
What does the entire sentence say, and who said it?
The original formatting was messed up by me. I was he one who pointed out where the researchers are from.
First off, one will see Stanford University at the bottom of the front page of the Official ENCODE site.
I was pointing out, ENCODE is regarded as providing relevant expert views by the medical and pharmaceutical industry, not Jerry Coyne and friends.
Here are the news release mentioning the experiments that ENCODE conducted:
https://www.encodeproject.org/news/
Those are among thousands of experiments being run that generate data for use by medical researchers. Graur calls this data “excrement” by “ignoramuses”.
Compare the quantity and quality of ENCODE and Roadmap Experiments to the experiments that Jerry Coyne conducts in his lab or Gaurs lab. Gruar and Coyne’s labs are dwarfed be ENCODE/RoadmapEpigenomics.
Creationists don’t have to “big up” ENCODE/RoadmapEpigenomics, it commands a half billion research budget, how much of a research budget does Coyne’s fruitfly evolution lab command in comparison. Which experiments command more attention in the medical and pharmaceutical industry?
Here is an ENCODE researcher at University of Washington (where Joe Felsenstein teaches) by the name of Stamatoyannopoulos:
http://www.stamlab.org/
Graur refers to Stamatoyannopoulos:
Ok, back to a selected point by point:
Henikoff:
‘
Now let’s see what the author of the New Yorker article said:
Let me show why Henikoff just embarrassed himself.
The New Yorker article was simply restating information available in peer-reviewed publications like this one.
http://cshperspectives.cshlp.org/content/6/2/a018606.full
The author of the above paper is a co-author with Nobel Laureate Yamanaka! Does that paper sound more like it agrees with Henikoff or the author of the New Yorker article. I think the author of the New Yorker article wins that exchange and Henikoff looks like someone just trying to trash talk.
What do you suppose he thinks of Intelligent Design Creationists?
Sal,
Is Graur right or wrong about Intelligent Design not being responsible for biology?
Compare what was actually written in the New Yorker article:
with this discussion at Wiki about Yamanka factors (in agreement with Takhashi’s article above)
Contrast with Henikoff’s attempt at a hack job:
Huh? Transcription factors according to Henikoff have nothing to do with epigenetic modifications? He looks the fool for saying stuff like that. That’s what he gets for trying to trash talk Allis’ work.
The paper you cite demonstrates unequivocally that Henikoff was correct in his criticism, which was (in case you didn’t understand it) that Mukherjee cited Yamanaka factors, which are transcription factors, in a misleading manner. The fact that you think otherwise is evidence supporting Henikoff’s thesis.
DNA_jock,
Henikoff misrepresented what Mukherjee said. Transcription factors like Yamanaka factors can erase epigenetic marks.
What is specifically wrong about this?
Henikoff’s misrepresentation:
BULL!
What did Mukherjee say that suggests he did “not realize they were transcription factors”?
Henikoff is misrepresenting and knocking down statements Mukherjee didn’t make.
The target of the Yamanaka experiment was erasing the epigenetic marks. The Yamanka factors were the instrument to artificially induce erasure of the marks. Mukherjee accurately used the notion of the Yamanaka factors (“identity switching factors”) that erase epigenetic marks.
Here is a howler from Henikoff:
Nothing in Mukherjee’s article suggest he thinks transcription factors are etching of marks on histones or DNA, but henikoff is dead wrong to suggest transcription factors like Yamanaka factors aren’t responsible for histone modifications. Howler! From Wikipedia, this is common knowledge that transcription factors can affect histones:
Jock, if we put in “transcription” for “identity switching” in this NYer quote:
can you explain the error or misleading aspect in the remark (I mean like for a total layperson)? Thanks.
Regarding the “ignoramuses” (to quote Dan Gruar) who pump out “excrement” in regards to the genome and epigenome, here is a listing of the ENCODE/RoadmapEpignomic “crooks” (to quote Gruar again) who are no doubt among Allis’ fan club.
This was taken from a paper describing experiments to discover more of the kinds of histone modifications Allis pioneered. Judge for yourself if these are “crooks”, “ignoramuses”, pumping out “excrement” for science journals as Dan Graur says, or whether Allis has pioneered a field that is valuable to medical research.
http://genome.cshlp.org/content/22/9/1813.full
Yamanaka factors are subclass of transcription factors. Here is a depiction of a transcription factor attaching (binding) to a gene in the DNA.
Yamanaka factors for whatever reason are able to induce erasure of epigenetic marks. Nothing in Mukhurjee’s article deviates from what is accepted in the mainstream as attested to by the Wikipedia article cited above and the more scholarly article written by Yamanaka’s co-author Takahashi.
Takahashi’s article is the same one in Allis’ textbook on epigenetics which is an accepted textbook in classes taught at the NIH and Johns Hopkins University.
I saw little cause to in that paragraph by Mukherjee that should draw the sort of criticism-based-on-misrepresentation-and-innuendo by Henikoff. Henikoff looks to me like he just made something up in saying Mukherjee doesn’t know the difference between a transcription factor and a histone. It’s a false insinuation, imho, intended to discredit Mukherjee.
Like Dan Graur’s criticism of the work of the ENCODE/RoadmapEpigenomics consortiums, something about this flap over Mukherjee’s article has an unwholesome air of vindictiveness and jockeying for prestige and money.
Way on the sideline periphery are the creationists and evolutionists picking their allies (creationists being friendly to the NIH ENCODE/Epigenomics projects and it’s Jesus-loving NIH Director Francis Collins, and evolutionists like Graur hating them).
Sal, are you ever going to get to the point?
You invited questions, and I asked some, but you haven’t responded.
He answered mine.
I don’t know if he’s right, but I haven’t heard from anybody else on the matter.
Exactly. And if you understood Henikoff’s criticism (which you evidently do not…), the point is that Mukherje wrote a 6,000 word article about epigenetics without ever once mentioning transcription factors. Mukherjee gives his lay audience the impression that “Yamanaka factors” do all sorts of different stuff – “Most important, epigenetic marks were erased and rewritten, resetting the landscape of active and inactive genes.”
This is the tail wagging the dog.
The reality:
These proteins all bind to DNA in a sequence-specific manner and regulate the initiation of transcription. As a consequence of this transcription, some histones may get modified. (c-myc is unusual here in that, in addition to its traditional TF role, it also interacts directly with HATs).
Mukherjee’s highly misleading implication is that the “most important” thing that Yamanaka factors do is to modify histones.
Instead of whining about how mean you think a few soundbites were, why don’t you address the meat of Ptashne and Greally’s critique? Or Zhang et al, that I cited above?
walto,
I hope the intervening comments have answered your question to me.
If not, then I offer the following edit to the offending passage:
“After a decades-long hunt for identity-switching factors, Yamanaka found a cocktail of DNA-binding proteins that would do the trick. ” …followed by couple of thousand words about the primacy of transcriptional activators over histone modifications…
The main question that I would like to ask Sal is — aside from providing additional detail about how development works — what is conceptually new about recent work?
You haven’t shown, or even discussed, how any of this impinges on the percentage of junk DNA, nor how any of it affects evolution, nor how any of it affects DNA as the carrier of inheritance.
Sal, how happy do you think Francis Collins would be to know that some creationists support him?
Or, how happy would they be to know that Sal is telling everyone that they support his interpretation of their work.
How about it,Sal? Invite your profs to read your posts.
Yes, that’s helpful, thanks.
You know, when I post anything at all about any aspect of evolutionary theory here, I feel a little like some 90-year-old Brit in 1909 might have felt if, based on a newspaper article he stumbled upon in his local lending library wrote a letter to the St. Petersburg Times with the heading, ‘Is the Concept of Tsarism Really So Bad?’
It’s all so political!
Double post. Sorry.
stcordova,
Sure it is, but why are you (and every fkin Creationist on the Net) taking their side when they stray outside their area of expertise?
I dare say. Doesn’t answer the question though. They can obviously stand on their own 884 feet and answer critics; why are Creationists weighing in? Numbers don’t make people right on everything, which is really the issue, not whether they are right on anything.
Perhaps we could get a straight answer to the following question: Creationists side with ENCODE on everything they say because …
Who cares? Maybe they have a crush on somebody. Why push guilt by association in response to whatever the hell reflected glory they’re evidently depending on? I’m sure there’s better responses to this stuff than raspberries.
Experts still have to be right.
Speaking of being right about evolution, it looks as if endosymbiosis may not be a unique event.
http://www.npr.org/sections/health-shots/2016/05/12/477691018/look-ma-no-mitochondria
Could be that borrowing functional bits, including mitochondrial functions, can happen on more than one level.
walto,
Me.
There are, and I have laboriously typed out many such on this very site, mostly to deaf ears. So I’m curious why those ears are so deaf to the counter-argument on this particular issue.