Epi-what?

A recent article in the New Yorker on epigenetics is drawing rage from scientists far and wide, because it apparently lacks a few key perspectives.  I decided to read the article and attempt to weigh in since this is an area I am particularly interested in and one that I follow pretty closely.  Before I comment though, I would like to acknowledge the excellent standing of the writer of the article, Siddhartha Mukaherjee, on account of the Pulitzer Prize he received for his wonderful book on cancer, The Emperor of All Maladies. This was a fine book and the recognition bestowed on the author seemed well placed to me, so what had gone wrong with the New Yorker article?

I read the article and for the most part enjoyed it.  Mukherjee weaves personal observations of his mother, a twin, with conversations on epigenetics with researchers  in the field.   These scientists have certain views, not all of which are accepted as mainstream.  One, for example, talks about an "epigenetic code"  which is not a broadly accepted concept. In the writing of the article, it seems Mukherjee down-played the role of transcription factors in the bigger epigenetics picture, and focused mainly on histone modification and methylation.  These terms are mumbo jumbo to most readers, but they are important to scientists. In the processes that modify the instructions of the genetic code there is a series of (sometimes unfortunate, sometimes not) events of which methylation and histone modification are rather downstream.  Transcription is nearer the top.  The fact that Mukherjee pretty much ignored transcription was what got him into hot water. Rather he waxed lyrical about the histones (because they are super interesting- little coils of DNA that expand and contract depending on how tightly they are bound), and about methylation.  He implied a stronger role for these elements than is currently accepted. Scientists from the field and a fair few outside of it called foul, and Mukherjee wrote rebuttals and explanations.  The article is an excerpt from a full book on genes that is yet to be published. He did not have time to go into all the details, he said, and quickly published an apology for omitting transcription factors and overextending speculations. However, reading his response to his critics it seems he wasn't so much sorry, as irritated.   And perhaps he had good reason.  As I read the complaints it seems like the critics all got together and decided on a response, then all said the same thing.  It seemed a bit witch-hunty to me but then I am not an epigenetics scientist.  I do think part of it had to do with what is extrapolation based on very early evidence, versus what the scientific establishment have accepted as fact. One critic even likened him to Deepak Chopra which seemed a bit excessive (I like Chopra for his edgy thinking, but obviously the Dawkins of the world do not share my affection).  I confess that I am way more comfortable with speculation than most scientists. I believe it is an essential part of the scientific process that gets us to eventual truth.

I would recommend you read the New Yorker article and decide for yourselves whether the author is scientifically accurate enough.  I felt he was, based on my limited understanding, but agree he over-emphasized some aspects over others. He also implied epigenetic changes can have lasting change which is a hotly debated area right now. My bigger interest though, is how easy it is for a writer to go from superstar to scourge of the earth; from Pulitzer Prize winner to scientific incompetent, in one fell swoop.  It happens very, very quickly, and I suspect Mukherjee will not recover easily from this experience. In the take downs there are also a fair few jabs at the New Yorker itself, for placing literacy juice over scientific fact.  Again, I have little beef with the New Yorker's style. I am a fan of the magazine and don't expect hard core science, but rather anecdotal tales with science in a supporting role.   Did Mukherjee take anecdote too far?  Read the article, and see for yourselves.  Let me know what you make of it.


References:

Same but Different

How epigenetics can blur the line between nature and nurture, by Suddharta Mukherjee

A couple of criticisms appear in these links:

https://whyevolutionistrue.wordpress.com/2016/05/05/the-new-yorker-screws-up-big-time-with-science-researchers-criticize-the-mukherjee-piece-on-epigenetics/ 

https://whyevolutionistrue.wordpress.com/2016/05/07/laffaire-mukherjee-the-last-word/

Moveable feasts: How cancer takes the party on the road

A few years ago I wrote a brief blog about the importance of metastasis to cancer's lethality.  About 90% of cancers deaths are due to metastasis. Metastasis is the process that leads to secondary cancers that arise in areas of the body that are remote to the primary tumor.  Cancer is at its most vicious once it's on the road and looking for new sites to settle, so research into the mechanisms of metastasis are crucial to finding potential new treatments.

In this week's Nature magazine there are a couple of very interesting articles that describe how complex this metastasis process appears to be.  One is about how breast cancer cells escape from the primary tumor in the first place, and the other is about the 'seeding' process that allows the roving cancer cells to set up shop in remote tissue in lethal metastatic prostate cancer. A third article by Hong et al, in Nature Communications also demonstrates the complexity of metastasis in prostate cancer using a similar analysis. What follows is my take-away from the articles. This particular Nature issue has other interesting cancer articles so I recommend looking at the whole thing (may require a subscription for the main papers though).

The view of cancer as just cells gone wild is changing to one of cells behaving badly in a somewhat organized way. The party contains cells that are just like the first cancer cells in the tumor (clonal cells), but also other cells that represent 'sub-clones',  derived from the primary cancer cells but with subtly different, and often competing, personalities.  The distribution of these personalities at the primary cancer site may well determine its fate. Some of these characters appear to be particularly good at evading chemotherapy. Others are good at creating the environment that allows them to escape from the primary tumor, and set out for new pastures through the bloodstream or the lymphatic system.  The Nature article by Wagenblast et al, shows that these Houdini cells (my term), are aided and abetted in their escape by two proteins called Serpine2 and Slpi that they express on their surface.  In breast cancer cells at least, these proteins cause the cells to acquire characteristics of endothelial cells (the cells that line the blood vessels) which means they can form connections between the tumor and blood vessels, thus providing the escape route. As such, they are potential target for new cancer treatments.  The proteins are also anticoagulants which may help keep the blood flowing and the escape hatch open. These so-called vascular mimics are very aggressive cells with what appears to be a focussed mission to get the team on the move. As Mary Hendrix points out in her review of the paper, their presence is a clear advantage for the tumor, but not so much for the patient- those people who show vascular mimicry in their cancers tend to have a poorer clinical outcome.  While the study used breast cancer cells in mice, the same proteins have been found on metastatic lung cells in humans.  Whether the current findings will apply to metastasis more broadly across other tumor types is not known, but it's a good hypothesis that deserves more attention in my view.

The second study of interest in this issue is by Gundem et al and this looks at the evolution of the remote metastatic sites in patients with lethal metastatic prostate cancer.  While prostate cancer is common, associated metastasis is much less common. Using whole-genome sequencing, the studies showed both clonal and subclonal cells to be present in the primary tumor. Hypothetically, the subclones may compete for dominance and in the presence of chemotherapy, those who have the resistance personalities may be able to prevail, changing the composition and fate of the overall tumor.  The studies also showed that at least two subclones were able to seed one metastatic site meaning that is is not only the primary tumor clones that seed the distant sites, but rather clusters of diverse cells. The diversity of these cluster may determine whether the seeding is successful or not. There are many circulating tumor cells but successful metastatic is relatively rare which suggests that the subclasses may cooperate during seeding process by leveraging distinct properties that at the moment are not understood. Michael Shen in his review of the current studies, suggested that disseminated single cells could settle in a remote site and remain dormant until cooperative metastatic cells arrive to help them take hold.  It was also interesting that some of the secondary tumor cells may have come from other secondary tumors as well as the primary tumor meaning that metastases could be reseeded several times from the both the primary and metastatic sites.  

A separate study by Hong et al, found similar results to Gundem et al, and also showed metastatic seeding occurs in temporal waves.  They also found that cells from the primary prostate tumor, can persist in the circulation in the long term, even after the primary tumor has been surgically removed.  This is wildly interesting to me and I am going to research this further for a future blog.  I've always felt understanding temporality is a major key to understanding disease.  This is all groundbreaking stuff and opens up new areas of research that could result in new treatments.   I am left wondering if the temporal waves, or the subclonal signatures could be influenced by epigenetic changes from environmental impacts?  I am off to explore!

 Nature. 16th April 2015. 

Smart parents raise smart kids-right?

Smart parents raise smart kids, but not for the reasons we all might think.  Making them do their homework, impressing on them the importance of reading, studying, paying attention in class, and generally being committed students, all give our kids the best chance at being smart-right?  Well, it turns out yes, and no.  While all the tricks we parents think we must teach our kids help them make the most of what they've got, what they end up having appears to be as much to do with what we gave them at conception, than whatever we've done for them since.
A large group of scientists running a project with the intriguingly mysterious name, Project ENIGMA, have discover that intelligence is probably affected by our genes to a greater degree than previously thought.  The researchers, from all across the globe pooled their resources and their data to look for correlations between DNA and specific brain disorders (the usual- dementia, depression, Parkinson's, schizophrenia, etc).  While they came up with some interesting data on the diseased brain, what they inadvertently uncovered was a connection between DNA and brain size, and even more curiously, between DNA and intelligence as measured by standardize IQ tests.  IQ tests look for a certain type of intelligence so the data has to be viewed with that in mind, but it appeared that a variant in a gene called HMGA2 may be able to explain differences in individual intelligence.  The difference in the HMGA2 was in one DNA letter only; at one point in the gene, the base thymine (usually represented as T), was replaced by a cytosine base (C).   People that had the letter C instead of T at a certain location on the gene were more likely to have a larger brain and score higher on the IQ tests.
This finding isn't really an answer to the question of inherited intelligence, but rather the starting point for many questions.  Correlation is not the same as causation and the size of the brain, the high IQ score and the genetic change, may all be linked by something else in the brain not yet found.  However, the association is interested and the methodology used by Project ENIGMA even more interesting.  Because it is difficult to obtain enough tissues in any one lab, to conduct thoroughly sound statistical analysis on genetic variations that only occur in a certain percent of the population, the researchers banded together to pool their brains, literally.  All the brain samples and data from each of the participating labs were used to look for associations between genes, brain size, disease risk and IQ test ability.  Without the collaboration, the project would not have been possible. The nature of the work and the extent of the collaboration has attracted hundreds of financial contributors that has ensured the success of the work.
So does this mean we should kick back as parents and let nature take its course?   Not at all.  Intelligence is only useful when harnessed appropriately, and usually (with notable exceptions), that happens through the cultivation of good study habits and the development of some level of self- discipline. When I was doing my PhD, someone told me that it would require one third pure hard work, one third organizational skills and one third smarts.  I never forgot that, and found it to be just about spot on not only for my PhD, but for pretty much every piece of work I've done since. Intelligence is only part of the picture.

[JOURNAL REFERENCE-Jason L Stein et al. Identification of common variants associated with human hippocampal and intracranial volumes. Nature Genetics, 2012; DOI: 10.1038/ng.2250]

The relevance of genomic information

I don't know about you but I am overwhelmed and energized by the recent explosion in interest in the relevance of gene knowledge to human difference. For several years we have been sequencing and profiling genes looking for aberrant genes and SNPs in the hope of identifying critical variants that explain disease risk, behavior, physical looks and so on. Only recently have we reached the point of understanding where can begin to realize the enormity of the task ahead in personalized meedicine. Recent papers in Nature have told us that the one-gene one-product hypothesis is only true for about 6% of our 20,000 or so genes. This is comforting given that our absolute gene number seems to equate with a mere nematode worm. The worm's genes can't seem to multi-task as well as ours do so it seems we are more evolved after all. Phew. This multi-tasking nature of most of our genes arises because a particular gene is separated along the DNA by areas of non-relevant DNA (to that gene at least) which means the gene can be read in a number of ways depending on the physical configuration of the DNA and possibly according to the influence of regulatory genes. This leads to 'alternative splicing' which results in different proteins being produced by a given gene. This finding is fascinating. No wonder SNP research has yielded so little in terms of identifying disease causing genes. The SNP is such a minoscule part of the altnertive splicing universe. Every issue of Science and Nature lately seems to have a plethora of articles and editorials on gene findings and their relevance to humans. I sense we are on the verge of a tipping point that will lead to novel hypotheses about the degree of determinisim we can reasonably expect from our genes. I'm excited about what we will come up with but hesitant because I know how long it can take to shift a paradigm. I hope we do not continue with more of the same and hope for some different conclusions. The alternative splicing effect should give us pause for thought; a reason to step back and question our assumptions about the role of genes and what we can expect from personalized medicine research in the future. What a fortunate time to be a biologist after all those years of sequencing and collecting the data--we can now begin to really look for the story our genes seem ready to tell.

Huntington's disease--a case of genetic determinism

After yesterday's blog I said that most genes are not deterministic and the some genes confer some risk of some disease either alone or in combination with other genes. The Huntington's Disease gene is different. Inherit this one and the disease is yours. For a truly gripping account of one man's decision to test or not test, check out the June 16th New York Magazine, Mind Bomb by Kevin Baker. Kevin's mother has the disease and he has watched her decline over the past 13 years. Is he witnessing his future? You have to read the article to find out but regardless of his fate, the very act of taking the test is fraught with caution, process and bravery. This test is not one that is undertaken lightly. As one nurse put it, once you know, you cannot not know.
Read the article. It's an enlightening tale of what all of our futures could be like as we learn more about what our own genes mean for our future health.

Dem old bones

Shortness is related to arthritis, the latest news tells us. Too tall, or too short, it seems we are at risk of osteoarthritis if we are anything other than average when it comes to height. The reasons are not clear but this news is quite exciting to researchers as there are so few genes for osteoarthritis, whereas there are a great many for height. Careful digging into the genetic and environmental causes of both may shed light on the increased risk. If short people get arthritis, what does this mean from an evolutionary perspective? Often, a genetic disadvantage in this day and age, may have been a significant advantage in years gone by.

I'm short, so I should perhaps be concerned. Or maybe not. I could have a gene test; a gene for Growth Differentiation Factor 5, or GDF5, has been associated with the development of cartilage, and also with arthritis susceptibility in Europeans and Asians. If I have the risk variant, what would I do about it? Nothing, because there is really nothing that can be done. I have no family history, although my son does appear to have somewhat mobile joints. Perhaps I passed a dicky gene on? To understand my genetic risk is becoming easier for bones and lots of other pending problems. Gene testing is relatively cheap and beginning to offer significant information, albeit un-validated for the most part. The HapMap, a complete map of human genetic variation, is about to be released in the journal Nature Genetics (online, Jan 13th 08). For sure, as these types of sweeping tests become more popular, doctors, diagnostics companies and drug companies will have to start coming up with ethical responses to the questions that will arise when folks begin to see they have genes rendering them susceptible to dementia, cancer and the like. How will we deal with the knowledge of the risk, with no preventatives to soothe us while we wait for the inevitable? Dangerous and heady times are ahead. But what an amazing time to be alive, and what a great opportunity for those of a business mind-set.