Debunking myths on genetics and DNA

Monday, December 10, 2012

Whole genome harvesting

You think the human genome, with its three billion base pairs and 23 chromosome pairs, is too complex to unravel? Turns out, the wheat genome is six times as big and it's hexaploid, in other words, instead of chromosome pairs it's organized in chromosome sextets!

I've recently discussed genetically modified organisms, crops in particular, and while I still can't provide a definite answer on whether they are absolutely good or absolutely bad, one thing struck me as relevant as I was researching the topic: between climate changes and an exponentially growing population, we are making drastic changes to our planet and resources. While Mother Nature is usually able to buffer changes and constantly adapt to new environments, the changes human kind is bringing upon the planet are happening at such a fast rate that natural adaptation is unable to keep up.

I think at some point we will have to face a hard choice: either starve or give in to GMOs, where by GMOs I mean crops that are bioengineered to yield more in harsher conditions. Again, I'm not saying we should all embrace GMOs as they are healthy and good for us. I really don't know. What I'm saying is that we may not have a choice: in 2009 the FAO estimated that in order to meet the ever-growing demand, wheat production has to increase by 60% by 2050. In the 20th century, the Green Revolution met the increase in demand with the technology known at the time. Today, given the FAO estimate, we may face the need of a second Green Revolution.

With this in mind, you understand the importance of sequencing the wheat genome, a task that is complicated by the complexity of the genome itself. Its three sets of chromosome pairs originated first from the hybridization of two diploid wild grasses, which originated tetraploid wheats (two sets of chromosome pairs) like durum wheat. After thousand years of domestications, these underwent a further hybridization, yielding the hexaploid wheats commonly used today to make bread. Domestication led to a bottleneck in genome variety, nonetheless, the wheat genome has a high percentage of repeats (roughly 80%, mostly retroelements) that yield great variation in length and gene order, making it difficult to sequence.

Despite these obstacles, two papers [1,2] in the latest issue of Nature report using both whole-genome 454 sequencing and shotgun sequencing to assemble the genome of bread wheat and barley. Both sequencing methods have the shortcoming of being applicable to very short regions, and therefore additional work is required to reassemble the full genome out of the various short sequences.

Interestingly, the wheat genome appears to implement a lot of the variation mechanisms I've been extensively discussing here on the blog:
"Several classes of plant DNA transposons and retroelements create and amplify gene fragments, disrupt genes and create pseudogenes, which can influence gene expression through epigenetic mechanisms [1]."
Similarly, in barley:
"Abundant alternative splicing, premature termination codons and novel transcriptionally active regions suggest that post-transcriptional processing forms an important regulatory layer. Survey sequences from diverse accessions reveal a landscape of extensive single-nucleotide variation [2]."
Brenchley et al. [1] conclude:
"Major efforts are underway to improve wheat productivity by increasing genetic diversity in breeding materials and through genetic analysis of traits43. The genomic resources that we have developed promise to accelerate progress by facilitating the identification of useful variation in genes of wheat landraces and progenitor species, and by providing genomic landmarks to guide progeny selection. Analysis of complex polygenic traits such as yield and nutrient use efficiency will also be accelerated, contributing to sustainable increases in wheat crop production [1]."

[1] Brenchley, R., Spannagl, M., Pfeifer, M., Barker, G., D’Amore, R., Allen, A., McKenzie, N., Kramer, M., Kerhornou, A., Bolser, D., Kay, S., Waite, D., Trick, M., Bancroft, I., Gu, Y., Huo, N., Luo, M., Sehgal, S., et al. (2012). Analysis of the bread wheat genome using whole-genome shotgun sequencing Nature, 491 (7426), 705-710 DOI: 10.1038/nature11650

[2] Mayer, K., Waugh, R., Langridge, P., Close, T., Wise, R., Graner, A., Matsumoto, T., Sato, K., Schulman, A., et al. (2012). A physical, genetic and functional sequence assembly of the barley genome Nature DOI: 10.1038/nature11543

Friday, December 7, 2012

The simulated brain

His name is Spaun, which stands for Semantic Pointer Architecture Unified Network, and he's a brain -- a simulated, brain. His 2.5 million neurons, organized in subsystems that simulate different brain areas, allow Spaun to perform tasks such as image recognition and recalling sequences, and respond through a motor arm. For example, Spaun can recognize numbers on a screen and write them on a piece of paper.

Spaun is the brain child (pun intended!) of authors Eliasmith et al. [1]. It models three specific brain areas: the prefrontal cortex for memory, the basal ganglia to select actions, and the thalamus. Spaun's functional architecture consists of a working memory that, given a visual input, compresses the information and translates the input into firing patterns. The next step is the action selection step, which results in a motor output through the robotic arm. Spaun's memory doesn't just store information, but it also correlates new information with the old one. A nice feature of the model is that different neuron parameters can be chosen from random distributions in order to simulate different population behaviors. This simulates the human brain so well that Spaun expresses a common human behavior: the tendency to remember best the first and last items in a list.

On the other hand, Spaun exhibits noteworthy deviations from human brains: while it can get better and better at a particular task, it cannot learn a completely new task. Another shortcoming is that Spaun's attention and eye position are fixed, so that, contrary to a real human brain, it cannot control the input.

As the authors explain:
"Anatomically, many areas of the brain are missing from the model. Those that are included have too few neurons and perform only a subset of functions found in their respective areas. Physiologically, the variability of spiking in the model is not always reflective of the variability observed in real brains. However, we believe that, as available computa- tional power increases, many of these limitations can be overcome via the same methods as those used to construct Spaun."

[1] Eliasmith, C., Stewart, T., Choo, X., Bekolay, T., DeWolf, T., Tang, Y., & Rasmussen, D. (2012). A Large-Scale Model of the Functioning Brain Science, 338 (6111), 1202-1205 DOI: 10.1126/science.1225266

Tuesday, December 4, 2012

Make a donation to NOAH, get a free 8x12 print

As many of you know, my research is on HIV, with a focus on HIV vaccine design. I work in Bette Korber's group, and through Bette, I came to learn about NOAH, an organization she co-founded.

Taking care of AIDS orphans has been one of the most prominent issues in Sub-Saharan Africa, where two-thirds of the people affected by HIV/AIDS live. Mother-to-infant transmissions are highly preventable yet, sadly, the drugs are expensive and not always available in Africa. A staggering 2.5 million African children have been orphaned by AIDS, and many of them are born HIV-positive. There are villages where a whole generation has disappeared because of AIDS.

NOAH takes care of these children without taking them away from their homes. NOAH is not an orphanage. The organization provides schooling, day care and food for the kids, while the kids continue to live in their village with older relatives. $80 covers one child for one year.

For the whole month of December, if you make a donation of $30 or more, I will send you an 8x12 print of one of my pictures. Follow this link to donate, forward the email receipt to eegiorgi (at), include your shipping address, and pick the picture of your choice from my G+ album (click on "Photo details" to see the file name).

Donations are tax-deductible.


Monday, November 19, 2012

Proteins as gene carriers

By now you probably know everything about pluripotent stem cells, right? They are the hot topic in genetics right now, to the point that the fear of being scooped has pushed some people to lie about their results. Pluripotent stem cells are cells that have the ability to divide into a specialized cell and another undifferentiated cell. This of course is greatly useful in repairing damaged organs and/or regenerating tissue, and has great potential in medicine.

Lately there has been a lot of buzz on the notion that pluripotency could be re-induced in already differentiated cells. Studies have shown that four reprogramming factors can indeed reprogram fibroblast cells into pluripotent stem cells when over-expressed.

But how to over-express these factors?

The typical route is to transfect the genes into the cells by means of a viral vector. Basically, the genes are delivered into the cell using a retrovirus. Though effective, this poses the question of side effects: whenever you introduce foreign DNA into a cell you have the potential to silence secondary genes or disrupt the usual gene regulation. Unanticipated epigenetic changes in the cell can occur. A recent study [1] shows a safer alternative: cell-permeant proteins, or CPPs. These are small proteins that can cross the cell membrane and carry peptides inside the cell in a process called "protein transduction," thus offering a valid alternative to viral vectors.

By comparing the two methods (CPPs and viral delivery) on human fibroblast cells, Lee et al. noticed that gene expression was achieved much faster when using the viral vector. Puzzled by this difference, they wondered what was so special about the viral route that made the gene delivery so much more efficient. There had to be something in the viral vector that aided the delivery of the genes. Lee et al. hypothesized that this could be linked to the fact that the viral vector somehow activated an inflammatory pathway in the cells which in turn aided the delivery of the genes. So the next question was: can we enrich the CPPs so they too activate the inflammatory pathway?

Indeed they could! They used TLR3 agonists, molecules that activate the TLR3, or Toll-like receptor 3, a receptor that recognizes double-stranded RNA generated by retroviruses and thus activates inflammatory pathways. Once combined with the TLR3 antagonists, over-expression of the reprogramming factors was achieved faster through CPPs than it was with the viral vectors, validating the hypothesis that the gene delivery has to be achieved via the activation of the immune pathway. In fact, the contrary was also true: when TLR3 was knocked down (biology jargon to say that the gene was silenced), the viral vector was also inefficient in delivering the genes.
"TLR3 activation enables epigenetic alterations, including changes in methylation status of the Oct4 and Sox2 promoters as well as changes in the expression of epigenetic effectors, that promote an open chromatin configuration. The knowledge that the activation of innate immune response affects nuclear reprogramming permitted us to enhance the efficiency and yield of human induce pluripotent stem cells by using reprogramming factors in the form of CPPs."
Lee et al. conclude:
"Our observations highlight a previously unrecognized role for innate immunity activation in nuclear reprogramming. The viral vectors constructs used to induce pluripotency are more than mere vehicles for the reprogramming factors. Innate immune activation causes striking changes in epigenetic modifiers that favor an open chromatin configuration. These changes enable a fluidity of cell phenotype that contributes to successful nuclear reprogramming."

[1] Lee, J., Sayed, N., Hunter, A., Au, K., Wong, W., Mocarski, E., Pera, R., Yakubov, E., & Cooke, J. (2012). Activation of Innate Immunity Is Required for Efficient Nuclear Reprogramming Cell, 151 (3), 547-558 DOI: 10.1016/j.cell.2012.09.034

Thursday, November 15, 2012

Is creativity an illness? But then... what is an illness?

Are you creative? Do you ever feel that when your creativity strikes you become absolutely compulsive about your "inspiration," and totally depressed when, for some reason, your inspiration wanes? It always strikes me to read about how some of the most beautiful works of art were created: their creators were obsessed, compulsive, borderline dysfunctional. Gabriel Garcia Marquez sold his car and had his family live on credit for eighteen months so he could write One hundred years of solitude. Brunelleschi's obsession was the dome of Santa Maria del Fiore, Antoni Gaudi's obsession was La Sagrada Familia. It seems to me that obsessions may ruin your life (or most likely the life of your closest ones) when you have them, but they may also lead to the most wonderful things.

So, is creativity a good thing or is it an illness?

My friend and collaborator Tanmoy Bhattacharya brought to my attention an interesting BBC post that discussed the issue. The article came up in a Facebook discussion because it raised the question: "How do you define illness? When, exactly, does a behavior trespass the normality threshold and becomes an illness?" I really liked Tanmoy's take on the issue, and I asked him permission to repost it here on the blog. It's the best thing I could get since he won't do a guest blog for me. :-)

I think he raises excellent points on the complexity of the brain, its stimuli as well as its constraints. I enjoyed reading it, I hope you will too. And if after reading this you have questions for Tanmoy, go ahead and post them in the comments and I will forward them to him.

TB: In a system as complex as the brain, which interacts with such diverse environments, it is difficult to define health and disease. There has been a long standing hypothesis that certain brain functions like deductive logic and creativity are kept in check evolutionarily because the same "structure" that can give rise to very highly creative adaptations in one environment would give rise to maladaptive behavior in a different environment. The interest in the research is, therefore, understanding the architectural limits on the brain, not to stigmatize writers or expect every bipolar to pen out a story about an old man and the sea.

EEG: That's a very interesting theory. All greatest masterpieces required such great energy and dedication from their creators that these individuals had to, at some level, become unsociable, as focused as they were on their creation. I can see how, at a species level, "being socially fit" puts a constraint on the amount of time and "obsession" the brain can dedicate to a certain task.

TB: I do not believe that we yet have a definition of illness which is "biologically" meaningful. Sure, there is a diagnostic manual that tells a doctor today when to diagnose a particular mental illness, but it is more an expression of "social" reality than a "biological" reality. So, for example, the discussion of whether homosexuality is a disease is not argued on any grounds about what it does or does not do to the person, but rather whether the majority of doctors consider it within the "normal" spectrum of behavior. No wonder its classification changed from a disease to a non-disease as the social acceptability of homosexuality grew: not because such acceptance lessened the mental load on the person with the trait (it is now not considered a disease even when the person with the trait lives in a non-accepting community), but because it became "socially" acceptable as a "normal" behavior. Currently, there is a similar debate about whether bereavement distress should be considered normal even when it leads to behavior sufficiently aberrant to otherwise merit a diagnosis of clinical depression. In other words, the question is not whether the person is depressed after a loss: the question is whether it is a disease (possibly temporary like say getting the 'flu is a disease) or whether it is not a disease because it is "normal". The classification is not done based on any kind of biological reality, except whether it is considered normal; which is determined by methods of social science, not biology.

Does this concept of normality depend on a biological reality? In other words, is there a way, other than surveying doctors (the social science method), to figure out whether some one is abnormal? Remember that we know pretty much that all of us are different in many ways, if you defined me abnormal simply because I am unique (which I certainly am), then everyone would be abnormal. One could always say that one should not look at the totality (which made everyone unique), but trait by trait, and ask whether I have traits that very few other people have? Defining abnormality this way would, of course, make Picasso abnormal; but during a mass hysteria, it would classify everyone as normal. We again see that this definition fails to capture the abnormality that is relevant to defining disease.

I claim that the only way people have found to capture the relevant abnormality is by taking the design stance: human brains (and bodies) are supposed to be "for" something. When the organ (or the totality) is carrying out this function, it is normal; when it fails to carry out this function, it is abnormal. Note that this does not solve the underlying problem: someone still has to define the function, but that turns out to be an easier problem.

We could define a disease objectively as a malfunction if we could define function objectively. And, here, biology can bring an insight: the function of brains (and bodies) is to survive and use the environment, physical, biological, and social, to further the fundamental goals of long term survival of the traits. This is usually called reproduction, but it is far more subtle: for example, one can help raise grandchildren and contribute to the long term survival; under appropriate conditions, one can help other helpful members of one's community to help survival of the helpfulness trait. The mathematics is not simple, but recent work has made much of this clearer, and it is far more than pure reproduction. The part relevant to this discussion is that for a social animal this survival depends a lot on social calculations as well as other considerations.

So, then, we can define function as being able to properly calculate and take appropriate action; but that depends on the environment one faces. The same trait of fast decisive action to take the life of an unexpected person is wonderful in times of violent combat but completely malfunctional in a peaceful society. Similarly, it is easy to show that a mental make up that helps everyone, whether or not they are helpful to others, is malfunctional in the sense that it does not help its own survival except in societies that pays a high moral premium on that. Now, since most traits will find themselves in various environments, the malfunctional has to be defined as an intermediate: it should not be "fatal" in any of the environments that an individual is likely to face. But, this depends on the environments one is "likely" to face.

Given this situation, therefore, most traits tune themselves to intermediate values, because extreme values are typically extremely ill suited in some environments one is likely to face. And, all this is further constrained by the possible organization of the brain: for example, it is completely possible that the brain is composed of two parts, one that can analyze and model its environment in terms of an "open-loop" system controlled by impersonal physical laws which constrain and guide change, and a social system that can alternately assign agency (or "will") to parts of the environment. If this simple separation of thought patterns is an useful approximation, the division of resources between the two will affect a lot of behavior: a lot of resources devoted to the physical system will make one unable to understand complicated social dynamics; whereas too high a reliance on the social system might make one unable to understand that physical phenomenon often do not have wills and desires. Both of these taken to an extreme are obviously malfunctional, and, therefore, diseased: one can think of autism or schizophrenia as examples illustrating such symptoms. But, where exactly one stops being analytical and starts being high-functional autistic will depend on what environment one is defining with respect to: when the norm is highly complex social environments, one will probably classify some highly analytic people as diseased because they cannot function in society (i.e., the "mad scientist" or "computer geeks" will get classified as "mad" or "autistic"), whereas when complex physical systems but with little social structure are the norm, some people who see willful patterns in the universe will find themselves considered ill (e.g., a "religious fanatic" will be considered "mad").

So, what have we done through all this argument? We started by arguing that DSM (diagnostic manual) definitions depend on a certain standard of normal and are not objective. Through the chain of arguments, I have tried to establish that the former (i.e. dependence on the standard of normal) is inherent part of the problem, and cannot be removed except in the trivial sense that some things have never been normal. I have also argued, however, that this dependence does not need to be subjective: what is important is not what the "doctors" have experienced as normal, but rather the environments that the *person* being diagnosed has experienced and is likely to experience.

The interesting question is that supposing we take a bunch of brains and tune up their creativity (by changing whatever neurotransmitter chemistry or electrochemical connections that we need to). Now, in some environments and depending on the rest of the circuits in the brain, this will work perfectly fine and be very useful in understanding and modeling otherwise-hard-to-model systems (somewhat similar to a physical effect called "annealing"). If the same tuning is done to a different brain which does not have the same set of controls, this tuning could lead to a bipolar disorder. Basically this hypothesis would say that creativity needs to be balanced by other control systems, so any means of independent inheritance will quite often lead to getting the creativity structures without the control structures, leading to madness. Under this hypothesis, creative people are not insane, but biology would dictate that they are at a higher risk of having insane relatives (children/siblings/etc.) than less creative people.

But, there is a different possibility as well: the "control" unit hypothesized in the previous post may not be inherited much, but developed based on experiences; or its need may be dependent on the environment. In this case, the only difference between creative people and people with some forms of insanity would be the environments they have faced or will face. Creative people can then look at bipolars and paraphrase Bradford "But for the grace of environment, there go I". We do not know if either of these hypotheses are correct, but I hope I have explained why I find it interesting to ask these questions, and why the data presented in the article is consequently interesting.

Monday, November 12, 2012

I haven't abandoned the blog!

Sorry for the absence, I've been, and still am, extremely busy. Unfortunately I have to slow down the frequency of posts. But I'm still here!

We've had our first snow and I found a new photographic challenge: snowflakes. They're harder than water refractions, with the added difficulty that after a few clicks my fingers are frozen.

For those of you who know how I spend my nights... yes, you've guessed it, my muse struck again.

I'll leave you with a question to muse: What do you think the future of the Internet will be, say twenty years from now?

Tuesday, November 6, 2012

Don't forget to vote today (US)

It is not titles that honour men, but men that honour titles.
Niccolò Machiavelli

Don't vote based on ideal principles. Be pragmatic. Vote for our children, vote for the best of the Country.

When you vote, remember the past and look up to the future.


Monday, October 29, 2012

GMOs love me, GMOs love me not..

I've been asked to discuss genetically modified foods and I confess I've been procrastinating. Why? Because I don't have an answer on whether or not GMOs are good or bad, and I can't offer one. But, what I can do is offer a few thoughts. Food for thought is usually super-natural, organic, and pesticide-free, so here it goes. :-)

1. What are GMOs?
Technically, all domesticated plants and animals are "genetically modified" since, rather than letting the species evolve through natural selection, mankind has steadily selected offsprings according to some man-made criteria. However, today's technology allows us to artificially modify an organism's genome. The difference between the two is not just in time scale: when selecting crops, or, more in general, any organism, generation after generation based on phenotype, uncharacterized genes are introduced in the species. Genetically engineering, or bioengineering, however, introduces a few well-characterized genes (often from a different species) into the organism. In a way, this is no news: gene therapy creates genetically modified organisms. Humanized mice are created in labs to test drugs and other therapies. The question of whether or not GMOs are good arises in the food industry. Are they safe to eat?

2. The Cartagena Protocol on Biosafety
As California gets ready to cast its vote on Proposition 37 [1] (which would require foods to denote their GMO content on the labels), it is good to review what currently is in act to "protect" us from possible hazards. From Wikipedia:
"The Cartagena Protocol on Biosafety is an international agreement on biosafety, as a supplement to the Convention on Biological Diversity. The Biosafety Protocol seeks to protect biological diversity from the potential risks posed by genetically modified organisms resulting from modern biotechnology."
"The Biosafety Protocol makes clear that products from new technologies must be based on the precautionary principle and allow developing nations to balance public health against economic benefits. It will for example let countries ban imports of a genetically modified organism if they feel there is not enough scientific evidence that the product is safe and requires exporters to label shipments containing genetically altered commodities such as corn or cotton."

3. Why are foods genetically modified?
For a number of reasons, some good and some not so good. Some are just practical reasons in a world that, whether we like it or not, is getting more and more "globalized": the first bioengineered produce was a tomato designed to have a prolonged shelf life. Some crops are genetically modified to resist harsher herbicides and pesticides. Others, are genetically modified to desist bugs from eating them. For example, genes producing Bt toxins have been introduced in cotton and corn. These toxins kill caterpillars that would otherwise eat up the whole crop. Notably, the modification benefits not only the genetically modified crops, but, since it reduces the global population of harmful caterpillars, it also benefits the non-modified crops.

I expect foods that can resist herbicides to be soaked in chemicals. On the other hand, if a crop is genetically modified so its flowers/fruits/seeds no longer offer a viable environment to certain parasites, I expect those foods to be pesticide-free. Yes, I'll take a few modified genes over harmful chemicals. Bottom line: NOTING WHETHER OR NOT A CERTAIN FOOD CONTAINS GMOs DOES NOT HELP. What you should really demand in a label is WHY SUCH FOOD WAS MODIFIED AND WHAT WAS ACHIEVED THROUGH THE BIOENGINEERING. Notice that while the Food and Drug administration currently does not impose any GMO labeling, their guideline recommendations state that the GMO content be noted, as well as the reason why the food was modified, and what was achieved through the modification.

4. Genetic homogeneity is bad
Rice is one of the most consumed crops in the world. Again, from Wikipedia:
"As of 2009 world food consumption of rice was 531,639 thousands metric tons of paddy equivalent (354,603 of milled equivalent), while the far largest consumers were China consuming 156,312 thousands metric tons of paddy equivalent (29.4% of the world consumption) and India consuming 123,508 thousands metric tons of paddy equivalent (23.3% of the world consumption). Between 1961 and 2002, per capita consumption of rice increased by 40%."
Rice is also highly "domesticated", as it has been selected over thousands of years to fit human needs. Currently, there are 20 different kinds of rice, but, according to FAO, the Food and Agriculture Organization, "It is estimated that not even 15 percent of the potential diversity has been utilized." This is a THREAT to food security. If a pesticide-resistant parasite were to attack rice crops, it'd be lethal to the vast majority of rice varieties currently harvested. Heavy use of pesticides favors the selection of pesticide-resistant organisms, while domestication favors genetic homogeneity in crops. This is NOT a good combination. Another reason why, between GMOs and pesticides, I'd favor GMOs. And if GMO research can prevent a pesticide-resistant organism to wipe out 50% of the world-wide food, hey, who's to complain?

5. Knowledge is NOT power if that knowledge is poorly understood
We live in a strange era when technology leaps forward at a higher speed than our ability to comprehend its output, especially in the field of genetics. We have loads of data we don't quite know how to store, let alone analyze. It's getting cheaper and cheaper to have a full human genome typed and companies are advocating that we do it for every individual. But are we capable of understanding the data? Last week I posted a shocking story of a boy discriminated because he carries a recessive mutation for a disease he doesn't have and he's at no risk of contracting (that's what recessive means). The Internet is full of bogus info on genes, genetics, mutations, etc. There's more noise than ever, giving people the illusion that they know when in fact they don't. I fear that the same will happen for GMOs. Once those labels come out, will people be able to understand what they mean? If Prop 37 will only require a "content" statement without a "reason", for example, will the information be really useful or will it just generate a stigma?

You now see why I cannot tell you whether GMOs are good or bad. They can be both! (Aren't we all?)

Food always has a higher impact than other things, but if you think about it, there are so many things that we've introduced in our daily lives in the past few decades that we simply don't know whether or not they are good IN THE LONG RUN: wi-fi, for example. Cell phones. Chemicals in skin products, from sun protection to cosmetics. I'm afraid the next generation will be the test. So the real question is: do we want to experiment with our children as guinea pigs? Sadly, when you put it in these terms, it seems to me it's too late to go back. The experiment has already begun.

If these few thoughts weren't depressing enough, read Pamela Ronald's review, referenced below [2]. One of the great points Ronald makes is that we are changing our climate and environment much faster than ever before (thanks to climate change and an exponentially growing population). Natural selection can't keep up with the pace, hence
"an important goal for genetic improvement of agricultural crops is to adapt our existing food crops to increasing temperatures, decreased water availability in some places and flooding in others, rising salinity, and changing pathogen and insect threats."
The review is clearly biased in favor of GMOs and it lists several benefits from such procedures. While advocating for adequate testing on every newly modified organisms, it also reports that all genetically modified crops tested so far have been deemed safe and substantially no different than conventionally selected crops "in terms of unintended consequences to human health and the environment."

Bottom line: I can't tell you what to vote on Prop 37 and I can't tell you whether or not you should avoid GMOs. Just read as much as you can and be sure to form your own opinion.


[1] Baker, M. (2012). Companies set to fight food-label plan Nature, 488 (7412), 443-443 DOI: 10.1038/488443a

[2] Ronald, P. (2011). Plant Genetics, Sustainable Agriculture and Global Food Security Genetics, 188 (1), 11-20 DOI: 10.1534/genetics.111.128553

Monday, October 22, 2012

Lorenzo's oil got upgraded to stem cell research

Have you seen the 1992 movie Lorenzo's oil? The film portrays the true (and sad!) story of Lorenzo Odone, who, at age 6, was diagnosed with adrenoleukodystrophy, one of the most common forms of leukodystrophies, a family of degenerative diseases that affects the growth of the myelin sheath. Myelin wraps around nerve fibers creating a fatty covering that increases the speed at which impulses propagate. Leukodystrophy is a genetic disorder caused by mutations in the genes that code myelin proteins. When myelin is defective, or not produced in sufficient quantities, it starts degrading, causing the progressive loss of signaling along the nerve. Eventually, the nerve dies.

As shown in the movie Lorenzo's oil, Lorenzo's parents refused to accept the common prognosis they were given at the time for their son (progressive paralysis and death within 2-3 years). Their determination led them to discover an oil mix able to alleviate the symptoms of the disease.

Two papers published in Science Translational Medicine now show that stem cell therapy can partially regenerate neurological function.

Uchida et al. [1] show that stem cell transplantation is effective in the regeneration of the myelin sheath in mouse models. The researchers transplanted human central nervous system stem cells (HuCNS-SCs) into the brains of mice with defective myelination in the central nervous system. The transplanted stem cells generated functional myelin in the mice's central nervous system

In the same issue, Gupta et al. [2] describe how they transplanted the same cells (HuCNS-SCs) into the frontal lobe of four young boys that were affected by Pelizaeus–Merzbacher disease (PMD), a form of leukodystrophy. The transplant was followed by a 9-month regimen of immunosuppression to minimize the chances of rejection. One year after the transplant, magnetic resonance imaging (MRI) showed that the transplanted cells had engrafted and successfully myelinated brain cells. The researchers conclude that "modest gains in neurological function were observed in three of the four subjects. No clinical or radiological adverse effects were directly attributed to the donor cells."

Sadly, Lorenzo died in 2008, one day after his thirtieth birthday. His story, though, was and still is an inspiration to many.

[1] Uchida, N., Chen, K., Dohse, M., Hansen, K., Dean, J., Buser, J., Riddle, A., Beardsley, D., Wan, Y., Gong, X., Nguyen, T., Cummings, B., Anderson, A., Tamaki, S., Tsukamoto, A., Weissman, I., Matsumoto, S., Sherman, L., Kroenke, C., & Back, S. (2012). Human Neural Stem Cells Induce Functional Myelination in Mice with Severe Dysmyelination Science Translational Medicine, 4 (155), 155-155 DOI: 10.1126/scitranslmed.3004371

[2] Gupta, N., Henry, R., Strober, J., Kang, S., Lim, D., Bucci, M., Caverzasi, E., Gaetano, L., Mandelli, M., Ryan, T., Perry, R., Farrell, J., Jeremy, R., Ulman, M., Huhn, S., Barkovich, A., & Rowitch, D. (2012). Neural Stem Cell Engraftment and Myelination in the Human Brain Science Translational Medicine, 4 (155), 155-155 DOI: 10.1126/scitranslmed.3004373

Saturday, October 20, 2012

From ABC news: "Boy Ordered to Transfer Schools for Carrying Cystic Fibrosis Gene Mutation"

I'm totally speechless.

Read the article and tell me it's a Halloween prank.


Friday, October 19, 2012

Why extra fat is bad, even when it's in the "right" spots.

Have you been thinking of going on a diet but haven't found the right motivation yet? How about this one: fat feeds tumor cells and enhances their growth. And another question, for the ladies this time: have you ever wondered how those annoying love handles would look so much better inside a bra? No, I don't mean to put a bra around my waist, rather to move that bit of fat up to my chest . . .

Somehow the two things are related. Stay with me and I'll explain.

Numerous studies have shown that obesity not only increases cancer risk, but it's also linked to accelerated progression in numerous types of cancers. A group of researchers from the University of Texas Health Science Center at Houston investigated the reason for such poorer prognosis in obese patients. In a paper published in Cancer Research [1], Zhang et al. show that white adipose tissue (WAT) facilitates tumor growth in mice, and the association was independent of the mice's diet. To show this, instead of overfeeding the mouse to make it grow the fat tissue, they transplanted into the animals adipose stromal cells (ACS) -- cells that can be thought of as the "progenitors" of adipose cells. The transplanted cells increased the proliferation of white adipose tissue. Furthermore, once recruited into tumors, they increased tumor vascularization.

Tumors are basically an uncontrolled growth of cells. It takes a lot of resources to keep cells growing, and new blood vessels are created to "feed" the growth. Zhang et al. showed that the transplanted adipose cells were mobilized in the mouse model and promoted the creation of new blood vessels, thus effectively "feeding" the tumor and promoting its progression. Yuck, right?
"Our results indicate that obesity can accelerate tumor growth irrespective of concurrent diet. [. . .] Our data indicate that ASCs recruited by tumors become perivascular or differentiate into intratumoral adipocytes [1]."
Why did I mention love handles and bras? Because Yoshimura et al. published a paper [2] in 2008 in which they do to humans what Zhang et al. did to mice. For a good reason, of course: in [2] Yoshimura et al. illustrate a new technique called cell-assisted lipotransfer in which they use the aforementioned ACS, the adipose progenitors, to perform cosmetic breast augmentation. Seems like a brilliant idea: no implants needed, no lipoinjection (which carries a high risk of necrosis), just isolate the fat cells, let them grow, then transfer them back. No scars, no complications.
"Final breast volume showed augmentation by 100 to 200 ml after a mean fat amount of 270 ml was injected. Postoperative atrophy of injected fat was minimal and did not change substantially after 2 months. Cyst formation or microcalcification was detected in four patients. Almost all the patients were satisfied with the soft and natural-appearing augmentation [2]."
But now you see why this could present a potential problem: many breast cancer patients seek tissue regeneration after a mastectomy, and while Yoshimura's technique seems innovative and promising, Zhang et al. warn against its possible risks as the fat transfer could potentially feed remaining cancer cells and yield devastating results.

And the moral of the story is: go on a diet, get rid of the love handles, and be happy with a smaller size bra. ;-)

[1] Zhang, Y., Daquinag, A., Amaya-Manzanares, F., Sirin, O., Tseng, C., & Kolonin, M. (2012). Stromal Progenitor Cells from Endogenous Adipose Tissue Contribute to Pericytes and Adipocytes That Populate the Tumor Microenvironment Cancer Research, 72 (20), 5198-5208 DOI: 10.1158/0008-5472.CAN-12-0294

[2] Yoshimura, K., Sato, K., Aoi, N., Kurita, M., Hirohi, T., & Harii, K. (2007). Cell-Assisted Lipotransfer for Cosmetic Breast Augmentation: Supportive Use of Adipose-Derived Stem/Stromal Cells Aesthetic Plastic Surgery, 32 (1), 48-55 DOI: 10.1007/s00266-007-9019-4

Tuesday, October 16, 2012

Reprogrammable cells

Can't remember if I already shared the above picture... it's my favorite sunset shot so far, so forgive me if it's a deja vu.

The Nobel Prize in medicine this year was awarded to John Gurdon and Shinya Yamanaka for pioneering the reprogramming of cells into an embryonic-like state. Embryonic stem cells are cells that undergo asymmetric division, as they divide into an undifferentiated cell and into a specialized cell. This way, they can grow indefinitely while maintaining their undifferentiated state and, at the same time, keep the ability to differentiate into all three germ layers, the cells formed during embryogenesis.

When still a PhD student, in 1958, Gurdon cloned a frog using the nucleus of a cell taken from the intestine of a tadpole. It took another 38 years before the first mammal was cloned: the first cloned sheep, Dolly, was born in 1996 from an unfertilized egg whose nucleus had been replaced with the nucleus of an adult cell. In this case, the adult cell, by being placed into the egg, was effectively "reprogrammed" into an embryonic stem cell. Up until Gurdon's work was published in 1962, general belief was that once cells specialized, they could not revert. The discovery that cells can actually undergo "reprogramming" under special circumstances is quite significant because it gives hope that we can achieve tissue regeneration and treat degenerative diseases or spinal cord injuries.

In 2006 Yamanaka and his colleague Kazutoshi Takahashi published a paper in Cell [1] in which they showed that, activating four genes, they were able to reprogram adult fibroblasts from mouse embryonic cells. They called the new cells induced pluripotent cells, or iPS, and found that they expressed embryonic-state cell markers. In fact, once in the proper environment, they contributed to embryonic development.

Yamanaka is a strong believer that this research will eventually lead to successful regeneration therapies. In fact, he plans to start a bank of induced pluripotent stem cells obtained from 75 different cell lines. Is this the beginning of a new era? A word of caution comes from a paper published in PNAS at the end of 2010 [2]: in this paper, Serwold and colleagues derived mice from reprogrammed T-cells (cells from the immune system) and showed that roughly half of the mice generated this way spontaneously developed T-cell lymphomas.

The mice were generated by transferring T-cell nuclei into enucleate oocytes. As they mature, T-cells undergo genomic rearrangements, and while normally these rearrangements occur in T-cells only during a specific stage of their development, such rearrangements were observed in all somatic cells in the cloned mice. In [2] Serwold et al. show that these rearrangements undergo T-cell lymphomagenesis: in other words, they cause cancer. Though T-cells are not the only cells that currently can be reprogrammed, this study clearly shows that different cell lines can yield different outcomes, some quite deleterious.
"This study suggests that precautions should be taken to ensure that the identity of the reprogrammed cell of origin is known, and that T cells, and probably also B cells, are not inadvertently turned into therapeutic iPS cells. Recent studies have used human blood-derived T cells as sources of iPS cells, and these cells promise to be valuable tools for studying human immune development and disease; however, the results presented here indicate that extra caution is warranted regarding the therapeutic use of such T cell-derived iPS cells [2]."

[1] Takahashi, K., & Yamanaka, S. (2006). Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors Cell, 126 (4), 663-676 DOI: 10.1016/j.cell.2006.07.024

[2] Serwold, T., Hochedlinger, K., Swindle, J., Hedgpeth, J., Jaenisch, R., & Weissman, I. (2010). T-cell receptor-driven lymphomagenesis in mice derived from a reprogrammed T cell Proceedings of the National Academy of Sciences, 107 (44), 18939-18943 DOI: 10.1073/pnas.1013230107

Thursday, October 11, 2012

Gene therapy goes... topical

The paper I'm discussing today is so cool, I don't know how I missed it when it came out last July. As the name implies, gene therapy is a technique used to "fix" defective genes either by replacing them with fully functional ones or by silencing them with the use of antisense RNA.

Defective genes either fail to produce the proteins they code for, or produce defective proteins, thus causing genetic disorders. A defective gene can be silenced (so that it will no longer produce the defective protein) using antisense RNA. The antisense RNA binds to the mRNA from the defective gene, thus preventing it from being translated into the protein. Small interfering RNAs, or siRNAs, are short double-stranded RNAs that can successfully deliver antisense RNA to the target genes and effectively suppress gene expression.

There are several ways to deliver either DNA or RNA, each with advantages as well as disadvantages. Viral vectors (genetically modified viruses that instead of carrying viral DNA they carry the therapeutic DNA or RNA to be delivered inside the cell) are great ways to deliver genes, but have to overcome the barrier imposed by the host's immune system. Furthermore, some vectors may have toxic side effects. Other delivery means include conjugate agents, i.e. particles such as lipids, polymers, or nanoparticles that bind to the RNA and have high penetrability. Though such therapies have been quite promising, they pose a challenge: they can be toxic when delivered intravenously or orally, while the topical route is inefficient because the skin won't let through anything greater than a few daltons.

That's too bad, though, because the skin would be the less invasive and easiest way to deliver therapy. Just imagine it: applying genes in the morning just like a daily moisturizer! :-)

In [1] Zheng et al. show that by conjugating siRNAs with inorganic gold nanoparticles, they can defeat the epidermic barrier and successfully reduce the expression of the target genes.
"Recently, we introduced spherical nucleic acid nanoparticle conjugates (SNA-NCs, inorganic gold nanoparticles densely coated with highly oriented oligonucleotides) as agents capable of simultaneous transfection and gene regulation. [. . .] SNA-NCs enter almost 100% of cells in more than 50 cell lines and primary cells tested to date, as well as cultured tissues and whole organs."
The researchers measured uptake, safety, and gene suppression efficacy of SNA-NCs in human keratinocytes, a cell line that constitutes 95% of the epidermis. They found no morphological difference between treated skin cells and controls. Furthermore, they studied the ability of SNA-NCs to aid the silencing of EGFR, or epidermal growth factor receptor, a cell-surface receptor that has been shown to be mutated and up-regulated in several types of cancers, including lung, anus, and 30% of skin cancers. After 3 weeks of treatment, EGFR expression was suppressed by 65% in hairless mice. Human skin is known to be thicker and more difficult to penetrate, so the treatment was also tested on 3D raft cultures that simulate in vivo human epidermis. EGFR mRNA expression was lowered by 52% and EGFR protein expression by 72%.

Zheng et al. conclude:
"Our data from blood serum and mouse skin show that siRNAs, when densely conjugated to the nanoparticles in the form of SNA-NCs, are minimally stimulatory and have far fewer off-target effects than the free siRNAs of the same sequence introduced by traditional methods. Furthermore, our studies show rapid clearance from skin, minimal accumulation in viscera, and no evidence of histological changes in internal organs after topical delivery. The low immunogenicity and few off-target effects, coupled with low toxicity and high efficacy, point to a significant advantage for using SNA-NC technology to introduce siRNAs."

Dan Zheng, David A. Giljohann, David L. Chen, Matthew D. Massich, Xiao-Qi Wang, Hristo Iordanov, Chad A. Mirkina, & Amy S. Paller (2012). Topical delivery of siRNA-based spherical nucleic acid nanoparticle conjugates for gene regulation PNAS DOI: 10.1073/pnas.1118425109

Tuesday, October 9, 2012

Albuquerque International Balloon Fiesta

Every year in October, Albuquerque hosts the International Balloon Fiesta. Here are some shots from the dawn patrol.

Thursday, October 4, 2012

Limb regeneration: a lesson from salamanders

As much as we would love to enlist limb regeneration among modern science's best accomplishments, so far it is still very much confined to science fiction. That doesn't mean it won't happen, though. Key to limb regeneration is cellular reprogramming that allows differentiated cells to return to a germline-like (undifferentiated) state. Genes involved in embryonic development need to be reactivated in order to restart the same process that created the limb during the growth of the embryo.

The vertebrates with the best ability to regenerate limbs are salamanders, making these little critters the most studied in the field. When a salamander loses a limb, a layer of epidermis grows to cover the wound, and beneath this layer new, undifferentiated cells start proliferating, forming a mass called blastema. Recent research shows that this first wave of cell dedifferentiation may recapitulate events occurring during embryogenesis.
"The cells in the limb blastema are believed to be a heterogeneous collection of dedifferentiated cells that have been reprogrammed to achieve varying levels of developmental potential exhibited by the cells involved in embryogenesis [1]."
Germline stem cells are cells that give rise to gametes (the reproductive cells) and have the ability to divide into another stem cell as wells as a more differentiated cell. This mechanism, called asymmetric division, is controlled by a protein called PIWI through small, non-coding RNAs called piRNAs. In [1] Zhu et al. showed that when salamanders regenerate a limb, a germline-like state is established in the growing tissue. In particular, they found that germline-specific genes were expressed in the regenerated limb. In order to show this, they looked specifically at the PIWI proteins.

Zhu and colleagues found a significant amount of upregulated transposable elements in the regenerated limbs. If you remember, transposable elements is a segment of DNA that can move from one locus to another within the genome of the same cell. During the limb regeneration process, transposable elements can impart a deleterious amount of instability, which is counteracted by a corresponding upregulation of the PIWI genes. Conversely, when the PIWI genes were knocked down (i.e. their expression was reduced) in the blastema, limb growth following the amputation was significantly reduced compared to controls.

Zhu et al. conclude
"In the future, further characterization of the subpopulations of these reprogrammed cells with additional germline-specific markers might provide more insight into exactly how far cellular dedifferentiation can proceed and whether there are indeed a small number of cells that could be isolated before a certain developmental threshold and exhibit true pluripotency when isolated from the influence of the partially programmed blastemal cells in the proximity."

[1] Wei Zhu, Gerald M. Pao, Akira Satoh, Gillian Cummings, James R. Monaghan, Timothy T. Harkins, Susan V. Bryant, S. Randal Voss, David M. Gardiner, & Tony Hunter (2012). Activation of germline-specific genes is required for limb regeneration in the Mexican axolotl Developmental Biology DOI: 10.1016/j.ydbio.2012.07.021

Monday, October 1, 2012

The sleep conundrum

How many hours of night sleep do you get? Are you ever surprised at how many more/less hours other people sleep? Well, if you are, you might find comfort knowing that the variation in number of sleep hours across species is huge and, so far, very much an evolutionary mystery.

Common thought is that sleep provides us with a much necessary "recharging" and common that it has an evolutionary advantage. However, it takes time away from foraging/preying and mating, and makes individuals more vulnerable to predators. Furthermore, the huge spread of sleeping hours across species makes it harder to pin point whether it does have a selective advantage or not.

A new paper published in Science [1] shows that at least in one bird species, pectoral sandpipers (Calidris melanotos), being able to sleep less confers an advantage: males compete for fertile females over a period of 3 weeks, during which they sleep very little. Lesku et al. showed that males who slept less were the ones who produced the most offsprings.

The researchers' main hypothesis was that
"variability in sleep duration observed across the animal kingdom reflects varying ecological demands for wakefulness, rather than different restorative requirements. According to this hypothesis, animals can evolve the ability to dispense with sleep when ecological demands favor wakefulness."
Pectoral sandpipers mate with multiple females and are not involved in the raising of their offsprings. Therefore, their mating success is determined solely on how many fertile females they have access to. Furthermore, in the high Arctic, the sun never sets during mating period. Males are awake longer than females and engage in courtship behaviors, competitions with other males, and defending their territory (don't know why, high school just came to mind. . . ahem). Their total wake time was a strong predictor of how many offsprings they ended up having. A caveat the researchers put forward is that this conclusion seems to be at odds with the hypothesis that there is a genetic basis to the duration of sleeping times, as shorter durations would be selected and hence the variation in wake time would progressively lessen. Instead, still a great variation was observed across males.

The bit that I found most intriguing is that at lower latitudes at some point darkness falls, and since most of courtship is based on visual displays, this limits the daily amount of time males can engage in such activities. However, the study was conducted in Alaska, where the sun never sets during mating season. I wonder if the same study on pectoral sandpipers that live at lower latitudes would have yielded the same results. I also can't help but wonder: this behavior was obviously selected by a favorable environment, in this case the fact that the sun never sets during summer days. I think it was this past summer that my dad said, "Would birds have ever learned to fly if it weren't for the wind?"

[1] John A. Lesku, Niels C. Rattenborg, Mihai Valcu, Alexei L. Vyssotski, Sylvia Kuhn, Franz Kuemmeth, Wolfgang Heidrich, & Bart Kempenaers (2012). Adaptive Sleep Loss in Polygynous Pectoral Sandpipers Science DOI: 10.1126/science.1220939

Thursday, September 27, 2012

Healthy habits are easier when you stop thinking about it

Raising health awareness has done little so far in actually improving global health. Humans seem to be stubbornly attached to certain behaviors, even when fully aware that such behaviors pose a health risk.

Currently, the four most prevalent noncommunicable diseases are diabetes, cardiovascular disease, lung disease, and cancer. The risk of death from any of the four can be significantly lowered by changing basic behaviors such as lowering the consumption of calories, alcohol and tobacco, while increasing physical activity and the consumption of fruits and vegetables. It sounds simple, in theory, but certain behaviors are so engrained in the society that despite widespread campaigns, we still haven't been able to change people's habits. And not surprisingly so, since much of our behavior is often automatic rather than dictated by consciousness. As Marteau et al. state in [1], not even personalized risk assessments like gene variants and other biomarkers have succeeded in dissuading people from certain behaviors.

We are complex beings, constantly shifting from full awareness and reflective, goal-driven behavior, to more automated actions where deep thoughts are far removed. The former behavior is more costly in terms of metabolic resources and energy. The latter is more efficient in our daily routine, but it has the disadvantage of taking over even when the consequences are undesired. For example, lab animals that have been trained to repeat certain behaviors, they will keep repeating them even when unpleasant consequences are introduced in the experimental setting. Therefore, in order to prevent noncommunicable diseases, Marteau et al. argue that we need to target automatic behaviors rather than conscious ones.

How can this be achieved?

Well, for example making fruits and vegetables very easy to find at the store, and relegate the so-called junk food to some hidden, desolate aisle that requires extra walking to get to. Also (and I know I'm totally going against common economy rules here), making fruits and vegetables cheaper than junk food would cause a huge switch in people's eating habits. If this may sound much of an utopia (yeah, I can see that), here are a few more practical things that can be changed: make stairs accessible from everywhere in a building, and hide the elevators. Make the elevators really slow that it's a lot more practical to take the stairs. Make tobacco and alcohol harder to find (though I have doubts about alcohol, since I grew up in a country where alcohol is on the table every day and somehow we seem to handle alcohol addictions better than other countries with lots of rules and prohibitions). Use smaller serving portions and smaller (but taller) glasses and plates. Marteau et al. even suggest "standing desks" in classrooms to have students spend more calories (this one made me smile).

Here's my two cents. As you know, I grew up in Italy, a country that very much cherishes food and spending social time at the table. I think Italians have pushed things to the far extreme, and now, when I go back to visit, after about one hour of sitting at the table "socializing" I get a little restless. Much of the overweight problems in Italy come from spending too much time at the table. After a while you don't feel hungry anymore but you just keep eating because food is being offered to you.

On the other hand, I see that the United States have the exact opposite problem. There's no definite time of when to eat lunch or dinner, and when you look around you see people eating at any given time of the day. This is just my personal opinion, of course, but I truly believe that introducing fixed eating times in the day can greatly help towards healthier eating behaviors. Also, we should learn from our children. When they are full they stop eating. Parents tend to get edgy and force them to eat more, whereas maybe it should be opposite, it should be the children telling the parents to stop eating so we can all get back into the habit of eating only when we're hungry. Unfortunately, because eating is so much part of our social life and social celebrations, in real life, things tend to get more complicated.

[1] Theresa M. Marteau, Gareth J. Hollands, & Paul C. Fletcher (2012). Changing Human Behavior to Prevent Disease: The Importance of Targeting Automatic Processes Science

Monday, September 24, 2012

ENCODE sheds light on non-coding variants

Back when I started studying human genetics, we were still doing single-gene associations. Namely, we would type a bunch of variants in a single gene and then do a case-control association study to see which, if any, of those variants marked an increase in disease risk. That's how breast cancer markers such as BRCA1 and BRCA2 have been found.

When the Human Genome Project was completed in 2003, scientists started looking for disease risk alleles across the whole genome. The findings were puzzling: more than 90% of the diseases-associated variants fell in non-coding regions. Why? One issue I've previously discussed is that when looking at tens of thousands of loci, you need huge sample sizes and often, when huge sample sizes aren't feasible, these studies are underpowered. Another possible explanation lies in epistasis, and the detected signal may be the effect of some unknown correlation.

However. You knew there was going to be a "however", right? Because thanks to the ENCODE project we now know that if a genetic variant falls in a non-coding region, it doesn't mean it has no effect whatsoever. ENCODE is bound to shed new light on these numerous non-coding risk alleles that genome-wide association studies (GWAS) studies have found.

Last time I discussed DHSs, or DNase I hypersensitive sites. These are chromatin regions where many regulatory elements have been found. In [1], Maurano et al. show that many of the non-coding variants associated with common diseases are concentrated in regulatory DNA marked by DHSs. The researchers performed genome-wide DNase I mapping across 349 cell and tissue types. As discussed last week, regions of DNase I accessibility harbor regulatory elements. The researchers also examined the distribution of 5654 non-coding SNPs (single base variants) that had been significantly associated to some disease or trait in genome-wide studies.

These the main findings:
"Fully 76.6% of all noncoding GWAS SNPs either lie within a DHS (57.1%, 2931 SNPs) or are in complete linkage disequilibrium (LD) with SNPs in a near-by DHS (19.5%, 999 SNPs)."
To be in linkage disequilibrium means that the variant is typically inherited together with a DHS site. Suppose the true causal variant is at locus A, but you haven't typed locus A, you've typed locus B, and A and B are inherited together. Then B is going to light up as strong signal in your statistical analysis. So, what Maurano et al. are saying in the above paragraph is that the non-coding SNPs either turned up in a DHS site, or they found evidence that they were strongly correlated with one of such sites.
"Many common disorders have been linked with early gestational exposures or environmental insults. Because of the known role of the chromatin accessibility landscape in mediating responses to cellular exposures such as hormones, we examined if DHSs harboring GWAS variants were active during fetal developmental stages. Of 2931 noncoding disease- and trait-associated SNPs within DHSs globally, 88.1% (2583) lie within DHSs active in fetal cells and tissues. Of DHSs containing disease-associated variation, 57.8% are first detected in fetal cells and tissues and persist in adult cells (“fetal origin” DHSs), whereas 30.3% are fetal stage–specific DHSs.
And finally:
"Enhancers may lie at great distances from the gene(s) they control and function through long-range regulatory interactions, complicating the identification of target genes of regulatory GWAS variants."
GWAS variants control distant genes that need not even be on the same chromosome. Furthermore, these variants in DHSs sites tend to alter allelic chromatin state, thus modulating the accessibility of genes to transcription factors. Disease-linked variants were found to alter such accessibility, resulting in allelic imbalance (one allele gets transcribed more than the other one), possibly explaining their role in altering the disease risk or quantitative trait.

[1] Matthew T. Maurano, Richard Humbert, Eric Rynes, Robert E. Thurman, Eric Haugen, Hao Wang, Alex P. Reynolds, Richard Sandstrom, Hongzhu Qu, Jennifer Brody, Anthony Shafer, Fidencio Neri, Kristen Lee, Tanya Kutyavin, & Sandra Stehling-Sun (2012). Systematic Localization of Common Disease-Associated Variation in Regulatory DNA Science DOI: 10.1126/science.1222794