We've taken another baby step away from the current one size fits all health care system.

We may finally be at the threshold of the age of personalized medicine. In a recent study, scientists were able to predict that a man was at a higher risk for developing Type 2 diabetes and over a two-year period tracked his health as he developed the disease. And even better, because they caught it so early, they were able to stave off the diabetes with lifestyle changes. This man’s glucose levels have returned to normal.

Wow. This story highlights the promise of at least one aspect of personalized medicine. By looking at someone’s DNA, you can predict what might go wrong with someone and so keep an eye out for early symptoms. Or maybe even start out with the right lifestyle changes that will keep the disease from developing in the first place.

This study also showed that intensely studying a single person can yield potential benefits for lots of other people. The researchers saw that just before the test subject’s glucose levels spiked, he had a viral infection. No one was really looking for viruses that trigger Type 2 diabetes in people. Now they will. (Keep in mind we don’t yet know if the two are connected or if it was just a coincidence.)

The study also points to the obstacles we still need to overcome to realize the full potential of personalized medicine. The top ones I could think of off the top of my head are our own ignorance, the inconvenience, the expense, and our lack of willpower.

The researchers were able to predict an increased risk for diabetes as well as an increased risk for high triglycerides but very little else. There is certainly more information lurking in his DNA…we just don’t understand our DNA well enough to tease it out yet.

Soon your treatments will be tailored for you based on at least partly on what's in your DNA.

Another related issue is whether we actually do know enough to make good predictions or if we just got lucky here. In other words, was his developing Type 2 diabetes a coincidence or was he really at a higher risk for getting it? He didn’t have any classic risk factors but given that so many people in the U.S. have the disease, it could have been chance. Doing many more studies on lots of different people should give us some idea about how predictive our DNA really is right now.

Besides our still sketchy knowledge, we also have to deal with the expense and inconvenience of this form of personalized medicine. The test subject had over twenty blood draws over a two year period that each looked at tens of thousands of different things. Not many people would put up with so many blood draws. And the expense of looking at all those different molecules is prohibitively expensive.

A better knowledge of our risks can help with the second point. Once we understand our DNA better and so know what are most likely risks are, we’ll be able to test for fewer molecules which should make the whole thing more affordable. This may also solve the first problem too.

Maybe in the future we’ll look at few enough molecules or the tests will be sensitive enough to get the information we need from a simple finger prick. Then we’d all be like folks with diabetes, self testing our blood on a regular basis. And hopefully in the more distant future, we’ll have some sort of implant that reads the information for us automatically without the need for a blood draw.

All of these are technical hurdles that will almost certainly be overcome at some point. The last obstacle, though, is much more difficult. It deals with human nature.

One reason this is such a powerful story is that the test subject was able to get his glucose under control without the use of medicines. This is not only good for him but it suggests that this form of personalized medicine may prove to be cost effective sooner rather than later. Keeping his Type 2 diabetes at bay will probably save tons of money over his life time. Perhaps even enough to justify the cost of his testing.

But to control his glucose levels, he had to make radical changes to his diet and exercise regime. He had to eat a whole lot less sugar and fat and exercise a lot more. Sound familiar?

Everyone should be doing this stuff anyway but most of us don’t. Will we have the willpower to realize the full potential of personalized medicine? Or will things pretty much stay the same except with more frequent scolding from our doctors?

Of course, catching a disease early and getting patients their medications early when it could do the most good is obviously wonderful too. Just not as cost effective.

What a visit to the doctor in the near future might look like.

Brand new mutations are helping scientists figure out what is going on in autism.

Autism is incredibly frustrating from a genetic point of view. Every study clearly shows that genetics plays an important role in this disease. But when these studies try to find a cause, they keep coming up short.

And this isn’t because scientists aren’t trying hard. They are. In most of the recent studies they are comparing thousands of people’s DNA at millions of different spots. If there was a simple explanation, they would have found it.

One thing they have managed to find from all of these studies is that a minority of cases result from brand new mutations that most likely happen in either the sperm or the egg before fertilization. While these are not going to be that useful as a diagnostic test, they may prove useful as a way of figuring out which genes to focus on. And maybe even for coming up with new ways to treat autism.

With that in mind, a whole slew of researchers set out to find more of these sorts of de novo mutations (as they are called). They reported their findings in the following three back to back to back reports in the journal Nature:

Study 1
Study 2
Study 3

These researchers took on the herculean task of looking at every letter of every gene of over two thousand people. Basically they compared the genes of parents to the genes of their autistic children (and a few of their unaffected siblings). They were able to confirm that some previously identified genes were important in autism and they identified a few new ones.

They were also able to confirm that older dads pass on more of these sorts of mutations than younger dads or moms of any age. Which makes sense if you think about sperm production.

Each time a sperm is made, its DNA needs to get copied. And each time DNA is copied, there is a chance for a mistake to creep in. So it makes sense that the older the dad, the more mutations he’ll have in his sperm.

Frustratingly, this was the most solid finding in the study. But fret not. They also learned some important things about how autism happens too.

The most important thing they found was that many children with autism shared mutations in related genes. Their mutations affected different genes that all impacted the same or related biochemical pathways.

An fMRI scan of an autistic brain.

This is what we might expect from something as complicated as brain development. To pull off something like this, various genes are going to need to fire off at the right time in the right order. If any one of the genes in a similar pathway misfires, you can end up with similar problems. And this is undoubtedly happening in some cases of autism.

This is a very important finding for identifying where to focus research so new treatments can be found for this disease. Isolated genes can be hard to target because sometimes scientists don’t know what they are doing or how they fit into the grand scheme of things. But researchers can identify a pathway, then they can identify chemicals that can tweak that pathway that can one day become medicines.

These studies also highlight yet again what a hideously complex disease autism is. Your risk depends on what versions of lots of different genes you have. Some versions will increase your risk and some will decrease it. Your chances are a summation of all of these risks.

As if this wasn’t complicated enough, autism is more than genes. Something in the environment has to trigger the disease and we know very little about these triggers (except that vaccines aren’t one). Now add to this the fact that different genetic combinations are going to be susceptible to different triggers and you begin to see why this has been such a challenge to geneticists.

Figuring out why a particular person ended up with autism is really hard. But even if we can’t figure it out for a particular person, these studies are important for finding what is happening in the brains of people with autism. And hopefully by knowing that, we can find new treatments.

A long video about the biology of autism. Well worth your time.

Cow DNA tells us domestication is an incredibly difficult undertaking.

Everyone knows about how genetics is changing how we look at and treat human disease. But what may be less appreciated is what it can tell us about human history.

From studying human genetics, we know that all humans started out in Africa. We also know that early humans interbred with Neanderthals and Denisovans before wiping them out. Now a new study looking at cow DNA is teaching us about our agricultural history.

Scientists compared DNA from 8,000 year old cow bones to 20 or so different modern cow breeds from all over the world. When they plugged their data into various computer models, the one that made the most sense had modern cows coming from about 80 wild founders around 10,000 years ago.

Not only does this help explain why cows all look so much alike, it also tells us that our ancestors only managed to domesticate cows once. This is despite the fact that ancient, wild cows (called aurochs) were wandering all over Africa, Asia, and Europe at the time.

The obvious conclusion is that domesticating a cow is really hard. And this makes sense if you think about the animals.

No, I don’t mean placid Bessie out in the pasture. Ancient cows were huge, ornery creatures that would have been hard to capture alive and hard to breed. Archeological evidence suggests it took hundreds of years or more of breeding to get animals that were smaller and more docile.

Archeological evidence also helps us pinpoint where the first domestication probably happened – the Middle East. But because archeological data is often incomplete, it can’t tell us whether this was the only cattle domestication event in human history. We need genetics to confirm this (or at least to confirm that modern cows came from a single domestication event).

Genetics is providing insight into human history that we could not have gotten in any other way. We can see that humans lost their fur and gained darker skin around 1.2 million years ago. And we can see that our original pale skin from back when we were hairy came back into fashion in Europe about 6000-12,000 years ago.

And cows aren’t the first animal’s DNA to tell us something about ourselves either. By looking at lice DNA, scientists think that humans first started wearing clothes about 170,000 years ago.

It is amazing what we can learn about our history from looking at the DNA of ourselves and our associated creatures. Soon we’ll learn even more. As the price of deciphering our DNA goes down, the number of questions we can answer will go up dramatically. The only thing that will keep us back is figuring out the best questions to ask.

Dr. Mina Bissell of LBNL Life Sciences in her laboratory. Photo courtesy of Lawrence Berkeley National Laboratory. © 2010 The Regents of the University of California, Lawrence Berkeley National Laboratory.

The human body is comprised of about ten trillion cells. These cells are constantly bombarded with damaging factors, like radiation from the sun, that cause some of the cells to mutate. Even healthy people produce many genetically impaired cells every day, but our bodies successfully eradicate these cancer-prone cells so the majority of people live cancer-free lives. How is this possible?

We all know that the human body has a highly developed immune system that detects and destroys invading pathogens and tumor cells. Researchers at Lawrence Berkeley National Laboratory (LBNL) have demonstrated that there is also a second line of defense against cancer: neighboring healthy cells.

Dr. Mina Bissell is a Distinguished Scientist with LBNL and one of the world’s leading researchers on breast cancer. Her group recently found that normal breast cells provide an innate defense mechanism against cancer, by secreting interleukin 25 (a protein known to play a key role in the immune response to inflammation) to actively and specifically kill breast cancer cells without harming normal ones.

Overall Bissell’s research has focused on the importance of factors other than genetics in the development of breast cancer, demonstrating the critical role that a breast cell’s microenvironment plays in whether it develops normally or whether it turns cancerous. A cell’s microenvironment includes other surrounding cells, like cancer-killing normal breast cells, and a supporting structure known as the extracellular matrix. This extracellular matrix (ECM) consists of a complex network of fibrous and globular proteins surrounding the breast cell. Bissell has shown that a healthy ECM is critical for a breast cell to function normally. If the ECM is damaged, this can lead to breast cancer.

As Bissell explained to attendees at an American Association for Cancer Research conference in 2009, “No cell is an island. All cells are surrounded by their own unique microenvironment. It is quite clear that the context in which a cell exists determines what that cell can do.”

Surprisingly, Bissell has also demonstrated that malignant breast cancer cells can “revert” back to function like normal breast cells by manipulating their microenvironment. A reverted cell’s genetic makeup (genotype) indicates that it is still cancerous, but the actual observed properties (phenotype) are that of a normal breast cell. Bissell explained at an LBNL lecture, “Clearly the genome is a mess, but we manipulate the cells to make them think they are normal. They revert to a normal phenotype.”

Image courtesy of Friends of Berkeley Lab.

Her studies also imply that there may be a better way to treat breast cancer. Bissell argues that therapies that modulate the microenvironment have the potential to make malignant cells appear normal or to at least help tumor cells remain dormant.

Dr. Bissell will discuss her pivotal breast cancer research at LBNL’s Science at the Theater: Health Detectives upcoming lecture. Four LBNL scientists will explain how they are uncovering the mysteries of disease. This free public lecture will be held at the Berkeley Repertory Theater on April 23 at 7 pm.

Vincent van Gogh's Sunflowers (1889)

Most admirers of van Gogh's iconic "Sunflower" paintings gaze upon the golden inflorescences without any awareness of the scientific conundrum they pose. But researchers from the University of Georgia have finally cracked the case with a paper published in PLoS Genetics.

Buttercup taken at Tilden Park in Berkeley, California by Calibas

Orchid taken by Alex Tievsky SaveThePoint

The puzzle begins with the fact that all flowers are either radially or bilaterally symmetrical. A buttercup is an example of radial symmetry; it looks the same no matter how you rotate it. An orchid, on the other hand, has bilateral symmetry, like a human face–the left and right sides look the same, but you can tell whether it's right side up or upside down.

Here's the sneaky thing: a lot of seemingly radially symmetrical flowers are actually clusters of tiny bilaterally symmetrical flowers, or florets. In fact, this is true of one of the biggest flower families, Asteraceae, which includes such familiar friends as dandelions, daisies and, yes, sunflowers.

Then the sunflower makes things extra complicated by building its cluster out of two kinds of florets: bilaterally symmetrical ray florets, and radially symmetrical disk florets. This may sound confusing, but it's obvious as soon as you look for it: the classic sunflower is a ring of petals (ray florets) surrounding a big disk that will become filled with seeds (fertilized disk florets). The ray florets are infertile–they're just there to help attract pollinators.

Now at last we can consider van Gogh, and his double-flowered sunflowers. They're mutants.

A double-flowered mutant has no true disk florets, only concentric rings of ray florets–a profusion of petals. Consequently, the plant loses a lot of its fertility. You might wonder, can the opposite occur? Indeed, in tubular-rayed mutants ray florets are replaced with radialized, fertile disk florets.

Mark Chapman and his colleagues have just discovered that one particular gene, called HaCYC2c, causes both mutations. If HaCYC2c is over-expressed, it creates double-flowered van Goghs. If the gene's function is lost, however, you get tubular-rayed flowers.

I particularly love this study because at least the first part of their methods is totally accessible to anyone who's studied Mendelian crosses in high school biology. See:

Okay, maybe it's a bit tricky. If you want to puzzle it out but you're rusty on Mendel, here's a primer.

Of course, you don't need to understand Mendelian crosses–or the super-sophisticated genetic mapping that Chapman et al. use later–to appreciate van Gogh's art. Nor do you need to be an Impressionist fan to appreciate sunflower genetics.

But I think we can all appreciate that it's not often a famous painting is included in Figure 1 of a scientific paper.

Figure 1. Entire inflorescences (A, C, E) and individual florets (B, D, F) from wildtype (A, B), double-flowered (C, D) and tubular (E, F) sunflower individuals. Florets are arranged left to right from the inner florets to the outer florets. (G) “Sunflowers (Still Life: Vase with Fifteen Sunflowers)” by Vincent van Gogh (1888) with double-flowered heads pointed out with arrows. Panel G was obtained from Steve Dorrington on flickr (available at http://flic.kr/p/8SsPYb) and is distributed under the terms of the Creative Commons Attribution 2.0 Generic (CC BY 2.0) License.

Yeast like these are teaching us how simple evolution really is. Image courtesy of Masur, Wikimedia Commons

Unless, of course, the changes are easier than we imagine. For example, what if it is pretty easy to go from a single celled beast to a multi-cellular one? Or what if you can get increased complexity through easy to come by DNA changes? Then maybe it becomes easier to wrap your head around evolving complexity.

A couple of new studies in baker’s yeast are showing us just how easy it can be to build up complexity. Going from a single celled yeast into a multi-cellular one is pretty easy under the right conditions. And fairly common DNA changes can lead to increased complexity.

Taken together these two studies show us that increased complexity is easier to get than many people think. Certainly simpler than creating a 747 from a tornado in a junkyard!

I won’t have the space to deal with both studies in this blog. So I’ll talk about multicellularity in this one and then tackle the other, more complicated mutation example in my next entry.

How to Make a Multi-Cellular Yeast

In the first study, Ratcliff and coworkers used a very clever technique to end up with multi-cellular yeast. Basically they shook yeast in a big flask and only let those that were at the bottom reproduce.

After just two months, you had these beautiful beasts:

These aren’t just yeast cells stuck to each other either. That wouldn’t really be multi-cellular life.

No, as you can see in the video, these new creatures give birth to smaller multi-cellular instead of reverting back to single celled yeast. Being multicellular is now the default state of these yeast. This is true even when you stop the flask experiment and let them grow “naturally.”

Not only do they give rise to little versions of themselves, but they also have specialized cells within the snowflake cluster. For example, certain cells are willing to die so that the little juvenile snowflakes can separate from mom. No self-serving single cell would suicide like that unless it were part of a larger organism.

So the transition from one to many cells may be simpler than we thought. Which makes sense if current theories about life’s evolution are true. Scientists think multicellularity evolved dozens of times over the last few billion years.

Animals are doing surprisingly well in the radioactive areas around Chernobyl.

Imagine people’s worst fears are realized and the nuclear power plant at Diablo Canyon here in California has a Chernobyl-style meltdown. The effects on people are obvious: high rates of thyroid and other cancers, permanent resettlement elsewhere, increased rates of birth defects and so on. But as the area around Chernobyl is showing, the effects on the environment may be more subtle.

Over the break I watched a Nature special called, "Radioactive Wolves". This is a documentary about wildlife in a radioactive exclusion zone around Chernobyl.

Even though the area around Chernobyl is still so contaminated that humans can only go in for limited amounts of time, the wildlife appears to be doing surprisingly well. Birth defects are higher than in surrounding areas but life is thriving. Wolves are doing great, beavers have returned and everything looks hunky dory.

This seemed strange to me. I would think that so much radiation should be having pretty severe effects on these animals. And as noted in this in this NIH study, for certain individuals it definitely is.

The difference is in perspective. For the individual, the area around Chernobyl is terrible. Your kids have a higher rate of being stillborn or having birth defects, you have a much higher rate of developing various cancers, and so on. But for the species as a whole, things aren’t so bad. The higher background radiation appears to hardly be affecting their numbers at all.

Now this isn’t to say that the initial fallout wasn’t catastrophic to wildlife. It was. Untold numbers of animals died a terrible death in Chernobyl’s aftermath.

For the lucky survivors and new immigrants, though, Chernobyl is a different story. It is a chance to live a life without human interference. At least for now it looks like the high background radiation is preferable to man for these animals.

It is important that scientists keep studying this ecosystem though. The DNA of the animals in this area are under constant attack from the radiation. There may come a tipping point where the genetic burden becomes too high and populations start to crash. We’ll have to wait and see.

Additional Reading: NY Times Review of Radioactive Wolves

For now, sulfites are able to kill the yeast that might spoil this wine.

Wine sometimes tastes a bit funky because it was contaminated during fermentation with a yeast called Brettanomyces bruxellensis. This yeast can give wine a variety of interesting flavors like “…horse sweat, Band Aids, barnyard, and burnt plastic…”

Winemakers usually keep this from happening by killing off the yeast with those dreaded sulfites. But for awhile now, people in the know have been worrying about the emergence of a sulfite-resistant form of this yeast. And this is a well-founded fear.

Yeast, like bacteria, are fast growing microorganisms with lots of variation in their DNA. If you hit a population like this with something that kills them (like sulfites for B. bruxellensis or antibiotics for bacteria), some small percentage are probably going to be resistant. These resistant strains can then grow and replace the sensitive ones. The end result is sulfite-resistant yeast ruining our wines.

To try to head off this problem, a group of scientists in Australia has figured out this yeast’s DNA. The hope is that scientists will be able to use this data to determine how B. bruxellensis might evolve into a more resistant form.

Note that despite much trumpeting online, they haven’t really solved any problems with this knowledge yet. They have merely created the tool that might let them solve a potential future problem. And given how cheap and easy DNA sequencing is these days, it isn’t necessarily even an impressive feat of technological prowess.

Still, it may one day prove useful in allowing winemakers to more quickly defeat a sulfite-resistant strain. Which can only be a good thing for wine making.

I don’t want to end this before saying a nice word or two about B. bruxellensis. This yeast can spoil wines but it isn’t all bad.

For example, it gives Belgian beers their special taste. And some winemakers actively seek it to give their wine a bit of a “brett” taste.

Still, a sulfite-resistant form would definitely be a bad thing for most winemakers. So scientists should definitely stay vigilant and be ready to come up with quick solutions using this new tool (and whatever other ones they can find) when sulfite-resistant B. bruxellensis begin to appear.

Line up the blocks and help cure a disease!

Yesterday I discovered an online game called Phylo. No, it isn’t about Greek pastry dough. It has more to do with phylogenetics.

Phylogenetics is the study of how living things are related to each other. It takes advantage of the fact that DNA changes slowly over time. So the more distantly related two things are, the less DNA they will share.

You can also learn which bits of DNA are important for life (or good health) by seeing which ones stay pretty constant between various animals. Since these don’t change, they are probably being used for something. And if they get changed in people who have a disease, then they may be involved in that disease.

This is what Phylo is based on. In the game, you are trying to line up the DNA sequences of various animals to figure out which DNA is important and which isn’t. This is a lot harder than it sounds.

While DNA changes slowly, we’re dealing with some pretty long spans of time since two animals shared a common ancestor (i.e. had the same DNA). This means that the DNA can start to look pretty different.

Phylo gives you sets of DNA sequences that haven’t been lined up yet and has you try to line them up manually. Apparently people are still better at this than computers!

Lining these sequences up helps scientists discover which DNA bases are important. It can also give us some basic knowledge about the evolutionary relationship of various animals.

Since DNA sequence might be a little intimidating, Phylo uses colored blocks instead. Each different colored block represents one of the four DNA bases.

In the game, you have a certain amount of time to align the sequences as best you can. You want the fewest mismatches with the fewest gaps.

The game is pretty hard but it is engaging. And you’re doing your bit for science which is always a good thing.

What would make it even better is more explanation about what I am doing and why it is important. This could make the game fun and educational. (I’d also like to see the DNA sequences I aligned but that’s probably just me.)

So give it a whirl and do your bit to help humanity. And have a little fun in the process.

Nice list of top 5 citizen scientist games (Phylo is number 1).

By changing how a worm like this uses its genes, scientists have added the equivalent of 20 human years to their lifespans. And to their children and grandchildren's lifespans too.

One of the workhorses of longevity studies is a little flatworm called C. elegans. Lots of scientists have come up with lots of ways to get these little guys to live longer than usual. With enough tinkering, they can live 4 or 5 times as long as their wild brethren. That's like a human living to be 500!

Work out of the Brunet lab at Stanford added another chapter to this story. They tinkered with how a worm uses its genes and found these worms lived around 25% longer than normal.

That’s old news though. What is exciting is the new study that shows that the worms passed this trait down to their kids and their grandkids.

These worms inherited the pattern of how their genes were used from the original parent. And now the kids and grandkids live longer than worms that have the exact same set of genes. All because of how an ancestor used its genes.

The effect is not forever though. The great grandkids live about as long as worms with the same set of genes. Apparently gene usage patterns reset to normal after three generations or so.

As you’ve probably gathered, how our genes are used is as important as the genes we have. For example, every cell has the same set of genes. And yet, a blood cell is wildly different than a muscle cell. The difference comes from how each cell type uses the genes it has.

A big part of how a gene gets used is determined by various chemical marks attached to either nearby DNA or to the spools around which the DNA is wrapped. In this case it has to do with chemical marks on one of the worm’s spools or histones.

Most of the marks on DNA and histones are wiped clean once they are packed into a sperm or egg. This allows the fertilized egg to develop into all the different cell types needed for an adult worm (or person).

The specific changes made in this study linger for awhile longer and take three or so generations to wipe out. We don’t exactly know why yet but we do know that this is what we often see with epigenetic changes like this.

Epigenetics is a fancy way of saying something is passed down without a change in the gene itself. As scientists delve deeper and deeper into genetics, they are finding many traits that behave this way. Even in people.

So is this going to make us live longer? Not yet.

But it does suggest that if we can one day mimic these effects with a chemical, we may end up living longer. And so will our kids even if they don’t pop the same pill.

Cynthia Kenyon talks about one particular long lived mutant.