Life expectancy nearly doubled in developed countries over the 20th century. Prior to 1950 the increase was due mostly to a decrease in infant mortality. After the 1950s it was a decline in old age mortality that provided the main life-prolonging force. Improvements to the social and physical environments and breakthroughs in healthcare underlie both phases of mortality decrease. A person born in the US at the turn of the 20th century could expect to live 49.2 years. Their ancestor born in 2003 could reasonably expect to see their 77th birthday. But while average lifespans continue to lengthen, the oldest of the old appear to be encountering a rather powerful limiting factor. As reported recently in Slate, the number of oldest supercentenarians – people 110 and older – has stayed at around 80 over the past few years. And the age at which they die hasn’t changed over the past few decades. Data from japan is used to illustrate this. In 1990 there were 3,000 people 100 or older, the oldest of them being 114. Twenty years later the number of people aged 100 and over had grown to around 44,000, but the oldest was still 114. Robert Young, a gerontologist working for the Guinness Book of World Records, estimates that “the odds of a person dying in any given year between the ages of 110 and 113 appear to be about one in two. But by age 114, the chances jump to more like two in three.”

Number of people living to 110 years or older in Switzerland.

This phenomenon of everyone getting older but the oldest dying at the about the same age is called “rectangularization of the mortality curve.” A mortality curve tracks the probability that a person will be alive at a certain age. At birth the value is 100%. By year one it begins to slope downward, and around 70s, 80s, 90s it drops at a faster rate. In decades past the curve looked like a ski slope, hitting zero around 114. But the fact that more people are living longer lifts the curve and pushes it to the right so it looks more like a cliff than a ski slope – and more like a rectangle.

During our last Google+ Hangout we got a chance to hang with longevity researcher Aubrey de Grey, author of “Ending Aging” who once proclaimed “the first person to live to 1,000 was probably born by 1945.” We asked him about rectangularization, why it was that the whole ski slope doesn’t just move to the right but instead comes crashing down at around age 114. “This is a fascinating phenomenon and nobody has really much idea of what’s going on. What we do know is that it’s absolutely essential to not jump to conclusions about what’s going on. Time and time again over the decades past demographers have been brutally misled by short-term phenomena, by statistics gathered only over a few years. Blips happen for all manner of impenetrable reasons. In this case we’re talking about people born in a small segment of time, around 1900, and most of them born in particular countries and going through certain types of life they might not have gone through had they been born 20 years previously or 20 years later. There are many factors called ‘cohort effects’ that can cause early life phenomena to have an influence on longevity.”

Bottom line: don’t believe the hype.

“At this point I’m not exactly losing sleep over the phenomenon you’re talking about. I think that we’re probably going to see a resumption of the trend of everything just moving to the right eventually.”

Rectangularization of mortality.

De Grey also adds that medical developments could make rectangularization, much speculated upon in the study of aging, a moot discussion. “I don’t really care about whether I’m right or not about what I just said because it’s all going to become completely irrelevant when we have therapies that repair the damage of aging. Those therapies are going to make the whole concept of life expectancy…the way it’s calculated today will no longer exist.”

Because therapies will make life expectancies of the future, de Grey argues, so much longer than they are today, today’s estimates will become irrelevant. Like trying to compare Mark McGuire, Sammy Sosa and Barry Bonds to Babe Ruth. After steroids, all bets are off.

But that’s not stopping researchers from looking for the “longevity gene.” Sampling the genetic material of centenarians, researchers have seen strong correlations with several genes and the likelihood of living to 100. Two of them are involved in fat metabolism, a third in calcium metabolism.

Despite de Grey’s skepticism, if there really is a genetically-programmed limit around 114, seems to me that would make it all the more imperative to make good on the Longevity Dividend, the range of benefits to both individuals and society were we to stay healthier longer. Some argue that extending life will only postpone the inevitable disease and frailty that comes with old age. But if there were an upper bound that is not affected by current longevity trends, then the longest lifespans will not get longer but the period of age-related disease and frailty would be shortened.

Just my two cents for what they’re worth. Regardless of whether or not the upper bound is real, I agree with de Grey. Biologically, it is a fascinating phenomenon.

[image credits: dailygalaxy.com via Urban Times, Journal of Epidemiology & Community Health, and Archives of Gerentology & Geriatrics]
image 1: old age
image 2: longevity
image 3: rectangularization

Yoshiki Sasai's group induced embryonic stem cells to grow into the three-dimensionally complex pituitary.

Chalk up another part of the body that can be grown from stem cells – at least in mice. Scientists in Japan have induced mouse embryonic stem cells to form a pituitary in the lab. The team’s success not only holds promise for people with pituitary defects, but it’s another of just a few examples in which stem cells have been coaxed into a complex, three dimensional structure.

The pituitary is often referred to as the “master” gland of the endocrine system, as it secretes hormones that control the hormones secreted by other endocrine glands. It’s formation during embryonic development is closely tied to the hypothalamus, another important endocrine structure located close by at the base of the brain. The challenge to the Japanese researchers was to maintain the proper developmental environment for the nascent pituitary outside the brain. It was a complex challenge. They added hypothalamus tissue to the mix so that the stem cells would receive guiding signals from it, but the brain obviously consists of much more than these two areas. In order to “trick” the pituitary into thinking it was still in the brain, they also added several growth-inducing molecules including Sonic Hedgehog (it was discovered by a graduate student back when the game was popular) that are required for proper development in the brain. It worked. In just a couple weeks, the new structure not only looked like a pituitary, it produced all the major molecular markers that a normal pituitary produces. For the icing on the cake, the researchers placed the lab-grown pituitary into mice which had their own pituitaries destroyed. A week after transplantation, the mice showed increased blood levels of hormones produced by the pituitary.

The study was led by Yoshiki Sasai at the Riken Institute in Japan and was published in Nature.

The same researchers had previously used embryonic stem cells to from layers of the brain cortex.

The pituitary hormones regulate important physiological processes such as growth, puberty, and reproduction. A dysfunctional pituitary can lead to deficits in these hormones, which in turn can lead to disorders such as stunted growth, hypothyroidism, and infertility. Combined pituitary hormone deficiency affects about 1 in 3,000 to 4,000 live births. And although hormone-replacement therapy can be effective for many, it does not return hormone levels to their normal levels.

But don’t expect doctors to start replacing dysfunctional pituitaries with ones grown in the lab anytime soon. Most importantly, we’ll first need to see if the procedure can be repeated with human embryonic stem cells. And even now questions remain regarding the lab-grown mouse pituitaries. Embryonic stem cells differentiated in the lab don’t always run the course to full maturity. The pituitary in the current study still looks like the pituitary from an embryo. Only time will tell whether or not it can develop to an adult pituitary. But even if we can’t start swapping out dysfunctional pituitaries, the lab-grown structures should be a valuable tool for researchers studying pituitary defects. They can be looked at under a microscope and labeled with molecular markers, lending themselves to analyses far more difficult to perform in whole animals.

It could also pave the way for future stem cell research. The pituitary’s development in the study, as it is in normal development, was dependent on molecular signals sent from the hypothalamus. Now that the researchers have established that it is possible to run this inductive process, they have opened the door for the induction of even more complex organs in the lab.

The victories for regenerative medicine just keep on coming. Last year a group at UC Irvine in California successfully grew a retina from human embryonic stem cells. Like the pituitary, the retina needed to be grown with a specific three-dimensional organization in order to function properly. The starting clumps of stem cells required no intervention from the scientists to become retinas. The cell-to-cell communication during “development” was all they needed. Sasai’s group also induced embryonic stem cells to form a retina this past spring, although theirs was grown from mouse cells.

The group is currently working to improve transplantation of pituitary tissue in their mouse model. They’re also working on methods to repeat the experiment with human stem cells. Sasai expects it will be about three years to create a human pituitary. When they do, the field of regenerative medicine will be that much closer to its holy grail of growing all organs in the lab and transplanting them into humans.

[image credits: RIKEN]
image 1: cortex
image 2: Sasai

An Electric Sheep created by Phssthpok

“Dr. Chandra, will I dream?”
“I don’t know.”

If Hal was somehow able to dream after Dr. Chandra shut it down in Arthur C. Clarke’s “2010,” the results may have looked something like Electric Sheep. Not at all resembling the sheep we might conjure as we fall asleep, these beautiful digital images spring to life when a computer drifts off to its screensaver slumber. But they’re more than just a randomized graphics program. Each Electric Sheep is the collective creation of thousands of computers around the world. The Electric Sheep shown in the pictures and videos below are a striking artistic demonstration of the creative powers of the Internet and open source sharing. By voting on sheep they find visually appealing, humans cause these virtual creatures to compete, procreate and evolve. What strange and beautiful results may arise as these visual creations become ever better examples of digital lifeforms?

The sheep are the creation of Scott Draves, and they’re the conceptual descendants of one of Draves’ earlier creations called Flames. While getting his PhD in computer science at Carnegie Melon he wrote an graphics algorithm to create fractal patterns by treating each pixel as a variable and by introducing parameters that numbered in the thousands. The original flames were created using plugins for Adobe Photoshop and AfterEffects. He published the algorithm online and, well, the Flames spread. Today the Flame algorithm is installed on millions of computers worldwide.

created by user cqfd93

But the Flames were static and Draves wanted to animate them. Inspired by the then new SETI@home project that used the Internet to tap the computing power of millions of idle home computers, Draves utilized a similar approach to be able to carry out the demanding computations required to animate his Flames.

As the number of people who downloaded the free screensaver grew, so did the number of frames able to be rendered on Draves’ server. The iterative algorithm was able to take the newly added pixel data and rearrange them in new ways, with new colors and motions. And that’s how Electronic Sheep were born.

created by user Bananablob

A GUI app called Apophysis that makes it easier for users to create sheep replaced the original Flame plugins. Now thousands of sheep are being created by thousands of people with an artistic bent. The current version gives each sheep about ten seconds of life before another sheep begins swirling in its place. If you think of the sheep as virtual lifeforms – as Draves does – you can – also like Draves – think of each sheep’s algorithm as its genetic code. As he wrote for the computer graphics interest group SIGGRAPH, new animations “have a chance to contribute their genes to the reproductive system of the Electric Sheep, which was already based on Darwinian evolution with mutation and crossover.”

created by user brood

Here’s the first of several videos to give you a look at what these beautiful “lifeforms” look like.

But every good Darwinian system needs a ‘Survival of The Fittest’ force. That’s where the users come in. As users watch the sheep morph on their computer screen they can vote for the ones they like and vote down the ones they don’t. In this example of art imitating life, the fittest sheep – that is, the most popular ones – get to “mate” and contribute their virtual DNA to produce unique offspring. Draves calls it a “process of death and rebirth.” You can download the sheep program here to try your hand at some artificial natural selection.

Electric Sheep are not merely a very ingenious and labor-intensive way to make pretty screensavers, they’re part of Draves’ open source philosophy. “I think we as a society should support free information to a much greater extent than we do,” he said in an interview with The Creators Project. “My artwork is a manifestation of this philosophy.”

Creative Commons, the online affiliate that provides licensing to open share software including Electric Sheep, share Draves’ philosophy. They believe that only through universal access to research, education, and culture can the full power of the Internet be realized. They envision a day when the Internet is used to harness the creative input of thousands or millions of users to not only make better sheep, but improve our lives through an evolutionary honing of knowledge.

But just as the products of biological evolution aren’t always the most straightforward solutions (see David Linden’s “The Accidental Mind” for a healthy dose of humility about our brains’ hodgepodge architecture), Draves is often not a big fan of what his sheep-creating collective breathes life into. What he refers to as the “Las Vegas Effect” lead people to choose sheep that are flashy, have lots of bright colors and move fast. This prompted Draves to sift through the sheep himself and pull out the ones he finds aesthetically pleasing. He’s compiled them into limited edition videos that he sells to help fund the not-yet-profitable Electric Sheep.

The fact that sheep evolve on their own means Draves has satisfied his own impetus to explore whether or not computers can create something new, something greater than the sum of their input. “Genetic code,” “evolution,” “creation”…sounds like Draves just might have a touch of God complex. “That’s right,” he told The Creators Project when asked if he was playing God. “I’ve created a universe and the rules for this universe and then inside it sort of has a population that lives there.” He follows that by admitting he wanted to create virtual life, and that “it’s not clear how successful one can be. …Can computers think? Can a computer be creative?”

And then he starts to get metaphysical for real.

“…is humanity the only possible vehicle for what really amounts to a soul? …some people believe that all material matter follows the rules of physics, and if you can figure out what physics is, a computer can follow the rules and therefore you can simulate life in a computer. So it really becomes a profound question that we are, as a society, really just starting to struggle with.”

Right now, the Electric Sheep are still just screensavers, albeit evolving and unpredictable screensavers. But what limits are there to the kind of reiterative, informatics cross-breeding underlying their generation? What might be produced when the same process is applied to more functional outputs? What if we woke up one morning and our computers were doing things that we didn’t fully understand.

And so on.

How that might play out is as unpredictable as the sheep. Until it happens, I suppose we’re left merely to wonder…and perhaps dream.

[image credits: created by Brother Lewis, Phssthpok, Bananablob, brood, and cqfd93 at electricsheep.org]
[video credits: electric sheep.org via YouTube]
images: electric sheep
video 1: electric sheep 1
video 2: electric sheep 2
video 3: electric sheep 3

As published in the Public Library of Science One, the cows were cloned by somatic cell nuclear transfer–the same method Ian Wilmut used to clone Dolly–in which the nucleus of a somatic (body) cell is transferred into an egg that has had its nucleus removed. Prior to inserting the somatic nucleus into the enucleated egg, Dr. Li’s group infected it with a virus carrying the human gene for lysozyme. Lysozyme is an enzyme found in large quantities in human breast milk that can lyse–or split open–the cell walls of harmful bacteria in the gut. In addition to its antibacterial effects, lysozyme works to boost the body’s immune response to infection. The immunological benefits imparted by lysozyme is an important reason why breast milk is so healthy for developing babies. It’s absent in most of the baby formulas commonly-used to supplement or substitute breast milk. Compared with the 200-400 μg/ml concentration found in human breast milk, cow milk normally contains lysozyme at 0.05-0.22 μg/ml. The milk from Dr. Li’s cows increased their lysozyme concentrations more than a hundred-fold to 26 μg/ml.

In addition to lysozyme, Dr. Li’s group successfully cloned cows that produce milk containing two additional human proteins, called lactoferrin and alpha-lactalbumin, both of which strengthen the baby’s immune system. Like lysozyme, these proteins are found in normal cow milk but increasing their amounts will provide a further boost to the baby’s immune system. On top of that the cow benefits as well. The human genes will help protect the cow’s udder from infection, possibly decreasing the need for feeding the animals antibiotics, a major contributor to the harrowing pattern of antibiotic resistance that is spreading worldwide.

In addition, the SKYLAB scientists made the cow milk more human-like in other ways. They increased milk fat content by about 20 percent and changed the concentrations of certain milk solids to more closely match the concentrations found in human milk. The group victoriously proclaimed that their demonstration proves that it is possible to “humanize” cow milk.

So when should we expect to find the cow-produced, human-like milk to show up at a supermarket near us? Dr. Li, collaborating with a Beijing company called GenProtein Biotechnology Company, projects it’ll take about ten years until the milk is made available to the public. How palatable consumers find the humanized milk is anybody’s guess at this point. The ChinaDaily reported that Dr. Li describes the milk as having a “stronger” taste than normal milk–but he didn’t specify normal cow milk or normal breast milk. We’ll just have to wait and see, I guess.

The researchers claim that the Holstein cows used in the study produced "humanized" milk but were otherwise normal.

Probably more important than taste, much of the West is still weary–if not outright hostile–to the idea of GM foods. Countries of the European Union are especially incensed–61% of Europeans are opposed to GM foods. As suggested by the Wall Street Journal, however, the increasing costs of non-GM food imports may cause Europeans to soon change their tune. Obviously, GM foods don’t face the same opposition in China which is moving rapidly towards the commercial use of genetically-modified foods such as rice and corn. The fact that they were able to combine two techniques that are still controversial in the West–GM and animal cloning–with the public visibility that they did is indicative of a country that is ready to take the lead on this area of research, and reap all of the economic benefits that will inevitably follow. Dr. Li alludes to this in the paper, saying, “Thus, our study not only describes transgenic cattle whose milk offers the similar nutritional benefits as human milk but also reports techniques that could be further refined for production of active human lysozyme on a large scale.”

If you’re totally grossed out by the idea of putting human genes into animals you should know that Dr. Li’s cows are far from the first organisms to have human genes and produce human proteins. Since the 1980s we’ve cultured the bacteria, E. coli, carrying the human gene for insulin (a protein). The easily mass-produced and purified insulin from bacteria was an improvement over the purified cow insulin we’d been using up until then to treat persons suffering from type 2 diabetes. Soon thereafter we enlisted the help of E. coli to produce human growth hormone. But bacterial systems are limited in their capacity to correctly fold and modify the comparably complex proteins produced by mammals. Because of this mammalian systems have been developed to express our mammalian genes. Human antithrombin III has been grown up and purified from farm animal milk; the human antibody interferon β1a has been grown up in chicken egg whites. To date the list of human genes grown in these and other mammalian systems includes monoclonal antibodies, blood factors, hormones, growth factors, cytokines, enzymes, milk proteins, collagen and fibrinogen.

Add to the list human lysozyme produced in cow milk. If we’re already growing it in cow milk, and we’re already drinking cow milk, is it really that much of a stretch to combine the two in a nice, refreshing, frothy cup?

Or baby bottle?

Dr. Li and his research team hopes that their readily available “humanized” milk will provide an option for women who can’t breast feed or don’t want to breast feed their child. Unfortunately it seems that fear-mongering, anti-GM groups often cause the benefits of GM to get overlooked, this despite the majority of American citizens eating GM foods without even realizing it. My guess is that when GM foods bear fruit that is both more plentiful and more resilient, sticking with non-GM foods will become an increasingly pricey proposition. The EU is finding this out right now. Will they relax their restrictions on GM, or will China be left to move easily forward into this as-yet unexploited market?

[image credits: supercoolbaby.com, Maksim via wikicommons]

image 1: supercoolbaby
image 2: wikicommons

Coming To A Town Near You - Antibiotic Resistant Bacteria

Be afraid. Be very afraid.

The widespread misuse of antibiotics is rapidly rendering them powerless against infection. Common infections that are easily cured today are going to become deadly, and it’s going to happen sooner than you think.

The World Health Organization reports:

* Each year there are about 440,000 new cases of multi-drug resistant tuberculosis, resulting in at least 150,000 deaths.

* Resistance to antimalarial drugs chloroquine and sulfadoxine-pyrimethamine is now widespread in most malaria-endemic countries, leading to the resurgence of malaria in areas where the disease had previously been eradicated.

* A large proportion of infections contracted in hospitals are caused by highly resistant bacteria such as methicillin-resistant Staphylococcus aureus (MRSA).

* An estimated 25,000 patients in the European Union die each year from drug-resistant infections.

Antibiotic resistance arises when the treatment doesn’t kill off the entire infectious population. The genetic variability of bacteria will inevitably make a small portion of them resistant to the antibiotic. These surviving members are then left to proliferate and spread their antibiotic-resistant genes. It’s a consequence that today’s caregivers are aware of, but the rate at which antibiotic resistance is spreading indicates that we’re not doing enough to slow the process.

And maybe we should. Like, now.

Dr. Margaret Chan, the Director-General of WHO, paints a dire picture of our immediate future. Speaking on April 7th in a World Health Day 2011 podcast, Dr. Chan says, “The message on this World Health Day is loud and clear. The world is on the brink of losing these miracle cures. In the absence of urgent corrective and protective actions, the world is heading towards a post-antibiotic era, in which many common infections will no longer have a cure and, once again, kill unabated.” Watch the video and sit in the hot seat and receive a tongue-lashing on behalf of the entire developed world from Dr. Chan.

So why are antibiotics losing their effectiveness? There are a number of factors, but most of the blame lies with doctors who overprescribe the drugs. They’ll prescribe antibiotics that kill many types of bacteria even when the patient is at risk for a specific infection. Unbelievably, doctors will give antibiotics to patients to “cure” conditions that are caused by viruses, such as the common cold. Why? Because the patient insists. Sometimes the doctor is simply too busy to explain to the patient why he or she doesn’t need antibiotics, but in the interest of time prescribes it anyway. Are you kidding me?! Documentation of these incredible findings can be found here.

She looks all smiles but WHO Director-General Margaret Chan means business as she sounds the alarm to halt practices that promote the spread of antimicrobial resistance.

But doctors don’t shoulder all of the blame. Patients who don’t stick with the prescribed regimen and complete the treatment properly end up feeling better, but they also allow some of the infectious microbes to remain alive, thus allowing them and their resistant genes to proliferate.

Another huge problem is the use of antibiotics in food-producing animals. Approximately half of all antibiotic production is used in agriculture. Factory farms routinely give perfectly healthy animals antimicrobials both as a precaution against infection and to promote growth. This leads to resistant bacteria, which can then spread to humans through the consumption of meat, direct animal contact, or via environmental spread through, for example, contaminated water. This is confounded with veterinarians in many countries whose income is largely derived from the drugs they sell. Try convincing one of these vets to take a hit in the pocketbook for the good of humanity.

For World Health Day 2011 the WHO issued a six point Policy Package To Combat Antimicrobial Resistance. A common component among the points was multi-agency cooperation: governments, doctors, patients, and communities worldwide all working together to stem the antimicrobial tide. A lovely thought, but it’s not going to happen.

Antimicrobial resistance is inevitable. What’s not inevitable is, as Dr. Chan says, “losing these miracle cures.” But, in my opinion, the WHO package is simply unfeasible. At the risk of drawing Dr. Chan’s ire, I think our best hope in fighting antibacterial resistance lies, not in the well-wishing hands of policymakers, but in the self-serving craft of economic forces. Think of MRSA as a new disease, which it is. Drug companies will want to cash in. Each time a new strain crops up that’s resistant to current antibiotics drug companies will scramble over each other to get their pills out first. Speaking of MRSA, Merck and Intercell are well underway testing a vaccine for it in humans. See what I mean?

Granted, a microbial arms race is not the ideal approach. It would be great if governments and doctors and patients would all work together in perfect harmony with the world’s long-term health in mind.

World peace would be great too.

Our misprescribing of pneumonia taught us a lesson: too many antibiotics = high resistance. It seems that we, unlike the microbes, are slow to learn. (From Albrich et al., 2004)

The business of making antibiotics, however, is a lot tougher than it used to be. Since the use of antibiotics became widespread in the 1940s the field has seen a single Golden Age come and go. In the 1950s and 60s developers produced multiple classes of antibiotics to battle diverse types of infections. In the 1980s and 90s they were able only to make improvements within a class. The trend begs the question of whether or not we’ve hit the wall on this one. Have we tapped our antibiotic reserves for every last drop? If so, shouldn’t we be witnessing, now, the emergence of a resistant time bomb? The SARS virus and the avian flu outbreaks had us staring nervously at news flashes as the boundaries of the impending epidemic crept from the far reaches of the globe toward our homes.

And then they were gone.

Vaccines were developed for both, but the SARS epidemic was contained years before its vaccine ever saw the light of day. A rapid, global public health response is what stopped it. Apparenlty, we can work together–when we’re scared to death.

I use SARS as an example of how changing habits in the absence of a cure-all antibiotic or vaccine can be extremely effective–within 6 weeks of its discovery the SARS virus had infected thousands of people on 6 continents–and may be the reason we have yet to see a resistant time bomb. The fear of H1N1 put hand sanitizer dispensers in the lobbies of hospitals, schools, and offices. And they’re there to stay. Now you feel guilty walking by one and not squirting your hands sterile. Everyone’s doing it. Everyone’s hands are sterile.

Simple improvements in hygiene practices are probably behind the recent decrease in MRSA infections contracted in hospitals (healthcare-associated MRSA). Likewise, the continued problem of MRSA infections that begin outside of hospitals (community-associated MRSA) is probably due to a lack of change.

It’s hard to predict the future. The world’s a pretty crowded place. Will the resistance time bomb explode and a microbial butterfly in China cause a pandemic hurricane in Canada? Has the bomb already gone off but we just don’t know it? Or are we kind of like the little buggers themselves, modifying our strategy to counter theirs using a mixture of greed and panic instead of genetic variation?

I’m going to try and not worry about it too much. I’ll take all of my pills (only the necessary ones!), sanitize my hands, and drink my orange juice. And I’ll try to shut out Dr. Chan’s shrill voice, over and over again heeding her warning “…once again, kill unabated…once again, kill unabated.”

[image credits: Fabio Pozzebom, Agencia Brasil via wikicommons; CDC]

image 1: wikicommons_chan

image 2: CDC_emerging_infectious_diseases

video: Dr. Chan

Scientists were finally able to grow sperm in the lab–from mice. It still remains to be shown if the same procedure can be used for humans.

In an amazing technical feat researchers in Japan have accomplished something that has stymied the field for the past half century: they successfully grew sperm in the lab. They then used the sperm to impregnate female mice and produce a healthy litter. The breakthrough holds promise for millions of men worldwide with infertility.

Published recently in Nature, the work was pioneered by Takehiko Ogawa and colleagues at Yokohama City University. The procedure involves taking biopsies of mouse testes, breaking them up into 1 to 3 mm pieces, placing them on agarose that has been partially soaked with a special medium, and letting them be for two months. If all goes according to plan, the chemicals in the medium would induce the gonadal stem cells to differentiate into mature sperm. Getting the ingredients of that medium right has been the major confound since efforts to produce sperm in the lab began in the 1960s.

To make their lives easier they used mice genetically modified with green fluorescence protein (GFP) that would only become activated in cells that had differentiated into viable sperm. The researchers could then just look through the microscope and all of the stem cells that had successfully differentiated to sperm would glow green.

Imagine, after years of frustration, peering into the microscope and seeing a lovely field of glowing green. But Ogawa and his crew didn’t pop the champagne just yet. The ultimate proof was then to see if their homegrown sperm was healthy and functional–could they be used to successfully fertilize an egg and produce normal, healthy offspring. Using two different methods they fertilized 23 and 35 oocytes, respectively. The dams gave birth to 7 and 5 live offspring who survived to adulthood and were able to produce offspring of their own.

Now it’s time to break out the champagne.

Sperm is often stored frozen in sperm banks for future use. To simulate this scenario Ogawa’s team cryopreserved the sperm in liquid nitrogen for 4 to 25 days. When the cells were thawed and cultured, expression of the GFP marker confirmed that they resumed full spermatogenesis in culture. They have yet to demonstrate that the freeze-thaw cycle leaves their cultured sperm intact well enough to produce healthy offspring that are in turn able to produce healthy offspring. It remains possible that freezing and thawing the cells left some as yet undetected structural damage, for example, or caused some epigenetic changes–changes in the molecules bound to genetic material that affects gene expression. Nevertheless, their demonstration is already an amazing accomplishment.

After 50 years of effort researchers have finally discovered a way to grow sperm in the lab. Swapping out a commonly-used culture ingredient may be the key.

Given the increasing number of successfully grown tissues and stem cell acrobatics flipping one type of cell to another, we may be getting the impression that simply growing sperm in a dish isn’t all that groundbreaking. You half expect those rambunctious miniature tadpoles to will themselves alive on their own. But growing sperm–a gamete–is much more complicated than growing somatic—rest of the body–cells. It is a sequential, multistep process involving a complex list of players named primordial germ cell, spermatogonium, primary spermatocyte, secondary spermatocyte, spermatid and mature sperm. Each of these stages requires an equally complex battery of signals provided by the non-germ cells that surround them. The whole process of going from stem to sperm cell takes over 60 days in humans; in mice (and most other mammals) it takes over a month. Successfully commanding a month long differentiation is a daunting challenge. We’d been trying since the 1960s but, until Ogawa’s study, we’d failed every time.

Through much trial and error, the researchers happened upon a key modification to their protocol that seemed to make all the difference. When trying to grow sperm in a dish a researcher would typically use fetal bovine serum (FBS), serum from the blood of newborn calves. FBS is a widely-used supplement in mammalian tissue cultures. Ogawa’s group had had some success with FBS in the past, but they decided to try replacing their FBS with what’s called knockout serum replacement (Ko-SR). This was a strange move, as Ko-SR is essentially FBS with most of the ingredients that promote the differentiation of cells removed. It’s typically used by stem cell researchers who want their stem cells to remain in an undifferentiated state. Surprisingly, and to the delight of Ogawa and colleagues, the Ko-SR had just the opposite effect: it promoted the differentiation of the sperm stem cells into mature sperm. It’s still unclear why the Ko-SR worked, but Ogawa suspects it’s due to one of the differentiation-inducing ingredients that still remains, called parvalbumin. If this turns out to be true, not only would the study give us a new tool to treat infertility, it will teach us something new about the basic biology of sperm maturation.

The team’s 12 newborn mice mark the triumph of a half a century’s effort. When you’re dealing with biological complexity slow and incremental is not only the pace of progress, it’s safer. It would be a tragedy if we were to give a man who had already conquered cancer the hope of having children, only to hand him the devastation of an unhealthy child. Taking genes into our own hands is risky business (let’s not forget that Dolly had progressive lung disease) and it remains to be seen whether or not the strategy can be used to make human sperm and to make human beings. Nevertheless, the team’s 12 newborn mice are a testament to power of relentless tinkering. And as they continue to tinker I don’t expect we will have to wait another sixty years to hear their good news.

[image credit: Rama and pdimages.com/web9 via wikicommons]

image 1: wikicommons_mouse

image 2: wikicommons_sperm

Dr. Michael G. Kaplitt, co-founder of Neurologix, helped design the study in which the GAD gene was infused directly into the brains of Parkinson's disease patients.

Imagine that for years you’ve lived with a body that is in constant, uncontrollable, motion: your hands tremble, your head bobs, you walk in short, shuffling steps, and muscles all over your body spasm so strongly it’s painful. Everyday activities, like going to the bathroom, are incredibly difficult and time consuming. You may look at others who walk, talk, and go to the bathroom with an ease that is impossible for you to imagine and think, that’s “them,” not me.

This description is close to what life is like for many of the estimated 7 to 10 million people worldwide who have Parkinson’s disease. But a recent study may offer hope that a change for the better could very well arrive in their near future. Neurologix, Inc., a biotech company based in Fort Lee, New Jersey, successfully used cutting edge gene therapy to improve tremors, rigidity and other motor control problems in a group of Parkinson’s disease patients, and with minimal side effects. If the therapy makes it to the market, it will be the first of its kind to do so.

For the uninitiated, gene therapy is a procedure in which a group of cells within the body are upgraded with one or more new genes to modify their behavior. The targeted cells are literally reprogrammed to exhibit new behaviors and abilities. A modified virus, called a vector, is used to safely infiltrate the targeted cells and inject them with the new genes. The viruses are mutated in a way to prevent their replicating out of control or causing any harm to the patient.

In the treatment being tested by Neurologix, gene therapy is used to modify neurons within the substantia nigra, an area in the back of the brain responsible for motor control. The gene therapy is aimed at increasing the expression of the GAD gene within the targeted neurons. GAD is used within neurons to make GABA, an inhibitory neurotransmitter which is decreased in the Parkinsonian substantia nigra, resulting in overactive neuron firing. In theory, with increased GAD production, GABA production will also increase, putting a “break” on the overactive neurons in Parkinson’s patients.

Injecting a virus into a specific part of the brain is no small feat. To inject the GAD gene, the experimental half of the patients (22) had, under local anesthesia, holes drilled into their skulls into which catheters were inserted. They then went to a recovery room where the GAD gene was injected into the catheter. The remaining control patients (23) underwent sham “surgeries” in which researchers imitated the sounds of the procedure and pretended to drill a hole in the skull and insert a catheter. Six months following surgery, the motor skills of both patient groups were assessed. The placebo group showed an improvement of 11 percent. The group that received GAD improved their motor scores by 23 percent with all but two showing improvement.

The current study was the second trial that Neurologix has subjected their GAD gene therapy to. Phase 1 took place in 2007 and included 22 patients. The study showed similar improvements but did not include a control group. Now that the therapy has passed phase 2 Neurologix is trying to raise the $30 to $40 million it will cost to conduct a phase 3 trial which requires 1,000 to 3,000 test subjects. After phase 3 is completed it is then up to the FDA to decide whether or not the therapy is safe enough to be sold.

The millions of people across the world with Parkinson’s disease and their families will be watching closely.

Neurons in the substantia nigra become impaired or die in people with Parkinson's disease. A new gene therapy, aimed at returning function to substantia nigra neurons, shows promising results.

If it does make it to market, Neurologix’s GAD gene will be a significant advancement in the treatment of Parkinson’s disease. At present the best treatment for Parkinson’s disease is levodopa. You might recall levodopa—also called L-dopa—bringing comatose patients to life in the movie “Awakenings” based on the book by Oliver Sacks. Like the GAD gene therapy, giving patients levodopa does not help prevent or reverse the actual disease but improves quality of life by alleviating symptoms. Unfortunately, as we saw in “Awakenings,” levodopa must be taken continually and its effectiveness typically wears off over time.

The Promise of Gene Therapy

To give cells back that which they have lost—as in returning Parkinsonian substantia nigra neurons to normal levels of dopamine or GABA production—is the ultimate hope in gene therapy. You permanently correct the cell itself rather than continually feeding it what it lacks. Making gene therapy ever more attractive is our increasing ability to identify the abnormal genes associated with inherited disease like Alzheimer’s disease, manic-depression, intestinal cancer, heart disease, diabetes, and many others.

But the road for gene therapy has been a rocky one. A four-year old girl with a severe combined immunodeficiency (SCID)—known to most as “bubble boy disease” from the famous story of David Vetter who was forced to live in a sterile bubble—became the first person to receive gene therapy in 1990. She got better, but whether or not it was due to the therapy or something else has since been a matter of debate. Much effort since then has been aimed at developing safe and effective gene therapies, but the fact is no gene therapy has yet been approved by the FDA.

Gene therapy is further complicated by the fact that diseases are, well, complicated. The advances in genetics that so enticingly point us to potential targets for gene therapy also reveal that many common disorders such as arthritis, diabetes, Alzheimer’s disease, and heart disease are brought on by the combined effects of multiple genes.

But a string of successes in the past few years clearly shows that gene therapy is marching forward:  in 2008, vision was restored to patients with blindness brought on by a defective gene; in 2009, 8 of 10 patients were cured of a form of bubble boy disease; in 2010, a man was cured of a kind of blood disease called beta-thalassaemia. Each of these success stories were the result of targeting single genes. The wait may be longer for gene therapies to tackle the more complicated, multi-gene diseases, but its evident that some diseases are amenable to a single gene approach. I think we can expect exciting results in the near future. Research is currently underway to develop gene therapies for Huntington’s disease, blood disorders, various forms of cancer, Sickle cell anemia, cystic fibrosis and deafness.

To be clear, both the GAD gene and levodopa treat the symptoms, not the underlying causes of Parkinson’s disease. Until we know exactly why the neurons of the substantia nigra become impaired we won’t be able to develop a cure. For now we’ll have to be okay with that. And if Neurologix one day soon delivers a gene that improves quality of life for millions of the world’s people with Parkinson’s disease, I think they’ll be okay with that too.

[image credits: wearingoff.org, New York Presbyterian/Weill Cornell Medical College via The Dana Foundation]

Dr. Sheng Ding pioneered a method by which skin cells are converted to heart cells without going through an induced pluripotent stem cell state.

It’s faster, more powerful, and user-friendly. No, I’m not talking about the latest generation tablet, I’m talking about the latest upgrade in stem cell research. The transformation of adult cells from one type to another is common enough. We’ve reported on researchers successfully transforming skin cells into heart, blood, and intestinal cells. This process typically involves converting the adult cell to a pluripotent, stem cell state, from which it can differentiate into one of the specialized forms. As if the cell one day realized that it never really wanted to grow up to be a skin cell, scientists could help revert it back to its infant—or, embryonic—state so it could have another go at life. A recent study by scientists at the Scripps Research Institute in La Jolla, California showcases a different method that bypasses this initial transformation to the stem cell state. Apparently you can teach an old dog new tricks.

Over the last decade scientists have had increasing success in converting skin cells and other types of cells into something different, including heart and blood cells. Efforts are underway across the world to improve the techniques and clinical viability of these cell conversions. The work by Dr. Sheng Ding and his colleagues at Scripps qualifies as a major improvement. The road ahead still requires much work, but it’s clear that each day mankind moves closer to producing cells of every type, custom made for your body.

Faster

The novelty of the new research coming out of Scripps is not going from skin cells to heart cells beating in a dish—that stuff’s becoming old hat—but that they accomplished it in just 11 days. It is normally a two step process that requires four to five weeks. It also requires a lot more work, owing to the step where the skin cells are converted to induced pluripotent stem cells (iPS). This is done by introducing four genes recently discovered to reprogram differentiated adult cells to embryonic stem cell-like pluripotency. The four genes encode transcription factors, proteins in the cell nucleus that regulate the expression of other genes. Typically the four genes are active for two to four weeks before the differentiated cell is converted to an iPS cell. Ding’s group modified this protocol by allowing the genes to work for as little as four days before deactivating them. The result are skin cells “pushed” in the direction of the induced stem cell state without actually becoming iPS cells. Turns out that’s enough, which, aside from saving time, reveals something new about stem cell biology. The current work was performed using skin cells from mice and it remains to be seen if the shortcut can be applied to human skin cells. Nevertheless, to render the iPS cell stage unnecessary is a major paradigm change for the field and it will be interesting to see if the new paradigm bolsters progress in the near future.

You can see the beating cells in a video below from newsy.com’s coverage of the study:

More Powerful

In addition to being faster, Ding’s protocol boosts efficiency. The old protocol yields an estimated maximum of approximately 0.2 heart cells for every skin cell plated. Skipping the iPS cell stage yields a whopping 1.2 heart cells per skin cell. In the paper the team speculates that the increased efficiency is due to the generation of mitotically active cells which are able to divide and multiply. Resembling heart precursor cells, they speculate further that “these intermediate cells, if successfully isolated and stabilized in culture, could become an expandable and renewable source for not just cardiomyocytes, but many other terminally differentiated cardiovascular cells as well.” In the paper they extend this thought, suggesting that the principle of a versatile intermediate might be important, not only for creating the numerous types of cells that go into making a heart, but for stem cell applications in all tissues.

Scientists at the Scripps Research Institute needed just eleven days to convert skin cells in beating heart cells.

User Friendly

The four genes that researchers use to produce the iPS cells is risky because these same genes can turn cells into tumors. Inactivating them after only a few days instead of a couple weeks reduces this risk. And, like any self-respecting technology, an upgrade is in the making. Because they can turn cells cancerous, stem cell researchers have been searching for a way to reprogram differentiated cells into iPS cells without using the four genes altogether. Demonstrating that the genes are only needed for a few days instead of weeks simplifies the problem and makes the genes easier to replace.

To be sure, stem cell research has a lot of ground to cover before it becomes an effective treatment for disease. For example, the current study was done in mice and it remains to be seen whether or not the shortened protocol produces the same results in human cells. I find it impressive, however, that the four genes widely used by researchers to convert fully-differentiated, adult cells into embryonic-like, pluripotent stem cells were discovered less than five years ago. Since then iPS cells have been gotten by converting other cells besides skin, including cells from the stomach and liver. The current study was the first that we are aware of to bypass the iPS cell stage for differentiation to heart cells, but this shortcut has already been taken for differentiation into blood cells. It is exciting to note that human cells were used in that study.

But in case you hadn’t heard, stem cells have already been used in tissue replacement therapies. We’ve previously reported on tracheal transplants of two women. This involved a donor trachea (from a cadaver) that was coated with a layer of the patients’ stem cells which fostered regrowth of the trachea. Because the new layer of cells originated from the patient the risk of an immune response against the new trachea was minimized.

From my vantage point, it seems that stem cell therapies are inevitable. I also believe that the day is long in coming. Unfortunately it seems that many people have been set up to hope for miracles after the hyperbolic political battles over stem cell research in the past. But therapies rarely come from sudden miracles. Instead it is the incremental advances and shifts in paradigm, such as that achieved by Ding and his colleagues, that will bring us the stem cell therapies we are hoping for.

[image credit: The Scripps Research Institute]

[video credits: newsy.com]

A researcher at the Autism Genome Project

Autism remains one of the most poorly understood and troubling developmental disorders of modern medicine. But the genetic revolution could turn that around. Recent research by the groundbreaking Autism Genome Project has identified key mutations and susceptibility genes involved in the disorder. Down the road, this could pave the way for new treatments.

Autism is a poorly understood developmental disorder that can cause impaired social interaction, poor communication skills, repetitive behaviors, and a variety of other symptoms. Cases differ in both the degree of severity and particular symptoms, and include three subtypes along what is called the autism spectrum: Autism, Asperger’s syndrome, and Pervasive Developmental Disorder. Symptoms usually appear before a child is three years old, and the disorder is four times more common in boys than in girls. The National Institute of Health estimates that three to six children in every 1,000 will have some type of autism spectrum disorder.

The cause of autism is notoriously elusive, and has been the focus of extensive research over the past few decades. Because disorders along the autism spectrum are diagnosed behaviorally, they exhibit variable symptoms (thus, the “spectrum”) and aren’t the result of any single cause. Controversies over autism’s causes have debated whether its roots are biological or environmental (see, for example, the debunked vaccine theory). Recent research is showing that like most psychological disorders, autism is linked to a combination of genetic, epigenetic, and environmental factors. However, the emerging consensus is that autism is primarily genetically mediated, with high rates of heritability for both autism (0.7) and Asberger’s syndrome (0.9). Until recently, the actual genetic mutations responsible for autism’s heritability have remained unclear.

Enter the Autism Genome Project (AGP). A large-scale collaborative research project of 120 scientists from 60 different institutions, the AGP aims to identify the genetic architecture involved in susceptibility to autism and spectrum disorders. This month, the AGP reported in Nature its findings that strengthen a growing consensus about the role of genetic mutations called copy number variations (CNVs) in the autism spectrum. Because we all inherit 23 chromosomes from each parent, most folks have two copies of each gene or DNA segment – one from mom, one from dad. A copy number variation is when a particular stretch of base pairs (sometimes encompassing whole genes) differs from the expected number of two. Sometimes a deletion leaves one set, and sometimes a duplication gives three or more sets.

The AGP research found a higher prevalence of CNVs in autistic populations than in controls, which are thought to disrupt genetic mechanisms important for neuronal and intellectual development. Only about 1% of the general population has these CNVs, and they are only present in 3.3% of autism cases. The AGP study also identified four susceptibility genes involved in cellular proliferation, synapse development, and signaling between neurons – but again, mutations on these genes are rare and aren’t common to all individuals with autism. These findings support a growing consensus that autism is caused by a multitude of “rare variants,” none of which directly cause the disorder but which collectively increase an individual’s susceptibility.

AGP started in 2002 and has undergone two major phases of research. Phase One concluded in 2007 and identified two genes involved in autism susceptibility.  Phase Two – the research presented here – compared the genomes of 996 people with autism against those of 1,287 control subjects, including many family studies. Future research at AGP will focus on identifying further CNVs within the population by increasing the sample sizes for comparison.

It’s important to recognize the complexity of factors that lead to autism, as well as our limited understanding of these relationships. Autism is not caused by just a few genes or mutations, and we aren’t going to find a “smoking gun” cause anytime soon. In many ways, autism research exemplifies the growing recognition that genetic determinants are messy and complex (recently covered here at the Hub). Instead of straightforward common genetic causes, the susceptibility to autism is linked to unique combinations of mutations, only some of which show significant overlap between individuals.

This recent AGP research is promising in two ways. First, it opens up new avenues for further research, showing scientists where to look to more clearly understand the causal mechanisms that underlie autism’s genetic component. Second, it paves the way for new treatment options by clarifying which molecular mechanisms are broken in the autistic genome – malfunctions we can develop drugs to counteract. This research is most valuable within the broader context of autism’s epigenetic and environmental factors, and its implications for treatment should be combined with equally important therapy-based approaches.

Autism remains one of the most troubling and elusive psychiatric disorders in the modern age. There is still a great deal we don’t know. Is the disorder becoming more common, or is it being diagnosed more consistently? What environmental factors increase the risk of autism, and how do they interact with genetic components? There are many avenues for further research, and it is promising to see such massively collaborative efforts as the AGP. Slowly, as we shed more light on the autistic genome, we can develop treatments more appropriate to the disorder and the individual patient. And because the costs of genetic testing are dropping rapidly, research should continue to accelerate and put new therapeutic options within reach. Maybe someday, we can look back on this pioneering research as the foundations for a cure.

Check out this short video from ABC covering the recent findings:

[image credit: Autism Genome Project]

New computer analysis out of Stanford is poised to help researchers find important genes in a fraction of a second.

A little bit of computer science has made it possible to reduce the time it takes to find new genes from years to milliseconds. Researchers at Stanford University have developed a simple processing method for searching through large public databases of genes and their associated proteins. Using Boolean logic (if A then B, if not C then D, etc), they have found a way to suggest which genes are responsible for different stages of complex chemical processes in our cells. Biologists generally take months or even years to find these genes experimentally, but the Boolean search method takes just a fraction of a second. When widely applied, the new method may accelerate genetic research enormously. Great things happen when sciences come together.

There are tens of thousands of human genes, and we are still uncertain about the exact function of most of them. Still, scientists have mapped the proteins associated with many genes. Large databases of these gene and protein relations have been compiled, and many are open for researchers to access. There’s huge amounts of data here, but turning it into meaningful scientific insight isn’t easy. Luckily, Debashis Sahoo of Stanford had an insight: the expression of genes is often asymmetric. Some gene X will be expressed only when gene A is not expressed. This lets you use Boolean logic to sort genes.

From that simple beginning Sahoo was able to perform a sort of Boolean analysis to determine a gene’s importance in a given protein pathway. Say you know that a protein pathway starts with gene A and ends with gene B, but you don’t know much about what happens in between. You can look through all the other genes and their associated proteins and find some that are not expressed at the same time as A, but are expressed with B. These genes may code for proteins that occur inside the pathway.

That’s an overly simplified explanation of the Boolean net Sahoo assembled, but you get the idea. With the right “if then” filters Sahoo was able to sort through thousands of genes in a fraction of a second, finding just those which are likely to be important in a given protein pathway. Instantly he had a short list of genes that geneticists could examine.

Does all this computer science sorting actually yield results? You bet. As discussed in PNAS, Sahoo and other Stanford researchers used their Boolean method to comb through a database looking for genes related to B-cells (an immune system cell). They found 62 genes that could be involved in B-cell pathways. They then looked at the DNA from 41 strains of mice that had been modified to have disruptions (so called ‘knockouts’) in those genes. In 26 of those 41 strains, they found mice that had disrupted B-cells. In other words, the Boolean method looks to have been better than 50% accurate in suggesting genes that might be related to B-cell development. That may not sound like much, but think of all the thousands of genes that they looked through – it’s like finding needles in a haystack!

And this is just the beginning. Those databases of genes and proteins can now be sorted using Sahoo’s method to suggest new genes to explore. Think of an important gene out there, say the FOXO3A gene that may code for longevity. Wouldn’t you like to know which genes are related to it? Oh yes. Scientists are starting to suspect that most of the characteristics we want to investigate (longevity, intelligence, resistance to a certain disease) may rely on a very complex interaction of different genes. Sahoo’s Boolean method will help researchers hunt down related genes quickly and effectively and get a better understanding of those complex interactions. We already have large stores of genetic information, biobanks, ready to be sequenced and examined. As that information becomes available, analytical methods will allow us to turn raw data into meaningful insights into how we should perform genetic research. In the end, genetics is an information science and with the right application of computer skills we’ll be able to accelerate its progress. That’s going to mean quicker and better results. So give some praise to computer logic. If Boolean Then awesome.

[image credit: Singularity Hub]
[source: Stanford News, PNAS]

Dr. Cliff Reid, CEO Complete Genomics, Has A Master Plan To Sequence 1 Million Genomes!

Without a doubt the hottest company in the genomics sector right now is gene sequencing powerhouse Complete Genomics. In just the last four years the company has come out of nowhere to dominate the market for low cost sequencing of human genomes in large quantities. Although Complete Genomics is now slated to sequence an incredible 5,000 human genomes in 2010, this is nothing compared to what the company has in store for the years ahead.  Just days ago, in a Singularity Hub exclusive interview with Complete Genomics CEO Dr. Cliff Reid, we have learned that the company is now hoping to sequence 50,000 genomes in 2011 and a whopping 1 million genomes by 2014. Considering that by the end of 2009 only about 100 or so human genomes had ever been sequenced, most of them by – you guessed it – Complete Genomics, this represents an enormous shift in the industry. In the rest of this post I will share with you the juicy details from the interview, followed by the full video of our conversation at the end.

Although companies like 23andme or Illumina have been hogging much of the headlines in genomics recently, the real story may be that Complete Genomics is about to rewrite the game for the entire industry. Simply put, Complete Genomics is the first company to realize that sequencing human genomes is a brute force computational problem that is best overcome through large scale centralization.

Traditionally if a research team wanted to sequence a human genome they would be forced to purchase expensive machines from the likes of Illumina to do the job. These machines, such as Illumina’s latest HiSeq 2000, might cost half a million dollars or more up front, require the hiring and training of several staff to operate and maintain the instruments, and require several different types of expensive, specialized materials as continuous inputs. What’s more, these expensive and wonderful machines might end up sitting around much of the time unused in between projects. In a world that demands the sequencing of millions of human genomes in the coming years, this model of distributing individual sequencing machines is simply too costly and inefficient.

Enter Complete Genomics: Master of Centralization and Scale

In the next decade we may sequence the genome of nearly every person in the developed world. With 6 billion people in the world and approximately three billion base pairs per genome we are talking about an enormous task of scale and computation.  Years ago Complete Genomics realized that centralization in a dedicated sequencing facility was the answer to this challenge.  Today they are bringing their vision to reality.

Instead of building individual machines that can be shipped off to laboratories, Complete Genomics is turning the traditional industry model upside down and doing the sequencing itself.  Researchers send Complete Genomics a sample of human DNA in the mail, allow them to process it in their sequencing center, and shortly thereafter they will ship back the sequencing results at a cost and speed that is crushing the rest of the industry.

What do I mean by “crushing”?  In November of last year Complete Genomics announced that they had sequenced 3 human genomes at an average cost of materials below $5000 apiece, shattering all previous records by nearly a factor of ten!  Last year Complete Genomics was charging its customers $20,000 per genome and this year they will be charging $10,000 or less.  We can expect the company’s costs and the prices it charges its customers to continue to drop dramatically in the next few years. The $1,000 genome is indeed within sight.

Complete Genomics is essentially turning genomic sequencing into an assembly line process with all of its associated advantages. Equipment can run pretty much 24/7 without interruptions, thereby maximizing the output and return from multimillion dollar investments. A small staff can be trained to run an entire facility of sequencing machines. This significantly reduces the human cost of training and labor.  Reagents and other supporting materials can be purchased in bulk on the cheap.

Further streamlining the process and the costs, Complete Genomics is only sequencing human genomes. This is a huge differentiator that people often overlook, yet it is crucial to the competitive advantage of the company. When working with multiple organisms, there are unique factors such as reagents, read sizes, genetic coding idiosyncrasies, and preparation methods that must be accounted for. By focusing solely on human genomes Complete Genomics is further optimizing its operations for low cost and high efficiency.

Although originally slated to go live this January, Cliff Reid says that Complete Genomics’ first large scale sequencing center is now going to launch on April 1. It is because of this delay that Complete Genomics’ is only targeting 5,000 genomes this year instead of its original target of 10,000. Of course 5000 genomes is still nearly 50 times the number of genomes that have ever been sequenced to date by all companies/institutions combined. Not bad!

Can They Really Sequence 1 Million Genomes In 5 Years?

Although Complete Genomics is aiming for 1 million genomes by 2014, we need to take this target with a grain of salt.  Given that the company is set to deliver only half as many genomes in 2010 as originally planned, who is to say that their 2014 roadmap won’t fall equally short?  Yet to focus on a specific number really misses the point.  The key takeaway here is that Complete Genomics is finally ushering in the long awaited era of cheap, high volume genomes through assembly line centralization and scale.  The model seems to be a winner, and even if Complete Genomics were to somehow stumble, it is likely that competitors would be quick to follow suit.

A comparison to Henry Ford’s pioneering of the car assembly line with its hugely successful Model T naturally comes to mind, and this is not lost upon Cliff Reid.  Ford famously said “Any customer can have a car painted any color that he wants so long as it is black.”  In homage to Ford, Reid joked during the interview that “We’ll sequence any organism, as long as its human”.

United States Today…Tomorrow The World

Over the course of the next year Complete Genomics will be creating plenty of waves in the industry with the world’s first human only large scale sequencing facility here in Mountain View, California.  This facility will single handedly sequence on the order of 50,000 human genomes in the next 18 months.  Impressive – yes – but to get to 1 million genomes in the next 5 years Complete Genomics is going to need more large scale sequencing facilities.  Many of them will need to be in other parts of the world, such as Asia.

Can you say CAPEX?  According to Reid, the current plan is to build up to 10 sequencing facilities in the next several years, each of them able to produce anywhere from 50,000 to 100,000 genomes per year.  Although increased output and redundancy of operations is a key driver of these added facilities, Reid points out that political jurisdiction is an equally important driver.  Governments will be reluctant to see their citizens’ genomic data crossing borders.  If Complete Genomics wants to be in the game of sequencing genomes of major countries in Asia they are going to have to go inside those countries to get the job done, and that is indeed where they will go.

More Genomes, More Money – An IPO?

As we can see it is going to take a fair amount of capital for Complete Genomics to pull off its master plan.  Last quarter the company secured $45 million in funding despite a very harsh economic environment.  Yet this cash infusion is only a temporary measure to get the first facility or two up and running.  More money will be needed to realize the company’s vision for ten or more sequencing facilities and 1 million genomes in the next 5 years.  As Reid says in the interview “Fast growing companies like ours inhale cash”.

Of particular interest to many following this hot story is whether or not an IPO is in the company’s future.  Although legally and strategically we can’t expect Reid to full out confirm an upcoming IPO, he does the next closest thing with the following comment “We also hear rumblings of the public offering market becoming open to certain companies and we would consider that a very attractive option”.  Investors get your cash reserves ready – an IPO looks like a strong possibility.

What Will We Do With 1 Million Genomes?

One argument I often hear from people is that all this genome sequencing business is a big waste of time.  After all, more than fifty genomes have already been sequenced and what do we have to show for it?  Where is the medical revolution that genetics was supposed to unleash?  Will there really be a demand for the sequencing of 1 million genomes in the next 5 years even if Complete Genomics can provide it?

Are you freaking kidding me!  One million genomes is just the tip of the iceberg folks!  Over the next decade or two we will probably sequence tens of millions of human genomes, and – yes – this data WILL be useful.

As with nearly all hot topics of the day – and genomics is no exception – our imaginations have gotten ahead of the technology.  The medical revolution promised by genomics will indeed become reality, but it will take many more years than people thought.  It turns out that by sequencing only a handful of human genomes there is only so much information that can be learned.

As Reid likes to say, there are only about 1,000 major human diseases out there.  One million sequenced human genomes will allow us to study the genetics of each of these 1,000 diseases, each with a pool of 1,000 genomes for comparison.  The information that will be teased out of this data will indeed produce the medical revolution that we have all been waiting for, but first we need tens of thousands of genomes to perform the required analysis.  We need the data!

We Are A Data Company

One of the things I love most about Complete Genomics is their laser focus on doing only one thing: sequencing human genomes.  This extreme focus on doing one thing well has been a proven secret to success for many of the world’s greatest companies.  It will be no different with Complete Genomics.

Rat genomes, worm genomes, and the genomes of countless other organisms will need to be sequenced in the coming years, but Complete Genomics is going to ignore those.  As Complete Genomics comes to dominate the market for human genome sequencing in the next several years, it may be the sequencing of non-human genomes that will provide a still enormous market for the Illuminas of the world.  It is great when a field is so large that there is enough room for pretty much everybody to win.  Genomics is one such field.

In the end Complete Genomics is a data company.  Their gift to the world in the next several years will be to deliver the vast store of data that is locked within the individual genomes of each of us.  Understanding this data, and converting this understanding into real world medical solutions, however, is not Complete Genomics’ game.  They will leave that task to other companies.  Complete Genomics is not the company that will directly give us the medical revolution that we have been waiting for, but indirectly their role is equally important.  They are giving us the data.

But Who Are The Customers?

Although it is exciting to think of average citizens purchasing their genomes directly from Complete Genomics, at least for the next two years Reid explains that the real demand will come from research institutions and corporations.  These organizations have the budgets, not to mention the desire to discover the next big medical breakthrough, to justify the purchase of thousands of genomes from Complete Genomics.  Individuals will indeed be able to purchase there own genomes in the future, but at least for the next 2 years individuals will not be the key driver in this business.

Complete Genomics – We’re Watching You!

Well, that about sums up the interview with Complete Genomics, but our coverage of the company has only just begun.  We will be monitoring the progress of the company closely, so stay tuned for more posts in the future.  In the meantime, be sure to check out a full video of the interview below:

23andMe is a personal genetics firm that allows individuals to test their genome for key genetic markers. These markers take the form of SNPs (pronounced ‘snips’), single nucleotide polymorphisms. A standard test that grants you access to information about ancestry, health, and traits costs you about $399. A research version is available for just $99. Basically all you do for either option is spit in a special tube and then mail it to the company.

The 23andMe research revolution is pretty straight forward. The company needs volunteers and sponsors to help in genetic testing for 10 diseases: migraines, psoriasis, severe food allergies, arthritis, celiac, lymphoma/leukemia, multiple sclerosis, ALS, epilepsy, and testicular cancer. Sponsors get to vote on which disease will be prioritized. Besides sending in some spit, volunteers will also be entering a lot of health information online in order to find correlations between genes and diseases. There’s no guarantees that the genetic testing and correlations will lead to any worthwhile data, but you have to admire 23andMe for getting out there and shaking things up.

The research revolution isn’t 23andMe’s first foray into a democratic approach to genetic testing for diseases. As we mentioned a few months ago, they sought out 10,000 volunteers for a Parkinson’s study. While there may be some statistical problems with the way that 23andMe solicits volunteers (everyone has to have at least $99, right?) the activism portion of their approach is laudable. With this new push for research, there’s a good chance that some insight will be made into at least one of the ten diseases mentioned.

In the future, other diseases will be added to the list, and past data will be leveraged into the new tests. That’s a lot of bang for your genetic buck. Just to show you how easy the submission process is, here’s a video from health advocate and strategist Jen S. McCabe:

You know, I don’t want to turn this post into a wholesale endorsement for 23andMe and their research revolution, but I’m definitely in favor of it. The idea of democratizing research while still keeping it meaningful is tremendously motivating. 23andMe is setting the basis for future debates on genetics just by affirming an individual’s rights to know more about their own genetic code. As we’ve said in previous stories, the company sits at the crossroads of genetic testing and internet community building that will be a powerful meme going forward. Even if this particular research revolution doesn’t yield results, one eventually will.

MIT is helping synthetic biologists by providing the Registry of Standard biological Parts.

While there are some costs associated with getting genes from the Registry, it’s not really a store. The registered segments of DNA are stored and shipped on a looser “get some, give some” exchange. Those users who request and utilize these biological parts are expected to share some of their results and innovations with everyone else. Sort of the biological equivalent of the take-a-penny-leave-a-penny tray at the corner store.

Before you start sending your genetic requests to MIT, I should point out that the Registry is for established scientists only. Do-it-yourself biologists need not apply. Most of those who receive parts are from academic labs, and/or forming a team to participate in iGEM, MIT’s annual genetic engineering competition. Still, the wide range of users gives this registry a scope that promises to catapult synthetic biology into its next phase of evolution.

She’s a brick house…
When you’re asking for a few parts, the Registry is your place to sift through and plan. But how can you be sure to splice the DNA into your creation in the right way? How can you make sure that each part you want is complete? Thankfully, the registry conforms to the Biobricks ™ standard. Biobrick ™ is a way of standardizing interchangeable genetic parts, allowing each developer to work separately but still design DNA that can work with everyone else’s creations. It was developed by a not-for-profit company composed of researchers from MIT, Harvard, and UCSF. Just to be clear, Biobricks™ and the Registry aren’t the same thing: one’s a standard and the other is storage. Together, however, the two are helping create a common vocabulary of genetic innovations.

We’ve seen this same approach in the field of robotics with Willow Garage. In fact, it’s pretty much the open-source template that software gurus have been creating for some time. In biology, the result is less open-source and more open-community. Even while working in different labs with different goals, by participating in the registry and using a standardized way of formulating synthetic genes scientists are helping raise the capabilities of the entire group. Along with a wiki for the parts, and some basic tutorials, this group dynamic allows relative new-comers to the field to get up to speed quickly. The more researchers who operate at this high level, the more innovations are bound to be produced. In essence, the Registry is a positive feedback loop, allowing each development to lead to other developments in an increasing spiral of genetic know-how.

Besides the social benefits, however, the Registry also provides some rock-solid advantages when using their standards. Most parts come with a gene (ccdB) that helps filter out specimens that haven’t received the gene. When you are modifying the DNA of an organism, it’s not always easy to tell which (if any) organisms have added the new genes correctly. By utilizing ccdB, scientists can kill off those cells which don’t have the modified genetics and keep the ones that do. There are many other “genetic tools” included in the database that help you manipulate DNA by moving, promoting, connecting, etc. In this way, the Registry isn’t just a gene depot, it’s also a genetic appliance warehouse.

There is still a lot of room for improvement. Most of the sequences described in the Registry still haven’t been built yet. Those genes that are complete may be categorized poorly. Even with intimate knowledge of synthetic biology, wading through the Registry can take considerable time and effort. Hopefully MIT will organize the system better. Establishing guidelines and rewards that will encourage development is likely in the future. Already, the Biobricks ™ foundation is working on a licensing scheme for new parts.

The Future is for Designers

We can’t talk about synthetic biology without mentioning some of its dangers. As tirelessly as researchers at MIT, and other institutes, pursue better living through genetics, other more nefarious work is undoubtedly being pursued elsewhere. Genetically modified warfare, or terrorism, is an unfortunate consequence of a better understanding of biology. Even discounting intentional biological attacks, we still have accidents to worry about. While developing a bacterium that could consume industrial waste, scientists could create one that devoured living tissue. The doomsday scenarios are endless but also avoidable. Part of the reason to standardize biological parts is to better control how those parts are integrated into living things. Better understanding, and standardized techniques will cut down on accidents. And the best defense against biological terrorism is a comprehensive set of genetic material that can be adapted to combat new threats. Far from increasing our risks, the Registry is a tool for avoiding them.

As the Registry and the Biobricks™ standard develop though, scientists will spend less time building genetic parts and more time experimenting with genetically modified creations. In the short term that of course means more cool micro-organisms that can perform miraculous tasks. Just looking at last year’s iGEM competition we see vaccines, teeth-cleaning yogurt, bio-sensors, bacterial computers, and many more crazy and wonderful creations. In the long-term, the rapid testing of genetic manipulations will eventually lead to an understanding of how every part of an organism is built. In the end, scientists won’t be using the Registry parts to modify cells, they’ll design an organism from the ground up.

That is when the real change will begin. Evolution has provided each organism you want to modify with a huge set of genes, expressed or otherwise, that have allowed it to succeed. Most of these genes don’t have anything to do with what humans want: a cell that performs a certain task. By starting from the ground up, scientists could create organisms that just do what they are told, nothing else.

Natural organisms also come from ecosystems complete with predators, a preferred food source, and environmental requirements. A designed organism could be free of all of those. No more worrying if the bacteria creating your bio-fuel will be wiped out by a virus. Did you build something that could harm people? Well, design it to die at 98 degrees Fahrenheit and maybe we won’t have to be concerned. Of course, genetic design isn’t a panacea. There will undoubtedly be new concerns and new limitations. The Registry of Standard Biological Parts, however, will help accelerate genetic engineering to overcome these concerns and limitations.

Obviously restricting access to these potentially powerful and dangerous genes is a good thing, and it makes a lot of sense for the Registry to be available only to scientists. However, as genetic engineering becomes more accessible to more people, I hope that the Registry follows suit. Already we’ve seen that the DIYbio communityis a growing and responsible set of amateur enthusiasts. I would be excited to see what such at-home researchers could produce if given access to a wider set of genetic information. For now, most DIY types wouldn’t even be able to utilize parts from the Registry (they don’t have the equipment or expertise), but that may change as wet-labs themselves become standardized and easier to understand.

What will we create with the building blocks of life?

Whether in the hands of traditional scientists or in a broader context of genetic enthusiasts, the standardization of genetic parts is a remarkable undertaking. MIT could possibly becomes the global hub for genetic engineering for no other reason than Biobricks™ and the Registry. As with so many other developments in technology, the way in which new creations are made has become as important as the creations themselves. As a child, I built spaceships, secret bases, and x-ray glasses out of Legos. Using the genetic equivalent, I have no doubt that synthetic biologists will construct even cooler creations.

Can a baby unlock the gene for strength?

The first super baby was born in Germany in 2004. Though his name was never released, pictures demonstrated that his young physique contained almost twice as much muscle as other infants. Look ahead to fall 2005 in Michigan, Mr. and Mrs. Hoekstra adopt a young boy named Liam. Soon he is growing muscle at an astounding rate. Hanging on rings in an iron cross position by 5 months, pull-ups by 9 months, Liam is the second super baby. His condition, now known as myostatin-related muscle hypertrophy, makes him hungry, lean, and strong. Check out his pic after the break.

With Liam, scientists had further proof that a genetic mutation could exist that causes a human to naturally build muscle. Without even trying, Liam has little to no body fat, can lift seven pound weights arms extended (he only weighs 30 lbs himself) and has a six-pack. Now nearly four, Liam is taking gymnastic lessons, but this is more of an outlet for his energy than an explanation for his physique. No doubts, it’s the lack of myostatin that’s helping him get ripped.

The Protein to End All Proteins

Blocking myostatin has been shown to have drastic effects in animals besides humans. Myostatin tests in labs have pumped up mice to Schwarzenegger proportions.

A whippet named Wendy has a bizarre condition that has slowed her myostatin production. A strain of cattle known as Belgian Blue are predisposed to genetic conditions that lower production of the protein…and wow, you can really tell. Looking at photos of each of these animals, I’m struck by the near absurdity of their pumpitude. It’s like someone delivered free weights to the zoo.

Which isn’t to say that the applications for myostatin blocking would be absurd. Doctors, like Louis Kunkel of the Children’s hospital in Boston, have long been searching for cures for muscular dystrophy (MD). It’s the most common genetic disease and few sufferers live into adulthood. A myostatin blocker could help these children survive and perhaps even live normal lives. All it might take is relatively small changes in the level of the protein: 20-50%.

Yet, when you type ‘myostatin’ into Google you don’t get web forums dedicated to curing muscular dystrophy. Since breakthrough research in 1997 by Alexandra McPherron, Se-Jin Lee, and Ravi Kambadur proved the effects of myostatin, the discussion has focused on one topic: blocking myostatin in order to get ripped. Buff. Cut. However you grunt it, the draw to have a treatment that creates muscle – with little exercise needed – is outweighing the medical pursuits of MD cures. Now that a human gene has been linked to myostatin-blocking I can only predict that such demand will increase dramatically. Expectedly, we can already buy “myostatin-blockers” as a workout supplement. Obviously I can’t comment on the veracity of the claims made by these products, but you may want to think about them in the same vein as other…er… ‘enlargement’ offers you receive online.

Women and Super Children First

Liam’s mother has been hesitant to allow press into her young son’s life. Rightfully she fears that the coverage would turn Liam’s existence into a circus. Even further ensconced in anonymity, the German super baby isn’t giving any interviews. However well they are shrouded from the public, however, they’ve been genetically sampled and will undoubtedly continued to be so as they mature. Using this information, it isn’t a matter of if the myostatin gene could be modified in others, it’s a matter of when.

So the demand is there, and the possibility is coming. What will it mean to have an available genetic treatment which will permanently make you able to build muscle with little effort? First, I hope it means sufferers form MD, AIDS, and other debilitating diseases will find relief from muscle atrophy. Secondly, maybe not trivially, it will mean a new series of anti-doping tests at sporting competitions. But most importantly it will be a sign that humans will be on their way to modifying their bodies to fit their lives and not the other way around.

From the 1997 study on myostatin blocking in mice. The hypertrophy is clear in two subjects.

Just recently Singularity Hub has discussed growing organs, stem cell treatments, and a wide host of bionic body augmentations (even some not so bionic additions). Genetic manipulation, however, is the real holy grail of the body-crafting endeavor. Much of this pursuit has focused on understanding the purpose of each sequence of the genome. But with myostatin, some of that completeness may be deemed unnecessary. Most people won’t be interested in discovering what a million different genes do when changing just one gives them the ideal athletes body.

That’s a recipe for disaster. We hope that Liam Hoekstra has a happy healthy life. But knowing the effects of other kinds of unchecked growth can include tough health problems, we should expect a price to come with his amazing gifts (remember Andre the Giant). We do not fully know the effects of myostatin on smooth and cardiac muscle. Organ development may benefit or be stunted. As researchers are quick to point out, ligament and tendon strength do not necessarily coincide with muscle strength. Liam already has to work a little harder on getting the flexibility common to children his age.

Myostatin genes have to be understood in the larger context before they can become part of a general genetic engineering lexicon. Let’s hope the demand for work-free muscle building will be controlled enough to wait for it. Before we can build super men we’ll have to get them the old fashioned way: by letting super babies grow up on their own.

Good luck, Liam, and watch out for Lex Luthor.