Friday, December 31, 2004

In vitro Gene Production... on a chip!

Traditionally, when you're interested in studying a gene or want to use it in a fancy molecular construction, you have only one option : molecular cloning techniques which are (very) time consuming. Recently, Gene production techniques emerged; they are relatively rapid (turnaround ranging from 2 weeks to a month or more, for more complicated construction) but very expensive (1.5$ / basepair... so a pricetag in the thousands is not uncommon). The technique involve oligo synthesis and stitching by PCR; it work well for small genes, but large constructions can be problematic. Researchers at the University of Michigan just invented a way to easily build genes using a microfluidics approach.

It's fast and inexpansive. And it will open loads of interesting research avenues and speed up drug development for sure!

Synthetic biology: Researchers mass produce genes on a chip

Imagine that the bricks used to build a house cost $1,000 each—building a home would be cost prohibitive. Similarly, the bricks to build living organisms—genes and genetic assemblies—can cost thousands of dollars to make in the lab, which is also cost prohibitive.

But now, scientists have developed a way to make the materials for genes on a microchip in mass quantities, for a fraction of the current cost.

The technology enabled scientists to make an important part of the genome for an E-coli bug, and to reproduce the instructions for making proteins. This has significant applications in vaccine production, gene therapy, and DNA sensors and diagnostics.

"The significance of our paper is that for the first time, we have a mechanism for us to make the genes in high accuracy, very inexpensively, and to make those genes containing the information for the protein factory in an organism" that synthesize all other proteins in the body, said Erdogan Gulari, a University of Michigan professor, the Donald L. Katz Collegiate Professor of Chemical Engineering and co-author of a paper on the topic to appear in the Dec. 23 Nature. "This is the starting point to making a complete functioning organism that can produce energy, neutralize toxins, and produce medically useful proteins, for the benefit of human health and quality of life."

The paper, entitled "Accurate multiplex gene synthesis from programmable DNA microchips," was co-authored by researchers from the U-M College of Engineering, Harvard University, University of Houston and Atactic Technologies Inc. The technology is currently licensed to Atactic, a company founded by a U-M College of Engineering alum Xiaochuan Zhou, professors Xiaolian Gao of University of Houston, and Gulari.

If made the standard way, a typical gene can cost several thousand to hundreds of thousands of dollars, Gulari said. That's because the cost of putting together each nucleotide, the building block of DNA and RNA, comes to $2 to $7 dollars each. Genes contain thousands to tens of thousands of oligonucleotides, short chains of nucleotides that join together to make genes. So synthesizing all the genes of even the most primitive living organism, which has several thousands of genes, could cost millions of dollars and take years, Gulari said.

The new method uses technology similar to that used to make computer chips, Gulari said, and generates oligonucleotides in thousands of tiny reaction wells and releases the sequences synthesized, Gulari said.

Scientists start with a thumbnail-sized silicon or glass chip containing microchannels and microwells loaded with low-cost reagent. The wells are so tiny, Gulari said, that thousands of them can be filled by half a drop of water. By shining tiny pixels of light at selected areas on the chip in a predetermined manner, scientists made thousands of gene fragments of specific sequences each containing 30-70 nucleotides. They then collected them in a solution microtube, and stitched them together in the desired order to produce the genes by enzymes.

The benefits of synthetic genes are tremendous, Gulari said.

"For instance, these products can be used to improve DNA sensor and diagnostics for comprehensive and more sensitive genetic analysis, and to produce the blueprint for novel proteins," Gulari said. "Some of these proteins would be too toxic to obtain from natural sources, so the synthetic version is much safer. Some of these man-made proteins have novel functions which do not exist in nature, which potentially can be a new generation of vaccines or therapeutics."

For instance, Gulari said, 50 percent of drug molecules are based on proteins and antibodies, and there are over 371 new products currently in clinical trials targeting more than 200 diseases. Consequently, there is great interest in creating humanized antibodies for early detection of infection and for medicines. For these applications, millions of new proteins and antibodies must be tested, he said. This technology makes that possible.

A forerunner to the synthetic gene was the birth of recombinant DNA (the joining of DNA from different species and subsequently inserting the hybrid DNA into a host cell) about 30 years ago. Recombinant DNA, Gulari said, has become an indispensable tool for advancing biochemical and biomedical sciences for improving health care and disease treatment. Without recombinant DNA technologies, there wouldn't be insulin, alpha interferon (cancer drug), a hepatitis B vaccine, and many human growth hormones and other therapeutic proteins, he said.

New course in bioinformatics

The University of Manchester has put up a free, web-based bioinformatics course. It even has an introductory "quiz" asking questions related to the course material (I personally scored 50%, which I consider to be quite an achievment; some questions are very tough / obscure). Of course, bioinformatics is a very large field, covering very differents topics (microarrays, sequence analysis, structure prediction, databases, etc), and EMBER doesn't cover all these topics. I recommend it to anyone interested in bioinformatics, even if its just to check what the field is about.

From IST results (press release) : Bioinformatics is a relatively new scientific discipline concerning the use of computers in biological research, especially gene sequencing. The EMBER project has developed learning resources to support training in this important new field.

Although there are trained biologists and computer scientists, there are very few computer-literate biologists or biology-literate computer scientists which are needed for this cross-disciplinary field. The shortage of trained bioinformatics professionals has been paralleled by a shortage of suitable training courses.

"The term bioinformatics was coined to describe the sequence information that was emerging from the new genome projects," says Professor Terri Attwood of University of Manchester and coordinator of the IST project EMBER. "It covered things like protein sequence analysis, DNA sequence analysis and protein structure analysis. The seed partners in EMBER were University of Manchester (UK) and The Netherlands-based Expert Centre for Taxonomic Identification (ETI). ETI were the technical developers of the learning resources."

The University of Manchester has been running a M.Sc. in Bioinformatics since 1999. It includes a set of practical activities in a package called BioActivity, which had been made freely available and used in training courses worldwide. Maintaining such a course is surprisingly time consuming, with updates, revisions, hyperlink checking, and so on. EMBER was set up to convert BioActivity into a commercial product that could be maintained by technical publishers.

There are three components to EMBER: a more up-to-date and professionally produced website; a stand-alone version of the materials on CD-ROM, and a book that accompanies the course. The course is publicly available on the project website, and a simple password system is used to keep track of who is using it. "This enables course coordinators to manage presentations of the course and student cohorts, and enables students to login to the correct version of the course," adds Attwood.

"We have between 20-40 students taking the M.Sc. in bioinformatics every year," comments Attwood. "Although it's difficult to give precise figures for the numbers of students in other institutions, we know that EMBER is being used in Finland, Switzerland, Portugal, Belgium to name just a few. The University of Manchester is in discussion with two of the project partners with a view to extend the collaboration beyond the end of the project. I expect that take-up will be good."

Thursday, December 30, 2004

An unusual RNA structure in the SARS virus offers a promising target for antiviral drugs

Even if SARS is a distant memory (from last year), research is still ongoing to understand this deadly virus. Californian researchers determined by X-Ray spectroscopy the 3D structure of the RNA genome of SARS. They found unusual and very interesting features, including a rare 90 degree bend. The unique structure could be a promising target for future drug development against SARS.

The article can be downloaded for free at the PLoS journal website.

The press release from the University of California (Santa Cruz) : Research on the genome of the virus that causes severe acute respiratory syndrome (SARS) has revealed an unusual molecular structure that looks like a promising target for antiviral drugs. A team of scientists at UCSC has determined the three-dimensional shape of this structure, an intricately twisted and folded segment of RNA. Their findings suggest that it may help the virus hijack the protein-building machinery of infected cells.

The SARS virus is a type of RNA virus, meaning that its genetic material is RNA rather than the more familiar DNA found in the chromosomes of everything from bacteria to humans. All RNA viruses have relatively high mutation rates, making their genomes highly variable. In HIV, for example, this high rate of mutation contributes to the rapid appearance of drug-resistant strains of the virus. In SARS and related viruses, however, one segment of the RNA genome--known as the s2m RNA--remains virtually unchanged.

"Because viral evolution has not been able to tamper with this sequence, it is clear that it must be of vital importance to the viruses that have it, but no one knows exactly what its function is," said William Scott, an associate professor of chemistry and biochemistry.

Scott's lab used the technique of x-ray crystallography to solve the structure of this RNA element with nearly atomic resolution, revealing where every one of the many thousands of atoms that make up the structure is situated. The results showed several unique and interesting features of the s2m RNA, including a distinctive fold that appears to be capable of binding to certain proteins involved in regulating protein synthesis in cells.

"The structure gives us strong hints about the function, because it forms a fold that has been implicated in binding a certain class of proteins," Scott said. "The structure itself also provides a starting point for designing antiviral drugs that might bind to this RNA and prevent it from doing whatever it is that is vital to the life cycle of the virus."

The UCSC researchers are publishing their findings in the journal PLoS Biology (, Volume 3, Issue 1). The first author of the paper is Michael Robertson, a postdoctoral researcher in Scott's lab. Robertson and Scott purified large amounts of s2m RNA, crystallized it, bombarded the crystals with x-rays, and determined the structure from the resulting pattern of x-ray scattering.

The other coauthors, in addition to Scott, are Manuel Ares, professor of molecular, cell, and developmental biology and a Howard Hughes Medical Institute (HHMI) professor; Haller Igel, a research associate in the Ares lab; David Haussler, professor of biomolecular engineering and a HHMI investigator; and Robert Baertsch, a graduate student working with Haussler.

All of the authors are affiliated with UCSC's Center for Molecular Biology of RNA. The strong interdisciplinary connections within the RNA center were a key to making the project possible, Scott said. The investigation brought together bioinformatics experts Baertsch and Haussler, who performed the computational sequence analysis of the genomes of SARS and related viruses; molecular biologists Igel and Ares, who cloned and chemically characterized the s2m RNA; and RNA crystallography experts Robertson and Scott.

"It's true that exciting discoveries are often made at the interfaces between disciplines, but it's rare that you see it happening in such a vivid way. This is a great example of interdisciplinary science at work," said Harry Noller, Sinsheimer Professor of Molecular Biology at UCSC and director of the RNA center.

Different types of RNA perform a variety of critical tasks in all living cells. Messenger RNA is the intermediary that carries genetic information from the DNA in the chromosomes to the cellular protein factories, called ribosomes, where the genetic information is translated into proteins. The ribosomes themselves are made primarily of ribosomal RNA.

The SARS s2m RNA is in an untranslated section at one end of each of the messenger RNAs that direct the production of viral proteins in infected cells.

"It hangs on the tail end of the messenger RNA like a little molecular knob," Noller said.

Noller, an expert on the ribosome, noticed that a sharp, 90-degree bend in the s2m RNA structure is similar to a part of the ribosome. "It may only be a superficial resemblance, but you don't often see this kind of right-angle bend in RNA," Noller said.

This part of the ribosome and the proteins that bind to it are involved in the regulation of protein synthesis, leading Scott and his coauthors to hypothesize that the s2m RNA, by mimicking the ribosomal binding site, may serve to hijack the host cell's protein-synthesis machinery for use by the virus. This hypothesis will have to be tested by further studies, which are already under way in Ares's lab.

"The precise function is something they're going to figure out, no doubt about it, and it's bound to be something of major importance," Noller said. "When you see a whole class of viruses that have this absolutely conserved structural element, it tells you there's something really interesting going on here."

Sequence analysis by Haussler and Baertsch found that viruses in two families--coronaviruses (which include the SARS virus) and astroviruses--share the s2m element. About 75 percent of this sequence is absolutely invariant between viral species. Furthermore, an analysis of 38 different SARS variants found absolutely no variation within the s2m sequence.

Other scientists had previously noticed this highly conserved element in astroviruses and a few other viruses, and had given it the s2m name. But no one had any idea what the s2m RNA does that would explain why it is so highly conserved, Haussler said.

According to Scott, the UCSC team's investigation represents a novel approach in the field known as structural genomics. A more common approach in structural genomics is to determine the three-dimensional shape of a novel protein and compare it to the shapes of proteins with known functions to find clues to the function of the unknown protein.

"We have taken the methodology of conventional structural genomics and extended it to investigate the structure of the RNA genome itself," Scott said.

Ultimately, this research could lead to the development of antiviral drugs that would bind to the s2m RNA and prevent it from carrying out its function. Such drugs might be effective against a range of coronaviruses and astroviruses. While the SARS virus is the most deadly of these, other coronaviruses are common causes of respiratory infections in humans and other animals. Although none of the other human coronaviruses have the s2m RNA, several important animal pathogens do and would be susceptible to a drug that targets s2m.

Astroviruses, meanwhile, are a leading cause of gastrointestinal infections, second only to rotaviruses as a cause of childhood diarrhea. In developing countries, diarrhea is a major cause of death in children. A drug that blocks s2m could help alleviate this suffering, as well as provide another tool in the fight against SARS.

Tuesday, December 28, 2004

Affymetrix Microarray System Cleared by FDA

Affymetrix, owner and inventor of the GeneChip technology, just took a big step forward with the approval by the FDA of their first system intended for clinical use, the 3000Dx. Apparently, the system will be used to provide in vitro clinical genotyping, such as the Roche AmpliChip(TM) CYP450 Test, which allow "... diagnostic laboratories to identify certain naturally occurring variations in the drug metabolism genes, CYP2D6 and CYP2C19. These variations affect the rate at which an individual metabolizes many common drugs used to treat diseases including depression, schizophrenia, bi-polar disorder, cardiovascular disease treated with beta-blockers and others. Knowledge of these variations, when considered with other contributing factors, can help a physician select the best drug and set the right dose for a patient sooner, as well as avoid drugs that may cause the patient to suffer adverse reactions." The in vitro genotyping market powered by microarray is developing fast and the competition is fierce, but currently Affymetrix got the best microarray technology (by far).

The GCS 3000Dx System is the first microarray system to be cleared by the FDA for diagnostic testing

SANTA CLARA, Calif., Dec. 27 /PRNewswire-FirstCall/ -- Affymetrix Inc., (Nasdaq: AFFX) announced today that the U.S. Food and Drug Administration (FDA) has granted regulatory clearance for the Affymetrix GeneChip(R) System 3000Dx (GCS 3000Dx), an instrumentation system to analyze in vitro diagnostic (IVD) microarrays. This is the first microarray instrument to be cleared by the FDA for IVD use, providing a standardized platform for nucleic acid diagnostics.

"The FDA's decision is another milestone demonstrating Affymetrix' commitment to clinical diagnostic products," said Stephen P.A. Fodor, Ph.D., Chairman and CEO of Affymetrix. "We can now offer our diagnostic partners an FDA-cleared system to develop and commercialize high-quality, reproducible diagnostic assays, such as Roche Diagnostics' AmpliChip CYP450 Test which is currently offered in the European Union."

The FDA clearance of the Affymetrix GCS 3000Dx follows the August ISO certification of the instrument and array manufacturing facilities and the September IVD CE marking of the instrument system in the European Union.

Affymetrix will continue to work with its Powered by Affymetrix(TM) partners to develop genotyping and gene expression diagnostics on the GCS 3000Dx system. Under the program, Affymetrix provides technology and expertise for products that are developed, branded and marketed by the diagnostic partners.

The FDA's clearance of the GCS 3000Dx system represents a major step toward the use of microarrays for clinical applications. Since the early 1990s, Affymetrix microarrays have become the industry-standard in medical research, used in nearly 3,000 peer reviewed publications in areas such as disease classification, mutation detection and gene expression analysis. The detailed molecular snapshot that microarrays reveal should provide clinicians with more accurate diagnoses and allow for more effective treatments.

More about the 3000Dx and the test who got approved by the FDA

World's First Diagnostic Microarray System Launched by Affymetrix in European Union

GeneChip(R) System 3000Dx CE Marked in Accordance With In-Vitro Diagnostic Directive

SANTA CLARA, Calif., Sep 1, 2004 /PRNewswire-FirstCall via COMTEX/ -- Affymetrix, Inc., (Nasdaq: AFFX) today announced the availability of the world's first microarray instrument system for clinical diagnostics. The GeneChip System 3000Dx (GCS 3000Dx) is CE marked (Conformite Europeene) for in-vitro diagnostic use, enabling clinical laboratories in Europe to analyze microarray diagnostics, such as the Roche AmpliChip(TM) CYP450 Test. This test, which looks for genetic variations that can affect drug efficacy and cause adverse drug reactions, is the first CE marked microarray product launched through the "Powered by Affymetrix" partnership between Roche Diagnostics and Affymetrix.

Availability of the new GCS 3000Dx system offers Roche Diagnostics and other "Powered by Affymetrix" partners an in-vitro diagnostic platform to develop innovative genotyping and gene expression array-based assays that will help improve patient care and treatment.

"Affymetrix is committed to developing tools that will improve quality of life," said Stephen P.A. Fodor, Ph.D., Chairman and CEO of Affymetrix. "With the availability of microarray-based diagnostics, the healthcare community will now have access to a scalable and affordable technology to analyze the genetics of human disease."

The GCS 3000Dx instrumentation system comprises a GCS 3000Dx Scanner with AutoloaderDx, a fluidics station FS450Dx, and GCOSDx software. This is an extension of the same platform that has been used in over 1,000 clinical research publications, but is now configured for diagnostic use in the European Union. The release of the Amplichip CYP450 Test builds on Affymetrix' recent ISO certification for its instrumentation and microarray manufacturing facilities, an achievement confirming the highest levels of quality assurance.

The Roche AmpliChip CYP450 Test will allow diagnostic laboratories to identify certain naturally occurring variations in the drug metabolism genes, CYP2D6 and CYP2C19. These variations affect the rate at which an individual metabolizes many common drugs used to treat diseases including depression, schizophrenia, bi-polar disorder, cardiovascular disease treated with beta-blockers and others. Knowledge of these variations, when considered with other contributing factors, can help a physician select the best drug and set the right dose for a patient sooner, as well as avoid drugs that may cause the patient to suffer adverse reactions.

Affymetrix is working to develop a network of partnerships with companies like Roche Diagnostics to create novel array-based diagnostics in both the clinical and non-clinical markets. With this "Powered by Affymetrix" program, Affymetrix manufactures the microarrays and instruments, while the diagnostic partner develops and commercializes the test. The resulting microarray products enable users to examine genomic sequences in more detail than ever before, providing a more efficient and complete method to diagnose a wide range of conditions and create potential solutions.

Monday, December 27, 2004

Beacon Designer 4.00 Released

If you're into qRT-PCR, Premierbiosoft's Beacon Designer (a primer-design software) version 4 just got released. Lots of primer design programs are available on the web, but this one is specialized in qRT-PCR (SYBR green primers, dual-labeled taqman primers and probes, molecular beacons).

It analyze secondary structures of the template / probes, it avoid regions of homology by doing BLAST analysis automatically... it's really a wonderful piece of software (I used it extensively in the past). Primers it design just work, everytime. Many new features we're added in version 4. Unfortunaly, it got a big problem... the price is very high! 2000$ USD for a license... you have to design quite a few primers to justify buying this (excellent) software. Too bad they don't have a toned down, cheaper version... Note that they do have a limited demo available for download. New features in version 4.00 include :

1. SYBR Green primer design mode: design primers for SYBR Green assays. If you choose, these primers can be exported to the TaqMan®, FRET or molecular beacon design modes to design compatible dual labeled probes.

2. FRET probe design mode: design optimal FRET probes and compatible primers using this module. A list of alternate primer and probes is made available for you to choose the most appropriate set for your needs. You will also be able to evaluate pre-designed primers and probes.

3. Ability to generate an attractive report of the designed assays: You will now be able to create an attractively formatted report for the assays you designed. It should be helpful in record keeping and for sharing information with colleagues. The report helps visualize the positions of the primers and probes on the sequence, includes a list of the alternate primers and probes, displays primers, probe, amplicon and sequence properties and the design parameters used.

4. Includes support for the latest databases available at NCBI for BLAST search.


First whole week of vacation time since last year (at the same period), and I can't say I don't appreciate it :) I found an internet connection (dial-up... soooo slow) so I can post news again. In this period of , I want to express my gratitude for the support and interest you've shown for this blog. I'm considering moving to a .com domain (still have to decide to keep TheScientistBlog or to switch to another name) with MovableType instead of Blogger as a platform. Blogger is nice with the free hosting and all, but it has several limits for my purpose. If you have suggestions, the discussion forum is online with 2 users (counting me!) :)

Happy Holidays everyone!

P.S. : As many surfers interested in bioinformatics come from India, a special thought for those in the Indonesia-India region who got hit by the earthquake / tsunami... I lived throught a very minor flood, compared to a tsunami... so I can try to imagine what you're living throught.

Friday, December 24, 2004

Scientists Identify Protein Critical to Melanoma Growth

Some hope for a Melanoma (skin) cancer cure : researchers from the Dana-Farber Cancer Institute (Boston) discovered that this particular type of cancer is dependant on a cell cycle control protein, CDK2. The finding suggest that by cutting off the supply of this protein, cancerous cells would be unable to multiply; it wouldn't destroy the cancer, but inhibit its growth.

Newswise press release : Researchers at Dana-Farber Cancer Institute and Children’s Hospital Boston have discovered that malignant melanoma, the potentially lethal skin cancer, can’t grow without a steady supply of a protein that normal cells can do without.

The findings, which are published in the December issue of Cancer Cell, suggest that drugs that cut off melanoma cells’ supply of the protein, called CDK2, might curb the growth of the dangerous skin cancer in patients, and with relatively low toxicity.

In theory, such a drug would leave normal cells unharmed and have many fewer side effects compared to standard chemotherapy.

Working with melanoma cells grown in the laboratory, the researchers, led by David E. Fisher, MD, PhD, Director of the Melanoma Program at Dana-Farber and the paper’s senior author, showed that adding a chemical that quashed the activity of CDK2, the gene that manufactures CDK2 protein, dramatically slowed the growth and proliferation of the cancer cells. Unlike conventional chemotherapy drugs, a CDK2 inhibitor drug wouldn’t be aimed at killing melanoma cells, only halting their growth.

Fisher said that CDK2-inhibiting drugs exist, and he hopes that the research results will soon lead to clinical trials of them in patients with melanoma.
The study’s lead author is Jinyan Du, PhD, who carried out the project while working as a student in Fisher’s lab at Dana-Farber. Fisher is also a pediatric oncologist at Dana-Farber/Children’s Hospital Cancer Care.

The CDK2 gene and its protein (an enzyme) are one of several regulators of the cell cycle: That is, they help determine when a cell should be “resting” and when it should begin dividing to make more of itself. When cells become malignant, it is in part because their normal controls on growth and division are disabled, and they proliferate abnormally. Overactive CDK2 has been found in many types of cancer, making it a prime candidate for designer drugs that would turn down CDK2 activity and, it was hoped, slow the runaway growth of cancer cells.

Recent research, however, had thrown cold water on the notion. Studies have shown that tumor cells in a variety of cancers weren’t dependent on CDK2 for growth. Thus, blocking its activity had little effect on the out-of-control cells.
The scientist’s report today is all the more striking, because it reveals that melanoma does require the CDK2 enzyme for growth. Why this is so isn’t clear, but the finding revives the strategy of using CDK2 inhibitors as a potential treatment – even if only for this one form of cancer. And, since it’s been previously shown that normal cells can divide and grow normally without CDK2 (other types of CDK molecules apparently can take over the job) “this is good news, because it means there may be little toxicity to a person who would receive a CDK2 inhibitor to treat melanoma,” says Fisher, who is also an associate professor of pediatrics at Harvard Medical School.

Melanoma will cause about 7,900 deaths this year in the United States, according to the American Cancer Society. Its incidence has been rising rapidly over the past several decades: about 55,000 cases are expected in 2004. Most cases caught early can be cured, but if melanoma cells penetrate the skin deeply, the cancer is highly prone to spread with life-threatening consequences despite treatment with surgery, chemotherapy and radiation.

The new findings stem from Fisher’s longtime work on a gene called MITF that regulates the development of skin pigment-producing cells called melanocytes. Regulatory genes like MITF act on other genes in a chain-of-command fashion. When Fisher’s group looked for genes regulated by MITF, they found a pigment gene called SILVER, and they noted that, surprisingly, it was located just a stone’s throw, genetically, from CDK2 on the chromosome.

“It was dumb luck,” says Fisher, and it led him and his colleagues to recognize that both SILVER and CDK2 were under the control of MITF. In all other body cells besides melanocytes, CDK2 is not subservient to MITF: To the researchers this was an important clue.

“If the control of CDK2 expression is so different in the development of melanocytes, then maybe the requirement for CDK2 in melanoma is different than in other cancers,” he says – and the new findings confirm this idea.

The fact that melanoma cells, unlike other cancer cells, become “overdependent” on the CDK2 protein while normal cells don’t need much of it provides a “therapeutic window.” That is, a drug that suppresses melanoma growth by shutting down CDK2 in theory could control the cancer yet have little toxic effect on the body.

In addition to Fisher and Du, the paper’s other authors include researchers from the Broad Institute at the Massachusetts Institute of Technology and Harvard University, MIT, and Massachusetts General Hospital.

The research was funded by the National Institutes of Health.

Dana-Farber Cancer Institute is a principal teaching affiliate of the Harvard Medical School and is among the leading cancer research and care centers in the United States. It is a founding member of the Dana-Farber/Harvard Cancer Center (DF/HCC), designated a comprehensive cancer center by the National Cancer Institute.

Children's Hospital Boston is home to the world's largest research enterprise based at a pediatric medical center, where its discoveries have benefited both children and adults for over 100 years. More than 500 scientists, including eight members of the National Academy of Sciences, nine members of the Institute of Medicine and 10 members of the Howard Hughes Medical Institute comprise Children's research community. Children's also is the primary pediatric teaching affiliate of Harvard Medical School. For more information about the hospital visit:

Thursday, December 23, 2004

Bioinformatics at SourceForge

Got programming skills? Wanna get involved in Bioinformatics? SourceForge got a whole category devoted to bioinformatics, with lots of active open-source projects going on and needing some help. Microarray and mass spectrometry data analysis, 3d molecular visualization, sequence assembly and alignement, the list goes on and on... Get involved : open-source development is a very good thing for the scientific community.

Scientists Show How Proteins Beat the Evolutionary Stakes

Molecular evolution - how did it arise? How did all these proteins aquire their (in some case multiple) functions? A new study published in Nature Genetics shed some light on the subject. They simulated evolution by mutating multiple-functions proteins, to see if all functions were affected at the same time and to gain a better understanding of evolution.

According to Newswise : Evolution is something of a gamble: in order to stay a step ahead of a shifting environment, organisms must change or risk extinction. Yet the instrument of this change, mutation, carries a serious threat: mutations are hundreds of times more likely to be harmful to the organism than advantageous. Now, in a paper published online Nov. 28 in Nature Genetics, a team of scientists at the Weizmann Institute of Science has shown one way that evolving organisms may be hedging their bets.

Dr. Dan Tawfik, who headed the team from the Biological Chemistry Department, believes that proteins with so-called promiscuous or moonlighting activities can provide nature with ready-made starting points for the evolution of new functions. Proteins that have evolved to perform a given function often have the ability to take on other, often completely unrelated tasks as well. For example, one of the enzymes studied by the group, PON1, is known to remove cholesterol from artery walls, as well as to break up a certain chemicals used as pesticides. Yet its main function is to act as a catalyst for the removal of a class of compounds called lactones that have no connection at all to the other two.

To investigate what kind of evolutionary advantage promiscuity offers, the team created a speeded-up version of evolution in the lab. Mutations were introduced into the genes coding for various proteins in a completely random manner. Evolutionary pressure was then simulated by selecting those mutants with higher levels of activity in one of the promiscuous traits.

After several rounds of mutation and selection, the scientists looked at their enzymes to see what had changed. As expected, they had managed to increase the activity they were selecting for by as much as a hundredfold and more. But how did increasing one skill affect the others?

Interestingly, the levels of the other promiscuous activities also underwent drastic changes. In most cases, the levels dropped dramatically, though in some there was a significant increase. However, the primary function of the enzymes, the one for which they had originally evolved, changed hardly at all. “This is particularly surprising when you consider that all of these activities take place at the exact same site on the enzyme,” says Tawfik.

This phenomenon makes sense when viewed in evolutionary terms. “Two contradictory things are necessary for the survival of organisms,” he says. “First of all, an organism needs to be robust in the face of mutation – it needs to undergo as little change as possible in its functioning in spite of mutations. But, evolutionary adaptation requires some mutations to induce new traits. It appears that the organism can have it both ways: the main function remains robust while the promiscuous functions are extremely responsive to mutation.”

The scientists believe that promiscuity may be an intermediate phase for some evolving proteins. In the face of further evolutionary pressure, the protein line could split, diverging into two distinct genes. This multi-tasking may also partly explain another phenomenon that has been puzzling biologists: rapidly emerging drug and antibiotic resistance, and enzymes that have adapted to break down man-made chemicals that have only been around for 50 years. Natural evolution, according to standard theory, should take thousands and hundreds of thousands of years to work. The key may be in promiscuous functions that have never been under selection pressure. These latent “underground” skills may provide the evolutionary shortcut needed for rapid adaptation.

Dr. Dan Tawfik's research is supported by the Y. Leon Benoziyo Institute for Molecular Medicine, the Dolfi and Lola Ebner Center for Biomedical Research, the Estelle Funk Foundation, the Dr. Ernst Nathan Fund for Biomedical Research, the Henry S. and Anne Reich Family Foundation, The Harry and Jeanette Weinberg Fund for Molecular Genetics of Cancer and the Eugene & Delores Zemsky Charitable Foundation Inc.

Dr. Tawfik is the incumbent of the Elaine Blond Career Development Chair.

The Weizmann Institute of Science in Rehovot, Israel, is one of the world's top-ranking multidisciplinary research institutions. Noted for its wide-ranging exploration of the natural and exact sciences, the Institute is home to 2,500 scientists, students, technicians and supporting staff. Institute research efforts include the search for new ways of fighting disease and hunger, examining leading questions in mathematics and computer science, probing the physics of matter and the universe, creating novel materials and developing new strategies for protecting the environment.

Tuesday, December 21, 2004

Genomatix Suite 3.3 released

Today version 3.3 of Genomatix popular in silico promoter analysis suite got released. Notable quote from their mailing list press release :

"We are currently writing up a paper about a very promising approach to micro array analysis, which has now many times proved to be successful in identifying small sets of co-regulated genes. The strategy includes identification of involved TF binding sites and framework construction leading to regulatory network analysis. If you would like to know more on that, just let me know. I can send you a explanatory Powerpoint presentation."

Can't wait to see what its about... Microarray analysis would greatly benefit from a biologically accurate co-regulation analysis package. It would help to build hypothesis about regulation mechanism (which transcription factor is likely to be involved, which signalization cascade, which receptor / pathway, etc). Changelog available in the full post for those who are interested.

New features and main changes from Release 3.2:
ElDorado / Gene2Promoter

* New genomes:
o Canis familiaris (Genomatix build 1)
o Pan troglodytes (Genomatix build 1)
* Comparative Genomics includes chicken, dog, and chimpanzee in addition to human, mouse, and rat.
* Gene2Promoter allows input of gene names. Optionally, all orthologous genes can be included in the output.


* Shorter startup time, reduced memory usage and increased filter speed.
* Improved graph layout algorithm for large cluster layouts.
* Search for MeSH identifiers in BiblioSphere filters enabled.
* Comparative Genomics analysis can be started from gene context menu.
* New feature "Show/hide ghosts": input genes that don't comply with filter criteria (ghosts) can be hidden.
* Preselection of most likely candidate in ambiguous gene lists.
* Input of larger gene lists (up to 1000 genes) supported.

Monday, December 20, 2004

Success of Experimental Herpes Vaccine Builds Momentum for Human Clinical Trials

More news from the Harvard Medical School : "A new study provides evidence that a herpes vaccine developed by a Harvard Medical School researcher is a strong candidate for testing in humans. The study, published online Dec. 14 in the Journal of Virology, compared three different experimental vaccines for herpes simplex virus 2 (HSV-2), the virus that causes most cases of genital herpes."

HSV-2 infects one in five Americans, and its prevalence has reached 50 percent in some developing countries, where it also seems to be helping to fuel the spread of HIV. HSV-2 infection, though incurable, typically does not cause major health problems, but can be life-threatening in immunocompromised people and newborn babies infected by their mothers.

Lead author Stephen E. Straus, MD, senior investigator in the Medical Virology Section in the Laboratory of Clinical Infectious Diseases at the National Institute of Allergy and Infectious Diseases of the National Institutes of Health, tested the vaccines in two established animal models of herpes infection. The HMS vaccine, developed by David Knipe, HMS professor of microbiology and molecular genetics, called dl5-29, outperformed the other two vaccines, one of which has already been tested in humans.

Straus said that the results argue strongly for taking dl5-29 into human trials. "Based upon d15-29's biological and immunological properties, it appears to be the most compelling new vaccine candidate for genital herpes," he said.

Straus said that dl5-29 seemed especially promising because it solves a critical problem that is believed to underlie the failure of previous candidate vaccines. The dominant approach to herpes vaccine development over the past 20 years has been the delivery of one or two pure glycoproteins found in the outer envelope of the virus in order to induce an antibody response. But in trials of HSV-2 vaccines, a healthy antibody response has not seemed sufficient to protect against infection. One version of the herpes glycoprotein vaccine failed to protect research subjects from HSV-2 infection, while a second version of the glycoprotein vaccine failed to protect men, but showed a protective effect only in the subset of women who also had not been infected previously with HSV-1, the common cause of cold sores.

In contrast, dl5-29 is a live, mutant strain of HSV-2 that is missing two of the genes necessary for it to replicate and persist inside its host. "The proteins that are expressed are able to induce immune responses but the virus can't spread," said Knipe, who is a coauthor on the paper. Normally, HSV-2 infects the cells lining genital areas, but makes its way into nearby sensory neurons, where it persists in a latent state. Because dl5-29 actually enters host cells and expresses many of its proteins within them, it not only elicits a broad spectrum of antibodies but also stimulates T cells, which directly attack infected host cells and release cytokines that further strengthen the immune response. The clinical trials of previous herpes vaccines suggested that T cells as well as antibodies must be activated to launch an effective defense.

Straus compared dl5-29 with a glycoprotein vaccine previously tested in humans and a third vaccine comprising a naked circular strand of DNA encoding the glycoprotein. Naked DNA vaccines have generated interest in recent years for their potential to elicit a stronger cellular immune response than by simply injecting the protein. Straus said that he tested dl5-29 against "the best tested standard vaccine plus the competing new concept in the field, DNA vaccines," in order to get a better sense of how well the dl 5-29 vaccine performed. His team tested the vaccines both in mice and in guinea pigs. The latter is the best model of human HSV-2 disease because it is the only one that mimics many of the aspects of the human disease, such as a recurring infection interspersed with periods of latency. The researchers studied how well the vaccines worked prophylactically-to prevent infection-and therapeutically to help control an existing infection.

Straus and his colleagues at the NIH found that in all measures dl5-29 performed as well or better than the other two candidates. It was as effective as the glycoprotein vaccine in preventing acute and recurrent disease in guinea pigs. Moreover, when given therapeutically to previously infected guinea pigs, dl5-29 reduced the rate of recurrent infections slightly better than the other candidates. A key finding was that dl5-29 also induced a substantially stronger T cell response than either of the two other vaccines.

Additionally, dl5-29 stimulated stronger antibody responses in animals than either of the other vaccines. Straus and Knipe said this result was surprising because it was thought that the large quantities of a single glycoprotein, as used in many recent human trials, was enough to stimulate sufficient levels of antibodies. Knipe said that as a live virus, dl5-29 produces many more viral proteins, and perhaps the resulting broader antibody response is important in preventing infection.

Because many other candidate vaccines have prevented infection in animals but failed in humans, the results do not guarantee success. But Straus observed that the stronger T cell response provides a major theoretical advantage for dl5-29 that could translate into greater clinical effectiveness in people. "The vaccine induced very good levels of immunity of the antibody type. It induced far better levels of immunity of the cellular type. It was enormously safe, and didn't seem to persist in the animals," said Straus. "With dl5-29, we believe there are now sufficient data to justify clinical studies."

DNA For Information Processing and Data Storage

Researchers from the University of Minnesota found a way to use DNA as a scaffold to build circuits to process or store data. They developped a technique allowing them to assemble DNA in predetermined patterns and create a circuit template. Using special DNA containing docking sites at regular intervals, they can then direct the assembly of microscopic electrical components. Interesting quote : "The scaffolding has the potential to self-assemble components 1,000 times as densely as the best information processing circuitry and 100 times the best data storage circuitry now in the pipeline." Another cool biotechnology application emerging from out-of-the-box thinking.

From the press release : The DNA molecule—nature’s premier data storage material—may hold the key for the information technology industry as it faces demands for more compact data processing and storage circuitry. A team led by Richard Kiehl, a professor of electrical engineering at the University of Minnesota, has used DNA’s ability to assemble itself into predetermined patterns to construct a synthetic DNA scaffolding with regular, closely spaced docking sites that can direct the assembly of circuits for processing or storing data. The scaffolding has the potential to self-assemble components 1,000 times as densely as the best information processing circuitry and 100 times the best data storage circuitry now in the pipeline. Members of the team first published their innovation in 2003, and they have now refined the technique to allow more efficient and more versatile assembly of components. The new work, which was a collaborative effort with chemistry professors Karin Musier-Forsyth and T. Andrew Taton at Minnesota and Nadrian C. Seeman at New York University, is reported in the December issue of Nano Letters, a publication of the American Chemical Society.

“There’s a need for programmability and precision on the scale of a nanometer—a billionth of a meter—in the manufacture of high-density nanoelectronic circuitry,” said Kiehl. “With DNA scaffolding, we have the potential for arranging components with a precision of one-third of a nanometer.

“In a standard silicon-based chip, information processing is limited by the distance between units that process and store information. With DNA scaffolding, we can lay out devices closely, so the interconnects are very short and the performance very high.”

The DNA scaffolding is made from synthetic DNA “tiles” that spontaneously assemble in a predetermined pattern to form a sheet of molecular fabric, much like corduroy. The ripples in the fabric are formed by rows of sticky DNA strands that occur at regular intervals in the scaffolding and function as a strip of Velcro® hooks that fasten to nanocomponents coated with matching DNA strands. The nanocomponents could be metallic particles designed to process or store data in the form of an electrical or magnetic state, or they could be organic molecules—whatever would best process or store the information desired.

In the earlier work, members of the Kiehl team made DNA scaffolding with regularly spaced gold nanocomponents pre-woven into the fabric, completing the synthesis all in one operation. Now, the team first makes DNA scaffolding with regularly spaced sticky DNA strips and then adds the nanocomponents, which stick to the DNA strips in rows. This allows them to use optimal synthetic methods for both steps. It’s analogous to using strips of Velcro® in cloth: It’s much easier to get a useful pattern by first weaving cloth and Velcro® strips together, and then attaching beads or other objects to the strips later, than it is by adding the objects during the weaving process.

The new procedure also lets the team add any one of various nanocomponents--such as other metals, organic molecules or tiny electronic devices—at a later time, depending on what is needed for the application. The result is a more perfect scaffolding, better and more regular attachment of electronic units, and more diversity in the types of units and the types of circuitry that can be made.

“We can now assemble a DNA scaffolding on a preexisting template, such as a computer chip, and then—on the spot—assemble nanocomponents on top of the DNA,” said Kiehl. The nanocomponents can hold electrical charge or a magnetic field, either of which would represent a bit of data, and interactions between components can act to process information. Circuitry based on regular arrays of closely spaced components could be used for quickly recognizing objects in a video image and detecting motion in a scene — slow and difficult tasks for conventional computer chips. The technology could help computers identify objects in images with something approaching the speed of the human eye and brain, Kiehl said. The technology could also be used for various other applications, including chemical and biological sensing, in which case the strips would be designed to stick to the tiny objects or molecules to be detected.

Discovery of first demethylase molecule, a long-sought gene regulator

DNA methylation plays a big role in gene regulation - methylation adds a small "tag" (a methyl group) to cytosine (the "C" letter of DNA) - this "silence" the gene in higher eukaryotes and it don't get expressed. It allows for time/space dependant expression. Until now, enzyme who mediate the addition of the methyl group were known, but not those charged of removing it (to unsilence the gene). Researchers from the Harvard Medical School just found the first one.

From the press release : "Researchers have discovered an enzyme that plays an important role in controlling which genes will be turned on or off at any given time in a cell. The novel protein helps orchestrate the patterns of gene activity that determine normal cell function. Their disruption can lead to cancer."

The elusive enzyme, whose presence in cells was suspected but not proven for decades, came to light in the laboratory of Yang Shi, HMS professor of pathology, and is described in a study published in the Dec. 16 online edition of Cell and appearing in the Dec. 29 print edition.

"This discovery will have a huge impact on the field of gene regulation," said Fred Winston, an HMS professor of genetics who was not involved with the work. "Shi and his colleagues discovered something that many people didn't believe existed."

The enzyme, a histone demethylase, removes methyl groups appended to histone proteins that bind DNA and help regulate gene activity. "Previously, people thought that histone methylation was stable and irreversible," said Shi. "The fact that we've identified a demethylase suggests a more dynamic process of gene regulation via methylation of histones. The idea of yin and yang is universal in biology; our results show that histone methylation is no different."

In the cell, yarnlike strands of DNA wrap around protein scaffolds built of histones. The histones organize DNA into a packed structure that can fit into the nucleus, and the packing determines whether the genes are available to be read or not. Acetyl, methyl, or other chemical tags appended to the histones determine how the histones and DNA interact to form a chromatin structure that either promotes gene activity or represses it.

Some histone tags, particularly acetyl groups, are known to be easily added and removed, helping genes to flick on and off when needed. But the addition of methyl groups was considered a one-way process that could only be reversed by the destruction of histones and their replacement with new ones. Part of the reason scientist believed this was that no one had isolated a demethylase, despite an active search.

The Shi lab was not among those in the hunt, but they stumbled onto the demethylase while probing the function of a new gene repressor protein. Postdoctoral fellow Yujiang Shi had exhausted the likely possibilities for how the mystery protein worked to suppress gene activity, so one day he tried an unlikely experiment. He had the purified protein in a test tube and decided to feed it methylated histones. His finding, that the enzyme could efficiently chew off the methyl group, leaving behind intact, unmodified histone left the postdoc Shi shaking with excitement. "Forty years ago some scientists speculated that histone demethylases existed," he said. "At first, I thought it was impossible that this protein was it." After reproducing the results using several different biochemical techniques, he began to feel comfortable that they had found the first demethylase.

Their enzyme didn't remove just any methyl group from histone. Instead, it removed a very specific methyl found on lysine 4 (K4) of histone 3 (H3). H3K4 methylation is associated with active transcription, so its removal would be consistent with the gene repression function they had identified.

Now that the first demethylase has been recognized, researchers will certainly find more. "This cannot be the only demethylase," said Shi.

Genes turning on at the wrong time or in the wrong place is a hallmark of cancer cells. In some tumors, high levels of methylation of H3K4 seem to play a role in activating genes that drive abnormal cell growth. The discovery of this H3K4 demethylase suggests a way to counterbalance this progrowth signal in some tumors. And if previous experience with histone deacetylases is any guide, the demethylases could one day be targets for cancer therapeutics.

"These findings will impact every walk of biology," said David Allis of Rockefeller University, a leader in studying the regulation and biological roles of histone tags. "Histone modifications are highly dynamic on-off switches that the cell throws a lot. These modifications affect everything DNA does, and getting the enzyme means you've got one upstream point of regulation. This will open up a wealth of new experiments."

Emerging Class of Viruses Found to Change Shape to Infect Humans

Interesting discovery about the method used by some (positive-stranded RNA) flaviviruses to get prefentially translated by the cellular machinery. It involve a protein complex with a specific 3d structure(instead of a poly-A tail) which cellular enzyme recognize. It could potentially lead to specific inhibitors targeting this class of virus.

Press release : "The binding of a viral RNA and a viral protein brings about a physical transformation that dupes host cells into enthusiastically copying the invading pathogen, according to a team of researchers from Harvard Medical School (HMS) and Massachusetts Institute of Technology (MIT).

In the December 17 issue of Science, collaborators led by professor Lee Gehrke of the Harvard-MIT Division of Health Sciences and Technology publish dramatic three-dimensional images of RNA-protein interactions in alfalfa mosaic virus (AMV), a safe model for investigating single-strand, positive-sense RNA viruses. AMV's dangerous relatives include flaviviruses that cause dengue fever, Japanese encephalitis and West Nile disease.

Gehrke and other molecular virologists knew that AMV was not infectious unless its genomic RNAs bound viral protein, but the details were unknown. Laura Guogas, a graduate student in Gehrke's lab, decided to seek answers with x-ray crystallography."

"What Guogas found is "stunning and unexpected," says James Hogle, an structural biologist and professor of biological chemistry and molecular pharmacology at Harvard Medical School. He and David Filman, also of HMS, contributed to this study.

RNA binding turned the viral coat protein from a floppy coil into a tight, springy helix. The RNA, a smooth strand punctuated by bumpy "hairpin structures," developed a kink that looks like a mountain turn on the Tour de France. The researchers attribute this kink to the formation of additional links between the two sides of the hairpins, another surprise from the three-dimensional structure. RNA and protein fold together in a way that locks them into place.

This distinctive, stable structure turns one end of the viral RNA into a handsome stranger. "It sticks out like a beacon compared with other RNAs in the cell," says Gehrke, who proposes that the host cell's replicating enzyme "jumps right on" and begins making more copies of the infecting virus.

Ordinarily, the translation of viral RNA into protein is triggered by a string of a particular RNA building block, adenosine, at one end of a typical RNA, a so-called "poly-A tail" that flaviviruses lack. AMV substitutes the striking RNA-protein complex that Guogas identified; other viruses in the family probably form different structures that make the ends of their RNA attractive to the cell's replicating machinery.

Future research will look for ways to translate differences between cellular and flavivirus RNAs into vaccines and treatments for dengue fever, West Nile virus, and similar emerging infections. The researchers hope to build on the synergy between biochemistry and structural biology demonstrated by Guogas's study. "This project is a great example of the role a talented student can play in a collaboration between two labs with complementary interests and expertise," says Hogle."

Sunday, December 19, 2004

Mymetics Receives NIH Approval to Advance HIV Vaccine Into Late Preclinical Studies

"Mymetics Corporation announced today that it has received approval from the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health (NIH) to begin advanced preclinical testing of the Company's trimeric gp41 vaccine in nonhuman primates. The study will run through the end of the third quarter of 2005, testing the second generation of HIV vaccines in development. Based on the results of the NIH-approved study, Mymetics expects to initiate advanced toxicology testing by the end of 2005 in preparation for filing an Investigational New Drug (IND) application in 2006."

Their vaccine approach is based on a concept not generaly accepted in the scientific community : they think that gp41 (HIV protein responsible for fusion with the cell to be infected) "subtly" mimics IL-2, an immune system protein. For this reason, they think that the immune system is somewhat unable to produce effective antibodies against HIV. I can't say I'm convinced by this, but the vaccine approach (trimeric gp41) might be worth trying. However, note gp41 is usually masked by gp120, so antibodies can't easily reach it.

Dr. Sylvain Fleury, Chief Scientific Officer of Mymetics, commented, "We are greatly pleased to have received our first formal recognition from the NIH for our program in HIV vaccinology. We have had strong results in this program to date, including neutralizing antibodies capable of blocking transcytosis and primary T cell and macrophage infections by primary HIV isolates. Our goal with the NIH-approved study is to further examine the potential of our gp41 vaccine approach and to gain additional data, including immunogenicity and stability, relevant to advancing our next-generation product."

Mymetics' Chief Executive Officer, Christian J.-F. Rochet, stated, "The NIAID-approved studies will place us among a small group of research teams poised to enter the clinic with a promising vaccine candidate. As we have previously announced, we anticipate signing in 2005 a partnership agreement with a major health sciences company to assist us in human clinical trials, for which we plan to file by 2006, pending the results of our preclinical program."

In 2003, Mymetics created its first-generation trimeric gp41 HIV vaccine candidate. The Company is now developing next-generation vaccines that contain a more complete sequence of the wild-type gp41, including key neutralizing epitopes. A primary objective in the Company's vaccine design program is to impair the molecular mimicry between gp41 and the IL-2 cytokine (Interleukin-2) of the infected host. Mymetics' researchers discovered this mimicry in 1997 and believe that it is a major reason underlying the shutting down of the immune system seen in patients with HIV and AIDS.

Mymetics Corporation is a biotechnology company focused on the development of human and animal vaccines and therapies in the field of retroviral and viral autoimmune diseases, including HIV-1 infection. The Company's key discovery is a fundamental though subtle three-dimensional mimicry between the viral envelope glycoprotein gp41 of HIV-1 and the IL-2 cytokine (Interleukin-2) of the infected host. Based on this understanding of molecular mimicry, Mymetics has been able to engineer gp41 proteins capable of eliciting neutralizing antibodies against primary HIV-1 strains and has also designed specific therapeutic molecules which have the potential to prevent and/or delay the disease. Mymetics' platform technology can also be applied to other retrovirus-related diseases that involve similar mimicries, including certain oncoviruses often associated with human leukemia.

Saturday, December 18, 2004

First U.S. SARS Vaccine Trial Opens at NIH

Powerful research tools that speed up vaccine development have led to the start today of human tests for a preventive vaccine against the respiratory disease SARS. The disease killed hundreds of people around the world before it was brought under control in 2003 with aggressive conventional public health measures.

Researchers at the Vaccine Research Center, National Institute of Allergy and Infectious Diseases (NIAID), part of the National Institutes of Health (NIH), will conduct the trials. The experimental vaccine against SARS, or severe acute respiratory syndrome, will be tested on 10 healthy volunteers at the NIH Clinical Center in Bethesda, MD. The clinic will do periodic follow-up exams on each volunteer for 32 weeks.

“This experimental vaccine is an outstanding achievement by NIAID researchers,” said Health and Human Services Secretary Tommy G. Thompson. “It is a model for research that could greatly shorten the time needed to create vaccines to be tested against other diseases.”

“The Vaccine Research Center, a cutting-edge facility established here at NIH just five years ago, encompasses the entire spectrum of vaccine development from basic research to clinical testing,” says NIH Director Elias A. Zerhouni, M.D. “This is why our team at NIAID has been able to develop this vaccine at an unprecedented pace, using technological discoveries that were not available just a few short years ago.”

The primary goal of the study is to determine if the experimental vaccine is safe in people. A secondary goal is to assess how well the vaccine stimulates the immune system to produce antibodies and cellular immunity, in this case, focusing on the SARS spike protein. The spike protein protrudes from the virus' outer envelope and helps it bind to cells it infects.

SARS was spotted first in China in November 2002. The virus sickened 8,096 and killed 774 worldwide by July 2003, according to the World Health Organization (WHO). SARS was brought under control with classic public health techniques: epidemiological investigations, patient isolations, quarantines of exposed people and stringent restrictions on travel.

The sudden appearance of SARS, its severity, and its ability to be spread far and fast by international travelers, spurred medical researchers. NIAID researchers developed the vaccine with unprecedented speed. Just 21 months passed from when international health officials recognized SARS as a new infectious disease to the opening of the NIAID human clinical vaccine trial. It often takes decades for scientists to develop a successful vaccine against an infectious disease.

“In the case of SARS, we have dramatically cut vaccine development time with powerful new tools from two different fields, molecular biology and information technology,” says Anthony S. Fauci, M.D., director of NIAID.

Instead of using weakened or inactivated virus, which is typical for vaccine development, the new vaccine is composed of a small circular piece of DNA that encodes the viral spike protein. Scientists modified the DNA to minimize the risk of it combining with the SARS virus or other viruses of the SARS type, called coronaviruses.

Scientists expect that the DNA will direct human cells to produce proteins very similar to the SARS spike protein. The immune system should recognize these proteins as foreign and then mount a defense against them. If the vaccinated person ever encounters the actual SARS virus, his or her immune system will be primed to neutralize it.

“It is truly remarkable that less than two years ago we were facing an unknown global health threat, and now we are testing a promising vaccine that may help us to counter that threat should it re-emerge,” Dr. Fauci said.

After SARS was identified as a disease, researchers worked hastily to identify the cause of the mysterious respiratory ailment and to develop therapies and vaccines. By April 2003, NIAID-funded researchers in Hong Kong were the first to show that SARS is a viral disease. They soon proved that a newly emerging coronavirus causes SARS. By May, an international collaboration of researchers had decoded the genetic sequence of the SARS coronavirus, opening many avenues of research to develop diagnostic tests, therapies and vaccines.

An NIAID team, lead by NIAID Vaccine Research Center Director Gary J. Nabel, M.D., used the available SARS coronavirus genomic information to develop a vaccine based on the gene for the SARS spike protein. The vaccine performed very well in mice, reducing the levels of virus in the lungs of infected mice by more than a million-fold, Dr. Nabel and colleagues reported in Nature in March 2004.

“Two years ago, we didn't know that this virus existed. Today, we begin clinical trials of a promising vaccine candidate. We owe the speed of this research to modern molecular genetics. The technology enables us rapidly to translate scientific discoveries into clinical interventions and improves our ability to battle these ever-evolving, highly lethal microbes,” says Dr. Nabel.

Under a contract with NIAID, Vical Inc. of San Diego, CA, is producing the SARS vaccine for the NIAID clinical trial. For more information on the SARS vaccine trial, phone the Vaccine Research Center’s toll free number 1-866 833-LIFE, or visit the Vaccine Research Center web site.

Chinese researchers began human testing of a SARS vaccine in May of this year. The Chinese vaccine trial uses an inactivated SARS virus vaccine developed through conventional vaccine technology.

While the bulk of SARS cases were in China, Hong Kong and Singapore, eventually cases also occurred in Canada, Europe and the United States, according to WHO. There were 27 probable SARS cases in the United States. No U.S. residents died of the disease, according to WHO.

NIAID is a component of the National Institutes of Health, an agency of the U.S. Department of Health and Human Services. NIAID supports basic and applied research to prevent, diagnose and treat infectious diseases such as HIV/AIDS and other sexually transmitted infections, influenza, tuberculosis, malaria and illness from potential agents of bioterrorism. NIAID also supports research on transplantation and immune-related illnesses, including autoimmune disorders, asthma and allergies.

Study identifies key aspect of immune response against HIV

BOSTON - December 8, 2004 - An international research team has identified immune-system genes that appear to play a key role in the body's defense against HIV, the virus that causes AIDS. The findings may lead to ways of circumventing the virus's ability to avoid vaccines by rapid mutation. The study in the Dec. 9 issue of Nature also describes how HIV infection is driving human evolution, since individuals with protective versions of the identified genes are more likely to survive and pass those genes along to children. Including researchers from the University of Oxford and the University of KwaZulu-Natal in South Africa, the investigation is a result of a program established by the Partners AIDS Research Center at Massachusetts General Hospital (MGH).

"This study identifies the genetic battleground where the struggle between HIV and the human immune response occurs," says Philip Goulder, MD, PhD, of the Partners AIDS Research Center at MGH, the study's principal investigator. "The findings will help in understanding precisely how the immune system can succeed or fail against HIV, a prerequisite for a rational approach towards design of an HIV vaccine." Goulder also has an appointment at the Peter Medawar Building for Pathogen Research at Oxford.

The human immune system learns to recognize and attack virus-infected cells through the activity of human leukocyte antigen (HLA) Class 1 molecules, which sit on the surface of cells. When new viruses are being produced within an infected cell, Class 1 molecules grab fragments of viral proteins from within the cell and display them at the cell surface, thereby alerting the body's "killer" T cells that something foreign is within the cell and it should be destroyed. Three genes called HLA-A, HLA-B, and HLA-C encode Class 1 molecules, and it is known that the HLA-B genes are extremely diverse, with more than 560 versions or "alleles" having been identified. The current study was designed to test the theory that the diversity of HLA Class 1 molecules could reflect differences in the killer T cell activity controlled by those molecules.

The researchers analyzed blood samples from 375 HIV-infected patients at the Doris Duke Medical Research Institute at the University of KwaZulu-Natal to determine whether particular HLA Class 1 molecules control the killer T cell response against the virus. They found that an individual's version of HLA-B made a significant difference in how well the immune system responds against HIV, whereas the version of HLA-A or HLA-C inherited did not matter.

To examine the impact of Class 1 molecules on blood viral levels, the team studied more than 700 chronically infected African patients and again found that particular versions of HLA-B were associated with high or low plasma virus levels. Additional tests that looked at levels of the helper T cells that are destroyed by HIV and that analyzed samples from Australian patients infected with a different strain of virus all supported the conclusion that the form of the HLA-B molecule patients inherit makes a significant difference in how well their immune systems cope with HIV infection.

Evidence of the virus's impact on human evolution was found in an analysis of HLA-B alleles in HIV-infected mothers and their infants. Not only are HIV-infected women who have a protective version of HLA-B more likely to survive, they are also less likely to pass the virus along to their children. From an evolutionary standpoint,
that finding suggests a trend towards greater frequency of the protective alleles in a population over time.

"We have known for some time that HLA-B molecules are evolving more rapidly than other types, but it has been unclear why this is happening," says Goulder. "These data suggest an explanation for the more rapid evolution of HLA-B in response to other infectious diseases, not only HIV. This is an exciting time for infectious disease research because we are witnessing the evolutionary fight between the human immune system and the HIV virus happening right now, rather than over a period of thousands of years." Goulder is an assistant professor of Medicine at Harvard Medical School.

"The AIDS crisis will only be solved with the development of an effective vaccine," says Bruce Walker, MD, director of the Partners AIDS Reseach Center at MGH and a co-author of the current study. "This study's results help to focus this effort by telling us what the most effective immune responses are." Walker is also a Howard Hughes Medical Institute researcher and a professor of Medicine at Harvard Medical School.

Scientists Discover Crucial Enzyme

VANCOUVER, B.C. – December 16, 2004: UBC scientists have discovered an enzyme in mammals crucial to the transportation of proteins within cells. This discovery opens up new avenues of understanding of the mechanisms underlying the function of neurons and new approaches to therapy for neurodegenerative diseases such as Alzheimers and Huntington Disease.

The enzyme, HIP14, is a palmitoyl transferase that adds signaling molecules to proteins resulting in their transportation to specific cellular locations where they perform essential functions. This process known as palmitoylation is extremely important for the normal functioning of the nervous system where proteins are transported rapidly within nerve cells known as neurons.

Until now, scientists did not know how mammalian proteins become palmitoylated. During their study of Huntington Disease, Dr. Michael Hayden’s team at the Centre for Molecular Medicine and Therapeutics (CMMT) had previously identified a protein called HIP14 and recognized that it might play a role in palmitoylation. To further understand this mechanism, the Hayden team formed a partnership with Dr. Alaa El-Husseini and his team at the Brain Research Centre who are experts in the field of protein palmitoylation. Through this unique collaboration between experts in complementary fields, they discovered that HIP14 is indeed a mammalian palmitoyl transferase.

The teams also discovered that in the absence of the HIP14 enzyme, proteins were not transported to locations in the cell where they are needed. This change in protein trafficking is thought to result in severe neuronal dysfunction and may be a mechanism underlying diseases such as Alzheimers and Huntington Disease.

This research was funded by the Michael Smith Foundation for Health Research (MSFHR), the Canadian Institute of Health Research (CIHR) and the High Q Foundation.

News - Press release style

Party till 3 in the morning + help my girlfriend move (very) early today = exhausted, so today's news are gonna be press release style. The party was a blast : I won two bottle of wine, cinema tickets and the first prize for the center's best article (as second author, but its still nice!). Still have shopping to do today (I'm always late for that). I'll be back tomorrow or late tonight.

Wednesday, December 15, 2004

Christmas party

Its (already) Christmas time; I have my lab's party tomorrow so don't expect news until Friday! As you can see I added a Google Adsense Search box, which allow to search through the whole site. I'm extremely sorry about those of you who got pop-up ads this week; I have tracked the cause of it to the Bravenet minipoll module, which I promptly remove. Future polls will be in the forum, which I encourage you to join to discuss news / ask questions.

Happy Pre-Christmas party for those of you who have one this week, remember : don't abuse of alcohol and chocolate, be careful while driving, and have fun!

Tuesday, December 14, 2004

In vivo molecular imaging

Molecular imaging is a field that gained major attention in biology in the past few years, starting with the discovery/isolation of GFP, a protein able to emit (green) fluorescence when excited (for example with a laser). Nowadays, a panoply of fluorescent proteins / chemicals allow for precise tracking of different biological components (proteins, organelles, DNA) when coupled to antibodies or fused with proteins. Pretty covers for science magazines, very useful in the lab, everyones happy! Recent developments in the field include in vivo imaging (in real, live animals) and sequence-specific RNA tracking (with molecular beacons). Both were put to good use by a research team of the Abramson Cancer Center in a study related to cancer.

They were able to track the effectiveness of chemoterapy (Doxorubicin chemo, to be precise) using molecular beacons in mice. Doxorubicin kill tumoral cells by a p53 induced mechanism. p53 is a transcription factor; they tracked the expression of a p53-dependant gene, p21/Waf. Molecular beacons are hairpin DNA structures with a fluorophore on one end and a quencher (a molecule that absorb fluorescence and reemit it at another wavelenght) at the other end. When the hairpin is closed, no fluorescence is emitted. When the beacon anneal to a specific sequence, the fluorophore is too far from the quencher, and fluorescence is emitted. Cells express p21 via a Doxirubicin, p53 induced mechanism and go fluorescent. And you get to see it in the animal while its alive (and getting cured of its cancer!), so you can track the response throught time and space. Wonderful application of biotechnology! Here's the press release :

Molecular Tailoring of Chemotherapy with Novel Imaging Techniques Molecular Beacons, Gene Silencing, and Reporter Genes Studied to Better Predict Response to Chemotherapy
(Philadelphia, PA) – Researchers at the Abramson Cancer Center of the University of Pennsylvania are applying a host of imaging techniques to develop better ways to look noninvasively at the molecular characteristics of tumors. The experiments, now in human cell cultures and mouse models, are aimed at better forecasting early response to chemotherapy so that treatment choices can be adjusted.

"Right now in cancer therapy, with the exception of relatively uncommon examples of cancers for which we have tumor markers, we don't have reliable ways of predicting who is going to respond early on to chemotherapy," says Wafik El-Deiry, MD, PhD Associate Professor, Departments of Medicine, Genetics, and Pharmacology. "Currently cancer patients get their chemo and you can't tell if they're responding for several weeks. We need to have tests that will tell us if patients are going to respond to the chemo or the radiation soon after it's first given, and whether these responses are going to last."

Two recent papers in Cancer Biology & Therapy and Cancer Research describe the work of the El-Deiry laboratory. One approach is to use a molecular beacon, a molecule that can be activated within cells due to a specific context, such as in this case, the response to chemotherapy. The beacon recognizes a characteristic change in chemo-treated tumor cells, physically opens up and fluoresces, which can then be measured. "The beacon goes right into the living cell and if it opens up, emitting fluorescence, we can detect the glow," says El-Deiry.

Human lung-cancer cells were treated with the chemotherapeutic agent doxorubicin (Adriamycin), which causes cellular DNA damage. Doxorubicin works through the tumor suppressor protein p53, which ultimately kills many types of cancer cells. "We engineered a molecular beacon to detect expression of a gene called p21, that is turned on directly by p53 when cells are exposed to Doxorubicin," says El-Deiry.

The cells that were exposed to Doxorubicin activated the p53-responsive molecular-beacon tag and emitted a strong fluorescence. From this El-Deiry and colleagues hope to develop a scan that could detect a patient's likely reaction to certain chemotherapies: Strong fluorescence equals a good response to the chemotherapy. They hope to make what he refers to as a "beacon cocktail" that can predict response by monitoring multiple genes simultaneously as well as additional intracellular events in the process of cell death.

In another study El-Deiry and colleagues combine imaging techniques and a mouse model for colon cancer. "In this research, we're combining two very powerful emerging technologies," says El-Deiry. "This is the first example, to my knowledge, of the use of inducible gene silencing and non-invasive bioluminescence imaging in a mouse model for cancer." Gene silencing is a technique that allows researchers to control the expression of any gene in a given cell by introducing small RNA sequences targeting the gene of interest. Inducible refers to the ability to control whether or not the silencing RNA is expressed in the cell so that investigators can compare gene activity to tumor growth, as El-Deiry did in this study. This approach allows researchers to regulate gene expression by what they feed the mice. In this study, the KILLER/DR5 receptor, another protein that responds to chemotherapy by killing cancer cells, is silenced in colon tumors in the mice.

They also labeled the cells with a reporter gene called firefly luciferase, which gives off light. "The use of a reporter like firefly luciferase marks the tumor cells so we can see them by another imaging technique," explains El-Deiry. "Fireflies that we see in the evening carry out the same chemical reaction with their own luciferase protein to give off the light." The imager detects the light and captures its intensity to provide a measurement of the size of the tumor.

"The bioluminescence imaging technology has provided a breakthrough that allows scientists to examine the size of a tumor in living mice with high sensitivity," says El-Deiry. "Since the reporter gene is always on and only in the tumor cell, it's essentially measuring tumor volume. Using the reporter gene along with the KILLER/DR5 silencer, we show for the first time that when we turn off KILLER/DR5, we get bigger tumors."

While the beacon or beacon cocktails have the potential to be used in the clinic to detect mutations in cancer cells or the activation of genes that predict therapeutic response, the major advance with the bioluminescence imaging is in accelerating preclinical drug development. The gene silencing allows precise molecular characterization of targets that are relevant for therapeutic response while the imaging allows non-invasive assessment of drug activity towards implanted tumors. This approach saves time and money because it is possible to see the effects of drugs in living mice without sacrificing them and it also requires fewer mice in experiments.

Because the KILLER/DR5 receptor is involved in the process of cell death by chemotherapy, El-Deiry is also gaining insight into which drugs use it and which drugs work by other mechanisms. "This is important because to maximize tumor killing and to attempt to bypass or reverse resistance to chemotherapy, we need to harness all the ways cancer cells can be killed," he says.

The KILLER/DR5 receptor is engaged by a therapeutic agent currently being developed called TRAIL (Tumor necrosis factor-Related Apoptosis Inducing Ligand). TRAIL is produced normally by natural killer cells and controls tumor spread by binding to a tumor's death-inducing receptor KILLER/DR5. "However, in cancer patients with suppressed immunity and for reasons we still don't understand, there isn't enough TRAIL being produced or effectively delivered by the natural killer cells at the site of tumors and so tumors are not suppressed," says El-Deiry. "The hope is that if TRAIL is administered to patients alone or in combination with chemotherapy, this may in the clinic lead to some benefit." TRAIL looks promising in animal studies but clinical studies that are due to start in the next year or so will determine how toxic TRAIL is and begin to see whether it really works in cancer patients.

The work was funded by NIH grants including a multi-institutional Network for Translational Research in Optical Imaging imaging grant from the National Cancer Institute. The Network provides support for imaging resources to accelerate translational research on cancer.

Saturday, December 11, 2004

Food poisoning virus grown in-lab for the first time

Washington University of St Louis report that researchers developed a method to gros norovirus in-lab, which is a first. It will lead to a better understanding of the virus and the diseases it induce (gastroenteritis or "stomach flu"). The virus is a very common cause of food poisoning. Using the in vitro cultured virus, researchers were able to determine that it had specific tropism for Macrophages and Dendritic cells, components of the immune system responsible of warning T and B cells that something is wrong (by presenting antigens, usually small peptides of degraded pathogens) and activating the immune response.

Noroviruses (genus Norovirus, family Caliciviridae) are a group of related, single-stranded RNA, nonenveloped viruses that cause acute gastroenteritis in humans. Norovirus was recently approved as the official genus name for the group of viruses provisionally described as “Norwalk-like viruses” (NLV).

PLoS biology announcement

PLos Biology Paper : Replication of Norovirus in Cell Culture Reveals a Tropism for Dendritic Cells and Macrophages

More information about norovirus.

CDC Technical Data Sheet

Usual Press release :

Scientists first to grow common cause of food poisoning in the lab

Nov. 30, 2004 — Scientists at Washington University School of Medicine in St. Louis have become the first to successfully grow a norovirus in the lab. In humans, noroviruses are a highly contagious source of diarrhea, vomiting and other stomach upset that made headlines two years ago after a series of repeated outbreaks on cruise ships. These viruses are a major cause of human disease worldwide.

Researchers showed that the mouse norovirus MNV-1 could be grown inside cells from mice with defective immune systems. Their findings make it much easier to learn about the mouse virus and may help other researchers seeking to duplicate the accomplishment with human forms of the virus.

In a study published this week in the online journal Public Library of Science-Biology, scientists who developed the new technique report it may already have led them to a good target for vaccine development.

"By looking at the mouse virus we'd grown in the lab, we were able to identify a part of the capsid, the virus' protein shell, that is essential to its ability to cause disease," says senior author Skip Virgin, M.D., Ph.D., professor of pathology and immunology and of molecular microbiology. "If this part of the capsid has an equivalent in human noroviruses, altering or disabling it may give us a way to produce forms of the viruses that are weak enough to serve as vaccines."

According to the U.S. Centers for Disease Control and Prevention, noroviruses are involved in about half of all food poisoning cases and annually cause about 23 million cases of acute gastroenteritis in the United States.

Norovirus disease is characterized by frequent vomiting and diarrhea over the course of one to two days. The most infamous norovirus, the Norwalk virus, was first identified after a 1968 outbreak at a school in Norwalk, Ohio. The Norwalk virus also caused a series of repeated outbreaks on cruise ships in 2002 and in military personnel in Afghanistan.

Although such infections rarely lead to serious or life-threatening illness in the United States and other Western countries, they spread rapidly, are difficult to prevent from spreading and can create considerable discomfort. Dehydration from the diarrhea and vomiting induced by the virus sometimes leads to hospitalization in the elderly, the young or those with weakened immune systems. In the developing world, these viruses are a major cause of human illness.

All previous attempts to culture human noroviruses in tissues in the laboratory have been unsuccessful.

"As a group, noroviruses have defied characterization for decades because there just hasn't been a way to get the virus to grow outside of a human host," Virgin says.

In 2003, Christianne Wobus, Ph.D., and Stephanie Karst, Ph.D., two postdoctoral fellows in Virgin's lab, identified MNV-1, the first known mouse norovirus. Virgin's group showed that the mice's ability to fight MNV-1 relied heavily on the innate immune system, the branch of the immune system that attacks invaders soon after they enter the body.

In the new paper, Virgin's group reveals that MNV-1 likes to infect cells of the innate immune system. In tests in mice, the researchers found the virus thrived in macrophages, immune system cells that normally engulf and destroy pathogens, and in dendritic cells, sentry-like cells that pick up and display proteins from pathogens.

"We think there may be dendritic cells just beneath the lining of the human gut that are providing the gateway the virus needs to cause disease," Virgin says.

To grow the virus in the lab, researchers took dendritic cells and macrophages from mice with defective innate immune systems and exposed them to the virus.

"The virus grew beautifully," Virgin says. "It's a very facile and robust system."

Comparisons of MNV-1 and human noroviruses have revealed many similarities in gene sequence, structure and overall arrangement of the genome. But Virgin acknowledges that differences between mouse and human physiology may significantly alter MNV-1's interactions with its host. For example, mice do not appear to be able to vomit. Additionally, researchers aren't sure yet whether MNV-1 can make mice with normal immune systems sick.

"The bottom line is that this mouse model provides us with a very useful way to examine certain similar aspects of the noroviruses," Virgin says. "Among other things, we'll be using it to look at how the capsid protein enables infection, viral replication processes and the receptors on host cells that enable the virus to infect specific cell types."

Computational DNA crystal

California Institute of Technology released today news that computation can be embedded in a DNA crystal, a cute mix of biotechnology and nanotechnology. They designed DNA molecules which specific sequence cause them to crystallize in a pattern of progressively smaller "triangles within triangles," known as a Sierpinski triangle.

The whole paper (complete with pretty figures and images) can be found here (PLoS Biology).The importance of this work is explained by the researchers :

"In fact the work is the first experimental demonstration of a theoretical concept that Winfree has been developing since 1995--his proposal that any algorithm can be embedded in the growth of a crystal. This concept, according to Winfree's coauthor and Caltech research fellow Paul W. K. Rothemund, has inspired an entirely new research field, "algorithmic self-assembly," in which scientists study the implications of embedding computation into crystal growth."

Nanotechnology applications aside, the authors contend that the most important implication of their work may be a better understanding of how computation shapes the physical world around us. "If algorithmic concepts can be successfully adapted to the molecular context," the authors write, "the algorithm would join energy and entropy as essential concepts for understanding how physical processes create order."

From the press release : Caltech computer scientists embed computation in a DNA crystal to create microscopic patterns

PASADENA, Calif.--In a demonstration that holds promise for future advances in nanotechnology, California Institute of Technology computer scientists have succeeded in building a DNA crystal that computes as it grows. As the computation proceeds, it creates a triangular fractal pattern in the DNA crystal.

This is the first time that a computation has been embedded in the growth of any crystal, and the first time that computation has been used to create a complex microscopic pattern. And, the researchers say, it is one step in the dream of nanoscientists to master construction techniques at the molecular level.

Reporting in the December issue of the journal Public Library of Science (PLoS) Biology, Caltech assistant professor Erik Winfree and his colleagues show that DNA "tiles" can be programmed to assemble themselves into a crystal bearing a pattern of progressively smaller "triangles within triangles," known as a Sierpinski triangle. This fractal pattern is more complex than patterns found in natural crystals because it never repeats. Natural crystals, by contrast, all bear repeating patterns like those commonly found in the tiling of a bathroom floor. And, because each DNA tile is a tiny knot of DNA with just 150 base pairs (an entire human genome has some 3 billion), the resulting Sierpinski triangles are microscopic. The Winfree team reports growing micron-size DNA crystals (about a hundredth the width of a human hair) that contain numerous Sierpinski triangles.

A key feature of the Caltech team's approach is that the DNA tiles assemble into a crystal spontaneously. Comprising a knot of four DNA strands, each DNA tile has four loose ends known as "sticky ends." These sticky ends are what binds one DNA tile to another. A sticky end with a particular DNA sequence can be thought of as a special type of glue, one that only binds to a sticky end with a complementary DNA sequence, a special "anti-glue''. For their experiments, the authors just mixed the DNA tiles into salt water and let the sticky ends do the work, self-assembling the tiles into a Sierpinski triangle. In nanotechnology this "hands off" approach to manufacturing is a desirable property, and a common theme.

The novel aspect of the research is the translation of an algorithm--the basic method underlying a computer program--into the process of crystal growth. A well-known algorithm for drawing a Sierpinski triangle starts with a sequence of 0s and 1s. It redraws the sequence over and over again, filling up successive rows on a piece of paper, each time performing binary addition on adjacent digits.

The result is a Sierpinski triangle built out of 0s and 1s. To embed this algorithm in crystal growth, the scientists represented written rows of binary "0s" and "1s" as rows of DNA tiles in the crystal--some tiles stood for 0, and others for 1. To emulate addition, the sticky ends were designed to ensure that whenever a free tile stuck to tiles already in the crystal, it represented the sum of the tiles it was sticking to.

The process was not without error, however. Sometimes DNA tiles stuck in the wrong place, computing the wrong sum, and destroying the pattern. The largest perfect Sierpinski triangle that grew contained only about 200 DNA tiles. But it is the first time any such thing has been done and the researchers believe they can reduce errors in the future.

In fact the work is the first experimental demonstration of a theoretical concept that Winfree has been developing since 1995--his proposal that any algorithm can be embedded in the growth of a crystal. This concept, according to Winfree's coauthor and Caltech research fellow Paul W. K. Rothemund, has inspired an entirely new research field, "algorithmic self-assembly," in which scientists study the implications of embedding computation into crystal growth.

"A growing group of researchers has proposed a series of ever more complicated computations and patterns for these crystals, but until now it was unclear that even the most basic of computations and patterns could be achieved experimentally," Rothemund says.

Whether larger, more complicated computations and patterns can be created depends on whether Winfree's team can reduce the errors. Whether the crystals will be useful in nanotechnology may depend on whether the patterns can be turned into electronic devices and circuits, a possibility being explored at other universities including Duke and Purdue.

Nanotechnology applications aside, the authors contend that the most important implication of their work may be a better understanding of how computation shapes the physical world around us. "If algorithmic concepts can be successfully adapted to the molecular context," the authors write, "the algorithm would join energy and entropy as essential concepts for understanding how physical processes create order."

Winfree is an assistant professor of computation and neural systems and computer science; Rothemund is a senior research fellow in computer science and computation and neural systems. The third author is Nick Papadakis, a former staff member in computer science.