Synthetic Biology May Spawn Biohackers 320
nusratt writes "EE Times reports 'Design automation systems tailored to the task of genetic engineering . . . can lead to the accidental or deliberate creation of pathogenic biological components.' Design of molecular machines is analogous to doing system-on-chip work, and hackers 'will not need a detailed knowledge of biochemistry to effectively create complex biochemical machines.' A Harvard genetics professor says, 'Even if we don't have bioterrorists and teen-age biohackers, we will still create things that do not have the properties that we thought they would . . . Even if you are genetically resistant and recently immunized, you will have problems with artificial biological agents.' He also says that there are two big differences between this risk and nuclear weapons: (1) building weapons is harder; (2) synth-bio work is more accident-prone. Oh great, just great: script-kiddies with smallpox . . ."
Re:Explain it to me (Score:5, Informative)
Re:Designed vs Evolved (Score:5, Informative)
The two really interesting parts are the adaptive immune system where there are the cell mediated (killer T-cell) and humoral (antibody) immune responses.
Both use the principle of making molecules that will stick to bad molecules, and if they do triggering a cascade of events that eventually winds up destroying the bad molecule and the things affected by it, and leaving healthy tissue behind untouched (we hope).
The really really good part is that you're right, if the viral coat proteins have the same chemical surface at specific regions called epitopes, then the same antibodies will be able to bind all of them, even if they're different in other places.
Most molecules have several epitopes on them, although sometimes you have to bind most or all of these before a response is raised.
Viruses in the wild beat this by mutating every time they reproduce inside a cell by using error prone replication techniques. After all, if you make a billion particles and only 2% work, you can still infect your next host quite smartly.
That means that two individuals with the same disease, one catching it off the other, might have sufficiently different viral particles that an immunisation against one set of epitopes is ineffective. That's what happens with the common cold.
Re:Defenses (Score:5, Informative)
The one thing about HIV is that it's very succeptible to oxidation. Any kind of oxidising agent, like bleach or strong disinfectants or even some mouthwashes, will render a puddle of HIV infected blood safe to clean up within minutes. Faster if you mix it, but please dont.
To get airborne, the HIV would first need to borrow some viral trickery from other diseases to reproduce in the lungs and mucous membranes as well as its usual home of in the lymphatic system, and then once expelled on a person's breath it would need a new coat to protect it from the toxic levels of oxygen in the air.
All this, while keeping the size of the genome down to a managable length so you can stuff it into its protein coat.
If you can engineer both those capabilities into HIV, you would have Airborne AIDS. Quite a puzzle though.
Re:Prey (Score:1, Informative)
Re:Defenses (Score:4, Informative)
After all the point of having a lot of different kinds of Major Histocompatibility Complex alleles in the population is that somebody in the population will have the right combination of MHC genes to be a responder to an arbitary infection and so survive to breed.
The flip side of this is that many people are prone to getting autoimmune disease as a consequence of getting certain infections. Crohn's disease is likely triggered by a bacteria. [kennedy.name]
Certain HLA antigens are bad to have. Such as HLA B27 [nih.gov] which makes one a sitting duck for autoimmune disease. People with that can get Reiter's syndrome [nih.gov] (a form of autoimmune arthritis) from something as simple as food poisoning. As bad as HLA B27 sounds, it is likely to provide protection against something, much like sickle cell trait protects against malaria.
Biological diversity means there is less likelyhood of a large scale wipeout of the population, but also that there will be many people who get diseases due to things like having a bad HLA antigen (such as B27).
Any protection from viruses that HLA antigens could provide likely could be circumvented, as HLA antigens are not secret at all. They are use in diagnosing autoimmune disease, matching organ transplants, etc.
It is roughly equivalent to a computer virus writer having access to all the patterns that an anti-virus program is designed to detect.
Re:Designed vs Evolved (Score:4, Informative)
My guess would be that it takes lots molecules of the same protein to raise an immune response, and only one viral particle is required to start an infection. A shotgun vaccine might have to include large quantites of protein injected into your blood to work, more than your body could tolorate.
That's only a guess though. A modified version of the idea (like getting a lot of different haptens each bearing one epitope conjugated onto a carrier protein to raise an immune response) may yet work.
Re:Yikes (Score:5, Informative)
So what if the circuit designers don't need to know all the physics behind what they're doing. They do need expertise in circuit design. In fact it amazes me sometimes how many people doing molecular biology don't even understand the chemistry behind what they're doing -they just follow the recipe. They do, however, know how to "design" what they're doing.
Yes, many proteins have a somewhat "modular" structure, but just sticking these "modules" together is most likely going to give you a misfolded protein that does nothing but get immediately degraded or end up in the cellular equivalent of the junk heap (if it doesn't kill the cell expressing it first).
There's all kinds of information in the article that IMO sounds much scarier and easier to do than it really is. From my vantage point it seems like it would be much harder to build a single working protein from pieces than to build an atomic bomb. It can take months to engineer a simple mutations and get a protein that's properly expressed.
Considering how much hard work it takes for experts, using very expensive equipment an reagents to do this kind of thing, I'm not too worried about BioHackers quite yet.
Re:Defenses (Score:4, Informative)
To mount an effective immune response, you'd have to be able to come up with a vaccine that pre-empts HIV infection and prevents it from killing those T-Cells. The HIV vaccine is still being persued.
We can only hope that engineered diseases would at least give us the human immune system as toe hold in the fight against them.
If the engineered disease happened to be caused by microscopic particles made of diamond that no protease could cut, we would be truly in trouble.
for perspective, consider that asbestos dust can never be expelled from the lungs and can never be degraded, because it is chemically and physically able to defeat the body's normal ways of clearing pulmonary debris. If antibodies could deactivate it and then macrophages could just eat it, the way antibodies and macrophages sometimes deal with proten based threats, it wouldn't be a problem.
Re:Designed vs Evolved (Score:4, Informative)
(1) you don't get an overwhelmingly higher concentraion of the effective antigens over the ineffective ones, so the system is not effectively triggered.
(2) Even if it was possible to immunize against 100s of pathogens at once, it wouldn't be desireable. Clonal selection is a great tool the body uses to keep a vast reserve library in storage until needed. Inducing a memory response for certain antigens could remove (or reduce the incidence of) memory cell sets for other antigens. So, with the shotgun, you are now immune to a few hundred antgigens that don't realy pose a threat, while you have now lost your memory cell sets to a dozen types of rhinovirus, rotavirus, and a few strains of influenza (theorectially, and my apologies for the oversimplification).
Re:Designed vs Evolved (Score:3, Informative)
Numerous studies have shown that when a virus first infects a human it is generally pretty deadly, but over time they adapt to NOT be so deadly. Afterall, the less deadly/less severe the symptoms the more likely the virus will be passed on...simple evolution. Of course, somethings work like HIV which is naturally nearly 100% fatal, but doesn't kill or cause any symptoms to appear for a long time. These would have to be considered very successful viruses as they allow the propagation of both the host species and the virus.
Twelve Monkeys (Score:5, Informative)
Looking at the picture of Prof. George Church -- the aforementioned Harvard geneticist [harvard.edu], one is struck by the resemblance with the guy Terry Gilliam cast as the "environmentalist" genetic engineer synthesized a pathogen to kill all humans in The Twelve Monkeys [imdb.com].
George Church is probably one of the least likely geneticists to hop on a world-wide jet tour to deliver a misanthropic virus he's synthesized.
The problem with all this isn't so much the creation of new, deadly pandemics -- nature does a good enough job of that. The real problem is the way amplification of international transport has been behind almost every major pandemic from the Plague which followed on the heels of the Mongol Empire's wide stretch [google.com] -- to the pandemic of the first World War [google.com].
Globalization has already given us the AIDS [google.com] epidemic and the SARS scare [google.com]. It may have given us autism's recent explosive growth [laboratory...states.com] and a lot more we don't even know about.
No one is being held liable for this increased risk imposed on an unaware population -- this despite the fact even identifiable corporations have externalized the costs of their risk-taking on the public and walked away with higher corporate profits as a result. Not even Ralph Nader has guts to touch this.
Re:Prey (Score:2, Informative)
Re:This will never happen, guys. (Score:3, Informative)
Alanine is very small as far as molecules go; it's one of the 4 key blocks for DNA
Alanine (which _is_ a simple molecule) is one of the 21 most common amino acids used to make peptides, enzymes, and proteins.
You were thinking of Adenine (which is _not_ a simple molecule) which is one of the four DNA base pairs. Every sequence of three DNA bases translates into one amino acid at the ribosome.