Pages

Saturday, December 10, 2011

Bioluminescent Light Bulbs?

There is a news blurb going around about using flasks of bioluminscent bacteria to light your house.  I mentioned this type of thing in the Foxfire post, but Philips has been working on this for some time.  The lighting system consists of a wall of hand-blown glass flasks, coupled to methane lines that are sourced locally (ie, from the users trash).  The bacteria have been engineered to glow when methane is present, so users can control the lighting.  Although this is impractical for general use at the moment, it is probably the beginning of the next revolution in lighting. Some of the articles talk about the most likely first application, which would be for safety lighting near buildings or on roadways.  I think it would be cool to have trees or shrubs engineered to glow as well... it would be awesome to have glowing bushes near our front walkway!  If you want to see how the bioluminscent bacteria fit into the Microbial Home, see the link at Philips' web site ( link), it's pretty trippy stuff.  If you want to try your hand at growing bioluminescent algae at home (or for a cool science fair project!), here is a good starting place (link).  If you want to see what a team from Cambridge University did for their iGEM project last year (hint: it's on making bioluminscent products) check out this link.

  Philips is not the only company looking at commercializing bioluminescence for non-medical applications.  A company called BioLume (link), based in Research Triangle Park, NC is trying to put bioluminescent proteins in food.  Yeah, that's right.  Glowing food.  They use examples like candy and alcohol as likely products, as well as makeup.  Sounds like a Rave Gone Wild! The company has IP around many different bioluminescent proteins (mostly luciferases) found in marine life.  I imagine that they formulate it in a way that the enzyme becomes active when there is a change in the environment.  They mention a calcium-induced reaction of a enzyme-substrate fusion, as well as fusions with fluorescent proteins.  I'm sure that the proprietary chemistry and photophysics involved in these products is really cool!  I do hope the metabolized product is non bioluminscent... there is nothing more scary that glowing pee!

Wednesday, November 2, 2011

Red wine in a pill: Metabolic effects of resveratrol in humans

Remember back in July I talked about a future where simply popping a couple of resveratrol tablets gave the same benefit as a walk around the block?  That future may be closer than we think!  A recent paper in the journal Cell Metabolism describes the results of a very small study of resveratrol in humans and the data is pretty exciting.  Let me repeat that caveat... this was a small study!  Still, the metabolic effects described in this work clearly emphasize the need for further research on this and other magical polyphenols.

The paper published by Timmers et al. (abstract) is the first to study the metabolic effects of resveratrol (RES) in a clinical setting.  Eleven obese men were given either RES (at 150mg/day) or placebo for 4weeks, followed by a 4-week washout and then the treatment was switched (this is known as a crossover study).  Patients and doctors were both blind as to what treatment was being administered and during treatment, a variety of metabolic tests were conducted.  There is a lot of data in the paper (and it looks like the pdf is free, so check it out yourself!) but let me hit a few highlights:

1) Patients taking RES show an increase in mitrochondrial efficiency, particularly in fatty acid oxidation of muscle fats, and decreased levels of triglycerides and glucose levels.  Significantly, these changes are seen at the gene level, suggesting that it is the overall metabolic pathway that is improved, not just a downstream clearance of metabolic markers.  A similar pattern of changes has been noted in athletes undergoing endurance training (they reference Dube et al, 2008 and Meex et al, 2010).  This is fairly consistent with the Momken paper I blogged about back in July, ie, RES acts like an exercise mimetic.

2) Changes in glucose and insulin levels are modest.  Timmers et al. report a statistically significant drop in serum glucose and insulin levels in the patients taking RES, but this effect is pretty modest.  There is also a shift in peak glucose and insulin levels after a liquid test meal, suggesting some changes in glucose homeostasis, but they could not draw definitive conclusions.  If you recall, the rat data from Momken et al. was also pretty weak with regard to insulin/glucose levels.

3) RES also showed other health benefits:  This study also demonstrated a significant effect of RES on lowering systolic blood pressure and mean arterial blood pressure, as well as decreases in resting energy expenditure and sleeping metabolic rate.  The later two effects are also seen in studies of calorie restriction and further illustrate the metabolic changes induced by RES.  Calorie restriction is also associated with increased lifespan in animals, so these observations may be pointing to another possible health benefit of RES.  They also observed a decrease in markers of inflammation, further suggesting an overall improvement in health. Although this is interesting, I still think the catechins are the more potent player here... I'd like to think that as the flavinoids polymerize during aging, the wine gets better and better for your heart. It would be interesting to see how some of these molecules perform in a study similar to this.

Taken together, this paper highlights some of the metabolic effects of RES in humans and may offer some insight into the health benefits of this polyphenol.  Much like the rat study, however, this is a very high dose (the equivalent of >100 glasses of wine per day) and so who knows if there are long term side effects at this dose.  The fact that they see statistical significance with only eleven patients is also very surprising.  Clinical studies usually need hundreds, or even thousands of patients to provide enough statistical power to draw conclusions like this.  Personally, I would find taking a pill much less satisfying that enjoying a nice glass of Cabernet. Since that glass of Cab is a veritable grab bag of Redox goodies, I think it is also very likely that there are many other 'good' polyphenols in wine that scientists haven't studied as rigorously as RES.  So as dozens of trick-or-treaters descended upon our neighborhood on Halloween night, I had to raise my glass to our ancient ancestors who discovered the wonderful winemaking process, and the scientists who now try and tease apart how it does what it does. I may have also stolen a chocolate or two... in the interests of science, you know.

Tuesday, October 25, 2011

Halloween winemaking magic at Bruliam Wines

Just in time for Halloween, Kerith Overstreet at Bruliam Wines has a great blog post this week about working with her spooky 2011 harvest (beating it into submission, actually).  Sounds like a challenging year!  She describes the redox chemistry that goes on during the early fermentation process and actually has a graph from her lab!  I first blogged about the magic of winemaking after Kerith's great talk on the subject (here is my post) and even she refers to the process as magic in her latest post (but she also refers to wife swapping, Alanis Morissette, and Hanukkah miracles, so who knows what state of mind she was in as she wrote this).  Anyhow, a very fun, informative read directly from the mysterious front lines of winemaking.  Enjoy it with a glass of good Cabernet, the official wine of the dead.

Keriths latest blog post:  (link)

Thursday, October 20, 2011

21st Century mummy

Just a quick pointer to a cool article on a recent attempt at mummification (here's the link). Stephen Buckley, a chemist at York Univeristy in England, has spent two decades studying how ancient Egyptians made mummies.  He studied tissue samples and chemical traces left on canopic jars in an effort to reproduce the method.  He then tested the process in his shed, using pig's legs as a proxy for human flesh (there's a DIYbio project for you!).  I'm not sure if this guy is married, but even here in the Dark Lab, this work would be pushing the limits.  Anyhow, this year he felt that he was ready for prime time.  He placed an ad looking for suitable volunteers and the lucky person was... Alan Billis, a London cab driver.  Terminally ill with lung cancer, he went through the mummification process after he died. By all accounts, it was a success and the body will be kept for at least a year to study.  Hopefully, Alan is hanging out with a bunch of cool, Egyptian princesses.  Thousands of years from now, archaeologists will argue over whether our society placed a high value on cab drivers, or if Mr. Billis was simply a member of the ruling elite.  They will come up with grand theories on how he lived, how he died, and why he was the only surviving mummy of the period.  Should make for an interesting read.


Cat mummy at the British Museum (link)

To me, it is truly amazing that with all of today's technology, it is difficult to reproduce the mummification procedure.  The Egyptians likely had years of empirical data to build from and since it was considered a sacred ritual for the upper class, there was significant motivation for young priests to be good at making a mummy.  I can see a room full of young mummification interns, trying to preserve rats or some other suitable test animal.  After months of work, the mentor unwraps the package, only to find a rancid, decomposing corpse. "Aw, man," the student whines (or whatever the angsty teen expression was at that time).  He then slouches his way over to the stone quarry.

Wednesday, October 5, 2011

A gene important for creating zombie caterpillars

Advancements in zombie science are coming fast and furious!  An email from a colleague (and fellow reader) noted that I missed a recent article in the journal Science concerning zombies.  The article, titled "A Gene for an Extended Phenotype", seemed pretty innocuous, however, after going back and reading it more carefully it turned out to be a pretty cool discovery.

First, a bit about zombie caterpillars.  There have been several documented reports of zombie-like behavior in moth larvae.  This report is focused on the infection of the gypsy moth (Lymantria dispar) by a baculovirus (known as LdMNPV).  During the various stages of molting, larvae typically hang out on the ground and away from The Very Hungry Birdie, but climb up into the trees at night to feed on leaves.  After infection by the baculovirus, their behavior changes (noticing a pattern here?).  As the virus replicates and ravages the inside of the caterpillar, the infected host climbs up into the leaves during the daylight hours and eventually dies.  The body then liquefies, and virus-laden particles rain down on the uninfected victims below.  Yeah, you can't script horror much better than that.

Ok, so Hoover et al. (from Penn State, see abstract) were interested in identifying which genes were important for the change in behavior.  To do this, they infected caterpillars with wild type baculovirus, as well as virus that had been genetically engineered to be missing certain genes.  The caterpillars were placed in 1 liter soda bottles equipped with a fiberglass screen for climbing (in true DIYbio fashion!!).  Interestingly, when a gene called egt was removed, the caterpillars died at ground level, suggesting that the behavioral control of the virus had been altered.  To make sure it wasn't an artifact of the mutated virus, they re-engineered the mutated constructs so that the egt gene was present again and the zombie caterpillars climbed up the mesh and died.  It would appear that the egt gene in the virus has evolved to make the caterpillars engage in high-risk behavior, and to place the caterpillars in a location where rain/gravity/hungry birdies can maximize viral spread. Genius.

The next step is to figure out the mechanism.  Interestingly, Hoover et al. mention that the egt gene encodes an enzyme which deactivates a hormone (20-hydroxyecdysone) involved in the process of molting.  It is intriguing that the virus blocks the molting process in order to give itself time to replicate inside the host, but it was hard for me to understand how the modification of the hormone leads to behavioral changes.  Perhaps entomologists already know that part.  Coincidentally, the hormone is also reported to have a variety of biological effects in humans, even though we do not molt and lack the endogenous receptor. What would the modified enzyme do in a human?  Until we find this stuff out, it's probably a good idea to check the trees around your house, particularly if you hear a dripping sound... and if your neighbor has been missing awhile.

Yes, there are zombies all around us.

Monday, September 19, 2011

Biopunks help solve structure of key viral enzyme

I've been meaning to blog about the program FoldIt for some time.  It's a program out of David Baker's lab (link to University of Washington) where users can manipulate protein structures to improve folding but in a unique twist, the improvement is 'scored', just like a game.  A problem is posed on the website and thousands of players compete to see who can find the best solution. They also have regular competitions to see how well players can predict the structure of a protein from just the amino acid sequence (the Holy Grail of the protein folding world).  I've played on and off for a couple of years and it's pretty fun... but it's a lot like my day job, so if I take the time to play a video game it is usually something like Rock Band.  Still, FoldIt is a pretty easy game to play (the number of rules is limited and the GUI is very intuitive) so the real challenge is in the player's ability to use logic and their skill in 3D visualization. I'll have a more detailed post about it and its less-familiar cousin EteRNA (the RNA equivalent of FoldIt) later. When you are ready to play, click here (it's free, easy, and fun... really!)

This week in the journal Nature Structural and Molecular Biology (I have had several rejection letters from this fine journal) is an article (abstract) describing the use of FoldIt to solve a scientific problem at least ten years old!  Crystallographers have been trying to determine the structure of the monomeric form of MPV protease. MPV stands for Mason Pfizer Virus, a virus that causes an AIDS-like syndrome in monkeys.  Although it is not HIV per se, the protease is a key enzyme in the production of a mature virus and so developing drugs using this structure could be useful in developing an HIV therapy.  This protein was one of those rare cases where good crystals could be grown but interpreting the x-ray data back into the final 3D picture of the protein could not be done. So, researchers looked to crowd-sourcing as a means to solve this problem.  For three weeks, FoldIt players had the chance to optimize a 3D model, based on an NMR structure of the monomer.  Teams competed to see who could come up with the best solution (based on how well the different structural pieces of the protein fit together).  After all that tweaking and optimizing, more than a million different models were created.  Crystallographers used the best of these models as a starting point and one of them successfully generated a solution to the x-ray data using molecular replacement.

Who were the people who provided the key to success?  The top groups were listed as co-authors by their team name (FoldIt Contenders Group and FoldIt Void Crushers Group) and by a few handles listed in the article (spvincent, grabhorn, mimi) so not much is known about these folks, but I'll bet many of them had no knowledge of crystallography or biochemistry and probably had no clue what this enzyme did.  However, the fact that thousands of players worked together on this problem and were ultimately successful highlights the power of crowd-sourcing and that top-flight science can be accomplished through a "game".  I think this paper will also motivate others, both players and scientists, to leverage the power of biopunks!

Tuesday, September 13, 2011

How about a GFP cat to go with your GFP beagle?


The GFP cat:
From Figure 2 of Wongsrikeao et al.

Well, that didn't take long.  I blogged last month about a paper describing GFP beagles that were able to glow under the control of a tetracycline promotor.  Now, a new Nature Methods paper describes a GFP cat!  The basic point of the paper (here's the abstract) was to demonstrate gamate-targeted transgenesis in cats and to use this ability to make a transgenic feline model for HIV research.  The HIV part was interesting, as they introduced the gene for a protein from the rhesus macaque known as TRIM5.  I studied TRIM5 pretty extensively when I was doing HIV research as it is a species-specific restriction factor that is effective at stopping HIV replication.  Cats don't have an analogous TRIM protein (that we know of) so by introducing this protein into the cat, researchers can study the transmission of FIV (the cat version of HIV).

The cool part was the GFP expression.  Wongsrikeao et al. wanted to see if they could introduce multiple genes, and since GFP is a convenient marker, they could also study the presence of the transgenes in progeny cats.  As I mentioned in the Beagle post, fluorescent proteins have been introduced in animals previously (see here and here for cats) but the efficiency in the present work is better.  I'm certainly no expert on transgenics, but the general trend over the last few years is clear... we are moving from the realm of the nearly impossible to the land of the challenging but doable.  Glowing cats, dogs, hamsters, bunnies, you name it are going to be popping up in your pet store relatively soon.  Lost your dog?  No problem... just look for the glowing blob in the woods.  How about a government program to develop a GFP opossum?  Think of how much road kill would be eliminated if you could see these animals before they got up close and personal with your tire.  Think about how cool the forest would be if all of the little woodland creatures glowed bright green. Think about how easy it would be to hunt.  I wonder how screwed up the food chain would get. I guess we would have to make GFP plants for the little GFP bunnies to hide in. Man, this would make a really cool dystopian/biopunk story!

Tuesday, August 30, 2011

Can gut bacteria make zombies?


L. rhamnosus: Not a zombie-
producing bacteria...yet.

If T. gondii weren't enough to worry about, now there is evidence that bacteria in our gut can influence brain function.  Lactobacillus and other probiotic organisms have long been speculated to have beneficial in vivo effects, and are most commonly known for improving the health of the digestive tract.  L. acidophilus is probably the most widely known, since it is used to make yogurt, but there are many other types of lactobacilli with alleged health benefits ranging from lowered cholesterol to improved mood.  Some of these benefits are speculative, and for many years any benefit (such as improved gut health) was thought to be due to local effects or secreted chemicals.  However, the impact of these bacterial colonies may be much more far-reaching.

A Proceeding of the National Academy of Sciences (PNAS) paper published by Heujtz et al. last January (abstract) showed that microbial colonization in the mouse gut led to the activation of signaling pathways involved with motor control and emotional response.  This was the latest in a growing body of literature that suggests gut bacteria could influence how we think and act.  Now, in the August edition of PNAS, Bravo and colleagues take this one step further (abstract).  They show that Lactobacillus rhamnosus can directly influence the expression of GABA receptors in the brain.  GABA is the primary neurotransmitter for regulating many physiological and psychological activities in humans.  For example, caffeine inhibits GABA and results in an overall increase in neurotransmitter activity.  In contrast, alcohol and sedatives tend to increase GABA activity, leading to reduced neurotransmitter activity. Bravo et al. showed that feeding mice L. rhamnosus reduced GABA expression in some areas of the brain, while increasing it in others.  The overall effect was to make the mice more calm.  Here in the Dark Lab, we would test this by subjecting the rats to endless episodes of Jersey Shore, Barney and Friends, and The Jonas Brothers, and then asking how long it takes before they fall into convulsions.  Bravo measured stress-induced hyperthermia (rise in core body temperature from stress) after a battery of different tests, including  forced swimming and mazes (less barbaric than listening to the Barney jingle, but presumably effective at producing stress) and showed that the L. rhamnosus-fed animals exhibited less stress during these activities.

The final point of the paper was what I found most interesting.  Some of the animals had the vagus nerve cut prior to the start of the experiment.  This nerve is a direct link between the gut and the brain and is responsible for transmitting signals about hunger and satiation.  In these animals,there was absolutely no effect from consuming L. rhamnosus.  No changes in GABA expression and no behavior changes versus the control animals.  This means that the probiotic bacteria that colonize in the gut might actually use this nerve to signal directly to the brain.  Although these bacteria appear to provide a health benefit, I can certainly imagine other strains that are more nefarious.  Yeah, I'm talking about zombies again.  This phenomenon is not all that different from what is seen in T. gondii or the zombie ants... microbial agents that manipulate the brain function of the host.  If a probiotic strain can mimic the effect of caffeine-overstimulation or, even worse, caffeine-deprivation, then that would be a whole new kind of scary.  Anyone who has seen me before that first cup of coffee knows what I'm talking about... it's bad enough to give a zombie nightmares!

Friday, August 19, 2011

Zombie rats are horny!

You may remember a very early post on zombies, where I talked about a parasite known as Toxoplasma gondii.  I just read a new article published in PLoS that is really mind-blowing.  It also should fall in the category of science known as what-kind-of-PhD-do-I-need-to-study-this-shit.  Very interesting, very strange, and very cool.  Here's the abstract to check out for yourself (abstract).  As you know, rats that are infected with T. gondii lose their fear of cats.  This is important for the parasite because part of its life-cycle occurs in the gut of the cat.  This report takes that one step further.  House et al. show that as part of this shift in rat behavior, the rats are actually attracted to cat urine.  Not attracted like "this smells like roses" but rather "yowsa, hot babes" (rats do say 'yowsa'... I've heard them).  Yes, the parasite changes the way the rat brain responds to the smell of cat urine.  Neurons in the ventromedial hypothalmus, dorsomedial part (mercifully abbreviated as VMHdm), which normally are active in response to fear, are silent when infected rats are exposed to urine.  In contrast, the area in the brain the DOES light up is the posterodorsal medial amygdala.  These are the same neurons that light up when rats watch porn (or are exposed to estrous females, but somehow that doesn't sound as sexy).  Their conclusion is that T. gondii makes rats think that if they follow the cat pee, they will get laid.  Although that may work in some dive bars, for these rats it ultimately leads to just a single dinner date.

These results are also consistent with other findings that show an increased level of dopamine in infected rats.  Dopamine is, of course, the primary molecule in behavioral reward so it seems reasonable that this pathway would also be effective at shifting rat behaviors.  They cite a 2006 paper by Webster et al. (abstract) where it was shown that dopamine receptor antagonists prevent the attraction to cat urine.  Could this receptor be the first therapeutic target for treating zombies?  I'll have to propose that at the next New Target meeting. 

The ability of T. gondii to alter brain function and behavioral response is incredibly cool and a little bit scary.  Why scary?  Because it is estimated that at least one third of the human population has been exposed to the zombie-inducing T. gondii (and no, it's not just pop stars and politicians).  Even subtle changes in behavior on that scale can lead to massive changes in society.  Although it is unclear how well this study translates to humans, it does suggest that in the event of a zombie apocalypse, it's probably not a good idea to hide out in the girls' bathroom.

Wednesday, August 10, 2011

Highlights from the Protein Society Symposium

A week ago, I attended the 25th annual symposium of the Protein Society in Boston.  I've gone to this meeting three times and this was the best yet.  Very, very cool stuff.  I also got to see a lot of old friends from grad school, including my graduate advisor.  It was great to hear him talk fondly about the 'good ol' days' (ie, when I was his student) as I distinctly remember them being less fond and more frantic.  I think finishing a PhD thesis is the mental equivalent of giving birth to an elephant while running a marathon, but for several obvious reasons I will never be able to test the hypothesis.  I also met some cool new people.  I talked politics with a girl from Poland in a loud Irish pub and met another girl who is some kind of flute prodigy from a well-known coffee empire.  At the reception, I met a guy from, well, some European country who had done a postdoc in San Diego and I knew many of the trails he had hiked.  We tried watching the Red Sox game from the 50th floor of the Prudential building... great view but when the outfielders look like fleas on a green dog, it's really hard to see what is happening.  We kind of made it up as we went along and since the Polish chick didn't know the game, it was all good.  He still owes me pictures of Fenway.  The poster session was crazy, with two overlapping sessions and the very friendly (but bored) vendor who bribed me with chocolate every time I passed her booth.  I was also a poster judge this year, so I missed most of that session, but tracked the presenters down later to ask questions.  It's a little awkward at the coffee breaks, since everyone is staring at the nametags trying to find people they want to talk to.  I would try and catch a glimpse of their badge over the rim of my coffee and hope they were not offended when I simply walked away.  It's a very unusual hierarchy at conferences.  There is absolutely no guessing about where you stand in the pecking order.

Anyhow, I can't talk about the things I found most exciting because I was there for work, and work stuff has to stay off the radar.  However, let me briefly describe two (not work-related) things that were pretty cool.  One was a talk by Della David at UCSF on protein aggregation as a part of aging.  I don't know a lot about this field, but one of her early slides really caught my attention.  She was discussing the role of protein "aging" in inducing aggregation using C. elegans (a worm) as a model.  As the worm aged, she showed that the concentration of over 400 different proteins increased in the insoluble fraction.  In simpler terms, if you take all of the proteins out of the worm, many of them are soluble but some fraction are in an aggregated form, which is not soluble.  Although the total amount of protein seemed fairly constant with age, the proportion of aggregated protein increased and seemed to disrupt the natural process of homeostasis.  Then came the kicker... to show that this was an active process (that is, controlled by a cellular system) she used a C. elegans that had been engineered to have a specific mutation in the Daf-2 receptor.  These worms had twice the life span of a normal worm.  Whoa!  Sign me up for that mutation! Imagine living 160 years!  It turns out that the fraction of aggregates is independent of the lifespan, suggesting that the process is controlled.  Two things here... Daf2, which is part of the insulin/IGF-1 signaling pathway, can regulate lifespan (possibly related to the observation that mice that eat less live longer?) and that protein aggregation as a result of aging could also be controlled.  Here is a review on insulin/IGF-1 signaling in aging (abstract) and here is David's recent open-access paper covering some of this story (PLOS paper).  Listen folks, please hurry up with this important work... I'm not getting any younger.

The other talk I liked was by Ken Dill (a long-time favorite of mine and also from UCSF).  I'm used to him talking about transfer free energies and lattice models for proteins but this time he was talking about the stability of the proteome.  He (and others) have shown pretty convincingly that on a macro scale, protein stability is roughly dependent on the length of the protein.  (Seems simple but it has taken decades to model it in a way that makes physical sense).  Armed with this model, he determined the stability of the entire proteome and found that it is only marginally stable.  Over 500 proteins have stabilities less than 3 kcal/mol, which means they are barely folded and functional.  The implication of this result is that even slight increases in temperature can cause many of these proteins to unfold.  The resulting denaturation catastrophe overwhelms the cell and causes cell death.  This is the most plausible explanation yet for why slight increases in temperatures cause such problems (even for humans, an increase in body temp of 7-8 degrees can be fatal).  I asked him about the proteome of thermophilic bacteria and whether it might explain their ability to survive extreme temperatures and he said he is working on that now.  I'm guessing that might explain some of the adaptability, although the detailed mechanism is still a mystery.  For you DIYbio people out there, this model provides a pretty simple way to do this type of analysis yourself. The simplicity of the model, and the fact that minor ensemble changes can be magnified into major improvements for the organism tell me that life might be lurking everywhere there is an energy gradient (I'm looking at you, Titan).  On the flip side, it shows how sensitive life can be to slight changes in the environment. Here are the links to the articles (proteome stability and Dill's model)

  Dill also started off with a joke: "There are three kinds of mathematicians... those that can count, and those that can't."  Nothing like a geek joke to start off a talk... but hopefully his material will be better in San Diego next year.

Tuesday, August 2, 2011

GFP Beagles: Disease model or designer pet?

A recent paper in the journal genesis describes the production of transgenic beagles that glow when exposed to UV light (abstract).  The eGFP transgene was introduced into a beagle embryo using a similar (but much improved) technique that was used to clone Dolly the sheep.  Making a puppy that glows is not really new, as it has been demonstrated by the same group in 2009 (abstract) and others (in dogs, as well as other small mammals) but there are two cool things about the recent work.  First of all, they put the transgene under a promotor.  This means that the puppy does not glow green until the gene gets switched on, in this case by the drug doxycycline.  Feeding the dog low levels of doxycycline induced the expression of the GFP protein (green glow) and the effect could be turned off simply by removing the drug from the dog's diet.  This is pretty easy to do in small animals (like rats and mice) but pretty complicated in large animals. 

  In a separate article, also in genesis, they report the transmission of the transgene to offspring (abstract).  This was also interesting, as the GFP-containing females were totally fertile and had normal pregnancies and births.  The dads were wild type beagles so roughly 50% of the offspring carried the transgene.  This is consistent with stable germline transmission.  This result suggests that we are really not far off from having designer dogs.  The initial cloning will be hard (efficiency is still in the 1-5% range) but if the transgene is stable, simple breeding techniques should be enough to make zillions of glowing puppies!  Although this is a crude example, one can envision a vastly different world where hair color, facial and body features, and perhaps even personality traits could be engineered.  Genetic defects and disease determinants could be engineered out.  One could build the perfect dog.  You did realize I was still talking about dogs... right?

Friday, July 15, 2011

Foxfire: Chemistry of the undead


Ghostly mushrooms

I am currently approaching the half way mark in my current work-in-progress (WIP-2) and have been making particularly heavy use of foxfire.  I didn't start out to write about that... in fact, this book started out as a humorous middle grade adventure and quickly turned into a dark, YA biopunk.  Write what you know, I suppose.  One of the emerging themes is the struggle between the protagonists and the technologically-advanced fungi that exists in their (slightly dystopian) world.  So, I found it a little coincidental that a report came out a week ago by Marina Capelari and colleagues about a type of ghost mushroom that had been re-discovered in a Brazilian rainforest after being extinct for over 150 years (abstract in the journal Mycologia).  The mushroom, formerly known as Agaricus gardneri, is notable for its bright and constant bioluminescence.  To understand why this is unusual, here is a brief description of how foxfire comes to be:

Bioluminescence is generally accepted to come from a 2-step reaction.  A chemical called luciferin (L) is first reduced (to LH2) and this reaction is catalyzed by an enzyme called reductase.  NADH is a molecule (di-nucleotide, actually) that is a cofactor in many redox reactions.  Its basic function is to move protons around (you're a geek if you noticed the chemistry pun).

L + 2NADH <--> LH2 + 2NAD+

Reduced luciferin is then oxidized (to LO) by an enzyme called luciferase.  This process also produces a photon of light and is the source of the creepy glow.

LH2 + O2 <--> LO + H2O+ LIGHT


Illudin S: Potential substrate for ghost fungi
Why am I cryptically showing fungal luciferin as L, instead of showing the chemical structure?  Could it be that my chemistry skills are so bad, I couldn't tell the difference between L and LH if my life depended on it?  Well, yes -- but it is also true that the luciferase substrate in fungi is not well characterized. The luciferin for A.gardneri is probably a member of the sesquiterpene family, most likely an illudin.  Some of these compounds have been studied as anticancer agents but the illudins tend to be extremely toxic (possibly another reason they are called ghost mushrooms!).  Interestingly, other luciferins (such as those found in fireflies, shrimp, etc) have totally different chemical structures, which gives them different biological properties and unique spectral characteristics (ie, different colors, brightness, etc).  Changes in the luciferin structure, amino acid substitutions in the active site of luciferase, and varying levels of oxygen or water can each contribute to changes in the emission of light.  What is unusual about A. gardneri is that unlike other species, the bioluminescence is almost constant.  In fireflies, the luciferin is released when they want to blink and in the case of other species, they light up only after contact (probably a means of self-defence).  So why does this mushroom glow all the time? No one knows yet.  The biochemistry of these things is almost as mysterious as seeing their eerie ghostly glow on some rotting tree stump at midnight.

However, it provides a great real-world example of the potential technology for my story.  It doesn't take much imagination to think that these mushrooms could be engineered to be very bright or to respond in controlled ways depending on environmental input.  A basic example from my WIP is that these types of fungi are used for lighting underground.  No electricity required, no pollution, and little maintenance.  They are almost the perfect type of lighting... or are they?  Anyhow, I thought it was a pretty clever idea early on until I found out that Ben Franklin used foxfire from mushrooms to light the inside of one of the first submarines.  Was there anything this guy didn't know about?  I guess he's going to have to go on my list of card-carrying biopunks.







Friday, July 1, 2011

Resveratrol from red wine: An exercise mimetic?

I've posted before about the magic of polyphenols in wine.  A new paper out by Iman Momken et al. in the FASEB Journal (abstract)  now suggests that one of these polyphenols can protect against muscle wasting and bone loss as a result of inactivity.  The group suspended rats by the tail to prevent their hind limbs from significant weight-bearing exercise in an attempt to model the situation during spaceflight (the main focus of the paper).  One group was treated with resveratrol (aka RES, a red wine polyphenol) at 400 mg/kg per day and compared to a control group receiving no treatment or normal rats (no leg suspension).  Over a two-week period, they studied both the physiological changes in the muscle and bone, as well as biochemical pathways involved to better understand the biological function of the polyphenol. 

The physical benefits were fairly clear.  They observed significantly reduced muscle atrophy and much less bone demineralization in RES-treated rats, suggesting that the compound was protective.  The interesting aspect was in the biochemical details.  It is well known that extreme lack of muscle usage (for example, in cases of long-term bed rest) can induce insulin resistance in humans.  In this study, they claimed that rats treated with RES did not lose insulin resistance. However, I thought this was the least convincing data in the paper.  Some of the differences were significant, but I thought the overall effect on insulin/glucose levels was pretty modest.

A much more convincing effect was observed in the bone and muscle.  They monitored a number of biochemical parameters and found that suspension of the hind limb led to significant changes in the morphology and function of muscle tissue.  All of these changes were consistent with atrophy.  Rats in the control group did not have this effect.  In the RES-treated group, suspension of the hind limb was found to produce little or no changes to muscle.  Then they show that specific biochemical pathways are involved in the protective effects of resveratrol, particularly those involved in oxidative stress and fatty acid metabolism. In plain English, they found that even though there was no weight-bearing exercise to stimulate cellular activity, resveratrol was able to preserve these activities and prevent muscle degradation.  That is, it acted almost like an exercise mimetic.

Does this mean we can forget the gym and just drink our way to better health? Can we have a glass or two of wine while watching Buffy the Vampire Slayer and call it exercise?  Probably not.  The amount of resveratrol in a typical glass of wine is less than a milligram.  The 250g rats in this study received 100 mg, so the observed benefit came from the equivalent of 100 glasses of wine per day.  Your muscles are really gonna need that resveratrol if you spend every day passed out next to the TV.  However, it does further illustrate the potential health benefits of these wonderful polyphenols.  Maybe on those cold, snowy days in winter you can just pop a RES pill and get the same benefit as a walk around the neighborhood. For you health nuts, you can chase it down with a glass of good Cabernet.

Saturday, June 25, 2011

Preliminary functional annotation of O104:H4 genes/proteins by Era7 Bioinformatics

Just a quick follow-up to my last post... here is a pdf of a paper (link) from the Oh No Sequences group at Era7 Bioinformatics that lists the full functional annotation of O104:H4.  It represents an amazing amount of work and is a great reference for anyone studying the O104:H4 strain. Major kudos to the ONS group!!

Monday, June 20, 2011

Using O104:H4 EHEC data... an example

I've had a few requests for an example of how to work with the new EHEC data.  I agree it can be very overwhelming to have  hundreds of Megabytes of genomic data, so here is a fairly simple example of what one might do and what you might encounter.  Suppose you had a drug (antibody, peptide, small molecule) and you knew it hit a protein called EprK.  EprK is an approximately 250 amino acid protein that is part of the Type III Secretion System (T3SS).  The T3SS is the cell-surface protein complex that attaches the pathogenic bacteria to the host cells.  Blocking proteins like EprK is one possible way to prevent EHEC pathogens from attacking normal cells and causing disease. Your drug works on other EHEC strains (such as O157:H7, the strain responsible for the 2006 outbreak in the US) but will it work on O104:H4?  Testing it directly is the best way to know, but obtaining the new strain is likely to be very difficult.  Another option is to go to the sequence data. 

I went to one of the sites that has the new sequence information (based on 'crowdsourcing' from various labs) on O104:H4 (I used the oh no sequences blog -- the blog for the R&D section of era7 bioinformatics) and found the identifier code for the EprK protein (here's the link).  Some of the data has been annotated based on sequence homology and EprK was one that has been identified.  Using this code, I found the DNA sequence and copied it to the clipboard.  Then I went to the NCBI website (link) and pasted the DNA sequence into the search box to do a BLAST search of all microbial genomes that have been sequenced.  There were dozens of hits, and nearly all of them were EprK proteins from various strains.  I found the O157:H7 strain and the alignment is impressive.  More than 95% of the DNA bases are identical between the two, suggesting that the two proteins are very similar. I've included the BLAST results of my search below using O104:H4 EprK (Query, top strand) and it's alignment with O157:H7 EprK (bottom strand). So, your drug probably works on the new strain too.  If you want the amino acid sequence of the O104:H4 strain, simply take the DNA sequence to ExPaSy (link) and translate it.  It actually took me a bit to get the protein sequence because there is a frameshift mutation in the O104:H4 sequence read.  If you scroll down to my alignment and find the part highlighted in red, you will see there is an extra adenosine (an 'A' base) in the O104 sequence.  This throws off the protein translation.  I assume it is a mis-read in the O104 sequence (a common mistake when the sequencing machine reads through a string of the same base) and deleted it when I translated from DNA to protein.  The resulting amino acid sequence (pasted below) is very similar to EprK from other EHEC strains.  I'll double check this and follow up with them.

     Anyhow, I don't think there is a structure for the EprK protein, but if there was, you could use the existing structure as a model and make the amino acid changes seen in the O104:H4 strain to give you a decent starting point for the structure-based design of new drugs.

 Find a pathogenic protein of interest and try this yourself... it's not too hard.  When the topic of EHEC comes up at the next party, you can impress your friends by saying you blasted several virulence factors and found them to be quite similar/different from strains of previous outbreaks.  I would do this myself but, oddly enough, I don't get invited to parties anymore.  Anyhow, as a final disclaimer... although I have tried to be careful please verify anything I have posted before use.



Query  1       GTTGAGGATGAATATAACTAATTGGATCATATATAATCTTTCTTAGGGCAAGATTCATAA 
               |||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||
Sbjct  443403  GTTGAGGATGAATATAACTAATTGGAGCATATATAATCTTTCTTAGGGCAAGATTCATAA 

Query  61      CGCTCTCATATGTCTACTTAATTTTCAACCTGACTAAATTAGTTAGAATGGCCCTATACT 
               || |||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sbjct  443343  CGTTCTCATATGTCTACTTAATTTTCAACCTGACTAAATTAGTTAGAATGGCCCTATACT  443284

Query  121     TCCATAACAGCCAGCAAGTCGCTACGGATATTAATGCAAGTAAGATAGAAACCGGCATAG 
               ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sbjct  443283  TCCATAACAGCCAGCAAGTCGCTACGGATATTAATGCAAGTAAGATAGAAACCGGCATAG  443224

Query  181     CCTTATCATAAGCAAAAACAGGTTCGCTAATTTCATATGTTGGTGCTTGCTCAATAATGT 
               ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sbjct  443223  CCTTATCATAAGCAAAAACAGGTTCGCTAATTTCATATGTTGGTGCTTGCTCAATAATGT  443164

Query  241     CTCTTCGTTTTGACAATACAACAGAAATATTTTCATATTGTACGCTTGCAGAGCTATTAA 
               ||||||||||||||||||||||||||||||| |||||||||||||||||| | |||||||
Sbjct  443163  CTCTTCGTTTTGACAATACAACAGAAATATTCTCATATTGTACGCTTGCAAAACTATTAA  443104

Query  301     CAATAAATCTCTTGATATCATTTATTTTTATTTCTGGGTTGATATCTTTTTCATATACTG 
               ||||||||||||| || |||||||||||||||||||| ||||||||||||||||||||||
Sbjct  443103  CAATAAATCTCTTTATGTCATTTATTTTTATTTCTGGATTGATATCTTTTTCATATACTG  443044

Query  361     CAAGTACAGAAATATGAATTGGTAAAGCAGTTTTACCACTATCGCCATTATCAACATCGT 
               ||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||
Sbjct  443043  CAAGTACAGAAATATGAATTGGTAAAGCAGTTTTACCACTATCGCCAGTATCAACATCGT  442984

Query  421     AACTAACATGTACTCTCGAAGAAATAATGCCATCCATAATTTTGAGAGATTGCTCTAACC 
               |||||||||||||||||||||||| ||| |||||||||||||||||||||||||||||||
Sbjct  442983  AACTAACATGTACTCTCGAAGAAACAATACCATCCATAATTTTGAGAGATTGCTCTAACC  442924

Query  481     GCTGCTCAATAGCAGAATATAGCCTTGCTTTTTCCGCTCGTGGAGATGAAAACGAGTGCA 
               ||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||
Sbjct  442923  GCTGCTCAATAGCAGAATATAGCCTTGCTTTTTCCGCTCGTGGAGATGAAA-CGAGTGCA  442865

Query  541     TCTGCAGGGAACATCTGCGATATTTGAATATCAGGCTTACCCGGTAGATTGTAGATTTTT 
               |||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||
Sbjct  442864  TCTGCAGGGAACATCTGCGATATTTGAATATCAGGCTTACCCGGGAGATTGTAGATTTTT  442805

Query  601     AGCCAATCCACCGCAGAAGCAAAATCCGTTGGTTCGACAAATATTGAAAATCCTGTTTTG 
               ||||||||||||||||||||||||||||||||||| ||| | ||||| ||||| ||||||
Sbjct  442804  AGCCAATCCACCGCAGAAGCAAAATCCGTTGGTTCAACATAGATTGAGAATCCAGTTTTG  442745

Query  661     CCTTGATCCTTCTTTTCAGCATTAATATTATGTCTTTGTAAAACAGCAAGGACATCATTA 
               ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sbjct  442744  CCTTGATCCTTCTTTTCAGCATTAATATTATGTCTTTGTAAAACAGCAAGGACATCATTA  442685

Query  721     GCTTGCTGTTGATCAAGATGGTTCAATAATTCCTGCTGCTTGCAGCCGCACAACAGCAGG 
               ||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||
Sbjct  442684  GCTTGCTGTTGATCAAGATGGTTCAGTAATTCCTGCTGCTTGCAGCCGCACAACAGCAGG  442625

Query  781     ATAAACAATAATA  793
               |||||||||||||
Sbjct  442624  ATAAACAATAATA  442612


Predicted amino acid sequence for O104:H4 EprK protein, (corrected for gap): 

L L F I L L L C G C K Q Q E L L N H L D Q Q Q A N D V L A V L Q R H N I N A E K K D Q G K T G F S I F V E P T D F A S A V D W L K I Y N L P G K P D I Q I S Q M F P A D A L V S S P R A E K A R L Y S A I E Q R L E Q S L K I M D G I I S S R V H V S Y D V D N G D S G K T A L P I H I S V L A V Y E K D I N P E I K I N D I K R F I V N S S A S V Q Y E N I S V V L S K R R D I I E Q A P T Y E I S E P V F A Y D K A M P V S I L L A L I S V A T C W L L W K Y R A I L T N L V R L K I K

Tuesday, June 14, 2011

Amazing new paper on Zombie Ants

As some of you may know, I’m a big fan of zombies.  Not the feet-dragging, flesh-rotting stereotypes found on B-grade horror movies (although they can be cool too) but the ones found in real life.  The ones that make you wonder whether human zombies are for real.  The science behind these phenomena is fascinating but absolutely terrifying.  Creatures that suddenly exhibit irrational behavior or complete odd and highly specific tasks.  (Don’t worry, your girlfriend is (probably) not a zombie.)  I already wrote a bit about T. gondii (link) but a recent article in the journal BMC Ecology (abstract) describes an even more horrifying example. Zombie ants. I’m thinking this would make a great sequel to A Bug’s Life.
 

Zombie ant with fruiting body

It starts with a simple fungal infection and before long the ant is no longer following the well-marked ant trails through the Thai rainforest.  It starts staggering and has the occasional convulsion but instead of heading to rehab, it falls out of the tree and onto the forest floor.  At solar noon, the ant stops its random stagger and makes a bee-line to a nearby sapling.  It clamps its mandibles into a leaf (almost always a primary vein, under the leaf, facing NNW, about 25 cm high) and dies.  Bizarre? Yes, but to the fungus it is all part of a diabolical plan (cue music).  In order to reproduce, the fungus (Ophiocordyceps unilateralis) requires a very specific temperature and humidity.  An environment not present in the canopy (where the ants are) but uniformly at about 25 cm from the forest floor.  What’s an evil fungus to do? In order to get there, the fungus hijacks the ant and manipulates its brain by releasing various chemicals and poisons as well as making specific morphological changes to the mandibles.  All of these activities are designed to get the ant out of the canopy, go to a specific environment, and have the ant remain attached there after death.  Then the fungus sprouts a fruiting body out of the ant’s head to release spores.  All in all, the amazing transformation from ant to fruiting body takes about 2-3 weeks.  Many of the details are still a mystery but the Hughes paper begins to shed some light on this process.  A process, incidentally, that is very ancient.  Another paper by Hughes (abstract) describes fossils from the Tertiary Period (from about 50 million years ago) that bear mandible scars on primary veins of leaves.  Could these be the echoes of ancient zombie ants?  Could our own legends be the echoes of human zombies?  I wouldn’t worry too much unless your spouse’s ‘honey-do’ list becomes very bizarre or your girlfriend’s new hat looks suspiciously like a fruiting body.

Thursday, June 9, 2011

Rapid characterization of the EHEC outbreak by “crowdsourcing”

Now we’ll be moving from papayas and fish to something a bit more sinister: EHEC O104:H4.  That is the name of the E. coli responsible for the recent German outbreak.  When a new outbreak begins sickening patients, researchers all around the world are mobilized to try and characterize the pathogen.  Since new strains often have similarity to well-understood strains, one of the critical first steps is to sequence parts the genome.  When SARS was first flaring up in Asia, I worked for a sizable biotech company focused on developing drugs for viral diseases.  Very early data suggested that the SARS virus may have a similar pathogenesis to related coronaviruses, particularly with regard to viral entry.  However, we couldn’t design drugs to combat SARS until we had the DNA sequence for that part of the genome.  Once that became available, I used our in-house analysis software to design the initial set of lead drugs and we were off and running.
                Sequencing the entire genome is extremely time consuming, but BGI (formally known as the Beijing Genomics Institute) is utilizing ‘crowdsourcing’ to help assemble the EHEC genome faster (here’s the press release).  Using open source software, Twitter feeds (@BGI_Events), and several sites for uploading data, they hope to pull together data from researchers around the world in an organized, efficient manner.   Here’s the bioproject link for this work at NCBI (link).  This exchange of data is great for biopunks because one can analyze the data almost in real time and there is a significant potential for finding interesting and important aspects of the EHEC strain, based on sequence similarities/differences with other strains.  Mike the Mad Biologist had a blog post a couple days ago that offers a glimpse of the type of analysis people are doing (link).  The more eyes there are on the data, the quicker the strain can be characterized and as I have mentioned before, the potential of using ‘citizen scientists’ or ‘crowdsourcing’ for efforts of this type are enormous.  With the advent of rapidly accessible data, and the power of on-line DNA analysis tools, the gap between the scientist and everybody else has never been smaller.

Tuesday, June 7, 2011

Using DNA 'barcodes' to combat fish fraud


Snapper fillets

Now that Memorial Day has passed, we tend to do a lot more grilling here in the Dark Lab.  The weather here in SoCal falls into a predictable perfection and any given evening is perfect for throwing something on the grill.  So, I head out to my local grocery store and look for a nice fish… snapper maybe?  Looking at the package, it’s definitely a fish but is it really snapper?  I can’t tell.  In fact, studies show that up to 70% of fish sold as snapper is actually something else.  The FDA tries to monitor fish but they are probably more focused on safety rather than accuracy.  However, there has been a lot of press lately about mislabeling of fish.  Last week, the New York Times ran an article (link) with some shocking statistics about how frequently fish are mislabeled.  According to a report by the non-profit group Oceana (2.3 Mb pdf here), for every three packs of fish you buy, one of them will be wrong. 
Oceana references a number of scientific studies, including a paper by Wong and Hanner (abstract), who use a PCR-based approach to analyze the DNA sequences of fish in the marketplace.  They found that some substitutions are obvious fraud.  For example, fish labeled as red snapper (sold at $3 per pound) was actually redfish (that would cost 72 cents a pound).  Fish labeled as white tuna sushi was actually tilapia.  These are flagrant mistakes, and it is not at all clear whether this is done on purpose or is the product of the complex network of processors and middle-men that are required to bring a fish out of the sea and to your dinner plate.  However, some mistakes are less apparent… for example, Atlantic halibut was labeled as Pacific halibut.  No big deal, right?  What if you knew that Atlantic halibut was endangered?  Would you still buy it? This type of mislabeling suggests some fishermen may be catching more than their quota of threatened or endangered fish and packaging them as something else.  Another recent article goes into more detail about the social and financial implications of fish fraud (abstract).
Want to know what fish you are buying?  It’s a great DIYbio project.  If your hackerspace has the ability to do DNA sequencing (or you can send sequencing samples via the hack shack) then checking your fish can be pretty easy.  You will probably want to sequence several spots in the genome and will need sequencing primers for each (which are cheap and easy to design).  Once you have the DNA sequence from your fish, you can use an online tool called Blast (link) to search the genome database for your sequence and it will tell you what species it is from.  If you already know the sequence (from the primer design, for example), then you can simply align the correct sequence with your fish’s DNA and see if you get a perfect match.  This method will give you a pretty good idea if you have the right fish as long as there are differences in the DNA sequence between the various species.  Sometimes, they can be very similar.
If you have access to a hackerspace with a PCR machine (and the reagents!) and a way to run an agarose (DNA) gel, there are several other options.  You can do an AFLP analysis, which is a very sensitive way to look for polymorphisms (changes) in DNA.  A recent paper by Maldini (abstract) outlines the approach and applies it to identifying fish.  They claim that even closely related fish can be identified.  Another PCR-based option is to amplify a gene using a species-specific primer.  In this case, you see good amplification (ie, a band on a gel) only when the DNA of that species is present.  Two advantages with the PCR approach are that you don’t need much DNA and it doesn’t need to be all that pure (both are big advantages for the biohacker). One thing you will need is a set of PCR primers for the species of fish you are buying.  I hope that someday these will also be readily available in any decent hackerspace, but until then, you will have to get them yourself.  The Wong and Hammer paper has some primers listed and primers for key genes from the most common types of market fish are freely available on the internet.  If they can’t be found directly, you can also design them from the fish’s genome.  Genbank (link) has some of this information but another good source is the website for the Fish Barcode of Life (link).  This great organization is trying to catalog all fish, including those we eat.  Eventually, they will have links to the genome of every fish so you can use that for primer design.  As an added DIYbio bonus, they are also looking for additional data from people like YOU!  Not with the DNA sequencing, but with the development of range maps that indicate where the different species of fish are found.  This is a great opportunity for all you fishermen out there (go here to see how you can report a sighting).  It’s also a way for biopunks to make important contributions to this effort while doing a little home-based food surveillance. 

So, did you notice?  The snapper picture is mislabeled... it's actually tilapia. At least you didn't pay 10 bucks for this blog post.

Wednesday, June 1, 2011

Biopunks in USA Today

There’s a pretty nice blurb on the DIYbio movement in today’s USA Today.  It covers the basics and highlights a few of the controversies but doesn’t describe many practical applications.  It touches on DNA sequencing and open access to lab equipment and basic molecular biology tools (expression vectors and strains, common reagents, etc) and hints at one cool application (the blue yogurt).  Nice plug for BioCurious (a hackerspace in the Bay area) and the OpenPCR machine.  They even talk about the risk of making “unstoppable Franken-microbes”.  Not that we would ever do such a thing in the Dark Lab, but I do know of three fictional teen biopunks who had a basement experiment go horribly wrong…
You can read the USA Today article here but you’ll have to wait awhile to read what happened to the teens.
 
 

Blogger