Transioning from academia to industry

Behind the Front Lines: Transitioning from Academia to Industry


By Elaine To, PhD

Apologies, dear Scizzlers, for my absence from the blogosphere! Since you last heard from me, I’ve graduated, moved across the country, and started my postdoctoral stint at InCube Labs, a medical device startup incubator. While this isn’t the traditional big pharma path that comes to mind when one talks about “going into industry,” I’d like to share the differences I perceive between the academic environment and the one I’m currently in.

On my first day, I expected to be given a project immediately and was ready to start reading the relevant literature. After a tour, being shown my cubicle, and setting up the laptop I was assigned, I was told to talk to people. I had a general idea which of the company’s ongoing projects I would be contributing to, but no specific tasks. So I talked to my coworkers, got to know them, and learned their roles within the company. I attended engineering group meetings, participated in manufacturing, and caught up on current progress with all products by reading the stored data. This lasted for two weeks before I figured out what niche in the company’s needs I could fill. Once I started working on that, I was assigned an additional task that required my specific skillset. This experience highlights multiple things:

1)     Most likely, one of the reasons the company hired you is because they don’t want their competition to hire you. They value your skillset and recognize that if you’re on the job market, someone else may value you as well. Thus it’s in their best interest to keep you happy and ensure your needs are met. This is why I wasn’t given any tasks with specific deliverables until I showed I was ready: they erred on the safe side and didn’t want to overwhelm me with my own project while I was still settling in and catching up.


Tied into this is the work-life balance that my company encourages. A typical workday is 9 to 10 hours and I haven’t yet seen anyone stay late into the night to finish an experiment. To be fair, I try to do a lot of the reading in the comfort of my own home and I’m sure my coworkers do as well. However, overall there is less pressure to constantly be seen working in the lab. Need to visit the doctor? Pick up a child from school? Just need to be home early to prepare something special for your family? It’s understood and accepted. There is never an implication that your project will fail and you will never graduate because you have other priorities in your life.


If you are unhappy, you will leave, and the company doesn’t want their valued employees doing so.

2)     Getting along with your co-workers is crucial. Not only is science done collaboratively with multiple individuals working on the same task, but nobody wants to work with somebody they don’t like. It’s an uncomfortable environment that impairs productivity. Be positive, chat with everyone, and be willing to help.


3)     Being proactive and taking the initiative is quintessential. In startup environments, every employee wears multiple hats. You may have been hired for a specific skillset, but you will need to be willing to do anything the company needs that you can take care of. However, there won’t always be somebody telling you to work on a specific aspect because your supervisors may not know you have the skills necessary for it. You know your skills best and it’s up to you to apply them in the roles that create the most value for the company.

And some additional things I’ve noticed:

4)     Every company has competition, and thus every company has a non-disclosure agreement. Gone are the days of discussing your data and poring over it with your fellow graduate student friends. True, if I hid enough of the details, I could probably get valuable feedback without revealing anything, but why take the risk? When people ask me what projects we work on I am careful to only talk about what has already been revealed in articles published in the media. Additionally, (individuals who have worked with patient samples in graduate school and are familiar with HIPAA will recognize this) we must trace our trash and throw anything with business secrets into the confidential disposal bins.


5)     Networking most certainly does not end when you get a job—it’s a lifetime duty. However, as a graduate student I often felt that I was the one hoping for help from others. Now, I am in a position to provide the help not just to my fellow recent graduates, but also to individuals who have seen much more of industry than I have. Yet, I still benefit greatly from the perspectives of those individuals with more experience and they appreciate the viewpoint of somebody who isn’t so far removed from the academic realm. In networking, everybody’s on equal footing, whether you’re the senior engineer with 15 years of experience or the industrial postdoc who just started a month ago.

fountain of youth is found in youth blood

Is the Fountain of Youth in the Blood of Youth?


By Elaine To

Although the rumors of the Countess of Bathory bathing in virgin blood to restore her beauty are unfounded, there may actually be scientific merit in the myth. In two separate publications in the same issue of Science, researchers investigating aging have shown that blood, brain, and muscle tissues in elderly mice can be regenerated by the GDF11 protein found in the blood of young mice.

The researchers begin by creating parabiotic pairs of mice who are joined surgically to share a circulatory system. Pairs were created between mice of the same age (isochronic) or different ages (heterochronic), resulting in young-young (Iso-Y), old-old (Iso-O), or young-old pairs (Het-Y and Het-O). The young mice were 2 months of age and the old mice were 15 months old. After 5 weeks of being surgically joined, the mice were separated and their characteristics analyzed for the hallmarks of aging. Throughout the study, the Iso-O mice are compared to the Iso-Y and Het-Y to understand the effects of aging. Then the Het-O mice are compared to the other three and they are found to consistently resemble the Iso-Y and Het-Y mice more closely than the Iso-O mice.

In the first publication, the researchers focus on effects on neural and vascular cell and tissue regeneration. Neural progenitor cells identified by Ki67, Sox2, and Olig2 markers are noticeably decreased in the Iso-O mice, and Het-O mice recover these cell populations. Examining the olfactory bulb revealed that Iso-O mice have fewer newborn neurons, and Het-O mice have more than the Iso-O mice, though they don’t quite recapitulate neuron counts of the Het-Y and Iso-Y mice. When mice were exposed to small amounts of an aroma, the Iso-O mice ignored the stimulus while both the Het-O and Iso-Y mice spent time exploring the scent. This indicates that the Het-O and Iso-Y mice had greater sensitivity towards the smell.

Examining vasculature in the brain showed that Iso-O mice had lower blood vessel volume than Iso-Y and Het-Y mice, whereas Het-O mice were similar to Iso-Y and Het-Y. Iso-O mice also have much decreased cerebral blood flow versus Iso-Y, which is equal between Het-O and Iso-Y mice.

It became clear that the regeneration factor was in the blood serum, and GDF11 was identified as the responsible protein. 21-month old mice treated with GDF11 had increased blood vessel volume and Sox2 progenitor cells versus the untreated controls. GDF11 also increased the number of phosphor-SMAD2/3+ cells, showing that GDF11 acts through the SMAD/TGF beta signaling pathway.

The second publication focused on muscular degeneration in elderly mice. The stem cells responsible for muscle regeneration, especially after injury, are known as satellite cells. The Iso-Y, Het-Y, and Het-O mice had greater amounts of satellite cells than the Iso-O. Aging satellite cells were also associated with increased DNA damage and phosphorylation of histone H2AX. The Het-Y, Iso-Y, and Het-O mice had much less of both as compared to the Iso-O mice.

Treating elderly mice with GDF11 increased the frequency of satellite cells and decreased their DNA damage versus the untreated mice. There were also more regenerating muscle fibers after an injury. Electron micrographs of muscles showed fewer swollen mitochondria and more regular muscle fiber patterning in the GDF11 treated mice. There were greater amounts of PGC-1alpha, a regulator of mitochondrial biogenesis, and a greater LC3-II over LC3-I ratio, indicating enhanced autophagy. Lastly, the GDF11 treated mice had greater endurance while running and demonstrated greater grip strength.

This collection of papers provides strong evidence for GDF11’s ability to regenerate neurons, blood vessels, and muscles. While it may have potential to greatly improve quality of life for elderly individuals, further studies must first prove that these effects occur in humans as well and examine long term effects.

Disclaimer: we do not condone bathing in the blood of virgins, or anyone else for that matter!


DNA is made of A, C, T, G…X, and Y?


By Elaine To


In biology classes, everybody is taught that deoxyribonucleic acid (DNA, AKA the genetic information of a cell) has four and only four nucleotide bases. Adenine (A) and thymine (T) base pair together and cytosine (C) and guanine (G) base pair together. For the first time ever, researchers have expanded the genetic alphabet to include two additional bases: dNaM (X) and d5SICS (Y).

The researchers have previously shown that DNA polymerases, the enzymes responsible for replicating DNA, successfully replicate DNA containing the dNaM-d5SICS base pair. However these reactions were not carried out within living cells. The researchers decided to try this in the bacterium Escherichia coli due to the simplicity of the cells. Multiple factors had to be optimized in preparation for carrying out these reactions inside cells.

Firstly, the unnatural bases must be present inside the bacteria for DNA polymerase to use them as raw materials. Cells normally obtain A, C, T, and G from breaking down food or recycling previously used nucleotides. Both these pathways were not options for X and Y, so the researchers first tried passive diffusion across the cell membrane. Once X and Y diffused into the cell, they could then be phosphorylated by naturally occurring enzymes to their triphosphate form, which is the form that DNA polymerases recognize and use. The phosphorylation was unsuccessful.

The researchers then explored the idea of transporting the triphosphate forms (XTP and YTP) directly into the cells. Uptake of XTP and YTP by nucleotide triphosphate transporters from multiple other species was screened. The PtNTT2 transporter from the diatom Phaeodactylum tricornutum was most efficient at bringing XTP and YTP into the cells.

The next issue was the instability of XTP and YTP in the culture medium, especially when the E. coli were actively growing. Tests were first carried out on the natural triphosphate ATP. It was determined that addition of KPi to the culture medium increased ATP stability significantly and that KPi had the same effect on XTP and YTP.

And with that, the researchers were ready to generate their E. coli organism containing X and Y. They prepared two circular pieces of DNA, known as plasmids, which are easy to transport into bacteria. One plasmid contained the gene for the PtNTT2 transporter and the other contained a gene with an A-T base pair replaced by X and an analog of Y. Since YTP is the provided substrate, any newly produced plasmid will contain X and Y. This distinguishes it from the original template plasmid containing X and the Y analog.

After inserting both plasmids into the bacteria and growing them in KPi, XTP, and YTP containing medium, the plasmids were extracted from inside the cells. Analyzing the total nucleotide content with mass spectrometry showed that Y was clearly present. X was not detected, but it is known to fragment poorly and thus be difficult to detect with mass spectrometry.

To check the incorporation of XTP and YTP into the extracted plasmid, it was replicated in a PCR reaction using the natural nucleotides, YTP, and biotinylated XTP as substrates. The new product should contain biotin and thus react with streptavidin, which binds very strongly to biotin. As expected, streptavidin bound to the PCR product, confirming that the X-Y base pair is in the plasmid.

Sequencing of the plasmid shows that the nucleotide sequence is correct up until the expected location of the X-Y base pair. The sequencing reaction terminates at this location because there is no X nor Y provided in the sequencing reagents. This proves that X-Y is present in the right location in the plasmid.

In a series of landmark experiments, the researchers have shown replication of DNA containing an unnatural base pair inside living cells. The next step to be undertaken is the transcription of this DNA to mRNA and then hopefully translation into a functional protein. It is conceivable that the incorporation of X-Y into mRNA will soon transpire due to the similarity of DNA and RNA. Subsequently, work that has already been done in incorporating unnatural amino acids could be leveraged to facilitate the use of X-Y in codons that result in proteins.

Is an Industrial Postdoc Right For You?

Is an Industrial Postdoc Right For You?


By Elaine To

In recent years, postdoctoral opportunities in industry have been on the rise. Notorious for higher pay, access to greater resources, and providing a leg up for future industry positions, these positions offer many advantages over traditional academic postdocs. However, they are not for everyone; certain aspects are cause for caution. Here’s a list of questions you need to answer when considering such a position and some tips for searching out opportunities.

Questions you need to answer:

1. Are you comfortable with closing the door to an academic career? Or do you want to leave it open?

Industrial postdocs are excellent opportunities if your goal is to stay in industry, but depending on the postdoctoral program and company, returning to academia may be difficult. Academic scientists need quality publications during their postdoctoral training, and not every project in an industrial setting may be designed to do that. Additionally, it is sometimes in the best interest of the company to dissuade publications, keeping discoveries and technologies confidential for profitable development. Thus many industrial postdocs get patents, which are a mark of success for the postdoc while also protecting the company’s business aspirations. However, publications tend to be more valuable to an academic hiring committee than patents. Patents often have a large number of authors and it can be difficult to discern each individual author’s contribution. Lastly, it is unlikely that you will be able to turn your postdoctoral research project into your lab’s research focus as a PI. This is not to say all industrial postdocs close the door to academia, but a warning to choose wisely with these caveats in mind.

2. What is the company’s goal in taking you on as a postdoc?

Be wary of companies that may just be trying to take advantage of you and pay less for what should be a full time employee position. Postdoctoral positions, whether in academia or industry, are meant to be training periods. Ensure that you are not being hired to perform a single specialized assay for the entire period of time. You should still be exposed to new techniques and concepts, even if you start out doing what you’re familiar with. Try to determine who you will be reporting to and gauge whether they have the time and scientific knowledge to mentor you in your chosen project(s).

An additional question that will help you understand the company’s goal is what are the differences between a postdoc and a full employee at this company? Will you be able to work on what interests you, or only what’s assigned to you? Are all the benefits the same?


3. After the postdoctoral position ends, is there the possibility to join the company afterwards?

Some companies actively refuse to hire their postdocs. In a roundabout way, this is actually beneficial for the postdoc. If a well-known company gains a reputation for hiring their postdocs, the postdocs they don’t hire may have the following stigma attached to their subsequent job search: “Well if XXX company didn’t hire you after your postdoc with them, you must not be very good.” Other companies will not guarantee a job after the postdoc, though it is likely if you perform well. These companies are viewing the postdoc period as a time to both train you and test you. While this state of affairs offers greater security, be wary and have a solid plan in case they do not hire you. If the company is small, it is entirely possible that they do not have a position open at the end of your postdoctoral training.

How to search out industrial postdoc opportunities:

If you read my previous post on how to jumpstart your non-academic job search, the major point I emphasized was the value of networking. That is just as applicable here, whether the company you’re targeting has a formal postdoctoral program or not.

Large companies including Genentech and Novartis have formal postdoctoral programs. The positions are often advertised online and easy to find. Compared to these announcements soliciting applications, the idea of networking into a position is scary, especially when we’ve spent our lives filling out such applications to advance our career (undergraduate, graduate, fellowships, etc). However, postdoctoral positions in industry, particularly at large reputable companies, are highly desired. Just like every other job announcement, the human resources department will be bombarded with hundreds of qualified applicants for that single position. Unlike the previous graduate or fellowship applications we’ve filled out, the human resources department may not have the scientific expertise required to adequately judge your fit for the position. If somebody in your network already works with the company, you can ask that individual to contact the scientist behind that advertised position and see if s/he is willing to do an informational interview with you. At the end of that interview, if you decide you still want to apply for that position, let the professor know to keep an eye out for your resume.

Smaller companies may not have formal programs and may have never even thought of taking on postdocs. Networking is also your way in here, though you could use it to ask for either a full time position or a postdoctoral position, depending on your career goals. Companies are sometimes more willing to take on a postdoc because it is a smaller investment with the same potential for gain. Even if they have no advertised positions online, ask if they are interested in discussing the possibility of a postdoctoral position. You may be able to inquire directly without a networking connection, though networking always helps. This cold calling method may also work with scientists in larger companies, if you’re able to identify who you’d like to work with.

My experience is an example of networking success and how much the job search relies on serendipity. Early on in my search, a chance met friend from an online dating site connected me to a biotech recruiter who worked closely with several companies in my targeted geographical area. After a brief rundown of my skills, she took my resume and passed it on to the company that ultimately offered me a postdoctoral position. Without my friend, or the recruiter (who I now consider a friend as well), the company and I would have never found each other. I’m both grateful and thrilled for the opportunity. Be on the lookout for further posts on the industrial postdoc experience after I start!

5 Ways to Jump Start your Non-Academic Job Search


By Elaine To

So you’re nearing the end of your PhD. It’s a time for celebration, right? Unfortunately, the shining dream of becoming a professor in academia that you started with has faded. It’s no longer your career path, and all those seminars on how to land the perfect postdoctoral position no longer apply to you. Searching for a job in the real world sounds scary at first, especially in the current economy. Here are some tips for getting started:

1)     Build and update your LinkedIn profile

You should be able to copy/paste the statements from your resume into the “Experience” section. You can be more detailed and include more accomplishments than in your resume, but remember to use action statements and communicate things clearly. This is also the area to include the details of any volunteer work or leadership positions. Make sure to write a concise summary of your background and goals for the “Summary” section. This is what most people will read when deciding whether to delve deeper into your profile. Your current job title is just as important. Lastly, ensure the photograph is a professional well lit depiction of your smile.

2)     Build your network

Once your LinkedIn profile is setup, join groups in your field or that are associated with your university. Add individuals who you know in a professional or friendly capacity, but be wary of adding anyone you don’t know. In this way, you’ll increase the number of people you are connected with. Networking is crucial, as much of the non-academic PhD market is unadvertised. Leave no stone unturned, check your undergraduate and graduate alumni, friends, and colleagues. Talk to the career offices and use your PI’s network.

3)     Do informational interviews

Other people love talking about their experiences. You’d be surprised how many people are willing to help out somebody who was previously in their shoes! Within your network, alumni groups, or 2nd and 3rd tier connections, look for somebody currently working in a position or company you’d be interested in. Ask for a 10-15 minute phone chat to learn more about their experiences. Often the person you chat with will also ask about your background and career goals. During this chat, DO NOT ASK IF THEY KNOW OF AN OPENING. Ask about their day to day life, what skills are necessary to succeed in their position, how they got there, what it’s like to work for this company, etc. At the end, thank them for their time and ask if you can do anything for them, and also if they know of anyone else who might be a useful contact for you. The fact that you are asking about their current position lets them know that you are looking for a job, and they will keep an eye out. If you can, try to meet the individual over coffee; it will be a more personal connection.

4)     Search for job postings

If you match more than 70% of what a job posting asks for, it is worth applying to the job. Make sure to tailor your resume and cover letter for that specific announcement. If you do not match, job posting searches still let you know which companies are hiring. If it’s a company that interests you, return to LinkedIn and see if any of the employees in the company are a 2nd tier connection. Ask your shared connection for an introduction, and then follow up with an informational interview. Don’t be afraid to check smaller startup companies, who may be more willing to take someone without the specific skills on their specific instrument with this specific model system. Looking at the portfolios of venture capital firms will give you an idea of which startups are well funded and likely to be hiring.

5)     Keep an open mind

Through my informational interviews, I learned about many more career options for scientists. Before I began, I wasn’t aware that venture capital firms hired scientists to be analysts, some law firms hire scientists without patent bar registration to be technical advisors, that technology transfer existed as a field, and that there are many routes into science policy in addition to the AAAS fellowships. Learn as much as you can about any potentially interesting field before deciding not to pursue an opportunity within it.

There will be many cycles through these 5 tips, just like the many cycles that your resume and LinkedIn profile will undergo. As you continue, you will gain a greater understanding of which of your skills are desirable and how to market yourself, and use this knowledge to refine your approach. The informational interviews, whether over the phone or in person, are also excellent low pressure practice for real job interviews. Good luck! It won’t be easy, but there is a light at the end of the tunnel! Trust me.

Marathon Diaries: Season Finale

By Elaine To


After hours of toil, pain, and sweat, I am proud to say that this Sunday I finished my first full marathon. I’ll celebrate later; let’s first talk about how you can achieve the same.

So you’ve done it. For months, you ran during your incubation periods and woken up at unimaginably early hours on the weekends just for your long run. It’s now race day and time for you to cross the finish line. When months of training coalesce into that single glorious moment, you don’t want anything to stop you from achieving your goal. How can you ensure that you run the best race ever? It’s not much of a surprise, but the pre-race preparation doesn’t just begin the day before! Here are some tips for preparing:

Days before:

1) Hydrate well in the week leading up to the race. It’s not just something to do a couple days before! Make sure you’re getting 8 glasses of water a day.

2) Your training should have tapered in the 2-3 weeks prior to the race. Any runs should still provide you an adequate workout, but be comfortable. Now is not the time to push your limits!

3) Get a good night’s sleep on the two nights prior. You’re likely to be nervous and unable to sleep well the night before, so build up a buffer with the previous night.

4) On the day immediately before, pick up your race packet and make sure you have 4 safety pins for the bib. Double check your running gear, because if you’re missing anything, the expo that accompanies the packet pick up is the place to buy it.

5) The day before the race is also the time to carbo-load. You want your glycogen stores to be as full as possible, so eat a lot of pizza and pasta. Avoid anything you don’t regularly eat or that has a chance of causing stomach problems.

6) Pin your bib onto your shirt and pack your gear check bag the night before. This isn’t something you want to worry about on the morning of. Good things to put into the gear check bag are warm clothes and granola bars.

7) EAT BREAKFAST. It should be mainly carbohydrate based.

During the race:

1) You can come to the start line wearing extra clothes so that you stay warm until you start. Most races donate clothes that are left behind to charity. If you don’t have anything you’re willing to toss, buy something from a thrift store.

2) Don’t be stressed or nervous! You should be running at a pace that allows you to talk to those around you so don’t be afraid to chat it up with your fellow runners. This will prevent you from trying to run too fast, especially in the beginning. Near the end, your fellow runners are also a great source of encouragement.

3) Do not drink at every water stop unless drinking water every 1.5 miles was normal during your training. If you overhydrate you may deplete your electrolytes, and even if you’re drinking the sports drink, you may get too full. Have faith in your training and follow that. To alleviate the fear of needing water but being far away from the next water stop, I would suggest carrying a bottle that you can refill at each stop. Also make sure you eat during the race, whether its energy gels or provided snacks. Again, be careful to follow what was normal for your training—do not overeat.

4) Make it a goal to high five or thank every single volunteer that you see. They are donating their time so you can run and they are behind you all the way! The morale boost you will get is absolutely crucial for the second half. On this note, you can try to wear something that makes you stand out so you can tell when cheers are directed towards you. Some wear flamboyant clothing such as tutus or hats, many have sentiments on their shirts. In the past my shirts have said “Hopkins,” and supporters have used that. The Virginia Beach bibs actually had our names printed in bold font; it’s an amazing experience to hear “Go Elaine! You’re rocking it!” from complete strangers.

5) If this is your first full marathon, it’s almost guaranteed you will be in pain during the second half. Don’t be afraid to walk and take stretch breaks. Finishing a couple minutes later is better than injuring yourself. On hilly courses, the uphill is particularly painful but reprieve comes during the downhill stretches. Flat courses use the same muscles constantly, so in some ways they are more difficult. I found that even the brief moments spent in a porta-potty revitalized me and helped me come out much stronger and faster.

6) Hopefully your family and friends have signed up for runner tracking so that every time you cross a checkpoint they get a notification. Think about this! Think about all the people who are behind you, wishing for your success, and who will congratulate you in the end! Don’t lose hope!

Crossing the finish line & afterwards:

1) Hold your hands up high, smile, and make sure you get that nice photo finish!

2) Hobble towards the gear check and food lines. Grab your medal(!) on the way. Most races give out mylar sheets so you can stay warm until you get your gear. It’s a good idea to wrap yours securely around yourself before you try to pick up the water and snacks provided.

3) Walk, walk, walk. The more you walk and stretch now, the less pain you’ll have later.

4) Eat a hearty meal that includes both carbs and protein as soon as possible. Rebuild and refuel.

And that’s it! Pat yourself on the back, happily receive the congratulations from your friends and family, take a nap, and then go back to lab. With one big race completed, it’s time to bring that other race to completion: your doctoral thesis!

The Great Pokémon Experiment


By Elaine To

After 16 days, 7 hours, and 44 minutes, one of the most immersive social experiments of the year came to an end. I’m talking, of course, about the Twitch Plays Pokémon (TPP) stream.

TPP was an attempt to play one of the original Pokémon video games using input provided by anyone currently in the chat room. Because the early generations of Pokémon accepted a limited number of commands (start, select, up, down, left, right, a, b), and due to the continuing popularity of the franchise, it was the ideal format to unleash on the internet community and see what would happen. Essentially, over 30,000 individuals were trying to control and complete a single video game, a task made even more difficult by bots spewing random commands, trolls who wanted to watch the world burn, and a 20-30 second lag delay between input acceptance and input execution. TPP rapidly gained in popularity, garnering articles in the mass media and a webcomic from XKCD, spawning a highly active subreddit, and possessing 36,505,439 total views with 194,697 likes. Although the game has been beaten and is taking a break before it starts the next Pokémon game in the series, the inactive stream currently has 29,714 viewers.

Few are aware of the spinoff, RNG Plays Pokémon. Instead of taking input from individuals in chat, RNG took input from a random number generator. The stream has currently been running for 10 days, 19 hours, 49 minutes and still has the arduous trial of Victory Road (puzzles and strong trainers) before it can even think of challenging the Elite 4 and final champion (5 notoriously strong trainers) to beat the game. RNG is running at ~500% normal speed, so in relation to TPP, RNG has played for 54 days, 3 hours, and 5 minutes without completing the game. It has 331 current viewers, 657,641 total views, and 3,396 likes, and the subreddit does not have its own unique culture.

I followed both streams as they happened, and thought this could make an interesting scientific analysis, comparing the purely random approach (RNG) with an approach where everyone has a voice but not the same goal (TPP).

Firstly, neither approach got anywhere completely on its own. When TPP got stuck in a floor maze for 25 hours, the creator implemented a new democratic system that allowed players to vote on which command was next. Votes could also be used to switch between this democratic system and the previous anarchic system. Because players who wanted to progress in the game outnumbered those who didn’t, democracy became the solution whenever the disorganization of anarchy couldn’t conquer an obstacle. Many insisted that democracy removed some of the novelty of the experiment, but without it, TPP might never have made it past the maze, certain ledges, or a later puzzle on Victory Road. RNG’s first run resulted in disaster, after he released its Charizard (really powerful) to leave itself with only a level 7 Magikarp (really useless) and no money to buy more pokeballs to catch new Pokémon. RNG restarted, but the new character then wouldn’t leave her room for at least 2 hours. In the current attempt, RNG has finally acquired all 8 gym badges with the help of 10 interventions to point him in the right direction. Additionally, the RNG input isn’t completely random; the creator decreased the frequency of start and increased the frequency of A for faster progress.

In both games, Pidgeot was a great boon to the team. TPP’s level 69 Pidgeot, nicknamed “Bird Jesus,” was consistently the highest leveled Pokémon owned until TPP captured the legendary Pokémon Zapdos. RNG’s Pidgeot was level 44 when it was released, but it is one of only two Pokémon that RNG has ever owned that were above level 35. This is a highly effective endorsement of Pidgeot’s power even in the face of disorganized or random input. You can bet I’m putting one in my party next time I play a game of Pokémon.

Let’s look at the initial starting Pokémon that each game received. Both picked the fire type starter, perhaps because that is the closest option presented. While RNG’s Typhlosion is currently level 100 (it’s carrying the whole team), TPP’s Charmeleon “Abby” accidentally got released 4 days into the game at level 34, prompting much sadness. I’d say RNG succeeded here.

The use of the Masterball also provides an interesting contrast. The Masterball is a pokeball that immediately captures any Pokémon and is only available once in the game. With a normal pokeball, the player would have to whittle down a Pokémon’s health without knocking it out and even then it’s not a guaranteed capture. TPP managed to use the Masterball on Zapdos during its only chance to capture the powerful legendary bird (in anarchy mode, to boot!). RNG wasted it on the easy to find Tentacool. Woops.

Releasing Pokémon was a plague to both games, resulting in mounting fear every time the game character approached a PC. Here’s a brief explanation for those who are unfamiliar: the player can only carry 6 Pokémon around with him at a time. All other Pokémon are stored in an electronic system and are accessible through a PC in the game. However, the PC interface is also where Pokémon can be released. RNG’s release of his Charizard in his first run doomed him, and in the current run he has released 15 out of the 22 Pokémon he has ever captured. Thankfully, only his Pidgeot was a particularly valuable team member. TPP’s releases have been much more tragic. The first release was 4 days in, wherein TPP released Rattata lovingly nicknamed “Jay Leno” along with the aforementioned Abby. After receiving a Hitmonlee and nicknaming it “C3KO,” TPP released it 37 minutes later while trying to withdraw it from the PC. On day 10, now dubbed “Bloody Sunday,” TPP released an unprecedented 12 Pokémon while trying to withdraw the freshly caught legendary Zapdos. Among these 12 were the beloved Cabbage (level 26 Gloom), DUX (level 31 Farfetch’d), and BigDig (level 29 Raticate). Although BigDig was known for randomly using Dig to return to the Pokémon center and temporarily resetting progress, he was greatly missed when TPP got stuck in the Pokémon mansion (see below). Thankfully, Zapdos was successfully withdrawn, though it was also one nerve-racking A away from release. Overall, TPP released 20 out of the total 31 captured Pokémon.

It is worth noting that TPP was playing Pokémon Red, part of the first generation, whereas RNG’s current run is in Pokemon Crystal, part of the second generation. RNG did not have to deal with the floor maze that impeded TPP for so long, and by the time RNG reached the difficult maze in his version, the coder knew to intervene and help him through. RNG also did not have the chance to get stuck (source: reddit) in the Pokémon Mansion, where TPP trapped himself after battling a roving trainer. The only way out was for TPP to roam in the 2 spaces he was permitted, encountering wild Pokémon and battling them until his Pokémon fainted and he respawned at a Pokemon center.

Despite these advantages, RNG did not progress as far into the game as TPP even though he has been playing for three times longer. It is encouraging that if we ignore the 500% run speed of RNG, he progressed further in ~11 days than TPP did. Both games were fraught with disaster and backtracking, but both games also progressed further than I have in my current Pokémon game. (To be fair, I don’t play 24 hours a day, though I started this game over 2 years ago.) It seems the disorganized multi-input multi-goal approach is more efficient than the purely random approach for accomplishing goals. Additionally, TPP was much more effective at building a cult following. The popularity likely arose from the fact that each individual could participate and influence the game. Watching a game character stumble around aimlessly for hours is more exciting when one of your inputs contributed to the madness.

As I write this, TPP plans to start a Pokémon game from the second generation in 15.5 hours. This may provide a more effective comparison to RNG, as the puzzles and obstacles will now be the same. RNG may stop wasting time battling level 2 Pidgey with his level 100 Typhlosion and start heading in the direction of Victory Road…eventually.

Praise Lord Helix.

Microscopy, Mice, and HIV


By Elaine To

Monkeys infected with simian immunodeficiency virus (SIV) have been the traditional animal model for the study of HIV pathology. However, SIV does not result in the same immunodeficiency that HIV does, and monkeys are expensive to care for. Mice without immune cells can be engrafted with human immune cells and used instead. The specific model used by Ladinsky et al transfers human fetal thymic and liver tissues along with hematopoietic stem cells. These mice, known as BLT mice, reconstitute human immune cells in significant levels in many tissues, and HIV infection results in T cell depletion.

Ladinsky et al. use a powerful microscopy technique known as cryoelectron tomography in addition to immunofluorescence to understand the characteristics of HIV infection in the small and large intestines in BLT mice. The interior of the small intestine has an upper layer that includes the villi, known as the lamina propria. Between the villi in the lower layer are intestinal crypts, and this is where the majority of HIV viruses were located. Any villi that had evidence of HIV were also adjacent to an infected crypt.

Looking closer, the researchers were able to see individual viruses in the process of budding out of infected cells. It was possible to distinguish between mature and immature viruses based on the differences in internal structures. An examination of the viral pools located outside cells showed that 90% of the pools were mostly mature, but 10% were mostly immature. This is in contrast with previous studies in cell culture showing all viruses found outside cells are mature, indicating a difference in virus maturation or diffusion between cells organized in tissues and cells cultured in vitro.

After some searching, an isolated infected cell was located that was responsible for the production of all viruses in the nearby region. The single cell produced 63 viruses, but the microscopic methods only saw viruses in the same plane as this cell. Regions above and below the cell could not be examined, so the real number of viruses that can result from a single infected cell is likely much more than just 63.

Antibodies targeting CD4 showed that uninfected cells have CD4 on their outer membranes, whereas infected cells have CD4 on their inner endoplasmic reticulum. This supports the previous finding that the HIV protein Vpu causes the internalization of CD4 to prevent newly released HIV viruses from reattaching to the host cell.

There was also evidence of the virological synapse, the phenomenon that happens when a virus budding out of an infected cell immediately contacts a neighboring cell and infects it. Two of the proteins that help bring two cells close together, LFA-1 and ICAM, were found at the cell-cell junction near the actively budding virus.

Lastly, the researchers looked for evidence of the ESCRT proteins, which are known to help release viruses from infected cells. The ESCRT components hCHMP1B, hCHMP2A, and hALIX were found on the thin membranous necks of actively budding viruses. Some budding viruses with thick necks appeared to be in an early stage of budding, and displayed spoke-like projections originating from the virus. These were proposed to be the early components of ESCRT.

Overall, the combination of advanced microscopy with the BLT mouse model revealed new aspects of the process of HIV infection, and showed that conclusions drawn from in vitro cell culture cannot always be assumed to be true in whole animals. Further evidence was also gained for the virological synapse and use of ESCRT proteins to facilitate the spread of HIV within whole organisms.

Marathon and Graduate School Diaries: Mid Training


by Elaine To

As I reach peak mileage in my marathon training, I’ll continue the series with some tips for getting through the toughest part of your training plan! One thing I’ve learned so far is that long distance running requires a certain level of physical fitness, but it is primarily a mental challenge. Running is a sport where when you’re challenging yourself with longer, faster runs, your muscles will hurt and you will want to quit halfway. With that in mind, one of my biggest tips is to plan your routes carefully. Try to plan routes that just go in and out of a trail, because you can generally keep your mind pumped until you reach the halfway point.  If after the halfway point, there is an easy way to cut the run short, you could might take it, so eliminate that option!

One of my personal tricks plays on my inborn desire to reach for the A (thanks to my parents). I frequently calculate how much of the run is complete, congratulate myself for getting that far, and then tell myself “but there’s only X% left for that 100% A+!” The mental struggle sounds exactly like what I do for my PhD as well, except the runs are easier, since all I have to do is try. I’m sure everyone reading this knows of the failed experiments that are the bread and butter of any graduate school experience!

Diet is another factor to pay attention to. Firstly, try to always have the same meal before long runs. Each run trains your body to perform that level of physical activity on that given amount of fuel. Maintaining consistency, especially on race day, is important for peak performance. There are many online resources for foods that will help to improve your running ability. Pick one that contains foods you like! Additionally, as runs get longer, you are guaranteed to hit the wall. Avoid this by carrying an energy source on your runs, such as granola bars or energy gels that you eat after a certain amount of mileage. It is important to get your body used to consuming fuel while running, because you are going to need to do this on race day. Overall, the theme here is to use your early training as the period to test out new diet items and strategies. Use the mid training to optimize what worked the best in peak mileage. In late training while tapering your mileage, keep using what worked and finally use this formula on race day. Sounds a lot like assay optimization to me!

The last item I want to address is time management. Graduate school is and should always be the primary time commitment, but you can fit marathon training around it! To get exercise in where you can, take the stairs instead of the elevator. If you have an incubation period longer than 20 minutes, go for a brief run. If the weather’s bad outside, I sometimes train on the treadmill while reading papers. Reading my feeds on Scizzle during my commute or lunch also saves valuable work hours for experiments, allowing runs to start before it gets dark. The most difficult thing to schedule is the longer runs, which are best done on weekend mornings. Weekends (assuming you don’t have experiments running into them) have much more scheduling flexibility, and if the run is in the morning, you won’t lose out on free time.

All tips are drawn from my experience, but each individual’s training will be different. Use what you will, and have a good run!

Do you have any tips?

Backdoor Targeting of the Cancer Causing Protein K-Ras


Elaine To

When targeting a specific protein with a small molecule drug in order to treat a disease, scientists often use a molecule that mimics the natural substrate of the enzyme and targets the active site. However, this approach has met with limited success in the case of the oncogenic GTPase K-Ras. GTPases are regulatory proteins that act like binary switches for cellular pathways. In its “on” state, K-Ras is bound to GTP and activates signaling cascades responsible for cell growth, survival, and differentiation. When GTP gets hydrolyzed to GDP, K-Ras is turned “off.” Mutations that prolong the lifetime of GTP when bound to K-Ras, such as the G12C (glycine at position 12 is changed to cysteine) mutant, are highly oncogenic and lead to cancer. The high affinity of K-Ras for GTP and GDP makes drug targeting of the K-Ras active site difficult, but researchers Ostrem, Peters, et al. have discovered an alternate site on K-Ras that can be targeted for cancer therapies.

The researchers set out to find a small molecule that could specifically bind to the oncogenic G12C mutant protein while avoiding the wild type K-Ras by screening a disulfide library, which would be expected to react with the thiol group of the cysteine. Intact protein mass spectrometry revealed which compounds bound to the G12C mutant without targeting the wild type. The two strongest binders were unaffected by the presence of excess GDP, indicating that they do not compete with GDP for binding. X-ray crystallography showed that one of the strong binders was binding in a previously allosteric pocket of K-Ras.

In order to further characterize the novel allosteric site, the researchers examined libraries containing electrophiles, acrylamides, and vinyl sulphonamides for G12C K-Ras binding. Co-crystals of potent binders with K-Ras revealed that the switch-I and switch-II domains of the protein are disrupted, which also disturbs magnesium ion binding. Previously studied mutations in the residues that coordinate the magnesium ion result in a preference for GDP over GTP, thus the researchers tested the compounds for this activity as well. Indeed, exchange assays reveal a shift in K-Ras’s preference from GTP to GDP when the potent electrophiles are bound. Additionally, the compounds can block nucleotide exchange by exchange factors, though EDTA still effectively catalyzes the exchange of GDP for GTP.

It was also noted that the potent compounds occupied a position normally reserved for G60 when K-Ras is active. Known mutants of G60 have impaired binding to partner effector proteins such as Raf. Studies in cell lines show that compound binding impairs the association of K-Ras with Raf. Lastly, in order to show the effectiveness of the identified compounds as chemotherapeutic drugs, the researchers treat various cancer cell lines, some of which contain the G12C mutation. As expected, the cells with the mutation demonstrated significantly decreased viability in the presence of the compounds.

Overall, this is an elegant approach to small molecule drug development that fortuitously revealed a novel regulatory site of K-Ras. Drugs that target this site can be designed specifically for oncogenic mutations, and do not have to overcome the significant barrier of trying to out compete GDP and GTP for binding. The extensive crystal structure and enzymatic characterizations lay the groundwork for further drug development on K-Ras and may open up a whole new class of chemotherapeutic drugs.

Cutting out HIV: One Step Closer to the Cure



Elaine To

Currently, individuals who test positive for HIV are put on highly active antiretroviral therapy (HAART), a cocktail of multiple drugs that inhibit different aspects of the viral life cycle. While there are drugs that prevent the integration of the viral genome into the host cell genome, there is no known mechanism to remove the viral genome post-integration. This is also the reason we cannot completely eradicate HIV from infected individuals—even after HAART treatment, the viral genome persists in inactive memory T cells. In order to address this, Hauber et al. re-engineered the commonly known Cre recombinase enzyme, directing the novel Tre recombinase to target sequences in the HIV long terminal repeat regions. These regions flank the viral genome, allowing Tre recombinase to cut the targeted sequences within these regions and excise the viral genome from the host cell’s genome.

Lentiviral transduction was used to deliver the Tre recombinase vector into cells. The vector was designed to place Tre under the control of a Tat dependent promoter, ensuring only the infected cells that express the HIV protein Tat will express Tre. Flow cytometry was used to analyze HeLa cells infected with HIV that contained blue fluorescent protein. Cells transduced with the Tre vector had fewer blue fluorescing cells while the blue fluorescing population remained stable in cells transduced with a control vector. Immunoblots confirmed the protein expression of Tre in the Tre transduced cells. Additionally, the time course of Tre expression matched the time course of the decreasing blue fluorescence seen in the flow cytometry experiment. PCR and DNA sequencing checked that the exact DNA sequence intended to be cut out was removed in the Tre transduced cells.

Viral gene delivery comes with a fear of deleterious effects on the host cells. The researchers first examined this possibility in Jurkat cells using a re-designed vector that constitutively expresses Tre. Between the Tre and control transduced cells, there were no differences in tubulin expression, growth rate, apoptosis, or cell cycle progression. When this constitutive Tre vector was transduced into CD4+ T cells isolated from a human donor, the cells displayed similar activation and cytokine secretion profiles as compared to the control vector. The Tat dependent and constitutive Tre vectors were both transduced into hematopoietic stem cells (HSCs) without any change on the abilities of the HSCs to differentiate into the expected cell lineages. Karyotyping and comparative genomic hybridization revealed that CD4+ T cells have no Tre dependent genomic aberrations. Lastly, Tre was shown to be incapable of cutting DNA sequences within the host genome that are similar to the targeted HIV LTR sequences.

The core experiments behind this paper are the in vivo studies done in Rag knockout mice, which can be transplanted with human immune cells and used as a humanized animal model. CD4+ T cells were isolated from human donors, transduced with the Tat dependent vector, and transplanted into the mice, which were then exposed to HIV. The mice displayed lower viral counts and higher frequencies of human T cells versus the control transduction vector. Similar results were obtained when mice were given Tre transduced HSCs. Thus, the researchers elegantly show that their engineered Tre recombinase can alleviate the symptoms of HIV infection. However, reliable methods of gene delivery are yet in development, and the inactive memory T cells harboring the latent HIV reservoir do not express Tat, precluding Tre expression. If combined with methods that activate viral protein expression in the presence of HAART, Tre recombinase therapy may yet play an important role in the cure of HIV.

The Most Scizzling Papers of 2013


The Scizzle Team

Bacteriophage/animal symbiosis at mucosal surfaces

The mucosal surfaces of animals, which are the major entry points for pathogenic bacteria, are also known to contain bacteriophages. In this study, Barr et al. characterized the role of these mucus associated phages. Phages were more commonly found on mucosal surfaces than other environments and adhere to mucin glycoproteins via hypervariable immunoglobulin like domains. Bacteriophage pre-treatment of mucus producing cells provided protection from bacterial induced death, but this was not the case for cells that did not produce mucus. These studies show that there may be a symbiotic relationship between bacteriophages and multicellular organisms which provides bacterial prey for the phages and antimicrobial protection for the animals.


Interlocking gear system discovered in jumping insects

Champion jumping insects need to move their powerful hind legs in synchrony to prevent spinning. Burrows and Sutton studied the mechanism of high speed jumping in Issus coleoptratus juveniles and found the first ever example in nature of an interlocking gear system. The gears are located on the trochantera (leg segments close to the body’s midline) and ensure both hind legs move together when Issus is preparing and jumping. As the insect matures, the gear system is lost, leaving the adults to rely on friction between trochantera for leg synchronization.


HIV-1 capsid hides virus from immune system

Of the two strains of HIV, HIV-1 is the more virulent and can avoid the human immune response, whereas HIV-2 is susceptible. This may be due to the fact that HIV-2 infects dendritic cells, which detect the virus and induce an innate immune response. HIV-1 cannot infect dendritic cells unless it is complexed with the HIV-2 protein Vpx, and even then the immune response isn’t induced until late in the viral life cycle, after integration into the host genome. Lahaye et al. found that only viral cDNA synthesis is required for viral detection by dendritic cells, not genome integration. Mutating the capsid proteins of HIV-1 showed that the capsid prevents sensing of HIV-1 cDNA until after the integration step. This new insight into how HIV-1 escapes immune detection may help HIV vaccine development.


Transcription factor binding in exons affects protein evolution

Many amino acids are specified by multiple codons that are not present in equal frequencies in nature. Organisms display biases towards particular codons, and in this study Stamatoyannopoulos et al. reveal one explanation. They find that transcription factors bind within exonic coding sequences, providing a selective pressure determining which codon is used for that particular amino acid. These codons are called duons for their function as both an amino acid code and a transcription factor binding site.


Chromosome silencing

Down syndrome is caused by the most common chromosomal abnormality in live-born humans: Trisomy 21. The association of the syndrome with an extra (or partial extra) copy of chromosome 21 was established in 1959. In the subsequent fifty years a number of advances have been made using mouse models, but there are still many unanswered questions about exactly why the presence of this extra chromosome should lead to the observed defects. An ideal experimental strategy would be to turn off the extra chromosome in human trisomy 21 cells and compare the “corrected” version of these cells with the original trisomic cells. This is exactly what a team led by Jeanne Lawrence at the University of Massachusetts Medical School has done. Down syndrome is not the only human trisomy disorder: trisomy 13 (Patau syndrome) and trisomy 18 (Edward’s syndrome), for example, produce even more severe effects, with life expectancy usually under one to two years. Inducible chromosome silencing of cells from affected individuals could therefore also provide insights into the molecular and cellular etiology of these diseases.


Grow your own brain

By growing organs in a dish researchers can easily monitor and manipulate the organs' development, gaining valuable insights into how they normally develop and which genes are involved. Now, however, a team of scientists from Vienna and Edinburgh have found a way to grow embryonic “brains” in culture, opening up a whole world of research possibilities. Their technique, published in Nature, has also already provided a new insight into the etiology of microcephaly, a severe brain defect.

[box style="rounded"]Scizzling extra: In general, 2013 was a great year for growing your own kidneyspotentially a limb and liver. What organ will be next? [/box]


Sparking metastatic cell growth

A somewhat controversial paper published in Nature Cell Biology this year showed that the perivescular niche regulates breast tumor cells dormancy. The paper showed how disseminated breast tumor cells (DTC) are kept dormant and how they can be activated and aggressively metastasize. Based on the paper, this is due to the interaction of interaction with the microvascularate, where thrombospondin-1 (TSP-1) induces quiescence in DTC and TGF-beta1 and periosstin induces DTC growth. This work opens the door for potential therapeutic against tumor relapse.


Fear memories inherited epigenetically

Scientists showed that behavioral experiences can shape mice epigenetically in a way that is transmittable to offspring.  Male mice conditioned to fear an odor showed hypomethylation for the respective odor receptor in their sperm; offspring of these mice showed both increased expression of this receptor, and increased sensitivity to the odor that their fathers had been conditioned on.  Does this suggest that memories can be inherited?


Grid cells found in humans

Scientists have long studied rats in a maze, but what about humans?  An exciting paper last August demonstrated that we, like out rodent counterparts, navigate in part using hippocampal grid cells.  Initially identified in the entorhinal cortex of rats back in 2005, grid cells have the interesting activity pattern of firing in a hexagonal grid in the spatial environment and as such are thought to underlie the activity of place cells. Since then grid cells have been found in mice, rats, and monkeys, and fMRI data has suggested grid cells in humans.  This paper used electrophysiological recordings to document grid cell activity in humans.


Sleep facilitates metabolic clearance

Sleep is vital to our health, but researchers have never been entirely sure why.  It turns out part of the function of sleep may be washing waste products from the brain, leaving it clean and refreshed for a new day of use.  Exchange of cerebral spinal fluid (CSF), which is the primary means of washing waste products from the brain, was shown to be significantly higher when animals were asleep compared to waking.  This improved flow was traced back to increased interstitial space during sleep, and resulted in much more efficient clearance of waste products.  Thus, sleep may be crucial to flushing metabolites from the brain, leaving it fresh and ready for another day’s work.

[box style = "rounded"] Robert adds: As a college student my friends and I always had discussions about sleep and it was also this mysterious black box of why we actually need to sleep. Studies could show the effects of lack of sleep such as poor cognition and worse memory but this paper linked it to an actual mechanism by which this happens. First of all I found it very impressive that the researchers trained mice to sleep under the microscope. On top of that showing the shrinkage of the neurons and the flow of cerebrospinal fluid which cleans out metabolites finally linked the cognitive aspects of sleep deprivation to the physical brain. [/box]


Poverty impedes cognitive function

People who are struggling financially often find it difficult to escape poverty, in part due to apparently poor decision making.  Investigators demonstrated that part of this vicious cycle may arise from cognitive impairment as a direct result of financial pressures.  The researchers found that thinking about finances reduced performance on cognitive tasks in participants who were struggling, but not in those who were financially comfortable.  Furthermore, farmers demonstrated poorer cognitive performance before harvest, at a time of relative poverty, compared to after harvest when money was more abundant.


Gut Behavior

2013 has definitely been the year of the gut microbiome! Studies have shown that diet affects the composition of trillions of microorganisms in the human gut, and there is also a great deal of evidence pointing towards the gut microbiome affecting an individual's susceptibility to a number of diseases. Recently published in Cell, Hsiao and colleagues report that gut microbiota also affect behavior, specifically in autism spectrum disorder (ASD). Using a mouse model displaying ASD behavioral features, the researchers saw that probiotic treatment not only altered microbial composition, but also corrected gastrointestinal epithelial barrier defects and reduced leakage of metabolites, as demonstrated by an altered serum metabolomic profile. Additionally, a number of ASD behaviors were improved, including communication, anxiety, and sensorimotor behaviors. The researchers further showed that a particular metabolite abundant in ASD mice but lowered with probiotic treatment is the cause of certain behavioral abnormalities, indicating that gut bacteria-specific effects on the mammalian metabolome influence host behavior.

Your skin - their home

A paper published in Nature examined the diversity of the fungal and bacterial communities that call our skin home. The analysis done in this study revealed that the physiologic attributes and topography of skin differentially shape these two microbial communities. The study opens the door for studying how the pathogenic and commensal fungal and bacterial communities interact with each other and how it affects the maintenance of human health.


Discovery of new male-female interaction can help control malaria

A study published in PLOS Biology provided the first demonstration of an interaction between a male allohormone and a female protein in insects.  The identification of a previously uncharacterized reproductive pathway in A. Gambiae has promise for the development of tools to control malaria-transmitting mosquito populations and interfere with the mating-induced pathway of oogenesis, which may have an effect on the development of Plasmodium malaria parasites.

[box style = "rounded"]Chris adds: "My friend chose this paper to present at journal club one week, because he thought it was well written, interesting etc etc. Unbeknownst to him, one of the paper’s authors was visiting us at the time. We sit down for journal club and one of the PIs comes in, sees the guy and exclaims (with mock exasperation) “you can’t be here!” Me and the presenter look at one another, confused. He presents journal club, and luckily enough, the paper is so well written, that he can’t really criticize it!" [/box]


Using grapefruit to deliver chemotherapy

Published in Nature Communications, this paper describes how nanoparticles can be made from edible grapefruit lipids and used to deliver different types of therapeutic agents, including medicinal compounds, short interfering RNA, DNA expression vectors, and proteins to different types of cells. Grapefruit-derived nanovectors demonstrated the ability to inhibit tumor growth in two tumor animal models. Moreover, the grapefruit nanoparticles used in this study had no detectable toxic effects, could be manipulated or modified to target specific cells/ tissues, and were economical to create. Grapefruits may have a bad reputation for interfering with drugs, but perhaps in the future we will be using grapefruit products to deliver drugs more effectively!



In May, a new technique called CLARITY to effectively make tissue transparent through a new fixation technique was published in Nature. This new process has allowed them to clearly see neuron connection networks not possible before because they can now view the networks in thicker tissue sections. This new advancement will help researchers be able to better map the brain, but this new technology can also be to create 3-D images of other tissues such as cancer. This new ability allows us to gain better insight into the macroscopic networks within a specific tissue type.


Crispier genome-editing

This year, the CRISPR technique was developed as an efficient gene-targeting method. The benefit of this method over the use of TALENS or a zinc-finger knockout is it allows for the rapid generation of mice that have multiple genetic mutations in just one step. The following review gives even more information on this new technique and compares its usefulness to that of TALENS and zinc-finger knockouts. Further, just couple of weeks ago, two back-to-back studies in Cell Stem Cell using the CRISPR-Cas9 system to cure diseases in mice and human stem cells.  In the first study the system was used in mice to correct the Crygc gene that causes cataracts; in the second study the CRSPR-Cas9 system was used to correct the CFTR locus in cultured intestinal stem cells of CF patients. These findings serve as a proof-of-concept that diseases caused by a single mutation can be “fixed” with genome editing using the CRISPR-Cas9 system.

What was your favorite paper this year? Let us know! And of course - use Scizzle to stay on top of your favorite topics and authors.

Marathon + grad school ≠ impossible!


Elaine To

Graduate students are expected to spend a significant amount of time in lab running experiments and reading literature. Marathon training is a large time commitment, and very necessary to ensure a strong finish and avoid injury. It seems impossible that someone who isn’t already an elite runner would attempt to train for a full marathon (26.2 miles or 42.2 km) while working towards a PhD. And yet, that is my current goal!

It all started approximately one year ago when my new year resolution was to push myself into shape. Previously, working out and staying fit had not been priorities in my life. The marathon team at my university had just added a half marathon (13.1 miles or 21.1 km) training plan and I wanted to set a long term goal for myself that I wasn’t sure I would accomplish and then accomplish it (hopefully). During group runs I figured it wouldn’t matter that I ran slower than the rest of them since they’d be running more miles anyway. The first 4 mile training run very nearly discouraged me from the entire thing until I realized it wasn’t about running fast: it’s about running at a comfortable pace that allows you to finish. In that way, distance running became a metaphor for graduate school for me.

Since then I’ve run two half marathons, but after the second one I realized I needed to aim for the next challenge. Hence, I’m now training for my first full marathon! I run slightly faster now than when I first started, but it’s still slower than the rest of the running team, so my training time commitment sometimes feels greater. We’ll have a short series here on Scizzle’s blog following me as I progress over the next couple months and providing tips on how to juggle training, lab, and life! How is this different from the rest of the marathon blogs out there? 1) I am far from an elite runner and 2) I’m a graduate student!

So if your new year resolutions include taking on running then here are 3 tips for starting on the path to accomplish your goal:

[ordered_list style="upper-roman"]

  1. Find a running group to run with. Not only do they help hold you to the training plan when experiments get in the way, but they can also show you a variety of nice running trails in the area. If there isn’t a group associated with your academic institution, local running stores can often make recommendations.
  2. Buy a good pair of running shoes from a runner’s specialty store. Yes, it will be expensive, but the store will analyze your running gait and recommend the best shoe to help you avoid injuries. It’s a worthwhile investment especially since you can also wear the shoes into lab.
  3. Make sure you have clothing appropriate for the weather you’ll be training in. Many races take place in the spring and autumn, which means training occurs in the winter or summer. Depending where you live, this can range from comfortable to downright painful.


As for me? I’m currently running one long run every weekend and supplementing with 1-2 shorter runs during the week when I get out of lab early enough or during long incubations. This past Saturday we did 11 miles and we’ll soon be running distances greater than I’ve ever run before. Stay tuned!

Two for the Price of One: Anti-viral and a Recreational Drug


Elaine To

Efavirenz is a frequently prescribed drug for treating HIV with side effects including depression, paranoia, hallucinations, and delusions. While efavirenz is prescribed orally, recent reports show that people have been crushing efavirenz pills and smoking the powder as a recreational drug. The researchers in this article set out to characterize the psychoactive effects of efavirenz, both on a molecular and a behavioral scale. They study efavirenz both in vitro and in vivo to show that the drug’s properties most closely resemble those of lysergic acid diethylamide (LSD) and that this is due to their shared targeting of the serotonin (5HT) pathway.

Hallucinogenic drugs are known to act through binding to specific receptors within the brain, which activates or deactivates certain neural pathways. Knowing what receptors a drug of interest can bind to is therefore a clue for its activity—and the first question that the researchers answer. Using a ligand displacement assay, they show that efavirenz can bind to 5HT2A and 5HT2C. Efavirenz also increases phospholipase C activity and this increase is abrogated by pretreatment with a 5HT antagonist, showing that efavirenz is a 5HT agonist. Other known agonists of the 5HT system are the natural ligand serotonin, antidepressants, and psychoactive drugs including mescaline and LSD.

The researchers followed up on the receptor pharmacology studies with in vivo studies in mice and rats. This allows a comparison of the whole body effects between efavirenz and other psychoactive drugs. Mice that are given efavirenz are less mobile when exposed to a new environment, similar to LSD. However, LSD has this effect at a dose as low as 3 mg/kg whereas the efavirenz dose needed to be 30 mg/kg before an effect was seen. Efavirenz also induces head twitching in wild type mice, but not in 5HT2A knockouts. Head twitching is a characteristic trait of rodents whose 5HT2A pathway has been affected.

The next set of experiments uses rats that are trained to distinguish between an injection of a drug and a saline negative control. Each rat is given an injection of the drug or saline before being placed in an experimental chamber containing two levers. If drug was injected, depressing one lever would release food pellets while depressing the other did nothing, and vice versa for saline. Rats that were trained on 0.1 mg/kg LSD responded similarly on the LSD lever when injected with 30 mg/kg efavirenz, and rats trained on 18 mg/kg efavirenz responded on the efavirenz lever when given 0.1 mg/kg LSD. There were no such associations with rats trained on ecstasy, cocaine, or carisoprodol.

Since rats known to self-administer cocaine did not do so with efavirenz, it was concluded that efavirenz is not an addictive drug. Additional evidence comes from place conditioning studies, where rats who preferentially spend time on a floor associated with cocaine infusions do not behave similarly with a floor associated with efavirenz infusions.

Overall, these studies show that efavirenz acts through the 5HT2A pathway to exert psychoactive effects similar to those of LSD. It is less potent than LSD, however, requiring larger doses to achieve the same effects on mice and rats. Both drugs are also non-addictive. LSD is not associated with long term adverse effects, so hopefully smoking efavirenz would be a similar case. However, efavirenz is often prescribed for HIV treatment in a single pill that contains additional HIV drugs whose activities on the brain are not as well characterized. Consequences aside, the fact that a drug meant to inhibit reverse transcriptase can also act as a psychoactive agent is fascinating, and demonstrates the complexity of pharmacology.