Caltech

Finally, A BETTER ENZYME!!

So, remember how I said that I was doing a re-screen of my top 8 mutants from the recent screen?  Well, as it turns out, we did end up getting some pretty good results!

Each column represents a different mutant, with the 2nd and 3rd columns representing the parent enzyme.  Even by eye-balling it, you can probably see that there are some columns that visibly look more blue than the parent enzyme.  When we analyzed the absorbance, we found four enzymes with at least 80% improvement from the parent, and we found one enzyme that had about 150% improvement!!

With such (surprisingly) great numbers, we have to ask ourselves why we didn't get nearly as great results from the first screen.  The most obvious answer is differences in the parent signal.  Obviously, our parent hasn't changed, but the parent signal is comparatively lower in the rescreen plate.  This would have to be further investigated.  Also, its important to note that in a concrete sense, not a lot of sugar is being produced... so small variations can translate into a much bigger difference in percentage improvement. 

Obviously, further analysis would be needed to verify these results.  The plasmids would have to be retrieved and sequenced, and the proteins purified... however SURF IS OVER.  So, the tragedy of the situation is that I won't be able to complete the purification as I'm leaving campus... but I still feel good knowing that my co-mentor is at least one step ahead in the right direction.

My project was full of surprises... but in the end I'm happy to say that we were successful.  (How successful we won't really know until later).  I have come to the conclusion (that many have come to before me) that SURF is really too short.  The last 10 weeks honestly flew by.. and who know?... if I had an addition week or two, I might have even been able to come to some definitive results. Ah well, there is always next year right?

As my SURF Is now officially over, I guess its time for good-byes.  I hope you enjoyed reading about my research as much I enjoyed writing about it.  If you have any questions please let me know.  If you have any questions about research at Caltech, or about life at Caltech in general, remember I'm just an email away: smishra@caltech.edu.  

I hope you all enjoy what's left of summer and have a great school year, and I hope to see many of you at Caltech someday!

The Final Stretch

Hi Everyone,

With only a couple days of SURF left, I am in the stages of wrapping up my project.  Firstly, I told you last time about half-life assays that I've been doing to determine a better temperature at which to screen.  Here is a picture that sort of shows the results.

The top two rows represent an assay I did two days ago, and so the color has started to lift.  The results are easier to see with the bottom two rows.  The lower one of these two rows represents a sugar standard for the sugar assay; in other words, the more "blue" it is, the more sugar that is present in the sample.  In order to determine the half-lives of my parent enzyme at 85 and 90 degrees C, I first incubated purified enzyme at the desired temperatures for different times.  In the third (and first) row the first four have to do with the 85 degrees C temperature, and the second four samples represent the 90 degrees C temperature.  The two dark blue samples of the third row represent the control, when the enzymes had not been incubated at the high temperatures at all.  Much to our surprise, even in 5 minutes, the enzymes seemed to have lost all activity, as can be seen by the lack of "blueness" in the other six wells.  This suggests that these temperatures are too high to screen at, however, we still have to take into consideration that in this scenario, the substrate was added after the incubation, which makes the enzymes more susceptible than in our screens, when the enzymes are incubated at high temperatures with the substrate already added.  We did it the way we did for the half-life assay in order to see whether or not the enzymes specifically were becoming denatured; we didn't want to have the added concerns of intact enzymes not binding to the substrate, which would have been a scenario we would have had to consider if we incubated at the high temperatures with substrate.

My co-mentor, and other members of the lab, will be investigating this problem even after I'm done. However, to wrap up my project, I'm doing a rescreen of the best eight mutants of my last screen. You probably remember my telling you that those eight mutants were probably not statistically significant. Nevertheless, we decided that we might as well check them out as not, and this would provide a nice ending to my project.  So, I'm reinvestigating those eight mutants today, too see if any of them actually prove to be better in a second screen. 

The summer seems to have flown by... I can't believe that tomorrow is my last day in the lab!  I will let you guys know the final verdict tomorrow! :)

Close to the End

Hi Everyone!

Last time I told you about looking at different screening conditions... well I did a few test screens and we found that we were getting better signal at a higher temperature than we had consistently been screening, and that we got better results if we used diluted enzyme.  So, last week, I went back to the beginning, and re-transformed my library and picked colonies to grow up the enzymes.  I picked 32 plates worth, and then screened them on Thursday and Friday of last week.  As it turns out, we only found 8 mutants that were slightly better, but they were again not significant because they were within the standard deviation of the average parent signal.

As this is the last week of SURF, I will probably not be making any more forward progress on finding a mutant.   However, that doesn't mean that I'm going to be spending this week twiddling my thumbs...  In fact today, I'm looking at the half-lives of my parent enzyme at 85 and 90 degrees Celsius.  I'm trying to see whether the enzymes are dying at 85 degrees Celsius, or whether they are intact, and just not binding to the substrate.  Basically, we are still in the phase of trouble-shooting... so maybe we can figure out why we're not getting a better enzyme.

So, it seems like I've been having quite a few problems in the last several weeks... but the truth is, I'm not alone.  SURF is supposed to be a glimpse into real research... and research does not happen without its fair share of speed bumps along the way.  In fact, it might not be too off base to research is a synonym for trouble-shooting.

However, my last week has not been all consumed in research.  The SURFSAC recently took a trip to Santa Monica Beach.  Here is a picture of our group posing at the beach.  (Photo credit: Xida Zheng, Caltech '12)

The trip was a lot of fun; it was actually the last major summer event that SURFSAC organized.  Other events were a pool party, broomballing, and a trip to the Huntington Gardens, among others.

Alright, well, that's all for now. :)

Safety... more than just a formality

Last night, I had to go back to lab after dinner in order to take out some samples from incubation.  However, while I was walking to my building, I noticed a lot of lights, and a fire truck parked on the street.  As I got closer, I saw two more fire trucks, and a bunch of people standing outside my building.  There had been a fire in one of the labs.  After talking to a few of the people standing outside, I found out that the fire had been in MY lab.  I got to see the breadth of the damage this morning:

Apparently, some ethanol caught fire under the hood.  You can probably see from the picture that nobody's going to be using this hood for a while.  Anyway, I thought I'd take this opportunity to highlight how important safety is in the lab.  We were really lucky that no one got hurt last night, because the fire spread really quickly.  You never really realize how soon things can go wrong until something like this happens... speaking for myself, I fell asleep at the safety talk at the beginning of SURF, because I, like everyone else, thought that something like this could never happen to me.  Well, on to less depressing things... like the results of my screen.  I told you last time how I was screening from the lower mutation rate library.  I went ahead and screened 3000 mutants from that library, and found 18 mutants that had improvement in the 15-30% range.  In order to verify these results, I created a re-screen plate, which had 3 copies of each of the 18 mutants, and I screened this plate again.  However, this time, the results were sort of puzzling.  There was improvement in comparison to the parent, but the differences in improvement between different mutants was about the same as the differences in improvement between clones of the same mutant on the plate.  So, we came to the conclusion that any improvement was not really significant outside the realm of statistical noise.  So, we are going back and trying to work with our screening conditions once more.  Tomorrow, I will do test plates and see if we can get better results by incubating the enzymes with substrate at 85 or 90 degrees Celsius.  As always, I will let you guys know how that goes.

In addition, as promised, I wanted to talk a little bit about the work that I've been doing regarding specific activity.  This isn't directly related to my project, but it still has the same spirit.   Since my screening has been basically been a sugar assay, I have been screening for enzymes with better "total activity."  This encompasses both thermostability, as well as specific activity.  Specific activity is what you would think of when you think of activity; it basically measures how well well the enzyme hydrolyzes cellulose.  My co-mentor has been looking at the specific activities of all of her parents (from each of the 4 generations) and seeing if there has been improvement in specific activity, or if its just been in thermostability.  The measurements that I've been taking have been suggesting that there has not been much improvement in specific activity; however, we have basically created a much more stable enzyme.  So, food for thought (for my mentor and me anyway) is to find a way to screen in such a way as to isolate those enzymes with better specific activity. 

Anyway, so I go back to picking and growing up yeast colonies again tomorrow, so that I have more mutants to work with, once I decide on some different screening conditions.  I will give you an update as soon as I have something to tell. :)

My plates got eaten... by fungi :P

So technically, saying that my plates got "eaten" by fungus isn't entirely correct, but take a look for yourself.

If you look closely, you'll see little white circular dots.  Those represent my yeast colonies.  The fuzzy looking things are fungi.  As I've mentioned before, we don't like fungus because they outgrow yeast, and we need the yeast to grow properly to release the mutant enzymes.  The interesting thing about the above plate is that when I was picking the colonies, there was no fungus, where you see the fungus now; this kind of poses a problem for me, because it poses the possibility for unwanted contamination... oh well... I'll guess I'll have to just wait and see.

So, you might be wondering why I've gone back to picking and growing up colonies, when I told you about possibly having three better enzymes in my last post.  Well, when I looked at the sequencing information for the three enzymes, I found out that two of the enzymes only had one mutation, and the other one had three.  The two individual mutations are actually mutations that we've seen before (in previous generations) so they're no help.  After looking more closely at the location of the 3 mutations of last mutant, we found out that the mutations were "silent" and had no affect on the activity.

Basically, if I cut to the chase, after picking and screening 4500 mutants total, and purifying and sequencing the best ones, my results returned NOTHING!!

My initial reaction was just AHHHH... the summer is over half-way done and I still don't have a single mutant that is significantly better than the parent.  As such, in the last couple of days, my co-mentor and I have gone back to the drawing board.  After talking to other members of the lab, we were turned toward the direction of possibly working with our screening protocol.

The basis of the screening protocol, as I've talked about before, is just that we take excreted enzymes and test them against substrate by doing a sugar assay.  In our screening protocol, we incubate the enzymes with the substrate at 77 degrees Celsius for 2 hours for the enzymes to hydrolyze the cellulose.  The problem with our parent enzyme is that it is already incredibly stable, with a half-life of about 3 hours at 77 degrees Celsius.  This means that most of our enzymes (since they are only a few mutations away from the parent) are alive and working for the entire time.  The problem that this poses is that there is a lot of sugar produced.  If you look a few posts ago, you'll remember the pictures of my plates after the sugar assay.  (These are the plates with yellow and blue wells.)  Put simply, the more sugar there is, the more "blue" the well is.  Anyone would agree that (even for a spectrophotometer, which takes my absorbance readings), it is easier to detect differences between say yellow and light blue than say a light blue and a dark blue.  Therefore, in order to get a more accurate reading of how the mutants do in relation to the parent enzyme, we decided to try and reduce the signal that would show up in the sugar assay.  We have been trying a few different ways of doing this.  For, instance we've been trying to dilute the concentration of enzyme and we have been increasing the incubation temperature to 85 degrees Celsius instead of 77, to see if we can decrease the signal that way.

Along with that, since my last screen was unsuccessful, I had to start picking again to restart (hence the picture of the plate up top).  This time, however, I decided to pick from the 100 uM Manganese chloride library.  (Remember my error prone PCR libraries and the 4 different concentrations of manganese chloride that I used to alter the mutation rate?  Well, I'm now working with the one that had the second lowest mutation rate, to see if I have better luck.  This was the other concentration that I was debating between way earlier, if you remember.)  

I'm now working on screening these, and I will let you know if I make some more progress when I finish.  I will also tell you about a new mini-project that I have started regarding specific activity.  Until then, take care!

Update on my work

First and foremost, sorry for my boring title; I couldn't really think of anything more interesting.  Also, sorry that I haven't posted in like, oh I don't know, 8 DAYS!  I wish I had a good excuse, but I don't really, except for the fact that I've been depressed that things haven't really been working too well for me recently.

So, I told you last time that we ended up choosing the best three mutants of basically our entire screen.  (I actually said four in the last post; i have no idea why, because I actually meant three, sorry!)  So the next step for these three mutants was for me to grow up larger yeast cultures, one that contained the gene for each of these mutants.  I then extracted the plasmids (containing the genes) from these yeast.  One of the first steps in looking at these mutants is to make sure that initially, in our 96-well plates, there was not some contamination between neighboring wells.  In other words, we need to make sure that what we think is in the three wells that we got our best mutants from is actually in the wells, nothing more and nothing less.  To make sure of this we have to get the three genes sequenced.  However, there is not enough DNA when we take the plasmid out of the yeast directly for sequencing, so I had to extract the plasmids out of the yeast, electroporate them into E. coli (so that the bacteria can produce many copies of the plasmid) and then again extract the plasmids out of the E. coli and send them in for sequencing.

But, in doing this, I ran into quite a few speed bumps.  For instance, in my initial plasmid extraction from yeast, there were remnants of ethanol from the protocol in my final extracted DNA, so when I did the transformation and then plated the E. coli colonies, it didn't work.  I then tried again after I concentrated the DNA.  I ended up doing several transformations.  In the end, I did succeed in getting three colonies for each mutant. I took these 9 E. coli colonies, grew them up (so the plasmids could replicate) and then I extracted the plasmids from these colonies and sent them in for sequencing.  The reason I took three colonies for each mutant was so that I could compare the sequences, in order to see whether there is a consensus sequence.  If there is, then it is evident that there wasn't a contamination in the original well, but if they are different sequences, then we have to see that which of the mixture of mutants in the well was the best mutant.

So, I will come back tomorrow with the sequencing information and let you know what happens. :)

A better enzyme... possibly?

I finally have some good news!  Even though my initial screen of the 3000 mutants did not yield any improvement, my secondary screen with 1600 mutants ended up going much better.  I got a mutant that gave me 37% better performance than my parent (original) enzyme, and a couple of other that were in the 20s in percentage improvement.

However, in the whole screen process, there is a lot of room for error.  So, over the weekend, I set a rescreen plate, where I took the best 11 mutants of the 1600, and re-grew up the yeast cultures, and rescreened the mutant enzymes today.  After rescreening these enzymes, I found out that the one that I initially thought was 37% actually showed a 12% improvement in the rescreen.  All the enzymes in the rescreen had a much lower improvement in performance.

This sort of struck me as weird; as with everything, sometimes, numbers can be deceiving.  Upon further inspection, I found that the parent signal in the rescreen plate was on average much higher than the parent signal that was used in the original screening plates to normalize the improvement percentages.  The increased parent signal was the reason that that the improvements seemed much lower.  The weird thing about this whole situation is that obviously, my parent hasn't changed.  Anyway, after discussing the situation with my co-mentor, we came to the conclusion that the parent signal in the rescreen plates is probably the one to trust, because the ones we used in the screen plate had been the result of a transformation that had given my co-mentor issues too.

Anyway, we still have a long way to go before we can decide on a better enzyme.  In an attempt to see whether these enzymes are actually showing improvement, we have to make sure that each well that we have indeed only has one type of enzyme, so that we're sure that the activity we're measuring is the product of one enzyme and not multiple.  As such, we chose the best 4 enzymes of the 11 today and I will be extracting the plasmid out of the original yeast cultures that yielded these enzymes.  I will then transform the plasmids into the E. coli because E. coli cells only take up one plasmid, so we can select against the presence of multiple enzymes.  After reproducing the plasmid, I will send them in for sequencing, so that i can analyze the actual DNA sequences.  I will let you know my next steps as I get there.

A few words about sequencing-- here at Caltech, it's fairly convenient, because we have a central facility on campus where we can submit our samples for sequencing, and we get emailed the results within two days.  So far, I've only had to submit 5 samples in for sequencing, which I did late last week.  Throughout my whole procedure so far, I have done nothing to really determine how many mutations were actually made on the parent enzyme.  So, in order to get a quantitative idea of the mutation rate, we chose 5 random mutants and submitted them in for sequencing, along with our parent, in order to determine on average how many mutations away from the parent each mutant is.  The program Sequencher is actually pretty neat, and lets me compare my samples right to the parent.  Here is a screen shot of Sequencher, when I was analyzing the samples. 

The chromatographs that are shown in the lower window represent the different signals of each nucleotide, which is how the sequence was actually determined.  The upper window shows all the sequences, and the little dots on the bottom represent where there are mutations.  After looking at the 5 mutants, I found the most similar mutant to have 5 nucleotides differing from those of the parent, while the most dissimilar mutant to have 15 nucleotides different from those of the parent.  While these results are from five random mutants, they don't give us a whole lot of information, but they give us a general idea of how many mutations is enough to create a different enzyme that is also functional.

In the coming days, I'll be starting a different chapter of my project, as I narrow my search for a better enzyme.  I am most likely done (or at least almost done) with my wide-scale screens, and I will be working with the actual purified enzymes.  Anyway, I will keep you posted as I move forward (and more importantly figure out what I'm doing... :P).  Cyu then!

Sometimes... things just don't work out

You might have guessed from the title of my post that I don't have very good news.  I told you last time how I've picked 3000 colonies.  After the colonies got a chance to grow in media, the released enzymes in the supernatant were suspended in solution with our substrate.  You might remember (or even if you don't, I'll recap) that when I was looking for the best mutation rate, I had 400 mutants, 100 each representing a different mutant.  I had done a sugar assay to determine which mutation rate that I wanted.  I kind of skimmed over that process, but I'll go into more detail now.

After I picked the 3000 colonies, I put them into the shaker to incubate in media for 2 days.  After this period, I took out the plates and spun them down in the centrifuge so that the cells pelleted, and all that was left in the supernatant was the released enzymes.  The enzymes (supernatant) from my original plates were then transferred to new plates that had substrate in them.  These plates were incubated for an hour and a half at an impermissible temperature so that the enzymes could bind to the substrate.  Afterwards, the plates were washed with a buffer, so that all the other sugars were removed. The enzymes with the bound substrate were then incubated at a high temperature so that the enzymes could break down the cellulose.  A sugar assay was then done to determine how much sugar was actually produced after the break-down.  This whole process (done in one day) takes 10-11 hours with 16 plates.  So, I did this twice, splitting up my 32 plates into two sets of 16.

Before I forget though, I have a story from my first day of screening.  When I was spinning down a few of my plates (I had to do 4 at a time due to the nature of our centrifuge), the edges of one of my plates broke, probably because I was a bit sloppy in aligning the plates.  Anyway, even though that was kind of a problem, it wasn't too big a deal, because none of my wells had been affected.  So, when I went to transfer my supernatant to my plates with the substrate, I went to go use the pipetting robot.  As cool as the pipetting robot is, it is still a machine... that messes up.  I mentioned before that the pipetting robot can pipette 96 wells all at once.  If you look closely at the picture I posted before, you'll see that there are specific positions for plates, so the machine knows where to pipette from and what to pipette to.  The problem that arose with the edges of my plate being broken was that it didn't align quite right for the robot, and so all of the pipettes slightly missed their target wells, which resulted in the pipettes getting stuck and causing a huge mess.  I'm not going to pretend like I didn't freak out, because I did.  Anyway, the short story is that I had to throw that plate out... so I finished the protocol with only 15 plates that day. :(  Anyway, here is a picture of the products of the sugar assay from that day:

So, like I said before, the wells that are darker represent enzymes that produced a lot of sugar, while the lighter wells show the opposite.  Of course, we use a spectrophotometer to compare the absorbance values to really compare.  I forgot to take a picture after my second day of screening, but things went smoothly for the most part that day (basically... I started AND ended with 16 plates :D). 

Now, we get to the bad news.  Basically, the parent that I used for the beginning of my round of directed evolution was not actually a natural cellulase; rather, my parent was the the best enzyme that my co-mentor had done in her last round (of which she has already done two).  Therefore, basically, my library represents the third generation of mutants.  After looking at the absorbance readings of all roughly 3000 mutants and comparing them to my parent, we got a few mutants that had activity 15% better than that of the parent, but most of the mutants were not much better at all.  In previous generations, my co-mentor has gotten 50 or 100% better mutants.  So, we were thoroughly disappointed.  However, if you remember the yeast plate that I showed you in the last post, you'll see that there are still PLENTY of mutants left.  So, tomorrow, I go back a few days and will be picking roughly 1500 mutants (16 plates) and will screen these again to see if we have any better luck.

I will give you an update next time, but until then, comment away!

Picking colonies... two days of fun!

So... after a couple days of picking colonies, I've come to realize the 3000 isn't as small of a number as it seems.  When you look at a square plate (like the one shown below) for long enough, you start to lose your mind ever so slightly.

So, before I get too far, I should probably backtrack a bit and get you guys up to speed.  I've already talked about how we decided which mutant library to use; we used the one that had a slightly higher mutation rate than the average, but one that wasn't so high that we killed off a bunch of enzymes.  However, you'll probably remember that we only had modeled each library with 100 mutants each.  In order to really get a better enzyme though, it was important to represent the library with more mutants.

In order to go about picking 3000 mutants, we retransformed our library into yeast with the linearized vector and grew up the resulting yeast colonies to get the result that is above.  We had initially plated 4 big square plates filled with mutants.  However, much to our disappointment, two of the plates were contaminated with a fungus.  Fungi are similar enough to yeast in the sense that we can't really select against them like we can bacteria, and fungi grows much faster than yeast, causing it to outgrow the yeast.  Needless to say, we couldn't use those plates.  However, two of the plates (one of which is shown above) were still good enough to use.  In order to pick colonies, I used toothpicks to gently swipe individual colonies and inoculate them in 96-well plates, like the one shown here.

After picking up a colony with a toothpick, I put it in a well of a plate.  After finishing a plate, and shaking it a bit to make sure the colonies were incorporated in the media in the wells, I took out the toothpicks and sealed the plates.  The finished products looked something like this.

You can't see all the plates, but I ended up doing 32 plates, 16 each of two days.  It took a looong time, but it was very satisfying at the same time.  At the end, I felt like I'd really accomplished quite a bit.  I put the plates in a shaker for 48 hours.  My next step will be to do a screen with all of these mutants so that I can scale down the 3000 to roughly the top 50.  I will tell you about the screen in the next post. :)

I have been really negligent in talking about this, but better late than never I suppose.  This summer, apart from my SURF project of course, I have also been involved with SURF SAC, which stands for the SURF Student Advisory Council.  In short, SURF SAC is in charge of making sure that SURF students have lives outside of the lab when they're on campus for the summer.  Yesterday, for instance, SURF SAC had a 4th of July Barbeque (on the 2nd of July... but whatever).  I wish I had taken pictures... but I was in lab. :D  SURF SAC has different activities going on every night; for instance, Tuesdays are sports nights, and Thursdays are movie nights.  Here's a picture from last week's movie night.  (We watched Slumdog Millionaire.)

Well, that's all I have for now.  I will show you the products of my screen in my next post.  Until then, enjoy the weekend.

Progress... even in the face of mishaps

Yesterday was not a good day in any sense.  In the morning, while I was in lab, I needed to keep some samples cold.  Sounds simple enough right?  I looked around the lab to find an ice bucket so I could go fill it with some ice.  The bench that I share with my co-mentor (shown below) didn't have one so I looked around.

I found an ice bucket on a nearby bench.  I asked the owner of the bench if it would be okay for me to use it.  Getting the okay, I began to pick up the ice bucket.  At first glance, it seemed like it was empty.  However, upon closer inspection, I realized that it had some melting ice.  Right then, my clumsiness decided to kick in, and the bucket slipped out of my hands.  The papers and equipment on that bench got drenched, and I made a huge mess.  On top of everything, I almost ruined someone's computer.

The day definitely did NOT get better.  In fact, when I got back to my room in the evening, and was making myself dinner, I burned three of my fingers while making two-minute noodles and consequently spent the subsequent 3 hours with my fingers on ice.  I know what you're thinking... but in my defense, the world was not created for the uncoordinated.  Everything put aside, though, I have been making progress with my project in the lab.  I talked last time about determining what concentration to use for error-prone PCR, in order to create my mutant enzymes.  To go about doing this, I needed to have actual enzymes to test.  However, when doing error-prone PCR, in essence, I just got a bunch of mutated DNA sequences which serve as the genes that encode for my enzymes, not the enzymes themselves. In order to get the enzymes, I had to utilize yeast.  

I first grew up competent yeast cells.  They are considered "competent" because they can readily take up plasmids.  Plasmids are circular DNA, and consist of a backbone and a DNA insert.  Our DNA insert is just the mutant genes for our enzymes.  Our particular backbone contains genes for the yeast to process uracil.  We insert the plasmids into the yeast cells by means of electroporation.  This is a cool concept; we basically shock the cells so that their membranes are perturbed and they take up the plasmids that are in the solution.  However, like everything in science, this is not 100% efficient. In order to make sure that we only grew the yeast cells that actually had the plasmid, we plated the cells in plates that didn't have uracil, which is necessary for yeast survival.  Because the backbone of our plasmids gave the yeast cells that did successfully electroporate the ability to make uracil for themselves, only these survived.

After growing these colonies, we got plates that had many many colonies, each with potentially a different mutant gene for the cellulase enzyme (because our replication process for the gene was so random).  Of course we had 4 different sets of plates, one for each of our initial concentrations of manganese chloride.  We then hand picked each colony with a toothpick and put them in 96-well plates.  We did one 96-well plate for each concentration.  (We basically hand-picked almost 400 yeast colonies from petri dishes using tooth picks!!  Sounds fun, huh?)  Afterwards, we let our colonies grow out in media, so that the yeasts could secrete the enzymes.

After this period, we spun the cells down so that the cells and other unwanted stuff was pelleted, and we took the supernatant, that contained the enzymes that we wanted.  We added this to new 96-well plates that had our substrate.  If you've every pipetted, you know that it can take a while, especially when you have something like 400 total samples.  BUT, it doesn't take so long when you have the pipetting robot!  (shown below)

The pipetting robot is just as cool as it sounds.  It can pipette an entire 96-well plate in one shot, therefore making my job much easier :D!  I got to use the the robot several times, as after we transferred the enzyme supernatant to the substrate and let them incubate, we had to wash the bound enzymes and substrate to get rid of excess sugars. After several washes, we needed to let the bound enzymes to actually break down the substrate, releasing individual sugars.  We then used a sugar reagent assay to determine how much sugar was produced.  The products of the four concentrations are shown below.