This is our first day in the lab, we are all excited to get off to a running start with the ideas we had talked about at the end of the spring term! We began with a brainstorming session and discussed a few different box designs that we hope to represent as caDNAno files (using either the honeycomb or square lattice options), such as:
Boxes that are permanently opened
Boxes that are permanently closed
Boxes that can be irreversible opened
Boxes that can be reversibly opened and closed again
Boxes with varying numbers of hinges
After our brainstorming sessions, I designed a six sided box with parallel helices and a hole in one side. My initial inclination is to create a staple-lock-staple mesh over the designed hole to contain cargo that is loaded into the box before folding (aka. before we add the staples that will help my box take on its desired confirmation). This mesh could (irreversibly) be removed using a key strand or a designed restriction enzyme site at designed staple-staple interfaces.
It was a good experience spending the morning acquainting myself with caDNAno. I foresee this skill being helpful as we move on in the summer! However, it was a bit intimidating at first. Below is one of the scariest looking caDNAno screens after hitting auto-staple
Below is a screenshot of the cleaner, and much less intimidating, CanDo output. It looks like my design was a success! Now to design a cover for the hole that can be opened by a separate mechanism...
This morning I continued to work on my caDNAno file from yesterday, this time attempting to ensure that all crossovers occurred after a minimum of 8 nucleotide lengths on any given helix (to grant the structure extra stability). Given that this was nearly impossible, I decided to start again, this time designing a six-sided box with a hinged lid.
Later this morning, we made 2D rectangular DNA origami with materials already available in the Yin lab and attempted to photograph it on the AFM. While our imaging wasn't successful, it was useful to go through the protocol for folding 2D origami with Adam (a useful process to understand as we begin thinking about 3D structure design). Additionally, it was helpful to walk through using the AFM (atomic force microscope) with Adam, as we plan to use this technology to image our structures once we begin folding them.
This afternoon we met with Tom and Wei, postdocs here in the Yin lab at the Wyss Institute. We discussed our initial box ideas and some new ones. Here is what we learned/spoke about:
For a five nanometer gold particle, the interior of any surface we design should have a diameter of at least 20 nm to account for a hydrodynamic shell present about the particle.
The hydrodynamic diameters of particles can be measured using a Dynamic Light Scattering Device (DLS)
The m13 sequence (which we will be using for our origami) is 7,249 nucleotides long; any design we create in caDNAno should have a scaffold of no more than 7,249 nucleotides
The possibility of designing a spherical "box" that could be opened along its equator, an extension of the ideas explored by Han, et. al
The possibility of designing a box with a spring mechanism for opening a hinged lid
This morning I began work on Tom's idea of a spring-loaded box. I used caDNAno to essentially design a five sided box without the lid (I plan to design the lid shortly). However, the .json file that I have for this five-sided box can be used for either my first idea of being covered by a staple-staple or staple-lock-staple mesh or for Tom's idea of a lid with a spring mechanism.
The CanDo Output above indicates the stability of my structure, where red is indicative of the most flexibility (least stability) in my structure. It is to be expected that the regions of my design that are least supported by staples to nearby helices show the most flexibility, as seen above. One concern I have, however, is whether, if the edges of the opening in my box vibrate too much (after a future cover to be designed is removed), my gold particle cargo will be able to escape from the opening.
Later in the afternoon, we devoted a significant portion of time to discussing spherical and rounded designs. We believe that a spherical structure provides the most efficient use of scaffold in maximizing the interior space of our structure. We also talked about the possibility of designing a "tuna-can-like" structure as a hollow cylinder with a frisbee-like origami structure attached to the top and bottom. The additional advantage of these two structures is that they do not have any explicit holes in their surface for cargo to escape. This is unlike the "Denmark box," which is composed of six separate 2D origami structures attached at the seams such that a 3D box is formed. To point us in the right direction towards designing curved 3D origami structures, we are reading a recent publication by Han, et al. in Science about designing 3D Origami with Complex Curvatures, an article that can be found here.
This morning I got to work on the ideas we talked about during yesterday's meetings. First I read Han's article in detail, trying to understand how he came to design these shapes. We also decided to e-mail Han and ask for some guidance.
One preliminary goal I set for the next few days is to replicate at least one of Han's designs in a caDNAno file. I first began working on the frisbee shape, but soon decided to work on designing a hemisphere as it seemed more interesting and useful to me (especially because I figured that once I designed a hemisphere, a sphere wouldn't be too far away). In re-creating the caDNAno file the supplementary info of Han's paper was invaluable. On page S122, he included a high resolution layout of his hemisphere with bases numbered and a colorful outline of the location of staples, staple crossovers, and scaffold crossovers.
Designing the caDNAno file for the hemisphere was a lot more time consuming than had been the time it took to design my box designs earlier this week. However, Professor Yin's gracious contribution of Apple Cinema Displays, certainly made the process of reading through the Supplementary Info much easier!
This afternoon, we went to the Yin Lab group meeting, which was very interesting! This is my first lab experience, and therefore this was my first group meeting. I am really looking forward to presenting our preliminary ideas to the group next Thursday and receiving feedback from group members.
This morning I returned to the my work on the hemisphere. Luckily it was a success and I was able to recreate the figures contained within the Science article I referenced yesterday.
While my caDNAno file was accurate, CanDo didn't necessarily agree (it actually kind of looks like an infinity symbol).
caDNAno File Screenshot & CanDo File Screenshot
Today I spent a lot of time exploring with Maya, a 3D design software program free to college students! It's quite an interesting program, and seems rather complex with its many features. However, after a few tutorials I was able to begin designing what will hopefully begin to look like a sphere.
This afternoon, we attended the Shih group meeting. This, too, was an interesting look into the work of other scientists around us here at the Wyss Institute.
It's hard to believe that a week has already gone by here in the lab! But if this week is an indication of what is to come (as I am sure it is), we are going to have a great summer! I can't believe how much I have learned already through my interactions with others in the lab, my explorations in the software we will be using this summer, and my introduction to some of the imaging equipment here in the lab.
This morning I began work on a program that specifies whether the staple strand at a particular point on the sphere we have replicated is either on the interior or exterior face of the sphere. The input of my program is the coordinates of a specific base in terms of phi and theta on the surface of a sphere in our Maya 3D model of the sphere, and the output is the caDNAno location of that base, along with the direction of the staple strand at that location on the helix. Ultimately, our goal is to give the user the option to input in a variety of formats, namely: (1) caDNA position, (2) Supplementary Info coordinates (see here), (3) theta,phi coordinates, or (4) x,y,z coordinates, and receive an output that specifies staple orientation, as well as coordinates in the other three systems
Today I continued work on my huge spreadsheet of direction values as Shwinn began writing the code for a Hashmap function in Python that mapped each base value to coordinates on Sherrie's caDNAno file and Evan's model of our sphere. Essentially, I am creating a data table separated into 24 "column groups" that correspond to the twenty four helical rings that make up our sphere. Each "column group" specifies the coordinates of a base in terms of theta and phi coordinates, caDNAno coordinates, and supplementary info coordinates. Additionally, using local helicity provided in the supplementary info of Han's paper, I am assigning each base an orientation of either up, down, out or in (where up and down correspond to crossovers with an adjacent helix). To help determine the orientation of a base, I used a "right-handed" water bottle covered in string, and used local helicity and known crossovers. The theta and phi coordinates of each base match a 3D model of the sphere created in Maya by Evan.
My aid in determining base orientation!
Evan's 3D Model of the Sphere
Today I finished the Excel file and it is huge! You should take a look:
Later, I decided that it would be cool to learn some basic code in Python so that I could better understand Shwinn's project (which will hypothetically use Python to reference my data table and generate an output based on user input). Since I have never dealt with code, outside of a few excel functions (if that even counts), Python was a bit overwhelming at first. So instead, I began with C and Java! After an hour or two I learned the basic format and function types for these languages. While not yet a SphereCAD interface, I was able to figure out how to write hello world functions!
This morning I did some error checking on the Excel document that I completed yesterday by picking random bases and making sure the orientations and positions in the different coordinate systems were correct. I wanted to make sure that I had used my water bottle model correctly! I am nearly certain that this document is ready to be put to use in designing our spheres.
In order to begin the process of attaching cargo to the inside of our sphere, I worked on isolating the staple strands located along the seam, breaking connections across the seam, and reorienting the locations of these staples such that at the breakpoints they are facing outwards. My ultimate goal is to manipulate these strands so that they may be cleaved by a restriction enzyme (via a specific palindromic sequence), UV light (via the incorporation of azobenzene or a photo cleavable spacer), or via DTT (via the incorporation of a disulfide bond).
This afternoon we had a group meeting with Professor Yin, Professor Shih, Adam, and several graduate students and postdocs from both labs. We talked about some new ideas for box opening mechanisms and attachment schemes for cargo, including the possibility of using a complete torus shape. As to the discussion of cargo attachment to structures, they emphasized the warning that even a 5nm particle is not 5nm in practice due to the effects of its hydrodynamic radius; Wei (a post-doc in the Yin lab) went as far to suggest that, after accounting for hydrodynamic size, a 5nm particle will be roughly 20nm in diameter.
On a slightly off-topic note… Congrats Bruins! (2011 Stanley Cup Champions)
This morning I spent a lot of time working on the Project section of our wiki. Check it out at HarvarDNAnos Project Page! We are re-designing this page so that it will give visitors a quick, easy-to-understand glimpse into what our project is all about!
Today, I also investigated the activity of the Restriction Enzyme NOT1 and began designing a mechanism to incorporate its recognition sequence into the seam of the sphere, in order to facilitate opening of the sphere. In specific, I designed a construct to test whether or not a restriction enzyme motif would be sufficient to open our sphere. Note in the image below, if the hairpin is open we should be able to detect this on a gel, and this will show that a given amount of pre-cut excess is sufficient to allow a R.E. to cut, but short enough such that the sphere could still open after cleavage.
Testing NOT1 Restriction Enzyme
Our Week 2 Team Photo!
Today I spend the majority of the morning experimenting with Illustrator and Photoshop, as the Yin lab graciously purchased each of us a license for CS5. These programs are going to be of upmost value to us as we begin communicating our designs to others, especially on this wiki!
After getting acquainted, I was able to produce by first Illustrator image, a depiction of Restriction Enzyme Scheme as a means to opening the sphere.
Opening a Sphere Using the R.E. NOT1
As we come closer to loading origami structures with cargo, today we spent time learning how to use some of the analytical devices in the lab that will help us to evaluate our progress.
First we worked with Steve on the Dynamic Light Scattering Device (DLS), which measures the (hydrodynamic) size distribution of particles. This morning we worked to determine the hydrodynamic radius of gold nanoparticles that Adam ordered approximately two months ago. The products were reported to be 5nm in size when Adam ordered them, and we see a corresponding spike the size distribution around 10nm (which is an appropriate hydrodynamic diameter for 5nm gold particles), along with a peak on the several micron scale. This second peak is indicative of aggregates in our sample. Steve has suggested synthesizing our own 5nm gold particles and will help us to do so tomorrow morning.
5nm gold nanoparticles ordered commercially (2 months old); there is aggregation in this sample
Later in the day, we worked with Adam on the TEM and looked at E.coli cells infected with bacteriophages. The TEM will be of most use to us after we have attached cargo to our structures as gold provides an extreme contrast in this electron microscope. However, it is also possible to see DNA under TEM after staining with urinal formate, but even so the contrast is not as significant as it is with gold particles.
This morning I also began working on preparing solutions of several pH ranges for synthesis of our disulfide linked strands. We are searching for the optimal pH for which the disulfide bonds are able to form, and then be subsequently cut by DTT.
Whilst doing other things, we also ran a series of gels containing samples from yesterday of our attempt to make a nanoparticle chain after we had conjugated some of our particles with DNA complementary to an ultramer strand. While we had a little difficulty managing the thin acrylamide gel the first time, we were able to get a much cleaner result our second time around.
Our First Acrylamide Gel of an Attempt to Make a Nanoparticle Chain
This morning I worked on determining the hydrodynamic radius of five of the species of gold nanoparticles we synthesized yesterday with Steve. Below you will find some data outputs from the DLS machine, which overall showed that the more gold particles from a stock solution of 5nm particles introduced to a given volume of AuCl yielded smaller gold aggregates (more particles = more locations for the aqueous gold to bind, such that an individual pre-formed gold particle will be smaller if there are more pre-formed gold particles in solution:
Hydrodynamic Radius of Particles after Addition of 500, 200, 50, 20 and 5 uL of 15nm Au Particle Stock Solution
This afternoon I worked on updating the wiki, in particular this Lab notebook page, and used Illustrator to create a model for the Disulfide Model of opening our sphere.
Disulfide Linkers as a Mechanism to Open the Sphere
Later this afternoon we also had the chance to present at the Yin Group meeting. We received useful feedback from the group members and were pleased that they thought our project goals for the summer were good ones (and somewhat within reach). With respect to the restriction enzyme mechanism I have been working on, I was given the following suggestions:
Look into use of restriction enzymes that cut at sites downstream of their recognition sequence
Consider using cargo other than gold nanoparticles if working with disulfide linkers, as it is possible for gold to cleave sulfide bonds
This morning I prepared a conjugation of two 5' thiol strands for our disulfide linker test that we will be preparing next week in two solutions (one of pH 7 and the other of pH 8). While we are not certain we will be using disulfide linkers, we still wish to find an optimal pH to complete the conjugation and subsequent cutting for potential future experiments with disulfide linkers.
This morning I also met with one of our faculty mentors, Peng Yin, to talk about our future plans for the BioMod team. I enjoyed hearing his input and was glad to have the opportunity to talk to him about
This afternoon I also helped Sherrie and Evan add DNA to gold nanoparticles, via the conjugation protocol.
Our Week 3 Team Photo!
This morning I prepared the disulfide test reaction by mixing the disulfide bonded strands with DTT for an hour. Unfortunately, the image I obtained from the Typhoon was distorted because my DNA ladder spilled over into some of my other wells.
We also attempted to isolate Gold nanoparticles with discrete numbers of attached oligos using an agarose gel, but unfortunately we think the gel was too thick and after five hours very little banding had occurred. We plan to re-do this experiment with Steve on Wednesday.
Preparing my samples!
Today I began designing the interior handles for attaching nanoparticles inside the sphere. I used my excel spreadsheet with staple orientations to determine locations where I could design interior facing handles. We also began folding the sphere today!
Meanwhile, I also began to retest the disulfide reduction reaction by DTT from yesterday. The typhoon was constantly in use though, so I plan to image my gel first thing tomorrow morning!
The first thing I did this morning was image the results of my disulfide reaction on the Typhoon, and it looks promising!
The above image shows Disulfide linkage and subsequent cutting by DTT at pH 7 & 8. The first non-ladder lane (reading left to right) contains disulfide linkage (un-cut) at pH 7, the second lane contains cut disulfide bonds at pH 7, the third lane contains un-cut bonds at pH 8, and the fourth lane contains cut disulfide bonds at pH 8. The large bands in all four lanes contain un-conjugated thiols (or cut disulfide bonds). In lanes 1 and 3 we can see faint bands, which we believe represent conjugated disulfides. Our future goal is to beef up the percentage of conjugated disulfides prior to cutting by DTT by finding an optimal pH or solvent mixture.
This afternoon we also used the AFM to image our spheres after purification through a 2% agarose gel.
Personally, I modeled the strand displacement mechanism for releasing cargo inside the sphere:
This morning we worked with Wei on the TEM and imaged our spheres. Unfortunately, we were unable to get a quality image of a folded sphere because of some problems with our staining technique and an incorrect concentration of buffer used in our folding reaction.
This afternoon after talking with Peng and William for the Biomod team meeting, we decided to drop the idea of using disulfide linkers in our sphere because of the ability of gold nanoparticles to cleave dithiol bonds, which in our case would facilitate the formation of our disulfide bonds.
TEM image of potential sphere, but likely only a water droplet
This morning I made an agarose gel again with the following specifications: 1 mL 1.2 M MgCl2, 8 uL Sybr gold, 120 mL .5x TBE, 2.4g agarose to purify our prepared sphere samples. This gel will, hopefully, help us to isolate our folded spheres, and ultimately get a clearer look at them under TEM/AFM early next week.
Afterwards, I again made AuNP with Steve and used the DLS to determine the hydrodynamic radius of my synthesized particles, it appears that I was able to prevent aggregation!
Afterwards, I began the protocol for phosphene stabilization of my synthesized Gold Nanoparticles so that they will be ready for conjugation with DNA strands on Tuesday!
After discussing our intentions to use disulfide linkers in origami structures that we hoped to have contain gold nanoparticles as cargo, William and Peng brought it to our attention that gold nanoparticles cleave disulfide bonds. Since we wish to continue to use gold nanoparticles as cargo in our structures due to the ease with which we can see it under TEM, we will no longer pursue the use of disulfide linkers in our structures.
The above files are links to the DLS measurements of my samples from various times. The first document, "PrePhosphene_4DaysLater" shows that there was a degree of aggregation of AuNP after sitting at 4 degrees C for four days without the addition of phosphene. The second measurement shows that the addition of phosphene stabilized the particles and prevented aggregation. The third measurement file, "PostSaltAddition" shows minimal aggregation of AuNP, but this is probably a result of using the centrifuge (which is required in the procedure).
We also began re-folding the sphere in the proper buffer conditions, which matched the exact protocol specified by Han in his supplementary information!
This morning we spent a lot of time working on the wiki!
In the afternoon we used both the AFM and TEM to image our recently re-folded sphere, we have some awesome images that show our open spheres on both TEM and AFM:
TEM Image of our unpurified closed spheres
TEM Image of our unpurified open spheres
This afternoon I also developed a program, SphereCADBasic, which helps users determine whether a staple is pointing in, out, up, or down at a specified location on the sphere modeled in Han's paper. The program can use either caDNAno coordinates, coordinates from Han's paper, or Theta/Phi coordinates from our Maya3D image, which is in development.
Download SphereCADBasic and open with Microsoft Excel or a compatible spreadsheet viewer (e.g. OpenOffice, NOTE: GoogleDocs cannot open this file)
Use any of the three available input options: caDNAno coordinates, Supplementary Info Coordinates, or Maya 3D Coordinates
This morning I worked with Evan in preparing TEM samples of our boxes, spheres and gold nanoparticles in preparation for attaching U1C sequence to gold nanoparticles
Later this afternoon we worked with Adam on the AFM so that we could examine purified open spheres, unfortunately we were not able to see much under AFM, it appeared to be a lot of noise/scanning artifacts.
This afternoon we also attended the Yin Group Meeting, we will be presenting our current progress at next week's meeting.
This morning Evan and I prepared our attachment of Au to the U1C DNA Sequence, following the conjugation protocol suggested by Steve
We also attempted to use the TEM to image our samples prepared yesterday but accidentally disengaged the vacuum and were no longer able to use the machine for the day :/ We hope to receive formal training in use of the TEM by early next week!
This morning I spent a lot of time working on getting some kinks out of SphereCADv.2, please check it out and send me feedback. The .zip file contains all necessary files to run the caDNAno file and a .txt file containing instructions on how to run the program.
Additionally, Evan and I began the process of gel purification of our AuNP conjugated to DNA strands. Unfortunately, the gel we obtained appeared to only have a large smear. Ideally, we were hoping to see several distinct bands in our gel that indicated a discrete, successively increasing number of DNA strands attached to our gold nanoparticles.
We attempted gel purification of our AuNP conjugated to DNA strands. Unfortunately, the gel we obtained appeared to only have a large smear. Ideally, we were hoping to see several distinct bands in our gel that indicated a discrete, successively increasing number of DNA strands attached to our gold nanoparticles.
When we first arrived Evan and I used Amicon filters to purify the fluid obtained from the sucrose extraction that Adam and Ralf prepared before we arrived. We used a 1000MW cutoff filter, which unfortunately does not seem to have been effective. Luckily, we still have conjugate AuNP in a separate tube on the slow shaker. In the meantime though, I prepared an additional 200 uL of conjugated AuNP for use on our sphere and box designs.
This morning our second sphere order also arrived! I spend some time re-suspending some of our ordered strands and I updated a new formula sheet with the protocol for diluting strands here.
After a second attempt to separate the conjugated particles via sucrose extraction we found a similar smear to yesterday. However, we decided to cut out the smear band and use Amicon filters to purify the fluid obtained (we also used the same filters to attempt purification of particles not introduced to the gel). We used a 1000MW cutoff filter, which unfortunately was effective. Luckily, we still have conjugate AuNP in a separate tube on the slow shaker. In the meantime though, I prepared an additional 200 uL of conjugated AuNP for use on our sphere and box designs.
NOTE: As of July 18, we have reasoned that the cause of the smear and failure of the Amicon filter was that instead of coating our particles with 1 (or only a few) DNA strands, we had accidentally saturated the surface of our particles with DNA. We will proceed by adding far less DNA in the conjugation protocol, see here
Today I learned how to tuse Microsoft Visual Basic and began writing Command Line User Interface programs for SphereCAD
I now have a working .dll file that just requires users to type in the command "base finder" upon which they will be asked to specify coordinates, which will then be graphed as a "sphere" or ball on the 3D image in AutoCAD.
This morning I found a slight an error in my initial SphereCAD reference formulas in Excel, but was able to work out the kinks and get a working demo ready for the lab meeting this afternoon
This afternoon we had both a BioMod team meeting with Peng and William, as well as a presentation at the Yin Group Meeting. The group meeting went well and everyone agrees that we are moving in the right direction toward achieving our goals.
Tonight i went to the midnight premiere of Harry Potter!!!
Today I went with Sherrie and Evan to Jeff Lichtman's course of optical microscopy at Biolabs in Cambridge. Certainly an interesting class, though not necessarily directly applicable to our work here at Biomod.
Today I was busy trying to figure out what has been going wrong with our AuNP-DNA conjugation!
As a preliminary step to determining what was wrong, Steve and I took DLS measurements of our prepared AuNP-DNA complexes and found that the radius of the particles was roughly 40nm, much largely than we had anticipated
Additionally, Steve and I decided to run a 3% agarose gel of a sample he had formerly prepared and purified along with my AuNP-DNA conjugates. Unfortunately, we did not see any banding in my sample; rather, we found a similar smear to the ones we have seen before
This led Steve and I to include that rather than having one or two oligos attached to the surface of our AuNP's, we had actually saturated the surface of our AuNP with DNA, causing oligos to be extending in all directions from the particles
This means that in our preparation of the conjugates, we were probably adding way too much oligo!
So.... this afternoon I prepared a fresh 100 mL batch of 5nm AuNP per the usual procedure, and afterwards added 20mg of phosphine buffer per this procedure. Before adding phosphine, I took DLS measurements of my sample and found that the hydrodynamic radius of my particles was roughly 11nm (consistent with an actual particle size of 5nm).
Then... I sat down with Adam and read more of this procedure, which was in fact a little more detailed than the procedure Evan and I had followed last time in producing our AuNP-DNA conjugates
It turns out, upon close reading of the above protocol that we had been using an improper extinction coefficient in measuring the concentration of our AuNP after salting and phosphene addition, which led our perceived concentration of AuNP to be 10x larger than it actually was....
While we normally multiply the results of an A520 uv-vis nanodrop by ten, this paper indicates that we do not have to multiply by ten at all. This is likely the reason why we are seeing that our AuNP are oversaturated by oligos....
Tomorrow, steve and I will carefully work out the stoichiometry and I'll make sure to try a synthesis that uses 10 fold less concentrated oligo (..and per adam's suggestion maybe 100 fold less concentrated as well)! Hopefully, this will solve our problem
Also, I plan to add a carbon stabilizing agent that Steve told me about today after i have added the correct concentration of oligo such that our AuNP is compatible with a solution containing Magnesium, and thereby compatible with DNA Origami!
Some additional things I learned today:
When making gels... measure out how much water is lost and account for it (about 30 mls for 140 ml gel for 3min in microwave)
Stop to shake every minute when microwaving
To clean glassware and/or stir bars that came in contact with gold nanoparticles prepare about 50-100mls of an aqua regia solution (which contains 1 part nitric acid to 3 parts hydrochloric acid, both at their maximum concentrations)
Swirl the solution in glassware for about 30 seconds and then wash a minimum of fifteen times with water
The aqua resia solution and the first two water wash solutions should be disposed of in acid containers
Finally, below are my goals for week seven:
Find a way to purify AuNP-DNA conjugates (and make them correctly)
Attach (or attempt to attach) AuNP to open spheres in a two-pot folding mechanism
Attach (or attempt to attach) AuNP to open spheres in a one-pot folding mechanism
A depiction of what our ideal nanoparticle would look like, with only one attached DNA oligo
A depiction of what we think our nanoparticle looks like after adding too much DNA
First thing this morning I worked with Evan as he used the AFM to image his boxes, below is a personal, shortened protocol for using the AFM:
Clean Mica surface with scotch tape
Prepare tip in fluid cell
Put 5uL 5x buffer followed by 5uL sample on the mica surface
Put Mica into AFM
Add 35uL of additional 5x buffer to make meniscus
Load fluid cell and clamp in with knob in back
Use knobs just above and behind fluid cell holder to focus on tip (vertical just above, horizontal just behind
Try to maximize sum to around 5, may also use slider to adjust mirror angle
I spent the greater portion of this morning working with Steve with the intent to develop an optimized procedure for attaching DNA to our synthesized gold nanoparticles for use as cargo in our origami structures, below I detail how to proceed after adding phosphene to your solution of gold nanoparticles and letting the solution sit overnight:
Image your gold nanoparticles using TEM
Glow discharge isn't specifically necessary in preparing your samples, but may be helpful in the case of a dilute sample
No staining is necessary, as AuNP alone provide enough contrast, additionally staining may lead to phosphene appearing in your TEM images, which would cause an over-approximation of the size of your particles
Add 3uL of a dilute sample of your gold nanoparticles (the solution should be a light pinkish color)
You may either leave the sample until the water dries, or wick off remaining water after about five minutes, either procedure is valid but the first will result in more particles left on your grid (which is a good thing if you placed a highly dilute sample on the grid)
Take multiple images of the particles at about 100,000x zoom that give you a view of approximately 500 particles:
It is not essential to use 100,000x zoom, but you do need well focused images and a uniform zoom across all images
Determine the size of your gold nanoparticles using ImageJ
Open your multiple images using stacks in ImageJ
Set the scale by zooming in, selecting the draw line tool, drawing a line over the scale bar, clicking analyze, then set scale, then type into the dialog box the size of your scale bar in the appropriate unit of measurement
Crop your image to remove the labels at the bottom
Cropping one image will crop all
Clear image of weird looking noise by selecting the region and then selecting edit, clear (you will have to do this on each layer)
Set the contrast by selecting image, adjust, threshold, set contrast, then manipulating the slide bars until it appears that the majority of your particles are highlighted red, click apply, and then make sure that neither of the two checkboxes in the dialog box that appears are checked before proceeding to click okay
Set the measurement to analyze the size of your particles by clicking analyze, set measurements, and deselecting all check boxes except fit ellipses and ferric diameter
Analyze the size of your particles by selecting analyze, analyze particles, adjusting to an appropriate size range (as square units) that you are expecting for your particle (for 5nm, use 4-400 … aka a 2-20nm particle range), select exclude edges, display results, and show outlines, click OK
Save the image obtained and export your data table to excel
Determine the average size of your particles by taking the average of the major axes of your ellipses
Determine the sphericity of your particles by dividing the average of your minor axes lengths by the average of your major axes lengths (1 will represent a perfect sphere)
After extraction and examination of the particles with one attached strand, add the specified amount of thiolated polyethylene glycol (PEG) molecule to your solution, as to stabilize your particles with their attached oligo for use in DNA Origami
In general the PEG1000 molecule available in the Shih lab would ideally coat the entire particle, preventing the possibility for aggregation and adding between 4 and 8 nanometers of length to the diameter of the particle (each peg1000 is roughly 2-4 nm long)
It is also possible to use a thiolated poly t in place of a PEG
On TEM, AuNPs with PEGs appear to have halos
Add 1.5 fold - 20 fold excess of AuNP to origami folded/folding structure, possibly with 15-50nM final concentration of NaCl
According to Wei, salt is absolutely necessary to prevent aggregation
A depiction of saturating the surface of our nanoparticle with either PEGs or Poly-T's & 1 DNA Oligo
Later this afternoon I worked with Adam in taking TEM images of our spheres for further size evaluation tomorrow
Image taken by TEM of our gold nanoparticles
Phosphene is a small organic molecule that degrades over time
Adding salt can neutralize the opposing charge of gold particles, allowing particles to get close but possibly aggregate
Thiols knock off phosphene, exposing some of the gold, allowing salt to get really close and cause potential aggregation
Wei does not find the addition of PEGs/PolyT's necessary (but this may be due to the nature of his use of particles) and just believes salt is necessary to prevent aggregation... we will try both
When designing the strand attached to our gold nanoparticles, we should put the poly-T sequence on the thiolated end, because either the PEGs or the attached Poly-T strands will block access to the first 2-4nm (5-10 bases) of our DNA that we hope to attach to a handle interior to our DNA structures
A depiction of the new thiolated strand design
This morning I finished evaluating my spheres per the appropriate protocol using ImageJ and calculated that the average diameter of our gold nanoparticles was 6.61 nanometers
TEM Images of our synthesized particles
Histogram showing the size distribution of our synthesized particles
Afterwards, I decided to check the sample for aggregation before beginning the salting protocol and obtained the following result from DLS. With an average PDI well below .1, there was no sign of aggregation! Therefore, I began the salting protocol and prepared my samples for oligo addition
Later in the day, I prepared a new strand sequence to be ordered that I will be attaching to my gold particle with the poly-T region now flanking the thiolated end. The new sequence is:
Microsoft Excel 2011 does not come with the capability to create histograms, but this website provides an applet that allows you to do so
This morning I helped Sherrie take DLS measurements of her spheres, seeing as we have already imaged them under TEM and AFM. We figured that a third mechanism for comparing the two confirmations might shed light on how to proceed with future experiments. However, as you can see in the links below, the data we obtained seems pseudo-random. To take my measurements I used the 50 uL cuvette and used an SOP that claimed to be valid for DNA Origami (however, this may only be true for 2D Origami...). I plan to re-take these measurements tomorrow morning, and consult with Steve or others on how to proceed if unsuccessful again.
The number measurements seem to have the largest sense of a correlation if any, see below for a side-by-side comparison of all four measurements. We may just be imagining this but if you look to the SD_Closed graph (bottom right) as compared to the SD_Open graph (bottom left) and the Closed graph (top right), it appears that the first peak is closed spheres, while the second (and much larger) peak corresponds to spheres that did not close. However, the top left graph doesn't seem to fit in...
Seeing as I must now wait for my strands to come in before beginning the conjugation protocol for gold nanoparticles and DNA (and ultimately their introduction into our origami containers), I plan to spend the next few days working on re-designing our wiki and helping with other extraneous experiments in the lab!
In particular, take a look at the re-organized protocols page! I have updated the way the gold nanoparticles section is organized. Hopefully this will make it easier for our future experiments using gold nanoparticles both this summer and beyond.
This morning I re-took the DLS measurements of our sphere samples, to a similar result as yesterday. However, I do believe that the shapes of our graphs are approaching the right shape given the physical states/conformations of our samples. I do, however, plan to talk to others in the lab about the appropriate-ness of using DLS on 3D Origami structures, and how to go about doing so.
Again it appears as if the number measurements seem to have the largest sense of a correlation if any, see below for a side-by-side comparison of all four measurements. Here, again, the comparison of the SD_Closed graph (bottom right) with the SD_Open graph (bottom left) and the Closed graph (top right), causes one to believe that the first peak is closed spheres, while the second (and much larger) peak corresponds to spheres that did not close. In this image the relative distribution of the open spheres (top right) even makes sense, but the relative scale does not (or at least not yet...).
As mentioned yesterday, I will be spending the next few days re-organizing our wiki as I wait for my thiolated strands to come in, today I spent a handful of time working on:
Today, one of the thiolated strands came in! Unfortunately though, it was the one with the PolyT on the 3' rather than the 5' end (hopefully my strand with the 5' Poly-T comes in tomorrow). Nonetheless, I began the protocol for conjugating gold nanoparticles to DNA. I began by re-suspending my strands in solution from IDT, and then mixed my oligo and particles in a 1:1 molar ratio solution. Now I must wait another 40 hours until they are ready.
In my continuing work on the wiki, today I cleaned up my Lab notebook page (specifically by adding more detail to some of my earlier entries), the Protocols page, and created the SphereCAD page.
This morning we decided to experiment with the Azobenzene strand we have around in the lab. Azobenzene, when incorporated into complementary strands of double stranded DNA, enables DNA to switch between double and single stranded states dependent upon the introduction of UV or ambient light. UV light prompts the DNA to be single stranded by preventing helix formation, while the re-introduction of ambient light allows to a helix to form again, and double stranded DNA results.
This morning I also re-suspended my new batch of oligos with the 5' poly-T; I then began the conjugation protocol, leaving my oligos on the slow shaker for 40 hours.
Before leaving this afternoon, I helped Sherrie to fold a fresh batch of open spheres that we put in the thermocycler overnight.
Started protocol for sucrose extraction, 20uL sample and 5uL 40% sucrose and let run for 2.5 hours (sucrose is used to increase the density of our samples)
Clear banding occurred; I isolated the particles with only one attached nanoparticles and mixed them in about 40 fold excess with my open sphere samples
Notes for tomorrow... make bigger lanes, 2 parallel gels.. let run for about three hours
Concentrating our sucrose extract from the gel: 14,000g for 15 mins, 400uL water 14,000 g for 10 mins
66nm monoconjugated, about 40 uL
Used old stuff with poly t on wrong side, 40uL particles, 14uL open PC handle spheres, 6uL of 300nmolar NaCl (final concentration of 30nm), left for 3 hours
set up two parallel gel boxes with three combination of 5' polyt and the one version of 3'polyt, only saw banding after 2 hrs in the one with 3'polyt, others appear identical to unconjugated particles
tem image of spheres + nps, not enough stain?
Today I went to Florida! :)
Today I found aggregation in unconjugated particles under DLS. However, after consulting Steve and Wei, decided that after purification this apparent aggregation should not have an effect on our results.
Later today I purified the auNP+sphere complex Evan performed last Friday. i attempted to image these samples, but was largely unsuccessful.
Below is a very faint gel showing the result of Evan's adding gold to spheres for me
This morning I first concentrated conjugated gold
Later, I attempted to make a nanoparticle chain by adding the ultramer strands and handle strands in a 1:9 ratio given that each ultramer binds nine handles, each of which ideally carry one gold nanoparticle. I then left this mixture at room temperature for three hours, and plan to purify in a 3% agarose gel.
After purification, I attempted to image my nanoparticle chain but could not find conclusive proof that I was successful in doing so. I did find what could have been bands of nanoparticles, but this could have just been due to random chance, especially given the relatively few instances I found this to be the case.
Before going home this afternoon, I prepared Open Spheres with strand displacement cargo handles for the thermocycler for use later this week.
Given the aggregation in my former sample of nanoparticles (and my limited supply), I decided to make more nanoparticles this morning
While attempting to purify my spheres this morning that I had prepared yesterday, I noticed that I had forgotten to add scaffold (see below). Therefore I prepared a new batch of strand-displacement cargo handle open spheres for use tomorrow.
This afternoon I also took TEM image of my newly synthesized gold nanoparticles. They seem fairly uniform in size with only a few aggregates present. The approximate diameter of each particle is 6.46 nanometers.
Today's My BIRTHDAY!
Today I purified my spheres from yesterday (this time correctly folded!) via amicon filter and gel purification.
I measured the concentration of conjugates I prepared on Tuesday. The Au-handle complexes with 5' polyT and 3' polyT were both roughly 1.2uM
I verified conjugates formed by running my solution on a 3% agarose gel for about two hours
I added conjugated gold to my purified spheres in the following ratio: 44uL sphere at 5nm, 16 uL 1.2 uM gold, .6 uL of 5 M NacCl (50 mM NaCl total). this is 42X gold
I saw aggregation after two hours visibly as black powder at bottom of pcr tube.
I redisigned my experiment to have less gold (18X) and less salt 15mM, 39uL sphere at 5nm, 6 uL 1.2 uM gold, .6 uL of 5 M NacCl (50 mM NaCl total)
Did this twice, once with 5'polyt, once with 3'polyt
Once again I found more aggregation
I made more spheres
I prepared more conjugation reactions with my newly synthesized particles.
Today I spent a lot of time updating our team wiki.
I also had the chance to TEM open spheres this afternoon, but Adam does not believe that what I believe to look like open spheres were actually open spheres. He is convinced I was only seeing water droplets.
Today I folded more sphere and attempted to combine more nanoparticles and spheres. Again I found aggregation in my samples.
Today I consulted Steve and Wei on the problems I am encountering with aggregation. We have come to the conclusion that my particles are not stable enough to be mixed with my DNA origami. I was able to verify this by mixing my gold particles with just running and folding buffer and found that within an hour aggregation was apparent.
Today I separated my last batch of conjugated nanoparticles and purified them to see if adding purified particles to my spheres would be more effective. Unfortunately, it was not.
Today we made our final presentation at Yin Group Meeting before returning home for the summer!
Today I met with Peng and said goodbye to Sherrie and Evan! We also discussed our plans moving forward to continue experiments and finalization of our presentation during the remainder of the summer and into the fall semester.