Second entry.

In my previous entry, I mentioned that we were analyzing the Apoferritin data to see if it was possible to achieve high resolution with time-resolve cryo-EM. It is very exciting to report that we were able to achieve a resolution of 3 Angstroms, which is the highest resolution yet! Since then, I have been working with time-resolve cryo-EM in effort to capture elongation factor G. When I was screening the grid, the first week, I saw an unusually large amount of contamination of ice on the grid and no droplets. However, after collecting data with Apoferritin, we expected to see a large number of drops and minimal contamination.
 
To pinpoint the source of the problem, we decided to test the polymix buffer, since it was the buffer the sample was dissolved in. However, polymix buffer is not the ideal buffer to use for time-resolved cryo-EM as it is very viscous so it will not spray as well as low salt PBS buffer: the buffer used with Apoferritin. The polymix buffer showed evidence of large amounts of contamination and round drops that did not spread. We hypothesized that the contamination could possibly be due to the salt build-up in the spray-nozzle. Between every grid preparation, we started to clean the nozzle with water. In addition, some of the holes had ice that resembled the texture of snakeskin; the ice should normally appear to be very smooth. We figured that the texture of the ice could be solved by checking the temperature of the cryogen - a mixture of one-part ethane and two-parts propane that the grid is plunged into after the sample is dispensed on it - before each grid is prepared; the temperature should ideally be under -150° Celsius. To make the drops spread more, the girds were plasma cleaned for longer time to make them more hydrophilic. After fixing these issues, the resulting grids looked very clean.
 
After the issues with the polymix buffer were resolved, we prepared grids with the EF-G sample. During data collection, we found many droplets and minimal contamination at the time point of 560 milliseconds. We found that the optimal place to collect data with the most particles with ribosomes was near the grid bar, which is near the edge of the square, since the grid bar would absorb some sample and help the drop spread. Data collection resulted in over 80,000 particles. These particles underwent data processing in which particles were picked, classified, and refined. Finally, we had a 3D map of the ribosome. This map demonstrated a resemblance of an elongation factor, but the resolution was not high enough for us to conclude that it is definitely EF-G. There are two types of elongation factors, EF-G and EF-Tu. The purpose of this experiment is to observe EF-G, however, the map had a resemblance of both elongation factors. We have to continue to prepare grids at different time points and increase the concentration of EF-G in hopes to capture a high-resolution image of EF-G.
 
During the past two weeks, I have been, in parallel, learning about ribosome purification. We are trying to extract ribosomes from bacterial E. coli cultures to study initiation factor 2 using time-resolved cryo-EM. Regardless of the many set-backs such as a broken centrifuge, we were able to complete the entire process once. Although we were not able to see many peaks which meant we did not have polysomes, 70S ribosomes, 50S subunits or 30S subunits, we became accustomed to the process of ribosome purification. We are currently repeating the entire process, in hopes, to finally purify ribosomes that can be analyzed.

This micrograph demonstrates a grid with the EF-G sample. There is evidence of ice contamination, no droplets and broken carbon. This image was taken using the F20 electron microscope.

This micrograph demonstrates a grid with just polymix buffer. There is evidence of minimal ice contamination, two big drops and no broken carbon. This image was taken using the F20 electron microscope.