Leonid Shmuylovich
Professor Joyce Wong
9/1/02
Summary of summer work:
The first project that I worked on was the set up of the LS50 Perkin Elmer Fluorimeter. The goal was to quantify the protein concentration on a gel by fluorescently labeling the protein and measuring the fluorescence in the fluorimeter. This turns out to be quite difficult with the LS50 because it is a cuvette-based fluorimeter. We therefore tried using a plate reader that could measure fluorescence. The plate reader at the Langer Lab did not have the proper filter to measure the fluorescently labeled fibronectin, so we just measured FITC beads that were dissolved in gel. The fluorescent readings of the beads dissolved in gel were consistently lower that of FITC beads dissolved in water (Fig. 1).
It seemed that perhaps the gel was quenching the fluorescent signal. We tested this by measuring the fluorescence of FITC beads dissolved in a gel as it’s gelling. There was no decrease of fluorescence as the gel formed (data is in file “LS50 Fluorescence Quench” in “Leo” folder). We didn’t figure out why the plate reader gave a consistently lower signal for the gels than for beads dissolved in liquid, but because the plate reader was not easy to get to, we decided to pursue other avenues in protein quantification.
Since the LS50 requires a homogeneous sample, we decided to try to break up the gel and dissolve it in liquid. Sonicating a gel that has FITC beads in it breaks it up but not enough to make a homogenous sample. HCl and heat break down the gel but destroy any fluorescent signal that the dissolved beads would have.
Another approach was to re-engineer the cuvette so that it could hold a coverslip. A slit was cut in the cuvette so that a 10mm slide could make a 60 degree angle with the wall of the cuvette when one end was in the slit and the other end was at the corner. (Fig 2)
According the technical support at Perkin Elmer this orientation is ideal for solid samples. This method was not easy, however, because the 10 mm coverslips are fragile and hard to work with. There is very little fluorescent protein on them and the signal is fairly weak. The data can be found in my lab notebook.
Eventually we gave up on using the LS50 Fluorimeter for quantifying protein and moved on to the ELESA method. In this method biotinylated fibronectin (b-fibronectin) is attached to the gel and quantified with an HRP-streptavidin assay. This method was also used to quantify biotinylated RGD peptide. Initial experiments produced data that suggested that the protein concentration went down as you added more protein to the gel. We fixed this problem by mixing b-fibronectin with regular fibronectin. This method produced much better results. It is likely that when we used 100% b-fibronectin the HRP-streptavidin couldn’t bind to the available sited because the HRP-streptavidin would interfere with itself. When we decreased the percent of b-fibronectin to .05%, the steric effect was probably gone. We used the same method for quantifying GRGDSPC-(Biotin).
Fig 3 compares the data obtained from 100% b-fibronectin to 0.05% b-fibronectin.
There were other problems with the ELISA that needed to be addressed. Initially we made 18 mm gels on 25 mm slides leaving a portion of the glass exposed. Since protein binds well to glass, it was possible to get a large signal from protein that wasn’t even bound to the gel. We changed our methods for making gels and started covering the entire coverslip with the gel. Also we tested whether any protein was binding to the underside of the coverslip and giving us a boosted signal. We took a coverslip and before the final detection step glued it upside down. This resulted in no signal. Therefore there was no binding of protein to the underside of the coverslip.
In addition to developing the optimum conditions for the ELISA, I learned how to biotinylate and fluorescenate proteins and peptides. I wrote protocols for all the methods I learned and they are saved in my folder under the “Protocols” folder.
In the last few days I used Biotin-4 Fluorescein to quantify streptavidin concentrations. The data can be found in the Streptavidin Flourescence folder under the file name “8_15 Data”. I attempted to determine the sensitivity of the assay by titrating smaller and smaller amounts of streptavidin with Biotin-4 Fluorescein. In each case the predicted mass of streptavidin was lower than the actual value but not enough trials were done to determine if there is a correction factor that could be used to consistently obtain the correct weight.