BME 273: Senior Design

11

Castellon

BME 273: Senior Design

New Technology for Protein Separation

Cathy Castellon

Advisor: Dr. Haselton

April 23, 2002

Abstract

While the world around us is rapidly advancing in the biomedical field, proteomics is a forerunner in the race. There is a compelling reason to pursue the human proteome; doing so will bring the industry that much closer to understanding the molecular basis of disease1. Everyday proteins are separated to compare their expression from an arbitrary reference state of a cell, tissue, or organism, to the profile of a non-standard condition. This process utilizes the SDS-page method, which separates proteins based on size and isoelectric point2. According to innovative workbench (Appendix 1), to find a way to separate proteins not based on size and isoelectric point will decrease the time and effort needed. The new technology outlined separates proteins based on hydrophobic characteristics. Micro-sizing this process using soft lithography allows us to have a greater resolution of proteins, decrease the work space for experimentation, and reduce time and labor. The glass slides used will also decrease the labor time since the process is fairly simple and easy.

Introduction

Proteomics is a rapidly emerging technology in which SDS-Page is the leading method used to separate proteins for the purpose of comparing the expression of protein profiles from an arbitrary reference state of a cell, tissue, or organism, to the profile of a non-standard condition. Proteins are the master molecules of all living things. Enzymes and hormones are composed of proteins3. Structural, membrane, contractile, protective, transport, and storage proteins are also vital and present in all living things3. Proteins are made up of amino acids and each amino acid is composed of an alpha carbon chains with four groups attached: a carboxylic acid group, an amine group, a hydrogen atom, and an “R”group that make it distinct from other amino acids3. These amino acids then combine to form peptides through peptide bonds3. Each protein is composed of one or more peptides3. The current technology, SDS-Page presents an enormous amount of problems, which include: time consumption, resolution problems, and labor intensive, and demands large areas for processing. SDS-Page separates proteins on a polyacrylamide gel based on isoelectric point and size (molecular weight) 2. This process usually takes up to 78 hours. That does not include the time necessary to analyze the protein and separate it from the gel. The new technology developed will provide an automated system that uses micro-fluid techniques to separate proteins based on hydrophobic properties. It will utilize soft lithography to create parameters for the micro-channel. This process drastically reduces the demand for time and space. The automation of this process will reduce the labor intensity of the current technique. With innovation and creativity our design will allow a more efficient and improved technique for protein separation.

Methods

The main goal of the new technology is to adapt a micro-fluid technique that will automate the protein separation process. The main parameters of the design include: developing a hydrophobic/ hydrophilic gradient on a glass slide, separating the proteins using an HPLC column to ensure proper separation, creating a flow channel using soft lithography, and fluroesently labeling proteins for accurate detection. These experimentations were adapted from existing protocols and developed through innovative design.

Experimenting with different concentrations of silanes the hydrophobic/ hydrophilic gradient was successfully produced. The two silanes used were 3-glicidoxypropyltrimethoxy and octyltrichloro4 in a 1:1000 ratio with toluene (experiment 1) and hexane (experiment 2). The glass slides were cleaned with a sodium hydroxide (35g), water (140mL), and 100% ethanol (210mL). This method required extreme caution highlighted by design safe because of the toxic chemicals involved. This process was desirable to ensure the purity of the glass surface. The hydrophobic and hydrophilic solutions were then injected into a micro-fluid pump, which delivered the desired gradient with the following conditions: run program 5, which allows (x)ul/minute of flow for (y) minutes. The variables (x, y) allows for multiple experimentation with different slides. A hydrophobic and hydrophilic control slide was also produced with each experiment. The slides were allowed to sit out over night to ensure the silane/ glass slide reaction. The following day the contact angles of the slides were measured to determine the accuracy of the gradient. The instruments and software programs used to measure these angles were a microscope, the OPTRONIC, a TV, a VCR, and XCAP software. A small drop of 3XSSC was pipetted on to the slide and a picture of the drop was taken. Different concentrations of the silanes were also experimented with to develop the most desirable gradient.

Next the proteins, lysozyme and cytochrome c (Figure 1) 5,6, were chosen according to their contrast in hydrophobic characteristics to be tested on the silane slides. This selection was derived from the Kyte-Dolittle hydrophobicity scale7. The proteins were then separated using an HPLC column to validate the hydrophobic characteristics. PD-10 columns were used in the initial experiment. The device used to automate the HPLC column demands a strict procedure. The power switch must first be turned on, and then the Xe lamp button should only be tapped once (turned on). The computer program used to analyze the information is called FluroSoftware. The parameters to be set on the program include: time scan 4800 seconds, PMT voltage 950 (which is the most sensitive), flows rate change 0.75mL/sec. 100ul of the protein mixture sample were injected into the column and the proteins separated.

The following process allows a micro-fluidic chamber to be used in the experiment. Soft lithography involves five main steps (Figure 2) 8. First a substrate (either a glass slide or silicon) is coated with a photoresist, which is a light sensitive, viscous chemical. The substrate is spun at a high speed while to photoresist is poured onto it’s surface, which allows a thin uniform layer of photoresist to spread across the surface. The photoresist is then allowed to harden for a few minutes . The next step is to apply the mask and expose the photoresist to a light source. Previously a mask of film was created with the desired channel design that included posts to aid mixing and accentuate the separation in the final design. This exposure to light changes the chemical composition of the photoresist and will allow the developer to wash away the undesired components. The next step is to expose the photoresist to the developer. The sample is then baked on a hot plate and is now termed the “master.” The next step involves preparing the PDMS and making a mold from the master. The Sylgard 184 Silicone elastomer base and curing agent protocol and materials (Appendix 2) were used to produce the PDMS. The PDMS was poured onto the master and baked for approximately 20-30 minutes depending on the desired volume. A rubber PDMS sample with the desired channel design imprinted on the surface is ready for the prototype4.

The final procedure in the experiment involves fluroscently labeling the proteins for detection by the flurometer. Two different protocols were used to label the different proteins. Alexa Fluro 350 Protein Labeling Kit was used to label the cytochrome C and Alexa Fluro 430 Protein Labeling Kit was used to label the lysozyme (Appendix 3).

The methods outlined in this section were then combined to formulate the final micro-channel used to separate the proteins based on their hydrophobic characteristics. The PDMS imprinted with the detailed channel design was attached to the glass slide with the hydrophobic/ hydrophilic gradient and clamped with a plexi-glass device. A sample of blue food coloring dye was then inserted into the inlet of the PDMS and collected from the two outlet reservoirs.

Future work with this experiment will include a flow chamber that will monitor the pressure of fluid flow and the flow rate. A spectrophotometer will then be used to test each reservoir and measure the labeled protein signal. Once the data shows the proteins have separated based on their hydrophobic characteristics, the labeling technique will no longer be needed.

Results

The silane slides in Figure 3 illustrates that the gradient placed on the glass slides was distributed evenly and efficiently for the simple micro-channel design. The silanes were initially combined with toluene, however the reactivity of the silanes degraded throughout the semester and we were unable to continue perfecting the gradient for the detailed channel design. As the silanes began to lose their reactivity, hexane was introduced as a solvent for the silane mixture because it has a higher evaporating component and was thought to be a simpler method to lay down the silanes. These results however were not promising because there was not an evenly distributed gradient.

The HPLC column separation was unsuccessful due to fact that the column size purchased was too large for the proteins chosen. Our result for this particular portion of the experiment will rely on literature reviews, excerpts from Shodex’s Hydrophobic Interaction Chromatography. Figure 4 illustrates that cytochrome c (1st peak) is the least hydrophobic and that lysozyme (5th peak) is more hydrophobic.

The soft lithography technique used was extremely successful. The detailed design channel (Figure 5), which originated from Strook’s chaotic mixer design, is the mask used in the soft lithography technique10. The 2X2 cm lanes were placed on the silane silanes and the posts were used to aid mixing and accentuate the separation. When the PDMS mixture produced too many bubbles the results were not satisfactory, therefore a vacuum chamber was used to remove this undesired problem.

The Alexa Fluro protocols used to label the proteins also produced accurate measurements. Figure 6 illustrates that as the flurometer reached the excitation wavelength of cytochrome c, the intensity of the label increased. The emission curve shows that once the protein was excited it emitted at a higher wavelength. The same results can be seen in Figure 7 for the labeling of lysozyme. The figures also illustrate that lysozyme has a higher intensity measurement than cytochrome c. This is due in particular to the fact that cytochrome c has fewer amines in it’s chemical composition and gives off a smaller signal.

The micro-channel assembly produced no results. The channel continuously leaked an accurate flow and flow rate measurements could not be made.

Discussion

Although the micro-channel prototype did not produce sufficient results, the separate steps in the process demonstrate a positive progression towards the separation of proteins based on hydrophobic characteristics using a micro-fluid technique. The initial glass slides produced a promising gradient. Cleaning the slides was a necessary process to ensure the purity of the glass surface. Silanes are composed of a Si and OH group. These molecules attract or repel the protein depending on its components. A hydrophobic silane will repel a hydrophobic protein and likewise a hydrophilic protein will attract a hydrophobic protein.

The HPLC column separation did not work in our experiment however generally these columns are used for hydrophobic interaction chromatography and are packed with polyhydroxmethacylate gel bonded with a phenyl group9. Hydrophobic interaction chromatography is an analytical method in which samples are eluted as the salt concentration of the eluent is gradually reduced9. Therefore, samples are dissolved in the initial eluent or in a solution, which has a high salt concentration than the initial eluent2. We did not follow these guidelines outlined in Shodex’s paper; perhaps this could have been a source of error.

The soft lithography process allowed a micro-fluidic chamber to be used in the experiment. The PDMS molds were fairly simple to produce and the results were extremely desirable. Initially the Sylgard Silicone Elastomer produced undesirable bubbles in the final product. A vacuum chamber was used to remove the bubbles and the PDMS cured in a uniform manner.

The protein-labeling step is necessary in the primary experimentation of this design. Ideally, when the design accurately produced separation based on hydrophobic characteristics the label will not be needed and the labor will be reduced. Removing this step will also ensure the purity of the sample. When the protein is labeled it loses its stable structure and this would be undesirable realistically. This step just provides accurate data for further experimentation.

Further experimentation with this design will hopefully yield protein separation based on hydrophobic characteristics. Future challenges include monitoring the pressure of flow through the micro-channel, monitoring the flow rate, and testing each reservoir for the desired signal. A nano-positioner will be the device used to control the flows and a spectrophotometer will measure the labeled protein signal.

Conclusion

The current technology for protein separation is an inefficient and slow process. This new technology for protein separation based on hydrophobic properties shows an enormous promise for a faster and more efficient solution. The silanes slides used for the separation are simple and easy to produce. This process is also desirable because of the micro-design of the apparatus. Micro-sizing this process using soft lithography will allow it to have a greater resolution of proteins, decrease the work space for experimentation, and reduce time and labor. Automating and reducing the time for this rapidly emerging field will allow us to keep up with the world around us and continue making progress in biomedical technology.

References

1.  http://www.chiresource.com/newsarticles/issue3_1.ASP

2.  Dutt, Michael J. and Kevin H. Lee. “Proteomic Analysis.” Current Opinion in Biotechnology. 2000, 11:176-179.

3.  Greg Stone, Graduate Student, Vanderbilt University 2002. Power Point Presentation

4.  http://www.unitedchem.com/1024x768/Uct2.htm (ordering information)

5.  http://crystal.uah.edu/~carter/protein/xray.htm

6.  http://metallo.scripps.edu/PROMISE/1BBH.html

7.  Kyte, J. and RF Dolittle. “A Simple Method for Displaying the Hydropathic

Character of a Protein.” Journal of Molecular Biology. 157 (1): 105-132 1982.
8.

http://mstflab.vuse.vanderbilt.edu/projects/microfluidics/soft_

lithography_intro.html

8. 

9. http://www.sdk.co.jp/shodex/english/dc010603.htm

10.  Stroock, Abraham D., Stephan K.W. Dertinger, etal. “Chaotic Mixer for

Microchannels.” Science. Vol 295, 647-651.