Getting Started

Introduction

So you've read a journal article or heard a talk about a technique that allows one to analyze more than 500 proteins or peptides in a single experiment.  After reading or hearing about this technique you might wonder:

• Is this technique too good to be true or is this the future of systems biology?
• Will this technique make no mark on the field or will it revolutionize the study of proteins and their functions?
• How can I use this technique to study my research problem?

Although I have my opinions about the first two questions, this website is not the place for offering them.  As a substitute, however, this website offers a thorough explanation for the final question.  So without further delay, let's get started.

Picking the Peptides/Proteins

The first thing that one has to do is decide what peptides and/or proteins are important in the system under study.  In my case, for example, I am interested in determining the protein autoantigens that are the targets in Multiple Sclerosis.  Multiple Sclerosis is an autoimmune disease where autoreactive immune cells erode the myelin sheath.  Thus, the peptides I am interested in come from proteins found in myelin.  These proteins include Myelin Oligodendrocyte Glycoprotein (MOG), Myelin Basic Protein (MBP), Protelipid Protein (PLP), Oligodendrocyte-Specific Protein (OSP), Myelin-Associated Glycoprotein (MAG), and others.

You may ask, did I simply pick find a neurobiology book and start randomly picking proteins in the myelin sheath?  The answer is, of course not.  These proteins represent a collection of common autoantigens that have been reported in the scientific literature.  Therefore, I relied upon what other people have shown to be important.  This reliance on others data serves two purposes.  First, it provides a starting place to build up the list of potential "target" proteins.  Second, the antigens serve as controls for validating the technique because it confirms the findings of other groups.

In addition to the putative autoantigens, I also add my best-guess-peptides to the list.  These best-guesses are based on thinking about potential mechanisms for autoimmunity and then coming up with reasonable proteins to test these hypothetical mechanisms.

Finally, I add a host of control peptides or proteins to the list.  These include vaccines, immunoglobulin, fluorescently labeled peptides, and others specific to multiple sclerosis.

Remember that the power of microarrays is their ability to survey many peptides or proteins simultaneously.  The key is not to get every peptide that will be recognized.  One simply wants to get enough potential targets so a pattern or profile starts emerging.  The pattern is key because that shows where to focus one's effort.  Now you know the secret to picking a good list of relevant peptides or proteins for your own microarray experiment.

Obtaining the Peptides/Proteins

The second thing that one has to do is obtain the peptides or proteins from the list generated in previous step.  There are a couple alternatives for going about this:

Isolate from the tissue - Yikes! The big benefit of this method is one gets native proteins that are most likely already known to be autoantigens.  The big downside, however, is the method is time intensive, thus making it difficult to build a large list.

Recombination - This method is nice because one can by the peptides or proteins pre-made.  One downside is that it becomes expensive when one wants hundreds of proteins.  The other downside is that whole proteins a mute about which epitope is important. 

Synthesis - This method allows for creating custom peptides (a huge benefit).  The big downside is that it can be expensive.  Fortunately, Sigma-Genosys offers a discount on synthesis of many peptides per order (minimum of 48).

Reconstituting the Peptides/Proteins

The first thing that one has to do is decide what peptides and/or proteins are important in the system under study.  In my case, for example, I am interested in determining the protein autoantigens that are the targets in Multiple Sclerosis.  Multiple Sclerosis is an autoimmune disease where autoreactive immune cells erode the myelin sheath.  Thus, the peptides I am interested in come from proteins found in myelin.  These proteins include Myelin Oligodendrocyte Glycoprotein (MOG), Myelin Basic Protein (MBP), Protelipid Protein (PLP), Oligodendrocyte-Specific Protein (OSP), Myelin-Associated Glycoprotein (MAG), and others.

You may ask, did I simply pick find a neurobiology book and start randomly picking proteins in the myelin sheath?  The answer is, of course not.  These proteins represent a collection of common autoantigens that have been reported in the scientific literature.  Therefore, I relied upon what other people have shown to be important.  This reliance on others data serves two purposes.  First, it provides a starting place to build up the list of potential "target" proteins.  Second, the antigens serve as controls for validating the technique because it confirms the findings of other groups.

In addition to the putative autoantigens, I also add my best-guess-peptides to the list.  These best-guesses are based on thinking about potential mechanisms for autoimmunity and then coming up with reasonable proteins to test these hypothetical mechanisms.

Finally, I add a host of control peptides or proteins to the list.  These include vaccines, immunoglobulin, fluorescently labeled peptides, and others specific to multiple sclerosis.

Remember that the power of microarrays is their ability to survey many peptides or proteins simultaneously.  The key is not to get every peptide that will be recognized.  One simply wants to get enough potential targets so a pattern or profile starts emerging.  The pattern is key because that shows where to focus one's effort.  Now you know the secret to picking a good list of relevant peptides or proteins for your own microarray experiment.

Obtaining the Peptides/Proteins

The second thing that one has to do is obtain the peptides or proteins from the list generated in previous step.  There are a couple alternatives for going about this:

Isolate from the tissue - Yikes! The big benefit of this method is one gets native proteins that are most likely already known to be autoantigens.  The big downside, however, is the method is time intensive, thus making it difficult to build a large list.

Recombination - This method is nice because one can by the peptides or proteins pre-made.  One downside is that it becomes expensive when one wants hundreds of proteins.  The other downside is that whole proteins a mute about which epitope is important. 

Synthesis - This method allows for creating custom peptides (a huge benefit).  The big downside is that it can be expensive.  Fortunately, Sigma-Genosys offers a discount on synthesis of many peptides per order (minimum of 48).

Reconstituting the Peptides/Proteins

Peptides or proteins often show up at one's door in a lyophilized condition.  In this case, we need to reconstitute them into a solution so they will be soluble and willing to transfer onto glass slides.  There is no magic method for reconstituting peptides or proteins.  The overall charge is a major factor and one can calculate the net charge to determine whether the peptide is acidic, basic, or neutral.  From this information one is guided to select different solutions as good starting points for solubilziing the peptide or protein.  For more information about this procedure check out the Sigma-Genosys suggestions.

In the end, however, the peptide simply goes into solution the easy way or the hard way.  Furthermore, the final solution for the peptide makes a difference in whether it will be conducive to the slide surface.  Thus, there are really two problems with reconstituting the peptide.  The first is getting the peptide into solution and the second is having a solution that is conducive to the surface chemistry of the slide the peptide is supposed to bind to.  Therefore, it's better to settle on a solution that works well with the slide's surface chemistry and sacrifice some in the solubility department.  It's the trade-off one must make in using this technique.

One other item to consider is the final concentration for the peptide once it finally goes into solution.  The optimal concentration for our prints is 0.2mg/mL.  This concentration assures that the binding sites on the slide are completely saturated without using an excessive amount of protein.

Below is my recipe for putting peptides into solution.  It works well with more than 90% of the peptides I've put into solution.  In those cases, where it doesn't it may be beneficial to try the more sophisticated charge-counting approach described above.

• Reconstitute in 1X PBS (final concentration of 1mg/mL)
• If the peptide doesn't go into solution after some PBS and a little mixing:
• Check the pH and adjust it to pH 7.0 (one can also use NaOH or HCl to make adjustments toward a more appropriate pH based on the net charge)
• If the peptide doesn't go into solution after adjusting the pH to 7.0 (or it's optimal value):
• Add DMSO (up to 30%) and mix after each addition
• f the peptide still doesn't go into solution (happens rarely) then one just has to do the best they can at making sure the peptide is well-mixed and uniformly distributed.

--Brian Kidd, 2004