HOW DOES AN SDS PAGE GEL REALLY WORK?                      Dyche Mullins 2/1/2002


The system most people use for separating proteins by gel electrophoresis was formulated by Laemmli (Nature 227:680-685 [1970]).  It is such a commonly used laboratory technique that nowadays we take it for granted and I dare say that many of us don’t know how a discontinuous pH gel system actually works.


First the sample. Most SDS PAGE sample buffers contain the following: SDS (sodium dodecyl sulphate, also called lauryl sulphate), b-mercaptoethanol (BME), bromophenol blue, glycerol, and Tris-glycine at pH 6.8. BME is added to prevent oxidation of cysteines and to break up disulfide bonds. Bromophenyl blue is a dye that is useful for visualizing your sample in the well and tracking its progress through the gel. Glycerol is much more dense than water and is added to make the sample fall to the bottom of the sample well rather than just flow out and mix with all the buffer in the upper reservoir. The interesting components are the buffer and the SDS.


SDS is an ionic detergent that binds to the vast majority of proteins at a constant ratio of 1.4 gm SDS/gm protein. A few proteins like tubulin do not bind at this ratio and this is one reason why some proteins migrate anomalously (there are other reasons as well so you shouldn’t put too much faith in the apparent molecular weight estimated from an SDS PAGE gel).  Since SDS is an anionic detergent it imparts a negative charge to all the proteins in your sample. More importantly, these charges swamp the inherent charge of the proteins and give every protein the same charge-to-mass ratio. Because the proteins have the same charge-to-mass ratio, and because the gels have sieving properties, mobility becomes a function of molecular weight. But what about running gels, stacking gels, electrode buffer, and all these different pHs?


The velocity of a charged particle moving in an electric field is directly proportional to the field strength and the charge on the molecule and is inversely proportional to the size of the molecule and the viscosity of the medium. Adding a gel with sieving properties (that is a gel where the resistance to the motion of a particle increases with particle size) increases the differences in mobility between proteins of different molecular weights. This is the basis of separation. The problem now becomes how to line up all the proteins in an orderly fashion at the starting gate. That’s where the discontinuous pH part comes in.


Laemmli gels are composed of two different gels (stacker and running gel), each cast at a different pH. In addition, the gel buffer is at a third, different pH.  The running gel is buffered with Tris by adjusting it to pH 8.8 with HCl. The stacking gel is also buffered with Tris but adjusted to pH 6.8 with HCl. The sample buffer is also buffered to pH 6.8 with Tris HCl (note all the chloride ions – they will become important in a minute).  The electrode buffer is also Tris, but here the pH is adjusted to a few tenths of a unit below the running gel (in this case 8.3) using only glycine – nothing else. We run our gels at constant voltage.


So here’s what happens when you turn on the power.  Glycine is a weak acid and it can exist in either of two states, an uncharged zwitterion, or a charged glycinate anion (that is to say, negatively charged). At low pH it is protonated and thus uncharged. At higher pH it is negatively charged. When the power goes on the glycine ions in the running buffer want to move away from the cathode (the negative electrode) so they head toward the sample and the stacking gel. The pH there is low and so they lose a lot of their charge and slow down. Meanwhile, in the stacker and sample the highly mobile chloride ions (which are also negatively charged) move away from the cathode too. This creates a narrow zone of very low conductance (in other words very high electrical resistance) in the top of the stacking gel. Because V=IR almost all of the voltage that you put across the gel (110 Volts is typical for stacking) is concentrated in this small zone. The very high field strength makes the negatively charged proteins move forward. The trick, however, is that they can never outrun the chloride ions. If they did they would find themselves in a region of high conductance and very low field strength and would immediately slow down. The result is that all the proteins move through the stacker in a tight band just behind the moving front of chloride ions. Behind them, the pokey glycine ions straggle along as best they can (they do move, but with lower mobility than the chloride ions).


The effect of this moving zone of high voltage is that all the proteins reach the running gel at virtually the same time so that migration of the proteins is truly a function of molecular size and not some complicated function of how carefully you loaded the gel and when you started the voltage.


When the big caravan of ions hits the running gel everything changes. The pH goes way up and the glycine becomes deprotonated (and thus more negatively charged). The mobility of the glycine goes way up and the mobility of the proteins goes way down (due to the sieving properties of the gel). The result is that the glycine races past the protein and the proteins are no longer in a narrow zone of very high resistance (and very high electric field). They find themselves in a much more relaxed, uniform electric field where they can chill out a bit. Move at their own pace.






Tips for running good gels:


1.      After pouring the running gel, carefully overlay it with ethanol or another imiscible liquid. This will give you a nice flat surface. Also, since polymerization of acrylimide is inhibited by oxygen it will speed up polymerization.

2.      For the mini-gels we run the minimum protein loading per well (single band) is 0.1 µg for standard Coomassie staining and 2 ng for silver staining. I haven’t tested it but my impression is that Simply Blue staining is within a factor of two as sensitive as standard Coomassie staining.

3.      The maximum protein loading per well (for a mixture of proteins of different sizes) is about 40 µg. If you exceed amount this your gel will look like crap.

4.      KCl causes SDS to precipitate. If you samples contain KCl you should dilute them or methanol precipitate them and resuspend them in 1X sample buffer. With low concentrations of KCl (<200 mM) you can run them on the gel but you should loaed every lane with sample buffer containing the same concentration of KCl (even if they are blanks). This will help the gel run a little less anomalously.

5.      If your sample buffer turns yellow, it is at the wrong pH. Add NaOH or




Safety Notes:

- Acrylimide is extremely toxic, causing central nervous system paralysis. It can be absorbed through unbroken skin. If skin comes in contact with acrylimide solution or powder, wash immediately with soap and a lot of water. Unpolymerized acrylimide should be polymerized with excess catalyst and disposed of with solid waste. DO NOT POUR UNPOLYMERIZED ACRYLIMIDE DOWN THE SINK.

- Amonium Persulfate should be made up fresh or used from a relatively fresh stock. It goes bad after a week or two in the refrigerator. It can be disposed of by dilution with water and pouring down the sink.

- TEMED should be stored in the refrigerator in dark glass bottles. A bottle should be good for about a year, maybe longer.