BSCI 1510L Literature and Stats Guide: Electrophoresis concepts

Last week, we predicted the types of fragments that would be formed as the result of digests of the product of our PCR amplification.  This week we will visualize these DNA fragments using a method called electrophoresis, which not only makes the fragments visible, but which allows us to determine their size.

Electrophoresis is one of the most important experimental techniques in modern biology.  It is a workhorse technique for separating substances for the purpose of identification and isolation.  Although there are many variations on electrophoresis, they all share basic features that will be described in the following section.

Electrophoresis is a method for separating a mixture of charged molecules in solution by passing it through a porous gel.  Differences in the charge, size, and shape of each molecule determines how quickly it will migrate toward the anode (+ pole) if it is negatively charged, or cathode (- pole) if it is posi­tively charged.   Initially, all types of molecules begin at a single spot on the gel (the origin), but as time passes each type of molecule forms a discrete band located a different distance from the origin, with bands of faster moving molecules located further from the origin (Fig. 1). 

Fig. 1.  Movement of particles of different size in gel electrophoresis.  The electric field varies in the vertical direction.

A number of mixtures can be separated on a single gel by placing them in a line perpendicular to the electric field at the origin.  The path of each sample through the gel is called a lane. 

Electrophoresis can be done in gels formed in tubes, slabs, or on a flat bed.  A slab gel is formed in a glass sandwich made of two flat glass plates separated by two spacer strips at the edges and clamped together to make a water-tight seal.  Both tube and slab gels are mount­ed vertically.  The gel in a flat bed unit is poured on a horizontal surface and has no cover plate on it.  Gels are mounted between two buffer chambers containing separate electrodes so that the only electrical connection between the two chambers is through the gel. 

During electrophoresis, the rate of migration or mo­bility of a charged molecule through a gel is dependent on:

• strength of the electric field
• net charge of the molecule
• size and shape of the molecule
• ionic strength, viscosity and temperature of the medium in which the molecules are moving.

Some of these factors can be controlled by us when we run the gel, while others are determined by the characteristics of the molecules being separated. 

 We can make all the molecules move proportionally faster through the gel if we increase the electric field in the gel by increasing the voltage applied to the electrodes.  However, as the voltage increases, the temperature of the gel increases through resistive heating.  This is generally bad.  Convection in the gel can cause mixing of the fragments.  Temperature variations can also cause irregularities in the pore size of the gel.  During electrophoresis, uneven heat distribution can cause samples in the warmer center of the gel to migrate faster than at the samples nearer the cooler edges. The uneven heat distribution produces a "smile" effect (Fig. 2) and makes comparison of samples loaded in different regions of the gel difficult. Thus, the problem of heat generation places a practical limit on how high the voltage can be in running a gel.

Fig. 2. A "smile" resulting from uneven heating of the gel matrix.  The origin is at the top of the gel and each vertical strip is a lane containing a different sample.

The charge of molecules is influenced by the pH of their medium.  For example, the charge of a protein comes from the pH-dependent ionization of carboxyl and amino groups.  By changing pH of the solution containing the protein and subsequently the charge of the protein itself, a protein can be made to travel toward either the cathode (negative electrode) or anode (positive electrode).  Likewise, the speed at which a protein travels through a gel can be affected by pH.  Obviously, it is very important to control the pH of solutions used in electrophoresis.  For this reason, the solutions used in electrophoresis are buffers - mixtures that limit the fluctuation of pH around a set value. 

If proteins are electrophoresed under certain conditions their native charge differences can be eliminated so that they can be separated based entirely on their size. Under other conditions, native charge differences can be accentuated to separate similarly-sized proteins that vary by small differences in amino acid sequence caused by genetic polymorphism.  Because nucleic acids are always negatively charged at pHs used for electrophoresis, electrophoresis is used to separate them by size alone. 

Molecule size is generally proportional to mass, so molecules with smaller masses usually move through the pores of the gel faster than those with large masses.  This relationship is fairly straightforward for linear DNA fragments but will not necessarily be so for circular DNA molecules.  For example, a plasmid (consisting of a closed loop of DNA) in the compact supercoiled form moves through a gel much faster than another plasmid of the same mass that is in the relaxed (open loop) form. 

Samples are electrophoresed on or through matrices such as paper, cellulose acetate, starch gel, agarose, or polyacrylamide.  The most commonly used matrices are agarose and polyacrylamide.  Because the pores of an agarose gel are large, agarose is used to separate macromolecules such as nucleic acids and large protein complexes.  Polyacrylamide, which makes a smaller-pore gel, is used to separate most proteins and small oligonucleotides (short, single stranded nucleic acids containing about 25 or fewer nucleotides).  Whichever matrix is chosen, it is important that the matrix be electrically neutral and chemically inert. 

If we let molecules pass through a longer gel, they will travel at different speeds for a longer time, producing a better separation.  However, there are practical limits to the size of a gel, and gel sizes are usually fixed for a particular rig.   An alternative to having a longer gel is to decrease the pore size of the gel.  Usually this is done by increasing the concentration of the support matrix (e.g. agarose, starch, or acrylamide) or in the case of acrylamide by increasing the amount of crosslinker.  It then takes longer for the molecules to move through the gel, but the separation is achieved on a shorter gel.  This slower movement of molecules through the gel can be compensated for somewhat by increasing the voltage across the gel, although heating of the gel places a limit on this.

Unlike changing the voltage across the gel, which simply changes the velocity of all molecules proportionally, changing the pore size of the gel has a different effect on different-sized molecules.  Changing the pore size of the gel is the only way to control the mass range of molecules capable of being separated by a gel of given length.

Generally, the bands on a gel are not visible while it is running.  A dye marker (usually included in the sample loading buffer) that moves at about the same speed as the fastest molecule in the mixture is usually added to the sample so that progress of the run can be monitored.  After electrophoresis is complete, the sample bands must be visualized by staining.  The bands can also be cut from the gel and eluted for sequencing or other operations.

Protein gels are most frequently stained with Coomassie blue, or by photographic amplification systems using silver or other first-row transition metals.  Coomassie blue staining is only sensitive to about 1 microgram of protein per band, whereas the photographic amplification systems are sensitive to about 10 nanograms of protein per band.  Once the gel is stained it can be photographed or scanned by densitometry for a record of the position and intensity of each band.  Nucleic acids are usually stained with ethidium bromide, a fluorescent dye which glows orange when bound to nucleic acids and excited by UV light.  About 5 ng of DNA can be detected with ethidium bromide.  We use GelRed, which is structurally related to ethidium bromide but is less toxic.  Gels are often photographed for a record of the run. 

If the mixture of substances in a gel is complicated (such as a whole-cell extract), locating the band containing a single substance from among the hundreds or thousands of bands on the gel is like finding a needle in a haystack.  The solution of this problem is to probe the bands using complementary DNA or specific antibodies that bind to particular bands on the gel. Because gels are not robust enough to survive these additional procedures, the bands are first transferred to a nitrocellulose, nylon, or polyvinylidene difluoride (PVD) membrane by capillary action or by electro­phoresis using an electric field perpendicular to the surface of the gel. Since the membrane binds the sample in the same pattern as on the original gel, the blot is a faithful copy of the original.  Once transferred to a membrane, the bands are stable and can be stored for over a year.