User:Kinkreet/Protein Science/Protein Separation and Purification

Overview
Before the emergence of recombinant DNA technology in 1978, proteins was only purified from natural sources. With recombinant DNA, one can transfect a vector with the gene that encode the protein of interest into a rapidly-replicating host, such as E. coli, and so a bigger quantity of pure protein can be obtained.

Purified proteins are required for functional analysis of the protein, and also structural determination (X-ray crystallography and NMR spectroscopy) of the protein.

The basis of protein purification exploits the fact that different proteins have different shapes, sizes, surface charges, hydropathy and specific ligand-binding sites.

The time that passes between the sample injection and peak maximum is called the retention time. The area underneath each peak is proportional to the amount of corresponding analyte in solution.

Hydrophobic interactions
Entropy can be defined as a 'lack of order'; polar molecules such as water molecules are organized in a fluid but ordered manner around a hydrophobic patch, and is entropically unfavourable. If there are more than one hydrophobic molecule in the solution, the most entropically favourable result would be when the hydrophobic molecules aggregate together, so the surface exposed to the water molecules is minimal, and so minimize the level of order (or increase disorder). Therefore, hydrophobic molecules will aggregate together in a polar solvent. The same reasoning can be applied to hydrophobic patches of a protein, where the patches will interact with each other attractively and cause the polypeptide to fold so that the hydrophobic residues 'clump' together. If the hydrophobic patches are on the surface, then it will cause the proteins to aggregate with other hydrophobic entities.

Adding salt
If salt (charged ions) are added, then the polar molecules (the 'bulk water') will also arrange orderly around the ions, further decreasing entropy of the water (or increasing the entropic penalty). This makes it even more unfavourable for any proteins with hydrophobic patches to stay in solution. After enough salt has been added, the free energy of precipitation becomes negative and the proteins will precipitate out. Therefore, salting out separates proteins based on their hydrophobicity.

The Hofmeister series is a classification of ions in order of their ability to 'salt out' or 'salt in' proteins. The effects of these changes were first worked out by Franz Hofmeister, who studied the effects of cations and anions on the solubility of proteins.

Hofmeister discovered a series of salts that have consistent effects on the solubility of proteins and (it was discovered later) on the stability of their secondary and tertiary structure. Anions appear to have a larger effect than cations, and are usually ordered:



\mathrm{F^{-} \approx SO_{4}^{2-} > HPO_{4}^{2-} > acetate > Cl^{-} > NO_{3}^{-} > Br^{-} > ClO_{3}^{-} > I^{-} > ClO_{4}^{-} > SCN^{-}} $$



\mathrm{guanidinium > Mg^{2+} > Ca^{2+} > Sr^{2+} > Ba^{2+} > Li^{+} > Na^{+} > K^{+} > NH_{4}^{+}} $$

Salting out works best with divalent anions such as sulphates, especially ammonium sulphate, as it is highly soluble; this is a technique used for concentrating dilute solutions of proteins, as well as for fractionation based on solubility.

Protein Chromatography
There are two types of protein chromatography - adsorption/desorption (solid-liquid phase) and permeation (liquid-liquid phase). At any time during the purification process, the purity can be calculated by dividing the amount of protein of interest, by the total amount of proteins present. An enzymatic-based assay can be used to estimate the concentration of our specific protein (provided it is an enzyme); a dye-based assay (such as the Bradford assay) can be used to see how much protein we have at any one time. The two are used to estimate purity.

Adsorption/desorption
Adsorption/desorption works by first adding proteins to a surface, and then removing them differentially. This is dependent on the thermodynamics between the protein and the solid phase.

Because the surfaces of proteins are heterogeneous due to differences in shape, charge, hydrophobicity and specific binding sites, under one condition (determined by a buffer) a protein may adsorb onto the solid phase, while in a different condition it will desorb from the solid phase.

Types of adsorption/desorption chromatography
There are many types of adsorption/desorption chromatography techniques, the most common are ion exchange chromatography (IEC), chromatofocusing, hydrophobic interaction chromatography (HIC) and affinity chromatography (AC).

Ion-exchange chromatography
Ion-exchange chromatography (IEC) exploits the mutual ionic interactions between charges on the ligand and the solid phase. A protein have many ionisable groups (as weak acids and weak bases) which can be uncharged or differentially charged dependent on the condition, such as the pH, which the protein is in. For example, if the pH is below their isoelectric point (pI), then the protein becomes positively-charged; if above their pI, the protein becomes negatively-charged. The solid phase of the IEC is chosen to include a strong base or acid, which will not change its charge even in extreme conditions; this allows the user to change the conditions of the buffer to allow different proteins to adsorb and desorb onto the column.

The solid phases chosen for IEC have the following properties:
 * Hydrophilic to avoid non-specific binding to hydrophobic molecules
 * Highly porous to avoid size exclusion (we only want separation based on charges and nothing else)
 * Rigid to minimize compression due to flow
 * As the method requires a large range of pH, the solid phase must be stable for a large range of pHs
 * Can be derivatized (attached onto the solid support such as cellulose)
 * Capable of being made into a fine powder or bead

Common substituents on the solid phase includes Diethylaminoethyl (DEAE, a positively-charged weak base, or weak anion exchanger), diethyl-(2-hydroxypropyl)-aminoethyl (QAE, a positively charged strong base), Carboxymethyl (CM, a weak acid, or weak cation exchanger) and sulfopropyl (SP, a strong cation exchanger). DEAE contains a removable proton which means you can remove the charge from it, and so is not always positively-charged, whereas QAE have no such hydrogen and so is always positively-charged; QAE is more usually used. For similar reasons, SP, a strong acid is preferred to CM, a weak acid.

The column is first conditioned using a salt solution, such as NaCl; this allows the ions to attach onto the column. When the ligand is added, depending on the pH and ionic strength, the ligand will displace the column-bound ions and itself bind to the column. The pH determines what the charges on the ligand will be, and ionic strength determines how much ions are present to mask the charge on the ligand. Often, pH cannot be changed due to risk of denaturation, and so more ions are added, in a lesser concentration, to control rate of adsorption and desorption, without denaturing the protein.

The pI of a protein can often be calculated using bioinformatic software, and the working pH should not be more than 1 unit either side of the pI.

Other factors. As the molecular weight of the ligand increases, there is a smaller surface contact area-to-mass ratio, and so the ligand will not bind to the column as tightly as one with the same surface contact area but with a lower mass. The mass of the ligand is allowed to be dragged down the column. Empirically, the capacity to molecular weight relationship follows an approximate logarithmic pattern.

If the column is overloaded, the concentration of ligand will be relatively high; in this case, many of the ligand will simply never bind to the column, or bind rarely. This means the protein will migrate down without interaction with the column; thus larger proteins will migrate slower, as they experience more resistance. This results in a fractionation based partly on size and not charge (similar to a permeation separation technique).

Chromatofocussing
Chromatofocussing is a variance of IEC that gives a very high resolution, where resolution is only constrained by diffusions. It elutes with a proton gradient instead of a salt gradient. Ions are introduced to move the pH from one which favours binding to one which favours elution. As the ion will form a gradient highest at the top and lowest at the bottom, proteins towards the tail will be eluted first/faster, and so the trailing edge is allowed to catch up.

However, the acid cannot be simply added, this is because the protein is in a buffer solution, and the acid concentration does not have a linear relationship with pH in buffer. Thus, a buffer with many pKas (ideally infinite number of pKas) should be used to give a roughly linear relationship. Typically, polyaminopolycarboxylic acids are used, because amino acids have a range of ionisable groups, and thus can act as buffers in a high enough concentration.

Hydrophobic Interaction Chromatography (HIC)
On a globular protein, there will be patches of charged areas, and patches of hydrophobic areas; hydrophobic interaction chromatography (HIC) bases the adsorption and desorption on hydrophobic interactions between the column and the protein. The column is hydrophobic and made up of phenyl or acyl groups, with octyl groups being more hydrophobic and thus a stronger binding partner.

The proteins are loaded with a high salt solution, decreasing the solvent's capacity to tolerate hydrophobic proteins, and so the proteins will bind with the column to reduce the entropic penalty (the salting out effect). After binding, the proteins are eluted using organic solvents such as ethylene glycol or methanol, mixed with water.

Reverse Phase Chromatography (RFC). RFC is the same as HIC, only that the solid matrix (the column) consists of even more hydrophobic ligands such as C18 or biphenyl. This high level of hydrophobicity denatures the protein; and because it is so hydrophobic, organic solvents can be used throughout. RFC is carried out at high pressures, giving high resolution. The proteins are denatured and so usually this technique is not used with proteins, but with peptides and carbohydrates.

Affinity Chromatography
The previous three techniques bases adsorption and desorption on general properties of the protein such as charge and hydrophobicity; affinity chromatography bases the technique on specific groups ofthe protein.

Although not designed to be, affinity columns will no doubt have separation based on hydrophobicity and/or charge, the extent depends on the nature of the solid matrix and the ligand; this also leads to non-specific binding. Non-specific binding when the protein mixture is crude and at high concentrations can lower the specificity of the technique.

Due to its specificity, it is often the most efficient method for purification, and so when available, is usually the first step in the purification process. However, columns specific to the ligand is not always available, or is too costly.

Common ligands include substrate analogues and inhibitors (often used in pull-down assays), cofactors, lectins, dyes, antibodies and fusion proteins. Fusion proteins are the typical ligands used, they are proteins with tags at the C- or N-terminus which binds to the column; common tags include glutathionine-S-transferase, polyhistidine or maltose binding protein.

The column for affinity chromatography for any specific proteins (i.e. not the common tags) often require to be made (or derivatized) ourselves. First a matrix with many hydroxide groups are activated by cyanogen bromide (CNBr), and reacted with a spacer arm (containing a (CH2)6 group flanked by two reactive end groups - E1 and E2). E1 reacts with the matrix while E2 can bind to the ligand.

The ligand can bind using its structural and spatial complementarity, but it can also bind by forming a covalent bond with the spacer; this bond can be an amide, a carbamate, thioether or other bonds. The thioether bond allow better recovery of the original ligand, as the S-R ether bond is relatively weak.

Ligands can be eluted using an analogue of the ligand (elution by competition), but this can be expensive if the ligand is not common. This works well with fusion proteins containing common tags, and also with lectins.

If the protein is a fusion protein, the tag needs to be removed after to regain the original, desired protein. This can be done by genetically incorporating a proteolytic site between the tag and the protein. The site must be picked so that it is unique and not in the tag nor the protein, to ensure proteolysis occurs only between the tag and protein. Common cleavage sites include thrombin, Factor Xa, tobacco etch virus protease and enterokinase.

Overview
Adsorption and desorption happens at an equilibrium described by the Langmuir isotherm, which describes adsorption in relation to gas pressure (for gases) and concentration (for liquids), at a fixed temperature. It assumes that all sites are equivalent, are spread out as a monolayer, and each binding event is independent. However, because the Langmuir isotherm was originally developed for adsorption of gases onto a solid surface, it has shown a poorer fit when applied to solution systems. Some solutes have a higher affinity to pre-adsorbed surfaces, where water molecules are already pre-adsorbed onto it. Sohn and Kim proposed a modification to the isotherm by introducing a concentration dependent factor, which assumes that the concentration of the solute affects adsorption and desorption.

Derivation
At equilibrium, a protein, X, adsorbs and desorbs onto the adsorbant, M, according to$$:

The association constant, $$K_A$$ is defined as: $$\theta$$ can be defined as the ratio between occupied and total number of adsorption sites. As the concentration of adsorbed proteins is proportional to the number of occupied sites, The number of free adsorption sites is equal to $$1-\theta$$: Combining $$, $$ and $$, we get: To make this into an equality, a proportionality constant, $$\tau$$, is introduced: $$ can be expanded and rearranged to give the general form of the Langmuir isoform: where $$K_D=\frac{1}{K_A}$$ Because the concentration of proteins, $$[X]$$, is difficult to measure, we can make two small assumption in order to get exchange $$[X]$$ for a known term. We also do not want to know the fraction of the surface bound, given by $$\theta$$, but the fraction of protein bound, $$\alpha$$.

$$\theta$$, by definition:

where $$q$$ is the concentration of binding sites occupied, and may be different from $$[XM]$$ if there are multiple binding sites on the same protein. $$\Gamma$$ is the total concentration of binding sites.

The first assumption is to assume that the proportionality constant, $$\tau$$ is insignificant. Therefore, $$\tau [X] = [X]$$. $$[X]$$ can then be defined as $$[X]=P_T-q$$, where $$P_T$$ is the total concentration of protein (Free = Total - Bound).

Using the assumption and combining $$ and $$, we get:

By definition: or

Combining $$ and $$:

After expansion and elimination of one $$P_T$$ term, we are left with:

If we assume that the concentration of binding site far exceeds the total number of proteins, i.e. $$\Gamma \ggg P_T$$, or that there is a large excess of binding capacity. Then, terms with $$P_T$$ can be omitted as insignificant. And $$ simplifies to:

Application
$$K_D$$ is constant when in a certain environment, but will change with varying pH, solvent, ionic strength etc. When $$K_D$$ changes, so will $$\alpha$$

For adsorption/desorption chromatography, proteins are adsorbed by using conditions where $$0.8 \le \alpha \le 1$$, and desorbed using conditions where $$0.5 \ge \alpha$$

Remember that to use the Langmuir isoform in the form given in $$, we must make four key assumptions: monolayer coverage, sites equivalence, sites independence, and a large binding capacity. It is a good approximation when the concentration of binding sites is high, and the dissociation constant is low.

Although the Langmuir isotherm may be a good approximation, it does not take into account the inconsistency of the adsorbant, both in concentration and in adsorptive strength. It may be hard, also, to measure the concentration of binding sites accurately.

Dispersion
In chromatography, when the protein is added to the top of a column, the band spreads out, or disperses. Longituinal diffusion and eddy diffusion (turbulent diffusion) are the two main contributing factors to dispersion. Dispersion, σ, can be defined as the difference between the volume reading when protein concentration is at half the peak. The resolution is defined as:
 * $$v_2-v_1/(\sigma_1+\sigma_2/2)$$

where vn is the volume reading at the peak protein concentration of protein n. The concentration of proteins is measured by fluorescence or UV spectrometry, by measuring the intrinsic fluorescence of tryptophan and other amino acids.

Longitudinal diffusion occurs due to random molecular movements, where a population of the same molecules will move from an area of high concentration to an area of low concentration. A band will diffuse in both directions (front and back) from the centre of the band, and spread the band into a Gaussian shaped-profile. The longer the sample is in the column, the more spread-out the peak will be, because the band is given longer to diffuse.

Eddy diffusion occurs because each molecule can pass through the column using an infinite number of paths, where some paths are longer than others. The solid phase exists as finely grounded powder or beads, and their surfaces are uneven. When the solution flow pass through the solid phase, small uneven eddy currents will form, which increases the number of paths a molecule can take, such as it allows molecules to move in the opposite direction to the general flow. Molecules that take the shortest path will be eluted first, whereas those that take longer paths will be eluted later; eddy diffusion, therefore, also causes band broadening.

Therefore, a fast flow will increase eddy diffusion while reducing longitudinal diffusion; a slow flow will decrease eddy diffusion and increase longitudinal diffusion. The optimal flow rate can be described by the following empirical equation:


 * $$\sigma=\sqrt{L({2D/v}+2\alpha(1-\alpha)vt_d+2\lambda d_p)}$$

where $$L$$ is the length of the column, $$v$$ is the flow rate, $$D$$ is the diffusion coefficient, $$d_p$$ is the particle diameter, $$t_d$$ is the desorption time, $$\lambda$$ is the obstructive factor. This formular tells us that the longer the column, or the longer the time it takes an adsorbed particle to desorb, or the higher the obstructive factor, and larger or rougher the particles of the solid phase, the higher the dispersion. However, the flow rate must be chosen to minimize the $$2\alpha(1-\alpha)vt_d+2$$ and $${2D/v}+2$$ terms. Also note that the particle size cannot be too small, otherwise the column becomes impermeable.

Tailing and and fronting
In theory, longitudinal diffusion and eddy diffusion should lead to a band with a Guassian shape concentration distribution, however this is not the case.

Tailing. As the band migrate down the column, proteins at the front of the band encounters many free binding sites, and thus is likely to bind. Proteins at the back of the band encounter less free binding site, because many are already bound by proteins which have passed through it. This means the front will be slowed to a larger degree than the back, meaning we will have a sharpening at the leading edge.

Fronting. As we elute by adding in salt, it will form a gradient where the concentration is highest at the top of the column and lowest at the bottom. This means proteins nearer the top (at the tail) will be eluted faster than the leading edge. This results in a sharpening of the band at the trailing edge.

Tailing and fronting both distort the shape of the band from a Guassian distribution, but the two, to some degree, cancel each other out, to give a relatively clear band.

Permeation
Permeation is usually tried only after adsorption/desorption-type chromatography have failed to give satisfactory purification.

Permeation works by keeping the proteins in liquid and exploit diffusion rate differences between liquid phases; and thus permeation is based on kinetics - generally that larger proteins diffuses slower, as they experience more resistance in its movement. Contrast this to adsorption/desorption, which is based on the equilibrium of binding/dissociating. Although it separates the proteins based on a different parameter, the sample should already be relatively pure, with a high concentration of the desired protein and low concentration of other

Gel filtration/Size exclusion chromatography
In gel filtration chromatography, the column is filled with many hollow beads with pores on it. The beads and the volumes inside the beads contribute only 1-5% of the total volume of the column, the rest are extraparticular, or void, volume. Larger proteins are unable to enter the bead because it cannot pass through the pore, thus they will continue down the column. Smaller proteins can pass through the pore and into the bead, where it will be trapped for a short while before passing out through the pores again; this trapping slows the migration of smaller proteins down the column. Therefore, larger proteins will pass through the quickest, and smaller proteins the slowest.

One of the main advantages of this technique is that there is little or no interactions with the solid phase, and so the protein is not likely to be denatured.

Protein Electrophoresis
Whereas chromatography is used to separate the protein fractions for other uses (it is preparative), electrophoresis analyses the purity, concentration and abundance of the proteins. Electrophoresis uses an applied electric field to separate proteins. Most protein electrophoresis is carried out in a polymeric medium, because in free flow electrophoresis, increased rates of diffusion and the effects of convection currents broaden the bands and thus reduce resolution.

The rate of migration depends on many factors: field strength (controlled by the voltage used during run), ionic strength (controlled by buffer), net charge, temperature, molecular size and shape (the parameter we are trying to determine) and viscosity (controlled by the gel percentage).

After the proteins have finish their run, it can be detected on the gel with Coomassie Brilliant Blue dyes or by silver staining. It can also be transferred onto a nitrocellulose membrane and use Western Blot to visuallize it. With staining, all the proteins are stained, whereas in Western blots, only proteins which the antibody can bind are stained.

There are 5 types of protein electrophoresis: denaturing conditions, native conditions, isoelectric focussing (IEF), 2D and titration curves.

Denaturing gels uses denaturing conditions to straighten the polypeptide chain, and therefore, its migration depends only on the size of the protein. Native gels runs without denaturing detergents, and thus remains folded during the run, and so migration depends on the size and shape of the protein. (A denaturing and native gel can be ran alongside each other to give information about the quaternary structure, for example whether the protein is a dimer or trimer).

Isoelectric focusing - migration depends only on the pI.

In 2D electrophoresis, the gel is ran using two conditions (denaturing/native and pI), one after the other and orientated perpendicularly. This will separate the proteins based on pI in one dimension, and by size in the other dimension.

Titration curves electrophoresis depends on the number of titratable amino acid side chains.

Agarose gels are used for DNA electrophoresis; for proteins, polyacrylamide is used. Polyacrylamide gels are made by cross-linking acrylamide with methylenebisacrylamide in a typical ratio of 29:1. The acrylamide forms rows while the methylenebisacrylamide links the rows together. The gap between the rows forms the pores. The pore size is determined by the concentration of acrylamide to methylenebisacrylamide, with a higher percentage of methylenebisacrylamide meaning bigger pores.

Denaturing SDS-PAGE
In denaturing SDS-PAGE, the proteins are first denatured by heating to 95-100°C in Laemmli 2X buffer, which contains the reducing agent β-mercaptoethanol to reduce disulphide bonds, and sodium dodecyl sulphate (SDS) to straighten out the protein. The SDS is charged and binds to the protein, this effectively masks any intrinsic charges the protein has. The protein will have the same charge density and thus migration is only based on molecular weights.

Higher resolutions can be obtained using discontinuous gels, where the gel is made up of a stacking gel, and a running gel. The stacking gel is typically 5% acrylamide and so proteins of any molecular weight can pass through easily, and there is no molecular sieving. The running gel contains glycine at high pH, which is key to the high resolution.

As with any amino acids, glycine can exists as a zwitterion, where the carboxylate group have a negative charge but is balanced by a positive charge on the amine group. Glycine have a pI of ~5.97, at this pH, the amino acid have no overall charge. At high pH, as in the running gel, glycine becomes negatively charged as it loses its positive charge on the amine.

In the stacking gel, there are no molecular sieving, so all moieties are separated based on their charge. Therefore, Cl- will travel faster than glycine, because it is more negatively charged than glycine at pH 6.7 (pH of the stacking gel). This creates a boundary between glycine and Cl-. The proteins, bound by SDS, is negatively charged, but not as much as Cl-, and so will travel in the Gly/Cl- boundary. This focuses the proteins into a narrow band when it enters the running gel.

When it enters the running gel, the pH is increased and so glycine is now negatively charged and can migrate down. The gel percentage means molecular sieving can occur, and so the protein can now be separated based on molecular weight.

Native gel PAGE
The principle behind native gel is the same as denaturing gel, but it does not heat the proteins or use SDS to give the proteins charge, because SDS also denatures the protein. Instead, the proteins is ran at a high pH. Because most amino acids have a pI of 5-6.5 (except Asp [2.77], Glu [3.22], Lys [9.74], Arg [10.76] and His [7.59]), and most proteins have a pI of 4-7; therefore, when ran in a high pH (8-9.5) gel, the proteins will be negatively charged and thus will migrate towards the anode.

The speed of migration can be altered by altering the charge of the protein, this is done by altering the pH: with the pH close to the pI, the protein is less charged and will migrate slower. The separation also depends on the size of the protein, with larger proteins migrating slower. The conformation of the protein is also important: more linear proteins will travel faster than globular proteins.

Therefore, native gels separate proteins based on size, conformation and charge. It is often used to characterize complexes. Also note that no discontinuous gel is used.

There is an alternative native gel called blue native gel, where a Coomassie dye is used to bind to the protein to make it run more evenly. The dye is charged and will migrate, similar to SDS, but Coomassie dye do not denature and is not as harsh, and so the protein will still retain its conformation.

Isoelectric Focusing (IEF)
In IEF, the protein migrate through a pH gradient, created by ampholytes (such aspolyaminopolycarboxylic acids) which have a range of pKas. In an electric field, the differentially charged ampholytes and protons separate out, creating a pH gradient.

Proteins can be added anywhere in the column and will migrate to the pH which matches their pI in that position. If the protein is at a lower pH than its pI, it will becomes positively charged and move towards the cathode; likewise, it it is at a higher pH than its pI, it will move towards the anode. Eventually, it will reach a pH point where it has no net charge, and will not migrate; this pH is the protein's pI. As different proteins will have different pIs, IEF is able to separate proteins into different bands.

This technique have the advantage that its resolution is not affected by diffusion, as any proteins which diffuses away will migrate back. This autofocusing allows a high resolution (<0.1 pH units).

2D PAGE
In 2D PAGE, two separation techniques are used. Typically, the sample is separated based on a pH gradient first, before being rotated and migrate through a denaturing or native PAGE.

Because it uses two dimensions, it has high resolution; thus it is the basis of many proteomic studies. Furthermore, the dot on the gel can be cut out and analysed using mass spectrometry.

Titration PAGE
Titration PAGE is a type of 2D PAGE where the protein is first separated based on its pI in isoelectric focusing. One band is then taken and applied onto a gel across a broad pH gradient. If there are more than one proteins in the band (with the same pI), they will separate out as the degree at which pH affects the magnitude of its charges will be different.

Capillary Electrophoresis
The idea behind capillary electrophoresis is that it is carried out in a capillary. The capillary is thin (25-100 µm diameter and <50cm long) have a high surface area to volume ratio, and so any heat caused by the current is dissipated quickly, preventing heat damage to the sample; therefore, CE can be carried out at high voltages (10 - 30 kV) to allow for fast separations (1 - 45 minutes, thus reducing the effect of diffusion). Also because it is thin, a light can be shown through the capillary into a detector to detect absorbances and other parameters. The surface of the silicate glass capillary contains negatively-charged functional groups that attract positively-charged counterions in the buffer. The positively-charged ions migrate towards the negative electrode; because it is a capillary, the cations will carry solvent molecules in the same direction. This overall solvent movement is called electroosmotic flow. Separation is based on the effect of electrophoresis and electroosmotic flow. Due to electroosmotic flow, all sample components migrate towards the negative electrode, but with a different rate. Uncharged molecules will move at the same velocity as the electroosmotic flow (with very little separation). Positively-charged ions move faster and negatively-charged ions move slower due to electrophoretic effects. And thus separation is based on charge and mass. There are two buffer reservoirs, which maintains a non-denaturing condition for the proteins. There are three injection systems: gravity injection and pressure injection. Gravity injection first removes one of the buffer reservoirs, and replaces it with the sample reservoir, elevated above the other buffer reservoir. Because it is at a higher gravitational potential, sample will enter into the capillary. In pressure injection, pressure is applied to the sample, or a vacuum to the buffer reservoir, to act in an analogous way to gravity in gravity injection. After injection, the buffer reservoir is replaced and voltage applied. The size of injection are in the nanolitre to microlitre scale.

The main advantages is its high resolution, able to detect subtle post-translational modifications. The disadvantages of CE includes it only works with molecules which are easily dissolved in solvent, and it can be hard to reproduce.

Capillary isoelectric focusing (CIEF) is the same as IEF, only carried out in a capillary. Electroosmotic flow must be eliminated by coating the capillary with Teflon. After the protein has focused, all the samples in the capillary are recovered by mobilizing the whole volume using pressure or chemicals.