Science

Acids, bases and pH

Atomic structure

Biochemical nomenclature

Biotransformation

Carbohydrates

Carnivorous plants

Cell communication

Cell cycle

Chromatography and electrophoresis

Cladistics

Core metabolism

DNA replication

Electron microscopy

Enzymes

Enzyme mechanisms

Essays

Extracellular matrix

Genetics

Genomes

Kinetics

Lipids and membranes

Molecular biology methods

Moles

Natural selection

Nervous system

Nitrogen metabolism

Nuclear structure

Nucleic acids

Oxidative phosphorylation

Phloem

Photomorphogenesis

Photorespiration

Photosynthesis

Plant action potential

Plant circadian rhythms

Plant growth

Plant mineral nutrition

Plant secondary metabolites

Prokaryotes and eukaryotes

Proteins

Protein targeting

Reaction mechanisms

Seeds

Spectroscopy

Statistics

Thermodynamics

Transcription

Translation

Water potential

You are what you eat

Reference

Domains

Genetic code

Geological timeline

Model organisms

Molecules

Plants

Vegetable empire

Steve's Place

Search

Guestbook

Me

Perl

Science

Rants

Chromatography & Electrophoresis

Objectives

• Chromatography.
• Electrophoresis.

Chromatography

Chromatography is the most basic and widely used way of separating molecules in biochemistry. All chromatography relies on the differential partitioning (or sorption) of a molecule between two phases. In the most basic form of chromatography, liquid-liquid extraction, we simply take an impure water solution of a chemical, adjust the pH so as to make the chemical unionised, and then shake it with an immiscible non-polar solvent. Ionised compounds dissolve well in water and preferentially partition into it, but unionised compounds dissolve well in water-immiscible solvents like hexane, chloroform or dichloromethane (DCM) and preferentially partition into them. Since the chemical is unionised, it will be more soluble in the non-polar phase than in the polar phase (like-dissolves-like), and will preferentially partition into it, leaving much of the impurities behind in the water. We can then allow the solvents to separate, and use a separating funnel to separate the two layers, and discard the one we don't want.

We can repeat this process, purifying the compound further (but with reduced yield), until we eventually get bored, and evaporate off the solvent and allow the purified compound to crystallise. So the idea of liquid-liquid extraction is:

• Change pH to make chemical unionised.
• Shake with hexane. Chemical partitions into solvent.
• Separate off solvent with a separating funnel. Discard aqueous layer.
• Change pH to ionise chemical.
• Shake with water. Chemical partitions into water. Discard solvent layer.
• Repeat as required.

Morphine is an alkaloid which can be extracted into a water syrup from ground up opium poppy capsules. The pK_a of Morphine-H⁺ is 7.87 (note the charge: it's sort of like one of those basic amino acids). Hence to extract it:

• Adjust syrup pH to 9. Add DCM. Morphine becomes unionised and proportionates into DCM.
  • M-H⁺ (aq) → M (dcm).
• Keep DCM layer, discard aqueous layer which contains impurities. Add dilute H₂SO₄ at pH 1. Morphine ionises and proportionates into dilute acid layer.
  • M (dcm) → M-H⁺ (aq).
• Keep aqueous layer, discard DCM layer with more impurities. Adjust to pH 9 with sodium hydroxide. Morphine becomes unionised again, and becomes much less soluble in water. Hence we can simply concentrate the solution by rotary evaporation under vacuum until the morphine precipitates as a solid.
  • M-H⁺ (aq) → M (aq) → M (s).
• Add HCl to the precipitate to form morphine hydrochloride solution (this is the form in which morphine is traditionally supplied. Morphine less the HCl is termed the 'free base' form, q.v. crack cocaine).
  • M (s) → M-H⁺ (aq).
• Concentrate and crystallise solution.
  • M-H⁺ + Cl⁻ (aq) → M-HCl (s).

From the principles of liquid-liquid extraction, we can define a partition coefficient, the concentration in phase 1 divided by the concentration in phase 2. For example, the most widely used coefficient, in pesticide degradation studies, is the octanol/water partition coefficient:

K_ow = [A]_octanol ⁄ [A]_water

The partition coefficient can be used to determine expected yields and concentrations: if you mix a known volume of octanol with a known volume of a chemical in water solution, the concentrations in each phase can be determined very easily using K_ow.

From liquid-liquid extraction, we now move to real chromatography. The main difference is than in chromatography, there are still two phases, but in this case, one can move (i.e. is a solid) and the other cannot (i.e. is a fluid: gas or liquid). These are called the stationary and mobile phases. The mobile phase moves over the stationary phase, either by capillary action (TLC), or by pumping (GC, FPLC, HPLC). Most people have done paper chromatography with felt-tip inks. Here the paper is the stationary phase, and the mobile phase (water) moves across the paper by capillary action.

When you add a mixture of chemicals to a chromatographic system, the different time spent partitioned in each phase determines the speed of migration (with the felt tips, some inks are more soluble in water, and move more quickly across the paper). One that spends most of its time partitioned into the mobile phase will move more quickly than one that spends most of its time stuck to the stationary phase.

There are many chromatographic techniques: we will discuss them in turn.

• Thin layer chromatography (TLC). Easy, cheap and quick, not good for quantitative work.
• Liquid chromatography (LC, FPLC, HPLC). Low temperature, very expensive, good for both quantitative (preparative) and qualitative work.
• Gas chromatography (GC). Informative if rigged to mass spectrometer, simple in concept but generally unsuitable for heat-labile compounds.

In TLC, the stationary phase is a silica layer on a glass plate, or just plain old paper (as you will have used at school). The mobile phase can be anything, but is often methanol, water or toluene. The mobile phase moves by capillary action, rather than by pumping, which is the norm for chromatography. The detector is usually the eye: you can see the spots, although you may need to view the plate under UV light. You can use fluorescent plates (for nonfluorescent compounds, which will appear as dark spots), or nonfluorescent plates (for fluorescent compounds, which will appear as bright spots). Note that TLC is denaturing to some compounds, as the solvents are quite nasty, and the technique involves spotting compounds onto a plate, letting them dry, then running them in a tank full of air (i.e. it's rather oxidising). TLC is useless for proteins therefore.

Liquid chromatography (LC) is somewhat more complex: here the basic system is composed of a pump, an injection port, a column, a detector and an integrator. In the pump, an organic solvent, buffers or water are mixed together in a specific ratio by the pump and made to travel at a certain flow rate. The solvents (often termed running buffers) must be degassed before use, because the high pressure that the column runs under, and the low pressure that the detector runs under, lead to gassing out, i.e. bubbles in the detector, which ruins results. The mobile phase then runs through the sample port, where a volume of sample can be injected into the mobile phase with a sample loop of known volume. Samples must be prefiltered before injection. After injection of the sample, the mobile phase often passes though an internal frit filter and a guard column to prevent gunge getting onto the column proper. The column contains a resin that effects the separation in combination with the mobile phase. The detector detects some property of the mobile phase and any compounds contained within it (absorbance, refractive index, etc.) and feeds the result to an integrator: for example, the absorbance of the mobile phase is plotted against time, giving peaks on the paper of a chart recorder. After the run finishes, the area under the peaks is mathematically integrated, and a report is printed out. The area under each peak is proportional to the amount of the chemical that eluted to give that peak. Given a known sample loop volume it is also proportional to the concentration of the chemical in the original sample. The detector is usually calibrated with known concentrations of the chemical of interest. The first things to elute are usually the solvents in which you dissolved your samples. This gives a large messy peak called the solvent front. The commonest form of LC is high pressure LC, or HPLC.

In HPLC, the stationary phase is silica beads, or silica beads coated with long chain alkyl (fat) groups. An HPLC column is somewhat akin to a TLC plate wrapped up into a tube. The mobile phase can be anything, but is often acetonitrile, methanol, hexane or water. HPLC is very similar in theory to fractional distillation, and the mathematical techniques used to describe HPLC often talk about 'theoretical plates' and similarsuch words culled from the oil industry.

If the stationary phase is fat-covered (hydrophobic) silica, we use water as the mobile phase: water soluble compounds drop off the column quickest (elute first), whilst fatty compounds dissolve into the alkyl chains on the stationary phase and are retained for longer. To speed things up we can use a mixture of acetonitrile and water, because acetonitrile dissolves up fatty compounds well. This technique is called reverse phase HPLC.

If the stationary phase is normal (hydrophilic) silica beads, water soluble compounds will stick to the stationary phase, and elute last. The mobile phase is often hexane, so fatty compounds elute first. This is the opposite to RP-HPLC and is called normal phase HPLC. Despite the name, RP-HPLC tends to be the more widely used of the two techniques.

The time the compound spends on the column is called the retention time (R_t), and is characteristic for a given set up and a given compound. It is analogous to the R_f value in TLC. The detector used to monitor when the compound falls off the column is often a small UV spectrophotometer, although refractive index specs, diode arrays (which monitor several wavelengths at once), and mass spectrometers are also common, if expensive.

There are several variables that can be manipulated to change the elution profile:

• Flow rate. Increased flow rate leads to quicker elution, and tighter peaks (because the compounds in the columns have less time to diffuse), but it also reduces resolution, i.e. because the compounds spend less time differentially partitioned into the two phases, they will come off closer together.
• Gradient or isocratic. That is, ramping up the amount of acetonitrile, or whatever organic solvent you are using, during the elution (gradient), or not (isocratic).
• Column length. Long columns give greater resolution but broader peaks, for the same reasons as the flow rate.
• Column packing. This can be chiral, normal/reversed phase, etc. Normal phase is hydrophilic resin/hydrophobic solvent, reversed phase is hydrophobic resin/hydrophilic solvent.

In GC, the stationary phase is a thin hydrophobic silicone polymer film in a 30 m long (or longer) capillary tube. The mobile phase is helium gas and the detector is usually a mass spectrometer. Elution of the compounds is by increasing temperature: as the temperature increases, compounds that don't stick well to the GC column will be shaken off it by their increased thermal vibrations. The GC tube is kept in an oven for this, and we can alter how we heat the oven containing the capillary tube up to achieve different elution profiles. GC is fairly useless for heat labile compounds (e.g. proteins). Some compounds are not hydrophobic enough to stick to the GC column well. For these compounds, we can derivatise samples with hydrophobic side chains (e.g. trimethylsilylate them: stick fatty (CH₃)₃Si- groups to it) to increase binding of hydrophilic molecules to the hydrophobic column.

Fast protein liquid chromatography (FPLC) is used for proteins: this is basically a form of HPLC that runs under low pressure, with a 'resin' made from an inert substance like cellulose or dextran, with side groups attached to it to give it specific binding properties. The side chains we add determine the sort of chromatography:

• Hydrophobic LC separates proteins by the amount of hydrophobic amino acids the contain by using fatty alkyl chains or similar bound to the inert resin. This is useful for membrane proteins, as these contain hydrophobic amino acids.
• Ion exchange LC separates proteins by the number and sort of charged amino acids. Cation exchange resins have negatively charged groups as their side-groups (e.g. carboxylic acids groups like carboxymethyl), which allow them to bind and exchange cations (positively charged proteins, and other cations). To elute a protein, you increase the concentration of salt (NaCl) in the buffer that is pumped over the resin: the Na⁺ ions compete for binding with cationic proteins, until they drop off. Anion exchange is similar, but with cationic side chains (e.g. amines like DEAE), binding anionic proteins, which can be eluted with salt, here it is the Cl⁻ ions that knock the protein off.
• Ligand/affinity LC separates proteins by their specificity to certain substrates, dyes or antibodies. The resin has antibodies specific to a particular protein, metal ions the protein binds to, ligands (compounds the protein binds, like NADH, or certain dyes), nickel (histidine binds nickel, and proteins can be genetically modified by adding a histidine tag) or similar bound to it. A useful technique, but quite expensive, and it can be difficult to elute the proteins off!
• Size exclusion LC (or gel filtration) separates proteins by their size. Beads with tiny pores in them are packed into a column. Small compounds enter the beads and are retained, large compounds fall straight through the column. The beads are usually discarded. The technique is sometimes called gel filtration, and is very useful for 'desalting' proteins: eluate from ion exchange resins containing a protein of interest are usually full of the salt used to elute the protein. SEC can be used to remove the salt, as the Na⁺ and Cl⁻ enter the beads, and the protein, now much less salty, falls through.

Electrophoresis

Electrophoresis is somewhat similar to chromatography with electricity. Compounds with charges on them will migrate in an electric field at a rate proportional to their charge density, i.e. their charge divided by their mass (small, highly charged things move quicker than heft, slightly charged things). Electrophoresis is widely used to separate proteins and DNA.

Polyacrylamide gel electrophoresis (PAGE) is usually used for proteins. In PAGE, we make a polyacrylamide gel from a mixture of acrylamide and a bis-acrylamide cross linker. Proteins are loaded onto the gel and move through the gel under the influence of an electric field. All things being equal, the protein will move according to its charge density, but in actual fact in most PAGE techniques, it is the size of the pores in the gel that have to more profound effect on the rate of protein migration.

In native PAGE, we just add protein directly to the gel and run it. Native PAGE is non-denaturing so we can test for protein activity afterwards, but it doesn't tell you much more than how pure your protein sample is. Proteins move in a complex way depending on their charge density and the size of the gel pores.

In SDS PAGE, we denature proteins in such a way as to make them migrate in a predictable fashion. SDS-PAGE is a denaturing technique. First proteins are unravelled by boiling, then any disulfide bonds are broken with mercaptoethanol. Hence what you add to the gel will be subunits and not the original protein. We then add the negatively charged detergent sodium dodecyl sulfate (SDS), which linearises the protein chains and coats the protein with negative charge (stoichiometrically: 1 SDS per 2 amino acids). This swamps the natural charge on the protein, so all proteins will have the same charge density.

Consequently, when we load the proteins, they will move according to how easily they can navigate through the pores in the gel alone: this is proportional to (the logarithm of) their size alone.

SDS-PAGE can be used to determine:

• Size of proteins.
• Presence of subunits.
• Monitoring chromatography eluates for purity.

DNA electrophoresis is usually done in agarose, a seaweed sugar used as a 2% gel in hot buffer.

A gel is cast, loaded and run in much the same way as PAGE, only usually horizontally. DNA and RNA are visualised by staining with ethidium bromide, which forms a fluorescent complex with nucleic acids. As before, DNA moves according to its charge density. The charge is largely supplied by DNA's phosphate groups, so the charge is roughly equal to the length of DNA, and hence its mass. This means that yet again, the charge density (charge/mass) of all nucleic acids is about the same, so separation is effected purely by the size of the pores in the gel again: long bits get snarled up and move very little, small bits move faster. There are various ways of improving resolution, such as pulsed field electrophoresis, where the polarity across the gel is rapidly reversed: this is useful for separating large bits of DNA.

Uses of DNA electrophoresis:

• Monitoring PCR products.
• Detecting sequences by blotting.
• Monitoring expression of genes.
• Gene sequencing.

Blotting is a technique for moving DNA, RNA or protein from an agarose or polyacrylamide electrophoresis gel to 'paper' for further analysis. In principle it is very simple:

• Apply current across faces of gel.
• Catch the molecules on nitrocellulose held to the face of the gel.

After blotting, we can stain the blot with a radioactive probe, etc. There are three sorts of blot:

• Southern blot for DNA. The DNA is blotted onto a membrane, and then soaked in a radioactive DNA probe. This binds to specific bands, and allows detection by autoradiography (i.e. put the membrane onto a photographic film and let the decay of the radioactive probe develop an image on the film).
• Northern blot for RNA. Basically the same as a Southern blot, but running RNA instead of DNA.
• Western blot for protein. Blot onto nitrocellulose, then use a radiolabelled (or fluorescent) antibody against your protein to detect its presence on the membrane.

Isoelectric focussing is another form of electrophoresis that allows separation of protein by their isoelectric points. The isoelectric point (pI) of a protein is the pH at which is has no net charge: since a protein bristles with many different ionisable groups, at low pH, it will tend towards being positively charged, and at high pH, towards being negatively charged. At the pI, the number of positive and negative charges balance, and the protein has no net overall charge. At this pH, the protein will be unaffected by an electric field. isoelectric focussing uses a special gel that creates a fixed pH gradient. A current is then passed through the gel. A protein migrates through the gel according to its charge, but as it does so, the pH changes, and so to does the charge on the protein. Eventually the protein will come to rest at the point in the pH gradient corresponding to its pI. The gel may be left like this, or perpendicular current may be applied afterwards to separate the proteins in a second dimension by conventional PAGE techniques.

Test yourself

1. Why is the R_t of butan-1-ol shorter than that of octan-1-ol on a reversed phase HPLC column?
2. Write a plan of action for a liquid-liquid extraction of cocaine (pK_a 8.6) from raw coca leaf extract.
3. Suggest a suitable chromatographic method for the purification of α-amanitin, a small toxic peptide from toadstools.
4. In which direction does DNA migrate in an agarose gel? If electrophoresis separates compounds based on their charge density, why does it separate DNA molecules at all?

Answers