Amino Acid Analysis
 
 The amino acid compositions of proteins are routinely determined by completely hydrolyzing the peptide bonds of the polypeptide chain, and then determining quantitatively the constituent amino acids that were released.
1. Peptide Bond Hydrolysis
The traditional method of hydrolyzing polypeptide chains has been to incubate them anaerobically in 6 M HCl at approximately 110°C for 24–72 h (1). More modern methods use other acids, higher temperatures, and shorter periods of time. Most peptide bonds hydrolyze at similar rates, but those between the large nonpolar amino acid residues, particularly Val, Leu, and Ile, are hydrolyzed more slowly and require longer hydrolysis times or the addition of organic acids such as trifluoroacetic acid. Hydrolysis is presumably hindered sterically by the bulky side chains.
Any chemical procedure that hydrolyzes the peptide bonds of the backbone will also hydrolyze the chemically similar amide side chains of Asn and Gln residues, to produce the amino acids aspartic acid and glutamic acid, respectively. It is feasible to measure the total number of Asn and Gln residues by measuring the amount of ammonia released during the hydrolysis, but otherwise it is not possible to distinguish between Asp and Asn and between Glu and Gln after hydrolysis of the polypeptide chain. In this case, it is common practice to designate such uncertain residues by the three-letter abbreviations Asx and Glx, and by the one-letter abbreviations B and Z, respectively.
Trp residues are usually destroyed completely by acid hydrolysis, probably as a result of reaction with chlorine produced by oxidation of the HCl. They can be protected by the addition of thiol or sulfonic acid compounds or of phenol to scavenge the chlorine (2). Tyr residues are also susceptible to chlorination, but they are usually lost only partially. The thiol groups of Cys residues are oxidized and the amino acid partially destroyed by acid hydrolysis; this residue is best analyzed after performic acid oxidation of the protein to convert all the Cys residues to cysteic acid.
Some of the problems with acid hydrolysis can be overcome by using other procedures, such as hydrolysis by alkali or by proteinases. Other amino acids, notably Ser and Thr, are destroyed by alkaline hydrolysis, however, and total proteinase digestion to amino acids is not straightforward. Consequently, acid hydrolysis remains in common use.
2. Quantifying the Amino Acids
The identities and quantities of the various amino acids present in a protein hydrolyzates are normally determined by automated amino acid analyzers. The amino acids are separated chromatographically and quantified as they emerge from the column. Traditional methods used ion-exchange chromatography of the free amino acids, followed by detection with ninhydrin or fluorescent reagents such as fluorescamine. Proline does not react in the usual manner with such reagents, due to absence of an amino group, so special procedures are required to measure it. More rapid and sensitive methods now predominate, in which the amino acids are reacted with suitable reagents prior to the chromatographic separation, rather than after. The favored method at present is to react the amino acids with phenylisothiocyate (see Edman Degradation) and then to separate the colored derivatives by reverse-phase chromatography. With this procedure, a complete quantitative amino acid analysis can be carried out in just a few minutes with only picomole quantities of amino acids (3).
The relative numbers of aromatic residues (Phe, Tyr, and Trp) in intact proteins and peptides can usually be determined from the UV absorbance spectrum under conditions in which the polypeptide chain is fully unfolded so that its spectrum is the sum of its constituent residues (4).
 Amino acid analysis does not give directly the number of residues of each amino acid per polypeptide chain. The most accurate result is the molar ratios of the various amino acids. The true molecular weight of the polypeptide chain, in the absence of any non-amino acid moieties, must be known for the amino acid analysis results to be converted to the number of residues of each amino acid per chain. Only with very accurate results, or with very small proteins, are such values usually close to the actual integer values. An alternative procedure is to use progressive chemical modification of one type of amino acid side chain for counting residues.
References
1. R. L. Hill (1965) Adv. Protein Chem. 20, 37–107. 
2. L. T. Ng et al. (1987) Anal. Biochem. 167, 47–52. 
3. S. A. Cohen and D. J. Strydom (1988) Anal. Biochem. 174, 1–16.
4. H. Edelhoch (1967) Biochemistry 6, 1948–1954.