C hapter 4

Amino Acids

Proteins are the indispensable agents of biological function, and amino acids are the building blocks of proteins. The stunning diversity of the thousands of proteins found in nature arises from the intrinsic properties of only 20 commonly occurring amino acids. These features include (1) the capacity to polymerize, (2) novel acid-base properties, (3) varied structure and chemical functionality in the amino acid side chains, and (4) chirality. This chapter describes each of these properties, laying a foundation for discussions of protein structure (Chapters 5 and 6), enzyme function (Chapters 14-16), and many other subjects in later chapters.

4.1 · Amino Acids: Building Blocks of Proteins

Structure of a Typical Amino Acid

Figure 4.1  •  Anatomy of an amino acid. Except for proline and its derivatives, all of the amino acids commonly found in proteins possess this type of structure.

The structure of a single typical amino acid is shown in Figure 4.1. Central to this structure is the tetrahedral alpha (a) carbon (C_a), which is covalently linked to both the amino group and the carboxyl group. Also bonded to this a-carbon is a hydrogen and a variable side chain. It is the side chain, the so-called R group, that gives each amino acid its identity. The detailed acid-base properties of amino acids are discussed in the following sections. It is sufficient for now to realize that, in neutral solution (pH 7), the carboxyl group exists as –COO^- and the amino group as –NH₃⁺. Because the resulting amino acid contains one positive and one negative charge, it is a neutral molecule called a zwitterion. Amino acids are also chiral molecules. With four different groups attached to it, the a-carbon is said to be asymmetric. The two possible configurations for the a-carbon constitute nonidentical mirror image isomers or enantiomers. Details of amino acid stereochemistry are discussed in Section 4.4.

Amino Acids Can Join via Peptide Bonds

The crucial feature of amino acids that allows them to polymerize to form peptides and proteins is the existence of their two identifying chemical groups: the amino (-NH₃⁺) and carboxyl (-COO^-) groups, as shown in Figure 4.2. The amino and carboxyl groups of amino acids can react in a head-to-tail fashion, eliminating a water molecule and forming a covalent amide linkage, which, in the case of peptides and proteins, is typically referred to as a peptide bond. The equilibrium for this reaction in aqueous solution favors peptide bond hydrolysis. For this reason, biological systems as well as peptide chemists in the laboratory must carry out peptide bond formation in an indirect manner or with energy input.

Figure 4.2  •  The a-COOH and a-NH₃⁺ groups of two amino acids can react with the resulting loss of a water molecule to form a covalent amide bond. (Irving Geis.)

Iteration of the reaction shown in Figure 4.2 produces polypeptides and proteins. The remarkable properties of proteins, which we shall discover and come to appreciate in later chapters , all depend in one way or another on the unique properties and chemical diversity of the 20 common amino acids found in proteins.

Common Amino Acids

(click icon to study the titration behavior of the amino acids)

Figure 4.3  • The 20 amino acids that are the building blocks of most proteins can be classified as (a) nonpolar (hydrophobic), (b) polar, neutral, (c) acidic, or (d) basic. Also shown are the one-letter and three-letter codes used to denote amino acids. For each amino acid, the ball-and-stick (left) and space-filling (right) models show only the side chain. (Irving Geis)

(click icon to test your knowledge of the amino acids)

(click any amino acid on the right to see 3D structure)

The structures and abbreviations for the 20 amino acids commonly found in proteins are shown in Figure 4.3. All the amino acids except proline have both free a-amino and free a-carboxyl groups (Figure 4.1). There are several ways to classify the common amino acids. The most useful of these classifications is based on the polarity of the side chains. Thus, the structures shown in Figure 4.3 are grouped into the following categories: (1) nonpolar or hydrophobic amino acids, (2) neutral (uncharged) but polar amino acids, (3) acidic amino acids (which have a net negative charge at pH 7.0), and (4) basic amino acids (which have a net positive charge at neutral pH). In later chapters, the importance of this classification system for predicting protein properties becomes clear. Also shown in Figure 4.3 are the three-letter and one-letter codes used to represent the amino acids. These codes are useful when displaying and comparing the sequences of proteins in shorthand form. (Note that several of the one-letter abbreviations are phonetic in origin: arginine = “ Rginine ” = R, phenylalanine = “ Fenylalanine ” = F, aspartic acid = “ asparDic ” = D.)

Nonpolar Amino Acids

The nonpolar amino acids (Figure 4.3a) include all those with alkyl chain R groups (alanine, valine, leucine, and isoleucine), as well as proline (with its unusual cyclic structure), methionine (one of the two sulfur-containing amino acids), and two aromatic amino acids, phenylalanine and tryptophan. Tryptophan is sometimes considered a borderline member of this group because it can interact favorably with water via the N-H moiety of the indole ring. Proline, strictly speaking, is not an amino acid but rather an a-imino acid.

Polar, Uncharged Amino Acids

The polar, uncharged amino acids (Figure 4.3b) except for glycine contain R groups that can form hydrogen bonds with water. Thus, these amino acids are usually more soluble in water than the nonpolar amino acids. Several exceptions should be noted. Tyrosine displays the lowest solubility in water of the 20 common amino acids (0.453 g/L at 25°C). Also, proline is very soluble in

(click icon to test your knowledge of the amino acids)


(click any amino acid on the right to see 3D structure)

water, and alanine and valine are about as soluble as arginine and serine. The amide groups of asparagine and glutamine; the hydroxyl groups of tyrosine, threonine, and serine; and the sulfhydryl group of cysteine are all good hydrogen bond-forming moieties. Glycine, the simplest amino acid, has only a single hydrogen for an R group, and this hydrogen is not a good hydrogen bond former. Glycine’s solubility properties are mainly influenced by its polar amino and carboxyl groups, and thus glycine is best considered a member of the polar, uncharged group. It should be noted that tyrosine has significant nonpolar characteristics due to its aromatic ring and could arguably be placed in the nonpolar group (Figure 4.3a). However, with a pK_a of 10.1, tyrosine’s phenolic hydroxyl is a charged, polar entity at high pH.

Acidic Amino Acids

There are two acidic amino acids—aspartic acid and glutamic acid—whose R groups contain a carboxyl group (Figure 4.3c). These side chain carboxyl groups are weaker acids than the a-COOH group, but are sufficiently acidic to exist as –COO^-  at neutral pH. Aspartic acid and glutamic acid thus have a net negative charge at pH 7. These negatively charged amino acids play several important roles in proteins. Many proteins that bind metal ions for structural or functional purposes possess metal binding sites containing one or more aspartate and glutamate side chains. Carboxyl groups may also act as nucleophiles in certain enzyme reactions and may participate in a variety of electrostatic bonding interactions. The acid-base chemistry of such groups is considered in detail in Section 4.2.

Basic Amino Acids

Three of the common amino acids have side chains with net positive charges at neutral pH: histidine, arginine, and lysine (Figure 4.3d). The ionized group of histidine is an imidazolium , that of arginine is a guanidinium, and lysine contains a protonated alkyl amino group. The side chains of the latter two amino acids are fully protonated at pH 7, but histidine, with a side chain pK_a of 6.0, is only 10% protonated at pH 7. With a pK_a near neutrality, histidine side chains play important roles as proton donors and acceptors in many enzyme reactions. Histidine-containing peptides are important biological buffers, as discussed in Chapter 2. Arginine and lysine side chains, which are protonated under physiological conditions, participate in electrostatic interactions in proteins.

Uncommon Amino Acids

(click on any amino acid to see 3D structure)

Figure 4.4   •  The structures of several amino acids that are less common but nevertheless found in certain proteins. Hydroxylysine and hydroxyproline are found in connective-tissue proteins, pyroglutamic acid is found in bacteriorhodopsin (a protein in Halobacterium halobium), and aminoadipic acid is found in proteins isolated from corn.

Several amino acids occur only rarely in proteins (Figure 4.4). These include hydroxylysine and hydroxyproline, which are found mainly in the collagen and gelatin proteins, and thyroxine and 3, 3´, 5-triiodothyronine, iodinated amino acids that are found only in thyroglobulin, a protein produced by the thyroid gland. (Thyroxine and 3, 3´, 5-triiodothyronine are produced by iodination of tyrosine residues in thyroglobulin in the thyroid gland. Degradation of thyroglobulin releases these two iodinated amino acids, which act as hormones to regulate growth and development.) Certain muscle proteins contain methylated amino acids, including methylhistidine, e-N-methyllysine, and e-N,N,N-trimethyllysine (Figure 4.4). g-Carboxyglutamic acid is found in several proteins involved in blood clotting, and pyroglutamic acid is found in a unique light-driven proton-pumping protein called bacteriorhodopsin , which is discussed elsewhere in this book. Certain proteins involved in cell growth and regulation are reversibly phosphorylated on the -OH groups of serine, threonine , and tyrosine residues. Aminoadipic acid is found in proteins isolated from corn. Finally, N-methylarginine and N-acetyllysine are found in histone proteins associated with chromosomes.

Amino Acids Not Found in Proteins

Certain amino acids and their derivatives, although not found in proteins, nonetheless are biochemically important. A few of the more notable examples are shown in Figure 4.5. g-Aminobutyric acid, or GABA, is produced by the decarboxylation of glutamic acid and is a potent neurotransmitter. Histamine, which is synthesized by decarboxylation of histidine, and serotonin, which is derived from tryptophan, similarly function as neurotransmitters and regulators. b-Alanine is found in nature in the peptides carnosine and anserine and is a component of pantothenic acid (a vitamin), which is a part of coenzyme A. Epinephrine (also known as adrenaline), derived from tyrosine, is an important hormone. Penicillamine is a constituent of the penicillin antibiotics. Ornithine, betaine, homocysteine, and homoserine are important metabolic intermediates. Citrulline is the immediate precursor of arginine.

(click on any amino acid to see 3D structure)

Figure 4.5  •  The structures of some amino acids that are not normally found in proteins but that perform other important biological functions. Epinephrine, histamine, and serotonin, although not amino acids, are derived from and closely related to amino acids.

4.2 · Acid-Base Chemistry of Amino Acids

Amino Acids Are Weak Polyprotic Acids

From a chemical point of view, the common amino acids are all weak polyprotic acids. The ionizable groups are not strongly dissociating ones, and the degree of dissociation thus depends on the pH of the medium. All the amino acids contain at least two dissociable hydrogens.

            Consider the acid-base behavior of glycine, the simplest amino acid. At low pH, both the amino and carboxyl groups are protonated and the molecule has a net positive charge. If the counterion in solution is a chloride ion, this form is referred to as glycine hydrochloride. If the pH is increased, the carboxyl group is the first to dissociate, yielding the neutral zwitterionic species Gly⁰ (Figure 4.6).

Figure 4.6  • The ionic forms of the amino acids, shown without consideration of any ionizations on the side chain. The cationic form is the low pH form, and the titration of the cationic species with base yields the zwitterion and finally the anionic form. (Irving Geis)

Further increase in pH eventually results in dissociation of the amino group to yield the negatively charged glycinate. If we denote these three forms as Gly⁺, Gly⁰, and Gly^-, we can write the first dissociation of Gly⁺ as

and the dissociation constant K₁ as

Values for K₁ for the common amino acids are typically 0.4 to 1.0 x 10^-2 M, so that typical values of pK₁ center on values of 2.0 to 2.4 (see Table 4.1). In a similar manner, we can write the second dissociation reaction as

and the dissociation constant K₂ as

Typical values for pK₂ are in the range of 9.0 to 9.8. At physiological pH, the a-carboxyl group of a simple amino acid (with no ionizable side chains) is completely dissociated, whereas the a-amino group has not really begun its dissociation. The titration curve for such an amino acid is shown in Figure 4.7.

Figure 4.7  •  Titration of glycine, a simple amino acid. The isoelectric point, pI, the pH where the molecule has a net charge of 0, is defined as (pK₁+ pK₂)/2.

Example

What is the pH of a glycine solution in which the a-NH₃⁺ group is one-third dissociated?

Solution

The appropriate Henderson- Hasselbalch equation is

If the a-amino group is one-third dissociated, there is one part Gly^- for every two parts Gly⁰. The important pK_a is the pK_a for the amino group. The glycine a-amino group has a pK_a of 9.6. The result is

            pH  = 9.6 + log₁₀ (1/2)

pH = 9.3

Note that the dissociation constants of both the a-carboxyl and a-amino groups are affected by the presence of the other group. The adjacent a-amino group makes the a-COOH group more acidic (that is, it lowers the pK_a ) so that it gives up a proton more readily than simple alkyl carboxylic acids. Thus, the pK₁ of 2.0 to 2.1 for a-carboxyl groups of amino acids is substantially lower than that of acetic acid ( pK_{a =} 4.76), for example. What is the chemical basis for the low pK_a of the a-COOH group of amino acids? The a-NH₃⁺ (ammonium) group is strongly electron-withdrawing, and the positive charge of the amino group exerts a strong field effect and stabilizes the carboxylate anion. (The effect of the a-COO^- group on the pK_a of the a-NH₃⁺ group is the basis for Problem 4 at the end of this chapter.)

Ionization of Side Chains

As we have seen, the side chains of several of the amino acids also contain dissociable groups. Thus, aspartic and glutamic acids contain an additional carboxyl function, and lysine possesses an aliphatic amino function. Histidine contains an ionizable imidazolium proton, and arginine carries a guanidinium function. Typical pK_a values of these groups are shown in Table 4.1. The b-carboxyl group of aspartic acid and the g-carboxyl side chain of glutamic acid exhibit pK_a values intermediate to the a-COOH on the one hand and typical aliphatic carboxyl groups on the other hand. In a similar fashion, the e-amino group of lysine exhibits a pK_a that is higher than the a-amino group but similar to that for a typical aliphatic amino group. These intermediate values for side-chain pK_a values reflect the slightly diminished effect of the a-carbon dissociable groups that lie several carbons removed from the side-chain functional groups. Figure 4.8 shows typical titration curves for glutamic acid and lysine, along with the ionic species that predominate at various points in the titration. The only other side-chain groups that exhibit any significant degree of dissociation are the para-OH group of tyrosine and the -SH group of cysteine. The pK_a of the cysteine sulfhydryl is 8.32, so that it is about 12% dissociated at pH 7. The tyrosine para-OH group is a very weakly acidic group, with a pK_a of about 10.1. This group is essentially fully protonated and uncharged at pH 7.

                                                     
Figure 4.8  •  Titrations of glutamic acid and lysine.

4.3 · Reactions of Amino Acids

Carboxyl and Amino Group Reactions

The a-carboxyl and a-amino groups of all amino acids exhibit similar chemical reactivity. The side chains, however, exhibit specific chemical reactivities, depending on the nature of the functional groups. Whereas all of these reactivities are important in the study and analysis of isolated amino acids, it is the characteristic behavior of the side chain that governs the reactivity of amino acids incorporated into proteins. There are three reasons to consider these reactivities. Proteins can be chemically modified in very specific ways by taking advantage of the chemical reactivity of certain amino acid side chains. The detection and quantification of amino acids and proteins often depend on reactions that are specific to one or more amino acids and that result in color, radioactivity, or some other quantity that can be easily measured. Finally and most importantly, the biological functions of proteins depend on the behavior and reactivity of specific R groups.

Figure 4.9  •   Typical reactions of the common amino acids (see text for details).

            The carboxyl groups of amino acids undergo all the simple reactions common to this functional group. Reaction with ammonia and primary amines yields unsubstituted and substituted amides, respectively (Figure 4.9a,b). Esters and acid chlorides are also readily formed. Esterification proceeds in the presence of the appropriate alcohol and a strong acid (Figure 4.9c). Polymerization can occur by repetition of the reaction shown in Figure 4.9d. Free amino groups may react with aldehydes to form Schiff bases (Figure 4.9e) and can be acylated with acid anhydrides and acid halides (Figure 4.9f).

The Ninhydrin Reaction

Amino acids can be readily detected and quantified by reaction with ninhydrin. As shown in Figure 4.10, ninhydrin, or triketohydrindene hydrate, is a strong oxidizing agent and causes the oxidative deamination of the a-amino function. The products of the reaction are the resulting aldehyde, ammonia, carbon dioxide, and hydrindantin, a reduced derivative of ninhydrin . The ammonia produced in this way can react with the hydrindantin and another molecule of ninhydrin to yield a purple product ( Ruhemann’s Purple) that can be quantified spectrophotometrically at 570 nm. The appearance of CO₂ can also be monitored. Indeed, CO₂ evolution is diagnostic of the presence of an a-amino acid. a-Imino acids, such as proline and hydroxyproline, give bright yellow ninhydrin products with absorption maxima at 440 nm, allowing these to be distinguished from the a-amino acids. Because amino acids are one of the components of human skin secretions, the ninhydrin reaction was once used extensively by law enforcement and forensic personnel for fingerprint detection. (Fingerprints as old as 15 years can be successfully identified using the ninhydrin reaction.) More sensitive fluorescent reagents are now used routinely for this purpose.

Figure 4.10  •  The pathway of the ninhydrin reaction, which produces a colored product called “Ruhemann’s Purple” that absorbs light at 570 nm. Note that the reaction involves and consumes two molecules of ninhydrin.

Specific Reactions of Amino Acid Side Chains

A number of reactions of amino acids have become important in recent years because they are essential to the degradation, sequencing, and chemical synthesis of peptides and proteins. These reactions are discussed in Chapter 5.

            In recent years, biochemists have developed an arsenal of reactions that are relatively specific to the side chains of particular amino acids. These reactions can be used to identify functional amino acids at the active sites of enzymes or to label proteins with appropriate reagents for further study. Cysteine residues in proteins, for example, react with one another to form disulfide species and also react with a number of reagents, including maleimides (typically N-ethylmaleimide), as shown in Figure 4.11.

(Click each section to see the corresponding animation)

Figure 4.11  • Reactions of amino acid side-chain functional groups.

Cysteines also react effectively with iodoacetic acid to yield S- carboxymethyl cysteine derivatives. There are numerous other reactions involving specialized reagents specific for particular side chain functional groups. Figure 4.11 presents a representative list of these reagents and the products that result. It is important to realize that few if any of these reactions are truly specific for one functional group; consequently, care must be exercised in their use.

4.4 · Optical Activity and Stereochemistry of Amino Acids

Amino Acids Are Chiral Molecules

Except for glycine, all of the amino acids isolated from proteins have four different groups attached to the a-carbon atom. In such a case, the a-carbon is said to be asymmetric or chiral from the Greek cheir, meaning “hand”), and the two possible configurations for the a-carbon constitute nonsuperimposable mirror image isomers, or enantiomers (Figure 4.12).

Figure 4.12  •  Enantiomeric molecules based on a chiral carbon atom. Enantiomers
are nonsuperimposable mirror images of each other.

Enantiomeric molecules display a special property called optical activity—the ability to rotate the plane of polarization of plane-polarized light. Clockwise rotation of incident light is referred to as dextrorotatory behavior, and counterclockwise rotation is called levorotatory behavior. The magnitude and direction of the optical rotation depend on the nature of the amino acid side chain. The temperature, the wavelength of the light used in the measurement, the ionization state of the amino acid, and therefore the pH of the solution, can also affect optical rotation behavior. As shown in Table 4.2, some protein-derived amino acids at a given pH are dextrorotatory and others are levorotatory, even though all of them are of the L configuration. The direction of optical rotation can be specified in the name by using a (+) for dextrorotatory compounds and a(-) for levorotatory compounds, as in L(+)- leucine.

Nomenclature for Chiral Molecules

The discoveries of optical activity and enantiomeric structures (see the box, page 97) made it important to develop suitable nomenclature for chiral molecules. Two systems are in common use today: the so-called D, L system and the (R,S) system.

            In the D,_L system of nomenclature, the (+) and (-) isomers of glyceraldehyde are denoted as D-glyceraldehyde and L-glyceraldehyde, respectively (Figure 4.13). Absolute configurations of all other carbon-based molecules are referenced to D- and L-glyceraldehyde . When sufficient care is taken to avoid racemization of the amino acids during hydrolysis of proteins, it is found that all of the amino acids derived from natural proteins are of the L configuration. Amino acids of the D configuration are nonetheless found in nature, especially as components of certain peptide antibiotics, such as valinomycin, gramicidin, and actinomycin D, and in the cell walls of certain microorganisms.

Figure 4.13  •  The configuration of the common L-amino acids can be related to the configuration of L(- )-glyceraldehyde as shown. These drawings are known as Fischer projections. The horizontal lines of the Fischer projections are meant to indicate bonds coming out of the page from the central carbon, and vertical lines represent bonds extending behind the page from the central carbon atom.

            In spite of its widespread acceptance, problems exist with the D,L system of nomenclature. For example, this system can be ambiguous for molecules with two or more chiral centers. To address such problems, the (R,S) system of nomenclature for chiral molecules was proposed in 1956 by Robert Cahn, Sir Christopher Ingold, and Vladimir Prelog. In this more versatile system, priorities are assigned to each of the groups attached to a chiral center on the basis of atomic number, atoms with higher atomic numbers having higher priorities (see the box, page 99).

            The newer (R,S) system of nomenclature is superior to the older D,L system in one important way. The configuration of molecules with more than one chiral center can be more easily, completely, and unambiguously described with (R,S) notation. Several amino acids, including isoleucine, threonine, hydroxyproline, and hydroxylysine, have two chiral centers. In the (R,S) system, L-threonine is (2S,3R)-threonine. A chemical compound with n chiral centers can exist in 2ⁿ-isomeric structures, and the four amino acids just listed can thus each take on four different isomeric configurations. This amounts to two pairs of enantiomers. Isomers that differ in configuration at only one of the asymmetric centers are non-mirror image isomers or diastereomers. The four stereoisomers of isoleucine are shown in Figure 4.14. The isomer obtained from digests of natural proteins is arbitrarily designated L-isoleucine. In the (R,S) system, L-isoleucine is (2S,3S)-isoleucine. Its diastereomer is referred to as L-alloisoleucine . The D-enantiomeric pair of isomers is named in a similar manner.

Figure 4.14  •  The stereoisomers of isoleucine and threonine. The structures at the far left are the naturally occurring isomers.

4.5 · Spectroscopic Properties of Amino Acids

One of the most important and exciting advances in modern biochemistry has been the application of spectroscopic methods, which measure the absorption and emission of energy of different frequencies by molecules and atoms. Spectroscopic studies of proteins, nucleic acids, and other biomolecules are providing many new insights into the structure and dynamic processes in these molecules.

Ultraviolet Spectra

Many details of the structure and chemistry of the amino acids have been elucidated or at least confirmed by spectroscopic measurements. None of the amino acids absorbs light in the visible region of the electromagnetic spectrum. Several of the amino acids, however, do absorb ultraviolet radiation, and all absorb in the infrared region. The absorption of energy by electrons as they rise to higher energy states occurs in the ultraviolet/visible region of the energy spectrum. Only the aromatic amino acids phenylalanine, tyrosine, and tryptophan exhibit significant ultraviolet absorption above 250 nm, as shown in Figure 4.15. These strong absorptions can be used for spectroscopic determinations of protein concentration. The aromatic amino acids also exhibit relatively weak fluorescence, and it has recently been shown that tryptophan can exhibit phosphorescence—a relatively long-lived emission of light. These fluorescence and phosphorescence properties are especially useful in the study of protein structure and dynamics (see Chapter 6).

Figure 4.15  •  The ultraviolet absorption spectra of the aromatic amino acids at pH 6. (From Wetlaufer, D.B., 1962. Ultraviolet spectra of proteins and amino acids. Advances in Protein Chemistry 17:303–390.)

Nuclear Magnetic Resonance Spectra

The development in the 1950s of nuclear magnetic resonance (NMR), a spectroscopic technique that involves the absorption of radio frequency energy by certain nuclei in the presence of a magnetic field, played an important part in the chemical characterization of amino acids and proteins. Several important principles rapidly emerged from these studies. First, the chemical shift¹ of amino acid protons depends on their particular chemical environment and thus on the state of ionization of the amino acid. Second, the change in electron density during a titration is transmitted throughout the carbon chain in the aliphatic amino acids and the aliphatic portions of aromatic amino acids, as evidenced by changes in the chemical shifts of relevant protons. Finally, the magnitude of the coupling constants between protons on adjacent carbons depends in some cases on the ionization state of the amino acid. This apparently reflects differences in the preferred conformations in different ionization states. Proton NMR spectra of two amino acids are shown in Figure 4.16.

Figure 4.16  •  Proton NMR spectra of several amino acids. Zero on the chemical shift scale is defined by the resonance of tetramethylsilane (TMS). (Adapted from Aldrich Library of NMR Spectra.)

Because they are highly sensitive to their environment, the chemical shifts of individual NMR signals can detect the pH-dependent ionizations of amino acids. Figure 4.17 shows the ¹³C chemical shifts occurring in a titration of lysine. Note that the chemical shifts of the carboxyl C, C_a, and C_b carbons of lysine are sensitive to dissociation of the nearby a-COOH and a-NH₃⁺ protons (with pK_a values of about 2 and 9, respectively), whereas the C_d and C_e carbons are sensitive to dissociation of the e-NH₃⁺ group. Such measurements have been very useful for studies of the ionization behavior of amino acid residues in proteins. More sophisticated NMR measurements at very high magnetic fields are also used to determine the three-dimensional structures of peptides and even small proteins.

Figure 4.17 •   A plot of chemical shifts versus pH for the carbons of lysine. Changes in chemical shift are most pronounced for atoms near the titrating groups. Note the correspondence between the pK_a values and the particular chemical shift changes. All chemical shifts are defined relative to tetramethylsilane (TMS). (From Suprenant, H., et al., 1980. Journal of Magnetic Resonance 40:231–243.)

4.6 · Separation and Analysis of Amino Acid Mixtures

Chromatographic Methods

The purification and analysis of individual amino acids from complex mixtures was once a very difficult process. Today, however, the biochemist has a wide variety of methods available for the separation and analysis of amino acids, or for that matter, any of the other biological molecules and macromolecules we encounter. All of these methods take advantage of the relative differences in the physical and chemical characteristics of amino acids, particularly ionization behavior and solubility characteristics. The methods important for amino acids include separations based on partition properties (the tendency to associate with one solvent or phase over another) and separations based on electrical charge. In all of the partition methods discussed here, the molecules of interest are allowed (or forced) to flow through a medium consisting of two phases—solid-liquid, liquid-liquid, or gas-liquid. In all of these methods, the molecules must show a preference for associating with one or the other phase. In this manner, the molecules partition, or distribute themselves, between the two phases in a manner based on their particular properties. The ratio of the concentrations of the amino acid (or other species) in the two phases is designated the partition coefficient.
            In 1903, a separation technique based on repeated partitioning between phases was developed by Mikhail Tswett for the separation of plant pigments (carotenes and chlorophylls). Tswett, a Russian botanist, poured solutions of the pigments through columns of finely divided alumina and other solid media, allowing the pigments to partition between the liquid solvent and the solid support. Owing to the colorful nature of the pigments thus separated, Tswett called his technique chromatography. This term is now applied to a wide variety of separation methods, regardless of whether the products are colored or not. The success of all chromatography techniques depends on the repeated microscopic partitioning of a solute mixture between the available phases. The more frequently this partitioning can be made to occur within a given time span or over a given volume, the more efficient is the resulting separation. Chromatographic methods have advanced rapidly in recent years, due in part to the development of sophisticated new solid-phase materials. Methods important for amino acid separations include ion exchange chromatography, gas chromatography (GC), and high-performance liquid chromatography (HPLC).

Ion Exchange Chromatography

The separation of amino acids and other solutes is often achieved by means of ion exchange chromatography, in which the molecule of interest is exchanged for another ion onto and off of a charged solid support. In a typical procedure, solutes in a liquid phase, usually water, are passed through columns filled with a porous solid phase, usually a bed of synthetic resin particles, containing charged groups. Resins containing positive charges attract negatively charged solutes and are referred to as anion exchangers. Solid supports possessing negative charges attract positively charged species and are referred to as cation exchangers. Several typical cation and anion exchange resins with different types of charged groups are shown in Figure 4.18.

Figure 4.18  •   Cation (a) and anion (b) exchange resins commonly used for biochemical separations.

The strength of the acidity or basicity of these groups and their number per unit volume of resin determine the type and strength of binding of an exchanger. Fully ionized acidic groups such as sulfonic acids result in an exchanger with a negative charge which binds cations very strongly. Weakly acidic or basic groups yield resins whose charge (and binding capacity) depends on the pH of the eluting solvent. The choice of the appropriate resin depends on the strength of binding desired. The bare charges on such solid phases must be counterbalanced by oppositely charged ions in solution (“counterions”). Washing a cation exchange resin, such as Dowex-50, which has strongly acidic phenyl-SO₃^- groups, with a NaCl solution results in the formation of the so-called sodium form of the resin (see Figure 4.19). When the mixture whose separation is desired is added to the column, the positively charged solute molecules displace the Na⁺ ions and bind to the resin. A gradient of an appropriate salt is then applied to the column, and the solute molecules are competitively (and sequentially) displaced (eluted) from the column by the rising concentration of cations in the gradient, in an order that is inversely related to their affinities for the column. The separation of a mixture of amino acids on such a column is shown in Figures 4.19 and 4.20.

Figure 4.19 •   Operation of a cation exchange column, separating a mixture of Asp, Ser, and Lys. (a) The cation exchange resin in the beginning, Na⁺ form. (b) A mixture of Asp, Ser, and Lys is added to the column containing the resin. (c) A gradient of the eluting salt (e.g., NaCl) is added to the column. Asp, the least positively charged amino acid, is eluted first. (d) As the salt concentration increases, Ser is eluted. (e) As the salt concentration is increased further, Lys, the most positively charged of the three amino acids, is eluted last.

Figure 4.20 •  The separation of amino acids on a cation exchange column.

Figure 4.21, taken from a now-classic 1958 paper by Stanford Moore, Darrel Spackman, and William Stein, shows a typical separation of the common amino acids. The events occurring in this separation are essentially those depicted in Figures 4.19 and 4.20. The amino acids are applied to the column at low pH (4.25), under which conditions the acidic amino acids (aspartate and glutamate, among others) are weakly bound and the basic amino acids, such as arginine and lysine, are tightly bound. Sodium citrate solutions, at two different concentrations and three different values of pH, are used to elute the amino acids gradually from the column.

Figure 4.21 •  Chromatographic fractionation of a synthetic mixture of amino acids on ion exchange columns using Amberlite IR-120, a sulfonated polystyrene resin similar to Dowex-50. A second column with different buffer conditions is used to resolve the basic amino acids. ( Adapted from Moore, S., Spackman, D., and Stein, W., 1958. Chromatography of amino acids on sulfonated polystyrene resins. Analytical Chemistry 30:1185–1190.)         

A typical HPLC chromatogram using precolumn derivatization of amino acids with o-phthaldialdehyde (OPA) is shown in Figure 4.22. HPLC has rapidly become the chromatographic technique of choice for most modern biochemists. The very high resolution, excellent sensitivity, and high speed of this technique usually outweigh the disadvantage of relatively low capacity.

Figure 4.22 •  HPLC chromatogram of amino acids employing precolumn derivatization with OPA. Chromatography was carried out on an Ultrasphere ODS column using a complex tetrahydrofuran:methanol:0.05 M sodium acetate (pH 5.9) 1:19:80 to methanol:0.05 M sodium acetate (pH 5.9) 4:1 gradient at a flow rate of 1.7 mL/min. (Adapted from Jones, B. N., Pääbo, S., and Stein, S., 1981. Amino acid analysis and enzymic sequence determination of peptides by an improved o-phthaldialdehyde precolumn labeling procedure. Journal of Liquid Chromatography 4:56–586.)