Returning to the structure of HIV Protease, we find that many of the carboxylate side groups (RCO2-) are closely paired with ammonium side groups (RNH3+), as shown in figure 2.??. We expect that two charged groups will have a strong electrostatic attraction. At a distance of 2.0 and in the absence of any intervening atoms, the attraction of a +1 charge to a -1 charge yields a stabilization (indicated by the negative sign of the energy) of about 191 kJ/mole (or 46 kcal/mol). These numbers derive from Coulombs law.
Essentially, Coulomb's law states that the effect of charge is inversely proportional to distance. In other words, nearly 200 kJ/mole are required to separate the opposite charges to an infinite distance. This energy is about half of the strength of a C-C covalent bond. We must be cautious in these estimates, however, because the charges in a carboxylate anion and an ammonium cation are spread out over many atoms. Such delocalization will decrease the magnitude of the electrostatic attraction by increasing the average separation of the charges. Nevertheless, it is clear that electrostatic attractions, as well as repulsions, can supply substantial forces that impact protein structures.
The presence of charged groups has another less direct influence on the folding of HIV Protease. Charged groups interact strongly with polar molecules such as water. We can characterize this interaction as being a charge-dipole interaction. Consider the ammonium group in an aqueous solution. Maximum stablization of the ammonium group will occur when surrounded by water molecules oriented with their dipoles pointing toward the positively charged ammonium, as shown in Figure 2.22.
Figure 2.22. Water stabilized ammonium ion
If the protein cannot fold to place an ammonium group near to a carboxylate, then it will be most favorable for the charged groups to be located on the water accessible surfaces of the protein. Thus we expect that proteins will fold in such a way that charged and highly polar functional groups are on the water accessible exterior surfaces. Such groups are called hydrophilic (Greek for water loving). Similarly, nonpolar groups (the hydrophobic, or water hating, groups with high hydrocarbon content) will tend to aggregate in the oily interior of the protein. Refer back to Table 2.1 and note that the amino acids where R = alkyl and aryl are hydrophobic while amino acids with basic, acidic, alcohol, and thiol functional groups are hydrophilic. All of these features are revealed in the structures of HIV Protease as shown below.
Figure 2.23 Color coded structures of HIV Protease showing aggregation of polor side qroups (shown in green) on the protein exterior.
The overall folding of a protein is referred to as its tertiary structure. Because the ionization of carboxylates and amines is a critical element in stabilizing the tertiary structure of proteins and because the ionization occurs through loss or gain of a proton, we might expect that the structure of a protein is dependend on the pH of the solution. Indeed this is observed, proteins commonly have a well defined structure at near physiological pH but unfold to random structures as the pH is either increased or decreased significantly. For example, as the pH is decreased carboxylate functions will be protonated to form free acids, which are not charged and cannot engage in salt bridge formation. In removing the structural driving force of the salt bridges, many proteins will simply unfold to a disordered state. This allows us to understand why the ability of organisms such as fish to survive is so strongly affected by the pH of their envirionment.
2.10 Carboxylic Acids are Critical for the Activity of HIV Protease
The acidity of the O-H bond in carboxylic acids plays a crucial role in the mechanism of HIV Protease. The two amino acids in the active site of HIV Protease that are responsible for cleaving the amide bond in the substrate are both aspartic residues; aspartic acid is a carboxylic acid (Table 2.1). In the active form of HIV Protease, one of the carboxylic acids is protonated and the other is present as the carboxylate anion. As the reaction of HIV Protease with its substrate progresses, the neutral carboxylic acid becomes deprotonated and the carboxylate anion gains a proton to become neutral. The mechanism of HIV Protease will be presented later; for now, it is sufficient to understand that the carboxylic acid functional group is ideally suited for HIV Protease because the carboxylic acid is acidic and the carboxylate anion is stable.
Figure 2.24. HIV Protease with active site aspartic acid groups highlighted
2.11 Carboxylic Acids Form Strong Intermolecular Bonds
A distinctive feature of many carbonyl-containing compounds is that their boiling points are significantly higher than hydrocarbons of similar mass. Examine the table shown below and it is obvious that the presence of the carbonyl markedly affects the boiling points. A closer look reveals that amides and carboxylic acids have particularly high boiling points. Although boiling points may seem to have little to do with protein structure, in fact the physical interactions underlying these properties have much in common. We have already seen the interactions between functional groups on proteins strongly impact the structure. Boiling points, in part, are a measure of the strength of intermolecular interactions. For example, the boiling points of carboxylic acids relative to hydrocarbons reveal something about strength of the intermolecular forces.
Table 2.6. Boiling point comparisons for molecules of similar mass.
| Carbonyl Compound | Alkene |
|---|---|
 |
 |
 |
 |
 |
 |
Simple carbonyl containing compounds are polar by virtue of the polarity of the carbonyl. Intermolecular dipole-dipole interactions lead to stabilizing forces, hence higher boiling points. As illustrated below, molecules will tend to align so that maximum attraction between dipoles is achieved. This occurs when the neighboring molecules have their dipoles anti-parallel, or the positive end of one dipole lies close to the negative end of another dipole. However, notice how acetic acid and acetamide have substantially higher boiling points than the related ketone, 2-butanone, and ester, methyl acetate. Obviously there is more occurring in acetic acid and acetamide than simple dipole-dipole intermolecular forces.
Figure 2.25. Dipole-Dipole Interactions between Acetone Molecules
Close inspection of the structure of a carboxylic reveals a distinctive capability for dimerization via complementary hydrogen bonding. The hydrogen donor is typically H attached to either an O (i.e,. water, alcohols, and Br¿nsted acids), a N (amines, amides, and ammonium groups), or F (hydrofluoric acid). The hydrogen bond acceptors are lone pairs on O (i.e., water, alcohols, ethers, and all carbonyls), N (amine, C=N), and F (generally HF or F-). As a carboxylic acid contains both a hydrogen donor (the O-H) and a hydrogen bond acceptor (the carbonyl O) function, two hydrogen bonds may be formed by dimerization to form a weakly bound six-membered ring. Indeed, it is found that in the solid, liquid, and moderate pressure gas phases most carboxylic acids undergo mutual hydrogen bonding (Figure 2.26).
Figure 2.26. Dimerization of acetic acid via Hydrogen Bonding.
In the solid state, a more complex mode of hydrogen bonding occurs that leads to the formation of extended arrays of hydrogen bond networks as shown below. Similar arrays are commonly seen for amides as well.
Figure 2.27 Extended arrays of H-bonds in solid Carboxylic acids and amides.
2.12 How Do We Know that Hydrogen Bonds Form?
Hydrogen bonds are generally identified using two sets of criteria: geometric criteria and energetic criteria. In carbonyl compounds, the C=O---H angle is ideally 120o, and hydrogen bonds in the plane of the C=O bond are in the preferred geometry. Energetically, hydrogen bonds are favored if the equilibrium between bonded and non-bonded systems lies such that the hydrogen bonded system is lower in energy. Methods of detecting hydrogen bonds usually focus on one of the two criteria.
The most common method of detecting hydrogen bonds is by X-Ray Diffraction Crystallography. As you may recall from a previous chemistry course, X-ray diffraction is a method for locating the positions of atoms in a crystalline substance. Once the positions of the atoms in the structure have been located, then the distance A-H---B is examined, where B is potential hydrogen bond acceptor and A-H is the donor functionality. If the distance between H and B is significantly smaller than the sum of the van der Waals radii for the two atoms, then a hydrogen bond is present. For example, the van der Waals radius of O is around 1.4 and that of H is about 1.2 yielding a sum of 2.6. In molecules making hydrogen bonds, noncovalent O-H separation in the range of 1.6 to 1.8 are common.
Vibrational spectroscopy (infrared or raman) is a direct method for detecting hydrogen bonds. In infrared spectroscopy, the A-H (OH or NH) stretch undergoes several key changes upon hydrogen bond formation. First, the position of the maximum absorbance is shifted to lower wavenumbers. The non hydrogen bonded OH or NH peak is at about 3460 cm-1 while the hydrgoen bonded OH or NH peak is shifted to about 3320 cm-1. Second, the width of the absorbance is larger in the hydrogen bonded compound than for non hydrogen bonded material. Lastly, the intensity of the absorbance increases. This is because intensity is proportional to the change in the dipole upon stretching. Since the hydrogen bonded H is more polarized, there is a greater change in the dipole moment upon stretching and the peak is larger.
Nuclear Magnetic Resonance (NMR) spectroscopy and calorimetry have also been used to detect hydrogen bonds. Changes in chemical shift in the 1H NMR spectrum of some systems can be attributed to hydrogen bonding. The evolution of heat (measured by calorimetry) as hydrogen bonding partners are mixed has been correlated to hydrogen bonding as well.
2.13 Helices and Sheets in HIV Protease
Let's return to our ribbon structures of HIV Protease. You may notice that some of the ribbon sections are coiled like springs. Less obvious, but present nonetheless, are regions is which the ribbons run parallel to one another. These structural motifs, which are given the names a-helix and b-pleated sheet, are highlighted in the structures shown below. The twisting and aligning of a strand of amino acids to form and a-helices and b-pleated sheets is called the secondary structure of a protein.
Figure 2.28. Structures of HIV Protease with the regions of [[alpha]]-helix colored blue and the [[beta]]-pleated sheet regions colored red.
Such well defined structures are not likely to occur randomly; what are the forces that form the helical and sheet regions of proteins? It was first pointed out by Linus Pauling in the early 1950's that such regions could result from intramolecular hydrogen bonding in proteins. Proteins are polyamides (also called polypeptides). Because amides have both hydrogen donor (the N-H bond) and acceptor (the carbonyl O) functionalities, everything needed to form many hydrogen bonds is present in a single strand of peptide. A closer look at the helical and sheet regions of HIV protease reveals the networks of hydrogen bonds that hold these structures together. Although each hydrogen bond yields just 12-20 kJ/mol of stabilization, the presence of many of these interactions provides a strong driving force for these strands to form local regions of helical and sheeted structures.
Coiling into an a-helix is driven by the formation of hydrogen bonds between amides on neighboring turns of the coil. For most helical proteins maximization of hydrogen bond stabilization and minimization of repulsive steric interactions occurs when the carbonyl of the nth amino acid is an acceptor for the N-H bond of the n+4th amino acid. This hydrogen bonding pattern leads to helices in which each turn is separated by about 5.4. The a-helix is like a spiral staircase without a center pole. Each amino acid is a step and hydrogen bonds connect it weakly to the amino acids four steps above and four steps below. Although each hydrogen bond is weak and the formation of such an ordered structure is entropically disfavored, there are enough hydrogen bonds to make the coiled structure of the a-helix a very common structural motif.
Figure 2.29. Close up of [[alpha]]-helix in HIV Protease showing hydrogen bonding.
The b-pleated sheet also is stabilized by hydrogen bonding. In the b-pleated sheet hydrogen bonds are formed between two sections of a strand of protein that run parallel to one another. The hydrogen bonds are arranged like the ties in a railroad track; the rails correspond to parallel sections of the protein strands and the ties are the hydrogen bonds made by the amide functional groups.
Figure 2.30. Close up of [[beta]]-pleated sheet in HIV Protease showing hydrogen bonding.
