Probing Enzyme Phosphoester Interactions by Combining Mutagenesis and Chemical Modification of Phosphate Ester Oxygens

1.1 Importance of the Phosphoester Handle in Biological Systems Proteins, nucleic acids, lipids and various other smaller biomolecules are united by the presence of a common chemical entity of paramount and unique importance—the phosphate ester. In proteins, phosphomonoester groups are added as reversible post-translational modifications to serine, threonine, or tyrosine amino acid side chains, and serve as a switch for intracellular signaling cascades (phosphoserine, phosphothreonine or phosphotyrosine)1. Alternatively, phosphoamide linkages involving the nitrogens of histidine or lysine side chains serve as high energy phosphoryl donors in multistep phosphoryl transfer reactions (Fig. 1A)2. In nucleic acids, the phosphodiester linkage provides the exceptionally stable polymeric backbone upon which the coding information of the nucleic acid bases are carried (Fig. 1B). In this regard, the biological selection of phosphate ester linkages for long term information storage in DNA may in part reside in their resistance to hydrolysis (t1/2 = 30 million years)3, and the requirement for a charged linkage that promotes aqueous solubility and deters aggregation. In phospholipids, the negatively charged phosphate monoester head group imbues phospholipid bilayers with their aqueous affinity that is crucial for forming functional cell membranes (Fig. 1C). Finally, phosphorylated polyalcohols such as inositol play a key role in intracellular signaling cascades that allow regulation of a diverse array of cellular processes (Fig. 1D)4. Because the phosphate ester group is used in numerous ways in biology, a detailed molecular understanding of the interactions of proteins, enzymes and small molecules with these ubiquitous functional groups is of great interest. Figure 1 Various biological forms of phosphate esters. (A) Phosphorylated amino acids (B) Nucleic acids (C) Phospholipids (D) inositol-1-phosphate. The primary focus of this review is enzymes that interact with phosphate esters to promote catalysis. These interactions may be divided into two general classes: indirect and direct. The indirect interaction involves noncovalent bonding of the enzyme to an oxygen atom of a phosphate ester group that does not undergo a change in covalent bonding during the enzymatic reaction (Fig. 2). By definition, indirect interactions must involve substrates (such as DNA and RNA) that contain more than one phosphate ester group, and the specific binding energy of the enzyme with such groups can be used to drive catalysis in a variety of ways as discussed in detail below. In contrast, the direct interaction (at least within the scope of this review) involves noncovalent bonding to a phosphate ester center that is the direct target of a nucleophilic substitution reaction (Fig. 2). In terms of catalysis, both indirect and direct interactions can have comparable large effects in lowering the activation barrier for a given reaction. As will be illuminated throughout this review, the dense display of lone pair electrons on the phosphate ester oxygens, and the overall negative charge of the phosphate ester group provide unique opportunities for powerful hydrogen bonding and electrostatic interactions that may be used to either destabilize ground states or stabilize transition states, as required for enzymatic catalysis (Fig. 2)5. Figure 2 Examples of indirect and direct catalytic interactions of an enzyme with phosphate ester groups. The direct interaction (right) involves a phosphate group that undergoes nucleophilic substitution during the course of the enzymatic reaction. Hydrogen bonding ... 1.2 Chemical Substitutions at Phosphate Ester Oxygens One general goal in the field of chemical biology and enzymology is to develop chemical tools that can provide unique insights into the nature of protein-ligand interactions that go beyond those provided by structural methods such as X-ray crystallography or NMR spectroscopy. The idealized objective is to construct chemical congeners of naturally occurring biological functional groups that discretely probe one energetic aspect of an interaction. Thus, the change in a binding interaction of the congener in the ground state and transition state relative to the actual functional group provides a measure of the energetic importance and the nature of the native interaction. Of course, truly selective perturbations of a single interaction are seldom if ever realized, and one is always faced with the harsh reality that observed energetic perturbations to such complex systems as enzyme-substrate complexes will always be a mixture of short-range and long-range effects brought about by the cooperative nature of the binding interactions. Nevertheless, chemical perturbation approaches can be qualitatively and quantitatively useful if binding measurements are made for each stable complex along a reaction coordinate and the structures of the complexes are also available. Moreover, as is the major focal point of this review, combining chemical perturbations approaches with protein mutagenesis can provide more compelling insights than either approach alone. Frequently used congeners for phosphate diesters are shown in Figure 3 (analogous congeners can be made for monoesters and triesters of phosphate). The depicted chemical substitutions may be divided into two classes depending on whether a nonbridging or bridging oxygen is replaced. Stereospecific substitution of a nonbridging oxygen with sulfur or a methyl group is often performed to explore the importance of hydrogen bonding, steric fit, or electrostatic interactions with a specific oxygen. Contrastingly, substitution of a low pKa sulfur for a bridging oxygen is assumed to destabilize the ester linkage to nucleophilic attack by relinquishing the requirement for acid catalysis of expulsion of the high pKa oxygen leaving group. The chemical rationale for these effects of nonbridging and bridging substitutions is discussed in detail in section 2. We also note here that substitution of single nonbridging oxygen of a prochiral phosphate diester with another atom or functional group generates a stereocenter as depicted in Figure 3. Figure 3 Common nonbridging and bridging chemical substitutions of oxygen atoms of phosphate esters. In this figure, the stereochemical priority is R > R′. 1.3 Scope The literature where oxygen atoms of phosphate esters have been substituted to investigate properties of various biological systems is extensive and dates back at least 45 years6. This review focuses on studies that have used sulfur and carbon substitution of phosphate ester oxygens to probe the energetics of protein-ligand interactions, with a deliberate emphasis on enzymatic reactions. Notably, we specifically exclude studies in which the phosphate ester oxygen is coordinated to a metal ion in favor of systems where the enzyme itself makes direct contact with the phosphate ester. In this regard, studies using nucleotides fall outside this scope 7–9, and we refer the reader to elegant studies that cover the powerful approach of “metal rescue” in which the damaging effect of sulfur substitution for oxygen is ameliorated by switching to a soft metal that prefers bonding to sulfur10,11,12. In addition, the general approach of using stereospecific sulfur substitution to determine the stereochemical course of enzymatic nucleophilic substitution reactions is not covered to any significant extent and we refer the reader elsewhere for this methodology8,6,13. Our approach is to first provide an overview of the chemical basis for the energetic effects of sulfur and methyl substitution for oxygen, then present the free energy analysis that describes the combined energetic effects of phosphate ester substitution and mutagenesis, and finally, provide representative examples that exemplify the utility of the approach.