D-Amino Acids: Properties, Occurrence, and Therapeutic Relevance

The Mirror-Image Problem

All twenty canonical amino acids encoded by the genetic code are L-amino acids, with the exception of glycine, which lacks a chiral α-carbon. The choice of the L-configuration is ancient, conserved across all life, and the source of one of biochemistry's most reliable rules: ribosomes do not incorporate D-amino acids into proteins. The enzymes that process, modify, and degrade proteins are correspondingly complementary to the L-configuration, and a D-amino acid placed in a peptide sequence is invisible to most of these enzymes. This invisibility is both a biological curiosity and an exploitable property that has been put to work across antibiotic natural products, synthetic peptide design, and drug discovery.

D-Amino Acids in Nature

D-amino acids are not biological rarities confined to exotic organisms. Bacterial cell walls are built around peptidoglycan, a polymer that contains D-alanine and D-glutamate as essential structural components. These D-amino acids are incorporated by racemases and D-amino acid ligases that operate entirely outside the ribosomal system, and their presence in peptidoglycan is one reason why the bacterial cell wall is an excellent antibiotic target: the enzymes that process D-amino acid-containing substrates have no mammalian counterparts. During stationary phase, many bacteria release free D-amino acids including D-methionine, D-leucine, D-tyrosine, and D-tryptophan into their environment, where they function as signals that trigger biofilm disassembly and cell wall remodeling. ^[11]

In mammalian tissues, D-serine and D-aspartate are present at physiologically significant concentrations. D-serine is produced from L-serine by serine racemase in glial cells and is a co-agonist of the NMDA receptor glutamate receptor subtype, regulating synaptic plasticity and long-term potentiation. Its concentration is regulated by D-amino acid oxidase in the brain, and alterations in D-serine metabolism are associated with schizophrenia and other neurological conditions. D-aspartate is found in the pituitary and pineal gland and plays a role in hormone regulation. These observations established that D-amino acids are not simply biosynthetic accidents but are metabolically regulated signaling molecules in mammalian biology.

D-Amino Acids in Natural Peptide Antibiotics

Many naturally occurring antimicrobial peptides produced by bacteria and fungi incorporate D-amino acids to enhance stability and modulate biological activity. Gramicidin S, the cyclic decapeptide antibiotic produced by Bacillus brevis, contains two D-phenylalanine residues that induce the beta-turn geometry required for its bioactive conformation. Tyrocidine A, produced by the same organism, incorporates D-phenylalanine and D-ornithine. The polymyxins, lipopeptide antibiotics used as last-resort treatments for multidrug-resistant Gram-negative infections, contain D-phenylalanine and D-leucine. In each case, the D-amino acids serve structural functions that cannot be replicated by the corresponding L-residues, and the resulting structures are resistant to the proteolytic enzymes of the target organism.

D-Amino Acids in Synthetic Peptide Design

The principal synthetic application of D-amino acids is proteolytic stabilization. A peptide composed entirely or substantially of D-amino acids is resistant to the stereospecific proteases of plasma, the gastrointestinal tract, and lysosomes, because these enzymes evolved to cleave L-peptide bonds and cannot accommodate the opposite configuration in their active sites. This approach to stability requires no backbone modification and preserves the amide NH hydrogen bond donor, which is eliminated by N-methylation. For sequences intended for oral administration, intravenous delivery, or prolonged systemic half-life, partial or complete D-amino acid substitution is among the most straightforward routes to metabolic stability.

The retro-inverso strategy takes this logic a step further by reversing the sequence direction simultaneously with the stereochemical inversion. A retro-inverso peptide, in which the sequence is written C-to-N using D-amino acids rather than N-to-C using L-amino acids, presents a side chain topology similar to that of the original L-peptide when the backbone direction is taken into account, because the reversal of sequence direction partially compensates for the inversion of backbone polarity. Retro-inverso analogs of biologically active peptides have been used to produce stable mimetics of L-peptide hormones and receptor ligands, though the correspondence in biological activity is not always predictable from structural reasoning alone.

Mirror-Image Phage Display

One of the most elegant applications of D-amino acid chemistry in drug discovery is mirror-image phage display, a strategy developed to identify D-peptide ligands for targets that cannot be easily crystallized in complex with conventional L-peptide hits. ^[12] The approach exploits a fundamental symmetry principle: the interaction between a D-peptide ligand and an L-protein target is the mirror image of the interaction between the corresponding L-peptide and a D-protein target. By synthesizing the target protein chemically as its mirror image, a D-amino acid version with identical topology but inverted chirality, and then performing standard L-peptide phage display against this mirror-image target, one selects L-peptide binders for the mirror-image protein. The sequence of the winning L-peptide, when synthesized as its D-amino acid enantiomer, binds the natural L-protein target with the same affinity and selectivity as the L-peptide bound the D-target. The resulting D-peptide drug candidate is proteolytically stable, potentially orally active, and non-immunogenic relative to L-peptide analogs.

The mirror-image phage display strategy requires total chemical synthesis of the D-protein target, which is feasible only for proteins accessible to native chemical ligation or related convergent methods. Despite this constraint, the approach has produced D-peptide leads against HIV gp41, amyloid-beta aggregates, and several other targets of therapeutic interest, and it represents a distinctive contribution of D-amino acid chemistry to the drug discovery toolkit.