The Role of Unnatural Amino Acids in Drug Development

Over the past several decades, peptide therapeutics have undergone rapid development. From the earliest natural hormone analogs to today’s cyclic peptide drugs, targeted oncology therapeutics, and protein-protein interaction inhibitors, peptides have become an important platform in modern innovative drug discovery and development. However, peptides composed of natural amino acids also have certain inherent limitations, such as susceptibility to protease degradation, short in vivo half-life, poor oral absorption, and insufficient conformational stability.

To overcome these limitations, researchers have gradually turned their attention to unnatural amino acids (UAAs). By introducing specially designed amino acid residues into peptide sequences, the stability, selectivity, pharmacokinetic properties, and biological activity of drugs can be significantly improved. Today, unnatural amino acids have become an indispensable and important tool in the development of innovative peptide therapeutics.

Why Unnatural Amino Acids Are Needed

Proteins in nature are mainly composed of the twenty standard amino acids. These amino acids, shaped by billions of years of evolution, are highly suitable for building living systems, but they are not necessarily the best choice for drug development. The goal of drug discovery is not to replicate natural proteins, but to create molecules with optimal therapeutic effects. Therefore, researchers hope to break through the limitations of natural amino acids and use chemical synthesis to introduce new structural units, enabling peptides to acquire properties that do not exist in natural systems. Unnatural amino acids can alter the spatial conformation, hydrophobicity, charge distribution, and metabolic pathways of peptides, thereby endowing drugs with new functions. Many modern peptide therapeutics are, in fact, no longer natural peptides in the strict sense, but engineered molecules extensively modified with unnatural amino acids.

D-Amino Acids: A Classic Strategy for Improving Stability

Proteins in nature are almost entirely composed of L-amino acids, and therefore most proteases in the human body have also evolved to specifically recognize amino acids in the L-configuration. This characteristic provides an important opportunity for drug design. When researchers replace certain key positions with D-amino acids, the overall structure of the peptide may remain similar, but protease recognition can be significantly reduced. Because enzymes cannot effectively bind to and cleave these sites, the in vivo stability of peptides is usually markedly improved. This strategy has been widely applied in the development of antimicrobial peptides, cancer therapeutic peptides, and peptides for the treatment of metabolic diseases. Many studies have found that replacing only one or two key amino acids can increase plasma stability by several-fold or even tens of fold.

In addition to improving stability, D-amino acids can also alter local conformation. Because their chirality is opposite to that of natural amino acids, the introduction of D-amino acids often changes the folding pattern of the peptide chain, thereby affecting receptor-binding modes. In some cases, this change can even improve drug activity and selectivity. In recent years, an increasing number of studies have begun to adopt partial D-amino acid substitution strategies rather than simply reversing the entire sequence completely. This refined design can achieve optimal stability while maintaining activity.

N-Methyl Amino Acids: An Important Tool for Improving Pharmacokinetics

If D-amino acids are primarily used to improve stability, then N-methyl amino acids are more often used to optimize pharmacokinetic properties. N-methylation refers to replacing the amino hydrogen in a peptide bond with a methyl group. This seemingly small change can have a profound impact on the behavior of the entire molecule. First, N-methylation can reduce the ability of peptide bonds to participate in hydrogen-bond formation. Because many proteases rely on peptide-bond hydrogen-bond networks when recognizing substrates, N-methylation often improves resistance to enzymatic degradation. More importantly, N-methylation can significantly affect the spatial conformation of a molecule.

In many cyclic peptide drugs, N-methylated residues promote the formation of stable internal hydrogen-bond networks, causing polar groups to be hidden inside the molecule. This phenomenon can reduce the apparent polarity of the molecule, thereby improving its ability to permeate cell membranes. For the development of orally administered peptide therapeutics, this is an extremely important property. The traditional view holds that peptides are difficult to absorb through the intestine because of their high polarity and large molecular weight. However, appropriate N-methylation can help molecules achieve better membrane permeability while maintaining activity.

The well-known immunosuppressant Cyclosporine is one of the most successful representatives of this strategy. Its molecule contains multiple N-methyl amino acid residues, and these modifications play a key role in its excellent oral activity.

Combining Unnatural Amino Acids with Cyclic Peptide Design

In recent years, an important trend in the development of innovative peptide therapeutics has been the integration of unnatural amino acids with cyclic peptide engineering. Cyclization itself can improve conformational stability and protease resistance, while unnatural amino acids can further optimize the physicochemical properties of the molecule. When the two strategies are combined, the resulting effects often far exceed those achieved by either strategy alone. Many new cyclic peptide drugs simultaneously employ D-amino acids, N-methyl amino acids, and other specialized residues. By precisely controlling cyclization sites and the distribution of unnatural amino acids, researchers are continuously optimizing the stability, selectivity, and pharmacokinetic profiles of drug candidates. This design concept is driving cyclic peptide therapeutics into a new stage of development.

Other Important Unnatural Amino Acids

In addition to D-amino acids and N-methyl amino acids, an increasing number of specialized amino acids are being applied in drug development. Aib (α-aminoisobutyric acid) is widely used to stabilize α-helical structures because of its strong conformational constraint. Many long-acting peptide therapeutics use Aib modification to improve structural stability. Aromatic unnatural amino acids such as Bip, Nal, and Cha are commonly used to enhance hydrophobic interactions and improve receptor-binding ability. Amino acids bearing special reactive groups can be used to conjugate drugs, attach fluorescent probes, or form stable bridged structures. These modifications provide modern drug design with a degree of freedom far beyond that of natural amino acid systems.

With the development of solid-phase peptide synthesis technology, it is now possible to conveniently introduce hundreds of types of unnatural amino acids into a single peptide chain, thereby enabling more complex and precise molecular design.

AI Is Accelerating the Development of Unnatural Amino Acid-Based Drugs

Although unnatural amino acids provide enormous design space, they also increase the complexity of research and development. Traditional experience-based methods have difficulty predicting how a given modification will affect activity, stability, and pharmacokinetics. In the past, extensive experimental screening was often required to identify the optimal combination. In recent years, artificial intelligence technologies have begun to help researchers address this problem. By learning from large amounts of structural and activity data, AI models can predict the effects of different unnatural amino acid modifications on peptide properties and rapidly screen the most promising candidate strategies. Combined with structure prediction and molecular simulation technologies, researchers can already complete a substantial amount of optimization work before experiments begin. In the future, the integration of unnatural amino acid engineering with AI-driven design is very likely to become one of the most important directions in the development of innovative peptide therapeutics.

Looking Ahead

Unnatural amino acids have profoundly changed the development model for modern peptide therapeutics. They not only improve drug stability and half-life, but also provide new solutions for the development of oral peptides, cyclic peptide drugs, and difficult-to-drug targets. From D-amino acids to N-methyl amino acids, and from conformationally constrained residues to functionalized modification units, unnatural amino acids are continuously expanding the boundaries of drug design. With further advances in artificial intelligence, structural biology, and synthesis technologies, future peptide therapeutics will increasingly rely on these carefully designed specialized building blocks. For innovative drug discovery and development, unnatural amino acids are no longer merely auxiliary modification tools; they are becoming an important foundation for creating the next generation of peptide therapeutics.