Natural peptides generally play significant roles in human physiology, such as actions as neurotransmitters, growth factors, hormones and ion channel ligands [1-3]. Peptides often have selective and effective binding to specific cell surface receptors initiating intracellular effects, including ion channels or G protein-coupled receptors [3]. Compared to small molecule drugs, peptides display higher affinity and specificity to receptors and show lower toxicity [4]. During the past decades, peptides are widely used in medicine and biotechnology, and therapeutic peptide studies gain rapid development (Scheme 1a). Currently, there are >80 commercially available peptide drugs approved by US Food and Drug Administration (FDA) on the global market [5]. And, >140 peptide medicines are in clinical trials while >500 therapeutic peptides are in preclinical research. For instance, the peptide-based drug Lupron™ is employed to treat prostate cancer, obtaining global sales of more than US$2.3 billion in 2011 [6]. However, using naturally occurring peptides as therapeutic medicines often encounters some intrinsic weaknesses, such as metabolic instability and low membrane permeability [7,8]. Thus, a series of strategies has been developed to modify natural amino acids and peptides. Inserting unnatural amino acids into peptides or directly modifying natural peptides is highly popular for medicinal and biological chemistry as well as industry. This generally leads to improved bioactivity and pharmacokinetics [9,10]. Additionally, peptides conjugated with drugs, polymers or fluorophores are critical for a variety of biological and therapeutic applications. Peptides labelled with fluorophores are applied for realtime tracking of cells, metabolites and biomolecules under physiological conditions [11-13]. Peptides conjugated to drugs are used for selective target therapy, such as antibody-drug conjugates (ADCs) [14-16]. Peptides ligation to polymers are employed for increasing their circulation half-lives and stability [17-19]. Among them, chemo- and site-selective functionalizations of amino acids and peptides are essential in contemporary biochemistry and pharmaceutical chemistry, as the ability to fine-tune structural features, which is beneficial to modulate physicochemical and biological properties [20-22]. Therefore, classical condensations and (cyclo)additions are firstly employed in the derivatizations of amino acids and peptides [22,23]. With the advancement of synthesis technology, transition metal-catalyzed cross-coupling reactions have emerged as important strategies for amino acid and peptide modifications, largely extending the chemical scope [21,24].

Over the past decades, transition metal-catalyzed C—H activation is becoming a powerful tool for the construction of C—C and C-X (X = N, O, S, P, Si…) bonds in a high atom- and step-economical manner [25-32]. Compared to transition metal-catalyzed cross-coupling, this approach avoids prefunctionalization and triggers the transformations at inert C—H bonds. Recently, transition metal-catalyzed C—H activation has been successfully employed in the diversifications of amino acids and peptides [30,31], such as C—H arylation [33-35], olefination [36], allylation [37], alkynylation [38] and alkylation [39-41]. Through meticulous design, an array of novel cyclopeptides is synthesized via transition metal-catalyzed C—H activation and subsequent intramolecular macrocyclization [42-45]. These have been summarized in several excellent reviews, and are outside the scope of this review [42,43]. As domino reactions can largely increase the complexity and diversity of products, they have been widely utilized in the synthesis of heterocycles, biologically active molecules and natural products [46]. Complex domino reactions recently have been developed to synthesize functionalized amino acids and peptides bearing diverse heterocycles, largely enriching the structural complexity and diversity, as well as the physical/chemical properties and biological activities. To the best of our knowledge, there is currently no review dedicated specifically to the synthesis of unnatural amino acids and peptides via transition metal-catalyzed C—H activation/annulation. In this review, we aim to provide a comprehensive summary and documentation of reported methods employed for the preparation of unnatural amino acids and peptides with various heterocycles, through transition metal-catalyzed C—H activation/annulation (Scheme 1b).

Natural peptides are usually not be used as direct therapeutics due to their intrinsic weaknesses, such as poor physical and chemical stability, and a short circulating plasma half-life. Unnatural peptides are often essential for academic and industry circles, because of the enhanced pharmacokinetics and bioactivity in contrast with the natural ones. Thus, concise and efficient accesses to unnatural amino acids and peptides have been investigated. Among them, chemo- and site-selective methods are the key points in modern chemistry, due to the capacity to fine-tune structural characters that can help modulate biological and physicochemical properties. Compared to classical condensations and (cyclo)additions as well as transition metal-catalyzed cross-coupling reactions, transition metal-catalyzed C—H activation shows great advantages in the modifications of amino acids and peptides. Without substrate prefunctionalizations, diverse transformations can be performed in a high atom- and step-economical fashion. Recently, transition metal-catalyzed C—H activation/annulation has been employed in the late-stage derivatizations of amino acids and peptides. For Rh-catalyzed C—H activation/annulation, by using amide as directing group, an array of structurally novel amino acids and peptides with different kinds of heterocycles is achieved, such as bearing quinolone, isoquinolone, benzazepine, indolizinone, quinolizinone, pyrimido[1,6-a]indolone and polycyclic aromatic hydrocarbon oxonium ion scaffolds. Employing N-phenoxy amide or phosphinamide as directing group gives amino acids bearing a dihydrobenzo[d]isoxazole or a phosphaisoquinolin-1-one moiety, respectively. Utilizing phenol as directing group affords amino acids and peptides bearing a chromone core. For other transition metal-catalyzed C—H activation/annulation, such as Ru, Pd and Re, by using amide or sulfonamide as directing group, amino acids and peptides bearing isoquinolone or benzosultam scaffolds are respectively generated. Employing imine as directing group produces amino acids bearing an indane moiety.

Despite major advances, some limitations still remain. Firstly, the directing groups are mainly limited to amide-type groups, largely restricting the reaction types. Exploring other kinds of directing groups is necessary for the C—H activation/annulation process, resulting in novel reactions. Secondly, except alkyne and alkene, other coupling partners are seldom investigated. Expanding the scope of the coupling partners would deliver structurally novel unnatural amino acids and peptides. Thirdly, enantioselective C—H activation/annulation has not yet been reported for the construction of enantiopure unnatural amino acids and peptides. Especially, when the generated heterocycles own an extra chiral center, it will be more difficult to achieve the control over diastereoselectivity. Therefore, developing effective strategies for enantioselective C—H activation/annulation is demanded, which is also important for new peptide drug discovery. Finally, developing mild and efficient transition metal-catalyzed C—H activation/annulation for the synthesis of unnatural amino acids and peptides is another direction. We hope this review will be beneficial for researchers to better understand the chemistry behind transition metal-catalyzed C—H activation/annulation for the construction of unnatural amino acids and peptides.