New therapeutic modalities are great interests in pharmaceutical researchers. We believe that induction of selective degradation of target proteins by small molecules could become a new therapeutic modality of drug discovery. And we speculated that formation of an artificial (nonphysiological) complex of cIAP1 (cellular inhibitor of apoptosis protein 1) and a target protein would be induced by a hybrid small molecule composed of a cIAP1 ligand linked to a ligand of the target protein, and this would lead to cIAP1-mediated ubiquitination and subsequent proteasomal degradation of the target protein. This review article summarizes advances in chemical protein knockdown, that is, the small hybrid molecules induce decrease of the target proteins in living cells. This article also describes that this technology is capable of degrading receptors, enzymes, substrate binding proteins, and aggregation-prone proteins.

The molecular glue that connects two different proteins is a unique but not an entirely new chemical entity. One of the notable research achievements with this molecular class is sequential identification of molecular mechanisms of two immunosuppressive natural products FK506 and rapamycin. FK506 binds to FKBP12 to interact with and inhibit the protein phosphatase calcineurin. Likewise, rapamycin binds to FKBP12 to interact with and inhibit the phosphatidylinositol 3-kinase-related kinase mTOR. Importantly, the molecular glue story has recently been extended to synthetic small molecule space as well. Clinically important myeloma drug lenalidomide and other immunomodulatory imide drugs (IMiDs) were found to hijack the cullin 4 (CUL4)-RING E3 ubiquitin ligase system by binding to the one of DDB1- and CUL4-associated proteins (DCAFs), cereblon (CRBN), to redirect the substrate selectivity of CRBN. In consequence of the small molecule mediated neo-substrate recruitment, IMiDs induce the proteasomal degradation of the transcription factors IKZF1, IKZF3 and casein kinase 1α in a selective manner. We also reported that a series of anticancer sulfonamides such as E7070 (indisulam) and E7820 induce the protein-protein interaction between another DCAF protein DCAF15 and the slicing factor CAPERα (also known asRBM39), resulting in the selective proteaosomal degradation of CAPERα as a neo-substrate for the E3 ubiquitin ligase. All these findings may indicate a significant opportunity to gain a new insight into molecular glues in drug discovery, leading to further development of new modalities of bi-functional selective protein degraders with target protein knockdown function, represented by PROTAC (Proteolysis Targeting Chimera), Degronimide, and SNIPER (Specific and Nongenetic IAP-dependent Protein Eraser).

Boranophosphate (PB) DNAs and RNAs are promising candidates of oligonucleotide therapeutics because of their attractive features, such as high nuclease resistance and low toxicity. However, the difficulties in synthesizing boranophosphate oligonucleotides hamper the detailed investigation on their physiological and biological properties. Under these circumstances, we have developed efficient synthetic methods of boranophosphate oligonucleotides. Our methodology can be roughly divided into two categories. The first cateogory includes the use of P-boronated nucleotide monomers which facilitate the utilization of general acyl-type amino protecting groups. The second one utilizes oxazaphospholidine monomers those chiral auxiliary can be removed under acidic conditions. Notably, the latter method affords stereo-controlled boranophosphate oligonucleotides with all four nucleobases. Important properties such as a duplex forming ability to the complementary strand, RNase H activity and nuclease resistance are revealed using boranophosphate oligonucleotides synthesized by above mentioned methods.

Nucleic acid therapeutics based on chemically synthesized oligonucleotides such as antisense oligonucleotide (ASO) or siRNA, are attracting attention as a new drug discovery. In the development of conventional nucleic acid therapeutics had been problems in in vivo stability and permeability to cell membranes. However, in recent years, they have been improved by modification of nucleic acid structure by several approaches such as phosphorothioate or bridged nucleic acid (BNA). Especially, 2’-O,4’-C-ethylene-bridged nucleic acid (ENA) is highly promising modified nucleic acid which has discovered from Daiichi-Sankyo. Herein, we have developed a highly efficient manufacturing method of ENA phosphoramidites as key intermediates for the synthesis of ENA containing oligonucleotides to increase the productivity and to reduce the manufacturing cost and lead time. The basic concept of the new synthetic route is “divergent synthesis” for the preparation of four types of monomers such as A, G, C and T. Namely, we set a common intermediate as downstream as possible to reduce the overall reaction steps. So the challenge was stereoselective glycosylation reactions without utilizing neighboring group participation. In addition, by changing the routes drastically and telescoping the multiple reaction steps, the productivity was significantly increased.

To date, a number of chemically modified oligonucleotides (ONs) have been designed for use in nucleic acids-based therapeutics. In our group, we have been intensely working on the synthesis of 4’-thio and 4’-seleno ONs having sulfur or selenium atoms in place of furanose ring oxygen. To prepare 4’-thio and 4’-seleno ribonucleosides, we applied the Pummerer reaction of 4-thio or 4-seleno sugar with nucleobases, and the resulting 4’-thio and 4’-seleno ribonucleoside units were incorporated into ONs. Under the standard phosphoramidite conditions, 4’-thioRNA was obtained in good yields. While 4’-selnoRNA was given in very low yields. We carefully investigated ON synthesis containing 4’-selenoribonucleosides under standard phosphoramidite conditions. As a result, we found the unexpected strand-break occurred during oxidation step using I₂. On the basis of this finding, we succeeded in the first synthesis of a fully modified 4’-selenoRNA by using tert-butyl hydroperoxide as an alternative oxidant.

Oligonucleotide-based therapeutics are expected as novel therapeutic modalities, because they have a strong potential to cure the diseases which could not be targeted by conventional low-molecular-weight drugs. However, oligonucleotide has a polyanion structure derived from a phosphodiester backbone and thus, it cannot penetrate cell membranes. To develop the oligonucleotides-based therapeutics, this extremely low membrane permeability should be improved. In this study, we focused on the disulfide modification of oligonucleotides. Several disulfide derivatives are known to show efficient cellular uptake triggered by disulfide exchanging reaction with thiol group of membrane proteins. Therefore, we have synthesized the series of disulfide LD, CD or PD-conjugate oligonucleotides. All these disulfide modifications improved the membrane permeability of Antisense DNA and siRNA. Furthermore, several disulfide-conjugated oligonucleotides showed higher cellular uptake over lipofection method without any cytotoxicity. Surprisingly, the disulfide modified oligonucleotides are internalized directly into the cytoplasm with only 10 min. This ultrafast internalization of oligonucleotides is an advantage for the therapeutic application. Currently, we are trying to apply this novel disulfide-based membrane permeable oligonucleotide (MPON) for the development of oligonucleotide-based therapeutics.

Recent rapid advancements in unnatural base studies have opened the door to a new research area, Xenobiology. Unnatural base pairs that function as a third base pair in replication, transcription, and/or translation were created, toward the expansion of the genetic alphabet. After the first implementation of the concept by Alexander Rich in 1962, the arduous research to develop unnatural base pairs began in earnest from the late 1980’s. The challenge to create a ‘third base pair’ by understanding what governs the exclusive and selective base pairings of DNA in nature has captivated researchers, who have chemically synthesized a number of unnatural base pair candidates by repeating proof of concept studies, using their own ideas. Finally, several successful unnatural base pairs were applied to in vitro and in vivo biological experiments, toward their practical use in “real” biological systems. The representative three unnatural base pairs that function in PCR are the P-Z, 5SCIS-NaM, and Ds-Px pairs developed by Benner’s, Romesberg’s, and our groups, respectively. These unnatural base pairs can mediate the site-specific incorporation of unnatural components into DNA, RNA and proteins, allowing for the generation of new biopolymers with increased functionalities. This article describes the creation of unnatural base pairs and their applications to qPCR, DNA aptamers, semisynthetic organisms, and protein synthesis involving unnatural amino acids.

Genetic information is stored in the form of DNA base sequence, and its length is about 3 billion base pairs. While this DNA sequence encodes many proteins, miRNAs, long non-coding (lnc) RNAs, etc., it is also important in terms of strictly regulating the gene expression. In other words, transcription factors that bind to DNA and control gene expression are site-selectively regulated by recognizing specific DNA sequences and structures. Since various functions of cells in vivo are maintained under such control, it is effective for cell reprogramming, differentiation induction, disease treatment, etc. to control this expression artificially. To enable artificial control of gene expression, our group uses a base sequence-selective DNA-binding molecule “pyrrole-imidazole (PI) polyamide” to switch gene expression on/off “artificial gene switch” has been working on development. PI polyamide can be described to be a new modality different from conventional low molecular weight compounds, antibody drugs, and nucleic acid drugs that have recently attracted attention. In this article, we will explain the artificial gene switch approaches and their advantages, and introduce the potential of epigenetic drug discovery using them.

Antibody-drug conjugates (ADCs) have received considerable attention as next-generation antibody drug. ADCs bring cytotoxic agents to lesion site and release them. Traditional conjugation on Lys and Cys residues typically generates heterogeneous ADCs. In order to expand therapeutic index, homogeneous ADCs are highly desired. We achieved homogeneous ADC preparation via N-glycan in the Fc region using endo-β-N-acetylglucosaminidase (ENGase) and its mutant. The heterogenous original N-glycan was removed by ENGase, and azide carrying human-type glycan was attached by the ENGase mutant. Next, bio-orthogonal reaction between azide and strained alkyne in payload moiety afforded the homogeneous ADC. Moreover, we established reaction monitoring method using ultra-performance liquid chromatography with an amide-based wide-pore column and optimized the reaction conditions.

ADC treatment for solid-cancer has some limitation because abundant stroma hampers antibody access to the antigen on the cell surface. To overcome this limitation, we proposed the caner stromal targeting (CAST) therapy. We developed an antibody against insoluble fibrin and designed a plasmin-cleavable linker. The ADC showed potent antitumor effect in mice bearing stroma-rich tumor.

Antibody-drug conjugates (ADCs) have become a major class of cancer biopharmaceuticals and traditional ADCs have a stochastic distribution of cytotoxic drugs linked across several different sites of the antibody. The heterogeneous nature of resulting stochastic ADCs can cause diminished efficacy and increased toxicity, thus limiting the corresponding therapeutic index. To improve on traditional ADC technology, we developed and report here a novel chemical conjugation platform termed “AJICAP™” for the site-specific modification of native antibodies through the use of a class of IgG Fc-affinity reagents. Site-specific installation of thiol functional groups to well-defined lysine residues in IgGs followed by conjugation to these newly installed thiols with cytotoxic payloads was efficiently conducted to generate AJICAP™-ADCs. Conjugation site location was confirmed by peptide mapping and Q-TOFMS analysis showed that the Drug/Antibody Ratio (DAR) was approximately two. Several xenograft in vivo efficacy studies were conducted and indicated the AJICAP™-ADCs display significant tumor inhibition comparable to Kadcyla^®. Furthermore, a rat pharmacokinetic analysis and toxicology study indicated an enhancement of the maximum tolerated dose, indicating an expansion of the comparative AJICAP™-ADC therapeutic index compared to stochastic technology. AJICAP™ technology is a powerful platform to enable next-generation ADCs through the reduction of heterogeneity and enhanced of therapeutic index.

Antibody-drug conjugates (ADCs) are an emerging drug format of chemotherapy agents with durable therapeutic efficacy and high target specificity. They are expected to revolutionize current cancer treatment strategies and regimens. Currently, seven ADCs are approved by the U.S. Food and Drug Administration (FDA), and more than 100 ADCs are in clinical trials. ADCs consist of humanized or fully human monoclonal antibodies with extremely potent cytotoxic agents (payloads) attached via chemical linkers. The linker structure and conjugation method markedly influence ADC homogeneity, circulation stability, pharmacokinetics profile, therapeutic window, and treatment outcome. In most prior attempts to improve these parameters, linear chemical linkers were used. A disadvantage of this approach is that only one payload molecule can be incorporated per linker. However, the use of branched linkers that can incorporate multiple payload molecules has not been fully explored. In addition, the most common enzymatically cleavable linker used in current ADCs, while stable in humans, is labile in mouse circulation depending on the conjugation site and linker length. In this paper we describe branched linkers, efficient conjugation methods for constructing homogeneous branched ADCs, and a novel tripeptide linker that retains responsiveness to enzymatic drug release and stability in both mouse and human plasma. We also examine several aspects to be considered for constructing safer and more efficacious next-generation ADCs based on recent advances in the ADC field.

Despite peptides have been a considerable class as therapeutics from the early 20^th century, they were recognized an elusive modality due to the quick biodegradability and the low membrane permeability. In recent two decades, macrocyclic peptides have emerged as an expanded modality from standard class of peptides, owing their synthetic accessibility, high degree of specific binding, and the ability to target protein surfaces traditionally considered “undruggable”. From the structural point of view, unusual side-chains, backbone, and macrocyclic structure are resembled to membrane permeable natural product peptides such as cyclosporin A, thus expected to overcome intrinsic drawbacks of peptides. In this review, we refer such natural product-like macrocyclic peptide ligands to as “pseudo-natural macrocyclic peptides”, and describe advances of biotechnologies to discover de novo pseudo-natural macrocyclic peptide ligands and our recent research to further optimize such drug leads.

Recently, development of bioactive molecules based on polypharmacology has attracted much attention. Conjugation of bioactive molecules can afford biofunctional middle molecules possessing polypharmacological action. These conjugated compounds are called small molecule drug conjugates (SMDC), which are a class of medium-sized drugs. A similar conjugation strategy, a self-adjuvanting strategy, has been used for vaccine development.

Covalently linked antigen-adjuvant conjugates are used as self-adjuvanting vaccines. The antigen-adjuvant conjugates are taken up by the antigen-presenting cells via the interactions between innate immune receptors and their ligands, and the adjuvants activate the immune system and efficiently induce antibodies. Here, we developed cancer vaccine candidates consisting in tri-peptides of a tumor associated carbohydrate antigens (TACA), a lipopeptide adjuvant (a TLR2 ligand), and a T cell epitope.