Process Development of a Macrocyclic Peptide Inhibitor of PD-L1.

This article outlines the process development leading to the manufacture of 800 g of BMS-986189, a macrocyclic peptide active pharmaceutical ingredient. Multiple N-methylated unnatural amino acids posed challenges to manufacturing due to the lability of the peptide to cleavage during global side chain deprotection and precipitation steps. These issues were exacerbated upon scale-up, resulting in severe yield loss and necessitating careful impurity identification, understanding the root cause of impurity formation, and process optimization to deliver a scalable synthesis. A systematic study of macrocyclization with its dependence on concentration and pH is presented. In addition, a side chain protected peptide synthesis is discussed where the macrocyclic protected peptide is extremely labile to hydrolysis. A computational study explains the root cause of the increased lability of macrocyclic peptide over linear peptide to hydrolysis. A process solution involving the use of labile protecting groups is discussed. Overall, the article highlights the advancements achieved to enable scalable synthesis of an unusually labile macrocyclic peptide by solid-phase peptide synthesis. The sustainability metric indicates the final preparative chromatography drives a significant fraction of a high process mass intensity (PMI).

An optimized agonist peptide of protease-activated receptor 4 and its use in a validated platelet-aggregation assay.

Protease-activated receptor 4 (PAR4) is a promising drug target to improve the efficacy/safety window of antiplatelet agents. The native peptide GYPGQV, and the more-potent peptide AYPGKF, are PAR4-specific activators. However, these PAR4 agonist peptides (APs) elicit an agonist response, for example, platelet aggregation, at concentrations of 50 to 1000 µM in platelet-function assays, thereby limiting their utility to monitor the pharmacodynamic effects of PAR4 antagonists over a wide concentration range. Improved pharmacodynamic assays are needed for clinical development of PAR4 antagonists. We attempted to identify potent PAR4 APs to aid development of robust assays for optimization of PAR4 antagonists. Using an AYPG-based biased phage-display peptide library approach followed by chemical peptide optimization, A-Phe(4-F)-PGWLVKNG was identified. This peptide demonstrated an EC50 value of 3.4 µM in a platelet-aggregation assay, which is 16-fold more potent than AYPGKF. Using this new PAR4 AP, a platelet-rich plasma-aggregation assay using light-transmission aggregometry was developed and validated in a series of precision and reproducibility tests. PAR4 antagonist responses to PAR4 AP A-Phe(4-F)-PGWLVKNG (12.5 µM to 100 µM) were subsequently evaluated in this assay in vitro and ex vivo in a human study using BMS-986120, a PAR4 antagonist that entered clinical studies.

Ultra-high-throughput SPE-MALDI workflow: Blueprint for efficient purification and screening of peptide libraries.

At the forefront of synthetic endeavors in the pharmaceutical industry, including drug discovery and high-throughput screening, timelines are tight and large quantities of pure chemical targets are rarely available. In this regard, the development of novel and increasingly challenging chemistries requires a commensurate level of innovation to develop reliable analytical assays and purification workflows with rapid turnaround that enables accelerated pharmacological evaluation. A small-scale automation platform enabling high-throughput analysis and purification to streamline the selection of candidate leads would be a transformative advance. Herein, we introduce an automation-friendly solid-phase extraction-matrix-assisted laser desorption/ionization (SPE-MALDI) platform applied to the high-throughput purification and analysis of peptide libraries. This advance enabled us to purify peptides from microgram levels in less than a day with results comparable to traditional high-performance liquid chromatography-diode array detection-mass spectrometry (HPLC-DAD-MS).

Inhibition of Myc transcriptional activity by a mini-protein based upon Mxd1.

Myc, a transcription factor with oncogenic activity, is upregulated by amplification, translocation, and mutation of the cellular pathways that regulate its stability. Inhibition of the Myc oncogene by various modalities has had limited success. One Myc inhibitor, Omomyc, has limited cellular and in vivo activity. Here, we report a mini-protein, referred to as Mad, which is derived from the cellular Myc antagonist Mxd1. Mad localizes to the nucleus in cells and is 10-fold more potent than Omomyc in inhibiting Myc-driven cell proliferation. Similar to Mxd1, Mad also interacts with Max, the binding partner of Myc, and with the nucleolar upstream binding factor. Mad binds to E-Box DNA in the promoters of Myc target genes and represses Myc-mediated transcription to a greater extent than Omomyc. Overall, Mad appears to be more potent than Omomyc both in vitro and in cells.

Multiple Synthetic Routes to the Mini-Protein Omomyc and Coiled-Coil Domain Truncations.

The Myc transcription factor represents an "undruggable" target of high biological interest due to its central role in various cancers. An abbreviated form of the c-Myc protein, called Omomyc, consists of the Myc DNA-binding domain and a coiled-coil region to facilitate dimerization of the 90 amino acid polypeptide. Here we present our results to evaluate the synthesis of Omomyc using three complementary strategies: linear Fmoc solid-phase peptide synthesis (SPPS) using several advancements for difficult sequences, native chemical ligation from smaller peptide fragments, and a high-throughput bacterial expression and assay platform for rapid mutagenesis. This multifaceted approach allowed access to up to gram quantities of the mini-protein and permitted in vitro and in vivo SAR exploration of this modality. DNA-binding results and cellular activity confirm that Omomyc and analogues presented here, are potent binders of the E-box DNA engaged by Myc for transcriptional activation and that this 90-amino acid mini-protein is cell permeable and can inhibit proliferation of Myc-dependent cell lines. We also present additional results on covalent homodimerization through disulfide formation of the full-length mini-protein and show the coiled-coil region can be truncated while preserving both DNA binding and cellular activity. Altogether, our results highlight the ability of advanced peptide synthesis to achieve SAR tractability in a challenging synthetic modality.

Omomyc Reveals New Mechanisms To Inhibit the MYC Oncogene.

The MYC oncogene is upregulated in human cancers by translocation, amplification, and mutation of cellular pathways that regulate Myc. Myc/Max heterodimers bind to E box sequences in the promoter regions of genes and activate transcription. The MYC inhibitor Omomyc can reduce the ability of MYC to bind specific box sequences in promoters of MYC target genes by binding directly to E box sequences as demonstrated by chromatin immunoprecipitation (CHIP). Here, we demonstrate by both a proximity ligation assay (PLA) and double chromatin immunoprecipitation (ReCHIP) that Omomyc preferentially binds to Max, not Myc, to mediate inhibition of MYC-mediated transcription by replacing MYC/MAX heterodimers with Omomyc/MAX heterodimers. The formation of Myc/Max and Omomyc/Max heterodimers occurs cotranslationally; Myc, Max, and Omomyc can interact with ribosomes and Max RNA under conditions in which ribosomes are intact. Taken together, our data suggest that the mechanism of action of Omomyc is to bind DNA as either a homodimer or a heterodimer with Max that is formed cotranslationally, revealing a novel mechanism to inhibit the MYC oncogene. We find that in vivo, Omomyc distributes quickly to kidneys and liver and has a short effective half-life in plasma, which could limit its use in vivo.

Formation of α-chiral centers by asymmetric ß-C(sp3)-H arylation, alkenylation, and alkynylation.

The enzymatic ß-C-H hydroxylation of the feedstock chemical isobutyric acid has enabled the asymmetric synthesis of a wide variety of polyketides. The analogous transition metal-catalyzed enantioselective ß-C-H functionalization of isobutyric acid-derived substrates should provide a versatile method for constructing useful building blocks with enantioenriched α-chiral centers from this abundant C-4 skeleton. However, the desymmetrization of ubiquitous isopropyl moieties by organometallic catalysts has remained an unanswered challenge. Herein, we report the design of chiral mono-protected aminomethyl oxazoline ligands that enable desymmetrization of isopropyl groups via palladium insertion into the C(sp3)-H bonds of one of the prochiral methyl groups. We detail the enantioselective ß-arylation, -alkenylation, and -alkynylation of isobutyric acid/2-aminoisobutyric acid derivatives, which may serve as a platform for the construction of α-chiral centers.

Ligand-enabled ß-C-H arylation of α-amino acids using a simple and practical auxiliary.

Pd-catalyzed ß-C-H functionalizations of carboxylic acid derivatives using an auxiliary as a directing group have been extensively explored in the past decade. In comparison to the most widely used auxiliaries in asymmetric synthesis, the simplicity and practicality of the auxiliaries developed for C-H activation remains to be improved. We previously developed a simple N-methoxyamide auxiliary to direct ß-C-H activation, albeit this system was not compatible with carboxylic acids containing α-hydrogen atoms. Herein we report the development of a pyridine-type ligand that overcomes this limitation of the N-methoxyamide auxiliary, leading to a significant improvement of ß-arylation of carboxylic acid derivatives, especially α-amino acids. The arylation using this practical auxiliary is applied to the gram-scale syntheses of unnatural amino acids, bioactive molecules, and chiral bis(oxazoline) ligands.

Mcl-1 antagonism is a potential therapeutic strategy in a subset of solid cancers.

Cancer cell survival is frequently dependent on the elevated levels of members of the Bcl-2 family of prosurvival proteins that bind to and inactivate BH3-domain pro-apoptotic cellular proteins. Small molecules that inhibit the protein-protein interactions between prosurvival and proapoptotic Bcl-2 family members (so-called "BH3 mimetics") have a potential therapeutic value, as indicated by clinical findings obtained with ABT-263 (navitoclax), a Bcl-2/Bcl-xL antagonist, and more recently with GDC-0199/ABT-199, a more selective antagonist of Bcl-2. Here, we report study results of the functional role of the prosurvival protein Mcl-1 against a panel of solid cancer cell lines representative of different tumor types. We observed silencing of Mcl-1 expression by small interfering RNAs (siRNAs) significantly reduced viability and induced apoptosis in almost 30% of cell lines tested, including lung and breast adenocarcinoma, as well as glioblastoma derived lines. Most importantly, we provide a mechanistic basis for this sensitivity by showing antagonism of Mcl-1 function with specific BH3 peptides against isolated mitochondria induces Bak oligomerization and cytochrome c release, therefore demonstrating that mitochondria from Mcl-1-sensitive cells depend on Mcl-1 for their integrity and that antagonizing Mcl-1 function is sufficient to induce apoptosis. Thus, our results lend further support for considering Mcl-1 as a therapeutic target in a number of solid cancers and support the rationale for development of small molecule BH3-mimetics antagonists of this protein.

Synthesis of a stable isotopically labeled universal surrogate peptide for use as an internal standard in LC-MS/MS bioanalysis of human IgG and Fc-fusion protein drug candidates.

The synthesis of a 16-residue, stable isotopically labeled peptide is described for use as a LC-MS/MS (Liquid chromatography-mass spectrometry/mass spectrometry) internal standard in bioanalytical studies. This peptide serves as a single universal surrogate peptide capable of quantifying a wide variety of immunoglobulin G and Fc-fusion protein drug candidates in animal species used in pre-clinical drug development studies. An efficient synthesis approach for this peptide was developed using microwave-assisted solid phase peptide synthesis (SPPS) techniques, which included the use of a pseudoproline dipeptide derivative. The corresponding conventional room temperature SPPS was unsuccessful and gave only mixtures of truncated products. Stable-labeled leucine was incorporated as a single residue via manual coupling of commercially available Fmoc-[(13) C6 , (15) N]-l-leucine onto an 11-unit segment followed by automated microwave-assisted elaboration of the final four residues. Using this approach, the desired labeled peptide was prepared in high purity and in sufficient quantities for long-term supplies as a bioanalytical internal standard. The results strongly demonstrate the importance of utilizing both microwave-assisted peptide synthesis and pseudoproline dipeptide techniques to allow the preparation of labeled peptides with highly lipophilic and sterically hindered side-chains.

Annexin A2 is a natural extrahepatic inhibitor of the PCSK9-induced LDL receptor degradation.

Proprotein convertase subtilisin/kexin-9 (PCSK9) enhances the degradation of hepatic low-density lipoprotein receptor (LDLR). Deletion of PCSK9, and loss-of-function mutants in humans result in lower levels of circulating LDL-cholesterol and a strong protection against coronary heart disease. Accordingly, the quest for PCSK9 inhibitors has major clinical implications. We have previously identified annexin A2 (AnxA2) as an endogenous binding partner and functional inhibitor of PCSK9. Herein, we studied the relevance of AnxA2 in PCSK9 inhibition and lipid metabolism in vivo. Plasma analyses of AnxA2(-/-) mice revealed: i) a ∼1.4-fold increase in LDL-cholesterol without significant changes in VLDLs or HDLs, and ii) a ∼2-fold increase in circulating PCSK9 levels. Western blotting and immunohistochemistry of AnxA2(-/-) tissues revealed that the LDLR was decreased by ∼50% in extrahepatic tissues, such as adrenals and colon. We also show that AnxA2-derived synthetic peptides block the PCSK9≡LDLR interaction in vitro, and adenoviral overexpression of AnxA2 in mouse liver increases LDLR protein levels in vivo. These results suggest that AnxA2 acts as an endogenous regulator of LDLR degradation, mostly in extrahepatic tissues. Finally, we identified an AnxA2 coding polymorphism, V98L, that correlates with lower circulating levels of PCSK9 thereby extending our results on the physiological role of AnxA2 in humans.

Exploration of structure-activity relationships at the two C-terminal residues of potent 11mer Glucagon-Like Peptide-1 receptor agonist peptides via parallel synthesis.

We report the identification of potent agonists of the Glucagon-Like Peptide-1 receptor (GLP-1R) via evaluation of two positional scanning libraries and a two-dimensional focused library, synthesized in part on SynPhase Lanterns. These compounds are 11 amino acid peptides containing several unnatural amino acids, including (in particular) analogs of biphenylalanine (Bip) at the two C-terminal positions. Typical activities of the most potent peptides in this class are in the picomolar range in an in vitro functional assay using human GLP-1 receptor.

Identification of potent 11mer glucagon-like peptide-1 receptor agonist peptides with novel C-terminal amino acids: Homohomophenylalanine analogs.

We report the identification of potent agonists of the Glucagon-Like Peptide-1 Receptor (GLP-1R). These compounds are short, 11 amino acid peptides containing several unnatural amino acids, including (in particular) analogs of homohomophenylalanine (hhPhe) at the C-terminal position. Typically the functional activity of the more potent peptides in this class is in the low picomolar range in an in vitro cAMP assay, with one example demonstrating excellent in vivo activity in an ob/ob mouse model of diabetes.

Eleven amino acid glucagon-like peptide-1 receptor agonists with antidiabetic activity.

Glucagon-like peptide 1 (GLP-1) is a 30 or 31 amino acid peptide hormone that contributes to the physiological regulation of glucose homeostasis and food intake. Herein, we report the discovery of a novel class of 11 amino acid GLP-1 receptor agonists. These peptides consist of a structurally optimized 9-mer, which is closely related to the N-terminal 9 amino acids of GLP-1, linked to a substituted C-terminal biphenylalanine (BIP) dipeptide. SAR studies resulted in 11-mer GLP-1R agonists with similar in vitro potency to the native 30-mer. Peptides 21 and 22 acutely reduced plasma glucose excursions and increased plasma insulin concentrations in a mouse model of diabetes. These peptides also showed sustained exposures over several hours in mouse and dog models. The described 11-mer GLP-1 receptor agonists represent a new tool in further understanding GLP-1 receptor pharmacology that may lead to novel antidiabetic agents.

NEMO binding domain of IKK-2 encompasses amino acids 735-745.

NF-kappaB activation is mediated by the IKK signalsome. Though this signalsome is comprised of IKK-1, IKK-2, and NEMO/IKKgamma, it is the interaction between IKK-2 and NEMO that is critical to formation of a functional signalsome. More specifically, previous reports have indicated that this interaction involves the C-terminal LDWSWL residues of IKK-2 (called the Nemo Binding Domain (NBD)) and the N-terminus of NEMO. In an effort to characterize the IKK-2:NEMO interaction, we have investigated several NBD-containing peptides for their ability to bind NEMO and inhibit the critical IKK-2:NEMO interaction. The six residue NBD peptide, LDWSWL, showed modest binding to NEMO and little inhibition of the IKK-2:NEMO interaction, whereas peptides containing the NBD plus additional flanking amino acids (NBD-containing peptides) more effectively bound NEMO and inhibited the interaction. These longer NBD-containing peptides may be required to give the NBD an appropriate conformation for recognition by NEMO and/or to provide for additional interactions with NEMO.

An inhibitor binding pocket distinct from the catalytic active site on human beta-APP cleaving enzyme.

Beta-APP cleaving enzyme (BACE) is responsible for the first of two proteolytic cleavages of the APP protein that together lead to the generation of the Alzheimer's disease-associated Abeta peptide. It is widely believed that halting the production of Abeta peptide, by inhibition of BACE, is an attractive therapeutic modality for the treatment of Alzheimer's disease. BACE is an aspartyl protease, and there is significant effort in the pharmaceutical community to apply traditional design methods to the development of active site-directed inhibitors of this enzyme. We report here the discovery of a ligand binding pocket within the catalytic domain of BACE that is distinct from the enzymatic active site (i.e., an exosite). Peptides, initially identified from combinatorial phage peptide libraries, contain the sequence YPYF(I/L)P(L/I) and bind specifically to this exosite, even in the presence of saturating concentrations of active site-directed inhibitors. Binding of peptides to the BACE exosite leads to a concentration-dependent inhibition of proteolysis for APP-related, protein-based substrates of BACE. The discovery of this exosite opens new opportunities for the identification and development of novel and potentially selective small molecule inhibitors of BACE that act through exosite, rather than active site, binding interactions.