Breaking the oral barrier（3）

The discovery, optimization, and subsequent modification of PN-235 (icotrokinra, JN-2113)

The advantages of orally administered cyclic peptides have been detailed in previous articles, with PN-235 receiving significant attention as one of the very few cyclic peptide drugs currently available for oral administration. On July 21, 2025, Protagonist and Janssen announced that they had submitted a New Drug Application (NDA) to the FDA for Icotrokinra, for the treatment of moderate to severe plaque psoriasis in adults and children aged 12 and older. If approved, Icotrokinra is expected to become the world's first oral drug targeting IL-23R for the treatment of psoriasis. This article focuses on the discovery process and systematic optimization strategies of this molecule, using patent publication dates as a timeline, and also summarizes potential directions for further optimization.

As early as 2007, Valorisation Hsj, Societe En Commandite filed a patent application for IL-23R peptide antagonists (WO2009007849). This is one of the earliest patents involving IL-23R peptide antagonists in the publicly available information. This patent describes a series of peptides targeting IL-23R, including the agonist APG-2301 and the antagonists APG-2303, APG-2305, APG-2307, and APG-2309 (sequences shown in the figure below). The team used the amino acid sequence of IL-23R as a basis, combined with computer modeling techniques, to generate linear peptides corresponding to its hinge region (mimicking IL-23 and acting as IL-23R antagonists). Among these, APG-2305 and APG-2309 effectively inhibited STAT3 phosphorylation in mouse splenocytes in vitro, with IC50 values of 1 nM and 2 nM, respectively. Furthermore, in a rat IBD model, treatment with APG-2309 significantly alleviated inflammation-induced edema and redness, and significantly reduced mucosal leukocyte infiltration and vascular congestion, demonstrating good anti-inflammatory and therapeutic potential.

In 2013, Medical Diagnostic Laboratories (MDL) published a patent (US20130029907) describing the discovery of IL-23R-binding peptides using phage display technology, successfully identifying several peptide sequences that specifically bind to IL-23R. This research constructed an M13 phage display library containing random 12-mer peptide sequences and performed multiple rounds of screening using full-length soluble recombinant IL-23R and its splice variants as targets. Ultimately, 27 peptides with binding activity were identified. Notably, these peptide sequences showed low similarity to the IL-23 crystal structure reported by Beyer et al. (PDB ID: 3D85). Through sequence alignment, the researchers found that 66.7% of the peptides contained a conserved recognition sequence: -WX1X2X3W-, which was hypothesized to be a key functional motif for binding to IL-23R and blocking its downstream signal transduction (a representative sequence is shown in the figure below). Subsequently, the researchers evaluated the inhibitory activity of the peptides on IL-23 binding to IL-23R using competitive ELISA experiments, identifying Peptide 23 (a 12-mer peptide) as the most active peptide, which significantly inhibited the binding of IL-23 to IL-23R (IC50 = 0.85 µM).

Based on this, the team optimized the structure of Peptide 23, including extending or shortening the C/N termini, replacing internal amino acids with acidic/neutral/basic amino acids, and cyclization modifications. Competitive ELISA results showed that extending amino acids at both ends of Peptide 23 had little effect on its inhibitory activity, but shortening led to a decrease in inhibitory activity. Modifying the charge properties of the "X" residues in the core motif -WX1X2X3W- (e.g., replacing negatively charged residues with other negatively charged, polar uncharged, or positively charged residues) did not significantly affect its inhibitory activity, indicating that the charge characteristics of the residues in this region are not key determinants of binding activity. Replacing one of the tryptophan (W) residues in the core structure with phenylalanine (F) or tyrosine (Y) retained its inhibitory ability, while replacing it with alanine (A) reduced the activity, which is consistent with the sequence characteristics obtained from previous phage display screening.

Furthermore, shortening Peptide 23.15 while retaining the core -WQDYW- structure, although the core motif was preserved, significantly weakened its inhibitory activity. In addition, introducing cysteine (Cys) at both ends to form a cyclic peptide resulted in activity comparable to the linear peptide; however, reducing the size of the cyclic peptide to 7 or 9 peptides while maintaining the core structure significantly weakened its ability to inhibit IL-23R.

In two subsequent patents published by MDL (US20130172272 and US20160039878), researchers used the -WX1X2X3W- core structure as a scaffold and constructed and screened a targeted peptide library using ribosome display technology (peptide library sequence: XXXWXXYWXXXX). After eight rounds of screening, the peptides were further subdivided based on the -WX1X2YW- structure: Group A contained -WVDYW-, Group B contained -WQDYW-, and Group C contained peptides with other core sequences. ELISA activity tests showed that Peptide No. 2/7 in Group A had superior activity compared to Peptide No. 23.19 disclosed in the previous patent. In competitive ELISA tests, Peptide No. 7 had an IC50 of 58.0 nM, and Peptide No. 2 had an IC50 of 34.6 nM.

Subsequently, using Peptide NO.2 and Peptide NO.7 as starting points, three cyclic peptides were designed and synthesized: 2HT-AA, 7-CC, and 7HT-AA. Among these, the molecule 2HT-AA, based on the head-to-tail cyclization of alanine residues at both ends of Peptide NO.2, exhibited the best activity (structure shown in the figure below), with a competitive ELISA IC50 of 8.1 nM and a reporter activity of 341.8 nM (testing method: IL-23 stimulated a DB cell line transfected with a STAT3 reporter gene, and inhibitory activity was quantified by detecting reporter gene expression). This represented a 4-6 fold increase in activity compared to the linear peptide. In in vitro activity tests, 2HT-AA effectively inhibited IL-23-induced IL-17F secretion in human Th17 cells in a concentration-dependent manner and blocked IL-22 production in human PBMCs and rat splenocytes. Simultaneously, this cyclic peptide also downregulated IL-17F mRNA levels in rat splenocytes. Unfortunately, despite the promising pharmacological activity of 2HT-AA, no subsequent optimization reports of this cyclic peptide structure and its derivatives by MDL have been found.