PD-1

POM

Programmed cell death protein 1 (PD-1) is an immune checkpoint receptor critical for modulating the body’s immune system. The binding of PD-1 with its native ligand, Programmed death ligand 1 (PD-L1), which is over-expressed in the tumor microenvironment, can impede the immune system’s ability to destroy tumor cells. This makes PD-1 a well-established target for cancer treatment.

Several monoclonal antibodies (mAbs) designed to block the PD-1/PD-L-1 interaction have been developed in recent years - a therapeutic approach known as immune checkpoint therapy (ICT). This strategy has proved highly successful, leading to the widespread clinical use of various PD-1 targeting mAbs, with more currently being investigated.

The 3D structure of the PD-1 complex with the FDA-approved mAb cemiplimab (co-developed by Sanofi and Regeneron) has been recently solved. This study revealed the molecular basis of binding and highlighted the importance of PD-1 glycosylation in the therapeutic efficacy of cemiplimab. Such insights could be exploited to develop innovative cancer treatments.

The scientists determined the X-ray crystal structure of the complex of PD-1 extracellular domain with the single-chain variable fragment (svFv) of cemiplimab. Their analysis showed that the mAb primarily binds the PD-1 BC and FG loops (Image 1A). When comparing the cemiplimab/PD-1 complex with the structure of PD-1 bound to its native ligand PD-L1, they observed a significant overlap in the binding surface (Image 1B). Specifically, the variable light chain (VL) of cemiplimab sterically hinders the region where PD-L1 binds, thus inhibiting the PD-L1 mediated signaling.

 

Image 1. A) 3D structure of the PD-1/cemiplimab complex (PDB: 7WVM; PD-1 in pink, cemiplimab heavy variable chain (VH) in red, light variable chain (VL) in green). BC and FG loops on PD-1 are indicated by the pink arrows. Protein-protein interactions (PPIs) between cemiplimab and PD-1 are indicated as dashed lines and calculated with the software 3decision®. B) Overlay of the PD-1/cemiplimab complex (PDB: 7WVM; PD-1 in pink, cemiplimab heavy variable chain (VH) in red, light variable chain (VL) in green) and of the PD-1/PD-L1 complex structure (PDB: 4ZQK; PD-1 in yellow, PD-L1 in white). Superposition by chain is performed using the PD-1 chain as a reference for the alignment. BC and FG loops on PD-1 are indicated by the pink arrows as a reference. All pictures are produced with the 3decision® software.

 

Interested to know more about the protein-protein interactions display in 3decision? Check out this short video.


Comparative structural analysis with other PD-1/mAb complexes showed that cemiplimab binds PD-1 similarly to camrelizumab and MW11-317 (Image 2A), but differently from other antibodies like nivolumab and pembrolizumab (Image 2B).

 

Image 2. Overlap of the 3D structure of PD-1/cemiplimab complex (PDB: 7WVM, pink) with the PD-1 complex with the following mABs: A) camrelizumab (PDB: 7CU5, orange) and MW11-H317 (PDB: 6JJP, white); B) pembrolizumab (PDB: 5JXE, green) and nivolumab (PDB: 5WT9, cyan). Superposition is performed using the PD-1 chain as a reference for the alignment. Pictures are produced with the 3decision® software.

 

Interestingly, the efficacy of mAbs that share the same binding mode as cemiplimab is known to be influenced by glycosylation of PD-1 at the Asn58 on the BC loop. Therefore, the scientists analyzed the effect of PD-1 glycosylation on cemiplimab binding.

Biochemical analyses confirmed that the glycosylation of PD-1 at Asn58 was crucial for cemiplimab binding. In the 3D crystal structure, the PD-1 in complex with cemiplimab was not glycosylated due to the expression system used for producing the protein. Nevertheless, the cemiplimab’s heavy chain complementary region 2 (HCDR2) was found to be spatially close to Asn58, and the overall conformation of the PD-1/cemiplimab interface clearly resembled that of other glycosylation-dependent antibodies (Image 3). This provides a structural explanation of why glycosylation at this site is so crucial for cemiplimab binding.

 

Image 3. Overlap of the 3D structure of PD-1/cemiplimab complex (PDB: 7WVM, pink) with the PD-1/camrelizumab complex (PDB: 7CU5, white, glycan chain is green). Superposition is performed using the PD-1 BC loop as a reference for the alignment. The BC loop and Asn58 locations on PD-1 are indicated with a yellow arrow, the position of the HCDR2 on the mAbs is indicated with a light blue arrow. Picture produced with the 3decision® software.

 

This work showed how the 3D structure of the antigen-antibody complex elucidated the inhibition mechanism of the mAb and provided valuable insights into the role of PD-1 glycosylation in developing PD-1 targeting mAbs.


Reference: Lu D, Xu Z, Zhang D, Jiang M, Liu K, He J, Ma D, Ma X, Tan S, Gao GF, Chai Y. PD-1 N58-Glycosylation-Dependent Binding of Monoclonal Antibody Cemiplimab for Immune Checkpoint Therapy. Front Immunol. 2022 Mar 2;13:826045. doi: 10.3389/fimmu.2022.826045. PMID: 35309324; PMCID: PMC8924070.


Curious about more structural stories on targeting protein-protein interactions? Read our Protein of the Month featuring β-catenin.

Previous
Previous

HER2

Next
Next

CCR8