De novo design of artificial IL-2 mimic for the treatment of cancer [PreClinical]

1. IL-2 is a central cytokine that enhances the immune response, with considerable potential for the treatment of cancer.

2. A mimic of IL-2 binds its receptor more tightly and more potently than the natural cytokine, without adverse effects, in mouse models of melanoma and colon cancer.

Evidence Rating Level: 2 [Good]

Study Rundown: As a stimulator of the immune response, the cytokine IL-2 has considerable potential for cancer treatment. Therapeutic use of IL-2 presents several challenges. The main limitation is toxicity. Adverse effects of IL-2 administration are likely mediated by the interaction between IL-2 and the CD25 receptor, while clinically beneficial effects are a result of interactions between IL-2 and the IL-2 receptor β and IL-2 receptor γ (IL-2Rβ and IL-2Rγ).

Previous efforts to increase the efficacy of IL-2 have relied on mutagenesis experiments that introduce amino acid substitutions, or more substantial changes such as additional functional domains to the native protein in an effort to enhance its function. However, these changes can result in foreign epitopes, and cross-reactive antibodies can potentially react with endogenous IL-2. Changes to the native IL-2 structure can also result in altered pharmacodynamics and tissue penetration. Furthermore, IL-2 is only marginally stable at room temperature, making manufacture and storage difficult.

In their landmark study that represents a novel approach to designing protein-based therapeutics, Silva et al optimized a protein called neoleukin-2/15 (Neo-2/15) that circumvents several of the challenges in applying IL-2 therapeutics. Neo-2/15 binds human and mouse IL-2Rβγ with higher affinity and potency than the natural cytokine, and notably does not interact with CD25, avoiding several adverse effects associated with endogenous IL-2 administration. Unlike IL-2, Neo-2/15 is hyperstable, with efficacy in driving T cell survival even after one hour of boiling at 95°C. When administered in mouse models of melanoma and colon cancer, Neo-2/15 enhances tumor killing without exhibiting any detectable immunogenicity. The construction of this IL-2 mimic is particularly remarkable in that Neo-2/15 was constructed entirely from scratch, resulting in a functional mimic with little structural resemblance to the native protein.

Functionally, Neo-2/15 elicits expansion of CD8⁺ killer T cells, with a lower expansion of immunosuppressive T regulatory (T_reg) cells than mouse IL-2, leading to a higher CD8⁺ killer T cell: T_reg ratio. Increases in this ratio are generally associated with better anti-tumor responses. Clinical application of Neo-2/15 will require further testing including evaluation of its stability and immunogenicity in more human-like models. Nevertheless, this artificially engineered cytokine represents the first-of-its-kind in de novo design of a stable protein therapeutic with extensive clinical applications.

Click here to read the study in Nature

Relevant Reading: De novo protein design for therapeutic applications

In-Depth [pre-clinical study]: The objective of this study was to develop a stable IL-2 mimic. De novo protein design relied on computational modeling in which the structural elements (α-helices H1, H3, and H4) of IL-2 that interact with the desired receptor subunits (IL-2Rβγ) were fixed in space, and an idealized de novo globular protein structure was built to support these interactions. These idealized structures were also designed to replace the H2 helix, which interacts with the CD25 surface, an interaction which is thought to mediate IL-2’s adverse effects.

Two generations of mimics were built. Rosetta-based combinatorial flexible backbone sequence design was carried out for each backbone in complex with human IL-2Rβγ, and eight designs were tested for binding to fluorescently tagged IL-2 receptor at low-nanomolar concentrations. The second generation of designs aimed to improve thermal stability of the IL-2 mimic. Helix lengths were parametrically varied, redesigned, and experimentally characterized, followed by site-saturation mutagenesis and selection for high affinity binding to IL-2Rβγ. This process culminated in Neo-2/15, a 100-residue protein with little sequence similarity to human or mouse IL-2 (BLASTP sequence identity of 14% and 24% respectively). The X-ray crystal structure of Neo-2/15 was similar to the computationally predicted design model.

Functional characterization of Neo-2/15 showed high affinity binding to both human and mouse IL-2Rβγ (Kd ~ 19nM and Kd ~ 38nM, respectively), without interaction with CD25. Additionally, Neo-2/15 was more potent in activating human YT-1 cells and mouse primary T cells than native IL-2 (EC₅₀ = 49 pM versus 410pM for human YT-1 cells, EC₅₀ = 130pM versus 30nM for mouse primary T cells). In ex vivo primary cell cultures, Neo-2/15 was still able to drive T cell survival after 60 minutes of boiling at 95°C, demonstrating its high stability.

The therapeutic efficacy of Neo-2/15 was tested in the B16F10 (melanoma) and CT26 (colon cancer) mouse models. Treatment with Neo-2/15 resulted in dose-dependent delays in tumor growth in both models. Little or no immunogenicity was observed in naïve and tumor-bearing mice.

Image: PD

©2019 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.