Exploiting protease activation for therapy

Integrating Computational Design and Experimental Approaches for Next-Generation Biologics

Therapeutic protein engineering has revolutionized medicine by enabling the development of highly specific and potent treatments for a wide range of diseases. This review examines recent advances in computational and experimental approaches for engineering improved protein therapeutics. Key areas of focus include antibody engineering, enzyme replacement therapies, and cytokine-based drugs. Computational methods like structure-based design, machine learning integration, and protein language models have dramatically enhanced our ability to predict protein properties and guide engineering efforts. Experimental techniques such as directed evolution and rational design approaches continue to evolve, with high-throughput methods accelerating the discovery process. Applications of these methods have led to breakthroughs in affinity maturation, bispecific antibodies, enzyme stability enhancement, and the development of conditionally active cytokines. Emerging approaches like intracellular protein delivery, stimulus-responsive proteins, and de novo designed therapeutic proteins offer exciting new possibilities. However, challenges remain in predicting in vivo behavior, scalable manufacturing, immunogenicity mitigation, and targeted delivery. Addressing these challenges will require continued integration of computational and experimental methods, as well as a deeper understanding of protein behavior in complex physiological environments. As the field advances, we can anticipate increasingly sophisticated and effective protein therapeutics for treating human diseases.

Site-Specific Antibody Prodrugs via S-Arylation: a Bioconjugation Approach Toward Masked Tyrosine Analogues

The utility of antibody therapeutics is hampered by potential cross-reactivity with healthy tissue. Over the past decade, significant advances have been made in the design of activatable antibodies, which increase, or create altogether, the therapeutic window of a parent antibody. Of these, antibody prodrugs (pro-antibodies) are masked antibodies that have advanced the most for therapeutic use. They are designed to reveal the active, parent antibody only when encountering proteases upregulated in the microenvironment of the targeted disease tissue, thereby minimizing off-target activity. However, current pro-antibody designs are relegated to fusion proteins that append masking groups restricted to the use of only canonical amino acids, offering excellent control of the site of introduction, but with no authority over where the masking group is installed other than the N-terminus of the antibody. Here, we present a palladium-based bioconjugation approach for the site-specific introduction of a masked tyrosine mimic in the complementary determining region of the FDA approved antibody therapeutic ipilimumab used as a model system. The approach enables the introduction of a protease cleavable group tethered to noncanonical polymers (polyethylene glycol (PEG)) resulting in 47-fold weaker binding to cells expressing CTLA-4, the target antigen of ipilimumab. Upon exposure to tumor-associated proteases, the masking group is cleaved, unveiling a tyrosine-mimic (dubbed hydroxyphenyl cysteine (HPC)) that restores (>90% restoration) binding affinity to its target antigen.

Exploring the next generation of antibody-drug conjugates

Antibody-drug conjugates (ADCs) are a promising cancer treatment modality that enables the selective delivery of highly cytotoxic payloads to tumours. However, realizing the full potential of this platform necessitates innovative molecular designs to tackle several clinical challenges such as drug resistance, tumour heterogeneity and treatment-related adverse effects. Several emerging ADC formats exist, including bispecific ADCs, conditionally active ADCs (also known as probody-drug conjugates), immune-stimulating ADCs, protein-degrader ADCs and dual-drug ADCs, and each offers unique capabilities for tackling these various challenges. For example, probody-drug conjugates can enhance tumour specificity, whereas bispecific ADCs and dual-drug ADCs can address resistance and heterogeneity with enhanced activity. The incorporation of immune-stimulating and protein-degrader ADCs, which have distinct mechanisms of action, into existing treatment strategies could enable multimodal cancer treatment. Despite the promising outlook, the importance of patient stratification and biomarker identification cannot be overstated for these emerging ADCs, as these factors are crucial to identify patients who are most likely to derive benefit. As we continue to deepen our understanding of tumour biology and refine ADC design, we will edge closer to developing truly effective and safe ADCs for patients with treatment-refractory cancers. In this Review, we highlight advances in each ADC component (the monoclonal antibody, payload, linker and conjugation chemistry) and provide more-detailed discussions on selected examples of emerging novel ADCs of each format, enabled by engineering of one or more of these components.

Next generation of multispecific antibody engineering

Multispecific antibodies recognize two or more epitopes located on the same or distinct targets. This added capability through protein design allows these man-made molecules to address unmet medical needs that are no longer possible with single targeting such as with monoclonal antibodies or cytokines alone. However, the approach to the development of these multispecific molecules has been met with numerous road bumps, which suggests that a new workflow for multispecific molecules is required. The investigation of the molecular basis that mediates the successful assembly of the building blocks into non-native quaternary structures will lead to the writing of a playbook for multispecifics. This is a must do if we are to design workflows that we can control and in turn predict success. Here, we reflect on the current state-of-the-art of therapeutic biologics and look at the building blocks, in terms of proteins, and tools that can be used to build the foundations of such a next-generation workflow.

In silico evolution of autoinhibitory domains for a PD-L1 antagonist using deep learning models

There has been considerable progress in the development of computational methods for designing protein-protein interactions, but engineering high-affinity binders without extensive screening and maturation remains challenging. Here, we test a protein design pipeline that uses iterative rounds of deep learning (DL)-based structure prediction (AlphaFold2) and sequence optimization (ProteinMPNN) to design autoinhibitory domains (AiDs) for a PD-L1 antagonist. With the goal of creating an anticancer agent that is inactive until reaching the tumor environment, we sought to create autoinhibited (or masked) forms of the PD-L1 antagonist that can be unmasked by tumor-enriched proteases. Twenty-three de novo designed AiDs, varying in length and topology, were fused to the antagonist with a protease-sensitive linker, and binding to PD-L1 was measured with and without protease treatment. Nine of the fusion proteins demonstrated conditional binding to PD-L1, and the top-performing AiDs were selected for further characterization as single-domain proteins. Without any experimental affinity maturation, four of the AiDs bind to the PD-L1 antagonist with equilibrium dissociation constants (KDs) below 150 nM, with the lowest KD equal to 0.9 nM. Our study demonstrates that DL-based protein modeling can be used to rapidly generate high-affinity protein binders.

Impact of Propeptide Cleavage on the Stability and Activity of a Streptococcal Immunomodulatory C5a Peptidase for Biopharmaceutical Development

Posttranslational modifications of proteins can impact their therapeutic efficacy, stability, and potential for pharmaceutical development. The Group AStreptococcus pyogenesC5a peptidase (ScpA) is a multi-domain protein composed of an N-terminal signal peptide, a catalytic domain (including propeptide), three fibronectin domains, and cell membrane-associated domains. It is one of several proteins produced by Group AS. pyogenesknown to cleave components of the human complement system. After signal peptide removal, ScpA undergoes autoproteolysis and cleaves its propeptide for full maturation. The exact location and mechanism of the propeptide cleavage, and the impact of this cleavage on stability and activity, are not clearly understood, and the exact primary sequence of the final enzyme is not known. A form of ScpA with no autoproteolysis fragments of propeptide present may be more desirable for pharmaceutical development from a regulatory and a biocompatibility in the body perspective. The current study describes an in-depth structural and functional characterization of propeptide truncated variants of ScpA expressed inEscherichia colicells. All three purified ScpA variants, ScpA, 79ΔPro, and 92ΔPro, starting with N32, D79, and A92 positions, respectively, showed similar activity against C5a, which suggests a propeptide-independent activity profile of ScpA. CE-SDS and MALDI top-down sequencing analyses highlight a time-dependent propeptide autoproteolysis of ScpA at 37 °C with a distinct end point at A92 and/or D93. In comparison, all three variants of ScpA exhibit similar stability, melting temperatures, and secondary structure orientation. In summary, this work not only highlights propeptide localization but also provides a strategy to recombinantly produce a final mature and active form of ScpA without any propeptide-related fragments.

Engineering cytokines for cancer immunotherapy: a systematic review

Cytokines are pivotal mediators of cell communication in the tumor microenvironment. Multiple cytokines are involved in the host antitumor response, but the production and function of these cytokines are usually dysregulated during malignant tumor progression. Considering their clinical potential and the early successful use of cytokines in cancer immunotherapy, such as interferon alpha-2b (IFNα-2b; IntronA^®) and IL-2 (Proleukin^®), cytokine-based therapeutics have been extensively evaluated in many follow-up clinical trials. Following these initial breakthroughs, however, clinical translation of these natural messenger molecules has been greatly limited owing to their high-degree pleiotropic features and complex biological properties in many cell types. These characteristics, coupled with poor pharmacokinetics (a short half-life), have hampered the delivery of cytokines via systemic administration, particularly because of severe dose-limiting toxicities. New engineering approaches have been developed to widen the therapeutic window, prolong pharmacokinetic effects, enhance tumor targeting and reduce adverse effects, thereby improving therapeutic efficacy. In this review, we focus on the recent progress and competitive landscape in cytokine engineering strategies and preclinical/clinical therapeutics for cancer. In addition, aiming to promote engineered cytokine-based cancer immunotherapy, we present a profound discussion about the feasibility of recently developed methods in clinical medicine translation.

Preclinical Assessment of ADAM9-Responsive Mesoporous Silica Nanoparticles for the Treatment of Pancreatic Cancer

Pancreatic adenocarcinoma (PDAC) remains largely refractory to chemotherapeutic treatment regimens and, consequently, has the worst survival rate of all cancers. The low efficacy of current treatments results largely from toxicity-dependent dose limitations and premature cessation of therapy. Recently, targeted delivery approaches that may reduce off-target toxicities have been developed. In this paper, we present a preclinical evaluation of a PDAC-specific drug delivery system based on mesoporous silica nanoparticles (MSNs) functionalized with a protease linker that is specifically cleaved by PDAC cells. Our previous work demonstrated that ADAM9 is a PDAC-enriched protease and that paclitaxel-loaded ADAM9-responsive MSNs effectively kill PDAC cells in vitro. Here, we show that paclitaxel-loaded ADAM9-MSNs result in off-target cytotoxicity in clinically relevant models, which spurred the development of optimized ADAM9-responsive MSNs (OPT-MSNs). We found that these OPT-MSNs still efficiently kill PDAC cells but, as opposed to free paclitaxel, do not induce death in neuronal or bone marrow cells. In line with these in vitro data, paclitaxel-loaded OPT-MSNs showed reduced organ damage and leukopenia in a preclinical PDAC xenograft model. However, no antitumor response was observed upon OPT-MSN administration in vivo. The poor in vivo antitumor activity of OPT-MSNs despite efficient antitumor effects in vitro highlights that although MSN-based tumor-targeting strategies may hold therapeutic potential, clinical translation does not seem as straightforward as anticipated.

In silico evolution of protein binders with deep learning models for structure prediction and sequence design

There has been considerable progress in the development of computational methods for designing protein-protein interactions, but engineering high-affinity binders without extensive screening and maturation remains challenging. Here, we test a protein design pipeline that uses iterative rounds of deep learning (DL)-based structure prediction (AlphaFold2) and sequence optimization (ProteinMPNN) to design autoinhibitory domains (AiDs) for a PD-L1 antagonist. Inspired by recent advances in therapeutic design, we sought to create autoinhibited (or masked) forms of the antagonist that can be conditionally activated by proteases. Twenty-three de novo designed AiDs, varying in length and topology, were fused to the antagonist with a protease sensitive linker, and binding to PD-L1 was tested with and without protease treatment. Nine of the fusion proteins demonstrated conditional binding to PD-L1 and the top performing AiDs were selected for further characterization as single domain proteins. Without any experimental affinity maturation, four of the AiDs bind to the PD-L1 antagonist with equilibrium dissociation constants (KDs) below 150 nM, with the lowest KD equal to 0.9 nM. Our study demonstrates that DL-based protein modeling can be used to rapidly generate high affinity protein binders.

Design of a protease-activated PD-L1 inhibitor

Immune checkpoint inhibitors that bind to the cell surface receptor PD-L1 are effective anti-cancer agents but suffer from immune-related adverse events as PD-L1 is expressed on both healthy and cancer cells. To mitigate toxicity, researchers are testing prodrugs that have low affinity for checkpoint targets until activated with proteases enriched in the tumor microenvironment. Here, we engineer a prodrug form of a PD-L1 inhibitor. The inhibitor is a soluble PD-1 mimetic that was previously engineered to have high affinity for PD-L1. In the basal state, the binding surface of the PD-1 mimetic is masked by fusing it to a soluble variant of its natural ligand, PD-L1. Proteolytic cleavage of the linker that connects the mask to the inhibitor activates the molecule. To optimize the mask so that it effectively blocks binding to PD-L1 but releases upon cleavage, we tested a set of mutants with varied affinity for the inhibitor. The top-performing mask reduces the affinity of the prodrug for PD-L1 120-fold, and binding is nearly fully recovered upon cleavage. In a cell-based assay measuring inhibition of the PD-1:PD-L1 interaction on the surface of cells, the IC50s of the masked inhibitors were up to 40-fold higher than their protease-treated counterparts. The changes in activity we observe upon protease treatment are comparable to systems currently tested in the clinic and provide evidence that natural binding partners are an excellent starting point for creating a prodrug.

A lesion-selective albumin-CTLA4Ig as a safe and effective treatment for collagen-induced arthritis

BACKGROUND

CTLA4Ig is a dimeric fusion protein of the extracellular domain of cytotoxic T-lymphocyte protein 4 (CTLA4) and an Fc (Ig) fragment of human IgG₁ that is approved for treating rheumatoid arthritis. However, CTLA4Ig may induce adverse effects. Developing a lesion-selective variant of CTLA4Ig may improve safety while maintaining the efficacy of the treatment.

METHODS

We linked albumin to the N-terminus of CTLA4Ig (termed Alb-CTLA4Ig) via a substrate sequence of matrix metalloproteinase (MMP). The binding activities and the biological activities of Alb-CTLA4Ig before and after MMP digestion were analyzed by a cell-based ELISA and an in vitro Jurkat T cell activation assay. The efficacy and safety of Alb-CTLA4Ig in treating joint inflammation were tested in mouse collagen-induced arthritis.

RESULTS

Alb-CTLA4Ig is stable and inactive under physiological conditions but can be fully activated by MMPs. The binding activity of nondigested Alb-CTLA4Ig was at least 10,000-fold weaker than that of MMP-digested Alb-CTLA4Ig. Nondigested Alb-CTLA4Ig was unable to inhibit Jurkat T cell activation, whereas MMP-digested Alb-CTLA4Ig was as potent as conventional CTLA4Ig in inhibiting the T cells. Alb-CTLA4Ig was converted to CTLA4Ig in the inflamed joints to treat mouse collagen-induced arthritis, showing similar efficacy to that of conventional CTLA4Ig. In contrast to conventional CTLA4Ig, Alb-CTLA4Ig did not inhibit the antimicrobial responses in the spleens of the treated mice.

CONCLUSIONS

Our study indicates that Alb-CTLA4Ig can be activated by MMPs to suppress tissue inflammation in situ. Thus, Alb-CTLA4Ig is a safe and effective treatment for collagen-induced arthritis in mice.

Evolution of protease activation and specificity via alpha-2-macroglobulin-mediated covalent capture

Tailoring of the activity and specificity of proteases is critical for their utility across industrial, medical and research purposes. However, engineering or evolving protease catalysts is challenging and often labour intensive. Here, we describe a generic method to accelerate this process based on yeast display. We introduce the protease selection system A2M^cap that covalently captures protease catalysts by repurposed alpha-2-macroglobulin (A2Ms). To demonstrate the utility of A2M^cap for protease engineering we exemplify the directed activity and specificity evolution of six serine proteases. This resulted in a variant of Staphylococcus aureus serin-protease-like (Spl) protease SplB, an enzyme used for recombinant protein processing, that no longer requires activation by N-terminal signal peptide removal. SCHEMA-based domain shuffling was used to map the specificity determining regions of Spl proteases, leading to a chimeric scaffold that supports specificity switching via subdomain exchange. The ability of A2M^cap to overcome key challenges en route to tailor-made proteases suggests easier access to such reagents in the future.

Proteases and Their Potential Role as Biomarkers and Drug Targets in Dry Eye Disease and Ocular Surface Dysfunction

Dry eye disease (DED) is a multifactorial disorder that leads to ocular discomfort, visual disturbance, and tear film instability. DED is accompanied by an increase in tear osmolarity and ocular surface inflammation. The diagnosis and treatment of DED still present significant challenges. Therefore, novel biomarkers and treatments are of great interest. Proteases are present in different tissues on the ocular surface. In a healthy eye, proteases are highly regulated. However, dysregulation occurs in various pathologies, including DED. With this review, we provide an overview of the implications of different families of proteases in the development and severity of DED, along with studies involving protease inhibitors as potential therapeutic tools. Even though further research is needed, this review aims to give suggestions for identifying novel biomarkers and developing new protease inhibitors.

The Vascular Endothelium and Coagulation: Homeostasis, Disease, and Treatment, with a Focus on the Von Willebrand Factor and Factors VIII and V

The vascular endothelium has several important functions, including hemostasis. The homeostasis of hemostasis is based on a fine balance between procoagulant and anticoagulant proteins and between fibrinolytic and antifibrinolytic ones. Coagulopathies are characterized by a mutation-induced alteration of the function of certain coagulation factors or by a disturbed balance between the mechanisms responsible for regulating coagulation. Homeostatic therapies consist in replacement and nonreplacement treatments or in the administration of antifibrinolytic agents. Rebalancing products reestablish hemostasis by inhibiting natural anticoagulant pathways. These agents include monoclonal antibodies, such as concizumab and marstacimab, which target the tissue factor pathway inhibitor; interfering RNA therapies, such as fitusiran, which targets antithrombin III; and protease inhibitors, such as serpinPC, which targets active protein C. In cases of thrombophilia (deficiency of protein C, protein S, or factor V Leiden), treatment may consist in direct oral anticoagulants, replacement therapy (plasma or recombinant ADAMTS13) in cases of a congenital deficiency of ADAMTS13, or immunomodulators (prednisone) if the thrombophilia is autoimmune. Monoclonal-antibody-based anti-vWF immunotherapy (caplacizumab) is used in the context of severe thrombophilia, regardless of the cause of the disorder. In cases of disseminated intravascular coagulation, the treatment of choice consists in administration of antifibrinolytics, all-trans-retinoic acid, and recombinant soluble human thrombomodulin.

Mechanistic insights into the rational design of masked antibodies

Although monoclonal antibodies have greatly improved cancer therapy, they can trigger side effects due to on-target, off-tumor toxicity. Over the past decade, strategies have emerged to successfully mask the antigen-binding site of antibodies, such that they are only activated at the relevant site, for example, after proteolytic cleavage. However, the methods for designing an ideal affinity-based mask and what parameters are important are not yet well understood. Here, we undertook mechanistic studies using three masks with different properties and identified four critical factors: binding site and affinity, as well as association and dissociation rate constants, which also played an important role. HDX-MS was used to identify the location of binding sites on the antibody, which were subsequently validated by obtaining a high-resolution crystal structure for one of the mask-antibody complexes. These findings will inform future designs of optimal affinity-based masks for antibodies and other therapeutic proteins.