(685c) Engineering Highly Selective TIMP3-Based Metalloproteinase (MP) Inhibitors

Metalloproteinases (MPs) such as matrix metalloproteinases (MMPs) and a disintegrin and metalloproteinases (ADAMs), play a key role in several diseases like cancer, fibrosis, and neurodegenerative disorders. When MPs are not regulated by activators and inhibitors properly, they can disrupt the extracellular matrix (ECM) turnover, leading to tissue damage and disease progression. Moreover, MPs can impact critical signaling molecules, exacerbating the disease process. Originally, MMPs were known for their contribution to the advancement of tumor growth by changing the extracellular matrix. Yet, certain MMPs have protective functions in the development of cancer. Given the multifaceted role of MPs, developing selective inhibitors to regulate their proteolytic activity in different diseases could be a promising treatment approach. The main challenge lies in discovering inhibitors that can reliably block the proteolytic actions of MPs in a selective manner to prevent off-target effects. In this way, previous attempts with broad-spectrum inhibitors, especially small molecules have caused unintended side effects because they lack specificity. As a result, protein-based inhibitors like tissue inhibitors of metalloproteinases (TIMPs) show great potential for use in therapies since TIMPs play a crucial role in controlling MPs in the body and have successfully regulated enzyme activities linked to many diseases. Currently, there are promising studies to modify TIMP structures using molecular and protein engineering techniques, with a focus on important regions or loops that interact with target MPs to develop highly selective inhibitors. We aim to strategically target important regions or interacting loops to enhance the efficiency of TIMPs, potentially overcoming challenges in targeting pathological MPs such as MMP-9 and ADAM-17 which were shown to play critical roles in several diseases such as specific cancer, neurodegenerative disease, and other inflammations.

We are particularly interested in using TIMP-3 as an initial framework to develop a highly selective protein therapeutic, a protein scaffold targeting specific MPs such as MMP-9 and ADAM-17. TIMP-3 stands out from the other TIMPs with its ability to interact with various components of the extracellular matrix such as signaling molecules and inhibit a wide range of substrates. TIMP-3 also showed the lowest pro-tumorigenic role in cancers. Our observations suggest that TIMP-3 exhibits remarkable flexibility in its loops, facilitating interactions with diverse MPs with varying affinities in the sub-nanomolar range. This adaptability makes TIMP-3 a promising candidate for developing a more targeted variant. We have created a combinatorial library of mutants, by applying mutations on the C-connector and AB-loop loops of TIMP-3. These loops have demonstrated the ability to interact with residues within the conserved active site of the MMPs, as well as adjacent non-conserved residues. Leveraging this insight, the selectivity of this interaction towards MMP-9 and ADAM-17 can be mediated, while concurrently minimizing potential off-target effects for other MPs. To improve selective binding of TIMP-3 variants towards specific MPs and address its broad inhibitory effects, we successfully used a counter-selective screening strategy to eliminate non-binders to non-specific targets while selecting the high affinity binders. We screened this combinatorial library against MMP-9 in the presence of increasing amounts of MMP-3 as a competing binder in sequential rounds of FACS. Enrichment of TIMP-3 C-loop variants towards MMP-9cd and ADAM-17cd was achieved through three successful sequential FACS rounds, followed by next generation sequencing (NGS) analysis for high-affinity binders. The final library demonstrated significant improvement in binding affinity for MMP-9 and ADAM-17 and higher expression levels of TIMP-3 variants, along with reductions in binding to MMP-3. The NGS data revealed key mutations within these flexible loops of TIMP-3 variants, indicating positions where specific amino acids became more frequent across screening rounds, correlating with improved binding and expression. For example, the bioinformatic studies of the NGS data obtained from three rounds of FACS screening showed a consistent decline in the frequency of Cys68 within the C-loop was observed across sorting rounds, potentially influencing stability compared to wild-type TIMP-3. Selected variants showed competitive expression and binding levels up to 4-fold higher compared to the N-terminal domain of TIMP-1 (N-TIMP1) used as a positive control as a strong inhibitor of MMP-9 with inhibition constant in picomolar range. The underlying mechanism for selective inhibition to MMP-9 and ADAM-17 derived by specific epitopes in TIMP-3 flexible loops has not been fully explored. This lack of sequence-structure-function understanding presents a significant obstacle to identifying TIMP-3's interaction regions with specific target MPs and hinders efforts to engineer proteins using the TIMP-3 as a protein scaffold. We aim to study structural clues of TIMP-3 protein and its top variants in complex with MMP-9 and ADAM-17 to shed light on the molecular mechanisms of enzyme binding and inhibition, providing insights for protein design and engineering of the next generation of MP inhibitors. Moreover, a potential challenge we might face is identifying TIMP-3 variants with increased binding affinity to MMP-9 while showing reduced attraction to other MPs. Thus, we expanded our targeted design to other interactive loops such as the AB-loop in the N-terminal region of TIMP-3, to improve specificity in binding. This decision is influenced by the fact that the AB-loop interacts with less conserved regions near the active site, as seen in other TIMP/MMP complexes. We will present the results of screening, testing and analyzing highly selective TIMP-3 mutants to MMP-9 and ADAM-17 as potential therapeutic protein for the diseases where these MPs have a pathological role.