[VDP-XXX] Gian De Nicola (KCL) - p38-TAB1 Inhibitors

One-liner: Dr De Nicola is developing small molecule inhibitors for non-canonical p38 activation (TAB1-facilitated autoactivation) with the potential to target numerous age-related disorders, including inflammaging, senescence, and modulating stress responses. He is now seeking funding for hit-to-lead development.

Note: This proposal is based on the supporting documents provided by Gian De Nicola, as well as questions by VitaDAO’s squad and answers from Gian.*

Note Because the terms of funding are still under discussion, the purpose of publishing this proposal is solely to gather feedback from the community, for the time being.

Longevity Dealflow WG team

  • Senior Reviewers: Review not yet launched
  • Shepherd: Paolo Binetti
  • Other squad members: Rhys Anderson, Ryan Spangler
  • Sourced by: Rhys Anderson

Project PI

Gian De Nicola PhD, lecturer at King’s College London (KCL)

Simple Summary

P38 is a stress-inducible kinase which can become dysregulated with age, with persistent activation being shown to play a role in the development of numerous pathologies from senescence induction to neurodegeneration.

Previous attempts to develop inhibitors have failed due to toxicity associated with directly inhibiting p38 kinase activity. However, inhibition of non-canonical p38 activation by TAB1- facilitated autoactivation avoids this issue.

Gian identified the site of interaction by solving the p38-TAB1 crystal structure and made a TAB1 KI mutant mouse which perturbed the p38-TAB1 interaction, had no side effects and could attenuate cardiac ischaemia/reperfusion injury. Others have shown that this interaction is harmful in different settings including in human T-cells where it drives senescence and in skin where the interaction drives inflammation.

Gian has performed a fragment screen which resulted in ~6 scaffolds with 20-60uM binding affinity and is now looking for funding to do hit-to-lead development

Problem

There is substantial interest in the medical field in trying to inhibit p38. These inhibitors have been developed for several therapeutic areas ranging from cardiovascular indications, inflammatory conditions to neurodegenerative pathologies, and metabolic disorders.

However, the clinical trials that have been published to date, those that are ongoing and those that are planned, all use pharmacological inhibitors of p38 kinase activity which cause blanket/systemic inhibition of the kinase activity. This past experience revealed skin and liver toxicity. Adverse effects that are shared between inhibitors based on different scaffolds and with diverse modes of binding to the p38 active site. Based on the biological importance of this kinase, its ubiquitous expression and the shared toxicity profile of diverse inhibitors, the assumption is that the side effects are a manifestation of on-target toxicity.

Consequently, current trials are carefully designed to achieve only partial p38 inhibition at peak concentration. Despite this care in the controlled environment of a clinical trial, liver toxicity is still evident. Thus, p38 inhibition is attractive in several settings, however, toxicity is still apparent despite dosing regimens which have been carefully titrated / compromised to cause only partial / suboptimal inhibition.

An alternative strategy to develop p38 pathway/circumstance specific inhibitors is therefore needed.

Opportunity

In order to tap into the clinical potential of p38 inhibition while avoiding the side effects that plague it, Gian has identified the p38-TAB1 interaction as a more specific target. In confirmation of its relevance, Gian has published that inhibition of p38MAPK:TAB1 with a genetic mouse model can attenuate ischaemia-reperfusion injury.

Concretely, p38α a mitogen activated kinase interacts with TAB1, a scaffold protein, which promotes p38α autoactivation; active p38α (pp38α) then trans-phosphorylates TAB1. The role of TAB1 is therefore twofold: first it acts as an activator of p38α, secondly as a substrate of active p38α.

Previously, Gian solved the X-ray structures of the p38α-TAB1 complex both in their phosphorylated and non-phosphorylated states, described the structural rearrangements induced by TAB1 binding and identified a surface on p38α that is specific for the TAB1 interaction.

Next, Gian developed a new approach underpinned by a novel technology that combines X-ray crystallography and fragment screening that has allowed the identification of a novel class of small-molecule protein-protein interaction inhibitors able to disrupt the interaction between p38α and TAB1.

A single fragment in crystal X-ray based screen identified several fragments binding in the target region localised to 3 discrete, but adjacent, fragment binding sites overlapping the TAB1 binding site. A chemical programme using simultaneously a merging and growing approach of the initial fragments, produced a well-characterised scaffold that is chemically tractable and has micromolar affinity against the target as measured using isothermal titration calorimetry.

N K d (µM) ΔG(Kcal/mol) ΔH(Kcal/mol) -TΔS(Kcal/mol)
p38α-K914 1.2±0.1 60±8 -5.7 -0.9±0.2 -4.8
p38α-K1410 1.2±0.2 112±22 -5.4 -0.8±0.1 -4.6
KCL1337 0.9 ± 0.1 57.8 ± 9.3 –5.8 –0.8 ± 0.1 –5.0 ± 0.3
KCL1088 1.3 ± 0.1 25.1 ± 5.5 –6.3 –0.7 ± 0.1 –5.6 ± 0.1

X-ray structure of active p38α (surface representation in magenta) in complex with TAB1 (ribbon representation in cyan). Zoom in of the non-canonical site to highlight the overlap between TAB1 and the fragments binding to site F1 and F2.

Gian now seeks funding to continue developing small molecules with improved efficacy, leading to novel composition of matter IP.

Experimental plan and budget

The purpose of this application is to start a medicinal chemistry effort to improve the efficacy of the newly identified ligands and to test their value in biophysical assays.

Protein-ligand binding free energy calculations based on atomistic molecular dynamics simulations: $25k

Through a fee for service model, the laboratory of professor Franca Fraternali will run an in-silico screen informed by the available structural and biophysical data to design ligands with improved binding affinity.

Ligand synthesis : $25k

Guided by the outcome of the in-silico screen the de novo synthesis of ~10-15 ligands will be commissioned. Based on previous experience, on average the de-novo synthesis on a 10mg scale with >90% purity is around 2000 euros per ligand.

In vitro binding affinity and efficacy assays:

Recombinant synthesis and purification of p38alphaMAPK & TAB1 proteins. Manual/automated isothermal titration calorimetry (ITC) will be used to measure the binding affinity of the ligands whereas IVKA, which will use the autophoshorylation of p38alpha as a read out, will be used to assess efficacy.

Research Assistant 12 months: $95k

S/he will carry out the protein production and the biophysical and biochemical assays.

Consumables: $22k

Reagents for the expression and purification of the proteins and biophysical/biochemical assays.

Directly allocated and Indirect cost: $65k

These are the overheads as calculated last year.

Future

If successful in our programme we will identify lead molecules able to inhibit p38 autoactivation. Autoactivation of p38, both in isolation and TAB1 mediated, has been shown to be harmful in several disease settings. The next step of the project would be to seek funding to test whether the ligands we develop have any allosteric effect on TAB1 independent activation of p38 as well. This will significantly expand the therapeutic areas beyond ischaemia reperfusion injury, chronic low-grade inflammation and immune cell senescence. Then further funding will be required to test the ligands in each disease relevant model.

VitaDAO Funding Terms

Sponsored Development Agreement with terms to be agreed with TTO. Discussion ongoing.

Relevance to longevity

Numerous age-related factors can lead to chronic p38 activation, including elevated oxidative stress, DNA damage, inflammation, paracrine senescence, mitochondrial dysfunction and telomere attrition. Research from Judith Campisi’s lab has shown that senescence induction leads to persistent p38 activation, which if inhibited can attenuate the senescence associated secretory phenotype (SASP). Furthermore, TAB1-dependent activation of p38 can drive T-cell senescence. Numerous age-related diseases are associated with p38 activation such as neurodegeneration, cardiovascular disease osteoarthritis and sarcopenia amongst others.

p38 and aging

The role of p38 in the pathology of aging is complex and multifaceted. P38 is a type of protein kinase that plays a crucial role in cellular stress response and inflammation. It is involved in various cellular processes, including cell cycle regulation, apoptosis, senescence, antioxidant, and immune response. During the aging process, p38 signaling can become dysregulated, leading to increased activation and chronic inflammation. This chronic low-grade inflammation, also known as inflammaging, is a hallmark of aging and is associated with the development of age-related diseases. It can contribute to the accumulation of oxidative stress, DNA damage, and the decline of cellular repair mechanisms. P38 activation has also been linked to cellular senescence, a state of irreversible growth arrest associated with aging.

p38-TAB1 and aging

Inflammation regulation: p38-TAB1 inhibition can potentially help alleviate chronic inflammation associated with aging and age-related diseases. Excessive activation of p38-TAB1 signaling can contribute to chronic low-grade inflammation (inflammaging), so inhibiting this pathway may help dampen inflammation. The p38-TAB1 pathway is involved in cellular stress response mechanisms. Inhibiting p38-TAB1 may reduce cellular stress and the accumulation of oxidative damage, which can have beneficial effects on cellular function and overall health.

Team

Dr. Gian De Nicola is a lecturer in the Randall Centre for Cell & Molecular Biophysics at King’s College London. He is an expert in structural biology with a long lasting interest in p38alpha signalling in the heart. His group has solved the crystal structure of the p38-TAB1 complex and identified the TAB1 competitive ligands.

Prof Franca Fraternali is Professor of Integrative Computational Biology. She received her PhD in Physical Chemistry from the University of Naples, spending part of the PhD at the Polytechnic of Zurich (ETH). After post-doctoral experiences at the ETH and at the EMBL, in 2000, she became staff scientist at the Mathematical Biology Division of the National Institute for Medical Research in London. In 2012 she was awarded the Chair of Bioinformatics and Computational Biology. The group develops computational methods and large-scale data analysis tools for Structural Biology of Proteins and Nucleic acids; Systems Biology; Molecular Dynamics of folded and misfolded proteins; Statistical Analysis of Protein Interaction Networks. In 2022, Franca took up the position of Director of the Institute of Structural and Molecular Biology (a joint Institute between UCL and Birkbeck) and a Chair in the Division of Biosciences at UCL.

Strengths

  • Published role for the p38-TAB1 interaction in ischaemia-reperfusion injury, T-cell senescence, and skin inflammation, and has the potential to impact numerous age-related disorders. Also pre- and post-transplant organ ischemia is a potential avenue.
  • Promising hits that can be quickly advanced in a hit-to-lead campaign.
  • Potential for novel composition of matter patent filing by the end of the project
  • There are some approved drugs with the identified functional group.

Risks

  • Whilst there is a lot of literature on p38 autoactivation, there are only a few papers on the clinical or aging relevance of the TAB1-p38 protein-protein interaction itself.
  • Clinical trials for anything related to myocardial infarction or any relevant acute cardiac event are challenging to enrol for, and patient numbers need to be high. Most patients come into the ER, where time is prioritised over anything else.
  • There may be a difficulty in identifying other downstream preclinical models, but that’s in the future, when the target will likely be more flushed out.
  • Early-stage, solo venture

Given the high risk nature, I think the terms with TTO will be the most important part.

For the $25k cost, would it be more cost effective to screen a small molecule library with compounds related to the ones with micromolar affinity for a cell functional readout? Then you get functional data on your leads, and cell permeable molecules.

The need is to 1000-1000000x the affinity. I would assume off-targets are horrible at the amounts currently needed to dose.

If the goal is a patent in the US, a peptide or biologic would be better than a small molecule. Longer period of exclusivity. Think immunizing an alpaca or llama and getting the phage display to isolate nanobody sequences that disrupt this interaction would also be ~$20k. And then the B cells do all the chemistry for you. Plus that opens up a lot more targeting-- think mRNA for the nanobody in targeted liposomes or conjugation to cell penetrating peptides, ligands, etc. And the structural PI gets to solve the structures with the nanobody, which is more fun than some tiny little molecule.

There’s also the concern of cell accessibility. If the lead works, but fails to get into cells, it’s the delivery rabbit hole.

We have a wealth of X-ray structures of fragments bound to the target area, DOI: 10.1021/acs.jmedchem.0c00403 which gives us a good understanding of the chemical features needed for the interaction, in a sense we have already done the chemistry and this is the reason we prefer a small molecules approach versus a nanobody one.

The scaffold we are using is present in some approved drugs which makes us confident we will avoid the delivery rabbit hole.

The molecules we have identified are novel chemical entities, to narrow down as much as possible the number of the de-novo synthesis and the costs associated, we have decided to use a bioinformatic approach informed by the structural and biophysical data. The p38-TAB1 interaction is in the low micromolar range which makes a dosing in the tens of nanomolar range realistic.

1 Like