Fragments in the clinic: Lirafugratinib

With crystal structures of protein-ligand interactions becoming increasingly accessible, it is easy to forget that proteins do not exist as the static structures seen on page or screen. Indeed, back in 2018 we quoted Karplus quoting Feynman that “everything that living things do can be understood in terms of the jiggling and wiggling of atoms,” and even the smallest proteins have lots of atoms. In an open-access paper published in Proc. Nat. Acad. Sci. USA earlier this year, Heike Schönherr, David Shaw, and collaborators at Relay Therapeutics, D.E Shaw Research, Pharmaron, and Columbia University take advantage of these movements.

 

The researchers were interested in finding selective inhibitors of fibroblast growth factor receptor 2 (FGFR2), which is activated in many cancers. The four members of the FGFR family are so closely related that finding selective inhibitors is difficult. Inhibiting FGFR1 can lead to hyperphosphatemia, while inhibiting FGFR4 can cause diarrhea, side effects seen with the approved fragment-derived drug erdafitinib.

 

Although the structures of FGFR1 and FGFR2 are very similar, extended (25 µs) molecular dynamics simulations revealed that the so-called P-loop of the proteins behaved differently: in FGFR1 it became disordered, while in FGFR2 it remained more rigid. The researchers sought to take advantage of these differences with a covalent inhibitor.

 

The researchers started with a non-selective hinge-binding fragment, compound 1. Adding an acrylamide warhead led to a nanomolar inhibitor with modest selectivity for FGFR2. (All IC₅₀ values are measured after 30 minute incubations.) Growing the molecule into the so-called back pocket of the kinase led to compound 5, with nearly 100-fold selectivity for FGFR2 over FGFR1. 

 

 

The path from compound 5 to lirafugratinib (also called RLY-4008) looks straightforward but was anything but. First, the aryl acrylamide was a metabolic liability, so the researchers attenuated the reactivity by adding a methyl group. Mechanistic studies with this molecule revealed that while it had only a slightly better affinity (K_I) for FGFR2 than FGFR1, it had a k_inact value about 15-fold higher for FGFR2. Molecular dynamics studies suggested that the relevant cysteine in FGFR1 is locked in a position too far from the acrylamide to react, while the corresponding cysteine in FGFR2 may be able to more closely approach the acrylamide warhead.

 

Further optimization, guided by extended molecular dynamics simulations, led eventually to lirafugratinib with ~250-fold selectivity for FGFR2 over FGFR1 and >5000-fold selectivity over FGFR4. Remarkably, the noncovalent version of lirafugratinib, compound 11, shows dramatically lower affinity for both FGFR1 and FGFR2 and very little selectivity between them. The ligand seems to assume a different binding mode after covalent bond formation, which could explain these differences in selectivity.

 

Mouse studies of lirafugratinib showed tumor stasis or regression without increased serum phosphate levels. More importantly, early clinical data has shown “minimal hyperphosphatemia and diarrhea.”

 

This is a lovely example of structure and dynamics-based design (SDBD?). Commonly cited advantages of covalent drugs include improved potency and extended pharmacological effects, but this work shows that they can also achieve remarkable selectivity between closely related proteins, even when both proteins contain cysteine residues in the same location. Moreover, an open-access paper in Cancer Discov. that dives more deeply into the biology shows that lirafugratinib is selective across the kinome, inhibiting just two of 468 kinases other than FGFR2 by >75% at 500 nM.

 

The next time you’re trying to find a selective inhibitor for one member of a protein family, it may be worth taking a covalent approach, and paying close attention to dynamics along the way.

Fragments vs β-glucocerebrosidase

The protein β-glucocerebrosidase, also called GCase and GBA, is a lysosomal enzyme that cleaves glucosylceramide. People with inactivating mutations in both copies of GCase develop Gaucher’s Disease, which can be treated with a recombinant form of GCase. Heterozygous mutations increase risk for Parkinson’s Disease and for dementia with Lewy bodies, and though the mechanism is unclear, stabilizing the enzyme and/or boosting activity of residual GCase might help. This approach is described in a recent J. Med. Chem. paper by Nick Palmer and colleagues at Astex Pharmaceuticals.

 

The researchers started with a crystallographic screen of 440 fragments, resulting in a whopping 91 hits. In parallel, 1800 fragments (including the aforementioned 440) were screened using ligand-observed NMR, SPR, and thermal shift assays, and hits were confirmed crystallographically to yield another 15 structures. Astex has previously reported that multiple ligand binding sites are common in proteins, and GCase is no exception, with the 106 ligands binding to 13 distinct sites.

 

With this embarrassment of riches, prioritization became critical. Sites formed by crystal packing and shallow solvent-exposed sites were deprioritized, along with those near the active site, since ligands binding there might inhibit the enzyme. SPR was not well-suited to measuring ligand affinities due to non-specific binding, and ligand-observed NMR was similarly complicated due to multiple binding sites. However, isothermal titration calorimetry (ITC) proved to be effective, and this technique was used to narrow in on two binding sites.

 

Site A was particularly attractive: it had 31 fragment hits, one of which has a respectable dissociation constant of 12 µM. Screening of analogs did not lead to anything better, but merging this fragment with another Site A fragment led to compound 15. Interestingly, crystallography revealed that this molecule binds not at Site A but at Site B. Although the affinity is low, the ligand efficiency is respectable. The fragment also makes several polar interactions and has multiple vectors for growing the molecule.

 

 

Testing analogs of compound 15 led to compound 16, and growing led to compound 17, with low micromolar affinity. Further structure-based design ultimately led to compound 22, with low nanomolar affinity. The molecule increased GCase activity in a cellular assay, albeit at a fairly high (mid-micromolar) concentration. The molecule was found to be cell permeable with no efflux, so the source of the disconnect between affinity and cell activity is unclear.

 

This lovely example of structure-guided fragment-based ligand design holds several lessons. First, as noted above, finding fragments is often the easy part; selecting among them and figuring out what to do next can be challenging. Second, especially at the earliest stages of optimization, fragments can change not just their binding mode but their binding site entirely.

 

Finally, figuring out which sites will be best for high-affinity allosteric ligands isn’t necessarily straightforward. Of the 105 fragment hits at 13 sites, only four bound in Site B, yet this site turned out to be more fruitful than Site A, which had many more bound fragments. The researchers note that Site B had previously been identified as ligandable by FTMap, supporting the utility of computational approaches.

 

The researchers conclude, “we hope that our findings will be of use to the wider community.” Certainly from a best practices perspective the paper succeeds. And although the most advanced molecules described do not meet all the criteria for robust chemical probes, and it is unclear whether they will work with mutant proteins, they could still be useful to better understand the complicated biology of GCase.

A bright idea for rapid affinity measurements

Finding fragments that bind to a target is important but so is measuring their affinities. NMR methods can find even weak fragments, but accurately assessing affinities takes time. In a recent (open-access) J. Am. Chem. Soc. paper, Felix Torres, Roland Riek, and collaborators at the Institute for Molecular and Physical Science and NexMR provide a new, fast method.

 

The approach is based on photochemically induced dynamic nuclear polarization (photo-CIDNP), which we wrote about here; Felix also spoke about it at the FBDD-DU meeting in June. As the name implies, the technique involves illuminating NMR samples to electronically excite ligands, thus increasing the signal to noise ratio of the NMR signal by as much as 100-fold. Previous work focused on using the method to identify binders, even with cheap, benchtop NMR instruments.

 

The new paper describes how to quantitatively measure dissociation constants using photo-CIDNP. The theory gets a bit hairy, but the basic idea is that the more photochemically excited ligand that binds to the protein, the more the signal decreases. A series of samples are prepared with increasing concentrations of ligand and either no protein or a fixed concentration of protein. After measuring the NMR signals, the data are plugged into equations to derive the K_D values in a method called CIDNP-K_D.

 

As the researchers have previously noted, not every ligand can be photosensitized. However, dissociation constants can still be measured for these using competition experiments with previously characterized reporter ligands that can polarized, akin to using to NMR competition studies with ¹⁹F reporter ligands (see here).

 

So how well does the technique work? The researchers first turned to the PDZ2 domain of a phosphatase called hPTP1E, which is involved in cell proliferation. They measured the affinities of a series of peptides having 4 to 8 amino acid residues and compared these values to those obtained using two dimensional [¹H,¹⁵N]-HSQC chemical shift perturbation, the gold standard NMR technique. Affinities ranged from low micromolar to low millimolar, and there was reasonable agreement (generally within about two-fold) between both techniques. Most of the peptides contained tryptophan, which is suitable for photo-CIDNP, but CIDNP-K_D also worked in competition mode when non-tryptophan containing peptides were competed against peptides containing tryptophan. And the technique was fast, with each datapoint taking only 30 seconds for photo-CIDNP compared to as long as 80 minutes for HSQC NMR.

 

Next the researchers turned to fragments. They had previously conducted a screen against the oncology target PIN1 and identified a number of fragment hits, two of which had been characterized in detail. The affinities of these were measured by CIDNP-K_D, and the low millimolar values agreed with those from HSQC NMR.

 

Another neat application described in the paper is “CIDNP-based epitope mapping,” which is based on the fact that an excited proton on a ligand that is in close proximity to the protein will relax more rapidly than one that is distant from the protein. This phenomenon is similar to STD epitope mapping, and the two methods yielded similar information for the two PIN1 ligands: one region of each molecule was buried in the protein, consistent with crystal structures.

 

One drawback of the technique is that, because measurements require fast protein-ligand exchange, CIDNP-K_D is limited to relatively weak binders (K_D > 10 µM), but this is usually not a problem in the early stages of a fragment program. A full affinity measurement takes about 15 minutes, which compares very favorably to two hours using [¹H,¹⁵N]-HSQC and without the need for isotopically labeled protein. It would be interesting to run head-to-head comparisons with two ligand-based NMR techniques we wrote about last year, imaging STD NMR and R2KD, to see how they compare in terms of speed, accuracy, and generality. Please let us know if you’ve done so.

Fragments vs GPx4 – in reverse micelles

Membrane proteins account for more than half of drug targets, but the fraction is far smaller for fragment-derived drugs. In part this is because biophysical methods, the mainstay of FBLD, have been harder to apply to membrane proteins. A recent (open-access) paper in JACS Au by Courtney Labrecque and Brian Fuglestad at Virginia Commonwealth University tackles this challenge.

 

The researchers use an approach called membrane-mimicking reverse micelles, or mmRM: tiny water-filled bubbles surrounded by lipids and suspended in an organic solvent. We last wrote about reverse micelles back in 2019, where they were being used to study high local concentrations of water-soluble proteins and ligands. Here, the researchers turned to membrane proteins.

 

There are actually two types of membrane proteins: integral membrane proteins and peripheral membrane proteins. The former, as their name implies, have at least part of the protein anchored in the membrane at all times; GPCRs are a prominent class. Peripheral membrane proteins are water soluble but associate with the membrane, and this interaction is often required for folding or function. One example is glutathione peroxidase 4 (GPx4), which reduces oxidized lipids. It is an intriguing but  challenging cancer target, with the only ligands being fairly reactive covalent modifiers. Thus the researchers turned to mmRMs, hoping these could both stabilize the protein in a biologically relevant state and also present binding opportunities unavailable in standard screens.

 

A library of 1911 fragments from Life Chemicals was screened against mmRM-encapsulated GPx4 using ¹⁵N-¹H HSQC protein-detected NMR. Fragments were chosen to have high aqueous solubility (at least 1 mM in PBS) and were screened in mixtures of 10 at 400 µM per fragment. After deconvolution, 14 hits were identified, and dose-response titrations revealed that 9 had apparent dissociation constants < 1 mM, with the most potent having a K_d = 105 µM.

 

Three fragments were studied in greater detail, and these were chosen to have a range of hydrophobicities from clogD = -2.1 (most polar) to clogD = 2.1 (most hydrophobic). Chemical shift perturbation (CSP) analyses suggested that the two more lipophilic fragments bind to the membrane-interacting region of the protein, while the more polar fragment likely binds to a water-exposed site. SAR-by-catalog was applied to find analogs, some of which had increased affinity for the protein, with the best being around K_d = 15 µM.

 

Interestingly, the fragments showed minimal binding to GPx4 under normal aqueous conditions (ie, in the absence of the mmRMs), even at very high fragment concentrations. The researchers suggest this is because the fragments are binding to the membrane-bound state of the protein found in mmRMs, which may adopt a different conformation than that in the absence of membranes. Perhaps. But as prior work shows, it is possible to detect extraordinarily low affinity interactions inside reverse micelles, so maybe these are just very weak binders. Ultimately it remains to be seen whether these fragments will have practical applications. I hope so, and look forward to seeing how they progress.