Fragments vs HDAC2: new metal chelators

Histone deacetylases (HDACs) are epigenetic writers that – as their name suggests – remove acetyl groups from lysine residues in histones. They have been pursued as anticancer targets for decades; vorinostat was approved back in 2006 (and arguably has fragment origins). The catalytic site of HDACs contains a zinc ion, and many inhibitors include zinc-binding moieties, often hydroxamic acids. However, the high affinity of these metallophilic fragments leads to inhibition of other zinc-containing enzymes, causing toxicity. In a new ACS Med. Chem. Lett. paper, Emiliano Tamanini, Shin Miyamura, and colleagues at Astex and Otsuka provide alternative chemistries. (Emiliano presented part of this story at Pacifichem last year.)

 

The researchers were specifically interested in HDAC2, which has been implicated in neurodegenerative diseases such as Alzheimer’s. Because of the chronic nature of these conditions, selectivity was all the more important, as was brain penetration. Screens of the Astex fragment library, along with a set of known zinc-binding fragments, yielded 35 crystallographically validated hits. These included compound 3, which forms bidentate interactions with the catalytic zinc through the amine and carbonyl moieties. 

  

An additional fragment (compound 4, not shown) bound in a pocket at the “foot” of the catalytic site, which is normally partially blocked by a side chain residue. Yet another fragment bound near the entrance tunnel. Growing compound 3 in both directions led to compound 7, the first molecule of the series with measurable activity in a fluorescence assay. Rigidification and further growing led to compound 9, with low micromolar activity.

 

 

Interestingly, merging compound 9 directly with foot-pocket binder compound 4 did not improve activity, but adding a chlorine atom to fill a small subpocket increased affinity two-fold (compound 13). Finally, optimization of the moiety at the entrance tunnel (the isoindoline of compound 13) yielded compound 17, with high nanomolar potency.

 

Compound 17 caused increased acetylation of H4K12 in cells. More importantly, it showed good brain exposure when orally dosed in mice, and H4K12 acetylation was observed in mouse brain tissue. The compound hits HDAC1 and HDAC8 with similar potency as HDAC2, and these could be contributing to the biological effect. Nonetheless, while potency and selectivity still need to be improved, compound 17 is an attractive lead for further optimization. 

 

In a comment to the Notum work we highlighted back in May, Daniel Beck wondered whether “fragments are especially good starting points for CNS target campaigns.” This paper suggests the answer is yes.

Fragments in the clinic: 2022 edition

In the US we're about to celebrate Thanksgiving. One of the things I'm thankful for is the discovery of new medicines, so this seems like an appropriate time to update our tally of fragment-derived drugs.

 

As always, this table includes compounds whether or not they are still in development (indeed, some of the companies no longer even exist). Because of this, the Phase 1 section contains a higher proportion of compounds that are no longer progressing. The full list contains 58 molecules, up from 52 last year, with more than 40% approved or in active trials: not bad given that only about 10% of drugs that make it into the clinic are ultimately approved.

 

Drugs reported as still active in clinicaltrials.gov, company websites, or other sources are in bold, and the 36 that have been discussed on Practical Fragments are hyperlinked to the most relevant post. The list is almost certainly incomplete, particularly for Phase 1 compounds. If you know of others please leave a comment.

Drug	Company	Target
Approved!
Asciminib	Novartis	BCR-ABL1
Erdafitinib	Astex/J&J	FGFR1-4
Pexidartinib	Plexxikon	CSF1R, KIT
Sotorasib	Amgen	KRASG12C
Vemurafenib	Plexxikon	B-RAFV600E
Venetoclax	AbbVie/Genentech	Selective BCL-2
Phase 3
Capivasertib	AstraZeneca/Astex/CR-UK	AKT
Lanabecestat	Astex/AstraZeneca/Lilly	BACE1
Navitoclax (ABT-263)	Abbott	BCL-2/BCLxL
Pelabresib (CP-0610)	Constellation	BET
Verubecestat	Merck	BACE1
Phase 2
ASTX029	Astex	ERK1,2
ASTX660	Astex	XIAP/cIAP1
AT7519	Astex	CDK1,2,4,5,9
AT9283	Astex	Aurora, JAK2
AUY-922	Vernalis/Novartis	HSP90
AZD5991	AstraZeneca	MCL1
DG-051	deCODE	LTA4H
eFT508	eFFECTOR	MNK1/2
Indeglitazar	Plexxikon	pan-PPAR agonist
LY2886721	Lilly	BACE1
LY3202626	Lilly	BACE1
LY3372689	Lilly	OGA
LY517717	Lilly/Protherics	FXa
LYS006	Novartis	LTA4H
MAK683	Novartis	PRC2 EED
Onalespib	Astex	HSP90
PF-06650833	Pfizer	IRAK4
PF-06835919	Pfizer	KHK
PLX51107	Plexxikon	BET
S64315	Vernalis/Servier/Novartis	MCL1
VK-2019	Cullinan Oncology / Wistar	EBNA1
Phase 1
AG-270	Agios/Servier	MAT2A
ABBV-744	Abbott	BD2-selective BET
ABT-518	Abbott	MMP-2 & 9
ABT-737	Abbott	BCL-2/BCLxL
AT13148	Astex	AKT, p70S6K, ROCK
AZD3839	AstraZeneca	BACE1
AZD5099	AstraZeneca	Bacterial topoisomerase II
BI 1823911	Boehringer Ingelheim	KRASG12C
BI 691751	Boehringer Ingelheim	LTA4H
CFTX-1554	Confo Therapeutics	AT2 receptor
ETC-206	D3	MNK1/2
GDC-0994	Genentech/Array	ERK2
HTL0014242	Sosei Heptares	mGlu5 NAM
HTL0018318	Sosei Heptares	M1-receptor partial agonist
HTL9936	Sosei Heptares	M1-receptor partial agonist
IC-776	Lilly/ICOS	LFA-1
LP-261	Locus	Tubulin
LY2811376	Lilly	BACE1
Mivebresib	AbbVie	BRD2-4
MRTX1719	Mirati	PRMT5•MTA
Navoximod	New Link/Genentech	IDO1
PLX5568	Plexxikon	RAF
SGX-393	SGX	BCR-ABL
SGX-523	SGX	MET
SNS-314	Sunesis	Aurora
TAK-020	Takeda	BTK

The agony and ecstasy of thiazoles

Earlier this year we highlighted an analysis of rings found in drugs. Thiazoles are tied for thirteenth place, occurring in at least 30 drugs. (A substructure search in DrugBank pulls up 49.) They pack a lot of diversity into just 5 heavy atoms, with a nitrogen atom capable of acting as a hydrogen bond acceptor as well as a sulfur atom. But they can also be tricksy: an analysis several years ago found that 2-aminothiazoles are over-represented as hits in fragment screens but are often not advanceable. A new open-access paper in ACS Med. Chem. Lett. by Rok Frlan and collaborators at the University of Ljubljana confirms and broadens these conclusions.

 

The researchers assembled a library of 44 fragment-sized 1,3-thiazoles and five 1,3,4-thiadizaozles. These were then screened at 0.5 or 0.625 mM against four unrelated enzymes in biochemical assays. Two of the enzymes contain catalytic cysteine residues, and these had high hit rats: 14 hits for the SARS-CoV-2 3CL^pro enzyme and 26 for the E. coli MurA enzyme. In contrast, MetAP1a had only 3 hits, while DdlB had none. Are any of these hits real?

 

None of the compounds had been classified as PAINS, and aggregation was deemed unlikely for all but one compound based on chemical searches and the presence of detergent in the assays for MurA, DdlB, and 3CL^pro. One compound also seemed to interfere with the fluorescent assays and was ruled a false positive. So far, so good.

 

However, 8 of the compounds turned out to be unstable in aqueous buffer. Moreover, four compounds turned out to be redox active in at least one of three different assays. Redox cycling can generate reactive oxygen species, which inhibit cysteine-dependent proteins nonspecifically.

 

Next the researchers tested to see whether their fragments reacted with a small test thiol, 5-mercapto-2-nitrobenzoic acid. Shockingly, 19 of them did, and most of these inhibited at least one of the enzymes. Many of these contain potential leaving groups such as halogen atoms, but some didn’t, leaving the nature of the reaction unclear. Still, the results suggest that the fragments are more thiol-specific than protein-specific, and so another potential source of false leads.

 

When the researchers retested the ability of the fragments to inhibit the enzymes in the presence of the reducing agent DTT, only one of the CL^pro hits reproduced – and that was the compound that showed fluorescence interference. The results were not quite so bad for MurA, though many hits fell out.

 

Finally, the researchers tried to correlate reactivity with quantum-mechanical calculations using several different methods. Unfortunately, as they note, “no meaningful relationships were observed.” Laudably, data for all the compounds are provided, so interested readers are free to try their own analyses.

 

In the end it is not clear whether any of the hits will be useful, but the high correlation between pathological mechanisms and activity does not make one optimistic. As the first paragraph above makes clear, this does not mean that thiazoles should be avoided. Indeed, the researchers explicitly state that “we do not want to establish a general knockout criterion to exclude thiazole or thiadizaole screening hits from further development, but it is essential to evaluate their reactivity if they prove to be hits.” This is where orthogonal biophysical methods, such as crystallography, can distinguish true hits from artifacts.

Fragments vs IL17A: merging and linking

One of the talks at the Discovery on Target meeting last month described the discovery of small molecule inhibitors of IL17A, a pro-inflammatory cytokine. Antibody-based drugs against this protein are useful for psoriasis and other diseases, but they require regular injections. Also, because the antibodies stick around for a long time and dampen the immune response, they could leave patients less able to combat an infection. An orally available small molecule could solve both these problems, but blocking protein-protein interactions is generally difficult. Early progress towards this goal has been published (open access) in Sci. Reports by Eric Goedken and collaborators at AbbVie.

 

The researchers started by ¹³C-labeling the methyl groups of isoleucine, leucine, valine, and methionine. They then performed a 2D NMR screen ([¹H, ¹³C]-HSQC) of ~4000 fragments in pools of 12, with each fragment at 1 mM. This yielded multiple hits, including compound 4. Interestingly, the chemical shift perturbations (CSPs) caused by this fragment were distinct from those caused by known previously disclosed binders, suggesting a different binding site. In particular, one of these CSPs could be traced to a methionine residue. The full NMR assignment of the protein was not conducted, but modeling and mutagenesis narrowed the possibility down to a methionine near the C-terminus.

 

Surface plasmon resonance (SPR) experiments revealed that compound 4 had very low affinity, but two rounds of optimization led to compounds 5 and 6. At this point the researchers were able to solve the crystal structures of these molecules bound to IL17A. The protein exists as a homodimer, and the molecules bind symmetrically, with two copies per homodimer. Moreover, the two copies bind close to one another.

 

 

In addition to these fragments, the researchers had also identified compound 7, and a crystal structure revealed that this binds in a similar fashion. Merging compound 7 with compound 6 and linking two copies of the resulting monomer led ultimately to dimeric compound 10, with nanomolar affinity by both SPR and isothermal titration calorimetry (ITC). This molecule inhibited the binding of IL17A with its receptor in a biochemical assay and also showed low micromolar activity in a cellular assay.

 

This is a nice paper that bears some similarity to previous work we highlighted from AbbVie on a different cytokine, TNFα. There too fragment linking was used initially. As in the present case, that effort led to a drop in ligand efficiency and a significant increase in the size of the molecules, resulting in suboptimal pharmaceutical properties. Identifying drug-like small molecules has been an ongoing challenge for IL17A; peptide-based inhibitors and macrocycles have been found that bind to other sites on the protein, but many of these are also well beyond rule-of-five space. As the researchers conclude, “we look forward to seeing which of these sites prove to be the most amenable to producing optimized drug candidates.”