Plenty of eye-opening material was published this August! Among this month’s highlights are a small molecule with twice-a-year dosing from Gilead, a non-classical C-H—O hydrogen-bonding kinase inhibitor from Pfizer, an E1 ligase inhibitor with a cool substrate-assisted covalent mechanism from Takeda, and some STING agonists with well-tolerated oral activity and fascinating mechanisms from Merck and TSRI/Calibr. Links to the articles and more discussion on specific molecules below.

Small Molecules of the Month - August 2020

Subscribe to get a link to download a high-resolution PDF of August 2020's "Small Molecules of the Month."

send download link to:

I confirm that I have read and agree to the Terms of Use and Privacy Policy.

GS-6207 is an HIV-1 inhibitor that acts on the capsid protein monomer (p24), which normally self-assembles in the HIV virion to form the capsid core that is essential for virus infectivity. GS-6207 targets a protein-protein interaction between the small (24 kDa) p24 monomers, binding in a glue-like fashion to two separate monomers, and is able to overcome the high concentration of p24 in a virion (~4 mM!) to inhibit HIV activity with sub-nanomolar potency in cells (20-500 pM). Since the HIV capsid protein is highly conserved among HIV-1 variants, GS-6207 shows high synergy and no cross-resistance with approved anti-HIV agents. GS-6207 was found to be generally safe and well-tolerated in Ph. I and showed evidence of antiviral activity. Amazingly, this small molecule has the half-life one might expect of a well-engineered biologic drug, and is being dosed subcutaneously once every 6 months in the Ph. II/III study. Definitely a candidate for “molecule of the year!”

PF-06826647 is a TYK2-selective kinase inhibitor which binds to the catalytically active JH1 kinase domain, in contrast to BMS-986165, a TYK2 inhibitor in Ph. III for psoriasis that binds to the JH2 pseudokinase domain. TYK2-selective inhibitors are expected to have anti-inflammatory activity with reduced side effects such as anemia due to inhibition of other JAK-family members. Interestingly, PF-07826647 interacts with the hinge region of TYK2’s JH1 domain through a polarized aryl C-H—O hydrogen bond. PF-06826647 is currently in clinical development as a once-daily oral treatment for a number of inflammatory conditions including psoriasis and ulcerative colitis.

MSA-2 is an orally available agonist of the STING protein, which potently induces inflammation. Clinically evaluated cyclic dinucleotide-based STING agonists have typically been dosed intratumorally to promote anti-tumor inflammation but avoid systemic inflammation. The compound binds in a 2:1 stoichiometry with STING, and forcibly dimerized versions of MSA-2 are significantly more potent against STING. Interestingly, the compound demonstrates oral activity in cancer immunotherapy models at well-tolerated concentrations. The apparent tumor selectivity is hypothesized to be due to the weakly acidic nature of molecule. The mildly acidic tumor microenvironment favors the more permeable, protonated form of MSA-2, which can accumulate within the more basic tumor cells through a sink effect.

SR-717 is STING agonist with a surprisingly similar mechanism to MSA-2, published contemporaneously with MSA-2. The identification of a second carboxylic acid-type molecule with both 2:1 binding stoichiometry and apparent tumor selectivity support both papers’ MoA hypotheses.

Vixotrigine (BIIB074) is a voltage-gated sodium channel blocker that preferentially binds to the inactivated state of the sodium channels, conferring selectivity for channels involved in pain-associated high frequency nerve signalling over normal low frequency nerve signalling (use-/state-dependence). The use-dependent activity is believed to confer a better safety window over previous generations of sodium channel inhibitors. Two Ph. III clinical trials appear to be planned for trigeminal neuralgia with estimated start dates in mid-2021.

BI-3406 is a compound which binds to SOS1, preventing it from activating the oncoprotein KRAS. While inhibition of WT-KRAS and MEK alone are anticipated to be toxic, the combination of a SOS1 inhibitor with a MEK inhibitor appears to have a broader therapeutic index due to synergy in KRAS-driven tumors. A related undisclosed SOS1 inhibitor, BI 1701963, is in clinical development alone and in combination with a MEK inhibitor.

“compound 3d” is a prodrug which is phosphorylated in vivo to form a potent sphingosine-1-phopshate receptor modulator. BMS took two previous compounds into the clinic for MS with the goal of differentiating from fingolimod on safety. One potential issue for a previous compound was an overly long (12 day) half-life which could be undesirable in certain settings such as those involving drug-drug interactions. Therefore “compound 3d” was developed to have a shorter preducted human half-life of “only” 5 days.

“compound 12” is a selective inhibitor of integrin αvβ5, a potential target for sepsis due to its potential role in regulating vascular leakage. Selectivity among integrin family members can be challenging to achieve and can have significant safety implications – hence, most integrin drugs have been selective biologics. Since the biological role of many integrins is still not fully understood, selective tool molecules like compound 12 can be very useful in early discovery.

“compound 12j” is a potent agonist of the ghrelin hormone receptor, activation of which stimulates growth hormone secretion and appetite among other activities. Ghrelin receptor antagonists have previously been explored to treat disorders related to obesity such as diabetes, but an agonist could be useful in treating atrophic disorders like cachexia. Compound 12j has improved off-target selectivity versus a previously reported compound and has significant bioavailability in dogs, but was not advanced due to uncertainty about projected human oral exposure.

“compound 7f” is an antagonist of the endosomal TLRs 7/8/9, which induce inflammation in response to internalized single-stranded RNA or hypomethylated double-stranded DNA. Potency and selectivity within endosomal TLR’s (3, 7, 8, 9) and against cell-surface TLR’s (1, 2, 4, 5, 6) has been difficult to achieve. Compound 7f has single-digit nanomolar activity against TLRs 7 and 8 in human whole blood, and submicromolar activity against TLR9. Its oral bioavailability in rodents makes it a useful tool compound for studying TLR antagonism in different disease models.

DS34942424 is a potent oral analgesic discovered using the natural product conolidine as a starting point. While the authors were able to demonstrate that the compound does not exert its activity through the mu-opioid receptor, the mechanism of action is not understood. The identification of non-opioid analgesics are an area of intense interest and hopefully we will learn how this compound works in the near future.

“compound 12” is a selective negative allosteric modulator of the GluN2B subunit of NDMA receptors intended for treatment of mood disorders. In this program, poor solubility and permeability were a challenge for oral exposure, but was addressed with a nanosuspension formulation. The nanosuspension allowed them to get high enough margins in preclinical species to assess safety, and achieve a low predicted human dose of 120 mg. No dose limiting toxicities were observed up to 500 mg/kg/day for 14 days in rats (~125 mg/rat/day).

“compound 37” is an inhibitor of the ATG7 E1 ligase through an interesting substrate-assisted covalent mechanism. Rather than covalently modify ATG7 directly, the molecule reacts with the E1 ligase substrate, Ubl, to form a Ubl-inhibitor adduct which non-covalently inhibits ATG7. This mechanism has been employed by Takeda successfully in the past against a number of other related targets such as NEDD8-activating enzyme. Compound 37 is one of three molecules from this article demonstrating in vivo inhibition of the autophagy process.

“compound 20” is a gamma-secretase modulator (GSM) with an interesting, uncommon oxadiazine scaffold. Like many hot areas in drug discovery, the GSM area for Alzheimer’s has led to many creative chemical scaffolds. It’s always interesting to see how these new scaffolds behave. This compound demonstrates surprisingly low clearance, high bioavailability in rodents, and sufficient brain penetration to observe desired PD in vivo.

Links to Articles:

  1. GS-6207
  2. PF-06826647
  3. MSA-2
  4. SR-717
  5. Vixotrigine (BIIB074)
  6. BI-3406
  7. “compound 3d”
  8. “compound 12”
  9. “compound 12j”
  10. “compound 7f”
  11. DS34942424
  12. “compound 12”
  13. “compound 37”
  14. “compound 20”

Finally, for the ADME buffs out there, OpenBench Lab is taking compounds from the Molecules of the Month series, calculating their properties, and comparing it to the reported data in the papers. Their comparison’s for last month’s molecules can be found here. It will be interesting to see how current property prediction algorithms do on the newest chemical matter, which seems to be getting more complex every year!

Hope this was helpful. Explore for more.