Deregulation of the transcriptional repressor BCL6 enables tumorigenesis of germinal center B-cells, and hence BCL6 has been proposed as a therapeutic target for the treatment of diffuse large B-cell lymphoma (DLBCL). Herein we report the discovery of a series of benzimidazolone inhibitors of the protein-protein interaction between BCL6 and its co-repressors. A subset of these inhibitors were found to cause rapid degradation of BCL6, and optimization of pharmacokinetic properties led to the discovery of 5-((5-chloro-2-((3R,5S)-4,4-difluoro-3,5-dimethylpiperidin-1-yl)pyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (CCT369260), which reduces BCL6 levels in a lymphoma xenograft mouse model following oral dosing.
Deregulation of the transcriptional repressor BCL6 enables tumorigenesis of germinal center B-cells, and hence BCL6 has been proposed as a therapeutic target for the treatment of diffuse large B-cell lymphoma (DLBCL). Herein we report the discovery of a series of benzimidazolone inhibitors of the protein-protein interaction between BCL6 and its co-repressors. A subset of these inhibitors were found to cause rapid degradation of BCL6, and optimization of pharmacokinetic properties led to the discovery of 5-((5-chloro-2-((3R,5S)-4,4-difluoro-3,5-dimethylpiperidin-1-yl)pyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (CCT369260), which reduces BCL6 levels in a lymphoma xenograft mouse model following oral dosing.
BCL6 (B-cell lymphoma
6 protein) is a transcriptional repressor
that plays a key role in the formation and maintenance of germinal
centers during the process of antibody affinity maturation.[1−3] By binding to DNA via its zinc fingers and recruiting one of its
co-repressors (NCoR, SMRT, or BCOR) to its dimeric BTB domain, BCL6
represses genes involved in cell cycle control, cell death, differentiation,
and the DNA damage response. This action enables B-cells in the germinal
center to proliferate rapidly, evade growth checkpoint controls, and
tolerate high levels of DNA damage. These features are required for
the process of somatic hypermutation of antibodies but are also among
the hallmarks of cancer;[4,5] hence deregulation of
BCL6 can lead to lymphomagenesis. Most B-cell lymphomas arise from
germinal center B-cells, which are dependent on continued expression
of BCL6 for survival.[1] Recruitment of co-repressors
is essential for the repressive and oncogenic functions of BCL6, and
disruption of this interaction is sufficient to inhibit lymphoma cell
growth.[6,7] We sought to discover compounds that disrupt
the protein–protein interaction (PPI) between the BTB domain
of BCL6 and its co-repressors to alleviate BCL6-mediated gene repression,
inhibit lymphoma cell growth, and identify new treatments for BCL6-driven
lymphomas. Following on from earlier publications that identified
small molecule or peptidomimetic compounds with moderate potency against
BCL6,[8−10] the recent publication of high-potency small molecules
has established the ligandability of the BCL6BTB domain dimer.[11−13] Proteolysis targeting chimeras (PROTAC)[14] and non-PROTAC degrader compounds that trigger the proteosomal degradation
of BCL6[15] have also been reported. However,
to date, no compound has been reported to be suitable for use as an in vivo probe to investigate the effect of degradation of
BCL6. In this study we report the discovery not only of compounds
that can potently inhibit BCL6 function by disrupting the PPI with
its co-repressors but also of a subset of molecules that trigger the
rapid degradation of BCL6. Further optimization of this series led
to compound 1 (CCT369260), which shows degradation of
tumoralBCL6 in vivo following oral dosing in a lymphoma
xenograft mouse model.
Results
Chemistry: Synthesis of
Compounds
Initial hit compounds 2 and 3 and follow-up compound 4 were obtained from
commercial vendors, while the cyclopropyl compound 5 and
benzimidazolone 6 (CCT365386) were prepared
by single step nucleophilic aromatic substitution (SNAr)
reaction from available building blocks. Compounds with alkyl substituents
in the benzimidazolone-N3 position could be synthesized
by stepwise alkylation of nitro-compound 7, reduction,
and SNAr reaction, as exemplified in the preparation of 11a (Scheme ). In order to facilitate the optimization of this position (as shown
in Table ), we developed
conditions to allow the addition of this group as the final step (Scheme , steps d and e),
using intermediate 9. The use of alkyl halides or epoxides
as electrophiles under these conditions gave primarily the desired
regioisomer, although in most cases minor products corresponding to
alkylation at the 4-pyridylamine were also observed in the reaction
mixture and required HPLC purification to separate.
Scheme 1
Final Step Diversification
of Benzimidazolone N3 Position
Reagents and conditions: (a)
2-ethyloxirane, cesium carbonate, DMF, 120 °C, 1 h; (b) sodium
dithionite, ethanol/DMSO, rt to 90 °C; (c) 2,4-dichloropyridine-3-carbonitrile,
DIPEA, DMA, 120 °C, 30–45 min; (d) (2R)- or (2S)-ethyloxirane, cesium carbonate, DMF,
120 °C, 1 h, then 140 °C, 1 h; (e) alkyl bromide or tosylate,
cesium carbonate, DMF, 60 °C, 3 days or 120–140 °C,
1 h.
Table 2
Structure-Activity Relationships of
the Benzimidazolone-N3 Position (Labeled as R3 Group)
Data represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.
Kinetic solubility measured by NMR
in HEPES buffer (containing 5% DMSO) at pH 8. Where this was <30
μM, solubility by HPLC in PBS buffer and 1% DMSO at pH 7.4 was
also measured and is shown in parentheses.
Final Step Diversification
of Benzimidazolone N3 Position
Reagents and conditions: (a)
2-ethyloxirane, cesium carbonate, DMF, 120 °C, 1 h; (b) sodium
dithionite, ethanol/DMSO, rt to 90 °C; (c) 2,4-dichloropyridine-3-carbonitrile,
DIPEA, DMA, 120 °C, 30–45 min; (d) (2R)- or (2S)-ethyloxirane, cesium carbonate, DMF,
120 °C, 1 h, then 140 °C, 1 h; (e) alkyl bromide or tosylate,
cesium carbonate, DMF, 60 °C, 3 days or 120–140 °C,
1 h.In order to vary the substituent of the
5-amino group on the benzimidazolone
(labeled as R4 in Table ), it was necessary first to scale up the synthesis
of the 1,3-disubstituted 5-aminobenzimidazol-2-one. Cyclization of
commercially available nitrodianiline 12 using disuccinimidylcarbonate
proceeded in high yield and was followed by alkylation with tosylate 14a, then hydrogenation to reduce the nitro group and form
aniline 16 (Scheme ). These steps were readily amenable to scale up, and
multigram quantities were produced without the need for column chromatography.
Table 3
Structure–Activity
Relationships
of the R4-Group-Substituted Pyridine and Pyrimidines
Data
represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.
Kinetic solubility measured by NMR
in HEPES buffer (containing 5% DMSO) at pH 8. Where this was <30
μM, solubility by HPLC in PBS buffer and 1% DMSO at pH 7.4 was
measured and is shown in parentheses.
Scheme 2
Synthesis of Key Intermediate 16
Reagents and conditions: (a)
disuccinimidylcarbonate, MeCN, rt, 18 h, up to 11g scale, 86–99%
yield; (b) cesium carbonate, MeCN, reflux, 4 h, 12 g scale, 92% yield;
(c) TsCl, NEt3, DCM, 0 °C to rt, up to 40g scale,
60–72% yield; (d) 10% Pd/C, 1 atm H2, ethanol, 60
°C, 3 h, 6 g scale, quant.
Synthesis of Key Intermediate 16
Reagents and conditions: (a)
disuccinimidylcarbonate, MeCN, rt, 18 h, up to 11g scale, 86–99%
yield; (b) cesium carbonate, MeCN, reflux, 4 h, 12 g scale, 92% yield;
(c) TsCl, NEt3, DCM, 0 °C to rt, up to 40g scale,
60–72% yield; (d) 10% Pd/C, 1 atm H2, ethanol, 60
°C, 3 h, 6 g scale, quant.Aromatic nucleophilic
substitution or (for less reactive pyridine
substituents) palladium coupling reactions were used to prepare the
set of compounds shown in Table (Scheme ).
Scheme 3
Synthesis of Compounds from Table , Investigating Different
Substitution Patterns on Pyrimidine or Pyridine Rings
Reagents and conditions: (a)
heteroaryl chloride, DIPEA, NMP or DMF, 80–120 °C, 30
min; (b) 4-bromo- or iodopyridine, Xantphos, Pd2(dba)3, cesium carbonate, NMP/toluene, 140 °C, 1 h.
Synthesis of Compounds from Table , Investigating Different
Substitution Patterns on Pyrimidine or Pyridine Rings
Reagents and conditions: (a)
heteroaryl chloride, DIPEA, NMP or DMF, 80–120 °C, 30
min; (b) 4-bromo- or iodopyridine, Xantphos, Pd2(dba)3, cesium carbonate, NMP/toluene, 140 °C, 1 h.In order to explore substitution at the pyrimidine-2-position
(Tables and 5), dichloropyrimidine 18 was prepared
on multigram
scale. Final compounds were prepared from this, again via SNAr or metal-catalyzed reactions (Scheme ).
Table 4
Structure–Activity
Relationships
of the Pyrimidine and Pyridine 2-Position (Labeled as R3 Group)
Data represent
the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.
Data represent the geometric mean
of at least two replicates except where superscript 1 is indicated.
n.d. = not done.
Kinetic
solubility measured by NMR
in HEPES buffer and 5% DMSO at pH 8, and/or where data are shown in
brackets, by HPLC in PBS buffer and 1% DMSO at pH 7.4. #These compounds showed <10% degradation of BCL6 in single concentration
experiments at 10 μM, n = 2.
Table 5
Modifications to the Pyrimidine 2-Substituent
Data represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.
Kinetic solubility measured by NMR
in HEPES buffer and 5% DMSO at pH 8, and/or where data are shown in
brackets, by HPLC in PBS buffer and 1% DMSO at pH 7.4. #These compounds showed <25% degradation of BCL6 in single concentration
experiments at 10 μM. n.d. = not done.
Scheme 4
Synthesis of 2-Substituted Pyrimidine
Compounds from Tables and 5, Starting from Common Intermediate 18
Reagents and conditions: (a)
dimethylamine HCl, cesium carbonate, NMP, 180 °C, 1 h; (b) cyclic
amine, DIPEA, NMP, 140–180 °C, 1–2 h; (c) pyrazole,
cesium carbonate, NMP, 170 °C, 1 h; (d) 2,4-dimethyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thiazole,
sodium carbonate, PdCl2(PPh3)2, 1,4-dioxane/water,
130 °C, 30 min; (e) 1-methyl-2-(tributylstannyl)-1H-imidazole, PdCl2(PPh3)2, 1,4-dioxane,
90 °C, 18 h; (f) bicyclic amine, DIPEA, NMP, 140 °C, 2–8
h.
Synthesis of 2-Substituted Pyrimidine
Compounds from Tables and 5, Starting from Common Intermediate 18
Reagents and conditions: (a)
dimethylamine HCl, cesium carbonate, NMP, 180 °C, 1 h; (b) cyclic
amine, DIPEA, NMP, 140–180 °C, 1–2 h; (c) pyrazole,
cesium carbonate, NMP, 170 °C, 1 h; (d) 2,4-dimethyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thiazole,
sodium carbonate, PdCl2(PPh3)2, 1,4-dioxane/water,
130 °C, 30 min; (e) 1-methyl-2-(tributylstannyl)-1H-imidazole, PdCl2(PPh3)2, 1,4-dioxane,
90 °C, 18 h; (f) bicyclic amine, DIPEA, NMP, 140 °C, 2–8
h.The reduced reactivity of the pyridine
core in comparison to the
pyrimidine meant that metal-catalyzed reactions on bromo- or iodopyridines
or SNAr reactions using 2-fluoropyridines were used to
prepare the pyridyl examples in Table (Scheme ).
Scheme 5
Synthesis of 2,4,5-Substituted Pyridine Derivatives
Reagents and conditions: (a) 16, DIPEA, NMP, 180 °C, 1 h; (b) 3,5-dimethylpyrazole,
Xantphos, cesium carbonate. Pd2(dba)3, DMF/toluene,
140 °C, 1 h; (c) DIPEA, THF, 100 °C, 16 h; (d) 16, Xantphos, Pd2(dba)3, cesium carbonate, DMF/toluene,
80 °C, 1 h.
Synthesis of 2,4,5-Substituted Pyridine Derivatives
Reagents and conditions: (a) 16, DIPEA, NMP, 180 °C, 1 h; (b) 3,5-dimethylpyrazole,
Xantphos, cesium carbonate. Pd2(dba)3, DMF/toluene,
140 °C, 1 h; (c) DIPEA, THF, 100 °C, 16 h; (d) 16, Xantphos, Pd2(dba)3, cesium carbonate, DMF/toluene,
80 °C, 1 h.
Hit Discovery
A detailed description of our hit discovery
campaign for BCL6 will be presented elsewhere.[16] Briefly, we identified hit compounds 2 and 3 (Table )
from an in-house high-throughput screen using a fluorescence polarization
(FP) assay, based on the displacement of a fluorescently labeled peptide
derived from the BCOR co-repressor.[7] In
common with other hits from this screen, these compounds showed weak
potency (IC50 ∼100 μM). Initial attempts to
obtain ligand-bound X-ray structures were hindered by solubility which
was lower than their biochemical assay activity (Table ). To address this, we made
analogues with reduced lipophilicity and fewer aromatic rings. Compounds 4 and 5 demonstrated comparable potency to initial
hits and improved solubility (Table ), enabling X-ray structure determination.[17] Like the BCL6 co-repressors SMRT and BCOR,[7] compounds were found to bind at the dimer interface
of the BTB domain of BCL6 (Figure A and Figure B). Two interaction features are common to both 4 and 5: a hydrogen bond from an NH to the backbone carbonyl
of Met51 and the intercalation of their cyanopyridine moieties into
a cleft between Tyr58 and Asn21 on the protein surface, creating a
critical hydrophobic interaction between the cyanopyridine and the
side chain of Tyr58. Compound 4 forms additional interactions
via the oxygen atom of its benzimidazolone core to the backbone N–H
of Glu115 and to the backbone N–H of His116 via a mediating
water molecule. The cyclopropyl group of 5 makes only
weak hydrophobic interactions with the backbone of residues 53–55
(in blue in Figure , also observed for the benzimidazole of 4), suggesting
that the chlorocyanopyridine of 5 provides much of the
binding affinity of this compound. We therefore hypothesized that
combining the benzimidazolone core with the chlorocyanopyridine of 5 would lead to an increase in potency. Gratifyingly, the
resulting compound 6 (CCT365386) showed a significant
improvement in binding affinity, down to ∼10 μM (Table ).
Table 1
Potency and Solubility Data for Hit
and Early Lead Compounds
Data represent
the geometric mean.
See Supporting Information Table S1 for
full statistics. Potency was later confirmed in the TR-FRET assay
described in the manuscript; see Supporting Information Table S2.
Kinetic
solubility measured by NMR
in HEPES buffer (containing 5% DMSO) at pH 8. LE calculated as 1.4
× pIC50/HAC.
Figure 1
Structure-guided
merging of 5 (A, PDB code 6TOG) and 4 (B, PDB code 6TOF) yielded the more
potent 6 (CCT365386) (C, PDB code 6TOH). The surface of
the BCL6 dimer is shown as a grey transparent surface. The two individual
monomers are highlighted in gray and cyan ribbons except for residues
53–55 which are indicated in blue. Key protein residues are
shown in line representation. The compounds are shown as orange ball
and sticks, a selected water molecule is shown as a red sphere, and
H-bonds are shown as yellow dashed lines. In panel C the orange arrow
indicates an exit vector towards a large accessible pocket close to
our compounds. Additional images containing overlays of 4 and 5 with 6 are shown in the Supporting Information Figure S1.
Data represent
the geometric mean.
See Supporting Information Table S1 for
full statistics. Potency was later confirmed in the TR-FRET assay
described in the manuscript; see Supporting Information Table S2.Kinetic
solubility measured by NMR
in HEPES buffer (containing 5% DMSO) at pH 8. LE calculated as 1.4
× pIC50/HAC.Structure-guided
merging of 5 (A, PDB code 6TOG) and 4 (B, PDB code 6TOF) yielded the more
potent 6 (CCT365386) (C, PDB code 6TOH). The surface of
the BCL6 dimer is shown as a grey transparent surface. The two individual
monomers are highlighted in gray and cyan ribbons except for residues
53–55 which are indicated in blue. Key protein residues are
shown in line representation. The compounds are shown as orange ball
and sticks, a selected water molecule is shown as a red sphere, and
H-bonds are shown as yellow dashed lines. In panel C the orange arrow
indicates an exit vector towards a large accessible pocket close to
our compounds. Additional images containing overlays of 4 and 5 with 6 are shown in the Supporting Information Figure S1.The X-ray structure of 6 bound to BCL6 confirmed
the
hypothesized binding mode, combining the features of 4 and 5 (Figure C). Low microsomal clearance and high PAMPA permeability,
combined with good ligand efficiency for a protein–protein
interaction inhibitor hit compound (LE = 0.32 based on the FP assay)
further established 6 as a promising hit compound for
optimization. Despite these good properties, the suboptimal biochemical
activity and solubility meant that we did not observe significant
displacement of co-repressor SMRT from BCL6 in the NanoBRET cellular
target engagement assay (IC50 > 30 μM).[16]
Initial Hit Optimization
Having
discovered an attractive
starting point, our next objective was to improve the biochemical
potency. To avoid being constrained by the tight binding limit of
the FP assay, we developed new TR-FRET assay conditions with reduced
protein concentration (from 3 μM used in the FP assay to 1 nM),
allowing us to measure compound activities in the nM range.[16] From the X-ray structure of 6,
we identified two main areas for optimization: first, growing from
the benzimidazolone-N3 position into a relatively large
cavity, indicated by the arrow in Figure C, and second, by further optimization of
the interactions between the pyridine and the pocket formed by Tyr58.Our aim was to gain potency by extending from the N3 position and filling the pocket while ensuring that hydrogen bond
donor or acceptor groups in the binding site were appropriately satisfied.
In our design, we sought to exploit a hydrophobic patch close to Val18
(Figure C) with a n-butyl chain. This chain would pass close to Asn21, so
we incorporated an alcohol (donor) or nitrile (acceptor) group to
satisfy this interaction, which we hypothesized would also address
the moderate solubility. However, the resulting compounds 11a and 11b showed no improvement in potency (Table ). The X-ray structure of 11a shows both enantiomers
bound to BCL6 with similar occupancies (0.46 and 0.54 for enantiomers
(S) and (R), respectively, Figure ). We hypothesized
that the weak (3.25 Å) hydrogen bond between the hydroxyl of
the (S) enantiomer of 11a and the side
chain of Asn21 is not sufficient to outweigh the desolvation penalty[18,19] and is hence not contributing positively to binding. This was supported
by testing of single enantiomers 11c and 11d and des-hydroxy 11e, which all have comparable potency.
Figure 2
Binding
mode of enantiomers of 11a (PDB code 6TOO). Panel A and panel
B show the binding mode of the respective (S)- and
(R)-enantiomers of 11a. Panel C shows
the (S)-enantiomer and highlights the four water
molecules targeted to gain potency. In all panels the surface of the
BCL6 dimer is shown as a gray transparent surface, with the two individual
monomers highlighted as ribbons and colored in gray and cyan, respectively.
Selected residues are shown in line representation. Enantiomers of 11a are shown as orange ball and sticks, selected water molecules
as a red spheres, and H-bonds as yellow dashed lines.
Binding
mode of enantiomers of 11a (PDB code 6TOO). Panel A and panel
B show the binding mode of the respective (S)- and
(R)-enantiomers of 11a. Panel C shows
the (S)-enantiomer and highlights the four water
molecules targeted to gain potency. In all panels the surface of the
BCL6 dimer is shown as a gray transparent surface, with the two individual
monomers highlighted as ribbons and colored in gray and cyan, respectively.
Selected residues are shown in line representation. Enantiomers of 11a are shown as orange ball and sticks, selected water molecules
as a red spheres, and H-bonds as yellow dashed lines.Data represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.Kinetic solubility measured by NMR
in HEPES buffer (containing 5% DMSO) at pH 8. Where this was <30
μM, solubility by HPLC in PBS buffer and 1% DMSO at pH 7.4 was
also measured and is shown in parentheses.We sought instead to interact indirectly with BCL6
via the network
of water molecules (Figure C) which were observed in the 11a-bound crystal
structure, by changing the position of the hydroxyl group from the
C-2 to C-3 carbon of the butyl chain. Compound 11f showed
a 3-fold improvement in potency (Table ), and a crystal structure of 11f bound
to BCL6 (Figure A)
confirmed the expected binding mode with interactions to three water
molecules. The addition of a methyl group to form the achiral tertiary
alcohol17a provided a further improvement in potency,
consistent with its increase in lipophilicity, while maintaining the
binding mode to BCL6 (Figure B).
Figure 3
Binding mode of 11f and 17a. Panel A
and Panel B show the respective BCL6 binding modes of 11f (PDB code 6TOI) and 17a (PDB code 6TOJ). Compounds are shown as orange ball
and sticks, selected water molecules as red spheres, and H-bonds as
yellow dashed lines. The surface of BCL6 dimer is shown as a gray
transparent surface. The two individual monomers are displayed as
ribbons and colored in gray and cyan, respectively. Key residues are
shown as lines.
Binding mode of 11f and 17a. Panel A
and Panel B show the respective BCL6 binding modes of 11f (PDB code 6TOI) and 17a (PDB code 6TOJ). Compounds are shown as orange ball
and sticks, selected water molecules as red spheres, and H-bonds as
yellow dashed lines. The surface of BCL6 dimer is shown as a gray
transparent surface. The two individual monomers are displayed as
ribbons and colored in gray and cyan, respectively. Key residues are
shown as lines.17a represented
our first submicromolar compound.
Further characterization in the NanoBRET assay demonstrated that 17a is able to disrupt the protein–protein interaction
between full-length co-repressor and BCL6 in cells, albeit with a
∼10-fold drop-off in activity (NanoBRET IC50 = 8.8
μM, compared with TR-FRET IC50 = 0.86 μM).
To gain further binding affinity and hence improve cellular potency,
we explored modifications to the pyridine ring (Table ). The 3-nitrile group points into a small lipophilic pocket,
does not appear to be making polar interactions, and could be replaced
by a chloro atom (17b). We also assessed whether the
potentially reactive 2-chloropyridine in 17b could be
removed and found that resulting compound 17c maintains
activity and improves solubility.Data
represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.Kinetic solubility measured by NMR
in HEPES buffer (containing 5% DMSO) at pH 8. Where this was <30
μM, solubility by HPLC in PBS buffer and 1% DMSO at pH 7.4 was
measured and is shown in parentheses.The crystal structure of BCL6 bound to 17a shows the
pyridine ring making hydrophobic contacts with the electron rich π-system
of Tyr58, and the pyridinenitrogen engaging in a water-mediated H-bond
with Arg28 (Figure B). We hypothesized that a pyrimidine ring could maintain both these
features, and indeed 17d and 17e showed
comparable activity to their pyridine equivalents.Further functionalization
of the pyrimidine or pyridine ring was
explored in order to gain potency. The pyridine 2-position offers
only limited scope for further substitution; the 2-chloro group in 17a (Figure ) is largely filling the available space in this region. Substitution
in the pyridine 5-position could give rise to a steric clash with
the phenyl ring of the benzimidazolone, preventing the molecule from
adopting the binding conformation observed crystallographically: for
example, addition of a chlorine at this position in 17f leads to a 37-fold drop in activity compared to parent 17c. The remaining position (the 2-position on the pyrimidine) was hence
selected for further exploration. Initial examples 17g, 18, and 19 showed that substitution was
tolerated, and a set of 2-heterocyclyl- and heteroarylpyrimidines
was therefore prepared to probe this region (Table ). A variety of groups including aliphatic amines and both
carbon- or nitrogen-linked heteroaryl groups were well tolerated in
this position, consistent with the space available in this solvent-exposed
region of the binding site, with hydrophilic groups such as morpholine20a showing 4-fold better potency than the more hydrophobic
piperidine 20b.Data represent
the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.Data represent the geometric mean
of at least two replicates except where superscript 1 is indicated.
n.d. = not done.Kinetic
solubility measured by NMR
in HEPES buffer and 5% DMSO at pH 8, and/or where data are shown in
brackets, by HPLC in PBS buffer and 1% DMSO at pH 7.4. #These compounds showed <10% degradation of BCL6 in single concentration
experiments at 10 μM, n = 2.We attempted to increase potency
further by increasing steric bulk
to more completely fill this area of the pocket and derive additional
hydrophobic surface contacts. Addition of one or two methyl groups
to pyrazole (compare 23a, 23b, 23c, and 23d), piperidine (compare 20b, 24a, and 24b) and morpholine (compare 20a and 25a) subseries did indeed lead to an increase in
potency, as measured in the TR-FRET assay. For the most potent examples,
pyridine matched-pairs were prepared but were less potent than their
pyrimidine counterparts (compare 23d and 26a, or 24b and 26b, or 25a and 26c, Table ), in contrast to the matched-pairs for simpler analogues (compare 17a and 17d, and 17b and 17e; shown in Table ). This may be due to the different conformational preferences of
the added substituent: the additional steric clash resulting from
the pyridyl CH compared to the pyrimidine N is likely to lead to a
more twisted minimum energy conformation. Compounds with submicromolar
activity in TR-FRET were profiled in the cellular NanoBRET assay and
showed inhibition in the low micromolar range. Further substitution
on the morpholine moiety led to 25b, which was our first
compound to show both sub-100 nM activity in the TR-FRET assay and
submicromolar activity in the cellular NanoBRET assay.X-ray
structures obtained for 23d and 25b provided
possible explanations for the observed increases in potency.
The pyrazole group of 23d forms new interactions: a possible
cation−π stacking interaction with Arg24, and a hydrogen
bond from one of the pyrazolenitrogen atoms to Arg28, which adopts
an alternative conformation (Figure C and Figure D). In contrast, no new polar interactions were observed for 25b (Figure A and Figure B),
which adopts the same binding mode as parent 17a. We
hypothesize that the observed potency enhancement results from the
improved hydrophobic contacts with Tyr58 and the displacement of water
molecules in this region (Figure A and Figure B).
Figure 4
Binding mode of 25b and 23d. Panels A
and B show two different views of the binding mode of 25b (PDB code 6TON). Panels C and D
show two views of the binding mode of 23d (PDB code 6TOK). The panels show
the changed conformation of Arg28 and the new interaction with one
of the compound’s pyrazole nitrogen atoms. In all panels, the
surface of the BCL6 BTB dimer is shown as a gray transparent surface,
with the two individual monomers displayed as ribbons and colored
in gray and cyan. Key residues are shown as lines. Compounds are shown
as orange ball and sticks, selected water molecules as red spheres,
and H-bonds as yellow dashed lines.
Binding mode of 25b and 23d. Panels A
and B show two different views of the binding mode of 25b (PDB code 6TON). Panels C and D
show two views of the binding mode of 23d (PDB code 6TOK). The panels show
the changed conformation of Arg28 and the new interaction with one
of the compound’s pyrazolenitrogen atoms. In all panels, the
surface of the BCL6BTB dimer is shown as a gray transparent surface,
with the two individual monomers displayed as ribbons and colored
in gray and cyan. Key residues are shown as lines. Compounds are shown
as orange ball and sticks, selected water molecules as red spheres,
and H-bonds as yellow dashed lines.
Discovery of BCL6 Degraders
In contrast to the morpholine
analogues 25a and 25b, we were surprised
to see no cellular activity for dimethylpiperidine 24b (CCT368682) in the NanoBRET assay, despite submicromolar potency
in the TR-FRET assay, and good passive permeability (PAMPA Papp is high [36 × 10–6 cm/s]). A large drop-off from TR-FRET to NanoBRET potencies was
also observed for monomethylpiperidine 24a. Upon
detailed examination of the NanoBRET data for 24b, we
observed a dose-dependent reduction in total luminescence,
suggesting either cellular toxicity or depletion of one of the assay
components, for example, BCL6 degradation, as has been previously
reported for another series of inhibitors.[15] To investigate this further, we treated SU-DHL-4 and OCI-Ly1 cells
with 24b; Western analysis of cell lysates demonstrated
a concentration-dependent reduction in BCL6 protein levels, consistent
with compound-mediated BCL6 degradation (Figure ). We note also an apparent elevation in
concentration of BCL6 at compound concentrations below the binding
IC50 which is discussed further below.
Figure 5
Treatment of DLBCL cell
lines with 24b (CCT368682)
leads to degradation of BCL6. SU-DHL-4 or OCI-Ly1 cells were treated
with 24b at the indicated concentration for 4 h at 37
°C. Cells were collected, lysed, and examined by Western blot
for BCL6 and GAPDH protein levels. D: DMSO control.
Treatment of DLBCL cell
lines with 24b (CCT368682)
leads to degradation of BCL6. SU-DHL-4 or OCI-Ly1 cells were treated
with 24b at the indicated concentration for 4 h at 37
°C. Cells were collected, lysed, and examined by Western blot
for BCL6 and GAPDH protein levels. D: DMSO control.Degradation of BCL6 represents an attractive alternative
to inhibition.
As the protein is removed rather than just inhibited, activity at
substoichiometric concentrations of compound is possible as the small
molecule degrader may act catalytically. This removes the need for
complete continuous occupancy of BCL6, as the loss of function would
be maintained until the target is regenerated.[20] We decided to focus on optimization of this novel BCL6
degrader, aiming to identify compounds with suitable physicochemical
and pharmacokinetic properties to investigate in vivo degradation of BCL6 in a tumor xenograft model.To support
optimization, we developed an immunofluorescence assay
to quantify BCL6 degradation in SU-DHL-4 cells. DC50 values
from this assay are shown in Table . Initial SAR appeared tight, with relatively small
changes in chemical structure resulting in the abolition of degradation
activity (Table ):
only the monomethyl- and dimethylpiperidines 24a, 24b, and 26b were identified as degraders, with
DC50 values similar to or below their IC50 values
in the TR-FRET assay. No degradation was observed for pyrazole and
morpholine analogues.
Optimization of DMPK Properties
Piperidine degraders 24a and 24b were subject
to rapid metabolism
in mouse microsomes, which we expected would limit our ability to
achieve sufficient exposure to demonstrate degradation in
vivo. We noted that morpholine25a (Table ) had lower microsomal clearance, leading us to hypothesize
that metabolism of 24a and 24b may be occurring
on the more lipophilic piperidine group. We aimed to reduce microsomal
clearance and hence improve the likelihood of sufficient exposure[21] first by replacing hydrogen atoms with fluorine
to block the potential metabolic sites and second by lowering lipophilicity.Data represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.Kinetic solubility measured by NMR
in HEPES buffer and 5% DMSO at pH 8, and/or where data are shown in
brackets, by HPLC in PBS buffer and 1% DMSO at pH 7.4. #These compounds showed <25% degradation of BCL6 in single concentration
experiments at 10 μM. n.d. = not done.Trifluorination of the piperidine 3-methyl group showed
modest
improvement in microsomal stability (27a), and a more
substantial reduction in clearance was achieved by the introduction
of a difluoro group in the 4-position of the piperidine. The addition
of these fluorine atoms also provided an improvement in both degradation
(DC50) and biochemical inhibition (IC50), for
monomethyl (Table , compare 24a and 27c) and dimethyl (compare 24b and 1) piperidine analogues.In the
previously reported series of BCL6 degraders, only hydrophobic
groups were shown to cause degradation.[15] We therefore designed analogues with lower lipophilicity to investigate
whether degradation could be retained with more hydrophilic groups
in this region and as an alternative approach to lowering metabolic
clearance. We chose to modify the 3-position of the piperidine, as
we had already shown that compounds with hydrophilic groups in the
4-position (morpholine25a and piperazine 25c) were not capable of inducing BCL6 degradation. Hydroxymethyl 27d induced incomplete degradation, with the response plateauing
at ∼60%. Addition of the 4,4-difluoro moiety was found again
to improve both metabolic stability and the ability of the molecule
to trigger BCL6 degradation, with 27e (CCT369900) giving
potent and full (>85%) degradation while maintaining excellent
aqueous
solubility. With 27e we had thus identified an alternative
and, importantly, less lipophilic group that induced degradation.Further modifications demonstrated that the structural requirements
for induction of degradation are quite specific: methylation of the
OH reduced degradation (27f); alternative hydrophilic
groups such as nitrile 27g in this position did not enable
degradation; other discrete changes in structure such as moving the
trifluoromethyl group (27h) and constraining the ring
by forming bicyclic groups (28a, 28b) also
led to a loss of degradation activity.To confirm that degradation
was not cell line specific, we developed
an MSD (Meso Scale Discovery) assay to measure BCL6 levels. Compound 1 (CCT369260), along with analogues 27b and 27e, shows full (>85%) degradation of BCL6 in OCI-Ly1 and
Karpas 422 cells, at DC50 values comparable to those observed
in the SU-DHL-4 immunofluorescence assay (Table ). With degradation activity confirmed, we investigated whether
this translated into antiproliferative activity; examination of degrader
compounds in a 14-day proliferation assay revealed growth inhibition
in both SU-DHL-4 and OCI-Ly1 cell lines. Inhibitor 25b was less effective, suggesting that depletion of BCL6 gives a stronger
effect on proliferation than inhibition alone, consistent with previous
findings.[15] Degraders showed a 5- to 40-fold
reduction in activity in the BCL6 low-expressing[22] OCI-Ly3 cell line compared to OCI-Ly1 (Table ). Degrader 1 was
further profiled in a panel of BCL6-negative cell lines and showed
no antiproliferative activty (Supporting Information Figure S2). These findings suggest that the observed antiproliferative
activity is driven by on-target effects.
Table 6
Antiproliferative
Activity of Key
Compounds
Data represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.
Data represent the geometric mean
of at least two replicates except where superscript i is indicated. #This compound showed <10% degradation of BCL6 in a single
concentration experiment at 10 μM, n = 2.
Data represent the geometric mean
of at least three replicates. See Supporting Information Table S2 for full statistics.Data represent the geometric mean
of at least two replicates except where superscript i is indicated. #This compound showed <10% degradation of BCL6 in a single
concentration experiment at 10 μM, n = 2.
In Vivo Profiling of 1 (CCT369260)
On the basis of
its acceptable microsomal clearance and robust
effects in degradation and proliferation assays, compound 1 was selected for further profiling. A pharmacokinetic study was
carried out in female Balb/Cmice, dosing at 1 mg/kg iv (n = 3) and 5 mg/kg po (n = 3). All mice appeared
normal after dosing and 24 h after dose. Compound 1 demonstrated
moderate clearance (CL 20 mL min–1 kg–1) with mean oral bioavailability of 54%. Protein binding measurements
using equilibrium dialysis showed the compound is highly bound (0.07%
free in SCIDmouse plasma, n = 6). To enable comparison
of free concentrations in vivo with free levels in
the OCI-Ly1DC50 MSD assay, the free fraction in the DC50 assay medium was also measured (2.2% fraction unbound in
IMDM medium, n = 3). This was used to estimate that
a free concentration of greater than 1.1 nM (based on the OCI-Ly1DC50) is required to elicit degradation in vivo, corresponding to a total plasma concentration of 1.5 μM or
a total blood concentration of 2.8 μM (SCIDmouse blood to plasma
ratio of 1.9). PK studies were carried out at three doses (5, 15,
and 50 mg/kg po, Figure ) to determine if sufficient compound exposure could be achieved
in the SCIDmouse strain that would be used for xenograft studies.
Exposure increased with dose between 5 and 15 mg/kg; however, no improvement
was obtained by a further increase to 50 mg/kg. This is likely due
to compound solubility; the compound was dosed as a clear solution
in saline containing 10% DMSO and 5% Tween 20 at the lower doses but
was not completely soluble and hence was dosed as a suspension in
the same vehicle at the 50 mg/kg dose. The mean total blood concentration
at the 15 mg/kg dose was above the desired concentration for approximately
10 h, confirming that this compound was suitable for progression to
a PK/PD study at this dose to determine if, and at what concentration,
BCL6 depletion could be achieved.
Figure 6
(Left) Total SCID mouse blood concentrations
of 1 (CCT369260)
after po dosing at 5 (blue triangles), 15 (red circles), and 50 (green
squares) mg/kg vs predicted active concentration (dashed line). (Right)
Data for individual animals showing cmax, tmax, and total AUC10h in
SCID mouse at 5, 15, and 50 mg/kg doses. All experiments were carried
out according to the U.K. guidelines for animal experimentation.
(Left) Total SCIDmouse blood concentrations
of 1 (CCT369260)
after po dosing at 5 (blue triangles), 15 (red circles), and 50 (green
squares) mg/kg vs predicted active concentration (dashed line). (Right)
Data for individual animals showing cmax, tmax, and total AUC10h in
SCIDmouse at 5, 15, and 50 mg/kg doses. All experiments were carried
out according to the U.K. guidelines for animal experimentation.A PK/PD study was carried out in mice using a single
dose of 1 at 15 mg/kg po in an OCI-Ly1 DLBCL xenograft
model. A clear
decrease in levels of BCL6 in the tumor was observed up to 10 h after
dosing with compound 1 (Figure and Supporting Information Figure S3), with a maximal effect at ∼4 h. No visible
adverse events were observed in this study. Examination of the relationship
between compound 1 and BCL6 levels (Figure and Supporting Information Figure S4) shows that, consistent with predictions,
free concentrations of >1 nM degrader 1 lead to degradation
of BCL6. Less substantial degradation is observed at 0.5 h; this delayed
onset may be due to the time required for the compound to distribute
into tumor. After compound levels drop below the calculated free DC50 at the 10 h time point, BCL6 levels remain low but rebound
by 12 and 16 h, presumably due to the time required for BCL6 resynthesis.
This study demonstrates that degradation of BCL6 in tumors following
oral dosing with a small molecule is a feasible approach.
Figure 7
PK/PD study
with 1 (CCT369260) at 15 mg/kg po. Tumour
xenografts were prepared by subcutaneous injection of 1 × 107 OCI-Ly1 cells in female SCID mice, with dosing of compound
commencing 20 days after injection, to mice with xenografts between
0.5 and 0.8 cm3, as described in more detail in the Supporting Information. Sampling took place at
0.5, 2, 4, 6, 10, 12, 16 and 24 h postdose. All experiments were carried
out according to the U.K. guidelines for animal experimentation. BCL6
levels in tumor were quantified using capillary electrophoresis and
normalized to a GAPDH loading control and are shown as gray (vehicle-treated)
or black (compound treated) bars. Free compound levels at 0.5–24
h are shown (blue dots); dotted line indicates free DC50.
PK/PD study
with 1 (CCT369260) at 15 mg/kg po. Tumour
xenografts were prepared by subcutaneous injection of 1 × 107 OCI-Ly1 cells in female SCIDmice, with dosing of compound
commencing 20 days after injection, to mice with xenografts between
0.5 and 0.8 cm3, as described in more detail in the Supporting Information. Sampling took place at
0.5, 2, 4, 6, 10, 12, 16 and 24 h postdose. All experiments were carried
out according to the U.K. guidelines for animal experimentation. BCL6
levels in tumor were quantified using capillary electrophoresis and
normalized to a GAPDH loading control and are shown as gray (vehicle-treated)
or black (compound treated) bars. Free compound levels at 0.5–24
h are shown (blue dots); dotted line indicates free DC50.
Discussion and Conclusions
We describe the discovery of a series of benzimidazolone inhibitors
of BCL6, including 1 (CCT369260), an orally bioavailable
degrader of BCL6. Weakly active (100 μM) hits derived from high
throughput screening were merged to give benzimidazolone lead compound 6 (CCT365386). By filling a hydrophobic pocket and forming
hydrogen bonds to water molecules, we were able to both improve binding
affinity and incorporate hydrophilic groups to improve solubility.
Further potency improvements were achieved through additional hydrophobic
surface contacts in a solvent-exposed region of the pocket, resulting
in sub-100 nM inhibitors. A subset of these compounds, including 24b (CCT368682), were shown to cause rapid degradation of
BCL6, initially by Western blot and by an immunofluorescence assay
in SU-DHL-4, and later confirmed in different cell lines using a MSD
assay.We noted that our initial Western blot data for 24b showed an apparent increase in BCL6 levels at low concentrations.
A similar effect is also visible in previously presented work on BCL6
degraders,[15] but analysis of curves from
MSD experiments shows that this effect is not observed in all experiments
and, when observed, corresponds to a < 25% increase. It is noteworthy
that we and others[15] also observe an increase
in BCL6 levels when cells are treated with nondegrading inhibitors of BCL6. This suggests that binding may trigger a feedback mechanism
leading to increased expression of BCL6.Further optimization
of initially identified degrader 24b enabled us to identify
compounds with reduced metabolic clearance
including 27e (CCT369900) and 1 (CCT369260).
CCT369260 showed sub-100 nM activity in a degradation assay in SU-DHL-4
cells, and robust antiproliferative activity, in common with previously
identified degraders[15] but in contrast
to BCL6-targetting PROTACs.[14]Similar
to previously reported degraders,[15] we
found that degradation mediated by our compounds is proteasome-dependent.
It is noteworthy that we observe no significant difference in binding
mode (as observed by X-ray crystallography) between 1 (CCT369260) and closely related nondegrader 25a (Figure ), suggesting that
the inhibitor or degrader behavior of a compound may depend on the
nature of the substituents in this region of the molecule rather than
on a distinct binding mode.
Figure 8
Binding modes of the BCL6 degrader 1 (CCT369260) (A)
and inhibitor 25a (B). Crystal structures of the BCL6
BTB domain bound to the BCL6 degrader 1 (CCT369260) (A,
PDB code 6TOM) and inhibitor 25a (B, PDB code 6TOL) show a nearly identical
binding mode. In both panels, the surface of the BCL6 dimer is shown
as a gray transparent surface, with the two individual monomers displayed
in ribbons and colored in gray and cyan. Selected residues are shown
as lines. Compounds are shown as orange ball and sticks, selected
water molecules as red spheres, and H-bonds as yellow dashed lines.
Binding modes of the BCL6 degrader 1 (CCT369260) (A)
and inhibitor 25a (B). Crystal structures of the BCL6BTB domain bound to the BCL6 degrader 1 (CCT369260) (A,
PDB code 6TOM) and inhibitor 25a (B, PDB code 6TOL) show a nearly identical
binding mode. In both panels, the surface of the BCL6 dimer is shown
as a gray transparent surface, with the two individual monomers displayed
in ribbons and colored in gray and cyan. Selected residues are shown
as lines. Compounds are shown as orange ball and sticks, selected
water molecules as red spheres, and H-bonds as yellow dashed lines.One possible hypothesis for how degradation is
induced is via formation
of a ternary complex of the degrader, BCL6, and an E3 ligase. In this
context, we note that the degrader compounds including CCT369260 are
largely buried in the BTB domain, with a relatively small area remaining
solvent-exposed and available to bind to another component (for example,
an E3 ligase) and that in contrast to bivalent PROTAC-type degraders,
no hook effect is observed with increased concentration.[23−25] These data suggest that the degrader molecule alone does not bind
potently to an E3 ligase in the absence of BCL6. One related potential
mechanism by which a small molecule could promote proteosomal degradation
is through “hydrophobic tagging”; the degrader binds
to the target protein and presents as a hydrophobic patch on the protein
surface, mimicking a partially unfolded protein state, which is then
targeted for proteosomal degradation.[23,24,26] Should this be the case for our BCL6 degraders, then
the incorporation of hydrophilic groups in the solvent-exposed region
would not be expected to be tolerated, and a broad range of lipophilic
groups could be expected to trigger degradation. However, we show
first that certain hydrophilic groups are tolerated with degradation
retained and second that the structural requirements for degradation
are quite specific. These data together support an alternative hypothesis:
that the combination of the protein surface and the bound degrader
surface generates a “neo-substrate” or “neo-degron”
that can be recognized by an E3 ubiquitin ligase or other effector
of proteasomal degradation. This is similar conceptually to the mechanism
of action of IMiDs which bind strongly to an E3 ligase and generate
a “neo-enzyme” surface which modifies the substrate
specificity of the ligase, although in our case it is the substrate
surface, rather than that of the E3 ligase, that we hypothesize to
be modified.CCT369260 was progressed into PK studies, and we
showed that the
levels of compound needed to mediate degradation could be achieved in vivo following oral dosing, enabling us to demonstrate
depletion of BCL6 levels in tumors following oral dosing in a mouse
xenograft model. Ongoing work includes further optimization of CCT369260
and the use of our BCL6 degraders as in vitro and in vivo chemical tools to pharmacologically validate BCL6
as a target for the treatment of hematological cancer.
Experimental Section
All in vivo experiments
were carried out according
to the U.K. guidelines for animal experimentation. Cell lines were
supplied by the German Collection of Microorganisms and Cell Cultures
(DSMZ). Cell lines were authenticated by STR profiling and were routinely
screened for Mycoplasma, using an in-house PCR-based
assay.
General Synthetic Information
All anhydrous solvents
and reagents were obtained from commercial suppliers and used without
further purification. Evaporation of solvent was carried out using
a rotary evaporator under reduced pressure at a bath temperature of
up to 60 °C. Flash column chromatography was carried out using
a Biotage purification system using SNAP KP-Sil cartridges or on reverse-phase
mode using SNAP Ultra C18 cartridges. Semipreparative separations
were carried out using an Agilent 1200 series preparative HPLC instrument
over a 15 min gradient elution. Microwave-assisted reactions were
carried out using a Biotage Initiator microwave system. Final compounds
were purified to ≥95% purity. NMR data were collected on a
Bruker Avance 500 spectrometer equipped with a 5 mm BBO/QNP probe
or on a Bruker Avance Neo 600 spectrometer equipped with a 5 mm TCI
CryoProbe. NMR data are presented in the form of chemical shift δ
(multiplicity, coupling constants, integration) for major diagnostic
protons, given in parts per million (ppm) relative to tetramethylsilane
(TMS), referenced to the internal deuterated solvent. HRMS was assessed
using an Agilent 1200 series HPLC instrument and diode array detector
coupled to a 6120 time-of-flight mass spectrometer with dual multimode
APCI/ESI source or on a Waters Acquity UHPLC and diode array detector
coupled to a Waters G2 QToF mass spectrometer fitted with a multimode
ESI/APCI source.
Preparation of Compounds. 5-((5-Chloro-2-((3R,5S)-4,4-difluoro-3,5-dimethylpiperidin-1-yl)pyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (1, CCT369260)
To a mixture of 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18) (2.0 g, 5.0 mmol) were added (3R,5S)-4,4-difluoro-3,5-dimethylpiperidine hydrochloride (1.03 g, 5.6
mmol), NMP (20.2 mL), and DIPEA (2.64 mL, 15.1 mmol). The resulting
mixture was heated to 140 °C for 2 h, cooled to room temperature,
and then poured into ice–water. The resulting precipitate was
collected and washed thoroughly with water to remove residual NMP,
then further washed with ethanol to give a brown solid. Ethanol filtrate
was concentrated and washed with water, then ethanol to give a second
crop of solid. The two solid crops were combined and suspended in
water. Ethanol was added until the brown color was removed from the
solid, which was collected by filtration, washed again with water,
and dried under vacuum to give compound 1 (1.97 g, 3.87
mmol, 77%) as an off-white solid. HRMS (ESI +ve): found 509.2269,
expected 509.2243 for C24H32ClF2N6O2 [M + H]+. δH (600
MHz, CDCl3) 7.98 (s, 1H), 7.32 (d, J =
2.0 Hz, 1H), 7.22 (dd, J = 8.3, 2.0 Hz, 1H), 6.99
(s, 1H), 6.94 (d, J = 8.3 Hz, 1H), 4.60 (br d, J = 12.7 Hz, 2H), 4.04 (t, J = 7.4 Hz,
2H), 3.43 (s, 3H), 2.73 (t, J = 12.7 Hz, 2H), 2.32
(s, 1H), 2.02–1.90 (m, 2H), 1.89 (t, J = 7.4
Hz, 2H), 1.29 (s, 6H), 1.05 (d, J = 6.7 Hz, 6H).
A suspension of cyclopropylmethanamine (17
μL, 0.2 mmol), triethylamine (25 μL, 0.18 mmol), and 2,4-dichloronicotinonitrile
(30 mg, 0.17 mmol) in DMA (0.35 mL) was stirred at rt for 18 h. The
reaction was quenched with brine and extracted with EtOAc. The combined
organics were washed with water and brine, dried, and concentrated,
then purified by flash column chromatography (0–15% EtOAc in
cyclohexane) to give 5 (26 mg, 0.13 mmol, 72%). δH (500 MHz, CDCl3) 8.09 (d, J =
6.1 Hz, 1H), 6.48 (d, J = 6.1 Hz, 1H), 5.31 (br s,
1H), 3.16–3.10 (m, 2H), 1.19–1.07 (m, 1H), 0.74–0.64
(m, 2H), 0.36–0.30 (m, 2H).
To a mixture
of N1-methyl-4-nitrobenzene-1,2-diamine
(10.5 g, 62.8 mmol) in acetonitrile (150 mL) cooled to 0 °C was
added portionwise disuccinimidyl carbonate (21.7 g, 84.7 mmol) over
20 min. The mixture was stirred at rt for 18 h and then poured into
ice–water, forming a precipitate. The solid was collected by
suction filtration and washed sequentially with water, DCM, and diethyl
ether to give 7 (11.12 g) as an orange-red solid. HRMS
(ESI +ve): found 194.0572, expected 194.0560 for C8H8N3O3+ [M + H]+. δH (500 MHz, DMSO-d6) 11.43 (s, 1H), 8.03 (dd, J = 8.7, 2.3 Hz, 1H),
7.76 (d, J = 2.3 Hz, 1H), 7.31 (d, J = 8.7 Hz, 1H), 3.36 (s, 3H).
To a mixture of 5-amino-1-methyl-1H-benzo[d]imidazol-2(3H)-one
(750 mg, 4.6 mmol) and 2,4-dichloropyridine-3-carbonitrile (760 mg,
4.4 mmol) under argon was added DMA (10 mL) followed by DIPEA (0.90
mL, 5.19 mmol). The reaction mixture was heated at 120 °C under
microwave irradiation for 45 min then allowed to cool to rt and added
dropwise to a stirring mixture of methanol:water (1:1; 120 mL). The
resulting precipitate was filtered, washed with water (2 × 25
mL) and diethyl ether (2 × 30 mL), affording 2-chloro-4-[(1-methyl-2-oxo-3H-benzimidazol-5-yl)amino]pyridine-3-carbonitrile
(1.30 g, 99%, 4.3 mmol) as a beige solid. δH (500
MHz, DMSO-d6) 10.95 (br s, 1H), 9.39 (br
s, 1H), 7.99 (d, J = 6.2 Hz, 1H), 7.13 (d, J = 8.3 Hz, 1H), 6.95 (dd, J = 8.3, 1.9
Hz, 1H), 6.90 (d, J = 1.9 Hz, 1H), 6.65 (d, J = 6.2 Hz, 1H), 3.29 (s, 3H).
To a mixture of 1-methyl-5-nitro-1H-benzo[d]imidazol-2(3H)-one (240 mg, 1.242
mmol) and cesium carbonate (485 mg, 1.489 mmol) in DMF (3 mL) was
added 2-ethyloxirane (0.119 mL, 1.367 mmol), and the resulting mixture
was heated in the microwave to 120 °C for 1 h, then added to
water (10 mL). The resulting mixture was extracted with DCM, and combined
organics were dried over sodium sulfate, filtered, and evaporated
onto silica gel for purification by flash column chromatography (10
g silica, 40–60% ethyl acetate in cyclohexane). Fractions were
combined and evaporated to give the title compound (200 mg, 60%) as
a yellow-orange oil which solidified on standing. δH (500 MHz, DMSO-d6) 8.13 (d, J = 2.2 Hz, 1H), 8.05 (dd, J = 8.7, 2.3
Hz, 1H), 7.36 (d, J = 8.8 Hz, 1H), 4.90 (d, J = 5.4 Hz, 1H), 3.88 (dd, J = 14.2, 4.2
Hz, 1H), 3.82 (dd, J = 14.2, 7.4 Hz, 1H), 3.70 (m,
1H), 3.41 (s, 3H), 1.49 (m, 1H), 1.37 (m, 1H), 0.92 (t, J = 7.4 Hz, 3H).
To a suspension of 3-(2-hydroxybutyl)-1-methyl-5-nitro-1,3-dihydro-2H-benzo[d]imidazol-2-one (10) (34 mg, 0.13 mmol) in ethanol (1.5 mL) and DMSO (0.5 mL) was added
sodium dithionite (116 mg, 0.67 mmol). The resulting mixture was heated
to 90 °C for 5 h, then stirred at room temperature for 2 days,
then heated again to 90 °C for 2 h. Additional sodium dithionite
(116 mg, 0.67 mmol) was added, and the mixture was heated to 90 °C
for 2 h. Water (1 mL) was added, and the resulting cloudy suspension
was stirred at rt for 3 days, then diluted with water until all was
in solution. The pH was adjusted to 10 using 2 M sodium carbonate,
and the resulting mixture was extracted with DCM (×3), dried
by passing through a phase separator, and evaporated under reduced
pressure. The resulting material was diluted with DMA (0.5 mL), and
then 2,4-dichloronicotinonitrile (15 mg, 0.087 mmol) and DIPEA (25
μL, 0.143 mmol) were added. The resulting mixture was heated
in the microwave to 120 °C for 30 min, then partitioned between
DCM and water. The aqueous layer was acidified to pH 7 using 10% citric
acid, then extracted with DCM (×3). Combined organics were evaporated
and purified by HPLC (ACE 5 C18-PFP 250 mm × 21.2 mm column,
15 min gradient from 60:40 to 0:100 water:methanol, 0.1% formic acid
modifier) at a flow rate of 20 mL/min. The resulting yellow oil was
triturated with diethyl ether to give compound 11a (4.2
mg, 0.011 mmol, 8.5%) as pale yellow solid. HRMS (ESI +ve): found
372.1230, expected 372.1222 for C18H19ClN5O2+ [M + H]+. δH (500 MHz, CD3OD) 7.94 (d, J =
6.2 Hz, 1H), 7.22 (d, J = 8.2 Hz, 1H) overlapping
with 7.22 (d, J = 1.9 Hz, 1H), 7.07 (dd, J = 8.4, 1.9 Hz, 1H), 6.70 (d, J = 6.3
Hz, 1H), 3.98–3.89 (m, 1H), 3.89–3.80 (m, 2H), 3.47
(s, 3H), 1.67–1.55 (m, 1H), 1.48 (dp, J =
14.6, 7.3 Hz, 1H), 1.03 (t, J = 7.4 Hz, 3H).
To a mixture of cesium carbonate (70 mg,
0.21 mmol) and 2-chloro-4-((1-methyl-2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)amino)nicotinonitrile
(9) (25 mg, 0.08 mmol) in DMF (0.50 mL) was added 2-(bromomethyl)butanenitrile
(25 mg, 0.15 mmol). The resulting mixture was heated to 60 °C
for 3 days, then partitioned between water and DCM, and the aqueous
layer acidified to pH 5 using a 10% citric acid solution. Layers were
separated, the aqueous phase was extracted with further DCM, and combined
organics were evaporated under reduced pressure and then purified
by HPLC (ACE 5 C18-PFP 250 mm × 21.2 mm column, 15 min gradient
from 45:55 to 30:70 water:methanol, 0.1% formic acid modifier) at
a flow rate of 20 mL/min to give compound 11b as an off-white
solid. HRMS (ESI +ve): found 381.1209, expected 381.1225 for C19H18ClN6O+ [M + H]+. δH (500 MHz, CDCl3) 8.04 (d, J = 6.1 Hz, 1H), 7.08 (d, J = 1.9 Hz, 1H),
7.07 (d, J = 8.2 Hz, 1H), 7.02 (dd, J = 8.2, 1.9 Hz, 1H), 6.96 (s, 1H), 6.71 (d, J =
6.1 Hz, 1H), 4.11 (dd, J = 14.4, 5.7 Hz, 1H), 4.04
(dd, J = 14.5, 8.8 Hz, 1H), 3.48 (s, 3H), 3.11 (tt, J = 9.1, 5.4 Hz, 1H), 1.89–1.69 (m, 2H), 1.19 (t, J = 7.4 Hz, 3H).
To a mixture of 2-chloro-4-((1-methyl-2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)amino)nicotinonitrile
(9) (20 mg, 0.067 mmol) and cesium carbonate (23 mg,
0.07 mmol) in DMF (0.50 mL) was added (2S)-2-ethyloxirane
(7 μL, 0.08 mmol), and the resulting mixture was heated in the
microwave to 120 °C for 1 h. Two further batches of (2S)-2-ethyloxirane (7 μL, 0.08 mmol) were added, with
microwave heating to 140 °C for 30 min after each. The mixture
was partitioned between water and DCM, and the aqueous layer was acidified
to pH 5 using a 10% citric acid solution. Layers were separated, the
aqueous phase was extracted with further DCM, and combined organics
were evaporated under reduced pressure and then purified by HPLC (ACE
5 C18-PFP 250 mm × 21.2 mm column, 15 min gradient from 50:50
to 35:65 water:methanol, 0.1% formic acid modifier) at a flow rate
of 20 mL/min to give compound 11c (7.5 mg, 0.020 mmol,
30%) as an off white solid. HRMS (ESI +ve): found 372.1218, expected
372.1222 for C18H19ClN5O2+ [M + H]+. δH (500 MHz, CD3OD) identical to racemate 11a.
Prepared by the same method as its enantiomer 11c, starting from (2R)-2-ethyloxirane. HRMS
(ESI +ve): found 372.1218, expected 372.1222 for C18H19ClN5O2+ [M + H]+. δH (500 MHz, CD3OD) identical to racemate 11a.
To 2-chloro-4-((1-methyl-2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)amino)nicotinonitrile
(9) (25 mg, 0.084 mmol) and cesium carbonate (33 mg,
0.10 mmol) in DMF (0.85 mL) was added 1-bromobutane (9.5 μL,
0.089 mmol), and the resulting mixture was stirred at 140 °C
under microwave irradiation for 1 h. Water (10 mL) was added, and
the aqueous mixture was acidified with 1 M HCl. The aqueous mixture
was extracted with EtOAc (5 × 10 mL). The organic extracts were
combined, dried (Na2SO4), and concentrated in
vacuo, then purified by HPLC (ACE 5 C18-PFP 250 mm × 21.2 mm
column; 15 min gradient of 40:60 to 20:80 water:methanol, 0.1% formic
acid modifier) at a flow rate of 20 mL/min to give compound 11e (12.3 mg, 41%, 0.035 mmol) as an off-white solid. HRMS
(ESI +ve): found 356.1263, expected 356.1273 for C18H19ClN5O+ [M + H]+. δH (500 MHz, CDCl3) 8.03 (d, J =
6.0 Hz, 1H), 7.02 (d, J = 8.1 Hz, 1H), 6.98 (dd, J = 8.1, 1.7 Hz, 1H), 6.93 (br s, 1H), 6.87 (d, J = 1.7 Hz, 1H), 6.60 (d, J = 6.0 Hz, 1H),
3.88 (t, J = 7.3 Hz, 2H), 3.46 (s, 3H), 1.77–1.68
(m, 2H), 1.45–1.35 (m, 2H), 0.97 (t, J = 7.3
Hz, 3H).
To a solution
of 3-methylbutane-1,3-diol (27
g, 0.26 mol) and triethylamine (40 mL, 0.29 mol) in DCM (250 mL) at
0 °C was added dropwise 4-methylbenzenesulfonyl chloride (50
g, 84 mmol) in 200 mL of DCM over 1 h. The resulting mixture was stirred
at 0 °C for 4 h, then water was added (200 mL) and stirred for
45 min. Adjustment to pH 10 was made by addition of 2 M NaOH and separated.
Organic phase was washed sequentially with 200 mL of half-saturated
sodium bicarbonate solution (×2), water, 1 M HCl, brine, then
dried over magnesium sulfate, filtered, and concentrated to give a
yellow oil. This was dissolved in DCM/cyclohexane and purified by
flash column chromatography (0–40% ethyl acetate in cyclohexane)
to give the title compound (39.9 g) as a colorless oil. δH (500 MHz, CDCl3) 7.85–7.78 (m, 2H), 7.40–7.34
(m, 2H), 4.23 (t, J = 6.8 Hz, 2H), 2.47 (s, 3H),
1.88 (t, J = 6.8 Hz, 2H), 1.23 (s, 6H).
A solution of (3R)-butane-1,3-diol
(0.86 mL, 9.5 mmol) in dry dichloromethane (10 mL) under a nitrogen
atmosphere was cooled in a salt–ice bath (bath temp −12
°C). Triethylamine (2.25 mL, 16.1 mmol) was added followed by
a solution of 4-methylbenzenesulfonyl chloride (2 g, 10.5 mmol) in
dry dichloromethane (6 mL) over 10 min. The resulting mixture was
allowed to warm slowly to room temperature and stirred for 20 h, then
diluted with DCM, washed with 10% citric acid, sat. sodium bicarbonate
solution, brine, dried over sodium sulfate, filtered, and evaporated
under reduced pressure to give a clear oil. This was purified by flash
column chromatography (50 g silica, 10–30% ethyl acetate in
cyclohexane) to give 14b (1.65 g, 6.75 mmol, 71%) as
a clear oil. δH (500 MHz, CDCl3) 7.82
(br d, J = 8.3 Hz, 2H), 7.37 (br d, J = 7.9 Hz, 2H), 4.26 (ddd, J = 10.0, 8.7, 5.0 Hz,
1H), 4.13 (dt, J = 10.0, 5.5 Hz, 1H), 3.96 (dqd, J = 9.7, 6.2, 3.5 Hz, 1H), 2.47 (s, 3H), 1.85 (dddd, J = 14.5, 8.7, 5.8, 3.6 Hz, 1H), 1.71 (ddt, J = 14.2, 8.9, 5.0 Hz, 1H), 1.21 (d, J = 6.3 Hz,
3H).
A mixture of 1-methyl-5-nitro-1,3-dihydro-2H-benzo[d]imidazol-2-one (7, 9.05
g, 46.9 mmol), cesium carbonate (30.5 g, 93.7 mmol), and 3-hydroxy-3-methylbutyl
4-methylbenzenesulfonate (14a, 16.4 g, 63.5 mmol) in
acetonitrile (234 mL) was heated to reflux for 4 h. The mixture was
concentrated under reduced pressure. The residue was then diluted
with water (500 mL) and stirred for 30 min, forming a precipitate,
which was filtered and washed thoroughly with water. The collected
solid was dried under vacuum to give 15 (12 g, 43 mmol,
92%) as a red-brown solid. δH (500 MHz, CDCl3) 8.13 (dd, J = 8.7, 2.0 Hz, 1H), 7.95 (d, J = 2.0 Hz, 1H), 7.04 (d, J = 8.7 Hz, 1H),
4.13 (m, 2H), 3.50 (s, 3H), 1.93 (m, 2H), 1.34 (s, 6H).
Palladium on activated charcoal (10% Pd, 240 mg) was added
to a solution of 3-(3-hydroxy-3-methylbutyl)-1-methyl-5-nitro-1,3-dihydro-2H-benzo[d]imidazol-2-one (15) (6.3 g, 22.6 mmol) in ethanol (113 mL), which was then stirred at
60 °C under an atmosphere of hydrogen for 3 h. The mixture was
filtered through a pad of Celite and washed through with ethanol.
The resulting solution was then concentrated under reduced pressure
to give compound 16 (5.85 g) as a pale pink solid. δH (500 MHz, CD3OD) 6.90 (d, J =
8.1 Hz, 1H), 6.63 (d, J = 1.9 Hz, 1H), 6.58 (dd, J = 8.1, 1.9 Hz, 1H), 3.95 (m, 2H), 3.36 (s, 3H), 1.84 (m,
2H), 1.29 (s, 6H).
A mixture of 5-amino-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (16) (30 mg, 0.12 mmol), DIPEA (30 uL, 0.17 mmol), and 4,5,6-trichloropyrimidine
(26 mg, 0.14 mmol) in DMF (0.9 mL) was heated in the microwave to
120 °C for 30 min. The mixture was partitioned between water
and DCM, and the aqueous layer was acidified to pH 4 using a 10% citric
acid solution. Layers were separated, the aqueous phase was extracted
with further DCM, and combined organics were evaporated under reduced
pressure and then purified by reverse phase flash chromatography (Biotage
12g SNAP Ultra C18, 20–100% methanol in water, 0.1% formic
acid modifier) and then further triturated with diethyl ether to give
compound 17e as a beige solid (41 mg, 0.10 mmol, 86%).
HRMS (ESI +ve): found 396.0997, expected 396.0989 for C17H20Cl2N5O2+ [M + H]+. δH (500 MHz, DMSO-d6) 9.46 (s, 1H), 8.26 (s, 1H), 7.30 (d, J = 2.0 Hz, 1H), 7.21 (dd, J = 8.4, 2.0
Hz, 1H), 7.13 (d, J = 8.4 Hz, 1H), 4.45 (s, 1H),
3.91–3.84 (m, 2H), 3.33 (s, 3H, obscured by solvent), 1.74–1.66
(m, 2H), 1.16 (s, 6H).
A mixture of 5-amino-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (16) (1.61 g, 6.47 mmol), 2,4,5-trichloropyrimidine (0.88 mL, 7.69 mmol),
and cesium carbonate (4.21 g, 12.93 mmol) in DMF (15 mL) was heated
in the microwave to 120 °C for 30 min. The resulting mixture
was diluted with water, acidified to pH 5 by addition of 10% citric
acid, and extracted with DCM. The combined organics were evaporated
under reduced pressure, and the resulting sticky solid was dissolved
in a minimum volume of ethyl acetate and precipitated by addition
of diethyl ether. The resulting solid was collected by filtration
and washed with diethyl ether and dried under reduced pressure, giving
the title product (1.96 g, 72%, 4.69 mmol) as a solid. δH (500 MHz, DMSO-d6) 9.57 (s, 1H),
8.34 (s, 1H), 7.35 (d, J = 1.9 Hz, 1H), 7.19 (dd, J = 8.4, 1.9 Hz, 1H), 7.15 (d, J = 8.4
Hz, 1H), 4.44 (s, 1H), 3.92–3.86 (m, 2H), 3.33 (s, 3H), 1.76–1.69
(m, 2H), 1.17 (s, 6H).
A mixture of 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18) (20 mg, 0.05 mmol), dimethylamine hydrochloride (37 mg, 0.45 mmol),
cesium carbonate (164 mg, 0.50 mmol) in NMP (0.5 mL) was heated in
the microwave to 180 °C for 1 h. Mixture was partitioned between
DCM and water, aqueous layer was extracted with DCM, and combined
organics were washed with Brine. Organic layers were combined and
evaporated, and the resulting NMP solution was purified by HPLC (ACE
5 C18-PFP 250 mm × 21.2 mm column; 15 min gradient of 40:60 to
25:75 water:methanol, 0.1% formic acid modifier) at a flow rate of
20 mL/min to give compound 19 (8 mg, 0.0188 mmol, 37%).
HRMS (ESI +ve): found 405.1790, expected 405.1800 for C19H26ClN6O2+. δH (500 MHz, acetone-d6) 7.96 (s,
1H), overlapping with 7.98–7.94 (0.6H, partly exchanged NH),
7.73 (d, J = 1.7 Hz, 1H), 7.35 (dd, J = 8.1, 1.7 Hz, 1H), 7.05 (d, J = 8.1 Hz, 1H), 4.02
(t, J = 7.9 Hz, 2H), 3.38 (s, 3H), 3.13 (s, 6H),
1.85 (t, J = 7.9 Hz, 2H), 1.25 (s, 6H).
A mixture of morpholine (0.031 mL, 0.25 mmol), 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18, 20 mg, 0.05 mmol), and DIPEA (0.04 mL, 0.25 mmol) in NMP (1 mL)
was heated in the microwave to 180 °C for 1 h, then partitioned
between DCM and water. The aqueous layer was extracted with DCM, and
combined organics were washed with brine. The organic layers were
combined and evaporated, and the resulting material (containing residual
NMP) was purified by preparative HPLC (ACE 5 C18-PFP column (5 μm,
250 mm × 21.2 mm), 15 min gradient elution from 40:60 to 25:75
water:methanol (both modified with 0.1% formic acid) at a flow rate
of 20 mL/min) to give 20a (18 mg, 0.038 mmol, 76%). HRMS
(ESI +ve): found 447.1897, expected 447.1906 for C21H28ClN6O3+ [M + H]+. δH (500 MHz, CD3OD) 7.92 (s, 1H), 7.47
(d, J = 1.7 Hz, 1H), 7.31 (dd, J = 8.4, 1.7 Hz, 1H), 7.10 (d, J = 8.4 Hz, 1H), 4.00
(m, 2H), 3.70–3.67 (m, 4H), 3.66–3.62 (m, 4H), 3.41
(s, 3H), 1.84 (m, 2H), 1.28 (s, 6H).
A mixture of 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18) (20 mg, 0.05 mmol), 2,4-dimethyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thiazole
(12.07 mg, 0.05 mmol), sodium carbonate (10.7 mg, 0.10 mmol), and
bis(triphenylphosphine)palladium(II) chloride (1.8 mg,
0.0025 mmol) in 1,4-dioxane (0.40 mL) and water (0.40 mL) was heated
in the microwave at 130 °C for 30 min. The mixture was partitioned
between DCM and water and pH adjusted to 5 using 10% citric acid before
separation and extraction with further DCM. The organic layers were
combined and evaporated, and the resulting solution was purified by
HPLC (ACE 5 C18-PFP 250 mm × 21.2 mm column, 15 min gradient
from 40:60 to 25:75 water:methanol, 0.1% formic acid modifier) at
a flow rate of 20 mL/min to give compound 21 (2 mg, 0.0042
mmol, 8%) as a white solid. HRMS (ESI +ve): found 473.1501, expected
473.1521 for C22H26ClN6O2S+ [M + H]+. δH (500 MHz,
acetone-d6) 8.50 (br, 1H), 8.38 (s, 1H),
7.54 (d, J = 1.8 Hz, 1H), 7.34 (dd, J = 8.3, 1.8 Hz, 1H), 7.13 (d, J = 8.3 Hz, 1H), 4.07
(t, J = 7.9 Hz, 2H), 3.41 (s, 3H), 2.63 (s, 3H),
2.60 (s, 3H), 1.89 (t, J = 7.9 Hz, 2H), 1.24 (s,
6H).
A mixture of 1-methyl-2-(tributylstannyl)-1H-imidazole (21 mg, 0.06 mmol), 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18) (20 mg, 0.05 mmol), and bis(triphenylphosphine)palladium(II)
chloride (3.5 mg, 0.005 mmol) in 1,4-dioxane (1 mL) was heated for
18 h at 90 °C. The mixture was partitioned between DCM and water
and pH adjusted to 5 using 10% citric acid before separation and extraction
with further DCM. The organic layers were combined and evaporated,
and the resulting solution was purified by HPLC (ACE 5 C18-PFP 250
mm × 21.2 mm column, 15 min gradient from 40:60 to 25:75 water:methanol,
0.1% formic acid modifier) at a flow rate of 20 mL/min to give compound 22 (3 mg, 0.0068 mmol, 13%) as the formic acid salt. HRMS
(ESI +ve): found 442.1740, expected 442.1753 for C21H25ClN7O2+ [M + H]+. δH (500 MHz,, acetone-d6) 8.50 (br, 1H), 8.37 (s, 1H), 8.12 (s, 1H), 7.74 (br s, 1H), 7.64
(br s, 1H), 7.55 (d, J = 2.0 Hz, 1H), 7.35 (dd, J = 8.6, 2.0 Hz, 1H), 7.14 (d, J = 8.6
Hz, 1H), 4.04 (t, J = 8.3 Hz, 2H), 3.91 (s, 3H),
3.40 (s, 3H), 1.88 (t, J = 8.3 Hz, 2H), 1.24 (s,
6H).
Method as for 19, using 1H-pyrazole, heating in the microwave to 170 °C for 1 h. Purified
by HPLC (ACE 5 C18-PFP 250 mm × 21.2 mm column; 15 min gradient
of 60:40 to 0:100 water:methanol, 0.1% formic acid modifier) at a
flow rate of 20 mL/min to give compound 23a. HRMS (ESI
+ve): found 428.1552, expected 428.1596 for C20H23ClN7O2+ [M + H]+. δH (500 MHz, DMSO-d6) 9.42 (br,
1H), 8.45 (s, 1H), 8.38 (d, J = 2.2 Hz, 1H), 7.79
(s, 1H), 7.63 (d, J = 2.2 Hz, 1H), 7.42 (dd, J = 8.4, 2.2 Hz, 1H), 7.17 (d, J = 8.4
Hz, 1H), 6.52 (s, 1H), 3.91 (t, J = 7.6 Hz, 2H),
3.34 (s, 3H), 1.75 (t, J = 7.6 Hz, 2H), 1.14 (s,
6H).
5-((5-Chloro-2-(3-methyl-1H-pyrazol-1-yl)pyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (23b) and 5-((5-Chloro-2-(5-methyl-1H-pyrazol-1-yl)pyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (23c)
A mixture of 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18) (25 mg, 0.063 mmol), 3-methylpyrazole (0.05 mL, 0.62 mmol), and
cesium carbonate (100 mg, 0.31 mmol) in NMP (0.5 mL) was heated in
the microwave to 170 °C for 1 h. The mixture was partitioned
between DCM and water and pH adjusted to 5 using 10% citric acid before
separation and extraction with further DCM. The organic layers were
combined and evaporated, and the resulting solution was purified by
HPLC (ACE 5 C18-PFP 250 mm × 21.2 mm column; 15 min gradient
of 45:55 to 20:80 water:methanol, 0.1% formic acid modifier) at a
flow rate of 20 mL/min. Two products were obtained: the major, later
eluting regioisomer was assigned as compound 23b (18
mg, 0.039 mmol, 61%). HRMS (ESI +ve): found 442.1755, expected 442.1753
for C21H25ClN7O2+ [M + H]+. δH (500 MHz, CDCl3) 8.34 (s, 1H), 8.27 (d, J = 2.6 Hz, 1H), 7.63 (d, J = 2.0 Hz, 1H), 7.32 (s, 1H, NH), 7.18 (dd, J = 8.3, 2.0 Hz, 1H), 7.02 (d, J = 8.3 Hz, 1H), 6.23
(d, J = 2.6 Hz, 1H), 4.15–4.08 (m, 2H), 3.47
(s, 3H), 2.40 (s, 3H), 1.94 (t, J = 7.3 Hz, 2H),
1.30 (s, 6H). The minor, earlier eluting regioisomer was assigned
as compound 23c (2.5 mg, 0.005 mmol, 8.5%). HRMS (ESI
+ve): found 442.1756, expected 442.1753 for C21H25ClN7O2+ [M + H]+. δH (500 MHz,, CDCl3) 8.37 (s, 1H), 7.74 (d, J = 2.0 Hz, 1H), 7.63 (m, 1H), 7.29 (s, 1H, NH), 7.08 (dd, J = 8.3, 2.0 Hz, 1H), 6.98 (d, J = 8.3
Hz, 1H), 6.16 (s, 1H), 4.13–4.06 (m, 2H), 3.45 (s, 3H), 2.46
(s, 3H), 1.97–1.91 (m, 2H), 1.29 (s, 6H).
A mixture of 5-((2,5-dichloropyrimidin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (18) (30 mg, 0.076 mmol), 2,2,6,6-tetramethylmorpholine (22 mg, 0.15
mmol), and DIPEA (40 μL, 0.23 mmol) in NMP (0.67 mL) was heated
in the microwave to 140 °C for 2 h. Once cooled, the mixture
was diluted with DMSO (0.5 mL) and then purified by reverse-phase
chromatography eluting from 30% to 100% methanol in water (each containing
0.1% formic acid) to give compound 25b (32 mg, 0.064
mmol, 84%) as a pale brown solid. HRMS (ESI +ve): found 503.2521,
expected 503.2532 for C25H36ClN6O3+ [M + H]+. δH (600
MHz, CDCl3) 7.97 (s, 1H), 7.36 (d, J =
2.0 Hz, 1H), 7.24 (dd, J = 8.4, 2.0 Hz, 1H), 7.00
(s, 1H), 6.93 (d, J = 8.4 Hz, 1H), 4.04 (t, J = 7.2 Hz, 2H), 3.58 (s, 4H), 3.42 (s, 3H), 1.89 (t, J = 7.2 Hz, 2H), 1.28 (s, 6H), 1.23 (s, 12H).
A mixture of 3,5-dimethyl-1H-pyrazole
(3 mg, 0.03 mmol), Pd2(dba)3 (1.5 mg, 0.0017
mmol), Xantphos (4.1 mg, 0.007 mmol), cesium carbonate (14 mg, 0.04
mmol), and 5-((2-bromo-5-chloropyridin-4-yl)amino)-3-(3-hydroxy-3-methylbutyl)-1-methyl-1,3-dihydro-2H-benzo[d]imidazol-2-one (29, 9 mg, 0.02 mmol) in NMP:toluene (1:1 v/v, 0.8 mL) was heated in
the microwave to 140 °C for 1 h. The resulting mixture was diluted
with water, acidified to pH 5 by addition of 10% citric acid, then
extracted with DCM (5 mL × 3). Organic layer was collected and
passed through a Si-DMT palladium scavenging column, then evaporated
under reduced pressure. Product was purified by HPLC (ACE 5 C18-PFP
250 mm × 21.2 mm column; 15 min gradient of 40:60 to 25:75 water:methanol,
0.1% formic acid modifier) at a flow rate of 20 mL/min to give 26a (4 mg, 0.0084 mmol, 41%). HRMS (ESI +ve): found 455.1943,
expected 455.1957 for C23H28ClN6O2+ [M + H]+. δH (500
MHz, acetone-d6) 8.19–8.14 (m,
1.6H, including partly exchanged NH), 7.39 (s, 1H), 7.19–7.16
(m, 2H), 7.09 (dd, J = 8.2, 2.0 Hz, 1H), 5.97 (s,
1H), 4.03 (t, J = 7.9 Hz, 2H), 3.42 (s, 3H), 2.57
(s, 3H), 2.10 (s, 3H), 1.87 (t, J = 7.9 Hz, 2H),
1.23 (6H, s).
A solution of cis-3,5-dimethylpiperidine
(0.23 g, 2.0 mmol), 5-chloro-2-fluoro-4-iodopyridine (0.53 g, 2.0
mmol), and DIPEA (0.53 mL, 3.1 mmol) in THF (8 mL) was heated in a
sealed vial to 100 °C for 16 h. When cooled, water was added
to the THF solution and extracted with EtOAc. The combined organic
layers were washed with water twice and dried with sodium sulfate.
Flash column chromatography (4% ethyl acetate in cyclohexane) gave 30a (488 mg, 1.39 mmol, 68%) as a white solid. δH (500 MHz, DMSO-d6) 8.08 (s, 1H),
7.40 (s, 1H), 4.36–4.10 (m, 2H), 2.25 (dd, J = 12.9, 11.4 Hz, 2H), 1.75 (dtd, J = 10.6, 3.6,
1.9 Hz, 1H), 1.62–1.46 (m, 2H), 0.88 (d, J = 6.6 Hz, 6H), 0.84–0.70 (m, 1H).
Authors: Leandro C Cerchietti; Alexandru F Ghetu; Xiao Zhu; Gustavo F Da Silva; Shijun Zhong; Marilyn Matthews; Karen L Bunting; Jose M Polo; Christophe Farès; Cheryl H Arrowsmith; Shao Ning Yang; Monica Garcia; Andrew Coop; Alexander D Mackerell; Gilbert G Privé; Ari Melnick Journal: Cancer Cell Date: 2010-04-13 Impact factor: 31.743
Authors: Mariano G Cardenas; Wenbo Yu; Wendy Beguelin; Matthew R Teater; Huimin Geng; Rebecca L Goldstein; Erin Oswald; Katerina Hatzi; Shao-Ning Yang; Joanna Cohen; Rita Shaknovich; Kenno Vanommeslaeghe; Huimin Cheng; Dongdong Liang; Hyo Je Cho; Joshua Abbott; Wayne Tam; Wei Du; John P Leonard; Olivier Elemento; Leandro Cerchietti; Tomasz Cierpicki; Fengtian Xue; Alexander D MacKerell; Ari M Melnick Journal: J Clin Invest Date: 2016-08-02 Impact factor: 14.808
Authors: Jose M Polo; Tania Dell'Oso; Stella Maris Ranuncolo; Leandro Cerchietti; David Beck; Gustavo F Da Silva; Gilbert G Prive; Jonathan D Licht; Ari Melnick Journal: Nat Med Date: 2004-11-07 Impact factor: 53.440
Authors: Jeffrey L Gustafson; Taavi K Neklesa; Carly S Cox; Anke G Roth; Dennis L Buckley; Hyun Seop Tae; Thomas B Sundberg; D Blake Stagg; John Hines; Donald P McDonnell; John D Norris; Craig M Crews Journal: Angew Chem Int Ed Engl Date: 2015-06-17 Impact factor: 15.336
Authors: Matthew G Lloyd; Rosemary Huckvale; Kwai-Ming J Cheung; Matthew J Rodrigues; Gavin W Collie; Olivier A Pierrat; Mahad Gatti Iou; Michael Carter; Owen A Davis; P Craig McAndrew; Emma Gunnell; Yann-Vaï Le Bihan; Rachel Talbot; Alan T Henley; Louise D Johnson; Angela Hayes; Michael D Bright; Florence I Raynaud; Mirco Meniconi; Rosemary Burke; Rob L M van Montfort; Olivia W Rossanese; Benjamin R Bellenie; Swen Hoelder Journal: J Med Chem Date: 2021-11-30 Impact factor: 7.446
Authors: Rosemary Huckvale; Alice C Harnden; Kwai-Ming J Cheung; Olivier A Pierrat; Rachel Talbot; Gary M Box; Alan T Henley; Alexis K de Haven Brandon; Albert E Hallsworth; Michael D Bright; Hafize Aysin Akpinar; Daniel S J Miller; Dalia Tarantino; Sharon Gowan; Angela Hayes; Emma A Gunnell; Alfie Brennan; Owen A Davis; Louise D Johnson; Selby de Klerk; Craig McAndrew; Yann-Vaï Le Bihan; Mirco Meniconi; Rosemary Burke; Vladimir Kirkin; Rob L M van Montfort; Florence I Raynaud; Olivia W Rossanese; Benjamin R Bellenie; Swen Hoelder Journal: J Med Chem Date: 2022-06-02 Impact factor: 8.039
Authors: Owen A Davis; Kwai-Ming J Cheung; Alfie Brennan; Matthew G Lloyd; Matthew J Rodrigues; Olivier A Pierrat; Gavin W Collie; Yann-Vaï Le Bihan; Rosemary Huckvale; Alice C Harnden; Ana Varela; Michael D Bright; Paul Eve; Angela Hayes; Alan T Henley; Michael D Carter; P Craig McAndrew; Rachel Talbot; Rosemary Burke; Rob L M van Montfort; Florence I Raynaud; Olivia W Rossanese; Mirco Meniconi; Benjamin R Bellenie; Swen Hoelder Journal: J Med Chem Date: 2022-06-03 Impact factor: 8.039
Authors: Mikołaj Słabicki; Hojong Yoon; Jonas Koeppel; Lena Nitsch; Shourya S Roy Burman; Cristina Di Genua; Katherine A Donovan; Adam S Sperling; Moritz Hunkeler; Jonathan M Tsai; Rohan Sharma; Andrew Guirguis; Charles Zou; Priya Chudasama; Jessica A Gasser; Peter G Miller; Claudia Scholl; Stefan Fröhling; Radosław P Nowak; Eric S Fischer; Benjamin L Ebert Journal: Nature Date: 2020-11-18 Impact factor: 49.962