Structure-Based Design of Selective Salt-Inducible Kinase Inhibitors

pubs.acs.org/jmc

Article

Roberta Tesch,^#Marcel Rak,^#Monika Raab, Lena M. Berger, Thales Kronenberger, Andreas C. Joerger,

Benedict-Tilman Berger, Ismahan Abdi, Thomas Hanke, Antti Poso, Klaus Strebhardt, Mourad Sanhaji,

and Stefan Knapp *

Cite This: J. Med. Chem. 2021, 64, 8142 −8160

Read Online

sı

*

Metrics & More

Article Recommendations

Supporting Information

ACCESS

ABSTRACT: Salt-inducible kinases (SIKs) are key metabolic regulators.

The imbalance in SIK function is associated with the development of

diverse cancers, including breast, gastric, and ovarian cancers. Chemical

tools to clarify the roles of SIK in diﬀerent diseases are, however, sparse and

are generally characterized by poor kinome-wide selectivity. Here, we have

adapted the pyrido[2,3-d]pyrimidin-7-one-based p21-activated kinase

(PAK) inhibitor G-5555 for the targeting of SIK, by exploiting diﬀerences

in the back-pocket region of these kinases. Optimization was supported by

high-resolution crystal structures of G-5555 bound to the known oﬀ-targets,

MST3 and MST4, leading to a chemical probe, MRIA9, with dual SIK/

PAK activity and excellent selectivity over other kinases. Furthermore, we

show that MRIA9 sensitizes ovarian cancer cells to treatment with the mitotic agent paclitaxel, conﬁrming earlier data from genetic

knockdown studies and suggesting a combination therapy with SIK inhibitors and paclitaxel for the treatment of paclitaxel-resistant

ovarian cancer.

and decreases PI3K and AKT phosphorylation levels.¹³

INTRODUCTION

■

Interestingly, SIK2 knockdown re-sensitizes ovarian cancer

Salt-inducible kinases (SIK1-3) are members of the AMP-

activated protein kinase (AMPK) sub-family of the calcium/

calmodulin-dependent kinase (CaMK) group. These serine/

cells to taxane-based chemotherapy.¹⁴

We have recently shown that in samples from patients with

gastric cancer, there were increased levels of SIK2 mRNA and

threonine kinases act as regulators of energy homeostasis and

protein in advanced stages of the tumor compared with a

metabolic stress. The SIK family member SIK2, for example, is

lower-grade tumor, independent of the metastatic stage.¹⁵

activated in cells recovering from starvation, leading to

Overall, these data highlight the complex roles of SIK family

proteins in diﬀerent types of cancer, making them important

therapeutic targets. To clarify the multifaceted roles of these

kinases in disease and normal physiology, chemical tools are

urgently needed. A number of potent SIK inhibitors have been

reported and have been used to investigate the roles of SIK

kinases in cancer (Figure 1). Among those inhibitors are the

tyrosine kinase inhibitors dasatinib (1) and bosutinib (2),

which are currently used for the treatment of chronic

myelogenous leukemia by targeting Abelson murine leukemia

viral oncogene homologue 1 (ABL) and SRC family kinases,

which all bear a threonine residue at the gatekeeper position.¹⁶

SIK kinases also harbor a threonine gatekeeper and are

therefore inhibited by these and many other tyrosine kinase

phosphorylation and hence activation of the transcription

factor cAMP response element-binding protein (CREB1).

However, SIK family members also repress CREB1-mediated

gene expression by modulating the activity of the associated

transducer of regulated CREB1 (TORC).¹⁻³In addition to the

key function of SIK in regulating metabolism, an imbalance in

SIK has been observed in the context of several diseases,

especially in cancer, with both tumor-promoting and tumor-

suppressive roles being reported.⁴SIK2 is often deleted in

breast cancer, and downregulation of SIK1 has been linked to

not only a tumor suppressor role but also the development of

metastasis by promoting p53-dependent anoikis.⁵In contrast,

SIK3 is highly expressed in breast cancer patients, with an

eﬀect on the inﬂammation process in this type of cancer,⁶

while SIK2 is often ampliﬁed in diﬀuse large B-cell lymphoma

(DLBCL), increasing cell survival, and tumor progression.⁷In

leukemia, SIK3 is critical for cell proliferation by regulating

subcellular localization and activity of HDAC4.^8,9In ovarian

cancer, overexpression of SIK2 has been observed in an

adipocyte-rich environment of metastatic deposits, where it

enhances lipid synthesis, resulting in poor prognosis.¹⁰⁻¹²

Depletion of SIK2 in ovarian cancer delays mitotic progression

Received: December 11, 2020

Published: June 4, 2021

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

8142

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 1. Selectivity proﬁle of SIK inhibitors. (A) Chemical structures of known SIK inhibitors. (B) Selectivity proﬁle of G-5555 at a concentration

of 1 μM, using the scanMAX kinase assay panel of 468 kinases (Euroﬁns Scientiﬁc). The data are shown on a kinase phylogenetic tree, provided by

Euroﬁns Scientiﬁc. Inhibited targets are highlighted by red circles, with the circle radius representing % activity compared with untreated control.

Activities on mutant kinases are not shown. Data summarized in this ﬁgure can be found in Table S2. (C) Comparison of the selectivity proﬁle of

three SIK inhibitors using the KINOMEscan assay platform (Euroﬁns Scientiﬁc) performed for 468 kinases. Data are reported as percent of activity

remaining in the presence of an inhibitor relative to activity measured in the absence of an inhibitor (100%).

inhibitors. HG-9-91-01 (3) is a pan-SIK inhibitor with IC₅₀

values of 0.92, 6.6, and 9.6 nM for SIK1, SIK2, and SIK3,

respectively, but it also inhibits several other kinases with

important roles in tumorigenesis, including SRC, YES, BTK,

FGFR1, and ephrin receptors.¹⁷The inhibitor YKL-05-099 (4)

has been designed based on HG-9-91-01 as a more selective

chemical tool to investigate the roles of SIK2 in vivo, which

highlighted the possibility of a therapeutic window for SIK

inhibitors to achieve immunomodulatory eﬀects without

misbalancing the normal glucose metabolism.¹⁸However, 4

is still not suﬃciently selective for mechanism-based studies

because it inhibited more than 60 kinases by more than 90% at

1 μM in a panel of 468 kinases.¹⁸Recently, compound ARN-

3236 (5) has been reported as a panSIK inhibitor with modest

selectivity toward SIK2 over SIK1 and SIK3, with IC₅₀values

of less than 1, 21, and 6 nM, respectively.¹⁹Treatment of

ovarian cancer cells with ARN-3236 prevented centrosome

separation and led to sensitization to paclitaxel treatment.¹⁴

However, selectivity data for 5 are only available against 74

kinases, at a single concentration. This rather limited selectivity

screen has identiﬁed signiﬁcant activity of this compound

(more than 80% inhibition at 0.5 μM) against the kinases

JAK2, LCK, NUAK2, SRPK1, and VEGFR2 (Figure S1).¹⁹

Thus, all currently available SIK inhibitors display signiﬁcant

oﬀ-target activity, limiting their use as chemical probes.

We therefore aimed to develop a well-characterized chemical

tool for SIK family members that lacks activity on tyrosine

kinases and other oﬀ-targets. In addition, sequence con-

servation within the ATP-binding sites suggests that isoform

selectivity would be diﬃcult to achieve using ATP competitive

inhibitors. Searching the literature for a chemical starting point

for the development of SIK inhibitors with improved

selectivity, we focused on three criteria: (i) a small number

of reported oﬀ-targets, (ii) suﬃciently comprehensive

8143

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 2. Crystal structure of MST3 in complex with G-5555. (A) Binding mode analysis of G-5555 in MST3. The insets show the water network

around the amino group of the 1,3-dioxane moiety in the HRD region (left) and the interactions of the 6-methylpyridine moiety in the back-pocket

region (right). Selected hydrogen bonds are highlighted by dashed lines. (B) 2F_o−F_comitted electron density map around the ligand shown at a

contour level of 2.0 σ. (C) Comparison of the back-pocket regions of PAK1 (PDB code 5DEY), MST3 (PDB code 7B30), and SIK2 (homology

model).

selectivity proﬁling on a large set of kinases, and (iii) a

synthetically easily accessible scaﬀold. Of particular interest to

us was the p21-activated kinase (PAK) inhibitor G-5555 (6),

which has been developed by Ndubaku and co-workers.²⁰In a

comprehensive selectivity panel of 235 kinases, G-5555 (6)

showed not only inhibition of the PAK family members,

related STE kinases (KHS1, MST3, MST4, and YSK1), and

the tyrosine kinase LCK but also potent activity against

SIK2.²⁰The dual activity of G-5555 against STE kinases as

well as SIK is intriguing, considering that these kinases share

very little sequence homology, including key residue changes

such as the gatekeeper. We therefore hypothesized that

selectivity between the main known G5555 targets within

the STE family and SIK can be achieved. Another PAK1

inhibitor structurally related to G-5555 was developed by

Rudolph and co-workers, but this inhibitor was proﬁled against

a comparatively small panel of 96 kinases,²¹which motivated

us in selecting G-5555 (6) as a chemical starting point for this

study. Here, we report the development and biological

evaluation of a suitable chemical tool compound targeting

SIK based on scaﬀold 6. Improved potency and selectivity were

achieved by structural variation of the back-pocket binding

motif of compound 6, which was guided by crystallographic

data on STE oﬀ-targets and molecular dynamics (MD)

simulations.

RESULTS AND DISCUSSION

■

Extended Selectivity Proﬁle of SIK Inhibitor G-5555.

To validate compound G-5555 (6) as a suitable starting

scaﬀold for the development of selective SIK inhibitors, we

decided to proﬁle the selectivity of G-5555 more broadly using

the scanMAX kinase assay panel of 468 kinases (Euroﬁns

Scientiﬁc), revealing a few additional oﬀ-targets (Tables S1 and

S2). Comparing the screening data for 4 and 6 at 1 μM

compound concentration, compound 6 showed a signiﬁcantly

improved selectivity proﬁle (Figure 1). In addition, screening

8144

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 3. Stabilization of G-5555 during MD simulation and water thermodynamic proﬁle. (A) WaterMap simulation of the G-5555 binding

pocket in unbound MST3 highlighting favorable (green) and unfavorable (red) hydration sites during the simulation and their location relative to

bound G-5555. (B) Frequency of the interaction (in percent) between the inhibitor and a particular residue during the MD simulation and the

distance proﬁle between the catalytic lysine and the pyridine moiety of G-5555 in both SIK2 (green) and MST3 (purple).

compound 6 at lower concentrations (100 nM) revealed only

few oﬀ-targets (Figure S2). Thus, given the sequence and

structural diversity of SIK compared with STE kinases,

compound 6 represented an attractive starting point for the

development of selective SIK inhibitors.

revealed (i) that the chlorinated aromatic ring system is

perpendicular to the pyrido[2,3-d]pyrimidin-7-one and the 6-

methylpyridine ring system, positioning the pyridine ring near

the catalytic residue Lys65 (MST3 residue numbering), (ii) a

hydrogen bond interaction of the 5-amino-1,3-dioxane group

with Asn161, and most interestingly, and (iii) the formation of

a water-mediated hydrogen-bond network linking the glycine-

rich loop (P-loop) residue Ser46, the catalytic Lys65, the

pyridine ring of G-5555, and the carbonyl group of the

pyrido[2,3-d]-pyrimidin-7-one ring. Notably, in the reported

complex of G-5555 with PAK1 (PDB code 5DEY), the

glycine-rich loop has moved out of the ATP-binding pocket,

and the pyridine ring is packed between Val342 and Glu315.

Importantly, MST3 and MST4 harbor a bulkier isoleucine

(Ile109 and Ile97, respectively) instead of the valine (Val342)

present in PAK1 (Figure S4A). When comparing the

complexes G-5555-MST3/4 and G-5555/PAK1 with the SIK

family homology model, a number of diﬀerences in the back-

pocket region and other ATP-binding site residues became

apparent (Figure 2C). Most strikingly, SIK family members

bear a threonine residue at the gatekeeper position and a

Strategy for Designing Selective SIK Pyrido[2,3-

d]pyrimidin-7-one Inhibitors. Crystallizing any isoform of

SIK has been challenging so far due to poor expression yields

in insect cell expression systems and instability of the protein.

To provide a structural basis for the rational design of

compounds with high selectivity for SIK kinases, we therefore

determined crystal structures of G-5555 bound to the oﬀ-

targets MST3 and MST4 (PDB codes 7B30 and 7B36) Figures

2A,B and S3. The resolution of the structures was 2.1 Å in both

cases. These structures then served as surrogate models for

rational modiﬁcations and as a template for generating a SIK2

family homology model. We chose SIK2 as a representative

family member. However, all speciﬁc SIK2 model residues

discussed in the article and considered for the inhibitor design

are conserved among SIK family proteins. The binding mode

of G-5555 in both of these highly conserved kinase domains

8145

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 4. Design of a focused series of pyrido[2,3-d]pyrimidin-7-one analogues of the PAK1 inhibitor G-5555. Based on the back-pocket

interactions of G-5555 in both PAK1 and MST3/4, the pyridyl moiety was substituted at positions 3, 4, and 6 with diﬀerent electron-withdrawing

and -donating groups.

methionine at position +4 after the conserved glutamic acid on

the αC helix. In contrast, both MSTs and PAK1 have a

methionine in the gatekeeper position and either a leucine

(Leu86 in MST3) or a methionine (Met319 in PAK1) at

position +4 on the αC helix. The pyrido[2,3-d]-pyrimidin-7-

one ring of G-5555 forms hydrogen-bond interactions with the

hinge region in a similar way in both MST structures and the

PAK1 complex.

5555 mediated by Lys65 (Figure 3B). Through the course of

the simulation, the interaction between the Lys65 and the

pyridyl moiety, and Lys65 and Ser46 occurred in 96 and 92%

of the snapshots along the 4 μs simulation, respectively,

suggesting a major role of the MST3 glycine-rich loop in G-

5555 stabilization upon binding. Interestingly, the interaction

between the G-5555 pyridyl moiety and the equivalent lysine

(Lys45) in the SIK2 complex was more ﬂexible during the

simulation (Figure 3B). The overall movement of the P-loop in

the SIK2 model was analyzed by monitoring the distance

between Asn30 (conserved in all three SIK family members)

and the back pocket (Figure S5A). To assess ﬂexibility in the

back pocket, we monitored the distance between the

gatekeeper residues and the αC helix in both SIK2 and

MST3, showing that the dynamics of this region are more

constrained in MST3 than in SIK2 (Figure S5B) and

suggesting that certain modiﬁcations of the pyridyl moiety

may impair binding to MST3 but not SIK family proteins. On

the basis of this MD analysis and the characteristic amino acid

variations in the pyridyl moiety binding pocket, we

hypothesized that modiﬁcations at the ortho position of the

pyridyl moiety may destabilize the interaction with the back

pocket of MST but not SIK kinases.

To explore whether the water network observed was not due

to crystal packing eﬀects, we determined the water

thermodynamic proﬁle with a WaterMap simulation. The

water proﬁle was calculated from MD simulations with explicit

waters and analyzed as a cluster of hydration sites for

calculation of entropy, enthalpy, and free energy properties.

We identiﬁed conserved hydration sites that matched the

position of the water molecules in the crystal structure MST3/

G-5555 between Ser46 and Glu82 (hydration site 5) and

between Lys65 and the carbonyl group of the pyrido[2,3-d]-

pyrimidin-7-one ring (hydration site 4) (Figure 3A). These

enthalpically favorable hydration sites suggest that the water

network observed in the back-pocket region is not due to

crystal packing but comprises stable structural water molecules

that contribute to stabilizing the binding mode of G-5555. In

addition, we analyzed the binding site without the bound

compound and observed unfavorable hydration sites that

overlap with the binding region of G-5555. The optimal

geometry and space ﬁll of the inhibitor is therefore important

for eﬃciently displacing those unfavorable water molecules

from the binding pocket.

Based on the crystallographic data on the G-5555 binding

mode and the in silico calculations, we designed a ﬁrst series of

compounds in order to explore the potential of the pyrido[2,3-

d]pyrimidin-7-one scaﬀold for the development of selective

SIK inhibitors. For the optimization of back-pocket inter-

actions, we focused on substitution at positions 3, 4, and 6 of

the pendant pyridyl moiety (R1), while modiﬁcation in R2

addressed the optimization of the solvent-exposed region

(Figure 4).

To provide additional structural insights for inhibitor

optimization, we next carried out MD simulations on the

complex of MST3/G-5555 and the model SIK2/G-5555 (4

μs). Our simulation on the MST3 complex showed a stable

interaction between the Ser46 and the pyridyl moiety of G-

Synthesis of G-5555 Derivatives. Compounds 7−18 are

synthesized using a nine-step synthetic route, based on the

8146

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Scheme 1. Synthesis of Intermediate 22

a

Reagents and conditions: (a) H₂SO₄and ethanol, 15 h, 100 °C; (b) 4-amino-2-(methylthio)-pyrimidine-5-carbaldehyde, K₂CO₃, and DMF, 16 h,

120 °C; and (c) bis(pinacolato)diboron, [1,1′-bis(diphenylphosphino)-ferrocene]dichloropalladium(II) (Pd(dppf)Cl₂), KOAc, and dioxane, 18 h,

115 °C.

a

Scheme 2. Synthesis of Pyrido[2,3-d]pyrimidin-7-one Derivatives (7−18)

a

Reagents and conditions: (a) phthalic anhydride and DMF, 4 h, 90 °C; (b) 2-bromo-1,1-diethoxyethane, p-toluenesulfonic acid, and toluene, 24 h,

115 °C; (c) 2-bromopyridines, Pd(dppf)Cl₂, KOAc, and dioxane/H₂O (2:1), 18 h, 107 °C; (d) Cs₂CO₃and DMF, 18 h, 100 °C; (e) 3-

chloroperbenzoic acid (m-CPBA) and DCM, 3 h, RT; (f) o-anisidine, HCl, and ethanol, 17 h, 100 °C; and (g) methylamine, N,N′-

diisopropylethylamine (DIPEA), and ethanol, 17 h, 85 °C. For R1 and R2, see Table 1.

synthetic strategy published by Rudolph et al.,²¹as outlined in

Schemes 1 and 2. The synthesis started with the esteriﬁcation

of commercially available 2-(4-bromo-2-chlorophenyl)acetic

acid (19) with ethanol to obtain ethyl 2-(4-bromo-2-chloro-

phenyl)acetate (20) (step a, Scheme 1). A ring-closure

reaction of 20 with 4-amino-2-(methylthio)pyrimidine-5-

carbaldehyde under basic conditions (step b, Scheme 1) led

to the formation of the pyrido[2,3-d]pyrimidine scaﬀold 21. In

the following Miyaura borylation reaction (step c, Scheme 1),

the bromo substituent of 21 was replaced by a boronic acid

pinacol ester to obtain 22. The 5-amino-1,3-dioxane moiety,

required for the subsequent nucleophilic substitution, was

synthesized in a two-step reaction. The ﬁrst step was the

protection of the amine group of serinol (23) with a

phthalimide protecting group (step a, Scheme 2), leading to

24. In the following ring-closure reaction (step b, Scheme 2),

24 was treated with 2-bromo-1,1′-diethoxyethane and p-

toluenesulfonic acid in toluene to obtain the 5-amino-1,3-

dioxane ring 25. A Suzuki cross-coupling reaction (step c,

Scheme 2) enabled the derivatization of the lead structure by

substituting the boronic acid pinacol ester of 22 with diﬀerent

pyridine derivatives, leading to the formation of the

corresponding compounds 26−37. 25 was used in the

subsequent nucleophilic substitution reaction (step d, Scheme

2) with compounds 26−37 to obtain the corresponding

compounds 38−49. The last step of the synthetic route started

with an oxidation of the methylthioether of 38−49 with m-

CPBA (step e, Scheme 2), leading to a mixture of the

corresponding sulfoxide and sulfone product. The mixture was

used in the following nucleophilic aromatic substitution

without further puriﬁcation. Compound 7 was synthesized

under acidic conditions by the treatment of the corresponding

mixture of sulfoxide/sulfone product with o-anisidine (step f,

Scheme 2), and compounds 8−18 were synthesized under

basic conditions by treatment with methylamine (step g,

Scheme 2).

Structure−Activity Relationship of G-5555-Based

Compound Series. We evaluated the activity of the

compounds against SIK2 and SIK3 using the cellular target

engagement assay NanoBRET^22,23and compared these data

with the oﬀ-target proﬁle on MSTs and PAK1 based on the

melting temperature shift (ΔT_m) from a diﬀerential scanning

ﬂuorimetry (DSF) assay.²⁴In DSF assays, the melting point of

a protein in the presence and absence of a compound was

measured. In this assay format, the thermal denaturation of a

protein is measured with the help of a ﬂuorescent dye that

changes its ﬂuorescence properties upon binding to hydro-

phobic surfaces exposed in unfolded proteins. The relative

binding strength of a ligand is measured by an increase in the

protein melting temperature (ΔT_m) compared to the unbound

protein. A comparison with binding aﬃnities measured in

solution shows that ΔT_mvalues of a target protein linearly

correlate with the aﬃnity of ligands with conserved binding

modes.²³The pan-SIK inhibitors 3 and 4 were used as a

positive control together with G-5555 (6). The assay data

correlated well when comparing the melting temperature

proﬁles on oﬀ-targets with the IC₅₀values reported in the

literature. G-5555 showed melting temperature shifts (ΔT_m) of

8147

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Table 1. Selectivity Proﬁle of Pyrido[2,3-d]pyrimidin-7-one Series 7−18 as SIK2 Inhibitors

a

b

IC₅₀values were determined using the NanoBRET assay in a 10-point dose−response curve (duplicate measurements). Average of three

measurements plus the standard deviation of the mean. Literature K_dvalues for staurosporine on SIK2 and SIK3.^25,26

c

8148

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 5. Selectivity proﬁle of MRIA9 (10). (A) Optimization of pyrido[2,3-d]pyrimidin-7-one compound 6 (G-5555) by altering the substitution

pattern on the pendant pyridine ring yielded compound 10 (MRIA9) with increased cellular potency and selectivity for SIK. (B) Comparison of

selectivity data of MRIA9 (10) and G-5555 (6) against 443 kinases, including kinase mutants, as percentage of remaining activity. (C) Mapping of

kinases inhibited by MRIA9 (10), shown on the kinase phylogenetic tree illustrated with the web-based tool CORAL.²⁸Inhibited targets are

highlighted by circles, with the circle radius representing the in vitro IC₅₀values. The circles colored in gray highlight the oﬀ-targets that were only

weakly inhibited in the NanoBRET assay. Activities on mutant kinases are not shown. Data displayed in this ﬁgure are included in Table S5. (D)

Results of screening of 10 against 443 kinases (as percent of control < 50% at 1 μM) as well as in vitro and in cellulo IC₅₀values for the main kinase

a

b

targets. IC₅₀values were determined by a radiometric protein kinase assay (³³PanQinase activity assay, at Reaction Biology). IC₅₀values were

determined using the NanoBRET assay in a 10-point dose−response curve (duplicate measurements).

9.9, 6.6, and 6.6 °C for MST3, MST4, and PAK1, respectively.

The reported IC₅₀values for these targets are 43 nM, 20 nM,

and 3.7 nM for MST3, MST4, and PAK1, respectively.²⁰Our

ﬁrst attempt was to develop a hybrid of G-5555 (6) and YKL-

05-099 (4) because the ortho-methoxyphenyl moiety present

in 4 had been highlighted as an important moiety for achieving

selectivity for MSTs and PAK1.¹⁸This eﬀort led to compound

7, which was equipotent toward SIK2 and 4 times more potent

toward SIK3 in the NanoBRET assay compared with G-5555

(6) and YKL-05-099 (4), but it still potently inhibited MST3,

MST4, and PAK1 as shown by ΔT_mvalues of 9.5, 7.8, and 5.0

°C (Table 1). We conﬁrmed the lack of selectivity of

compound 7 using the scanMAX kinase assay panel of 468

kinases (Euroﬁns Scientiﬁc) at a concentration of 1 μM,

obtaining an S-score (35) of 0.18, a value that represents the

count of data points with higher than 35% inhibition divided

8149

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

by the total number of data points (excluding mutants) (Figure

S6A,B and Table S3).

PAK1, which has a larger methionine residue at the

corresponding position in the back pocket (Figure S8).

These data highlight the crucial role of substitutions at

positions 3 and 6 of the G-5555 core scaﬀold for modulating

SIK2 and SIK3 activity. Interestingly, the substitution of the

methyl group at position 6 with a triﬂuoromethyl group (CF₃)

in compound 15 abrogated activity for SIK2 (IC₅₀> 4 μM)

and lowered the potency for SIK3 by a factor of 20 (IC₅₀= 508

4 nM), without aﬀecting the thermal shifts for MST3/4 and

PAK1. The STE oﬀ-target proﬁle of the diﬂuoromethyl

analogue (CHF₂), 16, was similar to that of the parent

molecule G-5555, with, however, the aﬃnity for both SIK2 and

SIK3 reduced by about 2-fold, while maintaining potency for

MST3/4 and PAK1. The co-crystal structure of 16 with MST3

(PDB code 7B33) revealed that the acidic hydrogen of the

CHF₂group interacted with the side-chain hydroxyl of Ser46

in the glycine-rich loop (calculated distance of 2.4 Å between

the hydroxyl oxygen and the acidic hydrogen of the CHF₂

moiety). Ser46 in MST3 is replaced by an asparagine (Asn30)

in SIK2, which may explain the diﬀerential eﬀect of the CHF₂

and CF₃moieties on binding aﬃnity in the two kinases (Figure

S9).

Selectivity Proﬁle of the pan-SIK Inhibitor MRIA9.

With our focused series of SIK inhibitors based on the G-5555

scaﬀold, we were able to signiﬁcantly improve the cellular

potency of the core scaﬀold against SIK2 and SIK3 but,

importantly, also the selectivity over the main known oﬀ-

targets of the parent compound. MRIA9 showed the best

balance of SIK2/3 inhibition potency and activity against STE

kinases, and we therefore screened this compound at 1 μM

against 443 kinases in the radiometric ³³PanQinase activity

assay (Reaction Biology) (Table S5), which returned an S-

Score (35) of 0.018. In comparison, selectivity data of G-5555

measured at the same concentration resulted in an S-Score

(35) of 0.12, conﬁrming that our strategy was successful in

designing a more selective SIK inhibitor (Figure 5B).

Interestingly, a favorable balance between potency and

selectivity was achieved not only by the addition of a group

at position 3 of the pyridine ring but also by the removal of the

methyl group at position 6, which seems key for the increased

selectivity of this series.

Next, we followed up the kinome-wide screening by

determining IC₅₀values in enzyme kinetic assays for all the

oﬀ-targets identiﬁed with inhibition values higher than 50%

(Figure 5D). MRIA9 was equipotent toward SIK1-3 and

PAK2, whereas it was 20 times less potent against PAK1, with

an in vitro IC₅₀of only 580 nM. In addition, MRIA9 was only 5

times less potent against KHS1 and NLK compared with SIK

inhibition in vitro. However, when comparing the results from

cellular NanoBRET assays, MRIA9 had IC₅₀of 516 5, 180

40, and 127 23 nM on SIK1, SIK2, and SIK3, respectively,

and it did not target KHS1 and NLK in the cells. We also

achieved selectivity over MST4 in vitro and in cellulo as

expected based on our proﬁling data.

Unfortunately, we were unable to assess cellular PAK activity

in the NanoBRET assay due to a lack of suitable tracer

molecules for this kinase. Nevertheless, our results show that

MRIA9 is a highly potent dual pan-SIK and PAK2/3 inhibitor

with excellent cellular activity, especially against SIK3.

Together with a selective PAK inhibitor as a control, such as

the highly selective allosteric inhibitor NVS-PAK1,²⁷MRIA9

(10) can be used as a chemical probe for elucidating the

complex biology of SIK kinases.

We also determined the crystal structure of compound 7

with MST3, revealing that the ortho-methoxyphenyl ring

resulted in a small induced-ﬁt movement of the side chain of

Tyr113 by about 1.0 Å (PDB code 7B32) (Figure S6C).

Because substitutions in the solvent-exposed area of the ligand

did not improve selectivity, we focused on the methyl group at

position 6 of the pyridyl moiety. Removing this methyl group

in compound 8 resulted in lower potency against SIK2 (IC₅₀

=

777 187 nM) and SIK3 (IC₅₀= 117 25 nM), and ΔT_mof

5.7 °C and 3.9 °C for MST3 and MST4, respectively,

indicating that the methyl group is important for MST binding.

The addition of a ﬂuorine atom at position 3 in compound 9

increased the cellular potency against SIK2 and SIK3, but 9

still retained considerable activity against MST3/4 and PAK1

as judged by ΔT_mdata. The selectivity proﬁle of compound 9

was very similar to that of 6 (G-5555), showing no improved

selectivity (Figure S7B,C and Table S4). Consistent with the

data for compound 8 which demonstrated the importance of

the methyl group at position 6 of the pyridine ring for potency

against MST3/4, the non-methylated analogue 10 (MRIA9)

showed reduced ΔT_mvalues for MST3/4 by 5.9 and 3.4 °C

compared with G-5555. To our surprise, the ﬂuorine

substitution at position 3 seemed to compensate for the lack

of the methyl group at position 6, retaining high potency

toward SIK2 (IC₅₀= 180 40 nM) and SIK3 (IC₅₀= 127

23 nM). The chlorine analogue 11 had a similar cellular

potency in the NanoBRET assay (IC₅₀of 139

110 nM),

conﬁrming that halogen substitution at position 3 can account

for high potency toward both SIK2 and SIK3. The comparison

of the crystal structures of MST3 in complex with 9 (PDB

code 7B34) and MST3 in complex with 10 (MRIA9) (PDB

code 7B31) revealed the molecular basis for the importance of

the methyl group for MST3/4 inhibition potency. In the MST

complexes, the methyl group of the ligand points toward

Phe47 in the glycine-rich loop, forming additional hydrophobic

interactions. In addition, the methyl group in position 6

increases the electron-donating properties of the pyridine

nitrogen, thereby strengthening the hydrogen-bond network

with Lys65 and Ser46 (Figure S7A).

When introducing a methyl group in position 3 (compound

12), not only the potency toward SIK2 and SIK3 increased to

65 35 and 14 1 nM, respectively, but also the melting

temperatures for MST4 and PAK1 increased, suggesting

poorer selectivity. Considering the environment of the back

pocket of SIK2, PAK1, and MST3/4, we expected that the

addition of a bulkier group at position 3 might impair the

activity for PAK1, which was conﬁrmed with the 3-methoxy

analogues 13 and 14, which induced melting temperature

shifts, ΔT_ms, of only 2.9 and 3.1 °C. However, only one of

those two compounds with a 3-methoxy modiﬁcation, 14,

retained potent SIK2 activity, with an IC₅₀of 600 146 nM

compared with an IC₅₀> 2 μM for 13, again highlighting the

importance of the methyl group at position 6 for the activity

against both SIK2 and MST3/4. To our surprise, compound

13 maintained considerable potency against SIK3 but

presented a similar kinome proﬁle than G-5555 (data not

shown). The crystal structure of MST3 in complex with 14

(PDB code 7B35) showed that the methoxy group of 14 is

accommodated in the back pocket next to Leu86. Such an

arrangement would result in steric hindrance upon binding to

8150

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Low Concentration of MRIA9 Inhibits the Auto-

Phosphorylation Activity of SIK2 In Vitro. To further

validate the speciﬁcity and the potency of MRIA9 over G-5555

in inhibiting SIK2, we performed an in vitro kinase auto-

phosphorylation assay. Puriﬁed GST-tagged SIK2 was

incubated with increasing concentrations of both inhibitors

in the presence of radioactive [γ-³²P]ATP. The auto-

phosphorylation of SIK2 was monitored as an indicator of its

activity. Exposing SIK2 to increasing concentrations of MRIA9

led to a gradual decrease in auto-phosphorylation activity. The

half-maximal inhibitory concentration of SIK2 (IC₅₀) was

reached at a concentration of 0.3 nM of MRIA9 (10). G-5555

(6) did not show the same potency in this SIK2 auto-

phosphorylation assay, and the highest inhibitor concentration

tested (5 nM) (6) did only inhibit 57% of the activity of SIK2

(IC₅₀= 4.4 nM) (Figure S10). These data conﬁrmed the high

potency of MRIA9 (10) in inhibiting SIK2 auto-phosphor-

ylation activity.

MRIA9 Sensitizes Ovarian Cancer Cells to Paclitaxel

Treatment. After the characterization of MRIA9 as a potent

SIK2 inhibitor with the lowest oﬀ-target proﬁle characterized

so far, we further validated the compound as a potential

chemical tool for mechanistic and phenotypic studies. One

intriguing observation for the translational research of SIK

inhibitors was that SIK2 depletion by RNAi sensitizes ovarian

cancer cells to paclitaxel treatment.^13,14Based on these data,

we next sought to investigate the therapeutic potential of

MRIA9 (10) on the ovarian cancer cell survival and Taxol

sensitivity. To this end, the growth of the human ovarian

cancer cell line SKOV3 was monitored in a long-term

experiment (9 days) upon treatment with increasing

concentrations of MRIA9 (0.5 to 5 μM) in the presence and

absence of paclitaxel. Low concentrations of MRIA9 (0.5 and 1

μM) showed no signiﬁcant eﬀect on cell viability, and only a

high concentration (5 μM) reduced the viability somewhat

compared with control cells (Figure 6A). Remarkably, a strong

reduction in cell viability was observed when paclitaxel was

combined with MRIA9 (Figure 6A). These results are in

accordance with the proﬁle of MRIA9 on the NCI-60 panel,

which did not show a strong cytotoxic eﬀect alone (Figure

S11). To investigate the apoptotic response of cells that is

induced by the single treatment with MRIA9 or by the

combinatorial treatment with paclitaxel and MRIA9, we

monitored caspase 3/7 activation (Figure 6B). A moderate

caspase-3/7 activation was observed in cells treated with

paclitaxel alone. As expected, based on the limited eﬀect of

MRIA9 on cell survival, only the highest concentration used (5

μM) induced caspase activation. Caspase 3/7 activation was

signiﬁcantly enhanced, however, in combination treatment

with paclitaxel and MRIA9 (Figure 6B).

The three-dimensional (3D)-spheroid system is widely used

in studies involving the eﬃcacy assessment of new small-

molecule inhibitors in vitro. Therefore, we decided to

investigate the eﬀect of MRIA9-dependent inhibition of SIK2

on a 3D-spheroid model of HeLa cells. We evaluated the

viability of HeLa cell-derived spheroids by using the live/dead

cell viability ﬂuorescence-based assay upon a 48 h-treatment

with paclitaxel (2 nM) or MRIA9 (10) (5 μM) or the

combination 2 nM paclitaxel/5 μM MRIA9 (Figure 7). The

inhibition of SIK2 using MRIA9 as a single treatment increased

the fraction of dead cells compared with the depletion of SIK2

or with paclitaxel as a single agent (Figure 7). Remarkably, the

combination paclitaxel/MRIA9 induced a robust increase of

Figure 6. Incubation with MRIA9 sensitizes SKOV3 cells to paclitaxel

treatment through inducing pronounced apoptosis. (A) SKOV3 cells

were either treated with 1 nM paclitaxel, or increasing concentrations

of MRIA9, or their combinations for 9 days. Cell viability was

determined using the CellTiter-Blue Cell Viability Assay. Measure-

ments were statistically signiﬁcant as determined by Student’s t-test

(*p ≤ 0.05; **p ≤ 0.01). (B) Caspase-3/7-activity in whole lysates of

SKOV3 cells treated with 1 nM paclitaxel, increasing concentrations

of MRIA9, or their combinations was measured 48 h post-treatment

using the Caspase-Glo 3/7 Assay. Each bar graph represents the mean

value SD (*p ≤ 0.05; ***p ≤ 0.001).

the fraction of dead HeLa cells compared with paclitaxel alone

or even when compared with the combination paclitaxel and

SIK2 siRNA after 48 h of treatment [p ≤ 0.001 for the

combination paclitaxel (2 nM) and MRIA9 (5 μM)]. These

results indicate that the MRIA9-dependent inhibition of SIK2

enhanced the cytotoxicity of paclitaxel and argue for this

combination therapy as already demonstrated in studies using

less selective inhibitors with SIK activity and in genetic

knockdown studies.¹⁴

Metabolic Stability of MRIA9 and Pharmacokinetic

Proﬁle Prediction. As a ﬁrst evaluation of the pharmacoki-

netic properties of MRIA9 for further use in vivo, we analyzed

the metabolic stability against an activated microsome mix

derived from the liver of Sprague-Dawley rats as previously

reported.²⁹MRIA9 (10) was incubated at 37 °C for 15, 30,

and 60 min, and the amount of unmetabolized compound was

determined by high-performance liquid chromatography

(HPLC) at these time points, including time zero. MRIA9

showed excellent metabolic stability after 1 h incubation, with

more than 90% of the compound detected (Figure S12) and is

therefore a good candidate for more detailed PK studies in

vivo. Figure S12 also shows the results from diﬀerent

pharmacokinetic predictors for MRIA9 and the parent

compound G-5555. According to these predictions, both G-

5555 and MRIA9 are expected to show good cell permeability

in Caco-2 and MDCK cells as well as high oral bioavailability

8151

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 7. Combination of the SIK inhibitor MRIA9 with paclitaxel signiﬁcantly enhances cell death in HeLa cells. (A) Spheroid cultures grown

from HeLa cells were SIK2-depleted, treated with paclitaxel (2 nM), treated with the SIK inhibitor MRIA9 (5 μM), or treated with the

combination paclitaxel/5 μM MRIA9 for 48 h. The spheroids were stained for immunoﬂuorescence analysis with the live/dead viability/

cytotoxicity kit. (B) The relative fraction of dead cells was quantiﬁed by immunoﬂuorescence analysis and presented as a bar graph. The results are

presented as mean SD and statistically analyzed. ***p < 0.001, **p < 0.01.

reagent grade and used without further puriﬁcation. Reactions were

(F = 75−80%). The pharmacokinetic behavior of G-5555 was

experimentally assessed by Ndubaku et al.,²⁰revealing high

oral bioavailability in mice (F = 80%) and monkeys (F = 72%),

which is in agreement with our predictions.

proceeded under an argon atmosphere. For compound puriﬁcation by

ﬂash chromatography, a puriFlash XS 420 device with a UV-VIS

multiwave detector (200−400 nm) from Interchim with pre-packed

normal-phase PF-SIHP silica columns with particle sizes of 30 μm

(Interchim) was used. All compounds synthesized were characterized

CONCLUSIONS

1

by H and ¹³C NMR and mass spectrometry (ESI +/−). In addition,

■

The lack of available crystal structures of SIK proteins has

impeded the development of potent and selective SIK

inhibitors. We have therefore utilized an oﬀ-target approach

to develop a new series of pyrido[2,3-d]pyrimidin-7-one

inhibitors with SIK kinase activity. Starting from compound

6 (G-5555), we analyzed the binding mode of this inhibitor in

oﬀ-targets, combining crystallographic data on MST3/4 and

MD simulations, to rationally design a series of derivatives that

exploit diﬀerences in the back-pocket region of SIK family

proteins. From this approach, compound 10 (MRIA9) was the

most promising chemical tool candidate, with both cellular

potency and optimized selectivity. PAK2 was the only

remaining oﬀ-target inhibited with the same potency in vitro.

Further assessment of the cellular potency of MRIA9 toward

the PAK subfamily will help delineate the usage of this new

chemical tool in diﬀerent phenotypic assays. Importantly, we

have shown the potency of MRIA9 in sensitizing ovarian

cancer cells to paclitaxel, suggesting that a combination therapy

with paclitaxel and SIK inhibitors may result in improved

therapeutic eﬀects.

ﬁnal compounds were identiﬁed by high-resolution mass spectra

(HRMS) and their purity was evaluated by HPLC. ¹H and ¹³C NMR

spectra were measured on a DPX250, an AV400, or an AV500 HD

AVANCE III spectrometer from Bruker Corporation. Chemical shifts

(δ) are reported in parts per million (ppm). DMSO-d₆was used as a

solvent, and the spectra were calibrated to the solvent signal: 2.50

ppm (¹H NMR) or 39.52 ppm (¹³C NMR). Mass spectra (ESI +/−)

were measured on a Surveyor MSQ device from ThermoFisher.

HRMS were obtained on a MALDI LTQ Orbitrap XL from

ThermoScientiﬁc. Preparative puriﬁcation by HPLC was carried out

on an Agilent 1260 Inﬁnity II device using an Eclipse XDB-C18

(Agilent, 21.2 × 250 mm, 7 μm) reverse-phase column. A suitable

gradient (ﬂow rate 21 ml/min) was used, with 0.1% triﬂuoroacetic

acid (TFA) in water (A) and 0.1% TFA in acetonitrile (B) as a mobile

phase. The determination of the compound purity by HPLC was

carried out on the same Agilent 1260 Inﬁnity II device, together with

HPLC-MS measurements using an LC/MSD device (G6125B, ESI

pos. 100−1000). The compounds were analyzed on an Eclipse XDB-

C18 (Agilent, 4.6 × 250 mm, 5 μm) reverse-phase column, using

0.1% TFA in water (A) and 0.1% TFA in acetonitrile (B) as a mobile

phase. The following gradient was used: 0 min 2% B2 min 2% B

10 min 98% B15 min 98% B17 min 2% B19 min 2% B (ﬂow

rate of 1 mL/min). UV detection was performed at 254 and 280 nm.

All the ﬁnal compounds showed >95% purity.

EXPERIMENTAL SECTION

■

General Procedure I for Suzuki Cross-Coupling Reaction. 6-

(2-Chloro-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)-2-

(methylthio)pyrido[2,3-d]pyrimidin-7-ol (22) (1.0 equiv), potassium

acetate (2.0 equiv), [1,1′-Bis(diphenylphosphino)ferrocene]

Chemistry. The synthesis of compounds 7−18 will be explained

in the following. Further information regarding the analytical data for

compounds 7−37 can be found in the Supporting Information. All

commercial chemicals were purchased from common suppliers in

8152

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

dichloropalladium(II) (0.1 equiv) and the corresponding pyridine

derivative (1.0 or 1.1 equiv) were dissolved in a mixture of dioxane/

water (2:1; 12 mL) and stirred at 107 °C for 18 h. The solvent was

evaporated under reduced pressure and the remaining residue was

puriﬁed by column chromatography on silica gel using cyclohexane/

ethyl acetate as an eluent.

6-(4-Bromo-2-chlorophenyl)-2-(methylthio)pyrido[2,3-d]-

pyrimidin-7-ol (21). 4-Amino-2-(methylthio)pyrimidine-5-carbalde-

hyde (1.70 g, 10.1 mmol), 20 (3.07 g, 11.1 mmol), and potassium

carbonate (4.17 g, 30.1 mmol) were dissolved in anhydrous DMF (70

mL) and stirred at 120 °C for 16 h. Water (300 mL) was added to the

reaction solution, leading to the precipitation of 21 as a pale-yellow

solid. After ﬁltration, the aqueous layer was extracted with ethyl

acetate (3 × 30 mL), and the combined organic layers were dried over

MgSO₄. The solvent was evaporated under reduced pressure, and the

remaining residue was recrystallized from acetone. A combined yield

General Procedure II for Nucleophilic Substitution. The

corresponding 6-[2-chloro-4-(3-R1-4-R2-6-R3-pyridin-2-yl)phenyl]2-

(methylthio)pyrido[2,3-d]pyrimidin-7-ol (1.0 equiv), obtained from

general procedure I, 2-((2r,5r)-2-(bromomethyl)-1,3-dioxan-5-yl)-

isoindoline-1,3-dione (1.5 equiv), and cesium carbonate (3.0 equiv)

were dissolved in anhydrous N,N-dimethylformamide (DMF) (10

mL) and stirred at 100 °C for 18 h. The solvent was evaporated under

reduced pressure. To remove the unreacted starting material, a

column chromatography on silica gel was performed using DCM/

MeOH as an eluent. A mixture of the corresponding 2-((2r,5r)-2-((6-

(2-chloro-4-(3-R1-4-R2-6-R3-pyridin-2-yl)phenyl)-2-(methylthio)-7-

oxopyrido[2,3-d]pyrimidin-8(7H)-yl)methyl)-1,3-dioxan-5-yl)-

isoindoline-1,3-dione and 2-(((2r,5r)-2-((6-(2-chloro-4-(3-R1-4-R2-

6-R3-pyridin-2-yl)phenyl)-2-(methylthio)-7-oxopyrido[2,3-d]-

pyrimidin-8(7H)-yl)methyl)-1,3-dioxan-5-yl)-carbamoyl)benzoic acid

were obtained and used in the next step without further separation.

General Procedure III for Oxidation. The corresponding

mixture of (2-((2r,5r)-2-(6-(2-chloro-4-(3-R1-4-R2-6-R3-pyridin-2-

yl)phenyl)-2-(methylthio)-7-oxopyrido[2,3-d]pyrimidin-8(7H)-yl)-

methyl)-1,3-dioxan-5-yl)isoindoline-1,3-dione and 2-(((2r,5r)-2-((6-

(2-chloro-4-(3-R1-4-R2-6-R3-pyridin-2-yl)phenyl)-2-(methylthio)-7-

oxopyrido[2,3-d]pyrimidin-8(7H)-yl)methyl)-1,3-dioxan-5-yl)-

carbamoyl)benzoic acid, obtained from general procedure II (1.0

equiv), was dissolved in anhydrous dichloromethane (DCM) (10

mL). M-chloroperoxybenzoic acid (m-CPBA) (≤77%, 2.1 equiv) was

added in one portion, and the reaction solution was stirred at room

temperature for 3 h. The remaining m-CPBA was quenched by adding

saturated NaHCO₃solution (10 mL). The aqueous layer was

extracted with a mixture of DCM/MeOH (4:1; 3 × 15 mL) and

DCM (15 mL). The combined organic layers were dried over

Na₂SO₄, and the solvent was evaporated under a reduced pressure.

The remaining residue, consisting of the corresponding sulfoxide and

sulfone products, was used in the next step without further

puriﬁcation.

General Procedure IV for Nucleophilic Aromatic Substitu-

tion. The remaining residue of the corresponding product of general

procedure III (1.0 equiv) was diluted in ethanol (10 mL), and a

solution of methylamine in tetrahydrofuran (2 M, 4.0 equiv) and N,N-

diisopropylethylamine (DIPEA) (5.0 equiv) were added. The reaction

solution was stirred at 85 °C for 17 h. After the reaction was ﬁnished,

the solvent was evaporated under reduced pressure. The remaining

residue was puriﬁed by HPLC chromatography on silica gel (H₂O/

ACN; 0.1% TFA) to obtain the corresponding 8-(((2r,5r)-5-amino-

1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-R1-4-R2-6-R3-pyridin-2-

yl)phenyl)-2-(methylamino)pyrido[2,3-d]pyrimidin-7(8H)-one prod-

uct as a TFA salt.

Ethyl 2-(4-bromo-2-chloro-phenyl)acetate (20). 2-(4-Bromo-2-

chloro-phenyl)-acetic acid (19; 4.00 g, 16.0 mmol) was dissolved in

ethanol (80 mL) and cooled to 0 °C. Sulfuric acid (1.57 g, 16.0

mmol) was added, and the reaction solution was allowed to warm to

room temperature. The reaction was stirred at 100 °C for 15 h.

Afterward, the solvent was evaporated under reduced pressure, and

the remaining residue was dissolved in ethyl acetate (30 mL). The

organic layer was washed with saturated NaHCO₃solution (2 × 20

mL) and brine (20 mL). The aqueous layer was extracted with ethyl

acetate (20 mL), and the combined organic layers were dried over

MgSO₄. The solvent was evaporated under reduced pressure to aﬀord

20 as yellow liquid with a yield of 93% (4.15 g). ¹H NMR (500 MHz,

DMSO-d₆, 300 K): δ 7.73 (d, J = 2.0 Hz, 1H), 7.54 (dd, J = 8.2, 2.0

Hz, 1H), 7.38 (d, J = 8.2 Hz, 1H), 4.09 (q, J = 7.1 Hz, 2H), 3.79 (s,

2H), 1.18 (t, J = 7.1 Hz, 3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆,

300 K): δ 169.6, 134.9, 133.7, 132.4, 131.2, 130.3, 120.7, 60.5, 38.0,

14.0 ppm. MS-ESI⁺(m/z) [fragment + H]⁺; calcd, 205.0; found,

205.0.

1

of 93% (3.57 g) was achieved. H NMR (500 MHz, DMSO-d₆, 300

K): δ 12.69 (s, 1H), 8.88 (s, 1H), 7.95 (s, 1H), 7.83 (d, J = 1.9 Hz,

1H), 7.64 (dd, J = 8.2, 2.0 Hz, 1H), 7.38 (d, J = 8.2 Hz, 1H), 2.58 (s,

3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.2, 161.5,

157.0, 154.5, 136.6, 134.3, 134.1, 133.3, 131.5, 130.7, 130.2, 121.9,

108.8, 13.7 ppm.

6-(2-Chloro-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-

phenyl)-2-(methylthio)-pyrido[2,3-d]pyrimidin-7-ol (22). 21 (200

mg, 0.52 mmol), bis(pinacolato)diboron (173 mg, 0.68 mmol),

potassium acetate (154 mg, 1.57 mmol), and [1,1′-bis-

(diphenylphosphino) ferrocene]dichloropalladium(II) in complex

with DCM (30 mg, 0.04 mmol) were dissolved in anhydrous dioxane

(10 mL) and stirred at 115 °C for 18 h. The reaction solution was

ﬁltered over a pad of Celite. Ethyl acetate (20 mL) was added, and the

organic layer was washed with brine (20 mL). The aqueous layer was

extracted with ethyl acetate (2 × 20 mL), and the combined organic

layers were dried over Na₂SO₄. The solvent was evaporated under

reduced pressure, and the remaining residue was puriﬁed by ﬂash

chromatography on silica gel (cyclohexane/ethyl acetate 8:1to 2:1).

Recrystallization from DCM/cyclohexane was performed to obtain 21

1

as a pale-yellow solid with a yield of 89% (200 mg). H NMR (500

MHz, DMSO-d₆, 300 K): δ 12.67 (s, 1H), 8.91 (s, 1H), 7.96 (s, 1H),

7.71 (s, 1H), 7.67 (dd, J = 7.5, 1.0 Hz, 1H), 7.44 (d, J = 7.5 Hz, 1H),

2.59 (s, 3H), 1.32 (s, 12H) ppm. ¹³C NMR (126 MHz, DMSO-d₆,

300 K): δ 172.2, 161.2, 157.0, 154.2, 137.7, 136.5, 134.5, 132.8, 132.7,

131.6, 131.5, 130.7, 108.8, 84.2, 24.6, 13.7 ppm. MS-ESI⁺(m/z) [M +

H]⁺: calcd, 430.1: found, 430.6.

2-(1,3-Dihydroxypropan-2-yl)isoindoline-1,3-dione (24). Serinol

(23) (500 mg, 5.49 mmol) and phthalic anhydride (800 mg, 5.49

mmol) were dissolved in anhydrous DMF (5 mL) and stirred at 90

°C for 4 h. The solvent was evaporated under reduced pressure,

leading to the formation of a yellow solid. The solid was recrystallized

from 2-propanol/cyclohexane to obtain 24 as a white solid with a

1

yield of 87% (1.05 g). H NMR (500 MHz, DMSO-d₆, 300 K): δ

7.86−7.82 (m, 4H), 4.87 (t, J = 6.0 Hz, 2H), 4.27−4.21 (m, 1H),

3.83−3.77 (m, 2H), 3.68−3.64 (m, 2H) ppm. ¹³C NMR (126 MHz,

DMSO-d₆, 300 K): δ = 168.4, 134.2, 131.7, 122.8, 58.3, 56.5 ppm.

MS-ESI⁺(m/z) [M + Na]⁺: calcd, 244.1; found, 244.1.

2-((2r,5r)-2-(Bromomethyl)-1,3-dioxan-5-yl)isoindoline-1,3-

dione (25). 24 (2.50 g, 11.3 mmol), 2-bromo-1,1′-diethoxy-ethane

(1.20 g, 13.6 mmol), and p-toluenesulfonic acid monohydrate (0.43 g,

2.3 mmol) were dissolved in 60 mL of toluene and stirred at 115 °C

for 24 h. The organic layer was washed with a saturated NaHCO₃

solution (2 × 50 mL) and brine (50 mL). The aqueous layer was

extracted with ethyl acetate (3 × 20 mL), and the combined organic

layers were dried over Na₂SO₄. The solvent was evaporated under

reduced pressure, and the remaining residue was puriﬁed by ﬂash

chromatography on silica gel (cyclohexane/ethyl acetate 6:1 to 2:1)

1

to obtain 25 as a white solid with a yield of 73% (2.70 g). H NMR

(400 MHz, DMSO-d₆, 300 K): δ 7.88−7.84 (m, 4H), 4.84 (t, J = 4.0

Hz, 1H), 4.31−4.30 (m, 3H), 4.11−4.09 (m, 2H), 3.51 (d, J = 4.1 Hz,

2H) ppm. ¹³C NMR (100 MHz, DMSO-d₆, 300 K): δ 167.5, 134.5,

131.4, 123.1, 98.5, 65.8, 43.6, 32.2 ppm. MS-ESI⁺(m/z) [fragment +

K]⁺: calcd, 382.0; found, 382.3.

6-(2-Chloro-4-(6-methylpyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (26). Compound 26 was prepared

according to general procedure I, using 2-bromo-6-methyl-pyridine

(1.0 equiv). 26 was obtained as a white solid with a yield of 80% (88

1

mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.69 (s, 1H), 8.92

(s, 1H), 8.24 (d, J = 1.7 Hz, 1H), 8.11 (dd, J = 8.0, 1.7 Hz, 1H), 8.02

8153

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(s, 1H), 7.88 (d, J = 7.8 Hz, 1H), 7.82 (t, J = 7.7 Hz, 1H), 7.53 (d, J =

8.0 Hz, 1H), 7.29 (d, J = 7.6 Hz, 1H), 2.59 (s, 3H), 2.57 (s, 3H) ppm.

¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ = 172.2, 161.4, 158.1,

157.0, 154.2, 153.3, 140.4, 137.7, 136.6, 135.0, 133.4, 132.1, 131.4,

126.9, 124.9, 122.8, 117.7, 108.8, 24.3, 13.7 ppm. MS-ESI⁺(m/z) [M

+ H]⁺: calcd, 395.1; found, 395.0.

J = 8.0, 1.5 Hz, 1H), 7.67 (d, J = 8.3 Hz, 1H), 7.50 (d, J = 8.0 Hz,

1H), 7.48−7.45 (m, 1H), 3.92 (s, 3H), 2.59 (s, 3H) ppm. ¹³C NMR

(126 MHz, DMSO-d₆, 300 K): δ 172.1, 161.4, 156.9, 154.2, 153.8,

144.0, 140.8, 138.8, 136.5, 134.5, 132.4, 131.5, 131.3, 129.5, 127.7,

124.5, 120.3, 108.8, 55.9, 13.7 ppm. MS-ESI⁺(m/z) [M + H]⁺: calcd,

411.1; found, 411.0.

6-(2-Chloro-4-(3-methoxy-6-methylpyridin-2-yl)phenyl)-2-

(methylthio)pyrido[2,3-d]-pyrimidin-7-ol (33). Compound 33 was

prepared according to general procedure I, using 2-bromo-3-methoxy-

6-methyl-pyridine (1.1 equiv). 33 was obtained as a white solid with a

6-(2-Chloro-4-(pyridin-2-yl)phenyl)-2-(methylthio)pyrido[2,3-d]-

pyrimidin-7-ol (27). Compound 27 was prepared according to

general procedure I, using 2-bromo-pyridine (1.0 equiv). 27 was

1

obtained as a white solid with a yield of 20% (117 mg). H NMR

1

yield of 78% (230 mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ

(500 MHz, DMSO-d₆, 300 K): δ 12.69 (s, 1H), 8.92 (s, 1H), 8.71

(dd, J = 4.7, 0.8 Hz, 1H), 8.26 (d, J = 1.7 Hz, 1H), 8.13 (dd, J = 8.0,

1.7 Hz, 1H), 8.09 (d, J = 8.0 Hz, 1H), 8.02 (s, 1H), 7.94 (dt, J = 7.7,

1.8 Hz, 1H), 7.55 (d, J = 8.0 Hz, 1H), 7.44−7.41 (m, 1H), 2.59 (s,

3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.2, 161.4,

157.0, 154.2, 154.0, 149.7, 140.3, 137.5, 136.6, 135.2, 133.5, 132.2,

131.4, 127.0, 124.9, 123.4, 120.7, 108.8, 13.7 ppm. MS-ESI⁺(m/z)

[M + H]⁺: calcd, 381.1; found, 381.0.

12.67 (s, 1H), 8.92 (s, 1H), 8.01 (d, J = 1.6 Hz, 1H), 8.00 (s, 1H),

7.93 (dd, J = 8.0, 1.7 Hz, 1H), 7.53 (d, J = 8.5 Hz, 1H), 7.47 (d, J =

8.0 Hz, 1H), 7.27 (d, J = 8.5 Hz, 1H), 3.87 (s, 3H), 2.59 (s, 3H), 2.48

(s, 3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.1,

161.4, 156.9, 154.2, 151.7, 149.1, 143.0, 139.3, 136.4, 134.2, 132.3,

131.6, 131.2, 129.3, 127.6, 123.5, 120.7, 108.8, 55.8, 23.1, 13.7 ppm.

MS-ESI⁺(m/z) [M + H]⁺: calcd, 425.1; found, 425.0.

6-(2-Chloro-4-(6-(triﬂuoromethyl)pyridin-2-yl)phenyl)-2-

(methylthio)pyrido[2,3-d]-pyrimidin-7-ol (34). Compound 34 was

prepared according to general procedure I, using 2-bromo-6-

triﬂuoromethyl-pyridine (1.1 equiv). 34 was obtained as a yellow

6-(2-Chloro-4-(3-ﬂuoro-6-methylpyridin-2-yl)phenyl)-2-(methyl-

thio)-pyrido[2,3-d]-pyrimidin-7-ol (28). Compound 28 was prepared

according to general procedure I, using 2-bromo-3-ﬂuoro-6-methyl-

pyridine (1.0 equiv). 28 was obtained as a white solid with a yield of

1

60% (173 mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.70 (s,

solid with a yield of 96% (300 mg). H NMR (500 MHz, DMSO-d₆,

1H), 8.92 (s, 1H), 8.04 (s, 2H), 7.95 (s, 1H), 7.77 (s, 1H), 7.57 (s,

1H), 7.39 (s, 1H), 2.59 (s, 3H), 2.55 (s, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 172.2, 161.3, 159.4, 157.0, 154.8, 154.3,

141.4, 136.6, 135.4, 133.1, 131.9, 131.3, 128.8, 127.0, 125.4, 125.2,

124.7, 108.8, 23.5, 13.7 ppm. MS-ESI⁺(m/z) [M + H]⁺: calcd, 413.1;

found, 413.0.

300 K): δ 12.70 (s, 1H), 8.93 (s, 1H), 8.43 (d, J = 8.0 Hz, 1H), 8.28

(d, J = 1.7 Hz, 1H), 8.24 (t, J = 7.9 Hz, 1H), 8.18 (dd, J = 8.0, 1.7 Hz,

1H), 8.03 (s, 1H), 7.93 (d, J = 7.7 Hz, 1H), 7.61 (d, J = 8.0 Hz, 1H),

2.59 (s, 3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.3,

161.3, 157.0, 154.8, 154.3, 146.9, 146.7, 146.4, 146.1, 134.0, 138.6,

136.7, 136.2, 133.7, 132.4, 131.2, 127.3, 125.4, 124.8, 124.3, 122.6,

120.5, 120.0, 118.3, 108.8, 13.7 ppm. MS-ESI⁻(m/z) [M − H]⁻;

calcd, 447.0; found, 447.0.

6-(2-Chloro-4-(3-ﬂuoropyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (29). Compound 29 was prepared

according to general procedure I, using 2-bromo-3-ﬂuoro-pyridine

(1.0 equiv). 29 was obtained as a white solid with a yield of 94% (260

6-(2-Chloro-4-(6-(diﬂuoromethyl)pyridin-2-yl)phenyl)-2-

(methylthio)pyrido[2,3-d]pyrimidin-7-ol (35). Compound 35 was

prepared according to general procedure I, using 2-bromo-6-

diﬂuoromethyl-pyridine (1.1 equiv). 35 was obtained as a pale yellow

1

mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.70 (s, 1H), 8.92

(s, 1H), 8.60 (d, J = 1.7 Hz, 1H), 8.06 (s, 1H), 8.04 (s, 1H), 7.96 (d, J

= 8.0 Hz, 1H), 7.93−7.87 (m, 1H), 7.60−7.53 (m, 2H), 2.59 (s, 3H)

ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.7, 161.8, 158.8,

157.5, 156.7, 154.7, 146.5, 143.3, 137.1, 136.9, 136.0, 133.6, 132.4,

131.7, 129.3, 127.5, 125.8, 125.6, 125.4, 109.3, 14.2 ppm. MS-ESI⁻

(m/z) [M − H]⁻: calcd: 397.0; found, 397.2.

1

solid with a yield of 87% (260 mg). H NMR (500 MHz, DMSO-d₆,

300 K): δ 12.70 (s, 1H), 8.93 (s, 1H), 8.29−8.28 (m, 2H), 8.18−8.13

(m, 2H), 8.03 (s, 1H), 7.73 (d, J = 7.7 Hz, 1H), 7.59 (d, J = 8.0 Hz,

1H), 7.07 (t, J = 54.9 Hz, 1H), 2.59 (s, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 172.2, 161.3, 157.0, 154.2, 154.2, 152.3,

152.1, 151.9, 139.4, 139.2, 136.6, 135.8, 133.7, 132.3, 131.3, 127.2,

125.2, 122.7, 119.8, 115.7, 113.8, 111.9, 108.8, 13.7 ppm. MS-ESI⁺

(m/z) [M + K]⁺: calcd, 469.0; found, 468.8.

6-(2-Chloro-4-(6-methoxypyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (36). Compound 36 was prepared

according to general procedure I, using 2-bromo-6-methoxy-pyridine

(1.0 equiv). (36) was obtained as a white solid with a yield of 98%

(295 mg). ¹H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.67 (s, 1H),

8.92 (s, 1H), 8.24 (d, J = 1.7 Hz, 1H), 8.12 (dd, J = 8.0, 1.7 Hz, 1H),

8.01 (s, 1H), 7.85−7.82 (m, 1H), 7.68 (d, J = 7.4 Hz, 1H), 7.54 (d, J

= 8.0 Hz, 1H), 6.85 (d, J = 8.2 Hz, 1H), 3.98 (s, 3H), 2.59 (s, 3H)

ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.2, 163.3, 161.3,

156.9, 154.2, 151.7, 140.2, 140.0, 136.5, 135.1, 133.4, 132.1, 131.4,

126.8, 124.9, 113.5, 110.3, 108.8, 53.0, 13.7 ppm. MS-ESI⁺(m/z) [M

+ Na]⁺: calcd, 433.1; found, 433.1.

6-(2-Chloro-4-(3-chloropyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (30). Compound 30 was prepared

according to general procedure I, using 2-bromo-3-chloro-pyridine

(1.0 equiv). 30 was obtained as a white solid with a yield of 69% (200

1

mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.71 (s, 1H), 8.91

(s, 1H), 8.68 (dd, J = 4.7, 1.1 Hz, 1H), 8.11 (d, J = 8.1 Hz, 1H), 8.05

(s, 1H), 7.84 (d, J = 1.3 Hz, 1H), 7.75 (dd, J = 7.9, 1.4 Hz, 1H), 7.55

(d, J = 7.9 Hz, 1H), 7.51 (dd, J = 8.1, 4.6 Hz, 1H), 2.59 (s, 3H) ppm.

¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.2, 161.3, 157.0, 154.2,

153.6, 148.2, 139.4, 138.6, 136.7, 135.2, 132.6, 131.4, 131.3, 129.8,

129.3, 127.9, 124.6, 108.8, 13.7 ppm. MS-ESI⁺(m/z) [M + Na]⁺:

calcd, 437.0; found, 437.0.

6-(2-Chloro-4-(3-methylpyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (31). Compound 31 was prepared

according to general procedure I, using 2-bromo-3-methyl-pyridine

(1.0 equiv). 31 was obtained as a white solid with a yield of 73% (200

6-(2-Chloro-4-(4,6-dimethylpyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (37). Compound 37 was prepared

according to general procedure I, using 2-bromo-4,6-dimethyl-

pyridine (1.0 equiv). (37) was obtained as a white solid with a

1

mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.68 (s, 1H), 8.92

(s, 1H), 8.53 (dd, J = 4.7, 1.1 Hz, 1H), 8.04 (s, 1H), 7.77 (dd, J = 7.7,

0.8 Hz, 1H), 7.72 (d, J = 1.7 Hz, 1H), 7.62 (dd, J = 7.9, 4.7 Hz, 1H),

7.52 (d, J = 7.9 Hz, 1H), 7.35 (dd, J = 7.7, 4.7 Hz, 1H), 2.60 (s, 3H),

2.40 (s, 3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 172.1,

161.4, 156.9, 155.7, 154.2, 147.0, 141.9, 138.9, 136.6, 134.2, 132.6,

131.4, 131.3, 130.8, 129.6, 127.6, 122.9, 108.8, 19.6, 13.7 ppm. MS-

ESI⁺(m/z) [M + H]⁺: calcd, 395.1; found, 395.1.

1

yield of 74% (210 mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ

12.66 (s, 1H), 8.91 (s, 1H), 8.23 (d, J = 1.4 Hz, 1H), 8.10 (dd, J = 8.0,

1.5 Hz, 1H), 8.01 (s, 1H), 7.74 (s, 1H), 7.52 (d, J = 8.0 Hz, 1H), 7.12

(s, 1H), 2.59 (s, 3H), 2.52 (s, 3H), 2.37 (s, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 172.1, 161.4, 157.7, 156.9, 154.3, 153.2,

148.3, 140.5, 136.5, 134.9, 133.3, 132.0, 131.4, 126.9, 124.9, 123.5,

118.6, 108.8, 24.1, 20.5, 13.7 ppm. MS-ESI⁺(m/z) [M + H]⁺: calcd,

409.1; found, 409.1.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(6-

methylpyridin-2-yl)phenyl)-2-((2-methoxyphenyl)amino)pyrido-

[2,3-d]pyrimidin-7(8H)-one (7). Compound 7 was prepared accord-

6-(2-Chloro-4-(3-methoxypyridin-2-yl)phenyl)-2-(methylthio)-

pyrido[2,3-d]pyrimidin-7-ol (32). Compound 32 was prepared

according to general procedure I, using 2-bromo-3-methoxy-pyridine

(1.0 equiv). 32 was obtained as a white solid with a yield of 70% (200

1

mg). H NMR (500 MHz, DMSO-d₆, 300 K): δ 12.68 (s, 1H), 8.92

(s, 1H), 8.31 (dd, J = 4.6, 0.9 Hz, 1H), 8.06−8.00 (m, 2H), 7.94 (dd,

8154

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

ing to general procedures II and III. The remaining residue of general

procedure III containing the corresponding sulfoxide/sulfone product

(1.0 equiv) and o-anisidine (4.0 equiv) was dissolved in ethanol (5

mL). HCl (1 M, 0.1 mL) was added, and the reaction solution was

stirred at 100 °C for 17 h. Water (5 mL) was added to the reaction

solution, and the aqueous layer was extracted with a mixture of

DCM/MeOH (4:1, 3 × 10 mL) and DCM (10 mL). The combined

organic layers were dried over Na₂SO₄, and the solvent was

evaporated under reduced pressure. The remaining residue was

puriﬁed by HPLC chromatography on silica gel (H₂O/ACN; 0.1%

TFA) to obtain 7 as an orange solid with a yield of 13% (20 mg) over

three steps. ¹H NMR (500 MHz, DMSO-d₆, 300 K): δ 8.84 (d, J = 5.3

Hz, 1H), 8.24 (d, J = 1.7 Hz, 1H), 8.10 (dd, J = 8.0, 1.7 Hz, 2H), 8.02

(s, 1H), 7.99 (s, 2H), 7.97 (s, 1H), 7.89 (d, J = 7.8 Hz, 1H), 7.83 (t, J

= 7.7 Hz, 1H), 7.53 (d, J = 8.0 Hz, 1H), 7.30 (d, J = 7.5 Hz, 1H),

7.16−7.11 (m, 2H), 7.01 (t, J = 7.5 Hz, 1H), 5.05 (t, J = 5.3 Hz, 1H),

4.49 (d, J = 5.1 Hz, 2H), 4.19 (dd, J = 11.0, 4.4 Hz, 2H), 3.88 (s, 3H),

3.54 (t, J = 10.7 Hz, 2H), 3.38 (s, 1H), 2.58 (s, 3H) ppm. ¹³C NMR

(126 MHz, DMSO-d₆, 300 K): δ 160.9, 159.5, 159.0, 158.0, 155.1,

153.2, 150.5, 140.0, 137.9, 136.6, 135.6, 133.6, 133.0, 132.3, 127.5,

127.0, 126.0, 124.9, 124.5, 122.8, 120.2, 117.8, 111.3, 105.8, 97.3,

66.3, 55.8, 42.5, 42.4, 24.2 ppm. HRMS (FTMS + p MALDI): m/z

calculated for C₃₁H₃₀ClN₆O₄[M + H]⁺: 585.2012; found, 585.2019.

HPLC: t_R= 11.49 min; purity ≥ 95% (UV: 254/280 nm). LCMS-ESI

+ (m/z) [M + H]⁺: calcd, 585.2; found, 585.0.

42.1, 27.9 ppm. HRMS (FTMS + p MALDI) m/z: calcd for

C₂₄H₂₃ClFN₆O₃[M + H]⁺, 497.1499; found, 497.1491. HPLC: t_R=

11.71 min; purity ≥ 95% (UV: 254/280 nm). LCMS-ESI⁺(m/z) [M

+ H]⁺: calcd, 497.1; found, 497.0.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-

chloropyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]-

pyrimidin-7(8H)-one (11). Compound 11 was prepared according to

general procedures II-IV and obtained as a yellow solid with a yield of

1

15% (30 mg) over three steps. H NMR (500 MHz, DMSO-d₆, 300

K): δ 8.73−8.65 (m, 2H), 8.11−8.08 (m, 4H), 7.97 (s, 1H), 7.90 (s,

1H), 7.82 (d, J = 1.7 Hz, 1H), 7.73 (dd, J = 7.9, 1.7 Hz, 1H), 7.53−

7.49 (m, 2H), 5.14−5.06 (m, 1H), 4.56−4.50 (m, 2H), 4.22−4.20

(m, 2H), 3.60−3.56 (m, 2H), 3.39 (s, 1H), 2.92 (d, J = 4.5 Hz, 3H)

ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 166.5, 162.3, 159.9,

155.8, 154.1, 148.6, 139.1, 137.5, 133.8, 133.3, 132.1, 131.1, 130.3,

129.8, 129.3, 128.4, 125.1, 104.4, 97.8, 66.7, 42.8, 42.5, 28.4 ppm.

HRMS (FTMS + p MALDI) m/z: calcd for C₂₄H₂₃Cl₂N₆O₃[M +

H]⁺, 513.1203; found, 513.1202. HPLC: t_R= 11.73 min; purity ≥

95% (UV: 254/280 nm). LCMS-ESI⁺(m/z) [M + H]⁺; calculated:

513.1, found: 513.1.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-

methylpyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]-

pyrimidin-7(8H)-one (12). Compound 12 was prepared according to

general procedures II−IV and obtained as a white solid with a yield of

1

27% (42 mg) over three steps. H NMR (500 MHz, DMSO-d₆, 300

K): δ 8.77−8.69 (m, 2H), 8.36 (d, J = 7.8 Hz, 1H), 8.17 (s, 3H), 8.07

(s, 1H), 7.91 (s, 1H), 7.87 (d, J = 1.7 Hz, 1H), 7.85 (dd, J = 7.9, 5.5

Hz, 1H), 7.69 (dd, J = 7.9, 1.7 Hz, 1H), 7.62 (d, J = 7.9 Hz, 1H),

5.13−5.07 (m, 1H), 4.55−4.50 (m, 2H), 4.23−4.20 (m, 2H), 3.61−

3.57 (m, 2H), 3.39 (s, 1H), 2.93 (s, 3H), 2.45 (s, 3H) ppm. ¹³C NMR

(126 MHz, DMSO-d₆, 300 K): δ 161.4, 161.0, 159.0, 155.6, 151.5,

145.3, 142.1, 137.3, 137.2, 135.0, 134.9, 133.5, 132.2, 130.2, 128.1,

125.2, 124.1, 104.0, 97.5, 66.4, 42.5, 42.3, 28.0, 18.9 ppm. HRMS

(FTMS + p MALDI) m/z: calcd for C₂₅H₂₆ClN₆O₃[M + H]⁺,

493.1749; found, 493.1752. HPLC: t_R= 10.60 min; purity ≥ 95%

(UV: 254/280 nm). LCMS-ESI⁺(m/z) [M + H]⁺; calcd, 493.2;

found, 493.2.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(pyri-

din-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]pyrimidin-7(8H)-

one (8). Compound 8 was prepared according to general procedures

II−IV and was obtained as a white solid with a yield of 15% (31 mg)

1

over three steps. H NMR (500 MHz, DMSO-d₆, 300 K): δ 8.73−

8.66 (m, 2H), 8.24 (d, J = 1.7 Hz, 1H), 8.13−8.09 (m, 2H), 8.06−

8.05 (m, 3H), 7.99 (td, J = 7.8, 1.7 Hz, 1H), 7.88 (s, 1H), 7.53 (d, J =

8.0 Hz, 1H), 7.48−7.46 (m, 1H), 5.13−5.12 (m, 1H), 4.55−4.49 (m,

2H), 4.22−4.19 (m, 2H), 3.59−3.55 (m, 2H), 3.39 (s, 1H), 2.92 (d, J

= 2.2 Hz, 3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ

163.1, 161.6, 161.0, 159.3, 155.4, 153.8, 149.3, 139.4, 138.2, 137.0,

136.3, 133.8, 132.4, 127.1, 125.0, 124.2, 123.6, 121.0, 97.4, 66.3, 42.4,

42.1, 27.9 ppm. HRMS (FTMS + p MALDI): m/z calculated for

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-

methoxypyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]-

pyrimidin-7(8H)-one (13). Compound 13 was prepared according to

general procedures II-IV and obtained as a yellow solid with a yield of

C₂₄H₂₄ClN₆O₃[M + H]⁺: 479.1593, found: 479.1613. HPLC: t_R

=

10.74 min; purity ≥ 95% (UV: 254/280 nm). LCMS-ESI⁺(m/z) [M

+ H]⁺: calcd, 479.2; found, 479.0.

1

33% (41 mg) over three steps. H NMR (500 MHz, DMSO-d₆, 300

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-ﬂu-

oro-6-methylpyridin-2-yl)-phenyl)-2-(methylamino)pyrido[2,3-d]-

pyrimidin-7(8H)-one (9). Compound 9 was prepared according to

general procedures II−IV and was obtained as a yellow solid with a

yield of 11% (21 mg) over three steps. ¹H NMR (500 MHz, DMSO-

d₆, 300 K): δ 8.74−8.66 (m, 1H), 8.07−8.03 (m, 4H), 7.98−7.97 (m,

1H), 7.93 (d, J = 8.1 Hz, 1H), 7.88 (s, 1H), 7.77 (dd, J = 11.3, 8.5 Hz,

1H), 7.54 (d, J = 8.0 Hz, 1H), 7.38 (dd, J = 8.4, 3.3 Hz, 1H), 5.14−

5.06 (m, 1H), 4.56−4.49 (m, 2H), 4.22−4.20 (m, 2H), 3.59−3.55

(m, 2H), 3.40 (s, 1H), 2.92 (d, J = 4.5 Hz, 3H), 2.53 (s, 3H) ppm.

¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 161.8, 160.9, 159.5, 156.8,

K): δ 8.73−8.66 (m, 1H), 8.31 (d, J = 4.4 Hz, 1H), 8.05 (s, 3H), 8.01

(s, 1H), 7.97−7.96 (m, 1H), 7.92 (d, J = 8.0 Hz, 1H), 7.87 (s, 1H),

7.66 (d, J = 8.4 Hz, 1H), 7.48−7.44 (m, 2H), 5.13−5.06 (m, 1H),

4.56−4.49 (m, 2H), 4.22−4.21 (m, 2H), 3.91 (s, 3H), 3.59−3.55 (m,

2H), 3.39 (s, 1H), 2.92 (d, J = 3.8 Hz, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 161.7, 161.0, 159.6, 159.4, 155.3, 153.7,

144.2, 140.9, 138.4, 136.8, 135.3, 132.7, 131.4, 129.5, 127.6, 124.4,

120.2, 103.9, 97.4, 66.3, 55.8, 42.3, 42.1, 27.9 ppm. HRMS (FTMS +

p MALDI) m/z: calcd for C₂₅H₂₆ClN₆O₄[M + H]⁺: 509.1699; found,

509.1712. HPLC: t_R= 10.68 min; purity ≥ 95% (UV: 254/280 nm).

LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 509.2; found: 509.0.

155.3, 154.8, 154.3, 154.2, 141.5, 141.4, 137.0, 136.2, 136.1, 133.3,

132.1, 128.8, 128.7, 126.9, 125.4, 125.2, 124.7, 124.6, 124.1, 103.9,

97.4, 66.2, 42.3, 42.1, 27.9, 23.5 ppm. HRMS (FTMS + p MALDI):

m/z calculated for C₂₅H₂₅ClFN₆O₃[M + H]⁺: 511.1655, found:

511.1647. HPLC: t_R= 11.62 min; purity ≥ 95% (UV: 254/280 nm).

LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 511.2; found, 511.1.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-ﬂu-

oropyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]pyrimidin-

7(8H)-one (10). Compound 10 was prepared according to general

procedures II-IV and obtained as a yellow solid with a yield of 23%

(35 mg) over three steps. ¹H NMR (500 MHz, DMSO-d₆, 300 K): δ

8.73−8.66 (m, 1H), 8.60−8.59 (m, 1H), 8.09 (s, 3H), 8.05 (s, 1H),

7.98−7.97 (m, 1H), 7.95−7.94 (m, 1H), 7.92−7.88 (m, 2H), 7.57−

7.54 (m, 2H), 5.14−5.06 (m, 1H), 4.56−4.49 (m, 2H), 4.22−4.20

(m, 2H), 3.60−3.56 (m, 2H), 3.39 (s, 1H), 2.91 (d, J = 4.5 Hz, 3H)

ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ 170.0, 161.8, 160.9,

159.5, 158.3, 156.3, 155.3, 146.0, 142.9, 137.0, 136.4, 136.0, 133.4,

132.2, 128.9, 127.0, 125.3, 125.1, 125.0, 124.0, 103.9, 97.4, 66.3, 42.3,

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(3-

methoxy-6-methylpyridin-2-yl)phenyl)-2-(methylamino)pyrido-

[2,3-d]pyrimidin-7(8H)-one (14). Compound 14 was prepared

according to general procedures II−IV and obtained as a yellow

1

solid with a yield of 12% (17 mg) over three steps. H NMR (500

MHz, DMSO-d₆, 300 K): δ 8.73−8.66 (m, 1H), 8.08−8.04 (m, 3H),

7.99 (d, J = 1.7 Hz, 1H), 7.97−7.96 (m, 1H), 7.91 (dd, J = 8.0, 1.7

Hz, 1H), 7.86 (s, 1H), 7.59 (d, J = 8.5 Hz, 1H), 7.46 (d, J = 8.0 Hz,

1H), 7.32 (d, J = 8.5 Hz, 1H), 5.14−5.06 (m, 1H), 4.56−4.49 (m,

2H), 4.22−4.20 (m, 2H), 3.88 (s, 3H), 3.59−3.55 (m, 2H), 3.40 (s,

1H), 2.92 (d, J = 4.5 Hz, 3H), 2.49 (s, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 161.8, 161.0, 159.5, 155.3, 151.8, 148.8,

142.8, 138.2, 136.8, 135.3, 132.6, 131.4, 129.4, 127.6, 124.4, 123.8,

121.3, 103.9, 97.4, 66.2, 56.0, 42.3, 42.1, 27.9, 22.8 ppm. HRMS

(FTMS + p MALDI) m/z: calcd for C₂₆H₂₈ClN₆O₄, 523.1855 [M +

H]⁺; found, 523.1846 [M + H]⁺. HPLC: t_R= 10.66 min; purity ≥

95% (UV: 254/280 nm). LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 523.2;

found, 523.0.

8155

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(6-

(triﬂuoromethyl)pyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-

d]pyrimidin-7(8H)-one (15). Compound 15 was prepared according

to general procedures II−IV and obtained as a yellow solid with a

yield of 34% (40 mg) over three steps. ¹H NMR (500 MHz, DMSO-

d₆, 300 K): δ 8.73−8.66 (m, 1H), 8.41 (d, J = 8.0 Hz, 1H), 8.26 (s,

1H), 8.23 (t, J = 7.9 Hz, 1H), 8.16−8.12 (m, 4H), 7.98 (s, 1H), 7.92

(d, J = 7.7 Hz, 1H), 7.87 (s, 1H), 7.58 (d, J = 8.0 Hz, 1H), 5.13−5.06

(m, 1H), 4.54−4.49 (m, 2H), 4.22−4.20 (m, 2H), 3.60−3.56 (m,

2H), 3.39 (s, 1H), 2.91 (d, J = 2.8 Hz, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 161.8, 160.0, 155.5, 155.4, 154.9, 147.0,

146.7, 146.4, 146.2, 140.0, 138.2, 137.1, 137.0, 134.0, 132.7, 127.3,

125.3, 124.8, 124.2, 124.0, 122.7, 120.5, 119.9, 118.3, 103.9, 97.4,

66.3, 42.3, 42.1, 27.9 ppm. HRMS (FTMS + p MALDI) m/z: calcd

for C₂₅H₂₃ClF₃N₆O₃, 547.1467 [M + H]⁺; found, 547.1457 [M +

H]⁺. HPLC: t_R= 12.35 min; purity ≥ 95% (UV: 254/280 nm).

LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 547.1; found, 547.0.

hydrogens, and also ﬁx missing side-chain atoms. The missing

activation loop between Ala174 and Thr194 of MST3 was modeled

based on the MST3 (PDB 3A7J) coordinates and optimized using

Prime,³¹and Thr178 was kept phosphorylated. The homology model

of the human SIK2 kinase domain (UniProt: Q9H0K1, residues 20−

271) was generated based on a stable frame from the MST3 MD

simulation (described below) using Prime.³²All ligands for docking

were drawn using Maestro and prepared using LigPrep³³to generate

the 3D conformation, adjust the protonation state to physiological pH

(7.4), and calculate the partial atomic charges, with the force ﬁeld

OPLS3e.³⁴Docking studies with the prepared ligands were performed

using Glide (Glide V7.7)^35,36with the ﬂexible modality of induced-ﬁt

docking with extra precision (XP), followed by a side-chain

minimization step using Prime. Ligands were docked within a grid

around 12 Å from the centroid of the co-crystallized ligand, generating

ten poses per ligand.

Molecular Dynamics Simulations. Selected docking poses were

further validated by MD simulations, where ligand stability within the

proposed pocket and its interactions were evaluated. MD simulations

were carried out using a Desmond³⁷engine with the OPLS3e force-

ﬁeld,^34,38which leads to improved performance in predicting

protein−ligand binding aﬃnities. The protein−ligand systems were

placed in a cubic box with 13 Å from the box edges to any atom of the

protein, using PBC conditions, and ﬁlled with TIP3P³⁹water. Then,

all systems were equilibrated by short simulations under the NPT

ensemble for 5 ns, implementing a Berendsen thermostat and barostat

methods. A constant temperature of 310 K and 1 atm of pressure were

used throughout the simulation using a Nose−Hoover thermostat

algorithm and Martyna−Tobias−Klein Barostat algorithm, respec-

tively. After minimization and relaxing steps, production steps of at

least 1 μs were performed. All MD simulations were performed in at

least three independent runs with randomly generated seeds.

Trajectories and interaction data are available from the Zenodo

repository (accession code: 10.5281/zenodo.4288340). The stability

of the MD simulations was monitored by analyzing root mean square

deviation (rmsd) and root mean square ﬂuctuation (RMSF) of the

ligand and protein along the trajectory. Variation in the rmsd values

stabilized after 150 ns for all the ligands, which was related to the

stabilization of the activation loop in the case of MST3, or the

relaxation of the homology model, in the case of SIK2. For principal

component analyses, the backbone of each frame was extracted and

aligned using trj_selection_dl.py and trj_align.py scripts from

Schrodinger. Individual simulations from all the runs were merged

using trj_merge.py into a ﬁnal trajectory and CMS ﬁle, which was

further used for the generation of the principal components. Principal

components of protein C-alpha atoms were calculated using the

trj_essential_dynamics.py script. Most of the large movements were

captured in the ﬁrst two components, with the ﬁrst component having

27% and the second component having 17% of the total motion

protein−ligand interactions. Protein conformational changes were

analyzed using the Simulation Interaction Diagram tool (SID).

WaterMap Calculations. WaterMap calculations⁴⁰were per-

formed using Maestro, and the system was solvated in TIP3P water

box extending at least 10 Å beyond the truncated protein in all

directions. 5 ns MD simulation was performed, following a standard

relaxation protocol, and the water molecule trajectories were then

clustered into distinct hydration sites. Entropy and enthalpy values for

each hydration site were calculated using inhomogeneous solvation

theory.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(6-

(diﬂuoromethyl)pyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-

d]pyrimidin-7(8H)-one (16). Compound 16 was prepared according

to general procedures II-IV and obtained as a yellow solid with a yield

1

of 30% (30 mg) over three steps. H NMR (500 MHz, DMSO-d₆,

300 K): δ 8.73−8.66 (m, 1H), 8.28−8.27 (m, 2H), 8.16−8.10 (m,

5H), 7.98 (s, 1H), 7.88 (s, 1H), 7.72 (d, J = 7.7 Hz, 1H), 7.56 (d, J =

8.0 Hz, 1H), 7.06 (t, J = 55.0 Hz, 1H), 5.13−5.06 (m, 1H), 4.55−4.49

(m, 2H), 4.22−4.21 (m, 2H), 3.60−3.56 (m, 2H), 3.40 (s, 1H), 2.91

(d, J = 4.1 Hz, 3H) ppm. ¹³C NMR (126 MHz, DMSO-d₆, 300 K): δ

161.8, 160.9, 159.5, 155.3, 154.3, 152.3, 152.1, 151.9, 139.3, 138.7,

137.0, 136.7, 133.9, 132.5, 127.1, 125.1, 124.0, 122.6, 119.7, 115.7,

113.8, 111.9, 103.9, 97.4, 66.3, 42.3, 42.1, 27.9 ppm. HRMS (FTMS +

p MALDI) m/z: calculated for C₂₅H₂₄ClF₂N₆O₃, 529.1561 [M + H]⁺;

found, 529.1553 [M + H]⁺. HPLC: t_R= 12.06 min; purity ≥ 95%

(UV: 254/280 nm). LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 529.2;

found, 529.0.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(6-

methoxypyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]-

pyrimidin-7(8H)-one (17). Compound 17 was prepared according to

general procedures II−IV and obtained as a yellow solid with a yield

1

of 28% (43 mg) over three steps. H NMR (500 MHz, DMSO-d₆,

300 K): δ 8.72−8.67 (m, 1H), 8.21 (d, J = 1.7 Hz, 1H), 8.11−8.08

(m, 4H), 8.05 (s, 1H), 7.85 (s, 1H), 7.82−7.79 (m, 1H), 7.65 (d, J =

7.4 Hz, 1H), 7.50 (d, J = 8.0 Hz, 1H), 6.83 (d, J = 8.2 Hz, 1H), 5.13−

5.07 (m, 1H), 4.54−4.50 (m, 2H), 4.23−4.20 (m, 2H), 3.97 (s, 3H),

3.60−3.56 (m, 2H), 3.40 (s, 1H), 2.92 (s, 3H) ppm. ¹³C NMR (126

MHz, DMSO-d₆, 300 K): δ 163.4, 162.4, 161.1, 158.6, 155.6, 151.9,

140.3, 139.7, 136.8, 136.0, 133.8, 132.3, 126.9, 124.9, 124.7, 113.4,

110.3, 104.0, 97.4, 66.4, 53.0, 42.4, 42.2, 28.0 ppm. HRMS (FTMS +

p MALDI) m/z: calcd for C₂₅H₂₆ClN₆O₄, 509.1699 [M + H]⁺; found,

509.1698 [M + H]⁺. HPLC: t_R= 12.27 min; purity ≥ 95% (UV: 254/

280 nm). LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 509.2; found, 509.2.

8-(((2r,5r)-5-Amino-1,3-dioxan-2-yl)methyl)-6-(2-chloro-4-(4,6-

(dimethyl)pyridin-2-yl)phenyl)-2-(methylamino)pyrido[2,3-d]-

pyrimidin-7(8H)-one (18). Compound 18 was prepared according to

general procedures II−IV and obtained as a white solid with a yield of

1

22% (34 mg) over three steps. H NMR (500 MHz, DMSO-d₆, 300

K): δ 8.74−8.67 (m, 1H), 8.20 (d, J = 1.8 Hz, 1H), 8.07−8.03 (m,

4H), 7.99 (s, 1H), 7.90 (s, 1H), 7.89 (s, 1H), 7.58 (d, J = 8.0 Hz,

1H), 7.39 (s, 1H), 5.14−5.06 (m, 1H), 4.55−4.49 (m, 2H), 4.22−

4.19 (m, 2H), 3.60−3.55 (m, 2H), 3.39 (s, 1H), 2.92 (d, J = 4.1 Hz,

3H), 2.62 (s, 3H), 2.47 (s, 3H) ppm. ¹³C NMR (126 MHz, DMSO-

d₆, 300 K): δ 161.7, 161.0, 160.9, 159.7, 159.5, 156.1, 155.3, 151.5,

137.0, 136.9, 133.7, 132.4, 127.8, 125.7, 124.8, 123.9, 120.8, 103.9,

97.4, 66.3, 42.3, 42.1, 27.9, 22.2, 20.9 ppm. HRMS (FTMS + p

MALDI) m/z: calcd for C₂₆H₂₈ClFN₆O₃, 507.1906 [M + H]⁺; found,

507.1898 [M + H]⁺. HPLC: t_R= 10.68 min; purity ≥ 95% (UV: 254/

280 nm). LCMS-ESI⁺(m/z) [M + H]⁺: calcd, 507.2; found, 507.1.

Homology Modeling and Molecular Docking. System

preparation and docking calculations were performed using the

Statistical Analyses and Figures. Structural images were

generated using PyMol 2.4.3,⁴¹and graphs were plotted using either

GraphPad Prism version 8.4, GraphPad Software (San Diego,

California USA, www.graphpad.com) or the Seaborn package

implemented in Python 3.8.

In Vitro Kinase Assay. To measure kinase activity of SIK2, in vitro

kinase assays were performed as described in Matthess et al.⁴²

Samples were resolved by sodium dodecyl sulfate-polyacrylamide gel

electrophoresis and subjected to autoradiography. Recombinant

human active SIK2 was purchased from ProQinase Reaction Biology,

product no. 1371-0000-1.

̈

Schrodinger Drug Discovery suite for molecular modeling (version

2019.4). Protein−ligand complexes were prepared with the protein

preparation wizard³⁰to ﬁx protonation states of amino acids, add

8156

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Proliferation Assays. To measure the cell proliferation using the

Cell Titer-Blue Cell Viability Assay, 2,500 cells per well were seeded

in 96-well plates and incubated with the diﬀerent single or

combination treatments. Cells were incubated with 20 μL substrate

of the Cell Titer-Blue Cell Viability Assay for 4 h, and the light

absorbance was measured at 540 nm (Victor X4, Perkin Elmer).

Caspase 3/7 Activity. The activity of Caspase 3/7 was

determined using the Caspase-Glo 3/7 Assay (Promega). Upon the

incubation of the cells with the diﬀerent treatment regimens, 20 μL of

the substrate were added to cell lysates. After 30 min shaking at room

temperature in the dark, luminescence was detected (Victor X4,

PerkinElmer).

was added at a ﬁnal concentration of 0.5 mM. The cells were grown

overnight and harvested the next day by centrifugation at 5000 rpm

for 10 min. The cell pellet was resuspended in lysis buﬀer containing

50 mM HEPES pH 7.5, 500 mM NaCl, 30 mM imidazole, 5%

glycerol, and 0.5 mM TCEP. Protease inhibitor cocktail from Sigma

was added at a dilution of 1:5000. The resuspended cells were lysed

by sonication and cleared of DNA by the addition of 0.15%

polyethylenimine, followed by centrifugation at 23,000 rpm for 30

min. The supernatant was loaded onto a gravity column containing 5

ml of 50% Ni-NTA resin (Qiagen) pre-equilibrated with 40 ml of lysis

buﬀer. The column was washed with 100 ml of lysis buﬀer, and MST3

and MST4 were then eluted with lysis buﬀer containing 50, 100, 200,

and 300 mM imidazole. The fractions containing MST3 were pooled

and concentrated using a 10 kDa cutoﬀ ultraﬁltration device and

further puriﬁed on a size exclusion chromatography (SEC; HiLoad

16/600 Superdex 200) pre-equilibrated with SEC buﬀer containing

25 mM HEPES pH 7.5, 200 mM NaCl, 5% glycerol, and 0.5 mM

TCEP. In the case of MST4, the eluted fractions from the Ni-NTA

column containing MST4 were dialyzed in SEC buﬀer and submitted

to the cleavage of the tag by TEV protease overnight. The cleaved

MST4 sample was then loaded onto a gravity column containing 2 ml

of 50% Ni-NTA, and the ﬂow through was collected, concentrated,

and puriﬁed by SEC using a Hiload 16/600 Superdex 200 column

(same buﬀer as for MST3). Protein identity was conﬁrmed by

electrospray ionization mass spectrometry (MST3: expected 34,507.7

Da, observed 34,506.9 Da, MST4: expected 33,817.9 Da, observed

33,818.7 Da).

Crystallization and Structure Determination. Inhibitors were

added to the protein solution (10 mg/mL MST3 or 12 mg/mL to

MST4) at a ﬁnal concentration of 1 mM and incubated for at least 1 h

at 4 °C. The samples were then centrifuged to remove any insoluble

material, and crystals were obtained using the sitting drop vapor

diﬀusion method. Crystallization buﬀers are given in Table S7. The

crystals were cryoprotected with crystallization buﬀer complemented

with 25% ethylene glycol and ﬂash-frozen in liquid nitrogen.

Diﬀraction data were collected at 100 K at the Swiss Light Source

(beamline X06SA). The data were processed with XDS⁴³and scaled

with AIMLESS,⁴⁴and the structures were solved by molecular

replacement with the program Phaser⁴⁵implemented in the CCP4

suite⁴⁶using the structure of MST3 (PDB ID 3ZHP) as a search

model for MST3, and subsequently the new G-5555/MST3 complex

as a search model for MST4. The structures were then reﬁned using

iterative cycles of model building in COOT⁴⁷and reﬁnement with

REFMAC⁴⁸or PHENIX.⁴⁹Ligand dictionary ﬁles were obtained from

the GRADE server (http://grade.globalphasing.org). Data collection

and reﬁnement statistics are summarized in Table S7.

Spheroid (3D) Cultures. A cell suspension of 1500 cells/50 μL

was prepared and pipetted from the topside into a 96 well Perfect 3D

Hanging Drop plate (BioTrend). Plates were incubated at 37 °C for

several days until hanging drops had developed. The 3D culture was

harvested on a 96 well plate covered with 1% agarose by low-spin

centrifugation. The treatment of cells with the diﬀerent combinations

was performed as indicated. Cells were stained with the LIVE/DEAD

viability/cytotoxicity kit (Molecular Probes/Thermoﬁsher) for 30

min and inspected using a ﬂuorescence microscope AxioObserver.Z1

microscope with a HCX PL APO CS 10 × 1.4 UV objective (Zeiss,

̈

Gottingen). While the polyanionic dye calcein is retained in live cells,

producing an intense uniform green ﬂuorescence, EthD-1 enters cells

with damaged membranes, thereby producing a bright red

ﬂuorescence upon binding to nucleic acids in dead cells. The ratios

of viable/dead cells were calculated with the software ImageJ Fiji.

NanoBRET Assay. The assay was performed as described

previously.^22,23In brief, full-length kinases are obtained as plasmids

cloned in frame with a terminal NanoLuc-fusion (Promega), as

speciﬁed in Table S6. Plasmids were transfected into HEK293T cells

using FuGENE HD (Promega, E2312), and proteins were allowed to

express for 20 h. Serially diluted inhibitor and NanoBRET Kinase

Tracer K10 (Promega) at a concentration determined previously as

the Tracer K10 IC₅₀(Table S6) were pipetted into white 384-well

plates (Greiner 781207) using an Echo acoustic dispenser (Labcyte).

The corresponding protein-transfected cells were added and reseeded

at a density of 2 × 10⁵cells/mL after trypsinization and resuspending

in Opti-MEM without phenol red (Life Technologies). The system

was allowed to equilibrate for 2 h at 37 °C/5% CO₂prior to BRET

measurements. To measure BRET, NanoBRET NanoGlo Substrate +

Extracellular NanoLuc inhibitor (Promega, N2540) was added as per

the manufacturer’s protocol, and ﬁltered luminescence was measured

on a PHERAstar plate reader (BMG Labtech) equipped with a

luminescence ﬁlter pair [450 nm BP ﬁlter (donor) and 610 nm LP

ﬁlter (acceptor)]. Competitive displacement data were then graphed

using GraphPad Prism 8 software using a normalized 3-parameter

curve ﬁt with the following equation: Y = 100/(1 + 10^{(X‑logIC50)}).

Diﬀerential Scanning Fluorimetry Assay. DSF assays were

performed as previously described by Fedorov et al.²⁴Diﬀerences in

melting temperature are given as ΔT_mvalues in °C.

ASSOCIATED CONTENT

■

sı

* Supporting Information

Kinome-Wide Selectivity Proﬁle. Compounds 6, 7, and 9 were

tested against a panel of 468 kinases in the scanMAX kinase assay

performed by Euroﬁns Scientiﬁc. Compound 6 was tested at a

concentration of 100 nM and 1 μM, while compounds 7 and 9 were

tested at 1 μM. Compound 10 was tested against a panel of 443

kinases in the radiometric ³³PanQinase Activity Assay at a

concentration of 1 μM. The assay was performed by Reaction Biology.

Protein Expression and Puriﬁcation. MST3 was expressed

using a construct containing residues R4-D301 followed by a C-

terminal hexahistidine tag in vector pNIC-CH. MST4 was expressed

as a fusion protein with an N-terminal hexahistidine tag, followed by a

TEV cleavage site and MST4 residues M1-S300 in vector pNIC28-

Bsa4. The expression plasmids were transformed into BL21(D3)-R3-

pRARE2 cells, and cells were grown in 2× LB media containing 50 μg

mL⁻¹kanamycin and 34 μg mL⁻¹chloramphenicol and cultured

overnight at 37 °C. 10−12 mL of the overnight culture were

transferred into 1L of TB media containing 50 μg mL⁻¹kanamycin,

and cultures were grown at 37 °C with shaking until an OD₆₀₀of 1.5−

1.8 was reached. The temperature was reduced to 18 °C, and IPTG

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c02144.

Selectivity proﬁle of G-5555, 7, 9, 10, and SIK2 inhibitor

ARN-3236; NanoBRET assay information; data collec-

tion and reﬁnement statistics; analytical data for

compounds 7−38; comparison of the crystal structures;

sequence alignment of SIKs and main oﬀ-targets of G-

5555, MSTs, and PAK1; binding-pocket ﬂuctuation in

MD simulations; comparison of the selectivity proﬁles of

G-5555 and compound 7; binding mode of compounds

9, 10, and 6 in MST3; comparison of the MST3 and

PAK1 back pocket; in vitro kinase assay of MRIA9 and

G-5555; results of compound 10 tested against 60

cancer cell lines; and metabolism of MRIA9 (PDF)

Molecular formula strings with activity data (CSV)

8157

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Pharmacy, University of Eastern Finland, Kuopio 70210,

Finland; orcid.org/0000-0003-4196-4204

Klaus Strebhardt − Department of Obstetrics and

Gynaecology, School of Medicine, Johann Wolfgang Goethe

University, Frankfurt am Main 60590, Germany;

orcid.org/0000-0003-2173-9763

Mourad Sanhaji − Department of Obstetrics and Gynaecology,

School of Medicine, Johann Wolfgang Goethe University,

Frankfurt am Main 60590, Germany

AUTHOR INFORMATION

■

Corresponding Author

Stefan Knapp − Institute of Pharmaceutical Chemistry, Johann

Wolfgang Goethe University, Frankfurt am Main 60438,

Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany;

German Translational Cancer Network (DKTK) and

Frankfurt Cancer Institute (FCI), Frankfurt am Main

60438, Germany; orcid.org/0000-0001-5995-6494;

Email: knapp@pharmchem.uni-frankfurt.de

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.0c02144

Author Contributions

Authors

^#R.T. and M. Rak contributed equally to the work. R.T., M.

Rak, and S.K. designed the project; M. Rak, I.A., and T.H.

synthesized the compounds; R.T. expressed, puriﬁed, and co-

crystallized MST3/4; R.T. and A.C.J. reﬁned the structures of

MST3 and MST4 inhibitor complexes; T.K. and A.P.

performed MD simulations and WaterMap calculations; R.T.

and M.R. performed DSF assays; L.B. and B.-T.B. performed

NanoBRET assays; M. Raab and M.S. performed auto-

phosphorylation and proliferation assays and tested the eﬀect

of SIK inhibitors in cancer cell models; and K.S. and S.K.

supervised the research. The manuscript was written by R.T.,

M. Rak, A.C.J., and S.K. with contributions from all coauthors.

Roberta Tesch − Institute of Pharmaceutical Chemistry,

Johann Wolfgang Goethe University, Frankfurt am Main

60438, Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany;

orcid.org/0000-0003-3347-2111

Marcel Rak − Institute of Pharmaceutical Chemistry, Johann

Wolfgang Goethe University, Frankfurt am Main 60438,

Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany

Monika Raab − Department of Obstetrics and Gynaecology,

School of Medicine, Johann Wolfgang Goethe University,

Frankfurt am Main 60590, Germany

Lena M. Berger − Institute of Pharmaceutical Chemistry,

Johann Wolfgang Goethe University, Frankfurt am Main

60438, Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany;

orcid.org/0000-0002-7835-8067

Notes

The authors declare no competing ﬁnancial interest.

Coordinates and structure factors of the MST3/4-inhibitor

complexes are available in the Protein Data Bank (PDB) under

accession codes 7B30 (MST3-6 complex), 7B31 (MST3-10

complex), 7B32 (MST3-7 complex), 7B33 (MST3-16

complex), 7B34 (MST3-9 complex), 7B35 (MST3-14

complex), and 7B36 (MST4-6 complex).

Thales Kronenberger − Dept. of Internal Medicine VIII,

University Hospital Tubingen, Tubingen 72076, Germany;

̈ ̈

School of Pharmacy, University of Eastern Finland, Kuopio

70210, Finland

ACKNOWLEDGMENTS

■

R.T. was supported by the German Research Foundation

(DFG) grant 397659447. M.R. is grateful for the support by

Dr. Hilmer-Stiftung. We are also grateful for support by the

SGC, a registered charity (no. 1097737) that receives funds

from AbbVie, Bayer AG, Boehringer Ingelheim, the Canada

Foundation for Innovation, Eshelman Institute for Innovation,

Genentech, Genome Canada through Ontario Genomics

Institute [OGI-196], EU/EFPIA/OICR/McGill/KTH/Dia-

mond, Innovative Medicines Initiative 2 Joint Undertaking

[EUbOPEN grant 875510], Janssen, Merck KGaA (aka EMD

in Canada and US), Merck & Co (aka MSD outside Canada

Andreas C. Joerger − Institute of Pharmaceutical Chemistry,

Johann Wolfgang Goethe University, Frankfurt am Main

60438, Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany;

orcid.org/0000-0002-1232-0138

Benedict-Tilman Berger − Institute of Pharmaceutical

Chemistry, Johann Wolfgang Goethe University, Frankfurt

am Main 60438, Germany; Structural Genomics Consortium

(SGC), Buchmann Institute for Life Sciences, Johann

Wolfgang Goethe University, Frankfurt am Main 60438,

Germany; orcid.org/0000-0002-3314-2617

and US), Pﬁzer, the Sao Paulo Research Foundation-FAPESP

̃

Ismahan Abdi − Institute of Pharmaceutical Chemistry,

Johann Wolfgang Goethe University, Frankfurt am Main

60438, Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany

Thomas Hanke − Institute of Pharmaceutical Chemistry,

Johann Wolfgang Goethe University, Frankfurt am Main

60438, Germany; Structural Genomics Consortium (SGC),

Buchmann Institute for Life Sciences, Johann Wolfgang

Goethe University, Frankfurt am Main 60438, Germany;

orcid.org/0000-0001-7202-9468

and Takeda as well as support from the German translational

cancer network DKTK and the Frankfurt Cancer Institute

(FCI). A.C.J. was supported by the German Research

Foundation (DFG) grant JO 1473/1-3. The data collection

at SLS was supported by funding from the European Union’s

Horizon 2020 research and innovation program grant

agreement number 730872, project CALIPSOplus. We thank

the staﬀ at beamline X06SA of the Swiss Light Source for

assistance during data collection. We thank the CSC-Finland

for the generous computational resources provided. We thank

Amelie Tjaden for help with the cell-based assays and Astrid

Kaiser for performing the microsomal stability assays.

Antti Poso − Dept. of Internal Medicine VIII, University

̈ ̈

Hospital Tubingen, Tubingen 72076, Germany; School of

8158

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(12) Zhao, J.; Zhang, X.; Gao, T.; Wang, S.; Hou, Y.; Yuan, P.; Yang,

Y.; Yang, T.; Xing, J.; Li, J.; Liu, S. SIK2 enhances synthesis of fatty

acid and cholesterol in ovarian cancer cells and tumor growth through

PI3K/Akt signaling pathway. Cell Death Dis. 2020, 11, 25.

(13) Ahmed, A. A.; Lu, Z.; Jennings, N. B.; Etemadmoghadam, D.;

Capalbo, L.; Jacamo, R. O.; Barbosa-Morais, N.; Le, X.-F.; Vivas-

Mejia, P.; Lopez-Berestein, G.; Grandjean, G.; Bartholomeusz, G.;

Liao, W.; Andreeff, M.; Bowtell, D.; Glover, D. M.; Sood, A. K.; Bast,

R. C. SIK2 is a centrosome kinase required for bipolar mitotic spindle

formation that provides a potential target for therapy in ovarian

cancer. Canc. Cell 2010, 18, 109−121.

(14) Zhou, J.; Alfraidi, A.; Zhang, S.; Santiago-O’Farrill, J. M.;

Yerramreddy Reddy, V. K.; Alsaadi, A.; Ahmed, A. A.; Yang, H.; Liu,

J.; Mao, W.; Wang, Y.; Takemori, H.; Vankayalapati, H.; Lu, Z.; Bast,

R. C. A novel compound ARN-3236 inhibits salt-inducible kinase 2

and sensitizes ovarian cancer cell lines and xenografts to paclitaxel.

Clin. Cancer Res. 2017, 23, 1945−1954.

(15) Montenegro, R. C.; Howarth, A.; Ceroni, A.; Fedele, V.; Farran,

B.; Mesquita, F. P.; Frejno, M.; Berger, B.-T.; Heinzlmeir, S.; Sailem,

H. Z.; Tesch, R.; Ebner, D.; Knapp, S.; Burbano, R.; Kuster, B.;

ABBREVIATIONS

■

ABL, Abelson murine leukemia viral oncogene homologue 1;

AKT, protein kinase B; AMPK, AMP-activated protein kinase;

BTK, Bruton’s tyrosine kinase; CaMK, calcium calmodulin

kinase; CREB, cAMP response element-binding protein;

DLBCL, diﬀuse large B-cell lymphoma; DSF, diﬀerential

scanning ﬂuorimetry; FGF, ﬁbroblast growth factor; HDAC,

histone deacetylase; JAK2, Janus kinase 2; KHS1, mitogen-

activated protein kinase kinase kinase kinase 5; LCK,

lymphocyte-speciﬁc protein tyrosine kinase; MST, mammalian

STE20-like protein kinase; NUAK2, SNF1/AMP kinase-

related kinase; PAK, p21-activated kinase; PI3K, phosphoinosi-

tide 3-kinase; SIK, salt-inducible kinase; SRPK1, serine/

arginine protein kinase 1; STE, homologous kinases to the

yeast proteins STE20, STE11, and STE7; TORC, transducer

of regulated CREB1; VEGFR2, vascular endothelial growth

factor receptor 2; YSK1, serine/threonine-protein kinase 25

Muller, S. Identification of molecular targets for the targeted

̈

REFERENCES

■

treatment of gastric cancer using dasatinib. Oncotarget 2020, 11,

535−549.

(1) Sun, Z.; Jiang, Q.; Li, J.; Guo, J. The potent roles of salt-

inducible kinases (SIKs) in metabolic homeostasis and tumorigenesis.

Signal Transduction Targeted Ther. 2020, 5, 150.

(2) Conkright, M. D.; Canettieri, G.; Screaton, R.; Guzman, E.;

Miraglia, L.; Hogenesch, J. B.; Montminy, M. TORCs: transducers of

regulated CREB activity. Mol. Cell 2003, 12, 413−423.

(3) Katoh, Y.; Takemori, H.; Lin, X.-Z.; Tamura, M.; Muraoka, M.;

Satoh, T.; Tsuchiya, Y.; Min, L.; Doi, J.; Miyauchi, A.; Witters, L. A.;

Nakamura, H.; Okamoto, M. Silencing the constitutive active

transcription factor CREB by the LKB1-SIK signaling cascade.

FEBS J. 2006, 273, 2730−2748.

(4) Chen, F.; Chen, L.; Qin, Q.; Sun, X. Salt-inducible kinase 2: an

oncogenic signal transmitter and potential target for cancer therapy.

Front. oncol. 2019, 9, 18.

(5) Cheng, H.; Liu, P.; Wang, Z. C.; Zou, L.; Santiago, S.; Garbitt,

V.; Gjoerup, O. V.; Iglehart, J. D.; Miron, A.; Richardson, A. L.; Hahn,

W. C.; Zhao, J. J. SIK1 couples LKB1 to p53-dependent anoikis and

suppresses metastasis. Sci. Signal. 2009, 2, ra35.

(6) Amara, S.; Majors, C.; Roy, B.; Hill, S.; Rose, K. L.; Myles, E. L.;

Tiriveedhi, V. Critical role of SIK3 in mediating high salt and IL-17

synergy leading to breast cancer cell proliferation. PloS One 2017, 12,

No. e0180097.

(7) Nagel, S.; Leich, E.; Quentmeier, H.; Meyer, C.; Kaufmann, M.;

Zaborski, M.; Rosenwald, A.; Drexler, H. G.; Macleod, R. A. F.

Amplification at 11q23 targets protein kinase SIK2 in diffuse large B-

cell lymphoma. Leuk. Lymphoma 2010, 51, 881−891.

(8) Tarumoto, Y.; Lin, S.; Wang, J.; Milazzo, J. P.; Xu, Y.; Lu, B.;

Yang, Z.; Wei, Y.; Polyanskaya, S.; Wunderlich, M.; Gray, N. S.;

Stegmaier, K.; Vakoc, C. R. Salt-inducible kinase inhibition suppresses

acute myeloid leukemia progression in vivo. Blood 2020, 135, 56−70.

(9) Tarumoto, Y.; Lu, B.; Somerville, T. D. D.; Huang, Y.-H.;

Milazzo, J. P.; Wu, X. S.; Klingbeil, O.; El Demerdash, O.; Shi, J.;

Vakoc, C. R. LKB1, Salt-inducible kinases, and MEF2C are linked

dependencies in acute myeloid leukemia. Mol. Cell 2018, 69, 1017−

1027.

(16) Sundberg, T. B.; Choi, H. G.; Song, J.-H.; Russell, C. N.;

Hussain, M. M.; Graham, D. B.; Khor, B.; Gagnon, J.; O’Connell, D.

̌

J.; Narayan, K.; Dancík, V.; Perez, J. R.; Reinecker, H.-C.; Gray, N. S.;

Schreiber, S. L.; Xavier, R. J.; Shamji, A. F. Small-molecule screening

identifies inhibition of salt-inducible kinases as a therapeutic strategy

to enhance immunoregulatory functions of dendritic cells. Proc. Natl.

Acad. Sci. U.S.A. 2014, 111, 12468−12473.

(17) Clark, K.; MacKenzie, K. F.; Petkevicius, K.; Kristariyanto, Y.;

Zhang, J.; Choi, H. G.; Peggie, M.; Plater, L.; Pedrioli, P. G. A.;

McIver, E.; Gray, N. S.; Arthur, J. S. C.; Cohen, P. Phosphorylation of

CRTC3 by the salt-inducible kinases controls the interconversion of

classically activated and regulatory macrophages. Proc. Natl. Acad. Sci.

U.S.A. 2012, 109, 16986−16991.

(18) Sundberg, T. B.; Liang, Y.; Wu, H.; Choi, H. G.; Kim, N. D.;

Sim, T.; Johannessen, L.; Petrone, A.; Khor, B.; Graham, D. B.;

Latorre, I. J.; Phillips, A. J.; Schreiber, S. L.; Perez, J.; Shamji, A. F.;

Gray, N. S.; Xavier, R. J. Development of chemical probes for

investigation of salt-inducible kinase function in vivo. ACS Chem. Biol

2016, 11, 2105−2111.

(19) Lombardi, M. S.; Gilliéron, C.; Dietrich, D.; Gabay, C. SIK

inhibition in human myeloid cells modulates TLR and IL-1R signaling

and induces an anti-inflammatory phenotype. J. Leukoc. Biol. 2016, 99,

711−721.

(20) Ndubaku, C. O.; Crawford, J. J.; Drobnick, J.; Aliagas, I.;

Campbell, D.; Dong, P.; Dornan, L. M.; Duron, S.; Epler, J.; Gazzard,

L.; Heise, C. E.; Hoeflich, K. P.; Jakubiak, D.; La, H.; Lee, W.; Lin, B.;

Lyssikatos, J. P.; Maksimoska, J.; Marmorstein, R.; Murray, L. J.;

O’Brien, T.; Oh, A.; Ramaswamy, S.; Wang, W.; Zhao, X.; Zhong, Y.;

Blackwood, E.; Rudolph, J. Design of selective PAK1 inhibitor G-

5555: improving properties by employing an unorthodox low-pKa

polar moiety. ACS Med. Chem. Lett. 2015, 6, 1241−1246.

(21) Rudolph, J.; Murray, L. J.; Ndubaku, C. O.; O’Brien, T.;

Blackwood, E.; Wang, W.; Aliagas, I.; Gazzard, L.; Crawford, J. J.;

Drobnick, J.; Lee, W.; Zhao, X.; Hoeflich, K. P.; Favor, D. A.; Dong,

P.; Zhang, H.; Heise, C. E.; Oh, A.; Ong, C. C.; La, H.; Chakravarty,

P.; Chan, C.; Jakubiak, D.; Epler, J.; Ramaswamy, S.; Vega, R.; Cain,

G.; Diaz, D.; Zhong, Y. Chemically diverse group I p21-activated

kinase (PAK) inhibitors impart acute cardiovascular toxicity with a

narrow therapeutic window. J. Med. Chem. 2016, 59, 5520−5541.

(22) Robers, M. B.; Friedman-Ohana, R.; Huber, K. V. M.;

Kilpatrick, L.; Vasta, J. D.; Berger, B.-T.; Chaudhry, C.; Hill, S.;

(10) Miranda, F.; Mannion, D.; Liu, S.; Zheng, Y.; Mangala, L. S.;

Redondo, C.; Herrero-Gonzalez, S.; Xu, R.; Taylor, C.; Chedom, D.

F.; Karaminejadranjbar, M.; Albukhari, A.; Jiang, D.; Pradeep, S.;

Rodriguez-Aguayo, C.; Lopez-Berestein, G.; Salah, E.; Abdul Azeez, K.

R.; Elkins, J. M.; Campo, L.; Myers, K. A.; Klotz, D.; Bivona, S.; Dhar,

S.; Bast, R. C.; Saya, H.; Choi, H. G.; Gray, N. S.; Fischer, R.; Kessler,

B. M.; Yau, C.; Sood, A. K.; Motohara, T.; Knapp, S.; Ahmed, A. A.

Salt-inducible kinase 2 couples ovarian cancer cell metabolism with

survival at the adipocyte-rich metastatic niche. Canc. Cell 2016, 30,

273−289.

Muller, S.; Knapp, S.; Wood, K. V. Quantifying target occupancy of

̈

small molecules within living cells. Annu. Rev. Biochem. 2020, 89,

557−581.

(11) Chen, J.-L.; Chen, F.; Zhang, T.-T.; Liu, N.-F. Suppression of

SIK1 by miR-141 in human ovarian cancer cell lines and tissues. Int. J.

Mol. Med. 2016, 37, 1601−1610.

(23) Vasta, J. D.; Corona, C. R.; Wilkinson, J.; Zimprich, C. A.;

Hartnett, J. R.; Ingold, M. R.; Zimmerman, K.; Machleidt, T.;

8159

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Kirkland, T. A.; Huwiler, K. G.; Ohana, R. F.; Slater, M.; Otto, P.;

Cong, M.; Wells, C. I.; Berger, B.-T.; Hanke, T.; Glas, C.; Ding, K.;

̈

Drewry, D. H.; Huber, K. V. M.; Willson, T. M.; Knapp, S.; Muller, S.;

Meisenheimer, P. L.; Fan, F.; Wood, K. V.; Robers, M. B.

Quantitative, wide-spectrum kinase profiling in live cells for assessing

the effect of cellular ATP on target engagement. Cell Chem. Biol. 2018,

25, 206−214.

(24) Fedorov, O.; Niesen, F. H.; Knapp, S. Kinase inhibitor

selectivity profiling using differential scanning fluorimetry. Methods

Mol. Biol. 2012, 795, 109−118.

(25) Karaman, M. W.; Herrgard, S.; Treiber, D. K.; Gallant, P.;

Atteridge, C. E.; Campbell, B. T.; Chan, K. W.; Ciceri, P.; Davis, M. I.;

Edeen, P. T.; Faraoni, R.; Floyd, M.; Hunt, J. P.; Lockhart, D. J.;

Milanov, Z. V.; Morrison, M. J.; Pallares, G.; Patel, H. K.; Pritchard,

S.; Wodicka, L. M.; Zarrinkar, P. P. A quantitative analysis of kinase

inhibitor selectivity. Nat. Biotechnol. 2008, 26, 127−132.

(26) Davis, M. I.; Hunt, J. P.; Herrgard, S.; Ciceri, P.; Wodicka, L.

M.; Pallares, G.; Hocker, M.; Treiber, D. K.; Zarrinkar, P. P.

Comprehensive analysis of kinase inhibitor selectivity. Nat. Biotechnol.

2011, 29, 1046−1051.

(27) Karpov, A. S.; Amiri, P.; Bellamacina, C.; Bellance, M.-H.;

Breitenstein, W.; Daniel, D.; Denay, R.; Fabbro, D.; Fernandez, C.;

Galuba, I.; Guerro-Lagasse, S.; Gutmann, S.; Hinh, L.; Jahnke, W.;

Klopp, J.; Lai, A.; Lindvall, M. K.; Ma, S.; Möbitz, H.; Pecchi, S.;

Rummel, G.; Shoemaker, K.; Trappe, J.; Voliva, C.; Cowan-Jacob, S.

W.; Marzinzik, A. L. Optimization of a dibenzodiazepine hit to a

potent and selective allosteric PAK1 inhibitor. ACS Med. Chem. Lett.

2015, 6, 776−781.

Sacerdoti, F. D.; Salmon, J. K.; Shan, Y.; Shaw, D. E. Scalable

algorithms for molecular dynamics simulations on commodity

clusters; ACM/IEEE SC 2006 Conference (SC’06); IEEE, 2006 - 11/

17/2006, p 43.

(38) Harder, E.; Damm, W.; Maple, J.; Wu, C.; Reboul, M.; Xiang, J.

Y.; Wang, L.; Lupyan, D.; Dahlgren, M. K.; Knight, J. L.; Kaus, J. W.;

Cerutti, D. S.; Krilov, G.; Jorgensen, W. L.; Abel, R.; Friesner, R. A.

OPLS3: a force field providing broad coverage of drug-like small

molecules and proteins. J. Chem. Theory Comput. 2016, 12, 281−296.

(39) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R.

W.; Klein, M. L. Comparison of simple potential functions for

simulating liquid water. J. Chem. Phys. 1983, 79, 926−935.

(40) Abel, R.; Young, T.; Farid, R.; Berne, B. J.; Friesner, R. A. Role

of the active-site solvent in the thermodynamics of factor Xa ligand

binding. J. Am. Chem. Soc. 2008, 130, 2817−2831.

̈

(41) Schrodinger LLC. The PyMOL Molecular Graphics System

version 2.0, 2015.

(42) Matthess, Y.; Raab, M.; Knecht, R.; Becker, S.; Strebhardt, K.

Sequential Cdk1 and Plk1 phosphorylation of caspase-8 triggers

apoptotic cell death during mitosis. Mol. Oncol. 2014, 8, 596−608.

(43) Kabsch, W. XDS. Acta Crystallogr., Sect. D: Biol. Crystallogr.

2010, 66, 125−132.

(44) Evans, P. R.; Murshudov, G. N. How good are my data and

what is the resolution? Acta Crystallogr., Sect. D: Biol. Crystallogr.

2013, 69, 1204−1214.

(45) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M.

D.; Storoni, L. C.; Read, R. J. Phaser crystallographic software. J. Appl.

Crystallogr. 2007, 40, 658−674.

(46) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;

Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A. G.

W.; McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;

Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson, K. S.

Overview of the CCP4 suite and current developments. Acta

Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 235−242.

(47) Emsley, P.; Cowtan, K. Coot: model-building tools for

molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,

60, 2126−2132.

(48) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of

macromolecular structures by the maximum-likelihood method. Acta

Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240−255.

(49) Liebschner, D.; Afonine, P. V.; Baker, M. L.; Bunkóczi, G.;

Chen, V. B.; Croll, T. I.; Hintze, B.; Hung, L. W.; Jain, S.; McCoy, A.

J.; Moriarty, N. W.; Oeffner, R. D.; Poon, B. K.; Prisant, M. G.; Read,

R. J.; Richardson, J. S.; Richardson, D. C.; Sammito, M. D.; Sobolev,

O. V.; Stockwell, D. H.; Terwilliger, T. C.; Urzhumtsev, A. G.; Videau,

L. L.; Williams, C. J.; Adams, P. D. Macromolecular structure

determination using X-rays, neutrons and electrons: recent develop-

ments in Phenix. Acta Crystallogr., Sect. D: Struct. Biol. 2019, 75, 861−

877.

(28) Metz, K. S.; Deoudes, E. M.; Berginski, M. E.; Jimenez-Ruiz, I.;

Aksoy, B. A.; Hammerbacher, J.; Gomez, S. M.; Phanstiel, D. H.

Coral: clear and customizable visualization of human kinome data.

Cell Syst. 2018, 7, 347−350.

(29) Hanke, T.; Dehm, F.; Liening, S.; Popella, S.-D.; Maczewsky, J.;

Pillong, M.; Kunze, J.; Weinigel, C.; Barz, D.; Kaiser, A.; Wurglics, M.;

Lämmerhofer, M.; Schneider, G.; Sautebin, L.; Schubert-Zsilavecz,

M.; Werz, O. Aminothiazole-featured pirinixic acid derivatives as dual

5-lipoxygenase and microsomal prostaglandin E2 synthase-1 inhibitors

with improved potency and efficiency in vivo. J. Med. Chem. 2013, 56,

9031−9044.

(30) Madhavi Sastry, G.; Adzhigirey, M.; Day, T.; Annabhimoju, R.;

Sherman, W. Protein and ligand preparation: parameters, protocols,

and influence on virtual screening enrichments. J. Comput.-Aided Mol.

Des. 2013, 27, 221−234.

(31) Jacobson, M. P.; Pincus, D. L.; Rapp, C. S.; Day, T. J. F.; Honig,

B.; Shaw, D. E.; Friesner, R. A. A hierarchical approach to all-atom

protein loop prediction. Proteins 2004, 55, 351−367.

(32) Sherman, W.; Day, T.; Jacobson, M. P.; Friesner, R. A.; Farid,

R. Novel procedure for modeling ligand/receptor induced fit effects. J.

Med. Chem. 2006, 49, 534−553.

(33) Shelley, J. C.; Cholleti, A.; Frye, L. L.; Greenwood, J. R.;

Timlin, M. R.; Uchimaya, M. Epik: a software program for pK(a )

prediction and protonation state generation for drug-like molecules. J.

Comput.-Aided Mol. Des. 2007, 21, 681−691.

(34) Roos, K.; Wu, C.; Damm, W.; Reboul, M.; Stevenson, J. M.; Lu,

C.; Dahlgren, M. K.; Mondal, S.; Chen, W.; Wang, L.; Abel, R.;

Friesner, R. A.; Harder, E. D. OPLS3e: Extending force field coverage

for drug-like small molecules. J. Chem. Theory Comput. 2019, 15,

1863−1874.

(35) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;

Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;

Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: a new

approach for rapid, accurate docking and scoring. 1. method and

assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739−1749.

(36) Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.;

Frye, L. L.; Pollard, W. T.; Banks, J. L. Glide: a new approach for

rapid, accurate docking and scoring. 2. enrichment factors in database

screening. J. Med. Chem. 2004, 47, 1750−1759.

(37) Bowers, K. J.; Chow, D. E.; Xu, H.; Dror, R. O.; Eastwood, M.

P.; Gregersen, B. A.; Klepeis, J. L.; Kolossvary, I.; Moraes, M. A.;

8160

https://doi.org/10.1021/acs.jmedchem.0c02144

J. Med. Chem. 2021, 64, 8142−8160

Article Doi

Structure-Based Design of Selective Salt-Inducible Kinase Inhibitors

DOI: 10.1021/acs.jmedchem.0c02144

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.0c02144

Full text of DOI:10.1021/acs.jmedchem.0c02144

Products guided by the article

R&D Labs maybe for 1261606-45-4

Relevant to this article

Hot Product