Development of Pyridazinone Chemotypes Targeting the PDEδ Prenyl Binding Site

Article abstract of DOI:10.1002/chem.201603222

The K-Ras GTPase is a major target in anticancer drug discovery. However, direct interference with signaling by K-Ras has not led to clinically useful drugs yet. Correct localization and signaling by farnesylated K-Ras is regulated by the prenyl binding protein PDEδ. Interfering with binding of PDEδ to K-Ras by means of small molecules provides a novel opportunity to suppress oncogenic signaling. Here we describe the identification and structure-guided development of novel K-Ras–PDEδ inhibitor chemotypes based on pyrrolopyridazinones and pyrazolopyridazinones that bind to the farnesyl binding pocket of PDEδ with low nanomolar affinity. We delineate the structure–property relationship and in vivo pharmacokinetic (PK) and toxicokinetic (Tox) studies for pyrazolopyridazinone-based K-Ras–PDEδ inhibitors. These findings may inspire novel drug discovery efforts aimed at the development of drugs targeting oncogenic Ras.

Full text of DOI:10.1002/chem.201603222

DOI: 10.1002/chem.201603222
Full Paper
&
Chemical Biology
Development of Pyridazinone Chemotypes Targeting the PDEd
Prenyl Binding Site
Sandip Murarka+,[a] Pablo Martꢀn-Gago+,[a] Carsten Schultz-Fademrecht,[b] Alaa Al Saabi,[b]
Matthias Baumann,[b] Eyad Fansa,[c] Shehab Ismail,[d] Peter Nussbaumer,[b]
Alfred Wittinghofer,[c] and Herbert Waldmann*[a, e]
Abstract: The K-Ras GTPase is a major target in anticancer
drug discovery. However, direct interference with signaling
by K-Ras has not led to clinically useful drugs yet. Correct lo-
calization and signaling by farnesylated K-Ras is regulated by
the prenyl binding protein PDEd. Interfering with binding of
PDEd to K-Ras by means of small molecules provides a novel
opportunity to suppress oncogenic signaling. Here we de-
scribe the identification and structure-guided development
of novel K-Ras–PDEd inhibitor chemotypes based on pyrrolo-
pyridazinones and pyrazolopyridazinones that bind to the
farnesyl binding pocket of PDEd with low nanomolar affinity.
We delineate the structure–property relationship and in vivo
pharmacokinetic (PK) and toxicokinetic (Tox) studies for pyra-
zolopyridazinone-based K-Ras–PDEd inhibitors. These find-
ings may inspire novel drug discovery efforts aimed at the
development of drugs targeting oncogenic Ras.
Introduction
target the G12C mutation in K-Ras which occurs with high fre-
quency in lung cancer.[13–15]
Ras proteins are key regulators of proliferation and differentia-
tion[1,2] and are mutated in about 20–30% of human cancers,
with the K-Ras isoform most frequently mutated, for example,
in colon and pancreatic cancers.[3,4] Inhibition of signal trans-
duction through the Ras-Map kinase pathway in general is
considered a viable strategy for anticancer drug discovery.[5,6]
However, despite substantial efforts, interference with the
signal-transducing activity of the Ras proteins and their correct
localization, in particular by means of farnesyltransferase inhibi-
tors (FTIs), has not led to clinically useful drugs yet.[7–9] Recent
attempts to target the Ras protein directly yielded compounds
with submillimolar affinity that were inhibitors of the Ras-Sos
interaction,[10–12] and covalent inhibitors of Ras signaling that
The biological function of Ras proteins critically depends on
their correct subcellular localization, which in turn is regulated
by lipid modifications at their C terminus.[16,17] K-Ras, the major
oncogenic isoform of the Ras proteins carries a C-terminal S-
farnesylated cysteine methyl ester. Recently it was demonstrat-
ed that the prenyl binding protein PDEd binds K-Ras and the
Ras homologue RheB through the farnesylated carboxymethy-
lated C terminus,[18–20] sustains their spatial organization and
proper localization in cells, and critically regulates signaling of
oncogenic K-Ras.[19,21] Along these lines, we showed that inter-
ference with the K-Ras–PDEd interaction with the small mole-
cule inhibitor Deltarasin 1a (Figure 1) inhibits oncogenic Ras
signaling in cells and blocks proliferation of human pancreatic
ductal adenocarcinoma (hPDAC) cells harboring oncogenic K-
Ras in vitro and in vivo.[22]
[a] Dr. S. Murarka,+ Dr. P. Martꢀn-Gago,+ Prof. Dr. H. Waldmann
Max-Planck-Institut fꢁr Molekulare Physiologie
Abteilung Chemische Biologie, Otto-Hahn-Straße 11
44227 Dortmund (Germany)
However, despite the fact that Deltarasin binds the prenyl
binding pocket of PDEd with nanomolar affinity in vitro,
growth inhibition requires micromolar concentrations of the
compound, raising the question whether this observation is
chemotype-dependent or rather reflects the influence of an
underlying biological mechanism, which would be of utmost
importance for any further inhibitor development. To this end,
we recently documented the discovery of pyrazolopyridazi-
none-based, highly selective PDEd inhibitors that exhibit less
unspecific cytotoxicity than the previously reported Deltarasin
1a and demonstrate a high correlation with the phenotypic
effect of PDEd knockdown in a set of human pancreatic cancer
cell lines.[23]
E-mail: herbert.waldmann@mpi-dortmund.mpg.de
[b] Dr. C. Schultz-Fademrecht, Dr. A. Al Saabi, Dr. M. Baumann,
Dr. P. Nussbaumer
Lead Discovery Center GmbH, 44227 Dortmund (Germany)
[c] Dr. E. Fansa, Prof. Dr. A. Wittinghofer
Max-Planck-Institut fꢁr Molekulare Physiologie
Structural Biology Group, 44227 Dortmund (Germany)
[d] Dr. S. Ismail
Beatson Institute for Cancer Research, Bearsden, Glasgow G61 1BD (UK)
[e] Prof. Dr. H. Waldmann
Technische Universitꢂt Dortmund, Fakultꢂt Chemie
Chemische Biologie, Otto-Hahn-Straße 6, 44221 Dortmund (Germany)
[+] These authors contributed equally to this work.
Here we describe the identification and structure-guided de-
velopment of novel K-Ras–PDEd inhibitor chemotypes featur-
ing pyrrolopyridazinones and pyrazolopyridazinones that bind
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201603222.
Chem. Eur. J. 2016, 22, 1 – 12
1
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 1. Bisbenzimidazole-based inhibitors 1a (Deltarasin) and 1b and novel inhibitor chemotypes pyrrolopyridazinone 2 and pyrazolopyridazinone 3 identi-
fied through alpha screen.
to the farnesyl-binding hydrophobic pocket of PDEd with low
nanomolar affinity. We detail the synthesis of these PDEd inhib-
itor classes, and report the structure–property relationship
study for pyrazolopyridazinone-based K-Ras–PDEd inhibitors
with a view to identify a suitable candidate for in vivo studies.
We report in vivo pharmacokinetic (PK) and toxicokinetic (Tox)
studies using a set of inhibitors based on the pyrazolopyridazi-
none scaffold.
(3-dimethylaminopropyl)carbodiimide (EDC) as the coupling re-
agent.
The binding efficacy of the synthesized analogues was eval-
uated by means of a fluorescence polarization assay based on
a known PDEd ligand. Our results show that introduction of
a methyl substituent on the phenyl ring of the amide function-
ality leads to a two- to threefold increase of binding potency
(Table 1, entries 1–3). Whereas, ortho- and para-substitution im-
parted similar effects, meta-dimethyl substituted anilides
proved to be slightly less effective (entries 2–4). Electron-with-
drawing substituents on the phenyl ring showed marginal ef-
fects as compared to unsubstituted or methyl-substituted ana-
logues (entries 9–11). Replacement of the methyl substituent
with an electron-rich methoxy functionality yielded much
higher KD values irrespective of the substitution pattern. The
additional increase of electron density turned out to be detri-
mental (entries 5–8). Importantly, a benzyl amide demonstrated
a similar level of potency as compared to methyl-substituted
anilide derivatives (entry 12). We then sought to explore the ef-
ficacy of biosteric ester analogues as they can impart interest-
ing effects owing to their rigidity, different hydrogen-bonding
properties and conformational orientation. None of the tested
phenyl ester analogues showed potency higher than the corre-
sponding amide congener (entries 13–16) and by analogy
a benzyl ester showed potency similar to methyl-substituted
phenyl ester derivatives (entry 17). Considering the fact that
the benzyl amide showed slightly higher binding affinity and
since they are stable to cleavage by hydrolases in prospective
cellular investigations we decided to further investigate this
chemotype.
Results and Discussion
Our previously developed biophysical alpha screen assay[22]
was employed to search for alternative chemotypes that can
be evolved into individual inhibitors with a potency similar to
Deltarasin 1a. The screen revealed pyrrolopyridazinone 2 and
pyrazolopyridazinone 3 as inhibitors of the K-Ras–PDEd interac-
tion (Figure 1). High inhibitory potency in vitro was independ-
ently confirmed by means of a fluorescence polarization assay
as described previously.[22]
Development of pyrrolopyridazinone inhibitors
Since the initial hit based on the pyrrolopyridazinone scaffold
embodied an anilide functionality, we decided to prepare a col-
lection of structurally and electronically diverse anilide deriva-
tives and their corresponding bioisosteric ester analogues. The
targeted pyrrolopyridazinone-based amides and esters were
retrosynthetically disconnected to the carboxylic acid precursor
4 and corresponding amines or alcohols, which can be linked
by standard coupling methods (Scheme 1). The synthesis of
acid fragment 4 commenced with commercially available 5,
which was treated with 6 in the presence of sodium hydride
to afford tricarbonyl derivative 7 in 45% yield (Scheme 1). Sub-
sequent heating of 7 and 4-aminobutanoic acid in acetic acid
under microwave conditions at 1208C yielded pyrrole deriva-
tive 8 in 87% yield. The free acid in 8 was then esterified by
using thionyl chloride and MeOH yielding 9 in 97% yield. Sub-
sequently, ester 9 was subjected to Vilsmeier–Haack formyla-
tion to give aldehyde 10, which was then condensed with
phenyl hydrazine in acetic acid to form pyridazinone 12 in
93% yield. Finally, saponification of 12 gave the final acid pre-
cursor 4 in excellent yield. With acid 4 in hand, a range of
amides (2, 13–22, 40) and esters (24–28) were synthesized in
good yields (62–99%) at ambient temperature using 1-ethyl-3-
Since initial attempts to obtain a crystal structure of pyrrolo-
pyridazinone 23 and close analogues in complex with PDEd
failed, molecular modeling (Schrçdinger, Maestro suite) was
employed to predict the binding mode of 23 for optimization
of binding and potency. Our docking studies suggest that 23
binds deeply in the hydrophobic tunnel of PDEd with H-bond-
ing to Tyr149 and Arg61 (Figure 2). This binding mode closely
resembles the interactions of Deltarasin analogue 1b with
PDEd (Figure 2).[22]
Based on these results and since modeling also suggested
that substituents on the N-phenyl ring would clash with the
protein, the length of the linker between the two hydrogen-
bonding groups and the structure of the amide were varied.
Determination of KD values in fluorescence polarization assays
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
2
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 1. Synthesis of pyrrolopyridazinone-based amide (2, 13–22, 40) and ester (24–28) derivatives.
reducing the linker length drastically diminished the binding
affinity (compare entries 2 vs. 6 and 4 vs. 7). Shortening of the
linker does probably not allow H-bond formation between the
amide carbonyl and Tyr149. Replacement of the benzyl amide
phenyl ring by different electron-rich or electron-poor hetero-
cycles led to substantially higher KD values (entries 8–10). How-
ever, cycloalkylamides with alkyl groups of similar size as the
phenyl ring were potent inhibitors (Table 1, entries 13–16).
Substituted cylohexyl amide derived inhibitors bind to PDEd
with low nanomolar affinity (entries 13–16) presumably due to
hydrophobic interactions which increase with the ring size.
In an attempt to further improve the drug-like properties
and explore an isosteric[24] replacement of the amide group,
a series of 1,2,4- and 1,3,4-oxadiazoles 45–50 and 52–58 was
investigated (Scheme 2). The two-step one-pot synthesis of
1,2,4-oxadiazoles 45–50 was accomplished in good yields (55–
60%) by treating acid precursor 4 with the corresponding ami-
doxime derivatives 44 in the presence of EDC·HCl in diglyme,
followed by heating to 1108C for the cyclodehydration step .[25]
Similarly, 1,3,4-oxadiazoles 52–58 were obtained in 58–80%
yields by treatment of 4 with benzoylhydrazides 51 in the
presence of N,N,N’,N’-tetramethyl-O-(1H-benzotriazol-1-yl)uroni-
um hexafluorophosphate (HBTU) and subsequent ring closure
Figure 2. Comparison between the binding mode of inhibitors 23 and 1b.
a) Modeling (Schrçdinger, Maestro suite) of 23. b) Crystal structure of inhibi-
tor 1b in complex with PDEd showing H-bonding interactions with the
same amino acids, Tyr149 and Arg61, respectively, as predicted for inhibitor
23. A graphic of 1b in complex with PDEd highlighting the surface of the
pocket can be found in the Supporting Information, Figure S4.
revealed that introduction of a chlorine atom into the phenyl
ring results in an increased potency whereas a methoxy sub-
stituent is less advantageous (Table 2, entries 2–5). Importantly,
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
3
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 1. KD values for pyrrolopyridazinone-based anilides and esters as determined by fluorescence polarization experiments.
Entry
1
Nr.
X
R
KD [nm]
Entry
10
Nr.
X
R
KD [nm]
83Æ24
2
NH
189Æ40
21
NH
2
3
13
14
NH
NH
74Æ19
74Æ17
11
12
22
23
NH
NH
96Æ31
71Æ19
4
5
6
7
15
16
17
18
NH
NH
NH
NH
134Æ34
243Æ43
282Æ52
93Æ21
13
14
15
16
24
25
26
27
O
O
O
O
242Æ35
163Æ24
265Æ28
614Æ70
8
9
19
20
NH
NH
>103
17
28
O
250Æ51
133Æ30
Table 2. KD values for pyrrolopyridazinones with different linker and amide structure as determined by fluorescence polarization measurements.
Entry
1
Nr.
X
R
KD [nm]
71Æ9
Entry
9
Nr.
X
R
KD [nm]
23
À(CH2)3À
36
À(CH2)3À
428Æ80
2
3
29
30
À(CH2)3À
À(CH2)3À
119 Æ30
41Æ5
10
11
37
38
À(CH2)3À
À(CH2)3À
815Æ118
634Æ56
4
5
6
31
32
33
À(CH2)3À
À(CH2)3À
ÀCH2À
32Æ11
58Æ13
>103
12
13
14
39
40
41
À(CH2)3À
À(CH2)3À
À(CH2)3À
102Æ19
23Æ3
29Æ5
7
8
34
35
ÀCH(Me)À
À(CH2)3À
>103
15
16
42
43
À(CH2)3À
À(CH2)3À
22Æ4
15Æ6
132Æ9
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
4
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 2. Synthesis of 1,2,4- (45–50) and 1,3,4- (52–58) oxadiazole derivatives.
mediated by 4-methylbenzenesulfonyl chloride (TsCl) and diiso-
propylethylamine (DIPEA).[26]
KD value determination for these heterocycles revealed that
none of these isosteres reached the potency of the guiding
amides (Table 3). However, we were able to co-crystallize the
1,3,4-oxadiazole 52 in complex with PDEd (2.1 ꢂ resolution,
Figure 3a and c). Close analysis revealed that binding is medi-
ated by hydrophobic interactions and two H-bonds between
the carbonyl group of the pyridazinone and Arg61, and be-
tween the oxadiazole and Tyr149. These results validate the
guiding docking experiments described above. Further design
based on the co-crystal structure of 52 with PDEd suggested
the additional filling of the empty space in the hydrophobic
pocket of PDEd (Figure 3d). To meet this goal and in order to
retain the H-bond to Tyr149, N-alkylated benzimidazole 61 was
introduced (Figure 3b and e). Benzimidazole fragment 60 was
synthesized by a condensation reaction between acid 4 and
commercially available phenylenediamine 59 using polyphos-
phoric acid (Scheme 3). The intermediate 60 was either N-alky-
lated by using benzylbromide to obtain 61 or was modified by
Mitsunobu reaction using n-tributylphosphine and tetramethy-
lazadicarboxamide (TMAD) to deliver thiophenylated fragment
62 (Scheme 3).
Table 3. KD values for 1,2,4- (45–50) and 1,3,4- (52–58) oxadiazole deriva-
tives.
Entry
Nr.
45
46
47
Ar
KD [nm]
758Æ155
1
2
3
1036Æ212
1072Æ249
4
48
3039Æ964
5
6
7
8
9
49
50
52
53
54
1389Æ374
830Æ249
1068Æ139
1372Æ273
1368Æ307
Gratifyingly, the newly designed chemotypes, 61 and 62,
emerged as highly potent inhibitors that bind to PDEd with
very low nanomolar affinities (KD <5 nm), supporting the struc-
tural model (Figure 3e).
10
55
550Æ53
11
12
13
56
57
58
1072Æ233
947Æ182
2326Æ682
Development of pyrazolopyridazinone inhibitors
Structure-based development of the pyrazolopyridazinone hit
class was enabled by a crystal structure of inhibitor 3 in com-
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
5
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
empty space in the PDEd pocket could be filled, a series of this
chemotype was investigated.
In analogy to the disconnection of the pyrrolopyridazinone
scaffold, amides based on the pyrazolopyridazinone core were
synthesized by coupling acid precursor 72 with the corre-
sponding amine derivatives in the presence of EDC as the cou-
pling reagent (Scheme 4). The acid 72 was synthesized in six
linear steps starting from commercially available aniline deriva-
tives 63. The hydrazonyl chloride 66 was synthesized through
the Japp–Klingemann reaction by treating 65 with benzene di-
azonium derivative 64 that was generated in situ from the ani-
line derivative 63 (Scheme 4).[27] The pyrazole core 68 was con-
structed by mixing hydrazonyl chloride 66 with a solution of
enolate 67 at ambient temperature.[27] Following this proce-
dure, refluxing pyrazole 68 with hydrazine in ethanol afforded
pyrazolopyridazinone 69,[26] which was alkylated with methyl
4-bromobutanoate (70) to give 71 in good yield. Subsequently,
71 was hydrolyzed to obtain the corresponding acid precursor
72.
Figure 3. Development of pyrrolopyridazinone-based inhibitors with isosteric
amide group replacements. Structure and binding affinities (competitive
fluorescence polarization assay, see the Supporting Information) of: a) com-
pound 52, and b) compound 61. c) Crystal structure of inhibitor 52 in com-
plex with PDEd at 2.10 ꢂ resolution. d) Surface representation of PDEd and
overlay of oxadiazole 52 showing an empty space in the hydrophobic
tunnel. e) Surface representation of the PDEd structure and fitting of the de-
signed N-benzylated benzimidazole derivative 61 in the hydrophobic tunnel.
Tyr149 (Y149) and Arg61 (R61) are shown as sticks to highlight the hydro-
gen-bond interactions.
Indeed, benzyl amide 74 binds to PDEd with nanomolar af-
finity (Table 4, entry 3). Ring-size reduction (Table 4, entry 2) or
incorporation of a substituent on the phenyl ring was not fa-
vorable (Table 4, entries 4–7). Homobenzyl amides also proved
to be potent inhibitors (entries 8 and 9). Our modeling studies
suggested that a small substituent might be attached to the
N-phenyl moiety pointing deep into the farnesyl binding site.
As we expected, incorporation of a p-methyl substituent into
the pyrazole N-phenyl ring resulted in low nanomolar affinity
(Table 4, entries 2–4 and 7–9 vs. 10, 14–16, 18, 19). Additional
plex with PDEd (Figure 3a and c, 2.6 ꢂ) which revealed that
heterocycle 3 also employs H-bonds to Tyr149 and Arg61 for
binding. Since the binding mode of inhibitor 3 is analogous to
binding of inhibitor 52, a C3-linker was kept for further optimi-
zation of affinity. Considering that anilides are not desired in
medicinal chemistry investigations and since modeling indicat-
ed that benzyl amides could adopt a conformation in which
a simultaneous H-bond with Tyr149 could be formed and the
Scheme 3. Synthesis of N-alkylated benzimidazole derivatives 61 and 62 and their binding affinities as measured by fluorescence polarization.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 4. Synthesis of pyrazolopyridazinone-derived amide (3, 73–96) derivatives.
replacement of the benzyl amide by a smaller furyl-, a tetrahy-
drofuryl group or a pyridine was not favorable (entries 10–13).
Interestingly, para-methyl and para-isopropoxy substituted
benzylamides (entries 16 and 17) as well as several homoben-
zyl amides displayed low nanomolar affinities, regardless of the
substitution pattern on the phenyl backbone (entries 18–22).
Notably, cycloaliphatic amides proved to be potent inhibitors
with the activity dependent on their ring-sizes (entries 23–25).
While very potent molecules were identified, ligand efficien-
cy, solubility, and drug likeness of the compounds still left
room for further optimization.[28] Compound 93 was selected
for further optimization, because we assumed that the pres-
ence of an additional methyl group at the benzylic position in
93 would further stabilize the compound against metabolic
degradation. Furthermore, molecular modeling experiments in-
dicated that 93 will form the same hydrogen bonds as 3 and
may additionally form a hydrogen bond with glutamine-78
(Figure 4d). Based on this reasoning, 93 was subjected to
a panel of biochemical and pharmacological screens in order
to determine its selectivity profile and possible interactions
with other pharmacologically relevant proteins.[23] This com-
pound proved to be remarkably selective as compared to Del-
tarasin and was chosen for further development (Supporting
Information, Figure S1).[23]
Figure 4. Identification of pyrazolopyridazinone-based K-Ras–PDEd inhibitor
chemotype. Structure and binding affinities (competitive fluorescence polari-
zation assay, see the Supporting Information) of: a) pyrazolopyridazinone de-
rivative 3, and b) 93. c) Crystal structure of inhibitor 3 in complex with PDEd
at 2.60 ꢂ resolution. Tyr149 (Y149) and Arg61 (R61) are shown as sticks to
highlight the hydrogen-bond interactions. d) Predicted binding mode of 93
(best docking pose; Schrçdinger, Maestro suite). Besides Y149 and R61,
Gln78 (Q78) is also shown as stick to highlight the probable third H-bond in-
teraction. A graphic of 93 docked into PDEd highlighting the surface of the
pocket can be found in the Supporting Information, Figure S4.
In vitro, 93 revealed an acceptable early ADME (eADME) pro-
file with the exception of high instability in the presence of
mouse liver microsomes (Table 5, entry 1 and Table S1 in the
Supporting Information). Non-quantitative in silico prediction
of the most liable sites of 93 pointed at four metabolic hot-
spots that are likely prone to CYP450-mediated metabolism
(Supporting Information, Figure S2). Consistent with the in
silico prediction, we confirmed experimentally that these posi-
tions are attacked by CYP enzymes in mouse liver microsomes
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
7
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 4. KD values for pyrazolopyridazinones with different amide structures and N-phenyl substituents, as determined by fluorescence polarization meas-
urements.
Entry
1
Nr.
R1
H
R
KD [nm]
5Æ2
Entry
13
Nr.
R1
R
KD [nm]
3
84
CH3
149Æ18
2
3
4
73
74
75
H
H
H
136Æ55
42Æ15
81Æ21
14
15
16
85
86
87
CH3
CH3
CH3
17Æ9
6Æ2
18Æ6
5
6
76
77
H
H
99Æ15
95Æ18
17
18
88
89
CH3
CH3
8Æ3
6Æ2
7
8
78
79
H
H
96Æ14
29Æ7
19
20
90
91
CH3
CH3
5Æ3
6Æ2
9
80
H
19Æ6
21
22
92
93
CH3
CH3
5Æ2
10
11
12
81
82
83
CH3
CH3
CH3
50Æ16
56Æ6
8Æ4[a]
23
24
25
94
95
96
CH3
CH3
CH3
258Æ18
13Æ3
5Æ1
178Æ30
[a] (S)-93 KD =5(Æ3) nm; (R)-93 KD =12(Æ4) nm.
(Supporting Information, Figure S3). These studies also showed
that a variety of additional metabolites were formed by
CYP450-mediated metabolism in vitro. In a next step we per-
formed a structure–property relationship (SPR) study focusing
on improving the metabolic stability of 93 in order to identify
a suitable in vivo candidate.
of the phenyl ring was blocked with a fluorine atom (entry 4).
The metabolic stability of the corresponding compound in-
creased substantially, while similar KD values were obtained
(entry 4). Comparable levels of potency and metabolic stability
were observed when the methyl group on the phenyl 4-posi-
tion was replaced with a chloro functionality for the fluorinat-
ed homobenzyl derivatives (entry 5). Replacement of chloro/
methyl substituents with electron-rich methoxy groups drasti-
cally diminished the metabolic stability of the molecule 103
without altering the binding affinity (entry 6). It is important to
mention that fluorinated benzyl amide analogues containing
a methyl- or chloro- moiety (102, 103) on the phenyl ring
showed high metabolic stability in mouse liver microsomes
(entries 7 and 8). Compound 99 was found to be kinetically
more soluble as compared to 100 and hence was chosen for
in vivo pharmacokinetic (PK) and toxicokinetic (Tox) studies to
eventually find a suitable route of administration and a maxi-
mum tolerated dose to guide efficacy experiments in mouse
models of xenografted K-Ras-dependent and K-Ras-independ-
We first removed the methyl group on the eastern phenyl 4-
position that was identified as one of the metabolic sites for
degradation. The modification proved to be inefficient both in
terms of affinity and clearance, which in turn demonstrates the
importance of the methyl functionality (Table 5, entry 2). Keep-
ing the other functionalities intact, removal of the methyl func-
tionality on the central pyrazole ring drastically improved the
clearance, while potency remained unchanged (entry 3). This
result shows that this site is indeed highly susceptible towards
metabolic degradation and corroborated the in silico predic-
tion. Then we sought to investigate the metabolic hot-spots
on the remaining part of the molecule and accordingly, the
benzylic methyl functionality was removed and the 4-position
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
8
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 5. KD, IC50 value in PaTu8902/Panc1 CTG assay, kinetic solubility and clearance values for pyrazolopyridazinones.
Entry
Nr.
R1
R2
R3
R4
KD [nm]
PaTu8902/Panc1
CTG IC50 [mm]
SolRank
[mm]
Clint, mouse
[mLminÀ1 mgÀ1
]
1
2
93
97
CH3
H
CH3
CH3
CH3
CH3
8Æ4
12.5/>30
49.7
1746
1980
22Æ6
13.5/n.d.
364.1
3
4
5
6
7
8
98
CH3
CH3
Cl
H
H
H
H
H
H
CH3
CH3
CH3
CH3
CH3
CH3
6Æ3
8Æ4
4.7/n.d.
12.3/n.d.
14.3/n.d.
n.d.
83.6
99.9
20.3
21.9
8.7
447
40
99
100
101
102
103
10Æ5
6Æ3
27
H3CO
CH3
Cl
1260
29
129Æ22
125Æ16
n.d.
>30/>30
1.6
10
Attempts to develop a covalent inhibitor
ent PDAC cell lines (see Supporting Information Table S1 for
ADME profile of 99 and 100). Deltarasin showed a kinetic solu-
bility of 270 mm in the same SolRank assay.
Covalent drugs have witnessed resurgence in recent years.[29]
The design of selective covalent inhibitors is conceptually
highly attractive and requires a right balance between reactivi-
ty and selectivity. Selective irreversible covalent binding of an
inhibitor to the target of interest often displays increased bio-
chemical efficiency and high residence time compared to re-
versible binding, and hence the desired pharmacological effect
is achieved at lower drug concentrations. We therefore investi-
gated the possibility of irreversible inhibition of PDEd by form-
ing a covalent bond between a Michael acceptor and a cysteine
residue. Since PDEd features only one cysteine close to the hy-
drophobic farnesyl binding cavity (Cys56, Figure 7a), pyrrolo-
and pyrazolopyridazinone-based reversible inhibitors were
equipped with reactive Michael acceptors (for synthesis see
the Supporting Information).
Different vehicles (A=5% Tween-80, 50% NaCl, 45% H2O;
B=20% DMSO, 80% PEG200; C=15% DMSO, 9.5% Cremo-
phor EL/EtOH (1:1), 75.5% H2O) were used to generate suitable
compound formulations for oral and intraperitoneal (IP)
dosing. Whereas only very low compound levels were found in
plasma after oral dosage (data not shown), significantly higher
plasma concentrations were found after IP administration of
99 in vehicle A, B or C at 10 mgkgÀ1 (Figure 5a). However, at
this dose no major improvement of the plasma concentration/
time curves of 99 compared to 93 was observed. Nevertheless,
as slightly higher compound levels were measured for 99 with
vehicle B during the terminal phase, the DMSO/PEG formula-
tion was chosen for further studies. Surprisingly, when the IP
dose was increased to 30 mgkgÀ1 in DMSO/PEG, serious ad-
verse effects, such as hard breathing, crouching and low blood
pressure, emerged within minutes after the administration
(data not shown). Similar toxic side effects were observed for
93, although previously it was found to be well tolerated at up
to 250 mgkgÀ1 when formulated in vehicle A. Although, IP ap-
plication of 99 at a dose of 30 mgkgÀ1 in vehicle A was well
tolerated, it did not lead to an increase of the plasma exposure
in comparison to 93 (Figure 5b).
Unfortunately, the synthesized inhibitors showed significant-
ly diminished binding affinities compared to their related re-
versible analogues (Table 6). The ineffectiveness of these inhibi-
tors can be explained by considering the orientation of the
thiol function of Cys56. Although PDEd is very flexible, differ-
ent co-crystal structures suggest that the thiol points out of
the pocket, and thus is not exposed to inhibitors in the bind-
ing site (Figure 7b).
Next, we investigated intravenous dosing with formulation
10% DMSO, 90% serum. Indeed, a significant increase of the
plasma exposure as well as the terminal half-life was observed
for 99 in comparison to 93 after intravenous (IV) dosing in
mice (Figure 6).
Conclusions and Outlook
We have described the structure-guided development of novel
K-Ras–PDEd inhibitor chemotypes based on pyrrolopyridazi-
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
9
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 7. a) Orientation of thiol function of Cys56 showing that it points
away from the inhibitor (PDB ID: 5E80). b) Crystal structure of PDEd high-
lighting the Cys56 moiety. The pocket surface is shown in dark grey.
Table 6. Binding affinities of pyrrolo- and pyrazolo-derivatives featuring
Michael-acceptor functionalities.
Entry
1
Nr.
R1
R2
KD [nm]
>103
104
>103
200Æ101
>103
Figure 5. Plasma concentration/time profiles of pyrazolopyridazinone deriva-
tives. Compounds 93 and 99 were administered intraperitoneally (IP) at vari-
ous doses and in different formulations as indicated. Blood samples were
taken over a period of 24 h and analyzed for the total compound concentra-
tions in plasma. All animal work was conducted according to German na-
tional guidelines for animal care. Permissions for the conduct of the animal
experiments were granted by the Regierungsprꢃsident in Tꢄbingen, Germa-
ny.
2
3
4
105
106
107
ther developed for possible in vivo studies employing mouse
models.
Acknowledgements
This research was supported by theDeutsche Forschungsge-
meinschaft (Collaborative Research Center SFB 858) and the
Max Planck Society. The authors are grateful to Axel Choidas
and Bert Klebl, Lead Discovery Center GmbH, as well as Ste-
phan Hahn and Maike Hoffmann, Ruhr University Bochum, for
helpful discussions and experimental support. S.M., P.M.G., E.F.,
S.I., A.W., and H.W. are inventors on a Max-Planck patent appli-
cation concerning PDEd inhibitors.
Figure 6. PK profiling of pyrazolopyridazinones. Compounds 93 and 99 were
administered to NMRI mice IV at a dose of 3 mgkgÀ1 each. Blood samples
were taken over a period of 24 h and analyzed for the total compound con-
centrations in plasma. Left panel: plasma concentration–time profile over
24 h for 93 and 99. Right panel: comparison of PK parameters of 93 and 99
calculated from the measured total plasma concentrations. All animal work
was conducted according to German national guidelines for animal care.
Permissions for the conduct of the animal experiments were granted by the
Regierungsprꢃsident in Tꢄbingen, Germany.
Keywords: k-ras · PDEd · small molecules · structure-based
design · structure–property relationships
none and pyrazolopyridazinone scaffolds, which bind to the
farnesyl binding pocket of PDEd with low nanomolar affinity.
Structure–property relationship, in vivo pharmacokinetic (PK)
and toxicokinetic (Tox) studies on pyrazolopyridazinone-based
K-Ras–PDEd inhibitors revealed a candidate which may be fur-
[1] L. Brunsveld, J. Kuhlmann, K. Alexandrov, A. Wittinghofer, R. S. Goody, H.
Waldmann, Angew. Chem. Int. Ed. 2006, 45, 6622–6646; Angew. Chem.
2006, 118, 6774–6798.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
10
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[2] M. H. Gelb, L. Brunsveld, C. A. Hrycyna, S. Michaelis, F. Tamanoi, W. C.
Van Voorhis, H. Waldmann, Nat. Chem. Biol. 2006, 2, 518–528.
[3] J. L. Bos, Cancer Res. 1989, 49, 4682–4689.
[17] O. Rocks, A. Peyker, M. Kahms, P. J. Verveer, C. Koerner, M. Lumbierres, J.
Kuhlmann, H. Waldmann, A. Wittinghofer, P. I. H. Bastiaens, Science
2005, 307, 1746–1752.
[4] J. Downward, Nat. Rev. Cancer 2003, 3, 11–22.
[18] S. A. Ismail, Y.-X. Chen, A. Rusinova, A. Chandra, M. Bierbaum, L. Gremer,
G. Triola, H. Waldmann, P. I. H. Bastiaens, A. Wittinghofer, Nat. Chem.
Biol. 2011, 7, 942–949.
[19] A. Chandra, H. E. Grecco, V. Pisupati, D. Perera, L. Cassidy, F. Skoulidis,
S. A. Ismail, C. Hedberg, M. Hanzal-Bayer, A. R. Venkitaraman, A. Wit-
tinghofer, P. I. H. Bastiaens, Nat. Cell Biol. 2011, 14, 148–158.
[20] H. Zhang, X. H. Liu, K. Zhang, C.-K. Chen, J. M. Frederick, G. D. Prestwich,
W. Baehr, J. Biol. Chem. 2004, 279, 407–413.
[21] M. Schmick, N. Vartak, B. Papke, M. Kovacevic, D. C. Truxius, L. Rossman-
nek, P. I. H. Bastiaens, Cell 2014, 157, 459–471.
[22] G. Zimmermann, B. Papke, S. Ismail, N. Vartak, A. Chandra, M. Hoffmann,
S. A. Hahn, G. Triola, A. Wittinghofer, P. I. H. Bastiaens, H. Waldmann,
Nature 2013, 497, 638–642.
[23] B. Papke, S. Murarka, H. A. Vogel, P. Martꢀn-Gago, M. Kovacevic, D. C.
Truxius, E. K. Fansa, S. Ismail, G. Zimmermann, K. Heinelt, C. Schultz-Fa-
demrecht, A. Al Saabi, M. Baumann, P. Nussbaumer, A. Wittinghofer, H.
Waldmann, P. I. H. Bastiaens, Nat. Commun. 2016, 7, 11360.
[24] N. A. Meanwell, J. Med. Chem. 2011, 54, 2529–2591.
[25] G.-B. Liang, D. D. Feng, Tetrahedron Lett. 1996, 37, 6627–6630.
[26] P. Stabile, A. Lamonica, A. Ribecai, D. Castoldi, G. Guercio, O. Curcuruto,
Tetrahedron Lett. 2010, 51, 4801–4805.
[5] J. S. Sebolt-Leopold, R. Herrera, Nat. Rev. Cancer 2004, 4, 937–947.
[6] P. J. Roberts, C. J. Der, Oncogene 2007, 26, 3291–3310.
[7] A. D. Cox, S. W. Fesik, A. C. Kimmelman, J. Luo, C. J. Der, Nat. Rev. Drug
Discovery 2014, 13, 828–851.
[8] N. Berndt, A. D. Hamilton, S. M. Sebti, Nat. Rev. Cancer 2011, 11, 775–
791.
[9] D. Vigil, J. Cherfils, K. L. Rossman, C. J. Der, Nat. Rev. Cancer 2010, 10,
842–857.
[10] A. Patgiri, K. K. Yadav, P. S. Arora, D. Bar-Sagi, Nat. Chem. Biol. 2011, 7,
585–587.
[11] T. Maurer, L. S. Garrenton, A. Oh, K. Pitts, D. J. Anderson, N. J. Skelton,
B. P. Fauber, B. Pan, S. Malek, D. Stokoe, M. J. C. Ludlam, K. K. Bowman,
J. Wu, A. M. Giannetti, M. A. Starovasnik, I. Mellman, P. K. Jackson, J. Ru-
dolph, W. Wang, G. Fang, Proc. Natl. Acad. Sci. USA 2012, 109, 5299–
5304.
[12] Q. Sun, J. P. Burke, J. Phan, M. C. Burns, E. T. Olejniczak, A. G. Waterson,
T. Lee, O. W. Rossanese, S. W. Fesik, Angew. Chem. Int. Ed. 2012, 51,
6140–6143; Angew. Chem. 2012, 124, 6244–6247.
[13] J. M. Ostrem, U. Peters, M. L. Sos, J. A. Wells, K. M. Shokat, Nature 2013,
503, 548–551.
[14] M. P. Patricelli, M. R. Janes, L.-S. Li, R. Hansen, U. Peters, L. V. Kessler, Y.
Chen, J. M. Kucharski, J. Feng, T. Ely, J. H. Chen, S. J. Firdaus, A. Babbar,
P. Ren, Y. Liu, Cancer Discov. 2016, 6, 316–329.
[15] S. M. Lim, K. D. Westover, S. B. Ficarro, R. A. Harrison, H. G. Choi, M. E.
Pacold, M. Carrasco, J. Hunter, N. D. Kim, T. Xie, T. Sim, P. A. Jꢃnne, M.
Meyerson, J. A. Marto, J. R. Engen, N. S. Gray, Angew. Chem. Int. Ed.
2014, 53, 199–204; Angew. Chem. 2014, 126, 203–208.
[16] O. Rocks, M. Gerauer, N. Vartak, S. Koch, Z.-P. Huang, M. Pechlivanis, J.
Kuhlmann, L. Brunsveld, A. Chandra, B. Ellinger, H. Waldmann, P. I. H.
Bastiaens, Cell 2010, 141, 458–471.
[27] V. S. Matiichuk, M. A. Potopnyk, N. D. Obushak, Russ. J. Org. Chem. 2008,
44, 1352–1361.
[28] C. Abad-Zapatero, J. T. Metz, Drug Discovery Today 2005, 10, 464–469.
[29] J. Singh, R. C. Petter, T. A. Baillie, A. Whitty, Nat. Rev. Drug Discovery
2011, 10, 307–317.
Received: July 6, 2016
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
11
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Chemical Biology
S. Murarka, P. Martꢀn-Gago,
C. Schultz-Fademrecht, A. Al Saabi,
M. Baumann, E. Fansa, S. Ismail,
P. Nussbaumer, A. Wittinghofer,
H. Waldmann*
&& – &&
Development of Pyridazinone
Chemotypes Targeting the PDEd
Prenyl Binding Site
Reaching deep: The identification and
structure-guided development of novel
K-Ras–PDEd inhibitor chemotypes based
on pyrrolopyridazinones and pyrazolo-
pyridazinones that bind to the farnesyl
binding pocket of PDEd with low nano-
molar affinity is described. The results
may inspire novel drug-discovery efforts
aimed at the development of drugs tar-
geting oncogenic Ras.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
12
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!