Phthalazinones 2: Optimisation and synthesis of novel potent inhibitors of poly(ADP-ribose)polymerase

Article abstract of DOI:10.1016/j.bmcl.2005.10.081

We have previously described the discovery of poly(ADP-ribose)polymerase-1 (PARP-1) inhibitors based on a phthalazinone scaffold. Subsequent optimisation of inhibitory activity, metabolic stability and pharmacokinetic parameters has led to a novel series of meta-substituted 4-benzyl-2H-phthalazin-1-one PARP-1 inhibitors which retain low nM cellular activity and show good stability in vivo and efficacy in cell based models.

Full text of DOI:10.1016/j.bmcl.2005.10.081

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1040–1044
Phthalazinones 2: Optimisation and synthesis of novel
potent inhibitors of poly(ADP-ribose)polymerase
Xiao-ling Cockcroft,a,* Krystyna J. Dillon,a Lesley Dixon,b Jan Drzewiecki,a
Frank Kerrigan,b Vincent M. Loh, Jr.,a Niall M. B. Martin,a
Keith A. Meneara,* and Graeme C. M. Smitha
aKuDOS Pharmaceuticals Ltd 327 Cambridge Science Park, Milton Road, Cambridge, CB4 0WG, UK
bMaybridge, Trevillett, Tintagel, Cornwall, PL34 0HW, UK
Received 5 October 2005; revised 19 October 2005; accepted 22 October 2005
Available online 15 November 2005
Abstract—We have previously described the discovery of poly(ADP-ribose)polymerase-1 (PARP-1) inhibitors based on a phthala-
zinone scaffold. Subsequent optimisation of inhibitory activity, metabolic stability and pharmacokinetic parameters has led to a
novel series of meta-substituted 4-benzyl-2H-phthalazin-1-one PARP-1 inhibitors which retain low nM cellular activity and show
good stability in vivo and efficacy in cell based models.
Ó 2005 Elsevier Ltd. All rights reserved.
Poly(ADP-ribose)polymerase-1 (PARP-1) is an abun-
dant nuclear enzyme that binds to and is activated by
DNA single and double strand breaks. PARP-1 activa-
tion represents one of the earliest cell responses to
DNA damage leading to DNA repair via the base-exci-
sion repair (BER) pathway.1 Utilising nicotinamide ade-
nine dinucleotide (NAD+) as substrate, PARP-1
catalyses the formation of ADP-ribose homopolymers
onto PARP-1 and other protein acceptors. Due to its
central role in a number of cellular processes PARP-1
has been implicated in several important disease states
including cerebro-vascular disease, reperfusion injury
and cancer resistance.2,3
the homologous recombination DNA repair pathway,
were seen to be extremely sensitive to the effects of
PARP inhibition, leading to selective cell death.5,6
These results indicate that for cancer treatment PARP
inhibitors, whether as agents for combination with
In tumour biology inhibition of PARP-1 activity has
been shown to enhance the effects of radiation and che-
motherapeutic agents by suppressing the repair of
potentially lethal damage in cancerous cells.4 More re-
cent studies have indicated that in certain genetic back-
grounds tumour cells may be sensitive to the effects of
PARP inhibition without the need for the presence of
cytotoxic chemotherapeutics.5,6 For example, cells that
showed loss of BRCA1 and BRCA2, components of
Figure 1. The overlay of the proposed docking model of compound 1
and 2 (shown in cyan and magenta, respectively) in the catalytic site of
PARP-1. The protein is shown as a solid surface with Gly863, Met890
and Tyr907 through a transparent surface. Key hydrogen bonding
interactions are indicated by green dotted lines.
Keywords: PARP-1 inhibitor; Phthalazinone; DNA repair; Anti-can-
cer; BRCA2-deficient cell.
*
Corresponding authors. Tel.: +44 1403 248844; e-mail addresses:
xcockcroft@kudospharma.co.uk; kmenear@kudospgarma.co.uk
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.10.081

X. Cockcroft et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1040–1044
1041
Table 1. Structure–activity relationships of the meta-substituted 4-benzyl-2H-phthalazin-1-one and the isonipecotate series
O
O
NH
N
NH
N
O
R1
N
R2
R1
IC50 (nM)
PF50
R2
IC50 (nM)
PF50
a
O
O
*
1
2
20
2.3
2.1
9
9
2.1
4.4
N
*
N
H
O
O
N
50
10
8
N
*
H
*
*
O
O
3
4
46
40
2.2
1.2
11
12
15
8
2.5
2.8
N
N
N
H
*
O
O
N
O
*
N
H
*
S
O
O
N
N
N
O
5
6
8
3.4
2.1
13
14
16
15
4.1
3.9
O
*
*
O
*
O
O
N
N
N
15
O
*
O
O
NH
7
8
9
2.4
3.7
15
16
9
2.6
2.0
N
*
O
O
NH
16
16
N
H
*
*
a The PF50 value is the potentiation factor, calculated as the ratio of the IC50 growth curve for the alkylating agent, methyl methanesulfonate (MMS)
divided by the IC50 of the curve of MMS + PARP inhibitor.
O
O
O
CN
CN
i
H
H
O
+
+
iii
O
35a
O
34
36
NH
N
O
O
O
O
X
iv
CN
F
ii
O
O
OH
CN
OMe
OMe
P
6a X=H
6b X=F
O
37
F
35b
38
Scheme 1. Synthesis route for phthalazinone acid intermediates. Reagents and conditions: (i) NaOCH3, CH3OH, reflux; (ii) NEt3, THF; (iii)
NH2NH2ÆH2O, reflux; (iv) a—aq NaOH, THF; b—NH2NH2ÆH2O, reflux.

1042
X. Cockcroft et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1040–1044
Table 2. Structure–activity relationships of the homopiperazine series
Table 2 (continued)
Compound
R3
X
IC50 (nM)
PF50
O
30
31
Et
nPr
F
F
7
5
6.7
7.0
NH
N
O
32
33
F
F
5
5
6.3
6.7
N
R3
N
*
*
X
Compound R3
X
H
IC50 (nM) PF50
O
17
8
6
2.1
3.2
chemotherapies or as potential monotherapies, could be
of significant therapeutic importance.
S
CF
3
O
*
We have recently reported7 phthalazinone compound 1
(Table 1) as a potent inhibitor of PARP-1 with an
IC50 8 of 20 nM and a cellular PF50 ratio of 2.3. Howev-
er, compound 1 was metabolised rapidly in vitro in
mouse hepatic microsomes with an intrinsic clearance
of 12 ml/min/g.
18
H
O
S
O
*
N
O
O
19
20
H
H
12
8
2.1
2.5
It was postulated that the pendant anilide at the meta-
benzyl position was the most likely point of metabolic
instability. Reversal of the anilide moiety in 1 to give
amide 2 was predicted to retain activity whilst improv-
ing metabolic stability. Docking of 1 and 2 into the
known crystal structure of chicken PARP-1, PDB
code, 1pax,9 utilising Monte-Carlo (MacroModel ver-
sion 7.1)11 provides a proposed binding mode for
the compounds (Fig. 1). The phthalazinone ring of 1
and 2 forms a hydrophobic p-p stacking interaction
with the phenyl moiety of Tyr907. The amido NH
and carbonyl of the phthalazinone ring on both 1
and 2 interact identically with the backbone carbonyl
and NH of Gly863 through bidentate hydrogen bond-
ing, characteristic of this class of inhibitors.9,10,12 The
binding model of the amide and anilide carbonyl on
the meta-benzyl position of 1 and 2 predicts a hydro-
gen bonding interaction with the backbone NH of
Met890, possibly replacing a conserved water molecule
identified in the crystal structure of PARP-1. Indeed,
compound 2 gave similar PARP-1 inhibition but with
a considerably enhanced metabolic stability in vitro in
mouse hepatic microsomes, with an intrinsic clearance
less than 1 ml/min/g. Based on these observations a di-
verse library of amides was synthesized to explore the
SAR further. The synthesis was carried out using par-
allel synthesis methodologies by employing the phtha-
lazinone acid 6a (Scheme 1) as the key functional
intermediate. Preliminary structure–activity explora-
tion showed that a wide range of amines could be
accommodated in this region without losing potency
against the enzyme. More significantly, a number of
compounds from this library showed comparable or
improved cellular activity over and above compound
1, exemplified by entries 1–18 in Table 1. Two further
complementary libraries were subsequently designed
and synthesized. The first library (9–16) utilised the
free acid of 6 coupled to a diverse set of amines.
The second library (17–22, Table 2) further exploited
S
O
*
O
*
O
Cl
*
*
N
21
H
9
2.5
O
O
O
H
N
22
7
H
H
H
H
H
H
H
8
9
2.2
2.4
1.8
2.6
2.5
6.6
3.1
O
H
23
24
25
26
27
Me
Et
23
15
10
6
nPr
*
*
14
28
29
H
F
F
7
9
12.6
5.5
Me
(continued on next page)

X. Cockcroft et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1040–1044
1043
O
O
O
NH
NH
N
NH
N
i
ii
N
O
O
O
N
OH
N
O
H
N
R
O
O
6a
6
Compounds 9-16(Table 1)
O
O
O
NH
NH
N
NH
N
O
N
iii
iv
O
X
O
O
N
N
OH
N
X
X
N
R
O
6a X=H
6b X=F
7a X=H
7b X=F
Compounds 7, 17-33 (Table 2)
Scheme 2. Synthesis route for phthalazinone nipecotate and homopiperazine analogues. Reagents and conditions: (i) HBTU, ethyl isonipecotate,
Hunigs base, DCM; (ii) a—2 M NaOH, THF; b—HBTU, DMA, RHN2; (iii) HBTU, tert-butyl homopiperazine-1-carboxylate, Hunigs base, DMA;
(iv) a—6 M HCl, EtOH; b—RCOCl or RSO2Cl or RNCO, Hunigs base, DCM or RCHO, NaBH(CH3CO2)3, acetic acid, DCM.
7 through coupling with alkyl, acyl, sulfonyl or urea
groups. The SAR developed from these compounds
illustrates that a wide variety of substituents can be
tolerated at the terminal position of the molecule. This
observation can be rationalised if these side chains are
approaching the solvent surface (Fig. 1) and do not
significantly interact with the enzyme. However, it
was noted that hydrophobic residues are preferred at
this position particularly if cellular activity is to be
maintained. Of note is the overall enhancement of
PF50 values, with compound 10 being the most attrac-
tive in the series with a PF50 value of 4.1. Alkylation
of the homopiperazine template 7 provided a series of
compounds (23–27, Table 2), which exhibit good IC50
and PF50 values. Further introduction of fluorine at
the para position of the pendant benzene ring, aiming
to block possible metabolism at this position and ex-
tend half life, resulted in a class of potent inhibitors
with excellent cellular activity (28–33, Table 2). Whilst
potency at the enzyme was maintained with com-
pounds 28–33, the PF50 values were significantly in-
creased. In nearly every case, introduction of a
fluorine at this para position leads to more than a
The synthesis of the key intermediate phthalazinone
acids 6a and 6b is outlined in Scheme 1. For the des-fluo-
rine analogue (6a, X = H) simple nucleophilic addition
of phthalide 34 to the aldehyde 35a in the presence of
sodium methoxide, followed by nitrile hydrolysis and
cyclisation of the indanedione 36 with hydrazine hy-
drate, gave 6a. For fluoro analogue 6b, in which the flu-
oro ortho to the nitrile group is labile to nucleophilic
displacement, the alternative milder route via phospho-
nate 37 was required.
The nipecotate (9–16) and homopiperazine (7, 17–33)
analogues were synthesized in parallel from intermedi-
ates 6a or 6b using the two-step process as outlined in
Scheme 2.
In conclusion, the optimization, synthesis and prelimin-
ary biological evaluation of a novel class of phthalazi-
nones as potent and cellularly active PARP-1
inhibitors have been described. In particular, compound
28 has been characterised as a compound having attrac-
tive pharmacokinetic and physicochemical properties
for iv administration. Further evaluation of this class
of compounds is ongoing and will be reported in due
course.
`
doubling of the PF50 value vis-a-vis the des-fluoro
analogues.
Compound 28 was further characterised in an in vivo
pharmacokinetic model (rat, 10 mg/kg, iv) where it dem-
onstrated a T1/2 of 60 min, clearance 54 ml/min/kg and a
volume of distribution of 2.1 L/kg. The aqueous solubil-
ity of compound 28 at pH 7.4 buffer was 5.1 mg/ml.
Notably, 28 is exquisitely potent in specific killing of
cells defective in the tumour suppressor genes BRCA1
or BRCA2. The SF50 (dosage at which 50% of cell sur-
vived) of 28 is 15 nM for BRCA2-deficient cells and
2 lM for wild type cells.6
References and notes
1. Satoh, M.; Lindahl, T. Nature 1992, 356, 356.
2. Szabo, S.; Zingarelli, B.; OꢀConnor, M.; Salzman, A. L.
J. Clin. Invest. 1997, 100, 723.
3. Cantoni, O.; Cattbeni, F.; Stocchi, V.; Meyn, R. E.;
Cerrutti, P.; Murray, D. Biochem. Biophys. Acta 1989,
1014, 1.
4. Curtin, N. J. Expert Rev. Mol. Med. 2005, 7, 1.

1044
X. Cockcroft et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1040–1044
J.; Whittle, A. J. Bioorg. Med. Chem. Lett. 2005, 15,
5. Bryant, H. E.; Schultz, N.; Thomas, H. D.; Parker, K. M.;
Flower, D.; Lopez, E.; Kyle, S.; Meouth, M.; Curtin, N.
J.; Helleday, T. Nature 2005, 434, 913.
6. Farmer, H.; McCabe, N.; Lord, C. J.; Tutt, A. N. J.;
Johnson, D. A.; Richardson, T. B.; Santarosa, M.; Dillon,
K. J.; Hickson, I.; Knights, C.; Martin, N. M. B.; Jackson,
S. P.; Smith, G. C. M.; Ashworth, A. Nature 2005, 434,
917.
2235.
8. Dillon, K. J.; Smith, G. S.; Martin, N. M. B. J. Biomol.
Screen. 2003, 8, 347.
9. Ruf, A.; de Murcia, J. M.; de Murcia, G.; Schulz, G. E.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 7481.
10. Ruf, A.; de Murcia, G.; Schulz, G. E. Biochemistry 1998,
37, 3893.
7. Loh, V. M.; Cockcroft, X.; Dillon, K. J.; Dixon, L.;
Drzewiecki, J.; Eversley, P. J.; Gomez, S.; Hoare, J.;
Kerrigan, F.; Matthews, I. T. W.; Menear, K. A.; Martin,
N. M. B.; Newton, R. F.; Paul, J.; Smith, G. C. M.; Vile,
11. MacroModel version 7.1, Schro¨dinger Inc. 1999–2000.;
12. White, A. W.; Almasy, R.; Calvert, H. A.; Curtin, N. J.;
Griffin, R. J.; Hostomsky, Z.; Maegley, K.; Newell, D. R.;
Golding, B. T. J. Med. Chem. 2000, 43, 4084.