Optimization of a series of potent and selective ketone histone deacetylase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.09.003

Histone deacetylase (HDAC) inhibitors offer a promising strategy for cancer therapy and the first generation HDAC inhibitors are currently in the clinic. Herein we describe the optimization of a series of ketone small molecule HDAC inhibitors leading to potent and selective class I HDAC inhibitors with good dog PK.

Full text of DOI:10.1016/j.bmcl.2008.09.003

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5528–5532
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimization of a series of potent and selective ketone histone
deacetylase inhibitors
*
Giovanna Pescatore , Olaf Kinzel, Barbara Attenni, Ottavia Cecchetti, Fabrizio Fiore, Massimiliano Fonsi,
Michael Rowley, Carsten Schultz-Fademrecht , Sergio Serafini, Christian Steinkühler, Philip Jones
IRBM/Merck Research Laboratories, Via Pontina km 30,600, 00040 Pomezia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Histone deacetylase (HDAC) inhibitors offer a promising strategy for cancer therapy and the first gener-
ation HDAC inhibitors are currently in the clinic. Herein we describe the optimization of a series of ketone
small molecule HDAC inhibitors leading to potent and selective class I HDAC inhibitors with good dog PK.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 1 August 2008
Revised 29 August 2008
Accepted 3 September 2008
Available online 6 September 2008
Keywords:
HDAC inhibitors
Alkyl ketone
Zinc binding group
Histone deacetylase (HDAC) and histone acetyltransferase
(HAT) are nuclear enzymes involved in regulating gene expres-
sion.1 The HDAC enzyme family, of which eleven isoforms belong-
ing to classes I, II and IV are known, catalyzes deacetylation of the
tive HDAC 1 and 2 inhibitors such as 5.12 Other classes include
short chain fatty acids like 613 and cyclic peptides such as 814
and 9.15
e
-amino group of lysine residues located near the N-terminus of
O
nucleosome histone proteins. Deacetylated histones acquire a
net positive charge that interacts strongly with the negatively
charged DNA, thereby condensing the chromatin and restricting
accessibility to transcription factors, and ultimately changing
gene expression.2 In cancer the acetylation status of these histone
tails is aberrantly regulated and many recent studies have shown
that HDAC inhibition leads to anticancer effects, as a result of
inhibiting cell growth and inducing apoptosis, as well as causing
differentiation and inhibition of angiogenesis. Hence, HDAC inhi-
bition represents a novel approach to cancer chemotherapy,3 and
indeed vorinostat (1) (ZolinzaÒ, formerly known as SAHA) has
been approved for the treatment of the cutaneous T-cell lym-
phoma (CTCL),4 while several other HDAC inhibitors (HDACi)
along with vorinostat, are in clinical trials showing efficacy in pa-
tient with hematological and solid malignancies.5 Known HDACi’s
O
OH
H
N
N
H
OH
H
N
N
H
O
HN
N
1: Vorinostat (SAHA)
2: LBH-589
O
O
O
O
O
N
H
NH2
H
N
S
OH
OH
S
N
N
H
H
4: MS-275, R = H
5: R = Ph
O
O
3: PXD-101
R
O
N
N
N
NH2
H
N
6: Valproic Acid
H
N
7: MGCD-0103
(Fig. 1) cover
a
wide cross-section of structures6 including:
S
hydroxamic acids (typically broad-spectrum HDACi) such as 1,7
2,8 and 39 as well as aminobenzamides, for instance 410 and 711
together with their bis-aryl derivatives recently disclosed as selec-
O
H
O
O
N
H
O
N
NH
H
HN
O
H
N
NH
H
N
O
N
O
N
H
O
OMe
H
O
O
* Corresponding author. Tel.: +39 0691093278; fax: +39 0691093654.
E-mail address: giovanna_pescatore@merck.com (G. Pescatore).
Present address: NV Organon Schering-Plough Corporation, Medicinal Chemistry
Oss RK2107, PO Box 20, 5340BH Oss, The Netherlands.
O
8: FK-228
9: Apicidin
Figure 1. Known HDAC inhibitor.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.09.003

G. Pescatore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5528–5532
5529
N
S
O
Me
PRO (HeLa)
IC50 (nM)
HDAC1
IC50 (nM)
N
Me
Entry
9
N
H
N
HN
O
O
H
HN
O
O
OMe
9
10
11
44
55
59
290
430
880
NMe
10
11
N
H
16
Figure 2. Development of a series of small molecule alkyl ketone HDACi’s and the identification of lead 11.
Recently we have reported the discovery of a structurally novel
series of HDAC inhibitors based on the natural compound Apici-
din.16 This natural product contains the unusual ethyl ketone as
a zinc binding group (ZBG) yet it is a potent HDAC1 inhibitor
(IC50 = 44 nM) and displays good anti-proliferative activity (HeLa
cells IC50 = 290 nM). A related series of simplified acyclic deriva-
tives was identified and optimized to potent and selective HDAC
inhibitors such as 10 showing good enzymatic activity against
HDAC1 (IC50 = 55 nM) and submicromolar activity in the anti-pro-
liferative assay (HeLa cells IC50 = 430 nM) (Fig. 2). Unfortunately
the heterocyclic acylamino bond was shown to be unstable in ro-
dent plasma, and SAR studies were conducted to identify a replace-
ment to this labile bond by a suitable bioisostere, culminating with
Encouraged by these results we herein describe the further
development of this novel class of HDACi and report our work opti-
mizing enzymatic and anti-proliferative activities, together with
the efforts to address pharmacokinetic liabilities.
Guided by X-ray crystallographic analysis of
a related
hydroxamic acid inhibitor bound to HDAC8,17 SAR investigations
focused primarily on the optimisation of the two surface recogni-
tion domains, the imidazole substituent and the amino capping
group.
The synthesis of these derivatives starts from the readily pre-
pared
a-amino acid derivatives 12, 13 (scheme 1) which were first
alkylated with the suitable
a
-bromo ketone and then cyclized to
the corresponding imidazoles in refluxing xylene in the presence
of a large excess of NH4OAc. Deprotection and standard coupling
conditions yielded either the desired amides 27–51 or the corre-
sponding sulfonamides 21–26. A small exploration of the ZBG
was also conduced starting from the (2S)-8-tert-butoxy-2-[(tert-
butoxycarbonyl)amino]-8-oxooctanoic acid 14. Cyclization as pre-
viously yielded the imidazole intermediate 17 which was fully
deprotected by treatment with TFA. The resulting amino acid 20
was then coupled with 1-methylpiperidine-4-carboxylic acid and
the acid moiety converted into the Weinreb amide. Reaction with
the corresponding Grignard reagent provided the homologated n-
propyl and n-butyl derivatives 52 and 53.
16
the identification of imidazole derivatives exemplified by 11.
Compound 11 displays class I HDAC subtype selectivity and levels
of cellular activity in different cancer cell lines comparable to exist-
ing clinical candidates. Furthermore in an in vivo efficacy study 11
was shown to cause tumour growth inhibition in a colon HCT-116
carcinoma xenograft model comparable to that achieved with vori-
nostat and MS-275; to our knowledge this is the first example of an
unactivated small molecule alkyl ketone HDACi to be efficacious
in vivo. Unfortunately 11 was highly cleared in rats (Cl = 80 mL/
min/kg, Rat microsomes Clint = >300
l
L/min/mg) and showed only
modest oral bioavailability (F = 15%).
O
Br
R1
R2
R2
R2
O
N
N
R
R1
R1
i. Cs2CO3, DMF
TFA, DCM
HO
R
R
N
N
NHBoc
O
H
H
NHBoc
O
NH3+ TFA-
O
quant.
ii. NH4OAc, xylene
Reflux, 2hr, 80%
12: R = CH3,
15: R = CH3
16: R = C2H5
17: R = OC(CH3)3
18: R = CH3
19: R = C2H5
20: R = OH
13: R = C2H5
14: R = OC(CH3)3
18,19
20
O
OH
i.
HOBt, EDC, TEA
N
RACO2H
HOBt, EDC, TEA
40-60%
RASO2Cl, TEA
35-60%
Me
ii. NHCH3OCH3.HCl, HOBt, EDC, TEA
iii. R6MgBr, Et2O, -78˚C
5-35%, 3 steps
R2
R2
R2
N
R1
R6
N
N
R1
R1
N
R
R
H
N
N
HN
O
O
H
H
O
HN
O
O
HN
O
S
O
RA
RA
N
Me
21-26: R = CH3, R1 = Ph, R2 = H
27-36: R = CH3, R1 = Ph, R2 = H
52: R6 = C3H7
37-50: R = CH3, RA = CH(CH2)2-(CH2)2NMe
51: R = C2H5, R1 = Ph, R2 = H
53: R6 = C4H9
RA = CH(CH2)2-(CH2)2NMe
Scheme 1. Synthesis of compounds 21–53.

5530
G. Pescatore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5528–5532
Table 2
These compounds were screened in our primary assays for the
SAR of replacements of the 2-methyl-5methoxyindole group: amine derivatives
ability to inhibit recombinant human HDAC-1 and in 72 h anti-pro-
liferation against cervical HeLa cancer cell line.16
N
Me
Previous work in the bis-amide series based on lead 10 had
shown that the electron rich 2-methyl-5-methoxy indolylaceta-
mide could be replaced by smaller neutral and basic substituents.18
Indeed in the imidazole series these groups offered the optimum
combination of reduced molecular weight, cellular activity and
in vitro stability (Tables 1 and 2).
N
H
HN
O
O
RA
Entry
RA
HDAC 1
PRO (HeLa)
IC50 (nm)
Clint rat microsomes
(ll/min/mg)
a
a
IC50 (nM)
The indolyl group could be replaced by a wide range of substi-
tuted aryl amides and sulfonamides (Table 1) although none of-
fered any significant advantages in terms of activity compared to
lead 11. Little difference in activity was seen between amides
and the corresponding sulfonamides (e.g. 22 and 27). Electron defi-
cient aromatic rings like 22, 24 and 26 appeared to give greater
HDAC1 inhibition compared to their electron rich counterparts like
25 and 28. As previously seen in the bis-amide series one of the
most interesting derivatives was the thiazole moiety 29 which de-
spite slightly reduced potency against HDAC1 (IC50 = 140 nM)
showed better cellular activity than 11, with IC50 = 530 nM on HeLa
cervical cancer cells.
Having seen that none of these neutral derivatives offered sig-
nificant advantages in activity compared to 11, attention focussed
on the incorporation of basic fragments (Table 2). Small cyclic ami-
no groups, such as N-methyl pyrrolidine 30 and N-methyl piperi-
dines 31–33, as well more elaborated amines were evaluated.
Despite having only modest activity in the enzymatic assay, al-
most all these compounds showed little shift between enzyme and
cellular activities, typically 3- to 4-fold compared to 15-fold for 11.
Of the three piperidine analogs, the 2-substituted derivative
(31) was the least active and was rapidly degraded in rat liver
microsomes; the 3-derivative (32) was 3-fold more potent against
HDAC1 (IC50 = 220 nM) but still had only modest metabolic stabil-
ity, while the 4-derivative (33) showed similar enzyme activity and
30
280
710
1100
2700
501
NMe
Me
N
31
>300
NMe
NMe
32
33
34
220
250
500
400
980
154
68
2000
64
N
NMe
N
35
190
220
1300
1700
>300
143
N
36
a
Values are means of two or more experiments.
comparable cell-based activity to 11 (HeLa IC50 = 980 vs 880 nM)
but significantly improved microsomal stability (Clint = 68 L/
min/mg) with respect to both the other piperidine isomers and also
l
Table 1
Preliminary SAR on the 2-methyl-5methoxyindole group
to 11.
N
More sterically demanding amino derivatives were also toler-
ated, such as the [2,2,1]-bicyclohexylamine (36) indicating that
the surface of the enzyme is very spacious in that region.
Based on these results it was decided to use 33 as the lead for a
further round of optimization, this time focusing on the imidazole
substituent (Table 3).
Me
N
H
HN
O
O
X
RA
RA
HDAC 1 IC50 (nM) PRO (HeLa) IC50 (nm)
a
a
Entry
X
C
It was found that by varying the substituent on the imidazole
ring, both enzymatic and anti-proliferative activities could be af-
fected and a number of potent compounds were identified. Substi-
tution in the ortho position was poorly tolerated compared to the
meta and para positions; these latter two positions provided a 2-
to 3-fold boost in enzyme and cellular activities (e.g. 38 and 39
vs 40). Lypophilic substituents such as –Br, –Cl and –Ph (37, 38,
and 42) increased activity, whereas polar substituents such as –F,
–CN (41 and 44) and polar heterocycles such as 2-pyridyl (46) low-
ered the activity. Replacement of the phenyl group with naphthyl
gave compound 43 which showed a dramatic improvement both
in enzymatic and cell-based assays; it was 20-fold more potent
than 33 on HDAC1 and 3-fold more potent in HeLa cells.
Unfortunately the opposite trend in terms of metabolic stability
was observed with the more polar compound being the more met-
abolically stable in rat liver microsomes as observed for 37, 38 and
41. Replacement of the phenyl with polar heterocycles such as the
2-pyridyl group (46) were not tolerated both in enzymatic and
anti-proliferative assays. In contrast the 2-thiophene-derivative
(47) was equipotent on the enzyme although it proved to be 3-fold
less active than 33 in proliferation assay. By increasing the lypophi-
licity, cellular activity could be regained as the benzo-fused ana-
11
59
880
NH
MeO
21
22
23
24
25
SO p-CN-Ph–
SO m-CN-Ph–
SO o-CN-Ph–
SO m-NO2-Ph–
SO p-OMe-Ph–
180
110
270
55
1200
1100
7300
1300
1400
250
F
26
SO
130
1000
NO2
27
28
C
C
m-CN-Ph–
160
300
1100
1100
O
O
S
N
29
C
140
530
a
Values are means of two or more experiments.

G. Pescatore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5528–5532
5531
Table 3
Table 5
SAR of replacements of the phenyl group
PK profile for compound 43
R2
Clint lL/min/mg t1/2 (h) iv Vdss (L/Kg) iv Cl (mL/min/Kg) iv F (%) po
N
Rat
Dog
Hum 12
194
10
2.4
11.8
6.9
5.2
133
7
0
28
R1
N
Me
H
HN
O
O
N
Table 6
Me
SAR on the zinc binding group
Entry
Rt
R2
HDAC 1
IC50 (nM)
PRO (HeLa)
IC50 (nM)
Clint rat microsomes
(ll/min/mg)
N
a
a
R
N
37
38
39
40
41
42
43
44
45
p-Br-Ph
p-Cl-Ph
m-Cl-Ph
o-Cl-Ph
H
H
H
H
H
H
H
H
H
100
75
90
180
380
50
11
120
480
550
680
400
1300
2200
390
300
1500
1540
>300
246
H
HN
O
O
p-F-Ph
99
170
194
330
p-Ph-Ph
2-naphthyl
p-CN-Ph
m-CN-Ph
N
Me
a
a
Entry
R
HDAC IC50 (nM)
3
PRO (HeLa) IC50 (nM)
1
6
N
46
47
H
H
>500
240
9700
3200
43
51
52
53
Me
Et
Pr
11
12
16
74
17
12
27
10
800
1400
3500
300
210
200
690
Bu
180
a
Values are means of two or more experiments.
S
S
48
H
180
1300
to both human and dog plasma protein. When dosed in dog 43, as
expected, showed a good PK profile with a low clearance, a long
half-life and acceptable oral bioavailability (Cl = 7 mL/min/kg,
F = 28%, t1/2 = 11.8 h); however the compound was cleared rapidly
in rat and unsurprisingly no oral bioavailability was found
(Cl = 133 mL/min/kg, F = 0%) (Table 5). Based on intrinsic clearance
43 might have human PK similar to dog.
49
50
tert-Bu
H
Me
>5000
>1500
>25000
14700
Ph
a
Values are means of two or more experiments.
logue 48 displayed IC50 = 1.3 lM in HeLa cells. The addition of
excessive steric bulk, such as tert-butyl (49) produced a significant
loss of activity, as did disubstitution on the imidazole, as 4-methyl-
5-phenyl analogue (50) was 20-fold less active. We hypothesized
this could be due to a steric clash between the two substituents
forcing the phenyl ring to adopt a more out of plane orientation.
Based on its improved potency, 43 was further profiled; it dis-
played excellent inhibitory activity on HDACs 1, 2, 3 and 6 while
no or weak activity was found on the other isoforms (Table 4).
Comparison of 43 with vorinostat and our lead 11 demonstrated
that it was from 3- to 8-fold more potent in the enzyme assay
and had 2-fold improvement in the antiproliferative activity across
a panel of cancer cell lines. 43 was also evaluated in terms of
microsomal stability in three different species: it was found to be
Following the results from our previous studies, a rapid investi-
gation into the ZBG was conducted and the higher homologs of 43
containing the ethyl (51), n-propyl (52) and n-butyl (53) ketone
were profiled (Table 6). The ethyl analogue (51) proved to be the
optimal chain length gaining both enzymatic and cellular activities,
and moreover it displayed only a weak activity on HDAC 6 and con-
sequently good HDACs class I selectivity. The n-propyl derivative
(52) showed equivalent activity in the cell-based assay while a
slightly reduced HDAC activity was observed, still maintaining
class I selectivity. Further elongation of the chain (53) resulted in
an erosion of the inhibitory activity probably as a result of a steric
clash at the bottom of the binding cavity.
In summary we have reported the optimization work of a series
of ketone small molecule HDAC inhibitors which have led to com-
pounds such as 43 and 51 displaying low nanomolar activities
against HDAC 1,2 and 3 and good selectivity over the other class
I and II HDACs. These compounds showed anti-proliferative activ-
ities equivalent to existing clinical candidates despite their lack of
a well established ZBG. Moreover compound 43 demonstrated to
have good PK properties in dogs, with acceptable oral bioavailabil-
ity and a long half-life.
unstable in rat liver microsomes with a Clint of 194
whereas it showed lower turnover in human and dog liver micro-
somes (12 and 10 L/min/mg, respectively).
lL/min/mg
l
The plasma protein binding in the three species was also mea-
sured showing the compound 89% bound to rat protein, 96% bound
Table 4
Activity of compound 43 on HDAC isoforms and various cell line, IC50 (nM)
Entry
1
2
3
6
4,5,7,8 (lM)
Acknowledgment
43
11
11
59
30
29
110
82
17
120
57
10
340
43
>1
>1
>1
Vorinostat(1)
The authors thank Claudio Giuliano, Mark Jairaj, Fabio Bonelli,
Francesca Naimo and Anna Alfieri for their support of this work.
Cervical HeLa
Colon HCT-116
Lung A-549
43
11
300
880
460
590
500
1000
720
6300
1800
References and notes
Vorinostat(1)
1. Brittain, D.; Weinmann, H.; Ottow, E. Annu. Rep. Med. Chem. 2007, 42, 337;
Minucci, S.; Pelicci, P. G. Nat. Rev. Cancer 2006, 6, 38; Rodriquez, M.; Aquino, M.;
Values are means of two or more experiments.

5532
G. Pescatore et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5528–5532
Bruno, I.; De Martino, G.; Taddei, M.; Gomez-Paloma, L. Curr. Med. Chem. 2006,
13, 1119; Jones, P.; Steinkuhler, C. Curr. Pharm. Des. 2008, 14, 545.
2. Johnstone, R. W. Nat. Rev. Drug Discov. 2002, 1, 287.
Liu, J.; Lu, A. H.; Zhou, N. Z.; Robert, M. F.; Gillespie, J.; Wang, J. J.; Ste-Croix, H.;
Rahil, J.; Lefebvre, S.; Moradei, O. M.; Delorme, D.; MacLeod, A. R.; Besterman, J.
M.; Li, Z. Mol. Cancer Ther. 2008, 7, 759.
3. Glozak, M. A.; Seto, E. Oncogene 2007, 26, 5420; . Nat. Rev. Cancer 2006, 6, 38.
4. Grant, S.; Easley, C.; Kirkpatrick, P. Nat. Rev. Drug Discovery 2007, 6, 21.
5. Rasheed, W. K.; Johnstone, R. W.; Prince, H. M. Expert Opin. Invest. Drugs 2007,
16, 659.
6. Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. J. Med. Chem. 2008, 51, 1505.
7. Richon, V. M.; Webb, Y.; Merger, R.; Sheppard, T.; Jursic, B.; Ngo, L.; Civoli, F.;
Breslow, R.; Rifkind, R. A.; Marks, P. A. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5705.
8. Maiso, P.; Carvajal-Vergara, X.; Ocio, E. M.; Lòpez-Pérez, R.; Mateo, G.;
Gutiérrez, N.; Atadja, P.; Pandiella, A.; San Miguel, J. F. Cancer Res. 2006, 66,
5781.
12. (a) Moradei, O. M.; Mallais, T. C.; Frechette, S.; Paquin, I.; Tessier, P. E.; Leit, S.
M.; Fournel, M.; Bonfils, C.; Trachy-Bourget, M.-C.; Liu, J.; Yan, T. P.; Lu, A.-H.;
Rahil, J.; Wang, J.; Lefebvre, S.; Li, Z.; Vaisburg, A. F.; Besterman, J. M. J. Med.
Chem. 2007, 50, 5543; (b) Witter, D. J.; Harrington, P.; Wilson, K. J.; Chenard,
M.; Fleming, J. C.; Haines, B.; Kral, A. M.; Secrist, J. P.; Miller, T. A. Bioorg. Med.
Chem. Lett. 2008, 18, 726.
13. Chen, J. S.; Faller, D. V.; Spanjaard, R. A. Curr. Cancer Drug Targets 2003, 3, 219.
14. Furumai, R.; Matsuyama, A.; Kobashi, N.; Lee, K.-H.; Nishiyama, M.; Nakajima,
H. Cancer Res. 2002, 62, 4916.
15. Han, J.-W.; Ahn, S. H.; Park, S. H.; Wang, S. Y.; Bae, G.-U.; Seo, D.-W. Cancer Res.
2000, 60, 6068.
16. Jones, P.; Altamura, S.; De Francesco, R.; Gonzales Paz, O.; Kinzel, O.; Mesiti, G.;
Monteagudo, E.; Pescatore, G.; Rowley, M.; Verdirame, M.; Steinkuhler, C. J.
Med. Chem. 2008, 51, 2350.
9. Plumb, J. A.; Finn, P. W.; Williams, R. J.; Bandara, M. J.; Romero, M. R.; Watkins,
C. J.; La Thangue, N. B.; Brown, R. Mol. Cancer. Ther. 2003, 2, 721.
10. Suzuki, T.; Ando, T.; Tsuchiya, K.; Fukazawa, N.; Saito, A.; Mariko, Y.;
Yamashita, T.; Nakanishi, O. J. Med. Chem. 1999, 42, 3001.
11. (a) Zhou, N.; Moradei, O. M.; Raeppel, S.; Leit, S.; Frechette, S.; Gaudette, F.;
Paquin, I.; Bernstein, N.; Bouchain, G.; Vaisburg, A.; Jin, Z.; Gillespie, J.; Wang, J.;
Fournel, ; Yan, P. T.; Trachy-Bourget, M. C.; Kalita, A.; Lu, A.; Rahil, J. A.;
MacLeod, R.; Li, Z.; Besterman, J. M.; Delorme, D. J. Med. Chem. 2008, 51, 4072;
(b) Fournel, M.; Bonfils, C.; Hou, Y.; Yan, P. T.; Trachy-Bourget, M. C.; Kalita, A.;
17. Vannini, A.; Volpari, C.; Gallinari, P.; Jones, P.; Mattu, M.; Carfì, A.; De
Francesco, R.; Steinkuhler, C.; Di Marco, S. EMBO Rep. 2007, 8, 879.
18. Jones, P.; Altamura, S.; Chakravarty, P. K.; Cecchetti, O.; De Francesco, R.;
Gallinari, P.; Ingenito, R.; Meinke, P. T.; Petrocchi, A.; Rowley, M.; Scarpelli, R.;
Serafini, S.; Steinkuhler, C. Bioorg. Med. Chem. Lett. 2006, 16, 5948.