Synthesis, immunosuppressive activity and structure-activity relationship study of a new series of 4-N-piperazinyl-thieno[2,3-d]pyrimidine analogues

Article abstract of DOI:10.1016/j.bmcl.2009.12.098

The synthesis of a new series of 4-N-piperazinyl-thieno[2,3-d]pyrimidines is described. The synthetic route allows introducing structural variety at positions 2, 4 and 6 of the scaffold. Evaluation of their immunosuppressive activity in a Mixed Lymphocyte Reaction (MLR) assay revealed that the most potent compound has an IC₅₀-value of 66 nM and therefore deserves attention for further medicinal chemistry optimization.

Full text of DOI:10.1016/j.bmcl.2009.12.098

Bioorganic & Medicinal Chemistry Letters 20 (2010) 844–847
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis, immunosuppressive activity and structure–activity relationship
study of a new series of 4-N-piperazinyl-thieno[2,3-d]pyrimidine analogues
Mi-Yeon Jang a, Steven De Jonghe b, Kristien Van Belle b, Thierry Louat b, Mark Waer c, Piet Herdewijn a,
*
a Catholic University of Leuven, Rega Institute, Laboratory of Medicinal Chemistry, Minderbroedersstraat 10, 3000 Leuven, Belgium
b Catholic University of Leuven, Interface Valorisation Platform (IVAP), Kapucijnenvoer 33, 3000 Leuven, Belgium
c Catholic University of Leuven, Laboratory of Experimental Transplantation, Campus Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a new series of 4-N-piperazinyl-thieno[2,3-d]pyrimidines is described. The synthetic
route allows introducing structural variety at positions 2, 4 and 6 of the scaffold. Evaluation of their
immunosuppressive activity in a Mixed Lymphocyte Reaction (MLR) assay revealed that the most potent
compound has an IC50-value of 66 nM and therefore deserves attention for further medicinal chemistry
optimization.
Received 17 November 2009
Revised 21 December 2009
Accepted 22 December 2009
Available online 4 January 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Medicinal chemistry
Thieno[2,3-d]pyrimidine
Immunosuppressive activity
MLR assay
In the last 50 years, solid organ transplantation has become the
therapy of choice for end stage liver, lung and heart diseases. This
is mainly due to the fact that more effective immunosuppressive
medication became available.1 Cyclosporine and tacrolimus belong
to the most frequently used drugs for transplant patients.2 Their
immunosuppressive activity is linked to inhibition of calcineurin,
a serine-threonine phosphatase that activates intracellular gene-
promoting transcription factors involved in IL-2 activation. In order
to achieve calcineurin inhibition, cyclosporine binds to an intracel-
lular cyclophilin, whereas tacrolimus interacts with another
cytoplasmic protein, called FK-binding protein. The ultimate bio-
logical effect of calcineurin inhibition is a decreased IL-2 produc-
tion and a resultant decreased IL-2–mediated proliferation of
T-cells. Sirolimus (rapamycin)3 also binds to a cytosolic immuno-
philin, called FKBP12. However, the sirolimus–FKBP12 complex
does not inhibit calcineurin, nor does it inhibit pathways of IL-2
production. The sirolimus–FKBP12 complex binds proteins down-
stream of IL-2 in T-cell activation pathways, known as the mam-
malian target of rapamycin (mTOR) that prevent DNA and
protein synthesis in T-cells. Antiproliferative agents, such as aza-
thioprine and mycophenolate mofetil (MMF) are commonly used
in transplant centers.4 Azathioprine is an inhibitor of purine bio-
synthesis. This leads to inhibition of proliferation of many cell
types and azathioprine is therefore a quite toxic drug. Inhibition
of inosine monophosphate dehydrogenase by MMF leads to a
depletion of guanine nucleotides and an inhibition of proliferating
lymphocytes. In addition, corticosteroids have played a central role
in the maintenance of immunosuppression as well as the treat-
ment of acute rejection.
Despite intensive efforts to overcome rejection following organ
transplantation, the strong demand for effective and safe immuno-
suppressants still remains.5 Most of the currently available drugs
for transplantation are known to be accompanied with serious side
effects5,3 such as nephrotoxicity, neurotoxicity, hyperlipidaemia,
new-onset post-transplant diabetes mellitus and hypertension.6
Thus, development of potent and safe immunosuppressants has
been desired.
However, in the first decade of the new millennium, no new
medication specifically indicated for organ transplantation has
been approved. There are currently three small compounds in var-
ious stages of clinical development in renal transplantation.7
ISA247 is a semisynthetic cyclosporine analogue. It is more potent
than cyclosporine in the calcineurin enzymatic assay and in pre-
clinical animal models of organ transplantation. Furthermore,
studies in monkeys suggest that this compound has no nephrotox-
icity. Pfizer is currently developing CP-690550, a selective Janus ki-
nase 3 (JAK3) inhibitor. AEB071 is currently in clinical trials by
Novartis. This compound blocks early T-cell activation by selective
inhibition of protein kinase C.
Over the last few years, our group has been active in the search
for novel immunosuppressive agents based on a bicyclic, heteroar-
omatic scaffold. Immunosuppressive activity has been associated
* Corresponding author. Tel: +32 16 337387; fax: +32 16 337340.
E-mail addresses: piet.herdewijn@rega.kuleuven.be, chantal.biernaux@rega.kule
uve (P. Herdewijn).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.12.098

M.-Y. Jang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 844–847
845
with pteridines,8 5-deazapteridines9 (pyrido[2,3-d]pyrimidines)
and 8-deaza-pteridines10 (pyrido[3,2-d]pyrimidines). A common
element is that these are all bicyclic hetero-aromatic flat struc-
tures. One of the main conclusions of our structure–activity rela-
tionship (SAR) study was that a phenoxyacetyl-piperazinyl group
at position 4 is necessary to obtain potent immunosuppressive
activity. As part of further exploration of the SAR of these com-
pounds, we envisaged to use the thieno[2,3-d]pyrimidine scaffold
as core scaffold, as described in this Letter.
For evaluation of the immunosuppressive activity of the com-
pounds, an allogeneic Mixed Lymphocyte Reaction (MLR) assay
was used.11 The MLR assay is a fundamental benchmark test for
immunosuppressants activity. Most of the current immunosup-
pressive compounds are very potent in MLR assay and several of
them were discovered by this assay.12 Therefore, it is appropriate
to evaluate the immunosuppressive activity of new compounds
with a MLR assay. This screening is used as a predictive in vitro test
of in vivo transplant rejection. Peripheral blood lymphocytes from
two individuals are mixed together in tissue culture for several
days. Lymphocytes from incompatible individuals will stimulate
each other to proliferate significantly (measured by tritiated thy-
midine uptake) whereas those from compatible individuals will
not. In the one-way MLR test, the lymphocytes from one of the
individuals are inactivated (usually by treatment with mitomycin
or radiation) thereby allowing only the untreated remaining popu-
lation of cells to proliferate in response to foreign histocompatibil-
ity antigens.
Within the isomeric thieno[3,2-d]pyrimidine series, a versatile
starting material has been described in literature.13 6-Bromo-4-
chloro-thieno[3,2-d]pyrimidine can be regioselectively functional-
ized at positions 4 and 6. Palladium-catalyzed cross-coupling
reactions, occur exclusively at position 6, whereas the reaction
with amines results in displacement of the chlorine at position 4.
Therefore, it was envisaged that the corresponding 6-bromo-4-
chloro-2-substituted-thieno[2,3-d]pyrimidine could similarly act
as a versatile building block for the introduction of various substit-
uents at position 4 and 6 to build up thieno[2,3-d]pyrimidine
libraries. In order to prepare this building block, methyl 2-amino-
thiophene-3-carboxylate 1 was heated in formamide yielding
thieno[2,3-d]pyrimidin-4(3H)one 2 (Scheme 1).14 Conversion of
the 4-oxo group into chlorine was achieved in good yield with oxa-
lyl chloride and DMF (Vilsmeier reagent). Regioselective introduc-
tion of a bromine at position 6 was performed with n-BuLi and
carbon tetrabromide as bromine source at low temperature
(À78 °C). Two major products (4a and 4b) were isolated in a ratio
of 1:1, after purification by flash chromatography. The mass spec-
tral and 1H NMR data of 4b, indicate that a butyl group is present
on the thienopyrimidine scaffold. Heteronuclear Multiple Bond
Correlation (HMBC) spectroscopy was used to determine the exact
regiochemistry (Fig. 1). A direct coupling (1J) of the aromatic pro-
ton (Ha: d = 7.4 ppm) and a 13C signal at 122 ppm was detected.
The carbon signal of C(2) is usually seen above 150 ppm, whereas
the carbon at position 5 is observed around 120 ppm, indicating
that the aromatic proton is present at position 5. Only one HMBC
correlation (2J) between the methylene protons (Hb) of the n-butyl
substituent and a carbon (at d = 150 ppm) is observed, which sup-
ports the fact that the n-butyl group is attached to C(2). These NMR
data also confirm that bromination takes place at position 6 of the
thieno[2,3-d]pyrimidine scaffold. Standard reaction conditions for
Suzuki coupling of 4a and 4b with 4-fluorophenylboronic acid
afforded compounds 5a and 5b, respectively. The remaining chlo-
rine atom was displaced by 2-(4-chlorophenoxy)-1-(piperazin-1-
yl)ethanone15 under mild conditions yielding compounds 6a and
6b.
Although 6-bromo-4-chloro-thieno[2,3-d]pyrimidine 4a is a
versatile building block, due to synthetic difficulties (low yields
and the formation of side products), an alternative route was ex-
plored (Scheme 2), starting from an appropriate phenylacetalde-
hyde (either commercially available when R = H; or synthesized
by oxidation of 2-(4-fluorophenyl)ethanol with pyridinium chloro-
chromate if R = F). The condensation of aldehyde 7 with ethyl cya-
noacetate in the presence of a base and elemental sulfur (Gewald
reaction) furnished ethyl 2-amino-5-aryl-thiophene-3-carboxylate
8.16 Reaction of compound 8 with several nitriles under acidic
conditions or with chloroformamidine hydrochloride provided
the 4-oxo-thieno[2,3-d]pyrimidine analogues 9a–e in good yields.
For the introduction of the piperazine moiety at position 4, a con-
venient phosphonium-mediated SNAr reaction for the derivatisa-
tion of the lactam functionality is used.17 Treatment of 9a and 9b
with
benzotriazol-1-yloxytris(dimethylamino)-phosphonium-
hexafluorophosphate (BOP), DBU and piperazine leads to the for-
mation of the 7-N-piperazinyl-thieno[2,3-d]pyrimidine analogues
10a,b. The piperazine moiety can be further derivatised to amides
(compound 11g–o) by reaction with appropriate acid chlorides, or
to urea (compound 11p–s) by coupling with isocyanates. Alterna-
tively, it is also possible to introduce substituted piperazine deriv-
atives directly in one step by the BOP-mediated reaction
(compound 11a–f, t, u). An overview of the compounds 11a–u syn-
thesized according to this general synthetic scheme is shown in
Table 1, together with their MLR data. The IC50 value represents
Scheme 1. Reagents and conditions: (a) formamide, reflux, 3 h; (b) oxalyl chloride,
DMF, CH2Cl2, 0 °C to reflux, 2.5 h; (c) n-BuLi, CBr4, THF, À78 °C, 20 min, then rt, 2 h;
(d) 4-fluorophenylboronic acid, Pd(PPh3)4, K2CO3, dioxane/water (3:1 v/v), reflux,
3 h; (e) 2-(4-chlorophenoxy)-1-(piperazin-1-yl)ethanone, dioxane, NEt3, 70 °C,
24 h.
Figure 1.

846
M.-Y. Jang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 844–847
Scheme 2. Reagents and conditions: (a) ethyl cyanoacetate, S, NEt3, DMF, 50 °C to rt; (b) RCN, 4 M HCl in dioxane, rt then DMF, 100 °C or chloroformamidine hydrochloride,
dimethylsulfone, 130 °C; (c) BOP, DBU, amine, CH3CN, rt to 60 °C; (d) acid chloride or isocyanate, NEt3, DMF, rt; (e) 2 M NaOH, MeOH, rt; (f) 7N NH3 in MeOH, rt.
Table 1 (continued)
Table 1
Overview of synthesized compounds and their MLR inhibition dataa
R2
R4
R6
F
MLR IC50
6.99
(lM)
a
Compd
R4
N
O
11g
NH2
N
O
O
N
R6
R2
N
11h
11i
NH2
NH2
F
F
>10
>10
S
Compd
R2
H
R4
R6
F
MLR IC50
6.35
(lM)
a
O
O
6a
N
O
O
Cl
Cl
Cl
O
O
11j
NH2
NH2
H
H
0.33
0.78
O
O
6b
n-Bu
F
F
1.33
3.39
Cl
O
O
O
11k
11a
CH3
O
11l
NH2
H
0.072
O
O
O
O
O
O
O
O
O
O
OCH3
11b
11c
11d
Ph
F
F
F
>10
>10
>10
O
O
Cl
Cl
Cl
Cl
Cl
O
O
11m
11n
11o
NH2
NH2
NH2
H
H
H
0.066
0.32
0.21
F
CO2Et
CO2H
Br
O
O
CH3
O
O
11e
11f
CONH2
NH2
F
F
>10
0.7
11p
11q
NH2
NH2
F
2.8
N
H
CH3
H
0.93
N
H
CH3

M.-Y. Jang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 844–847
847
In agreement with previous research,8–10 phenoxyacetyl-piper-
azine derivatives show potent activities, while a nicotinoyl amide
(compound 11i) and an aliphatic amide (compound 11h) do not
show any activity at 10 lM. The closely related hydrocinnamoyl
side chain (compound 11g) is also less active. To probe the optimal
Table 1 (continued)
R2
R4
R6
H
MLR IC50
6.7
(lM)
a
Compd
Cl
O
11r
NH2
N
H
substitution pattern on the phenyl ring of the phenoxyacetyl side
chain,
a number of analogues were prepared (compounds
O
11k–o). The unsubstituted congener 11k, the p-bromo analogue
11n, and the m-methyl congener 11o show comparable activity
as p-chloro compound 11j, with MLR IC50 between 0.2 and
11s
11t
NH2
NH2
NH2
F
8.2
N
O
H
H
9.64
0.7 lM. The p-methoxy (compound 11l) and the p-fluoro (com-
pound 11m) analogue do show excellent in vitro MLR activity with
IC50 values of 72 nM and 66 nM, respectively, This is almost as po-
tent as Cyclosporine (IC50 = 54 nM), a clinically used immunosup-
pressive agent.
CH3
N
11u
8.12
O
CsAb
0.054
In summary, the discovery of a new series of immunosuppres-
sive agents based on a thieno[2,3-d]pyrimidine scaffold is de-
scribed. The preliminary SAR studies led to the identification of
compound 11m with an IC50 value of 66 nM in the cellular MLR
assay, which is equipotent to Cyclosporine. These compounds rep-
resent a valid starting point for a new generation of immunosup-
pressive drugs.
a
Values are means of two independent experiments.
Cyclosporine A.
b
the lowest concentration of the compound (expressed in
resulted in a 50% inhibition of the MLR.
lM) that
Inspection of the SAR at position 2 (substituent R2 in Table 1) re-
veals that an amino group is optimal for biological activity (MLR
References and notes
IC50 = 0.7 lM). An n-butyl group has very similar activity (only
1. (a) Golshayan, D.; Buhler, L.; Lechler, R. I.; Pascual, M. Transplant. Int. 2007, 20,
12; (b) Lechler, R.; Sykes, M.; Thomson, A. W.; Turka, L. A. Nat. Med. 2005, 6,
605.
2. Pratschke, J.; Tullius, S. G.; Neuhaus, P. Graft 2002, 5, 338.
3. Gummert, J. F.; Ikonen, T.; Morris, R. E. J. Am. Soc. Nephrol. 1999, 10, 1366.
4. Allison, A. C. Immunopharmacology 2000, 47, 63.
twofold less active than the amino congener), whereas a methyl
is not far off (fivefold less active than the amino congener). Other
substituents (hydrogen, phenyl, carboxylic acid and its amide
and ester analogue) are less active, displaying MLR IC50 values of
6 lM or more.
5. Smith, J. M.; Nemeth, T. L.; McDonald, R. A. Pediatr. Clin. North. Am. 2003, 50,
1283.
By direct comparison of the fluorophenyl (R6 = F) and phenyl
(R6 = H) congeners, it seems that fluorine substitution has a slightly
negative impact on activity. For example, compare the urea derived
6. (a) O’Shea, J. J.; Pesu, M.; Borie, D. C.; Changelian, P. S. Nat. Rev. Drug Disc. 2004,
3, 555; (b) Hong, J. C.; Kahan, B. D. J. Semin. Nephrol. 2000, 20, 108.
7. Vincenti, F.; Kirk, A. D. Am. J. Transplant. 2008, 8, 1972.
8. (a) Shen, C.; Dillissen, E.; Kasran, A.; Lin, Y.; Herman, J.; Sienaert, I.; De Jonghe,
S.; Kerremans, L.; Geboes, K.; Boon, L.; Rutgeerts, P.; Ceuppens, J. Clin. Immunol.
2007, 122, 53; (b) Waer, M.; Herdewijn, P.; De Jonghe, S.; Marchand, A.; Gao, L.
J. WO2005/039587.
piperazines 11q and 11p with IC50 values of 0.93
respectively, and the amide derived piperazines 11j and 11f with
MLR IC50 values of 0.33 M and 0.7 M, respectively.
lM and 2.8 lM,
l
l
The SAR of the piperazine moiety has been extensively investi-
gated. It seems that the second nitrogen atom of the piperazine
moiety needs to be nonbasic as compounds in which the second
nitrogen atom was basic show much less immunosuppressive
activity. Compound 11t is 12-fold less active than its direct ana-
logue with a carbonyl functionality (compound 11k). Insertion of
a methylene group between the carbonyl moiety and the nitrogen
atom of the piperazinyl group affords compound 11u with a MLR
9. De Jonghe, S.; Herdewijn, P.; Gao, L. J. WO2006/087229.
10. (a) De Jonghe, S.; Dolusic, E.; Gao, L. J.; Herdewijn, P.; Pfleiderer, W. WO2006/
069805.; (b) De Jonghe, S.; Dolusic, E.; Gao, L. J.; Herdewijn, P.; Pfleiderer, W.
WO2009/003669.
11. Waer, M.; Vanrenterghem, Y.; Van der Schueren, E.; Michielsen, P.; Vandeputte,
M. Transplant. Proc. 1987, 19, 1570.
12. Kino, T.; Hatanaka, H.; Hashimoto, M.; Nishiyama, M.; Goto, T.; Okuhara, M.;
Kohsaka, M.; Aoki, H.; Imanaka, H. J. Antibiot. 1987, 40, 1249.
13. Munchhof, M. J.; Beebe, J. S.; Casavant, J. M. Bioorg. Med. Chem. Lett. 2004, 14,
21.
14. El-Bahaie, S.; Assy, M. G.; Ibrahim, Y. A. Sulfur Lett. 1988, 8, 199.
15. 2-(4-Chlorophenoxy)-1-(piperazin-1-yl)ethanone was prepared from tert-
butyl piperazine-1-carboxylate by the reaction with 4-chlorophenoxyacetyl
chloride and triethylamine in dichloromethane followed by acidic cleavage of
Boc group by treatment of trifluoroacetic acid in dichloromethane.
16. Donkor, I. O.; Li, H.; Queener, S. F. Eur. J. Med. Chem. 2003, 38, 605.
17. (a) Wan, Z. K.; Wacharasindhu, S.; Levins, C. G.; Lin, M.; Tabei, K.; Mansour, T. S.
J. Org. Chem. 2007, 72, 10194; (b) Wan, Z. K.; Wacharasindhu, S.; Binnun, E.;
Mansour, T. S. Org. Lett. 2006, 8, 2425.
IC50 of 8.12 lM. For the urea substituted piperazine derivatives
(11p–s), most potent activity is associated with a m-tolylacetamide
substituent (compounds 11p and 11q). The choice of substituents
on the phenyl group is important as the activity of the p-chloro-
analogue 11r drops by a factor 7, as compared to the m-methyl
substituted analogue 11q. The aliphatic urea (compound 11s) lacks
considerable MLR activity.