Evaluation and optimization of antifibrotic activity of cinnamoyl anthranilates

Article abstract of DOI:10.1016/j.bmcl.2009.09.120

Tranilast is an anti-inflammatory drug in use for asthma and atopic dermatitis. In studies over the last decade it has been revealed that tranilast can reduce fibrosis occurring in the kidney during diabetes, thereby delaying and/or preventing kidney dysfunction. We report a structure-activity study aimed at optimizing the antifibrotic activity of tranilast. A series of cinnamoyl anthranilates were prepared and assessed for their ability to prevent TGF-β-stimulated production of collagen in cultured renal mesangial cells. We reveal derivatives with improved potency and reduced cellular toxicity relative to tranilast. 3-Methoxy-4-propargyloxycinnamoyl anthranilate reduces albuminuria in a rat model of progressive diabetes, and thus has potential as an innovative treatment for diabetic nephropathy.

Full text of DOI:10.1016/j.bmcl.2009.09.120

Bioorganic & Medicinal Chemistry Letters 19 (2009) 7003–7006
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Evaluation and optimization of antifibrotic activity of cinnamoyl anthranilates
Steven C. Zammit a, Alison J. Cox b, Renae M. Gow b, Yuan Zhang b, Richard E. Gilbert c, Henry Krum b,
a,
Darren J. Kelly b,d, , Spencer J. Williams
*
*
a School of Chemistry, Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, Parkville, VIC 3010, Australia
b Department of Medicine, University of Melbourne, St. Vincent’s Hospital, Fitzroy, VIC 3065, Australia
c Keenan Research Centre of the Li Ka Shing Knowledge Institute, St. Michael’s Hospital and University of Toronto, Ontario, Canada
d Fibrotech Therapeutics, Melbourne, VIC 3000, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 July 2009
Revised 28 September 2009
Accepted 30 September 2009
Available online 7 October 2009
Tranilast is an anti-inflammatory drug in use for asthma and atopic dermatitis. In studies over the last
decade it has been revealed that tranilast can reduce fibrosis occurring in the kidney during diabetes,
thereby delaying and/or preventing kidney dysfunction. We report a structure–activity study aimed at
optimizing the antifibrotic activity of tranilast. A series of cinnamoyl anthranilates were prepared and
assessed for their ability to prevent TGF-b-stimulated production of collagen in cultured renal mesangial
cells. We reveal derivatives with improved potency and reduced cellular toxicity relative to tranilast. 3-
Methoxy-4-propargyloxycinnamoyl anthranilate reduces albuminuria in a rat model of progressive dia-
betes, and thus has potential as an innovative treatment for diabetic nephropathy.
Keywords:
Diabetes
Antifibrotic
Albuminuria
Ó 2009 Elsevier Ltd. All rights reserved.
Transforming growth factor-b
Chronic kidney disease can be caused by diabetes and high-
blood pressure and has devastating impacts on health, increasing
the risk of cardiovascular mortality and morbidity, blindness and
amputation. Pathological fibrosis is a hallmark of progressive renal
disease and correlates closely with the loss of kidney function.
Platelet-derived growth factor (PDGF), connective tissue growth
factor (CTGF) and transforming growth factor-b (TGF-b) are cyto-
kines that have been implicated in the progression of renal disease
through the stimulation of fibrosis in the mesangium.1 Inhibition
of TGF-b ameliorates tubulointerstitial fibrosis in a rat model of
diabetic nephropathy,2,3 and other models of chronic renal
disease.4,5
Antifibrotic drugs may provide a novel strategy for the treat-
ment of progressive kidney disease, and indeed other disorders
associated with pathological fibrosis. A recent review highlighted
the fact that some 45% of all deaths in the developed world may
have an underlying pathology of aberrant fibrosis.6 Despite this,
there are no drugs in clinical use that are optimized for antifibrotic
action. One promising antifibrotic agent is the cinnamoyl anthrani-
late tranilast. Tranilast has been used in Japan and South Korea for
over 20 years to treat allergic disorders, hypertrophic scars and
scleroderma. This drug has been shown to inhibit the pro-fibrotic
growth factors TGF-b,7,8 PDGF8 and CTGF.9 However, the recent
PRESTO (prevention of restenosis with tranilast and its outcomes)
study highlighted several reversible abnormalities associated with
the use of tranilast at the elevated doses required for antifibrotic
action.10 These included hyperbilirubinemia, increased alanine
aminotransferase (ALT)/serum glutamic pyruvic transaminase
(SGPT), decline in haemoglobin, and increased serum creatinine.
In order to optimize the antifibrotic effects of tranilast, a series
of cinnamoyl anthranilate derivatives were synthesized and their
ability to prevent TGF-b stimulated collagen production in vitro
was investigated. Modifications included saturation of the double
bond connecting the two rings, and removal and/or replacement
of the methoxy (positions A and B) and carboxyl groups (position
C) (Fig. 1). The most active compounds were assessed for their abil-
ity to inhibit TGF-b in cultured mesangial cells at a range of con-
centrations and to inhibit the onset of albuminuria in a rat model
of progressive diabetes.
Two general approaches were used for the synthesis of the var-
ious substituted cinnamoyl anthranilates. In the first approach a
piperidine-catalyzed Knoevenagel reaction of a carboxyacetamido-
benzoic acid and a benzaldehyde derivative provided the substi-
tuted cinnamoyl anthranilates. In the second approach,
substituted cinnamic acids were converted to the corresponding
acid chloride and treated with an aniline derivative (Scheme 1).
Benzaldehyde precursors were either obtained from commer-
cial sources, or were synthesized by alkylation of precursor pheno-
lic benzaldehydes with alkyl halides or alkyl tosylates (the latter
derived in turn from the corresponding alcohols). 2-Carboxyace-
tamidobenzoic acid was prepared by reaction of anthranilic acid
and Meldrum’s acid. 2-Aminobenzamide was obtained by reaction
* Corresponding authors.
E-mail addresses: dkelly@medstv.unimelb.edu.au (D.J. Kelly), sjwill@unimelb.
edu.au (S.J. Williams).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.09.120

7004
S. C. Zammit et al. / Bioorg. Med. Chem. Lett. 19 (2009) 7003–7006
O
O
2
MeO
MeO
MeO
MeO
N
H
N
1
H
CO2H
R
tranilast
Compound
number
Double
R
Inhibition at 100
Bond
µM (%)
O
A
B
N
H
1
Yes
CO2H
70
C
(tranilast)
8
9
10
Yes
Yes
No
H
50
35
CONH2
CO2H
tranilast modifications
no inhibition
Figure 1. Structural modifications of tranilast.
Chart 2. Inhibitory activities for compounds on TGF-b stimulated production of
collagen synthesis in cultured mesangial cells.
of ammonia and isatoic anhydride.11 The alkene of tranilast was
saturated by reduction with hydrogen and palladium on carbon.
Antifibrotic activity was assessed by evaluating the ability of each
compound to reduce the incorporation of [3H]-proline into colla-
gen upon TGF-b stimulation. For initial screening purposes, com-
O
R1O
R2O
N
H
pounds were screened at a concentration of 100
More active compounds were rescreened at 30
prior to more detailed analysis.
l
M in triplicate.
CO2H
l
M in triplicate,
As shown in Chart 1, variation of the methoxy substituents at
the 3 and 4 positions of ring 1 results in a considerable variation
of biological activity. Removal of the methoxy substituents (com-
pounds 2–4), or replacement of a single methoxy substituent with
a hydroxy group (compounds 5 and 6), resulted in a reduction in
activity whereas dihydroxy derivative 7 demonstrated more po-
tent inhibition of TGF-b.
Chart 2 reveals that the carboxylic acid functional group of
tranilast and the
a,b-unsaturated system are requirements for
activity. Removal of the carboxylic acid or replacement of the acid
with a primary amide gave 8 and 9, which were both less active
than tranilast. Saturated derivative 10 was inactive.
O
2
3
N
H
1
6
Rn
CO2H
4
5
Compound
number
n
R
Inhibition at 100
µM (%)
1
2
3,4-(OMe)2
70
(tranilast)
2
3
4
5
6
7
0
1
1
2
2
2
-
no inhibition
no inhibition
4-OMe
3-OMe
3-OMe, 4-OH
3-OH, 4-OMe
3-OH, 4-OH
40
50
50
90
Chart 1. Inhibitory activities for compounds on TGF-b stimulated production of
Chart 3. Inhibitory activities for compounds on TGF-b stimulated production of
collagen synthesis in cultured mesangial cells.
collagen synthesis in cultured mesangial cells.
CHO
R1
O
O
a
HO2C
H2N
N
N
R1
b
H
H
CO2H
CO2H
CO2H
O
R
c
H2N
MeO
MeO
CO2H
MeO
MeO
COCl
MeO
MeO
N
H
d
R2
Scheme 1. Synthesis of cinammoyl anthranilate derivatives. Reagents and conditions: (a) Meldrum’s acid, toluene, reflux, 3 h; (b) piperidine, toluene, reflux, 30 min;
(c) SOCl2, cat. DMF, toluene, 50 °C, 1 h; (d) pyridine, 24 h.

S. C. Zammit et al. / Bioorg. Med. Chem. Lett. 19 (2009) 7003–7006
7005
*
125
100
75
50
25
0
125
100
75
50
25
0
125
1
11
12
14
125
100
75
50
25
0
*
100
75
50
25
0
–
+
3
10
30
100
–
+
3
10
30
100
–
+
3
10
30
100
–
+
3
10
30
100
125
*
*
125
100
75
50
25
0
*
125
100
75
50
25
0
19
20
22
29
125
100
75
50
25
0
100
75
50
25
0
*
*
–
+
3
10
30
100
–
+
3
10
30
100
–
+
3
10
30
100
–
+
3
10
30
100
Figure 2. Dose–response histograms of inhibition of TGF-b stimulated [3H]-proline incorporation by tranilast (1) and compounds 11, 12, 14, 19, 20, 22 and 29 measured at
concentrations of 3, 10, 30 and 100
l
M. Inhibition values were determined in triplicate. (À) and (+) refer to the absence and presence of stimulation by the cytokine TGF-b.
Error bars indicate standard error. (*) Some toxicity was observed at the indicated concentration.
The effect of alteration of the nature of the alkyl ethers on ring 1
20
was assessed next (Chart 3). This was of particular interest as the
PRESTO trial, which evaluated the prevention of restenosis by
tranilast, found that tranilast increased the risk of hyberbilirubin-
emia in patients with Gilbert’s syndrome, which was attributed
to tranilast and/or its metabolites inhibiting the phase II glucuron-
osyltransferase UGT1A1.12,13 Glucuronidation, 4-demethylation
and sulfation of the 4-demethylated product are major metabolic
routes of tranilast.14 Thus, modification of the alkyl groups at the
3 and 4 positions may lead to drugs that are more metabolically
stable and limit side effects. To limit the number of rotatable bonds
we investigated the effect of alkynyloxy and cycloalkyloxy groups.
As chain length increased so did inhibition of TGF-b stimulated col-
18
16
14
12
10
8
6
4
2
Table 1
Calculated metabolic stability parameters of test compounds based on degradation
profiles in human liver microsomes.15
0
Control
Control + 11
Diabetic
Diabetic + 11
Compound
Microsome-predicted
CLint (mL/min/kg)
Microsome-predicted EH
Figure 4. Albuminuria after four weeks of experimental diabetes in a streptozo-
tocin-treated hypertensive Ren2 rat model. Treatment with 11 commenced within
24 h of streptozotocin injection.
1 (Tranilast)
6.7
5.3
6.9
21.8
54.1
12.6
36.3
17.4
0.24
0.20
0.25
0.51
0.72
0.38
0.64
0.46
11
12
14
19
20
22
29
lagen production in mesangial cells. The C5 and C6 derivatives 14,
16, 22, 24, 25 and 29 showed the greatest activity. However, the
C6, C7 and C8 derivatives (18, 26, 30, 31 and 32) were toxic in cel-
lular assays and thus could not be assessed for TGF-b inhibition.
Compounds 11 and 12, and their regioisomeric derivatives 19
and 20, demonstrated the most promising increase in TGF-b inhib-
itory activity relative to tranilast without showing any cellular tox-
icity. Similarly, compounds 14, 22 and 29 also showed potent
inhibition of TGF-b whilst displaying only mild cellular toxicity
0.6
0.5
0.4
0.3
0.2
0.1
0
when dosed at 100 lM, and no observed toxicity at 30 lM.
To better characterize the antifibrotic activity of this series of
compounds, more detailed inhibition profiles were determined at
a range of concentrations. Figure 2 shows that the TGF-b inhibitory
activity of the homologous series 11, 12 and 14 increases as the
chain length increases. A similar result was seen for the regioiso-
meric compounds 19, 20 and 22. It appears that similar sized
substituents possess similar activities; thus the 2-pentynyl regioi-
somers 14 and 22 have similar potencies.
50
100
200
400
dose of 11 (mg/kg/day)
Figure 3. Oral bioavailability of 11 as assessed by blood plasma concentration
determined by LC–MS. Bar colors denote plasma concentration at varying times
after dosing: white, 1 h after dosing; grey, 4 h after dosing; black, 8 h after dosing.
The metabolic stability of tranilast relative to the most
promising structural analogues was assessed in human liver

7006
S. C. Zammit et al. / Bioorg. Med. Chem. Lett. 19 (2009) 7003–7006
microsomes. Compounds were incubated with microsomes and
the loss of parent compound assessed by LC–MS analysis. Micro-
some-predicted intrinsic clearance values are shown in Table 1.
For the pair of homologous series 11, 12 and 14, and 19, 20
and 22, clearance rates increased with increasing substituent
chain length. Comparison of the regioisomeric pairs: 11 and 19;
12 and 20; and 14 and 22; shows that clearance rates are higher
for the 3-substituted series. Given that tranilast is in clinical use,
the observation of similar predicted metabolic stability for com-
pounds 11 and 12 suggests that they are sufficiently stable to
merit further investigation.
Based on the activity data and the predicted intrinsic clearance
values, compound 11 was selected for preliminary bioavailability
studies. Varying amounts of compound 11 were given as a single
oral dose to Sprague-Dawley rats and blood plasma levels were as-
sessed by LC–MS. Maximal blood plasma levels were obtained 4 h
post oral, and significant amounts of 11 were still present 8 h after
dosing (Fig. 3).
The ability to reduce albuminuria was experimentally assessed
in a rat model of diabetic nephropathy (Fig. 4). Hypertensive Ren2
rats were exposed to streptozotocin to induce diabetes. After four
weeks of experimental diabetes, a 24 h urine collection was
assayed for albuminuria by radioimmunoassay. Untreated rats
developed significant albuminuria that was attenuated by oral
administration of 11 by oral gavage at 200 mg/kg/day. In contrast,
tranilast attenuates albuminuria only when administered at higher
doses of 400 mg/kg/day.16
In conclusion novel derivatives of the known antifibrotic, trani-
last (1), were prepared and assayed for their ability to inhibit TGF-b
in cultured mesangial cells. Several lipophilic derivatives possessed
increased activity relative to 1, especially compound 11, which also
showed excellent metabolic properties, and improved in vivo
activity. Compound 11 can reduce albuminuria in a hypertensive
diabetic rat model and therefore represents a promising lead in
the development of potential antifibrotic drugs.
Acknowledgment
We thank the National Health and Medical Research Council for
financial support. Denis Scanlon is thanked for HPLC analysis.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.09.120.
References and notes
1. Sharma, K.; Ziyadeh, F. N. Diabetes 1995, 44, 1139.
2. Gilbert, R. E.; Cox, A.; Wu, L. L.; Allen, T. J.; Hulthen, U. L.; Jerums, G.; Cooper, M.
E. Diabetes 1998, 47, 414.
3. Gilbert, R. E.; Wilkinson-Berka, J. L.; Johnson, D. W.; Cox, A.; Soulis, T.; Wu, L. L.;
Kelly, D. J.; Jerums, G.; Pollock, C. A.; Cooper, M. E. Kidney Int. 1998, 54, 1052.
4. Gilbert, R. E.; Akdeniz, A.; Allen, T. J.; Jerums, G. Nephrol. Dial. Transplant. 2001,
16, 2442.
5. Houlihan, C. A.; Akdeniz, A.; Tsalamandris, C.; Cooper, M. E.; Jerums, G.; Gilbert,
R. E. Diabetes Care 2002, 25, 1072.
6. Wynn, T. A. J. Clin. Invest. 2007, 117, 524.
7. Pinto, Y. M.; Pinto-Sietsma, S. J.; Philipp, T.; Engler, S.; Kossamehl, P.; Hocher,
B.; Marquardt, H.; Sethmann, S.; Lauster, R.; Merker, H. J.; Paul, M. Hypertension
2000, 36, 747.
8. Ward, M. R.; Sasahara, T.; Agrotis, A.; Dilley, R. J.; Jennings, G. L.; Bobik, A.
Atherosclerosis 1998, 137, 267.
9. Qi, W.; Chen, X.; Twigg, S.; Polhill, T. S.; Gilbert, R. E.; Pollock, C. A. Kidney Int.
2006, 69, 989.
10. Holmes, D.; Fitzgerald, P.; Goldberg, S.; LaBlanche, J.-M.; Lincoff, A. M.; Savage,
M.; Serruys, P. W.; Willerson, J.; Granett, J. R.; Chan, R.; Shusterman, N. H.;
Poland, M. Am. Heart J. 2000, 139, 23.
11. Staiger, R. P.; Wagner, E. C. J. Org. Chem. 1953, 18, 1427.
12. Danoff, T. M.; Campbell, D. A.; McCarthy, L. C.; Lewis, K. F.; Repasch, M. H.;
Saunders, A. M.; Spurr, N. K.; Purvis, I. J.; Roses, A. D.; Xu, C.-F. Pharmacogenom.
J. 2004, 4, 49.
13. Katoh, M.; Matsui, T.; Yokoi, T. Drug Metab. Disp. 2007, 35, 583.
14. Slobodzian, D. K.; Hsieh, J. Y.; Bayne, W. F. J. Chromatogr. 1985, 345, 345.
15. Obach, R. S. Drug Metab. Dispos. 1999, 27, 1350.
16. Mifsud, S.; Delly, D. J.; Zhang, Y.; Pollock, C. A.; Wilkinson-Berka, J. L.; Gilbert, R.
E. Nephron Physiol. 2003, 95, 83.