Azo-reductase activated budesodine prodrugs for colon targeting

Article abstract of DOI:10.1016/j.bmcl.2012.10.006

Budesodine is a synthetic glurocorticoid that undergoes substantial first pass metabolism, limiting systemic exposure. Its use in treatment of inflammatory bowel disease would benefit from a targeting strategy that could lead to a local topical effect, improving safety and increasing anti-inflammatory efficacy. A two-step prodrug strategy involving azoreduction/cyclization that we developed previously for prednisolone is here applied with some variations to budesonide. The budesodine prodrugs were tested using an in vitro azoreductase assay simulating human colonic microflora. The kinetics of amino steroid ester cyclization and its pH dependence was also evaluated. The stability of the prodrugs systems in simulated human duodenal and gastric fluid was evaluated to determine the likelihood of intact intestinal transit. The propionic acid derived prodrug 3 undergoes rapid activation by Clostridium perfingens and its putative reduction product cyclizes with acceptable rapidity when synthesized independently. These properties of 3 suggest that it has potential in management of ulcerative colitis.

Full text of DOI:10.1016/j.bmcl.2012.10.006

Bioorganic & Medicinal Chemistry Letters 22 (2012) 7573–7577
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Azo-reductase activated budesodine prodrugs for colon targeting
Juan F. Marquez Ruiz a, Kinga Kedziora a, Mary O’Reilly c, Jacqueline Maguire c, Brian Keogh c,
Henry Windle b, Dermot P. Kelleher b, John F. Gilmer a,
⇑
a School of Pharmacy and Pharmaceutical Sciences, Trinity College, Dublin 2, Ireland
b School of Medicine, Trinity College, Dublin 2, Ireland
c Opsona Therapeutics LTD, Trinity Centre for Health Sciences, Institute of Molecular Medicine, St James’s Hospital, Dublin 8, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
Budesodine is a synthetic glurocorticoid that undergoes substantial first pass metabolism, limiting sys-
temic exposure. Its use in treatment of inflammatory bowel disease would benefit from a targeting strat-
egy that could lead to a local topical effect, improving safety and increasing anti-inflammatory efficacy. A
two-step prodrug strategy involving azoreduction/cyclization that we developed previously for prednis-
olone is here applied with some variations to budesonide. The budesodine prodrugs were tested using an
in vitro azoreductase assay simulating human colonic microflora. The kinetics of amino steroid ester
cyclization and its pH dependence was also evaluated. The stability of the prodrugs systems in simulated
human duodenal and gastric fluid was evaluated to determine the likelihood of intact intestinal transit.
The propionic acid derived prodrug 3 undergoes rapid activation by Clostridium perfingens and its putative
reduction product cyclizes with acceptable rapidity when synthesized independently. These properties of
3 suggest that it has potential in management of ulcerative colitis.
Received 11 August 2012
Revised 1 October 2012
Accepted 3 October 2012
Available online 17 October 2012
Keywords:
Budesonide
Azoreductase
Cyclization dependent prodrug
Colon targeting
Trimethyl lock
Clostridium perfringens
Ó 2012 Elsevier Ltd. All rights reserved.
Glucocorticoids (GCs) and aminosalicylates (e.g., 5-ASA) consti-
tute the mainstays of treatment for inflammatory bowel diseases
(IBD) which includes ulcerative colitis (UC) and Crohn’s Disease
formulations that respond to the pH gradient in the GIT as well
as matrices that are specifically sensitive to the colonic
environment.5,6
(CD).1 Biological medicinal products such as antibodies to TNF
a
The prodrug approach can be an effective way to target the co-
lon as reflected in the clinical success of azo prodrugs such as osal-
azine. This is specifically activated by the activities associated with
the colonic microflora to which it is exposed following transit
through the intestinal tract. Prodrug approaches have also been
investigated for targeting budesonide, including compounds
employing as vectors glycosidases associated with the colonic
microflora.4,7–9
We have recently introduced a novel design for colon targeting
that exploits azoreductases to trigger drug release (Fig. 1b).10
Reduction of the prodrug leads to liberation of an anilide ester that
is poised to undergo ring closure, releasing the drug. The design
was developed for targeting prednisolone to the colon by attach-
ment of the promoiety to the steroid 21-OH group that depresses
intestinal absorption. A compound based on this approach was
shown to have low systemic penetration but high efficacy in atten-
uating the effects of dextran sulfate administration in a mouse
model of IBD.10 As budesonide is now a more clinically attractive
compound for IBD, we wondered if our approach could be adapted
to it given structural differences with prednisolone. Budesonide
consists of an epimeric mixture of propyl dioxalans at C22 (R/S)
that sees the incorporation of a new E-ring into the corticosteroid
nucleus (Fig. 1). These structural variations to the classical
pharmacophore occur proximal to the centre involved in the
are becoming more established but these remain expensive and
less widely available. Corticosteroids are especially useful for
inducing remission in acute cases of IBD. Their chronic use is how-
ever limited by serious systemic side effects. These range from cos-
metic problems such as Cushingoid features, to growth failure in
children, hypertension, osteoporosis, diabetes mellitus and glau-
coma as well as more rare CNS dysfunctions including depression
and psychosis. There have been intensive efforts made to develop
GC therapies for IBD treatment that minimize side effects.2
Originally developed for asthma, budesonide (1), is a ‘soft’ cor-
ticosteroid that is metabolically inactivated by hepatic P450 en-
zymes thus limiting systemic exposure.3 Budesonide use is
associated with correspondingly smaller effects on adrenocorticoid
function than conventional GCs, as reflected in response to adreno-
corticotropic hormone or depression of baseline cortisol. There
have been several efforts to develop technologies that can target
budesonide to the colon in order to increase this safety margin
but also to enhance efficacy, which may not be as high as GCs such
as prednisolone (2).4 Mainly these efforts have focused on pharma-
ceutical approaches and the development of delayed release
⇑
Corresponding author. Tel.: +353 1 896 2795; fax: +353 1 896 2793.
E-mail address: gilmerjf@tcd.ie (J.F. Gilmer).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.10.006

7574
J. F. Marquez Ruiz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7573–7577
O
21
OH
O
A
B
HO
OH
HO
OH
O
H
H
O
H
H
H
H
O
O
2
1
O
O
OH
NH2
ODRUG
NH2
ODRUG
n(CH2)
n(CH2)
O
N
N
HO
+
OH
O
OH
Azoreductase
activity
Intramolecular
lactamization
(CH2)n
O
DRUG-OH
N
H
Figure 1. (A) Structural formulae for budesodine (1), prednisolone (2); (B) the two-step azoreduction/cyclization design for azoreductase activated drug release.
lactamization and drug release in the cyclization activated drug de-
sign. Therefore, we have investigated the applicability of the azo-
reductase/cyclization approach to budesonide exploring various
side chains including a trimethyl lock design (5), which has been
shown to accelerate intramolecular cyclization in analogous sys-
tems Figure 2.11
The synthetic approach to the prodrugs 3 and 4 is illustrated in
Schemes 1 and 2. The synthesis of the tert-butyl protected phenyl-
propionate linkers 6 and 7 was carried out as described in our pre-
vious work.12,13 These were attached to budesonide producing the
tert-butyl protected prodrugs which were deprotected with TFA in
DCM. The dimethyl compound 5 required the synthesis of 10
(highlighted in Scheme 3). This was accomplished by the conden-
sation of the nitrosoacid 15 and the tert-butyl ester of 5-ASA 18 un-
der acidic conditions. The nitroso acid 15 was produced in a four
step sequence from methyl 2-nitrophenyl acetate (16). This was
dimethylated using NaH in dry DMF with MeI as alkylation agent.
A hydrolysis in basic conditions released the free acid 13. Reduc-
tion under H2 atmosphere with Pd/C (10%) was carried out in 2%
aqueous NaOH to yield the intermediate 15. This was oxidised
without purification using OxoneÒ. It was necessary to perform
the reduction under basic conditions in order to inhibit cyclization
of intermediate 14. The aminosalicylate 18 was produced from 16
following tertbutyl protection and reduction. The condensation be-
tween 15 and 18 was performed in glacial acetic acid as solvent at
room temperature. tert-Butyl protection was chosen for the 5-ASA
component, mindful that there would be competition between the
steroid and the salicylate -OH during esterification later in the se-
quence. Mitsunobu conditions were used to attach 10 to budeso-
dine at the primary 21-OH.14 Finally a common step for the three
budesodine prodrugs involved a rapid removal of the tert-butyl
group using TFA. The prodrugs were produced as diastereomeric
mixtures with two HPLC peaks due to the epimeric R/S C22 centre
of budesonide. Compound 3 was obtained as a bright orange so-
lid.15 Compound 20 (Fig. 3) was produced by esterification of 1
with the appropriate BOC amino acid using the Steglich reaction.
This was used following deprotection as illustrated in Figure 3, to
estimate the rate of cyclization of the key anilido ester 21.
O
O
O
O
O
HO
O
O
HO
O
N
O
N
O
N
N
O
O
HO
O
3
4
OH
O
OH
O
O
O
HO
O
N
O
N
O
5
OH
HO
O
Figure 2. Structural formulae of budesonide prodrugs. 5-ASA-based compound, 3; PABA-based compound 4, and trimethylock system applied to 5-ASA conjugate (5).

J. F. Marquez Ruiz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7573–7577
7575
O
O
OH
O
HO
a
O
b
3
N
O
O
N
+
1
N
N
O
O
O
HO
6
8
O
OH
O
Scheme 1. Synthetic route to 3. Reagents: (a) 2, DMAP, DCC in THF; (b) TFA, DCM.
O
O
O
O
+
a
O
HO
b
O
O
4
N
N
O
N
N
1
7
O
O
9
OH
O
Scheme 2. Synthetic route to 4. (a) 2, DMAP, DCC in THF; (b) TFA, DCM.
NO2
NH2
NO2
NO
NO2
d
OH
OH
O
OH
O
c
b
O
O
O
a
O
O
11
15
12
13
14
intermediate
NO2
OH
NH2
OH
NO2
18
17
16
e
f
OH
O
O
O
O
OH
O
OH
g
O
N
10
+
18
15
N
O
O
OH
1
h
O
O
O
HO
O
i
N
O
N
5
O
19
O
OH
O
Scheme 3. Synthetic route to 5. Reagents and Conditions: (a) CH3I, 15-crown-5, in DMF; (b) NaOH, reflux; (c) Pd/C, H2 in NaOH 2 N; (d) OxoneÒ; (e) DCC in THF, DMAP in t-
butanol; (f) Pd/C, H2; (g) AcOH; (h) 3, Ph3P, DIAD in THF; (i) TFA, DCM.
An important parameter influencing the potential of the 3–5 to
deliver budesodine into the colon is the stability towards hydroly-
sis under conditions found in the GIT. Stability was tested by incu-
bation in simulated duodenal and gastric juices (37 °C).16 The
prodrugs 3, 4 and 5 were unchanged over 24 h indicating that they
have the capacity to transit the GIT to the colon intact. The aqueous
stability of the budesonide analogs was greater than the predniso-
lone analogs developed previously, probably due to increased ste-
ric hindrance of the ester group at C-21. The critical aspect of this
study was the evaluation of the cyclization of the amino ester
products of azoreductase activity on 3 given the steric hindrance
of the C21 ester group in the budesonide case relative to that of
prednisolone.18
The lactamization of 21 was studied in the pH range 2.5–10
(37 °C) and at I = 0.12, using HPLC to measure the disappearance
of the parent and the appearance of 1 and dihydroquinolone
(DHQ) 22 (Fig. 3). Lactamization was initiated following deprotec-
tion of 20. The deprotection was carried out in DCM at 0 °C with

7576
J. F. Marquez Ruiz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7573–7577
O
NHBOC
O
O
HO
O
O
20
O
Boc deprotection
O
NH2
O
22
O
1
+
HO
O
Intramolecular
Lactamization
N
O
O
H
21
O
Figure 3. Intramolecular lactamization of 21 was studied in aqueous buffer (37 °C) following deprotection of 20 with TFA in DCM.
A
B
0.100
0.075
0.050
0.025
0.000
0.04
0.03
0.02
0.01
0.00
k
0
10 20 30 40 50 60 70 80 90 100
Time (min)
2
3
4
5
6
7
8
9
10
pH
Figure 4. (A) Typical cyclization progress curve for pH-rate profile for the lactamization of 21 (d) at pH 3, 37 °C and I = 0.15 with formation of budesonide (.) and DHQ (22)
(N); (B) shows the pH rate profile for that event. kobs values at each pH were estimated by extrapolation to zero buffer at constant ionic strength (n = 4).
B
C
A
13
12
11
10
9
15.0
12.5
10.0
7.5
14
12
10
8
8
7
6
6
5
5.0
4
4
3
2
2.5
2
1
0
0.0
0
0
4
8
12
16
20
24
0
4
8
12
16
20
24
0
4
8
12
16
20
24
Time (h)
Time (h)
Time (h)
Figure 5. Progress curves following incubation in a suspension of Clostridium perfringens at 37 °C (pH 7.4). (A) 3 (d); (B) 4 (ꢀ) and (C) 5 (j) and evolution of budesonide (N),
and corresponding linker byproducts, DHQ (22) (.); 3,3-dimethyl-1,3-dihydro-indol-2-one (h).
TFA for several minutes, the volatiles removed under a stream of
N2 and the residue dissolved immediately in the appropriate buffer
solution before starting kinetic studies. The process which is buffer
salt concentration dependent was repeated at several buffer con-
centrations and the rate at zero pH modifier obtained by extrapo-
lation. The reaction followed well defined pseudo first-order
kinetics across the pH range with the exclusive production of 1
and 22. The t1/2 for cyclization of 21 at pH 2.5 was 25.9 min; at
pH 3.44 the t1/2 was 29.3 min suggesting that the release of budes-
odine will occur at feasible rates where pH is low. The t1/2 mea-
surement for pH 5 and 5.5 were 52.7 and 73.2 min. Finally in the
pH range 7–7.5 the half-life was 201–222 min. At constant pH
(7.4), in the presence of MeOH, the cyclization rate was indepen-
dent of ionic strength in the range 0.12–0.6. An example of a curve
for the disappearance of 21 and production of 1 is shown in Figure 4
alongside the pH rate-profile for this event (Fig. 4B).
Lactamization is expected to be triggered following azoreduc-
tase activity which is likely to take place in the presence of the co-
lonic microflora. The pH conditions for these areas of the GIT for
patients in UC and CD have been studied. In the case of UC the pH
values in stomach, proximal small intestine and ileum are similar
in IBD patients and healthy subjects but the pH values in the colon
areas is lower in many IBD patients, especially when the disease is
active, (pH values for the right colon/caecum as low as 2.3–3.4 have
been measured).17 Paradoxically, while 3 exhibited greater stability
than the prednisolone prodrug in simulated intestinal fluid, the
rates of cyclization of the corresponding anilide esters was higher
in the budesonide case at the relevant pH range of 5–5.5.

J. F. Marquez Ruiz et al. / Bioorg. Med. Chem. Lett. 22 (2012) 7573–7577
7577
The capacity of the human colonic microflora to reduce the azo
References and notes
link of the prodrugs is essential to release budesodine from the
prodrugs because without reduction, the intermediates which trig-
ger the intramolecular lactamization would not be formed. There-
fore compounds 3–5 were incubated in the presence of the
azoreductase producing anaerobe Clostridium perfringens which is
generally present in the human intestinal tract modelled by assay
conditions that we described recently.18
1. Blonski, W.; Buchner, A. M.; Lichtenstein, G. R. Curr. Opin. Gastroenterol. 2011,
27, 346.
2. Mulder, C. J.; Tytgat, G. N. Aliment. Pharmacol. Ther. 1993, 7, 125.
3. Clissold, S. P.; Heel, R. C. Drugs 1984, 28, 485.
4. Nolen, H., 3rd; Fedorak, R. N.; Friend, D. R. J. Pharm. Sci. 1995, 84, 677.
5. Patel, N. V.; Patel, J. K.; Shah, S. H.; Patel, J. N. Pharmazie 2011, 66, 124.
6. Varshosaz, J.; Ahmadi, F.; Emami, J.; Tavakoli, N.; Minaiyan, M.; Mahzouni, P.;
Dorkoosh, F. Int. J. Prev. Med. 2010, 1, 115.
7. Cui, N.; Friend, D. R.; Fedorak, R. N. Gut 1994, 35, 1439.
8. Varshosaz, J.; Emami, J.; Fassihi, A.; Tavakoli, N.; Minaiyan, M.; Ahmadi, F.;
Mahzouni, P.; Dorkoosh, F. Int. J. Colorectal Dis. 2010, 25, 1159.
9. Varshosaz, J.; Emami, J.; Tavakoli, N.; Fassihi, A.; Minaiyan, M.; Ahmadi, F.;
Dorkoosh, F. Int. J. Pharm. 2009, 365, 69.
10. Ruiz, J. F.; Radics, G.; Windle, H.; Serra, H. O.; Simplicio, A. L.; Kedziora, K.;
Fallon, P. G.; Kelleher, D. P.; Gilmer, J. F. J. Med. Chem. 2009, 52, 3205.
11. Greenwald, R. B.; Choe, Y. H.; Conover, C. D.; Shum, K.; Wu, D.; Royzen, M. J.
Med. Chem. 2000, 43, 475.
12. Márquez Ruiz, J. F.; Gilmer, J. F.; Kelleher, D. U.S. and Ireland Patent, Ireland
2007/0475, 2007.
Compound 3 was an excellent substrate for the azoreductase
enzymes of C. perfringens, as shown in Figure 5 (A) releasing the ex-
pected amounts of 1. Compound 3 was completed reduced after
6 h, with no sign of hydrolysis, which would have produced the
acid. Compound 4 was not reduced and did not produce the parent
or linker (panel B). Instead when the samples were analysed by
HPLC an unknown peak close in retention time to 4 appeared
and decayed over time. We suspect that this was the hydrazine
intermediate of azoreduction which underwent re-oxidation pro-
ducing 4. Finally when 5 was incubated under similar conditions
(Panel C), the azo linkage of the prodrug was reduced releasing
budesodine and the expected cyclization product. However, pro-
cessing of the dimethyl compound was not sufficiently faster than
the phenylethyl linker compound to justify its longer synthesis.
We have shown that the two-step azoreductase/cyclization de-
sign developed for prednisolone can be successfully applied to the
more sterically hindered budesonide. Compound 3 exhibited excel-
lent aqueous stability but was shown to be capable of undergoing
rapid anaerobe-mediated activation and cyclization releasing
budesonide.
13. Ruiz, J. F.; Kedziora, K.; Keogh, B.; Maguire, J.; Reilly, M.; Windle, H.; Kelleher, D.
P.; Gilmer, J. F. Bioorg. Med. Chem. Lett. 2011, 21, 6636.
14. Mitsunobu, O. Y.; Bull, Y. Chem. Soc. Jpn. 1967, 40, 2380.
15. Data for 3 mp 136–138 °C. IR (KBr): 3468.24, 1618.44 and 1384.43 cmꢀ1 1H
.
NMR d (DMSO):17.58 (1H, s), 8.31 (1H, d, J 32.12 Hz), 8.07 (1H, d, J 7.52), 7.80
(1H, s), 7.55 (1H, s), 7.24 (1H, d, J 7.04 Hz), 6.79 (1H, d, J 8.04 Hz), 6.18 (1H, d, J
9.56 Hz), 5.92 (1H, s). 13C NMR d (DMSO):203.36–201.77 (C@O, C-20), 185.16
(C@O, C-3), 171.62–170.83 (C@O, C-26), 169.20 (C@O, C-130), 168.55 (C, C-5),
162.45 (C, C-100), 156.58 (CH, C-1), 149.74 (C, C-20), 143.08 (C, C-70), 138.66 (C,
C-10), 130.37 (CH, C-50), 129.98 (CH, C-60), 128.54 (CH, C-80), 128.14 (CH, C-2),
127.96 (CH, C-40), 127.25 (CH, C-120), 121.68 (CH, C-4), 119.31 (CH, C-110),
116.35 (CH, C-30), 115.04 (C, C-90), 107.35–103.70 (CH, C-22), 98.00–97.23 (C,
C-17), 82.91–81.18 (CH, C-16), 69.70 (CH, C-11), 68.11–67.11 (CH2, C-21),
54.99 (CH, C-9), 52.09 (CH, C-14), 49.43 (C, C-10), 43.69 (C, C-13), C-23 masked
by DMSO, 36.39 (CH2, C-27), 35.27–34.42 (CH2, C-12), 33.87–33.51 (CH2, C-6),
33.34–33.01 (CH2, C-7), 30.56–29.91 (CH, C-8), 26.47 (CH2, C-28), 20.82 (CH3,
C-19), 16.91 (CH2, C-15), 16.41 (CH2, C-24), 16.83 (CH3, C-18), 13.87–13.83
(CH3, C-25). HRMS: Found: (MꢀH)+ = 727.3231. Required: (MꢀH)+ = 727.3203.
16. Roger, E.; Lagarce, F.; Benoit, J. P. Int. J. Pharm. 2009, 379, 260.
17. Fallingborg, J.; Christensen, L. A.; Jacobsen, B. A.; Rasmussen, S. N. Dig. Dis. Sci.
1993, 38, 1989.
Acknowledgments
This work was supported by the Enterprise Ireland and OPSONA
Therapeutics LTD under the Innovations Partnership Program (IP/
2007/503).
18. Ruiz, J. F.; Kedziora, K.; Windle, H.; Kelleher, D. P.; Gilmer, J. F. J. Pharm.
Pharmacol. 2011, 63, 806.