Novel derivatives of ISO-1 as potent inhibitors of MIF biological function

Article abstract of DOI:10.1016/j.bmcl.2009.06.052

A series of novel 1,2,4-oxadiazole, phthalimide, amide and other derivatives of ISO-1 were synthesized and probed for inhibition of macrophage migration inhibitory factor (MIF) activity. Several compounds inhibited MIF enzymatic activity at levels better

Full text of DOI:10.1016/j.bmcl.2009.06.052

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4773–4776
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel derivatives of ISO-1 as potent inhibitors of MIF biological function
a
a
a
b
Sarala Balachandran a, , Atish Rodge , Pradip K. Gadekar , Vitthal N. Yadav , Divya Kamath ,
*
Anshu Chetrapal-Kunwar b, Pooja Bhatt b, Shaila Srinivasan b, Somesh Sharma a,b
,
Ram A. Vishwakarma a, Nilesh M. Dagia b
a Department of Medicinal Chemistry, Piramal Life Sciences, 1 Nirlon Complex, Off Western Express Highway, Goregaon (E), Mumbai 400 063, India
b Department of Pharmacology, Piramal Life Sciences, 1 Nirlon Complex, Off Western Express Highway, Goregaon (E), Mumbai 400 063, India
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel 1,2,4-oxadiazole, phthalimide, amide and other derivatives of ISO-1 were synthesized
and probed for inhibition of macrophage migration inhibitory factor (MIF) activity. Several compounds
inhibited MIF enzymatic activity at levels better than ISO-1. Of note, compounds 7, 22, 23, 24, 25 and
27 inhibited the spontaneous secretion/release/recognition of MIF from freshly isolated human periphe-
ral blood mononuclear cells and, more importantly, inhibited the MIF-induced production of interleukin-
6 (IL-6) and/or interleukin-1b (IL-1b) significantly better than ISO-1.
Received 18 March 2009
Revised 21 May 2009
Accepted 15 June 2009
Available online 17 June 2009
Keywords:
Oxadiazole
Phthalimide
Amide
Ó 2009 Elsevier Ltd. All rights reserved.
MIF
ISO-1
Anti-inflammatory
A growing body of evidence from in vitro, in vivo and, most
importantly, clinically relevant studies implicates the pleiotropic
pro-inflammatory cytokine macrophage migration inhibitory factor
(MIF) in the pathogenesis of a variety of inflammatory and autoim-
mune diseases.1,2 Therefore, a promising therapeutic approach to
diminish pathological inflammation is to inhibit the production
and/orbiologicalactivityof MIF. GiventhatMIF possesses enzymatic
properties,1,2 combined with the observations that inhibiting the
catalytic site of MIF leads to a marked reduction in the biological
function of MIF,3 an attractive pharmacological strategy to inhibit
inflammation is to target the catalytic pocket of MIF.2
to synthesize fluorinated analogs of ISO-1 introducing heterocycle
on isooxazole ring and/or making reverse amides. This led to the
synthesis of novel series of 1,2,4-oxadiazoles, phthalimide and
amide derivatives of ISO-1.
The key intermediates 2-(3-(-3-fluoro-4-methoxyphenyl)-4,5-
dihydroisoxazole-5-yl)-N0-hydroxyacetimidamide (5) and 3-(-3-flu-
oro-4-methoxyphenyl)-N0-hydroxy-4,5-dihydroisoxazole-5-carbox-
imidamide (9) required for synthesis of the 1,2,4-oxadiazoles were
synthesized utilizing a four-step reaction starting with 3-fluoro-4-
methoxy benzaldehyde.5,6 3-Fluoro-4-methoxy benzaldehyde (1)
was treated with hydroxylamine hydrochloride in the presence of tri-
ethyl amine to yield the desired oxime (2)4 in 95% yield. The oxime (2)
was further treated with N-chlorosuccinimide in DMF resulting in the
formation of corresponding chloro-oxime (3), which was treated
in situ with allyl cyanide in presence of triethyl amine to yield 2-(3-
(3-fluoro-4methoxyphenyl)-4,5-dihydroisoxazol-5-yl) acetonitrile
(4)in85%yield. Compound4wasfurtherreactedwithhydroxylamine
hydrochloride in the presence of sodium carbonate in ethanol–water
to yield 73%of the desiredkey intermediate55 (Fig. 2). With theobjec-
One of the earliest reported small molecule catalytic MIF inhib-
itor was from the isoxazole series.3 ISO-1, the pseudo-lead mole-
cule from this series (Fig. 1), inhibits the macrophage release of
TNF-a from LPS-stimulated mice and is moderately protective in
a clinically relevant model of sepsis when administered intraperi-
toneally.3 In a quest towards increasing the potency of ISO-1, mod-
ifications were carried out on the ISO-1 scaffold. Synthesis and
biological activity evaluation of derivatives of ISO-1 revealed that
mono fluorination ortho to the phenolic group of ISO-1 increased
the activity profile of the analogs4 and that amides of ISO-1 were
comparatively less active at inhibiting dopachrome tautomerase
activity.4 Given the observations of this earlier study, we decided
O
O
N
O
HO
Figure 1. ISO-1, one of the earliest reported MIF inhibitor.
* Corresponding author.
E-mail address: sarala.balachandran@piramal.com (S. Balachandran).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.06.052

4774
S. Balachandran et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4773–4776
HO
HO
O
N
O
N
CN
F
O
N
N
F
N
N
OH
F
N
O
CF3
O
N
F
F
F
F
N
O
O
N
e
a
f
b
Cl
c
d
O
O
CF3
NH2
HO
O
O
O
N
O
1
7
2
3
6
4
5
Figure 2. Synthesis of compounds 2–7. Reagents and conditions: (a) NH2OHꢀHCl/Et3N/CH2Cl2, rt, 15–16 h, 95%; (b) NCS/DMF, rt, 4–5 h; (c) allyl cyanide/Et3N, rt, 24–26 h,
85%; (d) NH2OHꢀHCl, Na2CO3/EtOH, H2O, rt, 6 h, 73%; (e) trifluoroacetic anhydride/pyridine, 100–120 °C, 8 h, 60%; (f) BF3–DMS/DCM, 78 °C, 15–16 h, 41%.
HO
HO
N
N
10 : R= CH3, R1 = CH3
14 : R= CH2CH2CONHC6H11 R1 = H
15 : R= CH2CH2COOCH3, R1 = H
16 : R= C6H4NCH3COOH,R1= CH3
17 : R= C6H3(OCH3)2, R1 = CH3
17i : R= C6H3(OH)2, R =1H
O
O
N
O
N
N
N
F
10i : R= CH3, R1 = H
O
CN
Cl
F
F
F
a
c
11 : R=CF3 ,R1 =CH3
b
NH2
N
R
O
R1
11i : R= CF3, R1 = H
O
O
O
12 : R= CH2CH2COOH, R1 = CH3
13 : R= CH2CH2CONHC6H11 R1 = CH3
9
3
8
Figure 3. Synthesis of compounds 8–17i. Reagents and conditions: (a) acrylonitrile/Et3N, rt, 24–26 h, 93%; (b) NH2OHꢀHCl, Na2CO3/EtOH, H2O, rt, 6 h, 82%; (c) (10) acetic
anhydride, pyridine, 100–120 °C, 8 h, 64%; (10i) compound 10, BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 53%; (11) compound 9, trifluoroacetic anhydride, pyridine, 100–120 °C, 8 h,
61%; (11i) compound 11, BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 53%; (12) compound 9, succinic anhydride/pyridine, 100–120 °C, 8 h, 64%; (13) compound 12, SOCl2, 40 °C, 1 h,
Et3N, cyclohexylamine, rt, 8 h, 80%; (14) compound 13, BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 53%; (15) compound 12, BBr3/DCM, ꢁ78 °C, 15–16 h, 60%; (16) compound 9, N-
methylisatoic anhydride/pyridine, 100–120 °C, 8 h, 67%; (17) compound 9, CDI/DMF, 2-3-dimethoxy benzoic acid, rt, 1–2 h, 80–100 °C, 6–7 h, 75%; (17i) compound 17, BF3–
DMS/DCM, ꢁ78 °C, 15–16 h, 56%.
tive of introducing a heterocycle on isoxazole ring, initially, interme-
diates (5 and 9) were reacted with different anhydrides (trifluoroace-
tic anhydride, acetic anhydride, succinic anhydride, N-methylisatoic
anhydride)usingpyridineat80–110 °Ctoyieldsubstituted1,2,4-oxa-
diazoles (6, 10, 11 and 16, respectively) (Figs. 2 and 3). In an alterna-
tive approach (to introduce a heterocycle on isoxazole ring),
intermediate 9 was reacted with 2,3-dimethoxy benzoic acid using
CDI in DMF at 80–100 °C for 6–7 h. to afford 17 in 75% yield. To dissect
the contributionof ‘methylation’to theobserved MIF inhibitoryactiv-
ity, demethylated analogs of several compounds were synthesized
(e.g.,10i, 11i, 14, 15, 17i)usingBF3–DMS.Thereductionofnitrilecom-
pounds 4 and 8 was carried out using hydrogenation in presence of
10% Pd–C in methanol which resulted in the isolation of amines 18i
and 18, respectively (Figs. 4 and 5). Subsequently, the amide deriva-
tives of ISO-1 were synthesized by treating these isoxazoline amines
with different acyl or aryl acid halides (pivaloyl chloride, isovaleryl
chloride and N,N-dimethlycarbamoyl chloride) in the presence of
triethylamine. After demethylation the reverse amides (19–25) were
obtained (Figs. 4 and 5). The phthalimide derivatives of ISO-1 (26–29)
were synthesized using oxime 2 in cycloaddition reaction with vari-
ous N-alkene phthalimides in the presence of NCS or sodium hypo-
chlorite (Fig. 6) followed by demethylation. The structures of
various synthesized compounds were assigned on the basis of differ-
ent spectral data (experimental details are provided in Supplemen-
tary data).
We investigated the MIF inhibitory potential of the synthesized
compounds and, in parallel, compared their activity with ISO-1.
Initially, dopachrome tautomerase assays were performed. Several
compounds from both 1,2,4-oxadiazole series (e.g., 7, 12, 13, 14, 17
and 17i) and phthalimide and amide derivatives of ISO-1 (e.g., 22,
O
N
O
O
N
N
19 : R= C(CH3)3
20 : R= CH2CH(CH3)2
21 : R= N(CH3)2
CN
F
F
F
O
NH2
a
HN
b
HCl
O
R
O
HO
18
8
Figure 4. Synthesis of compounds 18–21. Reagents and conditions: (a) H2, Pd/C, MeOH, rt, 2 h, 63% (b) (19) (i) pivaloyl chloride, Et3N, THF, rt, 2 h; (ii) BF3ꢀDMS/DCM, ꢁ78 °C,
15–16 h, 37%; (20) isovaleryl chloride, Et3N, THF, rt, 2 h; (ii) BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 47%; (21) (i) N,N-dimethyl carbamoyl chloride, Et3N, THF, rt, 2 h; (ii) BF3–DMS/
DCM, ꢁ78 °C, 15–16 h, 18%.
O
R
22 : R=C (CH3)3
23 : R= CH2CH(CH3)2
24 : R= CF3
O
N
O
N
CN
O
NH2
HCl
N
N
H
F
F
F
a
b
25 : R= FC6H4
O
HO
O
18i
4
Figure 5. Synthesis of compounds 18i–25. Reagents and conditions: (a) H2, Pd/C, MeOH, rt, 2 h, 60% (b) (22) (i) pivaloyl chloride Et3N, THF, rt, 2 h; (ii) BF3–DMS/DCM, ꢁ78 °C,
15–16 h, 18%; (23) (i) isovaleryl chloride, Et3N, THF, rt, 2 h; (ii) BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 22%; (24) (i) trifluoro acetic anhydride, Et3N, THF, rt, 2 h; (ii) BF3–DMS/DCM,
ꢁ78 °C, 15–16 h, 9%; (25) (i) 4-fluoro, benzoyl chloride, Et3N, THF, rt, 2 h; (ii) BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 8%.
O
O
O
HO
B
N
A
N
O
N
N
N
O
n= 2
F
O
F
n
n
O
N
O
N
n
O
a
F
O
b
O
HO
26 : n= 1
27 : n= 2
28 : n= 3
29 : n= 2
2
HO
Figure 6. Synthesis of compounds 26–29. Reagents and conditions: (a) (i) NCS, DMF, A, Et3N or sodium hypochlorite (bleach), THF, rt, 48 h; (ii) BF3–DMS/DCM, ꢁ78 °C, 15–
16 h, 16%; (b) NCS, DMF, B, Et3N, rt, 15 h; (ii) BF3–DMS/DCM, ꢁ78 °C, 15–16 h, 48%.

S. Balachandran et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4773–4776
4775
Table 1
Table 3
Potency of synthesized 1,2,4-oxadiazoles to inhibit MIF enzymatic activity in
tautomerase assay
Activity of synthesized compounds to inhibit the secretion/release/recognition of MIF
in ELISA assays using hPBMCs
Compd
IC50
(
lM)
Compound
IC50 (lM)
THP-1 cell lysates
Purified human MIF
ISO-1
7
25
15
60
ISO-1
6
7
69
>100
48
>100
60
100
60
12
30
25
50
>100
55
60
ND
>100
95
>100
>100
9.5
>100
>100
ND
12
13
14
15
17
17i
22
23
24
25
26
27
28
29
Not active
Not active
Not active
Not active
Not active
15
20
15
10
Not active
1.5
Not active
20
10
10i
11
11i
12
13
14
15
16
17
17i
>100
20
19
32
>100
ND, not done.
the secretion/release/recognition of MIF from hPBMCs at levels
equal to or better than ISO-1 (Table 3).
23, 24, 25, 26, 27, 28 and 29) consistently inhibited the dopa-
chrome tautomerase activity in THP-1 lysates at levels better than
that of ISO-1 (Tables 1 and 2). Only compounds that inhibited the
tautomerase activity in THP-1 lysates at levels equal to or better
than ISO-1 were further screened for their enzyme inhibition
potential using purified recombinant human MIF. Again, com-
pounds 7, 13, 17, 22, 23, 25, 26, 27, 28 and 29 inhibited the tautom-
erase activity of purified recombinant human MIF at levels better
than that of ISO-1 (Tables 1 and 2). These latter results (with puri-
fied human MIF) confirm the specificity of these novel 1,2,4-oxadi-
azoles, phthalimide and amide derivatives of ISO-1 to inhibit MIF
enzymatic activity. Note that the difference in the observed and
reported3,4 IC50 value for ISO-1 in dopachrome tautomerase assay
may be due to the different concentrations of purified human
MIF utilized in the two studies.
In spite of the growing interest in inhibiting MIF activity via its
catalytic pocket, the physiological relevance of the enzymatic
activity of MIF is yet to be elucidated.1,2 Based on our recent find-
ings,7 we sought to find compounds that inhibit not only the enzy-
matic activity of MIF but also (more importantly) the secretion/
recognition and biological function of MIF. To this end, initially,
only compounds that suppressed the enzymatic activity of MIF at
levels equal to or better than that of ISO-1 were evaluated for their
potential to inhibit the secretion/recognition of MIF from human
peripheral blood mononuclear cells (hPBMCs). One compound
from 1,2,4 oxadiazole series (i.e., 7) and several phthalimide and
amide derivatives of ISO-1 (e.g., 22, 23, 24, 25, 27 and 29) inhibited
The secretion of MIF from THP-1 cells is known to occur via a
non-classical pathway.8 Reagents modulating the ABC transporter,
but not protein synthesis inhibitors such as cycloheximide, are
known to inhibit the secretion of MIF.8 Whether ISO-1 and 7, 22,
23, 24, 25, 27, and 29 work in a similar manner and inhibit the
secretion of MIF by affecting the ABC transporter is currently
unknown and warrants further investigation. However, the data
generated from our recent study7 and earlier studies9 are consis-
tent with the hypothesis that the inhibitory activity of ISO-1 and
novel 1,2,4-oxadiazoles, phthalimide and amide derivatives of
ISO-1 in cell-based ELISA assays is, at least in part, due to these
compounds binding directly to MIF and preventing its recognition.
Since several of these compounds (e.g., 7, 22, 23, 24, 25, 27, and 29)
have a similar or lower IC50 for MIF inhibition in cell-based ELISA
assays (in comparison to ISO-1), it would be of interest to deter-
mine whether these compounds have an equivalent or stronger
affinity to interact with MIF in comparison to ISO-1.
It is well-known that tautomerase activity inhibitors down-reg-
ulate the production of pro-inflammatory cytokines.3 Given that
MIF induces the expression of pro-inflammatory cytokines,1,2 we
investigated if these compounds inhibit the purified human MIF-
induced expression of TNF-a, IL-6 and IL-1b from hPBMCs. Only
compounds that inhibited the enzymatic activity of MIF and also
inhibited the secretion/release/recognition of MIF (at levels equal
to or better than ISO-1) in ELISA assays were tested in these assays.
Pre-treatment of MIF with compounds 7, 22, 23, 24, 25 and 27 led
to a marked suppression in the induced production of IL-6 and/or
IL-1b which was significantly better than ISO-1 (Table 4). Pre-treat-
Table 2
Potency of synthesized phthalimide and amide derivatives of ISO-1 to inhibit MIF
enzymatic activity in tautomerase assay
Table 4
Compd
IC50 (lM)
Potency of synthesized compounds to inhibit MIF-induced cytokine production from
hPBMCs
THP-1 cell lysates
Purified human MIF
ISO-1
19
20
21
22
23
24
25
26
69
>100
ND
ND
ND
22
22
>100
20
Compd
(% Inhibition at 100 l
M)a
>100
>100
>100
28
29
49
15
29
0.5
2
TNF-
a
IL-1b
IL-6
ISO-1
7
25 25
34 8.5
40 10
25 2.5
46 0*
50 16
58 15
66 3.5*
52 8*
0
0
33 5.5*
39 8.5*
47 5.5*
49 13
54 0.5*
22 2*
14 14
22
23
24
25
27
29
53
53 12
56
31 11
39
6
36
0.8
7
0
27
28
22
2
9
27 26.5
29
15
50
a
Values are means S.E.M. from at least two separate experiments.
*
ND, not done.
Indicates p < 0.05 compared to ISO-1 control.

4776
S. Balachandran et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4773–4776
ment of MIF with 29 led to a relatively moderate, albeit similar to
ISO-1-mediated, suppression in the induced production of TNF-
efficacious in blocking the secretion of MIF and/or the biological
function of MIF. It is tempting to speculate that the conformational
changes of MIF caused by inhibitors binding to the enzymatic
pocket9 may contribute to differential potencies in biological
assays. Clearly, future studies are warranted to completely under-
stand the relationship between potential interactions (of various
compounds) with the binding pocket of MIF and the resultant
effects on biological function of MIF.
a,
IL-6 and IL-1b (Table 4). These data clearly demonstrate that, at
least a few, 1,2,4-oxadiazoles and phthalimide and amide deriva-
tives of ISO-1 inhibit the biological function of MIF at levels better
than ISO-1. Collectively, these results provide evidence that novel
derivatives of ISO-1 not only inhibit the enzymatic activity of
MIF, but also, more importantly, inhibit the secretion/release/rec-
ognition and biological function of MIF.
In summary, a number of derivatives of ISO-1 were synthesized
and screened for MIF inhibitory activity. In the enzymatic assays, it
was observed that the demethylated analogs were more active
than the corresponding methoxy analogs (e.g., 11i vs 11). In prior
studies in the literature, the X-ray crystal structure of MIF and
binding domains have been discussed.3 To diversify our chemistry,
we choose to complement and extend the findings of those prior
studies. In all cases the phenol part of the ISO-1 scaffold was kept
constant as the ortho fluoro phenol substituted isoxazoline was
found to be favorable for binding in the MIF active pocket. Upon
making modifications to the ester moiety of the ISO-1 scaffold, it
is evident that flexibility is preferred in this region. Indeed, potent
MIF inhibitory activity is observed with the phthalimide and amide
derivatives of ISO-1. In contrast, MIF inhibitory activity is relatively
absent from the heterocyclic derivatives. Of note, as would be
hypothesized, variations in chain length give rise to variation in
MIF inhibitory activity of synthesized compounds. It appears from
our data that a chain length of n = 2 is optimum for MIF inhibitory
activity (e.g., 23 and 25).
Acknowledgment
We are thankful to Professor John R. David (Harvard School of
Public Health) for critical review of this manuscript and insightful
comments.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.06.052.
References and notes
1. Baugh, J. A.; Bucala, R. Crit. Care Med. 2002, 30, S27.
2. Morand, E. F.; Leech, M.; Bernhagen, J. Nat. Rev. Drug Discovery 2006, 5, 399.
3. Al-Abed, Y.; Dabideen, D.; Aljabari, B.; Valster, A.; Messmer, D.; Ochani, M.;
Tanovic, M.; Ochani, K.; Bacher, M.; Nicoletti, F.; Metz, C.; Pavlov, V. A.; Miller, E.
J.; Tracey, K. J. J. Biol. Chem. 2005, 280, 36541.
4. Cheng, F. K.; Al-Abed, Y. Bioorg. Med. Chem. Lett. 2006, 16, 3376.
5. Leite, F. C. C. L.; Ramos, N. M.; Silva, B. P. J.; Miranda, L. P. A.; Fraga, A. M. C.;
Barreiro, J. E. II Farmaco 1999, 54, 747.
Interestingly, several compounds from 1,2,4 oxadiazole series
(e.g., 13, 14, 17, 17i) as well as phthalimide and amide derivatives
of ISO-1 (e.g., 26 and 28) which inhibited the enzymatic activity of
MIF, did not inhibit the spontaneous secretion/release/recognition
of MIF from hPBMCs. Compound 29 inhibits the enzymatic activity
of MIF and also suppresses the spontaneous secretion/release/rec-
ognition of MIF from hPBMCs. However, 29 only moderately inhib-
its the biological function of MIF. Combined, the results of our
study reveal that not all tautomerase activity inhibitors are equally
6. Jakopin, Z.; Roskar, R.; Dolenc, S. M. Tetrahedron Lett. 2007, 48, 1465.
7. Dagia, N. M.; Kamath, D. V.; Bhatt, P.; Gupte, R. D.; Dadarkar, S. S.; Fonseca, L.;
Agarwal, G.; Chetrapal-Kunwar, A.; Balachandran, S.; Srinivasan, S.; Bose, J.; Pari,
K.; B-Rao, C.; Parkale, S. S.; Gadekar, P. K.; Rodge, A. H.; Mandrekar, N.;
Vishwakarma, R. A.; Sharma, S. Eur. J. Pharmacol. 2009, 607, 201.
8. Flieger, O.; Engling, A.; Bucala, R.; Lue, H.; Nickel, W.; Bernhagen, J. FEBS Lett.
2003, 551, 78.
9. Senter, P. D.; Al-Abed, Y.; Metz, C. N.; Benigni, F.; Mitchell, R. A.; Chesney, J.; Han,
J.; Gartner, C. G.; Nelson, S. D.; Todaro, G. J.; Bucala, R. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 144.