Receptor agonists of macrophage migration inhibitory factor

Article abstract of DOI:10.1016/j.bmcl.2010.09.118

The cytokine MIF is involved in inflammation and cell proliferation via pathways initiated by its binding to the transmembrane receptor CD74. MIF also promotes AMPK activation with potential benefits for response to myocardial infarction and ischemia-reperfusion. Structure-based molecular design has led to the discovery of not only antagonists, but also the first agonists of MIF-CD74 binding. The compounds contain a triazole core that is readily assembled via Cu-catalyzed click chemistry. The agonist and antagonist behaviors were confirmed via study of MIF-dependent ERK1/2 phosphorylation in human fibroblasts.

Full text of DOI:10.1016/j.bmcl.2010.09.118

Bioorganic & Medicinal Chemistry Letters 20 (2010) 7033–7036
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Receptor agonists of macrophage migration inhibitory factor
William L. Jorgensen a, , Sunilkumar Gandavadi a, Xin Du b, Alissa A. Hare a, Alexander Trofimov a, Lin Leng b,
⇑
Richard Bucala b,
⇑
a Department of Chemistry, Yale University, New Haven, CT 06520, USA
b Department of Medicine, Yale University School of Medicine, New Haven, CT 06520-8031, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The cytokine MIF is involved in inflammation and cell proliferation via pathways initiated by its binding
to the transmembrane receptor CD74. MIF also promotes AMPK activation with potential benefits for
response to myocardial infarction and ischemia-reperfusion. Structure-based molecular design has led
to the discovery of not only antagonists, but also the first agonists of MIF–CD74 binding. The compounds
contain a triazole core that is readily assembled via Cu-catalyzed click chemistry. The agonist and antago-
nist behaviors were confirmed via study of MIF-dependent ERK1/2 phosphorylation in human fibroblasts.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 3 August 2010
Revised 21 September 2010
Accepted 23 September 2010
Available online 29 September 2010
Keywords:
MIF
Cytokine
Agonists
Structure-based design
AMPK activation
ERK phosphorylation
Click chemistry
Macrophage migration inhibitory factor (MIF) is a pleiotropic
cytokine that is expressed in multiple cell types including macro-
phages, endothelial cells, T-cells, and cardiomyocytes. It is involved
in the pathogenesis of inflammatory diseases as well as in tumor
growth and angiogenesis. The potential biomedical significance
of MIF regulation is under active investigation.1–3 MIF signal trans-
duction is initiated by binding to a receptor complex consisting of
CD74 and CD44.4–6 Though emphasis has been placed on discovery
of antagonists of MIF signaling,1–3 report of agonists would be
valuable both as probes of MIF biology and in therapeutic indica-
tions. For example, MIF activation leads to an immune adjuvant
effect.7 Furthermore, it has recently been shown that MIF stimu-
lates the AMP-activated protein kinase (AMPK) in heart muscle.8
AMPK plays a central role in cardiac response to ischemia through
promoting glucose uptake and limiting myocardial injury and
apoptosis.9 Since AMPK deficiency is deleterious during ischemia-
reperfusion, MIF–CD74 agonism also provides a potential therapy
for acute myocardial ischemia via enhanced AMPK activation.8
Enhancement of MIF signaling and AMPK activity in limiting
cardiac damage may be especially beneficial for older patients.10
Besides being a cytokine, MIF shows enzymatic activity as a
keto-enol tautomerase, which is likely vestigial in mammals.11
The 114-residue MIF monomer associates to form a symmetrical
trimer with three tautomerase active sites (Fig. 1).2,12,13 Though
there is evidence that the MIF–CD74 contact occurs in the vicinity
⇑
Corresponding authors. Tel.: +1 203 432 6278; fax: +1 203 432 6299 (W.L.J.).
E-mail addresses: william.jorgensen@yale.edu (W.L. Jorgensen), richard.bucala@
Figure 1. 1GCZ crystal structure (1.9 Å) for the MIF trimer with three copies of
chromene derivative 1 (space-filling) in the active sites.
yale.edu (R. Bucala).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.09.118

7034
W. L. Jorgensen et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7033–7036
of the tautomerase active site,14 no crystal structure for a MIF–
CD74 complex has been reported. Nevertheless, the availability
of crystal structures for MIF has enabled structure-based discovery
of small-molecule MIF tautomerase inhibitors.1,3 In conjunction
with a protein–protein binding assay,4 it has also been possible
to discover inhibitors of the more biologically significant
MIF–CD74 complexation. Indeed, virtual screening by docking led
to our identification of 11 structurally diverse inhibitors of
Substituted 1,2,3-triazoles were pursued in four series corre-
sponding to structures 3, 4, 5, and 6. These represent two pairs
of isomers, which are either 4-hydroxyphenyl substituted at the
1- or 4-position and with either a benzyl or heteroaryl group at
the other site. The synthetic routes are summarized in Scheme 1.
For the Cu(I)-catalyzed 1,3-dipolar cycloadditions,19 a one-pot pro-
tocol utilizing terminal alkynes and in situ formation of aryl or
benzyl azides from the corresponding bromides or iodides was
highly effective.20 The typical procedure used 10 mol % CuI,
10 mol % sodium ascorbate, and 15 mol % trans-N,N0-dimethyl-
1,2-cyclohexanediamine as ligand at 25–75 °C under argon. Com-
pounds of types 3, 5, and 6 were readily accessible, while for 4
avoidance of benzyl acetylenes required a longer sequence featur-
ing a Grignard reaction, catalytic hydrogenation of the resulting
alcohol, and hydrolysis of the THP ether. Most reactions proceeded
in 70–100% yield. The identity of all assayed compounds was con-
firmed by 1H (Bruker DRX-500) and 13C NMR and HRMS; purity
was normally >95% as determined by reverse-phase HPLC.
MIF–CD74 binding with activities in the
l
M regime.15 Optimiza-
tion of one series, featuring a N-benzyl-benzoxazol-2-one core,
has provided compounds with IC50 values as low as 7.5 nM in the
tautomerase assay and 80 nM in the MIF–CD74 binding assay.16
In parallel with the docking investigations, de novo design also
was pursued using the program BOMB,17 which can build combina-
torial libraries of analogs starting from a core placed in a binding
site. Common features were apparent in available crystal struc-
tures for MIF complexed with the chromene analog 12,18 and the
dihydroisoxazole 2 ((R)-ISO-1),13 which are both reported to be
7 lM tautomerase inhibitors. The crystal structures feature a
The MIF–CD74 binding assay was performed as previously
presented.4,15 It features biotinylated MIF and immobilized CD74
ectodomain (CD7473–232) with streptavidin conjugated alkaline
phosphatase processing p-nitrophenylphosphate as the reporter.
Most compounds also were evaluated in a MIF tautomerase assay
using 4-hydroxyphenylpyruvate (4-HPP) as the substrate.15,21 Com-
pounds were tested at concentrations between 1 nM and 1 mM
using human MIF, which was recombinantly prepared.22
hydrogen bond between the phenolic OH of the inhibitors and
the side-chain CO of Asn97, which forms a backstop for the active
site. There is also the potential for additional hydrogen-bonding
with the NH of Ile64 and OH of Tyr95 as well as aryl–aryl interac-
tions with Tyr36, Tyr95, and Phe113.
O
O
O
The assay results are summarized in Table 1. In cases where
some inhibition is observed, but it plateaus without reaching the
50% level, the maximum percent inhibition is reported. Such max-
HO
O
O
N
O
HO
O
1
2
ima were normally reached below 0.5
lM in the binding assay and
Thus, using phenol as the core hydrogen-bonded to Asn97, li-
braries were constructed in the motif HOPhHetR where Het is a
5- or 6-membered heterocycle and R is a small substituent.
Roughly 100 choices for Het and 50 for R were initially considered
using BOMB. Among well-scoring alternatives for Het, pyrazoles
and triazoles were promising as were various aryl, CH2Ar, and
OAr options for R. Since facile synthesis of 1,2,3-triazole derivatives
was envisaged,19 they became the focus. For example, the dime-
thoxybenzyl derivative in Figure 2 could be constructed to yield
a striking complex including hydrogen bonds with Asn97 and
in the M range for the tautomerase assay. As discussed previ-
l
ously,15 a simple correlation between the results for the two assays
is not expected. Docking calculations for all of the complexes were
also carried out using Glide 5.5 in the extra precision (XP) mode.23
In previous studies, it was demonstrated that Glide XP scores cor-
related well with experimental log Ki or log IC50 values for com-
plexes of 10 known MIF tautomerase inhibitors.15,16 The most
favorable XP score, ꢀ9.4, for the known inhibitors is for a coumarin
derivative with a reported Ki of 38 nM,2 and inhibitors with Ki val-
ues near 1
l
M gave XP scores near ꢀ9.0. In comparing these results
Tyr95,
p
p
–p
interactions with Tyr95 and Phe113, and possible
to the XP scores in Table 1, most of the triazole derivatives were
cation–
and/or cation–ether interactions with Lys32.
N
N
N
N
N
N
HO
HO
N
Ar
Ar
Br
N
N
1. NaN3, CuI, ligand, 75 ºC,
DMSO/H2O (5:1), 45 min
5
6
+
+
2. PTSA, MeOH, 65 ºC, 3h
X
3
X
OH
H
OTHP
I
3. TBAF, THF,
N
N
N
1.
0 ºC-rt, 30 min
THPO
4. (COCl)2, DMSO,
TEA, -78 ºC, 30 min
O
OTBDMS
N
OTHP
HO
5. XPhMgBr,
THF, 0 ºC,
10 min
N
N
6. H2/60 psi, Pd(OH)2,
MeOH, rt, 48 h
7. PTSA, MeOH, 3 h
X
4
I
N
N
N
1, 2. As above. 74%
+
HO
N
N
5a
OTHP
Scheme 1. Synthesis of aryl-1,2,3-triazoles.
Figure 2. Computed structure for 3g bound to MIF from BOMB after energy
minimization using MCPRO and the OPLS/CM1A force field.

W. L. Jorgensen et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7033–7036
7035
Table 1
Activities for 1,2,3-triazole derivatives from the MIF–CD74 and tautomerase (4-HPP)
assays, and Glide XP scoresa
Compd
X or Ar
IC50 or max % inhib
XP Score
CD74
4-HPP
3a
3b
3c
3d
3e
3f
3g
3h
3i
4a
4b
4c
4d
4e
4f
4g
4h
4i
—
2-Me
3-Me
2-OMe
3-OMe
4-OMe
2,3-diOMe
3-COOMe
3-COOH
—
3-Me
2-OH
3-OH
4-OH
2-OMe
3-OMe
4-OMe
2,3-diOMe
3,4-diOMe
3-Pyridinyl
1-Naphthyl
4-Isoquinolinyl
3-Pyridinyl
1-Naphthyl
4-Isoquinolinyl
CH2-2-pyridinyl
CH2-3-pyridinyl
NA
NA
NA
Agon
50
NA
0.9
1.2
NA
ꢀ7.2
ꢀ7.0
ꢀ7.2
ꢀ6.9
ꢀ7.4
ꢀ7.2
ꢀ7.9
ꢀ7.3
ꢀ7.6
ꢀ6.7
ꢀ6.8
ꢀ7.0
ꢀ6.8
ꢀ6.7
ꢀ6.4
ꢀ6.5
ꢀ6.0
ꢀ7.1
ꢀ6.6
ꢀ6.5
ꢀ5.9
ꢀ6.0
ꢀ7.1
ꢀ6.4
ꢀ6.0
ꢀ7.3
ꢀ7.3
Agon
0.75
2.5
3.5
30%
9%
475
1000
NA
690
35%
40%
65
530
70
Agon
NA
23%
14%
8%
36%
NA
22%
Agon
NA
Agon
16%
23%
12%
30%
20%
Figure 4. Computed structure for 5a bound to MIF. Details as in Figure 2. Negligible
energetic preference is computed for the triazole ring being oriented up as shown or
down (rotated 180°).
4j
5a
5b
5c
6a
6b
6c
6d
6e
Agon
970
100 nM for 3d, 5a, and 5c, but not increasing significantly more
at higher concentrations.
32
N
N
N
N
N
N
HO
HO
a
IC50 in lM. NA = inactive. Agon = agonist. XP scores from Glide v. 5.5 using the
O
1GCZ structure for MIF.
3d
5c
N
It would be fascinating to know the structural origins of the
agonist behavior for the three compounds. Reasonable structures
can be built for them bound in the tautomerase active site, as in
Figure 4. For the tautomerase activity, it is possible that binding
in one active site could enhance allosterically the activity in an-
other site. It is also possible that the molecules bind to an alterna-
tive surface site that leads to the agonist character in both
assays.18,24 Experimental structural studies are clearly desirable.
It may be noted that small changes in a ligand have also been
shown recently to interconvert agonist and antagonist character
for the protein-protein association of Hsp40 and Hsp90.25
Finally, the potentially contrasting effects of the MIF agonists 3d,
5a, and 5c versus the MIF antagonist 3g on MIF signal transduction
were explored in human target cells. The extracellular-signal-regu-
lated kinase (ERK) pathway is of particular interest as aberrations in
it are associated with hyperproliferative diseases as well as inflam-
matory disorders. Up-regulation of MIF increases ERK1/2 phosphor-
ylation in a CD74-dependent manner.4 Specifically, human primary
not expected to be strong MIF tautomerase inhibitors. The excep-
tions turned about to be 3g and 3h, which yielded IC50 values of
0.75 and 2.5 lM. As illustrated in Figure 2, the binding of 3g to
MIF may be enhanced by coordination of the ammonium group
of Lys32 with the 2,3-dimethoxyl fragment of 3g.
Compounds 3g and 3h also emerged as the most potent antag-
onists in the MIF–CD74 binding assay with IC50 values near 1 lM.
However, the most remarkable discovery was that 3d, 5a, and 5c
increase the amount of MIF that remains bound to CD74 after
washing. The increase in bound MIF depends linearly on the log
of the agonist concentration. For 3d, there is a 40% increase in
bound MIF at 500 nM, while there are 100% increases for 5a
and 5c at ca. 100 nM and 150 nM, respectively. The contrasting
behaviors for the antagonist 3g and agonist 5a are striking in Fig-
ure 3. The three compounds were also agonists in the tautomer-
ase assay with the activity increasing by 31%, 23%, and 39% at
Figure 3. Results of the MIF–CD74 binding assays for the antagonist 3g (left) and agonist 5a (right).

7036
W. L. Jorgensen et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7033–7036
also showed 1-lM potency in inhibiting the binding of MIF to its
MIF, 2.7 nM
+
receptor CD74. Most significantly, three molecules, 3d, 5a, and
5c, were discovered that enhance the binding of MIF and CD74.
The contrasting antagonist and agonist behaviors of 3g versus 3d,
5a, and 5c were then confirmed by monitoring the MIF-dependent
phosphorylation of ERK kinases in human fibroblasts. The potential
utility of the newly discovered agonists extends from fundamental
studies of the biology of MIF to therapeutic applications as immune
adjuvants or in limiting ischemic cardiac injury.
5a
5c
3d
5a
5c
3d
MIF Agonist, 8.0 nM
pERK1/2
Total ERK1/2
1.0
0.8
0.6
0.4
0.2
0
*
*
*
*
Acknowledgements
Gratitude is expressed to the National Institutes of Health
(AI042310, AR049610, AR050498, GM032136) and the Treat B.
Johnson Fund at Yale for support.
Figure 5. Effects of the agonists on MIF-dependent ERK1/2 phosphorylation in
human fibroblasts. The upper panel displays a representative western blot and the
lower panel shows the numerical ratio of phosphorylated to total kinase protein
*
determined by densitometric scanning for three separate experiments. P <0.05.
References and notes
1. Bucala, R. Nature 2000, 408, 146.
2. Orita, M.; Yamamoto, S.; Katayama, N.; Fujita, S. Curr. Pharm. Des. 2002, 8, 1297.
3. Garai, J.; Lorand, T. Curr. Med. Chem. 2009, 16, 1091.
4. Leng, L.; Metz, C. N.; Fang, Y.; Xu, J.; Donnelly, S.; Baugh, J.; Delohery, T.; Chen,
Y.; Mitchell, R. A.; Bucala, R. J. Exp. Med. 2003, 197, 1467.
5. Shi, X.; Leng, L.; Wang, T.; Wang, W.; Du, X.; Li, J.; McDonald, C.; Chen, Z.;
Murphy, J. W.; Lolis, E.; Noble, P.; Knudson, W.; Bucala, R. Immunity 2006, 25,
595.
6. Meyer-Siegler, K. L.; Iczkowski, K. A.; Leng, L.; Bucala, R.; Vera, P. L. J. Immunol.
2006, 177, 8730.
pERK1/2
7. Bacher, M.; Metz, C. N.; Calandra, T.; Mayer, K.; Chesney, J.; Lohoff, M.; Gemsa,
D.; Donnelly, T.; Bucala, R. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 7849.
8. Miller, E. J.; Li, J.; Leng, L.; McDonald, C.; Atsumi, T.; Bucala, R.; Young, L. H.
Nature 2008, 451, 578.
Total ERK1/2
9. Russell, R. R., III; Li, J.; Coven, D. L.; Pypaert, M.; Zechner, C.; Palmeri, M.;
Giordano, F. J.; Mu, J.; Birnbaum, M. J.; Young, L. H. J. Clin. Invest. 2004, 114, 495.
10. Ma, H.; Wang, J.; Thomas, D. P.; Tong, C.; Leng, L.; Wang, W.; Merk, M.; Zierow,
S.; Bernhagen, J.; Ren, J.; Bucala, R.; Li, J. Circulation 2010, 122, 282.
11. Fingerle-Rowson, G.; Kaleswarapu, D. R.; Schlander, C.; Kabgani, N.; Brocks, T.;
Reinart, N.; Busch, R.; Schütz, A.; Lue, H.; Du, X.; Liu, A.; Xiong, H.; Chen, Y.;
Nemajerova, A.; Hallek, M.; Bernhagen, J.; Leng, L.; Bucala, R. Mol. Cell Biol.
2009, 29, 1922.
12. Lubetsky, J. B.; Swope, M.; Dealwis, C.; Blake, P.; Lolis, E. Biochemistry 1999, 38,
7346.
13. Lubetsky, J. B.; Dios, A.; Han, J.; Aljabari, B.; Ruzsicska, B.; Mitchell, R.; Lolis, E.;
Al-Abed, Y. J. Biol. Chem. 2002, 277, 24976.
14. 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.
15. Cournia, Z.; Leng, L.; Gandavadi, S.; Du, X.; Bucala, R.; Jorgensen, W. L. J. Med.
Chem. 2009, 52, 416.
* *
* *
1.0
0.8
0.6
0.4
0.2
0
Figure 6. Effect of antagonist 3g on MIF-dependent ERK1/2 phosphorylation in
human fibroblasts. The upper panel displays a representative western blot and the
lower panel shows the numerical ratio of phosphorylated to total kinase protein
16. Hare, A. A.; Leng, L.; Gandavadi, S.; Du, X.; Cournia, Z.; Bucala, R.; Jorgensen, W.
L. Bioorg. Med. Chem. Lett. 2010, 20, 5811.
**
determined by densitometric scanning for three separate experiments. P <0.0025.
17. Jorgensen, W. L. Acc. Chem. Res. 2009, 42, 724–733.
18. McLean, L. R.; Zhang, Y.; Li, H.; Choi, Y.-M.; Han, Z.; Vaz, R. J.; Li, Y. Bioorg. Med.
Chem. Lett. 2010, 20, 1821.
19. Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem., Int. Ed.
2002, 41, 2596.
20. Andersen, J.; Bolvig, S.; Liang, X. Synlett 2005, 2941.
21. Stamps, S. L.; Taylor, A. B.; Wang, S. C.; Hackert, M. L.; Whitman, C. P.
Biochemistry 2000, 39, 9671.
fibroblasts were incubated with MIF (2.7 nM trimer) together with
vehicle control (DMSO) or with 3d, 5a, or 5c (each at 8 nM) for
30 min. The cells then were lysed and the intracellular contents of
phospho-ERK1/2 and total ERK1/2 were detected by specific
antibodies and western blotting (Fig. 5). The addition of these com-
pounds to MIF enhances ERK1/2 phosphorylation,4 while co-incuba-
tion of antagonist 3g under the same experimental conditions
strongly inhibits the process (Fig. 6).
22. Bernhagen, J.; Mitchell, R. A.; Calandra, T.; Voelter, W.; Cerami, A.; Bucala, R.
Biochemistry 1994, 33, 14144.
23. Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.; Greenwood, J. R.;
Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. J. Med. Chem. 2006, 49, 6177.
24. Cho, Y.; Crichlow, G. V.; Vermeire, J. J.; Leng, L.; Du, X.; Hodsdon, M. E.; Bucala,
R.; Cappello, M.; Gross, M.; Gaeta, F.; Johnson, K.; Lolis, E. J. Proc. Natl. Acad.
U.S.A. 2010, 107, 11313.
25. Wisén, S.; Bertelsen, E. B.; Thompson, A. D.; Patury, S.; Ung, P.; Chang, L.; Evans,
C. G.; Walter, G. M.; Wipf, P.; Carlson, H. A.; Brodsky, J. L.; Zuiderweg, E. R. P.;
Gestwicki, J. E. ACS Chem. Biol. 2010, 5, 611.
In summary, de novo design of small molecules to bind to the
MIF tautomerase active site was carried out using the program
BOMB and led to pursuit of aryl-1,2,3-triazole derivatives. Two
ca. 1 lM tautomerase inhibitors, 3g and 3h, were discovered that