JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M203220200 on May 7, 2002

J. Biol. Chem., Vol. 277, Issue 28, 24976-24982, July 12, 2002
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
277/28/24976    most recent
M203220200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lubetsky, J. B.
Right arrow Articles by Al-Abed, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lubetsky, J. B.
Right arrow Articles by Al-Abed, Y.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The Tautomerase Active Site of Macrophage Migration Inhibitory Factor Is a Potential Target for Discovery of Novel Anti-inflammatory Agents*

Jodi B. LubetskyDagger §, Angeles Dios, Jialian Han, Bayan Aljabari, Bela Ruzsicska, Robert Mitchell, Elias LolisDagger ||, and Yousef Al-Abed**

From the Departments of Dagger  Pharmacology and § Molecular Biophysics and Biochemistry, Yale University School of Medicine, New Haven, Connecticut 06510 and the  North Shore-Long Island Jewish Research Institute, Manhasset, New York 11030

Received for publication, April 4, 2002, and in revised form, May 2, 2002

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Macrophage migration inhibitory factor (MIF) is an immunoregulatory protein that is a potential therapeutic target for a number of inflammatory diseases. Evidence exists that an unexpected catalytic active site of MIF may have a biological function. To gain further insight into the role of the catalytic active site, a series of mutational, structural, and biological activity studies were performed. The insertion of an alanine between Pro-1 and Met-2 (PAM) abolishes a non-physiological catalytic activity, and this mutant is defective in the in vitro glucocorticoid counter-regulatory activity of MIF. The crystal structure of MIF complexed to (S,R)-3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid methyl ester (ISO-1), an inhibitor of MIF D-dopachrome tautomerase activity, reveals that ISO-1 binds to the same position of the active site as p-hydroxyphenylpyruvic acid, a substrate of MIF. ISO-1 inhibits several MIF biological activities, further establishing a role for the catalytic active site of MIF.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Macrophage migration inhibitory factor (MIF)1 was first defined in the 1960s as an activity capable of suppressing the random movements of macrophages during delayed-type hypersensitivity responses (1-3). MIF was "rediscovered" as an important pro-inflammatory cytokine after the identification of the MIF gene (4, 5). MIF is released from macrophages (6), T cells (1, 2, 7), and eosinophils (8) upon cellular activation and by the anterior pituitary gland as a consequence of the systemic stress response (9, 10). The pro-inflammatory properties of MIF are mediated, at least in part, by its ability to counteract the immunosuppressive effects of glucocorticoids (7, 9, 10).

MIF is a critical component of several inflammatory and autoimmune diseases, including arthritis (11-14) and glomerulonephritis (15), and is a critical mediator of both Gram-positive and Gram-negative sepsis (10, 16). Additional biological activities for MIF include the regulation of macrophage and T cell activation (6, 7), IgE synthesis (17), insulin release and carbohydrate metabolism (18, 19), cell proliferation (20), and inhibition of apoptosis (21). MIF has recently been implicated as an important factor for tumor progression and angiogenesis (22-24).

Three-dimensional x-ray crystallographic studies have shown that human MIF exists as a homotrimer and is structurally related to the bacterial enzymes 4-oxalocrotonate tautomerase and 5-carboxymethyl-2-hydroxymuconate isomerase (25, 26). MIF possesses the unusual ability to catalyze the tautomerization of the non-physiological substrates D-dopachrome and L-dopachrome methyl ester into their corresponding indole derivatives (27). More recently, phenylpyruvic acid, p-hydroxyphenylpyruvic acid (HPP), 3,4-dihydroxyphenylaminechrome, and norepinephrinechrome also have been found to be MIF substrates, but high Michaelis constant (Km) values suggest that these also are unlikely natural substrates for MIF (28-30). Human MIF·HPP and murine MIF·inhibitor crystal structures have identified a substrate-binding hydrophobic cavity that lies between two adjacent subunits of the homotrimer (31, 32). Pro-1 appears to be a critical residue for enzymatic activity, as site-directed mutagenesis that substitutes a serine for this proline (P1S) is devoid of D-dopachrome tautomerase activity (33). Similarly, a proline to glycine (P1G) MIF mutant is also catalytically reduced for both D-dopachrome and HPP tautomerase activity (31, 34). The physiological role of these or other catalytic activities remains in question, however, because the P1S mutant is reported to retain wild-type glucocorticoid counter-regulation activity and monocyte chemotaxis inhibition, respectively (33, 35). However, the P1G mutant is greatly impaired in its ability to stimulate superoxide generation in activated neutrophils (34). In addition, a proline to alanine mutant of MIF loses its ability to enhance matrix metalloproteinase mRNA levels (36).

The catalytic activity provides a useful tool for screening potential MIF inhibitors to study the role of the catalytic site in vitro and in vivo. We have generated a series of pharmacological inhibitors of MIF based on structure-activity studies of other inhibitors and a non-physiological substrate. An acetaminophen metabolite, N-acetylbenzoquinone imine, forms a covalent bond with MIF and inhibits MIF catalytic activity (37). We recently used rational design and synthesis of p-hydroxyphenyl Schiff bases as inhibitors of MIF enzymatic activity (38). These data suggest a potentially powerful approach to the rational design of small molecule pharmacological inhibitors of MIF.

To further investigate the link between cytokine activities and the catalytic activity of MIF, we have first turned to the study of a catalytically inactive mutant. Although the physiological substrate of MIF is not known, a mutant that blocks the binding site would inhibit catalytic activity at the Pro-1 active site. In this mutant an alanine is inserted between Pro-1 and Met-2 (PAM). Crystallographic studies of the PAM mutant indicate that this modification of the sequence results in the movement of the proline into the substrate-binding site, blocking the binding of any substrate (31). In a second approach, synthesis and analysis of phenylimine scaffolds as potential MIF antagonists have revealed that isoxazolines, which are more stable in aqueous solution, may serve as an attractive scaffold for further drug design. We have assessed the MIF antagonistic capacity of (S,R)-3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid methyl ester (ISO-1) and related derivatives in several bioassays, including MIF stimulation of cytoplasmic phospholipase A2 and modulation of glucocorticoids on TNFalpha , PGE2, and Cox-2 production. The structure of ISO-1 complexed to MIF provides insights into the design of the next generation of MIF inhibitors.

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Expression and Purification of MIF-- Expression and purification of the proteins were performed as previously described (31). Briefly, expression of wild-type and mutant proteins involves growing human pET11b-MIF or murine pET24a-MIF-transformed BL21(DE3) cells to an A600 of 0.7-1.0 and inducing with isopropyl-1-thio-beta -D-galactopyranoside to a final concentration of 1 mM. After a 5-h growth, the cells are harvested and resuspended in 20 mM Tris, 20 mM NaCl, pH 7.5, at 7-14% of the original volume of growth medium. Cells are lysed using a French press. After addition of protamine sulfate to a final concentration of 0.006%, cell debris was removed by centrifugation at 27,000 × g for 30 min. The supernatant was filtered using a 0.22-µm cellulose acetate filter and then applied to Mono S and Mono Q (Amersham Biosciences) anion-exchange columns. Under the buffer conditions used, most contaminating proteins adsorb to one of the chromatographic media, whereas wild-type and mutant MIF do not. The fractions containing MIF were pooled and concentrated using ultrafiltration and microconcentrators. Protein concentrations were determined by measuring absorbance at 280 nm (epsilon  = 1.22 M-1 cm-1). The resulting MIF sample is >95% pure and suitable for catalytic characterization and x-ray crystallography.

Low Endotoxin Protein Purification-- MIF has an unusually high affinity for endotoxin that must be removed to ensure that lipopolysaccharide (LPS) does not interfere with the results of the biological assays. The initial steps of purification remain as stated previously. The pooled protein fractions were applied to a Sep-Pak C8 column (Waters) as follows. Columns were attached to a vacuum manifold to assist in the endotoxin purification. The columns were wet with methanol and then washed with low endotoxin water. Approximately 1 mg of protein was applied to each column and run though the column slowly. The resin was washed with pure water and 30% acetonitrile. The protein was eluted with 5 ml of 60% acetonitrile into polystyrene tubes. The samples were kept in the -80 °C freezer until lyophilization.

For refolding, all buffers were prepared with low endotoxin water in LPS-free containers. Several lyophilized pellets of MIF were dissolved in 350 µl of 8 M urea in an Eppendorf tube and incubated for 1 h at room temperature. Four µl of 1 M dithiothreitol solution was added and the solution was incubated for an additional hour. Three consecutive times, 350 µl of 10 mM dithiothreitol were added, the contents were swirled, and the tube was incubated at room temperature for 30 min. The solution was then incubated at 4 °C for ~30 min and subsequently centrifuged at 14,000 rpm for 10 min at 4 °C. The solution was then dialyzed at 4 °C against 10 mM dithiothreitol in Tris-buffered saline, pH 7.4 (human) or pH 6.8 (murine), using a dialysis Slidealyzer cassette (Pierce) MWCO 7,000 overnight followed by dialysis against Tris-buffered saline over the next several days. The sample was collected from the cassette and centrifuged at 4 °C for 10 min in an Eppendorf centrifuge at 14,000 rpm. D-Dopachrome tautomerase activity was assessed and the final endotoxin concentration was determined using a quantitative chromogenic Limulus amebocyte lysis test kit (BioWhittaker).

Synthesis and Derivatization of ISO-1-- ISO-1 was synthesized in three steps as described previously (39, 40). Several derivatives of ISO-1 were prepared as follows. Compound 2 was synthesized as described previously (39, 40). For compound 3, ISO-1 (0.5 mmol) was dissolved in acetone and treated with K2CO3 (0.7 mmol) followed by MeI (0.7 mmol). The reaction mixture was refluxed for 5 h and quenched after completion by aqueous ammonium chloride solution. The crude mixture was purified by flash chromatography. Compound 3 was obtained in 78% yield. Compound 4 was produced by standard acylation of ISO-1 followed by oxidation using N-bromosuccinimide and azobisisobutyronitrile as described previously (39). The crude mixture was treated with sodium methoxide in methanol to free the hydroxyl group and the final compound was recovered in 65% yield after flash chromatography. Last, for compound 5, a solution of ISO-1 (0.45 mmol) in tetrahydrofuran at 0 °C was treated with LiAlH4. After complete consumption of the ester and the formation of a new product, the reaction was quenched and the crude product was purified using flash chromatography to furnish compound 5 in 95% yield.

Dopachrome Tautomerase Assay-- L-Dopachrome methyl ester was prepared at 2.4 mM through oxidation of L-3,4-dihydroxyphenylalanine methyl ester with sodium periodate as previously described (38). Activity was determined at room temperature by adding dopachrome methyl ester (0.3 ml) to a cuvette containing 50 nM MIF in 50 mM potassium phosphate buffer, pH 6, 0.5 mM EDTA and measuring the decrease in absorbance from 2 to 20 s at 475 nm spectrophotometrically. The inhibitors were dissolved in Me2SO at various concentrations and added to the cuvette with the MIF prior to the addition of the dopachrome.

Glucocorticoid Override Assay-- Human or murine (for the mutant study) mononuclear cells were isolated from whole blood by Ficoll density gradient centrifugation and monocytes were purified by adherence. Non-adherent cells were removed by washing each well twice in 24 h with RPMI, 10% heat-inactivated fetal bovine serum. 1 × 106 cells/well were preincubated for 1 h with dexamethasone (10-8 M), dexamethasone plus MIF (100 ng/ml wild-type or mutant MIF), or dexamethasone plus MIF and ISO-1 (0.1-20 µM) before the addition of 0.5 µg/ml LPS (Escherichia coli 0111:B4, Sigma). After 16 h, cell culture supernatants were collected for determination of TNFalpha and prostaglandin (PGE2) concentration by enzyme-linked immunosorbent assay (R & D Systems).

Western blot analysis of Cox-2 was carried out as described previously (41). Briefly, aliquots of 1 × 106 human mononuclear cells were lysed with Tris-buffered saline containing 1% Nonidet P-40, 0.5% deoxycholic acid, 0.1% SDS, and 2 mM EDTA. Cellular debris was pelleted and the supernatants were diluted with an equal volume of reducing SDS-PAGE sample buffer. Ten µl of lysate was electrophoresed through 10% polyacrylamide gels and transferred to nitrocellulose membranes for Western blot analysis using polyclonal goat anti-Cox-2 IgG and donkey anti-goat IgG-horseradish peroxidase (Santa Cruz Biotechnology).

Arachidonic Acid Release Assay-- MIF plasmid construction and transfection were performed as described previously (41). Briefly, RAW 264.7 macrophages were plated at 1 × 105 cells/ml in RPMI, 10% fetal calf serum supplemented with 1 nM [14C]arachidonic acid (PerkinElmer Life Sciences). After overnight culture, the cells were washed extensively with RPMI followed by re-addition of RPMI, 10% fetal calf serum. Cells were subsequently transfected with FuGENE 6 (Roche Molecular Diagnostics) and pcDNA/GS (Invitrogen) empty vector or FuGENE 6 and pcDNA/GS/MIF in the absence or presence of vehicle (Me2SO) or inhibitors. After 16 h cell supernatants were collected and 14C-labeled arachidonic acid was assessed by scintillation counting.

Crystallography-- Human wild-type MIF was concentrated to 10-15 mg/ml before crystallization. Crystals were grown using the hanging drop vapor diffusion method. For MIF·ISO-1 crystals, equal volumes (1 µl) of protein and 2 mM inhibitor dissolved in reservoir solution (2.0 M ammonium sulfate, 4% isopropyl alcohol, 0.1 M Tris, pH 8.0) were mixed. Crystals belong to the space group P3121 with unit cell dimensions of a = 95.80 Å, b = 95.80 Å, and c = 104.12 Å.

All crystals were equilibrated with cryoprotectant (25% glycerol in mother liquor) for 10-15 min and were frozen rapidly in a stream of N2 at -140 °C for data collection. Diffraction data were collected on an RAXIS-IIC image plate detector (Rigaku, Tokyo) with a Rigaku RU200 rotating anode x-ray generator. Data were processed by the DENZO/SCALEPACK package (42).

The structure of the trigonal crystal form of wild-type MIF complexed with ISO-1 was solved by Fourier differences using phases from uncomplexed MIF (25). Refinement was performed with X-PLOR 3.851 (43). The Rfree was implemented in the beginning of refinement and used throughout the refinement process as a guide for model improvement (44). The graphics program O (45) was used for the display of electron density maps. After rigid body refinement, a series of positional refinement rounds were performed in which strict non-crystallographic symmetry constraints were implemented. ISO-1 was built into the electron density of an omit map calculated with non-crystallographic symmetry averaging. The coordinates, topology, and parameter files for the inhibitor were constructed with INSIGHTII (Molecular Simulations, Inc., San Diego, CA) and XPLOR2D (46).

Group B-factor refinement, individual B-factor refinement, and bulk solvent correction, which enabled the use of lower resolution data throughout the subsequent rounds of refinement, were implemented. The trimer was generated, and further rounds of non-crystallographic symmetry-restrained refinement and model building were performed to obtain the final structure. Root mean square deviation calculations were performed with LSQMAN (Uppsala Software Factory, Uppsala, Sweden). Data and refinement statistics are shown in Table I.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Crystal parameters, data, and refinement statistics of the MIF·ISO-1 complex
Completeness, Rm, and < I/sigma (I)> are given for all data and for data in the highest resolution shell.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

A Catalytic Inactive Mutant Is Defective in Biological Activities-- LPS, a bacterial cell wall component, is an inflammatory product that induces the release of TNF from macrophage cell lines or primary monocytes. Glucocorticoids such as dexamethasone exert anti-inflammatory effects by inhibiting TNF release. Exogenously applied MIF counteracts this anti-inflammatory activity, providing a convenient in vitro bioassay (9). We have previously designed a catalytically inactive alanine insertion mutant between Pro-1 and Met-2 (PAM) and determined its three-dimensional structure. Both murine wild-type MIF and the PAM mutant were tested in this assay to determine whether the disruption of the catalytic active site of MIF inhibits its glucocorticoid override activity. Wild-type MIF inhibits glucocorticoid inhibition of cytokine production and leads to higher TNFalpha production in dexamethasone-treated, LPS-activated monocytes, whereas PAM does not (Fig. 1). Moreover, PAM shows slight antagonist activity, as the production of TNF-alpha is decreased relative to the sample containing LPS and dexamethasone.


View larger version (9K):
[in this window]
[in a new window]
  		Fig. 1.   Glucocorticoid regulating activity of wild-type MIF and PAM. MIF, but not the PAM mutant, has the ability to override the dexamethasone inhibition of LPS-induced TNF-alpha production from murine mononuclear cells. These data are the average ± S.D. of at least duplicate wells that were repeated three times. The double asterisk indicates p < 0.01 compared with LPS + dexamethasone. The single asterisk indicates p < 0.05 compared with LPS + dexamethasone and p < 0.0001 compared with LPS + dexamethasone + wild-type MIF.

ISO-1 Inhibits Dopachrome Tautomerase Activity of MIF-- The synthesis and analysis of additional phenylimine scaffolds as potential MIF antagonists have revealed that isoxazolines may serve as an attractive scaffold for drug design. ISO-1 contains the structural elements that appear to be important for binding to the MIF catalytic active site based on our analysis of MIF·HPP complex and structure-activity relationships with the amino acid Schiff base compounds (31, 38).

Accordingly, we synthesized ISO-1 in three steps as described previously (39, 40). The structure was confirmed by 1H and 13C NMR and mass spectroscopy. Several derivatives were prepared and tested for activity in the MIF dopachrome tautomerase assay (Fig. 2). ISO-1 inhibited MIF tautomerase activity in a dose-dependent manner with an IC50 of about 7 µM, but the non-hydroxylated phenyl analog (compound 2) was 10-15 times less potent (Fig. 3). The 4-methoxy analog (compound 3) showed no activity, reinforcing our earlier conclusion that a para-hydroxyl function is an important feature of our emerging pharmacophore for MIF tautomerase inhibitors, probably attributable to the formation of a hydrogen bond with Asn-97, as suggested by the MIF·HPP co-crystallization data. Oxidization of ISO-1 to isoxazole (compound 4) (39) eliminates the chiral center. Surprisingly, the isoxazole was totally inactive in the dopachrome tautomerization assay. The reduction of the isoxazoline (compound 5) also leads to a molecule that is inactive in the dopachrome tautomerization assay.


View larger version (9K):
[in this window]
[in a new window]
  		Fig. 2.   Modification of ISO-1.


View larger version (9K):
[in this window]
[in a new window]
  		Fig. 3.   Inhibition of MIF dopachrome tautomerse activity by ISO-1 and compound 2. The percent inhibition of dopachrome tautomerase activity is plotted against the inhibitor concentration for ISO-1 (black-triangle) and compound 2 (). ISO-1 inhibits dopachrome tautomerase activity with an IC50 of ~7 µM, whereas compound 2 is 10-15 times less potent.

ISO-1 Inhibits Arachidonic Acid Release-- MIF has been shown to stimulate cytoplasmic phospholipase A2 activation in fibroblasts, leading to an increase in arachidonic acid production by these cells (20). Transfection of a MIF-expressing plasmid into macrophages induced the production of arachidonic acid when compared with vector control, and this effect was attenuated by the addition of a neutralizing anti-MIF monoclonal antibody, suggesting that MIF mediates the production of arachidonic acid via autocrine or paracrine mechanism(s) (41). In accordance with this potential mechanism(s), treatment of the transfected cells with ISO-1 inhibited the release of arachidonic acid in a dose-dependent manner (Fig. 4). Of note, compound 2 only inhibits arachidonic acid production at the greatest concentration used (100 µM), whereas compounds 3 and 4 fail to inhibit the arachidonic acid production. The MIF inhibitory activities of ISO-1 and its derivatives were not the result of cell toxicity, because none of the tested compounds were toxic in these pharmacological doses, as assessed by trypan blue exclusion assays (data not shown).


View larger version (15K):
[in this window]
[in a new window]
  		Fig. 4.   ISO-1 inhibits the MIF induction of arachidonic acid release. RAW 264.7 macrophages were exposed to medium supplemented with [14C]arachidonic acid and transfected with pcDNA/GS/MIF plasmid or pcDNA/GS vector control. ISO-1 was found to inhibit arachidonic acid release in a dose-dependent fashion in the pcDNA/GS/MIF-transfected cells. In contrast, the other analogs (compounds 2, 3, and 4) failed to do so. Data are the mean ± S.D. of three independent experiments.

ISO-1 Inhibits MIF Glucocorticoid Regulating Activity-- As shown previously, a catalytically inactive PAM mutant is not capable of overriding the glucocorticoid inhibition of cytokine production by LPS-stimulated macrophages. We set out to test ISO-1 in this model. As shown in Fig. 5A, ISO-1 significantly antagonizes in a dose-dependent manner MIF-dependent inhibition of dexamethasone in this system. To address the specificity of this inhibitory effect on MIF, we tested other isoxazoline analogs (compounds 2-5) that are either catalytically less potent inhibitors or do not have any inhibitory effect and found that these compounds do not provide any anti-inflammatory activity (data not shown). These results are consistent with our hypothesis that a small molecule bound at the catalytic active site can neutralize the pro-inflammatory effects of MIF.


View larger version (31K):
[in this window]
[in a new window]
  		Fig. 5.   ISO-1 binding to the MIF active site down-regulates MIF glucocorticoid regulating activity on activated monocytes. The ability of ISO-1 to inhibit MIF regulation of glucocorticoid suppression of TNFalpha (A), PGE2 (B), and Cox-2 (C) production in human monocytes was assayed. The levels of TNFalpha , PGE2, and Cox-2 production decreased as the concentration of ISO-1 increased from 0.1 to 20 µM. The data shown are the mean ± S.D. and are representative of three independently performed experiments.

Recently, recombinant MIF has been shown to specifically up-regulate Cox-2 mRNA and activity, but not Cox-1, in human rheumatoid arthritis synovium and cultured fibroblast-like synoviocytes (47). Moreover, the level of LPS/interferon gamma -mediated Cox-2 expression and PGE2 release were reported to be 10-fold less in murine MIF-deficient macrophages than in murine wild-type macrophages. Furthermore, the addition of recombinant MIF to the MIF-/- macrophages partially restored LPS/interferon gamma -mediated Cox-2 expression and PGE2 production (41). In a similar assay, macrophages treated with LPS induced Cox-2 expression and PGE2 release, whereas treatment with the anti-inflammatory dexamethasone counteracts the effects of LPS. Addition of exogenous recombinant MIF to macrophages treated with LPS and dexamethasone restores the secretion of PGE2 to the levels observed with LPS treatment alone. The addition of ISO-1 to these macrophages inhibited the production of PGE2 at 20 and 10 µM, but no major effect was found at 1.0 and 0.1 µM (Fig. 5B). Similarly, Cox-2 secretion was effectively suppressed by ISO-1 in a dose-dependent manner (Fig. 5C). In each of the above assays, ISO-1 does have a slight inhibitory effect on cytokine secretion without the addition of recombinant MIF. This may be because of the inhibition of the native MIF that is secreted by the monocytes. Taken together, we can conclude that MIF is regulating the dexamethasone anti-inflammatory activity and that the active site is involved in mediating this activity.

Crystallography of the MIF·ISO-1 Complex-- To determine the structural basis for the inhibition of enzymatic activity of MIF, ISO-1 was co-crystallized with MIF. The structure of the trigonal crystal form was solved by using the uncomplexed monomer MIF as the starting coordinates (25). The monomer was refined with X-PLOR using non-crystallographic symmetry constraints against 2.0-Å resolution data. The R-isomer of ISO-1 was built into a 2Fo - Fc omit map. The S-isomer could not fit properly into the electron density. Therefore, only the R-isomer of ISO-1 binds to MIF. The complex between MIF and R-ISO-1 was subjected to additional refinement. The final structure has a crystallographic R-factor of 23.7% and a Rfree of 26.2% with a root mean square deviation of 0.010 Å and 1.392 degrees for bonds and angles, respectively. A comparison of the uncomplexed (25) and ISO-1-complexed MIF structures reveals that there are no major conformational changes that occur upon inhibitor binding with a root mean square deviation of 0.24 Å for the Calpha positions after superposition (Fig. 6).


View larger version (52K):
[in this window]
[in a new window]
  		Fig. 6.   Superposition of uncomplexed MIF and MIF·ISO-1. Backbone atom traces are shown for the trimer arrangements of uncomplexed MIF (black) and MIF·ISO-1 (gray). All atoms from the active site residues (Pro-1, Lys-32, Ile-64, Tyr-95, and Asn-97) are also displayed. This superposition reveals that there are no major conformational changes upon inhibitor binding. This figure was prepared with the program INSIGHTII (Molecular Simulations, Inc.).

Interaction of ISO-1 with Catalytic Active Site Residues-- Fig. 7A shows a 2Fo - Fc omit map of ISO-1 in the active site with the compound placed in the electron density. Three molecules of ISO-1 bind each MIF trimer molecule and lie at each interface between two subunits. A series of aromatic-aromatic interactions, hydrogen bonds, and van der Waals contacts provides the binding energy for ISO-1. ISO-1 binds to the same position of MIF as HPP and therefore has many similar interactions to MIF (Fig. 7B). Both ISO-1 and HPP have hydrogen-bonding interactions to Lys-32 and Ile-64 from one subunit and Asn-97 from an adjacent subunit. The phenyl ring of both small molecules is placed in the same orientation with respect to MIF because of a bifurcated hydrogen bond with the side chain amide nitrogen and carbonyl oxygen of Asn-97C. As in HPP, the phenyl ring of ISO-1 forms an interaction with the aromatic ring of Tyr-95C (where "C" refers to an adjacent subunit). The aromatic rings are stacked perpendicular to each other to have favorable charge-charge interactions that are based on polarization of the aromatic rings (48). The backbone nitrogen of Ile-64 is within hydrogen bonding distance of the nitrogen, whereas the side chain nitrogen from Lys-32 is within hydrogen bonding distances of both the carbonyl oxygen of the carboxymethyl group and the oxygen of the isoxazoline ring.


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 7.   Structure of ISO-1 and MIF. A, 2Fo - Fc omit map at 1sigma for ISO-1 in the catalytic active site of MIF. The R-isomer is modeled in the electron density. The catalytic active site residues are displayed. B, schematic of the interactions between protein residues of MIF and ISO-1. All distances, with the exception of Pro-1A, represent hydrogen bonds. A series of aromatic-aromatic interactions, hydrogen bonds, and van der Waals contacts provide the binding energy for ISO-1. Pro-1A, Lys-32A, and Ile-64A are from one subunit, and Tyr-95C and Asn-97C are from an adjacent subunit. This figure was prepared with Bobscript (52) and ChemDraw (CambridgeSoft Corp., Cambridge, MA).


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Although MIF has been classified as a cytokine, no receptor has been identified. However, the determination of the three-dimensional structure of MIF revealed that MIF shares a similar structure and active site to several bacterial enzymes (25, 49). MIF utilizes this active site in several non-physiological catalytic activities (27, 28, 50). This catalytic active site has been clearly identified through crystallographic studies of MIF complexed to a non-physiological substrate, HPP (31), and to the inhibitor, (E)-2-fluoro-p-hydroxycinnamate (32). A multiple sequence alignment of MIF orthologues and homologues indicates that the active site residues are among the most highly conserved residues (51). The catalytic activities of MIF, the structural similarity to microbial enzymes, and the pattern of invariant residues to preserve this site suggest that the catalytic site is important in the biology of MIF. However, there has been controversy about the significance of this catalytic active site in the biological activities of MIF. Mutational studies of the catalytic proline residue have been performed and reveal that there is a correlation between catalytic activity and some biological assays (34, 36). There are also data to suggest that there is no correlation between the catalytic activity and other assays (21, 33, 35).

To further investigate the role of the catalytic site in the biological activity of MIF, we have assayed a catalytically defective mutant that has an alanine inserted between Pro-1 and Met-2 (PAM). This mutant shows minimal (<0.1% of wild-type) catalytic dopachrome tautomerase activity that cannot proceed through this enzymatic site as the proline of PAM has been dislodged to occupy a position that would normally be reserved for the substrate. This mutant is also inactive in the glucocorticoid override activity of MIF, one of the activities that previously had been shown to not require the catalytic proline (33). Our data strongly suggest that the enzymatic site is important in this biological activity of the protein.

The involvement of the MIF catalytic site in biological activity has also been studied using small molecule inhibitors (37, 38). We have previously studied a series of pharmacological inhibitors of MIF that were designed utilizing the known information about the MIF-binding site and the known chemical structures of substrates of MIF. Here, we have investigated isoxazolines, another family of small molecule inhibitors of MIF tautomerase activity. The ISO-1 inhibitor was tested for inhibition of biological activity. The testing of ISO-1 in a number of biological assays reveals that inhibition of the catalytic activity correlates with the disruption of biological activity. For example, ISO-1 prevents the glucocorticoid override assay of MIF, as substantially less TNF is secreted from mononuclear cells incubated with LPS, dexamethasone, and MIF relative to control cells. In addition, less PGE2 and Cox-2 are produced, and there is a dose-dependent effect with ISO-1 for all of these results. The results using ISO-1 support the mutational studies that indicate that the biological activity of MIF is due to the region involving Pro-1.

The crystal structure of the MIF·ISO-1 complex reveals that many residues involved in the MIF·HPP interactions are also involved in inhibitor interaction. Structural analysis indicates that only the R-isomer binds to the MIF pocket and suggests that a purified chiral molecule would be even more potent.

The interactions of ISO-1 with MIF are also in agreement with structure-activity relationships using the D-dopachrome tautomerase inhibition data. Lys-32 makes several key interactions with ISO-1. For example, reduction of ISO-1 (compound 5) eliminates the D-dopachrome tautomerase inhibition presumably because of the inability to interact with Lys-32. Asn-97 also is an important residue as it forms a bifurcated hydrogen bond interaction with the inhibitor. Methylation of ISO-1 (compound 3) to disrupt the interaction with Asn-97 also results in an inactive small molecule. Although the oxidation of ISO-1 leads to an inactive small molecule (compound 4), there is no obvious structural explanation for why the oxidized form does not have inhibitory activity. The oxidized form can be modeled into the active site and appears to be capable of having the same hydrogen-bonding interactions as ISO-1.

These crystallographic and kinetic data represent valuable information for improving ISO-1 by rationally designing the next generation of MIF antagonists. Tyr-95 plays a role in substrate binding through a favorable charge-charge interaction of its aromatic ring with that of the hydroxyphenyl ring of ISO-1. The hydroxyl group of Tyr-95 is ~3.5 Å away from carbon 4 of the isoxazoline ring. Substitution of this carbon for nitrogen or the addition of a carbonyl group on this carbon may provide further interactions to increase the affinity with MIF. In addition, the methoxy group of ISO-1 is within 3.4 Å of the hydroxyl group of Tyr-36 of MIF. Changing this group to a hydroxymethyl group may lead to a hydroxyl group that is capable of a hydrogen-bonding interaction. This would also potentially lead to a stronger inhibitor.

Further experiments are necessary to discriminate between the possibilities that the enzymatic activity itself is involved in biological activity or that the "catalytic site" is acting as a binding site for an uncharacterized receptor. Nonetheless, the potential exists to develop therapeutically beneficial pharmacological inhibitors of this active site of MIF.

    ACKNOWLEDGEMENT

We thank Paul Pepin for technical assistance with crystallographic data collection.

    FOOTNOTES

* This work was supported by a Biomedical Research Grant from the Arthritis Foundation (to E. L.) and the Picower Foundation (to Y. A.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 1LJT) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

|| To whom correspondence may be addressed: Dept. of Pharmacology, Yale University School of Medicine, 333 Cedar St., SHM CE11, New Haven, CT 06510. Tel.: 203-785-6233; Fax: 203-785-7670; E-mail: elias.lolis@yale.edu.

** To whom correspondence may be addressed: North Shore-Long Island Jewish Research Institute, Manhasset, NY 11030. Tel.: 516-562-9461; Fax: 516-365-6097; E-mail: yal-abed@picower.edu.

Published, JBC Papers in Press, May 7, 2002, DOI 10.1074/jbc.M203220200

    ABBREVIATIONS

The abbreviations used are: MIF, macrophage migration inhibitory factor; ISO-1 (S, R)-3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid methyl ester; HPP, p-hydroxyphenylpyruvic acid; TNF, tumor necrosis factor; PGE2, prostaglandin E2; Cox-2, cyclooxygenase-2; LPS, lipopolysaccharide.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

1. Bloom, B. R., and Bennett, B. (1966) Science 153, 80-82[Abstract/Free Full Text]
2. David, J. R. (1966) Proc. Natl. Acad. Sci. U. S. A. 56, 72-77[Free Full Text]
3. George, M., and Vaughan, J. H. (1962) Proc. Soc. Exp. Biol. Med. 111, 514-521[Medline] [Order article via Infotrieve]
4. Bernhagen, J., Mitchell, R. A., Calandra, T., Voelter, W., Cerami, A., and Bucala, R. (1994) Biochemistry 33, 14144-14155[CrossRef][Medline] [Order article via Infotrieve]
5. Weiser, W. Y., Temple, P. A., Witek-Giannotti, J. S., Remold, H. G., Clark, S. C., and David, J. R. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 7522-7526[Abstract/Free Full Text]
6. Calandra, T., Bernhagen, J., Mitchell, R. A., and Bucala, R. (1994) J. Exp. Med. 179, 1895-1902[Abstract/Free Full Text]
7. Bacher, M., Metz, C. N., Calandra, T., Mayer, K., Chesney, J., Lohoff, M., Gemsa, D., Donnelly, T., and Bucala, R. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 7849-7854[Abstract/Free Full Text]
8. Rossi, A. G., Haslett, C., Hirani, N., Greening, A. P., Rahman, I., Metz, C. N., Bucala, R., and Donnelly, S. C. (1998) J. Clin. Invest. 101, 2869-2874[Medline] [Order article via Infotrieve]
9. Calandra, T., Bernhagen, J., Metz, C. N., Spiegel, L. A., Bacher, M., Donnelly, T., Cerami, A., and Bucala, R. (1995) Nature 377, 68-71[CrossRef][Medline] [Order article via Infotrieve]
10. Bernhagen, J., Calandra, T., Mitchell, R. A., Martin, S. B., Tracey, K. J., Voelter, W., Manogue, K. R., Cerami, A., and Bucala, R. (1993) Nature 365, 756-759[CrossRef][Medline] [Order article via Infotrieve]
11. Mikulowska, A., Metz, C. N., Bucala, R., and Holmdahl, R. (1997) J. Immunol. 158, 5514-5517[Abstract]
12. Leech, M., Metz, C., Santos, L., Peng, T., Holdsworth, S. R., Bucala, R., and Morand, E. F. (1998) Arthritis Rheum. 41, 910-917[CrossRef][Medline] [Order article via Infotrieve]
13. Leech, M., Metz, C., Hall, P., Hutchinson, P., Gianis, K., Smith, M., Weedon, H., Holdsworth, S. R., Bucala, R., and Morand, E. F. (1999) Arthritis Rheum. 42, 1601-1608[CrossRef][Medline] [Order article via Infotrieve]
14. Onodera, S., Tanji, H., Suzuki, K., Kaneda, K., Mizue, Y., Sagawa, A., and Nishihira, J. (1999) Cytokine 11, 163-167[CrossRef][Medline] [Order article via Infotrieve]
15. Lan, H. Y., Mu, W., Yang, N., Meinhardt, A., Nikolic-Paterson, D. J., Ng, Y. Y., Bacher, M., Atkins, R. C., and Bucala, R. (1996) Am. J. Pathol. 149, 1119-1127[Abstract]
16. Bozza, M., Satoskar, A. R., Lin, G., Lu, B., Humbles, A. A., Gerard, C., and David, J. R. (1999) J. Exp. Med. 189, 341-346[Abstract/Free Full Text]
17. Mikayama, T., Nakano, T., Gomi, H., Nakagawa, Y., Liu, Y. C., Sato, M., Iwamatsu, A., Ishii, Y., Weiser, W. Y., and Ishizaka, K. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 10056-10060[Abstract/Free Full Text]
18. Benigni, F., Atsumi, T., Calandra, T., Metz, C., Echtenacher, B., Peng, T., and Bucala, R. (2000) J. Clin. Invest. 106, 1291-1300[Medline] [Order article via Infotrieve]
19. Waeber, G., Calandra, T., Roduit, R., Haefliger, J. A., Bonny, C., Thompson, N., Thorens, B., Temler, E., Meinhardt, A., Bacher, M., Metz, C. N., Nicod, P., and Bucala, R. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 4782-4787[Abstract/Free Full Text]
20. Mitchell, R. A., Metz, C. N., Peng, T., and Bucala, R. (1999) J. Biol. Chem. 274, 18100-18106[Abstract/Free Full Text]
21. Hudson, J. D., Shoaibi, M. A., Maestro, R., Carnero, A., Hannon, G. J., and Beach, D. H. (1999) J. Exp. Med. 190, 1375-1382[Abstract/Free Full Text]
22. Shimizu, T., Abe, R., Nakamura, H., Ohkawara, A., Suzuki, M., and Nishihira, J. (1999) Biochem. Biophys. Res. Commun. 264, 751-758[CrossRef][Medline] [Order article via Infotrieve]
23. Chesney, J., Metz, C., Bacher, M., Peng, T., Meinhardt, A., and Bucala, R. (1999) Mol. Med. 5, 181-191[Medline] [Order article via Infotrieve]
24. Ogawa, H., Nishihira, J., Sato, Y., Kondo, M., Takahashi, N., Oshima, T., and Todo, S. (2000) Cytokine 12, 309-314[CrossRef][Medline] [Order article via Infotrieve]
25. Sun, H. W., Bernhagen, J., Bucala, R., and Lolis, E. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 5191-5196[Abstract/Free Full Text]
26. Sugimoto, H., Suzuki, M., Nakagawa, A., Tanaka, I., and Nishihira, J. (1996) FEBS Lett. 389, 145-148[CrossRef][Medline] [Order article via Infotrieve]
27. Rosengren, E., Bucala, R., Aman, P., Jacobsson, L., Odh, G., Metz, C. N., and Rorsman, H. (1996) Mol. Med. 2, 143-149[Medline] [Order article via Infotrieve]
28. Rosengren, E., Aman, P., Thelin, S., Hansson, C., Ahlfors, S., Bjork, P., Jacobsson, L., and Rorsman, H. (1997) FEBS Lett. 417, 85-88[CrossRef][Medline] [Order article via Infotrieve]
29. Matsunaga, J., Sinha, D., Solano, F., Santis, C., Wistow, G., and Hearing, V. (1999) Cell. Mol. Biol. 45, 1035-1040[Medline] [Order article via Infotrieve]
30. Matsunaga, J., Sinha, D., Pannell, L., Santis, C., Solano, F., Wistow, G. J., and Hearing, V. J. (1999) J. Biol. Chem. 274, 3268-3271[Abstract/Free Full Text]
31. Lubetsky, J. B., Swope, M., Dealwis, C., Blake, P., and Lolis, E. (1999) Biochemistry 38, 7346-7354[CrossRef][Medline] [Order article via Infotrieve]
32. Taylor, A. B., Johnson, W. H., Jr., Czerwinski, R. M., Li, H. S., Hackert, M. L., and Whitman, C. P. (1999) Biochemistry 38, 7444-7452[CrossRef][Medline] [Order article via Infotrieve]
33. Bendrat, K., Al-, Abed, Y., Callaway, D. J., Peng, T., Calandra, T., Metz, C. N., and Bucala, R. (1997) Biochemistry 36, 15356-15362[CrossRef][Medline] [Order article via Infotrieve]
34. Swope, M. D., Sun, H. W., Klockow, B., Blake, P., and Lolis, E. (1998) J. Biol. Chem. 273, 14877-14884[Abstract/Free Full Text]
35. Hermanowski-Vosatka, A., Mundt, S. S., Ayala, J. M., Goyal, S., Hanlon, W. A., Czerwinski, R. M., Wright, S. D., and Whitman, C. P. (1999) Biochemistry 38, 12841-12849[CrossRef][Medline] [Order article via Infotrieve]
36. Onodera, S., Kaneda, K., Mizue, Y., Koyama, Y., Fujinaga, M., and Nishihira, J. (2000) J. Biol. Chem. 275, 444-450[Abstract/Free Full Text]
37. 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., and Bucala, R. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 144-149[Abstract/Free Full Text]
38. Dios, A., Mitchell, R. A., Aljabari, B., Lubetsky, J., O'Connor, K., Liao, H., Senter, P. D., Manogue, K. R., Lolis, E., Metz, C., Bucala, R., and Al-Abed, Y. (2002) J. Med. Chem. 45, 2410-2416[CrossRef][Medline] [Order article via Infotrieve]
39. Xue, C. B., Wityak, J., Sielecki, T. M., Pinto, D. J., Batt, D. G., Cain, G. A., Sworin, M., Rockwell, A. L., Roderick, J. J., Wang, S., Orwat, M. J., Frietze, W. E., Bostrom, L. L., Liu, J., Higley, C. A., Rankin, F. W., Tobin, A. E., Emmett, G., Lalka, G. K., Sze, J. Y., Di, Meo, S. V., Mousa, S. A., Thoolen, M. J., Racanelli, A. L., Olson, R. E., et al.. (1997) J. Med. Chem. 40, 2064-2084[CrossRef][Medline] [Order article via Infotrieve]
40. Wityak, J., Sielecki, T. M., Pinto, D. J., Emmett, G., Sze, J. Y., Liu, J., Tobin, A. E., Wang, S., Jiang, B., Ma, P., Mousa, S. A., Wexler, R. R., and Olson, R. E. (1997) J. Med. Chem. 40, 50-60[CrossRef][Medline] [Order article via Infotrieve]
41. Mitchell, R. A., Liao, H., Chesney, J., Fingerle-Rowson, G., Baugh, J., David, J., and Bucala, R. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 345-350[Abstract/Free Full Text]
42. Otwinowski, Z., and Minor, W. (1997) Methods Enzymol. 276, 307-325
43. Brünger, A. T. (1997) X-PLOR 3.851 Edition , Yale University Press, New Haven, CT
44. Brünger, A. T. (1992) Nature 355, 472-475[CrossRef][Medline] [Order article via Infotrieve]
45. Jones, T. A., Zou, J. Y., Cowan, S. W., and Kjeldgaard, M. (1991) Acta Crystallogr. Sect. A 47, 110-119
46. Kleywegt, G. J., and Jones, T. A. (1995) Structure (Lond.) 3, 535-540[Medline] [Order article via Infotrieve]
47. Sampey, A. V., Hall, P. H., Mitchell, R. A., Metz, C. N., and Morand, E. F. (2001) Arthritis Rheum. 44, 1273-1280[CrossRef][Medline] [Order article via Infotrieve]
48. Burley, S. K., and Petsko, G. A. (1985) Science 229, 23-28[Abstract/Free Full Text]
49. Subramanya, H. S., Roper, D. I., Dauter, Z., Dodson, E. J., Davies, G. J., Wilson, K. S., and Wigley, D. B. (1996) Biochemistry 35, 792-802[CrossRef][Medline] [Order article via Infotrieve]
50. Kleemann, R., Kapurniotu, A., Frank, R. W., Gessner, A., Mischke, R., Flieger, O., Juttner, S., Brunner, H., and Bernhagen, J. (1998) J. Mol. Biol. 280, 85-102[CrossRef][Medline] [Order article via Infotrieve]
51. Swope, M. D., and Lolis, E. (1999) Rev. Physiol. Biochem. Pharmacol. 139, 1-32[Medline] [Order article via Infotrieve]
52. Esnouf, R. M. (1997) J. Mol. Graph. 15, 132-134[CrossRef][Medline] [Order article via Infotrieve]

Copyright © 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Immunol.Home page
D. Kamir, S. Zierow, L. Leng, Y. Cho, Y. Diaz, J. Griffith, C. McDonald, M. Merk, R. A. Mitchell, J. Trent, et al.
A Leishmania Ortholog of Macrophage Migration Inhibitory Factor Modulates Host Macrophage Responses
J. Immunol., June 15, 2008; 180(12): 8250 - 8261.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
B. E. Rendon, T. Roger, I. Teneng, M. Zhao, Y. Al-Abed, T. Calandra, and R. A. Mitchell
Regulation of Human Lung Adenocarcinoma Cell Migration and Invasion by Macrophage Migration Inhibitory Factor
J. Biol. Chem., October 12, 2007; 282(41): 29910 - 29918.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
G. V. Crichlow, K. F. Cheng, D. Dabideen, M. Ochani, B. Aljabari, V. A. Pavlov, E. J. Miller, E. Lolis, and Y. Al-Abed
Alternative Chemical Modifications Reverse the Binding Orientation of a Pharmacophore Scaffold in the Active Site of Macrophage Migration Inhibitory Factor
J. Biol. Chem., August 10, 2007; 282(32): 23089 - 23095.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
Y. Cho, B. F. Jones, J. J. Vermeire, L. Leng, L. DiFedele, L. M. Harrison, H. Xiong, Y.-K. A. Kwong, Y. Chen, R. Bucala, et al.
Structural and Functional Characterization of a Secreted Hookworm Macrophage Migration Inhibitory Factor (MIF) That Interacts with the Human MIF Receptor CD74
J. Biol. Chem., August 10, 2007; 282(32): 23447 - 23456.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
M. Winner, A. C. Koong, B. E. Rendon, W. Zundel, and R. A. Mitchell
Amplification of Tumor Hypoxic Responses by Macrophage Migration Inhibitory Factor-Dependent Hypoxia-Inducible Factor Stabilization
Cancer Res., January 1, 2007; 67(1): 186 - 193.
[Abstract] [Full Text] [PDF]