Proteolysis Targeting Chimera (PROTAC) for Macrophage Migration Inhibitory Factor (MIF) Has Anti‐Proliferative Activity in Lung Cancer Cells

Abstract Macrophage migration inhibitory factor (MIF) is involved in protein‐protein interactions that play key roles in inflammation and cancer. Current strategies to develop small molecule modulators of MIF functions are mainly restricted to the MIF tautomerase active site. Here, we use this site to develop proteolysis targeting chimera (PROTAC) in order to eliminate MIF from its protein‐protein interaction network. We report the first potent MIF‐directed PROTAC, denoted MD13, which induced almost complete MIF degradation at low micromolar concentrations with a DC50 around 100 nM in A549 cells. MD13 suppresses the proliferation of A549 cells, which can be explained by deactivation of the MAPK pathway and subsequent induction of cell cycle arrest at the G2/M phase. MD13 also exhibits antiproliferative effect in a 3D tumor spheroid model. In conclusion, we describe the first MIF‐directed PROTAC (MD13) as a research tool, which also demonstrates the potential of PROTACs in cancer therapy.


Introduction
Cancer treatment has improved enormously over the past decades,but unfortunately cancer remains one of the leading health problems worldwide.T wo important reasons that limit the success of cancer treatments are heterogeneity of the tumor and acquired therapy resistance. [1] To address these problems,i tis imperative to discover and exploit previously unrecognized molecular mechanisms that are involved in cell proliferation. Thep rotein macrophage migration inhibitory factor (MIF) has been implicated in the pathogenesis of cancers. [2] Overexpression of MIF has been detected in cancer types such as genitourinary cancer, [3] melanoma, [4] neuroblastoma, [5] and lung carcinoma. [6] Remarkably,d own-regulation of MIF expression by gene-knockout [7] or gene-knockdown [8,9] not only reduced tumor progression and metastases, but also induced antitumor immune responses. [10] These results indicate that targeting MIF could be ap romising strategy towards development of novel cancer therapeutics.
MIF exists as ah omotrimer in which each monomer consists of a1 14-amino acid peptide. [11] Initial evidence indicated an important role for MIF in inflammation and immune responses.Subsequently,MIF was also discovered to function as ah ormone, [12] achemokine [13] and as am olecular chaperone. [14] MIF exerts its functions mainly through protein-protein interactions with membrane-bound receptors or intracellular signaling proteins.O ne of those receptors is cluster of differentiation 74 (CD74), which is the cognate receptor for MIF. [15,16] Thei nteraction between MIF and CD74 triggers activation of the mitogen-activated protein kinase (MAPK) pathway and inhibits the p53 pathway,which results in cell growth. [3] In addition, other non-cognate binding partners such as CXCR4 also play key roles in cancer development. [17,18] Therefore,t he discovery of reagents that interfere with the interaction between MIF and CD74 or other binding partners is an attractive strategy to inhibit MIFinduced cellular signaling in relevant disease models.
Apart from its function as ac ytokine,M IF also harbors enzymatic activity to catalyze keto-enol tautomerization of substrates such as D-dopachrome and 4-hydroxylphenylpyruvate (4-HPP). [19] MIF exerts the tautomerase activity through its proline-1, which is an ucleophile. [20] So far, the physiological function of the enzymatic activity remains elusive.I nterestingly,s ome key amino acid residues in close proximity to the tautomerase active site are involved in binding to CD74 and CXCR4. [21][22][23] This implies that small molecule inhibitors of MIF tautomerase activity are able to interfere with the MIF-receptor interactions.B ased on this idea, several series of small-molecule inhibitors for MIF tautomerase activity have been developed. [24,25] One of the earliest discovered MIF inhibitors is ISO 1 (Figure 1A), which gained wide use as ar eference compound in MIF research. [26] However,the binding potencyofISO 1 for MIF is only in the micromolar concentration range.R esearch over the past two decades yielded several MIF inhibitors with nanomolar potency. Previously,our group and others reported structure-activity relationships (SARs) for 7-hydroxycoumarin derivatives based on inhibitor 2 ( Figure 1A). [27,28] We also found that compounds containing a7 -hydroxy-3,4dihydrobenzoxazin-2-ones backbone such as 3 can also provide potent inhibition against MIF. [27,29] Furthermore,t he Jorgensen lab discovered potent MIF inhibitors that contain ab iaryltriazole or pyrazole scaffold. [30,31] However,t he potencyt oi nhibit MIF tautomerase activity does not always correlate well with the potencyt oi nhibit the MIF-CD74 interaction or MIF induced signaling in cell-based studies. [32,33] Altogether,this suggests that development of molecules that merely bind to the tautomerase enzyme active site may not be enough to effectively interfere with MIF protein-protein interactions.I na ddition, proteasome-dependent MIF degradation induced by HSP90 suppression proved to be correlated with the inhibition of MIF activity on cell proliferation. [34] Therefore,w es eek to use proteolysis-targeting chimeras (PROTACs) as an alternative strategy to attenuate MIF functions by depletion of MIF protein.
ThePROTACstrategy has emerged as anovel concept in small-molecule drug discovery.This strategy employs aheterobifunctional molecule that binds both the protein of interest and an E3 ubiquitin ligase to form at ernary complex ( Figure 1B). [35] This enables hijacking the E3 ubiquitin ligase activity to ubiquitinate the protein of interest that is subsequently degraded by the ubiquitin-proteasome system. After degradation of the protein of interest the PROTACcan be recycled for an ew round of targeted degradation of the protein of interest, thus providing ac atalytic cycle.I mportantly,t he PROTACs trategy enables downregulation of intercellular protein levels of the protein of interest rather than just blocking one of its respective catalytic activities or interaction surfaces.A fter its development by the groups of Crews and Deshaies, [36] the PROTACstrategy has progressed enormously by virtue of the identification of potent and selective E3 ligase ligands such as pomalidomide 4 (Fig-ure 1C). [37] Over the past years,a ni ncreasing number of proteins have been targeted by PROTACs,including kinases, epigenetic editors,b romodomains,n uclear receptors and others. [38] However,P ROTACd evelopment has been largely limited to clinically validated targets for which marketed drugs are available.The next step to unleash the full potential of PROTACd evelopment is targeting the traditionally undruggable proteome,f or example,p roteins involved in protein-protein interactions. [39] In the present study,w er eport the first MIF-directed PROTACs by linking potent MIF binding molecules to pomalidomide as al igand for the cereblon Cullin RING E3 ubiquitin ligase complex. Through investigation of the structure-activity relationship,w ediscovered ap otent PRO-TACM IF degrader (MD13)w ith aD C 50 < 100 nM and aD max > 90 %i nA 549 cells.C ontrol experiments were performed to demonstrate that MD13 reduced the MIF level via cereblon ligand-induced degradation. Moreover, MD13 inhibited the growth of cancer cells in a2 Da nd a3 Dc ell culture systems.A ltogether,d evelopment of these MIF degraders indicates an ew strategy for treatment of cancers and also provides an ew class of tools to study MIF.

PROTAC Design and Synthesis
Compound 2 and 3 ( Figure 1A)a re inhibitors of MIF tautomerase activity with nanomolar potency.B ased on the known pharmacophoric features of MIF tautomerase inhibitors derived from crystal structures [28,29] and docking studies ( Figure 2), we presume that the aromatic hydroxyl functionality is deeply embedded in the tautomerase active site,where it is involved in two key hydrogen-bonding interactions with Asn97. Consequently,t he methoxyphenyl functionality in 2 and the ortho-dimethoxyphenyl functionality in 3 protrude out of the pocket and are solvent-exposed. Therefore,w e replaced the methoxy-functionalities by an amine to enable attachment of al inker by an amidation reaction. Both MIF tautomerase inhibitor 2 and 3 were used as MIF binding ligand in PROTACs for MIF degradation. Compound 2 was linked to the E3 ligase ligand pomalidomide to provide PROTACs indicated as group 1a nd compound 3 was linked to pomalidomide to provide PROTACs indicated as group 2 (Table S2).

PROTACs Retain Their Inhibitory Potency for MIF Tautomerase Activity
To verify binding of the resulting PROTACs to MIF,w e measured their ability to inhibit MIF tautomerase activity. Thec andidate PROTACsi nhibited MIF enzymatic activity with nanomolar inhibition constants (K i ). Theg roup 1P RO-TACs that contain a7 -hydroxycoumarin MIF binding core provided K i values between 117 to 999 nM, which is in the same range as the K i of MIF inhibitor 2 that was reported to be 370 nM (Table S2). [27] Interestingly,the PROTACs MD1-4 with one carbon atom between the triazole and the amide functionality are more potent MIF tautomerase inhibitors compared to PROTACsw ith two (MD5)o rt hree carbon atoms (MD6)i nt his position. The K i values for the PROTACs of group 2w ith a7 -hydroxy-3,4-dihydrobenzoxazin-2-one MIF binding core were all around 100 nM. This is very well in line with the potencyo ft heir parent inhibitor 3, which has a K i value of 150 nM. [27] Ther esults demonstrated that our design strategy for linkers did not or minimally perturb target engagement.

PROTACs Induce MIF Degradation
Previous studies have demonstrated that A549 cells express ah igh-level of MIF. [41] In addition, A549 cells have been successfully used for assessing activity of cereblon ligand-based PROTACsp reviously. [42] Therefore,A 549 is as uitable cell line for evaluating the effect of our putative MIF-directed PROTACs MD1 to MD12 on MIF protein levels.T he reduction in MIF levels were monitored in A549 cells that were treated with two different PROTACc oncen-trations (20 and 2 mM) for 12 hino rder to estimate the dose dependency, which is important for PROTACsbecause of the Hook effect. [35] TheM IF levels in cell lysates were analyzed using an enzyme-linked immunosorbent assay (ELISA).
Tr eatment with 20 mMo fa ny of the putative MIF PROTACs resulted in lower MIF protein levels in A549 cells compared to vehicle-treated controls (Table S2 and Figure S1). Theo nly exception is MD6 that did not trigger MIF reduction. Thes eries of PROTACs with 2 as warhead showed increasing potencyw ith increasing linker length to reach more than 50 %r eduction in the MIF protein levels upon treatment with 20 mM MD4 and MD5.H owever,a t ac oncentration of 2 mMn os ignificant degradation of MIF was observed for this series of compounds.
In the new series of PROTACsu sing 3 as MIF binding ligand, treatment with 20 mMo fe ither of the three compounds with aliphatic linkers (MD7-9)resulted in more than 50 %lower MIF-protein levels compared to control, whereas this was only observed for MD10 for the series of compounds with at riazole in the linker (Table 1a nd Table S2). Subsequently,t he potency of the PROTACs at 2 mMw as investigated, which demonstrated the highest potencyf or MD9 and MD10.B ased on these data MD9 was selected as the most promising starting point to develop MIF-directed PROTACs further.
Thec ellular effect of MD9 as aP ROTACw as further investigated. In comparison to the vehicle control, MD9 reduced the MIF levels in ad ose-dependent manner to provide amaximal degradation of more than 90 %with ahalfmaximal degradation concentration (DC 50 )a round 1.5 mM measured by both ELISA and western-blot ( Figure S2). The action mode of MD9 induced MIF degradation is investigated. MIF degradation becomes visible after 3hours of treatment and reached its maximum effect after 6and 9hours with 10 mM MD9 treatment. Thedegradation can be rescued with pretreatment of 1, 3, 4 or proteasome inhibitor Bortezomib, [44] which indicates that the action of MD9 depends on MIF Figure 2. Design of MIF targeting PROTACs. Optimal binding poses of MIF with 2 (A, PDB 1GCZ) [28] and 3 (B, PDB 5HVT). [29] MIF is shown as apale-green cartoon and the key residues forming the binding pocket are represented as sticks. Docking studies were performed with Discovery Studio and models were prepared with Pymol. Table 1: Optimizationo flinker length of MD9 and control compound with impaired cereblon-binding ligand.
binding as well as on CRBN E3 ligase binding,which triggers proteasome-mediated degradation ( Figure S3).

Development of aMIF-Directed PROTAC with Improved Potency
Although MD9 was identified as an effective MIFdirected PROTAC, its potency remains limited to the micromolar concentration range.T he structure-activity relationship (SAR) analysis of the PROTACl inkers of the second group suggests that alonger aliphatic linker (MD7-9)ismore favorable for MIF degradation. To further explore the SARs and to improve the efficacy of MIF PROTACs,w ed esigned and synthesized MD13 and MD14,w hich contain longer linkers than MD9.Both new PROTACs showed MIF binding constants (K i )i narange similar to the parental ligand 3 (Table 1). Subsequently,t he reduction of MIF levels upon treatment with MD13 and MD14 at concentration of 2a nd 0.2 mMw as investigated. MD13 treatment resulted in 91 % and 71 %l ower MIF protein levels at 2a nd 0.2 mM, respectively.T his indicates that MD13 is the most potent MIF PROTACint his series.Inline with current knowledge, the length of the linker appears to be critical for the potency of MIF-directed PROTACs and the linker length of MD13 seems to be optimal.
To further confirm the action of MD13 as aMIF-degrading PROTAC, we synthesized acontrol compound for MD13 containing aCRBN ligand with impaired CRBN binding.The imide nitrogen of the piperidine-2,6-dione functionality in the CRBN ligand is involved in ac rucial hydrogen bond with CRBN. [45] Methylation of this imide nitrogen will abolish CRBN binding. [46] We synthesized control compound MD15 with am ethylated pomalidomide as CRBN ligand. MD15 preserved the MIF binding potencyw ith a K i of 55 nM (Table 1). However, MD15 was not capable of inducing MIF degradation at both 2a nd 0.2 mM, whereas MD13 was. Collectively,t his result confirms that MD13 induced MIF degradation through binding to E3 ligase cereblon.

Characterization of MD13 as aMIF-Directed PROTAC
Since PROTACsa re relatively large heterobifunctional molecules,the efficacy of these compounds may be limited by poor cell permeability. [47] In order to estimate the cellular uptake of PROTAC MD13,the intrinsic fluorescence properties of the pomalidomide part of MD13 were employed for visualization of its subcellular localization. Clear localization of MD13 in the cytoplasm of A549 cells was observed after one-hour incubation ( Figure S5). In parallel, the subcellular localization of MIF and its decrease in situ upon MD13 treatment was visualized by confocal fluorescence microscopy using afluorescent secondary antibody.T reatment with 1 mM MD13 significantly depleted MIF in A549 cells ( Figure S5). Taken together,microscopic analysis revealed that MD13 can enter cells to effectively induce MIF degradation.
Theconcentration dependence of PROTAC MD13-mediated induction of MIF degradation in A549 cells was investigated using western-blot. MD13 effectively induced MIF degradation at nanomolar concentrations ( Figure 3A). TheMIF levels were normalized to the vehicle treated control and plotted to the respective MD13 concentrations.T his provided aD C 50 of around 100 nM and am aximal degradation of around 90-95 %a tc oncentrations higher than 1 mM ( Figure 3B). TheD C 50 of MD13 measured by ELISA assay was about 200 nM ( Figure 3C), which is in line with the result from western-blot. Interestingly,a"Hook effect" was observed in both two assays at 20 mMo fMD13.
Ac ontrol experiment was performed to compare MD13 as an active PROTACa nd MD15 as an inactive PROTAC ( Figure 3D). This demonstrated that MD15 was not able to reduce the MIF levels relative to the control, thus indicating that CRBN binding is involved in the effect of PROTAC MD13.
Theability of PROTAC MD13 to reduce the MIF levels in A549 cells was investigated further. Thek inetics of MIF degradation proved to be relatively fast. Degradation was already visible after 3-hour treatment, reaching > 92 % degradation after 6h( Figure 4A). Only as light recovery of the MIF levels was observed after 48 h. Combined treatment with PROTAC MD13 and the proteasome inhibitor Borte-

Angewandte Chemie
Research Articles zomib [44] inhibited the degradation of MIF ( Figure 4B). Pretreatment of cells with either MIF inhibitor 3 or CRBN inhibitor 4 to outcompete the formation of ternary E3 ligase-MIF complex rescued MIF from degradation (Figure 4C). TheP ROTAC MD13 also proved to be active in HEK293 cells,w here it induced more than 90 %M IF degradation at aconcentration of 200 nM ( Figure 4D). Taken together,t he results demonstrate that the activity of MD13 depends on binding to both MIF and CRBN as well as on proteasome activity and that near complete MIF degradation is observed in the low micromolar range,which indicates that MD13 is apotent MIF-directed PROTAC.

Anti-Proliferative Effect of MD13
After having identified MD13 acts as aP ROTACt hat effectively reduces the MIF levels,w ee mployed this PRO-TACt ov erify the role of MIF in proliferation of A549 cells. As afirst step,the toxicity of MD13 was investigated using the MTS assay,w hich indicated that MD13 did not inhibit cell viability at concentrations below 20 mMf or at reatment of 24 hours ( Figure S7). We next evaluated its effects on cell proliferation, which indicated that MD13 inhibited the growth of A549 cells in ad ose-dependent manner (Figure 5A). Thei nhibitory effect became visible at nanomolar concentrations and reached about 50 %i nhibition of cell proliferation at ac oncentration of 20 mM. In contrast, the inactive control compound MD15 showed almost no inhibition of the proliferation of A549 cells.M IF inhibitor 3 and CRBN inhibitor 4 were also included as controls,b oth of which had no effect on the proliferation of cells with concentrations up to 20 mM. Taken together,t hese experiments indicates that the MIF-directed PROTAC MD13 inhibited cell proliferation of A549 cancer cells. A3 Ds pheroid model was employed to investigate the effect of longer term MD13 treatment in am ore complex model for tumor growth. The3 Ds pheroid model was established using A549 cancer cells by am ethod adapted from Feng et al. [48] Thes pheroids were grown over a1 2-day period in absence or presence of PROTAC MD13.E ach spheroid was prepared from about 1000 A549 cells.A fter three-day incubation, these spheroids were treated with 1, 2, or 5 mMo fMD13 with 72 hours intervals over 12 days. Spheroid growth was monitored by measuring the diameter and this was compared to day 0oft he treatment. Thet umor spheroids in MD13 treated groups were significant smaller compared to the control group ( Figure 5B and Figure S9). With continuous exposure to 1, 2, or 5 mMo fMD13 for 12 days,t he growth of the spheroid tumor volume was inhibited by 42 %, 53 %, and 81 %c ompared with control group,r espectively.I nc ontrast, 5 mMo ft he PROTACinactive control compound MD15, 3,o r4 showed no significant influence on the spheroid tumor growth. Collectively,o ur results indicate that the MIF-directed PROTAC MD13 effectively inhibits proliferation of A549 cancer cells in aspheroid tumor model.

MD13 Arrests Cells at G2/M Phase of the Cell Cycle
Theeffect of MIF-directed PROTAC MD13 on cell cycle progression was further analyzed using flow cytometry (Figure 6). A549 cells were treated with MD13 at concentrations of 1, 2, or 5 mMfor the duration of 48 hbefore analysis using flow cytometry.Our results showed that MD13 dose-dependently induced cell cycle arrest at the G2/M phase in A549 cells.The proportion of cells at the G2/M phase is 12 %for the control group.T his percentage increases to 17 %, 19 %, and 23 %upon treatment with 1, 2, and 5 mM MD13,respectively. In contrast, little or no effect on the cell cycle was observed upon treatment with 5 mMo ft he inactive control MD15. These results indicate MD13 induces inhibition of cell cycle progression, which can explain the observed inhibition of cell proliferation.

MD13 Inhibits ERK Signaling
Thee ffect of treatment with MIF-directed PROTAC MD13 on MIF-related signaling pathways was investigated by assessment of ERK phosphorylation using western blot analysis.T reatment with 2 mM MD13 proved to inhibit ERK phosphorylation in A549 by about 50 %a fter 24-hour treatment, which persisted at 48 h. In contrast, the cells exhibited no significant decrease on the pERK levels after incubation with the control compound 3, 4 for 24 horMD15 for 6h,24h, or 48 h ( Figure 7). Thus treatment with the MIF-directed PROTAC MD13 inhibits ERK phosphorylation as aM IFrelated signaling event.

Conclusion
Overexpression of MIF was found to stimulate proliferation of cancer cells via activation of the ERK/MAPK pathway and inhibition of the p53 pathway. [49,50] Therefore, anumber of MIF targeting modalities have been reported as potential treatments,i ncluding mAbs, [51] peptides, [52] smallmolecule inhibitors [24,25] etc.T hese modalities have been successfully applied in animal models for MIF-related diseases. [52] However, there is no clinically approved MIFdirected drug available yet. Use of PROTACs that trigger degradation of the MIF protein provides novel opportunities that might be particularly relevant for MIF.Importantly,MIF is involved in protein-protein interactions,s uch as the MIF-CD74 receptor interaction for which the interactions site is known to be located in close proximity of the MIF tautomerase active site. [3] However,other protein-protein interactions might occur at different locations of the MIF protein, thus making approaches aimed at MIF tautomerase activity ineffective.I nt his perspective the value of MIF-directed PROTACs becomes clear, because the high affinity ligands identified for the MIF tautomerase active site can be employed to induce degradation of the MIF protein as awhole,thus diminishing MIF from its effector network.
Thedevelopment of MIF-directed PROTACs requires the synthesis of heterobifunctional ligands that are able to bind both MIF and E3 ubiquitin ligase.Optimization of the linker is required to achieve ap roper orientation to trigger ubiquitination and subsequent degradation. To synthesize MIF-targeting PROTACs,w et ethered MIF binder 2 [27,28] or 3 [27,29] with the cereblon E3 ligase ligand pomalidomide by av ariety of linkers constructed by click reactions or amidation coupling reactions.U pon exploration of the structureactivity relationships for MIF degradation, we identified MD9 as the first MIF-directed PROTAC. Further optimization of the linker length provides MD13 as aMIF-directed PROTAC with improved potency,w hich proved to trigger almost complete (90-95 %) degradation of MIF in the low micromolar range and aDC 50 of around 100 nM on A549 cells.The potencyo fMD13 is comparable to PROTACs directed at other protein targets. [38] Fluorescence microscopy demonstrated that MD13 effectively entered the cytosol and reduced the MIF protein levels by about 80 %w ithin 3hours.T he reduction in MIF protein levels upon treatment with 2 mMof the MIF-directed PROTAC MD13 was still observed after 48 hours.A sabona fide MIF-targeting PROTAC, MD13 should induce the degradation through the formation of at ernary complex, which is followed by ubiquitination and proteasome-mediated proteolysis.Accordingly,rescue assays were conducted using 3 as acompetitor for MIF binding, 4 as ac ompetitor of E3 ligase binding and bortezomib as ap roteasome inhibitor. Our results showed that 3, 4,a nd bortezomib were all able to abolish the MD13 triggered degradation, thus indicating that the activity of MD13 depends on MIF binding,C RBN-binding,a nd proteasome mediated degradation. Importantly,t he control compound MD15,w hich contains an impaired E3 ligase ligand, has no effect on MIF protein level. Taken together, MD13 proved to be an effective and potent MIF-directed PROTAC.  Effect of the MIF-directedP ROTAC on ERK phosphorylation in A549 cells. A) A549 cells were treated with 2 mMofMD13, MD15, 3, 4,or DMSO for 24 h, the pERK, total ERK and GAPDH was examinedb yimmunoblots. B) A549 cells were treated with MD13, MD15 or DMSO for 6, 24 or 48 h, the pERK, total ERK and GAPDH was examined. C) Quantification of the pERK level using pERK:ERK ratio, normalizedtocontrol group at time points indicated. GAPDH was used as al oading control on western blots. Data are shown as mean AE SD of three replicates. **p < 0.01 and ***p < 0.001 vs. vehicle group.
Thee ffect of MD13 on cell proliferation was evaluated using cell culture assays on A549 cells.Amonolayer cell culture assay demonstrated that MD13 inhibited proliferation of A549 cells to am aximum of about 50 %a t2 0mM. In contrast, the CRBN inactive control MD15,t he MIF tautomerase inhibitor 3,o rt he E3 ligase ligand 4 had no or little effect on cell proliferation. Also aspheroid cell culture assay was employed because such assays mimic the main features of solid human tumors,s uch as their structural organization, cellular layered assembling,h ypoxia, and nutrient gradients. [53] In this spheroid assay, MD13 inhibited the growth of the spheroid volume by 53 %and 81 %upon treatment with 2 and 5 mMo fMD13 respectively.T hese results indicate that depletion of MIF using MIF-directed PROTACs provides as trong reduction of cell proliferation, which is consistent with the results of siRNAm ediated MIF silencing. [9,54] Growth of cancer cells is characterized by ordered progression of the cell cycle. [55] MIF coordinates the cell cycle through the association with the Jab1/CSN5 subunit of the COP9/CSN signalosome, [56] which plays ac entral role in the assembly of SCF complexes by removal of Nedd8 from Cullin. [57][58][59] MIF knockout leads to DNAdamage and stalled replication. [60] Treatment of A549 cancer cells with the MIFdirected PROTAC MD13 increased the number of cells in the G2/M phase thus indicating inhibition of cell cycle progression. MIF as agrowth factor stimulates cell cycle progression through the MAPK pathway. [61,62] Our results also demonstrate that MD13 treatment attenuates the MAPK signaling by reducing ERK phosphorylation. This result is again in line with the effect observed upon siRNA-mediated downregulation of the MIF protein levels. [54] Collectively,t hese results indicate that the MIF-directed PROTAC MD13 reduces the MIF protein levels and inhibits cell proliferation in both 2Dand 3D-cell culture,which can be explained by inhibition of ERK phosphorylation and cell cycle progression.
In conclusion, we have developed ap otent MIF-directed PROTAC MD13 that induces MIF degradation in A549 and HEK 293 cells. MD13 effectively reduces the MIF protein level in A549 cells in at ime-, cereblon-, and proteasomedependent manner. Fluorescence microscopy demonstrates that MD13 enters A549 cells with concomitant reduction of the MIF levels. MD13 inhibited proliferation by about 50 %at micromolar concentrations in a2 Dc ell culture assay using A549 cells.A3D cell culture also using A549 cells showed an even more pronounced effect with 80 %r eduction of cell proliferation at 5 mM MD13.F ACSa nalysis demonstrated that MD13 treatment induced cell cycle arrest in the G2/M phase. MD13 treatment also inhibited ERK phosphorylation, thus indicating that MIF degradation also inhibits signaling pathways that respond to MIF signaling and promote cell proliferation. In conclusion, the MIF-directed PROTAC MD13 mediates MIF degradation, which consequently results in inhibition of cell proliferation in 2D and 3D cell cultures, which can be explained by cell cycle arrest and inhibition of the MAPK signaling pathway.Altogether,this study demonstrates that MIF-directed PROTACs are novel modalities in MIF-directed drug discovery for oncology and other MIF related diseases.