Aggregation InhibitorsDOI: 10.1002/anie.200704051

Rhodanine-Based Tau Aggregation Inhibitors in Cell Models ofTauopathy**Bruno Bulic, Marcus Pickhardt, Inna Khlistunova, Jacek Biernat, Eva-Maria Mandelkow,Eckhard Mandelkow,* and Herbert Waldmann*

The two histopathological hallmarks that characterize Alz-heimer�s disease (AD) are the extracellular amyloid plaquesthat are formed by b-amyloid fragments of the amyloidprecursor protein (APP), and intracellular neurofibrillarytangles and neuropil threads, which consist of the micro-tubule-associated protein tau forming paired helical filamentswith highly ordered structures, as recently corroborated by X-ray microcrystallography.[1] In addition, tau deposits similar tothose of AD occur in several “tauopathies”.[2,3] The normalfunction of tau is to stabilize the microtubule network for thetransport of vesicles and organelles in nerve cells, which isnecessary for the communication between cells and thus forbrain activity. When tau aggregates, it is thought that thetracks for transport (microtubules) break down and thetransport is interrupted.[4–6] Moreover, the relevance of tau forneurodegeneration induced by b amyloid has been demon-strated in a mouse model.[7,8] It would therefore be highlydesirable to find methods to keep tau in a functional state andprevent or reverse abnormal aggregation. The quest for curesfor Alzheimer�s disease is very intense. Available therapies todate make use of cholinesterase inhibitors and NMDAreceptor antagonists,[9,10] and newer approaches focus, for

example, on inhibition of tau phosphorylation and amino-peptidase activation.[11–14]

Thus, the development of tau aggregation inhibitors thatare also able to disaggregate filaments could provide analternative to existing strategies.[15–18] Herein we report theinvestigation of substituted rhodanines with these propertiesin vitro and in a cell model consisting of a neuroblastoma cellline that expresses tau in an inducible fashion with subsequentaggregation. In an initial high-throughput screen,[19,20] severalaggregation inhibitors were identified. From these hits,rhodanines (2-thioxothiazolidin-4-ones, Figure 1a) were

selected for synthesis of a collection because of their activityin the screening and because the rhodanine core in generalhas been shown to be a viable scaffold for the development ofbiologically active molecules. Thus, rhodanines are classifiedas nonmutagenic,[21] and a long-term study on the clinicaleffects of the rhodanine-based Epalrestat demonstrated thatit is well tolerated.[22] Several assays were performed todevelop compounds that would show strong effects on theassembly and disassembly of paired helical filaments (char-acterized by their IC50 and DC50 values, corresponding to thehalf maximal compound concentration necessary for inhib-ition of tau assembly into aggregates and disassembly ofpreformed filament aggregates, respectively). We performeda tau aggregation assay in vitro based on the fluorescencechange of the aggregate-specific stain thioflavin S (ThS) as areadout and using the three-repeat tau construct K19 derivedfrom the fetal isoform htau23[19,23] (see the SupportingInformation). In addition, we investigated the aggregationof the four-repeat construct K18, either with the wild-typesequence, or with two types of mutations. In the “proaggre-gation” mutant K18DK280, the absence of K280 stronglyenhances aggregation.[24] The opposite behavior is observedwith the “antiaggregation” mutant by the insertion of pro-lines, which act as breakers of b structure and thus preventaggregation (I277P, I308P).[25]

Figure 1. Variation of inhibitor structure. a) Structure of the hit com-pound. Variations of flanking regions of the central rhodanine core(part B) are possible in parts A and C. b) Variations of the core (R1 andR2) and on the flanking substituents (R3 and R4).

[*] Dr. M. Pickhardt,[+] Dr. I. Khlistunova, Dr. J. Biernat,Dr. E.-M. Mandelkow, Prof. E. MandelkowMax-Planck-Unit for Structural Molecular Biologyc/o DESY, Notkestrasse 85, 22607 Hamburg (Germany)Fax: (+49)408-971-6810E-mail: mand@mpasmb.desy.de

Dr. B. Bulic,[+] Prof. H. WaldmannMax-Planck-Institute for Molecular PhysiologyOtto-Hahn-Strasse 11, 44227 Dortmund (Germany)andCenter for Applied Chemical GenomicsOtto-Hahn-Strasse 15, 44227 Dortmund (Germany)E-mail: herbert.waldmann@mpi-dortmund.mpg.deandUniversitCt Dortmund, Fachbereich 3Otto-Hahn-Strasse 6, 44227 Dortmund (Germany)Fax: (+49)231-1332499

[+] These authors contributed equally to this work.

[**] This work was supported in part by grants from the DeutscheForschungsgemeinschaft (DFG), the Institute for the Study of Aging(ISOA), the European Union (“EuropCischer Fond fGr regionaleEntwicklung”), and the state of North Rhine-Westphalia. We thankSabrina HGbschmann and Ilka Lindner for excellent technicalassistance.

Supporting information for this article (synthesis and character-ization data) is available on the WWW under http://www.ange-wandte.org or from the author.

AngewandteChemie

9215Angew. Chem. Int. Ed. 2007, 46, 9215 –9219 � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& Take advantage of blue reference links &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

For the synthesis of a compound collection we employedan iterative approach and focused on the heterocycle as wellas its substituents. Initial attempts focused on immobilizationof an N-carboxyalkyl-substituted rhodanine core on a poly-meric resin by means of an ester bond, and subsequentKnoevenagel reaction with aromatic aldehydes equipped witha suitable functional group to allow formation of the biarylbond by means of a Pd0-catalyzed aryl coupling reaction.However, these attempts failed because the Pd0-catalyzedreaction was inhibited by the rhodanine heterocycle. Thus, thesolution-phase synthesis sequence shown in Scheme 1 was

employed; this sequence involves formation of biaryl alde-hydes and subsequent Knoevenagel condensation with therhodanine core as key steps (for a solid-phase synthesis ofrhodanines by a different strategy, see reference [28]). Thearyl aldehydes necessary for the Knoevenagel condensation(part C, Figure 1a) were obtained by Suzuki couplingbetween haloaromatics and functionalized boronic acids(Scheme 1).

Initially, we focused on the central heterocycle itself,replacing the original rhodanine core with other heterocycles(R1 and R2, Figure 1b). In these experiments, rhodanines(R1 =S and R2 = S), thiohydantoin (R1 = S and R2 =N),thioxooxazolidine (R1 = S and R2 =O), oxazolidinedione(R1 =O and R2 =O), and hydantoin (R1 =O and R2 =N)were employed (see the Supporting Information) and thefollowing trend in inhibition of tau aggregation was observed:rhodanine (1)> thiohydantoin (3)@ oxazolidinedione (7)>thioxooxazolidinone (9)> hydantoin (10). The rhodanineheterocycle appeared to be the most potent. The thioxogroup in rhodanines is known as a carboxylic acid bioisosterby size, low electronegativity, and ability to build hydrogenbonds.[26,27] On the basis of these observations, the rhodanine

heterocycle was kept, and substituents A and C were varied(Figure 1a).

We investigated the importance of the carboxylic acid (A,Figure 1a) and the influence of the substitution and length ofthe linker connecting the central core (B, Figure 1a) to thecarboxylic acid (Figure 2). Replacement of the carboxylicacid with an imidazole or a benzimidazole as well asesterification led to reduced disassembly activity (Table 1,entries 1–4). Furthermore, the length of the linker betweenthe carboxylic acid and the rhodanine core (B, Figure 1a) wasvaried. These experiments revealed that increasing the

distance up to two carbonbonds resulted in an appre-ciable increase in the com-pound�s inhibitory potencywithout markedly affectingthe disassembly activity(Table 1, entries 1, 5, and6). In subsequent experi-ments, biaryl part C of thecompounds (Figure 1a)was varied. The heteroaro-matic side chain (part C,Figure 1a) tolerated varia-tions, but modifications onthe furan heterocycle ledto reduced potency(Table 1, entries 1, 5, and7–9, and the SupportingInformation), probably asa result of both electronicand steric changes, asreplacement of the furanring in 16 for thiophene in22 reduced the potency.Very bulky substituents,such as adamantyl or fer-

rocene (Table 1, entries 10 and 11), at the end of side chain Cwere generally well tolerated, reducing the overall efficiencyof the compounds only slightly. Also, introduction of acharged group by means of a carboxylic acid did not influencethe potency considerably (Table 1, entries 12–14), underliningthe structural flexibility around this position.

After completing the synthesis of our focused library, weobserved that the efficiency of the most potent derivatives arein the nanomolar range for both inhibition and disaggregation(19, IC50 = 0.17 mm, DC50 = 0.13 mm, Table 1). Examples ofdose–response curves are shown in Figure 3, both for the

Scheme 1. Synthesis of a compound collection. Reagents: a) 5 mol% [PdCl2(PPh3)2] , 2m aq Na2CO3, 1,2-dimethoxyethane, 85 8C, 9 h. b) Ethyl isothiocyanatoacetate, Et3N, toluene, reflux, 4 h. c) Bis(carboxymethyl)tri-thiocarbonate, Na2CO3, iPrOH, reflux, 18 h. d) Bis(p-nitrophenyl)carbonate, Et3N, N,N-dimethylformamide,room temperature, 12 h then HCl aq, dioxane, reflux, 2 h. e) For R2=O, NH: NaOAc, dioxane, 90 8C, 2 h; forR2=S: piperidine, CH2Cl2, room temperature, 12 h. f) NaOH, dioxane/water, room temperature, 1 h. g) meta-Chloroperoxybenzoic acid, NaHCO3, CH2Cl2, room temperature, 24 h.

Figure 2. Structure–activity relationship.

Communications

9216 www.angewandte.org � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2007, 46, 9215 –9219

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& Take advantage of blue reference links &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

inhibition of aggregation (Figure 3a) and for the dissolutionof preformed paired helical filaments (PHFs; Figure 3b).

The curves exhibit the typical sigmoidal decrease ofaggregated protein at increasing compound concentrations.

Figure 4 illustrates examples ofPHFs, made from the constructK19, in the process of disassemblyafter incubation overnight withincreasing inhibitor concentrationsof 30. Filaments are seen in variousstages of shortening or breakingand reveal a concentration-depen-dent degree of dissolution.

The compounds were investi-gated in murine neuroblastomaN2a cells to determine whetherthey are also active in cellularsystems. In these experiments, tauexpression was induced by incuba-tion with doxycyclin. The resultsshown in Figure 5 were obtained byswitching on the tau expression bydoxycyclin, incubating the cells withcompound (15 mm), and scoring forThS fluorescence after 5 days. Toassess the depolymerization of pre-formed aggregates, the expressionof tau was induced for 5 days, thencompounds were added, and thelevel of ThS fluorescence was mea-sured after 2 more days. In theseexperiments the compounds wereapplied in 15 mm concentrationbecause they were not cytotoxic tothe cells at this concentration (seebelow).

To exclude that the cell assayresults might be impaired by poten-tial cytotoxicity of the aggregationinhibitors, all compounds wereassayed for cytotoxicity at 10 mm

concentration. Orientating experi-ments employed an established lac-tate dehydrogenase assay (LDH),which reports on the leakiness ofmembranes in degenerating cells.After the cells were incubated for24 hours with 10 mm compound, thedegree of cell lysis was determinedby LDH release. The assay revealedthat the compounds are not or atworst only very weakly cytotoxic(see the SupportingInformation).[19,29] Furthermore, weinvestigated whether some of themost potent compounds interferewith the physiological function oftau, that is, binding to microtubules.An in vitro assay of tubulin poly-

merization in the presence or absence of 14 and 30, whichshowed the highest activity in the cellular assay (inhibition ofca. 70% of the aggregation relative to the untreated control)and a concentration of 60 mm (i.e. four times higher than the

Table 1: Compound structures, IC50 and DC50 values, PHF inhibition in cells, and cytotoxicity.

Entry Compd R3[a] R4[a] IC50 [mm][b] DC50 [mm][b] Inhibition incells[c] [%]

1 1 0.82 0.10 20.40�5.37

2 4 4.36 1.80 n.d.

3 14 0.67 0.94 70.47�4.49

4 49 1.09 0.80 n.d.

5 16 0.47 0.30 21.55�13.82

6 23 1.22 1.04 n.d.

7 22 0.97 0.77 n.d.

8 27 5.03 1.66 n.d.

9 42 7.92 1.40 n.d.

10 44 0.69 0.42 n.d.

11 54 0.37 0.47 n.d.

12 33 0.58 0.52 n.d.

13 34 0.73 0.60 n.d.

14 35 0.78 0.58 n.d.

15 30 0.26 0.16 69.02�3.10

16 17 0.54 0.39 62.67�4.08

17 19 0.17 0.13 16.49�4.38

[a] Substituents R3 and R4 refer to Figure 1b (R1=R2=S). [b] The IC50 and DC50 values represent theassembly-inhibition and disassembly-inducing half-maximal concentrations, respectively, measuredin vitro. For each data point, experiments were performed in triplicate and averaged. The standard errorof the IC50 or DC50 values determined from the curves was 10–20%. [c] The values obtained byincubating the cells with 15 mm compound correspond to the level of inhibition of tau aggregationnormalized to a control without inhibitor (0%) in cells. n.d.: not determined.

AngewandteChemie

9217Angew. Chem. Int. Ed. 2007, 46, 9215 –9219 � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& Take advantage of blue reference links &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

concentration used in the cellular screen), revealed thattubulin polymerization is at most only marginally affected bythe rhodanines. In this established assay, tubulin at 30 mm

without tau served as a negative control because it is unableto self-assemble into microtubules below the critical concen-tration. In the presence of tau (10 mm), tubulin polymerizeswithin 4 minutes (Figure 6).[19]

Taken together, these results demonstrate that the rho-danines investigated by us inhibit tau aggregation and mostimportantly promote paired helical filament disassembly at100–600 nm concentration. They also display activity incellular assays without showing cytotoxicity (at the concen-tration investigated) or interference with the normal function

Figure 3. Dose–response curves. a) Inhibition of tau aggregation byrhodanine-derived compounds. The extent of aggregation is measuredby ThS fluorescence and plotted as a percentage of the untreatedcontrol. The black line represents the initial hit structure 1. b) Dose–response curves for disassembly of preformed PHFs by rhodanine-derived compounds.

Figure 4. Electron micrographs of K19 PHFs treated with differentconcentrations of 30 and 50 for 12 h at 37 8C, showing the breakdownof PHFs into smaller fragments for compound 30. Inactive compound50 was used for a negative control. The methods used are describedelsewhere.[6]

Figure 5. Cell assay of aggregation inhibitors. Expression of K18DK280tau in doxycyclin-inducible N2a cells leads to aggregate formation,which is inhibited by rhodanines. Cells were stained with aggregate-specific thioflavine-S (green, center column) and with tau antibodyK9JA (red, left column). Compounds 30 and 14 were applied for 5 daysat 15 mm (top row, without inhibitor; middle row, with 30 ; bottom row,with 14). Note the disappearance of the ThS stain relative to thecontrol without compound (“DMSO”). Scale bar: 20 mm.

Figure 6. Rhodanines have only a minor effect on tau-induced micro-tubule assembly. Tubulin dimer (30 mm) was incubated with hTau40wt(10 mm) in the presence or absence of rhodanine compounds (60 mm).Samples were incubated in a microtiter plate at 37 8C, and absorptionwas measured continuously at 350 nm and plotted versus time.Tubulin alone g, tubulin and hTau40wt l, rhodanine 14 c*,and rhodanine 30 c~.

Communications

9218 www.angewandte.org � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2007, 46, 9215 –9219

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& Take advantage of blue reference links &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

of tau to promote tubulin polymerization. The initial struc-ture-determining properties of the inhibitors observed in vi-tro, which do not fully translate into the activities seen in thecellular context, may result from different penetrationthrough the cell membrane. Nonetheless, our data suggestthat the compound class identified by us may hold substantialpromise for further development.

Compounds that induce disaggregation of already formedPHFs, prevent aggregation or in the best scenario combineboth properties are of particular interest for medicinalchemistry research focusing on AD.

Received: September 3, 2007Published online: November 5, 2007

.Keywords: aggregation inhibitors · Alzheimer’s disease ·medicinal chemistry · neurochemistry · tau proteins

[1] M. R. Sawaya et al., Nature 2007, 447, 453 – 457; see theSupporting Information. The fact that paired helical filamentshave a highly ordered core has been shown before, for example,by X-ray diffraction; see references [25], [30], and [31].

[2] V. M. Lee, J. Q. Trojanowski, J. Alzheimer%s Dis. 2006, 9, 257 –262.

[3] L. BuIe, T. Bussiere, V. Buee-Scherrer, A. Delacourte, P. R. Hof,Brain Res. Rev. 2000, 33, 95 – 130.

[4] A. Ebneth, R. Godemann, K. Stamer, S. Illenberger, B. Trinczek,E. Mandelkow, J. Cell Biol. 1998, 143, 777 – 794.

[5] K. Stamer, R. Vogel, E. Thies, E. Mandelkow, E. M. Mandelkow,J. Cell Biol. 2002, 156, 1051 – 1063.

[6] E. M. Mandelkow, E. Thies, B. Trinczek, J. Biernat, E. Man-delkow, J. Cell Biol. 2004, 167, 99 – 110.

[7] E. D. Roberson, K. Scearce-Levie, J. J. Palop, F. Yan, I. H.Cheng, T.Wu, H. Gerstein, G.QYu, L. Mucke, Science 2007, 316,750 – 754.

[8] B. Bandyopadhyay, G. Li, H. Yin, J. Kuret, J. Biol. Chem. 2007,282, 16454 – 16464.

[9] A. Martinez, A. Castro, Expert Opin. Invest. Drugs 2006, 15, 1 –12.

[10] R. A. Hansen, G. Gartlehner, K. N. Lohr, D. I. Kaufer, DrugsAging 2007, 24, 155 – 167.

[11] S. L. Karsten et al., Neuron 2006, 51, 549 – 560; see theSupporting Information.

[12] S. Le Corre et al., Proc. Natl. Acad. Sci. USA 2006, 103, 9673 –9678; see the Supporting Information.

[13] W. Noble et al., Proc. Natl. Acad. Sci. USA 2005, 102, 6990 –6995; see the Supporting Information.

[14] K. S. Hung, S. L. Hwang, C. L. Liang, Y. J. Chen, T. H. Lee, J. K.Liu, S. L. Howng, C. H.Wang, J. Neuropathol. Exp. Neurol. 2005,64, 15 – 26.

[15] C. M. Wischik, P. C. Edwards, R. Y. Lai, M. Roth, C. R.Harrington, Proc. Natl. Acad. Sci. USA 1996, 93, 11213 – 11218.

[16] C. Chirita, M. Necula, J. Kuret, Biochemistry 2004, 43, 2879 –2887.

[17] S. Taniguchi, N. Suzuki, M. Masuda, S. Hisanaga, T. Iwatsubo, M.Goedert, M. Hasegawa, J. Biol. Chem. 2004, 280, 7614 – 7623.

[18] C. E. Augelli-Szafran (Warner Lambert Co), WO0076988, 2000.[19] a) M. Pickhardt, Z. Gazova, M. von Bergen, I. Khlistounova,

Y. P. Wang, A. Hascher, J. Biernat, E. M. Mandelkow, E.Mandelkow, J. Biol. Chem. 2004, 280, 3628 – 3635. In thispaper, anthraquinones are described as tau aggregation inhib-itors. However, this compound class is significantly morecytotoxic than the rhodanines described herein; b) M. Pickhardt,G. Larbig, I. Khlistunova, A. Coksezen, B. Meyer, E. M.Mandelkow, B. Schmidt, E. Mandelkow, Biochemistry 2007, 46,10016 – 10023. The described phenylthiazolylhydrazides displaycomparable activity in cellular assay but a lower biochemicalpotency.

[20] M. Pickhardt, M. von Bergen, Z. Gazova, A. Hascher, J. Biernat,E. M. Mandelkow, E. Mandelkow, Curr. Alzheimer Res. 2005, 2,219 – 226.

[21] E. Zeiger, B. Anderson, S. Haworth, T. Lawlor, K. Mortelmans,W. Speck, Environ. Mutagen. 1987, 9, 1 – 110.

[22] N. Hotta et al., Diabetes Care 2006, 29, 1538 – 1544; see theSupporting Information.

[23] P. Friedhoff, A. Schneider, E. M. Mandelkow, E. Mandelkow,Biochemistry 1998, 37, 10223 – 10230.

[24] S. Barghorn, Q. Zheng-Fischhofer, M. Ackmann, J. Biernat, M.von Bergen, E. M. Mandelkow, E. Mandelkow, Biochemistry2000, 39, 11714 – 11721.

[25] M. von Bergen, P. Friedhoff, J. Biernat, J. Heberle, E. M.Mandelkow, E. Mandelkow, Proc. Natl. Acad. Sci. USA 2000,97, 5129 – 5134.

[26] G. A. Patani, E. J. LaVoie, Chem. Rev. 1996, 96, 3147 – 3176.[27] S. Miyamoto, Chem-Bio Inf. J. 2002, 2, 74 – 85.[28] C. H. Lee, M. M. Sui, Tetrahedron Lett. 2000, 41, 5729 – 5732.[29] I. Khlistunova, J. Biernat, Y. P. Wang, M. Pickhardt, M.

von Bergen, Z. Gazova, E. Mandelkow, E. M. Mandelkow, J.Biol. Chem. 2005, 281, 1205 – 1214.

[30] M. von Bergen, S. Barghorn, L. Li, A. Marx, J. Biernat, E. M.Mandelkow, E. Mandelkow, J. Biol. Chem. 2001, 51, 48165 –48174.

[31] J. Berriman, L. C. Serpell, K. A. Oberg, A. L. Fink, M. Goedert,R. A. Crowther, Proc. Natl. Acad. Sci. USA 2003, 100, 9034 –9038.

[32] A. J. Leo, Chem. Rev. 1993, 93, 1281 – 1306.

AngewandteChemie

9219Angew. Chem. Int. Ed. 2007, 46, 9215 –9219 � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& Take advantage of blue reference links &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

Supporting Information

© Wiley-VCH 2007

69451 Weinheim, Germany

Supporting Information: Discovery of rhodanine-based tau aggregation inhibitors in cell

models of tauopathy**

Bruno Bulic, Marcus Pickhardt, Inna Khlistunova, Jacek Biernat, Eva-Maria Mandelkow,

Eckhard Mandelkow,* and Herbert Waldmann*

htau24

1 441P1 P1 2 3 41

1 441P1 P1 3 41htau23

K18

3 41K19

PHF6

2 3 41

PHF6*

VQ I I NK VQIVYK

VQIVYKVQ I I NK

VQPVYKVQPI NK

280

anti-aggregation

Wildtype

pro-aggregation

K18

K18ΔK280

K18ΔK280/I277P/I308P

(M)Q244 K274 V306 E372

htau241 441

P1 P1 2 3 41

1 441P1 P1 3 41htau23

K18

3 41K19

PHF6

2 3 41

PHF6*

VQ I I NK VQIVYK

VQIVYKVQ I I NK

VQPVYKVQPI NK

280

anti-aggregation

Wildtype

pro-aggregation

K18

K18ΔK280

K18ΔK280/I277P/I308P

(M)Q244 K274 V306 E372

Supplementary Figure 1: Tau-isoforms and constructs used in the PHF inhibition assay. From top to bottom: htau24 (4 repeats, no inserts), htau23 (smallest tau isoform, 3 repeats, no inserts), construct K18 (repeat domains with 4 repeats), construct K19 (repeat domain with 3 repeats, R2 absent). Red boxes indicate hexapeptide motifs important for beta structure and aggregation. The "pro-aggregation mutant" K18ΔK280 strongly enhances the rate of aggregation, and "anti-aggregation mutant" K18ΔK280/I277P/I308P prevents β-structure and aggregation.

Supplementary Figure 2: Determination of toxicity of compounds on N2a cells by LDH assay. After incubation of the cells for 24 h with 10 µM compound, the effects on the viability of neuronal cells (after 5 days) are plotted in percentages of the maximum value obtained by destroying cells completely by 5 % Triton X-100; the value of the untreated DMSO control is set as 0% cytotoxicity.

Supplementary Figure 3: Effects of aggregation inhibitors on cells by thioflavin-S staining. The data obtained by incubating the cells with 15 µM compound are normalized to control without inhibitor (100%). Black bar indicates the initial screening substance. The two best compounds 14 and 30 are marked in red, showing inhibitory levels of 70.5% and 69.0%.

Supplementary Table 1. Compound structures, IC50 and DC50 values, PHF inhibition in cells and cytotoxicity.

Entry

Comp-ound

R1 R2

R3

R4

IC50(µM)1

DC50 (µM)1

LDH2

% ±

Inhibition in cell3

% ±

ClogP4

1 1 S S

O

HO O

Cl 0.82 0.10 12.85 20.82 20.4 5.4 0.59

2 2 S N

O

MeO

O

Cl 3.79 0.94 22.79 6.06 N/D N/D 3.36

3 3 S N

O

HO

O

Cl 6.10 0.37 15.21 6.14 N/D N/D 0.14

4 4 S S

O

EtO O

Cl 4.36 1.80 24.55 3.83 N/D N/D 4.35

5 5 S N

O

EtO

O

Cl 3.29 1.86 24.84 6.73 N/D N/D 3.88

6 6 S O

O

EtO

O

Cl 8.94 7.12 -2.03 3.81 N/D N/D 4.13

7 7 O O

O

HO

O

Cl 3.50 2.20 -0.85 2.13 N/D N/D -0.01

8 8 O O

O

EtO

O

Cl 2.56 2.30 -0.48 3.70 N/D N/D 3.73

9 9 S O

O

HO

O

Cl 3.10 2.41 -1.14 3.50 N/D N/D 0.40

10 10 O N

O

EtO

O

Cl 22.57 54.31 -1.45 3.92 N/D N/D 2.64

11 11 S S

O

EtO

HN

B(OH)2

2.83 8.73 -0.32 2.75 N/D N/D 1.80

12 12 S S

O

HO

O

Cl 1.41 1.02 5.30 3.48 49.6 6.1 0.59

13 13 S S

O

HO

COOH

O

Cl 0.42 1.86 2.38 2.16 51.7 6.8 -2.50

14 14 S S

N

HN

O

Cl 0.67 0.94 6.14 5.55 70.5 4.5 5.30

15 15 S S

N

HN

O

1.40 0.80 2.99 5.69 65.6 7.9 6.50

16 16 S S

O

HO

O

Cl 0.47 0.30 3.10 7.80 21.5 13.8 0.60

17 17 S S

N

HN

O

N 0.54 0.39 6.50 5.65 62.7 4.1 5.75

18 18 S S

O

HO

O

Cl 0.65 0.48 N/D N/D N/D N/D 0.34

19 19 S S

O

HO O

0.17 0.13 16.49 4.38 N/D N/D 1.77

20 20 S S

O

HO O O COOH 0.17 0.10 1.20 10.50 36.1 28.2 -3.10

21 21 S S

O

HO O

N 0.24 0.19 1.95 4.49 61.1 26.2 1.03

22 22 S S

O

HO

S

Cl 0.97 0.77 N/D N/D N/D N/D 1.07

23 23 S S

O

HO

O

Cl 1.22 1.04 13.19 9.55 N/D N/D 0.65

24 24 S S

O

HO O

0.94 0.60 8.42 6.53 N/D N/D 1.82

25 26 S S

O

HO O

N 0.23 0.13 18.97 5.14 N/D N/D 1.09

26 27 S S

O

HO N

5.03 1.66 7.58 2.76 N/D N/D 1.15

27 29 S S

O

HO O

0.24 0.12 55.28 2.31 N/D N/D 1.47

28 30 S S

O

HO O O COOH

0.26 0.16 8.28 9.43 69.0 3.1 -3.40

29 31 S S

O

HO O

N 0.21 0.13 10.67 9.75 39.1 5.9 0.73

30 32 S S

O

HO N

5.03 1.15 6.31 3.86 N/D N/D 0.79

31 33 S S

O

OH O O COOH

0.58 0.52 13.68 1.77 N/D N/D -2.23

32 34 S S

O

OH O

0.73 0.60 64.26 10.84 N/D N/D 2.60

33 35 S S

O

OH O

N 0.78 0.58 14.68 5.64 N/D N/D 1.90

34 36 S S

O

OH

O

Cl 1.97 2.15 22.02 3.19 N/D N/D 1.46

35 37 S S

O

OH N

35.52 3.20 42.81 3.43 N/D N/D 1.96

36 38 S S

O

OH 146.5 12.10 17.74 4.34 N/D N/D 3.25

37 39 S S

O

OH 14.77 1.15 19.05 3.40 N/D N/D 3.25

38 40 S S

O

OH

O O

N-adamantyl 2.41 2.70 60.59 7.15 N/D N/D 2.13

39 41 S S

O

OH O

O

N-adamantyl 2.77 2.70 54.54 9.58 N/D N/D 2.13

40 42 S S

O

HO 7.92 1.40 26.70 5.19 N/D N/D 2.38

41 43 S S

O

HO 19.60 1.10 40.59 7.25 N/D N/D 2.38

42 44 S S

O

HO O O

N-adamantyl 0.69 0.42 59.57 4.52 N/D N/D 1.26

43 45 S S

O

HO O

O

N-adamantyl 2.09 1.92 8.01 3.73 N/D N/D 1.26

44 46 S S

HN

N O O COOH

0.75 0.20 16.86 3.22 N/D N/D 0.125

45 47 S S

HN

N O

7.70 3.20 10.14 9.71 N/D N/D 5.00

46 48 S S

HN

N O

N 1.85 0.84 12.80 10.69 N/D N/D 4.25

47 49 S S

HN

N

O

Cl 1.09 0.80 5.91 3.19 N/D N/D 3.82

48 50 S S

HN

N N

116.8 83.12 6.25 1.89 N/D N/D 4.32

49

51 S S HN

N 9.38 13.17 20.84 7.69 N/D N/D 5.61

50 52 S S

HN

N 12.42 59.16 10.28 3.96 N/D N/D 5.61

51 53 S S

HN

N O

O

N-adamantyl 4.49 1.92 16.30 5.62 N/D N/D 4.48

52 54 S S O

HO

O

Fe 0.37 0.47 N/D N/D N/D N/D N/D

The R1, R2, R3 and R4 substituents are referring to the core and flanking regions on Figure 1b. 1The IC50 and DC50 values represent respectively the assembly-inhibition and disassembly-inducing half-maximal concentrations measured in vitro. For each data point, experiments were done in triplicate and averaged. The s.e.m. of the IC50 or DC50 values determined from the curves was approx. 10-20%. 2The LDH values describe the cytotoxicity of the compounds in N2a cells compared to the negative DMSO control (set to 0%). Determined after incubation of the cells for 24 h with 10 µM compound 3The values obtained by incubating the cells with 15 µM compound correspond to the level of inhibition of tau aggregation normalized to control without inhibitor (0%) in cells. N/D: not determined. 4The calculated logarithms of water-octanol partition coefficients (ClogP) were obtained using ChemDraw Ultra 10.0 software (CambridgeSoft, Cambridge, MA).[32]

References: [1] M.R. Sawaya, S. Sambashivan, R. Nelson, M.I. Ivanova, S.A. Sievers, M.I. Apostol, M.J. Thompson, M.

Balbirnie, J.J.W. Wiltzius, H.T. McFarlane, A.O. Madsen, C. Riekel, D. Eisenberg, Nature 2007, 447, 453-457.

[11] S.L. Karsten, T.K. Sang, L.T. Gehman, S. Chatterjee, J. Liu, G.M. Lawless, S. Sengupta, R.W. Berry, J. Pomakian, H.S. Oh, C. Schulz, K.S. Hui, M. Wiedau-Pazos, H.V. Vinters, L.I. Binder, D.H. Geschwind, G.R. Jackson, Neuron 2006, 51, 549-560.

[12] S. Le Corre, H.W. Klafki, N. Plesnila, G. Hüblinger, A. Obermeier, H. Sahagun, B. Monse, P. Seneci, J. Lewis, J. Eriksen, C. Zehr, M. Yue, E. McGowan, D.W. Dickson, M. Hutton, H.M. Roder, Proc. Nat. Acad. Sci. USA 2006, 103, 9673-9678.

[13] W. Noble, E. Planel, C. Zehr, V. Olm, J. Meyerson, F. Suleman, K. Gaynor, L. Wang, J. LaFrancois, B. Feinstein, M. Burns, P. Krishnamurthy, Y. Wen, R. Bhat, J. Lewis, D. Dickson, K. Duff, Proc. Nat. Acad. Sci. USA 2005, 102, 6990-6995.

[22] N. Hotta, Y. Akanuma, R. Kawamori, K. Matsuoka, Y. Oka, M. Shichiri, T. Toyota, M. Nakashima, I. Yoshimura, N. Sakamoto, Y. Shigeta, Diabetes Care 2006, 29, 1538-1544.

Synthesis of rhodanine derivatives:

a. General procedure for the preparation of substituted aldehydes by palladium-

catalysed Suzuki-Miyaura coupling starting from 5-formyl-2-furanboronic acid.

To a solution of iodo- or bromoarene (1 mmol, 1 equiv.), 5-formyl-2-furanboronic acid (1.3

mmol, 182 mg), (Ph3P)PdCl2 (5 mol%, 50 µmol, 35 mg) in degassed DME (10 ml) was added

under argon 2M aqueous Na2CO3 (3.5 equiv., 2.3 ml). After stirring 9 hours at 85 °C, the

reaction mixture was cooled to room temperature and extracted with EtOAc (3 x 15 ml). The

combined organic phases were washed with brine, dried over MgSO4, filtered and

concentrated in vacuo. The crude product was purified by column chromatography on silica

gel using hexane/EtOAc (10:1 to 7:3, vol/vol) as eluent to afford the corresponding

substituted aldehyde (yield: 75-95 %).

b. General procedure for the Knoevenagel condensation between substituted aldehydes

and rhodanines.

To a solution of substituted aldehyde (0.1 mmol, 1 equiv.) and substituted rhodanine (0.1

mmol, 1 equiv.) in dry CH2Cl2 (10 ml) were added two equivalents of piperidine (0.2 mmol,

20 µl). The reaction mixture was stirred 12 hours at room temperature, forming a precipitate

when using acid-substituted rhodanines. The precipitate was recovered by filtration and

recrystallised from acetone/water (yield: 80-90%, purity>90% by HPLC). For ester-

substituted rhodanines, the reaction mixture was neutralised with acetic acid and extracted

with CH2Cl2 (3 x 10 ml), the combined organic phases washed with brine, dried over MgSO4,

filtered and concentrated in vacuo. The crude product was filtered over a silica gel pad (3 cm

x 6 cm) eluting first with hexane/EtOAc (7:3, vol/vol, 60 ml) followed by washing with

dichloromethane (100 ml). Concentration in vacuo gave analytically pure product (yields: 80-

90%). For ester hydrolysis, the purified product was dissolved in a dioxane/water mixture (1/1

vol/vol) and treated with aq.NaOH (2M, 3 equiv.) at room temperature for one hour. Dioxane

was removed under reduced pressure, the reaction mixture acidified with 1N HCl solution and

extracted with EtOAc (3 x 15 ml). The combined organic phases were concentrated in vacuo,

giving analytically pure compounds (total yields: 70-80 %)

HPLC was carried out with a HPLC Hewlett-Packard Series 1100 machine (Macherey-Nagel

Nucleodur C4 gravity 5 µM, 125 x 4 mm, with a linear gradient of 90% water/10%

acetonitrile to 100% acetonitrile over 15 min, 1 ml/min flow rate).

c. Compound Synthesis.

Synthesis of 2-((Z)-5-((5-(3-chlorophenyl)furan-2-yl)methylene)-4-oxo-2-thioxothiazolidin-3-

yl)ethanoic acid 1. Following the procedure b the product was obtained by reaction between

2-(4-oxo-2-thioxothiazolidin-3-yl)ethanoic acid ethyl ester (0.1 mmol, 21.9 mg) and 5-(3-

chlorophenyl)furan-2-carbaldehyde (0.1 mmol, 20.6 mg) in 76% yield (28 mg) as an orange

powder. mp: 175-176 °C; 1H NMR (400 MHz, DMSO-d6): δ = 4.69 (s, 2H), 7.39 (d, J = 4

Hz, 1H), 7.46 (d, J = 3.6 Hz, 1H), 7.47 (m, 1H), 7.60 (t, J = 8 Hz, 1H), 7.74 (s, 1H), 7.80 (m,

1H), 7.91 (t, J = 2 Hz, 1H); MALDI-MS (m/z): 380.5 [M+H]+; HRMS: calc. for

C16H11ClNO4S2 (M+H): 379.9812, found: 379.9808

Synthesis of (Z)-ethyl 2-(-5-((5-(3-chlorophenyl)furan-2-yl)methylene)-4-oxo-2-

thioxothiazolidin-3-yl)acetate (4). Following the procedure b the product was obtained by

reaction between ethyl-2-(4-oxo-2-thioxothiazolidin-3-yl)acetate (0.1 mmol, 21.9 mg) and 5-

(3-chlorophenyl)furan-2-carbaldehyde (0.1 mmol, 20.6 mg) in 76% yield (28 mg) as an

orange powder. mp: 140-141 °C; 1H NMR (400 MHz, CDCl3): δ = 1.29 (t, J = 6 Hz, 3H),

4.26 (q, J = 6 Hz, 2H), 4.88 (s, 2H), 6.88 (d, J = 3.6 Hz, 1H), 6.98 (d, J = 3.6 Hz, 1H), 7.37

(m, 1H), 7.43 (t, J = 8 Hz, 1H), 7.53 (s, 1H), 7.71 (m, 2H); MALDI-MS (m/z): 408.5 [M+H]+

Synthesis of (Z)-3-((1H-benzimidazol-2-yl)methyl)-5-((5-(3-chlorophenyl)furan-2-

yl)methylene)-2-thioxothiazolidin-4-one (14). Following the procedure b the product was

obtained by reaction between 3-((1H-benzimidazol-2-yl)methyl)-2-thioxothiazolidin-4-one

(0.1 mmol, 26.3 mg) and 5-(3-chlorophenyl)furan-2-carbaldehyde (0.1 mmol, 20.6 mg) in

73% yield (33 mg) as an orange powder. mp: 156-157 °C; 1H NMR (400 MHz, DMSO-d6): δ

= 5.49 (s, 2H), 7.20-7.22 (m, 2H), 7.41 (d, J = 3.7 Hz, 1H), 7.47 (d, J = 3.7 Hz, 1H), 7.50 (m,

1H), 7.52-7.55 (m, 2H), 7.61 (t, J = 8 Hz, 1H), 7.79 (s, 1H), 7.70-7.83 (m, 1H), 7.92 (t, J = 1.7

Hz, 1 H); MALDI-MS (m/z):452.5 [M+H]+; HRMS: calc. for C22H15ClN3O2S2 (M+H):

452.0288, found: 452.0276

Synthesis of (Z)-3-((1H-benzimidazol-2-yl)methyl)-5-((5-(diphenyl)furan-2-yl)methylene)-2-

thioxothiazolidin-4-one (15). Following the procedure b the product was obtained by reaction

between 3-((1H-benzimidazol-2-yl)methyl)-2-thioxothiazolidin-4-one (0.1 mmol, 26.3 mg)

and 5-(diphenyl)furan-2-carbaldehyde (0.1 mmol, 24.8 mg) in 70% yield (34 mg) as an

orange powder. mp: 248-249 °C; 1H NMR (400 MHz, DMSO-d6): δ = 5.54 (s, 2H), 7.23-7.26

(m, 2H), 7.45-7.49 (m, 2H), 7.50 (d, J = 3.7 Hz, 1H), 7.53 (d, J = 3.7 Hz, 1H), 7.55-7.60 (m,

3H), 7.83 (d, J = 1.18 Hz, 1H), 7.85 (d, J = 1.18 Hz, 1H), 7.88 (s, 1H), 7.99 (d, J = 8.8 Hz,

2H), 8.06 (d, J = 8.8 Hz, 2H); MALDI-MS (m/z): 494.5 [M+H]+; HRMS: calc. for

C23H20N3O2S2 (M+H): 494.0991, found: 494.0979

Synthesis of 3-((Z)-5-((5-(3-chlorophenyl)furan-2-yl)methylene)-4-oxo-2-thioxothiazolidin-3-

yl)propanoic acid (16). Following the procedure b the product was obtained by reaction

between 3-(4-oxo-2-thioxothiazolidin-3-yl)propionic acid (0.1 mmol, 23.3 mg) and 5-(3-

chlorophenyl)furan-2-carbaldehyde (0.1 mmol, 20.6 mg) in 75% yield (29 mg) as an orange

powder. mp: 176-177 °C; 1H-NMR (400 MHz, DMSO-d6): δ = 2.62 (t, J = 6.5 Hz, 2H), 4.22

(t, J = 6.5 Hz, 2H), 7.36 (d, J = 3.6 Hz, 1H), 7.44 (d, J = 3.6 Hz, 1H), 7.48 (m, 1H), 7.59 (t, J=

8 Hz, 1H), 7.68 (s, 1H), 7.78 (m, 1H), 7.89 (t, J = 1.6 Hz, 1H); MALDI-MS (m/z): 394.5

[M+H]+; HRMS: calc. for C17H13ClNO4S2 (M+H): 393.9975, found: 393.9962

Synthesis of (Z)-3-((1H-benzimidazol-2-yl)methyl)-5-((5-(4-1H-pyrrol-1-yl)phenyl)furan-2-

yl)methylene)-2-thioxothiazolidin-4-one (17). Following the procedure b the product was

obtained by reaction between 3-((1H-benzimidazol-2-yl)methyl)-2-thioxothiazolidin-4-one

(0.1 mmol, 26.3 mg) and 5-(4-1H-pyrrol-1-yl)phenyl)furan-2-carbaldehyde (0.1 mmol, 23.7

mg) in 75% yield (36 mg) as a yellow powder. mp: 210-211 °C; 1H NMR (400 MHz, DMSO-

d6): δ = 5.51 (s, 2H), 6.30 (t, J = 2.1 Hz, 2H), 7.20-7.26 (m, 2H), 7.38 (d, J = 3.7 Hz, 1H),

7.44 (d, J = 3.7 Hz, 1H), 7.49 (t, J = 2.1 Hz, 2H), 7.54-7.57 (m, 2H), 7.79 (s, 1H), 7.82 (d, J =

9 Hz, 2H), 7.94 (d, J = 9 Hz, 2H); MALDI-MS (m/z): 483.7 [M+H]+; HRMS: calc. for

C26H19N4O2S2 (M+H): 483.0944, found: 483.0939

Synthesis of 3-((Z)-5-((5-(diphenyl)furan-2-yl)methylene)-4-oxo-2-thioxothiazolidin-3-

yl)propanoic acid (19). Following the procedure b the product was obtained by reaction

between 3-(4-oxo-2-thioxothiazolidin-3-yl)propionic acid ethyl ester (0.1mmol, 23.3 mg) and

5-(diphenyl)furan-2-carbaldehyde (0.1 mmol, 24.8 mg) in 80% yield (35 mg) as a red powder.

m.p. 191-192 °C; 1H NMR (400 MHz, DMF-d6): δ = 2.95 (t, J = 8 Hz, 2H), 4.54 (t, J = 8 Hz,

2H), 7.59 (d, J = 3.6 Hz, 1H), 7.62 (d, J = 3.6 Hz, 1H), 7.67-7.71 (m, 3H), 7.87 (s, 1H), 7.97

(d, J = 1.2 Hz, 1H), 7.99 (d, J = 1.2 Hz, 1H), 8.10 (d, J = 8.8 Hz, 2H), 8.21 (d, J = 8.8 Hz,

2H); MALDI-MS (m/z): 436.8 [M+H]+; HRMS: calc. for C23H18NO4S2 (M+H): 436.0678,

found: 436.0672

Synthesis of 5-(4-(5-((Z)-3-(2-carboxy-ethyl)-4-oxo-2-thioxo-thiazolidin-5-ylidene methyl)-

furan-2-yl)-phenyl)-furan-2-carboxylic acid (20). Following the procedure b the product was

obtained by reaction between 3-(4-oxo-2-thioxothiazolidin-3-yl)propionic acid ethyl ester (0.1

mmol, 23.3 mg) and 5-(4-(5-formylfuran-2-yl)phenyl)furan-2-carboxylic acid (0.1 mmol, 28.2

mg) in 70% yield (32 mg) as a yellow powder. mp: 217-218 °C; 1H NMR (400 MHz, DMF-

d6): δ = 2.95 (t, J = 8 Hz, 2H), 4.54 (t, J = 8Hz, 2H), 7.46 (d, J = 3.8 Hz, 1H), 7.55 (d, J = 3.8

Hz, 1H), 7.59 (d, J = 3.4 Hz, 1H), 7.62 (d, J = 3.4 Hz, 1H), 7.86 (s, 1H), 8.19 (d, J = 12.8 Hz,

4H); MALDI-MS (m/z): 470.7 [M+H]+; HRMS: calc. for C22H16NO7S2 (M+H): 470.0369,

found: 470.0359

Synthesis of 4-((Z)-5-((5-(diphenyl)furan-2-yl)methylene)-4-oxo-2-thioxothiazolidin-3-

yl)butanoic acid (24). Following the procedure b the product was obtained by reaction

between 4-(4-oxo-2-thioxothiazolidin-3-yl)butanoic acid ethyl ester (0.1 mmol, 24.7 mg) and

5-(diphenyl)furan-2-carbaldehyde (0.1 mmol, 24.8 mg) in 71% yield (32 mg) as a yellow

powder. mp: 193-194 °C; 1H NMR (400 MHz, DMSO-d6): δ = 1.88 (q, J = 7 Hz, 2H), 2.28 (t,

J = 7.1 z, 2H), 4.07 (t, J = 7 Hz, 2H), 7.35(d, J = 3.9 Hz, 1H), 7.38 (d, J = 3.9 Hz, 1H), 7.40

(m, 1H), 7.46 (t, J = 14.8 Hz, 2H), 7.66 (s, 1H), 7. 74 (d, J = 0.8 Hz, 1H), 7.76 (d, J = 0.8 Hz,

1H), 7.87 (d, J = 8.6 Hz, 2H), 7.92 (d, J = 8.6 Hz, 2H); MALDI-MS (m/z): 450.8 [M+H]+;

HRMS: calc. for C24H20NO4S2 (M+H): 450.0834, found: 450.0821

Synthesis of 5-(4-(5-((Z)-3-(4-carboxybutyl)-4-oxo-2-thioxo-thiazolidin-5-ylidene)methyl)-

furan-2-yl)-phenyl)-furan-2-carboxylic acid (30). Following the procedure b the product was

obtained by reaction between 5-(4-oxo-2-thioxothiazolidin-3-yl)pentanoic acid ethyl ester (0.1

mmol, 25.9 mg) and 5-(4-(5-formylfuran-2-yl)phenyl)furan-2-carboxylic acid (0.1 mmol, 28.2

mg) in 75% yield (37.3 mg) as a yellow powder. mp: 219-220 °C; 1H NMR (400 MHz,

DMSO-d6): δ = 1.48 (m, 2H), 1.53 (m, 2H), 2.22 (t, J = 8 Hz, 2H), 4.05 (t, J = 7Hz, 2H), 7.32

(d, J = 3.3 Hz, 1H), 7.39 (d, J = 3.3 Hz, 1H), 7.48 (d, J = 2.5 Hz, 1H), 7.50 (d, J = 2.5 Hz,

1H), 7.65 (s, 1H), 7.88 (d, J = 13.0 Hz, 4H); MALDI-MS (m/z): 498.2 [M+H]+; HRMS: calc.

for C24H20NO7S2 (M+H): 498.0682, found: 498.0664

Synthesis of 5-((Z)-5-((5-(4-1H-pyrrol-1-yl)phenyl)furan-2-yl)methylene)-4-oxo-2-

thioxothiazolidin-3-yl)pentanoic acid (31). Following the procedure b the product was

obtained by reaction between 5-(4-oxo-2-thioxothiazolidin-3-yl)pentanoic acid ethyl ester (0.1

mmol, 26.1 mg) and 5-(4-1H-pyrrol-1-yl)phenyl)furan-2-carbaldehyde (0.1 mmol, 23 mg) in

71% yield (31 mg) as an orange powder. mp: 181-182 °C; 1H NMR (400 MHz, DMSO-d6): δ

= 1.50-1.53 (m, 2H), 1.63-1.67 (m, 2H), 2.24 (t, J = 7.2 Hz, 2H), 4.02 (t, J = 7.1 Hz, 2H), 6.30

(t, J = 2.1 Hz, 2H), 7.35(d, J = 3.8 Hz, 1H), 7.38 (d, J = 3.8 Hz, 1H), 7.48 (t, J = 2.2 Hz, 2H),

7.67 (s, 1H), 7.81 (d, J = 9 Hz, 2H), 7.91 (d, J = 9 Hz, 2H); MALDI-MS (m/z): 453.5

[M+H]+; HRMS: calc. for C23H21N2O4S2 (M+H): 453.0943, found: 453.0934

HPLC-traces recorded for selected compounds: HPLC was carried out with a HPLC Hewlett-Packard Series 1100 machine (Macherey-Nagel

Nucleodur C4 gravity 5 µM, 125 x 4 mm, with a linear gradient of 90% water/10%

acetonitrile to 100% acetonitrile over 15 min, 1 ml/min flow rate).