Fatty Acyl Benzamido Antibacterials Based on Inhibition of DnaK-catalyzed Protein Folding
2006; Elsevier BV; Volume: 282; Issue: 7 Linguagem: Inglês
10.1074/jbc.m607667200
ISSN1083-351X
AutoresMarkus Liebscher, Günther Jahreis, Christian Lücke, Susanne Grabley, Satish Raina, Cordelia Schiene‐Fischer,
Tópico(s)Protein Structure and Dynamics
ResumoWe have reported that the hsp70 chaperone DnaK from Escherichia coli might assist protein folding by catalyzing the cis/trans isomerization of secondary amide peptide bonds in unfolded or partially folded proteins. In this study a series of fatty acylated benzamido inhibitors of the cis/trans isomerase activity of DnaK was developed and tested for antibacterial effects in E. coli MC4100 cells. Nα-[Tetradecanoyl-(4-aminomethylbenzoyl)]-l-asparagine is the most effective antibacterial with a minimal inhibitory concentration of 100 ± 20 μg/ml. The compounds were shown to compete with fluorophore-labeled σ32-derived peptide for the peptide binding site of DnaK and to increase the fraction of aggregated proteins in heat-shocked bacteria. Despite its inability to serve as a folding helper in vivo a DnaK-inhibitor complex was still able to sequester an unfolded protein in vitro. Structure activity relationships revealed a distinct dependence of DnaK-assisted refolding of luciferase on the fatty acyl chain length, whereas the minimal inhibitory concentration was most sensitive to the structural nature of the benzamido core. We conclude that the isomerase activity of DnaK is a major survival factor in the heat shock response of bacteria and that small molecule inhibitors can lead to functional inactivation of DnaK and thus will display antibacterial activity. We have reported that the hsp70 chaperone DnaK from Escherichia coli might assist protein folding by catalyzing the cis/trans isomerization of secondary amide peptide bonds in unfolded or partially folded proteins. In this study a series of fatty acylated benzamido inhibitors of the cis/trans isomerase activity of DnaK was developed and tested for antibacterial effects in E. coli MC4100 cells. Nα-[Tetradecanoyl-(4-aminomethylbenzoyl)]-l-asparagine is the most effective antibacterial with a minimal inhibitory concentration of 100 ± 20 μg/ml. The compounds were shown to compete with fluorophore-labeled σ32-derived peptide for the peptide binding site of DnaK and to increase the fraction of aggregated proteins in heat-shocked bacteria. Despite its inability to serve as a folding helper in vivo a DnaK-inhibitor complex was still able to sequester an unfolded protein in vitro. Structure activity relationships revealed a distinct dependence of DnaK-assisted refolding of luciferase on the fatty acyl chain length, whereas the minimal inhibitory concentration was most sensitive to the structural nature of the benzamido core. We conclude that the isomerase activity of DnaK is a major survival factor in the heat shock response of bacteria and that small molecule inhibitors can lead to functional inactivation of DnaK and thus will display antibacterial activity. The rapid emergence of bacterial strains that are resistant to current antibiotics requires the development of antimicrobial compounds based on novel mechanisms of action. De novo protein synthesis and assisted protein folding are processes essential for viability of cells thus representing attractive targets for the design of antibacterials. The hsp70 chaperone DnaK is a folding helper enzyme showing ATPase and secondary amide peptide bond cis/trans isomerase (APIase) 2The abbreviations used are: APIase, secondary amide peptide bond cis/trans isomerase; Fmoc, N-(9-fluorenyl)methoxycarbonyl; HPLC, high-performance liquid chromatography; DTT, dithiothreitol; MOPS, 4-morpholinepropanesulfonic acid; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; MIC, minimal inhibitory concentration; IPTG, isopropyl 1-thio-β-d-galactopyranoside; HATU, N-[(dimethylamino)-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene]-N-methylmethanaminium hexafluorophosphate N-oxide; RCMLA, reduced carboxymethylated lactalbumin. activity that assists together with its co-chaperones the de novo protein folding of ∼20% of the Escherichia coli proteins larger than 30 kDa (1.Schiene-Fischer C. Habazettl J. Schmid F.X. Fischer G. Nat. Struct. Biol. 2002; 9: 419-424Crossref PubMed Scopus (81) Google Scholar, 2.Langer T. Lu C. Echols H. Flanagan J. Hayer M.K. Hartl F.U. Nature. 1992; 356: 683-689Crossref PubMed Scopus (793) Google Scholar, 3.Hartl F.U. Hayer-Hartl M. Science. 2002; 295: 1852-1858Crossref PubMed Scopus (2799) Google Scholar, 4.Teter S.A. Houry W.A. Ang D. Tradler T. Rockabrand D. Fischer G. Blum P. Georgopoulos C. Hartl F.U. Cell. 1999; 97: 755-765Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar). Whereas the deletion of the DnaK gene alone is not deleterious for E. coli cells, combined deletion of the ribosome-associated peptidylprolyl-cis/trans-isomerase trigger factor and DnaK genes is lethal under normal growth conditions (4.Teter S.A. Houry W.A. Ang D. Tradler T. Rockabrand D. Fischer G. Blum P. Georgopoulos C. Hartl F.U. Cell. 1999; 97: 755-765Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar, 5.Deuerling E. Schulze-Specking A. Tomoyasu T. Mogk A. Bukau B. Nature. 1999; 400: 693-696Crossref PubMed Scopus (408) Google Scholar). Furthermore, DnaK is essential under various stress conditions like heat, oxidative stress, and nutritional deprivation (6.Paek K.H. Walker G.C. J. Bacteriol. 1987; 169: 283-290Crossref PubMed Google Scholar, 7.Rockabrand D. Arthur T. Korinek G. Livers K. Blum P. J. Bacteriol. 1995; 177: 3695-3703Crossref PubMed Google Scholar, 8.Spence J. Cegielska A. Georgopoulos C. J. Bacteriol. 1990; 172: 7157-7166Crossref PubMed Google Scholar). In E. coli cells lacking DnaK, heat shock treatment at 42 °C causes strong protein aggregation comprising 150–200 different protein species and ∼10% of the amount of pre-existing E. coli proteins (9.Mogk A. Tomoyasu T. Goloubinoff P. Rudiger S. Roder D. Langen H. Bukau B. EMBO J. 1999; 18: 6934-6949Crossref PubMed Scopus (515) Google Scholar). During infection pathogens encounter stress conditions generated by the host defense mechanisms to eliminate the infection. They respond by the rapid acceleration in the rate of expression of heat shock and other stress proteins (10.Garbe T.R. Experientia. 1992; 48: 635-639Crossref PubMed Scopus (42) Google Scholar). The DnaK/DnaJ chaperone machinery of Salmonella enterica has been shown to be involved in bacterial invasion of epithelial cells, and the DnaK/DnaJ-depleted mutant could not survive or proliferate at all within macrophages (11.Takaya A. Tomoyasu T. Matsui H. Yamamoto T. Infect. Immun. 2004; 72: 1364-1373Crossref PubMed Scopus (72) Google Scholar). In the case of the pathogen Brucella suis it was shown that DnaK but not DnaJ is required for intracellular multiplication (12.Kohler S. Teyssier J. Cloeckaert A. Rouot B. Liautard J.P. Mol. Microbiol. 1996; 20: 701-712Crossref PubMed Scopus (79) Google Scholar). DnaK of Listeria monocytogenes is essentially required for survival under high temperatures and acidic conditions, and it is essential for the efficient phagocytosis of the facultative intracellular pathogen with macrophages (13.Hanawa T. Fukuda M. Kawakami H. Hirano H. Kamiya S. Yamamoto T. Cell Stress Chaperones. 1999; 4: 118-128PubMed Google Scholar). The DnaK homologues HscA and HscC seem to have more specialized functions in the cell; HscA appears to be involved in the assembly of Fe/S proteins (14.Hoff K.G. Silberg J.J. Vickery L.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 7790-7795Crossref PubMed Scopus (200) Google Scholar, 15.Silberg J.J. Hoff K.G. Tapley T.L. Vickery L.E. J. Biol. Chem. 2001; 276: 1696-1700Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar), and HscC seems to play a role in the response to special types of stress (16.Kluck C.J. Patzelt H. Genevaux P. Brehmer D. Rist W. Schneider-Mergener J. Bukau B. Mayer M.P. J. Biol. Chem. 2002; 277: 41060-41069Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). It has been already shown that proline-rich cationic oligopeptides such as pyrrhocoricin appear to kill responsive bacteria by binding to DnaK (17.Kragol G. Lovas S. Varadi G. Condie B.A. Hoffmann R. Otvos Jr., L. Biochemistry. 2001; 40: 3016-3026Crossref PubMed Scopus (387) Google Scholar). Dimeric analogues of pyrrhocoricin showed improved features with respect to broad applicability and stability in mammalian sera (18.Cudic M. Condie B.A. Weiner D.J. Lysenko E.S. Xiang Z.Q. Insug O. Bulet P. Otvos Jr., L. Peptides. 2002; 23: 2071-2083Crossref PubMed Scopus (87) Google Scholar). However, the relationship between the formation of the DnaK/pyrrhocoricin and the antibacterial action remained unexplained. From gene deletion studies, the potential of small molecule inhibitors of the APIase activity of DnaK to act as antimicrobial agents can be inferred. However, no small molecule inhibitors or even non-peptidic binders for the APIase site of DnaK have been reported so far. The existence of a distinct APIase site in DnaK serving the catalysis of the cis/trans isomerization of secondary amide peptide bonds opened up the new possibility of assaying compounds for inhibiting this essential DnaK function. Here we report that the fatty acylated benzamido derivatives such as Nα-[fatty-acyl-(4-aminoalkylbenzoyl)]-l-α-amino acids reversibly block the APIase site of DnaK, inhibit DnaK-mediated refolding of firefly luciferase, increase protein aggregation in the cell, and exhibit antibacterial activity against E. coli MC4100 strain. To examine structure-activity relationships, we modified the α-amino acid residue, the benzoic acid linker, as well as the fatty acid part of the compounds (see Fig. 1). Materials−Monoclonal anti-DnaK antibody was purchased from Stressgen (Victoria, Canada). Kool™ NC-45™ Universal RNA polymerase templates and E. coli RNA polymerase core enzyme were purchased from Epicenter (Madison, WI). E. coli MC4100 cells (araD139 Δ(argF-lac)205 flb-5301 pstF25 rpsL150 deoC1 relA1) were obtained from DSMZ (Braunschweig, Germany). The E. coli MC4100 ΔdnaK52::cat sidB1 (ΔdnaK) cells were provided by B. Bukau (BB1553), pET11DnaK was provided by F. U. Hartl, and pPL900-GrpE and pPL871-DnaJ were provided by N. Dixon. DnaK was purified as described by Zylicz et al. (19.Zylicz M. Georgopoulos C. J. Biol. Chem. 1984; 259: 8820-8825Abstract Full Text PDF PubMed Google Scholar). As an additional purification step DnaK has been treated with alkaline phosphatase and subsequently applied to a gel-filtration column as described by Theyssen et al. (20.Theyssen H. Schuster H.P. Packschies L. Bukau B. Reinstein J. J. Mol. Biol. 1996; 263: 657-670Crossref PubMed Scopus (199) Google Scholar). DnaJ and GrpE were purified according to already published procedures by Zylicz et al. (21.Zylicz M. Ang D. Liberek K. Georgopoulos C. EMBO J. 1989; 8: 1601-1608Crossref PubMed Scopus (203) Google Scholar). Synthesis−The fatty acylated benzamido derivatives were synthesized by solid-phase peptide synthesis with the synthesizer Syro II (MultiSynTech GmbH, Witten, Germany) using 0.15 mmol of amino acid pre-loaded Wang resins (compounds 1–18) and 0.15 mmol of Rink amide resin for the peptide amide 19. All resins were obtained from NovaBiochem (Läufelfingen, Switzerland). Synthesis was done by Fmoc strategy with Fmoc amino acids as building blocks and PyBOP (NovaBiochem) and N-methylmorpholine as coupling reagents in dimethylformamide. Piperidine/N,N-dimethylformamide (20%) was the standard cleavage mixture used for Fmoc detachment. The resin was treated twice for 10 min. All couplings were performed using a 4-fold excess of Fmoc amino acid derivative, PyBOP, and N-methylmorpholine (eight equivalents) in N,N-dimethylformamide. For all amino acid derivatives a 1.5-h double coupling protocol was used. The activation reagent HATU (AB Applied Biosystems, Foster City, CA) was used instead of PyBOP for coupling of the fatty acids in fatty acylated benzamido derivatives 11 and 16. After detachment of the fatty acylated benzamido derivatives from the resin and site-chain deprotection with trifluoroacetic acid/water (95:5) at room temperature for 12 h, the fatty acylated benzamido derivatives were purified by reversed-phase-HPLC on a Gilson 306 equipment with a column SP 250/10 Nucleosil 100-7 C8 (Macherey-Nagel, Düren, Germany) using a water (0.1% trifluoroacetic acid)/ acetonitrile gradient. The purified fatty acylated benzamido derivatives were than lyophilized. Conditions for analytical HPLC were: column LiChro-CART® (LiChrospher® 100, RP8, 5 μm) 125 × 4 mm (Merck, Darmstadt, Germany) using a water (0.1% trifluoroacetic acid)/acetonitrile gradient 5–100% in 30 min; flow rate 1 ml/min; detection at 220 nm. The identity of the fatty acylated benzamido derivatives were verified by electrospray ionization mass spectrometry on a VG-BIO-Q Triple Quadrupole Tandem Electrospray mass spectrometer (Fisons Instruments, San Carlos, CA). All 1H NMR spectra were obtained at 500.13 MHz on a Bruker DRX500 spectrometer (Bruker Biospin Fällanden, Switzerland). The measurements were carried out in CD3OD. Two-dimensional total correlation spectroscopy and two-dimensional nuclear Overhauser effect spectroscopy spectra were recorded at 25 °C. Identification of Compound 1 in E. coli Cytosol−E. coli MC4100 cells were incubated with 100 μg/ml 1 for 12 h at 42 °C in Mueller-Hinton broth. The cells were washed extensively with 10 mm Tris/HCl, pH 8.0, 50 mm NaCl, 0.1 mm EDTA, 1 mm dithiothreitol, 5% (v/v) ethanol until no compound 1 could be detected in the buffer after washing. Cells were lysed by sonication. After centrifugation (4300 rpm, 15 min, 4 °C) cell lysate was subjected to analytical HPLC (column LiChroCART® (LiChrospher® 100, RP8, 5 μm) 125 × 4 mm (Merck, Darmstadt, Germany) using a water (0.1% trifluoroacetic acid)/acetonitrile gradient 5–100% in 30 min; flow rate 1 ml/min; detection at 220 nm). The identity of compound 1 was verified by electrospray ionization mass spectrometry. Pure compound 1 was used as the reference. As a control, E. coli MC4100 cells without treatment with compound 1 were subjected to the described procedure. Luciferase Refolding Assay−Refolding experiments were done essentially as described by Szabo et al. (22.Szabo A. Langer T. Schroder H. Flanagan J. Bukau B. Hartl F.U. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10345-10349Crossref PubMed Scopus (444) Google Scholar). Briefly, luciferase from Photinus pyralis (Promega) was incubated for 1 h in denaturation buffer (30 mm Tris/HCl, pH 7.4, 5 mm dithiothreitol, 6.0 m guanidinium chloride) at 10 °C at a final concentration of 2.08 μm. Refolding was initiated by dilution of denatured luciferase into renaturation buffer (10 mm MOPS, pH 7.8, 50 mm KCl, 1 mm ATP, 5 mm MgCl2,1 μm bovine serum albumin, 436 nm DnaK, 160 nm DnaJ, and 436 nm GrpE) to a final concentration of 10.4 nm at 10 °C and 30 °C. Before the refolding reaction was started, compounds 1–18 at different concentrations were incubated in renaturing buffer for 10 min. After 1 h of refolding, luciferase activity was determined with a luminometer (Luminoskan, Ascent) using luciferase assay buffer (20 mm Tricine, pH 7.8, 5 mm MgCl2, 0.1 mm EDTA, 3.3 mm dithiothreitol, 270 μm coenzyme A, 500 μm d-luciferin, 500 μm ATP). The IC50 values have been calculated for a one site competition model with SPSS Sigmaplot 8.0. Enzymatic Measurements−The APIase activity of DnaK was determined using RNaseT1 P39A variant as substrate as described in Schiene-Fischer et al. at 15 and 30 °C (1.Schiene-Fischer C. Habazettl J. Schmid F.X. Fischer G. Nat. Struct. Biol. 2002; 9: 419-424Crossref PubMed Scopus (81) Google Scholar). Steadystate ATPase measurements were performed as described in Montgomery et al. (23.Montgomery D.L. Morimoto R.I. Gierasch L.M. J. Mol. Biol. 1999; 286: 915-932Crossref PubMed Scopus (129) Google Scholar). Compound 1 at 150 μm has been incubated with 1 μm DnaK for 10 min at 30 °C. The reaction was started with 330 μm ATP. The time course was followed for at least 10 min. Determination of MIC Values−All compounds were dissolved in ethanol/10 mm potassium phosphate, pH 7.8, to 2 mg/ml (final ethanol concentration, 13% (v/v)). After incubation for 12 h at 37 °C insoluble particles were separated by centrifugation at 16,000 × g for 15 min. The final concentration was determined UV-spectrometrically by using the characteristic absorption at 236 nm. MICs were determined in triplicate in Mueller-Hinton medium by a liquid growth inhibition assay. In a sterile microtitration plate, 100 μl of the compound in varying concentrations, or 10 mm potassium phosphate, pH 7.8, 13% (v/v) ethanol as a control, were added to 50 μl of a mid-logarithmic growth phase culture of E. coli MC4100 cells. Plates were incubated for 24 h at 42 °C. Samples were spread on agar plates and incubated overnight at 37 °C. The number of colony forming units was counted, and the MIC value was determined as the concentration of substance that causes a 99.9% inhibition of cell growth (NCCLS criteria (24.Amsterdam D. Lorian V. Antibiotics in Laboratory Medicine. 4th Ed. Lippincott, Williams, & Wilkins, Baltimore1996: 52-111Google Scholar)). Erythrocyte Hemolysis−The hemolytic activity of fatty acylated benzamido derivatives was determined by spectrometric analysis as described before (25.Dathe M. Schumann M. Wieprecht T. Winkler A. Beyermann M. Krause E. Matsuzaki K. Murase O. Bienert M. Biochemistry. 1996; 35: 12612-12622Crossref PubMed Scopus (350) Google Scholar) using human erythrocytes. Freshly isolated human erythrocytes were washed three times with Tris-buffered saline (10 mm Tris/HCl, pH 7.4, 150 mm NaCl) prior to the assay. The erythrocytes (2.5 × 108 cells/ml) were treated with fatty acylated benzamido derivatives at final concentrations ranging from 3.1 to 50 μg/ml and incubated under gently shaking at 42 °C for 30 min. The samples were subjected to ice for 2 min and centrifuged at 380 × g for 5 min. The amount of hemoglobin released from the cells was determined by measuring the absorbance of the supernatants at 550 nm after 10-fold dilution with 0.5% NH4OH. Saponin (100 mg/liter) was used to obtain a positive control hemolysis value (100% hemolysis), and 10 mm potassium phosphate buffer, pH 7.8, was used as a negative control. Inhibitor concentrations causing 50% hemolysis (EC50) were derived from the dose-response curves with SPSS SigmaPlot 8.0. Heat Shock Experiments−E. coli MC4100 ΔdnaK pTrc99A-DnaK cells were grown to mid-exponential growth at 30 °C in Terrific Broth. Cells were incubated for 1 h in the presence of 0, 0.2, 0.5, or 1.0 mm IPTG. Cells were subjected to 57 °C for 210 s in the absence or additional presence of 1.35 mm compound 1 added immediately before heat shock. Immediately afterward heat shock cells were cooled on ice for 2 min. Serial dilutions were plated onto LB plates with selection marker supplemented with the indicated IPTG concentrations. The plates were incubated for 12 h at 30 °C. For analysis of DnaK produced by E. coli MC4100 ΔdnaK pTrc99A-DnaK, cells were lysed by sonification after 1-h induction in the presence of varying IPTG concentrations, and samples were analyzed by 12.5% SDS-PAGE and immunoblotting using anti-DnaK antibody. Aggregation Assay−E. coli MC4100 wild type were grown to mid-exponential growth at 30 °C. 0.13 mg/ml or 0.066 mg/ml compound 1 or 2, respectively, was added, and the cells were cultured further for 15 min. Cells were subjected to heat shock at 45 °C for 30 min and subsequently grown at 30 °C for 3 h. E. coli MC4100 ΔdnaK cells in the absence of inhibitors was treated equally and thus used as a control for protein aggregation in the absence of DnaK. Aggregated proteins were isolated and separated from contaminating membrane proteins as described. Briefly, cells were harvested by 10-min centrifugation at 5000 × g and 4 °C. Pellets were resuspended in buffer A (100 mm Tris/HCl, pH 7.5, 100 mm KCl, 2 mm EDTA, 15% (w/v) sucrose, 1 mg/ml lysozyme) and incubated for 30 min on ice. Cell lysis was performed by adding a 7-fold excess of buffer B (100 mm Tris/HCl, pH 7.5, 1 mm EDTA), followed by sonification. After centrifugation at 2,000 × g for 15 min at 4 °C, an insoluble cell fraction was isolated by subsequent centrifugation at 15,000 × g for 20 min at 4 °C. The pellets were frozen, resuspended in 400 μl of buffer B by sonification, and centrifuged (15,000 × g, 20 min, 4 °C). The washed pellets were again resuspended in 320 μl of buffer B by sonification, subsequently 80 μl of 10% (v/v) Nonidet P-40 was added, and aggregated proteins were isolated by centrifugation (15,000 × g, 20 min, 4 °C). This washing procedure was repeated. Nonidet P-40-insoluble pellets were washed with buffer B and resuspended in 50 μl of buffer B by sonification (26.Tomoyasu T. Mogk A. Langen H. Goloubinoff P. Bukau B. Mol. Microbiol. 2001; 40: 397-413Crossref PubMed Scopus (276) Google Scholar). Fractions corresponding to identical cell numbers were analyzed by 12.5% SDS-PAGE followed by staining with Coomassie Brilliant Blue. Pull-down Assay of DnaK with Compound 19−MC4100 cells were grown overnight in LB medium at 42 °C. 200 μl of streptavidin beads was washed with buffer C (20 mm Tris/HCl, pH 7.5, 50 mm KCl, 1 mm MgCl2) and blocked with 1 mg/ml bovine serum albumin for 1 h at 4 °C. MC4100 cells were lysed by sonification and incubated with 0.7 mm compound 19 for 1 h at room temperature in 1200 μl of buffer C. Streptavidin beads were added, and the mixture was further incubated for an additional hour at room temperature. Streptavidin beads were pelleted and washed twice with buffer C. DnaK was eluted from beads by adding compound 1 with a final concentration of 1.92 mm.5 μl of the different samples was analyzed by 12.5% SDS-PAGE and immunoblotting using anti-DnaK antibody. Controls showed 0.1 μlof E. coli lysate and 0.2 μg if recombinant DnaK. Characterization of the Interaction of DnaK with Nα-[Tetradecanoyl-(4-aminomethylbenzoyl)]-l-isoleucine−As exemplified by Nα-[tetradecanoyl-(4-aminomethylbenzoyl)]-l-isoleucine (1) compounds were synthesized by stepwise solid-phase synthesis using the Fmoc strategy on Wang resin preloaded with isoleucine. After deprotection myristic acid was coupled using PyBOP as reagent. Type 1 anionic compounds consisted of three building blocks, a fatty acid moiety, an aromatic linker, and an α-amino acid residue covering a free carboxylate group (Fig. 1A). Compound 1 inhibited the APIase activity of DnaK with an IC50 value of 2.7 μm as determined by DnaK-catalyzed RNase T1 P39A refolding assay (Fig. 2A) (1.Schiene-Fischer C. Habazettl J. Schmid F.X. Fischer G. Nat. Struct. Biol. 2002; 9: 419-424Crossref PubMed Scopus (81) Google Scholar). Similarly, it decreased the yield of native protein from the guanidinium chloride-denatured firefly luciferase in a DnaK/DnaJ/GrpE-assisted refolding assay in a dose-dependent manner with an IC50 value of 9.5 μm (Fig. 2B). According to the refolding pathway, inhibition of the APIase function of DnaK translates into reduction of final folding yield (27.Schiene-Fischer C. Habazettl J. Tradler T. Fischer G. Biol. Chem. 2002; 383: 1865-1873Crossref PubMed Scopus (17) Google Scholar). The ATPase activity of DnaK remained unaffected up to a concentration of 150 μm of 1. To ascertain the specificity of the refolding assay for APIase function the effect of 2 was examined. Its inhibitory potency was low in both the RNase T1 P39A variant and the luciferase refolding assay. In both assays, compound 2 up to concentrations of 20 μm and 200 μm, respectively, had no effect. Reversibility of inhibition was examined by dialysis of the DnaK-1 complex formed by incubation of DnaK (0.56 μm) and 1 (10.5 μm). The refolding activity of DnaK completely recovered from the inhibited enzyme when dialyzed against 10 mm MOPS buffer, pH 7.8, 50 mm KCl, 5 mm MgCl2, regardless of whether the incubation time was prolonged or the concentration of 1 was further increased. Next we analyzed if 1 binds to DnaK at the peptide binding site of DnaK, which constitutes a channel defined by loops from the beta sandwich structure of the C-terminal substrate-binding domain (28.Zhu X. Zhao X. Burkholder W.F. Gragerov A. Ogata C.M. Gottesman M.E. Hendrickson W.A. Science. 1996; 272: 1606-1614Crossref PubMed Scopus (1064) Google Scholar). Therefore, our competition assay correlated with the decrease in fluorescence at 460 nm in the presence of increasing amounts of 1 when administered to a solution of a DnaK-bound σ32-derived peptide (σ32-132QRKLFFNLRKTKQ142-C-IAANS) that was labeled with the fluorescent probe IAANS. This 14-mer oligopeptide represented a segment of the transcription factor σ32, which was shown to specifically bind to the peptide binding site of DnaK with a substantial increase in fluorescence emission upon DnaK binding (29.McCarty J.S. Rudiger S. Schonfeld H.J. Schneider-Mergener J. Nakahigashi K. Yura T. Bukau B. J. Mol. Biol. 1996; 256: 829-837Crossref PubMed Scopus (95) Google Scholar). The concentration-dependent decrease of fluorescence signal upon addition of compound 1 indicated the ability of 1 to displace the fluorescent labeled peptide from the DnaK-peptide complex by competitive binding to the same site of the protein (Fig. 3, A and B). Competition by 1 revealed an IC50 value of 29.1 μm, which translates into a Kd value of 2.6 μm for the DnaK-1 complex as was calculated according to the equation described by Craig (30.Craig D.A. Trends Pharmacol. Sci. 1993; 14: 89-91Abstract Full Text PDF PubMed Scopus (123) Google Scholar). This value is close to the Ki-value directly measured using the APIase assay. Competition assays were repeated with the peptide NRLLLTG, controlling for method of σ32-Q132-Q144-C-IAANS displacement. Displacement of σ32-Q132-Q144-C-IAANS by NRLLLTG from DnaK resulted in an IC50 of 1.3 μm indicating a slightly higher affinity for DnaK of this peptide when compared with those of 1. To determine whether the enzymatically inactive DnaK-1 complex is still able to serve as a holding chaperone and thus can sequester unfolded proteins, the influence of 1 on the formation of DnaK complexes with reduced carboxymethylated lactalbumin (RCMLA) was investigated. Already under native conditions this protein variant is in its unfolded state (2.Langer T. Lu C. Echols H. Flanagan J. Hayer M.K. Hartl F.U. Nature. 1992; 356: 683-689Crossref PubMed Scopus (793) Google Scholar). Again, the peptide NRLLLTG was utilized as a control, because it was already shown by x-ray crystallography to cover a considerable part of the peptide binding cleft of DnaK (32.Stevens S.Y. Cai S. Pellecchia M. Zuiderweg E.R. Protein Sci. 2003; 12: 2588-2596Crossref PubMed Scopus (92) Google Scholar). After incubating DnaK with RCMLA in the absence or presence of different concentrations of 1 (Fig. 3C) or NRLLLTG (Fig. 3D) at 37 °C, DnaK-RCMLA complexes were separated from free DnaK and free RCMLA by native PAGE. The peptide NRLLLTG completely prevented DnaK-RCLMA complex formation at concentrations higher than 25 μm. However, up to a 500 μm concentration 1 was not able to inhibit the formation of this complex. Attempts were made to directly identify the cellular targets of fatty acylated benzamido derivatives in cell lysate. Cell lysates were prepared from E. coli MC4100 cells grown at 42 °C and were incubated with compound 19, a biotinylated derivative of 1 in which the Ile residue was replaced by a C-terminally biotinylated Glu residue. Complexes of 19 with cell proteins were extracted by streptavidin beads, washed extensively, and analyzed by Western blotting with anti-DnaK antibody. Evidence for a DnaK-19 interaction in the cell was established by detection of DnaK in the eluate after washing the beads with a buffer solution containing 1.92 mm of 1 (Fig. 4). Because under heat shock conditions DnaK was described to have important functions in preventing or reversing thermally induced protein damage (3.Hartl F.U. Hayer-Hartl M. Science. 2002; 295: 1852-1858Crossref PubMed Scopus (2799) Google Scholar, 9.Mogk A. Tomoyasu T. Goloubinoff P. Rudiger S. Roder D. Langen H. Bukau B. EMBO J. 1999; 18: 6934-6949Crossref PubMed Scopus (515) Google Scholar, 33.Schroder H. Langer T. Hartl F.U. Bukau B. EMBO J. 1993; 12: 4137-4144Crossref PubMed Scopus (501) Google Scholar), we analyzed the influence of 1 on bacterial growth at 42 °C. Antibacterial activity was detected characterized by a MIC of 380 ± 50 μg/ml against E. coli MC4100 cells at 42 °C (Table 1). At 37 °C, bacterial growth was unaffected at concentrations up to 1.4 mg/ml. The analysis of the bactericidal effect of 1 at 30 °C revealed a MIC of 860 ± 100 μg/ml using E. coli MC4100 Δtig cells and a MIC ≫ 5 mg/ml toward wild-type E. coli MC4100 cells at 30 °C. Uptake studies of 1 by reversed-phase HPLC separation of the cytosol combined with electrospray ionization mass spectrometry analysis verified its intracellular availability. The inability of E. coli cells to regrow on agar plates after application of compound 1 in concentrations above MIC for 12 h in LB medium at 42 °C shows that the inhibitor is bactericidal for E. coli. To exclude membrane effects of 1 on E. coli cell viability, membrane permeability in the presence of 200 μm 1 or 2 was assayed by measuring the activity of β-lactamase released to the medium in comparison to the total activity (supplemental Table S1). No significant difference in the amount of extracellular β-lactamase compared with the total β-lactamase activity was observed after growth of the cells in the presence of buffer or in the presence of compounds 1 or 2.TABLE 1Antibacterial and hemolytic activities of compound 1 and its analogues with substituted fatty acids in position 1CompoundFatty acidInhibition of assisted refolding IC50aCompound concentrations causing 50% (IC50) inhibition of assisted refolding of luciferase.Hemolytic activity EC50bCompound concentrations causing 50% (EC50) hemolysis determined with fresh human erythrocytes.MC4100 MICcMinimal inhibitor
Referência(s)