Journal Home
Search for

Volume 584, Issue 4, Pages 645-651 (19 February 2010)


View previous. 2 of 34 View next.

Gentamicin inhibits HSP70-assisted protein folding by interfering with substrate recognition

Edited by Miguel De la Rosa

Soh Yamamotoa, Shunsuke Nakanoa, Kensuke Owaria, Kazuhiko Fuziwaraa, Nobuaki Ogawaa, Michiro Otakab, Kumiko Tamakib, Sumio Watanabeb, Atsushi Komatsudac, Hideki Wakuic, Ken-ichi Sawadac, Hiroshi Kubotaa, Hideaki ItohaCorresponding Author Informationemail address

Received 29 September 2009; received in revised form 24 November 2009; accepted 10 December 2009. published online 22 December 2009.

Abstract 

We previously reported that gentamicin (GM) specifically binds to heat-shock protein with subunit molecular masses of 70kDa (HSP70). In the present study, we have investigated the effects of GM binding on HSP70-assisted protein folding in vitro. The C-terminal, and not the N-terminal of HSP70 was found to bind to GM. GM significantly suppressed refolding of firefly luciferase in the presence of HSP70 and HSP40, although the ATPase activity of HSP70 was unaffected by GM. A surface plasmon resonance analysis revealed that GM specifically interferes with the binding of HSP70 to a model peptide that mimics the exposed hydrophobic surface of the folding intermediates. These results indicated that GM inhibits the chaperone activity of HSP70 and may suppress protein folding via inhibition of HSP70 in vivo.

Structured summary

MINT-7384283: HSP40 (uniprotkb:P25685) binds (MI:0407) to HSP70 (uniprotkb:P34930) by surface plasmon resonance (MI:0107)

MINT-7384430: RNaseA (uniprotkb:P61823) binds (MI:0407) to HSP70 (uniprotkb:P34930) by surface plasmon resonance (MI:0107)

Article Outline

Abstract

1. Introduction

2. Materials and methods

2.1. Expression vectors

2.2. Protein purification

2.3. GM-affinity column chromatography

2.4. Surface plasmon resonance (SPR) assay

2.4.1. Measurement of HSP40 and HSP70 interaction

2.4.2. Measurement of HSP70 interaction with immobilized peptide

2.4.3. Measurement of GM interaction to HSP70

2.5. Measurement of protein aggregation

2.6. Luciferase refolding assay

2.7. Measurement of ATPase activity

3. Results

3.1. GM-binding domain of HSP70

3.2. GM inhibits HSP70-assisted protein folding

3.3. GM inhibits HSP70–substrate peptide interaction, but not HSP70–HSP40 interaction

4. Discussion

Acknowledgment

Appendix A. Supplementary data

References

Copyright

1. Introduction 

return to Article Outline

Aminoglycoside antibiotics are widely used drugs for the treatment of severe Gram-negative bacterial infections, and gentamicin (GM) is an aminoglycoside antibiotic frequently used for this purpose. However, aminoglycosides have many nephrotoxic side effects [1]. GM induces a complex phenomenon characterized by an increase in the blood urea nitrogen and serum creatinine concentration, leading to severe proximal renal tubular necrosis [2], [3]. Thus, its clinical use is often limited by the nephrotoxicity. Although the role of reactive oxygen metabolites in GM nephropathy has been suggested [4], molecular mechanisms by which GM disturb renal functions is poorly understood.

We previously reported that heat-shock protein with subunit molecular masses of 70kDa (HSP70), a molecular chaperone required for protein folding, was induced in injured tubular epithelial rat kidneys with GM-induced acute renal failure [3]. We also found that GM co-localizes with HSP70 in the lysosome of GM-treated rat kidneys in vivo and HSP70 can be purified from kidney extract using a GM-affinity column in vitro [5]. Aggregation prevention activity of HSP70 was suppressed in the presence of GM, and this drug induced a conformational change in the HSP70. These observations suggest that GM may inhibit the activities of HSP70 in the kidney. However, whether GM affect the HSP70-assisted protein folding is unknown. Moreover, little is known about how GM affects the HSP70 activities by binding.

The HSP70 molecule consists of two domains, i.e., a highly conserved amino-terminal nucleotide-binding domain (NBD) of 45kDa and a carboxyl-terminal substrate binding domain (SBD) of 25kDa. The SBD is further divided into a β-sandwich subdomain (β-PDB) and an α-helical subdomain (α-PDB) [6], [7], [8]. NBD and SBD are connected by a flexible linker region that contributes to allosteric regulation of NBD and SBD [9], [10]. The NBD has two structurally similar lobes (I and II), and each of them are further divided into two small subdomains (IA, IB, IIA and IIB). These subdomains are separated by a deep cleft that captures ATP and cations (one Mg2+ and two K+ ions) contacting the four subdomains [6]. The peptide-binding moiety is composed of two β-sheets with loops in the β-PDB and two helices (helices A and B) in the α-PDB [7], [11]. Helix B works as the “lid”, which closes the cavity through a salt bridge and two hydrogens bound to a loop of β-PDB. The ATP-bound form of HSP70 rapidly binds and releases a peptide [12], [13]. The binding of ATP to the NBD causes a conformational change, which results in structural alterations in the carboxyl-terminal domain leading to substrate release [14], [15]. In contrast, the ADP-bound form of HSP70 slowly binds the peptide but the binding is more stable. The ATP dependent HSP70 chaperone activity is tightly regulated by the various co-chaperones including J-domain proteins (e.g., HSP40) and nucleotide exchange factors [16], [17], [18], [19].

In the present study, we investigated how GM affects the HSP70 activities in vitro and found that GM inhibits the HSP70-assisted protein folding by blocking substrate binding through specific binding to the peptide-binding domain. We discuss the role of the GM–HSP70 interactions during the GM-induced perturbation of cellular activities including the chaperone-assisted protein folding.

2. Materials and methods 

return to Article Outline

2.1. Expression vectors 

Human HSP40 (DNAJB1, Accession No. NM_006145) cDNA was obtained from the Kazusa DNA Research Institute, Japan, and amplified by a polymerase chain reaction using oligonucleotides (5′-CTCGAGTATTGGAAGAACCTGCTCAAGTA-3′ and 5′-CATATGGGTAAAGACTACCAGACG-3′) as primers. The amplified HSP40 cDNA was subcloned into the NdeI–XhoI sites of the pCR2.1 (Life Technologies Corporation, CA, USA) and pET29a (MERCK, Darmstadt, Germany) vectors.

2.2. Protein purification 

HSP40 was expressed in Eschericha coli BL21 (DE3) pLysS cells as a C-terminally 6×his-tagged protein. Cells were grown in LB medium containing kanamycin and chloramphenicol at 37°C, and expression of the 6×his-tagged HSP40 was induced with 0.1mM isopropyl-1-thio-β-d-galactopyranoside. Cells were resuspended in 10mM Tris–HCl (pH 7.4) and lysed by sonication on ice. The supernatant was recovered after centrifugation at 20000rpm for 10min at 4°C, and applied to a Ni-NTA affinity column (GE Healthcare, Uppsala, Sweden) equilibrated with buffer A (0.3MNaCl and 10mM Tris–HCl, pH 7.4) supplemented with 20mM imidazole. After washing with 50mM imidazole in buffer A, proteins were eluted with a linear gradient of 100–500mM imidazole in buffer A. The HSP40 peak fractions were pooled and loaded onto a hydroxyapatite HTP column (Bio-Rad, CA, USA) equilibrated with 100mM potassium phosphate buffer (pH 7.0). After washing, the proteins were eluted with a linear gradient of 100–500mM potassium phosphate (pH 7.0). After concentration by ultrafiltration, the HSP40 peak fractions were applied onto a Superdex 200 HR column (GE Healthcare) equilibrated with buffer B (5% glycerol, 0.1M NaCl in 25mM HEPES–KOH, pH 7.4). The HSP40 fractions were dialyzed against buffer C (5% glycerol, 1mM DTT in 25mM HEPES–KOH, pH 7.4) and stored at −80°C.

HSP70 was purified from porcine brains as follows. The porcine brains were homogenized in 3 volumes of 1mM PMSF, 5mM EDTA, and 15mM β-mercaptoethanol in 10mM Tris–HCl (pH 7.4) on ice and centrifuged at 13000rpm for 30min at 4°C. The supernatants were fractionated by ammonium sulfate precipitation and the 45–90% ammonium sulfate fractions were collected. After overnight dialysis against 10mM Tris–HCl (pH 7.4), the proteins were applied onto a Q-sepharose column (GE Healthcare) equilibrated with 10mM Tris–HCl (pH 7.4). After washing with 0.1M NaCl in 10mM Tris–HCl (pH 7.4), the proteins were eluted with 0.3M NaCl in 10mM Tris–HCl (pH 7.4). After the addition of 50mM KCl and 5mM MgCl2 (final concentrations), the HSP70 peak fractions were load onto an ATP-sepharose column (Sigma, St. Louis, MO, USA) equilibrated with buffer D (0.3M NaCl, 5mM MgCl2, 50mM KCl and 10mM Tris–HCl, pH 7.4). After washing in buffer D, the column was further washed in buffer E (5mM MgCl2, 50mM KCl and 100mM Tris–HCl, pH 7.4). The proteins were eluted with a linear gradient of 0–20mM ATP in buffer E. After concentration by ultrafiltration, the HSP70 peak fractions were separated on a Superdex 200 column (GE Healthcare) equilibrated with buffer B. The HSP70 fractions were dialyzed against buffer C and stored at −80°C.

The amino-terminal and carboxyl-terminal domains of HSP70 (HSP70N and HSP70C, respectively) were purified as previously described [20].

2.3. GM-affinity column chromatography 

GM-Sepharose was prepared using GM (Sigma) and CH-activated Sepharose 4B (GE Healthcare) according to the manufacturer’s instructions. The 6×his-tagged HSP70N or HSP70C was applied to a GM-Sepharose column equilibrated with 10mM Tris–HCl (pH 7.4) and washed with 3 column volumes of 0.15M NaCl in 10mM Tris–HCl (pH 7.4). The proteins were eluted with 5mM GM, 0.15M NaCl in 10mM Tris–HCl (pH 7.4) and analyzed on SDS/PAGE (9% gel) followed by silver staining.

2.4. Surface plasmon resonance (SPR) assay 

2.4.1. Measurement of HSP40 and HSP70 interaction 

All SPR measurements using a performed on a BIAcore 2000 instrument (GE Healthcare) at 25°C, and 25mM HEPES–KOH (pH 7.4) buffer containing 0.005% Tween 20, 5mM MgCl2, and 150mM KCl was used as the running buffer at the flow rate of 10μl/ min. His-HSP40 (0.1μM) dissolved in 10mM potassium phosphate buffer (pH 7.0) was immobilized on CM5 sensor chips (GE Healthcare) through the amine coupling reaction, which was performed according to the standard procedure. HSP40 (∼4000 RU) was immobilized on the chip by this procedure, and 40μl of HSP70 solutions (31.3–4000nM) with or without 10mM GM was loaded on the chip using running buffer supplemented with 1mM ATP. Regeneration of the sensor chip surface was achieved by three 1-min pulses of 1M urea in the running buffer (10μl). The dissociation constant (KD) and observed reaction rate constant (Kobs) were obtained by non-linear curve fitting based on a 1:1 binding with a drifting baseline using BIAevaluation 3.0 software.

2.4.2. Measurement of HSP70 interaction with immobilized peptide 

An HSP70-binding model peptide (CKETAAAKFERQHMDSSTS), which consists of the amino-terminal 1–18 residues of RNase A [21], [22] and an additional cysteine for immobilization on the sensor chip, was purchased from Biosynthesis, USA. Ethylenediamine was coupled on the surface of the C1 sensor chip (GE Healthcare) by subsequent injection of a 0.1M 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide/0.05M N-hydroxysuccinimide solution and 0.1M ethylenediamine (70μl each). Carboxyl groups on the sensor surface were blocked with 1M monoethanolamine, and NHS-PEG6-maleimide (Pierce, Rockford, USA), a spacer for coupling with the cysteine residue, was introduced using a 5mM solution. After the model peptide (1mM) was injected using the running buffer, the free maleimide residues are blocked with 1M l-cysteine. The peptides immobilized on the surface of the sensor chip were ∼210 RU. For the binding experiments, HSP70 (1μM) solutions containing 0–10mM GM were loaded onto the sensor chip. Regeneration of the sensor chip surface was achieved by a 30-s pulse (5μl) of 10mM NaOH and a 1-min pulse (10μl) of 0.5M NaCl in running buffer.

2.4.3. Measurement of GM interaction to HSP70 

Porcine brain HSP70 (1μM) was dissolved in 100mM sodium acetate buffer (pH 4.0) and immobilized on CM5 sensor chips (GE Healthcare) by amino coupling. HSP70 (∼15000 RU) was immobilized on the chip, and GM solutions (19.5μM to 20mM) were loaded on the chip using running buffer. Regeneration of the sensor chip surface was achieved by 1-min pulse (10μl) of 0.5M NaCl in running buffer. The dissociation constant (KD) was obtained by non-linear curve fitting based on a steady-state affinity using BIAevaluation 3.0 software.

2.5. Measurement of protein aggregation 

Thermal aggregation of recombinant firefly luciferase (0.15μM) (Promega) in 25mM HEPES–KOH (pH 7.4) at 42°C was monitored by optical density at 360nm using a Pharmacia Ultrospec 3000 UV–Vis spectrophotometer equipped with a semi-micro-cuvette (0.5ml) with a path length of 10nm and a temperature control unit. In this experiment, the absorbance of 0.05 was observed by maximal aggregation and used as the value for 100% aggregation. In the presence of porcine HSP70 (0.3μM) and various concentrations of GM, aggregation of firefly luciferase was analyzed after incubation of 2min, and 50% inhibitory concentration (IC50) was calculated using origine 6.1 software (Origine Laboratory, Northampton, USA).

2.6. Luciferase refolding assay 

Recombinant firefly luciferase (6.25μM) (Promega, Madison, USA) was incubated for 30min at 30°C in denaturing buffer (6M guanidine–HCl, 5mM MgCl2, and 50mM KCl in 25mM HEPES–KOH, pH 7.4). For refolding, the denatured luciferase was diluted 125-fold into renaturation buffer (2mM DTT, 2mM ATP, 5mM MgCl2, and 50mM KCl in 25mM HEPES–KOH, pH 7.4) in the presence of 1μM HSP40 and 2μM HSP70. The refolding reaction mixture was incubated at 30°C up to 70min in the presence or absence of GM. At each sampling time point, aliquots (1μl) were mixed with 1mg/ml BSA/25mM HEPES–KOH, pH 7.4 (19μl) and the luciferase activity was determined using luciferase assay reagent (Promega) and Infinite M200 (TECAN, Mannedorf, Switzerland). The refolding activity was evaluated by the percent luciferase activity versus the native enzyme.

2.7. Measurement of ATPase activity 

HSP70 (0.5μM) was incubated with 0.5mM ATP in a buffer containing 1mM DTT, 5mM MgCl2, 50mM KCl and 25mM HEPES–KOH (pH 7.4) at 37°C up to 120min. After incubation, the samples (50μl) were transferred to a 96-well plate and BIOMOL GREEN reagent (100μl) (Enzo Life Science, New York, USA) was added. After the samples were incubated for 30min at room temperature, the absorbance at 650nm was determined. The free phosphate concentrations were estimated using an experimentally determined standard curve.

3. Results 

return to Article Outline

3.1. GM-binding domain of HSP70 

The GM-binding assay of the HSP70 fragments prepared by partial digestion with TPCK-trypsin suggested that the GM-binding domain of HSP70 may localize in a carboxyl-terminal region [5]. To exactly determine the GM-binding domain, the 45kDa HSP70N and 35kDa HSP70C domains were expressed in E. coli and purified from the bacteria. These recombinant proteins were applied onto a GM-affinity column and eluted with 5mM GM. The SDS–PAGE analysis of the eluates indicated that HSP70C strongly binds to the GM-affinity column (Fig. 1B). In contrast, HSP70N exhibited no significant binding to the column (Fig. 1A). These results clearly indicated that the GM-binding site of HSP70 is located in the 35kDa carboxyl-terminal region, which contains the SBD domain of 25kDa for the substrate peptide recognition [23]. Furthermore, we determined the dissociation constant (KD) between HSP70 and GM by SPR experiments. HSP70 purified from porcine brain was immobilized on the surface of a sensor chip, and various concentrations of GM were injected (Fig. 1C). A plot of GM concentrations versus equilibration value (Req) (Fig. 1D) provided a KD value of 3.64±0.22×10−3M.


View full-size image.

Fig. 1. Hsp70N, not HSP70C, binds to GM. HSP70N (A) and HSP70 C (B) were applied to a GM-affinity column and eluted with 5mM GM. The pass-through, wash, and eluate fractions were analyzed by SDS/PAGE (9% gel) followed by silver staining. (C) A SPR sensorgram for GM–HSP70 interaction. Increasing concentrations of GM (19.5μM, 39μM, 78μM, 156μM, 313μM, 625μM, 1.25mM, 2.5mM, 5mM, 10mM and 20mM) were loaded on an HSP70-immobilized sensor chip. (D) A plot of equilibration value (Req) versus GM concentration. RU, resonance unit.


3.2. GM inhibits HSP70-assisted protein folding 

We previously indicated that GM diminishes the aggregation prevention activity of HSP70 using rhodanase as a substrate [5]. To examine whether GM affects the protein folding by HSP70, we analyzed the refolding of firefly luciferase as a model substrate. Luciferase was denatured using 6M guanidine–HCl, and refolding was initiated by dilution in renaturation buffer containing HSP70 purified from the porcine brain. The spontaneous refolding was 5.4±2.3% and the refolding rate in the presence of HSP70 was 5.4±1.4% after incubation for 70min (Fig. 2A), indicating that HSP70 alone has no effect on the luciferase refolding. Although the refolding rate slightly increased to 12.4±1.9% in the presence of HSP40, it was significantly increased to 44.8±5.2% in the simultaneous presence of HSP70 and HSP40. These results indicate that HSP70 and HSP40 are both required for the luciferase refolding because they work as a chaperone system, in good agreement with previous reports [24]. In the presence of HSP70 and HSP40, the refolding of luciferase was strongly suppressed by GM in a concentration-dependent manner; 25.8±3.8% and 10.5±1.4% in the presence of 5mM and 10mM GM, respectively (Fig. 2B). In contrast, the inhibitory effect by GM on the luciferase refolding was very low in the presence of HSP40 alone; the refolding rate was 9.8±1.8% and 5.0±1.0% in the presence of 5mM and 10mM GM, respectively (Fig. 2C). In addition, GM at 10mM had no effect on the enzymatic activity of the native luciferase (data not shown). These results indicate that GM inhibited protein refolding assisted by the HSP40–HSP70 chaperone system, probably by direct binding to the HSP70C domain.


View full-size image.

Fig. 2. GM inhibits luciferase refolding by HSP70–HSP40 chaperoning system. (A) Firefly luciferase was denatured by 6M guanidine–HCl and diluted 125-fold into renaturation buffer for refolding in the presence of the indicated HSPs. (B) Effect of GM on luciferase refolding in the presence of HSP70 and HSP40. (C) Effect of GM on luciferase refolding in the presence of HSP40. Mean and S.D. of at least three independent measurements are shown. The data shown by gray and light blue lines in panels B and C are taken from panel A for comparison.


We also analyzed the ATPase activity of HSP70 in the presence of GM, because the binding of substrates to HSP70 is known to stimulate the ATPase activity [23], [25]. However, the ATPase activity of HSP70 was unaffected in the presence of GM. Thus, although this drug binds to the substrate binding domain of HSP70, this binding does not affect the activity of the ATPase domain (Fig. 3). The GM and substrate polypeptides may differ in their ability to induce a conformational change in HSP70. Taken together, these observations suggest substrate recognition, but not the ATPase activity, of HSP70 is affected by the GM binding to a carboxyl-terminal region.


View full-size image.

Fig. 3. GM has no significant effect on the ATPase activity of HSP70. HSP70 was incubated with ATP in the presence of GM. The grey and white columns indicate ATP hydrolyzed after incubation for 60 and 120min, respectively. Mean and S.D. of at least three independent measurements are shown.


3.3. GM inhibits HSP70–substrate peptide interaction, but not HSP70–HSP40 interaction 

GM inhibited the refolding of luciferase by the HSP40–HSP70 chaperone system (Fig. 2B), and there were two possibilities explaining this result; GM inhibited the HSP70–substrate interaction or HSP70–HSP40 interaction. We first tested whether GM affects HSP70–substate interaction by measuring aggregation prevention activity of HSP70 using luciferase as a substrate. Luciferase was aggregated at 42°C, but the aggregation was completely inhibited by the presence of HSP70 (Fig. 4A). Aggregation prevention activity of HSP70 was diminished by GM in a concentration-dependent manner (Fig. 4B), and IC50 of the GM activity was calculated to be 140±7μM (Fig. 4C). These results strongly suggested that GM specifically inhibited substrate recognition of HSP70.


View full-size image.

Fig. 4. GM inhibits the aggregation prevention activity of HSP70. (A) Firefly luciferase solution was incubated diluted in the presence of GM (red line) or HSP70 (blue line) at 42°C up to 2000s. (B) Effect of GM on aggregation prevention activity of HSP70 against luciferase. (C) The 50% inhibitory concentration (IC50) of GM for aggregation prevention activity of HSP70. Mean and S.D. of at least three independent measurements are shown. The data shown by red and blue lines in panels B was taken from panel A for comparison.


Next, we examined the effects of GM on the HSP70 binding to the substrate by SPR using a model peptide containing the amino-terminal 1–18 residues of RNase A, because this peptide is known to bind to HSP70 [21], [25]. This model peptide was first immobilized on the surface of a sensor chip, then HSP70 solutions containing 0–10mM GM were injected. A sensorgram for the HSP70-model peptide system in the absence of GM is shown in Fig. 5A, which gives a KD of 2.26×10−8M. The KD value represents a strong interaction between HSP70 and the model peptide. In the presence of GM, the SPR signals significantly decreased with the change in the GM concentration. This result indicates that GM blocks the binding of HSP70 to the immobilized model peptide.


View full-size image.

Fig. 5. GM significantly inhibits HSP70–peptide interaction but not HSP70–HSP40 interaction as analyzed by SPR. (A) A sensorgram of HSP70 interaction with model peptide in the presence of GM. HSP70 solutions containing different concentrations of GM were loaded onto peptide-immobilized sensor chip. (B) Sensorgrams of HSP70 interacting with HSP40 in the presence or absence of ATP and GM. The HSP70 concentration was fixed at 1μM. (C) A plot of observed reaction rate constant (Kobs) versus HSP70 concentration. Open circle and closed square indicate the presence and absence of GM, respectively. Mean and S.D. at least three independent measurements are shown.


Finally, we investigated the effects of GM on the HSP40–HSP70 interaction using an HSP40-immobilized chip. Since the HSP40–HSP70 interaction is known to be ATP dependent [28], an SPR assay was performed in the presence and absence of ATP. No significant HSP40–HSP70 interaction was detected in the absence of the ATP, regardless of the GM addition (Fig. 5B and unpublished data). In the presence of ATP, HSP70 significantly interacted with HSP40 in the absence of GM and KD was estimated to be 2.13±0.24×10−6M. In the presence of 10mM GM, the KD was 2.79±0.84×10−6, indicating that GM has a slight or no effect on the HSP40–HSP70 interaction (Fig. 5B and C, Supplementary Fig. 1 and Table 1). Taken together, these observations indicate that GM inhibits the chaperone activity of HSP70 by interfering with the substrate recognition.

4. Discussion 

return to Article Outline

In the present study, we demonstrated that GM strongly suppresses protein folding by the HSP40–HSP70 chaperone system without affecting the ATPase activity (Fig. 2, Fig. 3). Moreover, GM significantly inhibited the aggregation prevention activity of HSP70 (IC50=140±7μM), although the effect on the HSP40–HSP70 interaction was very low (Fig. 4, Fig. 5). These observations indicate that GM significantly inhibits the protein folding activity of HSP70 by modulating substrate recognition.

We also found that GM binds to the carboxyl-terminal peptide-binding domain, but not to the amino-terminal nucleotide-binding domain of HSP70 (Fig. 1A and B). GM inhibited binding of HSP70 to a model peptide (Fig. 5A). These results suggest that GM has an inhibitory effect on peptide recognition by the carboxyl-terminal peptide-binding domain. However, KD value of HSP70–GM interaction was estimated to be 3.6×10−3M (Fig. 1C and D), indicating a weak or transient interaction. Since IC50 for the inhibition of HSP70-dependent aggregation prevention was significantly lower than the KD value, the weak or transient interaction may induce structural alteration of HSP70 rather than direct competition to substrate peptides. It may also be possible that GM interferes HSP70–substrate binding by interacting with HSP70 and substrate proteins.

The peptide-binding domain in the carboxyl-terminal region of HSP70 is known to recognize the hydrophobic core of folding intermediates or the unstructured backbone of the non-native protein [26], [27]. These interactions stimulate productive folding of newly-synthesized polypeptides or refolding of denatured proteins by preventing any non-essential interaction of hydrophobic cores in a protein and aggregation of partially folded proteins. In our previously study, binding of GM to HSP70 resulted in a conformational change of HSP70 [5]. However, the ATPase activity of HSP70, which is known to be stimulated by the presence of the substrate protein, was unaffected by GM. Thus, the conformational change may be limited in the carboxyl-terminal region and the effect of the conformational change may not be transferred to the amino-terminal region.

Although a number of inhibitors have been reported for HSP90, including geldanamycin derivatives, novobiocin and cisplatin [29], [30], [31], there are only a few inhibitors of HSP70. We previously reported that geranly-geranyl-acetone (GGA) inhibits the aggregation prevention activity of HSP70 by binding to the carboxyl-terminal region [20]. Liebscher et al. reported that N-α-[tetradecanoyl-(4-aminomethylbenzoyl)]-l-asparagine, a fatty acylated benzamido derivative, prevents bacteria growth by inhibiting the protein folding activity of the bacterial homologue of HSP70, DnaK [32]. As this substance competed with the substrate peptides for the DnaK binding, its binding site appears to be the substrate recognition domain at the carboxyl-terminus. An hydrophobic interaction is probably important to the binding of GGA and the fatty acylated benzamido derivative to HSP70/DnaK, because these substances have a hydrophobic straight carbon chain and the inhibitory effect of the latter is increased with extension of the carbon chain. In contrast, GM has no hydrophobic straight carbon chain. Thus, GM may differ from the other two substances in the mode of interaction with HSP70 at least in part, although these three substances block binding of the substrate proteins to the carboxyl-terminal region of HSP70/DnaK.

GM is administered at 4–5mg/kg every 36–48h on clinical use (extended interval method), because concentration of GM in blood decreases 2.5mg/kg every 12h. The peak concentration and trough concentration of GM in blood using the extended interval method are 5–12μg/ml (approx. 11–26μM) and under 2μg/ml (approx. 4μM), respectively [33]. Most of administered GM (approx. 70–90%) is excretes into urine, and specifically accumulates in the cortex of kidney via megalin receptor within 30min after GM administration. The accumulated GM persists in the kidney for a long time (half-life>100h) [34], [35]. GM is highly accumulated in lysosomes (approx. 90%), providing a very high concentration (30mM or more) [36]. Approximately 10% of GM localizes with cytoplasm, nuclear, mitochondria and endoplasmic reticulum [37]. Concentration of GM to lysosome is thought to cause destruction of this organelle [34], [38], [39], and it probably increases GM concentration in the cytosol. Thus, the GM concentration in the cytosol of kidney cells in patients can be comparable to the GM concentration to inhibit 50% of HSP70-dependent aggregation prevention activity (140μM).

The HSP70 family chaperones are highly conserved proteins that play essential roles in the protein metabolism and quality control by controlling the protein folding, translocation, degradation and gene expression [26], [40]. Thus, inhibition of the HSP70 activities, including folding of the newly-synthesized proteins and refolding of misfolded proteins by GM may result in perturbation of the cellular activities essential for cell survival. Since GM co-localizes with HSP70 in the enlarged lysosomes of the rat kidneys with GM-induced acute tubular injury in vivo [5], GM may affect the HSP70 activity required for kidney cells and cause nephrotoxicity via the HSP70 inhibition.

Acknowledgments 

return to Article Outline

S.Y. was supported by a Sasakawa Scientific Research Grant from the Japan Science Society, and H.I. was supported by a Grant-in-Aid for Scientific Research (Exploratory Research: 16651056) from the Japanese Ministry of Education, Science, Sports and Culture.

Appendix A. Supplementary data 

return to Article Outline


View full-size image.

Fig. S1. A sensorgram of HSP40-HSP70 in the presence or absence of GM. (A and B) Increasing concentrations of HSP70 (31.3, 62.5, 125, 250, 500, 1000 nM) were loaded onto HSP40 immobilized sensor chip with (B) or without GM (A) in the presence of 1mM ATP. RU, resonance unit. Note that the running buffer used for creating the post-injection current does not contain GM in all the SPR experiments.


Table S1. The kinetic parameter of HSP40–HSP70 interaction in the presence or absence of GM. The GM+ and GM- indicate HSP40–HSP70 interaction in the presence of 10 mM GM and in the absence of GM using SPR analysis (see “Materials and methods” section). Kon, Koff, KA, KD are association rate constant, dissociation rate constant, association constant and dissociation constant, respectively. Mean and standard deviation of at least three independent measurements are shown.

References 

return to Article Outline

[1]. [1]Parlakpinar H, Tasdemir S, Polat A, Bay-Karabulut A, Vardi N, Ucar M, et al. Protective role of caffeic acid phenethyl ester (cape) on gentamicin-induced acute renal toxicity in rats. Toxicology. 2005;207:169–177. CrossRef

[2]. [2]Al-Majed AA, Mostafa AM, Al-Rikabi AC, Al-Shabanah OA. Protective effects of oral arabic gum administration on gentamicin-induced nephrotoxicity in rats. Pharmacol. Res. 2002;46:445–451. MEDLINE | CrossRef

[3]. [3]Komatsuda A, et al. Altered localization of 73-kilodalton heat-shock protein in rat kidneys with gentamicin-induced acute tubular injury. Lab. Invest. 1993;68:687–695. MEDLINE

[4]. [4]Walker PD, Barri Y, Shah SV. Oxidant mechanisms in gentamicin nephrotoxicity. Renal Fail. 1999;21:433–442.

[5]. [5]Miyazaki T, et al. 73-kDa molecular chaperone HSP73 is a direct target of antibiotic gentamicin. J. Biol. Chem. 2004;279:17295–17300. MEDLINE | CrossRef

[6]. [6]Flaherty KM, DeLuca-Flaherty C, McKay DB. Three-dimensional structure of the ATPase fragment of a 70K heat-shock cognate protein. Nature. 1990;346:623–628. MEDLINE | CrossRef

[7]. [7]Mayer MP, Bukau B. Hsp70 chaperones: cellular functions and molecular mechanism. Cell Mol. Life Sci. 2005;62:670–684. CrossRef

[8]. [8]Morshauser RC, Wang H, Flynn GC, Zuiderweg ER. The peptide-binding domain of the chaperone protein Hsc70 has an unusual secondary structure topology. Biochemistry. 1995;34:6261–6266.

[9]. [9]Swain JF, Dinler G, Sivendran R, Montgomery DL, Stotz M, Gierasch LM. Hsp70 chaperone ligands control domain association via an allosteric mechanism mediated by the interdomain linker. Mol. Cell. 2007;26:27–39. MEDLINE | CrossRef

[10]. [10]Bertelsen EB, Chang L, Gestwicki JE, Zuiderweg ER. Solution conformation of wild-type E. coli Hsp70 (DnaK) chaperone complexed with ADP and substrate. Proc. Natl. Acad. Sci. USA. 2009;106:8471–8476. CrossRef

[11]. [11]Zhu X, Zhao X, Burkholder WF, Gragerov A, Ogata CM, Gottesman ME, et al. Structural analysis of substrate binding by the molecular chaperone DnaK. Science. 1996;272:1606–1614. MEDLINE

[12]. [12]Schmid D, Baici A, Gehring H, Christen P. Kinetics of molecular chaperone action. Science. 1994;263:971–973. MEDLINE

[13]. [13]Greene LE, Zinner R, Naficy S, Eisenberg E. Effect of nucleotide on the binding of peptides to 70-kDa heat shock protein. J. Biol. Chem. 1995;270:2967–2973. MEDLINE | CrossRef

[14]. [14]Palleros DR, Reid KL, Shi L, Welch WJ, Fink AL. ATP-induced protein–Hsp70 complex dissociation requires K+ but not ATP hydrolysis. Nature. 1993;365:664–666. MEDLINE | CrossRef

[15]. [15]Freeman BC, Myers MP, Schumacher R, Morimoto RI. Identification of a regulatory motif in Hsp70 that affects ATPase activity, substrate binding and interaction with HDJ-1. EMBO J. 1995;14:2281–2292. MEDLINE

[16]. [16]Bukau B, Weissman J, Horwich A. Molecular chaperones and protein quality control. Cell. 2006;125:443–451. MEDLINE | CrossRef

[17]. [17]Fan CY, Lee S, Cyr DM. Mechanisms for regulation of Hsp70 function by Hsp40. Cell Stress Chaperones. 2003;8:309–316. MEDLINE | CrossRef

[18]. [18]Polier S, Dragovic Z, Hartl FU, Bracher A. Structural basis for the cooperation of Hsp70 and Hsp110 chaperones in protein folding. Cell. 2008;133:1068–1079. CrossRef

[19]. [19]Xu Z, Page RC, Gomes MM, Kohli E, Nix JC, Herr AB, et al. Structural basis of nucleotide exchange and client binding by the Hsp70 cochaperone Bag2. Nat. Struct. Mol. Biol. 2008;15:1309–1317. CrossRef

[20]. [20]Otaka M, et al. The induction mechanism of the molecular chaperone HSP70 in the gastric mucosa by Geranylgeranylacetone (HSP-inducer). Biochem. Biophys. Res. Commun. 2007;353:399–404. CrossRef

[21]. [21]Wang C, Lee MR. High-level expression of soluble rat hsc70 in Escherichia coli: purification and characterization of the cloned enzyme. Biochem. J. 1993;294(Pt 1):69–77.

[22]. [22]Terlecky SR, Chiang HL, Olson TS, Dice JF. Protein and peptide binding and stimulation of in vitro lysosomal proteolysis by the 73-kDa heat shock cognate protein. J. Biol. Chem. 1992;267:9202–9209. MEDLINE

[23]. [23]Flynn GC, Chappell TG, Rothman JE. Peptide binding and release by proteins implicated as catalysts of protein assembly. Science. 1989;245:385–390. MEDLINE

[24]. [24]Wegele H, Wandinger SK, Schmid AB, Reinstein J, Buchner J. Substrate transfer from the chaperone Hsp70 to Hsp90. J. Mol. Biol. 2006;356:802–811. MEDLINE | CrossRef

[25]. [25]Tapley TL, Cupp-Vickery JR, Vickery LE. Sequence-dependent peptide binding orientation by the molecular chaperone DnaK. Biochemistry. 2005;44:12307–12315.

[26]. [26]Hartl FU, Hayer-Hartl M. Molecular chaperones in the cytosol: from nascent chain to folded protein. Science. 2002;295:1852–1858. CrossRef

[27]. [27]Rudiger S, Buchberger A, Bukau B. Interaction of Hsp70 chaperones with substrates. Nat. Struct. Biol. 1997;4:342–349. MEDLINE | CrossRef

[28]. [28]Suh WC, Lu CZ, Gross CA. Structural features required for the interaction of the Hsp70 molecular chaperone DnaK with its cochaperone DnaJ. J. Biol. Chem. 1999;274:30534–30539. MEDLINE | CrossRef

[29]. [29]Soti C, Nagy E, Giricz Z, Vigh L, Csermely P, Ferdinandy P. Heat shock proteins as emerging therapeutic targets. Br. J. Pharmacol. 2005;146:769–780. MEDLINE | CrossRef

[30]. [30]Donnelly A, Blagg BS. Novobiocin and additional inhibitors of the Hsp90 C-terminal nucleotide-binding pocket. Curr. Med. Chem. 2008;15:2702–2717. CrossRef

[31]. [31]Ishida R, et al. Cisplatin differently affects amino terminal and carboxyl terminal domains of HSP90. FEBS Lett. 2008;582:3879–3883. Abstract | Full Text | Full-Text PDF (357 KB) | CrossRef

[32]. [32]Liebscher M, Jahreis G, Lucke C, Grabley S, Raina S, Schiene-Fischer C. Fatty acyl benzamido antibacterials based on inhibition of DnaK-catalyzed protein folding. J. Biol. Chem. 2007;282:4437–4446. MEDLINE | CrossRef

[33]. [33]Pacifici GM. Clinical pharmacokinetics of aminoglycosides in the neonate: a review. Eur. J. Clin. Pharmacol. 2009;65:419–427. CrossRef

[34]. [34]Zietse R, Zoutendijk R, Hoorn EJ. Fluid, electrolyte and acid–base disorders associated with antibiotic therapy. Nat. Rev. Nephrol. 2009;5:193–202.

[35]. [35]Schmitz C, Hilpert J, Jacobsen C, Boensch C, Christensen EI, Luft FC, et al. Megalin deficiency offers protection from renal aminoglycoside accumulation. J. Biol. Chem. 2002;277:618–622. MEDLINE | CrossRef

[36]. [36]El Mouedden M, Laurent G, Mingeot-Leclercq MP, Tulkens PM. Gentamicin-induced apoptosis in renal cell lines and embryonic rat fibroblasts. Toxicol. Sci. 2000;56:229–239. MEDLINE | CrossRef

[37]. [37]Sandoval RM, Molitoris BA. Gentamicin traffics retrograde through the secretory pathway and is released in the cytosol via the endoplasmic reticulum. Am. J. Physiol. Renal Physiol. 2004;286:F617–F624. MEDLINE

[38]. [38]Servais H, Van Der Smissen P, Thirion G, Van der Essen G, Van Bambeke F, Tulkens PM, et al. Gentamicin-induced apoptosis in LLC-PK1 cells: involvement of lysosomes and mitochondria. Toxicol. Appl. Pharmacol. 2005;206:321–333. CrossRef

[39]. [39]Priyamvada S, Priyadarshini M, Arivarasu NA, Farooq N, Khan S, Khan SA, et al. Studies on the protective effect of dietary fish oil on gentamicin-induced nephrotoxicity and oxidative damage in rat kidney. Prostaglandins Leukot. Essent. Fatty Acids. 2008;78:369–381. Abstract | Full Text | Full-Text PDF (463 KB) | CrossRef

[40]. [40]Itoh H, Tashima Y. The stress (heat shock) proteins. Int. J. Biochem. 1991;23:1185–1191. CrossRef

a Department of Life Science, Faculty of Engineering and Resource Science, Akita University, 1-1 Tegata Gakuen Town, Akita City 010-8502, Japan

b Department of Gastroenterology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunkyo-Ku, Tokyo, Japan

c Third Department of Internal Medicine, Akita University School of Medicine, Akita 010-8543, Japan

Corresponding Author InformationCorresponding author. Fax: +81 18 883 3041.

PII: S0014-5793(09)01067-9

doi:10.1016/j.febslet.2009.12.021


View previous. 2 of 34 View next.