Chromium(III) ion and thyroxine cooperate to stabilize the transthyretin tetramer and suppress in vitro amyloid fibril formation

FEBS Letters 580 (2006) 491–496

Chromium(III) ion and thyroxine cooperate to stabilize the transthyretin tetramer and suppress in vitro amyloid fibril formation Takashi Satoa,1, Yukio Andob,1, Seiko Susukia,1, Fumi Mikamia, Shinji Ikemizua, Masaaki Nakamurab, Ole Suhrc, Makoto Anrakua, Toshiya Kaid, Mary Ann Suicoa, Tsuyoshi Shutoa, Mineyuki Mizuguchie, Yuriko Yamagataa, Hirofumi Kaia,* b

a Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1 Oe-Honmachi, Kumamoto 862-0973, Japan Graduate School of Medical Sciences, Department of Laboratory Medicine, Kumamoto University, 1-1-1 Honjo, Kumamoto 860-0811, Japan c Department of Medicine, University of Umea˚, S-901 85 Umea˚, Sweden d Pharmaceutical Research Center, Nipro Co., 3023 Nojicho, Kusatsu, Shiga 525-0055, Japan e Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceutical University, Toyama 930-0194, Japan

Received 3 November 2005; revised 29 November 2005; accepted 15 December 2005 Available online 22 December 2005 Edited by Jesus Avila

Abstract Transthyretin (TTR) amyloid fibril formation, which is triggered by the dissociation of tetrameric TTR, appears to be the causative factor in familial amyloidotic polyneuropathy and senile systemic amyloidosis. Binding of thyroxine (T4), a native ligand of TTR, stabilizes the tetramer, but the bioavailability of T4 for TTR binding is limited due to the preferential binding of T4 to globulin, the major T4 carrier in plasma. Here, we show that Cr3+ increased the T4-binding capacity of wild-type (WT) and amyloidogenic V30M-TTR. Moreover, we demonstrate that Cr3+ and T4 cooperatively suppressed in vitro fibril formation due to the stabilization of WT-TTR and V30M-TTR. Ó 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: Amyloid; Transthyretin; Thyroxine; Cr3+; Familial amyloidotic polyneuropathy

1. Introduction Transthyretin (TTR), which is present in human plasma and cerebrospinal fluid, is a homotetrameric protein of 55 kDa. TTR binds and transports thyroxine (T4) and the retinol binding protein [1]. In certain individuals, TTR is converted into an insoluble fibrillar structure called amyloid. The precise mechanisms underlying the conversion of TTR into amyloid fibrils are unknown, but the extensive b-sheet structure of TTR might be responsible for its amyloidogenic potential [2]. These amyloid fibrils putatively cause senile systemic amyloidosis (SSA) and familial amyloidotic polyneuropathy (FAP) by virtue of the neurotoxic effects of amyloid or by means of physical interference with normal organ function [3]. Tetrameric TTR is not itself amyloidogenic, but dissociation of the tetramer into a compact non-native monomer with low conformational stability can lead to amyloid fibril formation [4]. Several of the 50 FAP-associated TTR single-site mutations have a normal tet* Corresponding author. Fax: +81 96 371 4405. E-mail address: [email protected] (H. Kai). 1

These authors contributed equally to this work.

rameric structure under physiological conditions [5]; however, these mutations significantly destabilize the tetramer [6,7]. Preventing the conformational changes which initiate amyloid fibrillization could intervene in the pathogenesis of FAP [8]. Small inhibitor molecules that bind to unoccupied T4-binding sites of TTR and the halide ions, chloride and iodide, were shown to enhance tetrameric TTR stability [9,10]. We screened a number of metal ions that can affect in vitro amyloid formation and we report here that Cr3+ enhanced the effect of T4 on the thermostability of both normal- and V30M-TTR tetramers and suppressed tetramer dissociation induced by low pH.

2. Materials and methods 2.1. Purification of wild-type TTR and amyloidogenic V30M TTR from human plasma Serum wild-type TTR (s-WT-TTR) was purified as described previously [11]. Serum amyloidogenic V30M-TTR (s-V30M-TTR) was prepared using fraction IV obtained from human plasma of homozygotic V30M FAP patients by Cohn’s ethanol fraction method. The subsequent purification was essentially as described by Ando et al. [11]. 2.2. Preparation of recombinant WT-TTR (r-WT-TTR) Two primers, 5 0 -AACATATGGGTCCGACCGGTACCGGTGA3 0 and 5 0 -AAGTCGACTTATTCTTTCGGGTTGGTAA-3 0 , were designed to amplify the wild-type TTR gene by PCR. The PCR product was digested and ligated into NdeI/SalI pre-digested pET-22b(+) vector (Novagen, Darmstadt, Germany). The TTR plasmid was used to transform Escherichia coli strain, BL21(DE3) STAR (Invitrogen, Carlsbard, CA). Protein expression was induced by isopropyl-b-D thiogalactopyranoside (IPTG) for 3 h before harvesting cells. Cell pellet was resuspended in 20 mM phosphate buffer saline (PBS, pH 7.0), and lysed by sonication at 4 °C. The supernatant was obtained, filtered and applied onto a DEAE Sepharose Fast Flow column (Amersham). Bound proteins were eluted with a 20 mM phosphate (pH 7.0)/200 mM NaCl buffer. TTR was further purified by reverse-phase high-performance liquid chromatography (Cosmosil 5C4-AR-300 column, Nacalai Tesque, Japan) with an acetonitrile gradient. The fractions containing TTR were pooled and dialyzed extensively against 20 mM NH4HCO3 solution, followed by lyophilization. Purity of TTR was assessed by SDS–PAGE. 2.3. [125I]T4 and 51Cr3+ binding studies Serum WT-TTR and V30M-TTR having a concentration of 100 lg/ mL in PBS buffer were used for T4 and Cr3+ binding analyses. The concentration of [125I]T4 used was 0.6 nM for binding with s-WT-TTR and

0014-5793/$32.00 Ó 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2005.12.047

492

T. Sato et al. / FEBS Letters 580 (2006) 491–496

6.0 nM for s-V30M-TTR. Binding of [125I]-labeled T4 (1500 lCi/lg, Amersham Biosciences, NJ) to TTR was carried out at a total volume of 25 lL at 4 °C for 14–18 h in the absence or presence of CrCl3 (300 or 500 lM). Free and protein-bound [125I]T4 were separated using CentriSep columns (Princeton Separations, NJ) and the bound [125I]T4 was determined using an auto-well gamma counter (ARC-2000, Aloka). The affinity constants and the number of binding sites were calculated using Scatchard analysis. For binding assay with Cr3+, 5 lM of [51Cr]Cl3 (872 lCi/lg, Perkin–Elmer, MA) was incubated with TTR protein solution in the absence or presence of 360 nM T4, in a manner as described above. All incubations were done in duplicate. 2.4. Thioflavine T-binding assay Serum or recombinant WT-TTR, or serum V30M-TTR was incubated at 37 °C for 5 days in PBS (pH 4.0) with the indicated concentration of Cr3+ and 360 nM to 10 lM T4 or 0 to 10 lM diflunisal. TTR samples were prepared to a final concentration of 0.2 mg/mL. Thioflavine T-binding assays were performed on 2.5 lg/mL TTR samples by adding freshly prepared 10 lM thioflavine T to 50 mM glycine buffer (pH 9.0). Fluorescence emission spectra were obtained with excitation and emission wavelengths of 450 and 482 nm, respectively. Fluorescence measurements were performed with a F-4500 Hitachi spectrofluorometer (Hitachi, Tokyo, Japan). 2.5. High-sensitivity differential scanning calorimetry Differential scanning calorimetry (DSC) was performed with a differential scanning calorimeter (MC-2, MicroCal, Northhampton, MA) with cell volumes of 1.22 mL using heating rates of 1 K/min, as described previously [12]. TTR samples were prepared at a concentration of 25 lM in 20 mM sodium phosphate and 150 mM NaCl at the indicated pH in the presence or absence of Cr3+. The data obtained from DSC were applied to non-linear fitting algorithms to calculate the thermodynamic parameters, thermal denaturation temperature (Tm), calorimetric enthalpy (DHcal) and van’t Hoff enthalpy (DHv), from the temperature dependence of excess molar heat capacity, Cp, by employing Origine scientific plotting software (OriginLab Co.). 2.6. Far circular dichroism Far circular dichroism (CD) was monitored with JASCO J-720 (Nihon Bunko, Tokyo, Japan). Serum WT-TTR samples were incubated at 37 °C for 2 days in PBS buffer (pH 4.0) with or without Cr3+ (500 lM). Samples were then analyzed at 25 °C using a bandwidth of 1.0 nm, a time constant of 1 s, a step resolution of 0.5 nm and a scan speed of 5 nm/min. CD measurements were carried out in triplicate and spectra were reported as the means of three scans in the range of 220–240 nm.

3. Results 3.1. Effect of Cr3+ on the binding of T4 to serum WT-TTR and V30M-TTR It has been shown that binding of the natural ligand T4 stabilizes the tetrameric TTR [13] and that certain metal ions af-

fect amyloidogenesis [14], therefore, we first determined the effect of Cr3+ on T4 binding to TTR. Scatchard analysis of purified s-WT-TTR and radiolabeled T4 resulted in a rectilinear curve with a measured affinity (Ka) of 149.5 L/mmol. The number of binding sites per mol TTR in the eluate was 1.16 (Table 1). Surprisingly, Cr3+ dose-dependently increased the concentration of bound T4. In the sample containing 500 lM Cr3+, the bound T4 and the number of binding sites per mol TTR were approximately twice than that without Cr3+ (Table 1 and Fig. 1A). However, the binding affinity for T4 was diminished by the addition of Cr3+ (Table 1). We also examined the effect of Cr3+ on T4-binding to V30M-TTR and found that the Ka was slightly increased upon the addition of 500 lM Cr3+. Moreover, the number of bound T4 per mol V30M-TTR was increased 3-fold in the presence of Cr3+ (Table 1). These data suggested that Cr3+ could increase the maximal binding capacity of WT-TTR and V30M-TTR for T4. We next asked whether Cr3+ itself could possibly bind to WT-TTR. In Scatchard analysis using purified s-WT-TTR and [51Cr3+], we observed that the molar ratio of Cr3+ binding sites to TTR was 2.16. The presence of T4 had no effect on the number of bound ligands per mol TTR (Table 1 and Fig. 1B). However, the Ka was diminished from 1.40 (T4, 0 nM) to 0.60 (T4, 360 nM). These results indicated that Cr3+ might bind to WT-TTR and that T4 negatively affects the binding affinity but not the binding capacity of WT-TTR towards Cr3+. 3.2. Cr3+ and T4 cooperatively suppressed amyloid fibril formation To investigate the effect of Cr3+ and T4 on TTR amyloid fibril formation, we performed thioflavine T-binding assay. It was previously shown that a T4 concentration of more than 10 lM was needed for T4 alone to show a suppressive effect on amyloid fibril formation of WT-TTR [13]. However, 0.36 lM T4 together with low Cr3+ concentrations (10– 50 lM) significantly suppressed amyloid fibril formation of s-WT-TTR, when incubated at pH 4.0 for 5 days (Fig. 2A). Amyloid fibril formation of r-WT-TTR was also suppressed by Cr3+ in combination with T4 (0.36 lM), albeit at a higher concentration of Cr3+ (500 lM) (Fig. 2B). To understand the discrepant effect of Cr3+ on s-WT-TTR and r-WT-TTR, we determined the level of T4 remaining in the TTR samples purified from human plasma used in this study. We found that s-WT-TTR samples contained 40 nM T4 in bound form (data not shown). The presence of this natural ligand most probably contributed to the stabilization of the tetrameric s-WT-TTR.

Table 1 Summary of scatchard analysis of T4 and Cr3+ binding in serum TTR Bound ligands (lM)

Bound ligands/tetramer TTR

Ka (L/mmol)

T4 binding WT-TTR WT-TTR with Cr3+ (60 lM) WT-TTR with Cr3+ (300 lM) WT-TTR with Cr3+ (500 lM) V30M-TTR V30M-TTR with Cr3+ (500 lM)

2.16 2.37 3.06 3.97 0.29 1.00

1.16 1.28 1.65 2.14 0.16 0.54

149.5 105.8 86.0 47.2 18.8 29.7

Cr3+ binding WT-TTR WT-TTR with T4 (360 nM)

4.00 4.17

2.16 2.25

1.40 0.60

T. Sato et al. / FEBS Letters 580 (2006) 491–496

Fig. 1. Binding of T4 to WT-TTR was increased in a Cr3+ concentration-dependent manner. (A) 100 lg/mL of s-WT-TTR and 0.6 nM [125I]T4 were incubated with or without Cr3+. Free and proteinbound [125I]T4 were separated by CentriSep column and bound [125I]T4 was determined using a gamma counter. The affinity constants and the number of binding sites were calculated using Scatchard analysis. (B) 100 lg/mL s-WT-TTR and 5 lM [51Cr3+] were incubated with or without T4 using the method described above. All incubations were done in duplicate.

Previous studies showed that several non-steroidal antiinflammatory drugs (NSAIDs), which also bind to TTR as mimic compounds of T4, suppressed amyloid fibril formation [9,15]. We therefore examined the combined effects of Cr3+ and an NSAID, diflunisal, on s-WT-TTR fibril formation. A concentration higher than 1 lM was needed for diflunisal alone to show suppressive effect on amyloid fibril formation of s-WT-TTR, but Cr3+ (10 lM) enhanced the effect of diflunisal at this concentration when co-incubated at pH 4.0 for 5 days (Fig. 2C). When we examined the effect of Cr3+ on s-V30M-TTR, we found that 10 lM Cr3+ in the presence of 0.36 lM T4 was insufficient in inhibiting fibril formation (Fig. 2D). V30MTTR has a very low binding affinity for T4 [16] (Table 1) so it is possible that higher concentration of T4 might be needed for suppressing fibril formation. As expected, at 10 lM T4, fluorescence intensity was reduced by >50% of the control when Cr3+ was present compared to that without Cr3+ (Fig. 2D). Collectively, our data suggest that Cr3+ and T4 or diflunisal may cooperatively suppress the amyloid fibril formation of WT- and V30M-TTR.

493

3.3. Effect of Cr3+ on the thermostability of TTR To evaluate the effect of Cr3+ on the potential conformational stability of s-WT- and V30M-TTR, we used differential scanning calorimetry, which is a useful technique for characterizing the energetics of thermal unfolding of proteins [12]. Cr3+ (500 lM) increased the excess heat capacity, Cp, of sWT-TTR at pH 7.4 and concomitantly, an increase in the calorimetric enthalpy was observed, DHcal is 457 kcal mol
1 and 99.6 kcal mol
1 with and without Cr3+, respectively (Fig. 3A and Table 2). WT-TTR at pH 5.0 had a lower thermal denaturation than at pH 7.4, but Cr3+ shifted the thermogram towards a higher temperature and almost completely stabilized WT-TTR incubated at pH 5.0 against heat-related unfolding (Fig. 3B and Table 2). More importantly, at pH 6.0, the DHcal of V30M-TTR in the presence of Cr3+ was 2-fold higher than that without Cr3+ (Table 2). The Cp of V30M-TTR at pH 6.0 was 50% lower than at pH 7.4, but Cr3+ increased the heat capacity such that at pH 6.0, this sample had the same Cp value as at pH 7.4 (Fig. 3C), indicating that Cr3+ stabilized the amyloidogenic V30M-TTR against thermal-induced denaturation. Moreover, the DHv/DHcal ratio of V30M-TTR at pH 6.0 (>1) suggests that aggregation of the denatured form may occur, but the addition of Cr3+ inhibited its aggregation (Table 2). Collectively, these data indicated that Cr3+ stabilizes WT-TTR and V30M-TTR against thermal unfolding at low pH. Because certain Cr3+ complexes have been found to induce a structural transition of human orosomucoid from the native twisted b-sheet to a more compact a-helix [17], we used circular dichroic analysis to examine the effect of Cr3+ on the structural transition of s-WT-TTR. Cr3+ had no effect on the secondary structure of s-WT-TTR, even at a high concentration (500 lM) (Fig. 3D). This result suggested that the effect of chromium on TTR might be mediated through increased T4 binding, not a direct induction of structural transition of TTR.

4. Discussion Our findings demonstrate that Cr3+ and T4 cooperatively suppressed amyloid fibril formation by stabilizing the tetrameric structure of WT-TTR and V30M-TTR (Figs. 2 and 3). A possible mechanism for this observation may be the increased T4 binding to TTR, induced by Cr3+, as determined by Scatchard analysis (Fig. 1A and Table 1). Previous studies demonstrated that the most dramatic reduction of TTR fibril formation occurs when both T4 binding sites are occupied by inhibitors that bind to TTR [15]. It is likely that Cr3+, by facilitating the binding of T4, contributes to the inhibition of amyloid fibril formation. Although we used here CrCl3, we ruled out the possibility that under the present conditions, Cl
itself suppressed TTR amyloid fibril formation, as reported previously [18] because in our preliminary screening of metal ions (Fe3+, Cr3+, Zn2+, Co2+, Cu2+, Cd2+, Mn2+, Ca2+), all the compounds used were chlorides, but among these, only Cr3+ had a suppressive effect (data not shown). Furthermore, we used here 500 lM Cl
concentration, which is so much lower than the reported effective Cl
concentration (0.5–1.8 M) [18]. We cannot yet, however, determine exactly how Cr3+ increases TTR binding capacity. It has been reported that

494

T. Sato et al. / FEBS Letters 580 (2006) 491–496

A

B

s-WT-TTR

100 75 50 25 0

0

10 50 Cr3+

C

0

10 50 ( M) Cr3+ + T4 (360 nM)

Fluorescence intensity

(%)

75 50 25 0

0 10 100 500 Cr3+

0 10 100 500 ( M) Cr3+ + T4 (360 nM)

s-WT- TTR

75 50 25 0 0.36

1

10

(%)

0

0.36

1

10

M)

diflunisal + Cr3+ (10 M)

diflunisal

Fluorescence intensity

100

100

0

D

r-WT- TTR

(%) Fluorescence intensity

Fluorescence intensity

(%)

s-V30M-TTR

100 75 50 25 0 0

0.36 T4

10

0

0.36

10

M)

T4 + Cr3+ (10 M)

Fig. 2. Cr3+ and T4 cooperatively suppressed amyloid fibril formation. (A, B) s-WT-TTR or r-WT-TTR was incubated for 5 days at 37 °C in PBS buffer (pH 4.0) with the indicated concentration of Cr3+ and 360 nM T4 and analyzed by thioflavine T-binding assay. Data shown are representative of 3 independent experiments. (C) Thioflavine T-binding assay was performed on 2.5 lg/mL serum WT-TTR in PBS buffer (pH 4.0) with the indicated concentration of diflunisal in the presence or absence of Cr3+. (D) V30M-TTR was incubated for 5 days at 37 °C in PBS buffer (pH 4.0) with the indicated concentration of T4 and 10 lM Cr3+and analyzed by thioflavine T-binding assay. Error bars show S.D. of results from three independent determinations.

electrostatic repulsion between two pairs of Lys15 residues in the TTR destabilized the tetrameric structure and that anion shielding of these charge repulsions can modulate tetramer stability and amyloid fibril formation [18]. It is known that T4 can bind to the two binding sites of TTR but binding of the second ligand is much weaker than that of the first molecule [19]. We have roughly determined the crystal structure of WT-TTR ˚ resolution. Our initial findings revealed mawith Cr3+ at 1.8 A jor peaks, which may signal the position of Cr3+, next to the Glu54 residue of TTR on the anomalous difference Fourier maps (data not shown). The T4-binding site is topologically close to Glu54, as crystallographic studies have demonstrated [20]. Cr3+ may electrostatically neutralize this area and increase the binding of T4 to TTR. Further crystallographic studies are needed to clarify this hypothesis.

Previous reports have shown that tetramer dissociation is the rate-limiting step for amyloidogenesis in vitro [21], providing a potential explanation for the observed correlation between the kinetic and thermodynamic stability of TTR tetramers and their amyloidogenicity [22]. It is evident from the calorimetric analyses that Cr3+ effected a change in the heat-induced dissociation pattern of s-WT-TTR at pH 7.4 and endowed heat-tolerance to TTR (Fig. 3A). WT-TTR and V30M-TTR, at pH 5.0 and 6.0, respectively were less stable than at pH 7.4, consistent with previous observations that lower pH renders the tetramer unstable [23], but Cr3+ stabilized these tetramers against thermal denaturation at low pH (Fig. 3B and C). Ligand binding is known to give an increment in the thermal stability of a protein due to an improved structural ordering and compactness of the protein [24]. It is therefore, possible that Cr3+, having

T. Sato et al. / FEBS Letters 580 (2006) 491–496

B

None (pH 7.4)

30

Cr3+ (pH 7.4)

20

10

Cr 3+ (pH 5.0)

10

0 80

0 80

90

100

110

None (pH 7.4) None (pH 5.0)

20

Cp (Kcal /K mol of monomer) 120

90

temperature (˚C)

C

120

110

D 5000 None (pH 7.4)

20

No metal Cr 3+

None (pH 6.0) Cr3+ (pH 6.0)

0

10

temperature (˚C)

250

240

110

100

230

-10000

90

80

220

0

200

-5000

190

Cp (Kcal /K mol of monomer)

100

temperature (˚C)

210

Cp (Kcal /K mol of monomer)

A

495

wavelength

Fig. 3. Effect of Cr3+ on the thermostability of TTR. Excess heat capacity of: (A) s-WT-TTR at pH 7.4, (B) s-WT-TTR at pH 5.0 or 7.4, (C) s-V30MTTR at pH 6.0 or 7.4, with or without Cr3+, was determined by differential scanning calorimetry on 25 lM TTR samples and calculated using Origine scientific plotting software. (D) The effect of Cr3+ on the secondary structure of s-WT-TTR was examined by far CD. TTR samples were incubated at 37 °C for 2 days in PBS buffer (pH 4.0) with or without Cr3+ (500 lM). Far CD was monitored with JASCO J-720. All spectra were recorded from 220 to 240 nm. Data shown are representative of 3 independent measurements.

Table 2 Thermodynamic parameters of serum WT-TTR and V30M-TTR

WT-TTR (pH 7.4) WT-TTR with Cr3+ (pH 7.4) WT-TTR (pH 5.0) WT-TTR with Cr3+ (pH 5.0) V30M-TTR (pH 6.0) V30M-TTR with Cr3+ (pH 6.0)

Tm

DHcal (kcal/mol)

DHv/DHcal

100.7 102.2 92.8 100.5 91.7 92.4

99.6 457 91.3 101 59.8 128

1.21 0.13 0.84 0.99 2.29 0.73

Acknowledgments: The authors’ work was supported by grants from the Amyloidosis Research Committee, the Pathogenesis and Therapy of Hereditary Neuropathy Research Committee, Surveys and Research on Specific Diseases, the Ministry of Health and Welfare of Japan, the Charitable Trust Clinical Pathology Research Foundation of Japan, and Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan.

References

Cr3+ concentration: 500 lM.

increased the binding of T4 to TTR would preserve the tetramer integrity and render a higher thermodynamic stability to TTR, thus, inhibiting amyloid fibril formation. The use of small molecules to stabilize proteins that easily misfold such as TTR has been well-documented [9,13,25]. We note here that Cr3+ can enhance the suppressive effect of 1 lM diflunisal on in vitro fibril formation (Fig. 2C). Our current findings that Cr3+, an essential trace element in humans with no known toxicity [26], has the potential of stabilizing tetrameric TTR, highlights the possible role of ions in TTR stability through shielding of electrostatic repulsions [18]. It would be interesting to verify these data in vivo. The present results and similar studies could contribute to the growing knowledge on the prevention of the onset of amyloidogenesis.

[1] Monaco, H.L., Rizzi, M. and Coda, A. (1995) Science 268, 1039– 1041. [2] Sousa, M.M. and Saraiva, M.J. (2003) Prog. Neurobiol. 71, 385– 400. [3] Benson, M.D. (1989) Trends Neurosci. 12, 88–92. [4] Quintas, A., Saraiva, M.J. and Brito, R.M. (1999) J. Biol. Chem. 274, 32943–32949. [5] Connors, L.H., Richardson, A.M., Theberge, R. and Costello, C.E. (2000) Amyloid 7, 54–69. [6] McCutchen, S.L., Colon, W. and Kelly, J.W. (1993) Biochemistry 32, 12119–12127. [7] McCutchen, S.L., Lai, Z., Miroy, G.J., Kelly, J.W. and Colon, W. (1995) Biochemistry 34, 13527–13536. [8] Sacchettini, J.C. and Kelly, J.W. (2002) Nat. Rev. Drug Discov. 1, 267–275. [9] Klabunde, T., Petrassi, H.M., Oza, V.B., Raman, P., Kelly, J.W. and Sacchettini, J.C. (2000) Nat. Struct. Biol. 7, 312–321. [10] Hornberg, A., Hultdin, U.W., Olofsson, A. and Sauer-Eriksson, A.E. (2005) Biochemistry 44, 9290–9299. [11] Ando, Y. et al. (1997) Biochim. Biophys. Acta 1362, 39–46.

496 [12] Shnyrov, V.L., Villar, E., Zhadan, G.G., Sanchez-Ruiz, J.M., Quintas, A., Saraiva, M.J. and Brito, R.M. (2000) Biophys. Chem. 88, 61–67. [13] Miroy, G.J., Lai, Z., Lashuel, H.A., Peterson, S.A., Strang, C. and Kelly, J.W. (1996) Proc. Natl. Acad. Sci. USA 93, 15051– 15056. [14] Bush, A.I. (2003) Trends Neurosci. 26, 207–214. [15] Hammarstrom, P., Wiseman, R.L., Powers, E.T. and Kelly, J.W. (2003) Science 299, 713–716. [16] Murrell, J.R., Schoner, R.G., Liepnieks, J.J., Rosen, H.N., Moses, A.C. and Benson, M.D. (1992) J. Biol. Chem. 267, 16595–16600. [17] Shrivastava, H.Y. and Nair, B.U. (2001) Biochem. Biophys. Res. Commun. 285, 915–920. [18] Hammarstrom, P., Jiang, X., Deechongkit, S. and Kelly, J.W. (2001) Biochemistry 40, 11453–11459. [19] Neumann, P., Cody, V. and Wojtczak, A. (2001) Acta Biochim. Pol. 48, 867–875.

T. Sato et al. / FEBS Letters 580 (2006) 491–496 [20] Wojtczak, A., Neumann, P. and Cody, V. (2001) Acta Crystallogr. D Biol. Crystallogr. 57, 957–967. [21] Jiang, X., Smith, C.S., Petrassi, H.M., Hammarstrom, P., White, J.T., Sacchettini, J.C. and Kelly, J.W. (2001) Biochemistry 40, 11442–11452. [22] Hammarstrom, P., Jiang, X., Hurshman, A.R., Powers, E.T. and Kelly, J.W. (2002) Proc. Natl. Acad. Sci. USA 99, 16427–16432, (Epub 2002 September 25). [23] Lai, Z., Colon, W. and Kelly, J.W. (1996) Biochemistry 35, 6470– 6482. [24] Gonzalez, M., Bagatolli, L.A., Echabe, I., Arrondo, J.L., Argarana, C.E., Cantor, C.R. and Fidelio, G.D. (1997) J. Biol. Chem. 272, 11288–11294. [25] Peterson, S.A., Klabunde, T., Lashuel, H.A., Purkey, H., Sacchettini, J.C. and Kelly, J.W. (1998) Proc. Natl. Acad. Sci. USA 95, 12956–12960. [26] Anderson, R.A. (1997) Regul. Toxicol. Pharmacol. 26, S35– S41.