- Email: [email protected]
Binding of Cu2+ to S-adenosyl-l -homocysteine hydrolase
JOURNAL OF
Inorganic Biochemistry Journal of Inorganic Biochemistry 98 (2004) 977–983 www.elsevier.com/locate/jinorgbio
Binding of Cu2þ to S-adenosyl-L -homocysteine hydrolase Yanjie Li a, Jiejin Chen a, Jing Liu a, Xiaoda Yang a
a,b,* ,
Kui Wang
a,b
Department of Chemical Biology, School of Pharmaceutical Sciences, Peking University Health Science Center, Beijing 100083, PR China National Research Laboratories of Natural and Biomimetic Drugs, Peking University Health Science Center, Beijing 100083, PR China
b
Received 14 October 2003; received in revised form 16 February 2004; accepted 18 February 2004 Available online 19 March 2004
Abstract S-Adenosylhomocysteine (AdoHcy) hydrolase regulates biomethylation and homocysteine metabolism. It has been proposed to be a copper binding protein playing an important role in copper transport and distribution. In the present work, the kinetics of binding and releasing of copper ions was studied using fluorescence method. The dissociation constant for copper ions with AdoHcy hydrolase was determined by fluorescence quenching titration and activity titration methods using ethylenediaminetetraacetic acid (EDTA), nitrilotriacetic acid (NTA), and glycine as competitive chelators. The experimental results showed that copper ions bind to AdoHcy hydrolase with a Kd of 1011 M. The association rate constant was determined to be 7 106 M1 s1 . The releasing of copper ions from the enzyme was found to be biphasic with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . It is suggested that copper ions do not bind to the substrate binding sites because the addition of adenine substrate did not compete with the binding of copper to AdoHcy hydrolase. Interestingly, it was observed that EDTA could bind to AdoHcy hydrolase with a dissociation constant of K1 ¼ 8:0 105 M and result in an increased affinity (Kd ¼ 1017 M) of binding of copper ions to the enzyme. Ó 2004 Elsevier Inc. All rights reserved. Keywords: Copper; S-Adenosyl-L -homocysteine hydrolase; Kinetic
1. Introduction S-Adenosylhomocysteine (AdoHcy) hydrolase (EC 3.3.1.1) catalyzes the reversible hydrolysis of S-adenosylhomocysteine (AdoHcy) to adenosine (Ado) and L homocysteine (Hcy) [1,2]. AdoHcy hydrolase plays an important role in regulating the levels of two important biological molecules, AdoHcy and Hcy. AdoHcy is the product of all adenosylmethionine (AdoMet)-dependent biological transmethylation, and strongly inhibits various AdoMet-dependent transmethylation enzymes. AdoHcy hydrolase contributes to the regulation of methyl transfer reactions in proliferating cells [3]. On the other hand, the activity of AdoHcy hydrolase may affect the level of Hcy. The abnormally elevated level of plasma Hcy (hyperhomocysteinemia) appears to be an important independent risk factor for cardiovascular disease [4] and possible risk factor for amyloid diseases *
Corresponding author. Tel.: +8601082801539; fax: +860106601 5584. E-mail address: [email protected] (X. Yang). 0162-0134/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2004.02.013
[5]. Consequently, AdoHcy hydrolase is an attractive target for antiviral [1,2], antiparasitic [1,2], antiarthritic, immunosuppressive and antitumor [6] drugs. The X-ray crystallographic structures of human and rat AdoHcy hydrolases have been determined [7,8] and the catalytic mechanisms of the enzyme have been clarified [9–13]. Although the structure and catalysis of AdoHcy hydrolase do not involve metal ions, it has been shown by Ettinger and co-workers [14–16] that this enzyme is a copper binding protein in vivo, which may play important roles in regulating tissue copper levels and the intracellular distribution of copper. Ettinger and co-workers [15] reported that AdoHcy hydrolase binds copper ions reversibly with a high affinity of Kd ¼ 1017 M in the presence of EDTA (ethylenediaminetetraacetic acid); however, the enzyme-bound copper quickly exchanges with copper isotope [14]. We have reported that copper inactivates the AdoHcy hydrolase with an apparent inhibition constant (Ki ) of 14 nmol L1 [17]. This inhibitory effect of copper ions might be related to: (i) copper ions increase the intracellular release of homocysteine [18] and (ii) hyperhomocysteinemia was
978
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
usually associated with increased serum copper level [19]. Therefore, elucidation of the mechanism for copper inactivation of AdoHcy hydrolase is important to better understand the relationship of Hcy metabolism with copper ions as well as other biological effects of copper ions and complexes, such as antiviral and antiarthritic activities. The mechanism of these activities has not been clarified. In the present work, the binding and releasing of copper ions, as the first step of the interaction of copper ions with AdoHcy hydrolase, were studied by fluorescence titration and stopped-flow method. The association rate constant was determined to be 7 106 M1 s1 . The releasing of copper ions from the enzyme was found to be biphasic with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . The dissociation constant is determined to be about 1011 –1012 M. An observed dissociation constant of 1017 M was obtained when using EDTA as the competitive chelator. This reaction was probably due to the binding of EDTA to the enzyme.
2. Materials and methods 2.1. Materials Escherichia coli JM109 transformed with a plasmid (pPROK-1) encoding human placental AdoHcy hydrolase was a gift from the University of Kansas, USA. Adenosine (Ado), homocysteine (Hcy), adenosylhomocysteine (AdoHcy), and 3-(N-morpholino) propanesulfonic acid (MOPS) were purchased from Sigma Co., USA. PD-10 Desalting Columns were obtained from Amersham Pharmacia Biotech. All other chemicals were molecular biological or analytical reagent grade. 2.2. Preparation of AdoHcy hydrolase AdoHcy hydrolase was expressed in E. coli JM109 according to the method described previously [20]. The JM109 bacterial cells were collected by centrifugation with 6000g at 4 °C for 15 min and washed with 0.9% NaCl. Then the cells were suspended in 25 mM tris(hydroxymethyl)aminomethane–HCl (Tris–HCl) buffer (containing 2 mM ethylenediaminetetraacetic acid (EDTA), pH 8.0) and then broken by sonication (400 W, 3 s 140, interval 3 s). Cell-free extract (CFE) was obtained by centrifuging the cell lysate at 15,000g for 60 min. Then the enzyme was isolated from CFE by Odiethylaminoethyl-cellulose (DEAE-cellulose) ion exchange chromatography (1.5 15 cm, pre-equilibrated and washed with 10 mM phosphate buffer containing 1 mM EDTA, pH 7.2 (PBE), eluted with 20 mM, pH 5.0 malic acid, at a flow rate 3.5 mL/min) and gel filtration with a SephacrylTM S-200 column (2.5 50 cm, eluted with 10 mM PBE, pH 7.2, at a flow rate 1.0 mL/min).
The protein concentration was determined by Bradford method [21]. The purity was assayed using SDS–polyacrylamide gel electrophoresis (SDS–PAGE). 2.3. Fluorescence quenching titration AdoHcy hydrolase samples were prepared at the concentration of 320 nM in 20 mM MOPS (pH 7.4) containing various concentrations of competitive chelators (including glycine, NTA, and EDTA) ranged from 40 to 1000 lM in the presence or absence of 100 lM adenine. Aliquots of 5 lL CuSO4 solution (0.1–10 mM) were added to the samples to a final concentration ranging from 0.1 lM to 1 mM. The quenching of intrinsic fluorescence of the enzyme was monitored on a Shimadzu RF-5301 PC fluorescence spectrometer at kex=em ¼ 280/340 nm. All experiments were performed at room temperature. 2.4. Activity titration In a total volume of 150 lL of 20 mM MOPS buffer (pH 7.4), 1 lg of AdoHcy hydrolase was incubated with 0.8 lM of copper ions and different concentrations of chelators (EDTA or NTA or glycine) at 37 °C for half hour. After incubation, 50 lL of AdoHcy (400 lM) containing 400 lM of 5,5-dithiobis-(2-nitrobenzoic acid) (DTNB) and 1 unit of adenosine deaminase were added to the enzyme solution. The residue activity of AdoHcy hydrolase was then assayed spectrophotometrically in the hydrolytic direction by measuring the rate of the product (Hcy) formed as the method as described previously [22]. The assays were performed at 37 °C on a Sunrise (Tecan) microplate reader and monitored at 412 nm. 2.5. Fluorescence stopped flow experiments Equal volume of AdoHcy Hydrolase (200 nM) and metal buffer (EDTA-Cu and NTA-Cu) at various concentrations of free copper were mixed at room temperature on a Cary Eclipse fluorescence spectrometer equipped with a SPF-20 stopped flow at kex=em ¼ 280/340 nm. Five individual runs were averaged for data analysis. 2.6. Release of copper ions from AdoHcy hydrolase At the concentration of 100 nM in 20 mM MOPS (pH 7.4), AdoHcy hydrolase samples were incubated with equal moles of copper ions at 37 °C for half hour. Then a chelator (EDTA or NTA) was added to a concentration of 50 lM. Aliquots were removed and the fluorescence of AdoHcy Hydrolase was measured on a Shimadzu RF-5301 PC fluorescence spectrometer at kex=em ¼ 280/340 nm.
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
F ¼ A1 ð1 ekð1Þ x Þ þ A2 ð1 ekð2Þ x Þ F0
2.7. Data analysis The apparent dissociation constant (KdðappÞ ) of copper ions with AdoHcy hydrolase was calculated by fitting the relative fluorescence quenching Q to a OneSite-Binding model (Eq. (1)) using a Microcal Origin program, in which F is the fluorescence intensity of the sample, F0 is the fluorescence intensity in absence of copper ions, Qmax is the maximum fluorescence 0 quenching and [M ] means the sum concentration of 0 non-enzyme-bound copper ions ([M ] ¼ [M] + [ML]): 0
Q¼
F0 F Qmax ½M ¼ : F0 ½M0 þ KdðappÞ
Kd ½L: KL
ð4Þ
in which kð1Þ and kð2Þ refer the first order rate constant of releasing of copper ions from the enzyme; A1 and A2 are amplitude parameters.
3. Results 3.1. Dissociation constants for binding of copper to AdoHcy hydrolase
ð1Þ
In the presence of competitive ligands, if the ligands do 0 not bind to the enzyme, then KdðappÞ ¼ [P][M ]/[MP] (P refers to the free enzyme and MP refers to copper–enzyme complex) and Eq. (2) could be deduced. Hence, the real dissociate constant Kd values were obtained by fitting the KdðappÞ values to Eq. (2) using a Microcal Origin program, where KL values are the dissociation constants of copper ions to the competitive ligands, which are 1.3 1010 [23] for NTA, 2.51 109 [24] for glycine, and 1.4 1016 [14] for EDTA, at pH 7.4 KdðappÞ ¼ Kd þ
979
ð2Þ
The dissociation constants (Kd ) of copper ions with AdoHcy hydrolase were also obtained from the activity titration. In a system containing copper, enzyme and ligand, the concentration of free copper ions is determined by the ratio of concentration of copper complex [ML] to the concentration of free ligand [L], that is [M] ¼ [ML]KL /[L]. The binding of copper ions to the enzyme results in loss of enzyme activity with a stoichiometry of 1:1 for copper to enzyme subunit [14,15,17]. The relative residual activity (Act/Act0 ) is proportional to the ratio of free enzyme concentration to the total enzyme concentration ([P]/Cp ). The concentration of copper–enzyme complex [MP] could be calculated from the total enzyme concentration (Cp ) and the loss of enzyme activity ([MP] ¼ Cp (Act0 ) Act)/Act0 , where Act is the activity of AdoHcy hydrolase in presence of copper and Act0 is the activity of AdoHcy hydrolase in absence of copper). Then, the free copper concentration [Cu2þ ] for each point of the titration assay could be calculated from [MP] and total concentrations of ligand and copper ions using a MatLab 6.0 program. Therefore, the dissociation constants were calculated by fitting the data to Eq. (3) using a Microcal Origin program ½Cu2þ Act ¼ Act0 1 : ð3Þ Kd þ ½Cu2þ Data for enzyme fluorescence recovery upon releasing of copper ions were fitted to biphasic exponential process [Eq. (4)] using a Microcal Origin program,
Fluorescence titration of AdoHcy hydrolase by Cu2þ (data not shown) indicated that each subunit of the enzyme had one high affinity binding site and multiple weak binding sites. Figs. 1–3 show quenching of fluorescence of AdoHcy hydrolase by Cu2þ in the presence of different concentrations of glycine, NTA, and EDTA. The dissociation constants were calculated as described in Section 2 and were found to be (1.26 0.08) 1011 M, (1.50 0.04) 1011 M, and (5.8 0.5) 1017 M using glycine, NTA, and EDTA as competitive chelators, respectively. In the presence of substrate adenine, the dissociation constants were found to be (2.1 0.3) 1011 (glycine), (3.0 0.3) 1011 (NTA), and (5.9 0.5) 1017 (EDTA), respectively. In the linear plotting of Kapp vs. concentrations of ligand, intercepts close to zero were obtained for NTA and glycine ligands (the inset of Figs. 2 and 3); however, a positive intercept (28 7 lM in the presence of adenine, 36 8 lM in the absence of adenine) (Fig. 1(c)) was obtained when using EDTA as the competitive ligand. The dissociation constants were also determined by activity titration method as described in Section 2 and the results were shown in Fig. 4. The dissociation constants were calculated to be (7.8 0.4) 1012 M, (6.9 0.2) 1012 M, and (5.1 0.5) 1016 M with glycine, NTA, and EDTA as competitive chelators. The results of dissociation constants obtained above are summarized in Table 1. 3.2. The association rate of copper ions to AdoHcy hydrolase AdoHcy hydrolase was mixed with copper ions at different concentration levels (in either Cu–NTA or Cu– EDTA buffer system). The binding of copper ions to the enzyme was monitored by fluorescence stopped-flow method. The apparent first order rate constants (k) were calculated by fitting the data to the first order exponential model. The intrinsic association rate constant (kon ) was estimated by plotting of the first order rate constants obtained by using NTA–Cu metal buffer system versus free copper ion concentrations (Fig. 5). kon is the slope of the fitting line and was found to be
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
80
60
40
20
0
10
(a)
0
1
10
10
2
10
3
10
100
80
0
200 400 600 800 1000
[NTA] µM
60
40
20
0
4
-1
10
Concentration of non-enzyme-bound copper ions,[M'] (µM)
0
10
1
10
2
10
3
10
4
10
Concentration of non-enzyme-bound copper ions,[M'] ( µM)
Fig. 2. Quenching of intrinsic fluorescence of AdoHcy hydrolase by copper ions in the presence of various concentrations of nitriloacetic acid (NTA). The concentrations of NTA are: 100 lM (), 200 lM (s), 500 lM (M), and 1 mM (O).
100
Relative Fluorescence Quenching, Q
120 100 80 60 40 20 0
Kd(app)(µM)
Relative Fluorescence Quenching, Q
100
80
60
90
20
0 -1
0
10
(b)
10
1
2
10
3
10
4
10
10
5
10
Concentration of non-enzyme-bound copper ions,[M'] (µM) 130 120 110
80 70 60
6
Kd(app)(µM)
40
Relative Fluorescence Quenching, Q
Relative Fluorescence Quenching, Q
980
5 4 3 2 1 0
0
200 400 600 800 1000
[Gly] (µM)
50 40 30 20 10 0
Kd(app) (µM)
100 -2
-1
0
1
2
3
90
10
80
Concentration of non-enzyme-bound copper ions (µM)
70
10
10
10
10
10
Fig. 3. Quenching of intrinsic fluorescence of AdoHcy hydrolase by copper ions in the presence of various concentrations of glycine. The concentrations of glycine are: 100 lM (), 200 lM (s), 500 lM (M), and 1 mM (O).
60 50 40 30
20
(c)
40
60
80
100
120
140
160
180
200
220
[EDTA] (µM)
Fig. 1. Quenching of intrinsic fluorescence of AdoHcy hydrolase by copper ions in the presence of various concentrations of EDTA. (a) in the absence of adenine, (b) in the presence of adenine. The concentrations of EDTA are: 40 lM (), 100 lM (s), 150 lM (M), and 200 lM (O). (c) Linear plotting of KdðappÞ is liner versus the concentration of EDTA in the presence () and absence (s) of adenine.
(7 4) 106 M1 s1 using NTA–Cu buffer system. The kon value could not be determined using the EDTA–Cu metal buffer due to the slow ligand exchange.
3.3. The rate of release of copper ions from AdoHcy hydrolase The release of copper ions appears when the ligand is added and that it is followed by the increase of intrinsic fluorescence of free enzyme. Fig. 6 shows the time course of recovery of enzyme fluorescence upon release of copper ions. The curve using EDTA as copper acceptors overlaps well with NTA as the acceptor (both shown in Fig. 6). Data fitting showed that release of copper ions from the enzyme is a biphasic process. The dissociation rate constants were found to be kð1Þ ¼ ð2:8 1:1Þ 103 s1 and kð2Þ ¼ ð1:7 0:6Þ 105 s1 .
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
Apparent First Order Constant, k (s-1)
120
% of Act / Act0
100
80
60
40
20
0 -20
10
-18
10
-16
10
-14
10
-12
10
-10
10
-8
10
-6
10
Concentration of free copper ions,[M] (M) Fig. 4. Plot of the activity of AdoHcy hydrolase versus the concentration of free copper ions when using EDTA (), NTA (s), and glycine (M) as competitive chelators, respectively. AdoHcy hydrolase (1 lg) was incubated in a total volume of 150 lL at 37 °C for half hour with 800 nM of copper ions and different concentrations of EDTA, NTA or glycine ranged from 40 to 1000 lmol/L and then assayed for the activity in the hydrolytic direction. The concentration of free copper ions for each point was calculated using a MatLab 6.0 program as described in Section 2.7.
981
0.70
0.65
0.60
0.55
0.50 -9
0.0
-8
5.0x10
1.0x10
-8
1.5x10
Concentration of Free Copper Ions (M) Fig. 5. Plot of apparent association rate constant versus the concentration of free copper ions. The time courses of fluorescence quenching of 100 nM AdoHcy hydrolase upon mixing with Cu-NTA buffer were monitored on the stopped flow spectrofluorometer at kex ¼ 280 nm. The apparent rate constant was estimated by fitting the data to the first order exponential function.
1.2
1.0
For determination of dissociation constants of metal ions from their high affinity sites at biological macromolecules, competitive chelators such as EDTA and glycine are ordinarily used to provide a suitable metal buffer system. As shown in Table 1, the dissociation constants obtained by fluorescence titration method were in good agreement with those obtained by activity titration method. When EDTA was used as a competitive ligand, the dissociation constant was also identical to that obtained by ultrafiltration reported previously by Ettinger and co-workers [14]. These results indicate that the three methods are equivalent in the determination of the binding constants of copper ions to AdoHcy hydrolase. In addition, these results are consist with the previously reported observation [14,17] that only one copper binds to each subunit of AdoHcy hydrolase and
0.8
F / F0
4. Discussion
0.6
0.4
0.2
0.0 0
20
40
60
80
Fig. 6. Time course of release of copper ions from AdoHcy hydrolase. AdoHcy hydrolase at the concentration of 100 nM was incubated with equal mole copper ions at 37 °C for half hour. Then a chelator, EDTA () or NTA (O), was added to a concentration of 50 lM. Aliquots were removed and the fluorescence of AdoHcy Hydrolase were measured at kex=em ¼ 280=340 nm.
Table 1 Summary of dissociation constants obtained using different methods and different competitive chelators Method of deter.
Kd (M) EDTA
NTA
Glycine
Fluorescence titration
(5.8 0.5) 10 (5.9 0.5) 1017a
(1.50 0.04) 10 (3.0 0.3) 1011a
(1.26 0.08) 1011 (2.1 0.3) 1011a
Activity titration
(5.1 0.5) 1016
(6.9 0.2) 1012
(7.8 0.4) 1012
Ultrafiltration
(5.8 0.4) 1017b
a b
17
100
Time (h)
11
The dissociation constants obtained in the presence of 100 lM adenine as described in Section 2. Data was from Ettinger and co-workers [14].
982
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
that this results in both the inhibition of enzyme activity and the quenching of intrinsic fluorescence of the enzyme. The dissociation constants in the presence of adenine (Table 1) were found to be almost the same as those in the absence of substrates. Since adenine is a competitive inhibitor of AdoHcy hydrolase and stabilizes AdoHcy hydrolase in the closed conformation [8,9], these results suggest that copper ions did not bind to the substrate binding sites of AdoHcy hydrolase and the global conformation of the enzyme did not affect binding of copper ions. It is interesting that there is a significant difference between the dissociation constant obtained using EDTA as competitive ligands and those using NTA and glycine as the ligands. With NTA or glycine as the ligand, the dissociation constants, Kd , were close to 1011 M; however, a Kd of approximately 1017 M was found when EDTA was the competitive ligand. Because the enzyme conformation was unlikely to affect copper binding, the association of the ligand to protein could be a possible reason for the significant difference in the binding constants. It has been shown that chelators with tetracarboxylate structure, such as EDTA, bind to protein with Kd at micromolar level [25,26]. In a system containing copper ions, EDTA ligands, and enzyme, a multiple equilibrium mechanism in Scheme 1. M represents copper ions, P is the enzyme and L is the ligand EDTA. If MPL is much more stable than MP, the apparent dissociation constant (KdðappÞ ) is KdðappÞ ¼
½M0 ½P0 ð½M þ ½MLÞð½P þ ½PLÞ : ¼ ½MP0 ½MP þ ½MPL
ð5Þ
Since the dissociation constants of copper ions to the competitive ligands are high enough (1.3 1010 for NTA, 2.51 109 for glycine, and 1.4 1016 for EDTA), it is reasonable to assume [M] [ML]. In addition, if L binds to the enzyme with relatively high affinity and thus K2 is small enough, then KdðappÞ would be proportional to the concentration of the ligand ([L]) as described by Eq. (6): KdðappÞ ¼
Kd0 K0 K1 þ d ½L: KL KL
P+M + L
ð6Þ
Kd
MP + L K2
K1
LP + M
Kd'
MPL
Scheme 1.
Plotting of KdðappÞ versus [L] gave a graph with its slope equal to Kd0 =KL with the intercept equal to K1 Kd0 =KL . This allows calculation of K1 and Kd0 to be (8 2) 105 M and (5.8 0.5) 1017 M, respectively (Fig. 1(c)). As shown in Fig. 1(c), plotting of KdðappÞ versus the concentrations of EDTA in presence/absence of substrate adenine gave positive intercept, indicating that the results fit better into the model described by Eq. (6) than Eq. (2). This result suggested that EDTA binds to AdoHcy hydrolase, which may account for the very highly observed binding affinity. Therefore, the intrinsic dissociation constant (Kd ) of copper–AdoHcy hydrolase complexes should be 1011 M, which is the value determined using NTA or glycine as ligand. From the thermodynamic cycle shown by Scheme 1, K2 (the dissociation constants of MPL) value was estimated to be K2 ¼ Kd0 K1 =Kd ¼ 1 109 M. The kinetic of binding/releasing of copper ions was in agreement with a Kd of 1011 M. Generally, Kd is associated with the rate constants of copper binding (kon ) and release (koff ) as: Kd ¼
koff : kon
ð7Þ
As described above, kon was determined to be 7 106 M1 s1 . Shown in Fig. 6, release of copper ions was a biphasic process with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . This result was not consistent with that the enzyme-bound copper undergoing quick exchanges with copper isotope in solution [14]. Because the slower step was rate determinant, the kð2Þ value could be assigned to the koff of copper release. Hence, the Kd could be estimated to be 2 1012 M, which was similar with those values obtained using NTA/glycine as competitive ligand. The biphasic process of copper ions releasing suggested that binding of copper ions undergo a transition process from less tight binding sites at the surface of the enzyme to the tight binding sites inside the enzyme structure. It has been reported that AdoHcy hydrolase is a homotetramer, with each subunit containing three domains: a catalytic/substrate binding domain (domain I), a NADþ binding domain (domain II), and a small carboxyl-terminal domain (domain C) [7,8]. A hinged domain movement results in large spatial rearrangements with respect to catalytic domain and NADþ binding domain. As copper ions are unlikely to bind to domain I described above, the copper binding sites are possibly in the rigid core structure consisting of domain II and III. Therefore, the rate of Cu2þ getting into its sites will be determined by the frequency of opening– closing of the enzyme structure. The frequency of domain movement has been determined to be 25 ns [27], which suggests a kon value of less than 4 107 M1 s1 . This is in good agreement with the experimental result of 7 106 M1 s1 .
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
Our results have shown that copper ions reversibly bind to AdoHcy hydrolase and inhibit the enzyme catalytic activity [17], suggesting copper might be a regulator of the enzyme. Since AdoHcy hydrolase is crucial in regulating biological methylation and Hcy metabolism, the actions of copper ions might provide a novel link to the antiviral or antiarthritic activities of copper compounds and the elevated plasma Hcy level in copper-deficient animals [28]. In summary, the kinetic of binding of copper ions to AdoHcy hydrolase was studied by fluorescence titration and stopped-flow method using EDTA, NTA, and glycine as competitive chelators. The association rate constant was determined to be 7 106 M1 s1 . The release of copper ions from the enzyme was biphasic with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . The dissociation constant was 1011 –1012 M. Copper ions do not bind to the substrate binding site because addition of substrate adenine did not affect binding of copper to AdoHcy hydrolase. In addition, binding of EDTA to the enzyme resulted in an observed dissociation constant of 5.8 1017 M, suggesting that ligands with tetracarboxylate could affect the binding affinity of metal ions to protein.
Acknowledgements We are grateful to Dr. Borchardt at the University of Kansas for providing the JM109 expressing AdoHcy hydrolase. This project was sponsored by CMB and SRF for ROCS/SEM.
References [1] D. Yin, X. Yang, R.T. Borchart, C. Yuan, in: Biochemistry: Applying Chemical Principles to the Understanding and Treatment of Diseases, John Wiley & Sons Inc, New York, 2000, pp. 41–47. [2] M. Turner, X. Yang, D. Yin, R.T. Borchardt, Cell Biochem. Biophys. 33 (2000) 101–125.
983
[3] P.S. Backlund, D. Carotti Jr., G.L. Cantoni, Eur. J. Biochem. 160 (1986) 245–251. [4] J.M. Moller, G.L. Nielsen, S. Ekelund, E.B. Schmidt, J. Dyerberg, Thromb. Res. (1997) 333–335. [5] R.W. Anthony, X. Huang, F.J. Michael, J.B. Colin, J. Neurochem. 76 (2001) 1509–1520. [6] J. Suarez, V. Chagoya de Sanchez, Int. J. Biochem. Cell Biol. 29 (1997) 1279–1284. [7] M.A. Turner, C.S. Yuan, R.T. Borchardt, M.S. Hershfield, G.D. Smith, P.L. Howell, Nat. Struct. Biol. 5 (1998) 369–376. [8] Y. Hu, J. Komoto, Y. Huang, Biochemistry 38 (1999) 8323– 8333. [9] J.L. Palmer, R.H. Abeles, J. Biol. Chem. 254 (1979) 1217– 1226. [10] R.H. Abeles, A.H. Tashjian, Biochem. Biophys. Res. Commun. 95 (1980) 612–617. [11] D.J. Porter, F.L. Boyd, J. Biol. Chem. 266 (1991) 21616–21625. [12] D.J. Porter, F.L. Boyd, J. Biol. Chem. 267 (1992) 3205–3213. [13] D.J. Porter, J. Biol. Chem. 268 (1993) 66–73. [14] K.E. Bethin, T.R. Cimato, M.J. Ettinger, J. Biol. Chem. 270 (1995) 20703–20711. [15] K.E. Bethin, N. Petrovic, M.J. Ettinger, J. Biol. Chem. 270 (1995) 20698–20702. [16] N. Petrovic, A. Comi, M.J. Ettinger, J. Biol. Chem. 271 (1996) 28335–28340. [17] J. Chen, Q. Liu, X. Yang, K. Wang, Chin. Sci. Bull. 47 (2002) 1176–1179. [18] B. Hultberg, A. Andersson, A. Isaksson, Toxicology 117 (1997) 89–97. [19] M.A. Mansoor, C. Bergmark, S.J. Haswell, I.F. Savage, P.H. Evans, R.K. Berge, A.M. Svardal, O. Kristensen, Clin. Chem. 46 (2000) 385–391. [20] C.S. Yuan, J. Yeh, S. Liu, R.T. Borchardt, J. Biol. Chem. 268 (1993) 17030–17037. [21] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254. [22] C. Yuan, D.B. Ault-Riche, R.T. Borchardt, J. Biol. Chem. 271 (1996) 28009–28016. [23] W.-B. Chang, K.-A. Li, Concise Handbook for Analytical Chemistry, Peking University Publishing House, Beijing, 1981. [24] A. Ringbom, in: Complexation in Analytical Chemistry, Interscience Publishers, New York, 1963. [25] A.C. Ribou, J. Vigo, P. Viallet, J.M. Salmon, Biophys. Chem. 81 (1999) 179–189. [26] P. Bryan, P. Alexander, S. Strausberg, F. Schwarz, W. Lan, G. Gilliland, D.T. Gallagher, Biochemistry 31 (1992) 4937–4945. [27] D. Yin, X. Yang, Y. Hu, K. Kuczera, R.L. Schowen, R.T. Borchardt, T.C. Squier, Biochemistry 39 (2000) 9811–9818. [28] T. Tsunenobu, H.H. Kyu, M. Yasuharu, E.J. Kelley, L.K. Carl, Biochim. Biophys. Acta 1427 (1999) 351–356.
Inorganic Biochemistry Journal of Inorganic Biochemistry 98 (2004) 977–983 www.elsevier.com/locate/jinorgbio
Binding of Cu2þ to S-adenosyl-L -homocysteine hydrolase Yanjie Li a, Jiejin Chen a, Jing Liu a, Xiaoda Yang a
a,b,* ,
Kui Wang
a,b
Department of Chemical Biology, School of Pharmaceutical Sciences, Peking University Health Science Center, Beijing 100083, PR China National Research Laboratories of Natural and Biomimetic Drugs, Peking University Health Science Center, Beijing 100083, PR China
b
Received 14 October 2003; received in revised form 16 February 2004; accepted 18 February 2004 Available online 19 March 2004
Abstract S-Adenosylhomocysteine (AdoHcy) hydrolase regulates biomethylation and homocysteine metabolism. It has been proposed to be a copper binding protein playing an important role in copper transport and distribution. In the present work, the kinetics of binding and releasing of copper ions was studied using fluorescence method. The dissociation constant for copper ions with AdoHcy hydrolase was determined by fluorescence quenching titration and activity titration methods using ethylenediaminetetraacetic acid (EDTA), nitrilotriacetic acid (NTA), and glycine as competitive chelators. The experimental results showed that copper ions bind to AdoHcy hydrolase with a Kd of 1011 M. The association rate constant was determined to be 7 106 M1 s1 . The releasing of copper ions from the enzyme was found to be biphasic with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . It is suggested that copper ions do not bind to the substrate binding sites because the addition of adenine substrate did not compete with the binding of copper to AdoHcy hydrolase. Interestingly, it was observed that EDTA could bind to AdoHcy hydrolase with a dissociation constant of K1 ¼ 8:0 105 M and result in an increased affinity (Kd ¼ 1017 M) of binding of copper ions to the enzyme. Ó 2004 Elsevier Inc. All rights reserved. Keywords: Copper; S-Adenosyl-L -homocysteine hydrolase; Kinetic
1. Introduction S-Adenosylhomocysteine (AdoHcy) hydrolase (EC 3.3.1.1) catalyzes the reversible hydrolysis of S-adenosylhomocysteine (AdoHcy) to adenosine (Ado) and L homocysteine (Hcy) [1,2]. AdoHcy hydrolase plays an important role in regulating the levels of two important biological molecules, AdoHcy and Hcy. AdoHcy is the product of all adenosylmethionine (AdoMet)-dependent biological transmethylation, and strongly inhibits various AdoMet-dependent transmethylation enzymes. AdoHcy hydrolase contributes to the regulation of methyl transfer reactions in proliferating cells [3]. On the other hand, the activity of AdoHcy hydrolase may affect the level of Hcy. The abnormally elevated level of plasma Hcy (hyperhomocysteinemia) appears to be an important independent risk factor for cardiovascular disease [4] and possible risk factor for amyloid diseases *
Corresponding author. Tel.: +8601082801539; fax: +860106601 5584. E-mail address: [email protected] (X. Yang). 0162-0134/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2004.02.013
[5]. Consequently, AdoHcy hydrolase is an attractive target for antiviral [1,2], antiparasitic [1,2], antiarthritic, immunosuppressive and antitumor [6] drugs. The X-ray crystallographic structures of human and rat AdoHcy hydrolases have been determined [7,8] and the catalytic mechanisms of the enzyme have been clarified [9–13]. Although the structure and catalysis of AdoHcy hydrolase do not involve metal ions, it has been shown by Ettinger and co-workers [14–16] that this enzyme is a copper binding protein in vivo, which may play important roles in regulating tissue copper levels and the intracellular distribution of copper. Ettinger and co-workers [15] reported that AdoHcy hydrolase binds copper ions reversibly with a high affinity of Kd ¼ 1017 M in the presence of EDTA (ethylenediaminetetraacetic acid); however, the enzyme-bound copper quickly exchanges with copper isotope [14]. We have reported that copper inactivates the AdoHcy hydrolase with an apparent inhibition constant (Ki ) of 14 nmol L1 [17]. This inhibitory effect of copper ions might be related to: (i) copper ions increase the intracellular release of homocysteine [18] and (ii) hyperhomocysteinemia was
978
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
usually associated with increased serum copper level [19]. Therefore, elucidation of the mechanism for copper inactivation of AdoHcy hydrolase is important to better understand the relationship of Hcy metabolism with copper ions as well as other biological effects of copper ions and complexes, such as antiviral and antiarthritic activities. The mechanism of these activities has not been clarified. In the present work, the binding and releasing of copper ions, as the first step of the interaction of copper ions with AdoHcy hydrolase, were studied by fluorescence titration and stopped-flow method. The association rate constant was determined to be 7 106 M1 s1 . The releasing of copper ions from the enzyme was found to be biphasic with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . The dissociation constant is determined to be about 1011 –1012 M. An observed dissociation constant of 1017 M was obtained when using EDTA as the competitive chelator. This reaction was probably due to the binding of EDTA to the enzyme.
2. Materials and methods 2.1. Materials Escherichia coli JM109 transformed with a plasmid (pPROK-1) encoding human placental AdoHcy hydrolase was a gift from the University of Kansas, USA. Adenosine (Ado), homocysteine (Hcy), adenosylhomocysteine (AdoHcy), and 3-(N-morpholino) propanesulfonic acid (MOPS) were purchased from Sigma Co., USA. PD-10 Desalting Columns were obtained from Amersham Pharmacia Biotech. All other chemicals were molecular biological or analytical reagent grade. 2.2. Preparation of AdoHcy hydrolase AdoHcy hydrolase was expressed in E. coli JM109 according to the method described previously [20]. The JM109 bacterial cells were collected by centrifugation with 6000g at 4 °C for 15 min and washed with 0.9% NaCl. Then the cells were suspended in 25 mM tris(hydroxymethyl)aminomethane–HCl (Tris–HCl) buffer (containing 2 mM ethylenediaminetetraacetic acid (EDTA), pH 8.0) and then broken by sonication (400 W, 3 s 140, interval 3 s). Cell-free extract (CFE) was obtained by centrifuging the cell lysate at 15,000g for 60 min. Then the enzyme was isolated from CFE by Odiethylaminoethyl-cellulose (DEAE-cellulose) ion exchange chromatography (1.5 15 cm, pre-equilibrated and washed with 10 mM phosphate buffer containing 1 mM EDTA, pH 7.2 (PBE), eluted with 20 mM, pH 5.0 malic acid, at a flow rate 3.5 mL/min) and gel filtration with a SephacrylTM S-200 column (2.5 50 cm, eluted with 10 mM PBE, pH 7.2, at a flow rate 1.0 mL/min).
The protein concentration was determined by Bradford method [21]. The purity was assayed using SDS–polyacrylamide gel electrophoresis (SDS–PAGE). 2.3. Fluorescence quenching titration AdoHcy hydrolase samples were prepared at the concentration of 320 nM in 20 mM MOPS (pH 7.4) containing various concentrations of competitive chelators (including glycine, NTA, and EDTA) ranged from 40 to 1000 lM in the presence or absence of 100 lM adenine. Aliquots of 5 lL CuSO4 solution (0.1–10 mM) were added to the samples to a final concentration ranging from 0.1 lM to 1 mM. The quenching of intrinsic fluorescence of the enzyme was monitored on a Shimadzu RF-5301 PC fluorescence spectrometer at kex=em ¼ 280/340 nm. All experiments were performed at room temperature. 2.4. Activity titration In a total volume of 150 lL of 20 mM MOPS buffer (pH 7.4), 1 lg of AdoHcy hydrolase was incubated with 0.8 lM of copper ions and different concentrations of chelators (EDTA or NTA or glycine) at 37 °C for half hour. After incubation, 50 lL of AdoHcy (400 lM) containing 400 lM of 5,5-dithiobis-(2-nitrobenzoic acid) (DTNB) and 1 unit of adenosine deaminase were added to the enzyme solution. The residue activity of AdoHcy hydrolase was then assayed spectrophotometrically in the hydrolytic direction by measuring the rate of the product (Hcy) formed as the method as described previously [22]. The assays were performed at 37 °C on a Sunrise (Tecan) microplate reader and monitored at 412 nm. 2.5. Fluorescence stopped flow experiments Equal volume of AdoHcy Hydrolase (200 nM) and metal buffer (EDTA-Cu and NTA-Cu) at various concentrations of free copper were mixed at room temperature on a Cary Eclipse fluorescence spectrometer equipped with a SPF-20 stopped flow at kex=em ¼ 280/340 nm. Five individual runs were averaged for data analysis. 2.6. Release of copper ions from AdoHcy hydrolase At the concentration of 100 nM in 20 mM MOPS (pH 7.4), AdoHcy hydrolase samples were incubated with equal moles of copper ions at 37 °C for half hour. Then a chelator (EDTA or NTA) was added to a concentration of 50 lM. Aliquots were removed and the fluorescence of AdoHcy Hydrolase was measured on a Shimadzu RF-5301 PC fluorescence spectrometer at kex=em ¼ 280/340 nm.
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
F ¼ A1 ð1 ekð1Þ x Þ þ A2 ð1 ekð2Þ x Þ F0
2.7. Data analysis The apparent dissociation constant (KdðappÞ ) of copper ions with AdoHcy hydrolase was calculated by fitting the relative fluorescence quenching Q to a OneSite-Binding model (Eq. (1)) using a Microcal Origin program, in which F is the fluorescence intensity of the sample, F0 is the fluorescence intensity in absence of copper ions, Qmax is the maximum fluorescence 0 quenching and [M ] means the sum concentration of 0 non-enzyme-bound copper ions ([M ] ¼ [M] + [ML]): 0
Q¼
F0 F Qmax ½M ¼ : F0 ½M0 þ KdðappÞ
Kd ½L: KL
ð4Þ
in which kð1Þ and kð2Þ refer the first order rate constant of releasing of copper ions from the enzyme; A1 and A2 are amplitude parameters.
3. Results 3.1. Dissociation constants for binding of copper to AdoHcy hydrolase
ð1Þ
In the presence of competitive ligands, if the ligands do 0 not bind to the enzyme, then KdðappÞ ¼ [P][M ]/[MP] (P refers to the free enzyme and MP refers to copper–enzyme complex) and Eq. (2) could be deduced. Hence, the real dissociate constant Kd values were obtained by fitting the KdðappÞ values to Eq. (2) using a Microcal Origin program, where KL values are the dissociation constants of copper ions to the competitive ligands, which are 1.3 1010 [23] for NTA, 2.51 109 [24] for glycine, and 1.4 1016 [14] for EDTA, at pH 7.4 KdðappÞ ¼ Kd þ
979
ð2Þ
The dissociation constants (Kd ) of copper ions with AdoHcy hydrolase were also obtained from the activity titration. In a system containing copper, enzyme and ligand, the concentration of free copper ions is determined by the ratio of concentration of copper complex [ML] to the concentration of free ligand [L], that is [M] ¼ [ML]KL /[L]. The binding of copper ions to the enzyme results in loss of enzyme activity with a stoichiometry of 1:1 for copper to enzyme subunit [14,15,17]. The relative residual activity (Act/Act0 ) is proportional to the ratio of free enzyme concentration to the total enzyme concentration ([P]/Cp ). The concentration of copper–enzyme complex [MP] could be calculated from the total enzyme concentration (Cp ) and the loss of enzyme activity ([MP] ¼ Cp (Act0 ) Act)/Act0 , where Act is the activity of AdoHcy hydrolase in presence of copper and Act0 is the activity of AdoHcy hydrolase in absence of copper). Then, the free copper concentration [Cu2þ ] for each point of the titration assay could be calculated from [MP] and total concentrations of ligand and copper ions using a MatLab 6.0 program. Therefore, the dissociation constants were calculated by fitting the data to Eq. (3) using a Microcal Origin program ½Cu2þ Act ¼ Act0 1 : ð3Þ Kd þ ½Cu2þ Data for enzyme fluorescence recovery upon releasing of copper ions were fitted to biphasic exponential process [Eq. (4)] using a Microcal Origin program,
Fluorescence titration of AdoHcy hydrolase by Cu2þ (data not shown) indicated that each subunit of the enzyme had one high affinity binding site and multiple weak binding sites. Figs. 1–3 show quenching of fluorescence of AdoHcy hydrolase by Cu2þ in the presence of different concentrations of glycine, NTA, and EDTA. The dissociation constants were calculated as described in Section 2 and were found to be (1.26 0.08) 1011 M, (1.50 0.04) 1011 M, and (5.8 0.5) 1017 M using glycine, NTA, and EDTA as competitive chelators, respectively. In the presence of substrate adenine, the dissociation constants were found to be (2.1 0.3) 1011 (glycine), (3.0 0.3) 1011 (NTA), and (5.9 0.5) 1017 (EDTA), respectively. In the linear plotting of Kapp vs. concentrations of ligand, intercepts close to zero were obtained for NTA and glycine ligands (the inset of Figs. 2 and 3); however, a positive intercept (28 7 lM in the presence of adenine, 36 8 lM in the absence of adenine) (Fig. 1(c)) was obtained when using EDTA as the competitive ligand. The dissociation constants were also determined by activity titration method as described in Section 2 and the results were shown in Fig. 4. The dissociation constants were calculated to be (7.8 0.4) 1012 M, (6.9 0.2) 1012 M, and (5.1 0.5) 1016 M with glycine, NTA, and EDTA as competitive chelators. The results of dissociation constants obtained above are summarized in Table 1. 3.2. The association rate of copper ions to AdoHcy hydrolase AdoHcy hydrolase was mixed with copper ions at different concentration levels (in either Cu–NTA or Cu– EDTA buffer system). The binding of copper ions to the enzyme was monitored by fluorescence stopped-flow method. The apparent first order rate constants (k) were calculated by fitting the data to the first order exponential model. The intrinsic association rate constant (kon ) was estimated by plotting of the first order rate constants obtained by using NTA–Cu metal buffer system versus free copper ion concentrations (Fig. 5). kon is the slope of the fitting line and was found to be
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
80
60
40
20
0
10
(a)
0
1
10
10
2
10
3
10
100
80
0
200 400 600 800 1000
[NTA] µM
60
40
20
0
4
-1
10
Concentration of non-enzyme-bound copper ions,[M'] (µM)
0
10
1
10
2
10
3
10
4
10
Concentration of non-enzyme-bound copper ions,[M'] ( µM)
Fig. 2. Quenching of intrinsic fluorescence of AdoHcy hydrolase by copper ions in the presence of various concentrations of nitriloacetic acid (NTA). The concentrations of NTA are: 100 lM (), 200 lM (s), 500 lM (M), and 1 mM (O).
100
Relative Fluorescence Quenching, Q
120 100 80 60 40 20 0
Kd(app)(µM)
Relative Fluorescence Quenching, Q
100
80
60
90
20
0 -1
0
10
(b)
10
1
2
10
3
10
4
10
10
5
10
Concentration of non-enzyme-bound copper ions,[M'] (µM) 130 120 110
80 70 60
6
Kd(app)(µM)
40
Relative Fluorescence Quenching, Q
Relative Fluorescence Quenching, Q
980
5 4 3 2 1 0
0
200 400 600 800 1000
[Gly] (µM)
50 40 30 20 10 0
Kd(app) (µM)
100 -2
-1
0
1
2
3
90
10
80
Concentration of non-enzyme-bound copper ions (µM)
70
10
10
10
10
10
Fig. 3. Quenching of intrinsic fluorescence of AdoHcy hydrolase by copper ions in the presence of various concentrations of glycine. The concentrations of glycine are: 100 lM (), 200 lM (s), 500 lM (M), and 1 mM (O).
60 50 40 30
20
(c)
40
60
80
100
120
140
160
180
200
220
[EDTA] (µM)
Fig. 1. Quenching of intrinsic fluorescence of AdoHcy hydrolase by copper ions in the presence of various concentrations of EDTA. (a) in the absence of adenine, (b) in the presence of adenine. The concentrations of EDTA are: 40 lM (), 100 lM (s), 150 lM (M), and 200 lM (O). (c) Linear plotting of KdðappÞ is liner versus the concentration of EDTA in the presence () and absence (s) of adenine.
(7 4) 106 M1 s1 using NTA–Cu buffer system. The kon value could not be determined using the EDTA–Cu metal buffer due to the slow ligand exchange.
3.3. The rate of release of copper ions from AdoHcy hydrolase The release of copper ions appears when the ligand is added and that it is followed by the increase of intrinsic fluorescence of free enzyme. Fig. 6 shows the time course of recovery of enzyme fluorescence upon release of copper ions. The curve using EDTA as copper acceptors overlaps well with NTA as the acceptor (both shown in Fig. 6). Data fitting showed that release of copper ions from the enzyme is a biphasic process. The dissociation rate constants were found to be kð1Þ ¼ ð2:8 1:1Þ 103 s1 and kð2Þ ¼ ð1:7 0:6Þ 105 s1 .
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
Apparent First Order Constant, k (s-1)
120
% of Act / Act0
100
80
60
40
20
0 -20
10
-18
10
-16
10
-14
10
-12
10
-10
10
-8
10
-6
10
Concentration of free copper ions,[M] (M) Fig. 4. Plot of the activity of AdoHcy hydrolase versus the concentration of free copper ions when using EDTA (), NTA (s), and glycine (M) as competitive chelators, respectively. AdoHcy hydrolase (1 lg) was incubated in a total volume of 150 lL at 37 °C for half hour with 800 nM of copper ions and different concentrations of EDTA, NTA or glycine ranged from 40 to 1000 lmol/L and then assayed for the activity in the hydrolytic direction. The concentration of free copper ions for each point was calculated using a MatLab 6.0 program as described in Section 2.7.
981
0.70
0.65
0.60
0.55
0.50 -9
0.0
-8
5.0x10
1.0x10
-8
1.5x10
Concentration of Free Copper Ions (M) Fig. 5. Plot of apparent association rate constant versus the concentration of free copper ions. The time courses of fluorescence quenching of 100 nM AdoHcy hydrolase upon mixing with Cu-NTA buffer were monitored on the stopped flow spectrofluorometer at kex ¼ 280 nm. The apparent rate constant was estimated by fitting the data to the first order exponential function.
1.2
1.0
For determination of dissociation constants of metal ions from their high affinity sites at biological macromolecules, competitive chelators such as EDTA and glycine are ordinarily used to provide a suitable metal buffer system. As shown in Table 1, the dissociation constants obtained by fluorescence titration method were in good agreement with those obtained by activity titration method. When EDTA was used as a competitive ligand, the dissociation constant was also identical to that obtained by ultrafiltration reported previously by Ettinger and co-workers [14]. These results indicate that the three methods are equivalent in the determination of the binding constants of copper ions to AdoHcy hydrolase. In addition, these results are consist with the previously reported observation [14,17] that only one copper binds to each subunit of AdoHcy hydrolase and
0.8
F / F0
4. Discussion
0.6
0.4
0.2
0.0 0
20
40
60
80
Fig. 6. Time course of release of copper ions from AdoHcy hydrolase. AdoHcy hydrolase at the concentration of 100 nM was incubated with equal mole copper ions at 37 °C for half hour. Then a chelator, EDTA () or NTA (O), was added to a concentration of 50 lM. Aliquots were removed and the fluorescence of AdoHcy Hydrolase were measured at kex=em ¼ 280=340 nm.
Table 1 Summary of dissociation constants obtained using different methods and different competitive chelators Method of deter.
Kd (M) EDTA
NTA
Glycine
Fluorescence titration
(5.8 0.5) 10 (5.9 0.5) 1017a
(1.50 0.04) 10 (3.0 0.3) 1011a
(1.26 0.08) 1011 (2.1 0.3) 1011a
Activity titration
(5.1 0.5) 1016
(6.9 0.2) 1012
(7.8 0.4) 1012
Ultrafiltration
(5.8 0.4) 1017b
a b
17
100
Time (h)
11
The dissociation constants obtained in the presence of 100 lM adenine as described in Section 2. Data was from Ettinger and co-workers [14].
982
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
that this results in both the inhibition of enzyme activity and the quenching of intrinsic fluorescence of the enzyme. The dissociation constants in the presence of adenine (Table 1) were found to be almost the same as those in the absence of substrates. Since adenine is a competitive inhibitor of AdoHcy hydrolase and stabilizes AdoHcy hydrolase in the closed conformation [8,9], these results suggest that copper ions did not bind to the substrate binding sites of AdoHcy hydrolase and the global conformation of the enzyme did not affect binding of copper ions. It is interesting that there is a significant difference between the dissociation constant obtained using EDTA as competitive ligands and those using NTA and glycine as the ligands. With NTA or glycine as the ligand, the dissociation constants, Kd , were close to 1011 M; however, a Kd of approximately 1017 M was found when EDTA was the competitive ligand. Because the enzyme conformation was unlikely to affect copper binding, the association of the ligand to protein could be a possible reason for the significant difference in the binding constants. It has been shown that chelators with tetracarboxylate structure, such as EDTA, bind to protein with Kd at micromolar level [25,26]. In a system containing copper ions, EDTA ligands, and enzyme, a multiple equilibrium mechanism in Scheme 1. M represents copper ions, P is the enzyme and L is the ligand EDTA. If MPL is much more stable than MP, the apparent dissociation constant (KdðappÞ ) is KdðappÞ ¼
½M0 ½P0 ð½M þ ½MLÞð½P þ ½PLÞ : ¼ ½MP0 ½MP þ ½MPL
ð5Þ
Since the dissociation constants of copper ions to the competitive ligands are high enough (1.3 1010 for NTA, 2.51 109 for glycine, and 1.4 1016 for EDTA), it is reasonable to assume [M] [ML]. In addition, if L binds to the enzyme with relatively high affinity and thus K2 is small enough, then KdðappÞ would be proportional to the concentration of the ligand ([L]) as described by Eq. (6): KdðappÞ ¼
Kd0 K0 K1 þ d ½L: KL KL
P+M + L
ð6Þ
Kd
MP + L K2
K1
LP + M
Kd'
MPL
Scheme 1.
Plotting of KdðappÞ versus [L] gave a graph with its slope equal to Kd0 =KL with the intercept equal to K1 Kd0 =KL . This allows calculation of K1 and Kd0 to be (8 2) 105 M and (5.8 0.5) 1017 M, respectively (Fig. 1(c)). As shown in Fig. 1(c), plotting of KdðappÞ versus the concentrations of EDTA in presence/absence of substrate adenine gave positive intercept, indicating that the results fit better into the model described by Eq. (6) than Eq. (2). This result suggested that EDTA binds to AdoHcy hydrolase, which may account for the very highly observed binding affinity. Therefore, the intrinsic dissociation constant (Kd ) of copper–AdoHcy hydrolase complexes should be 1011 M, which is the value determined using NTA or glycine as ligand. From the thermodynamic cycle shown by Scheme 1, K2 (the dissociation constants of MPL) value was estimated to be K2 ¼ Kd0 K1 =Kd ¼ 1 109 M. The kinetic of binding/releasing of copper ions was in agreement with a Kd of 1011 M. Generally, Kd is associated with the rate constants of copper binding (kon ) and release (koff ) as: Kd ¼
koff : kon
ð7Þ
As described above, kon was determined to be 7 106 M1 s1 . Shown in Fig. 6, release of copper ions was a biphasic process with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . This result was not consistent with that the enzyme-bound copper undergoing quick exchanges with copper isotope in solution [14]. Because the slower step was rate determinant, the kð2Þ value could be assigned to the koff of copper release. Hence, the Kd could be estimated to be 2 1012 M, which was similar with those values obtained using NTA/glycine as competitive ligand. The biphasic process of copper ions releasing suggested that binding of copper ions undergo a transition process from less tight binding sites at the surface of the enzyme to the tight binding sites inside the enzyme structure. It has been reported that AdoHcy hydrolase is a homotetramer, with each subunit containing three domains: a catalytic/substrate binding domain (domain I), a NADþ binding domain (domain II), and a small carboxyl-terminal domain (domain C) [7,8]. A hinged domain movement results in large spatial rearrangements with respect to catalytic domain and NADþ binding domain. As copper ions are unlikely to bind to domain I described above, the copper binding sites are possibly in the rigid core structure consisting of domain II and III. Therefore, the rate of Cu2þ getting into its sites will be determined by the frequency of opening– closing of the enzyme structure. The frequency of domain movement has been determined to be 25 ns [27], which suggests a kon value of less than 4 107 M1 s1 . This is in good agreement with the experimental result of 7 106 M1 s1 .
Y. Li et al. / Journal of Inorganic Biochemistry 98 (2004) 977–983
Our results have shown that copper ions reversibly bind to AdoHcy hydrolase and inhibit the enzyme catalytic activity [17], suggesting copper might be a regulator of the enzyme. Since AdoHcy hydrolase is crucial in regulating biological methylation and Hcy metabolism, the actions of copper ions might provide a novel link to the antiviral or antiarthritic activities of copper compounds and the elevated plasma Hcy level in copper-deficient animals [28]. In summary, the kinetic of binding of copper ions to AdoHcy hydrolase was studied by fluorescence titration and stopped-flow method using EDTA, NTA, and glycine as competitive chelators. The association rate constant was determined to be 7 106 M1 s1 . The release of copper ions from the enzyme was biphasic with a kð1Þ of 2.8 103 s1 and kð2Þ of 1.7 105 s1 . The dissociation constant was 1011 –1012 M. Copper ions do not bind to the substrate binding site because addition of substrate adenine did not affect binding of copper to AdoHcy hydrolase. In addition, binding of EDTA to the enzyme resulted in an observed dissociation constant of 5.8 1017 M, suggesting that ligands with tetracarboxylate could affect the binding affinity of metal ions to protein.
Acknowledgements We are grateful to Dr. Borchardt at the University of Kansas for providing the JM109 expressing AdoHcy hydrolase. This project was sponsored by CMB and SRF for ROCS/SEM.
References [1] D. Yin, X. Yang, R.T. Borchart, C. Yuan, in: Biochemistry: Applying Chemical Principles to the Understanding and Treatment of Diseases, John Wiley & Sons Inc, New York, 2000, pp. 41–47. [2] M. Turner, X. Yang, D. Yin, R.T. Borchardt, Cell Biochem. Biophys. 33 (2000) 101–125.
983
[3] P.S. Backlund, D. Carotti Jr., G.L. Cantoni, Eur. J. Biochem. 160 (1986) 245–251. [4] J.M. Moller, G.L. Nielsen, S. Ekelund, E.B. Schmidt, J. Dyerberg, Thromb. Res. (1997) 333–335. [5] R.W. Anthony, X. Huang, F.J. Michael, J.B. Colin, J. Neurochem. 76 (2001) 1509–1520. [6] J. Suarez, V. Chagoya de Sanchez, Int. J. Biochem. Cell Biol. 29 (1997) 1279–1284. [7] M.A. Turner, C.S. Yuan, R.T. Borchardt, M.S. Hershfield, G.D. Smith, P.L. Howell, Nat. Struct. Biol. 5 (1998) 369–376. [8] Y. Hu, J. Komoto, Y. Huang, Biochemistry 38 (1999) 8323– 8333. [9] J.L. Palmer, R.H. Abeles, J. Biol. Chem. 254 (1979) 1217– 1226. [10] R.H. Abeles, A.H. Tashjian, Biochem. Biophys. Res. Commun. 95 (1980) 612–617. [11] D.J. Porter, F.L. Boyd, J. Biol. Chem. 266 (1991) 21616–21625. [12] D.J. Porter, F.L. Boyd, J. Biol. Chem. 267 (1992) 3205–3213. [13] D.J. Porter, J. Biol. Chem. 268 (1993) 66–73. [14] K.E. Bethin, T.R. Cimato, M.J. Ettinger, J. Biol. Chem. 270 (1995) 20703–20711. [15] K.E. Bethin, N. Petrovic, M.J. Ettinger, J. Biol. Chem. 270 (1995) 20698–20702. [16] N. Petrovic, A. Comi, M.J. Ettinger, J. Biol. Chem. 271 (1996) 28335–28340. [17] J. Chen, Q. Liu, X. Yang, K. Wang, Chin. Sci. Bull. 47 (2002) 1176–1179. [18] B. Hultberg, A. Andersson, A. Isaksson, Toxicology 117 (1997) 89–97. [19] M.A. Mansoor, C. Bergmark, S.J. Haswell, I.F. Savage, P.H. Evans, R.K. Berge, A.M. Svardal, O. Kristensen, Clin. Chem. 46 (2000) 385–391. [20] C.S. Yuan, J. Yeh, S. Liu, R.T. Borchardt, J. Biol. Chem. 268 (1993) 17030–17037. [21] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254. [22] C. Yuan, D.B. Ault-Riche, R.T. Borchardt, J. Biol. Chem. 271 (1996) 28009–28016. [23] W.-B. Chang, K.-A. Li, Concise Handbook for Analytical Chemistry, Peking University Publishing House, Beijing, 1981. [24] A. Ringbom, in: Complexation in Analytical Chemistry, Interscience Publishers, New York, 1963. [25] A.C. Ribou, J. Vigo, P. Viallet, J.M. Salmon, Biophys. Chem. 81 (1999) 179–189. [26] P. Bryan, P. Alexander, S. Strausberg, F. Schwarz, W. Lan, G. Gilliland, D.T. Gallagher, Biochemistry 31 (1992) 4937–4945. [27] D. Yin, X. Yang, Y. Hu, K. Kuczera, R.L. Schowen, R.T. Borchardt, T.C. Squier, Biochemistry 39 (2000) 9811–9818. [28] T. Tsunenobu, H.H. Kyu, M. Yasuharu, E.J. Kelley, L.K. Carl, Biochim. Biophys. Acta 1427 (1999) 351–356.