Thermal Effects on the Activity and Structural Conformation of

Unless otherwise stated, catechol dioxygenase activity was determined spectrophotometrically using a Beckman DU-640 equipped with a thermojacketed ...
1 downloads 0 Views 3MB Size
Download PDF
J. Phys. Chem. B 2010, 114, 987–992

987

Thermal Effects on the Activity and Structural Conformation of Catechol 2,3-Dioxygenase from Pseudomonas putida SH1 Shir-Ly Huang,† Yuan-Chang Hsu,† Chun-Ming Wu,‡ Jeffrey W. Lynn,§ and Wen-Hsien Li*,‡ Department of Life Sciences and Center for Biotechnology & Biomedical Engineering, National Central UniVersity, Jhongli 32001, Taiwan, Department of Physics and Center for Neutron Beam Applications, National Central UniVersity, Jhongli 32001, Taiwan, and NIST Center for Neutron Research, NIST, Gaithersburg, Maryland 20899 ReceiVed: August 9, 2009; ReVised Manuscript ReceiVed: NoVember 19, 2009

A bacterium, Pseudomonas putida SH1, which can catabolize phenol, naphthalene, or cresol as the sole carbon and energy source, was isolated from a petroleum-contaminated site in Taiwan. The catechol 2,3-dioxygenase (C23O) was purified from this bacterial strain when grown on naphthalene as the sole carbon and energy source. The enzyme is composed of four identical subunits with a native molecular weight of 128 ( 5 kD. Small-angle neutron scattering (SANS) techniques were employed to study the thermal effects on the structural conformation of this enzyme in solution. The SANS measurements revealed distinct changes in the size of the enzyme between 50 and 80 °C, and the size was not restored during the subsequent cooling. The enzyme started to denature at 55 °C, and the structure was destroyed by the time the temperature reached 80 °C, at which the enzyme had become more than twice the original size. The optimal catalytic temperature of the enzyme was at 50 °C. The half-life of the activity at this temperature was 45 min. The enzyme activity increases starting from 25 °C and reaches its maximum at 50 °C, below which no obvious change in the size of the enzyme is found. Noticeable enlargement of the enzyme is revealed when the enzymatic activity starts to fall. By combination of SANS measurement and biochemical properties of the enzyme, this study demonstrates the correlation of enzyme size in solution and catalytic activity upon a heat treatment. In addition, for a protein composed of multiple subunits, the shape of the enzyme and the dissociation of the enzyme subunits in a thermal cycle were also demonstrated by SANS methodology. 1. Introduction Microorganisms are known to play important roles in the degradation and detoxification of toxic environmental pollutants. Bacterial degradation of aromatic compounds under aerobic conditions often involves conversion of the initial substrates into common intermediates, such as catechol or its derivatives. The catecholic intermediates are subject to attack by aromatic ringcleavage dioxygenase, which can be classified into two types according to the site of ring fission.1 Intradiol dioxygenases cleave ortho to the hydroxyl substituents and depend typically on nonheme Fe3+. Members of the intradiol dioxygenases usually consist of a different number of R and β subunits. Of these, the second type, extradiol dioxygenases, cleaves meta to the hydroxyl substituents and depends typically on nonheme Fe2+. Enzymes that belong to this group usually contain various numbers of identical subunits. Evolutionary studies have shown that these two classes of aromatic ring-cleavage dioxygenase genes have arisen from different ancestors.2 Among the extradiol dioxygenases, 2,3-dihydroxylbiphenyl 1,2-dioxygenases (2,3-DHBD) from Pseudomonas cepacia LB4003 and from Pseudomonas sp. strain KKS1024 and, more recently, catechol 2,3-dioxygenase (XylE) from P. putida mt25 have all been crystallized. Their three-dimensional structures * To whom correspondence should be addressed. E-mail: whli@ phy.ncu.edu.tw. Tel.: +886 3 4227151ext 65317. Fax: +886 3 2807404. † Department of Life Sciences and Center for Biotechnology & Biomedical Engineering, National Central University. ‡ Department of Physics and Center for Neutron Beam Applications, National Central University. § NIST Center for Neutron Research.

have been determined. All these structures reveal a conserved core region in the active site comprising three Fe2+ ligands (two histidine residues and one glutamate), one tyrosine, and two histidine residues. These results suggest that extradiol dioxygenases employ a common mechanism to recognize the aromatic ring of the substrate, catechol, and to activate the oxygen.5 The amino acid residues in the active site involved in the substrate binding and iron coordination are shown to be conserved among the 38 enzymes.6 A bacterial strain, P. putida SH1, was isolated from a soil sample of a petroleum-contaminated site. It has been demonstrated that this bacterium uses naphthalene, phenol, o-cresol, m-cresol, or p-cresol as its sole source of carbon and energy to grow.7 When the bacterium grows in a naphthalene-containing medium, a catechol 2,3-dioxygenase (C23O) is induced and is designated as C23O(SH1). This enzyme belongs to the family of extradiol dioxygenases. In this report, we have characterized the basic biochemical properties of C23O(SH1) and investigated the thermal effects on the activity and the structural conformation of C23O(SH1) in solution. Small-angle neutron scattering (SANS) techniques8-12 were used to study the C23O(SH1) structure in solution, the thermal effects on the protein size, and its correlation to enzyme activity. In addition, C23O is a homotetrameric enzyme. The general advantages of oligomerization for many enzymes are cooperativity, formation, and stabilization of the active sites of the monomeric subunit or maintaining the stability of the tertiary structure. This study was also anticipated to observe the thermal effect on the stability of individual subunits and the interactions among the subunits of an oligomeric enzyme.

10.1021/jp9078579  2010 American Chemical Society Published on Web 12/28/2009

988

J. Phys. Chem. B, Vol. 114, No. 2, 2010

Huang et al.

Figure 1. Amino acid sequence alignment of catechol 2,3-dioxygenases from P. putida SH1 together with that from P. putida mt-2 for a direct comparison. The blue symbols indicate the ligands for iron. The framed amino acid residues are involved in the enzyme active sites of both enzymes.

2. Materials and Methods 2.1. Bacteria and Media. The strain SH1 was grown at 30 °C in a minimal salts basal (MSB) medium13 containing 0.05% naphthalene as the sole carbon source. Bacterial growth was followed by turbidity determination by measuring the absorbance at 600 nm. Large-scale bacterial culturing was obtained from 10 L fermentation (BIOSTAT B, B. Bran Biotech International) with dissolved oxygen at 50%. Cells were harvested at the latelog phase, and the pellets were stored at -75 °C. 2.2. Enzyme Purification. The bacterial cell extract was prepared by suspending frozen cells in a 4-fold volume of 20 mM Tris-HCl, with a pH 8.0 containing 10% acetone and 20 µM phenylmethylsulfonyl fluorides at 4 °C. The suspended cells were then disrupted by sonication using a sonicor ultrasonic processor (Sonicor ultrasonic processor, Sonicor Instrument Co.). The broken cells were centrifuged at 104g for 15 min. The supernatant was further centrifuged at 105g for 1 h to obtain the crude cell extract. The C23O(SH1) was purified by a heat treatment at 50 °C for 30 min followed by centrifugation at 104g for 20 min to remove the precipitates. The supernatant was then loaded into a DEAE-Sepharose (Pharmacia, Sweden) anionic exchange column (5 × 2.8 cm) equilibrated with 20 mM Tris-HCl (pH 8.0) containing 10% acetone. It was then further eluted with a linear gradient of the same buffer solution containing 0.15-0.3 M KCl. The active fractions were pooled and concentrated by ultrafiltration. (NH4)2SO4 powder was added to 0.4 M before being loaded into a Phenyl-Superose hydrophobic interaction column (1 × 10 cm; Pharmacia, Sweden), which was equilibrated with 20 mM Tris-HCl (pH 8.0) containing 10% acetone and 0.4 M (NH4)2SO4. The elution was carried out with a linear gradient, 0.4-0 M (NH4)2SO4. The

active fractions were collected at 0.2-0.05 M (NH4)2SO4 and concentrated by ultrafiltration. The concentrated sample was then loaded onto a Sephacryl S-200 (Pharmacia, Sweden) column (1.6 × 90 cm) and eluted by 20 mM Tris-HCl (pH 8.0) containing 10% acetone. The active fractions were collected, concentrated, and stored at -75 °C. All the purification procedures were performed at 4 °C with a fast protein liquid chromatography system (Pharmacia). 2.3. Enzyme Activity Assay. Unless otherwise stated, catechol dioxygenase activity was determined spectrophotometrically using a Beckman DU-640 equipped with a thermojacketed cuvette holder and a circulating water bath. The activity was monitored by increasing absorbance at 375 nm upon addition of the enzyme to a 1 mL assay mixture containing 1 mM catechol (ε ) 42 800 cm-1 M-1) in 20 mM KH2PO4, pH 7.5, at 50 °C (modified from the recipe of Nakai et al.).14 The enzyme reaction was started by adding an appropriate amount of enzyme. One unit of enzyme activity was defined as the formation of 1 µmol of 2-hydroxymuconic semialdehyde per minute. The protein content was determined as described by Bradford15 using bovine serum albumin as a standard. 2.4. Molecular Weight and Subunit Determination. Enzyme purity was determined by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE).16 The molecular weight of the enzyme subunit was determined by SDSPAGE with a molecular weight standard kit (Bio-Rad Laboratories, CA), and the theoretical molecular weight was calculated from the amino acid sequence (Figure 1). A Sephacryl S-200 filtration was used to measure the molecular weight of the native C23O(SH1), where the High Molecular Weight Calibration Kit from Pharmacia (Sweden) was used as the standard.

Conformation of Catechol 2,3-Dioxygenase

J. Phys. Chem. B, Vol. 114, No. 2, 2010 989

TABLE 1: Purification of Catechol 2,3-Dioxygenase (C23O) from P. putida SH1a purification step crude extract 50 °C treatment DEAE-Sepharose Pheny-Superose Sephacryl S-200

protein activity specific activity recovery purification (mg) (unit)b (unit/mg) (%) (fold) 1658 1008 77 18.6 9.0

25 116 25 187 18 822 11 009 7220

15 25 246 592 802

100 100 75 44 29

1 1.6 16.3 39.0 52.7

a The activity of C23O(SH1) was assayed in a reaction mixture of 1 mL of 20 mM NaH2PO4, pH 7.5, containing 1 mM catechol at 50°C. b 1 unit ) formation of 1 µmol of 2-hydroxymuconic semialdehyde per minute.

2.5. Cofactor Analysis. The iron content of the pure enzyme was determined by the o-phenathroline method.14 The pure enzyme (0.5 mg in 1 mL) was mixed with 0.5 mL of 1 N HCl and heat-denatured at 80 °C for 10 min. An amount of 0.5 mL of the supernatant was then mixed with 175 µL of 0.2 M potassium diphthalate, 20 µL of 0.1 M o-phenathrolin, 295 µL of 20 mM Tris-HCl, with a of pH 8.0 containing 10% acetone, and 10 µL of 3 N NaOH. The concentrations of Fe2+ and Fe3+ were determined spectrophotometrically at A512nm and A396nm with (NH4)2Fe(SO4)2 · 6H2O and (NH4)Fe(SO4)2 · 12H2O as standard (0-10 mg/L), respectively. 2.6. Thermal Stability. An amount of 26 µg of C23O(SH1) was incubated in 1 mL of 20 mM Tris-HCl, with a pH of 8.0 and 10% acetone, at various temperatures and time intervals. For each activity assay, 5 µL of the mixture was taken from the main bath, and the analysis was performed under optimal catalytic conditions for the individual purified enzyme. The assay was performed in 1 mL of 20 mM KH2PO4, pH 7.5, at 50 °C. 2.7. Neutron Scattering. SANS measurements were performed at the NIST Center for Neutron Research using the 8 m SANS instrument on NG-1, with a mechanical velocity selector to extract neutrons of λ ) 6 Å. A 3He two-dimensional positionsensitive detector having a spatial resolution of ∼10 mm, with an evacuated flight path of 3.5 m between the sample and the detector, was employed. To take advantage of the scattering contrast of neutrons from hydrogen and deuterium, the purified C23O enzymes were dialyzed to deuterated 20 mM Tris-HCl, with a pH 8.0 buffer solution containing 10% acetone. For these measurements, the sample of concentration 8 mg/mL and the buffer were loaded into identical quartz cells that were mounted in a liquid-circulating temperature controller, operated between 5 and 80 °C. The Guinier plots17,18 and PRIMUS program19 were used to extract the radius of gyration Rg of the enzyme in solution. The scattering profile was calculated employing the GNOM program.20 The initial calculation was performed by taking the maximum size of the enzyme to be 2Rg, followed by a series of calculations assuming smaller sizes for the enzyme until the calculated scattering profile agrees well with the observed one, judged by the least-squares criterion. The overall shape of the enzyme was then reconstructed employing the modeling program GASBOR,21 based on the parameters generated by GNOM. Uncertainties where indicated are statistical in nature and represent one standard deviation. 3. Results and Discussion 3.1. Biochemical Properties. The enzyme was purified in homogeneity following the process listed in Table 1. Its biochemical properties are summarized and compared to XylE from P. putida mt-2 (Table 2). Accordingly, the native molecular weight of C23O(SH1) is 128 ( 5 kD (n ) 3), with a subunit

TABLE 2: Biochemical and Amino Acid Sequence Properties of Catechol 2,3-Dioxygenases from P. putida SH1 and P. putida mt-2 property molecular weight (kD) native enzyme (gel filtration) subunit (SDS-PAGE) subunit structure optimal catalytic temperatureb (°C) optimal catalytic pHc cofactor specific activity (unit)d,e amino acid sequence homology (%) identity similarity

C23O(SH1)

XylEa

128 ( 5 (n ) 3) 39 ( 1 (n ) 3) R4 50 7.5-8.0 Fe2+ 802

140 35 R4 50 6.5 Fe2+ 320

100 100

94 97

a

Nakai, et al. J. Biol. Chem. 1983. b C23O(SH1), in 20 mM Na2HPO4 at pH 7.5; C23O(mt-2) in 50 mM Na2HPO4/NaH2PO4 at pH 6.5. c C23O(SH1), in 20 mM Na2HPO4 at 50 °C; XylE, in 50 mM Na2HPO4/NaH2PO4 at 25 °C. d 1 unit ) formation of 1 µmol of 2-hydroxymuconic semialdehyde per minute. e C23O(SH1), in a reaction mixture of 1 mL of 20 mM NaH2PO4, pH 7.5, containing 1 mM catechol at 50 °C; XylE, in a reaction mixture of 50 mM Na2HPO4/NaH2PO4 at pH 6.5 at 25 °C.

molecular weight of 39 ( 1 kD (n ) 3) determined by SDSPAGE and the theoretical molecular weight calculated from amino acid residues (Figure 1), suggesting the enzyme is composed of four subunits (Table 2). The optimal temperature for the enzyme activity in 20 mM Na2PO4 with pH 8.5 was found to be at 50 °C, where the optimal pH was found to be 7.5-8.0, that is, alkalinic. Spectrophotometric analysis showed that the enzyme contained Fe2+ as a cofactor, as has been found in other active extradiol dioxygenases.2,3,5 Figure 1 shows the amino acid sequence alignment of C23O from P. putida SH1 and XylE (also a C23O) from P. putida mt-2. Both C23Os from Pseudomonas putida mt-2 and from strain SH1 contain 307 amino acid residues. They are highly conserved with 94% identity and 97% similarity at the amino acid sequence level. In particular, the coordinated residues for Fe2+, shown by the blue symbols in Figure 1, are the same. The residues in the frames represent the residues involved in the active sites which are also conserved for both enzymes. The 3D structure of the XylE from strain mt-2 is available.5 (The PDB code of C23O is 1MPY.) The close identification of amino acid sequences provides a good template for the structure analysis of C23O(SH1) by homology modeling. 3.2. Enzyme Stability. The growth temperature range for strain SH1 is around 20-40 °C. The optimal growth temperature range is 25-30 °C. Since the enzyme has an in vitro optimal catalytic temperature at 50 °C, a heat treatment range for monitor enzyme stability was set from 30 to 60 °C. The residual activity of C23O(SH1) as a function of the heating duration at various temperatures is shown in Figure 2(a), where the origin is set to be the time when the heat treatment started. As expected, activity was reduced by the thermal treatment, with the reduction rate depending strongly on the temperature. Interestingly, the loss of enzyme activity caused by thermal treatment can be expressed by an exponentially decaying function with a fractional exponent. The solid curves in Figure 2(a) indicate the fits of the data to the expression R(t) ) exp{ -(t/t0)R}, where t0 and R are the fitting parameters. Accordingly, t0 is the time constant of the thermal treatment and represents the duration needed to reduce the activity of the enzyme to 1/e (∼37%) of its initial capability; the exponent R signifies the decay rate of the enzymatic activity when t0 is used to scale the time axis, which is closely related to the mechanism of enzyme action. As listed in Figure 2(a), t0 was found to be strongly dependent on the

990

J. Phys. Chem. B, Vol. 114, No. 2, 2010

Huang et al.

Figure 3. ln I vs Q2 plots at 5 °C of native C23O(SH1). The solid line indicates the linear fit of the low Q data, giving Rg ) 36 ( 4 Å.

Figure 2. (a) Time dependence of the residue activity of C23O(SH1) at various temperatures. The solid curves indicate the fits of the data to the expression shown in the plot. (b) Temperature dependence of the half-life of the activity, where the solid curve indicates the results of the fit of the data to an exponentially decaying function shown in the plot.

temperature, indicating that the thermal treatment did affect the catalytic capability of C23O(SH1), as expected. In terms of the half-life, t1/2 ) t0(ln 2/R) ) t0(ln 2/0.581) ) 1.193t0. The variation of t1/2 with temperature is plotted in Figure 2(b), indicating a half-life of ∼7 h for the enzyme at 30 °C. Apparently, above 50 °C activity is effectively suppressed, as indicated by a reduction of the half-life up to ∼50 min. We note that the half-life for the XylE from the P. putida mt-2 strain is 74 min at 55 °C,22 and that of NahH from the G7 strain is very heat-labile, becoming completely inactivated in less than 1 min when incubated at 50 °C.23 The temperature dependence of the half-life can be described by an exponentially decaying function, as indicated by the solid curve shown in Figure 2(b), which marks the results of a fit of the data to t1/2(T) ) A exp(-T/ T0), where A and T0 are the fitting parameters. A value of T0 ) 10.7 ( 0.5 °C was obtained from the fit, indicating that the half-life of the enzyme would be reduced by a factor of e ) 2.718 whenever the temperature was raised by 10.7 °C. In addition, the same value of R ) 0.581(1) was obtained for the exponent at all temperatures studied, showing that it is a good physical parameter for describing the phenomenon and implying the mechanism of enzyme action was not altered by temperature change. 3.3. Structural Conformation. SANS measurements were used to study the structural conformation of C23O(SH1) in solution. Scattering data were collected at various temperatures, during a warming cycle from 5 to 80 °C followed by a cooling cycle back to 5 °C. The intensities of scattered neutrons from the sample, buffer, and empty cell were all measured at each temperature. The measuring time at each temperature is ∼2 h. The scattering intensities from the enzyme were isolated from the background by subtracting the transmission-corrected buffer pattern from the transmission-corrected sample pattern. The resulting two-dimensional data arrays were then radically

Figure 4. Variation of Rg with temperature throughout the thermal cycle together with the activity observed at several representative temperatures, for a direct comparison, where 1 unit in activity indicates formation of 1 µmol of 2-hydroxymuconic semialdehyde per minute.

averaged to obtain I(Q), the intensity variation with wave vector transfer Q ) 4π sin θ/λ, where 2θ is the scattering angle. It is known that the Guinier ln I(Q2) plot can be used to extract the apparent radius of gyration Rg of the molecules in dilute solution.17,18 Figure 3 shows the Guinier ln I(Q2) plot of native C23O(SH1) at 5 °C, where the solid line indicates the linear fits of the low Q data. A value of Rg ) 36 ( 4 Å is thus obtained. The variations of Rg of the enzyme with temperature throughout the thermal cycle, together with the activity observed at several representative temperatures, are illustrated in Figure 4. The enzyme activity increases starting from 25 °C and reaches its maximum at 50 °C, below which no obvious change in the size of the enzyme is found. Noticeable enlargement of the enzyme is revealed when the enzymatic activity starts to fall. The size of the enzyme increases significantly above 55 °C. The transition occurs between 50 and 80 °C, at which point the enzyme is 2.6 times its native size. Note that biochemical data reveal a significant reduction in activity and an extremely short half-life of enzymes in this transition regime, as can be seen in Figure 2. In constructing the shape of the enzyme, attention has been paid to search the possible conformations that the data may have indicated, especially for the patterns taken after the enzymes have been heated to above 50 °C. No structural symmetry or shape of the enzyme was proposed during the shape reconstruction. Eight separated modeling runs, which cover enzyme sizes from 40% larger to 20% smaller than the proposed Rg, were performed for each pattern taken before thermal cycle to above 50 °C. No significant differences in the shapes generated from

Conformation of Catechol 2,3-Dioxygenase

J. Phys. Chem. B, Vol. 114, No. 2, 2010 991

Figure 6. Variation of scattering intensity with wave vector transfer obtained at 80 °C. The solid curve indicates the fit of the data to the scattering profile of the mass distribution shown in the inset, revealing that the enzyme has become denatured which results in the disordered formation.

Figure 5. (a) Variation of neutron scattering intensity with wave vector transfer obtained at 5 °C. The solid curve indicates the fit of the data to the scattering profile of the mass distribution shown in the inset, revealing a compact elliptical structure for the native C23O(SH1). (b) Structural conformation obtained for C23O from P. putida mt-2.

these runs may be identified. The log I(Q) of the enzyme at 5 °C is shown in Figure 5(a). The solid curve in Figure 5(a) indicates the fit of the data to the scattering profile of the mass distribution shown in the inset to Figure 5(a), which reveals a compact elliptical structure with the three principle-axis parameters La:Lb:Lc ≈ 1:1:1.5 for the native C23O(SH1). The size of the C23O(SH1) enzyme in solution, as measured by the SANS method, is almost identical to that obtained from the crystalline structure of XylE, determined by X-ray diffraction as shown in Figure 5(b). Note that the radius of gyration Rg of an elliptical object with its principle-axis parameters La, Lb, and Lc is Rg2 ) (1/5)(La2 + Lb2 + L2c ). Recalling the Rg ) 36 ( 4 Å obtained from the Guinier fit, the average size of the longest axis of native C23O(SH1) in solution at 5 °C is Dc ) 2Lc ) 117 ( 6 Å, which is close to what was observed5 for XylE. The average sizes along the other two principle axes of the elliptical assembly are Da ) Db ) 0.67Dc ) 78 Å. No obvious change in scattering profile was found in the subsequent thermal cycle up to 50 °C, indicating that the size and shape of the enzyme remain essentially unaltered. We note that the enzyme stayed at each temperature studied for ∼2 h and that it took 12 h for the temperature to reach 50 °C. The biochemical data presented in Figure 2, on the other hand, show a significant reduction in the half-life of the activity when the enzyme was incubated at 50 °C. These observations indicate that the loss of enzyme activity is mainly at the microenvironmental level, likely in the active sites, but causes no significant change in its conformation. The loss in catalytic activity during heat treatment may be due to the removal of the catalytic iron, as has been observed24,25 in XylE. Sixteen separated modeling runs, which cover enzyme sizes from 50% larger to 30% smaller than the proposed Rg, were performed for each pattern taken after the enzymes have been heated to above 50 °C. Unacceptable shapes that consist of a lot of separated segments were frequently obtained when

Figure 7. Structural conformations of C23O(SH1) at several representative temperatures throughout the thermal cycle. The subunits begin to fall apart at 52.5 °C. At 62.5 °C, subunit A is still bound to subunit B, and subunit C is still bound to subunit D; however, subunit A+B and subunit C+D are barely connected. At 72.5 °C, the enzyme has become denatured which results in a much more disordered formation. Heat treatment up to 80 °C initiates an irreversible reaction in C23O(SH1).

assuming the maximum enzyme size to be larger than 1.2Rg, whereas assuming a maximum enzyme size that is smaller than 0.85Rg gives poor fit to the observed scattering profile. No significant differences may be identified among those shapes otherwise obtained. Figure 6 shows the log I(Q) obtained at 80 °C, where the solid curve indicates the fit of the data to the scattering profile of the mass distribution shown in the inset. The heat treatment results in the enzyme being much more disordered in structure. We believe that the enlargement of the enzyme size is due to the partial unfolding of the tetrameric subunits of the enzyme, which results in a very loosely linked structure, as has been observed26 in β-lactoglobulin A. Note that the scattering intensities in the low-Q regime increase as the temperature is increased, indicating that the degree of aggregation among the enzymes may have increased upon heating. The structural conformations of C23O(SH1) at several representative temperatures throughout the thermal cycle are illustrated in Figure 7. There are four identical subunits (marked A, B, C, and D, same as shown in Figure 5) in each C23O(SH1) enzyme. It appears that the subunits begin to fall apart at 52.5 °C, at which point subunits A and B as well as subunits C and D are still closely bound, although separation between subunits (A and B) and subunits (C and D) already forms a dumbbell shape. There are consequently two distinguishable interactions between the subunits of C23O(SH1): a face-to-face interaction,

992

J. Phys. Chem. B, Vol. 114, No. 2, 2010

such as that between subunit A and subunit B (same as C and D subunits), and a bottom-to-top interaction, such as that between subunit A+B and subunit C+D. At 62.5 °C, the size of the enzyme has been enlarged by ∼75%, and the A+B and C+D subunits become barely connected; however, A still binds to B and C still binds to D. It is then clear that the face-to-face interactions between two subunits are significantly stronger than their bottom-to-top interactions. We believe that it is the hydrogen bonds between the projecting loop of one subunit and the β-sheet at the N-terminal of another subunit that hold the two subunits (A and B; C and D) tightly together face-to-face. Note that Okuta et al. had found27 that a 4S5β-sheet of one of the monomers is critical for the formation of a quaternary structure. Information on the structure-function relationship of XylE from strain mt-2 has also suggested5 that the projecting loop and the 4S5β-sheet are crucial for subunit interaction. In XylE from strain mt-2, there are three residues, namely, Asp133, Val-134, and Asn-135, that form hydrogen bonds between the subunits.23 In C23O(SH1), by homology modeling, the faceto-face interactions between two subunits are primarily through the three hydrogen bonds from a projecting loop, namely, from Val134 of subunit A, as an example, to Thr201 of subunit B, from Asn135 (subunit A) to Leu283 (subunit B), and from Asn135 (subunit A) to Asp286 (subunit B). At 72.5 °C, the enzyme has become denatured which results in a much disordered formation. This disordering does not return to its native form when it is cooled back to 5 °C. Heat treatment up to 80 °C initiates an irreversible reaction to C23O(SH1). In this study, by the combination of SANS measurement and biochemical properties of an enzyme, we demonstrate the correlation between the enzyme size and catalytic activity. For a protein composed of multiple subunits, the shape of the enzyme and the dissociation of the enzyme subunits during a thermal cycle can be revealed using SANS methodology. C23O is an oligomeric enzyme. The maintenance of a proper quaternary structure is an essential prerequisite for it to be a functional macromolecule. The results of this study not only provide a basic understanding of the structural conformation of individual subunit of an oligomeric enzyme but also reveal how the subunits are dissociated upon heat denaturation. By construction of mutations on amino acid residues related to ion pairs, hydrogen bonds, or hydrophobic interaction at the subunit interface, the C23O enzyme may serve a useful model for studying the subunit interface interaction of an oligomeric enzyme at the molecular level.

Huang et al. Acknowledgment. Identification of commercial equipment in the text is not intended to imply recommendation or endorsement by NIST. The work at NCU was supported by grant Nos. NSC 98-2112-M-008-016-MY3 and NSC 98-2313B-008-001-MY3 from the National Science Council of Taiwan. References and Notes (1) Harayama, S.; Rekik, M. J. Biol. Chem. 1989, 264, 15328–15333. (2) Harayama, S.; Kok, M.; Neidle, E. L. Annu. ReV. Microbiol. 1992, 46, 565–601. (3) Han, S.; Eltis, L. D.; Timmis, K. N.; Muchmore, S. W.; Bolin, J. T. Science 1995, 270, 976–980. (4) Senda, T.; Sugiyama, K.; Narita, H.; Yamamoto, T.; Kimbara, K.; Fukuda, M.; Sato, M.; Yano, K.; Mitsui, Y. J. Mol. Biol. 1996, 255, 735– 752. (5) Kita, A.; Kita, S.; Fujisawa, I.; Inaka, K.; Ishida, T.; Horiike, K.; Nozaki, M.; Miki, K. Structure 1999, 7, 25–34. (6) Eltis, L. D.; Bolin, J. T. J. Bacteriol. 1996, 178, 5930–5937. (7) Huang, S.-L.; Chiang, F.-H. in Abstract of the 99th General Meeting of The American Society for Microbiology 1999, p. 413. (8) Svergun, D. I.; Koch, M. H. Curr. Opin. Struct. Biol. 2002, 12, 654–660. (9) Costenaro, L.; Zaccai, G.; Ebel, C. Biochemistry 2002, 41, 13245– 13352. (10) Loupiac, C.; Bonetti, M.; Pin, S.; Calmettes, P. Eur. J. Biochem. 2002, 269, 4731–4737. (11) Sugiyama, M.; Nakamura, A.; Hiramatsu, N.; Annaka, M.; Kuwajima, S.; Hara, K. Biomacromolecules 2001, 2, 1071–1073. (12) Lebedev, D. V.; Baitin, D. M.; Islamov, A. K.; Kuklin, A. I.; Shalguev, V. K.; Lanzov, V. A.; Isaev-Ivanov, V. V. FEBS Lett. 2003, 537, 182–186. (13) Stanier, R. Y.; Palleroni, N. J.; Doudoroff, M. J. Gen. Micobiol. 1966, 43, 159–171. (14) Nakai, C.; Hori, K.; Kagamiyama, H.; Nakazawa, T.; Nozaki, M. J. Biol. Chem. 1983, 258, 2916–2922. (15) Bradford, M. M. Anal. Biochem. 1976, 72, 248–254. (16) Laemmli, U. K. Nature 1970, 227, 680–685. (17) Guinier, A. Ann. Phys. 1939, 12, 161–236. (18) Guinier, A.; Fournet, G. Small-Angle Scattering of X-Rays; John Wiley & Sons, Inc.: New York, 1955. (19) Konarev, P. V.; Volkov, V. V.; Sokolova, A. V.; Koch, M. H. J.; Svergun, D. I.; PRIMUS, J. Appl. Cryst. 2003, 36, 1277–1282. (20) Svergun, D. I. J. Appl. Crystallogr. 1992, 25, 495–503. (21) Svergun, D. I.; Petoukhov, M. V.; Koch, M. H. Biophys. J. 2001, 80, 2946–2953. (22) Milo, R. E.; Duffner, F. M.; Muller, R. Extremophiles 1999, 3, 185–190. (23) Okuta, A.; Ohnishi, K.; Harayama, S. Gene 1998, 212, 221–228. (24) Bartels, L.; Knackmuss, H. J.; Reineke, W. Appl. EnViron. Microbiol. 1984, 47, 500–505. (25) Polissi, A.; Harayama, S. EMBO J. 1993, 12, 3339–3347. (26) Panick, G.; Malessa, R.; Winter, R. Biochemistry 1999, 38, 6512– 6519. (27) Okuta, A.; Ohnishi, K.; Yagame, S.; Harayama, S. Biochemistry 2003, 371, 557–564.

JP9078579