A highly stable NADP-dependent isocitrate dehydrogenase from Thermus thermophilic HB8: purification and general properties

133

Biochimica et Biophysica Acta. 990 (1989) 133-137 Elsevier

BBA 23049

A highly stable NADP-dependent isocitrate dehydrogenase from Thermus thermophilus HB8: purification and general properties Hidetaka Eguchi, Takayoshi Wakagi and Taira Oshima Department of Life Science, Tokyo Institute 0/ Technology, Kanagawa (Japan)

(Received 19 July 1988)

Key words: Enzyme purification; Isocitrate dehydrogenase; Thermophile; (Thermus)

NADP-dependent isocitrate dehydrogenase (EC 1.1.1.42) was purified to electrophoretic homogeneity from an extremely thermophilic bacterium, Thermus thermophilus HB8, and shown to be a dimeric protein of molecular weight 115000, with a pI of 5.5. The amino acid composition of the present enzyme was similar to that reported for other bacterial counterparts, except for a high Arg/Lys ratio and a low Cys content. Divalent cations, such as Mn2+ and Mg 2 +, were essential for activity. The optimal pH was 7.8 at 55 ° C. The K m values for NADP and n-isocltrate were 6.3 and 8.8 p.M, respectively, with a Vm ax of 77.6 p.moljmin per mg at 55°C. NAD was able to replace NADP with low efficiency. Backward reaction at 40 ° C indicated that the K m value for 2-oxoglutarate was 63 /LM with a Vmllx of 4% that of the forward reaction at that temperature. The enzyme was highly stable against high temperature and denaturing reagents. Introduction

Purification of IDH Cells (200 g wet weight) of T. thermophilus HB8 were grown at 75 0 C as described previously [5]. All the buffers used in the purification procedure contained 1

mM EDTA. The cells were suspended in 1 liter of 20 mM Tris-HCl, pH 7.5, containing 2 mM phenylmethylsulfonyl fluoride, and subjected to sonication using a Branson sonifier B15P at maximum power for 5 min. The disrupted cell suspension was centrifuged at 100000 X g for 20 min. The supernatant (crude extract) was brought to 75% saturation with ammonium sulfate. The resulting precipitate was collected by centrifugation, dialyzed and chromatographed on a DEAE-Sephacel column (3.6 X 26 em) with 2 liters of 20 mM sodium phosphate buffer gradient from pH 8.5 to 6.0. The fraction containing IDH activity was brought to 2 M with ammonium sulfate and applied to a Toyopearl HW55S column (1.6 X 17 em). The enzyme was eluted with a 1 liter gradient from 2 to 0 M ammonium sulfate in 20 mM sodium phosphate buffer, pH 8.0. The active fraction was dialyzed against 5 mM sodium phosphate buffer, pH 8.0 and concentrated in vacuo. The concentrated enzyme solution was applied to a TSK gel G3000SWxL HPLC equilibrated with 20 mM Trissulfate, pH 7.5 containing 0.2 M sodium sulfate. The active fraction was brought to 1 M with ammonium sulfate and applied to a Butyl Toyopearl 650S column (1 X 10 em). The enzyme was eluted with 500 ml gradients from 1 to 0 M ammonium sulfate and 0 to 10% (wIv) glycerol. The IDH thus prepared was stored at 4 0 C without loss of activity for at least 1 month.

Correspondence: T. Wakagi, Department of Life Science, Tokyo Institute of Technology, Midoriku, Yokohama. Kanagawa 227, Japan.

Other chemicals NADP-specific IDH from yeast, NADP, NADPH and NAD were purchased from Oriental Yeast Co. Ltd.

The NADP-dependent isocitrate dehydrogenase (IDH) (EC 1.1.1.42) catalyzes the oxidative decarboxylation of n-isocitrate to 2-oxoglutarate. The enzyme is widely distributed among living organisms [1]. Some microorganisms have two isozymes, both specific for NADP [2,3]. Bacterial IDH is of particular interest since the enzyme from Escherichia coli is controlled by a phosphorylation/ dephosphorylation reaction [4]. In the present study, an NADP-dependent IDH is purified from Thermus thermophilus, a gram-negative, extremely thermophilic bacterium [5]. General characterization of the enzyme including amino acid composition, subunit structure, and stability against heat and denaturing reagents is presented. IDHs have been reported for two families belonging to the genus Thermus, T. aquaticus [6] and T. flauus AT62 [7,8], without information on the structure of the enzymes such as the composition of subunit(s) and amino acids. Materials and Methods

0304-4165/89/$03.50 <0 1989 Elsevier Science Publishers B.V. (Biomedical Division)

134 M

o-threo- and DL-isocitrate were from Sigma. DEAE-

5

Sephacel was obtained from Pharmacia. Other materials for chromatography were from Tosoh Co. Enzyme assay IDH activity was followed by the formation of NADPH at 55 0 C in the standard assay mixture (final volume 700 Itl) containing 50 mM Hepes-Na, pH 7.8, 0.5 mM MnCI 2 , 0.57 mM nr.-isocitrate, 0.2 mM NADP, and sample enzyme. For determination of K m values, D-threo-isocitrate was used instead of the DL-mixture. The backward reaction was followed by oxidation of NADPH in an assay mixture containing 50 rnM HepesNa, pH 7.9, 50 mM sodium bicarbonate, 0.5 mM MnCI 2 , various concentrations of NADPH, 2-oxoglutarate, and enzyme, through which CO2 gas was bubbled. Polyacrylamide gel electrophoresis SDS-gel electrophoresis was performed according to Laemmli [9] using molecular weight marker proteins (Bio-Rad) as indicated in Fig. 1. Isoelectric focusing and determination of pI were as previously described [10]. Amino acid analysis The amino acid composition was determined for the samples hydrolyzed in vacuo at 110 0 C for 24 h on an lrika A3300 analyzer. Trp content was determined spectrophotometrically [11]. Cys was determined by the reaction with 5,5'-dithiobis(2-nitrobenzoic acid) [12]. Stability of the enzyme Thermal stability of IDHs from T. thermophilus and yeast was determined as follows; the enzyme (0.1 mgyrnl) was incubated in 50 mM sodium phosphate buffer, pH 7.6, at various temperatures for 10 min and then cooled in ice; the remaining activity was determined by the standard assay procedure except that yeast IDH activity was determined at 25 C. 0

Results and Discussion Purification; homogeneity and molecular weight of IDH A typical purification of IDH from T. therrnophilus HB8 is summarized in Table I. The enzyme was homo-

Fig. 1. "OS-polyacrylamide gel electrophoresis of the IDH purified from T. thermophilus HB8. S, purified enzyme (5 1'-8); M. marker proteins for molecular weight (as indicated).

geneous as judged by SDS-polyacrylamide gel electrophoresis (single protein band with a molecular mass of 57.5 kDa, Fig. 1). The molecular mass of the native enzyme was determined by gel filtration to be 120 kDa (Sephacryl S-200) and 95 kDa (Toyopearl 03000 HPLC) (data not shown). These values are higher than those reported for T. aquat icus (60-70 kDa) [6J and for T. flaous (80-90 kDa) [7,8]. The data indicate that T. thermophilus IDH is a dimeric protein with a molecular mass of about 115 kDa (based on SDS-polyacrylamide gel electrophoresis). IDH from E . coli [13], Bacillus stearothermophilus [14], IDH-I from Vibrio sp. [3J, and IDH from an alkalophilic Bacillus [15] have been reported to be dimeric proteins with molecular weights of 90000-98000 (see Table II).

TABLE I Summary oj purification oj IDH from T. thermophilus HB8 Step 1. Crude extract 2. (NH4hS04 ppt. 3. DEAE-Sephacel 4. HW55S Toyopearl 5. G3000SWxL HPLC 6. Butyl Toyopearl

Volume (ml)

Protein

Activity

Spec. act.

(rng)

(U)

(U /mg)

1050 260 545 45 17.3 10.0

8440 8060 621 65.7 39.8 23.2

6850 6370 4250 1740 1340 1030

0.81 0.79 6.84 26.5 33.6 44.5

Purification (-fold) 1

1 8.4

33 41 55

135 TABLE II

TABLE III

Amino acid composition of dimeric NADP-dependent IDH

Requirement for divalent cations

Residues per mol of subunit of T. thermophilus (Tt), B. stearothermophilus (Bs), E. coli (Ec), and Vibrio sp. (V).

The activities dependent on NADP and NAD were measured in assay mixtures containing 0.2 mM NADP and 1 mM NAD, respectively. The reaction was carried out in a standard assay mixture except that 1 mM of various metal salts (Cl- as counter ion) and 0.57 mM of nt-isocitrate were used. See Materials and Methods for details. The relative activity is compared to that of Mn2 +.

Amino acid

Tt

Bs

Ec

Lys His Arg Asx Thr Ser Glx Pro Gly Ala Cys Val Met lie Leu Tyr Phe Trp

31 7 31 42 33 19 67 25 52 53 2 38 6 20 42 14 18 2

34 5 18 38 24 13 51 20 39 45

31 5

M r (XlO- 3 )

pI

Ref. n

b

a

Ec

V

31 5

27 7 15 43 22 34 55 21 52 45 4 32

17

17

31 9 33 26 12 16 5

40 18 13 46 20 40 38 6 30 13 37 31 15 10 6

38 19 15 50 21 42 40 7 30 13 38 31 13 10 6

27 34 13 13 4

57.5

46.5

45.756

46.9

49

5.5

5.0

this work

±

14

b

4.4 17

13,16

Divalent cation

Activities dependent on NADP

NAD

Mn 2 + Mg 2 + Ca 2 + Sr 2 + Co 2 + NiH

100 71 47 17 4.4 0

100 67 47 69 80 0

11

4.9 3

and C02 + were more effective for NAD-dependent activity than for the NADP-dependent one (Table Ill).

Catalytic properties The optimal pH of the enzyme was about 7.8 at 55 0 C. About half of the maximum activity was observed at pH 6.4. The enzyme was quite thermophilic. Arrhenius

Deduced from DNA sequence. Half-cystine peak was too low to be calculated.

Amino acid composition and isoelectric point Table II also shows the amino acid composition and the pI value of the present IDH, compared with those reported for moderately thermophilic [14], mesophilic [13,16,17] and psychophilic [3] bacteria. A high ArgjLys ratio (greater than 1) is commonly found in Thermus enzymes [18], which may be derived from a high GC content (about 70%) of DNA, because an unusually high ratio of ArgjLys is not found for NADH dehydrogenase [5] nor IDH (manuscript in preparation) from another extreme thermophile, Sulfolobus acidocaldarius, the GC content of which is low (38%). An abundance of Glx is also reported for many thermophile proteins. However, in the present IDH some of Glu residues appear to be necessary to neutralize the positive charge of Arg, because the pI value (5.5) of the enzyme is similar to that (4.4-5.0) of its counterparts. A low Cys content is also usually found in thermostable enzymes. However, Cys residue(s) appear(s) to be located at the active site of Thermus IDH, because incubation of the enzyme with 0.01 mM p-chloromercuribenzoate caused complete loss of activity. In these respects, the present IDH resembles IDH from B. stearothermophilus [14].

80

o

0

60

40

'-0 o

....>, u 0 1IJ

>

0

.'t:

c

C]l

0 ...J

2

o

3 2.8

3.0

3.2 1/Txl0

Metal requirement Divalent cations were essential for maximal activity, with Mn2 + > Mg 2 + > Ca2+ in decreasing order. Sr H

o (Oe)

20

3.4 3

3,6

(K"')

Fig. 2. Temperature dependence of the reaction. The activity of IDH from T. thermophilus HB8 was assayed at the various temperatures indicated. See Materials and Methods for details.

136

plots (Fig. 2) revealed discontinuities in slope at 37 0 C. From the plots, activation energies of 10.5 kcaljmol and 23.4 kcaly'mol were calculated above and below 37 0 C, respectively. Such two-phase plots have been reported for T. flauus IDH, the discontinuity of which occurred at 55-60 0 C [8]. Above 90 0 C, the reaction rate decreased drastically. Kinetic parameters of the forward reaction were determined at 55 0 C, pH 7.8. The s ; values for NADP and D-threo-isocitrate were 6.3 and 8.8 p.M, respectively. The Vrnax value was 77.6 p.mol/min per mg. NAD can replace NADP with low efficiency; that is, the K rn values for NAD and D-threo-isocitrate were 3 mM and 80 p.M, respectively, with a Vrn ax of 47 ,umoljrnin per mg. The NAD-dependent activity was not due to a small amount of NADP contained in NAD, because NADP-specific IDH from pig heart (obtained from Sigma) did not show any activity with high concentrations (> 1 mM) of NAD. It has been reported that crude IDH from S. acidocaldarius, a thermoacidophilic archaebacterium, showed almost similar Vrnax for NAOand NAOP-dependent activities [19]. The backward reaction using N ADPB as a cofactor revealed that the K rn for 2-oxoglutarate was 63 p.M, and the Vrnax value was about 4% that of the forward reaction, at 40 0 C. The effects of organic salts related to the tricarboxylic acid and glyoxylate cycles on the present IDH were examined under similar condi tions to those for the standard assay. The enzyme activity was not significantly affected (less than 20% inhibition) by 1 mM each of citrate, 2-oxoglutarate, succinate, fumarate, and malate. Glyoxylate (1 mM) or oxaloacetate (1 mM) inhibited the enzyme to 0 and 9%, respectively. However, combination of both caused 42% inhibition. Such

100 '0' ~ ~ >

80

4.>

60

iii

I I

I I I

I I

40

>

Q)

4 I

U

It!

B

,

I \ I I

20

a::

0

'4

4254O 60 80 100 Temperature(OC)

0

,, ,

I

°

0.2

0.4

0.6

CGuanidi ne Hel]

0,8 (M)

Fig. 3. Stability of the IDH from T. thermophilus (0) and Sac, cereuisiae (LI). (A) The enzyme (0.1 mgy'rn! in 50 mM sodium phosphate buffer, pH 7.6) was incubated for 10 min at the temperature indicated and the remaining activity was determined. The value at 25 0 C was taken as 100% for each enzyme. (B) Effect of GdnHCI On the enzyme activity. The enzyme was assayed with the indicated concentrations of GdnHCI.

a concerted inhibition by oxaloacetate and glyoxylate has been reported for IDBs from other bacteria including Thermus [6,7]. Stability The present IDH was extremely stable against heat and guanidine hydrochloride (GdnHCI). Ten minutes incubation at 55 0 C completely inactivated the lOB from yeast, while it did not affect Thermus enzyme at all (Fig. 3A). The concentrations of GdnBCI required to cause 50% inactivation of the IOHs from T. thermophi/us and Saccharomyces cereuisiae was 0.27 and 0.07 M, respectively (Fig. 3B). The yeast enzyme was a dimeric protein with a subunit of about 45 kDa (data not indicated). Inactivation of these IDHs may arise not only from the unfolding of polypeptide but also from the dissociation of the oligomer into subunits. Further studies are required for the exact evaluation of the stability of the present IDH. Conclusion Apart from their high stability, Thermus IOHs share common features with most of the other bacterial IDBs so far reported: molecular weight, dimeric structure, pI, and amino acid composition, although minor variations can be pointed out as mentioned above. Acknowledgement

We want to thank Dr. Showbu Sato of Mitsubishikasei Institute of Life Science, for the amino acid analysis. References 1 Colman, R.F. (1983) Pept, Protein Rev. 1, 41-69. 2 Reeves, H.C., O'Neil, S., and Weitzman, P.D.J. (1986) FEMS Microbiol. Lett. 35, 229-232, 3 Ishii, A, Ochiai, T., Imagawa, S., Fukunaga. N., Sasaki, S" Minowa, 0., Mizuno, Y. and Shiokawa, H. (1987) J. Biochem, 102, 1489-1498. 4 Garnak, M. and Reeves, H.C. (1979) Science 203,1111-1112, 5 Oshima, T. and Irnahori, K. (1974) Int. J, Syst, Bacterial. 24, 102-112. 6 Ramaley, R.F. and Hudock, M.O. (1973) Biochim, Biophys. Acta 315, 22-36, 7 Saiki, T. and Arirna, K. (1975) J, Biochem. 77, 233-240, 8 Saiki, T., Mahmud, 1., Matsubara, N., Taya, K. and Arima, K. (1976) in Enzymes and Proteins from Thermophilic Microorganisms (Zuber, H., ed.), pp. 169-183, Birkhaeuser Verlag, Basel/Stuttgart, 9 Laemmli, U.K. (1970) Nature 227,680-685, 10 Wakao, H., Wakagi, T. and Oshima, T. (1987) J, Biochem. 102, 255-262. 11 Edelhoch, H. (1967) Biochemistry 6, 1948-1954. 12 Phillips, N.L, Jenness, R. and Kalan, E,n. (1967) Arch. Biochem, Biophys. 120, 192-196. 13 Burke, W,F., Johanson, R.A, and Reeves, H.C. (1974) Biochim, Biophys. Acta 351, 333-340.

137 14 Boward. R.C . and Becker. R.R. (1970) J. BioL Chern. 245, 3186-3194. 15 Shikata, S., Ozaki, K., Kawai, S., Ito, S. and Okamoto, K. (1988 ) Biochim, Biophys. Acta 95~, 282-289. 16 Vasquez. B. and Reeves, H.C. (1979) Biochirn. Biophys. Acta 578. 31-40.

17 Thorsness, P.E. and Koshland, D .E .• Jr. (1987) J. BioI. Chern. 262, 1042 2-10425. 18 Oshima, T. (1986) in Thermophiles - General, Molecular and Applied Microbiology (Brock. T.D .• ed.), pp. 137-158, Wiley-Interscience, New York . 19 Danson, M.J. and Wood, P.A. (1984) FEBS Lett. 172 . 289-293.