A specific l-asparaginase from Thermus aquaticus

ARCHIVES OF BIOCHEMISTRY Vol. 241, No. 2, September,

AND BIOPHYSICS pp. 571-576, 1985

A Specific L-Asparaginase MONIQUE

P. CURRAN, AND

School

of Science,

Received

ROY M. DANIEL,’ HUGH W. MORGAN

University

February

from Thermus aquaticus

of Wail&o,

Hamilton,

1, 1985, and in revised

form

GRAEME

New May

R. GUY,

Zealand 15, 1985

The L-asparaginase from an extreme thermophile, Thermus aquaticus strain T351, was highly substrateand stereospecific, with no activity against glutamine or D-asparagine. It had a high K, of 8.6 mM. In these aspects it closely resembled the corresponding enzymes from thermophilic bacteria. The enzyme had a molecular weight of 80!,000, an isoelectric point of 4.6, and a pH optimum of 9.5. It showed some substrate inhibition above 20 mM asparagine and was also inhibited by L-aspartic acid, D- and L-lysine (Ki of 5.2 and 1.25 mM, respectively), and D- and L-serine. The half-life of the enzyme at 85’C was 40 min. The Arrhenius plot showed a change in slope at 55°C. 0 1985 Academic Press, Inc. the facultative thermophile, Bacoagdans, and the obligate thermophile, Bacillus stearothermophilus (l),

lifetime in vivo. Since persistence and low K,,, are probably two important properties

Both

cillus

have been shown to possess L-asparaginase activity. However, these two thermophilic asparaginases both differ from most mesophilic enzymes in having high K, values a.nd high specificity, (l-3), i.e., lack of activity against D-asparagine or glutamine. It is not clear whether these differences are associated with thermophilly. It may be that they reflect differences in asparagine metabolism specifically associa.ted with growth at high temperatures. If so, these features should be particularly marked in extreme thermophiles, and this work on the properties of an L-asparaginase from an extremely thermophilic bacterium, Thermus aquaticus (ATCC :31674), enables a comparison with the en:zyme from thermophiles and mesophiles. Since enz:ymes from extreme thermophiles are likely to be resistant to proteolysis (4) and will have enhanced thermal stability, it is possible that L-asparaginases from this source will have a longer 1 To whom

correspondence

should

be addressed. 571

of L-asparaginases used as antineoplastic agents (3), it is of particular interest to know whether a high K, is a property likely to be associated with L-asparaginases from thermophiles and extreme thermophiles. MATERIALS

AND

METHODS

Bacterial growth T aquaticus (ATCC 31674, previously known as strain T351) was maintained, grown, and harvested as described previously (5). Asparaginase assay. The Nessler method of Bergmeyer (6) was used for cell-free L-asparaginase analysis except that 0.1 M Na2B&-HCl buffer, pH 9, was used instead of Tris buffer, and the supernatant was diluted with 0.1 M NaOH instead of distilled water, improving the sensitivity by about 40%. Each assay included a substrate blank (asparagine omitted), an enzyme blank (autoclaved enzyme used), and a standard (enzyme omitted, ammonium sulfate standard included), and these were treated identically to the assay tubes. Standard curves for ammonia, derived from a series of ammonium sulfate standards treated in the same way as the assays, were linear and passed through the origin. We found no evidence for any loss of ammonia from the assay system, or for any nonenzymatic hydrolysis of asparagine. All assays were carried out at 75°C unless otherwise stated. 0003-9861/85 Copyright All rights

$3.00

0 1985 by Academic Press, Inc. of reproduction in any form reserved.

572

CURRAN

Protein detemination. Protein was routinely assayed by the method of Bradford (7). Bovine serum albumin (fraction V, Sigma) was used as a standard. Enzyme puurification All purification steps were carried out at room temperature (20°C) unless otherwise stated. Cakes (250 g) of T aquaticus strain T351 were suspended in 400 ml 0.15 M NaCl in 0.05 M Tris-HCl, pH 8.2, 0.2 mg ml-’ DNAase. The suspension was kept on ice and disrupted by continuous recirculation through a Dynatech sonic dismembrator (Artex Model 300) for 4 h at a flow rate of 25 ml min-‘. The cell debris and unbroken cells were removed by centrifugation at 10,OOOo for 40 min. The supernatant was carefully removed and centrifuged at 195,OOOg for 4 h. The upper 7/B of the high-speed supernatant was removed, bulked, and then dialyzed for 18 h against 0.15 M NaCl in 0.05 M Tris-HCl buffer, pH 8.2. The dialyzed fraction was loaded on to a DEAE-Sepharose CL-6B column (15 X 2.5 cm) that had been equilibrated with 0.15 M NaCl in 0.05 M Tris-HCl buffer, pH 8.2, and eluted with a salt gradient (0.15-0.25 M NaCl in 0.05 M Tris-HCl buffer, pH 8.2). The active fractions were pooled, dialyzed against 0.25 M NaCl in 0.05 M TrisHCl buffer, pH 9.2, for 18 h, and loaded onto a QAESephadex A50 column (10 X 2.5 cm) equilibrated with 0.25~ NaCl in 0.05~ Tris-HCl buffer. The fractions were eluted with a linear salt gradient from 0.25 to 0.35 M NaCl in 0.05 M Tris-HCl buffer, pH 9.2. The fractions with L-asparaginase activity were desalted, equilibrated with 0.01 M NazHP04/ NaHzPO, buffer, pH 7, containing 1 mM P-mercaptoethanol buffer, pH 7, and concentrated using a PM 30 ultrafilter (Amicon). The enzyme was then applied to a hydroxylapatite (Bio-Rad Bio-Gel HTP) column (7 X 2.5 cm) which had been equilibrated with 0.01 M Naz HPOJ NaHzPO,, pH 7.0, and was eluted with 0.02 M phosphate buffer. Peak fractions containing L-asparaginase were desalted and concentrated using a PM 30 ultrafilter, applied to a Sephadex G-150 (96 X 2.5 cm) column which had been calibrated with proteins of known molecular weight, and eluted with 0.025 M Tris-HCl buffer, pH 8.2, containing 1 mM P-mercaptoethanol. Electrophoresis. Analytical isoelectric focusing was carried out on commercial PAG plates, pH 3.5-9.5 (LKB, Stockholm, Sweden), and stained according to the manufacturer’s instructions. Asparaginase activity was determined by the agar gel overlay method of Padjack and Padjack (8). SDS’-electrophoresis was carried out by the method of Laemli (9), and gels were stained with a solution of 50% aqueous methanol:acetic acid (glacial):Coomassie blue R250 (454:46:1.25, v/v/w).

’ Abbreviation

used: SDS,

sodium

dodecyl

sulfate.

ET

AL. RESULTS

AND

DISCUSSION

The purification results are summarized in Table I. By the criteria of isoelectric focusing and SDS-gel electrophoresis the enzyme was better than 99.5% pure. At heavy loads of purified L-asparaginase, isoelectric focusing and SDS-electrophoresis gels sometimes showed traces of two other protein bands, constituting less than 0.5% of the total protein. The L-asparaginase was free of racemase, D-asparaginase, L- and D-asparagine-oxo-acid aminotransferase, L- and D-glutaminase, L- and D-glUtam&? dehydrogenase, and L- and D-aSpartWZ!.

The specific activity (585) is comparable with that of L-asparaginases from guinea pig serum (470), Escherichia coli (280), Erwinia carotivora (650-700), and Pseudowmnas vulgaris (300) (lo), at the growth temperatures of the organisms concerned. At 37°C the specific activity of the T. aquaticus enzyme is only 70. The molecular weight of the L-asparaginase, determined by gel filtration on Sephadex G-150, and G-200, and by SDSelectrophoresis, was 80,000 + 2000. This molecular weight is the lowest reported for a bacterial L-asparaginase, but is similar to that from the thermophile, B. coagulans (85,000) (2). The isoelectric point of the enzyme was 4.6, and the pH optimum 9.5. The thermal decomposition of L- and Dglutamine at 75°C released such a high level of ammonia in both the control and test solutions that accurate determination of any glutaminase activity of the L-asparaginase enzyme was prevented at this temperature. No D- or L-glutaminase activity was detected when the L-asparaginase was incubated with glutamine at 6O”C, where ammonia release due to thermal decomposition was much less. L-ASparaginase activity was readily measurable at this temperature. At 75°C the enzyme catalyzed no ammonia release from D-aSparagiIM?, L-aspartic acid, D-aspartic acid, or L-glutamic acid. Few bacterial L-asparaginases lack any trace of D-asparaginase or L-glutaminase activity, and of those that do [including the en-

A SPECIFIC

L-ASPARAGINASE

FROM

TABLE PURIFICATION Total

protein (mg)

Step 1. High-speed supernatant 2. DEAE-Sephabrose CL-6B linear gradlient 3. QAE Sephadex A-50 linear gradlient 4. Hydroxylapatite linear gradient 5. G-150 Sephadex chromatography

Specific activity (units/mg)

Relative purification

100

1

2,780

127,880

46

86

18

1,620

123,120

'76

83

29

292

56,940

195

38

75

54

31,590

585

21

225

which

will

catalyze

Es -[II,---J”

0.3

0.4

FIG. 1. Effect of substrate concentration on Lasparaginase activity at 75°C (Lineweaver-Burk plot).

2.6

Yield (%I

148,000

-i-

0.2

I

Total activity (units)”

most which have been tested are inhibited by D-asparagine except the L-asparaginase from T. aquaticus (see below). This enzyme is therefore particularly substrate- and stereospecific. As we have pointed out for the D-aspara.ginase from the same organism (13), the stereospecificity of the enzyme may be a response to the higher rates of thermal racemization at the growth temperature of the organism. The L-asparaginase has a K, of 8.6 mM and is inhibited by asparagine concentrations above 20 mM (see Fig. 1). The K, is rather high compared to most bacterial L-asparaginases. It is interesting that all of the specific L-asparaginases mentioned

0.1

573

OF L-ASPARAGINASE

zymes from Pseudomonas geniculata (11); Mycobacterium tuberculosis (12); Bacillus stearothermophilus and possibly B. coagulans (1, 2); and T. aquaticus, this paper]

0

aquaticus

58,000

n Activity unit = amount of enzyme under the conditions of the assay.

-0.1

Thermus

the formation

of 1 Fmol

ammonia

min-’

at 75°C

above also have relatively high K,‘s, above lop3 M, whereas most other bacterial asparaginases have K,‘s of less than 1O-4 M. Although high specificity and a relatively high K, are not exclusive to thermophiles, it is interesting that all thermophiles so far examined possess them. Ammonia (1 mM) has no significant effect on Michaelis-Menten kinetics of Lasparaginase. Investigation of the inhibition of L-asparaginase activity by amino acids was complicated by interference with the Nesslerization reaction (Table II). Standard curves in the presence of 20 mM tryptophan, 2 mM cysteine, and 2 mM histidine were constructed for determining the effect of these on L-asparaginase activity. Table III shows the effect of amino acids on L-asparaginase activity. The Lineweaver-Burk plot obtained in the presence and absence of L-aspartate indicated that the L-asparaginase is competitively inhibited. A Dixon plot confirmed this (Fig. 2). The K, in the presence of 20 mM L-aspartic acid increased from 8.6 to 20 mM. D-Aspartic acid caused no inhibition. Lineweaver-Burk plots obtained in the presence and absence of Llysine and of D-lysine suggested noncompetitive inhibition, and this was confirmed with Dixon plots (Fig. 3). The Ki for Llysine is 1.25 MM, while that for D-lysine is 5.2 mM. The Lineweaver-Burk plot for L-asparaginase in the presence of L-lysine

574

CURRAN ET AL. TABLE II

TABLE III

INTERFERENCE WITH NESSLERIZATION BY AMINO ACIDS

Amino

acid

Percentage decrease at Aa6 nm caused by the amino acid

INHIBITION

Percentage activity

Inhibitor Zn2+

30 100 44 100 64

Note. Standards containing 0.5 mM (NH&S04 were prepared for each incubation mixture containing the respective amino acid. The ammonia level was determined by Nesslerization. The pH of all reaction mixtures was adjusted to 9.0. L-Asparagine, L-aspartic acid, L-threonine, L-serine, L-glutamic acid, Lmethionine, L-lysine, D-aspartic acid, D-lysine, and D-Serine showed no inhibition of the Nesslerization reaction.

(not shown) shows that it overcomes the substrate inhibition at high asparagine concentrations. The binding of L-lysine may alter the enzyme conformation, subunit interactions, or site-site interactions, so that the substrate can no longer inhibit at high concentrations. Relief of substrate inhibition by L-aspartic acid or L-serine (see below) did not occur. L- and D-serine also inhibit the L-asparaginase competitively (data not shown), which suggests the serine may bind to the substrate site. However, substrate binding is stereospecific but the serine inhibition is not. Alternatively, the L- and D-serines may bind to the same site as the L- and D-lysines. This site did not display stereospecificity but the binding of the L- and D-lysines to the enzyme resulted in noncompetitive inhibition. Or, the L- and D-serines may bind to a third site which does not display stereospecificity and which permits competitive inhibition.

84

1 mM 1 mM 5mM 5mM 1 mM 20 mM 20 mM 2mM 20 mM 2mM 20 mM 20 mM 20 mM 2 mM 20 mM 2mM

PO;L-Trytophan 20 mM L-Cysteine 20 mM 2mM L-Histidine 20 mM 2mM

OF L-ASPARAGINASE

u-Ketoglutarate Pyruvate Oxaloacetate L-Aspartic acid L-Lysine D-Lysine L-Serine D-Serine L-Cysteine L-Histidine

113 87 56 80 49 0 43 0 73 80 79 -.-a 101” -a 97’

Note. The following did not affect L-asparaginase activity: 1 mM NH:, K+, Na+, Mg’+, Ba*+, Ca’+, or Pb’+; and 20 mM L-methionine, L-tryptophan, Lglutamate, L-threonine, D-asparagine, or D-aSpartiC acid. a Figures for activity (ammonium concentration) were corrected for interference with the Nesslerization reaction (see Table II). In the case of 20 mM L-cysteine and L-histidine the interference was too strong to read the assay.

All combinations of L-aspartic acid, Llysine, and L-serine gave levels of inhibition greater than the sum of the inhibitory -

20mN Asparagine

1

I . -5 -h

I 0

5

10 Aspartic

FIG. 2. Effect of L-aspartate asparaginase activity at 75°C

15

20

25

30

acid I mtl I

concentration (Dixon plot).

on L-

A SPECIFIC

L-ASPARAGINASE

effects produced by the individual amino acids. While the increase in inhibition over that expected for cumulative inhibition was not. large (6-g%), it was consistent. This type of inhibition is defined by Dixon and Webb (15) as being cooperative. At 65”C, the half-life (t& of L-asparaginase was 8 h, at 85°C t1i2 was 40 min, and at 95°C tl12 was 2.5 min. An Arrhenius plot for L-asparaginase showed a change in slope at 55°C (Fig. 4); E, > 55°C was 13 kJ mol-‘, and E, < 55’C was 88 kJ m&‘. However, the K, for Lasparaginase determined at 45°C was not significantly different from that at 75”C, suggesting that if a conformational change occurs it does not affect binding groups at the active site. Furthermore, the inhibition of L-asparaginase by L-aspartic acid or L-lysine ‘was the same at 45°C as at 75°C. Sugimloto and Nosoh (16) reported that ORD and CD studies indicated that no gross structural change occurred in B. stearothermophilus fructose-1,6-diphosphate aldolase through its transition temperature. However, studies with EDTA suggested that structural changes occurred around the active site. Maramatsu and Nosoh (17) also showed that while ORD and CD studies showed no gross structural change in glucose-6-phosphate isomerase from B. stearothermophilus, the

FROM

2.0 21

Th.erwms

90 a

00 I' 2.8

aqua&us

575

70

60

so

2.9

3.0

3.1

)

40

3.2

3.3

‘“~(K-‘, FIG. 4. Effect of temperature on L-asparaginase activity (Arrhenius plot). Open circles indicate data which has been corrected to compensate for significant thermal denaturation occurring over the time period of the assay.

enzyme had differing K, values above and below the transition temperature. Overall, these results show that the Lasparaginase from T. aquaticus is a particularly stable enzyme, with features consistent with allosteric regulation. It resembles the enzyme from thermophiles in having a high K, and high specificity. This suggests differences in asparagine metabolism between these organisms and mesophiles. We speculate that L-asparaginases from thermophiles and extreme thermophiles tend to have high K,‘s and, despite

their

stability,

are therefore

likely to be useful as antineoplastic

un-

agents.

REFERENCES

L-Lyr;ine

ImHl

O-Lyrine

imI4l

FIG. 3. Effect of L- and D-lysine on L-asparaginase activity at 75°C (Dixon plot); 0, 5 mM asparagine; W, 20 mM asparagine.

1. MANNING, G. B., AND CAMPBELL, L. L. (1957) Canad. J. Microliol. 3, 1001-1009. 2. LAW, A. S., AND WRISTON, J. C. (1971) Arch. Biochem Biophgs. 147, ‘744-751. 3. COONEY, D. A., AND HANDSCHUMACHER, R. E. (1970) Amu, Rev. PharmacoL 10, 421-440. 4. DANIEL, R. M., COWAN, D. A. C., MORGAN, H. W.,

576

5. 6.

7. 8. 9. 10.

CURRAN

AND CURRAN, M. P. (1982) &o&em. .I 207, 641-644. HICKEY, C. W., AND DANIEL, R. M. (1979) J. Gen. Microbial. 114, 195-200. BERGMEYER, H. U. (1974) in Methods of Enzymatic Analysis, 2nd English ed. Vol. 1, pp. 435-436, Academic Press, New York. BRADFORD, M. M. (1976) Anal. B&hem. 72, 248254. PADJACK, E., AND PADJACK, W. (1972) Anal. B&hem. 60,317-320. LAEMMLI, U. K. (1970) Nature (London) 227,680685. RUYSSEN, R., AND LAUWERS, A. (1978) in Pharmaceutical Enzymes (Ryssen, R., and Lauwers, A., eds.), pp. 181-200, E. Story-Scientia, Gent.

ET

AL.

11. KITTO, G. B., SMITH, G., THIET, T-Q., MASON, M., AND DAVIDSON, L. (1979) J. Bacterial. 137,204212. 12. JAYARAM, H. N., RAMAKRISHWAN, T., AND VAIDYANATHAN, C. S. (1968) Arch. Biochem. Biophys. 126, 165-174. 13. GUY, G. R., AND DANIEL, R. M. (1982) B&hem. J 203, 787-790. 14. RATCLIFFE, H. D., AND DROZD, J. W. (1978) FEMS Lett. 3, 65-69. 15. DIXON, M., AND WEBB, E. C. (1979) Enzymes, 3rd ed., Longman, London. 16. SUGIMOTO, S., AND NOSOH, Y. (1971) Biochim. Biophys. Acta 235, 210-221. 17. MURAMATSU, N., AND NOSOH, Y. (1971) Arch Biochem. Biophys. 144, 245-252.