Eur. J. Biochem. 208,475-480 (3992) 8 FEBS 1992

Site-specific mutagenesis of Escherichia coli asparaginase I1 None of the three histidine residues is required for catalysis Andreas WEHNER', Etti HARMS', Michael P. JENNINGS3, Ifor R. BEACHAM3, Christian DERST ', Peter BAST4 and Klaus H. ROHM' I

Institut fur Physiologische Chemie, Philipps-Universitat, Marburg, Federal Republic of Germany Department of Biological Sciences, Purdue University, West Lafayette, USA Division of Science and Technology, Griffith University, Brisbane, Australia Fachbereich Chemie, Philipps-Universitat, Marburg, Federal Republic of Germany

(Received May 18, 1992) - EJB 92 0681

Site-specific mutagenesis was used to replace the three histidine residues of Escherichiu coli asparaginase I1 (EcA2) with other amino acids. The following enzyme variants were studied: [H87A]EcA2, [H87L]EcA2, (H87KIEcA2, IH183LIEcA2 and [H197L]EcM. None of the mutations substantially affected the K , for L-aspartic acid fl-hydroxamate or impaired aspartate binding. The relative activities towards L-Asn, L-Gln, and 1-asparticacid 8-hydroxamate were reduced to the same extent, with residual activities exceeding 10% of the wild-type values. These data do not support a number of previous reports suggesting that histidine residues are essential for catalysis. Spectroscopic characterization of the modified enzymes allowed the unequivocal assignment of the histidine resonances in 'H-NMR spectra of asparaginase 11. A histidine signal previously shown to disappear upon aspartate binding is due to His183, not to the highly conserved His87. The fact that [H183L]EcA2 has normal activity but greatly reduced stability in the presence of urea suggests that His183 is important for the stabilization of the native asparaginase tetramer. 'H-NMR and fluorescencespectroscopy indicate that His87 is located in the interior of the protein, possibly adjacent to the active site.

The asparaginases catalyze the hydrolysis of L-asparagine to L-aspartate and ammonia. Several asparaginases of mi-

crobial origin exhibit antitumor activity. Among these, asparaginase I1 of Escherichiu coli (EcA2) has been studied most thoroughly. Preparations of this enzyme are in use for the treatment of acute lymphoblastic leukemias [ 11. Kinetic studies of substrate turnover and oxygen-exchange have firmly established that EcA2 and related enzymes employ a ping-pong mechanism of catalysis with a covalent acyt enzyme intermediate [2, 31. Based on these and other data, it was proposed that the mechanism of asparaginase may resemble those of the serine proteinases [4]. Recently, Thrl2 of EcA2 was identified as an important catalytic residue [5]. However, the chemical nature of the covalent intermediate of EcA2 is still unresolved. Evidence from a number of kinetic and chemical modification studies indicated that two of the three histidine residues of EcA2 may be involved in catalysis. So, for instance, dyesensitized photo-oxidation [6] and modification by diethyl pyrocarbonate [7l inactivate the enzyme. Both reagents primarily affect histidine residues, although tyrosine, tryptophan, Correspondence to K . H . Rohm, Institut fur Physiologische Chemie, Philipps-Universitat, W-3550 Marburg (Lahn), Federal Republic of Germany Abbreviations. EcA2, E. coli asparaginase I1;AHA L-aspartic acid P-hydroxamate. Enzymes. Glutamic dehydrogenase (EC 1.4.1.3); a-chymotrypsin, (EC 3.4.21.1); subtilisin (EC 3.4.21.14); asparaginase I1 (EC 3 5.1.1).

methionine and other residues may also react. From the pH/rate profiles of EcA2-catalyzed asparagine hydrolysis, O'Leary and Mattes [8] concluded that an unprotonated histidine residue is required for catalysis. Independent evidence for the presence of a histidine residue within or near the active site emerged from spectroscopic data, which showed that the fluorescence of the single tryptophan residue of EcA2 is partially quenched by the protonation of a group with an apparent p K , of approximately 6 [9]. The proximity of this residue to the active site is suggested by the observation that binding of' asparagine and aspartic acid also result in fluorescence qnenching. Subsequent experiments [lo] showed that the quenching residue has anomalous thermodynamic features that are combtent with a subshntral conformational change, triggered by proton NMR studies furnished additional evidence that a histidine residue is associated with the active site [ll]. 'H-NMR signals originating from the C 2 protons of two different histidine imidazolyl groups were identified. By NMR titration pK, of 7.0 (His a) and 5.9 (His b) were determined. When aspartate, a high-affinity competitive inhibitor, was added, the His-a signal gradually broadened and eventually disappeared while His b was unaffected. This observation suggests that ligand binding to the active site alters the environment of His a in a way that reduces the mobility of its side. chain. In this study we employed site-directed mutagenesis to examine the role of all three histidine residues of EcA2. This approach became feasible by the recent development of ef-

476 ficient systems for the high-level expression of CcA2 from recombinant plasmids [121.

MATERIALS AND METHODS Mutagenesis and expression

Site-directed mutagenesis was performed by two different methods. [HX7A]EcA2, [H183L]EcA2 and [H197L]EcA2 were constructed by the phosphorothioate method of Sayers et al. [131. For mutagenesis a 1.35-kb EcoRI - Hind111 fragment was excised from expression vector pTWE1 [12] and cloned into the polylinker site of M13mp19. The following synthetic oligonucleotide primers were used (mutagenic changes underlined). H87A, 5'-C GTC ATT ACC GGT ACC-3'; H183L, 5'-GGT TAC ATT CTC ACC GG-3'; H197L, 5'GCA cc'r AAG CTT ACC AGC G-3'. Recombinant phage plaques were propagated and screened for the prescnce of the desired mutations by dideoxy sequencing or by plaque hybridization in the prescncc of trimethylammonium chloride [14] using the respective 32Plabelled mutagenic oligonucleotides as probes. The EcaRI Hind111 fragments of mutant phage clones were then re-introduced into EcnRI - HindIII-digested pT7-7 to provide mutant pTWE-type expression vectors. [H87L]EcA2 and [H87K]EcA2 were obtained using the polymerase chain reaction with the desired mutations incorporated into primers. The sequence 5'-TCTGTCCA TGGAGTTTTCAAAAAGAC-3' (NcoI site in italics) was combined with one of the following mutagenic primers (KpnI sites in italics, mutagenic changes underlined). H87K, 5'-C GTC ATT ACC AAG GGT ACC GAC-3'; H87L, 5'-C GTC ATT ACC C I C GGT ACC GAC-3'. The polymerase chain reactions were performed as described previously [12]. Their products were digested with NcoI and KpnI, then ligated into similarly digested pKP2. The nucleotide sequence of the amplified region was determined to confirm the presence of the desired mutation and the absence of errors introduced during amplification. Expression of wild-type and modified asparaginases from pTWE and pKP plasmids and their purification from periplasmic fluid have been described in detail [12]. All enzyme species were isolated using the standard purification protocol developed for wild-type EcA2. The only modification required applied to variant [H87K]EcA2. Due to its markedly higher isoelectric point, a pH gradient from pH 7 to pH 4 was used during chromatofocussing.

-

Assays Enzyme concentrations were derivcd from ultraviolet spectra using an absorption coefficient of A ;go, 7.4. During enzyme purification and while measuring pH/activity profiles, asparaginase activities towards L-asparagine were measured by direct ultraviolet spectroscopy at 225 nm [12]. Since the K , constant for L-Asn is too low to be readily determined, Laspartic P-hydroxamate (AHA; USB Co.) was the staiidard substrate for detailed kinetic studies. The discontinuous AHA assay used is based on the reaction of hydroxylrtmine liberated from AHA with 8-hydroxyquinoline at high pH [15]. The intensely green oxindole dye formed in the reaction (qO5 = 1.7 x lo4 M - ' . cm-') is detectable with high sensitivity. However, the useful working range of the assay is limited to pH 4.5 - 7, as hydroxylaminc is unstable above neutrality.

Routinely; 0.05- 1.5 mM AHA in 100 mM Mes/NaOH buffers, pH 5.0 or 6.0, was incubated with enzyme at 25°C. At regular intervals 100-4 aliquots were withdrawn and added to 50-pl volumes of 12% trichloroacetic acid in microccntrifuge tubes to stop the reaction. When all samplcs had been collected, 1 ml color reagent (2% 8-hydroxyquinolinc in ethanol/l M Na,CO,, 1:3 by vol.) was added to each tube. They were tightly closed andthen heated for exactly 1 min in a boiling-water bath. After allowing to cool for 10 min, the absorbance was read. Initial rates of AHA hydrolysis were calculatcd from parabolic fits to the A705vs. time data (5 points/curve). Kinetic parameters were estimated by non-linear regression analysis. Glutaminase activities were determined at pH 7.0 and 25 "C by a coupled-enzyme assay 1161. In this test. ammonia liberated from L-Gln is employed by the indicator enzyme, ~-glutamicdehydrogenase, to aminatc 2-oxoglutarate in a NADH-dependent reaction. At a sufficicntly large excess of L-glutamic dehydrogenase over asparaginase the rate of NADH consumption (followed at 340 nm) equals the rate of glutamine hydrolysis. Fluorescence measurements

The stability of the active asparaginase tetramer was monitored by measurements of its tryptophan fluorescence at various concentrations of urea. Upon denaturation, the fluorescence emission maximum of EcA2 is shifted from 323 nm to 350 - 360 nm. At the same time, fluorescence intensity is reduced by approximately 67% [17]. The enzyme solutions contained 20 - 50 pg protein/ml in 50 mM TrisiHCI, pH 7.4, and urca at 10 - 15 different concentrations ( I 5 M). They were incubated for 1 h at 2 5 T , then emission spectra of the samples were recorded in a Jasco FP-770 speclrofluorimeter with excitation at 285 nm. From these data, apparent free energies of unfolding, AC,, were estimated as described in [18]. Briefly, for a given urea concentration the frce energy of unfolding is given by AC, = - R . T . In . [(yf -J,)/ (y - y J ] where is the measured fluorescence, while yf and y, are the intensities of the folded and fully unfolded states. When AC, is plotted vs. urea concentration, straight lines are obtained. Extrapolation of those lines to a urca concentration of 0 yields an estimate of the stabilization energy in water, A G(H20). Fluorescence quenching by aspartatc was cxamined at 25°C by successively adding 2.5-pl volumes of a neutral 100 mM aspartate solution to 1 mg enzyme in 2.5 ml 50 mM Hepes/NaOH, 100 mM NaCl, pH 7.0. After each addition, the fluorescence intensity at 323 nm was measured. The effect of pH on fluorescence intensity was studied as follows. The enzyme was dissolved to a final concentration o f approximately 0.3 mgiml in a mixture of Mes, Mops and Taps (10 mM each) previously adjusted to pH 5. Then 1 -2-111 volumes of dilute NaOH were successively added to raise the pH and fluorescence emission spectra were recorded after each addition. When pH 8.5 was reached, dilute HCl was used to titrate the solutions down to pH 5 in the same stepwise fashion. The position of the emission maximum (323 nm) did not significantly change during the procedure. NMR measurements

'H-NMR spectra were recorded at 500 MHz with a Bruker AMX-500 spectrometer. The spectra were collected at 25 "C with unbuffered 1- 2 mM asparaginase solutions in 300 mM

477

A /I

!i

0)

100

u C

0

;

90

i

3 0

H183L

LL . 0

80

K

70

b -

WT

s:o

B.'5

B.'O

?.'5

?.'O

6.5

70

1

I

I

I

6

7

8

I

Chemicol shift (ppm)

Fig. 1. 'H-NMR spectraof wild-type and variant asparaginases.Spectra were recorded as described in Materials and Evlc~liod~ \iith 1-2 mM protein solutions in 300 mM NaCI/D,O at pH values around 5. Only thc aromatic regions of the spectra are shown. For the assigment of the resonances see text. The slightly different chemical shifts of His b are due to small pH differences between the samples. WT, wild type. C

NaCl/D,O, previously equilibrated with D 2 0 to exchange the labile protons, and adjusted to pH 5.0 by addition of DC1 in D 2 0 as described elsewhere [ll]. During data accumulation (200 - 1000 transients/spectrum) the residual HDO signal was suppressed by prior saturation. In some cases, an additional spin echo pulse sequence was employed to reduce the background of residual NIT protons. The data were digitized in 32-K data blocks and an exponential weighting function was applied before Fourier transformation. pH titrations and titrations with aspartate were performed as in [I ll.

RESULTS Enzyme variants The three histidine residues of EcA2 were mutagenized to other amino acids by site-directed alteration of the ansB gene. All enzyme species were readily expressed and purified by procedures previously developed for wild-type EcA2 [ 121. As expected, enzymes in which a His was replaced with Ala or Leu had slightly lower isoelectric points (4.65 - 4.7 compared to 4.85 with wild-type enzyme) while the isoelectric point of [H87K]EcA2 was increased to 5.4. NMR assignments

NMR spectra of the aromatic regions of several asparaginase variants are shown in Fig. 1. The spectra of wild-type enzyme and [H87A]EcA2 are indistinguishable. In agreement with previous results, both contain two different histidine C2 signals at about 8.8 ppm (His a) and 8.2 ppm, respectively (His b). By contrast, the His-a resonance is missing from the spectrum of [H183L]EcA2, while [H197L]EcA2 lacks His b. From these findings it is obvious that His a is identical with His183 while His b corresponds to His197. His87 did not produce a detectable 'H signal under the conditions of the experiment. The chemical shifts of Hisa and His b of [H87A]EcA2 and the shift of H i s b in the spectrum of

70

5

PH Fig. 2. pH-dependent fluorescence quenching. The pH dependence of fluorescence emission of (A) [H183L]EcA2, (B) [H197L]EcA2 and (C) (H87AIEcA2 is shown (0).For experimental details see Materials and Methods. The solid lines in (A) and (B) are titration curves fitted to the data for pH < 7. The estimated pK, were 6.06 0.09 (A) and 6.15 0.08 (B). The solid line in (C) is arbitrary. Rel., relative.

[H183L]EcA2 were studied as a function of pH. The pKa values derived from these titrations (7.0 for His183 and 6.0 for His1 97, data not shown) were not significantly different from those of the wild-type enzyme [ll]. pH dependence of fluorescence In order to identify the histidine residue responsible for the pH-dependent quenching of asparaginase fluorescence, we measured the pH dependence of fluorescence intensity with several asparaginase variants (Fig. 2). The profiles obtained with [H183L]EcA2 and [H197L]EcA2 as well as the apparent pKa values (approximately 6) were the same as reported for the wild-type enzyme 191. By contrast, the fluorescence emission intensity of [H87A]EcA2 was pH independent between pH 5 and pH 7 (Fig. 2c), indicating that the quenching curve reflects the protonation of the imidazolyl moiety of His87. The decrease of fluorescence intensity above pH 7.5 was unaffected by the mutations and thus is not related to histidine residues. It may be due the dissociation of tyrosine residues, two of which have abnormally low p K , values of approximately 8 1171. Kinetic properties Kinetic data for wild-type and variant asparaginases are summarized in Table 1. Surprisingly, none of the mutations

47 8

Table 1. Kinetic constants for wild-type and variant asparaginases. Activities were measured as follows. L-Asn, 10 mM substrate ( E 100 x K,) at 25°C and pH 7.0; L-Gln, 20 mM substrate (=: 10 x K,) at 2 S T and pH 7.0; AHA, maximal velocities determined at 25°C and pH 5.0. Reaction rates are given as turnover numbers, k,,,, calculated for one active site/subunit.

Asparaginase I1

k,,, for

L-Asn

5,

I

I

I

4 1

K, for AHA AHA

L-Gln

I

1

LL

(1

Wild type

31 11 10 3.9 30 8.8

[H8?A] [II87L] [H87K] [Hl83L] [H197L]

122 & 15

33 8.5 8.3 4.0 29 9.0

0.33

0.11 -

0.23 0.10

*

121 18 90 & 10 110 f 17 160 i 2s 125 & 26 A__

0.0

0 2 0.2

0 4 0.4

0.6

0.8

i 1 .o

Fig. 4. Aspartate binding by wild-type and variant enzymes. The concentralion dependence ol' fluorescence quenching at 323 nm was employed to monitor binding (see Materials and Methods). The percentage of quenching (- AF) is plotted vs. L - A sconcentration. ~ (0)Wildtype enzyme; ( 0 )[H87A]EcA2; (+) [f1183L]EcA2;).( [H19?L]EcA2.

+0 :

2o 0

i

quenching as a function of aspartate concentration) are shown in Fig. 4. Again, the modified enzymes exhibited essentially normal binding behaviour. The apparent binding constants of [H87A]EcA2 and [HI 83LIEcA2 were increased approximately threefold while that of [H197L]EcA2 was unchanged. Maximal quenching by aspartate approached 5% with all enzyme species examined.

t1 1 4

5

6

7

3

9

10

PH

Fig. 3. pH dependence of L-Asn hydrolysis. Activities wcrc measured a1 37°C in MesiNaOH (pH 5-7) and Tris/HCI buffers (pH 7.5 --9) with 5 mM L-Asn as the substrate. Relative values are shown with the respective maximum activities set to 100%. (0)Wild-type enryme; ( 0 )[H87A]EcA2;(+) [Hl83L]EcA2; ( W ) [H197L]EcA2.

produced a dramatic loss of activity. While [H183L]EcA2 showed almost normal activity, catalytic constants of the other enzymes were only reduced to 10-30% of wild-type EcA2. This was true for all substrates tested, i.e. the relative activities of the enzyme variants towards L-Asn, AHA and L-Gln were comparable to the ratios found with wild-type. Finally, with the exception of HI 83L with a slightly increased K,, none of the mutations had an appreciable effect on the K, for AHA. pH/rate profiles of L-asparagine hydrolysis are shown in Fig. 3. To facilitate comparison, the respective activities at pH 7 were set to 100% in each case. None of the mutations substantially altered the pH dependence of the reaction which had a maximum at approximately pH 7. Two of thc variants exhibited minor shifts of the profile towards higher pH ( < p H 0.5). Aspartate binding The fluorescence of Trp66 is a valuable probe to study aspartate binding to EcA2. Titration curves (i.e. the degree of

Conformational stability

Fig. 5 shows the results of urea denaturation experiments. Free energies of denaturation AG, are plotted vs. urea concentration. The plots are linear to a good approximation, indicating that the underlying simple two-state model is applicable. For wild-type enzyme, and [H87A]EcA2 and [H197L]EcA2, the midpoints of denaturation (at AG, = 0) were different, while extrapolation to urea concentrations of 0 yielded comparable values for LI G(H20) (approximately 50 kJ/mol). [H183L]EcA2 was much more sensitive to urea. Denaturation took place at 1-2 M compared to 3.5 -4.5 M for wild type ;in addition, the estimated energy of stabilization in water, AG(H20) was reduccd to 20 kJ/mol.

DISCUSSlON The results presented here do not support a number of reports inferring that histidine residues in EcA2 are directly associated with catalysis. While at least one of the histidine residues of the enzyme, His87, shows a high degree of evolutionary conservation, none of the mutations seriously impaired its catalytic properties. Alignments of relevant partial sequences are summarized in Fig. 6. His87 is located within an almost completely conserved region spanning residues 81 -96. By contrast, position 183 is highly variable with a histidine residue appearing in EcA2 exclusively. His1 97 is conserved in three high-affinity

479

6o

50 40

5

1

k

Fig. 5. Stabilities of wild-type and variant asparaginases in urea. Apparent free energies of unfolding dG, were derived from fluorescence emission intensities a t various urea concentrations (see Materials and Methods). Extrapolation of the plots to a urea concentration of 0 yields estimates of the respective unfolding energies in water, BG(H,O). (0)Wild-type enzyme; (m)[H87A]EcA2; (+) [H183L]EcA2; (U) [H197L]EcA2.

1457

I ECOA2 AgGA ErcA BsuA

YeaA2 EcoAl

GFVITHGTDTMEETAyFLd GvVITHGTDTMEETAfFLn GvVITHGTDTvEEsAyFLh GFVITHGTDTMayTsaaLs GaVVTHGTDTMEETAfFLd GFYIlHGTDTMayTAsaLs H183

EcoA2 AgGA ErcA BSUA YeaA2 EcoAl

I

H197

I

VNyGpLGyIhngRIdYqRtparkHTsd aqwGaLGtlvEgkpywfRssvKkHTnn neeGYLGvIignrTyYqnridKlHTtr iNypYiafInEdgIeYnkqvtePendt deqGYLGyfsnddvefyyppvKPngwq PNlppL.. .lEagIhirRintpPaphg

Fig. 6. Evolutionary conservation of asparaginase histidine residues. Alignments were made using the program MACAW [21]. EcoA2, E. coli asparaginase IT [22]; AgGA, Acinetabacter glutaminas$icans glutaminase-asparaginase [23]; ErcA, asparaginase from Erwinia chrysanthemi [24]; BsuA, asparaginase from Bacillus subtilis [25]; YeaA2, Yeast asparaginase I1 [26]; EcoAl ,E . coli asparaginase I [27]. Residues appearing in three or more of the sequences are shown in capitals, conservative replacements are indicated by italics.

asparaginases of bacterial origin but absent in all asparaginases with lower affinity, i.e. in the enzymes from Bacillus subtilis and yeast, as well as E. coli asparaginase I. The largely unchanged catalytic properties of our His87 variants came as a surprise since several independent findings suggest that His87 is associated with the active site. First, His87 is an absolutely invariable residue. Second, the failure of His87 to exhibit a sharp C2 resonance in 'H N M R shows that its side chain has restricted mobility. Such behaviour is typical of active-site residues buried within the protein. Third, the fluorescence of Trp66 is partly quenched by protonation of His87 (Fig. 2) and, at the same time, affected by substrate or product binding (Fig. 4). Taken together, these effects suggest that both Trp66 and His87 are located within or near the substrate-binding center. We cannot exclude, however, that

quenching is mediated by long-range interactions in the protein such as a conformational change. In order to characterize the role of His87 in detail, we exchanged it for three different amino acids. In [H87A]EcA2 the histidine imidazolyl group is deleted, in [H87L]EcA2 it is replaced with a non-polar side chain of roughly the same size, and in [H87K]EcA2 the positive charge at position 87 is retained. The fact that none of these replacements caused substantial changes of K , makes it unlikely that the side chain of His87 is in direct contact with the bound substrate or required for substrate binding as proposed by O'Leary and Mattes [8]. Furthermore, the moderate decrease of k,,, observed with all His87 variants excludes a function in catalysis like that of the histidines in the catalytic triad of serine proteases [19]. For instance, the conversion to alanine of a member of such a triad, His64 in Bacillus umyloliquefuciens subtilisin, reduced k,,, by a factor of approximately lo6 1201. If His87 had a similar role in the asparaginases a much more dramatic activity loss and a marked shift of the pH/rate profiles should have resulted from mutagenesis. Our present data do not provide conclusive evidence on the role of His197. Its NMR signal is sharp and does not respond to ligand binding. Both findings indicate that the side chain of His197 is exposed to the solvent. In agreement with this notion, the pK, of its imidazolyl group estimated by NMR titration is close to that of free histidine. The effects of the H197L mutation on both activity and stability in urea are moderate and similar to those found with H87A. A rather unexpected result of the present studies was the identification of His183 as the residue giving rise to the His-a signal in 'H NMR.The disappearance of this resonance upon aspartate binding clearly reflects a substantial change in the environment of His183 when the active site accepts a ligand. However, the normal activity of [H183L]EcA2is incompatible with any direct catalytic role of the imidazolyl group of His183. This discrepancy may be resolved by our finding that replacement of His183 with leucine drastically reduces the stability of the protein (Fig. 5). This is consistent with the assumption that His183 is involved in the assembly of the asparaginase tetramer by mediating intersubunit contacts. In such a case, the environment of His183 should be rather sensitive to conformational changes of the active tetramer. While there is no direct evidence for a conformational change during aspartate binding, minor rearrangements of the subunits may suffice to bring about the effects of aspartate on the 'H N M R signal of His183. A pair of anomalous tyrosine signals also respond to aspartate binding [111. The reason for their behaviour may be the same as with His183. In fact, the environment of one or more tyrosine residues is perturbed during dissociation of tetrameric EcA2 into subunits [17]. We cannot exclude, however, that the still unidentified aspartate-responsive tyrosine residue is directly associated with the active site. The present study was supported by the Deutsche Fnrschungsgerneinschaft (grant Ro 433/9-3 to K . H. R.) and by Public Health Service grant GM 12522 from the National Institute of General Medical Sciences of the USA. We thank J. Henseling for expert technical assistance.

REFERENCES 1. Clavell, L. A,, Gelber, R. D., Cohen, H. J., Hitchcock-Bryan, S., Cassady, J. R., Tarbell, N. J., Blattner, S. R., Tantravahi, R., Leavitt, P. & Sallan, S . E. (1986) Four-agent induction and intensive asparaginase therapy for treatment of childhood acute lymphoblastic leukemia, New Engl. J . Med. 315, 657-663.

2. Ehrman, M., Cedar, H. & Schwartz, J. H. (1971) LAsparaginase I1 of Escherichiu coli. Studies on the enzymatic mcchanisni of action, J. Biol. Chem. 246, 88 - 94. 3. Rohm, K. H. & Van Etten, R. L. (1986) The "0 isotope effect in 3C nuclear magnetic resonance spectroscopy: mechanistic studies on asparaginase from Escherichia coli, Arch. Biochem. Biophys. 244, 128 - 136. 4. Peterson, K.G., Richards, F. F. & Handschumacher, R. E. (1 977) Structure of peptide from active site region of Escherichiu cnti L-asparaginase, J . Bid. Chem. 252,2072 2076. 5. Harms, E., Wehner, A., Aung, H.4'. & Rohm, K. H. (1991) A catalytic role for threonine-12 of E. coli asparaginase I1 as established by site-directed mutagenesis, FEBS Lett. 285, 55 58. 6. Makino. H. & Inada, Y . (1973) Photo-oxidation of histidine residues in asparaginase in relation to its enzymic activity, Biochim. Biophys. Actu 295, 543 - 548. 7. Jayaram, H. N., Cooney, D. A. & Huang, C. Y . (1986) Interaction between L-aspartic acid and L-asparaginase from Escherichia coli: binding and inhibition studies, J . Enzyme Inhibition I , 151-161. 8. O'Leary, M. H. & Mattes, S. L. (1978) pH dependence of the kinetic parameters of L-asparaginase, Biochim. Biophys. Acta 522,238 - 242. 9. Homer, R. B. (1972) The pH dependence and quenching of the fluorescence of asparaginase from Escherichiu coli B, Biochim. Biophys. Acta 278, 395-398. 10. Homer, R. B. & Allsopp, S . R. (1976) An investigation of an unusual histidyl residue in Escheric:hiu coli B L-asparaginase through fluoresccncc quenching, Biochim. Biophys. Actu 434, 100- 109. 11. Bagert, U. & Rohm, K. H. (1989) On the role of histidine and tyrosine residues in E . coli asparaginase. Chemical modification and 'H-nuclear magnetic resonance studies, Biochim. Biophys. Actu 999, 36-41. 12. Harms, E., Wehner, A,, Jennings, M. P., Pugh, K. J., Beacham, I. R. & Rohm, K . H. (1991) Construction ofexpression systems for Eschcrichiu coli asparaginase TI and two-step purification of the recombinant enzyme from periplasmic extracts, Protein Purif. &Expression 2, 144-150. 13. Sayers, J. R., Schmidt, W. & Eckstein, F. (1988) 5'-3'-Exonucleases in phosphorothioate-based oligonucleotide-directed mutagenesis, Nucleic Acids Res. 16, 791 - 802. 14. DiLella, A. G. & Woo, L. C. (1983) Hybridization of genomic DNA to oligonucleotide probes jn the presence of tetramethylammonium chloride, Methods Enzyrnol. 99,447-451. -

-

15. Frear, D. S. & Burrell, R. C. (1955) Spectrophotometric method for determining hydroxylamine reductase activity in highcr plants, Anal. Chem. 27, 1664-1665. 16. Bergmeyer, H. U. (1974) Methoden der enzymatischen Anulyse, 3rd edn, Verlag Chemie, Weinheim. 17. Shifrin, S., Luborsky, S. W. & Grochowski, B. J. (1971) L-Asparaginase from Escherichia coli B. Physicochcmical studies of the dissociation process, J. Biol. Chem. 246, 7708 - 7714. 18. Pace, C. N., Shirley, B. A. and Thomson, J. A. (1989) in Protein structure. Apractical approach (Creighton, T. E., cd.) pp. 31 1 330, IRL Press, Oxford. 19. Blow, D. M, Birkioft, J. J. & Hartley, B. S. (1969) Role of a buried acid group in the mechanism of action of chymotrypsin, Nature 221, 337- 340. 20. Carter, P. &Wells, J. A. (1988) Dissecting the catalytic triad of a serine protease, Nature 332, 564- 568. 21. Schuler, G . D., Altschul, S. F. & Tipman, D. J. (1991) A workbench for multiple alignment construction and analysis, Proteins Struct. Funct. Genet. 9, 180-190. 22. Jennings, M. P. & Beacham, I. R. (1990) Analysis of the Escherichiu coli gene encoding L-asparaginase 11, ansB, and its regulation by cyclic AMP rweptor and FNR proteins, J . Bacteriol. 172, 1491 -1498. 23. Tanaka, S., Robinson, E. A., Appella, E., Miller, M., Ammon, H. L., Roberts, J., Weber, I. T. & Wlodawer, A. (1988) Structure of amidohydrolases. Amino acid sequence of glutaminaseasparaginase from Acinetobacter glutuminusificans, J . Biol. Chem. 263, 8583-8591. 24. Minton, N. P., Bullman, H. M. S., Scaven, M. D., Atkinson, T. & Gilbert, H. J. (1900) Nucleotide sequence of the Erwinia chrysanthemi NCPPB 1066 L-asparaginase gene, Gene (Amst.) 46,25-35. 25. Sun, D. & Setlow, P. (1991) Cloning, nucleotide sequence, and expression of the Bacillus subtilis ans operon, which codes for L-asparaginase and L-aspartase, J . Bacteriol. 173, 3831 -3845. 26. Kim, K. W., Kamerud, J. Q., Livingston, D. M. & Roon, R. J. (1988) Asparaginase 11 of Sacchuromyces cerevisiae. Characterization of the Asp3 gene, J . Biol. Chem. 263, 11 948 - 11953. 27. Jerlstrom, P. G., Bezjak, D. A,, Jennings, M. P. & Beacham, 1. R. (1989) Structure and expression in Escherichia coli K-12 of the L-asparaginase I-encoding ansA gene and its flanking regions, Gene (Amst.) 78, 37-46.

Site-specific mutagenesis of Escherichia coli asparaginase II. None of the three histidine residues is required for catalysis.

Site-specific mutagenesis was used to replace the three histidine residues of Escherichia coli asparaginase II (EcA2) with other amino acids. The foll...
638KB Sizes 0 Downloads 0 Views