Expression, Purification, and Kinetic Characterization - The Journal of ...

5 downloads 0 Views 387KB Size Report
Randy L. Stone$#, Howard ZalkinT, and Jack E. DixonS(1. From the $Department ..... All dpm determinations were made with a Beckman model. LS-3801 liquid ...
Protein Expression and PuriWcation 51 (2007) 227–234 www.elsevier.com/locate/yprep

Expression, puriWcation, and characterization of stable, recombinant human adenylosuccinate lyase Peychii Lee, Roberta F. Colman ¤ Department of Chemistry and Biochemistry, University of Delaware, Newark, DE 19716, USA Received 2 June 2006, and in revised form 5 July 2006 Available online 9 August 2006

Abstract The full length human adenylosuccinate lyase gene was generated by a PCR method using a plasmid encoding a truncated human enzyme as template, and was cloned into a pET-14b vector. Human adenylosuccinate lyase was overexpressed in Escherichia coli Rosetta 2(DE3)pLysS as an N-terminal histidine-tagged protein and was puriWed to homogeneity by a nickel–nitriloacetic acid column at room temperature. The histidine tag was removed from the human enzyme by thrombin digestion and the adenylosuccinate lyase was puriWed by Sephadex G-100 gel Wltration. The histidine-tagged and non-tagged adenylosuccinate lyases exhibit similar values of Vmax and Km for S-AMP. Analytical ultracentrifugation and circular dichroism revealed, respectively, that the histidine-tagged enzyme is in tetrameric form with a molecular weight of 220 kDa and contains predominantly -helical structure. This is the Wrst puriWcation procedure to yield a stable form of human adenylosuccinate lyase. The enzyme is stable for at least 5 days at 25 °C, and upon rapid freezing and thawing. Temperature as well as reducing agent (DTT) play critical roles in determining the stability of the human adenylosuccinate lyase. © 2006 Elsevier Inc. All rights reserved. Keywords: Adenylosuccinate lyase; ADL deWciency; Human adenylosuccinate lyase; His-tagged enzyme

Adenylosuccinate lyase (E.C. 4.3.2.2) catalyzes two reactions in the purine nucleotide biosynthesis pathway: the conversion of succinylaminoimidazole carboxamide ribotide (SAICAR)1 to aminoimidazole carboxamide ribotide (AICAR) and fumarate and the conversion of adenylosuccinate (S-AMP) to AMP and fumarate. The gene encoding adenylosuccinate lyase has been isolated from many sources including Escherichia coli, Bacillus subtilis, Pyrobaculum aerophilum, Thermotoga maritima, avian, mouse, and human [1–7]. The human gene is located in chromosome 22 [8–10]. The human gene, encoding a truncated ade-

*

Corresponding author. Fax: +1 302 831 6335. E-mail address: [email protected] (R.F. Colman). 1 Abbreviations used: SAICAR, succinylaminoimidazole carboxamide ribotide; AICAR, aminoimidazole carboxamide ribotide; SAICAr, succinylaminoimidazole carboxamide riboside; S-Ado, succinyladenosine; S-AMP, adenylosuccinate acid; DTT, dithiothreitol; DTNB, 5,5⬘-dithiobis(2-nitrobenzoic acid); E. coli, Escherichia coli; B. subtilis, Bacillus subtilis; Ni–NTA column, nickel–nitrilotriacetic acid column. 1046-5928/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2006.07.023

nylosuccinate lyase (459 amino acids) was Wrst isolated by Stone et al. [7], from human liver using avian liver cDNA as hybridization probe. The full length human adenylosuccinate lyase gene was later isolated by Fon et al. [10]. Adenylosuccinate lyase deWciency in human is characterized by the accumulation of dephosphorylated enzyme substrates (succinylaminoimidazolecarboxamide riboside; SAICAr and succinyladenosine; S-Ado) in patients’ cerebrospinal Xuid, plasma and urine, while no detectable level of the two compounds is observed in normal humans [11]. Clinically, patients with adenylosuccinate lyase deWciency are characterized by autism, mild to severe mental retardation, muscle wasting and epilepsy [11,12]. The major obstacle to systematically studying adenylosuccinate lyase and its roles in human disease has been producing suYcient active human enzymes (either wild-type or mutant enzymes). DiYculties in expressing active “truncated” human adenylosuccinate lyase in E. coli have been documented [13,14]. Catalytically active adenylosuccinate lyase could be obtained by expressing the enzyme as a

228

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234

fusion protein with thioredoxin or maltose binding protein [13,14]. However, complete separation of the aYnity tags from human adenylosuccinate lyase was not achieved. Thioredoxin-human adenylosuccinate lyase was only stable for hours in ice and the enzyme lost its activity after storage at either ¡20 °C or ¡80 °C. Instability of the puriWed enzyme further compromised the studies of the enzyme and its roles in human disease. Here, we report the cloning of full length human adenylosuccinate lyase into a pET-14b vector. The human enzyme was overexpressed as a 6-histidine-tagged protein in E. coli and was puriWed to homogeneity by a single nickel column. After thrombin digestion, the 6-histidine tag was separated from the human enzyme by gel Wltration chromatography. Finally, the initial characterization of the recombinant human enzyme is also presented in this paper.

(PCR) using plasmid pT7-7ASL2 as PCR template. For PCR, we purchased from Biosynthesis, Inc. a 119-base forward primer containing a 5⬘-end NdeI restriction site as well as 75-bases encoding the 25 amino acids missing from the pT7-7ASL (Met1-Glu25), and a 46-base reverse primer, which introduces a BlpI restriction site to the 3⬘-end of the human gene. pET-14b, a vector designed to produce target proteins with a thrombin cleavable N-terminal histidine tag was chosen for this study. The 1.5 kb PCR products were digested with NdeI and BlpI and were ligated into the 4.6 kb DNA fragment of pET-14b, which had previously been digested with NdeI and BlpI. The ligated DNA was used to transform E. coli TB1 cells. The clones containing the correct full length human adenylosuccinate lyase gene (pETN25HASL) were conWrmed by BigDye terminator cycle sequencing (DNA Sequencing and Genotyping Center, University of Delaware).

Materials and methods Expression and puriWcation of human adenylosuccinate lyase Materials pET-14b vector, E. coli Rosetta 2(DE3)pLysS and E. coli BL21(DE3)pLysS competent cells were purchased from EMD Bioscience (La Jolla, CA). Restriction enzymes, NdeI and BlpI, Vent DNA polymerase, T4 DNA ligase, dNTPs, and E. coli TB1 strain were obtained from New England Biolabs (Ipswich, MA). QuikChange XL sitedirected mutagenesis kit was purchased from Stratagene (La Jolla, CA). Ni–NTA resin, QIAprep spin miniprep kit, QIAquick PCR puriWcation kit, and QIAquick gel extraction kit were from Qiagen Sciences (Germantown, MD). Potassium monophosphate, potassium diphosphate, potassium chloride, agarose, EDTA, and LB broth were from Fisher ScientiWc (Pittsburgh, PA). S-AMP, DTT, imidazole, DTNB, deoxyribonuclease I from bovine pancreas (DNase), ribonuclease A from bovine pancreas (RNase), and Sephadex G-100 resin were purchased from SigmaAldrich (St. Louis, MO). Oligonucleotides were from Biosynthesis, Inc (Lewisville, TX). Human -thrombin was obtained from Enzyme Research Laboratories (South Bend, IN). Vector construction pT7-7ASL [7], which encodes a truncated human adenylosuccinate lyase gene (Met26-Leu484),2 contains an internal NdeI restriction site located around Pro471. In order to facilitate the cloning of full length DNA into pET14b vector, QuikChange XL site-directed mutagenesis kit was used to abolish the internal NdeI restriction site without aVecting the proline residue (CCATATG ! CCG TATG) (pT7-7ASL2). The full length human enzyme gene was generated by a polymerase chain reaction method

2 Numbering of amino acids is based on full length human adenylosuccinate lyase.

In order to overexpress the human enzyme in E. coli, pETN25HASL was transformed into E. coli Rosetta 2(DE3)pLysS. E. coli Rosetta 2(DE3) pLysS containing pETN25HASL was Wrst grown in LB broth at 37 °C until cell density (OD600) reached 0.4–0.6. Then, the cell culture was cooled to 25 °C and the human enzyme expression was induced with 0.4 mM IPTG at 25 °C overnight. The cell pellet (»3.7 g wet cell pellet per 2 L cell culture) from the overnight expressed cell culture was freeze– thawed at least twice to rupture the bacterial cell membrane and to allow lysozyme, produced by E. coli Rosetta 2(DE3)pLysS, to degrade the bacterial cell wall. The cell pellet was resuspended in cell lysis buVer (50 mM potassium phosphate buVer, pH 8.0, containing 300 mM KCl and 10% glycerol) based on 7 mL lysis buVer per gram of wet cell pellet. DNase (1 mg/mL) and RNase (10 mg/mL) were Wrst dissolved in the same cell lysis buVer; then, 0.1% (v/v) DNase and 0.1% (v/v) RNase were added to reduce the viscosity of the cell lysate. The soluble protein fraction (crude cell lysate) was separated from the cell debris by centrifugation at 14,000 rpm and was loaded on a Ni–NTA column (»12 mL resin per 2 L cell culture), which had been previously equilibrated with lysis buVer. Then, the Ni–NTA column was washed with 200 mL lysis buVer, followed by a second wash with 200 mL lysis buVer containing 20 mM imidazole to remove any loosely bound protein. The histidine-tagged human enzyme was eluted from the column by a gradient of 150 mL each of lysis buVer containing, respectively, 20 mM imidazole and 250 mM imidazole. The fractions with high adenylosuccinate lyase activity were pooled and concentrated to 1/10 its original volume using Amicon Ultra-15 (MWCO 10,000) and an Eppendorf centrifuge (model 5810R). The concentrated enzyme was dialyzed against enzyme storage buVer (50 mM potassium phosphate buVer, pH 7.0, containing 150 mM KCl, 1 mM DTT, 1 mM EDTA, and 10% glycerol). The entire puriWcation was conducted at room temperature.

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234

Cleavage and separation of 6-histidine tag from human adenylosuccinate lyase The human adenylosuccinate lyase was Wrst expressed as an N-terminal 6-histidine tagged recombinant protein containing a thrombin cleavage site (Leu-Val-Pro-Arg-Gly-SerHis, where thrombin cleaves at the bond between Arg and Gly) between the tag and the human enzyme. After thrombin cleavage only three extra amino acids (Gly-Ser-His) remain before the start of the normal human adenylosuccinate lyase sequence (Met-Ala-Ala). Thrombin, a serine protease containing four disulWde bonds [15], is completely inactive in our enzyme storage buVer. In order to remove DTT from the storage buVer, puriWed adenylosuccinate lyase was Wrst dialyzed against 50 mM potassium phosphate buVer, pH 8.0, containing 300 mM KCl and 10% glycerol. Thrombin (10 units/mg protein) was added to the enzyme solution and thrombin digestion was conducted at room temperature overnight. Separation of the thrombin and histidine-tag from the human enzyme was achieved by chromatographing the digestion mixture on a Sephadex G-100 column which had previously been equilibrated with 50 mM potassium phosphate buVer, pH 7.0, containing 150 mM KCl, 1 mM EDTA, 1 mM DTT, and 10% glycerol (storage buVer). Fractions with high enzyme activity and high purity on SDS–PAGE were pooled, concentrated to 1/10 its original volume as described in protein puriWcation, and stored in a freezer at ¡80 °C. Purity of recombinant human adenylosuccinate lyase The purity of adenylosuccinate lyase (either tagged or untagged form) was determined both by SDS–PAGE and N-terminal sequencing (Applied Biosystems gas-phase sequenator, Model Procise). Adenylosuccinate lyase activity assay The enzyme activity toward the substrate S-AMP was followed by the decrease in absorbance at 282nm as S-AMP is converted to AMP, and the activity was calculated using a diVerence extinction coeYcient of 10,000 M¡1 cm¡1 [16]. The standard assay solution contains 60M S-AMP in 50mM HEPES, pH 7.4, in a total volume of 1 mL. To determine the ratio of adenylosuccinate lyase activity toward SAICAR as compared to S-AMP, both SAICAR and S-AMP concentrations were Wxed at 90 M in 50 mM HEPES, pH 7.4. S-AMP at concentrations higher than 150 M exhibits a high absorbance at 282 nm, which complicates the determination of enzyme activity. Alternatively, the adenylosuccinate lyase activity was monitored at 290 nm and its activity was calculated using a diVerence extinction coeYcient of 4050 M¡1cm¡1 in an assay solution containing high S-AMP concentrations [17]. pH proWle of human adenylosuccinate lyase The pH proWle of the recombinant enzyme was determined by measuring the Vmax of the enzyme at various pHs.

229

The buVers used to obtain the desired pH were MES (pH 5.2–6.8), HEPES (pH 6.6–8.4), and TAPS (pH 7.4–9.2). Constant anion concentration (30 mM) was maintained in all buVer systems. The S-AMP in the assay solution was Wxed at 300 M. Free thiol group determination of human adenylosuccinate lyase The quantity of free thiols in adenylosuccinate lyase was determined by a modiWcation of the method of Poole [18]. Adenylosuccinate lyase (either active or inactive wild-type) was Wrst dialyzed against 50 mM sodium phosphate buVer, pH 8, containing 300 mM NaCl. The enzyme was mixed with 10% SDS in the same dialysis buVer to give a Wnal concentration of 1% SDS. The whole mixture was incubated at room temperature for 3 min; at the end of this time, a Wnal concentration of 1 mM DTNB was added to the reaction mixture and the solution was further incubated at room temperature for two more minutes. The free thiol content in adenylosuccinate lyase was determined from an increase in absorbance at 412 nm, using an extinction coeYcient of 14,150 M¡1 cm¡1 [19]. Secondary structure of human adenylosuccinate lyase Circular dichroism was used to determine the secondary structure of human adenylosuccinate lyase. PuriWed adenylosuccinate lyase (0.3–0.4 mg/mL) (either active or inactive form) was Wrst dialyzed against 50 mM potassium phosphate buVer, pH 8, containing 300 mM KCl, 10% glycerol and was used for CD measurement immediately. The enzymes were pipetted into a 0.1 cm path length quartz cell. The ellipticity of each sample was measured on Aviv Circular Dichroism Spectrometer, Model 215, by scanning the sample three times between 200 and 250 nm at 2 nm increments. The molar ellipticity, [], was obtained by using the equation: [] D /(10nCl), where  is the measured ellipticity in millidegrees, n is the number of amino acids per subunit (503 for the tagged human adenylosuccinate lyase), C is the molar concentration of the enzyme, and l is the cell path length in centimeters. Molecular weight assessment of human adenylosuccinate lyase Analytical ultracentrifugation was used to assess the oligomeric state of the recombinant human enzyme. Adenylosuccinate lyase in storage buVer was Wrst dialyzed against 50 mM potassium phosphate buVer, pH 7.0, containing 150 mM KCl, 1 mM EDTA, 0.1 mM DTT, and 10% glycerol. Enzyme (0.2–0.7 mg/mL) was loaded into an An-50Ti analytical rotor. Sedimentation equilibrium experiments were conducted at 25 °C on a Beckman Coulter ProteomeLab™ XL-I analytical ultracentrifuge, using speeds of 8000, 9000, and 11,000 rpm. The molecular weight of adenylosuccinate lyase was calculated using IGOR

230

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234

computer program (Wavemetrics, Inc.) as described previously [20,21]. Results and discussion Expression of recombinant truncated human adenylosuccinate lyase Originally, we cloned the truncated human adenylosuccinate lyase gene into pET-14b vector and expressed the N-terminal histidine-tagged recombinant protein (Met26Leu484) in E. coli BL21(DE3)pLysS. We were able to obtain high purity recombinant proteins as determined by SDS–PAGE and N-terminal sequencing (data not shown) and achieved a good protein yield (»10–15 mg/L cell culture) after a single Ni–NTA column chromatography. However, the truncated enzyme exhibited no detectable enzymatic activity. Various approaches such as changing the bacterial expression strains [E. coli BL21(DE3)pLysS, Rosetta 2(DE3)pLysS, and Rosetta-gami B(DE3)pLysS], cell membrane disruption method (lysozyme, French press, sonication), buVer systems (sodium phosphate, pH 8, potassium phosphate, pH 8, Tris chloride, pH 8), concentration of IPTG (from 50 M to 0.4 mM), adding reducing agents (DTT, -mercaptoethanol), or varying the puriWcation temperature (25 °C or 4 °C) did not yield active enzyme. Furthermore, the enzyme did not regain its activity after removal of the histidine tag by thrombin cleavage. Renaturation of guanidine hydrochloride-denatured enzyme under favorable refolding conditions [22– 25] also did not give active human enzyme. Placement of the 6-histidine tag at the C-terminal end of the enzyme not only produced an inactive human enzyme but gave recombinant protein which bound poorly to a Ni–NTA column and was eluted at low imidazole concentration (10 mM). The results suggest that the C-terminal 6-histidine tag is not well exposed to the environment. We also tried to express the truncated human enzyme directly from pT7-7ASL and puriWed the enzyme using Q Sepharose fast Xow column (Sigma-Aldrich) and Blue Sepharose 6 fast Xow column (Amersham Bioscience) [26,27]. Minute amounts of truncated human enzyme exhibiting low speciWc activity were obtained (»0.2 mg/L cell culture). Our diYculties in expressing suYcient active truncated adenylosuccinate lyase in E. coli are consistent with those reported by Race et al. [14]. Additionally, point mutations at the extra N-terminal 25 amino acids (M1L, A2V, and A3P) had been identiWed in adenylosuccinate lyase deWciency patients with severe mental retardation [14,28,29], suggesting that the N-terminal residues may play a critical role in human enzyme function. This result prompted us to reintroduce the extra 25 amino acids to our truncated human enzyme. The full length adenylosuccinate lyase gene was generated by PCR and was cloned into pET-14b vector (pETN25HASL). pETN25HASL was transformed into either E. coli BL21(DE3)pLysS or E. coli Rosetta 2(DE3)

pLysS to evaluate protein expression in various bacterial stains. E. coli Rosetta 2(DE3)pLysS contains extra t-RNAs to prevent codon usage bias and to enhance mammalian protein expression in E. coli. Under the same inducing conditions, human adenylosuccinate lyase was expressed better in E. coli Rosetta 2(DE3)pLysS than in E. coli BL21(DE3)pLysS, as revealed by SDS–PAGE of crude cell lysate (data not shown). Therefore, subsequent recombinant protein was expressed in E. coli Rosetta 2(DE3)pLysS. Expression and puriWcation of full length human adenylosuccinate lyase using Ni–NTA chromatography In the presence of 0.4 mM IPTG, human adenylosuccinate lyase was overexpressed in E. coli Rosetta 2(DE3)pLysS (Fig. 1, lane 2). The crude cell lysate was loaded on a Ni–NTA column at room temperature, followed by washing the column with lysis buVer and lysis buVer containing 20 mM imidazole. The human enzyme was eluted from the Ni–NTA column by a gradient of 20 mM to 250 mM imidazole. A summary of the puriWcation of the human enzyme is listed in Table 1. Usually the protein yield is about 15 mg protein per liter cell culture. The adenylosuccinate lyase eluted by an imidazole gradient exhibits a single band with a molecular weight of 57 kDa on SDS–PAGE (Fig. 1, lane 3), indicating that single Ni–NTA column chromatography is suYcient to separate the human enzyme from those of unwanted bacterial proteins. The purity of the human enzyme is further conWrmed by the presence of the correct residues by N-terminal amino acid sequencing (Gly-Ser-Ser-HisHis- His -His-His-His-Ser-Ser-Gly-Leu-Val-Pro-Arg-Gly). Additional ly, the N-terminal sequence reveals that the Wrst methionine (fMet) was removed by E. coli. The puriWed human enzyme has a speciWc activity of 3.6 mol/

97 kDa 66 kDa

45 kDa

30 kDa

Fig. 1. PuriWcation of human adenylosuccinate lyase. Lane 1, molecular weight markers, which contain carbonic anhydrase (30 kDa), ovalbumin (45 kDa), albumin (66 kDa), and phosphorylase b (97 kDa). Lane 2, crude cell lysate. Lane 3, puriWed human adenylosuccinate lyase.

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234

231

Table 1 Summary of histidine-tagged human adenylosuccinate lyase puriWcation Step

Total protein (mg)

Total enzyme units

Yield of enzyme unit (%)

SpeciWc activity (mol/min/mg)

Crude cell lysatea Imidazole gradient elution Final puriWed enzyme

434 34 31

172 124 114

100 72 66

0.4 3.6 3.6

From 2 L of 0.4 mM IPTG induced E. coli Rosetta 2(DE3)pLysS containing pETN25HASL plasmid.

min/mg, which is somewhat higher than that of B. subtilis recombinant histidine-tagged enzyme (2 mol/min/mg) [30]. Determination of free –SH group of human adenylosuccinate lyase Initially, we attempted to purify the human enzyme at 4 °C in the absence of DTT. The starting cell lysate exhibited a reasonable number of total enzyme units and we were able to purify the human enzyme to homogeneity with a good protein yield using Ni–NTA chromatography (data not shown). However, the resulting enzyme exhibited very low speciWc activity (0.03 mol/min/mg). In contrast, enzyme which was puriWed at 25 °C and was dialyzed against storage buVer containing 1 mM DTT after it was eluted from the Ni– NTA column, exhibits a speciWc activity of 3.6 mol/min/mg. The active enzyme has 12 free cysteines per enzyme subunit, as revealed by reaction with DTNB, which is close to the 13 cysteines per enzyme subunit determined from its amino acid sequence. In contrast, the full length enzyme with low speciWc activity contains only one free thiol group per enzyme subunit. The loss of adenylosuccinate lyase activity during puriWcation under oxidizing conditions at 4 °C may be due to the random formation of disulWde bonds among the cysteine residues in the enzyme (oxidized enzyme). The human adenylosuccinate lyase was initially eluted from the Ni–NTA column and was collected directly into empty test tubes. Precipitates were always observed in the elution fractions and the precipitates also exhibited a single band of similar molecular weight as the puriWed soluble human enzyme when tested on SDS–PAGE. The importance of maintaining reducing conditions prompted us to modify the puriWcation procedure by collecting the column fractions in tubes containing 10 mM DTT in 50 mM potassium phosphate, pH 7, containing 150 mM KCl, 1 mM EDTA, and 10% glycerol in each imidazole gradient elution collecting tubes. The Wnal concentration of DTT in each elution fraction is 1 mM. The earlier addition of DTT to the enzyme preparation greatly increases the Wnal protein yield to 40 mg protein per liter cell culture of mutant human adenylosuccinate lyases which exhibit comparable enzymatic activity to that of wild-type enzyme (unpublished data). Secondary structures of human adenylosuccinate lyase

family enzymes such as aspartase, fumarase, argininosuccinate lyase, and -crystallin, all contain extensive -helical structure [2,3,30,31]. The CD spectrum of active recombinant human enzyme exhibiting a predominant -helical structure suggests that expression of the human enzyme as a histidine-tagged protein does not alter its secondary structure (Fig. 2). On the other hand, the observed lower molar ellipticity of the oxidized enzyme (Fig. 2) indicates that the oxidation of the enzyme’s thiol groups decreases its characteristic -helical structure. Reactivation of oxidized human adenylosuccinate lyase In order to examine whether inactive adenylosuccinate lyase is capable of regaining its enzymatic activity, oxidized enzyme was incubated in buVers containing various combinations of DTT and EDTA. As shown in Fig. 3A, the speciWc activity of the oxidized human enzyme (0.03 mol/ min/mg) when incubated at 25 °C with DTT plus EDTA is increased 53-fold and with DTT alone is increased 50-fold to yield a speciWc activity of 1.6 mol/min/mg and 1.5 mol/ min/mg, respectively. On the other hand, the oxidized enzyme only slightly regains its activity when it was incubated under the similar buVer conditions but at 4 °C (Fig. 3B). The preference for milder room temperature was also observed for adenylosuccinate lyase isolated from rabbit muscle [27]. Studies conducted in this laboratory on B. subtilis adenylosuccinate lyase indicated that the weight average molecular weight of the bacterial enzyme decreases at low temperature (Ariyananda and Colman, unpublished data). Addition of EDTA to enzyme storage buVers 5000

0

Molar ellipticity

a

-5000

-10000

-15000

-20000 200

210

220

230

240

250

Wavelength (nm)

Although the crystal structure of human adenylosuccinate lyase has not yet been determined, adenylosuccinate lyases from bacteria, together with other fumarase super-

Fig. 2. Circular dichroism spectra of human adenylosuccinate lyases: enzyme with a speciWc activity of 3.6 mol/min/mg (䉭) and enzyme with a speciWc activity of 0.024 mol/min/mg (䊊).

232

Specific Activity (μmol/min/mg)

A

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234 3

97 kDa 2

66 kDa

1

45 kDa

0 0

5

100

150

200

250

30 kDa

Time (hr)

Specific Activity (μmol/min/mg)

B

0.5

0.4

Fig. 4. Thrombin cleavage of human adenylosuccinate lyase. Lane 1, molecular weight markers. Lane 2, histidine-tagged enzyme. Lane 3, thrombin cleaved and gel Wltration puriWed human enzyme.

0.3

0.2

prevents metal-catalyzed oxidation of DTT and extends the lifetime of DTT solutions. However, it appears that DTT at a concentration of 1 mM in the reactivation buVer is suYcient to maintain enzyme active for 240 h at 25 °C in the absence of EDTA. Our results suggest that temperature as well as reducing agent (DTT) play critical roles in determining the activity of human adenylosuccinate lyase. The observation that DTT and EDTA only partially restore the activity of the oxidized human enzyme implies that some irreversible denaturation other than oxidation of thiols in the human enzyme occurred during puriWcation.

sequence is about 57 kDa. Analytical ultracentrifugation reveals that the histidine-tagged enzyme exhibits a molecular weight of 222 kDa, indicating that the native enzyme is in a tetrameric form. Because of the molecular weight diVerence among the histidine tag (2 kDa), thrombin (37 kDa) and non-tagged adenylosuccinate lyase (220 kDa), the histidine tag and thrombin can be separated from the non-tagged enzyme by gel Wltration. Fig. 4 shows that the gel Wltration-puriWed non-tagged human enzyme (lane 3) exhibited a lower molecular weight on SDS–PAGE as compared to that of histidine-tagged enzyme (lane 2), suggesting the removal of the histidine tag from the human enzyme by thrombin cleavage. The N-terminal sequencing of the puriWed enzyme (same preparation as Fig. 4, lane 3) reveals the presence of three extra amino acids from the vector (Gly-Ser-His) followed by the human adenylosuccinate lyase amino acid sequence: Gly-Ser-His-Met-Ala-Ala-GlyGly-Asp-His-Gly-Ser-Pro-Asp. These results conWrm the complete separation of thrombin and histidine tag from human adenylosuccinate lyase by gel Wltration. In contrast to the human enzyme fused to a maltose binding protein, which lost half of its original activity after protease cleavage [13], no loss of enzymatic activity after thrombin cleavage is observed in our enzyme preparations. The puriWed non-tagged human enzyme exhibits a speciWc activity of 3.8 § 0.1 mol/min/mg, which is similar to that of uncleaved enzyme (3.6 § 0.3 mol/min/mg).

Cleavage and separation of histidine-tag from adenylosuccinate lyase

Characterization of recombinant human adenylosuccinate lyase

The calculated molecular weight of the histidine-tagged adenylosuccinate lyase monomer based on its amino acid

One of the obstacles to studying human adenylosuccinate lyase in vitro has been the instability of the puriWed enzyme

0.1

0.0 0

20

40

60

80

100

120

140

160

180

Time (hr)

Fig. 3. Reactivation of oxidized human adenylosuccinate lyase. (A) Enzyme was incubated at 25 °C in either 50 mM potassium phosphate buVer, pH 8.0, containing 300 mM KCl and 10% glycerol (buVer A) (䊊) or buVer A plus 1 mM EDTA () or buVer A plus 1 mM DTT (䊐) or buVer A plus 1 mM EDTA and 1 mM DTT (䉭). (B) Enzyme was incubated at 4 °C in either 50 mM potassium phosphate buVer, pH 8.0, containing 300 mM KCl and 10% glycerol (buVer A) (o) or buVer A plus 1 mM EDTA () or buVer A plus 1 mM DTT (䊐) or buVer A plus 1 mM EDTA and 1 mM DTT (䉭).

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234

0.8

Et/E0

0.6 0.4

0.2

0.1 0

20

40

60

80

100

120

140

Time (hr)

Fig. 5. Stability of human adenylosuccinate lyases (䊊: histidine-tagged enzyme; 䉭: non-tagged enzyme). The human enzymes in 50 mM potassium phosphate, pH 7.0, containing 150 mM KCl, 1 mM EDTA, 1 mM DTT, and 10% glycerol were incubated at 25 °C. At various times, small amounts of sample were withdrawn and the enzyme activities were assayed using standard assay method as described in Materials and methods. E0 is the enzyme activity at incubation time zero and Et is the enzyme activity at various incubation time.

[14]. We evaluated the stabilities of histidine-tagged and non-tagged human adenylosuccinate lyase by incubating the enzymes in enzyme storage buVer containing DTT and EDTA at 25 °C. Fig. 5 shows that the human enzyme does not lose its activity during 5 days incubation in enzyme storage buVer at 25 °C. Additionally, the human adenylosuccinate lyase remained fully active after three-cycles of freeze and thaw as long as the freezing (¡80 °C) and thawing (25 °C) were conducted rapidly on small enzyme aliquots. The results imply the importance of the presence of reducing agent in the storage buVer as well as of the temperature in determining the stability of human adenylosuccinate lyase. The kinetic parameters were determined for histidinetagged and non-tagged adenylosuccinate lyase. Both enzymes have similar apparent aYnities for the substrate S-AMP and have comparable Vmaxs (Table 2). The results indicate that the presence of the N-terminal histidine tag in human enzyme does not compromise its enzymatic functions. Thus, we used recombinant histidine-tagged human enzyme in our subsequent characterization experiments. Histidine-tagged human adenylosuccinate lyase exhibits similar apparent aYnity for SAICAR (Km D 1.33 § 0.38 M) to that for S-AMP (Km D 1.46 § 0.29 M); however, the human enzyme shows a higher activity using SAICAR as a substrate (Vmax D 5.65 § 0.18 mol/min/mg) than it does for S-AMP (Vmax D 3.88 § 0.02 mol/min/mg). Adenylosuccinate lyase deWciency patients accumulate dephosphorylated products of SAICAR and S-AMP in their body Table 2 Km values for S-AMP and Vmax values of histidine-tagged and non-tagged human adenylosuccinate lyases Enzyme

Km (S-AMP) (M)

Histidine-tagged Non-tagged

1.46 § 0.29 1.78 § 0.05

Vmax (mol/min/mg) 3.88 § 0.02 3.88 § 0.07

Xuids [11,12]. The ratio of the reaction rates of SAICAR and S-AMP with adenylosuccinate lyase can be used to predict the severity of the disease [11]. Our previous studies on mutant B. subtilis adenylosuccinate lyases, which mimic the human enzymes isolated from severe adenylosuccinate lyase deWciency patients, exhibited diVerent ratios of Vmax for SAICAR to Vmax for S-AMP as compared to that of wild-type bacterial enzyme [32,33]. The ratio of activities toward SAICAR as compared to S-AMP of the human enzyme, determined in assay solutions containing 90 M of either SAICAR or S-AMP, is 1.27. The more eYcient conversion of SAICAR than of S-AMP by the human adenylosuccinate lyase from our preparation is consistent with those of thioredoxin-human adenylosuccinate lyase fusion protein, as well as B. subtilis enzyme [14,32], but contrasts with those of truncated human enzyme [26] and maltose binding protein-human enzyme fusion protein [13]. Fig. 6 shows the pH dependence of Vmax of histidinetagged human adenylosuccinate lyase. The human enzyme exhibits high enzyme activity at pH values ranging from 7.2 to 8.4. The activity of the human enzyme drops substantially at pH below 7.2 with a pK1 of 6.4, which is slightly lower than the reported value of B. subtilis enzyme p(K1 D 6.78) [32]. However, in the alkaline range, the pH dependence of Vmax of the human enzyme shows a remarkable diVerence from that of B. subtilis adenylosuccinate lyase. The Vmax of the human enzyme decreases slightly at pH above 8.4 while the bacterial enzyme loses most of its activity at pH above 8.6. The calculated pK2 for the human enzyme is 9.5, which is substantially higher than that of the bacterial enzyme (pK2 D 8.37). Considering that human adenylosuccinate lyase shares only 25% amino acid sequence identity with the B. subtilis enzyme, the catalytically important ionizable amino acids in the human enzyme may be diVerent from those in the bacterial enzyme even though the two enzymes exhibit similarities in their secondary structures and substrate preferences. 6

5

4

Vmax

1.0

233

3

2

1

0 5

6

7

8

9

10

pH

Fig. 6. pH dependence of Vmax of human adenylosuccinate lyase. The enzyme activity was determined at various pHs in an assay solution containing 300 M S-AMP as described in Materials and methods.

234

P. Lee, R.F. Colman / Protein Expression and PuriWcation 51 (2007) 227–234

In conclusion, we demonstrate in this paper the cloning, expression and puriWcation of catalytically active human adenylosuccinate lyase. We revealed that human enzyme is fully active in the presence of reducing agent and at 25 °C. The enzyme is stable when it is stored in a potassium phosphate buVer containing KCl, DTT, EDTA, and glycerol at ¡80 °C. This is the Wrst puriWcation procedure to yield a stable form of human adenylosuccinate lyase. These studies lay the groundwork for our further investigation of mutant enzymes and their roles in adenylosuccinate lyase deWciency, as well as in determination of the enzyme’s structure by X-ray crystallography.

[14]

[15]

[16]

[17]

Acknowledgments [18]

We thank Dr. J.E. Dixon for providing us pT7-7ASL, Dr. Y.-C. Huang for N-terminal amino acid sequencing, Dr. S. Misquitta for her help in analytical ultracentrifugation, and S. Sivendran for providing us SAICAR. References [1] D.J. Ebbole, H. Zalkin, Cloning and characterization of a 12-gene cluster from Bacillus subtilis encoding nine enzymes for de novo purine nucleotide synthesis, J. Biol. Chem. 262 (1987) 8274–8287. [2] E.A. Toth, C. Worby, J.E. Dixon, E.R. Goedken, S. Marqusee, T.O. Yeates, The crystal structure of adenylosuccinate lyase from Pyrobaculum aerophilum reveals an intracellular protein with three disulWde bonds, J. Mol. Biol. 301 (2000) 433–450. [3] E.A. Toth, T.O. Yeates, The structure of adenylosuccinate lyase, an enzyme with dual activity in the de novo purine biosynthetic pathway, Structure 8 (2000) 163–174. [4] J. Aimi, J. Badylak, J. Williams, Z. Chen, H. Zalkin, J.E. Dixon, Cloning of a cDNA encoding adenylosuccinate lyase by functional complementation in Escherichia coli, J. Biol. Chem. 265 (1990) 9011–9014. [5] L.-J.C. Wang, W.E. O’Brien, Characterization of the cDNA and the gene encoding murine adenylosuccinate lyase, Genomics 28 (1995) 341–343. [6] M.R. Redinbo, S.M. Eide, R.L. Stone, J.E. Dixon, T.O. Yeates, Crystallization and preliminary structural analysis of Bacillus subtilis adenylosuccinate lyase, an enzyme implicated in infantile autisum, Protein Sci. 5 (1996) 786–788. [7] R.L. Stone, J. Aimi, B.A. Barshop, J. Jacken, G. Van den Berghe, H. Zalkin, J.E. Dixon, A mutation in adenylosuccinate lyase associated with mental retardation and autistic features, Nat. Genet. 1 (1992) 59–63. [8] M.L. Van Keuren, I.M. Hart, F.T. Kao, R.L. Neve, G.A. Bruns, D.M. Kurnit, D. Patterson, A somatic cell hybrid with a single human chromosome 22 corrects the defect in the CHO mutant (Ade-1) lacking adenylosuccinase activity, Cytogenet. Cell Genet. 44 (1987) 142–147. [9] O. Delattre, C.J. Azambuja, A. Aurias, J. Zucman, M. Peter, F. Zhang, M.C. Hors-Cayla, G. Rouleau, G. Thomas, Mapping of human chromosome 22 with a panel of somatic cell hybrids, Genomics 9 (1991) 721–727. [10] E.A. Fon, S. Demczuk, O. Delattre, G. Thomas, G.A. Rouleau, Mapping of human adenylosuccinate lyase (ADSL) gene to chromosome 22q13.1 ! q13.2, Cytogenet. Cell Genet. 64 (1993) 201–203. [11] G. Van den Berghe, J. Jaeken, Adenylosuccinate lyase deWciency, in: 8th ed., C.P. Scriver, A.L. Beaudet, W.S. Sly, D. Valle, B. Childs, K.W. Kinzler, B. Vogelstein (Eds.), The Metabolic and Molecular Basis of Inherited Diseases, vol. II, McGraw-Hill, N.Y., 2001, pp. 2653–2662. [12] E.K. Spiegel, R.F. Colman, D. Patterson, Adenylosuccinate lyase deWciency, Mol. Genet. Met. 89 (2006) 19–31. [13] S. Kmoch, H. Hartmannova, B. Stiburkova, J. Krijt, M. Zikanova, I. Sebesta, Human adenylosuccinate lyase (ADSL), cloning and charac-

[19]

[20]

[21]

[22] [23]

[24]

[25] [26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

terization of full-length cDNA and its isoform, gene structure and molecular basis for ADSL deWciency in six patients, Hum. Mol. Genet. 9 (2000) 1501–1513. V. Race, S. Marie, M.-F. Vincent, G. Van den Berghe, Clinical, biochemical and molecular genetic correlations in adenylosuccinate lyase deWciency, Hum. Mol. Genet. 9 (2000) 2159–2165. R.J. Butkowski, J. Elion, M.R. Downing, K.G. Mann, Primary structure of human prethrombin 2 and -thrombin, J. Biol. Chem. 252 (1977) 4942–4957. T.T. Lee, C. Worby, J.E. Dixon, R.F. Colman, IdentiWcation of His141 in the active site of Bacillus subtilis adenylosuccinate lyase by aYnity labeling with 6-(4-bromo-2,3-dioxobutyl)thioadenosine 5⬘-monophosphate, J. Biol. Chem. 272 (1997) 458–465. J.B. Palenchar, R.F. Colman, Characterization of a mutant Bacillus subtilis adenylosuccinate lyase equivalent to a mutant enzyme found in human adenylosuccinate lyase deWciency: asparagine 276 plays an important structural role, Biochemistry 42 (2003) 1831–1841. L.B. Poole, Flavin-dependent alkyl hydroperoxide reductase from Salmonella typhimurium. 2. Cystine disulWdes involved in catalysis of peroxide reduction, Biochemistry 35 (1996) 65–75. P.W. Riddles, R.L. Blakeley, B. Zerner, Ellman’s reagent: 5,5⬘-dithiobis(2-nitrobenzoic acid); A reexamination, Anal. Biochemistry 94 (1979) 75–81. J.P. Schneider, J.D. Lear, W.F. DeGrado, A designed buried salt bridge in a heterodimeric coiled coil, J. Am. Chem. Soc. 119 (1997) 5742–5743. J.K. Kretsinger, J.P. Schenider, Design and application of basic amino acid displaying enhanced hydrophobicity, J. Am. Chem. Soc. 125 (2003) 7907–7913. R. Rudolph, H. Lilie, In vitro folding of inclusion body proteins, FASEB J. 10 (1996) 49–56. H. Rogl, K. Kosemund, W. Kuhlbrandt, I. Collinson, Refolding of Escherichia coli produced membrane protein inclusion bodies immobilised by nickel chelating chromatography, FEBS Lett. 432 (1998) 21–26. R. Zahn, C. von Schroetter, K. Wuthrich, Human prion proteins expressed in Escherichia coli and puriWed by high-aYnity column refolding, FEBS Lett. 417 (1997) 400–404. H. Lilie, E. Schwarz, R. Rudolph, Advances in refolding of proteins produced in E. coli, Curr. Opin. Biotech. 9 (1998) 497–501. R.L. Stone, H. Zalkin, J.E. Dixon, Expression, puriWcation, and kinetic characterization of recombinant human adenylosuccinate lyase, J. Biol. Chem. 268 (1993) 19710–19716. S.U. Gite, R.F. Colman, AYnity labeling of the active site of rabbit muscle adenylosuccinate lyase by 2-[(4-bromo-2,3-dioxobutyl)thio] adenosine 5⬘-monophosphate, Biochemistry 35 (1996) 2658–2667. P. Edery, S. Chabrier, I. Ceballos-Picot, S. Marie, M.-F. Vincent, M. Tardieu, Intrafamilial variability in the phenotypic expression of adenylosuccinate lyase deWciency: a report on three patients, Am. J. Med. Genet. A120 (2003) 185–190. A.M. Marinaki, M. Champion, M.A. Kurian, H.A. Simmonds, S. Marie, M.F. Vincent, G. Van den Berghe, J.A. Duley, L.D. Fairbanks, Adenylosuccinate lyase deWciency-Wrst British case, Nucleosides Nucleotides Nucleic Acids 23 (2004) 1231–1233. T.T. Lee, C. Worby, Z.-Q. Bao, J.E. Dixon, R.F. Colman, His68 and His141 are critical contributors to the intersubunit catalytic site of adenylosuccinate lyase of Bacillus subtilis, Biochemistry 38 (1999) 22–32. J.L. Brosius, R.F. Colman, A key role in catalysis for His89 of adenylosuccinate lyase of Bacillus subtilis, Biochemistry 39 (2000) 13336–13343. J.B. Palenchar, J.M. Crocco, R.F. Colman, The characterization of mutant Bacillus subilis adenylosuccinate lyases corresponding to severe human adenylosuccinate lyase deWciencies, Protein Sci. 12 (2003) 1694–1705. S. Sivendran, D. Patterson, E. Spiegel, I. McGown, D. Cowley, R.F. Colman, Two novel mutant human adenylosuccinate lyase (ASLs) associated with autism and characterization of the equivalent mutant Bacillus subtilis ASL, J. Biol. Chem. 279 (2004) 53789–53797.