Biochimica et Biophvsica Acta, 1087 (1990) 205-211 Elsevier
205
BBAEXP 92177
Expression of active human hypoxanthine-guanine phosphoribosyltransferase in Escherichia coli and characterisation of the recombinant enzyme M i r a n d a L. F r e e 7, R o s s B. G o r d o n 2 D i a n n e T. K e o u g h 2, If or R. B e a c h a m 3, B r y a n T. E m m e r s o n 2 a n d J o h n d e J e r s e y 1 1 Department of Biochemistry and Centre for Molecular Biology and Biotechnology, University of Queensland, Queensland, e Department of Medicine, University of Queensland, Princess Alexandra Hospital, Woolloongabba, Queensland and "~Division of Science and Technology, Griffith University, Nathan, Queensland (Australia) (Received 19 April 1990)
Key words: Hypoxanthine-guanine phosphoribosyltransferase; Lesch-Nyhan; Gout; Recombinant enzyme; Gene expression; Human erythrocyte
A plasmid, pRG1, has been constructed by incorporating the coding sequence of human hypoxanthine-guanine phosphoribosyltransferase (EIPRT) into the expression vector pT7-7. Expression of human EIPRT has been achieved in H P R T - Escherichia coil cells transformed with pRGI and pGP1-2, as shown by: (1) exclusive labelling with [3SS]methionine of a polypeptide with the same mobility as purified human HPRT on SDS-PAGE; and (2) measurement of HPRT activity after cell lysis. Although the majority of the recombinant HPRT was present in the particulate fraction after cell lysis and centrifugation, sufficient HPRT activity was present in the supernatant fraction to allow comparison with the I-1PRT purified from human erythrocytes and the activity in human haemolysates and lymphoblast lysates. Small differences in electrophoretic mobility on native gels were found between HPRT activity from these sources. The K= values of recombinant HPRT for the substrates 5-phospho-a-D-ribosyl-l-pyrophosphate and guanine were compared with those of lymphoblast and erythrocyte HPRT. Introduction Hypoxanthine-guanine phosphoribosyltransferase (HPRT; EC 2.4.2.8) is a purine salvage enzyme which catalyses the conversion of the purine bases hypoxanthine and guanine to their respective mononucleotides. In man, complete deficiency of HPRT activity results in the Lesch-Nyhan syndrome [1] which is characterised by over-production of uric acid and severe neurological disorders. Partial deficiency causes hyperuricaemia which may lead to gouty arthritis and renal failure if untreated [2]. Human HPRT is a soluble cytoplasmic enzyme consisting of a tetramer of identical subunits, each of which is a single polypeptide chain of 217 amino acid residues, with the N-terminal alanine acetylated [3,4]. HPRT cDNA has been cloned, sequenced and transferred to several types of animal cells after incorporation into a retroviral vector [5-8]. Expression of the human HPRT gene in E. coli should allow the preparation of larger amounts of enzyme than are currently available from Correspondence: J. de Jersey, Department of Biochemistry, University of Queensland, Queensland 4072, Australia.
erythrocytes [9]. This should facilitate physical characterisation of the enzyme (e.g., by N M R spectroscopy and X-ray crystallography), as a representative of the 10 known phosphoribosyltransferases [10]. Expression in a bacterial system will also simplify studies of the catalytic mechanism by site-directed mutagenesis. In this paper, we describe a system for the expression of active human HPRT in E. coli and compare some of the physical and catalytic properties of the expressed enzyme with those of HPRT purified from human erythrocytes.
Materials and Methods Materials
Restriction endonucleases, DNA polymerases and ligases were purchased from New England Biolabs and used according to the manufacturer's instructions. Geneclean was obtained from Bio 101 (La Jolla, CA, U.S.A.). Ovalbumin, bovine -/-globulin, guanine and 5-phospho-t~-o-ribosyl-l-pyrophosphate ( P R i b - P P ) were products of Sigma and Sephacryl S-300 and chymotrypsinogen were obtained from Pharmacia LKB Biotechnology. [35S]methionine (1350 Ci/mmol) and 14C-labelled rainbow markers were purchased from
0167-4781/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
206 Amersham. Oligonucleotides were synthesized using an Applied Biosystems D N A synthesizer. Human lymphoblasts were transformed, grown and harvested by standard procedures [11].
harvested by centrifugation at 5000 × g for 10 rain at 4°C. The cells were resuspended in 1/10 of their original volume in 0.1 M Tris-HC1 buffer, 0.11 M MgCI 2, 1 mM/3-mercaptoethanol (pH 7.4) and stored at - 2 0 ° C .
P lasmid construction A plasmid containing 956 bp of human H P R T cDNA, pHPT31, was kindly supplied by Dr. C.T. Caskey. Recombinant plasmid pGP1-2, which has been described previously [12] and pT7-7, which includes a ribosomal binding site and ATG initiator codon, were kindly supplied by Dr. S. Tabor. The construction of pRG1 is described in the text. All plasmids were maintained in two E. coli strains, JM109 [13] and $o606 [14]; strains carrying plasmids were grown on LB media in the presence of appropriate antibiotics at a concentration of 50 ~g/ml. Restriction fragments for cloning experiments were purified from agarose gels using Geneclean according to the manufacturer's instructions.
Preparation of bacterial extracts Phenylmethylsulphonylfluoride (PMSF) and P RibPP were added to the cells to a final concentration of 1 mM before lysozyme (1 mg/ml) was added to lyse the cells. After incubation on ice for at least 45 rain, the ceils were freeze-thawed using liquid nitrogen and then centrifuged at 19000 x g for 20 min at 4°C to separate soluble and insoluble fractions,
Polymerase chain reaction pHPT31 was digested with PstI and the 956 bp fragment containing H P R T c D N A purified and used as the template for amplification by the polymerase chain reaction. Two oligonucleotide primers, one containing a 2 bp change from the target sequence to give an NdeI site immediately prior to the translation start codon, were used to generate an amplified fragment 413 bp long (Fig. 1). Reactions were performed using a denaturation temperature of 95°C for 1 min, an annealing temperature of 40°C for 40 s and an extension temperature of 70°C for 40 s; after 35 cycles of amplification, a further extension at 70°C for 5 min was performed to complete any fragments that were not yet full-length double-stranded. Cultures E. coli $o606 [ara, A pro-gpt-lac, thi, hpt, F - ] cells containing pGP1-2 and the reeombinant pRG1 were used. Standard expression system cultures were grown in an enriched medium (20 g/1 Baeto-tryptone, 10 g/1 yeast extract, 5 g/1 sodium chloride, 2 ml/1 glycerol and 50 mM potassium phosphate, pH 7.2) containing 50 # g / m l of both ampicillin and kanamycin in a Gallenkamp shaking incubator until a n A59o Of 1.5 was reached. Induction of expression system Expression was then induced by an increase in the culture temperature from 3 0 ° C to 42°C for 25 min; this caused derepression of the PL promoter controlling expression of T7 R N A polymerase from pGP1-2. The polymerase binds to the T7 R N A promoter on the recombinant plasmid (pRG1) enabling transcription of human H P R T cDNA [12]. Following induction, the cells were allowed to express at 37°C for up to 2 h and
H P R T activity assays and determination of K,,, values H P R T activity was determined using a continuous spectrophotometric assay measuring the conversion of guanine to G M P at 257.5 nm [15]. 1 unit of H P R T activity is defined as the amount of enzyme required to convert 1 nmol of guanine to G M P per rain. The ranges of substrate concentration used for the determination of Km values were 120-1600 t~M P R i b - P P (at a guanine concentration of 60 #M) and 2-20 ktM guanine (at a P R i b - P P concentration of 1100 ~tM), at a Mg 2+ concentration of 110 raM. Aqueous solutions of P R i b - P P were prepared immediately before use and kept on ice during each series of assays. The concentration of PRib-PP in these solutions was determined by the method of Kornberg et al. [16]. Protein estimation Protein concentration was determined by the Lowry method [17] with bovine serum albumin as the standard protein. Incorporation of [¢SS]methionine The method used was based on that of Tabor and Richardson [12]. Cultures were grown up in 5 ml of LB medium at 30°C until an A590 of 0.4-0.5 was reached. After harvesting, cells were washed with 5 ml of M9 medium and resuspended in 1 ml of M9 medium supplemented with 18 amino acids (each 10/~g/ml; excluding cysteine and methionine) and thiamine (20 ~tg/ml) and grown for a further 30-120 rain. Expression was induced as above except that rifampicin (20 # g / m l final concentration) was added to suppress bacterial protein synthesis. The temperature was decreased to 30°C and after 20 min growth, the cells were pulsed with 40/zCi [35S]methionine for 5 min, The cells were then harvested by centrifugation at 10 000 × g for 20 s. Polyacrylamide gel electrophoresis (PAGE) Bacterial extracts were heated at 95°C for 3 rain in 60 mM Tris-HC1 (pH 6.8), 1% sodium dodecyl sulphate (SDS), 1% /3-mercaptoethanol, 10% glycerol, 0.01% bromophenol blue and subjected to SDS-PAGE using
207 17.5% acrylamide gels [18]. Gels were stained with Coomassie Blue, then dried on a slab gel drier for 2 h at 80°C before being exposed for 12 h using Fuji RX100 X-ray film. PAGE of native samples was run at pH 8.8 (0.88 M alanine, 0.25 M Tris) using the Pharmacia PhastSystem equipped with a PhastGel Gradient 10-15 or Homogeneous PhastGel 12.5 and PhastGel native buffer strips. The gels were stained for activity by the method of Zannis et al. [19] and for protein by Coomassie Blue or silver staining.
Sephacryl S-300 chromatography A column of Sephacryl S-300 (22 × 2.5 cm; bed volume, 108 ml) was equilibrated with Buffer A (0.05 M Tris-HC1/0.01 M MgC12/1 mM/3-mercaptoethanol, p H 7.4). Where specified, PRib-PP (210/~M) was added to Buffer A. V~ values were determined for purified erythrocyte H P R T and for three marker proteins: bovine y-globulin, ovalbumin and chymotrypsinogen.
Results Cloning of human H P R T eDNA A recombinant plasmid, pRG1, containing 856 bp of human H P R T e D N A was constructed using pT7-7 as the vector (Fig. 1). An NdeI restriction site was created at the ATG start codon of human H P R T e D N A coding sequence so that translation would result in production of a protein without extra amino acids at the N-terminus. A 413 bp fragment of human H P R T e D N A was amplified as described in the Methods section. This fragment was purified from agarose and digested with NdeI and TaqI and the 956 bp PstI fragment from pHPT31 was digested with TaqI. The vector was digested with PstI and NdeI and a triple ligation set up with the vector, the NdeI/TaqI and the TaqI/PstI fragments, giving pRG1.
Expression of human H P R T gene in E. coli An H P R T - strain of E. coli, $o606, was transformed with pGP1-2, a plasmid containing the heat-inducible T7 R N A polymerase gene and a kanamycin resistance gene. Following antibiotic selection of the transformants, a second transformation was carried out with pRG1. Since pRG1 carries the bla gene, selection for E. coli containing both plasmids was possible by using ampicillin and kanamycin. In the initial experiments, a modification of the expression system described by Tabor and Richardson [12] was adopted. This system involves the addition of rifampicin in order to repress bacterial protein synthesis. Harvested cells were resuspended in 0.1 M Tris-HC1, 0.01 M MgC12, 1 mM fl-mercaptoethanol (pH 7.4) and following lysis with lysozyme and freeze/thawing, the lysate was sonicated. This procedure yielded approx.
100 units of H P R T activity per ml of supernatant after centrifugation at 19 000 × g for 20 rain at 4°C. Subsequently, it was found that the addition of rifampicin decreased the amount of recombinant protein produced, so its use was discontinued. The addition of PMSF (a serine proteinase inhibitor) and P R i b - P P (known to stabilise purified erythrocyte HPRT) to the resuspension buffer prior to cell lysis was found to result in higher H P R T activity. The highest levels of H P R T activity so far achieved were obtained under the following conditions: the cells were resuspended in 0.1 M Tris-HC1, 0.11 M MgC12, 1 mM B-mercaptoethanol (pH 7.4), containing 1 mM PMSF and 1 mM PRib-PP. The cells were then lysed by the addition of lysozyme and freeze/thawing. Routinely, the yield of H P R T activity in the supernatant was in the range of 700-1000 units m l - a.
Expression of H P R T in the presence of [35S]methionine Recombinant H P R T was labelled with [35S]methionine and run on an SDS-PAGE gel alongside H P R T isolated from erythrocytes. As shown in Fig. 2, the exclusively labelled expressed enzyme migrated the same distance as human erythrocyte HPRT. In a second series of experiments, three fractions containing labelled recombinant H P R T were obtained by lysing ceils as in the standard expression system followed by differential centrifugation. The three fractions were the soluble or supernatant fraction, and two insoluble fractions: a cell debris fraction from low speed centrifugation (5000 × g for 10 rain) and a membrane fraction from high speed centrifugation (19 000 × g for 20 min). When these fractions were subjected to SDS-PAGE, most of the labelled H P R T was found to be in the insoluble fractions (Fig. 3).
Microscopic examination of E. coli cells Cells from the standard expression system were examined by phase contrast microscopy. A large proportion of the cells examined contained dark spots in the polar regions, consistent with the formation of inclusion bodies.
Gel chromatography of expressed enzyme Recombinant H P R T from the standard expression system and a small amount of 35S-labelled recombinant H P R T were combined and subjected to gel chromatography on Sephacryl S-300. Fractions were assayed for both activity and radioactivity and the results are shown in Fig. 4. H P R T activity eluted with the same elution volume (V~) as that of purified erythrocyte HPRT, with a minor peak eluting later. In this experiment, only 26% of the H P R T activity applied to the column was recovered. As shown in Fig. 4, the radioactivity eluted from the colunm in a much broader peak than did the H P R T activity. In a subsequent experiment in which
208
PstI, P s i / ~
,Taql anneal / ATG olig'0nucieotides ~ P s t I
Psfl PstlP
NdeI PCR,
PstI/TaqI TaqI/Ndel ~/TaqI
Nde~ Taql
PstI~h"
ATG
bla
X.
,ATG//
HindIIt
\
~. XbaI IXh,~lxN BamHl .... ~ Smal \ EcoRl ~ ' N d e l
-
"~k
'1
a
~
bl
XbaI/ // ~ Ndel / TaqI\ Fig. 1. Constructionof pRG1. In the diagram the thick shaded areas indicate HPRT cDNA coding sequence and the thin shaded areas 3' and 5' non-coding sequence.
the column was equilibrated and eluted with buffer containing 0.21 mM PRib-PP, ~ 90% of the loaded activity was recovered from the column.
Stability of expressed HPRT In initial experiments, low levels of HPRT activity were obtained and the expressed enzyme was very unstable: most activity was lost within 1-2 days when the supernatant fraction (obtained by cell lysis and centrif-
ugation) was stored at either 5°C or - 2 0 ° C . A 2-fold increase in initial activity was observed when the Mg 2+ concentration of the resuspension buffer was increased from 10 to 110 raM; however, stability did not improve. The addition of the proteinase inhibitor PMSF (1 mM final concentration) and the substrate PRib-PP (1 mM final concentration) increased the initial activity of the expressed enzyme 7- to 10-fold and greatly improved its stability. When samples were stored at 5°C or - 2 0 ° C
209 a 200
b
c
d
),,'-
92.5 ,,"69 "'49
)"
........
...... ] ]]
e
!!I
o
b
-." 290 -'.( 92.5
~2°~ , ~ 3° ~
---, 69 46
o ,