Gene, 164 (1995) 9-15 © 1995 Elsevier Science B.V. All rights reserved. 0378-1119/95/$09.50
9
GENE 09221
Versatile, multi-featured plasmids for high-level expression of heterologous genes in Escherichia coli: overproduction of human and murine cytokines (Expression vector; mRNA structure; T7 RNA polymerase-based expression; phage ~ PL promoter; transcription termination; site-directed mutagenesis; affinity purification; histidine tag; translation initiation)
Nico Mertens, Erik Remaut and Walter Fiers Laboratory of Molecular Biology, Universityof Gent, B-9000 Gent, Belgium Received by J.K.C. Knowles: 9 January 1995; Revised/Accepted: 15 May 1995; Received at publishers: 20 July 1995
Abstract
We describe the construction, expression characteristics and some applications of a versatile dual-promoter expression plasmid for heterologous gene expression in Escherichia coli which contains both ~ PL and PT7 promoters. Furthermore, the plasmid is optimized to allow the expression of mature coding sequences without compromising the strength of the highly efficient PT7 or of the T7glO ribosome-binding site. The effect of the the naturally occurring RNA loops at both the 5' and 3' ends of the T7glO mRNA on expression was also examined. A double T7 RNA polymerase transcription terminator was inserted to ensure more reliable transcription termination and a higher expression level of the preceding gene. Further improvements involve a clockwise orientation of the promoters to minimize read-through transcription from plasmid promoters, a largely extended multiple cloning site, an antisense phage T3 promoter and a phage fl-derived, single-stranded replication origin. Variants of this plasmid allow for the production of fusion proteins with part of T7glO, a hexahistidine peptide and an enterokinase recognition site. The potential of these expression vectors is demonstrated by comparing the expression levels of a number of mammalian cytokines (human tumor necrosis factor, human immune interferon, human and murine interleukins 2, murine interleukin 4 and murine fibroblast interferon), using these expression plasmids.
INTRODUCTION
The use of Escherichia coli as a host for high-level expression is widespread both in molecular biology laboratories as in the biotechnology industry. Wellcharacterized expression systems have been described and successfully used. Nevertheless, despite the wide Correspondence to: Dr. W. Fiers, Laboratory of Molecular Biology, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium. Tel. (32-9) 264-5139; Fax (32-9) 264-5348; e-mail:
[email protected] Abbreviations: aa, amino acid(s); Ap, ampicillin; bp, base pair(s); CBB, Coomassie brilliant blue; EK, enterokinase; fl ori, intergenic region of phage fl; h, human; IFN, interferon; IL, interleukin; IPTG, isopropyl[3-o-thiogalactopyranoside; LB, Luria-Bertani (medium); X PL, leftward early major promoter of phage X; m, murine; MCS, multiple cloning site(s); nt, nucleotide(s); PAGE, polyacrylamide-gel electrophoresis; SSDI 0378-1119(95)00505-6
experience in this field, high-level expression has often been a result of trial and error. Therefore, the development of improved vectors favors successful expression experiments. The decision as to which promoter to use can be based on various criteria, such as promoter strength and control (Lanzer et al., 1988), economy and ease of utilization, or the conditions under which the Phil, promoter of the ApRgene on pBR322; Pollk, Klenow (large) fragment of E. coil DNA polymerase I; PT3, equivalent to PT7 from phage T3; PT7 or 010, T7 RNA polymerase-specific class-III promoter (derived from T7glO); RBS, ribosome-binding site(s); SDS, sodium dodecyl sulfate; TB, Terrific Broth (medium); T~, natural transcription terminator for T7 RNA polymerase derived from phage T7; T7glO, gene for the major capsid protein of phage T7; T7RNAP, T7 RNA polymerase; tfd, transcription terminator derived from phage fd; TIR, translation initiation region(s); TNF, tumor necrosis factor; [], denotes plasmidcarrier state.
10 promoter is to be used (e.g., temperature and choice of host strains or medium). Both X PL and PT7 a r e among the strongest promoters known in E. coli, they can be tightly regulated and allow a free choice of host strains or induction conditions. The PT7 can be activated in hosts carrying an inducible bacteriophage T7 RNA polymerase gene (T7gl) (Tabor and Richardson, 1985; Studier and Moffatt, 1986; Mertens et al., 1995). This system offers the advantage of achieving very efficient synthesis of RNA, insusceptible to premature transcription termination. Encoded proteins can be specifically labeled in the presence of rifampicin (Tabor, 1987). However, strains carrying a T7gl often express a residual level of the enzyme which can lead to a strong counterselecting pressure. Expression of the T7gl should, therefore, be very efficiently controlled when a heterologous gene and PT7 are combined (Mertens et al., 1995). But even more than transcription initiation, the initiation rate of translation determines the final expression level (Gold, 1990). High-level expression vectors therefore should contain strong translation initiation signals to optimize production yields. These signals often critically depend on their sequence context both 5' and 3' of the initiation A
B I NI II INI
codon (Zabeau and Stanley, 1982; Stanssens et al., 1985). This makes prediction of the optimal translation initiation region (TIR) for a given coding sequence difficult. A solution is to fuse the coding sequence to be expressed to a leader peptide derived from a very efficiently expressed gene, while leaving the original TIR largely intact. This strategy also permits to include a removable affinity tag in the leader peptide, allowing a one-step purification of the fusion protein. An elegant purification system, which has the advantage that it can also be used under proteindenaturing conditions, has been described based on the specific binding of vicinal histidine residues to metal chelating resins (St0ber et al., 1990). Here we describe an expression system that permits to use the tightly regulated k PL promoter and/or the strong and specific PT7 combined with the very efficient, A + Urich translation initiation region from the T7g10. We further demonstrate the usefulness of preserving the natural RNA loops at both 5' and 3' distal ends of the T7glO mRNA. The basic expression vectors were designed to include numerous features to facilitate cloning and expression in E. coli, and were successfully applied to the overproduction of a number of heterologous proteins.
C 1U_j i_2J 131 % synthesisrelativeto wild type mRNA Tc
20
40
60
80
100
'i
i
I
~
!
IIIII
IIIII
III
120
pLT10g 10TT glOB
pLT10g 10T
glOA
4.-- TT 4..'-T
pLT10gl 0 pLTg 10TT pLTg 10T pLTgl 0
Fig. 1. Effect of mRNA-terminal loops on T7glO expression kinetics. A glO-containing ClaI-PvulI fragment from phage T7 was cloned into pPlc321 (Remaut et al., 1981), opened with the same restriction enzymes. The 5' loop was deleted by first cloning the T7gl 0 in pET3 as a XbaI-PvulI fragment or in pT710 (N.M., unpublished results), a pT73-1ike vector (Tabor and Richardson, 1985) where a StuI site was included in the PT7 (Rosenberg et al., 1987) as a blunted XbaI-Pvull fragment (thereby omitting the 5' loop) and subsequently retransferring the T7glO with its promoter as a BgllIPvulI fragment into pPLc236 (Remaut et al., 1981). The Tdp terminator was deleted or duplicated in both variants as a CellI-PvulI fragment. All enzymes were used according to the suppliers' instructions, cloning and general analysis techniques were as described (Sambrook et al., 1989). (A) T7glO was induced using LMD5231 [pT7POL22][pLT10g10T] as an expression strain, which contained a tightly regulated T7gl under X PL control (Mertens et al., 1995). NI refers to protein derived from a noninduced strain, IN from a strain after a 2-h shift from 28 to 42°C. The protein extracts were analyzed using 0.1% SDS-15% PAGE and stained with CBB. (B) pLT10gl0T (lane 1), pLT10gl0 (lane 2) and pLT10glOTT (lane 3) plasmids were transcribed in vitro by T7 RNA polymerase (Boehringer-Mannheim, Mannheim, Germany) in the presence of trace amounts of [~z-azP] ATP. The transcripts were separated on a 2.2 M formaldehyde-1.4% agarose gel (Sambrook et al., 1989). Termination can be seen at the T~-terminators (T and TT) and at a region further downstream on the plasmid (Tc, clockwise termination region between Ap rt and ori). (C) LMD5231 [pT7POL23] containing one of the T7glO expression plasmid variants was induced by shifting the culture to 42°C. Cultures were labeled with 20 txCi [35S]methionine/ml for 10 min in rich medium. The proteins were analyzed using 0.1% SDS-15% PAGE and quantified using a phosphorimaging screen. The pLT10gl0T-containing expression strain showed a relative T7glO de novo synthesis of approx. 60% of total protein synthesis. The relative differences in T7glO synthesis due to variations in 5' or 3' mRNA structures are shown as a diagram compared to this reference.
11 RESULTS AND DISCUSSION
(a) Effect of T7glO mRNA structural features on protein expression The g l 0-coding region, including the T7RNAP-specific T ~ terminator, was cloned from phage T7 in a L PL expression plasmid ( R e m a u t et al., 1981), resulting in a vector which showed a relative accumulation of T 7 g l 0 up to 60% of the total protein after 2 h of induction (Fig. 1A). The 5'-terminal nt sequence p p p G G G A G A , which is part of the T7 promoters ( D u n n and Studier, 1983)is predicted to form a base-paired stem and loop structure with adjacent regions in highly translated T7 m R N A s (Rosa, 1981; Helke et al., 1993). It has been d o c u m e n t e d that a stable hairpin structure at the 5' end in some cases can serve as an efficient stabilizer against m R N A degradation (Fuerst and Moss, 1989; Chen et al., 1991). To evaluate the importance of this 5' loop on the expression level of T7glO, we changed this region in the original T7glO expression vector. Also the effect of the T 7 T ~ terminator was assayed either by deleting it or by inserting a second Td0 element behind the cloned gene. The termination efficiency of these constructs was assayed by in vitro transcription reactions p r o g r a m m e d with the p L T 1 0 g l 0 T and p L T 1 0 g l 0 T T @10 p r o m o t e r and
plasmids (Fig. 1B). These data as well as in vivo expression experiments (data not shown) indicated that the transcription termination at a single T 7 T ~ terminator was only about 60% efficient. Fig. 1C demonstrates the positive effect of both hairpin loops on the expression levels of T7glO. It is conceivable that less optimal gene-RBS combinations could benefit more from these structural R N A elements than T7glO, where the optimal expression kinetics reach the theoretical m a x i m u m obtainable in E. coll. However, the experiment does not indicate whether this difference in expression levels is due to either m R N A stability, more efficient translation initiation or changes in plasmid biology.
(b) Construction of the expression vectors We designed an expression vector using both ~, PL and PT7" These promoters are tightly regulated, belong to the strongest promoters k n o w n in E. coli and can be used to
overcome unwanted transcription termination in heterologous coding sequences. By choosing the appropriate host strain for expression, both promoters can be induced separately or simultaneously (Table I). Furthermore, the combined use of PT7 and rifampicin permits exclusive gene expression, which allows the specific synthesis of a labeled protein (Tabor, 1991). This offers important ana-
TABLE I E. coli strains and plasmids used for expression with pLT10T3-derived vectorsa Genotype description Strain LMD5231 MC1061~,
Plasmid pci857
pT7POL23
pT7POL26
INDb
Comments
Wild-type isolate, no known mutations (Latarjet and Caldas, 1952). Constructed by lysogenizing ~ in MC1061 (Casadaban and Cohen, 1980). Used for general cloning because of its high competence. Contains no T7RNAP and the ~, wt repressor efficiently represses L PL. L phage remains more stably integrated when grown at 28°C.
E. coil K-12 )~ FhsdR mcrB araD139 A(araABC-leu)7697 AlacX74 galU gaIK rpsL thi (~)
P15A KmR ~, c1857
)~ PL
pSC101KmR PL-N'tL(tfd)z'T7gl'c1857
~ PL+PT7
pSC101Kma PN2s/oz-N-tL(tfd)z-TTgl-
)~ PL
cI857-1aclq
and/or PT7
Plasmid compatible with ColEl-based expression plasmids which holds an autogenously regulated thermo-sensitive X repressor gene (Remaut et al., 1983). Tight repression when grown at 28°C, full induction at 42°C. Low-copy plasmid compatible with both PI5A and ColE1 replicons containing an attenuated T7gl which is tightly controlled by ~, PL and the addition of transcription terminators which can be reversed by using the included ~ N/nutL transcription anti-termination system. Also contains the autogenously regulated ci857 gene (Mertens et al., 1995). Low-copy-number vector similar to pT7POL23 but the TTgl is regulated by the IPTG-inducible PN25o2(Lanzer and Bujard, 1988). Also contains the c1857 and the laclq genes (Mertens et al., 1995).
~ PL and Pr7 can be induced separately or together, by temperature shift and/or by adding IPTG, respectively, simply by choosing the appropriate expression strain. A set of accessory plasmids can be used to convert any E. coil strain for expression with derivatives of the pLT10T3 vector (Mertens et al., 1995). In the presence of the anti-termination protein N, k PL-initiated transcripts become insensitive to normal E. coil termination signals; this is useful when stem-loop structures prevent E. coil RNA polymerase to efficiently transcribe a heterologous gene. T7RNAP-initiated transcripts are also known to ignore most E. coil transcription termination signals. b IND, promoter that can be induced when this strain is used. a
12 lytical as well as preparative advantages. Also, rifampicinbased, exclusive gene expression sometimes helps to increase the relative levels of expression (data not shown). The pLT10gl0T contains clockwise oriented L PL and PT7 promoters relative to the counter-clockwise replication origin in pBR322 as conventionally drawn. We have found this type of orientation to limit transcriptional read-through from promoters elsewhere on the plasmid (data not shown) in contrast to promoters in the same orientation behind the replication origin. This transcriptional read-through from distinct plasmid promoters has been found to be a disadvantage (Olenik et al., 1991; Somerville et al., 1991; Brown and Campbell, 1993; Nishihara et al., 1994) shared by several PTv-based expression vectors so far described. Also, the clockwise orientation of the promoter in pLT10T3 does not lead to enhanced production of 13-1actamase upon induction, thus avoiding both the accelerated destruction of the
A
B
MCS EcoRI
,~tll Spht
Ndel f
Xbal Xhol EcoR[ Kpnl Sinai Asull Cla I Hindlll
cAA A C C=G C=G A--U G.U
A--U
Nsil Sphl
Pst I
~lt"~
Xbal
PLT_10T3 "
3654
bp
Apal
\\\
Xba I
BamHI Xhol
no.
\ EcoRI \ \'\\ I Sacl s,, s~l
%\\
Sl
G=C
~(~--CCUCUAGAAAUAAUUUUGUUUAAC~"-GGAAGAAGGAGAUAUACAuAU GGGCC'C
NotJ
I(f
antibiotic and the possible overloading of the cellular secretion mechanism by pre-13-1actamase. Probably the most determining factor in obtaining successful synthesis of heterologous proteins is the choice of the translation initiation region. Suboptimal expression has been ascribed to formation of unfavorable secondary mRNA structures in the TIR (Iserentant and Fiers, 1980; Stanssens et al., 1985; de Smit and van Duin, 1990). We chose to use the A + U-rich leader and RBS of the very highly translated T7glO(Rosa, 1981; Studier and Moffatt, 1986; Olins et al., 1988), in order to disfavor the formation of stable RNA structures. For precise in-frame fusions of the mature part of a desired coding sequence to the ATG start codon, an ApaI recognition site was engineered overlapping this codon. After resecting the 3' sticky ends with Pollk, a blunt-ended start codon is formed, without any change in the preceding sequence. This is important, since even a single nt change in front
SD
Nde I
Apa I
C aa from
TTglO pkT]OTH pLFIOT4 pLF10T5
AT
Metal chelating affinity tail GAT CCA CAT CAC CAr
//J
~11
Enterokinase site
C A C CAT C A C G A C G A T G A C G A T AAGGCggCvQ
u pro his hi~ his his his his asp asp asp asp / y s ' " Bbel Ehel Narl Kasl
Fig. 2. Map and construction of the expression plasmids. (A) pLT10T was constructed by introducing an ApaI site overlapping the glO ATG start codon by linker mutagenesis. Subsequently the T7glO-coding region extending to the CellI site preceding the TO terminator was replaced by an ApaI-blunted NsiI fragment containing the multiple cloning site of pGEM7Zf+ (Promega, Madison, WI, USA). This resulting vector (pLT10T) was adapted to include a double terminator by duplicating a blunted CelII-PvulI T~-containing fragment in the PvulI site behind the first Tqb terminator, while also extending the multiple cloning site behind the HindIII site with a HindlII-BalI fragment from pGEMEX-1 (Promega), which also contains an antisense T3 promoter. Finally, the PvulI-PstI fragment containing the replication origin in the plasmid was substituted with a blunted XbaI-PstI fragment ofpMc58 (Stanssens et al., 1989) harboring the ColE1 ori, the fl ori and a double tfd terminator. This resulted in pLT10T3. (B) Sequence and model for the T7glO leader mRNA structure (Rosa, 1981). Possible interactions with the 3' end of the 16S rRNA and the 30S r-protein S1 are indicated. The ATG start codon can be made accessible by cutting with ApaI and resecting the 3' unpaired ends with Pollk. (C) pLF10T was constructed by blunting the first StyI site in T7glO with Pollk and fusing it to the similarly blunted ApaI site of pLT10T. Filling in the AatlI site and fusing it to a 5'-GATCCACATCACCATCACCATCACGACGATGACGATAAGGCGCC linker coding for six consecutive His residues, followed by four Asp and one Lys codon (encoding the recognition site for enterokinase) and a recognition site for the neoschizomers NarI, BbeI, KasI or EheI. Since these four enzymes recognize the same site but cut differently, fusions in all reading frames are simplified. After substitution of a blunted KasI-PstI fragment with a resected ApaI-PstI fragment of pLT10T3, pLF10T4 has the same features as pLT10T3, pLF10T5 and pLT10TH were constructed in a similar way, except that the fusion point in the T7g10 coding region was not after the 14th codon, but after the 121st codon by using the AccI site in glO or by fusing the linker sequence directly to the ATG start codon of g10.
13 of the AUG start codon can lead to a drastically lowered expression levels (Gross et al., 1990). The coding sequence of T7glO was replaced with a large and versatile MCS (Fig. 2). The size of pLT10T3 was kept low by only including relevant genetic elements. Also the rop gene was deleted, resulting in a high-copynumber vector which gives higher yields of plasmid DNA and fewer plasmid-free cells (Qoronfleh and Ho, 1993; Clos and Brandau, 1994). To overcome possible deleterious interference of the strong promoters with plasmid replication (St0ber and Bujard, 1982), a double T~ terminator, combined with a tandem tfd transcription terminator for E. coli RNA polymerase, was engineered between the MCS and the ori. Furthermore, a phage fl ori and an antisense-orientated phage T3 promoter were also added. To construct an expression vector directing the synthesis of afffinity-tagged fusion proteins, we fused a hexameric histidine-encoding fragment (St0ber et al., 1990) to either the ATG start codon or after codons 14 or 121 of T7glO, followed by a synthetic linker encoding the enterokinase recognition site (Maroux et al., 1971), allowing the precise release of the mature heterologous gene (Fig. 2C). This leader fragment was engineered in such a way that in-fase fusions can be made by inserting the same DNA fragment as used for cloning in pLT10T3. Fusions in the other two reading frames can easily be made due to the presence of a restriction site which can be cut by a convenient set of neoschizomers.
(c) Induction and specific detection of heterologous protein synthesis The heterologous genes to be expressed were engineered either by site-specific mutagenesis or by PCRbased mutagenesis (Sambrook et al., 1989) in such a way that the first codon of the mature coding sequence became accessible as a blunt-ended DNA fragment after cutting with a restriction enzyme and followed, where necessary, by treatment with Pollk to create blunted ends. They were then cloned in either pLT10T3 or pLF10T3. Induction was initiated by a shift from 28 to 42°C in a strain harboring a temperature-sensitive ~, repressor (Bernard et al., 1979; Remaut et al., 1983), or by inducing the synthesis of T7RNAP (Mertens et al., 1995). When using the latter system, particular attention should be paid to avoid basal level synthesis of the RNA polymerase, since continuous expression can be detrimental for the stable maintenance of a functional expression system (Studier et al., 1990; Studier, 1991; Mertens et al., 1995). In general, the expression levels of the heterologous genes were slightly higher when controlled by the T7RNAP/ promoter system than with the L PL promoter (Table II).
TABLE II Comparison of heterologous gene expression in different expression systems Expressed genea
Protein synthesis
(%)b
Protein accumulated (%)~ RBS behind )~ PL,
Promoter,
hlL-2 mlL-2 mlL-4 hlFN- 7 mlFN-fl hTNF
~. Pl.
PT7
T7glO
Other
40 29 15 15 13 18
45 44 21 17 8 32
20 35 10 52 10 43
51; 102 102 < 12; < 14 252; 103 < 14 355; 101; <12; <13
a Heterologous gene expressed (h, human; m, murine). b The genes were expressed using the pLT10T-derived vectors. ~ PL was induced by shifting the temperature from 28 to 42°C in LMD 523l[pc1857] or PT7 was induced by adding l mM IPTG to LMD5231 [pT7POL26]. The cultures were labeled 30 rain after induction for 10min with 20 gCi [35S]methionine/ml (Amersham), in LB medium. The relative synthesis of the heterologous proteins was quantified by scanning the gel with a phosphor-imager (Molecular Dynamics, Sunnyvale, CA, USA). ¢ Total accumulation of heterologous proteins after 3 h induction at 42°C in LMD5231 [pci857], The amount of heterologous protein was compared relative to the total cellular protein content by scanning a CBB-stained 0.1% SDS-15% PAGE gel with a normal light emitting densitometer (Hoefer, San Francisco, CA, USA), Protein accumulation was also compared with previously described ~, pL-RBS combinations (Remaut et al., 1986, 1987), including TIR derived from (1) phage ~t ner, (2) trpL, (3) phage MS2 pol, (4) lacZ and (5) phage T4 g32.
Most of the new constructs could be compared to previously described )~ PL expression vectors for the same genes, but containing a different RBS (Remaut et al., 1986; 1987). The overall level of )~ pL-induced protein synthesis as well as accumulation was in all cases tested significantly higher using the pLT10T3 vector (with the T7glO TIR) than with similar L PL expression vectors containing another RBS (Table II, Fig. 3) (see also Olins et al., 1988).
(d) Conclusions (I) We cloned the T7glO gene and analyzed the expression characteristics using a clockwise oriented )~ PLcontaining expression vector. Expression-promoting elements from the latter gene were subsequently used for construction of a versatile dual promoter expression plasmid, containing both the powerful, but efficiently regulatable )~ PL and PTV. (2) The naturally occurring, terminal RNA loops at the 5' and 3' end were found to have a beneficial influence on T7glO gene expression from a plasmid resident gene. (3) Transcriptional read-through from these strong promoters into the replication origin region was mini-
14
A
B Cellular extract
ihll~_ I iml.L21 imlL41
IhlFN~'IImlFNBIIhTNFII
NI
II M
II
r 1
I
I
2
I
I
3
I MAC eluate I I
4 Ibm5 I
I I
6
I [
7
I I
8
I
Fig. 3. High-level expression of heterologous proteins. Panel A shows total protein made in LMD5231 [pT7POL23] after 2 h induction by shifting the temperature to 42°C. Proteins were resolved using 0.1% SDS-15% PAGE and stained with CBB. The first six lanes are extracts from induced cells containing a pLT10T-derived expression vector for hlL-2, mlL-2, mlL-4, hlFN-?, mlFN-[3 and hTNF respectively. NI denotes noninduced (28°C) extract from LMD5231 [pT7POL23][pLT10HIFGT] as a control. Typically, up to 50 I.tg/a600 n m of heterologous protein could be obtained. Molecular mass markers (M) were 67, 43, 28, 18.4, 14.3 and 13.7 kDa. Panel B compares total protein extracts after induction of strains expressing hlL-2 (lanes 1 and 5) or fusions with a removable affinity tag (lanes 2 and 6) or the affinity tag fused to parts of T7gl0 (with 14 aa, lanes 3 and 7; with 121 aa, lanes 4 and 8) with the eluate after binding and washing on an immobilized metal ion affinity column under denaturing conditions. Methods: Overnight cultures were diluted 100-fold in TB medium and grown until A6o0 nm reached approx. 0.5. Next the cells were shifted to 42°C and grown for a further 3 h. Induced cells were opened by sonication. The pelleted inclusion bodies were washed with 1% Triton X-100 and subsequently solubilized in 6 M guanidine hydrochloride/1% [3-mercaptoethanol/10 mM Tris.HC1 pH 8.0 and loaded onto a 1 ml Ni 2 +-NTA column (Qiagen, Hilden, Germany). After sequential washing with buffer containing 8 M urea, 0.1% Tween-20, 5 mM dithiothreitol/10 mM imidazole/10 mM Tris.HC1 at pH 8.0 and 6.3, proteins were eluted with three column volumes of washing buffer at pH 5.0 containing 0.2 M imidazole.
mized by the insertion of a double transcription terminator sequence for both the phage and the bacterial RNA polymerase. Read-through transcription from other plasmid promoters was found to be negligible due to the clockwise orientation of the inserted promoters. The strong RBS from T7glO was made directly accessible to insertion of the mature coding region of a heterologous gene by creating an ApaI site overlapping the ATG start codon. Furthermore, the expression vector was provided with a versatile MCS, an counter-clockwise Pra, and an fl ori. (4) Variations of this vector were constructed to allow coding sequences to be fused in all reading frames to parts of T7glO, followed by a fragment coding for a hexahistidine affinity tag and an enterokinase recognition site. (5) Using the newly constructed vectors, high production levels were obtained for a number of mammalian cytokines. These vectors have the potential to considerably improve the expression level of other heterologous genes.
ACKNOWLEDGEMENTS
We are grateful to Dr. F.W. Studier (Brookhaven National Laboratory, Brookhaven, NY, USA) for kindly providing phage T7, Dr. S. Tabor (Harvard Medical
School) for providing pT7-3, Dr. J. van der Toorn (TU, Delft) for providing a wild-type E. coli K-12 isolate (LMD5231) and Dr. P. Stanssens (Corvas, Gent) for the use ofpMc58. This work was supported by the IWTCOT program, the FGWO and the National Lottery. We also thank M. Vandecasteele for his help in preparing the manuscript.
REFERENCES Bernard, H.-U., Remaut, E., Hershfield, M.V., Das, H.K., Helinski, D.R., Yanofsky, C. and Franklin, N.: Construction of plasmid cloning vehicles that promote gene expression from the bacteriophage lambda PL promoter. Gene 5 (1979) 59-76. Brown, W.C. and Campbell, J.L.: A new cloning and expression strategy for genes encoding proteins toxic to Escherichia coll. Gene 127 (1993) 99-103. Casadaban, M.J. and Cohen, S.N.: Analysis of gene control signals by DNA fusion and cloning in Escherichia coll. J. Mol. Biol. 138 (1980) 179-207. Chen, L.-H., Emory, S.A., Bricker, A.L., Bouvet, P. and Belasco, J.G.: Structure and function of a bacterial mRNA stabilizer: analysis of the 5' untranslated region of the ompA mRNA. J. Bacteriol. 173 (1991) 4578-4586. Clos, J. and Brandau, S.: pJC20 and pJC40- two high-copy-number vectors for the T7 RNA polymerase-dependent expression of recombinant genes in Escherichia coll. Protein Expr. Purif. 5 (1994) 133 137. de Smit, M.H. and van Duin, J.: Secondary structure of the ribosome-
15 binding site determines translational efficiency: A quantitative analysis. Proc. Natl. Acad. Sci. USA 87 (1990) 7668-7672. Dunn, J.J. and Studier, F.W.: Complete nucleotide sequence of bacteriophage T7 DNA and the locations of T7 genetic elements. J. Mol. Biol. 166 (1983) 477 535. Fuerst, T.R. and Moss, B.: Structure and stability of mRNA synthesized by vaccinia virus-encoded bacteriophage T7 RNA polymerase in mammalian cells. Importance of the 5' untranslated leader. J. Mol. Biol. 206 (1989) 333 348. Gold, L.: Expression of heterologous proteins in Escherichia coli. Methods Enzymol. 185 (1990) 11 14. Gross, G., Mielke, C., Hollatz, I., Blocker, H. and Frank, R.: RNA primary sequence or secondary structure in the translational initiation region controls expression of two variant interferon-13 genes in Escherichia coli. J, Biol. Chem. 265 (1990) 17627 17636. Helke, A., Geisen, R.M., Vollmer, M., Sprengart, M.L. and Fuchs, E.: An unstructured mRNA region and a 5' hairpin represent important elements of the Escherichia coli translation initiation signal determined by using the bacteriophage T7 gene 1 translation start site. Nucleic Acids Res. 21 (1993) 5705 571l. Iserentant, D. and Fiers, W.: Secondary structure of mRNA and efficiency of translation initiation. Gene 9 (1980) 1-12. Lanzer, M. and Bujard, H.: Promoters largely determine the efficiency of repressor action. Proc. Natl. Acad. Sci. USA 85 (1988) 8973-8977. Latarjet, R. and Caldas, J.: Restoration induced by catalase in irradiated microorganisms. J. Gen. Physiol. 35 (1952)455-470. Maroux, S., Baretti, J. and Desneulle, P.: Purification and specificity of porcine enterokinase. J. Biol. Chem. 246 (1971) 5031 5039. Mertens, N., Remaut, E. and Fiefs, W.: A tight transcriptional control mechanism for T7 RNA polymerase ensures stable high-level expression from T7 promoter-based expression plasmids. Bio/Technology 13 (1995) 175-179. Neidhardt, F.C., Ingraham, J.L. and Schaechter, M.: Physiology of the Bacterial Cell. A Molecular Approach. Sinauer, Sunderland, MA, 1990. Nishihara, T., lwabuchi, T. and Nohno, T.: A T7 promoter vector with a transcriptional terminator for stringent expression of foreign genes. Gene 145 (1994) 145-146. Olins, P.O., Devine, C.S., Rangwala, S.H. and Kavka, K.S.: The T7 phage gene I0 leader RNA, a ribosome-binding site that dramatically enhances the expression of foreign genes in Escherichia coll. Gene 73 (1988) 227 235. Qoronfleh, M.W. and Ho, T. A.: A modified pET vector that augments heterologous protein expression. Biotechnol. Lett. 15 (1993) 337-340. Remaut, E., Stanssens, P. and Fiers, W.: Plasmid vectors for highefficiency expression controlled by the PL promoter of coliphage lambda. Gene 15 (1981) 81 93. Remaut, E., Tsao, H. and Fiers, W.: Improved plasmid vectors with a thermoinducible expression and temperature-regulated runaway replication. Gene 22 (1983) 103 113. Remaut, E., Stanssens, P., Simons, G. and Eiers, W.: Use of the phage
lambda PL promoter for high-level expression of human interferons in Escherichia coli. Methods Enzymol. 119 (1986) 366-375. Remaut, E., Marmenout, A., Simons, G. and Fiers, W.: Expression of heterologous unfused protein in Escherichia coli. Methods Enzymol. 153 (1987) 416-431. Rosa, M.D.: Structure analysis of three T7 late ribosome-binding sites. J. Mol. Biol. 147 (1981) 55 71. Rosenberg, A.H., Lade, B.N., Chui, D, Lin, S.W., Dunn, J.J. and Studier, W.F.: Vectors for selective expression of cloned DNAs by T7 RNA polymerase. Gene 56 (1987) 125-135. Sambrook, J., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. Somerville, R.L., Ni Shieh, T.-L., Hagewood, 13. and Cui, J.: Gene expression from multicopy T7 promoter vectors proceeds at single copy rates in the absence of the T7 RNA polymerase. Biochem. Biophys. Res. Commun. 181 (1991) 1056-1062. Stanssens, P., Remaut, E. and Fiers, W.: Alterations upstream from the Shine-Dalgarno region and their effect on bacterial gene expression. Gene 36 (1985) 211 223. Stanssens, P., Opsommer, C., McKeown, Y.M., Kramer, W., Zabeau, M. and Fritz, H.-J.: Efficient oligonucleotide-directed construction of mutations in expression vectors by the gapped duplex DNA method using alternating selectable markers. Nucleic Acids Res. 17 (1989) 4441-4453. StOber, D. and Bujard, H.: Transcription from efficient promoters can interfere with plasmid replication and diminish expression of plasmid specified genes. EMBO J. 1 (1982) 1399-1404. Sttiber, D., Matile, H. and Garotta, G.: System for high-level production in Escherichia coli and rapid purification of recombinant proteins, applications to epitope mapping, preparation of antibodies, and structure-function analysis. In: Lefkovits, I. and Pernis, B. (Eds.), Immunological Methods, Vol. IV. Academic Press, New York, NY, 1990, pp. 121-152. Studier, F.W. and Moffatt, B.: Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J. Mol. Biol. 189 (1986) 113 130. Studier, F.W., Rosenberg, A.H., Dunn, J.J. and Dubendorff, J.W.: Use of the T7 RNA polymerase to direct the expression of cloned genes. Methods Enzymol. 185 (1990) 60-88. Studier, F.W.: Use of bacteriophage T7 lysozyme to improve an inducible T7 expression system. J. Mol. Biol. 219 (1991) 37 44. Tabor, S. and Richardson, C.C.: A bacteriophage T7 RNA polymerase/ promoter system for controlled exclusive expression of specific genes. Proc. Natl. Acad. Sci. USA 82(1985) 1074-1078. Tabor, S.: Expression using the T7 RNA polymerase/promoter system. In: Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (Eds3, Current Protocols in Molecular Biology. John Wiley, New York, NY, 1987, p. 16.2. Zabeau, M. and Stanley, K.K.: Enhanced expression of cro-13galactosidase fusion proteins under the control of the p~ promoter of bacteriophage L. EMBO J. 1 (1982) 1217 1224.