- Email: [email protected]
A leader open reading frame is essential for the expression in Escherichia coli of GC-rich leuB gene of an extreme thermophile, Thermus thermophilus
ELSEVIER
FEMS Microbiology
Letters I35 ( 1996) 137- 142
A leader open reading frame is essential for the expression in Escherichia coli of GC-rich ZeuB gene of an extreme thermophile, Thermus thermophilus Masami Ishida a,x, Tairo Oshima b ” Laboruto~
of Biochemistry of Marine of Molecular Biolog!.
h Deptrrtmrnt
Resources.
Tokw
Received 7 November
Utlil,ersie
of Pharmacy
Tokyo University
1995; accepted
of Fisheries,
Konm
4, Minato-ku,
and L(fe Science, Huchioji,
13 November
Tokyo 108, Japm
Tokyo 1 YZ-03. Jtrport
1995
Abstract gene of an extreme thermophile, Thrrmus thermophilus, in Escherichitr tuc. However, the expression was hardly improved, despite increased transcription. The expression under tat promoter was significantly improved by introducing a leader open reading frame in front of the gene. Similar improvement under a weak promoter, tet, with a leader open reading frame had been described previously. The present results provide evidence that the major limiting step in the expression of a GC-rich thermophile gene in E. coli is translation, and that the addition of a leader open reading frame is more crucial for high level expression of the gene than the use of a potent promoter. To improve
expression
efficiency
of the
/euB
coli, the gene was placed under a potent promoter,
Keywords:
Promoter
Gene expression:
K-rich
DNA; Leader open reading
frame; Extreme thermophile;
Thermus thermophilus:
E.tcherichitr
c.oli;
f(u
1. Introduction Most genes isolated from an extreme thermophile, Thermus thermophilus. are scarcely or poorly expressed in Escherichia coli, even when they are cloned into general expression vectors for E. coli. To improve the expression of Thermus genes in E. coli, we investigated the expression mechanism of the T. thermophilus 1euB gene, coding for 3-isopropylmalate dehydrogenase (EC 1.1.1.85), and its mutant genes in E. coli. In our previous study [l], it was speculated that the translation of 1euB was inhibited
_ Corresponding
author.
Tel.: + 8 I (3) 5463 0586; Fax: + 8 I
(3) 5463 0589. 037%1097/96/$12.00 0 1996 Federation SSDl 037X- 1097(95)00460-2
of European
Microbiological
by secondary structure(s) in the translational initiation region of the mRNA (G + C content of the thermophile DNA is about 70% [2]), and that the inhibition was prevented by introduction of a leader open reading frame (ORF). In the previous study [l], a weak promoter, tet, of the tetracycline resistance (Tet’) gene was used, and expression efficiency under a strong promoter was not tested. Since one of the general strategies to improve gene exprelssion is the use of a strong promoter such as tat, it must be confirmed whether or not the expression of the thermophile gene in E. coli is improved by the use of tat promoter. If the expression efficiency of’ 1euB is improved by the promoter exchange, another question is whether or not the efficiency is further imSocieties.
All rights reserved
M. Ishida, T Oshima/
138
FEMS Microbiology
Letters I35 (19961 137-142
plasmid, pCM4 [4], carrying the chloramphenicol acetyltransferase gene (designated as cat in this paper) was purchased from Pharmacia Biotech (Tokyo). E. coli strain JA221 (F-, hstR, AtrpE.5, leuB6, lacy, recA1) [5] was used as a host.
proved by the addition of a leader ORF. The present study was carried out to answer these questions.
2. Materials and methods
2.1. Reagents, plasmids, and bacterial strains
2.2. Molecular mids
3-Isopropylmalate was purchased from Wako Pure Chemical Industries, Ltd. (Tokyo). A T. thermophilus lea&expressible plasmid, pHB2, was described in our previous papers (Fig. 1) [1,3]. A
manipulations
of recombinant
plas-
Plasmid preparation and bacterial transformation were carried out according to the procedures described by Sambrook et al. [6]. The filling-in reaction
Lell
IPMDH activity
phenotype
[units (g of cei~s)J x103]
IPMDH
(4
pHB2
(B)
ptacll6
+t
45.4
IPMDH
pTHB12
<0.5
I
< 0.5
ptacll2
+
Cm reslstance [(pgof (c)
pHBZ
<50 Sm
I
0
Cm) ml-l]
Sm
CAT
Afl
I
pHB2-SCM
250
pHBP-ACM
250
pTHB-CAT
200
pTHBl2-CAT
450 Bm
ptac-CAT
600 -
:200bp
Fig. I. The effects of promoter strength on the expressions of the T. thermophilus kuB gene and the reporter cat gene. (A) The expression of leuB under a weak promoter, let, on pHB2. A leader ORF (Tet”) is in pHB2. (Bl The expression of 1euB under a strong promoter, tat. The thermophile DNA fragments containing 1euB on pHB2 were replaced downstream of the tat promoter in ptacI12, pTHB121, and ptacI16. (Cl The expression of cat under the tef promoter. The cat gene was inserted into the 1euB coding region (pHB2-ACM) or the 5’ non-coding region (pHB2-SCM). (D) The expression of cat under the tuc promoter. Closed boxes, open arrows, and shaded boxes indicate the coding regions of T. thermophilus 1euB (marked with ‘IPMDH’), the Tet”-leader ORFs, and the cat coding region (marked with ‘CAT’), respectively. Open triangle, closed triangle, and open and closed circles indicate tet promoter, tuc promoter and ribosome binding sites, respectively. Ail, AflII site; Bg, BgrII site; Bm, BamHI site; and Sm. SmaI site. The details of the recombinant bacterial cultivation, ‘Cm the assays for 3-isopropylmalate dehydrogenase, ‘IPMDH’, the leucine phenotype, ‘Leu’, and the resistance to chloramphenicol, resistance’, were described in Materials and methods.
M. Ishida.
T. Oshima / FEMS Microhiolog~
with the Klenow fragment was done as described by Cobianchi and Wilson [7]. The 0.8-kb BamHI fragment encoding car from pCM4 was inserted into the BamHI or Bglll site of the plasmids carrying T. thermophilus 1euB. The same BamHI fragment was also recloned into the BumHI site of pUC19, the composite plasmid was digested with Smul and Sull, and used as a SmuI-Sul I cut cartridge. pHB2 was cleaved with &III, end-flattened by the Klenow fragment, cleaved with Sull, and ligated with the Smul-Sull cut cartridge. 2.3. Bacterial growth and phenotypes
assay
E. coli JA221 was grown in L broth [8] or M9 minimal medium [6]. The supplementation level of amino acids to the medium was as described by Rodriguez and Tait [9]. Ampicillin sodium salt was added to 50 pg ml-‘, when required. A Leu phenotype of E. coli JA221 harboring a plasmid was tested on the minimal medium with ampicillin, and its growth rate was estimated by measuring the time-dependent increase of the apparent optical density at 600 nm in the exponential phase of growth. A chloramphenicol resistance phenotype of E. coli harboring a plasmid was tested in YT medium with chloramphenicol (100 Kg ml -’ ). The production level of chloramphenicol acetyltransferase (CAT) was
tacp 4”
CAT
Bm L I
Letters
139
135 C1996) 137-142
determined by the maximum concentration of chloramphenicol in YT medium, in which the E. coli cells grew after shaking for 20 h at 37°C. 2.4. Isopropylmulute
dehydrogenuse
assay
To estimate 3-isopropylmalate dehydrogenase activity, recombinant cells were grown at 37°C for 20 h. The cells were collected, washed and disrubted by ultrasonication in 0.1 M potassium phosphate buffer (pH 7.8). The cell-free extract was heated at 70°C for 30 min, centrifuged at 12 000 X g for 1.5 main, and the supernatant was used for the enzyme assay. Activity was measured according to Yamada et al. [ 101, except that the reaction was carried out at 30°C. One unit of activity was defined as 1 pmol of NADH produced per min.
3. Results 3.1. 1euB expression
under a strong promoter,
In order to test the expression 1euB under the strong promoter,
tat
of T. theryophilus tuc, the thermophile
DNA fragments containing the 1euB coding region without or with the 5’ non-coding region on pHB2 (Fig. 1A) were replaced downstream of the tuc
Cm’
Lell
phenotype
phenotype
IPMDH activity [units (g of cells~l xl@]
$9
Ef” IPMDH
ptacll6-2CAT
ptacllS-PCAT-IV
ptacll2-2CAT
~‘PMPH
-
+
: 200
-
bp
Fig. 2. The effects of introduction of a leader ORF (cat) on the expression of T. thermophilus la&. Chloramphenicol-resistant phenotype is marked as ‘Cm”. Other symbols and marks in the figure are the same as those in Fig. I. The details of the recombinant bacterial chltivation. and the assays for the phenotypes and enzyme activity are described in Materials and methods.
140
M. Ishida.
T. Oshima / FEMS Micmbiolog~
promoter (Fig. 1B). Only the leuB-coding fragment (a 1.15kb BamHI fragment) was cloned in ptacI12, while the 5’ non-coding fragments were accompanied with the gene fragment in pTHB 121 and ptacI16. None of these recombinant plasmids expressed the thermophile 1euB as shown in Fig. IB. No leader ORF was found in these plasmids with the tuc promoter. Although E. coli JA221 (ptacI12) could grow in the minimal medium (Fig. IB), the growth rate was lower than that of E. co/i JA221 (pHB2): the time-dependent increase of the apparent optical density of E. coli JA221 (ptacIl2) culture was about half that of E. coli JA221 (pHB2). The finding suggests that the expression efficiency of T. thermophilus 1euB on ptacI12 is so weak that enzyme activity is not detected in the heat-treated extract from the cells. To monitor the transcription activity on pHB2 and other plasmids with the tat promoter, the cat gene was inserted as a reporter gene into the plasmids IeuB (Fig. lC, D). In contrast to the thermophile (Fig. lB), cut was expressed from all plasmids with the tuc promoter (Fig. 1D). Stronger CAT was produced from the plasmids with the tuc promoter (Fig. 1D) than ,pHB2 with the tet promoter (Fig. lC>. However, the longer the distance between the promoter and the coding region, the lower was the expression (Fig. lD), suggesting that the 5’ non-coding region of T. thermophilus 1euB is inhibitory for the transcription in E. coli. These results strongly suggest that the expression efficiency of the thermophile .leuB is not related to promoter strength alone. 3.2. Efsects of introduction
of a cut leader ORF
The cut gene was introduced as a 6.57-bp leader ORF into the 5’ non-coding region of T. thermophilus 1euB under the tuc promoter in ptacI161CAT or ptac16-2CAT (Fig. 2). These insertions resulted in significant improvement of 3-isopropylmalate dehydrogenase production. Electrophoretic analysis confirmed the production of 3-isopropylmalate dehydrogenase (3.6 kDa) in the heat-treated extracts from both cells (data not shown). The inserted cut gene was also expressed, since the recombinant E. coli cells were resistant to chloramphenicol. On the other hand, when cat was inserted in
Lettrcc
135
(19961137-
142
the inverted orientation in ptacIl6-2CAT-IV, neither 3-isopropylmalate dehydrogenase nor CAT was synthesized. Also, when cat was introduced downstream of the coding region of 1euB in ptacl I2-2CAT, 3-isopropylmalate dehydrogenase was hardly detected, although CAT was synthesized (the growth of E. coli JA22 1 harboring ptacI12-2CAT in the presence of chloramphenicol was weaker than that of E. coli JA221 harboring ptacIl6-ICAT or ptacl62CAT). These results indicate that the insertion of cat as a leader ORF considerably improved the expression of T. thermophilus IruB, and suggest that the position and orientation of the ORF inserted is important.
4. Discussion The use of the potent promoter alone hardly improves 3-isopropylmalate dehydrogenase production. This fact strongly suggests that the major limiting step in the expression of T. thermophilus 1euB in E. coli is translation rather than transcription. This speculation is also supported by our previous finding that the gene was expressed on a plasmid without a leader ORF if the secondary structure around the translational initiation region of the thermophile LeuB was destructed [l]. The present analysis using the reporter cut gene shows that the transcription activity under the tuc‘ promoter was lowered by the addition of the thermophile DNA region (Fig. ID and Fig. 2). Such an inhibitory effect of the T. thermophilus DNA on the transcription may be due to the formation of secondary structures because of the high G + C content of the thermophile IeuB DNA [I I]. A computer analysis showed that many possible secondary structures ( AG 5 20 kcal molt ’ ), including the translational inhibitory structures [ 11, were found in various places of both 5’ non-coding and coding regions of T. thermophilus 1euB (data not shown), and some of them could function as transcriptional terminators or transcriptional pause sites in E. coli. On the other hand, T. thermophilus 3-isopropylmalate dehydrogenase was efficiently synthesized from the plasmids containing a leader ORF between the tuc promoter and the 1euB coding region, although the inhibitory secondary structure remained
M. Ishida. T. Oshimcl/ FEMS Micmhiolog~ Letters 135 f 1996)137- 142
unchanged. The introduction of a leader ORF was essential to express the thermophile 1euB despite the use of a weak, fer [ 11, or a strong, tat, promoter. The present study strongly suggests that for efficient expression of a GC-rich thermophile gene in E. co/i, the insertion of a leader ORF is more effective than a strong promoter. When two tandem ORFs are closely placed or overlap slightly, activation of the translation of a downstream gene by the upstream translation is known as translational coupling [ 12-181. In translational coupling, it is suggested that the ribosomes initiate the translation at the upstream ORF, disrupt the inhibitory secondary structure on mRNA of the downstream ORF, and re-initiate the translation at the downstream ORF [ 12.191. The leader ORFs used in the present study are considerably distant (about 0.2.5-0.5 kb) from the 1euB coding region (Fig. 2). The effect of the distance will be a subject of future study. In conclusion, the present study clearly demonstrates that the addition of a leader ORF is essential for enhanced expression of the thermophile gene even under the control of a strong promoter. Although it is difficult to test whether the E. coli ribosomes are capable of ‘scanning diffusion’ like eukaryotic ribosomes [20], the present study suggests that the E. coli ribosomes after translating the leader ORF migrate along mRNA to the downstream gene in a ‘scanning diffusion’ fashion, disrupt the inhibitory secondary structure, and then initiate the translation of it.
Acknowledgements This work was partly supported by a Grant-in-Aid for Scientific Research 60060004 from the Ministry of Education, Science, Sports, and Culture of the Japanese Government.
References [II Ishida, M. and Oshima, T. (1994) Overexpression
of genes of an extreme thermophile, Thermus thermophilus, in Esc~hrrichitr co/i cells. J. Bacterial. 176, 2767-2770.
141
Dl Oshima. T. and Imahori, K. (1974) Description of Thrnnus thrrntophilus (Yoshida and Oshima) comb. nov., a nonsporulating thermophilic bacterium from a Japanese thermal spa. Int. .I. Syst. Bacterial. 24, 102-l 12. [31 Tanaka. T., Kawano, N. and Oshima. T. (I98 I ) Cloning of 3-isopropylmalate dehydrogenase gene of an extrlme thermophile and partial purification of the gene ptoduct. J. Biochem. 89. 677-682. [Jl Close. T.J. and Rodriguez, R.L. (1982) Construy-tion and characterization of the chloramphenicol-resistance ‘gene cu-tridge: a new approach to the transcriptional mhpping ot extrachromosomal elements. Gene 20. 305-3 16. [51 Almond. J.. Beecroft, L.. Blair, E.. Meacock. P. and Windass, J. (1985) Transformation of competent E. cali cells by plasmid DNA. In: Basic Cloning Techniques: A Manual of Experimental Procedures (Pritchard, R.H. and Holland, I.B.. Eds.). pp. 3.5-38. Blackwell Scientific Publication% London. [61 Sambrook, J.. Fritach. E.F. and Maniati\. T. (19x91) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor. NY. [71 Cobianchi. F. and Wilson. S.H. (1987) Enzymes for modifying and labeling DNA and RNA. Methods Enzymol. 152. 93- I IO. and preservation of [81 Morrison, D.A. ( 1979) Transformation competent bacterial cells by freezing. Methods Enzymol. 68, 326-33 I DNA [91 Rodriguez. R.L. and Tait, R.C. (1983) Recombinant techniques: an introduction. Addison-Wesley Publi’shing Co., Massachusetts. K.. Kakimuma. K.. 1101 Yemada. T.. Akutsu. N.. Miyazaki. Yoshida. M. and Oshima. T. (1990) Purification, catalytic properties. and thermal stability of thrro-Ds-3.isapropylmalate dehydrogenase coded by /euB gene from a01 extreme thermophile. Tkrrrtus thrn~zophi/u.\ strain HB8. J. Biochem. 108, 449-456. 1111 Kagawa. Y., Nojima, H.. Nukima. N.. Ishizuka, M.. Nakajima, T.. Yasuhara, T., Tanaka. T. and Oshima, T. (lY84) High guanine plus cytosine content in the third letter ot codona of an extreme thermophile. J. Biol. Chem. 259. 2956-2960. 1121 Berkhout, B., Schmidt. B.F., van Strien, A.. van Boom, J.. van Westrenen. J. and van Duin, J. (1987) Lyaiz, gene of bacteriophage M I2 is activated by translation termination at the overlapping coat gene. J. Mol. Biol. IYS. 517-524. [I31 Comeau, D.E., Ikenaka. K., Tsung, K. and Inouye, M. (1985) Primary characterization of the protein products of the Esc,hrrichitr co/i on@ locus: structure and regulation of aynthesis of the OmpR and EnvZ protein\. J. Bacterial. 164. 578-584. [I41 Gold, L. and Stormo. G. (1987) Translational initiation. In: Escherichio co/i and Sahnonrlla typhimurium C!zllular and Molecular Biology (Neidhardt. F.C.. Ingraham. J.L., Low. K.B.. Magasanik. B. and Umbarger. H.E.. Eds.), pp. 13021307. American Society for Microbiology, Washington, D.C. 1151 Lijestroem. P.. Laamanen, 1. and Palva, E.T. (19X8) Structure and expression of the or@ operon, the regulatory locus for
142
M. Ishida.
the
outer
T. Oshima / FEMS Microbiology
membrane porin regulon in Salmonrlln fyJ. Mol. Biol. 201, 663-673. [I61 Schoner, B.E., Belagaje, R.M. and Schoner, R.G. (19861 Translation of a synthetic two-cistron mRNA in Eschrrichitr coli. Proc. Natl. Acad. Sci. USA 83, 8506-8510. [ 171 Spanjaard, R.A. and van Duin, J. (1989) Translational reinitiation in the presence and absence of a Shine and Dalgarno sequence. Nucleic Acids Res. 17, 550 l-5507. [ 181 Yanofsky, C. and Crawford, I.P. (19871 The tryptophan operon. In: Eschrrichio co/i and Salmonella tyhimurium Cellular and Molecular Biology (Neidhardt. F.C., Ingraham. phimurium.
Letters
135 C19%) 137- 142
J.L., Low, K.B., Magasanik, B. and Umbarger, H.E.. Ed.\.), pp. 1453-1472. American Society for Microbiology, Washington. D.C. [19] Schmidt, B.F., Berkhout, B., Overbeek. G.P., van Strien. A. and van Duin, J. (19871 Determination of the RNA secondary structure that regulates lysis gene expression in bacteriophage MS2. I. Mol. Biol. 195, 505-516. [20] Kozak. M. (19831 Compilation and analysis of sequences upstream from the translational start site in eukaryotic mRNAs. Microbial. Rev. 47, I-45.
FEMS Microbiology
Letters I35 ( 1996) 137- 142
A leader open reading frame is essential for the expression in Escherichia coli of GC-rich ZeuB gene of an extreme thermophile, Thermus thermophilus Masami Ishida a,x, Tairo Oshima b ” Laboruto~
of Biochemistry of Marine of Molecular Biolog!.
h Deptrrtmrnt
Resources.
Tokw
Received 7 November
Utlil,ersie
of Pharmacy
Tokyo University
1995; accepted
of Fisheries,
Konm
4, Minato-ku,
and L(fe Science, Huchioji,
13 November
Tokyo 108, Japm
Tokyo 1 YZ-03. Jtrport
1995
Abstract gene of an extreme thermophile, Thrrmus thermophilus, in Escherichitr tuc. However, the expression was hardly improved, despite increased transcription. The expression under tat promoter was significantly improved by introducing a leader open reading frame in front of the gene. Similar improvement under a weak promoter, tet, with a leader open reading frame had been described previously. The present results provide evidence that the major limiting step in the expression of a GC-rich thermophile gene in E. coli is translation, and that the addition of a leader open reading frame is more crucial for high level expression of the gene than the use of a potent promoter. To improve
expression
efficiency
of the
/euB
coli, the gene was placed under a potent promoter,
Keywords:
Promoter
Gene expression:
K-rich
DNA; Leader open reading
frame; Extreme thermophile;
Thermus thermophilus:
E.tcherichitr
c.oli;
f(u
1. Introduction Most genes isolated from an extreme thermophile, Thermus thermophilus. are scarcely or poorly expressed in Escherichia coli, even when they are cloned into general expression vectors for E. coli. To improve the expression of Thermus genes in E. coli, we investigated the expression mechanism of the T. thermophilus 1euB gene, coding for 3-isopropylmalate dehydrogenase (EC 1.1.1.85), and its mutant genes in E. coli. In our previous study [l], it was speculated that the translation of 1euB was inhibited
_ Corresponding
author.
Tel.: + 8 I (3) 5463 0586; Fax: + 8 I
(3) 5463 0589. 037%1097/96/$12.00 0 1996 Federation SSDl 037X- 1097(95)00460-2
of European
Microbiological
by secondary structure(s) in the translational initiation region of the mRNA (G + C content of the thermophile DNA is about 70% [2]), and that the inhibition was prevented by introduction of a leader open reading frame (ORF). In the previous study [l], a weak promoter, tet, of the tetracycline resistance (Tet’) gene was used, and expression efficiency under a strong promoter was not tested. Since one of the general strategies to improve gene exprelssion is the use of a strong promoter such as tat, it must be confirmed whether or not the expression of the thermophile gene in E. coli is improved by the use of tat promoter. If the expression efficiency of’ 1euB is improved by the promoter exchange, another question is whether or not the efficiency is further imSocieties.
All rights reserved
M. Ishida, T Oshima/
138
FEMS Microbiology
Letters I35 (19961 137-142
plasmid, pCM4 [4], carrying the chloramphenicol acetyltransferase gene (designated as cat in this paper) was purchased from Pharmacia Biotech (Tokyo). E. coli strain JA221 (F-, hstR, AtrpE.5, leuB6, lacy, recA1) [5] was used as a host.
proved by the addition of a leader ORF. The present study was carried out to answer these questions.
2. Materials and methods
2.1. Reagents, plasmids, and bacterial strains
2.2. Molecular mids
3-Isopropylmalate was purchased from Wako Pure Chemical Industries, Ltd. (Tokyo). A T. thermophilus lea&expressible plasmid, pHB2, was described in our previous papers (Fig. 1) [1,3]. A
manipulations
of recombinant
plas-
Plasmid preparation and bacterial transformation were carried out according to the procedures described by Sambrook et al. [6]. The filling-in reaction
Lell
IPMDH activity
phenotype
[units (g of cei~s)J x103]
IPMDH
(4
pHB2
(B)
ptacll6
+t
45.4
IPMDH
pTHB12
<0.5
I
< 0.5
ptacll2
+
Cm reslstance [(pgof (c)
pHBZ
<50 Sm
I
0
Cm) ml-l]
Sm
CAT
Afl
I
pHB2-SCM
250
pHBP-ACM
250
pTHB-CAT
200
pTHBl2-CAT
450 Bm
ptac-CAT
600 -
:200bp
Fig. I. The effects of promoter strength on the expressions of the T. thermophilus kuB gene and the reporter cat gene. (A) The expression of leuB under a weak promoter, let, on pHB2. A leader ORF (Tet”) is in pHB2. (Bl The expression of 1euB under a strong promoter, tat. The thermophile DNA fragments containing 1euB on pHB2 were replaced downstream of the tat promoter in ptacI12, pTHB121, and ptacI16. (Cl The expression of cat under the tef promoter. The cat gene was inserted into the 1euB coding region (pHB2-ACM) or the 5’ non-coding region (pHB2-SCM). (D) The expression of cat under the tuc promoter. Closed boxes, open arrows, and shaded boxes indicate the coding regions of T. thermophilus 1euB (marked with ‘IPMDH’), the Tet”-leader ORFs, and the cat coding region (marked with ‘CAT’), respectively. Open triangle, closed triangle, and open and closed circles indicate tet promoter, tuc promoter and ribosome binding sites, respectively. Ail, AflII site; Bg, BgrII site; Bm, BamHI site; and Sm. SmaI site. The details of the recombinant bacterial cultivation, ‘Cm the assays for 3-isopropylmalate dehydrogenase, ‘IPMDH’, the leucine phenotype, ‘Leu’, and the resistance to chloramphenicol, resistance’, were described in Materials and methods.
M. Ishida.
T. Oshima / FEMS Microhiolog~
with the Klenow fragment was done as described by Cobianchi and Wilson [7]. The 0.8-kb BamHI fragment encoding car from pCM4 was inserted into the BamHI or Bglll site of the plasmids carrying T. thermophilus 1euB. The same BamHI fragment was also recloned into the BumHI site of pUC19, the composite plasmid was digested with Smul and Sull, and used as a SmuI-Sul I cut cartridge. pHB2 was cleaved with &III, end-flattened by the Klenow fragment, cleaved with Sull, and ligated with the Smul-Sull cut cartridge. 2.3. Bacterial growth and phenotypes
assay
E. coli JA221 was grown in L broth [8] or M9 minimal medium [6]. The supplementation level of amino acids to the medium was as described by Rodriguez and Tait [9]. Ampicillin sodium salt was added to 50 pg ml-‘, when required. A Leu phenotype of E. coli JA221 harboring a plasmid was tested on the minimal medium with ampicillin, and its growth rate was estimated by measuring the time-dependent increase of the apparent optical density at 600 nm in the exponential phase of growth. A chloramphenicol resistance phenotype of E. coli harboring a plasmid was tested in YT medium with chloramphenicol (100 Kg ml -’ ). The production level of chloramphenicol acetyltransferase (CAT) was
tacp 4”
CAT
Bm L I
Letters
139
135 C1996) 137-142
determined by the maximum concentration of chloramphenicol in YT medium, in which the E. coli cells grew after shaking for 20 h at 37°C. 2.4. Isopropylmulute
dehydrogenuse
assay
To estimate 3-isopropylmalate dehydrogenase activity, recombinant cells were grown at 37°C for 20 h. The cells were collected, washed and disrubted by ultrasonication in 0.1 M potassium phosphate buffer (pH 7.8). The cell-free extract was heated at 70°C for 30 min, centrifuged at 12 000 X g for 1.5 main, and the supernatant was used for the enzyme assay. Activity was measured according to Yamada et al. [ 101, except that the reaction was carried out at 30°C. One unit of activity was defined as 1 pmol of NADH produced per min.
3. Results 3.1. 1euB expression
under a strong promoter,
In order to test the expression 1euB under the strong promoter,
tat
of T. theryophilus tuc, the thermophile
DNA fragments containing the 1euB coding region without or with the 5’ non-coding region on pHB2 (Fig. 1A) were replaced downstream of the tuc
Cm’
Lell
phenotype
phenotype
IPMDH activity [units (g of cells~l xl@]
$9
Ef” IPMDH
ptacll6-2CAT
ptacllS-PCAT-IV
ptacll2-2CAT
~‘PMPH
-
+
: 200
-
bp
Fig. 2. The effects of introduction of a leader ORF (cat) on the expression of T. thermophilus la&. Chloramphenicol-resistant phenotype is marked as ‘Cm”. Other symbols and marks in the figure are the same as those in Fig. I. The details of the recombinant bacterial chltivation. and the assays for the phenotypes and enzyme activity are described in Materials and methods.
140
M. Ishida.
T. Oshima / FEMS Micmbiolog~
promoter (Fig. 1B). Only the leuB-coding fragment (a 1.15kb BamHI fragment) was cloned in ptacI12, while the 5’ non-coding fragments were accompanied with the gene fragment in pTHB 121 and ptacI16. None of these recombinant plasmids expressed the thermophile 1euB as shown in Fig. IB. No leader ORF was found in these plasmids with the tuc promoter. Although E. coli JA221 (ptacI12) could grow in the minimal medium (Fig. IB), the growth rate was lower than that of E. co/i JA221 (pHB2): the time-dependent increase of the apparent optical density of E. coli JA221 (ptacIl2) culture was about half that of E. coli JA221 (pHB2). The finding suggests that the expression efficiency of T. thermophilus 1euB on ptacI12 is so weak that enzyme activity is not detected in the heat-treated extract from the cells. To monitor the transcription activity on pHB2 and other plasmids with the tat promoter, the cat gene was inserted as a reporter gene into the plasmids IeuB (Fig. lC, D). In contrast to the thermophile (Fig. lB), cut was expressed from all plasmids with the tuc promoter (Fig. 1D). Stronger CAT was produced from the plasmids with the tuc promoter (Fig. 1D) than ,pHB2 with the tet promoter (Fig. lC>. However, the longer the distance between the promoter and the coding region, the lower was the expression (Fig. lD), suggesting that the 5’ non-coding region of T. thermophilus 1euB is inhibitory for the transcription in E. coli. These results strongly suggest that the expression efficiency of the thermophile .leuB is not related to promoter strength alone. 3.2. Efsects of introduction
of a cut leader ORF
The cut gene was introduced as a 6.57-bp leader ORF into the 5’ non-coding region of T. thermophilus 1euB under the tuc promoter in ptacI161CAT or ptac16-2CAT (Fig. 2). These insertions resulted in significant improvement of 3-isopropylmalate dehydrogenase production. Electrophoretic analysis confirmed the production of 3-isopropylmalate dehydrogenase (3.6 kDa) in the heat-treated extracts from both cells (data not shown). The inserted cut gene was also expressed, since the recombinant E. coli cells were resistant to chloramphenicol. On the other hand, when cat was inserted in
Lettrcc
135
(19961137-
142
the inverted orientation in ptacIl6-2CAT-IV, neither 3-isopropylmalate dehydrogenase nor CAT was synthesized. Also, when cat was introduced downstream of the coding region of 1euB in ptacl I2-2CAT, 3-isopropylmalate dehydrogenase was hardly detected, although CAT was synthesized (the growth of E. coli JA22 1 harboring ptacI12-2CAT in the presence of chloramphenicol was weaker than that of E. coli JA221 harboring ptacIl6-ICAT or ptacl62CAT). These results indicate that the insertion of cat as a leader ORF considerably improved the expression of T. thermophilus IruB, and suggest that the position and orientation of the ORF inserted is important.
4. Discussion The use of the potent promoter alone hardly improves 3-isopropylmalate dehydrogenase production. This fact strongly suggests that the major limiting step in the expression of T. thermophilus 1euB in E. coli is translation rather than transcription. This speculation is also supported by our previous finding that the gene was expressed on a plasmid without a leader ORF if the secondary structure around the translational initiation region of the thermophile LeuB was destructed [l]. The present analysis using the reporter cut gene shows that the transcription activity under the tuc‘ promoter was lowered by the addition of the thermophile DNA region (Fig. ID and Fig. 2). Such an inhibitory effect of the T. thermophilus DNA on the transcription may be due to the formation of secondary structures because of the high G + C content of the thermophile IeuB DNA [I I]. A computer analysis showed that many possible secondary structures ( AG 5 20 kcal molt ’ ), including the translational inhibitory structures [ 11, were found in various places of both 5’ non-coding and coding regions of T. thermophilus 1euB (data not shown), and some of them could function as transcriptional terminators or transcriptional pause sites in E. coli. On the other hand, T. thermophilus 3-isopropylmalate dehydrogenase was efficiently synthesized from the plasmids containing a leader ORF between the tuc promoter and the 1euB coding region, although the inhibitory secondary structure remained
M. Ishida. T. Oshimcl/ FEMS Micmhiolog~ Letters 135 f 1996)137- 142
unchanged. The introduction of a leader ORF was essential to express the thermophile 1euB despite the use of a weak, fer [ 11, or a strong, tat, promoter. The present study strongly suggests that for efficient expression of a GC-rich thermophile gene in E. co/i, the insertion of a leader ORF is more effective than a strong promoter. When two tandem ORFs are closely placed or overlap slightly, activation of the translation of a downstream gene by the upstream translation is known as translational coupling [ 12-181. In translational coupling, it is suggested that the ribosomes initiate the translation at the upstream ORF, disrupt the inhibitory secondary structure on mRNA of the downstream ORF, and re-initiate the translation at the downstream ORF [ 12.191. The leader ORFs used in the present study are considerably distant (about 0.2.5-0.5 kb) from the 1euB coding region (Fig. 2). The effect of the distance will be a subject of future study. In conclusion, the present study clearly demonstrates that the addition of a leader ORF is essential for enhanced expression of the thermophile gene even under the control of a strong promoter. Although it is difficult to test whether the E. coli ribosomes are capable of ‘scanning diffusion’ like eukaryotic ribosomes [20], the present study suggests that the E. coli ribosomes after translating the leader ORF migrate along mRNA to the downstream gene in a ‘scanning diffusion’ fashion, disrupt the inhibitory secondary structure, and then initiate the translation of it.
Acknowledgements This work was partly supported by a Grant-in-Aid for Scientific Research 60060004 from the Ministry of Education, Science, Sports, and Culture of the Japanese Government.
References [II Ishida, M. and Oshima, T. (1994) Overexpression
of genes of an extreme thermophile, Thermus thermophilus, in Esc~hrrichitr co/i cells. J. Bacterial. 176, 2767-2770.
141
Dl Oshima. T. and Imahori, K. (1974) Description of Thrnnus thrrntophilus (Yoshida and Oshima) comb. nov., a nonsporulating thermophilic bacterium from a Japanese thermal spa. Int. .I. Syst. Bacterial. 24, 102-l 12. [31 Tanaka. T., Kawano, N. and Oshima. T. (I98 I ) Cloning of 3-isopropylmalate dehydrogenase gene of an extrlme thermophile and partial purification of the gene ptoduct. J. Biochem. 89. 677-682. [Jl Close. T.J. and Rodriguez, R.L. (1982) Construy-tion and characterization of the chloramphenicol-resistance ‘gene cu-tridge: a new approach to the transcriptional mhpping ot extrachromosomal elements. Gene 20. 305-3 16. [51 Almond. J.. Beecroft, L.. Blair, E.. Meacock. P. and Windass, J. (1985) Transformation of competent E. cali cells by plasmid DNA. In: Basic Cloning Techniques: A Manual of Experimental Procedures (Pritchard, R.H. and Holland, I.B.. Eds.). pp. 3.5-38. Blackwell Scientific Publication% London. [61 Sambrook, J.. Fritach. E.F. and Maniati\. T. (19x91) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor. NY. [71 Cobianchi. F. and Wilson. S.H. (1987) Enzymes for modifying and labeling DNA and RNA. Methods Enzymol. 152. 93- I IO. and preservation of [81 Morrison, D.A. ( 1979) Transformation competent bacterial cells by freezing. Methods Enzymol. 68, 326-33 I DNA [91 Rodriguez. R.L. and Tait, R.C. (1983) Recombinant techniques: an introduction. Addison-Wesley Publi’shing Co., Massachusetts. K.. Kakimuma. K.. 1101 Yemada. T.. Akutsu. N.. Miyazaki. Yoshida. M. and Oshima. T. (1990) Purification, catalytic properties. and thermal stability of thrro-Ds-3.isapropylmalate dehydrogenase coded by /euB gene from a01 extreme thermophile. Tkrrrtus thrn~zophi/u.\ strain HB8. J. Biochem. 108, 449-456. 1111 Kagawa. Y., Nojima, H.. Nukima. N.. Ishizuka, M.. Nakajima, T.. Yasuhara, T., Tanaka. T. and Oshima, T. (lY84) High guanine plus cytosine content in the third letter ot codona of an extreme thermophile. J. Biol. Chem. 259. 2956-2960. 1121 Berkhout, B., Schmidt. B.F., van Strien, A.. van Boom, J.. van Westrenen. J. and van Duin, J. (1987) Lyaiz, gene of bacteriophage M I2 is activated by translation termination at the overlapping coat gene. J. Mol. Biol. IYS. 517-524. [I31 Comeau, D.E., Ikenaka. K., Tsung, K. and Inouye, M. (1985) Primary characterization of the protein products of the Esc,hrrichitr co/i on@ locus: structure and regulation of aynthesis of the OmpR and EnvZ protein\. J. Bacterial. 164. 578-584. [I41 Gold, L. and Stormo. G. (1987) Translational initiation. In: Escherichio co/i and Sahnonrlla typhimurium C!zllular and Molecular Biology (Neidhardt. F.C.. Ingraham. J.L., Low. K.B.. Magasanik. B. and Umbarger. H.E.. Eds.), pp. 13021307. American Society for Microbiology, Washington, D.C. 1151 Lijestroem. P.. Laamanen, 1. and Palva, E.T. (19X8) Structure and expression of the or@ operon, the regulatory locus for
142
M. Ishida.
the
outer
T. Oshima / FEMS Microbiology
membrane porin regulon in Salmonrlln fyJ. Mol. Biol. 201, 663-673. [I61 Schoner, B.E., Belagaje, R.M. and Schoner, R.G. (19861 Translation of a synthetic two-cistron mRNA in Eschrrichitr coli. Proc. Natl. Acad. Sci. USA 83, 8506-8510. [ 171 Spanjaard, R.A. and van Duin, J. (1989) Translational reinitiation in the presence and absence of a Shine and Dalgarno sequence. Nucleic Acids Res. 17, 550 l-5507. [ 181 Yanofsky, C. and Crawford, I.P. (19871 The tryptophan operon. In: Eschrrichio co/i and Salmonella tyhimurium Cellular and Molecular Biology (Neidhardt. F.C., Ingraham. phimurium.
Letters
135 C19%) 137- 142
J.L., Low, K.B., Magasanik, B. and Umbarger, H.E.. Ed.\.), pp. 1453-1472. American Society for Microbiology, Washington. D.C. [19] Schmidt, B.F., Berkhout, B., Overbeek. G.P., van Strien. A. and van Duin, J. (19871 Determination of the RNA secondary structure that regulates lysis gene expression in bacteriophage MS2. I. Mol. Biol. 195, 505-516. [20] Kozak. M. (19831 Compilation and analysis of sequences upstream from the translational start site in eukaryotic mRNAs. Microbial. Rev. 47, I-45.