Appl Microbiol Biotechnol(1990) 33:657-663

Applied Microbiology Biotechnology © Springer-Verlag 1990

Regulated expression of heterologous genes in Bacillus subtilis using the Tnl0 encoded tet regulatory elements Manfred Geissendiirfer and Wolf gang Hillen

Lehrstuhl fiJr Mikrobiologie, Institut f~irMikrobiologieund Biochemie, Friedrich-AlexanderUniversit~t Erlangen-NiJrnberg, Staudtstrasse 5, D-8520 Erlangen, Federal Republic of Germany Received 11 April 1990/Acepted 7 May 1990

Summary. The Escherichia coli-derived tet regulatory elements from TnlO have been used to construct vectors allowing the regulated, inducible, high-level expression of foreign genes in Bacillus subtilis. While the wild-type tet promoters are inactive in B. subtilis, a synthetic mutant tet sequence with improved promoter consensus sequences and upstream poly A blocks shows activity in B. subtilis. The expression of an indicator cat gene is inducible by sublethal amounts of tetracycline, indicating that the Tet repressor protein and the tet operator sequences are functional. However, the inducibility and maximal expression are not sufficient in this construct. To improve these properties a tet operator sequence was placed between the - 3 5 and - 1 0 boxes of the B. subtilis-derived very strong xyl promoter. In the presence of a tetR gene this construct is about 100-fold inducible and has high promoter strength, but some basal expression. This is avoided by placing a second tet operator downstream resulting in no detectable basal expression at the expense of reduced inducibility. Using the system with a single tet operator inducible expression of glucose dehydrogenase from B. megaterium was obtained at a very high level, and inducible expression of human single-chain urokinase-like plasminogen activator was achieved at the same level as in E. coli. Unlike in E. coli, the product was not degraded up to 4 h after induction in B. subtilis. These results demonstrate that the regulated expression vector described here should be very useful for production of foreign gene products from B. subtilis cultures.

Introduction

Members of the genus Bacillus have attractive properties for the expression of industrially important heterologous genes. Among them are the ability to secrete large amounts of protein into culture broth, the appar-

Offprint requests to: W. Hillen

ent lack of strains with human pathogenicity and the large experience regarding fermentation of various species on an industrial scale. The cloning of genes in B. subtilis, however, is hampered by increased instability of many constructions. In particular, the overexpression of proteins may contribute to this effect. Therefore, some attempts have been reported to construct regulated expression systems for B. subtilis. In one attempt, a phage phi 105 promoter with a mutant temperature sensitive repressor allele was used (Osburne et al. 1985). Another construction involved providing the repressor gene on a plasmid with a temperature sensitive ori (Duvall et al. 1983). In both cases expression was inducible about 30-fold by raising the cultivation temperature. A different strategy makes use of well-characterized regulatory elements from Escherichia coli. A phage SPO-1 derived promoter and the lac operator were fused and the Lac repressor was provided by a fusion of the lacI gene to a B. subtilis promoter. Fusion of this hybrid promoter to the interferon gene yielded expression of interferon, which was inducible by isopropylthiogelactoside (Yansura and Henner 1984). Recently, the use of lambda Pc and PR promoters with the temperature sensitive cI allele to regulate transcription in B. subtilis has been reported (Breitling et al. unpublished data). We report here the use of the Tnl0-encoded tet operator (Wissmann et al. 1988) and tetR gene (Postle et al. 1984) to control transcription in B. subtilis. These elements were used in conjunction with the strong promoter of the B. subtilis-derived xyl operon (Ggrtner et al. 1988; Kreuzer et al. 1989). This resulted in a wellregulated, efficient expression of glucose dehydrogenase from B. megaterium (Heilmann et al. 1988) and human single chain urokinase-like plasminogen activator (scupa) (Holmes et al. 1985). Materials and methods General methods. Restrictionendonucleaseswere purchased from Boehringer, Mannheim, FRG; BRL, Eggenstein, FRG; or Biolabs, Schwalbach, FRG. T4 DNA ligase and E. coli DNA poly-

658 Table 1. Bacterial strains and plasmids Strains

Genetic markers

Reference or source

Escherichia coli RRI

hsdS20 (rff, mE), ara, proA, lacY, gaIK, rpsL20 xyl, mtl, supE44 trpC2, metBlO, lys-3

Bolivar et al. (1977)

Km r Krnr Km r K1TIr Km r, tetR (Tnl0) Ap ~, tetR (Tnl0) Km r Km r, tetR (Tnl0) Kn]r Km r Km r Km r, uro Km r, gdhA Ap r, Tc r, 9dhA

G/~rtner et al. (1988) This study This study This study This study Oehmichen et al. (1984) This study This study This study This study This study This study This study Meinhardt et al. (1989)

Bacillus subtilis BR151

Gordon et al. (1983)

Plasmids pWH331 pWH341 pWH332 pWH342 pWH333 pWH305 pWH343 pWH346 pWH352 pWH353 pWH354 pWH359 pWH353gdh pSA677

Km, kanamycin, Ap, ampicillin; Tc, tetracycline

merase (Klenow fragment) were from Boehringer. All reactions and polyacrylamide and agarose gel electrophoresis were done as described previously (Hillen et al. 1982). Sodium dodecyl sulphate-polyacrylamid gel electrophoresis was done according to Laemmli (1970). Western blotting was done exactly as described previously (Schmucker et al. 1989). Synthesis and cloning of oligonucleotides was as described by Wissmann et al. (1988). Nucleotide sequencing was done according to Sanger et al. (1977).

Bacterial strains and plasmids. The bacterial strains and plasmids used and constructed in this study are shown in Table 1. Culture and growth conditions. Bacilli were grown in Luria-Bertani (LB) medium (10 g peptone/l, 5 g yeast extract/l, 10 g NaCI/1, pH 7.4; plates were prepared with 1% agar. The final concentration of kanamycin (Km) was 25 mg/1. Transformation of B. subtilis was done according to Puyet et al. (1987). Assays of enzymatic activities. For determination of glucose dehydrogenase (Gdh) activity the respective strains were grown in LB supplemented with 25 mg/l Krn. Induction of Gdh expression was done at an optical density at 650 nm of 1.0 by adding tetracycline (Tc) to a final concentration of 0.5 Ixg/ml. At the times indicated in the respective table 1 ml culture medium was centrifuged to harvest the cells. The supernatants were washed in 1 ml of 10 mM potassium phosphate, pH 6.5, centrifuged and resuspended in 500 Ixl of the same buffer. The cells were sonicated six times for 15 s with waiting periods of 45 s on ice using a Labsonic 1510 from Braun, Melsungen, FRG, equipped with a 5-mm horn at 50 W. The resulting suspension was centrifuged for 10 min at 13 000 rpm and the supernatant used to determine Gdh activity. The assay was exactly done as described by Meinhardt et al. (1989). The urokinase activities were measured exactly as described by Surek et al. (1990), Holmes et al. (1985), and Plough and Kjeldgaard (1957). Induction of expression was done by adding Tc to a final concentration of 0.5 l~g/ml. Chloramphenicol acetyltransferase (CAT) activities were determined after induction with Tc at 0.5 I~g/ml exactly as described by Rodgriguez and Tait (1983).

Results and discussion T h e r e g u l a t e d e x p r e s s i o n system m a k e s use o f the T n l 0 - e n c o d e d tet regulatory elements. T h e i r activities i n B. subtilis were d e t e r m i n e d a n d i m p r o v e d b y fusing t h e m to the cat i n d i c a t o r gene a n d a s s a y i n g the exp r e s s e d C A T activity. T h e v a r i o u s c o n s t r u c t i o n s for tet r e g u l a t e d C A T exp r e s s i o n are d i s p l a y e d i n Fig. 1. T h e y were all c l o n e d p*

xyl/tet

..~

~

Z}---.-Z3--' p*

.-~

~

~ C . - - - - - . - - ~ p~* --

,

~

~

pWH353 pA ~ --

.~/~..._~

~1~'--~_-.---'~

pWH354

xyl/te t

PR --

PA --

,

pWH346

'

pWH333

PA P~

.~_-~..."~ ,

P.,_~ P..~* ~

,

pWH342 pWH332

Fig. 1. Schematic presentation of the regulatory elements used to construct expression vectors. The drawings on the left side indicate the tet operators (open boxes), the tet promoters (PA and PR) with their direction of transcription (thin arrows), the xyl/tet fusion promoters (direction of transcription indicated by heavy arrows), and the tetR genes (black boxes). The designations of the respective plasmids are given on the right side of the figure. The orientation of the insertions in these plasmids is such that the cat indicator gene is located on the right of the drawings. P~ and ~ denote the improved mutant tet promoters shown in Fig. 2 and P* denotes the further improved PR promoter

659

into the shuttle plasmids pWH331 described by G~irtner et al. (1988), or pWH341, which is essentially the same construction except that it contains the polylinker in reverse orientation. In the first step the activities of the TnlO encoded tetPA and Pr promoters in B. subtilis BR151 were determined. For that purpose at 138-bp AluI-TaqI DNA from TnlO with the ends converted to B a m H I sites (Meier et al. 1988) was inserted in both orientations in pWH331. This resulted in pWH332 containing a tetPA-cat transcriptional fusion and in pWH342 containing a tetPr-cat translational fusion. The tetR gene was then introduced into pWH332 as a XbaI-HpaI fragment from pWH305 (Oehmichen et al. 1984) between the XbaI and SmaI sites. Technically this was realized by a three fragment ligation of the X b a I - B a m H I tet regulatory sequence, the XbaI-HpaI tetR DNA and the S m a I - B a m H I vector DNA. This resulted in pWH333 (see Fig. 1). The CAT activities of strains transformed with these plasmids are presented in Table 2. In the absence of Tc both tet promoter-cat fusions show only slightly increased CAT activities over the promoterless control indicating very low activities of both promoters in B. subtilis. The introduction of the tetR gene in pWH333 does not reduce the CAT activity of pWH332 confirming the inactivity of the tet Pr promoter. Thus, no induction is observed in the presence of 0.5 gg/ml Tc. As with several other E. coli-derived promoters, the TnlO encoded tet promoters are inactive in B. subtilis and are not useful to construct expression systems in their wild-type sequence. In the next attempt the sequence of the tet regulatory region was changed with the goal of obtaining active promoters and increased tetR expression in B. subtilis. While the consensus sequences of vegetative promoters in B. subtilis and E. coli are the same, the B. subTable 2. Chloramphenicol acetyltransferase (CAT) activity expressed from tet-cat fusions B. subtilis strain -

BR151 BR151, pWH331 BR151 (pWH342) BR151 (pWH333) a

CAT activity a + Tc

Tc

0 58 210 105

--

-190 80

Given in units described by Rodriguez and Tait (1983)

T

AATTTT

* * *G

TT A C C PA

-35 C T AGACA T CA T T AA t G c c t C C

tilis promoters often contain a poly A block upstream of their - 3 5 region (Moran et al. 1982). Furthermore, efficient expression of a gene in B. subtilis requires a more pronounced Shine-Dalgarno sequence (McLaughlin et al. 1981; Shine and Dalgarno 1974). Therefore, the tet regulatory sequence was optimized with the following goals: the essential tet operator sequences should be maintained (Wissmann et al. 1988) to assure regulation, the homology of the promoters to the consensus sequences should be increased, a poly A block should be placed upstream of the - 3 5 region of tet PA to create a strong promoter, and a better ShineDalgarno sequence for tetR should be introduced to obtain expression of this regulatory protein. The resulting sequence along with indications of the regulatory elements are displayed in Fig. 2. It was synthesized and introduced in pWH333 resulting in pWH346. The nucleotide sequence was verified in the final construct (Sanger et al. 1977) and pWH346 was transformed to B. subtilis BR151. The CAT expression of that strain was monitored after induction with 0.5 gg/ ml Tc. The result is shown in Table 3. Before induction about 700 units (U) of CAT were expressed. This is about a tenfold increase over the promoterless background (see Table 1). Thus, the basal expression under non-induced conditions is rather high, indicating that this system is not efficiently shut down. After induction for 2 h the maximal expression of 5700 U was obtained. This corresponds to a roughly eightfold induction under these conditions. When analysed in E. coli (not shown) this construction was about 340-fold inducible to a maximal expression of 21000 U CAT. This result indicates that the construction can be efficiently regulated in E. coli. In B. subtilis, however, neither the regulation nor the maximal expression under inducing conditions are sufficient for an expression vector. On the other hand, the tet regulatory sequence cannot be further improved owing to constraints resulting from the overlapping operators and promoters. An improved system should have increased Tet repressor expression and increased promoter strength. This requires a dissection of the regulatory elements. Among the strongest promoters kwnon in B. subtilis is the xyl promoter which directs the expression of enzymes involved in xylose utilization (G~irtner et al. 1988). This was chosen for the next construction. To obtain the most efficient regulation, the tet operator sequence was placed between the - 3 5 and - 1 0 boxes.

-10

A A A A T A A A T T T G A C / : ~ I[~1#.,llI[~L,I I i I[¢f:1/r:Tcf:T¢~j T A T A A T T C A A G I l ~ #:t I[~:tt~lcf:l #:tcf:~cf:~ A A G T T T AT T T A A A C T GI =r~:~[¢f:! Ir:~ct iE¥:~q IEI I~l l~:~, T A T T A A G T T C r.,ItIcler:! f..~[t'~lI[~:tq ir.,i I[q i [ ~ IT T T C C TAG

TG T AGT AA T / A CGGA GG T

S.D. tetR

01 -10

02 PR2

Fig. 2. Nucleotide sequence of the mutant tet regulatory sequence adapted for expression in Bacillus subtilis. The promoters PA and Prz are indicated with their consensus sequences, The tet operators O~ and 02 are indicated in white letters on black background. The tet regulatory elements are described in Bertrand et al. (1983)

-35

and Hillen et al. (1983, 1984). The Sine-Dalgarno sequence of tetR is also indicated. Changed nucleotides are printed enlaroed and the respective wild-type nucleotides are given above the sequence. Stars denote inserted nucleotides

660 Table 3. CAT activities expressed as a function of induction time in B. subtilis BR151 (pWH346) Induction time (h)

0.0

0.5

1.0

2.0

3.0

CAT activity a

690

2600

3650

5700

5000

a Given in units described by Rodriguez and Tait (1983)

The resulting nucleotide sequence is displayed in Fig. 3. Furthermore, the tetPR promoter directing expression of tetR was improved by changing the - 3 5 region to the consensus sequence and providing an upstream poly A block. In addition, the Shine-Dalgarno sequence was further improved. These modifications are also indicated in Fig. 3. The oligonucleotide was prepared by chemical synthesis. For the construction of pWH353 (see Fig. 1) the X b a I - B a m H I synthetic oligonucleotide (see Fig. 3) with the tetPR promoter was cloned in pWH346 replacing the tet regulatory region. From this construction the promoter-tetR fusion was isolated as a ClaI-SmaI fragment and inserted into the filled in XhoI and ClaI sites of pWH341. The synthetic xyl-tet promoter-operator fusion was cloned into the XhoI and H i n d l I I sites of the resulting plasmid. In this construction, named pWH353, the xyl-tet element directs expression of the cat gene as a transcriptional fusion. It was transformed to B. subtilis BR151 and the expression of CAT was determined in the presence of either 0.2 or 0.4 Ixg/ml Tc. In addition, a second tet operator sequence was inserted as a 19-bp synthetic oligonucleotide into the filled in H i n d l I I site of pWH353 to yield pWH354. In this construct the expression of CAT should be regulated by two tet operators rather than just one. It was also transformed to B. subtilis BR151 and the CAT expression was determined in the presence of the inducer tetracycline. The results for both plasmids are displayed in Table 4. pWH353 showed a basal expression of about 110 U, which is roughly twice the amount obtained with the promoterless cat gene (see Table 2). After addition of either 0.2 or 0.4 ~tg/ml Tc the expression was induced. After 3 h the expression of CAT in the presence of 0.2 Ixg/ml Tc reached 7000 U while in the presence of 0.4 ~tg/ml Tc 11600 U were present at that time. This corresponds to induction efficiencies of

60-fold and 100-fold, respectively. Thus, the expression characteristic of this promoter system shows only little basal expression and good inducibility. The strength of the fully induced promoter corresponds to a medium level when compared to some strong B. subtilis promoters tested so far (G~irtner et al. 1988; Dhaese et al. 1984). pHW354, on the other hand, shows no detectable basal expression of CAT in the absence of inducer. It is concluded that the introduction of the second tet operator shuts the expression completely down. Induction after 4 h reached 180 U in the presence of 0.2 ~tg/ml Tc and 570 U in the presence of 0.4 ~tg/ml Tc. Induction factors cannot be determined due to the lack of basal expression. It is apparent that this very tight regulation is achieved at the expense of efficient induction. Therefore, this system may be useful for the cloning of genes encoding proteins the expression of which is lethal for the host. Taken together, the improvements of the tetR expression signals have been successful because the inducibility is over tenfold increased compared to the tet system in pWH346 (see Table 3). The resulting induced promoter strength in pWH353 is sufficient for an expression system. It is higher using 0.4 lxg/ml Tc as compared to 0.2 ~tg/ml.Tc for induction. We therefore determined the optimal inducer concentration (data not shown). Up to 0.5 ~tg/ml are sub-lethal amounts of Tc for this strain. However, in the presence of 0.5 ~tg/ml Tc the generation time of the strain doubled and induction was not increased

Table 4. CAT activities as a function of induction time expressed from the xyl/tet-cat fusions in B. subtilis BR151 at two inducer concentrations Induction time (h)

0.0 0.5 1.0 2.0 3.0

CAT activity a pWH353 pWH354 0.2 ~tg/ml Tc

pWH353 pWH354 0.4 ~tg/ml Tc

110 3400 4500 5900 7000

110 5600 6800 9600 11600

0 80 90 120 180

0 390 420 520 570

a Given in units as described by Rodriguez and Tait (1983)

C T A G A C A T C A T T A A T T C C T C C T T T TT G T T G A C E ~ K ~ ' : ~ [ ~ : ~ K ~ : t ~ : ~ : ~ | T A T TTGTCAAACTAGT T T T T TAT TTG TG T AGT AA T T AAGGAGGAAAAACAAC T G/(~:~¢~:~1# : ~ ~':¥:~ ~:~1I[~ I [ ~ T AAACAGT TT GA TCAAAAAA T AAACC T AG S.D. te._~tR

'

01

PR*

-10

-~

x¥1/tet

-35

-~o

TCGAGT T C A T G A A A A A C T A A A A A A A A T A T T G A C ~ ~'~I[~ I~-~1I[~:~1I~I[~-~1I ~ iA T A A T T A A A A T A C A A G T A C T T T T T G A T T T T T T T T A T A A C T ~ I ~ : ~ : ~ I~-~ ~ : ¥ ~ lEvI~ ~ T A T ~ AA T T T T A T TCGA

O~ Fig. 3. Nucleotide sequences of the improved tetR promoter (upper) and the xyl/tet fusion promoter (lower). The promoter consensus boxes and tet operator sequences are denoted as in Fig. 2. The extended Shine-Dalgarno sequence of the tetR gene is indicated

661 Glucose dehydrogenase (Gdh) activitya in B. subtilis BR152 transformed with pWH353 and pWH353 gdh

Induction time (h)

0 1 3 5 22

pWH353

_xy!/t~!

X~I

Table 5.

I~

~ .~,,~I

t ~ K

pWH353 gdh

Supernatant

Cells

Supernatant

Cells

0 0 0 0 0

0 0 0 0 0

0.0 5.1 2.2 31.1 4.4

1.2 10.1 18.8 21.9 15.5

i ::::: I yo. ~ Y

~1

22

~,~l C~l I "fill-in"

a Given in units according to Meinhardt et al. (1989)

over 0.4 ~tg/ml Tc. Therefore, higher concentrations of inducer were not investigated. The regulated expression system has been further characterized by the expression analysis of two heterologous genes. First, the glucose dehydrogenase gene (gdhA) from B. megaterium (Heilmann et al. 1988) was used as a non-problematic example. It is considered to be non-problematic because it has been expressed with high yield in E. coli (Heilmann et al. 1988) and in B. megaterium (Meinhardt et al. 1989) before. The second gene (puk) encodes for human scupa and is considered a problematic gene because expression in E. coli occurs only at moderate levels (Holmes et al. 1985; Surek et al. 1990). To obtain regulated expression of G d h from B. me9aterium the ClaI-HpaI D N A from pSA677 (Meinhardt et al. 1989) was isolated, filled in with Klenow polymerase and inserted into the SmaI site of pWH353. The resulting plasmid, pWH353 Gdh, contains a xyl-tet-gdhA transcriptional fusion. After transformation into B. subtilis BR151 the glucose dehydrogenase activity was determined under inducing conditions in the cells and also in the supernatant. The results are presented in Table 5. They indicate that pWH353 does not lead to G d h expression, pWH353 G d h shows a low level of expression without induction which was increased about 45fold after 5 h induction. After 22 h induction the activity of G d h somewhat decreased again. At the maximal level o f expression at 5 h after induction, about 60% of the activity was found in the supernatant while 40% was associated with the cell fraction. The total yield was about fivefold higher than that obtained with B. megaterium (Meinhardt et al. 1989). The reason for the presence of activity in the supernatant is not clear because the gdhA genes does not contain a signal sequence for export (Heilmann et al. 1988) and B. subtilis BR151 should be spo-. It must thus be assumed that non-specific lysis of host cells occurs under conditions of G d h overexpression. However, the results confirm that the vector pWH353 is very suitable for regulated, high level expression. The construction for scupa expression was obtained by isolating the XbaI-ClaI D N A from pWH1330 (Surek et al. 1990), filling in the producing ends with Klenow D N A polymerase and inserting it into the SmaI site of pWH353 to yield pWH359. A scheme of this construc-

Fig. 4. Construction scheme of pWH359. The relevant genes and restriction sites of the plasmids are indicated; Ato designates the respective terminator sequence from bacteriophage ~.. The sizes of the plasmids are given in bp

Table 6. Fibrinolytic activity expressed as a function of induction time Induction time (h)

0

1

2

B. subtilis strain

Proteolytic zone (mm)~

BR151 BR151 (pWH359)

0 0

-10

-10

3

0 11

" High molecular weight urokinase used as a standard gave the following proteolytic zones: 3.5 units (U)= 12.5 mm; 0.35 U = 5.5 ram; and 0.035 U=2.5 mm

tion is shown in Fig. 4. The expression of urokinase activity was determined by a plate fibrinolysis assay (Surek et al. 1990) and is given as the diameter of the proteolytic zone in Table 6. The maximal expression of urokinase activity was already found after 1 h of induction and corresponds to about 2.8 U of high molecular weight urokinase (HUK) used as a standard. This corresponds to about 370 U urokinase (Plough and Kjeldgaard 1957). The urokinase expression was verified by Western blotting as described (Surek et al. 1990). The result shown in Fig. 5 indicates that prourokinase was indeed present in the cells. Furthermore, even after 250 min of induction only very little degradation was found. This is a considerable improvement over expression in E. coli where heat shock induction at 37°C resulted in high proteolytic instability of the product (Surek et al. 1990). The amount expressed in B. subtilis

662

Fig. 5. Western blot of total proteins prepared from B. subtilis expressing urokinase. The left three lanes contain proteins of B. subtilis BR151 without a plasmid. The numbers on top of the gel give the time after induction. The next six lanes show the expression of single chain urokinase-like plasminogen activator. The rightmost lane contains 1 p~g of purified single-chain urokinase-like plasminogen activator (scupa). The analysis was done on a 10% polyacrylamide gel in the presence of sodium dodecyl sulphate. The open arrow on the left designates the position of scupa

u s i n g the e x p r e s s i o n vector p r e s e n t e d here was r o u g h l y the same as that expressed with o p t i m i z e d vectors i n E. coli. T h u s , for the e x p r e s s i o n of p r o u r o k i n a s e this vector-host system is even s u p e r i o r to E. coli. Acknowledgements. We thank M. Will and G. Schneider for excel-

lent technical assistance and Mrs. K. Garke for typing the manuscript. This work was supported by a grant from the Bundesministerium for Forschung und Technologie and by the Fonds der chemischen Industrie.

References

Bertrand KP, Postle K, Wray LV Jr, Reznikoff WS (1983) Overlapping divergent promoters control expression of TnlO tetracycline resistance. Gene 23:149-156 Bolivar F, Rodriguez RL, Greener PJ, Betlach MC, Heyneker HL, Boyer HW, Crosa JH, Falkow S (1977) Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2:95-133 Dhaese P, Hussey C, Montagu M van (1984) Thermoinducible gene expression in Bacillus subtilis using transcriptional regulatory elements from temperate phage d~105. Gene 32:181-194 Duvall EJ, Williams DM, Lovett PS, Rudolph C, Vasantha N, Guyer M (1983)Chloramphenicol-inducible gene expression in Bacillus subtilis. Gene 24:171-177 G~irtner D, Geissend6rfer M, Hillen W (1988) Expression of the Bacillus subtilis xyl operon is repressed at the level of tran-

scription and is induced by xylose. J Bacteriol 170:31023109 Gordon RE, Haynes WC, Pang CH-N (1983) The genus Bacillus. US Dept of Agriculture, Washington, DC Heilmann HJ, Magert HJ, Gassen HG (1988) Identification and isolation of glucose dehydrogenase genes of Bacillus megaterium M 1286 and their expression in Escherichia coli. Eur J Biochem 174:485-590 Hillen W, Klock G, Kaffenberger I, Wray LV Jr, Reznikoff WS (1982) Purification of the Tet repressor and tet operator from the transposon TnlO and characterization of their interaction. J Biol Chem 257:6605-6613 Hillen W, Gatz C, Altschmied L, Schollmeier K, Meier I (1983) Control of expression of the Tnl0-encoded tetracycline resistance genes. Equilibrium and kinetic investigations of the regulatory reactions. J Mol Biol 169:707-721 Hillen W, Schollmeier K, Gatz C (1984) Control of expression of the Tnl0-encoded tetracycline resistance operon. II. Interaction of RNA polymerase and Tet repressor with the tet operon regulatory region. J Mol Biol 172:185-201 Holmes WE, Pennica D, Blaber M, Rey MW, Guenzler WA, Steffens GJ, Heynecker HL (1985) Cloning and expression of the gene for pro-urokinase in Escherichia coli. Biotechnology 3:923-929 Kreuzer P, Gartner D, Allmansberger R, Hillen W (1989) Identification and sequence analysis of the Bacillus subtilis W23 encoded xylR gene and xyl operator. J Bacteriol 171:3840-3845 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 McLaughlin JR, Murray CL, Rabinowitz JC (1981) Unique features in the ribosome binding site sequence of the Gram-positive Staphylococcus aureus fl-lactamase gene. J Biol Chem 256:11283-11291 Meier I, Wray LV, Hillen W (1988) Differential regulation of the Tnl0-encoded tetracycline resistance genes tetA and tetR by the tandem tet operators O~ and Oa. EMBO J 7:567-572 Meinhardt F, Stahl U, Ebeling W (1989) Highly efficient expression of homologous and heterologous genes in Bacillus megaterium. Appl Microbiol Biotechnol 30:343-350 Moran CP, Lang N, Legrice SFJ, Lee G, Stephens M, Sonenshein AL, Pero J, Losick R (1982) Nucleotide sequences that signal the initiation of transcription and translation in Bacillus subtilis. Mol Gen Genet 186:339-346 Oehmichen R, Klock G, Altschmied L, Hillen W (1984) Construction of an E. coli strain overproducing the TnlO encoded Tet repressor and its use for large scale purification. EMBO J 3 : 539-543 Osburne MS, Craig RJ, Rothstein DM (1985) Thermoinducible transcription system for Bacillus subtilis that utilizes control elements from temperate phage ~b 105. J Bacteriol 163:11011108 Plough J, Kjeldgaard NO (1957) Urokinase: an activator of plasminogen from human urine. I. Isolation and properties. Biochim Biophys Acta 24:278-282 Postle K, Nguyen TT, Bertrand KP (1984) Nucleotide sequence of the repressor gene of the TnlO encoded tetracycline resistance determinant. Nucleic Acids Res 12:4849-4963 Puyet A, Sandoval H, Lopez P, Aguilar A, Martin JF, Espinosa M (1987) A simple medium for rapid regeneration of Bacillus subtilis protoplasts transformed with plasmid DNA. FEMS Microbiol Lett 40:1-5 Rodriguez RL, Tait RC (1983) Recombinant DNA techniques: an introduction. Addison-Wesley Publishing Co., Reading, Mass Sanger F, Nieklen S, Coulson AR (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sei USA 74:5463-5467 Schmucker R, Galland U, Will M, Hillen W (1989) High level expression of Acinetobacter calcoaceticus mutarotase in Escherichia coli is achieved by improving the translational control sequence and removal of the signal sequence. Appl Microbiol Biotechnol 30:509-514

663 Shine J, Dalgarno L (1974) The 3'-terminal sequence of Escherichia coli 16S ribosomal RNA: complementarity to nonsense triplets and ribosome binding sites. Proc Natl Acad Sci USA 71:1342-1346 Surek B, Wilhelm M, Hillen W (1990) Optimizing the promoter and ribosome binding sequence for expression of human single chain urokinase-like plasminogen activator in Escherichia coli and stabilization of the product by avoiding heat shock

response. Appl Microbiol Biotechnol, in press Wissmann A, Meier I, Hillen W (1988) Saturation mutagenesis of the Tnl0-encoded tet operator O1: identification of base-pairs involved in Tet repressor recognition. J Mol Biol 101:397406 Yansura DG, Henner DJ (1984) Use of the Escherichia coli lac repressor and operator to control gene expression in Bacillus subtilis. Proc Natl Acad Sci USA 81:439-443