- Email: [email protected]
A refined vector system for the in vitro construction of single-copy transcriptional or translational fusions to lacZ
Gene, 169 (1996) 65-68 ©1996 Elsevier Science B.V. All rights reserved. 0378-1119/96/$15.00
65
GENE 09490
A refined vector system for the in vitro construction of single-copy transcriptional or translational fusions to lacZ ([3-galactosidase; gene fusion vector; phage lambda; recombinant DNA)
R e b e c c a St. P i e r r e a n d T h o m a s L i n n Department of Microbiology and Immunology, Faculty of Medicine, Universityof Western Ontario, London, Ontario N6A 5C1, Canada Received by M. Bagdasarian: 23 August 1995; Revised/Accepted: 26 September 1995; Received at publishers: 30 October 1995
SUMMARY
New single-copy vectors based on ~, phage have been developed for creating either transcriptional (operon) or translational (gene) fusions to the lacZ gene. The improvements of these vectors over the previous )~TL61 vector include: (i) incorporation of a tetracycline-resistance-encoding gene (TcR) to permit direct selection of lysogens, (ii) lowbackground [3-galactosidase activity, (iii) the ability to accept DNA inserts up to 8 kb in size, and (iv) an expanded multiple cloning site (MCS). The new transcriptional fusion vector retains the RNase III processing site downstream from the MCS which ensures independent translation of lacZ. The set of three translational fusion vectors allow for convenient subcloning in any of the three translational reading frames.
Gene fusions have become indispensable tools in the analysis of gene expression (reviewed in Silhavy and Beckwith, 1985; Slauch and Silhavy, 1991). A wide assortment of systems based on either multicopy (plasmid) or single-copy (phage) vectors have been developed' to construct lacZ (a common reporter gene) fusions for use in prokaryotes. Although the construction of fusions with plasmid vectors is often technically easier, the use of fusions on multicopy plasmids can be problematic for a variety of reasons. These include: (i) the high copy number of the promoter and regulatory site(s) can potentially titrate out trans-acting factors, (ii) the copy number of many plasmid replicons is size dependent (Scott and Round; 1980; Adams and Hatfield, 1984) and (iii) some DNA regions present at high copy number or the high levels of the gene products produced can be
detrimental. These problems can lead to abnormal expression of the fusion and/or the selection of unanticipated mutations. These problems can be mitigated by using vectors based on phage )~, which allow a single copy of the fusion to be stably integrated into the host chromosome. We have previously developed a ~, vector (~,TL61) for the in vitro construction of transcriptional fusions to lacZ (Linn and St. Pierre, 1990). An important and unique feature of this vector is the insertion of an RNase III processing site between the MCS and the lacZ gene. Cleavage of the hybrid mRNA at this site produces a constant 5' end for the lacZ mRNA which helps ensure independent translation'of lacZ irrespective of what sequences are fused upstream (Linn and St. Pierre, 1990). We have continued to improve and refine this transcriptional fusion vector (~,TXF97) and have also constructed analogous translational fusion vectors 0~TLF97-1, -2, -3).
Correspondence to: Dr. T. Linn, Department of Microbiology and Immunology, Faculty of Medicine, University of Western Ontario, London, Ontario N6A 5C1, Canada. Tel. (1-519) 661-3426; Fax (1-519) 661-3499; e-mail: [email protected]
Abbreviations: [3Gal, [3-galactosidase; bp, base pairls); kb, kilobase(s) or 1000 bp; MCS, multiple cloning site(s); nt, nucleotide(s); R, resistance/ resistant; re, recombinant; Tc, tetracycline; u, unit(s); XGal, 5-bromo4-chloro-3-indolyl-13-D-galactopyranoside.
INTRODUCTION
SSDI 0378-1119(95)00787-3
66 lac A
..
T2
/,i
J
Y
Z
EXPERIMENTAL AND DISCUSSION
att i 21 n i n 5 , ~___ ,
I ,XFy7i lacY
•
lacZ
•£3
roll 1 ,,x~p,~.~ T1
Tc R
~
p
I~TLF97-1,2,3] T2
•C)
lacy
lacZ'
MCS
~
T1 ~
Tc R
j
Fig. I. Structure of the transcriptional and translational lacZ fusion vectors. The top diagram shows the overall structure of the phage genome and position of the lac region (not drawn to scale). The relative positions of the ~, A and J genes in the right arm of the vector and the phage attachment site (att), immunity 21 substitution 021) and nin5 deletion in the left arm of the vector are indicated. The expanded sections shown below detail the position of the Tc rt gene, the rrnBT1 (T1) terminator upstream from the MCS, the rrnBT2 (T2) terminator following lac Y, and the RNaselII processing site (RIII) in ~,TXF97. The arrows indicate the orientation of these segments. The orientation of the MCS (see Fig. 2) is indicated by the position of the SnaBI (Sn) and HindlIl (H) sites. Almost all of the sequence of XTXF97 can be deduced from published sequences and the steps in its construction. As s u m m a rized below only a few sites have one or more ambiguous nt. Starting from the left end of the genome, nt 1 to 19 399 are from XL47.1 (Loenen and Brammar, 1980), which should be identical to wild-type X (Daniels et al., 1983), except the BamHI site at nt 5505 has been removed by an uncharacterized mutation. Next, nt 19 400 to 19 492 of the vector correspond to (symbols=below) sequence CTAG followed by nt 6845 to 6760 of the rrnB operon (this segment contains rrnBT2) (Brosius et al., 1981) followed by ACG. Next nt 19493 to 19 5 0 6 = n t 1373 to 1360 of pBR322 (Watson, 1988). Next, nt 19 507 to 23 903 of the v e c t o r = t h e lac region, with lacy (nt 19 507 to 20760) (Buchel et al., 1980) and lacZ (nt 20812 to 23 883) (Kalnins et al., 1983), except the EcoRI site normally present at the 3' end of lacZ has been removed by an uncharacterized mutation. Then nt 23904 to 2 4 0 4 8 = n t 238 to 94 of pTL61T (Linn and St. Pierre, 1990), this includes the RNase III processing site and extends up to the MCS. Next, nt 24049 to 24 1 0 5 = n t 452 to 396 of pUC19, which is the pUC19 MCS. Then nt 24106 to 24 132 of the vector = the newly added section of the MCS which has the sequence 5 ' - G G C G C C T C G A G A T C A G A G A C C T A C G T A . Next nt 24133 to 24 329 of the vector = 5 ' - C C G A T C C C C A A T T C C followed by nt 6778 to 6610 of the rrnB operon (this segment contains rrnBT1) (Brosius et al., 1981) followed by G G A A T T G G G A A T T . Then nt 24330 to 24 522 = AA plus nt 4361 to 4171 of pBR322. Next nt 24 523 to 25 937 = the Tc R gene of pACYC184, nt 1453 to 2867 (Rose, 1988), except 1873 was changed from C to T by oligonucleotide mutagenesis to remove the BamHI site. This change does not alter the amino-acid sequence of the Tc R protein. The ~, sequences then resume with nt 25 938 to 33 776 of the v e c t o r = n t 26 532 to 34 366 of X, except nt 27 974 was changed from A to G by oligonucleotide mutagenesis to remove the BamHI site, and T C G A was added after nt 33 502 to remove the XhoI site. The change at nt 27 974 is within the int gene, but the amino-acid sequence of the protein is not altered. Next nt 33 777 to 35 625 of the vector = the approx. 1849-bp i 21 substitution. The sequence of this region has been determined except for two approx. 4-bp ambiguities, in the ell and cro genes. (Franklin, 1984; N.C. Franklin, personal communication). For purposes of this vector sequence they have both been represented as 4 N residues. This is followed by ut 35 626 to 37 509 of the v e c t o r = n t 38618 to 40501 of ~,. Finally, nt 37510 to 42704 of the vector = n t 43 308 to 48 502 of X. The ~, D N A missing between these two segments is the result of the nin5 deletion. Since the ~ sequences
(a) Features of kTXF97, the new transcriptional fusion vector (I) In order to measure 13Gal production from lacZ fusions carried on X vectors such as ~,TL61, the recombinant (re-) phage is used to form monolysogens in a Alac E. coli strain (Linn and St. Pierre, 1990). We have inserted the Tc R gene from pACYC184 into the right arm of the vector (Fig. 1) to permit direct selection of lysogens on plates containing 2 gg Tc/ml. This makes the isolation of lysogens simpler than cell transformation with drugresistance plasmids. (2) A monolysogen of the XTL61 vector has a background 13Gal activity of 71 u which can complicate the analysis of weak promoters. This level produces light pink colonies on MacConkey plates and light blue colonies on XGal plates. The majority of the background expression was found to result from a low level of transcription initiating somewhere to the right of the MCS that reads through into lacZ. The background level was reduced four-fold to 18 u by insertion of a 169-bp fragment carrying the efficient rrnBT1 transcriptional terminator just upstream from the MCS (Fig. 1). Since this level produces colorless colonies on MacConkey plates and faint blue colonies on XGal plates, re-fusions with even weak promoters can now be distinguished from the reconstituted vector. (3) Although XTL61 could incorporate DNA fragments up to 6 kb, this maximum amount was reduced by the insertion of rrnBT1 and the TcR gene upstream from the MCS. To compensate, we deleted non-essential regions of XTL61 including lacA with the downstream tandem copies of the complete rrnBt terminator region (Linn and St. Pierre, 1990) and 425 bp in the right arm of the vector (between nt 26 104 (EcoRI site) and 26 529 (PmlI site) of the X genome). A 94-bp fragment containing a single copy of rrnBT2 was inserted downstream from lacY (Fig. 1) to prevent transcription from the fusion continuing into the X left arm. As a result of these changes the size of XTXF97 is 42704 bp (Fig. 1), allowing it to incorporate DNA fragments up to 8 kb. (4) Finally, the MCS was altered from that of ~,TL61
in the fight arm of this vector originate from LNM616 (Mileham et al., 1980), the EcoRI sites normally present at nt 31 747, 39 168 and 44972 in the X genome all have been removed by uncharacterized mutations. The translational fusion vectors, LTLF97-1, -2 and -3, have the same sequence as XTXF97 except the MCS is attached directly to the ninth codon of the lacZ gene (see Fig. 2). Details of the construction of these vectors will be supplied on request. The sequences of XTXF97, XTLF97-1, ~,TLF97-2 and XTLF97-3 were submitted to G e n B a n k under accession Nos. U37692, U39284, U39285 and U39286, respectively.
67 TXF97 NarI SnaBI
SacI
XhoI
EcoRI
SmaI
XbaI
KDnI
BamHI
PstI SalI
HindIII EDhI
TACG~T~QGTCTCTGATCTCGAGGCGCCGAATTCGAGCTCGGTACCCGGGGATCCTCTAGAGTCGACCTGCAGGCATGCAAGCT~2~CTAACTAGCG --->
RIII
--->
lacZ
TLF97 - 1 NarI SnaB I
Xho I
TA CGT AGG
TCT CTG ATC
TCG AGG
SacI
SmaI
EcoR I
KDn I
CGC CGA ATT
CGA GCT CGG TAC
Xhol
PstI
BamH I
HindIII
Sa I I
SDh I
CCG GGG ATC CCT CGA GGT CGA CCT GCA GGC ATG
CAA GCT TGG GGG - - - >
TLF97
lacZ
-2 NarI
SnaBI
TAC GTA GGT
XhoI
CTC TGA TCT
SacI EcoRI
SmaI KDnI
CGA GGC GCC GAA TTC GAG
XbaI
PstI
BamHI
HindIII
SalI
CTC GGT ACC CGG GGA TCC TCT AGA GTC GAC
SDhI
CTG CAG GCA TGC AAG
CTT GGG GGG --->
lacZ
TLF97-3 NarI SnaBI
XhoI
T ACG TAG GTC TCT GAT CTC
Sa cI EcoRI
GAG GCG CCG AAT TCG AGC
SmaI KDnI
TCG GTA
XbaI BamHI
P s tI
HindI I I
SalI
CCC GGG GAT CCT CTA GAG TCG ACC
SDhI
TGC AGG
CAT GCA AGC
TTG GGG GGG --->
lacZ
Fig. 2. The nt sequence of the MCS in the transcriptional and translational lacZ fusion vectors. The restriction sites are overlined. Translational stop codons are underlined. )~TXF97 has stop codons in all three reading frames immediately downstream from the HindlII site. This ensures that any ribosomes continuing from the attached upstream sequences of a fusion will not read through into the RNase III site (RIII) disrupting its structure and thus prevent cleavage of the mRNA. The MCS sequences of the translational fusion vectors, )~TLF97-1, -2, -3, are indicated as triplets that are in the same translational reading frame as the lacZ gene. Following the G residues at the end of the MCS sequences for each of the translational fusion vectors is a junction sequence of three codons before the lacZ gene sequence begins at its ninth codon.
to facilitate the in vitro construction of fusions to lacZ. The sequence from the EcoRI site through the HindlII site is identical to the MCS of pUC19 (Fig. 2). An additional sequence was inserted upstream from the EcoRI site that contains the recognition sites for NarI, XhoI and SnaBI. As a result of these changes and in vitro mutagenesis to remove restriction sites elsewhere in the vector (see legend to Fig. 1) the XhoI, EcoRI, BamHI and XbaI sites of the MCS are unique and can be used to generate both intact right and left arms (Fig. 2,.Table I). Cleavage at the NarI, SmaI, SalI or HindlII sites will produce the complete left arm which includes lacZY, while cleavage at the SnaBI, SacI or KpnI sites will produce the complete right arm of the vector. As summarized in Table I, a wide variety of restriction endonucleases can be used to produce fragments from the DNA region of interest, whose upstream and downstream termini can then be directly ligated to the vector arms to produce the desired re-lacZ fusion. Details of the construction and recovery of re-fusion phage using ~ vector arms has been described previously (Linn and St. Pierre, 1990). (b) Translational fusion vectors, ~.TLF97-1, -2, -3 Starting with ~TXF97 a set of translational fusion vectors was constructed that carry the MCS in each of the three translational reading frames. This was accomplished by replacing the D N A segment of kTXF97 that extends from the MCS through the RNase lII processing site and into the 5' region of lacZ;with the MCS and 5' truncated lacZ segment from three translational fusion
TABLE I Restriction endonucleases that cleave in the MCS to produce complete left and/or right arms of the vectors Enzyme
Vectors arm(s) produced
SnaBI XhoI NarI
Right Right and left Left
EcoRI SacI
Right and left Right
KpnI
Right Left
Sinai
BamHI
XbaI b SalI
Right and left Right and left Left
HindlII
Left
Enzymes that generate compatible termini Any that produce blunt termini Sail AciI, BsaHI, BstBI, ClaI, HinPlI, HpalI, MaelI, Pspl406I, TaqI, (AccI; GT/CGAC) ~ ApoI, MfeI, Tsp509I
Any that produce blunt termini BclI, BgllI, BstYI, Sau3AI Avrll, NheI, SpeI XhoI
a AccI recognises several sequences, only the one listed generates termini compatible to those of the NarI site. u There is no XbaI site in TLF97-1.
plasmids (Jain, 1993). The kTLF97-1, -2 and -3 vectors are identical except for the presence of 5, 6 or 7 G residues, respectively, downstream from the HindlII site (Fig. 2). Also the XbaI site present in LTLF97-2 and -3 is replaced with an XhoI site in kTLF97-1 to avoid an in-frame stop codon present at the XbaI site (Jain, 1993). The background 13Gal level of the vectors is less than 0.1 u, allowing accurate measurement of even weak expres-
68 sion from re-fusions. Because of this extremely low background, lysogens of reconstituted vector are colorless even on plates containing XGal.
(c) Conclusions (1) The ~,TXF97, LTLF97-1, -2, -3 set of vectors with their expanded MCS greatly facilitate the direct in vitro construction of single-copy transcriptional or translational fusions to lacZ and should prove valuable in assessing the relative roles of transcription and translation in regulating the expression of a wide variety of genes. (2) Access to the nt sequence of these phage vectors expedites the confirmation and structure analysis of recombinant fusions by restriction site mapping. (3) The vectors insulate the MCS-IacZY region with flanking transcriptional terminators. (4) The presence of the Tc R gene allows for simple and direct selection of lysogens. ACKNOWLEDGEMENTS
We thank C. Jain for providing pLACZla, pLACZ2 and pLACZ3. This study was supported by the Canadian Medical Research Council. REFERENCES Adams, L.W. and Hatfield, G.W.: Effects of promoter strengths and growth conditions on copy number of transcription-fusion vectors. J. Biol. Chem. 259 (1984) 7399-7403.
Brosius, J., Dull, T.J., Sleater, D.D. and Noller, H.F.: Gene organization and primary structure of a ribosomal RNA operon from Escherichia coll. J. Mol. Biol. 148 (1981) 107-127. Buchel, D.E., Gronnelborn, B. and M011er-Hill, B.: Sequence of the lactose permease gene. Nature 283 (1980) 541-545. Daniels, D.L., Schroeder, J.L., Szybalski, W., Sanger, F., Coulson, A.R., Hong, G.F., Hill, D.F., Peterson, G.B. and Blattner, F.R.: Complete annotated lambda sequence. In: Hendrix, R.W., Roberts, J.W., Stahl, F.W. and Weisberg, R.A. (Eds.), Lambda II. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1983, pp. 519-676. Franklin, N.C.: Conservation of genome form but not sequence in the transcription antitermination determinants of bacteriophage ~., d~21 and P22. J. Mol. Biol. 181 (1984) 75-84. Jain, C.: New improved lacZ fusion vectors. Gene 133 (1993) 99-102. Kalnins, H., Otto, K., Rather, U. and MOiler-Hill, B.: Sequence of the lacZ gene of Escherichia coli. EMBO J. 2 (1983) 593-597. Linn, T. and St. Pierre, R.: Improved vector system for constructing transcriptional fusions that ensures independent translation of lacZ. J. Bacteriol. 172 (1990) 1077-1084. Loenen, W.A.M. and Brammar, W.J.: A bacteriophage vector for cloning large DNA fragments made with several restriction enzymes. Gene 10 (1980) 249-257 Mileham, A.J., Revel, H.R. and Murray, N.E.: Organization and expression of the f r d - D N A ligase region. Mol. Gen. Genet. 179 (1980) 227-239. Rose, R.E.:The nucleotide sequence of pACYC184. Nucleic Acids Res. 16 (1988) 355. Scott, J.R. and Rownd, R.H.: Regulation of plasmid replication. In: Alberts, B., Fox, C.F. and Trusser, F.J. (Ed.), Mechanistic Studies of DNA Replication and Genetic Recombination. Academic Press, New York, NY, 1980, pp. 1-8. Silhavy, T.J. and Beckwith, J.R.: Uses of lac fusions for the study of biological problems. Microbiol. Rev. 49 (1985) 398-418. Slauch, J.M. and Silhavy, T.J.: Genetic fusions as experimental tools. Methods Enzymol. 204 (1991) 213-247. Watson, N: A new revision of the sequence of pBR322. Gene 70 (1988) 399-403.
65
GENE 09490
A refined vector system for the in vitro construction of single-copy transcriptional or translational fusions to lacZ ([3-galactosidase; gene fusion vector; phage lambda; recombinant DNA)
R e b e c c a St. P i e r r e a n d T h o m a s L i n n Department of Microbiology and Immunology, Faculty of Medicine, Universityof Western Ontario, London, Ontario N6A 5C1, Canada Received by M. Bagdasarian: 23 August 1995; Revised/Accepted: 26 September 1995; Received at publishers: 30 October 1995
SUMMARY
New single-copy vectors based on ~, phage have been developed for creating either transcriptional (operon) or translational (gene) fusions to the lacZ gene. The improvements of these vectors over the previous )~TL61 vector include: (i) incorporation of a tetracycline-resistance-encoding gene (TcR) to permit direct selection of lysogens, (ii) lowbackground [3-galactosidase activity, (iii) the ability to accept DNA inserts up to 8 kb in size, and (iv) an expanded multiple cloning site (MCS). The new transcriptional fusion vector retains the RNase III processing site downstream from the MCS which ensures independent translation of lacZ. The set of three translational fusion vectors allow for convenient subcloning in any of the three translational reading frames.
Gene fusions have become indispensable tools in the analysis of gene expression (reviewed in Silhavy and Beckwith, 1985; Slauch and Silhavy, 1991). A wide assortment of systems based on either multicopy (plasmid) or single-copy (phage) vectors have been developed' to construct lacZ (a common reporter gene) fusions for use in prokaryotes. Although the construction of fusions with plasmid vectors is often technically easier, the use of fusions on multicopy plasmids can be problematic for a variety of reasons. These include: (i) the high copy number of the promoter and regulatory site(s) can potentially titrate out trans-acting factors, (ii) the copy number of many plasmid replicons is size dependent (Scott and Round; 1980; Adams and Hatfield, 1984) and (iii) some DNA regions present at high copy number or the high levels of the gene products produced can be
detrimental. These problems can lead to abnormal expression of the fusion and/or the selection of unanticipated mutations. These problems can be mitigated by using vectors based on phage )~, which allow a single copy of the fusion to be stably integrated into the host chromosome. We have previously developed a ~, vector (~,TL61) for the in vitro construction of transcriptional fusions to lacZ (Linn and St. Pierre, 1990). An important and unique feature of this vector is the insertion of an RNase III processing site between the MCS and the lacZ gene. Cleavage of the hybrid mRNA at this site produces a constant 5' end for the lacZ mRNA which helps ensure independent translation'of lacZ irrespective of what sequences are fused upstream (Linn and St. Pierre, 1990). We have continued to improve and refine this transcriptional fusion vector (~,TXF97) and have also constructed analogous translational fusion vectors 0~TLF97-1, -2, -3).
Correspondence to: Dr. T. Linn, Department of Microbiology and Immunology, Faculty of Medicine, University of Western Ontario, London, Ontario N6A 5C1, Canada. Tel. (1-519) 661-3426; Fax (1-519) 661-3499; e-mail: [email protected]
Abbreviations: [3Gal, [3-galactosidase; bp, base pairls); kb, kilobase(s) or 1000 bp; MCS, multiple cloning site(s); nt, nucleotide(s); R, resistance/ resistant; re, recombinant; Tc, tetracycline; u, unit(s); XGal, 5-bromo4-chloro-3-indolyl-13-D-galactopyranoside.
INTRODUCTION
SSDI 0378-1119(95)00787-3
66 lac A
..
T2
/,i
J
Y
Z
EXPERIMENTAL AND DISCUSSION
att i 21 n i n 5 , ~___ ,
I ,XFy7i lacY
•
lacZ
•£3
roll 1 ,,x~p,~.~ T1
Tc R
~
p
I~TLF97-1,2,3] T2
•C)
lacy
lacZ'
MCS
~
T1 ~
Tc R
j
Fig. I. Structure of the transcriptional and translational lacZ fusion vectors. The top diagram shows the overall structure of the phage genome and position of the lac region (not drawn to scale). The relative positions of the ~, A and J genes in the right arm of the vector and the phage attachment site (att), immunity 21 substitution 021) and nin5 deletion in the left arm of the vector are indicated. The expanded sections shown below detail the position of the Tc rt gene, the rrnBT1 (T1) terminator upstream from the MCS, the rrnBT2 (T2) terminator following lac Y, and the RNaselII processing site (RIII) in ~,TXF97. The arrows indicate the orientation of these segments. The orientation of the MCS (see Fig. 2) is indicated by the position of the SnaBI (Sn) and HindlIl (H) sites. Almost all of the sequence of XTXF97 can be deduced from published sequences and the steps in its construction. As s u m m a rized below only a few sites have one or more ambiguous nt. Starting from the left end of the genome, nt 1 to 19 399 are from XL47.1 (Loenen and Brammar, 1980), which should be identical to wild-type X (Daniels et al., 1983), except the BamHI site at nt 5505 has been removed by an uncharacterized mutation. Next, nt 19 400 to 19 492 of the vector correspond to (symbols=below) sequence CTAG followed by nt 6845 to 6760 of the rrnB operon (this segment contains rrnBT2) (Brosius et al., 1981) followed by ACG. Next nt 19493 to 19 5 0 6 = n t 1373 to 1360 of pBR322 (Watson, 1988). Next, nt 19 507 to 23 903 of the v e c t o r = t h e lac region, with lacy (nt 19 507 to 20760) (Buchel et al., 1980) and lacZ (nt 20812 to 23 883) (Kalnins et al., 1983), except the EcoRI site normally present at the 3' end of lacZ has been removed by an uncharacterized mutation. Then nt 23904 to 2 4 0 4 8 = n t 238 to 94 of pTL61T (Linn and St. Pierre, 1990), this includes the RNase III processing site and extends up to the MCS. Next, nt 24049 to 24 1 0 5 = n t 452 to 396 of pUC19, which is the pUC19 MCS. Then nt 24106 to 24 132 of the vector = the newly added section of the MCS which has the sequence 5 ' - G G C G C C T C G A G A T C A G A G A C C T A C G T A . Next nt 24133 to 24 329 of the vector = 5 ' - C C G A T C C C C A A T T C C followed by nt 6778 to 6610 of the rrnB operon (this segment contains rrnBT1) (Brosius et al., 1981) followed by G G A A T T G G G A A T T . Then nt 24330 to 24 522 = AA plus nt 4361 to 4171 of pBR322. Next nt 24 523 to 25 937 = the Tc R gene of pACYC184, nt 1453 to 2867 (Rose, 1988), except 1873 was changed from C to T by oligonucleotide mutagenesis to remove the BamHI site. This change does not alter the amino-acid sequence of the Tc R protein. The ~, sequences then resume with nt 25 938 to 33 776 of the v e c t o r = n t 26 532 to 34 366 of X, except nt 27 974 was changed from A to G by oligonucleotide mutagenesis to remove the BamHI site, and T C G A was added after nt 33 502 to remove the XhoI site. The change at nt 27 974 is within the int gene, but the amino-acid sequence of the protein is not altered. Next nt 33 777 to 35 625 of the vector = the approx. 1849-bp i 21 substitution. The sequence of this region has been determined except for two approx. 4-bp ambiguities, in the ell and cro genes. (Franklin, 1984; N.C. Franklin, personal communication). For purposes of this vector sequence they have both been represented as 4 N residues. This is followed by ut 35 626 to 37 509 of the v e c t o r = n t 38618 to 40501 of ~,. Finally, nt 37510 to 42704 of the vector = n t 43 308 to 48 502 of X. The ~, D N A missing between these two segments is the result of the nin5 deletion. Since the ~ sequences
(a) Features of kTXF97, the new transcriptional fusion vector (I) In order to measure 13Gal production from lacZ fusions carried on X vectors such as ~,TL61, the recombinant (re-) phage is used to form monolysogens in a Alac E. coli strain (Linn and St. Pierre, 1990). We have inserted the Tc R gene from pACYC184 into the right arm of the vector (Fig. 1) to permit direct selection of lysogens on plates containing 2 gg Tc/ml. This makes the isolation of lysogens simpler than cell transformation with drugresistance plasmids. (2) A monolysogen of the XTL61 vector has a background 13Gal activity of 71 u which can complicate the analysis of weak promoters. This level produces light pink colonies on MacConkey plates and light blue colonies on XGal plates. The majority of the background expression was found to result from a low level of transcription initiating somewhere to the right of the MCS that reads through into lacZ. The background level was reduced four-fold to 18 u by insertion of a 169-bp fragment carrying the efficient rrnBT1 transcriptional terminator just upstream from the MCS (Fig. 1). Since this level produces colorless colonies on MacConkey plates and faint blue colonies on XGal plates, re-fusions with even weak promoters can now be distinguished from the reconstituted vector. (3) Although XTL61 could incorporate DNA fragments up to 6 kb, this maximum amount was reduced by the insertion of rrnBT1 and the TcR gene upstream from the MCS. To compensate, we deleted non-essential regions of XTL61 including lacA with the downstream tandem copies of the complete rrnBt terminator region (Linn and St. Pierre, 1990) and 425 bp in the right arm of the vector (between nt 26 104 (EcoRI site) and 26 529 (PmlI site) of the X genome). A 94-bp fragment containing a single copy of rrnBT2 was inserted downstream from lacY (Fig. 1) to prevent transcription from the fusion continuing into the X left arm. As a result of these changes the size of XTXF97 is 42704 bp (Fig. 1), allowing it to incorporate DNA fragments up to 8 kb. (4) Finally, the MCS was altered from that of ~,TL61
in the fight arm of this vector originate from LNM616 (Mileham et al., 1980), the EcoRI sites normally present at nt 31 747, 39 168 and 44972 in the X genome all have been removed by uncharacterized mutations. The translational fusion vectors, LTLF97-1, -2 and -3, have the same sequence as XTXF97 except the MCS is attached directly to the ninth codon of the lacZ gene (see Fig. 2). Details of the construction of these vectors will be supplied on request. The sequences of XTXF97, XTLF97-1, ~,TLF97-2 and XTLF97-3 were submitted to G e n B a n k under accession Nos. U37692, U39284, U39285 and U39286, respectively.
67 TXF97 NarI SnaBI
SacI
XhoI
EcoRI
SmaI
XbaI
KDnI
BamHI
PstI SalI
HindIII EDhI
TACG~T~QGTCTCTGATCTCGAGGCGCCGAATTCGAGCTCGGTACCCGGGGATCCTCTAGAGTCGACCTGCAGGCATGCAAGCT~2~CTAACTAGCG --->
RIII
--->
lacZ
TLF97 - 1 NarI SnaB I
Xho I
TA CGT AGG
TCT CTG ATC
TCG AGG
SacI
SmaI
EcoR I
KDn I
CGC CGA ATT
CGA GCT CGG TAC
Xhol
PstI
BamH I
HindIII
Sa I I
SDh I
CCG GGG ATC CCT CGA GGT CGA CCT GCA GGC ATG
CAA GCT TGG GGG - - - >
TLF97
lacZ
-2 NarI
SnaBI
TAC GTA GGT
XhoI
CTC TGA TCT
SacI EcoRI
SmaI KDnI
CGA GGC GCC GAA TTC GAG
XbaI
PstI
BamHI
HindIII
SalI
CTC GGT ACC CGG GGA TCC TCT AGA GTC GAC
SDhI
CTG CAG GCA TGC AAG
CTT GGG GGG --->
lacZ
TLF97-3 NarI SnaBI
XhoI
T ACG TAG GTC TCT GAT CTC
Sa cI EcoRI
GAG GCG CCG AAT TCG AGC
SmaI KDnI
TCG GTA
XbaI BamHI
P s tI
HindI I I
SalI
CCC GGG GAT CCT CTA GAG TCG ACC
SDhI
TGC AGG
CAT GCA AGC
TTG GGG GGG --->
lacZ
Fig. 2. The nt sequence of the MCS in the transcriptional and translational lacZ fusion vectors. The restriction sites are overlined. Translational stop codons are underlined. )~TXF97 has stop codons in all three reading frames immediately downstream from the HindlII site. This ensures that any ribosomes continuing from the attached upstream sequences of a fusion will not read through into the RNase III site (RIII) disrupting its structure and thus prevent cleavage of the mRNA. The MCS sequences of the translational fusion vectors, )~TLF97-1, -2, -3, are indicated as triplets that are in the same translational reading frame as the lacZ gene. Following the G residues at the end of the MCS sequences for each of the translational fusion vectors is a junction sequence of three codons before the lacZ gene sequence begins at its ninth codon.
to facilitate the in vitro construction of fusions to lacZ. The sequence from the EcoRI site through the HindlII site is identical to the MCS of pUC19 (Fig. 2). An additional sequence was inserted upstream from the EcoRI site that contains the recognition sites for NarI, XhoI and SnaBI. As a result of these changes and in vitro mutagenesis to remove restriction sites elsewhere in the vector (see legend to Fig. 1) the XhoI, EcoRI, BamHI and XbaI sites of the MCS are unique and can be used to generate both intact right and left arms (Fig. 2,.Table I). Cleavage at the NarI, SmaI, SalI or HindlII sites will produce the complete left arm which includes lacZY, while cleavage at the SnaBI, SacI or KpnI sites will produce the complete right arm of the vector. As summarized in Table I, a wide variety of restriction endonucleases can be used to produce fragments from the DNA region of interest, whose upstream and downstream termini can then be directly ligated to the vector arms to produce the desired re-lacZ fusion. Details of the construction and recovery of re-fusion phage using ~ vector arms has been described previously (Linn and St. Pierre, 1990). (b) Translational fusion vectors, ~.TLF97-1, -2, -3 Starting with ~TXF97 a set of translational fusion vectors was constructed that carry the MCS in each of the three translational reading frames. This was accomplished by replacing the D N A segment of kTXF97 that extends from the MCS through the RNase lII processing site and into the 5' region of lacZ;with the MCS and 5' truncated lacZ segment from three translational fusion
TABLE I Restriction endonucleases that cleave in the MCS to produce complete left and/or right arms of the vectors Enzyme
Vectors arm(s) produced
SnaBI XhoI NarI
Right Right and left Left
EcoRI SacI
Right and left Right
KpnI
Right Left
Sinai
BamHI
XbaI b SalI
Right and left Right and left Left
HindlII
Left
Enzymes that generate compatible termini Any that produce blunt termini Sail AciI, BsaHI, BstBI, ClaI, HinPlI, HpalI, MaelI, Pspl406I, TaqI, (AccI; GT/CGAC) ~ ApoI, MfeI, Tsp509I
Any that produce blunt termini BclI, BgllI, BstYI, Sau3AI Avrll, NheI, SpeI XhoI
a AccI recognises several sequences, only the one listed generates termini compatible to those of the NarI site. u There is no XbaI site in TLF97-1.
plasmids (Jain, 1993). The kTLF97-1, -2 and -3 vectors are identical except for the presence of 5, 6 or 7 G residues, respectively, downstream from the HindlII site (Fig. 2). Also the XbaI site present in LTLF97-2 and -3 is replaced with an XhoI site in kTLF97-1 to avoid an in-frame stop codon present at the XbaI site (Jain, 1993). The background 13Gal level of the vectors is less than 0.1 u, allowing accurate measurement of even weak expres-
68 sion from re-fusions. Because of this extremely low background, lysogens of reconstituted vector are colorless even on plates containing XGal.
(c) Conclusions (1) The ~,TXF97, LTLF97-1, -2, -3 set of vectors with their expanded MCS greatly facilitate the direct in vitro construction of single-copy transcriptional or translational fusions to lacZ and should prove valuable in assessing the relative roles of transcription and translation in regulating the expression of a wide variety of genes. (2) Access to the nt sequence of these phage vectors expedites the confirmation and structure analysis of recombinant fusions by restriction site mapping. (3) The vectors insulate the MCS-IacZY region with flanking transcriptional terminators. (4) The presence of the Tc R gene allows for simple and direct selection of lysogens. ACKNOWLEDGEMENTS
We thank C. Jain for providing pLACZla, pLACZ2 and pLACZ3. This study was supported by the Canadian Medical Research Council. REFERENCES Adams, L.W. and Hatfield, G.W.: Effects of promoter strengths and growth conditions on copy number of transcription-fusion vectors. J. Biol. Chem. 259 (1984) 7399-7403.
Brosius, J., Dull, T.J., Sleater, D.D. and Noller, H.F.: Gene organization and primary structure of a ribosomal RNA operon from Escherichia coll. J. Mol. Biol. 148 (1981) 107-127. Buchel, D.E., Gronnelborn, B. and M011er-Hill, B.: Sequence of the lactose permease gene. Nature 283 (1980) 541-545. Daniels, D.L., Schroeder, J.L., Szybalski, W., Sanger, F., Coulson, A.R., Hong, G.F., Hill, D.F., Peterson, G.B. and Blattner, F.R.: Complete annotated lambda sequence. In: Hendrix, R.W., Roberts, J.W., Stahl, F.W. and Weisberg, R.A. (Eds.), Lambda II. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1983, pp. 519-676. Franklin, N.C.: Conservation of genome form but not sequence in the transcription antitermination determinants of bacteriophage ~., d~21 and P22. J. Mol. Biol. 181 (1984) 75-84. Jain, C.: New improved lacZ fusion vectors. Gene 133 (1993) 99-102. Kalnins, H., Otto, K., Rather, U. and MOiler-Hill, B.: Sequence of the lacZ gene of Escherichia coli. EMBO J. 2 (1983) 593-597. Linn, T. and St. Pierre, R.: Improved vector system for constructing transcriptional fusions that ensures independent translation of lacZ. J. Bacteriol. 172 (1990) 1077-1084. Loenen, W.A.M. and Brammar, W.J.: A bacteriophage vector for cloning large DNA fragments made with several restriction enzymes. Gene 10 (1980) 249-257 Mileham, A.J., Revel, H.R. and Murray, N.E.: Organization and expression of the f r d - D N A ligase region. Mol. Gen. Genet. 179 (1980) 227-239. Rose, R.E.:The nucleotide sequence of pACYC184. Nucleic Acids Res. 16 (1988) 355. Scott, J.R. and Rownd, R.H.: Regulation of plasmid replication. In: Alberts, B., Fox, C.F. and Trusser, F.J. (Ed.), Mechanistic Studies of DNA Replication and Genetic Recombination. Academic Press, New York, NY, 1980, pp. 1-8. Silhavy, T.J. and Beckwith, J.R.: Uses of lac fusions for the study of biological problems. Microbiol. Rev. 49 (1985) 398-418. Slauch, J.M. and Silhavy, T.J.: Genetic fusions as experimental tools. Methods Enzymol. 204 (1991) 213-247. Watson, N: A new revision of the sequence of pBR322. Gene 70 (1988) 399-403.