key: cord-0005215-fwav40fv
authors: Sekine, Yasuhiko; Nagasawa, Hiromichi; Ohtsubo, Eiichi
title: Identification of the site of translational frameshifting required for production of the transposase encoded by insertion sequence IS 1
date: 1992
journal: Mol Gen Genet
DOI: 10.1007/bf00279376
sha: 4128102a9f8fe15b8698d0390d88f69016e8804e
doc_id: 5215
cord_uid: fwav40fv

Previous genetic analyses indicated that translational frameshifting in the −1 direction occurs within the run of six adenines in the sequence 5′-TTAAAAAACTC-3′ at nucleotide positions 305–315 in IS 1, where the two out-of-phase reading frames insA and B′-insB overlap, to produce transposase with a polypeptide segment Leu-Lys-Lys-Leu at residues 84–87. IS 1 mutants with a 1 by insertion, which encode mutant transposases with an amino acid substitution within the polypeptide segment at residues 84–87, did not efficiently mediate cointegration, except for an IS 1 mutant which encodes a mutant transposase with a Leu-Arg-Lys-Leu segment instead of Leu-LysLys-Leu. An IS 1 mutant with the DNA segment 5′-CTTAAAAACTC-3′ at positions 305–315 carrying the termination codon TAA in the B′-insB reading frame could still mediate cointegration, indicating that codon AAA for Lys corresponding to second, third and fourth positions in the run of adenines is the site of frameshifting. The β-galactosidase activity specified by several IS 1- lacZ fusion plasmids, in which B′-insB is in-frame with lacZ, showed that the region 292–377 is sufficient for frameshifting. The protein produced by frameshifting from the IS 1-lacZ plasmid in fact contained the polypeptide segment Leu - Lys - Lys - Leu encoded by the DNA segment 5′-TTAAAAAACTC-3′, indicating that −1 frameshifting does occur within the run of adenines.

Insertion sequence IS 1 is the smallest active IS element in bacteria (Ohtsubo and Ohtsub 0 1978; Johnsrud 1979) and is involved in various kinds of genomic rearrangements, including the cointegration event between two Correspondence to: E. Ohtsubo replicons (Iida and Arber 1980; Ohtsubo et al. 1980 Ohtsubo et al. , 1981 . IS/ encodes two out-of-phase reading frames, insA and B'-insB, where B' is an open reading frame extending in-frame from the initiation codon ATG of the insB frame for 126 bp. The B' reading frame, which overlaps the 3' end of the insA frame, is in the -1 frame with respect to insA. Previous genetic analyses indicated that a frame shifting event occurs in the -1 direction within a run of six adenines, which lies in the overlap region between insA and B', to fuse insA and B'-insB by translation, producing the InsA-B'-InsB fusion protein that has IS 1 transposase activity (Sekine and Ohtsubo 1989) . The InsA protein, which is produced unless the frameshifting event occurs, may play a role as a negative regulator of transposition (Machida and Machida 1989; Sekine and Ohtsubo 1989; Zerbib etal. 1990 ). Since the efficiency of frameshifting determines the ratio between the amount of InsA and that of transposase, frameshifting is thought to be a mechanism which controls transposition of IS 1 (Sekine and Ohtsubo 1989; Escoubas et al. 1991) . The production of transposase encoded by other IS elements, such as IS3 and perhaps those related to IS 3, has been suggested to depend on -1 frameshifting within a run of adenines in these elements (Sekine and Ohtsubo 1991) . Recently, frameshifting in IS i50 which is related to IS3, has been demonstrated (V6gele et al. 1991) .

We present here further genetic analyses which support the concept of frameshifting in IS 1 and show that the precise site of frameshifting is codon AAA for Lys in the run of adenines in insA. We also present here the result of amino acid sequencing analysis showing that -1 frameshifting does occur in the run of adenines to produce transposase with the polypeptide segment Leu -Lys -Lys -Leu.

Bacterial strains andplasmids. Bacterial strains used were Escherichia coli K12 derivatives, MVl184 (A [lac- proAB] ara strA thi[q580 lacZAM15] A [srl-recA] ] dutt ungl thil retAt) (Kunkel et al. 1987) , MC1000 (F-araD139 A [ara teu] 7697 A lacX74 galU galK strA) (Casadaban and Cohen 1980) , JE5519 (F-fecAl man aroD argEH str nalA lac gat xyl mtt) (Ohtsubo et ai. 1981) , GC4670 (F-lon::TnlO thr leu tac'~) (a gift from S. Casaregola), and AD202 (F-ompT::kan araDt39 A[argF-lac] U169 rpsLl50 relA1 flbB5301 deoCt ptsF25 rbsR) (a gift from Y. Akiyama). The strain YS202, which is a londerivative of AD202, was constructed by P1 transduction of AD202 to tetracycline resistance, using a lysate grown on GC4670.

Plasmid pSEKI50 a pUC18 derivative, carries one copy of IS 1 between the KpnI and BamHI sites within the multiple cloning site segment (Sekine and Ohtsubo 1989) . Plasmid pSEK24, a derivative of pSEK15, car~qes an IS t mutant, IS •-24, with a base substitution at position 322 (Sekine and Ohtsubo 1989) ; plasmids pSEK31, pSEK32 and pSEK33, which were derived from pSEK15, have IS 1 mutants each with a 1 bp insertion, IS 1-31, IS 1-32 and IS •-33, respectively (Sekine and Ohtsubo 1989) . Plasmid pUC119 (Vieira and Messing 1987) was used to construct pSEK117, as will be described below. Construction of pSEKt7 and its derivatives will also be described below. Plasmid pHS1 is a temperaturesensitive replication mutant of the tetracycline-resistance plasmid pSCI01 (Hashimoto-Gotoh and Sekiguchi 1977) . Plasmid pR-pMLB (a gilt from D. Bastia) is a pBR322 derivative, from which IS 1 -tacZ fusion plasraids were constructed as described below.

Media. Culture media used were L broth, L-rich broth, 4~-medimn (Yoshioka et al. 1987) and 2 x Y T broth (Messing 1983) . qS-medium was used for transformation of plasmid DNA, and 2 x YT broth was used for mutagenesis in constructing mutant plasmids. L-agar plates contained 1.5% (w/v) agar (Eiken Chemical) in L broth. Antibiotics were added in L-agar plates, if necessary, at concentrations of 150 gg ampicillin/ml (Wako Junyaku), 5 or 10 gg tetracycline/ml (Sigma). Peptone dilution buffer (0.1% peptone (Kyokuto Seiyaku) in 0.3% NaC1) was used for dilution of cell cultures. (BamHI, BgtII,-HindIII, KpnI, PstI, SatI and SphI) , bacterial alkaline phosphatase, T4 DNA polymerase, T4 polynucleotide kinase, and T4 DNA ligase were purchased from Takara Shuzo. Restriction endonuclease BstEII was purchased from New England Biolabs. RNase A was purchased from Sigma. These enzymes were used in the buffers recommended by their suppliers.

DNA preparation. Strain MV1184 or MCI000 harboring a plasmid was grown in L-rich broth. The alkaline lysis method (Maniatis et al. 1982 ) was used to prepare plasmid DNA for cloning and nucleotide sequencing.

Nucleotide sequencing. Nucleofide sequences were determined by the dideoxynucleotide method (Sanger et al. 1977; Messing 1983 ) using a 7-DEAZA sequencing kit and M4 primer (Takara Shuzo). Synthetic oligodeoxyribonucleotide D~ described in Sekine and Ohtsubo (1989) was also used as primer for sequencing derivatives of pSEK117. The DNA chains were labeled with e-[3ZP]dCTP (i 5 TBq/mmoI, Amersham) and separated in 6 or 8% polyacrylamide gels containing 8 M urea.

Plasmid construction, pSEK17 has only one cleavage site for PstI in the insA coding region in IS 1, and was constructed by self-ligation of plasmid pSEK15 after digestion with SphI and SalI and treatment with DNA polymerase I (Klenow) to remove the PstI site flanked by t.he SphI and SalI sites present in the pUC18 sequence. Each pSEK17 derivative carrying an IS/ mutant with a substitution(s) or a I bp insertion was then constructed as follows. The KpnI-HindIII fragment in the cloning site segment in vector plasmid pUCI19 was replaced with the KpnI--HindIII fragment of pSEK17 containing the entire IS/ sequence, yielding pSEK117. Using pSEKll7 as template and oligodeoxyribonucleotides synthesized using a DNA synthesizer 380B (Applied Biosystems) as primers, the IS 1 sequence in pSEKI 17 was mutagenized by site-directed mutagenesis according to Kunkel et al. (1987) . The sequences mutated were confirmed by DNA sequencing. Then the PstI-BstEII fragment of IS/ in pSEK17 was replaced ~vith the PstI-BstEII fragment of each of the resulting pSEK117 derivatives.

Plasmids pSEK2055, pSEK207, pSEK9000 and pSEK6000 are IS t -l a c Z fusion plasmids having a DNA fragment of wild-type IS 1. These were constructed as follows. Using pSEKI 17 as template, two BglII recognition sites were introduced into appropriate positions (see Fig. 3 ) flanking the run of adenines within the IS t sequence, by site-directed mutagenesis as described above. The sequences mutated were confirmed by DNA sequencing. Each of the BglII fragrnents was then inserted into the BamHI site of vector plasmid pR-pMLB. Plasmids pSEK2055-I, pSEK207-I, pSEKg000-I and pSEK6000-I are I S i -l a c Z fusion plasmids having a DNA fragment of IS 1 with a single adenine insertion in the run of adenines. These were constructed in the same way as the I S / -l a c Z fusion plasmids, such as pSEK2055 etc. using, however, plasmid pSEK13I, which carries ISt-31 with a single adenine insertion in the run of adenines as template; pSEK131 itself was obtained by replacing the KpnI-HindIII fragment in the cloning site segment of vector plasmid pUC119 with the KpnI-HindIII fragment of pSEK31 which includes IS 1-31 (Sekine and Ohtsubo 1989) .

Purification of fl-galactosidase (LacZ) fusion proteins and amino acid sequencing. Strain YS202, harboring the IS l -l a c Z fusion plasmid pSEK9000 or pSEK9000-I, was grown in L-rich broth (301 for pSEK9000 and 250 ml for pSEK9000-I), containing 0.2% (w/v) glucose at 30 ° C until the OD6o o reached 0.5-0.6, and then the culture was incubated ~vith aeration at 40 ° C for 60 rain to induce the LacZ fusion protein. The protein was purified from these cells according to the procedure of In-amoto and Ohtsabo (1990) , except that a French press was used to disrupt the cells instead of sonication. After lyophilization, the protein was solubilized in water, subjected to SDS-polyacrylamide gel electrophoresis, and subsequently electroblotted onto PVDF-type membrane (ProBlott, Applied Biosystems), according to the method of Aebersold et al. (1986) . The protein band of interest was excised and subjected to amino acid sequencing analysis using an Applied Biosystems model 470A sequencer or model 477A sequencer fitted with an on-line Applied Biosystems 120A high performance liquid chromatography analyzer. Here, to avoid possible degradation of the protein we used strain YS202. This strain carries a mutation in the lon gene, which encodes a protease involved in the degradation of some unstable proteins (for a review, see Gottesman 1989) , and a second mutation in the ompT gene, which encodes an outer membrane-associated protease responsible for in vitro cleavage of several proteins, including the SecY protein (Akiyama and Ito 1990) , during their purification. Indeed when an ompT ÷ strain was used as a host, cleavage of the protein between two consecutive basic amino acids was observed.

Cointegration assay. Each of the ampicillin-resistance pSEK plasmids carrying wild-type IS 1 or mutant IS 1 was introduced by transformation into the E. coli K12 strain JE5519, which already harbored the tetracyclineresistance plasmid pHS1. Cointegration between a pSEK plasmid and pHS1 was assayed according to the method described in Sekine and Ohtsubo (1989) .

LacZ assay. Each of the ampicillin resistance plasmids carrying the IS i-lacZ fusion gene was introduced by transformation into MC1000. Liquid cultures of MC1000 harboring a plasmid were incubated overnight at 30 ° C in L-rich broth containing 100 gg ampicillin/ml, and diluted 1/100 into L-rich broth. After shaking at 30°C until turbidity at 600 nm had reached 0.15-0.2, the temperature was shifted to 40 ° C. When the turbidity of the culture was 0.8-0.9, LacZ activity was measured by the method described by Miller (1972) .

Our previous analyses of the cointegration ability of IS 1 mutants carrying a nonsense mutation in the insA or B'-insB reading frame have suggested that the run of six adenines at nucleotide positions 307-312, where the two frames overlap (see Fig. 3 ), is likely to contain the possible frameshift site (Sekine and Ohtsubo 1989) . Also, our analyses of several IS I mutants with single bp insertions within or close to the run of adenines which results in the placement of insA and B'-insB in the same reading frame, have shown that one IS I mutant with a single adenine insertion in the run of adenines (see mutant IS1-31 in Fig. 1) can mediate cointegration at a much with a single adenine insertion is supposed to produce wild-type IS i transposase with a polypeptide segment LKKL without frameshifting. In the other IS i mutants, which produce mutant transposases, only the codons altered and the amino acids substituted are indicated. Boldface letters indicate the mutated nucleotides.

The frequency of cointegration (per division cycle) mediated by wild-type IS 1 or each IS i mutant is shown with its relative value in parenthesis by taking the frequency for IS 1-31 as 100. Cointegration frequencies mediated by IS 1-31, IS/-32 or IS 1-33 were taken from Sekine and Ohtsubo (1989) higher frequency than does either wild-type IS 1 or any of the other mutants with a 1 bp insertion neighboring the run of adenines (see IS1-32 and IS1-33 in Fig. 1 ) (Sekine and Ohtsubo 1989) . The mutant IS1-31 is considered to produce, without frameshifting, active transposase in which the amino acid sequence at residues 84-87 is Leu--Lys-Lys-Leu (LKKL in Fig. 1 ), but the other mutants produce inactive transposase with FKKL or LKNL (see Fig. 1 ). These results have suggested that wild-type IS 1 produces transposase having the polypeptide segment LKKL by -1 frameshifting in the run of six adenines. To support this suggestion, we constructed four other mutants with a 1 bp insertion designed to alter the amino acid sequence in the possible IS i transposase. We then examined cointegration between each of these mutant plasmids as donor and pHSI, a temperature-sensitive replication mutant of a tetracycline-resistance plasmid pSC101 as recipient, by selecting for cells harboring cointegrates which can replicate in the presence of tetracycline at 42 ° C. Three mutants, IS 1-35, IS •-36 and IS 1-38 (Fig. 1) , which generate Sekine and Ohtsubo (1989) . See the legend to Fig. 1 for further details transposases with LEKL, LQKL and LKKV polypeptide segments, respectively, instead of LKKL, did not efficiently mediate cointegration. Conversely, one mutant (IS1-37 in Fig. 1 ) which generates a transposase with an LRKL segment, with a related amino acid substitution R (Arg) for K (Lys) at the 85th residue, mediated cointegration at an even higher frequency than that of IS1-31 (Fig. 1) . This supports the suggestion above that IS1 transposase with the polypeptide segment LKKL at residues 84-87 is the product of frame-shifting and used to mediate cointegration in wild-type IS 1.

To give rise to the transposase with the polypeptide segment LKKL from the DNA sequence 5'-TTAAAAAACTC-3' at nucleotide positions 305-315, frameshifting is likely to occur after recognition of the codon TTA at nucleotide positions 305-307 (designated 3°STTA) in insA as the 84th residue L (a4L in Fig. 2 ) in one of the following two ways (see Fig. 2 ): (i) codon 3°TAAA in B'-insB is recognized to give aSK; or (ii) codon 3°SAAA in insA is read as ~SK and then codon 31°AAA in B'-insB is recognized as a6K. To determine which is the case, we constructed an IS I mutant, IS1-81, with two substitutions, C for nucleotide T at 305 and T for nucleotide A at 307, which not only convert codon 3°STTA for 84L in insA to the synonymous codon CTT, but also convert codon 3°TAAA for the first K in B'-insB to the ochre codon TAA (Fig. 2) . This IS 1 mutant retained the ability to mediate cointegration (Fig. 2) . This indicates that the mutant produces active transpo-sase upon frameshifting, using the 3°SAAA codon for SSK as the last codon in insA at which -1 frameshifting occurs during translation to give rise to transposase. The other two IS 1 mutants were constructed as negative controls: IS/-62 with a substitution of A for T at position 306, which introduces an ochre codon in insA at a position upstream of the run of adenines, and IS/-24 with a T for C substitution at position 322 which introduces an amber codon in B'-insB downstream of the run of adenines. These mediated cointegration at greatly reduced frequencies (Fig. 2) as expected, since both of the termination codons introduced resulted in production of a truncated protein and, therefore, inactive transposase.

To determine the nucleotide sequence required for and the efficiency of, frameshifting in IS/, we constructed several IS 1-lacZ fusion plasmids having a DNA fragment of wild-type IS 1 containing the run of adenines. The fragment is flanked by the ATG of cro of bacteriophage ,t and the lacZ gene, such that insA is fused with the initiation codon ATGcro in-frame and B'-insB is fused with lacZ in-frame (Fig. 3) . The transcription of the fusion gene from the 2 PR promoter for cro is under the control of a thermosensitive repressor, the product of ci857 which is also carried by the fusion plasmid. The occurrence of -1 frameshifting during translation of insA would therefore result in the synthesis of the InsA-B'-InsB-LacZ fusion protein with /~-galactosidase (LacZ) activity. We also constructed IS i-lacZ fusion plasmids having a DNA fragment of IS ! with a single adenine insertion in the run of adenines, so that insA is fused with B'-insB-lacZ in-frame. The ratio of LacZ activity specified by the out-of-frame plasmid to that specified by the corresponding in-frame plasmid is considered to reflect the efficiency of frameshifting in the out-of-frame plasmid.

The LacZ activity measured after heat induction in lysates of cells harboring pSEK2055, having a DNA insert corresponding to IS i coordinates 63-377 and containing the entire insA and B' coding frames, was 1.73 units (Fig. 3) . On the other hand, LacZ activity specified by pSEK2055-I, a derivative of pSEK2055 with a single adenine insertion in the run of adenines, was 811 units (Fig. 3) . Thus, the efficiency of frameshifting in pSEK2055 was estimated to be 0.21% (Fig. 3) . LacZ activity specified by plasmid pSEK9000 having the region 292-377 was 17.9 units, while the LacZ activity specified by the corresponding in-frame plasmid pSEK9000-I was 5220 units (Fig. 3) . The efficiency of frameshifting in pSEK9000 was thus estimated to be 0..34% (Fig. 3) , which was alsmost the same level as that in pSEK2055. This suggests that the region 63-291, which is upstream of the run of adenines, is not required for frameshifting. The increase of LacZ activity specified by pSEK9000 (and pSEKg000-I), when compared with that specified by pSEK2055 (and pSEK2055-I) (see Fig. 3 ), may reflect instability of a portion of the InsA in boldface are those determined by sequencing the purified LacZ fusion protein produced; numbers above or below the amino acids indicate cycle numbers of Edman degradation of the LacZ fusion protein. The 9th lysine residue (K) of the protein specified by pSEK9000 is encoded by codon 3°SAAA in insA as described above (see Fig. 2 ). B Critical phenylthiohydantoin (PTH)-amino acids (in pmol) detected during each sequencing cycle of Edman degradation of the purified LacZ fusion protein produced from cells harboring pSEK9000 (a) or pSEKg000-I (b). The data are not corrected for injection, base line, and tailover. A number over a vertical line represents the major PTH-amino acid recovered from that cycle protein or a portion of the transcript encoded by pSEK2055 (and pSEK2055-I).

The efficiency of frameshifting in pSEK6000 carrying the region 292-353 was 0.25% (Fig. 3) which was reduced relative to that in pSEK9000; a similar reduction has been reported by Escoubas et al. (1991) . The degree of the reduction, however, seemed not to be so great, and the efficiency in pSEK6000 is the same as that in pSEK2055. Moreover, the efficiency in pSEK6100 hav-ing the region 292-332 was the same as that in pSEK2055 and pSEK6000 (data not shown). In the two plasmids, pSEK6000 and pSEK6100, several possible secondary structures seen in the region downstream of the run of adenines (Sekine and Ohtsubo 1992) are deleted. These show that the region 333 377 is not essential for frameshifting, suggesting that the contribution of the secondary structures downstream of the run of adenines to the efficiency of frameshifting is small (Sekine and Ohtsubo 1992) . Note here that the efficiency of frameshifting in pSEK207 having the region 63-314 was, however, 0.047% (Fig. 3) , a 4.5-fold decrease compared with that in pSEK2055. This suggests that the region 315-377, which is located downstream of the run of adenines, is required to stimulate frameshifting. The reduction of efficiency of frameshifting seen in pSEK207 was considered to be caused mainly by the elimination of the termination codon of insA, but not by the elimination of secondary structures, as described above and by Sekine and Ohtsubo (1992) .

To obtain direct evidence for frameshifting in IS 1, we determined the amino acid sequence of the protein produced in the form of the InsA-B' -InsB-LacZ fusion protein from plasmid pSEK9000, which contained a region of IS 1 sufficient for efficient frameshifting, as described above. In this plasmid, the codon for Leu at the 84th residue in IS 1 transposase becomes the 8th codon in the coding region for the LacZ fusion protein, where the ATG of cro is the first codon (see Fig. 4A ). The LacZ fusion protein, which was overproduced by heat induction in cells harboring pSEK9000, was readily purified using a LacZ-specific affinity column (see Materials and methods). The purified protein was subjected to 21 cycles of Edman degradation (Fig. 4A) , and the relevant phenylthiohydantoin (PTH)-amino acids detected in the first 13 cycles are shown in Fig. 4B (a). This shows that translation of the LacZ fusion protein was initiated at the ATGcro and continued in-frame along insA, but shifted into B'-insB-lacZ. The amino acid sequence detected at cycles 8-11, which was produced from the sequence 5'-TTAAAAAACTC-Y, was LKKL ( Fig. 4B, a) , as expected from the results obtained from genetic analyses described above.

We also overproduced, purified, and sequenced the LacZ fusion protein from cells harboring the corresponding in-frame plasmid pSEKg000-I (Fig. 4A) . The pattern of PTH-amino acids (Fig. 4 B, b) was the same as that from the protein specified by pSEK9000 (Fig. 4 B, a) , as expected.

We have demonstrated here that shifting of the reading frame from insA to B'-insB in IS 1 occurs within the run of six adenines in the -1 direction so as to produce IS 1 transposase with amino acid segment LKKL at residues 84-87, and that the last codon in insA recognized during translation of IS/transposase is 3°SAAA in the run of adenines for 85K. These results suggest that a tRNA for 85K, recognizing the codon 3°SAAA in insA, plays a key role in frameshifting in the -I direction. This suggestion is further supported by mutational analysis of the run of adenines (Sekine and Ohtsubo 1992) .

A run of adenines has been implicated in -1 frameshifting in other genes, such as the dnaX gene of E. coli (Blinkowa and Walker 1990; Flower and McHenry 1990; Tsuchihashi and Kornberg 1990) , genes encoded by IS150 (V6gele et al. 1991 ) and IS3 (Sekine and Ohtsubo 1991) , and the gag andpro genes of mouse mammary tumor virus (MMTV) (Hizi et al. 1987; Jacks et al. 1987; Moore et al. 1987) . In MMTV, -1 frameshifting occurs at the gag-pro overlap, with the sequence AAAAAAC identical to that of IS 1, to produce the transframe protein, such that the last codon in gag (0 frame) is codon AAC (Hizi et al. 1987) and is downstream by one codon when compared with the site in IS 1. This difference between IS 1 and MMTV and probably other retroviruses which are thought to use the AAAAAAC sequence for frameshifting (Rice etal. 1985; Shimotohno et al. 1985) might be due to the structural or functional differences between prokaryotic and eukaryotic molecules participating in the translational process. The run of six adenines in dnaX and IS150 is followed by a guanine residue instead of a cytosine residue, unlike IS1 and MMTV. In dnaX, the amino acid sequence of the protein produced by frameshifting from the DNA segment 5'-GCAAAAAAGAG-3' is-AKKE (Tsuchihashi and Kornberg 1990) , indicating that frameshifting occurs at one of the consecutive codons, AAA and AAG, for lysine (K) in the 0-frame. In IS 150, the amino acid sequence of the protein produced by frameshifting from the DNA segment 5'-CUAAAAAAGCU-3' was LKKA (V6gele et al. 1991) , indicating that frameshifting occurs at either codon CUA for leucine (L) or one of the consecutive codons, AAA and AAG, for lysine (K) in the 0-frame. Since the exact site for frameshifting is not clear in dnaX or IS I50, it is unknown at present whether the site of frameshifting in these genetic systems is the same as that in IS 1 or not.

Secondary structures downstream of the frameshift site have been demonstrated to stimulate frameshifting in many genetic systems (Jacks et al. 1987 (Jacks et al. , 1988 Brierley et al. 1989; Flower and McHenry 1990; Tsuchihashi and Kornberg 1990; Dinman et al. 1991 ; V6gele et al. 1991 ), but such structures seem not to be responsible for efficient frameshifting in IS 1 (for further experiments on, and discussion of the role of the secondary structures in IS/, see Sekine and Ohtsubo 1992) . We have shown here that the efficiency of frameshifting in IS i is 0.2-0.3%. This is very low when compared with other cases, for example, 5-25% in retroviruses (for a review, see Varmus and Brown 1989) , 40 50% in dnaX (Flower and McHenry 1990; Tsuchihashi and Kornberg 1990) , and 30% in IS150 (V6gele et al. 1991) . The lack of the secondary structures which stimulate frameshifting might result in such a low efficiency of frameshifting in IS 1. It is reasonable to assume that IS 1 adopts a low level of frameshifting, which results in a low level production of transposase, to avoid deleterious rearrangement of .the host chromosome containing IS 1.

Electroblotting onto activated glass

SecY protein, a membrane-embedded secretion factor of E. coli, is cleaved by the OmpT protease in vitro

Programmed ribosomal frameshifting generates the Escherichia coli DNA polymerase III ?, subunit from within the z subunit reading frame

Characterization of an efficient coronavirus ribosomal frameshifting signal: requirement for an RNA pseudoknot

Analysis of gene control signals by DNA fusion and cloning in Escherichia coli

A -1 ribosomal frameshift in a double-stranded RNA virus of yeast forms a gap-pol fusion protein

Translational control of transposition activity of the bacterial insertion sequence IS

The 7 subunit of DNA polymerase III holoenzyme of Escherichia coli is produced by ribosomal frameshifting

Genetics of proteolysis in Escherichia coli

Mutations to temperature sensitivity in R plasmid pSC101

Characterization of mouse mammary tumor virus gap-pro gene products and the ribosomal frameshift site by protein sequencing

On the role of IS/ in the formation of hybrids between the bacteriophage PI and the R plasmid NR1

Specific binding of TraY protein to oriT and the promoter region for the traY gene of plasmid R100

Two efficient ribosomal frameshifting events are required for synthesis of mouse mammary virus gag-related polyprotein

Signals for ribosomal frameshifting in the Rous sarcoma virus gag-pol region

DNA sequence of the transposable element IS 1

Rapid and efficient site-specific mutagenesis without phenotypic selection

Regulation of IS/ transposition by the insA gene product

Molecular Cloning: A laboratory manual

Complete nuleotide sequence of a milk-transmitted mouse mammary tumor virus: two frameshift suppression events are required for translation of gag and pol

Plasmids containing insertion elements are potential transposons

Mechanism of insertion and cointegration mediated by ISI and Tn3

Nucleotide sequence of an insertion element, IS 1

The gag and pol genes of Bovine leukemia virus: nucleotide sequence and analysis

DNA sequencing with chain-terminating inhibitors

Frameshifting is required for production of the transposase encoded by insertion sequence 1

Translational frameshifting in IS elements and other genetic systems

DNA sequences required for translational frameshifting in production of the transposase encoded by IS

Complete nucleotide sequence of an infectious clone of human T-cell leukemia virus type II: an open reading frame for the protease gene

Translational frameshifting generates the 7 subunit of DNA polymerase III holoenzyme

Retroviruses. In: Berg DE, Howe MM (ed) Mobile DNA

Production of single stranded plasmid DNA

High-level ribosomal frameshifting directs the synthesis of IS 150 gene products

Repressor gene finO in plasmids R100 and F: constitutive transfer of plasmid F is caused by insertion of IS 3 into finO

The regulatory role of the IS/-encoded InsA protein in transposition

Acknowledgements. We thank D. Bastia for kindly sending us plasmid pR-pMLB, S. Casaregola for strain GC4670, and Y. Akiyama 324 for strain AD202. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan. Y.S. is the recipient of a JSPS Fellowship for Japanese Junior Scientists.