FEMS MicrobiologyLetters 66 (1990) 135-140 Pubhshed by Elsevier
135
FEMSLE03789
Mu gem3 as a tool to investigate the influence of chromosome supercoiling on gene expression in Escherichia coli K12 E. Bianchi, I. Rubcrti, P. G h e l a r d i n i and L. Paolozzi 1 Cenffo Studx per el! AcJ~I Nuclewa del CNR. and I Dipartlmento dl Bsologsdl. Umverslld dl R o ~ "T~ Vergata~,Rome. Ira/y
Receivedl0 July 1989 Accepted 26 July 1989 Key words: Phage Mu; DNA superenlhng; Gene expression: ts Mutants
1. SUMMARY We have developed a rapid method to investigate the influence of chromosome supercoihng on gene expression m Eschenchra coh KI2. This method exploits the ability of the gem3 mutant of the bacteriophage Mu, even in the prophagtc state m immune cells, to mduce relaxaUon of the host chromosome. The expcnmants can thus be performed under physiological condnions, and without the use of the drugs. In theory, this method can be apphed to any bacteria! eerie. Here, we report the results obtained wnh four DNA rephcation and three cell di,asmn genes.
2. I N T R O D U C T I O N The degree of DNA supercoillng in E coh ts homeostatically regulated by the ¢quihbrtum between two tmzymauc activmes with opposing elloots: the DNA gyras¢ and ~ e DNA topo.somerase 1 [1,2 I.
Correspondence to" L Panloz~, Centre Stud1 per gh Aodl Nu¢lelct d¢l CNR, Dlpartanenlo dt Genetlca ¢ Btologla Molecolar¢, Umversa~dt Roma "La Saplenza', P~azzale Aide More 5. Reran, Italy
Both m wvo and in vitro studies indicate that the topological state of DNA plays an important role m several btolo~fal processes such as rephcaUon, recombination and gene expression (for reviews se¢ [1,2]). As far as this last process is concerned, random cloning of E. colt chromosomal DNA fragments into a promoter clomng vector has shown that the expression of about 40~ of the bactenal promoters was substantially modified after the mhibttton of DNA $yrase with coumermycm [3] Analogous results have been reported [4] for gene expression in Salmonella typtumurmm, using a techmque of in viva gene fusion by Mu-dlac. The use of mutants defootsve in either DNA gyrase (gyrA and g y r B ) or DNA topolsomerase (topA) genes and of specific inhibitors of DNA gyrasc has pcrmnted the identification of a number of genes which respond differentially to changes m DNA supereoding, such as the genes gyrA and g y r 8 [5] or the operons be1 [6] and histidine [7] The observauon that the product of the gene gem of bacteriophage Mu is able to modulate gene expression m E coh via an interaction with the host DNA gyrasc [g] allowed us to use the bacteriophage Mu as a reliable instrument to investigate the effect of DNA topology on the expressrun of specific bacterial genes. We have de-
0378-1097/89/$03 50 © 1989FedeTatlonof European MicrobsologlcalSocieties
136 veloped a method to study this effect on essential E colt genes exploiting the ability of Mu, even as a prophage, to cause a relaxation of the D N A and consequently to suppress ts mutations in genes activated by D N A relaxation. In fact, increased transcription results in a greater amount of the gene product and hence also of the amount of the ts protein statistically not-denatured at high temperature. Therefore, if the examined gene is positively regulated by D N A relaxation, the phenotype of a ts mutant in this gane should be more thermoreslstant when lysogenie for Mu than the non lysogen; m contrast, if the examined gene is positively regulated by D N A supercolhng, the expected result will be an increased thermosensitivlty.
3. MATERIALS A N D M E T H O D S 3.1. Bactertal stratus, phages and plasmM used m tMs work are listed in Table 1. 3 2. Me&a: LB and LB plates [9] were used for standard cultures and plating LB and LB plates without the addttton of NaCI ( L B ( - ) ) were also used for fts mutants. 3.3. Lysogemzatwn: Bactenai strains, grown at 3 0 ° C m LB to about l x l 0 ~ cells/nil, were lysogemzed with Mu c + gem3 as already described ]10]. 3.4. Eff|clency of plating was measured at vanous temperatures on bacterial strams grown as described above. 3.3. Kmencs of growth: Baetenal strains, grown as described above on LB or L B ( - ) , were shifted to the indicated temperature. Aliquots were Table l Bacterial stratus c6oo CRT 46 25 CRT 266 JW 113 PC 79 GC 2624 GC 2625 AX 655
Relevantgenotype thr, leg tht dnaA ts,~ dnaB Is:~s dnaE tssl t daaC ts /tsAts /tsZ tss, (tsl rs
Provtdedby L Paoloz~a R "thomas R Thomas R. Thomas A, Falasch R D'An R DAn R. D'An
withdrawn at various times, spread on LB plates and incubated at 30°C. During the growth at 4 0 ° C and 4 2 ° C of strain GC2625, lysogenic or not for Mu gem3, afiquots were withdrawn, observed at the optical microscope (300 x and 600 x ) and photographed with Kodak TMX100 fdm. 3.6. Plasm:dsupercodmg" Bacteiaal strains were transformed with pUC13 plasrmd D N A [11] as described [12]. Plasmid D N A was than extracted from each strain after overnight growth at 3 0 ° C and analyzed on agarose gel supplemented with cMoroquine (Sigma) as described [8].
4. RESULTS A N D DISCUSSION We have previously shown that Mu can suppress the effects of thermosens,:ttve mutations in the D N A ligase gene of E. cob via an interaction with the host enzymatte apparatus controlling the D N A topology. This suppression is greatest with the Mu gem3 mutant. In bactenal strains lysogenie for Mu gem3, we have also observed a substantaal increase (towards the relaxed state) in the average linking number of a resident plasoud [8]. The same effect in one of the stratus used in this study is shown m Fig. 1; m the lysogenie strain the average linloog number of pUC13 plasnud D N A is increased wtth respect to the non-lysogenic strain. 4.1. Effect of chromosome relaxatwn on DNA rephcaflon genes Thcrmosenstttve mutants m D N A replication genes dnad, dnaB, dnaC and dnaE were lysogenized with Mu gem3 and analyzed for thermosensitiwty at non-pcmussive or senu-permisstv¢ temperatures. Fig. 2 shows the kinetics of growth of the 4 mutants lysogenic or not for Mu gem3 at the most signihcattve temperature. The results mthcate that the thermoseusmvity produced by the ts mutatton m genes dnaA, dnaB and dnaE is partially suppressed by the presence of Mu gem3 a~ a prophage suggestmg that these genes could be posttively regulated by chromosome relaxation. On the other hand, dnaC gene express:on seems not to be affected by variation in D N A topology, since no
137
A
B
Fig l Supercolhngof pUCI3 plaslrad DNA isolated from CRC2624fts,4 strain lysogemcor not for Mu gem3 Electrophoresls was performed in the presence of O, 11 5 and 13 5/~8/ml of chloroqmne The Fig represL~tsIhe results of the agarosegel contmnmgIt 5 ~tg/'ml of chloroqmne and the analysis of Ihe topoasomerspaltern performed with an LKB Uhroscan (A) GC2624 (Mu gem3). (B) GC267A.
difference was observed between the lysogemc and the non-lysogenic strain at any temperature from 30 to 43°C. Ftg. 2C shows only one of the various growth kmettes obtained. It should be noted that the ts mutants in genes whose expression is stimulated by D N A relaxation are differentially suppressed by Mu gem3. Tiffs could be due to at least two not mutually exclusive reasons: (1) each promoter reaches mammal acuvauon at a specific superhehcal density [13] and Mu gem3 induces a modthcation m the average linking number that would probably not be the optimal for the different promoters of the exarmned genes; (2) the amount of remammg actxve protein necessary to assure the surv3val of the mutant could differ m the various ts stratus.
4.2. Effect of chromosome relaxation on cell dwsswn genes A number of cell dtvision genes, called fts, have been tdentifted m E colt by the selecuon of mutants carrying thermosensitive mutaUons which give rise to cell fdamentation at non permissive
temperature For our analysis, we have chosen three fts genes: ftsA, ftsl and ftsZ, that are clustered at 2 nun on the E. cob genetic map [14,15]. The temperature sensmve mutants in these genes were lysogemzed with Mu gem3 and analyzed for thetr thennosensitivity using different media since it was shown that the presence of salt m the growth medium is able to correct partially the ts phenotype of some fts mutants [16]. The results (Table 2) show that the presence of Mu gem3 renders the strata carrying the ftsA mutatton more thermosensitive than the nonlysogen. According to our hypothesis, this would suggest that the ftsA gene expression is activated by chromosome supercoifing. The thermosensittvity of both ftsl and ftsZ is instead partially corrected m the lysngenic strains. Analogous resuits have been obt='uned studying the ~neucs of growths at different non-pernusslve or semi-permtsstve temperatures (data not shown). In addttlon, the reduced thermosenslttvity m GC2625 strata lysogemc for Mu gem3 is also coupled to a reduction in the size of the fdaments produced by
138
'i
w.o
05 0
A
,
,
80
60
~
~
,
,
120
150
15
01
0
,
.
60
120
rains "-=-¢lr~A
rains
""-dr~A[l~ugam3)
"=-~B
ft.
"~-dr~B(t~utw,3]
Growth el 4 0 C
Growth at 37C N/NO
..j 180
C
w.o
D
0.1
0.01
(X6"
~0
0901 0
~r4
Growth at 40(3
, O0
. ' 120
180
mlM
Growth at 37G
Fzg 2 Y~neucs of ~rowth of various dnals stFaln$ and their |ys0gcmc d¢nvabves, (A) CRT 46 25 and CRT 46 25 (Mu gem3), (F CR"I'266 and CRT266 (Mu gem3), (C) PC79 and PC79 (Mu get.3), (D) JWI]3 and JWII3 (Mu gem3)
139 Table 2 Bactcnal slram
GC2624/rsArs GC2624(Mugem3) AX655ftslts A X f 5 5 ( M u gem3)
GC2625flsZts CK: 2625 (Mu gem3)
Efficiency of plating 37°C/30° C
40°C/30°C
420C/300 C
1 6 × 1 0 ~2 1 4 × 1 0 -5 40×10 -I 1 l 4 x ] O -2 1
135
FEMSLE03789
Mu gem3 as a tool to investigate the influence of chromosome supercoiling on gene expression in Escherichia coli K12 E. Bianchi, I. Rubcrti, P. G h e l a r d i n i and L. Paolozzi 1 Cenffo Studx per el! AcJ~I Nuclewa del CNR. and I Dipartlmento dl Bsologsdl. Umverslld dl R o ~ "T~ Vergata~,Rome. Ira/y
Receivedl0 July 1989 Accepted 26 July 1989 Key words: Phage Mu; DNA superenlhng; Gene expression: ts Mutants
1. SUMMARY We have developed a rapid method to investigate the influence of chromosome supercoihng on gene expression m Eschenchra coh KI2. This method exploits the ability of the gem3 mutant of the bacteriophage Mu, even in the prophagtc state m immune cells, to mduce relaxaUon of the host chromosome. The expcnmants can thus be performed under physiological condnions, and without the use of the drugs. In theory, this method can be apphed to any bacteria! eerie. Here, we report the results obtained wnh four DNA rephcation and three cell di,asmn genes.
2. I N T R O D U C T I O N The degree of DNA supercoillng in E coh ts homeostatically regulated by the ¢quihbrtum between two tmzymauc activmes with opposing elloots: the DNA gyras¢ and ~ e DNA topo.somerase 1 [1,2 I.
Correspondence to" L Panloz~, Centre Stud1 per gh Aodl Nu¢lelct d¢l CNR, Dlpartanenlo dt Genetlca ¢ Btologla Molecolar¢, Umversa~dt Roma "La Saplenza', P~azzale Aide More 5. Reran, Italy
Both m wvo and in vitro studies indicate that the topological state of DNA plays an important role m several btolo~fal processes such as rephcaUon, recombination and gene expression (for reviews se¢ [1,2]). As far as this last process is concerned, random cloning of E. colt chromosomal DNA fragments into a promoter clomng vector has shown that the expression of about 40~ of the bactenal promoters was substantially modified after the mhibttton of DNA $yrase with coumermycm [3] Analogous results have been reported [4] for gene expression in Salmonella typtumurmm, using a techmque of in viva gene fusion by Mu-dlac. The use of mutants defootsve in either DNA gyrase (gyrA and g y r B ) or DNA topolsomerase (topA) genes and of specific inhibitors of DNA gyrasc has pcrmnted the identification of a number of genes which respond differentially to changes m DNA supereoding, such as the genes gyrA and g y r 8 [5] or the operons be1 [6] and histidine [7] The observauon that the product of the gene gem of bacteriophage Mu is able to modulate gene expression m E coh via an interaction with the host DNA gyrasc [g] allowed us to use the bacteriophage Mu as a reliable instrument to investigate the effect of DNA topology on the expressrun of specific bacterial genes. We have de-
0378-1097/89/$03 50 © 1989FedeTatlonof European MicrobsologlcalSocieties
136 veloped a method to study this effect on essential E colt genes exploiting the ability of Mu, even as a prophage, to cause a relaxation of the D N A and consequently to suppress ts mutations in genes activated by D N A relaxation. In fact, increased transcription results in a greater amount of the gene product and hence also of the amount of the ts protein statistically not-denatured at high temperature. Therefore, if the examined gene is positively regulated by D N A relaxation, the phenotype of a ts mutant in this gane should be more thermoreslstant when lysogenie for Mu than the non lysogen; m contrast, if the examined gene is positively regulated by D N A supercolhng, the expected result will be an increased thermosensitivlty.
3. MATERIALS A N D M E T H O D S 3.1. Bactertal stratus, phages and plasmM used m tMs work are listed in Table 1. 3 2. Me&a: LB and LB plates [9] were used for standard cultures and plating LB and LB plates without the addttton of NaCI ( L B ( - ) ) were also used for fts mutants. 3.3. Lysogemzatwn: Bactenai strains, grown at 3 0 ° C m LB to about l x l 0 ~ cells/nil, were lysogemzed with Mu c + gem3 as already described ]10]. 3.4. Eff|clency of plating was measured at vanous temperatures on bacterial strams grown as described above. 3.3. Kmencs of growth: Baetenal strains, grown as described above on LB or L B ( - ) , were shifted to the indicated temperature. Aliquots were Table l Bacterial stratus c6oo CRT 46 25 CRT 266 JW 113 PC 79 GC 2624 GC 2625 AX 655
Relevantgenotype thr, leg tht dnaA ts,~ dnaB Is:~s dnaE tssl t daaC ts /tsAts /tsZ tss, (tsl rs
Provtdedby L Paoloz~a R "thomas R Thomas R. Thomas A, Falasch R D'An R DAn R. D'An
withdrawn at various times, spread on LB plates and incubated at 30°C. During the growth at 4 0 ° C and 4 2 ° C of strain GC2625, lysogenic or not for Mu gem3, afiquots were withdrawn, observed at the optical microscope (300 x and 600 x ) and photographed with Kodak TMX100 fdm. 3.6. Plasm:dsupercodmg" Bacteiaal strains were transformed with pUC13 plasrmd D N A [11] as described [12]. Plasmid D N A was than extracted from each strain after overnight growth at 3 0 ° C and analyzed on agarose gel supplemented with cMoroquine (Sigma) as described [8].
4. RESULTS A N D DISCUSSION We have previously shown that Mu can suppress the effects of thermosens,:ttve mutations in the D N A ligase gene of E. cob via an interaction with the host enzymatte apparatus controlling the D N A topology. This suppression is greatest with the Mu gem3 mutant. In bactenal strains lysogenie for Mu gem3, we have also observed a substantaal increase (towards the relaxed state) in the average linking number of a resident plasoud [8]. The same effect in one of the stratus used in this study is shown m Fig. 1; m the lysogenie strain the average linloog number of pUC13 plasnud D N A is increased wtth respect to the non-lysogenic strain. 4.1. Effect of chromosome relaxatwn on DNA rephcaflon genes Thcrmosenstttve mutants m D N A replication genes dnad, dnaB, dnaC and dnaE were lysogenized with Mu gem3 and analyzed for thermosensitiwty at non-pcmussive or senu-permisstv¢ temperatures. Fig. 2 shows the kinetics of growth of the 4 mutants lysogenic or not for Mu gem3 at the most signihcattve temperature. The results mthcate that the thermoseusmvity produced by the ts mutatton m genes dnaA, dnaB and dnaE is partially suppressed by the presence of Mu gem3 a~ a prophage suggestmg that these genes could be posttively regulated by chromosome relaxation. On the other hand, dnaC gene express:on seems not to be affected by variation in D N A topology, since no
137
A
B
Fig l Supercolhngof pUCI3 plaslrad DNA isolated from CRC2624fts,4 strain lysogemcor not for Mu gem3 Electrophoresls was performed in the presence of O, 11 5 and 13 5/~8/ml of chloroqmne The Fig represL~tsIhe results of the agarosegel contmnmgIt 5 ~tg/'ml of chloroqmne and the analysis of Ihe topoasomerspaltern performed with an LKB Uhroscan (A) GC2624 (Mu gem3). (B) GC267A.
difference was observed between the lysogemc and the non-lysogenic strain at any temperature from 30 to 43°C. Ftg. 2C shows only one of the various growth kmettes obtained. It should be noted that the ts mutants in genes whose expression is stimulated by D N A relaxation are differentially suppressed by Mu gem3. Tiffs could be due to at least two not mutually exclusive reasons: (1) each promoter reaches mammal acuvauon at a specific superhehcal density [13] and Mu gem3 induces a modthcation m the average linking number that would probably not be the optimal for the different promoters of the exarmned genes; (2) the amount of remammg actxve protein necessary to assure the surv3val of the mutant could differ m the various ts stratus.
4.2. Effect of chromosome relaxation on cell dwsswn genes A number of cell dtvision genes, called fts, have been tdentifted m E colt by the selecuon of mutants carrying thermosensitive mutaUons which give rise to cell fdamentation at non permissive
temperature For our analysis, we have chosen three fts genes: ftsA, ftsl and ftsZ, that are clustered at 2 nun on the E. cob genetic map [14,15]. The temperature sensmve mutants in these genes were lysogemzed with Mu gem3 and analyzed for thetr thennosensitivity using different media since it was shown that the presence of salt m the growth medium is able to correct partially the ts phenotype of some fts mutants [16]. The results (Table 2) show that the presence of Mu gem3 renders the strata carrying the ftsA mutatton more thermosensitive than the nonlysogen. According to our hypothesis, this would suggest that the ftsA gene expression is activated by chromosome supercoifing. The thermosensittvity of both ftsl and ftsZ is instead partially corrected m the lysngenic strains. Analogous resuits have been obt='uned studying the ~neucs of growths at different non-pernusslve or semi-permtsstve temperatures (data not shown). In addttlon, the reduced thermosenslttvity m GC2625 strata lysogemc for Mu gem3 is also coupled to a reduction in the size of the fdaments produced by
138
'i
w.o
05 0
A
,
,
80
60
~
~
,
,
120
150
15
01
0
,
.
60
120
rains "-=-¢lr~A
rains
""-dr~A[l~ugam3)
"=-~B
ft.
"~-dr~B(t~utw,3]
Growth el 4 0 C
Growth at 37C N/NO
..j 180
C
w.o
D
0.1
0.01
(X6"
~0
0901 0
~r4
Growth at 40(3
, O0
. ' 120
180
mlM
Growth at 37G
Fzg 2 Y~neucs of ~rowth of various dnals stFaln$ and their |ys0gcmc d¢nvabves, (A) CRT 46 25 and CRT 46 25 (Mu gem3), (F CR"I'266 and CRT266 (Mu gem3), (C) PC79 and PC79 (Mu get.3), (D) JWI]3 and JWII3 (Mu gem3)
139 Table 2 Bactcnal slram
GC2624/rsArs GC2624(Mugem3) AX655ftslts A X f 5 5 ( M u gem3)
GC2625flsZts CK: 2625 (Mu gem3)
Efficiency of plating 37°C/30° C
40°C/30°C
420C/300 C
1 6 × 1 0 ~2 1 4 × 1 0 -5 40×10 -I 1 l 4 x ] O -2 1
