Download Phenotypic Stability of trp Operon Recombinant

Journal of Gemral Microbiology (1980), 118, 253-261.
Printed in Great Britain
253
Phenotypic Stability of trp Operon Recombinant Plasmids in
Escherichia coli
By T A D A Y U K I I M A N A K A , H I R O S H I T S U N E K A W A
AND S H U I C H I AIBA
Department of Fermentation Technology, Faculty of Engineering, Osaka
University, Yamada-kami, Suita-shi, Osaka, Japan
(Received 18 July 1979)
The recombinant plasmids RSF2124-trp and pSC101-trp were examined for their phenotypic stability in Escherichia coli W3110 and its derivatives under various culture conditions. RSF2124-trp and pSC101-trp were stable in a trpAEI strain. In an amber mutant
of the tryptophan repressor gene, RSF2124-trp was fairly stable, whereas pSC101-trp was
unstable. All Trp- segregants from the pSC101-trp carrier had lost the entire plasmid. In a
mutant carrying the tnaA mutation, RSF2124-trp was unstable in rich media. Most Trpsegregants that appeared under these conditions were deleted in trp genes as well as in the
cI gene on the recombinant plasmid. pSClO1-trp in this tnaA mutant was also unstable. All
Trp- segregants had lost the plasmid. Studies of enzyme activities revealed that the greater
the activity of anthranilate synthase and tryptophan synthase in bacteria, the more segregants tended to appear in the stability test. RSF2124 and PSClOl without the trp gene were
completely stable in the same bacteria. The apparent instability of bacteria carrying
the recombinant plasmid could be explained by the lower growth rate compared with
bacteria carrying only the vector plasmid, resulting in the enrichment of Trp- bacteria
during culture.
INTRODUCTION
Plasmids may serve as powerful tools in enhancing the expression of a specific enzyme
and/or metabolite in the cultivation of micro-organisms when combined with gene manipulation or engineering techniques. However, it is necessary to look closely at the stability
of the recombinant plasmids before they can be used satisfactorily. Stability here is defined
by the ability of transformed cells to maintain the plasmid unchanged during their growth,
manifesting the expected phenotypic characteristics.
Hershfield et aE. (1974), Cohen et al. (1977) and Hobom & Hogness (1974) recently
emphasized the significance of plasmid stability, not only from the genetic viewpoint, but
also from the aspect of gene manipulation. The stability of a plasmid may be affected by the
genetic characteristics of host cells, the COPY number of the plasmid, culture conditions and
the gene(s) carried on the plasmid.
As a model system we took the typical recombinant plasmids RSF2124-trp and pSC101trp, which both carry the same insert, having a whole trp operon, the PL promoter and CI
gene of phage A, but are maintained at different copy numbers in E. coli cells (Nagahari
et al., 1977). We have studied factors responsible for the stability and instability of these
plasmids.
0022-1287/80/oooO-8855 $02.00 @ 1980 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
254
T. I M A N A K A , H. T S U N E K A W A A N D S. A l B A
METHODS
Materials. Ampicillin (Ap), tetracycline (Tc), chloramphenicol, chorismic acid, agarose, lysozyme and
RNase A were purchased from Sigma; proriase E was from Kaken Chemical Co., Tokyo, Japan; restriction endonuclease EcoRI, DNA of phage h and T4 ligase were from Miles Laboratories, Elkhart,
Ind., U.S. A.; Bacto-tryptone was from Difco. All other reagents were from Wako Chemical Industries,
Osaka, Japan.
Media and solutions. PBB medium (Matsubara et al., 1975) contained 10 g peptone, 10 g bonito extract and
2.5 g NaCl in 1 1 deionized water; it was adjusted to pH 7.2 with NaOH and solidified with 15 g agar 1-l.
vogel& Bonner's minimal medium (VB) for E. coli was prepared as described by Smith & Yanofsky (1 962).
Minimalmedium(MM) was VB medium supplemented with glucose (2.5 g 1-l) and Casamino acids (0.5 g 1-l).
MM agar contained 15 g agar l-l, and Casamino acids were reduced to 50 mg 1-I. MM and MM agar
were also supplemented with amino acids (each 100 pg ml-l), thiamin (10 ,ug ml-l) and thymine (10 pg
ml-1) when required (see Table 1). Tryptone broth ('I
contained
)
10 g Bacto-tryptone and 2.5 g NaCl
per litre; it was adjusted to pH 7.0 with NaOH. T agar and T top agar contained 12 and 5 g agar l-l, respectively. TS buffer contained 0.02 M-Tris (pH 8.0) and 0.15 M-NaCl. TES buffer contained 0.02 M-TriS (pH
8.0), 0.01 M-N~,EDTAand 0.05 M-NaCI. Tris/borate buffer (Meyers et a[., 1976) contained 0.089 M-TriS,
0.089 M-boric acid and 2.5 ~ M - N ~ , E D T A .
Bacterial strains andplasmids. Strains and plasmids used in this work are listed in Table 1. Escherichia colj
K12 KM960 which required tryptophan for growth was obtained by mutation of the strain from Matsubara
et al. (1975) with N-methyl-N'-nitro-N-nitrosoguanidinetreatment (Adelberg et al., 1965). The two plasmids
pSC101-trp and RSF2124-trp were originally in E. coli C600-1.
Plasmid preparatiorr. Escherichia coli carrying RSF2124-trp was grown ovei night in PBB and diluted 10fold in 100 ml M-9 medium (Clowes & Hayes, 1968) supplemented with Casamino acids (20 g 1-lj and thiamin (1Opg ml-l) or thymine (1Opg ml-l), when required. The culture was grown for 2 h a t 37 "Cwith shaking.
Chloramphenicol was then added to a final concentration of 170pg ml-l, and incubation continued for 15 h.
After harvesting the cells, Sarkosyl lysate was prepared by the method of Clewell (1972). For the preparation
of pSClOl-trp, which cannot be amplified by chloramphenicol, the procedure of Tanaka et al. (1977) was
used.
Sarkosyl lysates and lysates obtained by the procedure of Tanaka et al. (1977) were centrifuged to equilibrium in a CsCl gradient containing ethidium bromide (Tanaka et al., 1977). The fractions containing covalently closed circular (CCC) DNA were pooled, treated three times with 1 vol. 1-butan01to remove ethidium
bromide, and dialysed against TS buffer.
For the preparation of RSF2124 and pSC101, the plasmid (RSF2124-trp or pSC101-trp) DNA, after dialysis against TS buffer, was digested with restriction endonuclease EcoRI (see below) and then heated at 65 " c
for 10 min to inactivate the EcoRI. This was followed by treatment at 4 "C overnight with T4 ligase by the
procedure of Tanaka & Weisblum (1975). Escherichia coli K12 KM960 Trp- was then transformed with the
DNA solution. After selection of drug-resistant transformants (see below), subculture on MM agar and on
MM plus Trp agar was used to confirm that the trp gene was absent from the transformants. The transformants were subjected either to the procedure of Clewell (1972) using chloramphenicol or to that of Tanaka &
Weisblum (1975) to obtain the plasmid DNA of RSF2124 or pSC101. After dialysis against TES buffer, this
was stored at 4 "C for several weeks.
Transformation of E. coli with plasmid DNA and selection of transformants. Transformation was carried out
by calcium chloride treatment, following the method of Tanaka & Weisblum (1975). Transformants of RSF2124 and RSF2124-trp were selected on PBB plus Ap (1OOpg ml-l) agar, while those of pSClOl and pSCto1trp were selected on PBB plus Tc (30 pg ml-l) agar.
Phenotypic stability of plasmids. Each bacterial strain carrying a plasmid was first isolated as a single
colony on PBB agar containing the drug to which the strain was resistant. Then a fresh single colony
(not more than 1 d-old) was inoculated into 20 ml PBB plus drug as a preculture in a 100 ml shake flask and
incubated at 37 "C until late-exponential phase.
Transformants of E. coli W3110 trpAEf trpRam27 were precultured successively on MM plus drug agar
and in MM plus drug, because the phenotypic loss of selective markers in PBB was appreciable. MM plus
drug was also used for preculture of E. coli W3110 trpAEl trpR tnaA carrying RSF2124-trp or pSC101-trp.
The preculture was diluted with fresh MM and inoculated into various media at about 50 cells ml-l. This
small inoculum was necessary to observe plasmid stability after numerous cell divisions. The culture media
used were PBB, MM, and MM plus tryptophan (50 pg ml-I). After about 25 generations in these media, the
culture was diluted and samples were plated on PBB agar. One hundred colonies were transferred by replica
plating on to several media (MM, MM plus tryptophan, PBB plus drug, and PBB) to determine drug
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
255
Ph en0 typic stability of recombinant plasm ids
Table 1. Strains of Escherichia coli and plasmids used
E. coli strain
C600-1
K12 KM960
W3110
W3110
W3110
Relevant properties
Source/reference
rg mg leid thrl thil supE44
lacy1 tonA2l Trpstr his recA I su thy
(gal atth) d e z TrptrpAEI *t
trpAEI trpRam27*$
trpAEl trpR tnaA*§
K. Sakaguchi
(Nagahari et al., 1977)
This work
Plasmid
Phenotype
pSC101-trp
Tc' immh Trp+
RSF2 124-trp
Apr immh Trp-t immColEl
pSCl0l
RSF2 124
Tc'
Apr ColEl immColEl
F. Imamoto
F. Imamoto
F. Imamoto
Source/reference
K. Sakaguchi
(Nagahari et a/., 1977)
K. Sakaguchi
(Nagahari et a/., 1977)
This work
This work
*
Isolated originally by C. Yanofsky.
Deletion mutant of trpA-E, designated AEl.
$. trp repressor amber mutant, designated Ram.
5 Tryptophanase-deficient mutant, designated Tna.
resistance and the Trp+ phenotype. Concentrations of Tc and Ap used were 30 pg ml-l and 100 pg ml-l,
respectively.
Digestion of plasmid D N A with EcoRI. Plasmid DNA in TS buffer was digested at 37 "C for 1 h with
restriction endonuclease EcoRI by the method of Greene et al. (1974). The reaction mixture (26 1.1) contained
0.1 M-Tris/HCl (pH 7.5),0.05 M-NaCI, 0.01 M-MgCl,, about 1 pg DNA and 10 units EcoRI.
Agarose gel electrophoresis of DNA. DNA samples were subjected to electrophoresis in agarose (10 g I-l)
dissolved in Tris,/borate buffer. A dye solution ( 5 PI) (Meyers et al., 1976) composed of bromophenol blue
(0.7 mg ml-l), sodium dodecyl sulphate (70 mg ml-l) and glycerol (330 mg ml-l in water) was added to each
DNA sample (10 to 40 p1) prior to electrophoresis i n a vertical slab gel apparatus. h DNA digested with
restriction endonuclease EcoRI was used as an internal reference.
Phage preparation and immh test. hc190 (Kaiser, 1957), h uzuIu3(Ptashne & Hopkins, 1968) and himm21
(Liedke-Kulke & Kaiser, 1967), all from Dr F. Imamoto, were purified by the procedure of Doerfler &
Hogness (1968). Sensitivity to infection by these phages was assessed by the cross test on tryptone agar.
Escherichia coli KM605 carrying hdvl, from Prof. K. Matsubara (Matsubara & Kaiser, 1968), was used as a
reference.
ColEZ immunity test. Colicin E l immunity was tested by the method of Ozeki (1968). Escherichia coli
A745 from Ogawa (Sakakibara & Tomizawa, 1974) was used to produce colicin El.
Preparation of crude extract and assay of anthranilate synthase and tryptophan synthase activities. Bacteria
were grown until the late-exponential phase at 37 "C in 100 ml MM. Harvested cells were washed with cold
NaCl solution (0.9 ",, w/v), resuspended in 5 ml of 0.1 M-Tris/HCl (pH 7.8), and sonicated three times for 1
min (ultrasonic generator, model 4280, Kaijo Denki Co., Tokyo, Japan) at 10 kHz in ice/water. Cell debris
was removed by centrifuging at 18000 rev. min-l for 30 min (Sorval SS-34 rotor) and the supernatant was used
as crude extract.
Anthranilate synthase (EC 4 . 1 .3.27) activity in the crude extract was assayed by the method of Matsushiro
(1965) and tryptophan synthase (EC 4.2.1.20) activity by the method of Smith & Yanofsky (1962). The
protein content was measured by the method of Lowry. In parallel with the assay of anthranilate synthase and
tryptophan synthase activities, the phenotypic stability of RSF2124-trp or pSC101-trp was assessed. The
specificactivities of anthranilate synthase and tryptophan synthase were corrected by dividing by the fraction
of Trp+ cells in the stability test.
Assay of tryptophan. For the assay of intracellular tryptophan, harvested cells were washed with cold 0.9 74
NaCl, and amino acids (including tryptophan) were extracted with boiling water for 10 min. Extract or culture broth (1 ml) was well mixed three times with 6 ml toluene to remove indole. Tryptophan in the water
layer was assayed with xanthydrol as described by Dickman & Crockett (1956).
MIC
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
118
256
T. I M A N A K A , H. T S U N E K A W A A N D S. A I B A
Table 2. Stability of plasmid RSF2124-trp in mutants of E. coli W3110
Host cell
W3110 trpAEI
W3110 frpAEl frpRam27
W3110 trpAEI trpR tnaA
No. of
generations
Medium
Precult ure
(PBB Ap)
PBB
MM
MM +Trp
Precul t ure
(MM + AP)
PBB
MM
MM+Trp
Preculture
(MM + AP)
PBB
MM
MM+Trp
+
26.0
24.1
23.8
-
25.1
22.6
23.5
-
26-1
22.2
23-3
-
Phenotype of colonies
Trp +
AP'
(7;)
Trp+
AP*
(?:I )
7
TrpAP'
(9b)
TrpAP'
(OO)
100
96
1 00
0
100
0
0
0
0
0
0
0
0
4
0
0
100
94
98
100
0
0
0
0
0
2
0
0
0
4
2
0
40
7
99
0
0
0
0
46
93
1
100
14
0
0
0
0
KESULTS
Transformat ion with recombinant plasm id
Purified recombinant plasmid DNA obtained from E. coli C600-1 was introduced into
E. coli K12 KM960. The efficiency of transformation was very low (results not shown)
because the plasmid DNA was not modified in the former host strain E. coli C600-1 rkmG.
To increase the transformation efficiency, E. coli K12 KM960 Trp- was first transformed
to modify the plasmid DNA and then the recombinant plasmid prepared from the transformant was introduced into other strains. The transformation was about 1000 times more
efficient than that of plasmid from the modification-deficient strain of E. coli C600-1
Stability of recombinant plasmih in tryptophan repressor mutants and in a
tryptophanase mutant
The phenotypic stability of RSF2 124-trp was tested in E. coli W3 110 trpAE2 (subsequently
designated AEl) under various culture conditions, and in the presence or absence of
trpR and tnaA mutations. The results are shown in Table 2. Only MM plus Ap was used in
the preculture of strains W3 1 10 trpAE2 trpRam27 (designated Ram) and W3 110 trpAEl trpR
tnaA (designated Tna) to eliminate the segregants. If PBB plus Ap was used in the preculture,
the fraction of Trp+ Ap' cells was only 13yo for Ram and 0 yofor Tna. No such segregation
was observed with AEl. This suggests that RSF2I24-trp is unstable in both Ram and
Tna mutants, but not in AEl .
RSF2124-trp was stable in AEI , but some segregants appeared in the Ram mutant after
about 25 generations. The plasmid was quite unstable in the Tna mutant in the presence of
tryptophan. The percentage of Trp+ Ap' is shown in Fig. 1. (Other phenotypic expressions
in Table 2 are discussed later.) The phenotypic stability of pSC101-trp was also tested in
strains AEl, Ram and Tna (Fig. 1). With pSClOl-trp, the initial loss of genetic markers
(Trp+ Tcr) during preculture was not observed. The histograms of the stability of pSC101trp (Trp+ T c ~ )shown in Fig. 1 were obtained from similar results to those in Table 2.
pSC1Ol-trp was stable in AEl, the percentage of Trp+ Tcr was considerably reduced in Ram
and it was quite unstable in Tna.
Specific activities of anthranilate synthase and tryptophan synthase were measured for
the three mutants of E. coli W3110 carrying either RSF2124-trp or pSC101-trp (Table 3). In
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
257
Phenotypic stability of recombinant plasmids
100
AE I
Tna
Ram
r
a
a2
L
1
+
+
E
E
2
E
E
E
Growth medium
Fig. 1. Stability of RSF2124-trp and pSC101-trp in mutants of E. coli W3110: trpAEl (AE1);
trpAEl trpRam27 (Ram); trpAEl trpR tnaA !Tna). For descriptions of media, see Methods.
0,
pSC101-trp (Tc'); H, RSF2124-trp (Ap').
Table 3. Anthranilate synthase and tryptophan synthase activities in mutants of
E. coli W3110 carrying RSF2124-trp or pSC101-trp grown in MA4
Enzyme activities are expressed in units (mg protein)-l.
Plasmid carried
RSF2 1 24-trp
,
pSC101-trp
Host cell
Anthranilate
synthase
Tryptophan
synthase
Anthranilate
synthase
Tryptophan
syn thase
W3110 trpAE1
W3110 trpAEl trpRam27
W3110 trpAEl trpR tnaA
0.19
0.63
0.57
11
27
27
0.61
25
100
153
4-30
6.14
Ram and Tna carrying RSF2124-trp, these activities were both about three times those in
AEl, possibly because the first two strains lack repressors. The fact that the relative magnitude of the anthranilate synthase and tryptophan synthase activities for each strain in
Table 3 was almost the same for the three mutants might be consistent with normal expression of the trp operon by transcription from the main trp promoter.
Since the activities of tryptophan synthase assessed here were less variable than those of
anthranilate synthase, the former was used to represent trp operon expression. For the
enzyme assays, cells were grown in MM. In this medium, RSF2124-trp remains fairly stable,
irrespective of whether the plasmid is maintained in AEl, Ram or Tna (Fig. 1). However,
this stability is only apparent since only Trp+ cells are screened in MM.
Tryptophan synthase activity may not exceed 1 I units (mg protein)-l (Table 3) because of
the presence of a repressor in AE1 when cultured in rich media such as PBB or MM plus
Trp. This low tryptophan synthase activity is correlated with the stability of the RSF2 124trp plasmid in any of the three media (see below). In mutants deficient in the production of
17-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
258
T. I M A N A K A , H. T S U N E K A W A A N D S. A I B A
active trp repressor, the tryptophan synthase activity was 27 units (mg protein)-l, i.e. about
2.5 times that in the strain containing the repressor. This increase may be correlated with the
instability of the RSF2124-trp plasmid, since such increased protein synthesis is likely to
exert considerable stress on cell metabolism in terms of growth (see below).
The stability of pSC101-trp (Fig. 1) is consistent with the notion that the greater the
activity of anthranilate synthase and tryptophan synthase in cells, the more segregants
tended to appear. Activities of anthranilate synthase and tryptophan synthase in Ram and
Tna cells containing pSClOl-trp were higher than those in RSF2124-trp carriers (Table 3).
The above observation might appear to contradict the findings of previous workers
(Nagahari et al., 1977) that the copy number of RSF2124-trp per chromosome (11.9) is
about three times that of pSC101-trp. However, RSF2124-trp frequently exhibited the
deletion of the trp operon, whereas no such deletions were observed in pSCIO1-trp (see
below). Thus, the number of trp operons in RSF2124-trp carriers might be less than the
copy number of the plasmid in pSC101-trp carriers, which could account for the discrepancy.
Segregants from tryptophan repressor mutants
Phenotypic expression of segregants that appeared in RSF2 124-trp carriers in E. coli
W3110 trpAE1 and its derivatives are shown in Table 2. No Trp+ ApS segregants were detected. Most of the segregants deriving from AEI and Ram were Trp- ApS. Such cells have
lost the plasmid, as no CCC DNA was found in CsCl/ethidium bromide density gradient
centrifugations.
In contrast, CCC plasmid DNA was detected in Trp- Apr cells that segregated from Ram
and Tna. These plasmid DNAs were digested with restriction endonuclease EcoRl and examined by agarose gel electrophoresis using DNAs of phage h and RSF2124-trp as references. The finding of two fragments from RSF2124-trp corresponding to 10.8 x lo6 and
7.2 x lo6 daltons and six bands from DNAs of phage h (results not shown) were consistent
with previous observations (Hershfield et al., 1974; Nagahari et al., 1977). On the other
hand, plasmid DNAs extracted from the Trp- Apr segregants always had only one EcoRI
cleavage site. The molecular masses of these plasmids ranged from 4 x lo6 to 8 x 106daltons,
sizes which were equal to or smaller than the 7.2 x lo6daltons of RSF2124. All Trp- Apr segregants had lost immunity to A, and hence the cZ gene of phage A. These observations show that
the Trp- Apr segregants contain plasmids deleted in the region covering the cZ gene in the
recombinant plasmid. The deletion plasmids were subjected to the ColE 1 immunity test. All
were Apr immColEl+. Hershfield et al. (1974) considered Trp- cells to result from deletion
of a DNA fragment or the loss of ColEl-trp plasmid from an E. coli C600 trpR rec strain.
Though not shown in Table 2, segregants from pSC101-trp carriers of the three bacterial
strains were all Trp- TcS. Here again, in CsCl/ethidium bromide density gradient centrifugations no plasmid CCC DNA was detected in the Trp- TcS segregants; they must,
therefore, have lost the plasmid.
Stability of RSF2124 and pSClOl
RSF2124 and pSClOl without the trp operon were very stable in each of the strains used
in this work. Another plasmid obtained from the digestion of RSF2124-trp with restriction
endonuclease HindllI, of phenotype imrnh+Trp- Apr immColE1 +, was also quite stable in
these strains (results not shown); i.e. it is unlikely that the CZ gene of phage h affects the
stability of RSF2124-trp and pSC101-trp. The instability observed with RSF2124-trp and
pSClO1-trp might therefore be due to the presence of the trp operon in each of the recombinant plasmids.
Growth rates of cells carrying either the Trp + recombinant plasmid or the vector plasmid
Growth rates of cells carrying either the Trp+ recombinant plasmid or the vector plasmid
were examined in PBB at 37 "C, following the procedure used previously in the stability test
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
Phenotypic stability of recombinant plasmids
259
Table 4. Relative growth rates of E. coli W3110 mutants carrying
RSF2124-trp or pSC101-trp grown in PBB
The growth rates of strains carrying a recombinant plasmid were compared with the growth rates of
the corresponding strains carrying the vector plasmid. Strain designations: AE1, W3110 trpAEZ;
Ram, W3110 trpAEl tupRam27; Tna, W3110 trpAEZ trpR triaA.
Strain with recombinant plasmid
Strain with vector plasmid
AEl (RSF2124)
AEl (pSC101)
Ram(RSF2124)
Ram(pSC101)
Tna(RSF2124)
Tna(pSC101)
N
N
N
>
>
>
AE1 (RSF2124-trp)
AE1(pSC101-trp)
Ram(RSF2124-trp)
Ram(pSC101-trp)
Tna(RSF2124-trp)
Tna(pSC1Ol-trp)
N , Grows at a similar rate. > , Grows at a faster rate.
except for an increase in inoculum size (A,,, about 0.05). When Asg0 values for samples
taken at 0-5 h intervals were plotted semi-logarithmically against the time of cultivation ( t ) ,
the growth rate apparently decreased during culture even in the early phase of growth
( t = 0 to 2 h). This particular feature might have been due to the nature of PBB. It was
difficult to determine the growth rate precisely in each run (generation times extended from
0.7 to 1 h). However, when the growth rates of plasmid-carrying cells with and without the
trp operon were compared (As9,0-1 to 0.3) reliable estimates of relative rates were obtained
(Table 4). Growth rates were almost the same irrespective of the presence or absence of the
trp operon when the tryptophan synthase activities of the relevant cells were low (Tables 3
and 4). The disparities in growth rate shown in Table 4 were commensurate, by and large,
with greater tryptophan synthase activity. For example, the growth rate of Tna(pSC101)
was about 20y0 higher than that of Tna(pSC101-trp).
Extra- and intracellular concentrations of tryptophan
The extracellular concentration of tryptophan in MM remained almost unchanged in all
of the cultures, being 2 to 4 ,ug ml-l. In addition, the intracellular concentration of tryptophan in Tna was nearly the same as that in AEI or Ram [about 1 pg (mg dry wt cells)-l].
These results indicate that tryptophan biosynthesis is not actually stimulated in Tna due
to product inhibition even if the levels of the trp operon enzymes are raised.
DISCUSSION
Stability or instability of a plasmid in host cells may be affected by several factors, such as
growth rate, characteristics of the plasmid (copy number, relaxed or stringent type of
replication control, and physiological consequences of expression of genes on the plasmid) or
environmental stress imposed on the host cells. We have shown here that the recombinant
plasmids carrying the trp operon became unstable when tryptophan synthase levels were
raised. Overproduction of trp operon enzymes might have adversely affected the normal
metabolism and growth of the host cells. Tn fact, the stress caused by the overproduction of
these enzymes has resulted in the disappearance of the entire recombinant plasmids or
deletion of the trp operon from them. Several workers have reported differences in growth
rate between cells with and without a plasmid (Collins & Pritchard, 1973; Engberg &
Nordstrom, 1975); in the course of this work, we noticed the appearance of two types of
colonies, large and small in diameter. The large colonies did not carry the recombinant
plasmid whereas the small ones did, an observation which agrees with our previous
hypothesis.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
260
T. I M A N A K A , H. T S U N E K A W A A N D S. A I B A
In the case of high tryptophan synthase activity, Trp+ transformants grew more slowly in
PBB than Trp- cells carrying only the vector plasmid. Consequently, higher levels of trp
operon enzymes might have resulted in lower growth rate and thus yielded the relative
increase of Trp- cells during culture. The percentage of Trp+ Tcr in Ram(pSClO1-trp)
and Tna(pSClO1-trp) was low even in MM (Fig. 1). Although in this case, cells deprived of
the trp genes could not grow, extraordinarily high tryptophan synthase activities in Ram(pSC10l-trp) and Tna(pSClO1-trp) in MM (see Table 3) must have resulted in cross-feeding
of cells lacking trp genes with tryptophan and the faster growth rate of Trp- cells might have
enriched the Trp- segregants.
The fact that the growth rate of Tna(RSF2124) was faster than that of Tna(RSF2124-trp)
in PBB in spite of similar growth rates of Ram with and without the trp operon on the
plasmid (Table 4) might have been due to excessive stress imposed by the tnaA mutation on
cell metabolism. Accordingly, the marked instability of RSF2 124-trp in Tna cells when
cultured in media containing tryptophan (see Fig. 1) could be explained by the enrichment
of Trp- cells in the stability test.
We have seen the formation of deletions covering the trp genes in RSF2124-trp in Tna
cells. The appearance of deletion plasmids from RSF2124-trp in Tna cells could be attributed to the presence of a transposon (TnA) in RSF2124, as suggested by Holmans et al.
(1978) who proposed that a deletion occurs from a terminal of TnA opposite to the ori gene
on the plasmid. More detailed studies of the deletion patterns occurring in various strains of
E. coli are now in progress.
The authors are indebted to Professor K. Matsubara, Molecular Genetics Establishment,
Osaka University School of Medicine, for his critical review of the manuscript and valuable
suggestions, to Drs H. Ogawa, T. Ogawa and F. Imamoto of Osaka University, for the
provision of useful strains, bacteriophages and valuable suggestions, and to D r K. Sakaguchi,
Laboratory of Microbial Biochemistry, Mitsubishi-Kasei Institute of Life Sciences,
Machida, Tokyo, for his cooperation and provision of composite plasmids. Technical
assistance and valuable discussions with M. Tateishi and T. Tanaka, Department of Fermentation Technology are appreciated. This work was partly supported by Grant-in-Aid
from the Education Ministry, Japanese Government.
REFERENCES
ADELBERG,
E. A., MANDEL,M. & CHEN,G. G. C.
(1965). Optimal conditions for mutagenesis by
N-methyl-N '-ni t ro- N-ni t rosoguanidine in Escherichia coli K12. Biochemical and Biophysical Research Communications 18, 788-795.
CLEWELL,
D. B. (1972). Nature of ColEl plasmid
replication in Escherichia coli in the presence of
chloramphenicol. Joirrnul of Bacteriology 110,
667-676.
CLOWES,R. C. & HAYES,
W. (1968). In Experiments
in Microbial Genetics, p. 187. New York: John
Wiley.
F., CHANG,A. C. Y . &
COHEN,S . N., CABELLO,
TIMMIS,
K. (1977). DNA cloning as a tool for the
study of plasmid biology. In Recombinant Molecules: Impact on Science and Society, pp. 91--105.
Edited by R. F. Beers, Jr & E. G . Bassett. New
York: Raven Press.
COLLINS,J. & PRITCHARD,
R. H. (1973). Relationship between chromosome replication and F'lac
episome replication in Escherichia coli. Juurnal of
Molecirlar Biology 78, 143-1 55.
DICKMAN,
S.R. & CROCKETT,A. L. (1956). Reactions of xanthydrol. 1V. Determination of
tryptophan in blood plasma and in proteins.
Journal of Biological Chemistry 220, 957-965.
DOERFLER,
W. & HOGNESS,D . S. (1968). The strands
of DNA from lambda and related bacteriophages:
isolation and characterization. Joirriinl of Molecular Biology 33, 635-659.
ENGBERG,
B. & r \ l T ~ ~K.~(1 975).
~ ~ Replication
~ O ~ ,
of R-factor R1 in Escherichia coli K-12 at different growth rates. Journal of Bacteriolugy 123,
179-1 86.
GREENE,P. J . , BETLACH,M. C., EOYER,H. W. &
GOODMAN,
H. M. (1974). The EcoRI restriction
endonuclease. Methods in Molecular Biology 7, 87111.
HERSHFIELD,
V., EOYER,H. W., YANOFSKY,
C.,
LOVETT,M. A. & HELINSKI,
D. R. (1974). Plasmid
ColEl as a molecular vehicle for cloning and amplification of DNA. Proceedings of the Nationul
Academy of Sciences of the United States of
America 71, 3455-3459.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34
Phenotypic stability of recombinant plasmids
HOBOM,
G. & HOGNESS,
D. S. (1974). The role of
recombination in the formation of circular oligomers of the hdvl plasmid. Journal of Molecular
Biology 88, 65-87.
HOLMANS,
P., ANDERSON,
G. C. & CLOWES,R. C.
(1978). TnA-directed deletions and translocations
within the R6K plasmid. In Microbiology-1978,
pp. 38-41. Edited by D. Schlessinger. Washington,
D.C. : American Society for Microbiology.
KAISER,A. D. (1957). Mutations in a temperate
bacteriophage affecting its ability to lysogenize
Escherichia coli. Virology 3, 42-61.
LIEDKE-KULKE,
M. & KAISER,A. D. (1967). Genetic
control of prophage insertion specificity in bacteriophages h and 21. Virology 32, 465-474.
MATSUBARA,
K. & KAISER,A . D . (1968). hdv: an
autonomously replicating DNA fragment. Cold
Spring Harbor Symposia on Quantitative Biology
33, 769-775.
MATSUBARA,
K., TAKAGI,Y. & MUKAI,T. (1975).
In vitro construction of different oligomeric forms
of hdv DNA and studies on their transforming
activities. Journal of Virology 16, 479-485.
MATSUSHIRO,
A. (1965). On the transcription of the
tryptophan operon in Escherichia coli. Journal of
Molecular Biology 11, 54-63.
MEYERS,J. A., SANCHEZ,D., ELWELL,L. P. &
FALKOW,
S. (1976). Simple agarose gel electrophoretic method for the identification and charac-
26 1
terization of plasmid deoxyribonucleic acid.
Journal of Bacteriology 127, 1529-1 537.
NAGAHARI,K., TANAKA,T., HISHINUMA,F.,
KURODA,M. & SAKAGUCHI,
K. (1977). Control of
tryptophan synthetase amplified by varying the
numbers of composite plasmids in Escherichia coli
cells. Gene 1, 141-152.
OZEKI,H. (1968). Methods for the study of colicine
and colicinogeny. Methods in Virology 4, 565-592.
PTASHNE,
M. & HOPKINS,
N. (1968). The operators
controlled by the h phage repressor. Proceedings of
the National Academy of Sciences of the United
States of America 60, 1282-1287.
SAKAKIBARA,
Y. & TOMIZAWA,
J. (1974). Replication
of colicin El plasmid DNA in cell extracts. Proceedings of the National Academy of Sciences of
the United States of America 71, 802-806.
SMITH,D. H. & YANOFSKY,
C. (1962). Enzymes involved in the biosynthesis of tryptophan. Methods
in Enzymology 5 , 794-806.
TANAKA,
T. & WEISBLUM,
B. (1975). Construction of
a colicin El-R factor composite plasmid in vitro:
means for amplification of deoxyribonucleic acid.
Journal of Bacteriology 121, 345-362.
TANAKA,
T., KURODA,M. & SAKAGUCHI,
K. (1977).
Isolation and characterization of four plasmids
from Bacillus subtilis. Journal of Bacteriology 129,
1487-1494.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 16:39:34