Previous Article | Next Article 
J Bacteriol, February 1998, p. 655-659, Vol. 180, No. 3
Biology Department, Concordia University,
Montreal, Quebec H3G 1M8, Canada
Received 4 September 1997/Accepted 17 November 1997
Transcription of the Escherichia coli genes
serA and gltBDF depends on the
leucine-responsive regulatory protein, Lrp, and is very much decreased
in an lrp mutant. By the use of an Lrp-deficient host and
the lrp gene cloned under a plasmid-borne arabinose pBAD promoter, we varied the amount of Lrp present in the cell and showed
that both genes were transcribed in proportion to the amount of Lrp
synthesized. The affinity of serA for Lrp was four to five times greater than the affinity of gltD. Overproduction of
Lrp was lethal to the cell.
The conformation of DNA in the cell
results from the interaction of many cytoplasmic factors, including the
many DNA-binding proteins which affect its conformation. Some of these,
like H-NS and HU, bind with relatively low sequence specificity to many sites and are considered to be structural proteins setting the overall
DNA conformation (8). Others, like the lac
repressor, bind at one or very few, well-defined sites and are
considered to be regulatory proteins. Still others, the global
regulatory proteins, bind with high specificity but to many sites,
usually regulating expression of a number of genes whose products have a related metabolic function (10).
Leucine-responsive regulatory protein (Lrp), a 19-kDa protein binding
cooperatively as a dimer (28), is usually considered to be
one of the global regulatory proteins (3, 20, 21). However,
it has proved difficult to define a short, clear binding site for Lrp
(7). A 15-bp binding sequence has resulted from a Selex
search (6), but its biological significance is uncertain, the more so since Lrp protects sites of 80 to 100 bp against DNase I
digestion.
Lrp has a strong preference for AT-rich sequences and may even
recognize a DNA structure resulting from such sequences. We therefore
suggested that Lrp might not be a regulatory protein in the usual sense
but might determine DNA conformation. This would be consistent with
previous in vitro studies which showed great differences in expression
of a variety of seemingly functionally unrelated genes in
lrp mutants compared to lrp+ parents.
In this study, we examined the effects of in vivo variation of Lrp
concentration and showed that expression of two genes varies with the
Lrp concentration, each gene showing its own characteristic affinity.
This shows that the expression of various genes can be expected to vary
when Lrp concentration varies but does not clearly differentiate
between the structural and regulatory roles of Lrp.
Plan of experiments.
Expression of chromosomal
serA::lacZ and
gltD::lacZ fusions was determined by
Growth conditions, chemicals, bacterial strains, and
plasmids.
The Escherichia coli K-12 strains and
plasmids used in this study are listed in Table
1. Minimal medium and growth conditions are as previously described (1, 26). Plasmids were isolated according to Maniatis et al. (18).
0021-9193/98/$04.00+0
Copyright © 1998, American Society for Microbiology. All rights reserved.
Comparison of the Sensitivities of Two
Escherichia coli Genes to In Vivo Variation of Lrp
Concentration
ABSTRACT
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
INTRODUCTION
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
MATERIALS AND METHODS
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
-galactosidase assay of cells carrying the lrp gene on
the arabinose-controlled pBAD22 vector and grown with a variety of
arabinose concentrations. The arabinose levels were expressed in Lrp
units by comparison with expression of pBAD22lacZ in cells
grown in the same circumstances. Host cells carried a Tn10
insertion in lrp and a deletion of the entire ara
operon and were thus Lrp deficient and unable to degrade arabinose.
TABLE 1.
E. coli strains and plasmids used
Enzyme assay.
-Galactosidase activity was assayed in
whole cells according to the method of Miller and expressed in Miller
units (19).
Construction of arabinose-nonutilizing cells.
Strains were
made arabinose deficient in two steps. First a requirement for leucine
was transduced to the strain by using phage grown on
CP55leu::
placMu cells
(15). Leucine-requiring transductants were then transduced
to leucine independence by using phage grown on strain MEW308
(
ara) and screened for the inability to grow with
arabinose as a carbon source. This was not straightforward because some
stocks of the known ara strains carry a cryptic mutation
resulting in filamentation under some experimental conditions. We
therefore first separated the ara and filamentation
mutations, by transducing CP55 to leucine independence and selecting a
nonfilamenting ara strain as donor. We verified that this
strain was not inhibited by addition of arabinose to glycerol-grown
cultures.
Problems with plasmid maintenance.
The experiments described
here were carried out with a variant of pBAD22 carrying the
chloramphenicol resistance gene. Analogous experiments, not presented
here, with pBAD18 and pBAD22 carrying the -lactamase gene and with
ampicillin to ensure plasmid maintenance were not sufficiently
reproducible due to rapid breakdown of ampicillin and loss of plasmid
from a large proportion of cells.
Tests of plasmid maintenance.
Approximately 500 cells from
several cultures from each experiment were taken at the time of
-galactosidase assay, plated on LB, and replicated on LB and LB with
chloramphenicol (25 µg/ml). Experiments presented here showed more
than 90% plasmid retention.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Effect of variation in Lrp concentration on expression from
serA::lacZ.
Expression of
serA is known to be activated by Lrp and decreased in the
presence of leucine, so that the SerA activity of an Lrp-deficient
mutant is much lower than that of its lrp+
parent. We show here that this is also true for the ara
strains used in this work (Table 2).
|
-galactosidase (from serA)
was expressed even in the absence of arabinose but increased markedly with arabinose (0 to 30 µg/ml) and was not further increased by arabinose levels up to fourfold higher (Fig.
1A, curve 1). This is consistent with the
facts that the lrp mutant is not a serine auxotroph
(14) and that the serA gene has two promoters,
one of which, P1, is strongly induced by Lrp (16).
|
) and with both
L-serine and L-leucine (
); (B) strain MEW309
(lrp/pBADlrp::lacZ) grown
and assayed as described above in glycerol minimal medium with no
addition (
) and addition of L-leucine (
),
L-serine (-galactosidase activity with 10 µg of
arabinose/ml (Fig. 1B, curve 1). Addition of L-serine
decreased expression from the vector drastically
to 675 U at 10 µg/ml and 7,300 U at 70 µg/ml (Fig. 1B, curve 2). Addition
of L-leucine had very little effect at low arabinose concentrations but inhibited strongly at higher arabinose levels (Fig.
1B, curves 3 and 4).
This effect of L-serine on expression from the arabinose
promoter is interesting in itself and is so drastic that the curves of
Fig. 1A cannot be understood without correction. We therefore replotted
the curves of Fig. 1A against the appropriate curves of Fig. 1B and
show the results in Fig. 1C. It is clear that serA is
induced proportionally with Lrp in a curve that saturates at an Lrp
equivalent around 3,000 -galactosidase units. Leucine inhibited
expression at all levels testedaround 50% with no arabinose and over
70% at arabinose concentrations resulting in maximum expression.
Effect of variation in Lrp concentration on expression from gltD::lacZ. Lrp not only activates transcription of many genes but is indispensable for transcription of several, gcv and gltD included (9, 15), as is confirmed for the strains used here (Table 2). However, nothing is known about the relative sensitivities of Lrp-regulated promoters to Lrp.
We therefore carried out the same type of experiment as described above on a gene for which Lrp is indispensable, gltD. Expression of chromosomal gltD::lacZ increased similarly in proportion to arabinose from 0 to 20 µg/ml and also, with lower sensitivity, from 40 to 100 µg/ml (Fig. 2A, curve 1). To allow a direct comparison with the results from the serA::lacZ fusion, we repeated this experiment in the presence of L-serine. gltD expression was very much reduced (Fig. 2A, curve 2), as would be expected from the effect of L-serine on the vector. L-Leucine inhibited expression at all arabinose concentrations (Fig. 2A, curve 3).
|
Direct comparison of regulation of serA and gltD. The maximum expression of serA is much higher (900 U) than that of gltD (150 U) and is reached at a much lower Lrp level. To compare these in what may be a biologically more meaningful way, we expressed the serA::lacZ and gltD::lacZ data as percentages of the maximal induction seen (Fig. 3). This is not completely satisfactory for serA, since the level of expression is a result of increased use of P1 and decreased use of P2. In Fig. 3, curve 1, we plot the data for serA::lacZ with basal activity subtracted. In any case, this figure shows that serA is fully induced at much lower Lrp concentrations than is gltD.
|
Toxicity of Lrp to E. coli K-12.
It was reported
earlier that Lrp in high concentration inhibits growth of E. coli (2). Taking advantage of the wide range of
concentrations available from pBADara, we show that Lrp at high concentration is in fact lethal (Table
3). We grew strain MEW308
(ara/pBADlrp) in glycerol minimal medium,
added arabinose (100 µg/ml), and followed turbidity, viable count,
and plasmid retention. The culture increased somewhat in turbidity for
about 4 h after arabinose was added. Turbidity remained roughly
constant thereafter, but cells died rapidly, as judged by their ability to form colonies on LB. No significant plasmid loss was seen. We have
isolated a variety of mutants resistant to Lrp and are currently
characterizing them.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
In this study, we investigated the effect of changes in Lrp concentration on the expression of gltD and serA, two E. coli genes that require Lrp for transcription. By cloning the lrp gene under the control of the arabinose pBAD promoter, we made Lrp concentration dependent on the quantity of arabinose provided, using lacZ in a pBAD22lacZ fusion as a reporter gene to estimate the amount of Lrp produced.
Physiological implications of relative affinities of serA and gltD promoters for Lrp. Expression of serA was much more sensitive to Lrp than expression of gltD (Fig. 3). Half-maximal induction of serA was seen at 1,000 U of pBADlacZ compared to the more than 4,000 U needed for the same proportion of gltD expression. This means that at least in glycerol minimal medium in the presence of serine, other factors apart from Lrp being equal, if serA is not turned on, gltD cannot be turned on either. Conversely, if gltD is fully expressed, serA cannot be turned off. It is obvious that both genes respond to Lrp concentration. However, the difference in sensitivity ensures that there is no way of turning on gltD without turning on serA unless another regulatory factor is invoked.
A difference in sensitivity of promoters to a regulator is not unusual, cyclic AMP-cAMP receptor protein complex activating different genes at different concentrations (17). However, Crp regulates genes whose products are used as alternatives to each other; Lrp-regulated genes include many which are used simultaneously. If the sensitivities of other Lrp-regulated promoters also vary, then at different Lrp concentrations, the cell may achieve very different balances in the chemical reactions that it catalyzes. In fact, lrp transcription varies markedly with growth conditions, being decreased in LB and in the presence of amino acids and certain carbon sources and also affected by the ppGpp concentration of the cell (5, 13, 15). Insofar as the Lrp concentration regulates gene expression, one may expect subtle and not so subtle changes as a result of this variation in Lrp concentration.Activation of serA and gltD. It is clear that expression of both genes was highly dependent on Lrp and increased smoothly with the arabinose provided, in the range of 0.1 to 20 µg of arabinose/ml, the same range as cited in the original description of this vector (11), with expression saturated at about the same level as in the lrp+ parent but never exceeding that level. This is consistent with a steadily increasing response to increasing levels of activator, with a saturation level determined by intrinsic promoter strength.
Leucine inhibited transcription of serA and gltD in these studies, where lrp is transcribed from the pBAD promoter, as it does in wild-type cells, where it is transcribed from its own promoter. The leucine inhibition at lower arabinose concentrations is consistent with an equilibrium between free Lrp and leucine-bound Lrp, with leucine reducing the amount of Lrp to the same extent at all concentrations and thus reducing the amount of Lrp available for activation. However, if leucine were simply decreasing the available Lrp concentration, then one would expect that expression would increase at higher arabinose concentrations and the curves with and without leucine would eventually meetwhich they clearly do not.
In fact, the curves displayed in Fig. 1 and 2 result from the
interaction of a variety of factors. At very low arabinose and Lrp
levels, only the promoters with the highest affinity would be
transcribed. At higher Lrp concentrations, more and more promoters may
be affected. On the basis of in vitro experiments, it is generally thought that several dimers (28) of Lrp bind with high
cooperativity (16) at Lrp-regulated promoters like
ilvIH. However, our in vivo experiments do not show
cooperativity, either because they are insufficiently sensitive or
because in vivo Lrp may have to compete with several other nonspecific
binding proteins.
If cells grown with a high Lrp concentration transcribe some genes that
those grown at low concentration do not, they may also contain
different compounds, some of which may modulate the effects of leucine,
Lrp, or both. If one or more of these affects serA or
gltD transcription, or antagonizes lrp action,
this may be (part of) the explanation as to why the curves level off as they do.
Comparison with an earlier study on gltD expression. We have shown that transcription of both genes increased smoothly with the arabinose provided to reach a plateau which varied little as arabinose was increased further. The rise in activity was also seen in an earlier study of gltBDF by Borst et al. using lrp cloned under the control of an isopropylthiogalactopyranoside-inducible promoter, which showed a large decrease in expression at high Lrp concentrations (Fig. 4 in reference 2). This difference may be due to their using an insert of lacZ in gltB whereas we used one in gltD. If there is, as suggested (4), a minor promoter downstream of gltB, and if that promoter was activated at high Lrp concentration, this would explain the difference in results. It is also possible that this discrepancy is due to their use of an ampicillin-selected plasmid. At high Lrp concentrations, growth slows; there is more time for ampicillin to be degraded and plasmid lost, and the apparent GltD activity would be lower in proportion to the number of cells which lost plasmid. This was indeed our experience in preliminary experiments using ampicillin selection, which were not sufficiently reproducible to analyze and showed a wide variation in plasmid retention (data not shown).
Effect of leucine and serine on transcription from the pBAD22 promoter. The extent of the inhibition of transcription from pBAD by L-serine was a considerable surprise. L-Serine by itself does not serve as carbon source for our strain (22). However, when provided with limiting amounts of glucose, it does support growth (data not shown). Cells provided with both glycerol and serine may derive most of their carbon and energy from serine and thus cause catabolite repression at pBAD. We are currently investigating whether this is the case or whether some explanation based on arabinose uptake or plasmid copy number is more likely.
Lrp toxicity. We showed that Lrp overproduction not only slows growth as reported earlier (2) but also is actually toxic to the cell. The cells remain viable for several hours after arabinose is added, but most have died by 24 h of incubation. This may be due simply to overproduction of a positively charged DNA-binding protein clogging the works as suggested by Kurland and Dong (12). However, there may also be more specific effects of Lrp overproduction, an area which we are currently investigating.
Potential problems in uneven arabinose uptake. The analysis in this report depends on the assumption that the internal arabinose concentration is a direct function of the external concentration. Seigele and Hu have suggested that this is not the case (25). They report that in a population of cells grown in their medium, at low arabinose concentrations only some of them express green fluorescent protein cloned under the control of the pBAD promoter. They ascribe this to a differentiation in the population between some cells being fully induced for uptake and some cells not being induced at all.
Were this the case, our data would reflect an increase in the number of cells being turned on at different arabinose concentrations and could not be interpreted as we have done. We suggest a different interpretation based on our finding of the extreme sensitivity of the arabinose promoter to serine. The experiments of Seigele and Hu were done in the presence of a number of amino acids, among them serine at 40 µg/ml. This is a low concentration compared to the 500 µg/ml needed to saturate a culture of a serine-requiring mutant, and even a serine auxotroph degrades serine extensively (24). Because serine is the first amino acid used from a mixture of amino acids (23), we think that serine disappears rapidly in their experiments and that these experiments were done by accident at a critical level of serine which varies just enough from cell to cell so that some cells have enough serine to inhibit the arabinose promoter and some do not. This caveat would apply to any compound which might affect the arabinose promoter or might displace arabinose in the catabolite repression pecking order.| |
ACKNOWLEDGMENTS |
|---|
This work was supported by grant A6050 from the Canadian National Science and Engineering Research Council, for which we are extremely grateful.
We are grateful for ongoing discussions with R. D'Ari, G. Szamosi, and V. P. Mathur.
| |
FOOTNOTES |
|---|
* Corresponding author. Mailing address: Biology Department, Concordia University, 1455 de Maisonneuve Ave., Montreal, Quebec H3G 1M8, Canada. Phone: (514) 848-3410. Fax: (514) 848-2881. E-mail: Neweb{at}Vax2.Concordia.ca.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
| 1. |
Ambartsoumian, G.,
R. D'Ari,
R. T. Lin, and E. B. Newman.
1994.
Altered amino acid metabolism in lrp mutants of E. coli K-12 and their derivatives.
Microbiology
140:1737-1744 |
| 2. |
Borst, D. W.,
R. M. Blumenthal, and R. G. Matthews.
1996.
Use of an in vivo titration method to study a global regulator: effect of varying Lrp levels on expression of gltBDF in Escherichia coli.
J. Bacteriol.
178:6904-6912 |
| 3. |
Calvo, J. M., and R. G. Matthews.
1994.
Leucine-responsive regulatory protein: a global regulator of metabolism in Escherichia coli.
Microbiol. Rev.
58:466-498 |
| 4. | Castano, I. N., N. Flores, F. Valle, A. A. Covarrubias, and F. Bolivar. 1992. gltDF, a member of the gltBDF operon of Escherichia coli, is involved in nitrogen-regulated gene expression. Mol. Microbiol. 6:2744-2741. |
| 5. |
Chen, C. F.,
J. Lan,
M. K. Korovine,
Z. Q. Shao,
L. Tao,
J. Zhang, and E. B. Newman.
1997.
Metabolic regulation of lrp gene expression in Escherichia coli K-12.
Microbiology
143:2079-2084 |
| 6. |
Cui, Y.,
Q. Wang,
G. D. Stormo, and J. M. Calvo.
1995.
A consensus sequence for binding of Lrp to DNA.
J. Bacteriol.
177:4872-4880 |
| 7. | D'Ari, R., R. T. Lin, and E. B. Newman. 1993. The leucine-responsive regulatory protein: more than a regulator? Trends Biochem. Sci. 18:260-263[Medline]. |
| 8. |
Drlica, K., and Y. Rouviere-Yaniv.
1987.
Histone-like proteins of bacteria.
Microbiol. Rev.
51:301-319 |
| 9. |
Ernsting, B. R.,
M. R. Atkinson,
A. J. Ninfa, and R. G. Matthews.
1992.
Characterization of the regulon controlled by the leucine-responsive regulatory protein in Escherichia coli.
J. Bacteriol.
174:1109-1118 |
| 10. | Gottesman, S. 1984. Bacterial regulation: global regulatory networks. Annu. Rev. Genet. 18:415-441[Medline]. |
| 11. |
Guzman, L. M.,
D. Belin,
M. J. Carson, and J. Beckwith.
1995.
Tight regulation, modulation, and high-level expression by vectors containing the arabinose pBAD promoter.
J. Bacteriol.
177:4121-4130 |
| 12. | Kurland, C. G., and H. Dong. 1996. Bacterial growth inhibition by overproduction of protein. Mol. Microbiology 21:1-4[Medline]. |
| 13. |
Landgraf, J. R.,
J. Wu, and J. H. Calvo.
1996.
Effects of nutrition and growth rate on Lrp levels in Escherichia coli.
J. Bacteriol.
178:6930-6936 |
| 14. |
Lin, R. T.,
R. D'Ari, and E. B. Newman.
1990.
The leucine regulon of Escherichia coli: a mutation in rblA alters expression of L-leucine-dependent metabolic operons.
J. Bacteriol.
172:4529-4535 |
| 15. |
Lin, R. T.,
R. D'Ari, and E. B. Newman.
1992.
placMu insertions in genes of the leucine regulon: extension of the regulon to genes not regulated by leucine.
J. Bacteriol.
174:1948-1955 |
| 16. | Lin, R. T. 1992. . Characterization of the leucine/Lrp regulon of E. coli K-12. Ph.D. thesis. Concordia University, Montreal, Canada. |
| 17. | Lin, S., and A. D. Riggs. 1975. The general affinity of lac repressor for E. coli DNA: implications for gene regulation in procaryotes and eucaryotes. Cell 4:107-111[Medline]. |
| 18. | Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. . Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. |
| 19. | Miller, J. H. 1972. . Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. |
| 20. | Newman, E. B., R. D'Ari, and R. T. Lin. 1992. The leucine-Lrp regulon in E. coli: a global response in search of a raison d'etre. Cell 68:617-619[Medline]. |
| 21. | Newman, E. B., and R. T. Lin. 1995. Leucine-responsive regulatory protein, a global regulator of gene expression in E. coli. Annu. Rev. Microbiol. 49:747-75[Medline]. |
| 22. |
Newman, E. B.,
N. Malik, and C. Walker.
1982.
L-Serine degradation in Escherichia coli K-12: directly isolated ssd mutants and their intragenic revertants.
J. Bacteriol.
150:710-715 |
| 23. |
Pruss, B. M.,
J. M. Nelms,
C. Park, and A. J. Wolfe.
1994.
Mutations in NADH:ubiquinone oxidoreductase of Escherichia coli affect growth on mixed amino acids.
J. Bacteriol.
176:2143-2150 |
| 24. | Ramotar, D., and E. B. Newman. 1986. An estimate of the extent of deamination of L-serine in auxotrophs of Escherichia coli K-12. Can. J. Microbiol. 32:842-846[Medline]. |
| 25. |
Siegele, D. A., and J. C. Hu.
1997.
Gene expression from plasmids containing the araBAD promoter at subsaturating inducer concentrations represents mixed populations.
Proc. Natl. Acad. Sci. USA
94:8168-8172 |
| 26. |
Su, H., and E. B. Newman.
1991.
A novel L-serine deaminase activity in Escherichia coli K-12.
J. Bacteriol.
173:2473-2480 |
| 27. |
Tchetina, E., and E. B. Newman.
1994.
Identification of Lrp-regulated genes by inverse PCR and sequencing: regulation of two mal operons in Escherichia coli by leucine-responsive regulatory protein.
J. Bacteriol.
177:2679-2683 |
| 28. | Wang, Q., and J. H. Calvo. 1993. Lrp, a global regulatory protein of E. coli binds cooperatively to multiple sites and activates transcription of ilvIH. J. Mol. Biol. 229:306-318[Medline]. |
J Bacteriol, February 1998, p. 655-659, Vol. 180, No. 3
0021-9193/98/$04.00+0
Copyright © 1998, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Socias, S. B., Vincent, P. A., Salomon, R. A.
(2009). The Leucine-Responsive Regulatory Protein, Lrp, Modulates Microcin J25 Intrinsic Resistance in Escherichia coli by Regulating Expression of the YojI Microcin Exporter. J. Bacteriol.
191: 1343-1348
[Abstract] [Full Text] -
Berthiaume, F., Crost, C., Labrie, V., Martin, C., Newman, E. B., Harel, J.
(2004). Influence of L-Leucine and L-Alanine on Lrp Regulation of foo, Coding for F1651, a Pap Homologue. J. Bacteriol.
186: 8537-8541
[Abstract] [Full Text] -
Janes, B. K., Bender, R. A.
(1999). Two Roles for the Leucine-Responsive Regulatory Protein in Expression of the Alanine Catabolic Operon (dadAB) in Klebsiella aerogenes. J. Bacteriol.
181: 1054-1058
[Abstract] [Full Text]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»