Identification of promoter and stringent regulation of transcription of the
fabH
,
fabD
and
fabG
genes encoding fatty acid biosynthetic enzymes of
Escherichia coli
Identification of promoter and stringent regulation of transcription of the fabH , fabD and fabG genes encoding fatty acid biosynthetic enzymes of Escherichia coli
Sergey M.
Podkovyrov
and
Timothy J.
Larson*
Department of Biochemistry and Anaerobic Microbiology, Virginia Polytechnic
Institute and State University,
Blacksburg
, VA 24061-0308,
USA
Received December 15, 1995;
Revised and Accepted March 15, 1996
ABSTRACT
In
Escherichia coli
, amino acid starvation results in the coordinate inhibition of a variety of
metabolic activities, including fatty acid and phospholipid biosynthesis. By
using primer extension analysis we identified the
fabH
promoter responsible for transcription of the
fabH
,
fabD
and
fabG
genes encoding fatty acid biosynthetic enzymes. The response of the
fabH
promoter to amino acid starvation was determined
in vivo
. Transcripts originating from the
fabH
promoter were quantified by employing a ribonuclease protection assay. The
fabH
promoter was subject to
relA
-dependent stringent control and was repressed ~4-fold upon amino acid starvation. The results suggest that inhibition of
transcription initiation of lipid biosynthetic genes in starved cells
contributes to the stringent control of lipid biosynthesis.
INTRODUCTION
Escherichia coli
and other bacteria have effective adaptation mechanisms that help them survive
unfavorable environmental conditions such as nutritional stresses or
temperature shifts. An example of rapid adaptation is the cellular response to
amino acid starvation, termed the stringent response. In
E.coli
, amino acid deprivation results in the coordinate inhibition of a variety of
metabolic activities, including stable RNA synthesis, protein synthesis and
lipid synthesis (see
1
for review). Amino acid deficiency results in binding of codon-specified uncharged tRNA to ribosomes which activates the ppGpp synthetic
activity of the ribosomally bound RelA protein. Accumulation of ppGpp leads to
a highly specific inhibition of the transcription of stable RNA genes (
2
). In a relaxed
relA
strain ppGpp levels fail to increase with the onset of amino acid starvation
and stable RNA synthesis continues (
3
). ppGpp has been proposed to modify RNA polymerase, thereby altering the
pattern of transcription initiation from stable RNA promoters (
4
). Recently, direct interaction of ppGpp with
E.coli
RNA polymerase has been demonstrated (
5
).
One of the pleiotropic effects of the stringent response is an immediate
inhibition of fatty acid and phospholipid biosynthesis which occurs in
relA
+
but not in
relA
strains (
6
). Induction of expression of an unregulated, truncated
relA
gene situated on a multicopy plasmid leads to elevated ppGpp levels and
inhibition of
de novo
fatty acid and phospholipid synthesis (
7
). These data suggest that ppGpp is involved in the inhibition of fatty acid and
phospholipid synthesis, but little is known about the mechanisms of inhibition.
There are several reports that ppGpp can inhibit
in vitro
some enzymes participating in synthesis of fatty acids and phospholipids (
8
,
9
). Rock and co-authors demonstrated that
sn
-glycerol-3-phosphate acyltransferase is inhibited upon induction of ppGpp synthesis
in vivo
(
7
). Their data pointed to a direct biochemical interaction between the enzyme and
ppGpp. To our knowledge, all studies reported to date concerning the mechanisms
of inhibition of fatty acid and phospholipid synthesis dealt with effects of
ppGpp on the biosynthetic enzymes. In this paper, for the first time, we
present data regarding regulation of transcription of lipid biosynthetic genes
during amino acid starvation.
Recently we demonstrated that the
g30k
gene of unknown function, the
rpmF
gene encoding ribosomal protein L32, the
plsX
gene encoding a protein involved in membrane lipid synthesis and the
fabH, fabD
and
fabG
genes encoding several fatty acid biosynthetic enzymes comprise an operon (
10
; Fig.
1
). We found that, in addition to the operon promoters located upstream of
rpmF
, there is an internal promoter located within the
plsX
gene. The goal of the present study was to identify this promoter and to test
if it is subject to stringent control.
MATERIALS AND METHODS
Strains and growth media
All bacterial strains used in this study were
E.coli
K-12 derivatives. DH5[alpha]F'[[Phi]80
d lacZ
[Delta]
M15
[Delta](
lacZYA-argF
)
U169
deoR recA1 endA1
hsdR17
supE44
thi-1 gyrA96 relA1
] (Gibco BRL, Gaithersburg, MD) was used as the host for DNA manipulations.
TL504[[Delta](
lacZYA-argF
)
U169
zah-735
::Tn
10
] was derived from wild type strain MG1655(
11
) by P1 transduction with strain SH205 (
12
) as donor, with selection for tetracycline resistance. This
relA
+
[Delta]
lac
strain was used for isolation of RNA. The
fabH
promoter was assayed in parallel in both MC4100[
relA1
[Delta]
lac
] (
13
) and XZ132 [MC4100
relA
+
] (
14
). For experiments to test stringent control, A and B salts of Clark and Maaløe (
15
) were supplemented with 0.4% glucose, uracil (50 [mu]g/ml), thiamine (10 [mu]g/ml), and all 20 amino acids except serine (each at 40 [mu]g/ml). Cells were grown at 37oC to an A
600
of 0.5 and amino acid starvation was induced by addition of serine hydroxamate
to 400 [mu]g/ml. In all other experiments Luria-Bertani medium (
16
) was used. When needed, media were supplemented with 100 [mu]g/ml of ampicillin.
Oligonucleotides
Oligonucleotides were synthesized using an Applied Biosystems model 381A DNA
synthesizer and purified using oligonucleotide purification cartridges
(Cruachem, Dulles, VA) as recommended by the manufacturer. The following
oligonucleotides, with their 3' coordinates and references for the DNA sequences, were used in this
study:
Plasmids used in this study are listed in Table
1
. Plasmids pSPlac and pSPbla were constructed by cloning appropriate PCR
fragments into the
Eco
RI and
Bam
HI sites of pBluescript KS(+) (Stratagene, La Jolla, CA). pSP417 was the
template for PCR; primers 2 and 3 were used for amplification of the
bla
fragment and primers 4 and 5 were used for amplification of the
lac
fragment. T3 and T7 primers were used to sequence the fragment inserts of
pSPlac and pSPbla.
For plasmid DNA purification, Wizard Minipreps DNA Purification System was employed (Promega, Madison, WI). DNA fragments for
cloning were isolated from agarose gel by using Wizard PCR Preps DNA
Purification System (Promega). Insertions of recombinant plasmids were
sequenced by the chain termination method (
21
) with the Sequenase version 2.0 (Amersham, Arlington Heights, IL). PCR was performed in a standard reaction buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl) including 3 mM MgCl
2
, 0.2 mM of each deoxynucleotide, 1 [mu]M of each primer, 0.5 [mu]g plasmid DNA and 2 U
Taq
polymerase (AmpliTaq; Perkin-Elmer Cetus, Norwalk, CT) at 94oC for 30 s, 55oC for 30 s and 72oC for 30 s in a total of 35 cycles. All other standard
molecular biology techniques were based on methods described elsewhere (
16
).
Purification of RNA and primer extension analysis
Total RNA was isolated by using guanidine isothiocyanate for cell lysis and
rapid inactivation of cellular RNases (
22
) (all reagents for isolation of RNA were purchased from 5 Prime -> 3 Prime, Inc., Boulder, CO). The quality of RNA was determined by
visualization of distinct ribosomal RNA bands on a denaturing formaldehyde gel (
16
).
For primer extension analysis, primer 1 was 5' end-labeled using [[gamma]-
32
P]ATP and T4 polynucleotide kinase. Labeled primer, 1 pmol, was mixed with 5 [mu]g total RNA in a final volume of 8 [mu]l. The mixture was boiled for 2 min and cooled on ice. The hybridized
primer was extended by addition of all four dNTPs at 0.7 mM each, reverse
transcriptase buffer and 50 U Moloney murine leukemia virus reverse
transcriptase (both purchased from New England Biolabs, Beverly, MA) in a total
volume of 15 [mu]l, followed by incubation at 42oC for 30 min. The reaction was stopped by addition of 15 [mu]l gel loading buffer containing 95% formamide, 20 mM EDTA, 0.05%
bromphenol blue and 0.05% xylene cyanol. The extension products were resolved
on a 5% polyacrylamide sequencing gel.
vector for construction of
lacZ
transcriptional fusions
(24)
pSP413
fabH
promoter cloned into pSP417
this study
pSP17
rrnB
P1 promoter cloned into pSP417
this study
pSS20
lacUV5
promoter cloned into pSP417
S. Solow (this lab.)
pBluescript KS(+)
multiple cloning site flanked by T3 and T7 promoters
Stratagene
pSPlac
a
5' part of
lacZ
gene cloned into pBluescript KS(+)
this study
pSPbla
a
5' part of
bla
gene cloned into pBluescript KS(+)
this study
a
For coordinates of the cloned fragments see coordinates of oligonucleotides 4
and 5 (for
lac
), and 2 and 3 (for
bla
) in Materials and Methods.
In vitro
transcription
To prepare probes for RNase protection assays, plasmids pSPlac and pSPbla were
linearized using
Hin
dIII and transcribed
in vitro
in the presence of 0.5 mM each ATP, GTP and CTP, 50 [mu]M UTP, 50 [mu]Ci [[alpha]-
32
P]UTP and T3 RNA polymerase as described in Ambion technical bulletin for MAXIscript
in vitro
transcription kit (Ambion, Austin, TX). Following
in vitro
transcription, the template was destroyed by the addition of 2 U RNase-free DNase (Ambion). The unincorporated [[alpha]-
32
P]UTP was removed by passing the reaction mixture through a G-25 spin column (Boehringer Mannheim, Indianapolis, IN), and the eluate containing the probe was kept
at -70oC. To prepare molecular weight standards,
in vitro
transcription was performed with RNA Century marker template set (Ambion).
Ribonuclease protection assay and quantitation of RNA
Ribonuclease protection assays were performed by co-precipitation of
32
P-labeled probe (~1 * 10
4
c.p.m.) with sample RNA (~1-10 [mu]g). Hybridization and RNase digestion were conducted by using the
RPAII kit (Ambion). Protected RNA fragments were separated on an 8 M urea, 5%
polyacrylamide gel and detected by autoradiography. RNA was quantified by counting the radioactivity in the corresponding bands and by scanning the X-ray film using a Shimadzu CS-9000 scanning densitometer.
Assay of
[beta]
-galactosidase
Enzyme activity was determined in triplicate by using logarithmically growing
cells permeabilizied with sodium dodecyl sulfate and chloroform as described (
23
). One unit of enzyme activity was defined as described by Miller (
23
).
RESULTS
Identification of the
fabH
promoter
Recently we constructed a series of transcriptional fusions between different
parts of the
rpmF-plsX-fab
operon and
lacZ
, and employed insertional mutagenesis to study the transcriptional organization
of the operon (
10
). One of our conclusions was that there is an internal promoter (termed the
fabH
promoter) within the
plsX
gene responsible for ~40% of all transcripts for
fabH, fabD
and
fabG
. It follows from comparison of [beta]-galactosidase activities in the cells carrying transcriptional
fusions between different parts of the operon and
lacZ
that the
fabH
promoter is located between the
Ssp
I and the
Nru
I restriction sites (
10
; Fig.
1
). To further localize the position of the
fabH
promoter we used the
Hin
cII restriction site conveniently located between
Ssp
I and
Nru
I and cloned the
Ssp
I-
Hin
cII and the
Hin
cII-
Nru
I DNA fragments upstream of the promoterless
lacZ
gene of the plasmid vector pSP417 (
24
; the recombinant plasmids were named pSP411 and pSP413, respectively).
Expression of
lacZ
from plasmid pSP413 was much higher than that from pSP411 (18 500 U versus 1300
U) and comparable with that from the plasmids carrying the
Ssp
I-
Nru
I fragment (19 000 U;
10
), which localizes the
fabH
promoter within the 283 bp
Hin
cII-
Nru
I fragment (plasmid pSP413).
To map the
fabH
promoter more precisely we performed primer extension analysis. Total RNA was
isolated from exponentially growing TL504(pSP413) cells and hybridized with
primer 1 (see Materials and Methods). Primer extension products were run on the same gel with products of sequencing reaction of pSP413 with the same
primer. One extension product was seen (Fig.
2
) suggesting that there is only one promoter located within the
Hin
cII-
Nru
I fragment of pSP413. From the nucleotide sequence of the
plsX
gene (GenBank accession number M96793) and the results of primer extension
analysis (Fig.
2
), we concluded that the
fabH
promoter sequence is:
Stringent regulation of the
fabH
promoter
Three recombinant plasmids, pSP413, pSP17 and pSS20, were used in the
experiments to test stringent control of the
fabH
promoter. They all were derived from the same vector pSP417 (
24
), and carry the ampicillin resistant gene (
bla
) and different promoters cloned upstream of the
lacZ
gene. Plasmid pSP413 contains the
fabH
promoter. Plasmid pSP17 contains the P1 promoter of the
rrnB
ribosomal operon, a classical example of a stringently regulated promoter (
26
). We chose to clone the
rrnB
P1 core promoter (-42; +4) since it is 300-fold less active than the wild-type promoter with the upstream sequences, but is inhibited
by amino acid starvation to the same extent as the full-length promoter (
27
). P1 was assembled from the oligonucleotides 6 and 7 and cloned into the
Eco
RI and
Bam
HI sites of pSP417. Plasmid pSS20 contains the
lacUV5
promoter, which is not subject to stringent control (
1
) and serves as a negative control.
DISCUSSION
In our previous work we have shown that the
fabH, fabD
and
fabG
genes are part of the
rpmF-plsX-fab
operon (
10
). In the present study we mapped an internal
fabH
promoter of the operon located within the
plsX
gene. The three genes transcribed from the
fabH
promoter are
fabH, fabD
and
fabG
.
fabH
encodes 3-ketoacyl-ACP synthase III, the enzyme that catalyzes the first condensation
reaction of fatty acid biosynthesis (see
30
for a review of fatty acid biosynthesis). Malonyl-ACP required for this step is produced by the action of malonyl CoA-ACP transacylase (encoded by
fabD
). Mutants deficient in malonyl CoA-ACP transacylase require both saturated and unsaturated fatty acids for
growth (
31
).
fabG
encodes 3-ketoacyl-ACP reductase, the first enzyme participating in each cycle of chain
elongation. The FabD, FabH and FabG proteins catalyze the successive reactions
and organization of their genes into an operon is likely to be a means for
coordinate regulation. In
E.coli
, genes are often organized in operons for coordinate control of transcription
from the operon promoter. Some operons, however, have more complex regulatory
mechanisms such as transcription from multiple promoters (
32
), intra-operon attenuation (
33
) or differential decay of the polycistronic mRNA (
34
). Internal promoters have been discovered in a number of operons including an
operon containing genes for ribosomal protein, DNA primase and [sigma] factor of RNA polymerase (sigma operon;
35
) and an operon containing genes for ribosomal proteins and [beta] and [beta]' subunits of RNA polymerase (beta operon;
33
). The presence of promoters internal to the operon makes regulation of gene
expression more flexible permitting coordinate expression in some situations and discoordinate expression in others. For example, regulation of both RNA
polymerase and ribosomes is relatively coordinate upon changes in growth
conditions (
36
), but in the case of heat shock, the presence of an internal promoter in the
sigma operon allows discoordinate regulation by selective activation of
transcription of the
rpoD
gene encoding sigma factor (
35
,
37
). We suggest that the
fabH
promoter and the operon promoters located upstream of the
rpmF
gene may respond differently to some environmental signals, but at present
these signals are not identified.
The striking feature of the
fabH
promoter found in this work is its stringent regulation. By and large, studies
on stringently regulated promoters are limited to genes that encode products
involved in ribosome function. Recently stringent control has been demonstrated
for the
dnaA
(
38
) and
fis
(
39
) genes encoding the DnaA protein involved in DNA replication and the Fis
protein involved in a number of cellular processes including the
transcriptional activation of stable RNA synthesis. Our results show that
transcription of genes encoding fatty acid biosynthetic enzymes is also subject
to stringent control. Thus, the stringent control of transcription may be a
mechanism for inhibition of some anabolic cellular functions during amino acid
starvation.
One of the numerous effects of amino acid starvation on cellular physiology and
metabolism is a
relA
-dependent inhibition of fatty acid and phospholipid synthesis (
6
). A target for stringent control of lipid synthesis has not been defined
precisely, however. Rock and co-workers showed that overexpression of the
plsB
gene encoding
sn
-glycerol-3-phosphate acyltransferase relieves the inhibition of fatty
acid and phospholipid synthesis induced by accumulation of ppGpp (
7
). It should be noted that in their work ppGpp accumulation was achieved by
induction of expression of the
relA
gene located on a plasmid, and, in contrast to induction by amino acid
starvation, phospholipid biosynthesis was not completely abolished in induced
cells. Thus, cell responses to amino acid starvation and
relA
overexpression are different and a target for stringent control of lipid
synthesis in these two cases may be different. The
sn
-glycerol- 3-phosphate acyltransferase catalyzes the first step in
phospholipid biosynthesis by condensation of
sn
-glycerol-3-phosphate and fatty acylthioesters to yield lysophosphatidic
acid. Since the preceding step, formation of fatty acids, requires >90% of the
ATP consumed in lipid biogenesis, it appears to be likely that an early step in
fatty acid biosynthesis could be a primary site for stringent regulation. Our
data indicate that the
fabH
promoter is subject to stringent control. This means that transcription of
fatty acid biosynthetic genes (
fabH
,
fabG
and
fabD
) is inhibited upon amino acid starvation. Especially noteworthy in this regard
is that the FabH protein is thought to be a regulator of fatty acid
biosynthesis in bacteria (
40
,
41
). Our findings show that one potential mechanism of inhibition of fatty acid
biosynthesis upon amino acid starvation may be realized through ppGpp-dependent inhibition of transcription of the pathway genes. On the other
hand, an immediate effect of ppGpp inhibition may be caused by direct
biochemical interaction between ppGpp and the corresponding biosynthetic
enzyme(s). We suggest that due to the complexity of the changes taking place
during the stringent response, inhibition of fatty acid and phospholipid
biosynthesis is a complex event with controls exerted at both the
transcriptional and enzymatic levels. Also, some inhibitory effects of ppGpp
may be indirect or part of a regulatory cascade.
ACKNOWLEDGEMENTS
We thank H. Bremer for strain XZ132, S. Solow for plasmid pSS20, A. T. van Loo-Bhattacharya for synthesizing oligonucleotides and R. Gourse for helpful advice. This work was supported by US Public
Health Service Grant GM47270 from NIH.
REFERENCES
1 Cashel,M. and Rudd,K.E. (1987) In Neidhardt,F.C., Ingraham,J.L., Low,B.L., Magasanik,B., Schaechter,M. and Umbarger,H.E. (eds), Escherichiacoliand Salmonella typhimurium-Cellular and Molecular Biology. American Society for Microbiology, Washington, DC, pp. 1410-1438.
2 Stomato,T.D. and Pettijohn,D.E. (1971) Nature (London), 234, 99-102.
6 Cronan,J.E.,Jr. and Rock,C.O. (1987) In Neidhardt,F.C., Ingraham,J.L., Low,B.L., Magasanik,B., Schaechter,M. and Umbarger,H.E. (eds), Escherichiacoliand Salmonella typhimurium-Cellular and Molecular Biology. American Society for Microbiology, Washington, DC, pp. 474-497.
7 Heath,R.J., Jackowski,S. and Rock,C.O. (1994) J. Biol. Chem., 269, 26584-26590.MEDLINE Abstract
8 Merlie,J.P. and Pizer,L.I. (1973) J. Bacteriol., 116, 355-366.MEDLINE Abstract
14 Zhang,X. and Bremer,H. (1995) J. Biol. Chem., 270, 11181-11189.MEDLINE Abstract
15 Clark,D.J. and Maaløe,O. (1967) J. Mol. Biol., 23, 99-112.
16 Maniatis,T., Fritsch,E.F. and Sambrook,J. (1989) Molecular cloning: A Laboratory Manual. Second edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
17 Yanofsky,C., Platt,T., Crawford,I.P., Nichols,B.P., Christie,G.E., Horowitz,H., Van Cleemput,M. and Wu,A.M. (1981) Nucleic Acids Res., 9, 6647-6668.MEDLINE Abstract
23 Miller,J.H. (1992) A Short Course in Bacterial Genetics. A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
24 Podkovyrov,S.M. and Larson,T.J. (1995) Gene, 156, 151-152.MEDLINE Abstract
36 Iwakura,Y. and Ishihama,A. (1975) Mol. Gen. Genet., 142, 67-84.MEDLINE Abstract
37 Gross,C.A., Burton,Z., Gribskov,M., Grossman,A., Liebke,H., Taylor,W., Walter,W. and Burgess,R.R. (1982) In Rodriguez,R. and Chamberlin,M. (eds), Promoters: Structure and Function. Praeger, NY, pp. 252-262.
38 Chiaramello,A.E. and Zyskind,J.W. (1990) J. Bacteriol., 172, 2013-2019.MEDLINE Abstract
