Molecular Microbiology (2000) 36(3), 570±584
Enzyme I and HPr from Lactobacillus casei: their role insugar transport, carbon catabolite repression andinducer exclusion
Rosa Viana,1² Vicente Monedero,1,2²
ValeÂrie Dossonnet,2 Christian Vadeboncoeur,3
Gaspar PeÂrez-MartõÂnez1* and Josef Deutscher2
1Instituto de AgroquõÂmica y TecnologõÂa de Alimentos,
C.S.I.C., Ap. Correos 73, 46100 Burjassot, Valencia,
Spain.2Laboratoire de GeÂneÂtique des Microorganismes, INRA-
CNRS URA 1925, 78850 Thiverval-Grignon, France.3GREB, DeÂpartement de Biochimie et de Microbiologie,
Faculte des Sciences et de GeÂnie and Faculte de
MeÂdecine Dentaire, Universite Laval, Ste-Foy, QueÂbec,
Canada G1K 7P4.
Summary
We have cloned and sequenced the Lactobacillus
casei ptsH and ptsI genes, which encode enzyme I
and HPr, respectively, the general components of the
phosphoenolpyruvate±carbohydrate phosphotransfer-
ase system (PTS). Northern blot analysis revealed that
these two genes are organized in a single-transcrip-
tional unit whose expression is partially induced. The
PTS plays an important role in sugar transport in
L. casei, as was confirmed by constructing enzyme
I-deficient L. casei mutants, which were unable to
ferment a large number of carbohydrates (fructose,
mannose, mannitol, sorbose, sorbitol, amygdaline,
arbutine, salicine, cellobiose, lactose, tagatose, treha-
lose and turanose). Phosphorylation of HPr at Ser-46 is
assumed to be important for the regulation of sugar
metabolism in Gram-positive bacteria. L. casei ptsH
mutants were constructed in which phosphorylation
of HPr at Ser-46 was either prevented or diminished
(replacement of Ser-46 of HPr with Ala or Thr
respectively). In a third mutant, Ile-47 of HPr was
replaced with a threonine, which was assumed to
reduce the affinity of P±Ser±HPr for its target protein
CcpA. The ptsH mutants exhibited a less pronounced
lag phase during diauxic growth in a mixture of
glucose and lactose, two PTS sugars, and diauxie
was abolished when cells were cultured in a mixture
of glucose and the non-PTS sugars ribose or maltose.
The ptsH mutants synthesizing Ser-46±Ala or Ile-47±
Thr mutant HPr were partly or completely relieved
from carbon catabolite repression (CCR), suggesting
that the P±Ser±HPr/CcpA-mediated mechanism of CCR
is common to most low G1C Gram-positive bacteria.
In addition, in the three constructed ptsH mutants,
glucose had lost its inhibitory effect on maltose
transport, providing for the first time in vivo evidence
that P±Ser±HPr participates also in inducer exclusion.
Introduction
In Gram-positive bacteria, the phosphocarrier protein HPr,
one of the general components of the phosphoenol-
pyruvate (PEP)±sugar phosphotransferase system (PTS),
can be phosphorylated at two different sites. PEP-
dependent phosphorylation catalysed by enzyme I occurs
at the catalytic His-15. P±His±HPr transfers its phos-
phoryl group to the sugar-specific enzyme IIA 0s, which in
turn phosphorylate their corresponding enzyme IIB 0s. P-
enzyme IIB and the integral membrane protein enzyme
IIC catalyse the simultaneous uptake and phosphorylation
of carbohydrates (Postma et al., 1993). P±His±HPr can
transfer its phosphoryl group also to non-PTS proteins,
such as glycerol kinase (Charrier et al., 1997) or anti-
terminators and transcriptional activators possessing the
PTS regulation domain (PRD), which contains several
phosphorylation sites recognized by P±His±HPr (Tortosa
et al., 1997; StuÈ lke et al., 1998; Lindner et al., 1999). In all
cases, P±His±HPr-dependent phosphorylation leads to
the activation of the function of the non-PTS proteins and
this phosphorylation has been shown to serve as a
secondary carbon catabolite repression (CCR) mechan-
ism in Gram-positive bacteria (Deutscher et al., 1993;
KruÈger et al., 1996; Martin-Verstraete et al., 1998). In
Lactobacillus casei, the antiterminator LacT, which
regulates the expression of the lac operon, contains a
PRD and seems to be controlled by this mechanism.
Dephosphorylation of P±His±HPr and P±His±LacT in
cells growing in glucose-containing medium was assumed
to be partly responsible for the repressive effect exerted
by the presence of glucose on lac operon expression
(Gosalbes et al., 1997; Gosalbes et al., 1999).
The second, ATP-dependent phosphorylation of HPr in
Q 2000 Blackwell Science Ltd
Received 13 October, 1999; revised 18 January, 2000; accepted28 January, 2000. ²The first two authors contributed equally to thiswork. *For correspondence. E-mail [email protected]; Tel.(134) 963 900 022; Fax (134) 963 636 301.
Regulatory functions of enzyme I and HPr in L. casei 571
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
Gram-positive bacteria is catalysed by the bifunctional
HPr kinase±phosphatase (Galinier et al., 1998; Reizer
et al., 1998; Brochu and Vadeboncoeur, 1999; Kravanja
et al., 1999). In Bacillus subtilis, this phosphorylation,
which occurs at the regulatory Ser-46 (Deutscher et al.,
1986), is stimulated by fructose-1,6-bisphosphate and
inhibited by inorganic phosphate (Galinier et al., 1998).
P±Ser±HPr participates in the major mechanism of CCR±
carbon catabolite activation operative in bacilli and pre-
sumably other Gram-positive bacteria (Deutscher et al.,
1997). It functions as a corepressor for the catabolite
control protein CcpA, a member of the LacI±GalR family
of transcriptional repressors±activators (Henkin et al.,
1991). The complex formed between CcpA and P±Ser±
HPr has been shown to bind to catabolite response
elements (cre) (Fujita and Miwa, 1994; GoÈsseringer et al.,
1997; Kim et al., 1998; Galinier et al., 1999; Martin-
Verstraete et al., 1999), operator sites preceding or
overlapping the promoters or being located within the 5 0
region of catabolite repressed genes and operons (Hueck
et al., 1994).
Studies regarding the role of P±Ser±HPr in the CcpA-
mediated CCR mechanism in Gram-positive bacteria
have mainly been carried out with bacilli. A single Bacillus
subtilis ptsH1 mutation, which results in the replacement
of Ser-46 of HPr with an alanine, has so far been used to
study in vivo the involvement of Ser-46 phosphorylation of
HPr in CCR. This mutation was originally present in strain
SA003 (Deutscher et al., 1994) and has been transferred
in the meantime into many other strains. Similarly, hprK
mutants, which are defective in HPr kinase±phosphatase,
have been constructed only with B. subtilis (Galinier et al.,
1998; Reizer et al., 1998). To test whether the P±Ser±
HPr/CcpA-dependent CCR mechanism established for
bacilli is operative in a similar manner in other Gram-
positive bacteria, we have cloned and sequenced the L.
casei ptsHI operon and constructed various ptsH mutants,
including ptsH1. We studied the effect of these mutations
on diauxic growth and CCR. L. casei was an ideal
candidate for these studies, as it exhibits a strongly
expressed lag phase during diauxic growth. Cells grown in
a medium containing glucose and a less preferred carbon
source such as lactose, galactose, maltose or ribose,
stopped growing for up to 20 h when glucose had been
consumed, before they finally started to regrow on the
second carbon source (Veyrat et al., 1994).
Based on in vitro studies, P±Ser±HPr has been
suggested to also be implicated in inducer exclusion.
The Ser-46±Asp mutant HPr which structurally resembles
P±Ser±HPr (Wittekind et al., 1989), has been reported to
interact with the lactose- and glucose-specific non-PTS
permeases of Lactobacillus brevis (Ye and Saier,
1995a,b) and to inhibit the non-PTS sugar transporters
by uncoupling them from proton symport (Ye et al., 1994a,b).
The presence of the corresponding non-PTS substrate
was found to exert a co-operative effect on binding of Ser-
46±Asp mutant HPr to the non-PTS permeases. How-
ever, owing to the use of Ser-46±Asp mutant HPr in
place of the physiologically relevant P±Ser±HPr and
owing to the availability of only in vitro results, the
involvement of P±Ser±HPr in inducer exclusion in
Gram-positive bacteria remained questionable and was
still awaiting in vivo confirmation. This could not be
achieved with Bacillus subtilis strains, because until now
no transport system submitted to inducer exclusion has
been detected in this organism. As we had found that
maltose uptake by L. casei cells is strongly inhibited by
the presence of glucose, suggesting that the maltose
permease is regulated by an inducer exclusion mechan-
ism, we studied whether the above ptsH mutations would
influence the inhibitory effect of glucose on maltose
transport in L. casei.
Results
Purification and N-terminal sequence of HPr from L. casei
Purification of L. casei HPr was performed as described
under Experimental procedures by carrying out the
following steps: heat denaturation of a crude extract, gel
filtration of the supernatant containing the heat-resistant
HPr followed by reverse phase chromatography on a
high performance liquid chromatography (HPLC) column.
After reverse phase chromatography, HPr was found to
be nearly homogenous (more than 95% pure), as was
judged from SDS±PAGE on which 10 ml aliquots of HPr-
containing fractions had been separated. Aliquots (about
0.5 nmol) of the purified, lyophilized HPr were used to
determine the sequence of the first 21 amino acids by
automated Edman degradation. The following N-terminal
amino acid sequence was found for HPr from L. casei:
M±E±K±R±E±F±N±I±I±A±E±T±G±I±H±A±R±P±A±T±L.
The obtained sequence exhibited strong similarity to the
N-terminal sequence of HPr from other bacteria, notably
around the active centre His-15, the site of PEP-
dependent phosphorylation (data not shown), confirming
that the purified protein was indeed HPr of L. casei.
Cloning of PCR-amplified L. casei ptsHI fragments
To amplify L. casei DNA fragments containing ptsH and
part of ptsI, degenerate oligonucleotides were designed
based on the N-terminal sequence of HPr described
above and on strongly conserved regions in enzyme I,
which were detected by carrying out an alignment of
different enzyme I sequences (data not shown). Two
combinations of primers (PTS-H2±PTS-I3 and PTS-H2±
PTS-I4) gave PCR-amplified fragments of 1.6 kb and
572 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
0.3 k respectively. Sequencing of the PCR products
revealed that the deduced amino acid sequences
exhibited strong similarity to the sequences of known
enzyme I and HPr. As expected, both DNA fragments
began with the 5 0 end of ptsH and extended to the region
in ptsI encoding the conserved sequence chosen as a
basis for the second primer (Fig. 1). The larger of the two
fragments obtained with primer PTS-I3 was cloned into
pUC18, providing plasmid pUCR±H1.
Disruption of ptsI and cloning and sequencing of DNA
fragments containing flanking regions of L. casei ptsH and
ptsI
A 865 bp EcoRI fragment, which contained an internal
part of the ptsI gene, (Fig. 1) was obtained from plasmid
pUCR±H1 and subcloned into the suicide vector pRV300
(Leloup et al., 1997), providing plasmid pVME800. This
plasmid was used to transform the L. casei wild-type
strain BL23 and integration of the plasmid at the correct
location (ptsI::pVME800) was verified by PCR and
Southern blot. The phenotype of two clones carrying the
correct integration was tested. In contrast to the wild-type
strain, the two mutants could no longer produce acid from
fructose, mannose, mannitol, sorbose, sorbitol, amygda-
line, arbutine, salicine, cellobiose, lactose, tagatose,
trehalose and turanose. However, they could still meta-
bolise ribose, galactose, glucose, N-acetylglucosamine,
aesculine, maltose and gluconate, suggesting that in L.
casei PTS-independent transport systems exist for this
second class of sugars.
Restriction analysis of the ptsHI region was carried out
by Southern hybridization using DNA isolated from one
integrant (BL124) with the aim of identifying restriction
enzymes allowing cloning of the ptsH and ptsI genes
together with their flanking regions. Digestion of BL124
Fig. 1. Schematic presentation of the sequenced chromosomal L. casei DNA fragment containing the ptsHI operon. Indicated are the three ORFsdetected in this fragment, the promoter and terminator of the ptsHI operon and several important restriction sites. A 400 bp DNA sequence,including the ptsHI promoter, the ribosome binding site (both underlined), the transcription start site mapped by primer extension (marked with anasterisk) and almost the complete ptsH gene, is shown underneath the schematic presentation of the total sequenced DNA fragment. The arrowindicates the position of the oligonucleotide used in the primer extension experiment. The phosphorylatable His-15 and Ser-46 residues are boxed.Amino acid substitutions present in HPr encoded by the three ptsH alleles are written below the wild-type sequence at positions Ser-46 and Ile-47.The initially isolated PCR fragments H2±I4 and H2±I3 (flanked by inverted arrows) and the 865 bp EcoRI fragment, which was called E800 andsubcloned into pRV300, are shown above the schematic presentation of the total sequenced DNA fragment. The plasmid containing the 865 bpEcoRI fragment was used to construct a mutant with a disrupted ptsI gene, which allowed us to clone the flanking regions of the ptsHI operon byisolating plasmids pVMH1 and pVMS1, the inserts of which are completely or partly presented in this figure.
Regulatory functions of enzyme I and HPr in L. casei 573
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
DNA with SacI and religation of the obtained DNA
fragments allowed to isolate plasmid pVMS1 carrying an
about 9 kb insert. Partial sequencing of this insert
revealed that it contained the 3 0 part of ptsI and its
downstream region (Fig. 1). The same experiment carried
out with HindIII allowed us to isolate plasmid pVMH1
carrying a 2.4 kb insert comprising the complete ptsH
gene together with part of its promoter region and the
5 0 part of ptsI (Fig. 1). The sequence containing the
complete ptsH promoter and 560 bp of the upstream
region was subsequently obtained by reverse PCR. In
total, a continuous stretch of 4150 bp has been
sequenced and has been submitted to the GenBank
database under accession number AF159589. It con-
tained the complete ptsH and ptsI genes and an open
reading frame (ORF) located downstream of ptsI. The
stop codon of ptsH was found to overlap with the initiation
codon of ptsI by 1 bp, suggesting that these two genes
are organized in an operon. Whereas the encoded L.
casei HPr and enzyme I exhibited sequence similarities
ranging from 65% to 85% when compared with their
homologues in B. subtilis, Lactococcus lactis, Lactoba-
cillus sakei, Streptococcus salivaruis or Enterococcus
faecalis, the protein encoded by the ORF located down-
stream of ptsI exhibited similarity to the sugar permeases
XylE (Davis and Henderson, 1987) and GalP (Pao et al.,
1998) from Escherichia coli. No ORF could be detected in
the 560 bp region upstream from the ptsHI promoter.
Transcriptional analysis of the L. casei ptsHI operon
To determine the size of the ptsHI transcripts and to test
the effect of a man (prevents the uptake of glucose via the
PTS) and a ccpA mutation on ptsHI expression, Northern
blots were performed with RNA isolated not only from the
L. casei wild-type BL23, but also from the mutant strains
BL30 (man) (Veyrat et al., 1994), BL71 (ccpA) (Monedero
et al., 1997) and BL72 (man ccpA) (Gosalbes et al.,
1997), which were grown in medium containing either
glucose, lactose or ribose. Hybridization experiments
were carried out with either ptsH (Fig. 2) or ptsI-specific
probes (data not shown). With both probes, a mRNA band
of about 2.1 kb could be detected, which is in good
agreement with the size expected for the combined ptsH
and ptsI genes, confirming that these two genes are
organized in an operon and that transcription stops at the
stem±loop structure located downstream of ptsI (Fig. 1).
Densitometric measurement of the hybridizing bands in
the RNA isolated from cells of the different mutants
grown in glucose-, lactose-, or ribose-containing medium
showed that expression of the ptsHI operon was
moderately induced by glucose in the wild type and
ccpA mutant, while this effect was less pronounced
in the strains carrying the man mutation (Fig. 2). The
transcriptional start point of the transcript of L. casei ptsHI
operon was mapped by primer extension using the primer
PTS-P.E. (Fig. 1) and was found to be at the G located
82 bp upstream of the ptsH start codon (Fig. 1).
Construction of ptsH mutants altered at Ser-46 or Ile-47
To construct chromosomal L. casei ptsH mutants by
double crossing over, we first tried to obtain a ptsI mutant
strain carrying a frameshift mutation. For this purpose,
plasmid pVMH1 was partially digested with EcoRI and
made blunt end (filled in with the Klenow fragment) before
it was religated and used to transform E. coli DH5a. From
one of the resulting transformants, a plasmid (pVMR10)
could be isolated bearing a frameshift mutation at the
EcoRI site located at nucleotide 327 of the ptsI gene, as
was confirmed by restriction analysis and DNA sequen-
cing (insertion of four additional bp). Plasmid pVMR10
was subsequently used to transform L. casei BL23 and an
erythromycin-resistant ptsI1 integrant resulting from a
Campbell-like recombination was isolated. From this
strain, a ptsI mutant (ptsI1, BL126) could be obtained by
a second recombination. BL126 was erythromycin-sensi-
tive and exhibited a fermentation pattern identical to that
found for the ptsI::pVME800 mutant BL124. Interestingly,
no ptsHI mRNA could be detected in BL126 by Northern
blot analysis (data not shown).
Fig. 2. Northern blot showing hybridization bands obtained with aptsH-specific probe. Experiments were carried out with RNA isolatedfrom L. casei BL23 (wild type), BL30 (man mutant), BL71 (ccpAmutant) and BL72 (man ccpA double mutant). Lanes are markedaccording to the carbon source used to grow the bacterial cells: G(glucose), L (lactose) or R (ribose). Numbers on the left and rightmargins indicate the position of the two closest RNA standards andthe estimated size of the observed band in kb respectively. Thehistogram below shows the average relative band intensities and theirerrors from three different experiments. The average relative bandintensities were calculated as the peak area of the bands observed inthe photographic film after hybridization divided by the peak area ofthe 23S rRNA.
574 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
PCR-based site directed mutagenesis was carried out
with the L. casei ptsH gene present in plasmid pVMH1
(Table 1) to replace either Ser-46 with alanine, aspartic
acid or threonine, or Ile-47 with threonine. The four
resulting plasmids carrying the various ptsH alleles were
named pVMH2, pVMH3, pVMH4 and pVMH5, respec-
tively (Table 1), and were used to transform the L. casei
ptsI mutant BL126. Erythromycin-resistant recombinants
were obtained with all plasmids. As outlined in Fig. 3, the
Campbell-like first recombination can occur at three
different places (before the ptsH mutation, between the
ptsH and ptsI mutations and after the ptsI mutation)
resulting in three distinguishable DNA arrangements.
Whereas recombinants of type 1 and 2 were unable to
utilize lactose, recombinants of type 3 were able to grow
slowly on lactose. This was probably due to a readthrough
from a plasmid-located promoter allowing low ptsI
expression. For each ptsH allele, one `type 3' recombi-
nant was used for further experiments aimed to isolate
strains with a second recombination. Again, three
possibilities existed for the second recombination event
providing either the original ptsI mutant, a wild-type strain
or one of the desired ptsI1 ptsH strains (Fig. 3). In the last
case, the resulting strains were expected to contain a
functional PTS and to be able to grow on PTS substrates,
although slightly reduced growth on PTS substrates has
been reported for mutants in which Ser-46 or Ile-47 of HPr
were altered (Deutscher et al., 1994; Eisermann et al.,
1988; Reizer et al., 1989; Gauthier et al., 1997).
Erythromycin-sensitive clones able to ferment lactose
were therefore isolated after cells had been grown for 200
generations without selective pressure to allow the
second recombination to occur. As explained, these
clones could be either wild-type strains or ptsI1 ptsH
mutants (Fig. 3). Two types of erythromycin-sensitive
lactose-fermenting recombinants were obtained that
exhibited slightly different growth characteristics. Using
appropriate primers, the ptsH alleles of two clones of the
slower and faster growing recombinants were amplified by
PCR and sequenced. For each ptsH allele, the two faster
growing clones contained the wild-type ptsH, whereas the
slightly slower growing strains carried either the Ser-46±
Ala (ptsH1, BL121), the Ser-46±Thr (ptsH2, BL122) or the
Ile-47±Thr ptsH mutation (ptsH3, BL123). Only a strain
synthesizing Ser-46±Asp mutant HPr could not be
obtained with this method, although PCR amplification
followed by DNA sequencing was carried out with 15
erythromycin-sensitive clones constructed with plasmid
pVMH4. Because the Ser-46±Asp ptsH mutation is
known to drastically lower PTS-mediated sugar transport
Table 1. Lactobacillus casei: strains and plas-mids used in this study. Strain or plasmid Genotype or properties Origin
L. caseiBL23 wild type Bruce ChassyBL30 man Veyrat et al. (1994)BL71 ccpA Monedero et al. (1997)BL72 man ccpA Gosalbes et al. (1997)BL121 ptsH1 (S46AHPr) This workBL122 ptsH2 (S46THPr) This workBL123 ptsH3 (I47THPr) This workBL124 ptsI::pVME800 This workBL126 ptsI1 (frameshift introduced into the
first EcoRI site of ptsI)This work
PlasmidpUC18 Pharmacia-BiotechpRV300 pBluescript SK2 with the pAMb1 EmR gene Leloup et al. (1997)pUCR-HI pUC18 with 1.6 kb PCR fragment with
part of ptsH and ptsIThis work
pVME800 pRV300 with a 865-bp EcoRI internalptsI fragment
This work
pVMS1 pRV300 with 9 kb fragment downstreamfrom ptsI
This work
pVMH1 pRV300 with part of ptsI, complete ptsHand 105 bp upstream from ptsH
This work
pVMH2 pVMH1 derivative (codon 46 of ptsHis GCT for Ala)
This work
pVMH3 pVMH1 derivative (codon 46 of ptsHis ACT for Thr)
This work
pVMH4 pVMH1 derivative (codon 46 of ptsHis GAT for Asp)
This work
pVMH5 pVMH1 derivative (codon 47 of ptsHis ACC for Thr)
This work
pVMR10 pVMH1 derivative with a frameshift in thefirst EcoRI site of ptsI
This work
Regulatory functions of enzyme I and HPr in L. casei 575
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
and phosphorylation (Reizer et al., 1989), clones able or
unable to ferment lactose were tested.
The ptsH mutations affect CCR and diauxic growth
In order to test the effect of the different amino acid
substitutions in HPr on diauxie, the growth behaviour of
the mutants on basal MRS broth supplemented with 0.1%
glucose and 0.2% lactose was compared with that of the
wild type and a ccpA mutant (Fig. 4). As previously
demonstrated (Veyrat et al., 1994; Gosalbes et al., 1997;
Gosalbes et al., 1999), the L. casei wild-type strain
exhibited strong diauxic growth in the presence of these
two sugars with a lag phase of about 15 h separating
the growth phases on glucose and lactose, whereas in the
ccpA mutant strain this lag phase was reduced to 5 h. The
diauxic growth observed with the ptsH2 mutant was very
similar to that of the wild-type strain. By contrast, the lag
phase was only about 6 h for the ptsH1 mutant and in
between the wild type and ptsH1 mutant for the ptsH3
strain (10 h).
A similar gradation was found when the relief from
glucose-mediated repression of N-acetyglucosaminidase
activity was investigated. Whereas high activity of this
enzyme could be measured in ribose-grown wild-type
cells, glucose was found to inhibit its activity about 10-fold
(Fig. 5). Similarly to the ccpA mutant, the repressive effect
of glucose on N-acetylglucosaminidase had completely
disappeared in the ptsH1 mutant. Inhibition of N-acetyl-
glucosaminidase activity by the presence of glucose in the
growth medium was also clearly diminished in the two
other ptsH mutants (about twofold inhibition in the ptsH3
mutant and 2.5-fold inhibition in the ptsH2 mutant),
confirming the importance of Ser-46 phosphorylation of
HPr and of the amino acids in the vicinity of Ser-46 for
CCR in L. casei. Therefore, these two tests indicated that
there was a remarkable and progressive loss of catabolite
repression in the different mutants: wild-type , ptsH2 ,
ptsH3 , ptsH1 � ccpA.
The ptsH mutations affect inducer exclusion in L. casei
To test a potential effect of the ptsH mutations on inducer
exclusion of non-PTS sugars, we decided to study ribose
Fig. 3. Schematic presentation of possible recombination events during the construction of ptsH mutants with the ptsI1 strain BL126 and the pVMHplasmids containing the various ptsH alleles. Integration of the pVMH plasmids carrying a mutation in ptsH (indicated by the filled circle) into thechromosome of BL126 carrying a frameshift mutation in ptsI (indicated by the filled triangle) by Campbell-like recombination could take place atthree different locations, resulting in the three different DNA arrangements presented under `1st recombination'. Integrants obtained by the first andsecond type of recombination exhibited a Lac2 phenotype, whereas type 3 integrants could slowly ferment lactose (probably due to a readthroughfrom a plasmid-located promoter). Type 3 integrants obtained with each of the three pVMH plasmids were grown for 200 generations withoutselective pressure to allow the second recombination leading to the excision of the pVMH plasmids. Again, three possibilities existed for the secondrecombination providing either a lac2 strain (3a), a wild-type strain (3b) or the desired ptsH mutant (3c) (see 2nd recombination). As only the Ser-46±Asp ptsH mutation was found to exert an inhibitory effect on the phosphocarrier activity of HPr, erythromycin-sensitive, lactose-fermentingcolonies were isolated from `type 3 0 integrants obtained by single recombination. These clones contained either the wild-type ptsH (secondrecombination according to 3b) or one of the three desired ptsH alleles (second recombination according to 3c). Amplification of the ptsH gene byPCR and DNA sequencing allowed us to distinguish between these two possibilities and ptsH mutants with the three different ptsH alleles couldthus be obtained.
576 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
and maltose uptake in L. casei. When L. casei wild-type
cells were grown in a medium containing glucose and
either ribose or maltose, a diauxic growth behaviour
similar to that obtained with cells growing in the presence
of glucose and lactose was observed (data not shown).
However, whereas the lag time of the diauxic growth in
the presence of glucose and lactose was not or only partly
reduced in the ptsH mutants (Fig. 4), the diauxic growth
completely disappeared when the ptsH strains were
grown in a medium containing glucose and either maltose
or ribose (data not shown). These results suggest that
phosphorylation of HPr at Ser-46 plays an important role
in regulation of the utilization of these two non-PTS
sugars by L. casei. In order to distinguish whether this
effect was mediated via interaction of the CcpA/P±Ser±
HPr complex with cre sequences or via interaction of P±
Ser±HPr with a sugar permease according to the
proposed mechanism of inducer exclusion ( Ye et al.,
1994a, b; Ye and Saier, 1995a, b), sugar transport
experiments were performed. The uptake of ribose by
ribose-grown L. casei wild-type cells is shown in Fig. 6A.
The addition of glucose to ribose-transporting wild-type
cells caused no inhibition of ribose uptake but instead
increased the transport about fourfold.
In contrast to the stimulatory effect exerted by glucose
on ribose uptake, maltose transport was found to be
instantaneously arrested when glucose was added to L.
casei wild-type cells transporting maltose. Maltose uptake
was also completely abolished when glucose was added
to the cell suspension 10 min before the addition of
maltose (Fig. 6B). The ptsH1 (S46AHPr) mutant showed
a completely different behaviour to the wild-type strain
(Fig. 6C). Maltose uptake in this strain was slightly higher
and the addition of glucose caused a further increase of
the maltose transport rate. A similar stimulatory effect of
glucose on maltose transport was found for the ptsH2
(S46THPr) mutant (Fig. 6D), whereas no change in the
maltose transport rate following glucose addition was
observed for the ptsH3 (I47THPr) mutant (Fig. 6E). In the
ptsI mutant BL126, which is unable to transport glucose
via the PTS, the presence of glucose exerted no inhibitory
effect on maltose uptake (Fig. 6F). We measured glucose
uptake in the ptsI mutant BL126 and found that glucose
was transported 10 times more slowly than the wild-type
strain (data not shown). The slower glucose uptake and
metabolism was probably responsible for the failure of
glucose to elicit inducer exclusion in the ptsI mutant strain.
By contrast, in a ccpA mutant strain, glucose exerted an
inhibitory effect on maltose uptake identical to that
observed with the wild-type strain. This result clearly
established that CcpA is not involved in glucose-triggered
maltose exclusion.
To make sure that growing the cells for 30 min in
glucose-containing medium had no drastic effect on
expression of the maltose genes, we carried out inducer
exclusion experiments with cells that had not been
exposed to glucose. Under these conditions, addition of
glucose to maltose transporting cells exerted a strong
inhibitory effect on maltose uptake in the wild type and
ccpA mutant strains, although maltose continued to be
slowly taken up by these cells after the addition of
glucose. By contrast, the presence of glucose completely
arrested maltose uptake by cells that had been grown on
glucose for 30 min. However, with the ptsH1, ptsH2 and
ptsH3 mutants grown only on maltose, glucose exerted no
Fig. 5. The effect of the various ptsH mutations on CCR of N-acetylglucosaminidase. The N-acetylglucosaminidase activitiesexpressed in nmol of product formed per min and mg of cells (dryweight) and determined in the L. casei wild type (wt) and the ccpA,ptsH1 (S46 A), ptsH2 (S46T) and ptsH3 (I47T) mutant strains grownin MRS basal medium containing 0.5% glucose or ribose arepresented.
Fig. 4. Growth behaviour of L. casei wild type and ccpA and ptsHmutant strains in MRS basal medium containing 0.1% glucose and0.2% lactose. The symbols represent: filled circles, wild-type BL23;filled squares, ccpA mutant BL71; open circles, ptsH1 mutant BL121;filled triangles, ptsH2 mutant BL122; open triangles, ptsH3 mutantBL123.
Regulatory functions of enzyme I and HPr in L. casei 577
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
inhibitory effect at all on maltose uptake (data not shown),
clearly establishing that the failure of glucose to inhibit
maltose transport in the ptsH mutant strains is not related
to pregrowing the cells in glucose-containing medium.
The observed inhibition of maltose transport could have
been due to elevated secretion of maltose fermentation
products when glucose was added to wild-type cells. In
the ptsH mutants, this glucose effect might have been
less pronounced. To exclude this possibility, we measured
sugar consumption by resting cells that had been grown
on maltose and, for the last 30 min before harvesting the
cells, on maltose and glucose. The results presented in
Fig. 7A confirm that maltose is not utilized in the presence
of glucose by L. casei wild-type cells. Maltose consump-
tion stopped immediately when glucose was added and
the maltose concentration remained constant in the
medium as long as glucose was present. Maltose
consumption restarted only when glucose had completely
disappeared from the medium. By contrast, the addition of
glucose to ptsH1 mutant cells taking up maltose caused
only a short transient inhibition of maltose consumption,
which was followed by the simultaneous utilization of both
sugars (Fig. 7B). Reduced uptake of glucose by the ptsH1
mutant does not seem to be responsible for the absence
of the inhibitory effect of glucose, as in this strain glucose
was utilized slightly faster compared with the wild-type
strain. These results suggest that phosphorylation of Ser-
46 in HPr is necessary for the exclusion of maltose from L.
casei cells by glucose and probably other rapidly
metabolizable carbon sources and that P±Ser±HPr
Fig. 7. Maltose consumption by resting cells of the L. casei wild-typestrain BL23 and the ptsH1 mutant BL121 in the presence or absenceof glucose. The symbols represent: squares, maltose concentration inthe medium in experiments without glucose; diamonds, maltoseconcentration and circles, glucose concentration in the medium whenglucose was added 3 min after the experiment had been started.
Fig. 6. The effect of glucose on maltose and ribose uptake by wildtype and ptsH mutant cells. Ribose transport by the L. casei wild-typestrain BL23 was measured in the absence and presence of glucose(A). Maltose transport in the absence and presence of glucose wasdetermined in the L. casei wild-type strain BL23 (B), the ptsH1 mutantBL121 (C), the ptsH2 mutant BL122 (D), the ptsH3 mutant BL123 (E),the ptsI mutant BL126 (F) and the ccpA mutant BL71 (G). Thesymbols represent: squares, ribose or maltose uptake in the absenceof other sugars; diamonds, ribose or maltose uptake with glucose(10 mM final concentration) added after 10 or 4 min respectively;triangles, the cells were incubated for 10 min in the presence of20 mM glucose before the maltose uptake reaction was started.
578 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
plays an important role in the regulatory phenomenon
called inducer exclusion (Ye et al., 1994a, b; Ye and
Saier, 1995a, b).
Discussion
The major mechanism of CCR in B. subtilis (Deutscher
et al., 1997) is completely different from the c-AMP-
dependent CCR mechanism operative in Gram-negative
bacteria such as E. coli and Salmonella typhimurium
(Postma et al., 1993). The B. subtilis CCR mechanism
was thought to be widely distributed within low G1C
Gram-positive bacteria, as the components involved in
this mechanism, i.e. HPr kinase±phosphatase, HPr,
CcpA and the operator sites cre, could be found in most
low G1C Gram-positive bacteria. The participation of
CcpA in CCR has been established for several Gram-
positive bacteria. including in addition to bacilli Staphylo-
coccus xylosus (Egeter and BruÈckner, 1996), L. casei
(Monedero et al., 1997), Lactobacillus pentosus (Lokman
et al., 1997), Listeria monocytogenes (Behari and Young-
man, 1998) and L. lactis (Luesink et al., 1998). In addition,
spontaneous Streptococcus salivarius ptsH mutants have
been isolated in which the preferential use of glucose and
fructose over lactose and melibiose was abolished
(Vadeboncoeur et al., 1994; Gauthier et al., 1997). In
these mutants, Ile-47 or Met-48 of HPr were found to be
replaced with threonine or valine, respectively, suggesting
a role of HPr in CCR in this organism. Mutants carrying a
disrupted ptsH gene have been constructed with L. lactis,
which were subsequently transformed with plasmids
carrying the wild-type ptsHI operon or ptsHI operons
containing a ptsH allele encoding either Ser-46±Ala or
Ser-46±Asp HPr under control of a nisin-inducible
promoter. Overexpression of the plasmid-encoded ptsH
alleles in the L. lactis ptsH mutant suggested a role of
P±Ser±HPr in CCR of the galactose operon and in
carbon catabolite activation of the las operon in this
organism (Luesink et al., 1999), as CCR was observed
after transformation with the plasmid carrying the wild
type or the Ser-46±Asp ptsH but not with the Ser-46±Ala
ptsH.
In order to confirm the suggested general role of
P±Ser±HPr in CCR in Gram-positive bacteria, we cloned
the ptsH and ptsI genes of L. casei and constructed
chromosomal ptsH mutants altered at the ATP-dependent
phosphorylation site of HPr in a second Gram-positive
organism. As in most other bacteria, ptsH and ptsI seem
to be organized in an operon in L. casei, as a major
transcript corresponding to the size of the combined
genes could be detected. The hybridization bands
observed in the Northern blots indicated that there is
some glucose induction in the wild type and ccpA mutant
and all the mutants tested showed very low expression of
the operon on ribose, which together with the fact that the
ptsI (BL126) mutant does not express this operon is
suggesting that PTS activity may be required for induc-
tion. CcpA did not seem to be implicated in regulation of
ptsHI expression, although the man mutation had some
effect. ptsI mutants were constructed that had lost the
capacity to ferment a large variety of carbohydrates
including fructose, mannose, mannitol, sorbose, sorbitol,
amygdaline, arbutine, salicine, cellobiose, lactose, taga-
tose, trehalose and turanose. Some of these carbohy-
drates might not be taken up by the PTS, but their
transport or metabolism might rather depend on a
functional PTS, as has been observed for glycerol uptake
and metabolism in Gram-positive and Gram-negative
bacteria (Postma et al., 1993).
In order to study the importance of phosphorylation of
HPr at Ser-46 for CCR, diauxic growth and inducer
exclusion, we tried to construct four different L. casei ptsH
mutants in which HPr was expected to be phosphorylated
at Ser-46 to various degrees. Unfortunately, we did not
succeed in obtaining a strain producing the Ser-46±Asp
mutant HPr, which has been shown to resemble
structurally P±Ser±HPr and which would have mimicked
the presence of completely phosphorylated HPr in the
cell. This mutation seems to be toxic in L. casei cells, as
some of the clones potentially containing the Ser-46±Asp
mutation were found to have either undergone a second
mutation, changing, for example, the Asp in position 46 of
HPr to Tyr, or to have went through an abnormal second
recombination deleting the erythromycin resistance cas-
sette, but retaining both the wild type and the Ser-46±Asp
ptsH alleles. The other three L. casei ptsH mutants could
be constructed. The Ser-46±Ala ptsH mutation com-
pletely prevents ATP-dependent phosphorylation of HPr
(Eisermann et al., 1988), whereas the B. subtilis Ser-46±
Thr mutant HPr was found to be slowly phosphorylated by
ATP and the HPr kinase±phosphatase (Reizer et al.,
1989). By contrast the Ile-47±Thr mutant HPr was found
to be normally phosphorylated in S. salivarius cells
(Gauthier et al., 1997). Nevertheless, the Ile-47±Thr
ptsH mutation abolishes the preferential use of glucose
and fructose over lactose and melibiose and causes
derepression of a- and b-galactosidase, suggesting that
this mutation prevents the interaction of P±Ser±HPr with
CcpA in S. salivarius.
When introduced into L. casei, all three ptsH mutations
were found to diminish the repressive effect of glucose on
the activity of N-acetylglucosaminidase (Fig. 5). As with
the ccpA mutant strain BL71 complete relief from CCR
was observed for the ptsH1 mutant BL121 (Ser-46±Ala),
suggesting that CCR of N-acetylglucosaminidase in L.
casei is mediated via the CcpA/P±Ser±HPr-dependent
mechanism. The Ile-47±Thr mutation caused a less
pronounced relief from CCR, which was further weakened
Regulatory functions of enzyme I and HPr in L. casei 579
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
in the Ser-46±Thr mutant. The amount of ATP-phos-
phorylated Ser-46±Thr mutant HPr present in the cell is
difficult to predict. Although probably only slowly phos-
phorylated by the HPr kinase±phosphatase (Reizer et al.,
1989), the threonyl-phosphorylated form of the mutant
HPr might accumulate to relatively high concentrations in
the cell if the kinetic parameters for the dephosphorylation
reaction of the phosphorylated mutant HPr have changed
as well. This might explain why the Ser-46±Thr mutation
always had the weakest releasing effect on CCR and
diauxic growth. Nevertheless, the above results clearly
confirm that phosphorylation of HPr at Ser-46 is impli-
cated in CCR in L. casei and support the concept that the
CcpA/P±Ser±HPr-dependent CCR mechanism is widely
distributed within low G1C Gram-positive bacteria.
The ptsH1 and ptsH3 mutations also diminished the
lag time of the diauxic growth in glucose- and lactose-
containing medium, whereas the ptsH2 mutation (Ser-46±
Thr) had no such effect. A second mechanism involved in
CCR and regulation of diauxie of the L. casei lac operon
seems to be operative as even with a ccpA mutant strain
no complete relief from diauxic growth could be observed
(Fig. 4). This second mechanism probably depends on
P±His±HPr-catalysed phosphorylation of LacT, the tran-
scriptional antiterminator regulating the expression of the
lac operon in L. casei (Gosalbes et al., 1997; Gosalbes
et al., 1999). In B. subtilis, a similar P±His±HPr-
dependent control of the transcriptional regulators LicT
and LevR has been suggested to depend on phosphor-
ylation of HPr at Ser-46, as this regulation disappeared in
ptsH1 mutants (KruÈger et al., 1996; Martin-Verstraete
et al., 1998). The finding that the lag phase during diauxic
growth of a L. casei ptsH1 mutant is longer than in a ccpA
mutant might indicate that a third CcpA-dependent CCR
mechanism for the L. casei lac operon exists. Possibly as
a result of this complex regulation of the lactose system,
the effect of the ptsH mutations on diauxic growth was
clearer when the experiments were carried out with media
containing glucose and either maltose or ribose. With both
sugars, the diauxic growth observed with the wild-type
strain was completely abolished in the three ptsH
mutants.
Nevertheless, the repressive effect of glucose on the
metabolism of ribose and maltose seemed to be mediated
by two different mechanisms. The diauxic growth
observed with glucose and ribose is probably due to the
CcpA/P±Ser±HPr dependent CCR mechanism and not to
inducer exclusion, as the presence of glucose in the
transport buffer exerted a stimulatory effect on ribose
transport (Fig. 6A). A ptsI mutation was also found to
exert a stimulatory effect on ribose or arabinose uptake
and metabolism in L. sakei (Stentz et al., 1997). It is
possible that the presence of dephosphorylated PTS
proteins caused either by ptsI disruption or by the uptake
of a rapidly metabolizable PTS sugar leads to elevated
ribose uptake via an as yet unknown regulatory mechan-
ism. By contrast, maltose uptake was found to be
instantaneously arrested when glucose was added to
the transport buffer, suggesting that maltose metabolism
is regulated by an inducer exclusion mechanism (Fig. 6B),
although it cannot be excluded that expression of the
maltose operon is additionally regulated by the CcpA/P±
Ser±HPr dependent CCR mechanism. Whereas inducer
exclusion is mediated via enzyme IIAGlc in Gram-negative
bacteria, P±Ser±HPr has been suggested to carry out
this function in L. brevis ( Ye et al., 1994a, b; Ye and
Saier, 1995a, b). However, this assumption was based
only on in vitro experiments, which have been carried out
using the heterologous B. subtilis Ser-46±Ala mutant HPr
instead of L. brevis P±Ser±HPr. As a consequence, the
involvement of P±Ser±HPr in inducer exclusion in Gram-
positive bacteria remained questionable. The finding that
the repressive effect of glucose on maltose transport had
completely disappeared in the ptsH1 and ptsH2 mutants
provided for the first time in vivo evidence that P±Ser±
HPr is involved in inducer exclusion in Gram-positive
bacteria. This concept was further supported by the
recent construction of a L. casei hprK mutant unable to
phosphorylate wild-type HPr at Ser-46. As observed with
the ptsH1 and ptsH2 mutant strains, no inhibitory effect of
glucose on maltose uptake could be detected with the
hprk mutant (Dossonnet et al., 2000).
By contrast, this novel inducer exclusion mechanism is
independent of CcpA, as glucose was found to arrest
maltose transport in a ccpA mutant in a way similar to that
observed with the wild-type strain. Surprisingly, although
glucose exerted a short transient stop of maltose uptake
in the Ile-47±Thr ptsH mutant, maltose was subsequently
taken up at a rate similar to that observed in the
experiment carried out in the absence of glucose
(Fig. 6E). This suggests that the Ile-47±Thr ptsH mutation
affects not only the interaction of P±Ser±HPr with CcpA,
but that it also prevents the interaction with proteins of
non-PTS transport systems, which are subject to regula-
tion by inducer exclusion. In HPr of Gram-positive
bacteria, the highly conserved Ile-47, Met-48 and Met-
51 form a surface-exposed hydrophobic patch located
between the two phosphorylation sites Ser-46 and His-15
(Jia et al., 1994). To explain the loss of the repressive
effect of glucose and fructose on the metabolism of
lactose and melibiose in the Ile-47±Thr and Met-48±Val
S. salivarius ptsH mutants, it has been proposed that this
hydrophobic patch might be important for the interaction
of P±Ser±HPr with CcpA (Gauthier et al., 1997), which
was experimentally confirmed by NMR measurements
(Jones et al., 1997). As inducer exclusion is also
abolished in the Ile-47±Thr ptsH mutant, this hydrophobic
patch might as well be important for the interaction of
580 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
P±Ser±HPr with non-PTS permeases regulated by inducer
exclusion.
Experimental procedures
Strains, plasmids and culture conditions
The L. casei strains and plasmids used in this study are listedin Table 1. L. casei cells were grown at 378C under staticconditions in MRS medium (Oxoid) or MRS fermentationmedium (Adsa-Micro) containing 0.5% of the indicatedcarbohydrates. For diauxic growth experiments, L. caseistrains were grown in MRS basal medium containing in 1 l;polypeptone (Difco), 10 g; meat extract (Difco), 10 g; yeastextract (Difco), 5 g; K2HPO4.3H2O, 2 g; sodium acetate, 5 g;di-ammonium citrate, 2 g; MgSO4, 0.1 g; MnSO4, 0.05 g andTween 80, 1 ml. The basal medium was supplemented withdifferent sugars at a final concentration of 0.5%, but for thediauxic growth experiments the sugar concentrations werechanged as indicated in the text. E. coli DH5a was grown withshaking at 378C in Luria±Bertani (LB) medium. Transformedbacteria were plated on the respective solid media containing1.5% agar. The concentrations of antibiotics used for theselection of E. coli transformants were 100 mg ml21 ampi-cillin, and 300 mg ml21 erythromycin and for the selection ofL. casei integrants 5 mg ml21 erythromycin. The sugarutilization pattern of certain strains was determined with theAPI50-CH galeries (Biomerieux).
Recombinant DNA and transformation procedures
Genomic DNA from L. casei strains was purified using thePurogene DNA isolation Kit (Gentra Systems). Restrictionand modifying enzymes were used according to therecommendations of the manufacturers. For DNA sequen-cing, a Perkin Elmer ABI310 automatic sequencer was usedand the reaction conditions and reagents were as recom-mended by the manufacturer. General cloning procedures inE. coli were performed according to Sambrook et al. (1989).L. casei strains were transformed by electroporation with aGene-pulser apparatus (Bio-Rad Laboratories,) as describedbefore (Posno et al., 1991).
PCR amplification, cloning of PCR fragments and reverse
PCR
For PCR amplification of the two fragments comprising part ofthe ptsHI operon, we used a Progene thermocycler (Real)programmed for 30 cycles, including the following threesteps: 30 s at 958C, 30 s at 508C and 1 min at 728C, followedby a final extension cycle at 728C for 5 min. The followingdegenerate oligonucleotides were used for the amplificationof DNA fragments containing part of ptsHI from L. casei:PTS-H2 (5 0-ATG GAA AAR CGN GAR TTY AAY-3 0)(MEKREFN); PTS-I3 (5 0-GCC ATN GTR TAY TGR ATYARR TCR TT-3 0) (NDLIQYTMA); PTS-I4 (5 0-CCR TCN SANGCN GCR ATN CC-3 0) (GIAASDG); where R stands for A orG, Y for C or T, S for C or G and N for any nucleotide.Underlined in parentheses are the N-terminal amino acid
sequence of HPr and the conserved enzyme I sequences,which served to design the primers.
Cloning of PCR fragments was achieved with the Sure-Clone Ligation Kit (Pharmacia Biotech). Cloning of theregions flanking the insertion site of plasmid pRV300 wasperformed as follows. DNA (10 mg) from L. casei BL124(obtained after integration of plasmid pVME800 into thechromosome of BL23) was digested with SacI or HindIII,diluted 500-fold, religated with T4 DNA ligase and differentaliquots were used to transform E. coli DH5a. Plasmid DNAwas isolated from several transformants and subsequentlysequenced. The region upstream of ptsH containing thecomplete promoter could not be obtained with this technique.Reverse PCR was therefore used to get this missing part ofthe ptsHI operon. For this purpose, DNA isolated from the L.casei wild-type strain BL23 was cut with Pst I and religatedwith T4 DNA ligase (Gibco-BRL). Twenty nanograms of theligated DNA and two primers derived from the 5 0 part of ptsHand oriented in opposite directions were used to specificallyamplify using PCR a 2.3 kb fragment containing the upstreamregion of ptsH. The sequence comprising 560 bp upstreamfrom the ptsHI promoter has been determined in this fragment.
RNA isolation and Northern blot analysis
L. casei strains were grown in MRS fermentation mediumsupplemented with 0.5% of the different sugars to anOD550 of between 0.8 and 1. Cells from a 10 ml culturewere collected by centrifugation, washed with 50 mM EDTAand resuspended in 1 ml of Trizol (Gibco BRL). One gram ofglass beads (diameter 0.1 mm) was added and the cells werebroken by shaking the cell suspension in a Fastprep apparatus(Biospec) twice for 45 s. RNA was isolated according to theprocedure recommended by the manufacturer of Trizol andquantified with the RiboGreen kit (Molecular Probes) using aVersaFluorTM fluorometer (Bio-Rad) prior to being separated byformaldehyde-agarose gel electrophoresis and transferred toHybond-N membranes (Amersham).
The RNA probes used in the Northern blots were obtainedeither with the EcoRI fragment of ptsI present in pVME800 orwith the 205 bp HindIII±EcoRV fragment of ptsH cloned inpRV300. As these plasmids are pBluescriptII SK2 deriva-tives, antisense RNA could be synthesized in vitro fromlinearized plasmids with T3 or T7 RNA polymerase using thereagents from the Boehringer Mannheim digoxigenin-RNAlabelling kit. Hybridization (558C), washing and staining wereperformed as suggested by the supplier. For approximativequantification of the relative intensity of the hybridizing bandsin the Northern blots, rRNA bands observed after the transferin the methylene blue-stained membranes were used as aninternal standard for each sample. For this purpose, bands inthe films after the Northern blots and the stained 23S rRNAbands in the membranes were scanned with an HP ScanJet5100C and quantified by densitometry with the SCION IMAGE
software (Scion). Then, the relative band intensities in thefilms were calculated for each sample.
Primer extension experiments
Using 20 ml of reverse transcriptase buffer (Amersham)
Regulatory functions of enzyme I and HPr in L. casei 581
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
containing 0.5 mM dNTP, 15 mg of RNA was annealed for5 min at 658C with 0.2 pmol of a [32P]-5 0-labelled oligonu-cleotide with the sequence 5 0-ACT TGC TTG CCT GTA CC-3 0 (see PTS-P.E. in Fig. 1). The subsequent extensionreaction was carried out at 428C for 1 h using 10 U of AMVreverse transcriptase (Amersham). The cDNA product andthe products of sequencing reactions performed with thesame primer were loaded on a 6% polyacrylamide26 M ureagel and labelled bands were detected by autoradiography.
Enzymatic assays
For the N-acetylglucosaminidase assays, permeabilized L.casei cells were prepared following a previously describedmethod (Chassy and Thompson, 1983). The N-acetyl-glucosaminidase assays were carried out at 378C in a volumeof 250 ml, containing 10 mM potassium phosphate pH 6.8,1 mM MgCl2, 5 mM p-nitrophenyl N-acetyl-b-D-glucosami-nide (Sigma) and 5 ml of permeabilized cells. The reactionwas stopped with 250 ml of 5% Na2CO3 and the OD420 wasmeasured. Protein concentrations were determined with theBio-Rad dye-binding assay.
Protein purification
Cells from an overnight culture (1 l of MRS medium) werecentrifuged and washed twice with 20 mM Tris-HCl, pH 7.4.The cells were resuspended in 20 mM ammonium bicarbo-nate buffer, pH 8 (2 ml per gram of cell pellet), sonicated(Branson Sonifier 250) and then centrifuged to remove thecell debris. As HPr resists heat treatment, the supernatantwas kept at 708C for 5 min to precipitate most of the otherproteins. An additionnal centrifugation step was performed toremove the heat-denatured proteins. The supernatant wasloaded on a Sephadex G-75 column (42 cm 1 1.6 cm)equilibrated with 20 mM ammonium bicarbonate, pH 8,which was eluted with the same buffer, and fractions of1.5 ml were collected. To test for the presence of HPr inthese fractions, a mutant complementation assay with theStaphylococcus aureus ptsH mutant strain S797A wascarried out (Hengstenberg et al., 1969). HPr activity wasdetected in fractions 48±56. These fractions were pooled andconcentrated to a final volume of 500 ml.
Half of the partially purified HPr was separated by reversephase chromatography on a Vydac C-18 HPLC column(300 AÊ , 250 mm 1 4.6 mm; Touzart et Matignon). Solvent Awas an aqueous solution of 0.1% (v/v) trifluoroacetic acid andsolvent B contained 80% acetonitrile and 0.04% trifluoroace-tic acid. Proteins were eluted with a linear gradient from 5% to100% of solvent B in 60 min at a flow rate of 500 ml min21.
Fractions with a volume of about 500 ml were collectedmanually. The presence of HPr in the fractions was tested bya PEP-dependent phosphorylation assay containing 10 mMMgCl2, 50 mM Tris-HCl, pH 7.4, 10 ml aliquots of thefractions, 10 mM [32P]-PEP and 1.5 mg of B. subtilis enzymeI(His)6. Enzyme I(His)6 and HPr(His)6 of B. subtilis werepurified by ion chelate chromatography on a Ni±NTASepharose column (Qiagen) after expression from plasmidspAG3 and pAG2 respectively (Galinier et al., 1997).HPr(His)6 from B. subtilis was used as a standard in the
phosphorylation reactions. [32P]-PEP was prepared from[g-32P]-ATP via the pyruvate kinase exchange reaction(Roossien et al., 1983). The assay mixtures were incubatedfor 10 min at 378C and separated on 15% polyacrylamidegels containing 1% SDS (Laemmli, 1970). After drying thegels, radiolabelled proteins were detected by autoradiogra-phy. HPr was found to elute at 60% acetonitrile in fractions44±46. These fractions were pooled, lyophilized and aliquotscorresponding to < 0.5 nmol of HPr were used to determinethe first 21 N-terminal amino acids of HPr by automatedEdman degradation on a 473-A Applied Biosystems micro-sequencer.
Construction of ptsH mutants
Site-directed mutagenesis was performed in order to replacethe codon for Ser-46 of L. casei ptsH with a codon for Ala,Asp or Thr and the codon for Ile-47 with a codon for Thr. Forthis purpose, PCR amplification was carried out, using as atemplate the plasmid pVMH1 containing the L. casei wild-type pstH gene as well as the 5 0 part of the ptsI gene and asprimers the reverse primer of pBluescript (Stratagene) andone of the following oligonucleotides: 5 0ptsHS46A (5 0-AAGAGC GTT AAC TTG AAG GCT ATC ATG GGC G-3 0);5 0ptsHS46T (5 0-AAG AGC GTT AAC TTG AAG ACT ATCATG GGC G-3 0); 5 0ptsHS46D (5 0-AAG AGC GTT AAC TTGAAG GAT ATC ATG GGC G-3 0); 5 0ptsHI47T (5 0-AAG AGCGTT AAC TTG AAG TCT ACC ATG GGC G-3 0). In theseoligonucleotides, the codons for Ser-46 or Ile-47 werereplaced by the indicated codon (underlined). The resulting1.4 kb PCR fragments containing the ptsH alleles (fromcodon 40) and the 5 0 part of ptsI were digested with HpaI (theHpaI site present in ptsH before codon 46 is indicated initalics in the above primers) and SacI and used to replace thewild-type 1.4 kb HpaI±SacI fragment in pVMH1.
In order to confirm the presence of the mutations, thesequence of the ptsH alleles was determined in the fourconstructed plasmids. To eliminate mutations possiblyintroduced in the ptsI gene by the PCR amplification, the1.35 kb BalI±SacI fragment from pVMH1 was used toreplace the corresponding fragment in each of the fourplasmids containing the various ptsH alleles. A unique BalIsite is present 27 bp behind codon 46 of L. casei ptsH inpVMH1 and the pVMH1 derivatives carrying the differentptsH alleles.
Sugar transport and sugar consumption by resting cells
Cells were grown to mid-exponential phase in MRS fermen-tation broth containing 0.5% of the indicated sugars. Thirtyminutes before harvesting the cells, glucose was added to afinal concentration of 0.5% to fully induce the synthesis of theglucose-specific PTS transport proteins. Experiments werealso carried out without pregrowing the cells in the presenceof glucose. Cells were washed twice with 50 mM sodiumphosphate buffer, pH 7, containing 10 mM MgCl2 andresuspended in 50 mM Tris-maleate buffer, pH 7.2, contain-ing 5 mM MgCl2. Transport assays were carried out in 1 ml ofthis latter buffer containing 1% peptone and 0.6 mg of cells(dry weight). Samples were preincubated for 5 min at 378C
582 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
prior to adding [14C]-labelled sugars (0.5 mCi mmol21,Isotopchim, Ganagobie-Peyruis) to a final concentration of1 mM. Samples of 100 ml were withdrawn at different timeintervals, rapidly filtered through 0.45 mm pore-size filters,washed twice with 5 ml of cold Tris-maleate buffer and theradioactivity retained was determined by liquid scintillationcounting.
In order to follow the sugar consumption by the L. caseiwild type and ptsH1 mutant strains, cells were grown andharvested as described for the transport studies and 18 mg ofcells (dry weight) were resuspended in 5 ml of 50 mM sodiumphosphate buffer, pH 7. After 5 min incubation at 378C,maltose and glucose were added to a final concentration of0.04% and 0.2% respectively. Samples of 300 ml werewithdrawn at different time intervals, boiled for 10 min andclarified by centrifugation. The sugar content in the super-natant was determined with a coupled spectrophotometrictest using a-glucosidase and hexokinase/glucose-6-P dehy-drogenase as recommended by the supplier (Boehringer).
Acknowledgements
We are thankful to M.-M. Boutillon for amino acid sequencing, to
A. Lepingle for DNA sequencing and to M. Zagorec for supplying
plasmid pRV300. A. Galinier is acknowledged for valuable
discussions. This research was supported by the Spanish
CICYT project No. ALI98-714, the CNRS, the INRA, the INA-
PG and the European Community Biotech Programme contract
No. BIO4-CT96-0380 and by the Medical Research Council of
Canada. R.V. and V.M. were supported by a grant from the
Spanish Secretariat of Universities and Research. During his
postdoctoral stay in the Laboratoire de GeÂneÂtique des Micro-
organismes at the INRA-CNRS, V.M. received a grant from the
Spanish Ministry of Education and Culture.
References
Behari, J., and Youngman, P. (1998) A homolog of CcpA
mediates catabolite control in Listeria monocytogenes but notcarbon source regulation of virulence genes. J Bacteriol 180:
6316±6324.
Brochu, D., and Vadeboncoeur, C. (1999) The HPr (Ser) kinase
of Streptococcus salivarius: purification, properties, and
cloning of the hprK gene. J Bacteriol 181: 709±717.
Charrier, V., Buckley, E., Parsonage, D., Galinier, A., Darbon, E.,
Jaquinod, M., et al. (1997) Cloning and sequencing of two
enterococcal glpK genes and regulation of the encodedglycerol kinases by phosphoenolpyruvate dependent, phos-
photransferase system-catalyzed phosphorylation of a single
histidyl residue. J Biol Chem 272: 14166±14174.
Chassy, B.M., and Thompson, J. (1983) Regulation of lactose-
phosphoenolpyruvate-dependent phosphotransferase system
and b-D-phosphogalactoside galactohydrolase activities inLactobacillus casei. J Bacteriol 154: 1195±1203.
Davis, E.O., and Henderson, P.J.E. (1987) The cloning and DNAsequence of the gene xylE for xylose-proton symport in
Escherichia coli K12. J Biol Chem 262: 13928±13932.
Deutscher, J., Pevec, B., Beyreuther, K., Kiltz, H.-H., and
Hengstenberg, W. (1986) Streptococcal phosphoenolpyru-
vate-sugar phosphotransferase system: amino acid sequence
and site of ATP-dependent phosphorylation of HPr. Biochem-istry 25: 6543±6551.
Deutscher, J., Bauer, B., and Sauerwald, H. (1993) Regulation of
glycerol metabolism in Enterococcus faecalis by phosphoe-nolpyruvate-dependent phosphorylation of glycerol kinase
catalyzed by enzyme I and HPr of the phosphotransferase
system. J Bacteriol 175: 3730±3733.
Deutscher, J., Reizer, J., Fischer, C., Galinier, A., Saier, M.H. Jr,and Steinmetz, M. (1994) Loss of protein kinase-catalyzed
phosphorylation of HPr, a phosphocarrier protein of the
phosphotransferase system, by mutation of the ptsH gene
confers catabolite repression resistance to several catabolicgenes of Bacillus subtilis. J Bacteriol 176: 3336±3344.
Deutscher, J., Fischer, C., Charrier, V., Galinier, A., Lindner, C.,
Darbon, E., et al. (1997) Regulation of carbon metabolismin Gram-positive bacteria by protein phosphorylation. Fol
Microbiol 42: 171±178.
Dossonnet, V., Monedero, V., Zagorec, M., Galinier, A., PeÂrez-MartõÂnez, G., and Deutscher, J. (2000) Phosphorylation of HPr
by the bifunctional HPr kinase/phosphatase from Lactobacillus
casei controls catabolite repression and inducer exclusion, but
not inducer expulsion. J Bacteriol (In press.)
Egeter, O., and BruÈckner, R. (1996) Catabolite repression
mediated by the catabolite control protein CcpA in Staphylo-
coccus xylosus. Mol Microbiol 21: 739±749.
Eisermann, R., Deutscher, J., Gonzy-TreÂboul, G., and Hengsten-
berg, W. (1988) Site-directed mutagenesis with the ptsH gene
of Bacillus subtilis. Isolation and characterization of heat-stable
proteins altered at the ATP-dependent regulatory phosphoryla-tion site. J Biol Chem 263: 17050±17054.
Fujita, Y., and Miwa, Y. (1994) Catabolite repression of the
Bacillus subtilis gnt operon mediated by the CcpA protein. JBacteriol 176: 511±513.
Galinier, A., Haiech, J., Kilhoffer, M.-C., Jaquinod, M., StuÈ lke, J.,
Deutscher, J., et al. (1997) The Bacillus subtilis crh gene
encodes a HPr-like protein involved in carbon cataboliterepression. Proc Natl Acad Sci USA 94: 8439±8444.
Galinier, A., Kravanja, M., Engelmann, R., Hengstenberg, W.,
Kilhoffer, M.-C., Deutscher, J., et al. (1998) New protein kinaseand protein phosphatase families mediate signal transduction
in bacterial catabolite repression. Proc Natl Acad Sci USA 95:
1823±1828.
Galinier, A., Deutscher, J., and Martin-Verstraete, I. (1999)Phosphorylation of either Crh or HPr mediates catabolite
repression and binding of CcpA to the cre of the Bacillus
subtilis xyn operon. J Mol Biol 286: 307±314.
Gauthier, M., Brochu, D., Eltis, L.D., Thomas, S., and Vadebon-
coeur, C. (1997) Replacement of isoleucine-47 by threonine in
the HPr protein of Streptococcus salivarius abrogates the
preferential metabolism of glucose and fructose over lactoseand melibiose but does not prevent the phosphorylation of HPr
on serine-46. Mol Microbiol 25: 695±705.
Gosalbes, M.J., Monedero, V., Alpert, C.-A., and Perez-Martinez,G. (1997) Establishing a model to study the regulation of the
lactose operon in Lactobacillus casei. FEMS Microbiol Lett
148: 83±89.
Gosalbes, M.J., Monedero, V., and Perez-Martinez, G. (1999)
Elements involved in catabolite repression and substrate
induction of the lactose operon in Lactobacillus casei. J
Bacteriol 181: 3928±3934.
GoÈsseringer, R., KuÈster, E., Galinier, A., Deutscher, J., and
Hillen, W. (1997) Cooperative and non-cooperative DNA
binding modes of catabolite control protein CcpA from Bacillus
Regulatory functions of enzyme I and HPr in L. casei 583
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
megaterium result from sensing two different signals. J Mol Biol266: 665±676.
Hengstenberg, W., Penberthy, W.K., Hill, K.L., and Morse, M.L.
(1969) Phosphotransferase system of Staphylococcus aureus:
its requirement for the accumulation and metabolism of
galactosides. J Bacteriol 99: 383±388.
Henkin, T.M., Grundy, F.J., Nicholson, W.L., and Chambliss,G.H. (1991) Catabolite repression of a-amylase gene expres-
sion in Bacillus subtilis involves a trans-acting gene product
homologous to the Escherichia coli lacI and galR repressors.
Mol Microbiol 5: 575±584.
Hueck, C.J., Hillen, W., and Saier,M.H. Jr (1994) Analysis of acis-active sequence mediating catabolite repression in Gram-
positive bacteria. Res Microbiol 145: 503±518.
Jia, Z., Quail, J.W., Delbaere, L.T.J., and Waygood, E.B. (1994)
Structural comparison of the histidine-containing phosphocar-
rier protein HPr. Biochem Cell Biol 72: 202±217.
Jones, B.E., Dossonnet, V., KuÈster, E., Hillen, W., Deutscher, J.,and Klevit, R.E. (1997) Binding of the catabolite repressor
protein CcpA to its DNA target is regulated by phosphorylation
of its corepressor HPr. J Biol Chem 272: 26530±26535.
Kim, J.-H., Voskuil, M.I., and Chambliss, G.H. (1998) NADP,
corepressor for the Bacillus catabolite control protein CcpA.Proc Natl Acad Sci USA 95: 9590±9595.
Kravanja, M., Engelmann, R., Dossonnet, V., BluÈggel, M., Meyer,
H.E., Frank, R., et al. (1999) The hprK gene of Enterococcus
faecalis encodes a novel bifunctional enzyme: the HPr kinase/
phosphatase. Mol Microbiol 31: 59±66.
KruÈger, S., Gertz, S., and Hecker, M. (1996) Transcriptionalanalysis of bglPH expression in Bacillus subtilis: evidence
for two distinct pathways mediating carbon catabolite repression.
J Bacteriol 178: 2637±2644.
Laemmli, U.K. (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227: 680±685.
Leloup, L., Ehrlich, S.D., Zagorec, M., and Morel-Deville, F.
(1997) Single-crossover integration in the Lactobacillus sake
chromosome and insertional inactivation of the ptsI and lacL
genes. Appl Environ Microbiol 63: 2117±2123.
Lindner, C., Galinier, A., Hecker, M., and Deutscher, J. (1999)Regulation of the activity of the Bacillus subtilis antiterminator
LicT by multiple PEP-dependent, enzyme I- and HPr-catalysed
phosphorylation. Mol Microbiol 31: 995±1006.
Lokman, B.C., Heerikhuisen, M., Leer, R.J., van den Broek, A.,
Borsboom, Y., Chaillou, S., et al. (1997) Regulation ofexpression of the Lactobacillus pentosus xylAB operon. J
Bacteriol 179: 5391±5397.
Luesink, E.J., van Harpen, R.E.M.A., Grossiord, B.P., Kuipers,
O.P., and de Vos, W.M. (1998) Transcriptional activation of the
the glycolytic las operon and catabolite repression of the galoperon in Lactococcus lactis are mediated by the catabolite
control protein CcpA. Mol Microbiol 30: 789±798.
Luesink, E.J., Beumer, C.M., Kuipers, O.P., and de Vos, W.M.
(1999) Molecular characterization of the Lactococcus lactis
ptsHI operon and analysis of the regulatory role of HPr. JBacteriol 181: 764±771.
Martin-Verstraete, I., Charrier, V., StuÈ lke, J., Galinier, A., Erni, B.,
Rapoport, G., et al. (1998) Antagonistic effects of dual PTS-
catalysed phosphorylation on the Bacillus subtilis transcrip-
tional activator LevR. Mol Microbiol 28: 293±303.
Martin-Verstraete, I., Deutscher, J., and Galinier, A. (1999)Phosphorylation of HPr and Crh by HprK, early steps in the
catabolite repression signalling pathway for the Bacillus subtilis
levanase operon. J Bacteriol 181: 2966±2969.
Monedero, V., Gosalbes, M.J., and Perez-Martinez, G. (1997)Catabolite repression in Lactobacillus casei ATCC 393 is
mediated by CcpA. J Bacteriol 179: 6657±6664.
Pao, S., Paulsen, S.I.T., and Saier, M.H. Jr (1998) Major
facilitator superfamily. Microbiol Mol Biol Rev 62: 1±34.
Posno, M., Leer, R.J., van Luijk, N., van Giezen, M.J.F.,
Heuvelmans, P.T.H.M., Lokman, B.C., et al. (1991) Incompat-
ibility of Lactobacillus vectors with replicons derived from small
cryptic Lactobacillus plasmids and segregational instability ofthe introduced vectors. Appl Environ Microbiol 57: 1822±1828.
Postma, P.W., Lengeler, J.W., and Jacobson, G.R. (1993)Phosphoenolpyruvate: carbohydrate phosphotransferase sys-
tems of bacteria. Microbiol Rev 57: 543±594.
Reizer, J., Sutrina, S.L., Saier, M.H. Jr, Stewart, G.C., Peter-
kofsky, A., and Reddy, P. (1989) Mechanistic and physiological
consequences of HPr (ser) phosphorylation on the activities ofthe phosphoenolpyruvate: sugar phosphotransferase system in
Gram-positive bacteria: studies with site-specific mutants of
HPr. EMBO J 8: 2111±2120.
Reizer, J., Hoischen, C., Titgemeyer, F., Rivolta, C., Rabus, R.,
StuÈ lke, J., et al. (1998) A novel protein kinase that controls
carbon catabolite repression in bacteria. Mol Microbiol 27:1157±1169.
Roossien, F.F., Brink, J., and Robillard, G.T. (1983) A simpleprocedure for the synthesis of [32P]phosphoenolpyruvate via
the pyruvate kinase exchange reaction at equilibrium. Biochim
Biophys Acta 760: 185±187.
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular
Cloning A Laboratory Manual, 3rd edn. Cold Spring Harbor,NY: Cold Spring Harbor Laboratory Press.
Stentz, R., Lauret, R., Ehrlich, S.D., Morel-Deville, F., andZagorec, M. (1997) Molecular cloning and analysis of the ptsHI
operon in Lactobacillus sake. Appl Environ Microbiol 63: 2111±
2116.
StuÈ lke, J., Arnaud, M., Rapoport, G., and Martin-Verstraete, I.
(1998) PRD 2 a protein domain involved in PTS-dependent
induction and carbon catabolite repression of catabolic operonsin bacteria. Mol Microbiol 28: 865±874.
Tortosa, P., Aymerich, S., Lindner, C., Saier, M.H. Jr, Reizer, J.,and Le Coq, D. (1997) Multiple phosphorylation of SacY, a
Bacillus subtilis transcriptional antiterminator negatively con-
trolled by the phosphotransferase system. J Biol Chem 272:
17230±17237.
Vadeboncoeur, C., Gauthier, L., Gagnon, G., Leduc, A., Brochu,
D., Lapointe, R., et al. (1994) Properties of a Streptococcussalivarius spontaneous mutant in which the methionine at
position 48 in the protein HPr has been replaced by a valine.
J Bacteriol 176: 524±527.
Veyrat, A., Monedero, V., and Perez-Martinez, G. (1994) Glucose
transport by the phosphoenolpyruvate: mannose phospho-transferase system in Lactobacillus casei ATCC 393 and its
role in carbon catabolite repression. Microbiology 140: 1141±
1149.
Wittekind, M., Reizer, J., Deutscher, J., Saier, Jr, M.H. and Klevit,
R.E. (1989) Common structural changes accompagny the
functional inactivation of HPr by seryl phosphorylation or byserine to aspartate substitution. Biochemistry 28: 9908±9912.
Ye, J.-J., and Saier, M.H. Jr (1995a) Cooperative binding oflactose and the phosphorylated phosphocarrier HPr (Ser-P) to
the lactose/H1 symport permease of Lactobacillus brevis. Proc
Natl Acad Sci USA 92: 417±421.
Ye, J.-J., and Saier, M.H. Jr (1995b) Allosteric regulation of the
glucose: H1 symporter of Lactobacillus brevis: cooperative
584 R. Viana et al.
Q 2000 Blackwell Science Ltd, Molecular Microbiology, 36, 570±584
binding of glucose and HPr (ser-P). J Bacteriol 177: 1900±1902.
Ye, J.-J., Neal, J.W., Cui, X., Reizer, J., and Saier, Jr, M.H.
(1994a) Regulation of the glucose: H1 symporter by
metabolite-activated ATP-dependent phosphorylation of HPrin Lactobacillus brevis. J Bacteriol 176: 3484±3492.
Ye, J.-J., Reizer, J., Cui, X., and Saier, M.H. Jr (1994b) ATP-dependent phosphorylation of serine-46 in the phosphocarrier
protein HPr regulates lactose/H1 symport in Lactobacillus
brevis. Proc Natl Acad Sci USA 91: 3102±3106.
Top Related