An impor tant role for FlhDC in the control of
nucA expr ession, cell
division and flagellar synthesis in
Serratia marcescens
Jia-Hurng Liu(1), Meng-Jiun Lai(1), Sunny Ang (1), Jwu-Ching Shu (1), Poo-Chi Su (1),
Yu-Tze Horng(1), Wen-Ching Yi (1), Hsin-Chih Lai (1,2) (*), Kwen-Tay Luh (2),
Shen-Wu Ho(1,2), and Simon Swift (3)
(1)School and Graduate Institute of Medical Technology,
College of Medicine, National Taiwan University, Taipei, Taiwan, R.O.C. (2) Department of Laboratory Medicine,
National Taiwan University Hospital, Taipei, Taiwan, R.O.C.
(3) Institute of Infections and Immunity, University of Nottingham, Queen’s
Medical Centre, Nottingham, NG7 2UH and School of Pharmaceutical Sciences, University Park, Nottingham, NG7 2RD, U. K.
Key words: Serratia marcescens, flhDC, nucA expression, Cell division, Flagella synthesis.
running title:flhDC controls nucA expression and cell differentiation flhDC Genbank accession number: AF077334
* corresponding author/to whom proofs should be sent: tel. +886-2-23970800 extn 6931 / fax. +886-2-23711574 e-mail: [email protected]
SUMMARY
The gene products of the flhDC operon are r egulator y proteins exer ting global
effects in enter ic bacter ia, especially in the ar eas of motility and cell division. The role of the flhDC operon was investigated in Serratia marcescens. In addition to
cell division and flagellar synthesis, the regulation of expression of one of the vir ulence factor s, nuclease (encoded by the nucASm gene) was obser ved. Inter r uption of the chromosomal flhDC operon in S. marcescens CH-1 resulted in
non-flagellated cells with aber r ant cell-division that were nuclease negative. Expression of nucASm and the other phenotypes were restored to the flhDC mutant in-tr ans by flhDC in multi-copy plasmid. Fur ther study showed that flhDC in multi-copy induces the for mation of differentiated (polyploid aseptate
cells with per itr ichous flagella over -synthesis) cells in broth culture of minimal growth medium. Expression of flhDC shows evidence of positive auto-regulation,
and is growth-phase dependent (up-regulated in ear ly log phase). The regulation of a PflhDC::luxAB genomic fusion was dependent upon environmental
temper ature (inhibited when the bacter ial cell culture was shifted from 30 to 37 ℃) and osmolar ity (inhibited by increasing salt concentr ation), but was not influenced by glucose catabolite repression as has been descr ibed for Salmonella.
INTRODUCTION
The products of the flhDC operon, FlhD and FlhC, act as global gene regulators in enteric bacteria. The expression of many genetic determinants involved in the processes of cell division, cell differentiation, swarming/swimming motility and virulence is controlled by flhDC (Macnab, 1992; Liu and Matsumura, 1994; Givskov et al., 1995; Eberl et al., 1996; Pruss and Matsumura, 1996; Pruss et al., 1997; Givskov et al., 1998; Dufour et al., 1998). The flhDC operon was first noticed as an important regulator in the hierarchical system controlling the synthesis of the bacterial flagellum including Escherichia coli (Macnab, 1992). Subsequently, homologues have been described in Shigella species, Salmonella typhimurium, Serratia liquefaciens and Proteus mirabilis (Givskov et al., 1995; Al Mamun et al., 1996; Furness et al., 1997; Kutsukake, 1997). In addition to the essential function of flhDC operon in the control of bacterial motility, it was also found to regulate important functions outside bacterial flagellar system. Pruss and Matsuyama (1996) have shown that the transcriptional activation by FlhD/FlhC is also involved in the process of cell division and affects growth rate in E. coli. They further showed that in E. coli, while flagellar synthesis is dependent on the tetrameric FlhD/FlhC complex, cell division is dependent on cadA under the control of FlhD only (Pruss et al., 1997). Pleiotropic effects attributable to FlhD/FlhC regulation have also been observed in Serratia liquefaciens (Givskov et al., 1995), with the induction of phospholipase synthesis and swarm-cell differentiation (flagellar over-synthesis and cell elongation) observed when flhDC are over-expressed in an LB rich broth culture (Eberl et al., 1996).
We are interested in the roles that flhDC operon may play in the process of swarm-cell differentiation and virulence factor expression in Serratia marcescens, which is
an important opportunistic human pathogen and plays an important role in the infection (Khan et al., 1997; Peeters et al., 1997). In S. marcescens and other related bacteria, the differentiation of swarmer-cells involves a change from short motile vegetative cells with a few peritrichous flagella into multi-nucleate, aseptate swarm cells of a size up to 40~80 times the vegetative cell and and over-production of surface flagella (Harshey, 1994; Lai et al., 1997b). Although Eberl et al., (1996) have shown the important role that the flhDC operon performs in the differentiation of S. liquefaciens swarm-cells, their work did not rule out a role for uncharacterized factors in the media in this process. Indeed, using cell elongation, flagella over-synthesis and expression pattern of hag (the gene encoding the flagellin subunit) as the markers for swarmer-cell differentiation, we have demonstrated that cells with differentiation characteristics are observed in LB plate and broth cultures without the artificial over-expression of the flhDC operon (Lai et al., 1997b). It is thus possible that cell differentiation is growth–phase dependent and occurs when flhDC operon expresses at its peak.
S. marcescens produces a number of virulence factors including protease, chitinase and lipase, and most of them were substrate-regulated (Chen et al., 1992). Unlike most catabolic enzymes, the expression of nuclease gene, nucA, in S. marcescens is reported not to be regulated by substrate catabolite repression, but by 1) a SOS-like system (Ball et al., 1990) and 2) a growth-phase dependent system, which is independent of SOS induction (Chen et al., 1992). Significantly, the dominant genetic regulator of nucASm expression remains to be identified. As there is evidence that
expression of the virulence factor phospholipase was regulated by flhDC operon, we thus investigated the effect of flhDC operon on nucASm expression.
marcescens, we asked the following questions: (i) Does flhDC, like those reported from other bacteria, regulate cell division and flagellar synthesis in S. marcescens? (ii) Does flhDC play a dominant role in virulence factor (particularly nucASm) expression?
(iii) What are the factors that regulate the expression of flhDC? (iv) Are there any environmental factors, such as solid surface contact, or any putative signal molecules existing in the rich media that are essential to initiate the formation of differentiated swarm-cells? To answer these questions, we have cloned and sequenced the flhDC from S. marcescens and begun to correlate its relationship with cell division, morphological change, flagellar synthesis and expression of the nuclease gene nucASm.
Furthermore, by construction of an PflhDC::luxAB genomic fusion we have identified some factors influencing the activity of the flhDC promoter.
MATERIALS AND METHODS
Bacter ial str ains and plasmids
Table 1.
Bacterial strains Relevant genotype/phenotype Refer ence/sour ce
S. marcescens CH-1 The wild-type cells This Lab.
S. marcescens S-1 PhagSm::luxAB in the chromosome of S. marcescens
CH-1
Lai et al., (1997b). S. marcescens F-1 flhD::SM, derived from S. marcescens CH-1 This study
S. marcescens F-3 PflhDCSm::luxAB in the chromosome of S.
marcescens CH-1
This study
S. marcescens F-4 PflhDCSm::luxAB in the chromosome of S.
marcescens F-1
This study
S. marcescens N-1 PnucASm::luxAB in the chromosome of S.
marcescens CH-1
This study
S. marcescens N-2 PnucASm::luxAB in the chromosome of F-1 This study
S. marcescens S-2 flhD::SM in the chromosome of S-1 This study Plasmids
pCR 2.1 TA cloning vector; AmpR; KmR. ColE1 ori. Invitrogen
pZero 2.1 Cloning vector; KmR
; ColE1 ori. The ccdB
(Bernard, 1996) was used as the screening marker.
Invitrogen
pBCSK+ Cloning vector; CmR
; ColE1 ori. Stratagene
pBIISK+ Cloning vector; AmpR
; ColE1 ori. Stratagene
pACYC184 Cloning vector; TcR
; CmR
. p15A ori. Chang and Cohen (Chang and Cohen, 1978)
pUT miniTn5Km A suicide vector for conjugation and transposition of the inserted DNA after modification (Lai et al.,
1997b).
de Lorenzo et al., (1990). de
Lorenzo and Timmis (1994). pJH03 A 5kb Chromosomal BamHI fragment containing
the complete flhDCSm operon cloned in pBIISK+.
This study
pJH04 A PCR with primers
(GCATGCGTGTACATCCATACACG-3’ and 5’-AACAATGTGGATGGAAGGTGG-3’) amplified a 1.5kb DNA fragment bearing the complete
flhDCSm operon including its native promoter. This
fragment was T-cloned into pCR2.1 and the gene direction was with Placin the vector.
This study
pJH05 A 1.5kb EcoRI fragment (derived from pJH04)
bearing the complete flhDCSm operon with its
native promoter cloned into pBCSK+. The gene direction was with Plac in the vector. Copy number
of this plasmid is supposed to be between 15~20.
Media and growth conditions
Luria-Bertani (LB) medium was used for routine growth of both E. coli and S. marcescens. E. coli was routinely grown at 37o
C and S. marcescens at 30o
C. Cell density was assayed by measuring absorbance at 600 nm. Minimal growth media (MGM): [5x M9 salt 200 ml, 1 M MgSO4 2 ml, 1M CaCl2 0.1 ml, 20 % glucose 20 ml, d.d. H20 up to 1,000 ml] (Maniatis et al., 1989).
DNase, phospholipase and plate gelatin assay
DNAse Test Agar medium (DIFCO) was inoculated with 100 µl of a saturated overnight bacterial culture. After culture for 6 to 8 h, levels of the nuclease production were estimated by measurement of the halo size around bacterial colonies formed after addition of 1 ml of 1N HCl. For a quantitative analysis the microplate assay described by Chen et al., (1992) was performed. The phospholipase assay was performed on egg-yolk agar plate (DIFCO) (Koneman et al., 1997). The plate gelatin assay was performed on 1.5 % LB plate containing 0.1 % gelatin (SIGMA) (Lai, 1994).
Enzymes and chemicals
Restriction and modification enzymes were purchased from Boehringer Mannheim.
Taq polymerase and PCR related products were from Perkin Elmer or Takara Biomedicals. Other chemicals were purchased from SIGMA.
DNA manipulation and analysis
DNA was sequenced by an ABI 373A DNA Sequencer with the Taq DyeDeoxy Terminator Cycle sequencing kit (Perkin-Elmer). DNA/amino acid sequence analysis
was achieved by GCG (version 9.1). Unless mentioned specifically, standard protocols were used for Southern hybridization, cloning, the isolation of plasmid and chromosomal DNA, transformation, PCR, restriction digests, elution and ligation of DNA (Maniatis et al., 1989). Hybridization and washes were carried out at 68o
C. DNA detection was achieved by DIG system (Roche). S. marcescens CH-1 chromosomal DNA library was constructed by lambda DashII system according to the protocol supplied by the manufacturer (Stratagene).
Inactivation of flhD
A DNA fragment containing the complete flhD structural gene and a partial flhC were amplified in a PCR reaction from S. marcescens CH-1 chromosomal DNA template using the amplimers (5’-CATCCATACACGTTGGTTTACGCT-3’/5’TGCTCGTCCAGCAGTTGAGGAATA-3’). The fragment was then cloned into pCR2.1 vector. A SmaI digested streptomycin-resistant gene (SM) was inserted into the SalI site of the amplified DNA fragment by blunt-ended ligation. Through transfer by modified pUT suicide vector (Lai et al., 1997b) and homologous recombination (de Lorenzo and Timmis, 1994), the wild-type flhD was replaced by the recombinant DNA fragment to create a new strain designated F-1. Similar procedures were performed on strain S-1 (main genetic feature, i.e. PhagSm::luxAB, see Table 1) (Lai et
al., 1997b) to form S-2.
Constr uction of isogenic str ains N-1 and N-2
The procedures were similar to those described in Lai et al., (1997b) for the
creation of a chromosomal PhagSm::luxAB fusion. Primers (5'-CTGTTGGACGCCGTTTTTATTT-3' and 5'-ATTCATATCCTCAATAAGTTAA-3')
were designed from the promoter region of nuclease gene (Chen et al., 1992), and a 170 bp promoter region of the nuclease structure gene nucASm was PCR amplified, T-cloned, sequenced, and then ligated in the same direction into the upstream region of bacterial luxAB reporter genes as a NotI/SfiI fragment. The PnucASm::luxAB fragment was randomly transferred into the chromosome of wild-type S. marcescens through conjugation and transposition.
Detection of lucifer ase activity
A cell suspension, 100 µl, was added to Con's tubes (Genetek, Taiwan) containing 900 µl 0.85 % normal saline before measurement using the Autolumat LB 953 luminometer (EG & G). N-decyl aldehyde 100 µl [1 % (v/v) in ethanol] was used as the enzyme substrate. The result was expressed in relative light units (R. L. U.).
Cell differ entiation assays
Fixed amounts of bacteria were harvested hourly after a 1/100 dilution of an overnight culture into fresh MGM broth. Mean cell length was estimated by phase-contrast microscopy BH2 (Olympus) of at least 50 cells fixed in 1 % formaldehyde/0.85 % normal saline. For microscopic examination, bacteria were fixed on slides by mild flaming. Flagella silver stain followed the procedures of Koneman et al. (1997).
RESULTS
Identification of S. marcescens flhDC oper on
A pair of primers flhDF2 (5’-CATCCATACACGTTGGTTTACGCT-3’) and flhCR2 (5’-TGCTCGTCCAGCAGTTGAGGAATA-3’) designed from the conserved nucleotide acid sequence region of flhD/flhC from E. coli (GenBank accession number AB001340) and S. liquefaciens (Givskov et al., 1995) were used for PCR with S. marcescens chromosomal DNA as the template. An amplification product of 1093 bp was obtained and sequenced after cloning. This DNA fragment contained a high nucleotide sequence identity (95 %) with flhDC of S. liquefaciens. This DNA fragment was used to probe a λ-DASHII (Stratagene) phage library of partial EcoRI fragments from the wild-type S. marcescens CH-1 chromosome. Five phage clones hybridized and restriction mapping by Southern blot hybridization using the 1093 bp DNA as the probe was performed. A 5Kb BamHI fragment high-lighted was subcloned from phage clone JH-1.
A 1.7 kb DNA fragment containing the complete flhD and flhC genetic determinants was sequenced. The physical map, DNA and the translated amino acid sequences were shown in Fig. 1. As predicted, no putative promoter sequence was observed in the intergenic region of flhD and flhC, suggesting that similar to those of E. coli and S. liquefaciens, flhD and flhC also form an operon (flhDCSmand hereafter).
The sequence of flhD and flhC are predicted to encode a protein of 116 amino acids (a.a.) (nucleotide sequence 656~1003), and 194 a.a. (nucleotide sequence 1006~ 1587), respectively. There is no open reading frame found on the complementary strand or immediately upstream or down stream of the coding sequence.
with the protein sequences indicated similarity to a variety of FlhD/FlhC proteins. The highest similarity to known FlhD/FlhC was, as expected, with the S. liquefaciens FlhD/FlhC (Givskov et al., 1995). There were 99/95 % identity and 99/96 % similarity for FlhD/FlhC at the protein level between S. marcescens and S. liquefaciens. Identity with the E. coli FlhD/FlhC was 70/83 % (76/84 % similarity) at the protein level.
Inactivation of the chromosomal flhD gene by homologous recombination
To unravel the possible physiological functions that flhDCSmmay play, a flhD
knock-out mutant strain was constructed. A streptomycin-resistant antibiotic marker was inserted into the chromosomalflhD as described in the materials and methods. A mutant strain named S. marcescens F-1 was selected. To confirm that insertional mutagenesis into flhD had occurred in strain S. marcescens F-1, PCR was performed
using (5’- CATCCATACACGTTGGTTTACGCT-3’ and
5’-TGCTCGTCCAGCAGTTGAGGAATA-3’) as primers and F-1 chromosomal DNA as the template. Compared with the strain CH-1 as the control, a 3.1 kb amplified DNA fragment was observed in S. marcescens F-1 (Fig. 2a), indicating the F-1 mutant genotype. Further experiments by Southern blot hybridization confirmed a double cross-over event had taken place with the replacement of flhD by flhD::SM (Fig. 2b).
Cellular motility, growth and elongation were defective in flhDCSm mutant
S. marcescens F-1 was used for characterization of flhDCSmfunction in cell division
and flagellar synthesis. It was first inoculated into 0.4 % and onto 0.8 % LB agar to assess the swimming and swarming behaviors. As predicted, no phenomena of
swarming and swimming were observed (Fig. 3a). Flagella silver stain further showed no evidence of flagella (data not shown). However, the red pigment prodigiosin and the transparent zone supposed to be the surfactant serrawettin (Matsuyama et al., 1995; Lindum et al., 1998) were not affected in S. marcescens F-1, suggesting that they were not regulated by flhDCSm. The growth rate, cell elongation, and expression
profile of hag (the flagellin structural gene of S. marcescens. Harshey et al., (1989) throughout growth in LB broth culture were compared for S. marcescens CH-1 and F-1 cells. Fig. 3b showed that the growth rate expressed as O. D. (A600) of F-1 was
higher than that of CH-1 cells, and similar phenomenon was also observed in E. coli flhD mutant (Pruss and Matsumura, 1996). Measurement of cell elongation showed that significant difference in cell length was observed, especially during the period of early exponential phase, when the length of F-1 cells was about two thirds of the length of CH-1 cells (Fig. 3b). The hag promoter activity in S. marcescens S-2 was significantly decreased compared with that of S. marcescens S-1, (1/1000 activity was left). These data showed that, as expected, in S. marcescens swarmer-cell differentiation is regulated by flhDCSmoperon.
NucA activity was abolished in flhDCSm mutant
The role of flhDCSm in the control of secreted virulence factors, such as the
expression of nucASm was compared in S. marcescens CH-1 and N-2. A qualitative DNase plate assay (Koneman et al., 1997) was first performed and showed a difference between the nuclease activity of these two strains was observed (about 1/2 activity was left). As expected, following similar procedures described from S. liquefaciens by Givskov et al., (1995), the phospholipase-negative S. marcescens CH-1 cells were also observed in the flhD mutant (Fig.4a). Plate gelatin degradation assay
(Lai, 1994) to see the effect of flhDCSm operon on the secreted protease activity did
not show significant difference in the transparent zones around bacterial colonies. A microplate assay was performed to quantify the nuclease activity. The result showed that the relative nuclease activity detected in the spent supernatant of S. marcescens CH-1 was higher than that of F-1 for about 4~8 times. These results demonstrate that flhDCSmhas an important role in the regulation of nuclease synthesis.
To determine whether regulation of nucASmby flhDCSm occurs at the transcriptional
level, a PnucASm::luxAB reporter system was constructed in S. marcescens CH-1. This
approach offers high sensitivity of reporting of the nucASm promoter activity, not
obtained with alacZ reporter system, where mitomycin has to be added for detection of nucASm promoter activity (Chen et al., 1992). Also, by using this assay, it is easier
and faster than measuring the nuclease activity directly. The bacterial strains (N-1 and N-2) containing the promoter region of nucASm in front of luxAB in the chromosome
were constructed. Intensity of light emission was measured following the growth process in LB broth culture. A significant difference in bioluminescence emission was observed between N-1 and N-2 cells and only 1/10 of N-1 PnucASmactivity was left in
the S. marcescens N-2 mutant (Fig. 4b), suggesting that expression of nucASm was
regulated by FlhD/FlhCSm directly or indirectly at the promoter level. For complementation, flhDCSm constructed in pJH05 was transformed into S. marcescens
N-2 cells. Chloramphenicol resistant transformants were assayed for nucASm expression and cell differentiation. As expected, cell division, flagellar synthesis and swimming/swarming ability were restored to slightly better than the level of wild-type cells. The nucASm expression pattern was also restored to normal in the flhDCSm mutant S. marcescens N-2 (Fig. 4b), showing that flhDCSm operon was the genetic
Cell differ entiation was obser ved in multi-copy flhDCSm in MGM br oth cultur e
The pJH05 was transformed into S. marcescens CH-1. Following the growth in MGM broth culture at 30o
C, cellular morphology was observed closely. The result in Fig. 5 showed that elongated cells of cell length up to 40~80 vegetative cells without septum formation were observed at early-to-mid log phase, suggesting that the flhDCSm operon has a dominant role in the cell differentiation process. We have
previously hypothesized that there may be putative signal factors existing in the rich broth media, which in combination with flhDC Sm operon, stimulate the formation of
differentiated cells. The over-expression of flhDCSmsuggests that either a signal is not
required for formation of differentiated cells or over-expression of flhDCSmovercomes
the need for a signal.
Expr ession of flhDCSm is self-r egulated
To monitor the expression pattern of flhDCSmthroughout growth,S. marcescens F-3
(PflhDCSm::luxAB) was constructed. Following the growth curve at 30 o
C in LB broth culture, bioluminescence measurements (indicating flhDCSm transcription) were taken
(Fig. 6). Expression from PflhDCSmis growth-phase dependent, with the peak activity
occurring in the early log phase. To determine whether FlhD/FlhC auto-regulates, an flhD knock-out mutation was constructed in S. marcescens F-3 and designated S. marcescens F-4. The patterns of light emission were measured in S. marcescens F-3 and F-4 (Fig. 6) and show that a 30 % ~50 % decrease in PflhDCSm occurs in the flhD
mutant. Using a different reporter system, a similar result was found in Salmonella typhimurium (Kutsukake, 1997). The results show that flhDCSm autoregulation does
the regulation.
Expression of flhDCSm is influenced by temper ature, osmolar ity, but not by glucose catabolite r epression
Our data showed that the remaining 50 % of flhDCSm activation was not achieved
by autoregulation. As the expression of nuclease gene nucASm, the synthesis of flagella
and the swarming phenotype are regulated by environmental temperature and osmolarity [Li et al., 1993, Lai et al., unpublished data], we investigated the role of these important environmental parameters in the regulation of flhDCSm. The
PflhDCSm::luxAB and PhagSm::luxAB DNA fusions were used as the reporter systems
and the patterns of bioluminescence emission of S. marcescens F-3 and S-1 were monitored closely throughout growth in MGM broth culture. Results in Fig. 7a showed that the activity of PflhDCSm in F-3 at 37℃ was decreased for 25 % compared
with that at 30℃, and expression of PhagSm in S-1 was decreased for about 50 %.
Although it has been reported that expression of flhDCSm is regulated by the
EnvZ/OmpR signal transduction system in E. coli (Shin and Park, 1995), the ompR mutation did not affect the motility phenotype or flhD expression in S. typhimurium (Kutsakake 1997). EnvZ and OmpR are the sensor and response regulator proteins of a two-component system that controls the porin regulon of Escherichia coli in response to osmolarity (Hsing and Silhavy, 1997). To investigate the role of osmolarity in the control of PflhDCSm, S. marcescens F-3 was used. Increase of NaCl
concentration (8.5, 100, 250, and 500 mM) in MGM broth cultures inhibits the cell growth and down-regulates the expression of PflhDCSm (Fig 7b).
It has also been reported that a small increase in the glucose concentration of the medium up to 27.8 mM (0.5 %) completely inhibited the synthesis of flagella in E.
coli (Adler and Templeton, 1967), a phenomenon known as the glucose catabolite repression (Emmer et al., 1970). However, a similar phenomenon was not observed in our experiments including E. coli strains (Lai et al., 1997a). For confirmation, S. marcescens F-3 was inoculated into MGM broth cultures containing different concentrations of glucose (0.4 %, 0.8 %, 1.2 %, and 2.0 %), and the light emission pattern was recorded. The result in Fig. 7c showed that instead of inhibition, increase of glucose concentration generally up-regulates the expression of PflhDCSm. This
DISCUSSION
It is becoming increasingly apparent that flhDCSm has an important global role in
the regulation of physiology in Gram-negative bacteria. Here we report that in S. marcescens flhDC (flhDCSm) activates the expression of nuclease gene nucASm in addition to previously described activation of expression of phospholipase gene, control of cellular motility, and cell division. This result shows that FlhD/FlhC is a multiple-functional transcriptional activator involved in the process of cell differentiation, swarming and virulence factor expression.
Many extracellular proteins made by S. marcescens are partly co-ordinated in the regulation of their expression. The nuclease, chitinase, and phospholipase are all found at increased levels when bacterial growth slows down. However, the signals specifying growth-phase regulation have not yet been determined for these proteins. In this work we extend our studies to nuclease regulation. Our results show that in addition to growth–phase and SOS regulation, nucASm is activated by FlhDCSm at the
promoter level. To add further complexity, expression of nucASm is co-ordinately regulated with cell division and flagellar synthesis. As it is reported that regulation of nucASm expression is growth-phase dependent, and occurs at the transcriptional level
in S. marcescens (Chen et al., 1992), it is quite possible that this effect actually takes place via the expression of flhDCSm operon. Furthermore, as the expression of both
flhDCSm and nucASm are influenced by osmolarity and temperature in S. marcescens
(Lai et al., unpublished data), such an effect might also work through flhDCSm.
It is not clear whether FlhDSm alone or FlhD/FlhCSm controls the expression of nucASm. What is known is that the flagellar synthesis is controlled by FlhD/FlhC in E.
coli, and that cell division is regulated by FlhD alone (Pruss et al., 1997). We are currently investigating whether both FlhD/FlhC are essential for nucASm expression.
Another question we hope to address is whether there is direct interaction between FlhD/FlhC and the nucASm promoter. DNA footprinting will be performed for such
purpose.
It is interesting to note that differentiated swarmer-cells were observed in S. marcescens containing multi-copy of flhDCSm, in the absence of other environmental
factors such as solid agar surface contact and signal molecules. This observation suggests that initiation of cell differentiation in S. marcescens might be growth–phase dependent. In fact, in LB rich broth media, differentiated cells could also be observed in early log phase, when flhDCSm, and then the hag were over-expressed (Lai et al.,
1997b). More experiments have to be performed to confirm the supposition.
Our data showed that expression of flhDCSm was growth-phase dependent,
autoregulated, and influenced by temperature and osmolarity, but no glucose catabolite repression was observed in S. marcescens. These data suggested that a feedback regulation was observed from flhDCSm, and that the environmental factors
affecting flhDCSmexpression might indirectly affects the phenotypes regulated by
flhDCSm. The data also confirms our previous observation that the glucose effect
observed on the inhibition of cell swimming is not observed in enterobacteria (Lai et al., 1997a). Although from the nucleotide sequence data, a consensus CRP binding site was found in the upstream region of flhDCSm, our result showed that no significant
effect was observed from glucose catabolite repression, suggesting that such effect does not play a dominant role in the regulation of flhDCSm. Similar effect was also
observed from S. typhimurium (Kutsukake, 1997). Also, it was found that a σ28
consensus sequence was found between the nucleotide sequence 182~208, suggesting that the PflhDCSm might be regulated by FliA (the σ
28
factor) of S. marcescens. Indeed, some genetic evidence was also observed in S. typhimurium (Kutsukake,
1997). To confirm this, an experiment was performed to see whether an interaction between FliA and PflhDCSm exists in S. marcescens. S. marcescens does not swarm at
37oC and our result showed that the promoter activity of flhDCSm was decreased for
about 25 % when the environmental temperature was shifted from 30 to 37 o
C. This may partly explain the mechanism of swarming inhibition at higher temperature in S. marcescens.
Acknowledgements
This work was supported by the National Science Council grants NSC88-2314-B002-287, and NSC88-2314-B002-330, Taiwan, R. O. C. We thank Wei-Ming Chen and Ya-Lan Cheng for technical and facility assistance.
Refer ences
Adler, J. & Templeton, B. (1967), The effect of environmental conditions on the motility of Escherichia coli. J. Gen. Microbiol., 46, 175-184.
Al Mamun, A.A., Tominaga, A. & Enomoto, M. (1996), Detection and
characterization of the flagellar master operon in the four Shigella subgroups. J. Bacteriol., 178, 3722-3726.
Ball, T.K., Wasmuth, C.R., Braunagel, S.C. & Benedik, M.J. (1990),
Expression of Serratia marcescens extracellular proteins requires recA. J. Bacteriol., 172, 342-349.
Bernard, P. (1996), Positive selection of recombinant DNA by CcdB.Biotechniques, 21, 320-323.
Chang, A.C.Y. & Cohen S.N. (1978), Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid.J. Bacteriol., 134, 1141-1156.
Chen, Y.C., Shipley, G.L., Ball, T.K. & Benedik, M.J. (1992), Regulatory mutants and transcriptional control of the Serratia marcescens extracellular nuclease gene.Mol. Microbiol., 6, 643-651.
Dufour, A., Furness, R.B. & Hughes, C. (1998), Novel genes that upregulates the Proteus mirabilis flhDC master operon controlling flagellar biogenesis and swarming. Mol. Microbiol., 29, 741-751.
Eberl, L., Christiansen, G., Molin, S., Givskov, M., Guynn, L.J., Dai, W. & Benedik, M.J. (1996), Differentiation of Serratia liquefaciens into swarm cells is
controlled by the expression of the flhD master operon. J. Bacteriol., 178, 554-559.
receptor protein of E. coli: Its role in the synthesis of inducible enzymes. Proc. Natl. Acad. Sci., 66, 480-487.
Furness, R.B, Fraser, G.M., Hay, N.A. & Hughes, C. (1997), Negative feedback from aProteus class II flagellum export defect to the flhDC master operon controlling
cell division and flagellum assembly.J. Bacteriol., 179, 5585-5588. Givskov, M., Eberl, L., Christiansen, G., Benedik, M.J. & Molin, S. (1995),
Induction of phospholipase- and flagellar synthesis in Serratia liquefaciens is controlled by expression of the flagellar master operon flhD. Mol. Microbiol., 15, 445-454.
Givskov, M., Ostling, J., Eberl, L., Lindum, P.W., Christensen, A.B., Christiansen, G.,
Molin, S. & Kjelleberg, S. (1998), Two separate regulatory systems participate in control of swarming motility of Serratia liquefaciens MG1. J Bacteriol., 180, 742-745.
Harshey, R.M. (1994), Bees are not the only ones: swarming in Gram-negative bacteria. Mol. Microbiol., 13, 389-394.
Harshey, R.M., Estepa, G. & Yanagi, H. (1989), Cloning and nucleotide sequence of a flagellin-coding gene (hag) from Serratia marcescens 274. Gene, 79, 1-8. Hsing, W. & Silhavy, T.J. (1997), Function of conserved histidine-243 in phosphatase
activity of EnvZ, the sensor for porin osmoregulation in Escherichia coli. J. Bacteriol., 179, 3729-3735.
Khan, E.A., Wafelman, L.S., Garcia-Prats, J.A. & Taber, L.H. (1997),Serratia marcescens pneumonia, empyema and pneumatocele in a preterm neonate. Pediatr. Infect. Dis. J., 16, 1003-1005.
(1997), Color Atlas and Textbook of Diagnostic Microbiology. Fifth Edition. J. B. Lippincott Company. Philadelphia, U. S. A.
Kutsukake, K. (1997), Autogenous and global control of the flagellar master operon, flhD, in Salmonella typhimurium. Mol. Gen. Genet., 254, 440-448.
Lai, H.C. (1994), Molecular studies on the swarming migration of Proteus mirabilis. Ph.D thesis, Wolfson College, Cambridge University. England, U.K.
Lai, H.C., Shu, J.C., Ang, S., Lai, M.J., Fruta, B., Lin, S., Luh K.T. & Ho, S.W. (1997a), Effect of Glucose Concentration on Swimming Motility in Enterobacteria. Biochem. Biophy. Res.Commu., 231, 692-695.
Lai, H.C., Lai, M.J., Lin-Chao, S., Luh, K.T. & Ho, S.W. (1997b), Population cell differentiation of Serratia marcescens on agar surface and in broth culture. Chin. J. Microbiol. Immunol., 30, 242-254.
Li, C., Louise, C.J., Sh, W. & Adler, J. (1993), Adverse conditions which cause lack of flagella inEscherichia coli. J. Bacteriol. 175, 2229-2235.
Lindum, P.W., Anthoni, U., Christophersen, C., Eberl, L., Molin, S. & Givskov, M. (1998), N-acyl-l-homoserine lactone autoinducers control production of an extracellular lipopeptide biosurfactant required for swarming motility of Serratia liquefaciens MG1. J. Bacteriol., 180, 6384-6388.
Liu, X.Y. & Matsumura, P. (1994), The FlhD/FlhC complex, a transcriptional activator of the Escherichia coli flagellar class II operons. J. Bacteriol., 176, 7345-7351.
de Lorenzo, V., Herrero, M., Jakubzik, U. & Timmis, K.N. (1990), Mini-Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in Gram-negative eubacteria. J. Bacteriol., 172, 6568-6572.
de Lorenzo, V. & Timmis, K.N. (1994), Analysis and construction of stable phenotypes in Gram-negative bacteria with Tn5- and Tn10-derived
minitransposons. Bacterial pathogenesis part A. Identification and regulation of virulence factors.Methods. Enzymol., 235, 386-405.
Macnab, R.M. (1992), Genetics and biogenesis of bacterial flagella.Annu.Rev. Genet., 26, 131-158.
Maniatis, T., Sambrook, J. & Fritsch, E.F. (1989), Molecular cloning. A
laboratory manual, second edition. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, N. Y.
Matsuyama, T., Bhasin, A., & Harshey, R.M. (1995), Mutational analysis of
flagellum-independent surface spreading of Serratia marcescens 274 on a low-agar
medium.
J. Bacteriol., 177, 987-991.Peeters, A., Vandercam, B., Sindic, C.J., Hantson, P.& Mahieu, P. (1997)
Community-acquired Serratia marcescens meningitis. J. Infect.Dis., 35, 303-304.
Pruss, B.M. & Matsumura, P. (1996), A regulator of the flagellar regulon of Escherichia coli, flhD, also affects cell division. J. Bacteriol., 178, 668-674. Pruss, B.M., Markovic, D. & Matsumura, P. (1997), The Escherichia coli flagellar
transcriptional activator flhD regulates cell division through induction of the acid response gene cadA. J. Bacteriol., 179, 3818-3821.
Shin, S. & Park, C. (1995), Modulation of flagellar expression in Escherichia coli by acetyl phosphate and the osmoregulator OmpR.J. Bacteriol., 177, 4696-4702.
Figur e Legends
Fig. 1. The physical map and sequence characterization of the flhDCSm genetic locus.
(a) The physical map of S. marcescens flhDCSm genetic locus. PI, PstI; EI, EcoRI; SI, SalI; XI,
XhoI; SpI, SphI.
(b) DNA and predicted protein sequences of the flhDCSmlocus. A potential Shine-Dalgarno (S.D.)
sequence was indicated. A sequence TAAAAT, similar to the E. coli consensus –10 box, is found between the nucleotide sequence 413~418. The sequence, TTGCGC, similar to the E. coli consensus – 35 box, is found between positions 387~392 (The shaded area). Aσ28 consensus sequence was found
between 182~208, and a putative consensus CRP binding site was found between 311~331, with the conserved sequence being boxed.
Fig. 2. Confirmation of inactivation of flhDSm by streptomycin (SM) through homologous
recombination.
(a) A pair of primers (5’CATCCATACACGTTGGTTTACGCT-3’ and 5’-TGCTCGTCCAGCAGTTGAGGAATA-3’) were used in the PCR for confirmation of insertion mutagenesis. S. marcescens CH-1 chromosomal DNA (lane 1), pJH04 (lane 2), pCR2.1flhDCSm::Sm
(lane 3), S. marcescens F-1 chromosomal DNA (lane 4), and S. marcescens S-2 chromosomal DNA (lane 5) were used as the template. The arrows indicate the DNA size markers.
(b) Southern blot hybridization using the PCR amplified 1.1 Kb DNA fragment (the same as a) as the probe. Chromosomal DNA of S. marcescens CH-1 (lane 1, 4, 7), S. marcescens F-1 (lane 2, 5, 8) and S. marcescens S-2 (3, 6, 9) were digested by SalI, EcoRI/HindIII, and PstI separately as indicated. Arrows indicate the DNA size markers.
Fig. 3. The flhDSmmutant phenotypes.
(a) The swimming of F-1 (A) and CH-1 (B), and swarming of F-1(C) and CH-1 (D) in LB plates at 30℃. A and B, 0.4 % agar; C and D, 0.8 % agar.
(b) The growth curve [CH-1 (O); F-1 ( )] and mean cell length units [CH-1 (filled column); F-1 (empty column)] of S. marcescens CH-1 and F-1 following the growth process in LB broth culture at 30℃. Results are the means of at least 3 independent experiments (SEM < 5 %).
Fig. 4. Effect of flhDCSmon the phospholipase activity and nucASmexpression.
(a) The phospholipase–negative cells (upper arrow) were observed in flhD mutant (S. marcescens F-1) after O/N culture on egg-yolk agar plate (Koneman et al., 1997). The lower arrow indicates the wild-type control cells.
(b) Expression of nucASmwas activated by flhDCSm. Following the growth process in LB broth
culture at 30℃, the light emission of S. marcescens 1/ pBCSK+ ( ), 2/ pBCSK+ (O), and N-2/pJH05 (Ž) were measured and the specific activity (spe. act.) was expressed as (R.L.U./O.D.). Results are the means of three independent experiments (SEM < 6 %).
Fig. 5. Cells with differentiated characteristics were observed in S. marcescens CH-1/pJH05 in MGM broth culture.
S. marcescens CH-1/pJH05 was inoculated in MGM broth culture at 30℃. Following the growth process, cells were harvested for observation of cellular morphology. The cells shown in the left (A) were taken from early log phase (O.D. 0.2). The cells shown in the right (B) was taken from S. marcescens CH-1/ pBCSK+ control cells (O.D. 0.2).
Fig. 6. PflhDCSm was self-regulated.
Following the growth process in LB broth culture at 30℃, the light emissions of S. marcescens F-3 ( ) and F-4 (O) were measured and expressed as specific activity (spe. act., R.L.U./O. D.). Results are the means of three independent experiments (SEM < 6 %).
Fig. 7. Effect of temperature, osmolarity and glucose concentration on the expression of PflhDCSm.
(a)S. marcescens F-3 (PflhDCSm::luxAB) was cultured in LB broth at either 30 (O) or 37℃ ( ). The
light emission was measured following the growth process. For comparison, S. marcescens S-1 (PhagSm::luxAB) was also assayed at 30 (Ž) or 37℃ ( ).
(b) S. marcescens F-3 was cultured at increasing NaCl concentration [8.5 (O), 100 ( ), 250 (Ž) and 500 ( ) mM] in LB broth, followed by measuring the light emission. (c) S. marcescens F-3 was
cultured at increasing glucose concentration [0.4 % (O), 0.8 % (Ž), 1.2 % ( ) and 2.0 % ( )] in MGM broth culture. Results are the means of three independent experiments (SEM < 5 %). spe. act., specific activity (R. L. U./O. D.).
