OUP user menu

The expression of the flagellum of Legionella pneumophila is modulated by different environmental factors

Klaus Heuner, Bettina C. Brand, Jörg Hacker
DOI: http://dx.doi.org/10.1111/j.1574-6968.1999.tb13603.x 69-77 First published online: 1 June 1999


Legionella pneumophila, the aetiologic agent of legionnaires' disease, contains a single, monopolar flagellum which is composed of one major subunit, the FlaA protein. Expression studies using a reporter gene fusion of the flaA promoter with the luxAB gene revealed that the flaA expression is not only temperature regulated but is also influenced by the growth phase, the viscosity and the osmolarity of the medium, and by amino acids.

  • Legionella pneumophila
  • flaA expression
  • Reporter gene
  • Environmental factor

1 Introduction

Flagellar motility, via a single monopolar flagellum, may be a potential virulence factor in the infection of Legionella pneumophila in Acanthamoeba castellanii [1], in Hartmannella vermiformis and to a limited degree in human macrophage-like U937 cells [2]. It was shown that L. pneumophila is exclusively flagellated in the final phase of its intracellular life cycle [1, 3]. The flagellin is the major subunit of the flagella of L. pneumophila, and it has recently been shown that various L. pneumophila strains and isolates of species other than L. pneumophila are able to produce flagella [4]. Previous reports from this laboratory have shown that the temperature-dependent expression [5] of the gene flaA, encoding the flagellum subunit, is regulated at the transcriptional level [4], possibly directed by the alternative σ28 factor [6]. Many bacteria modulate their repertoires of gene expression to cope with various changes in the environment such as nutrients, temperature, cell density and osmolarity. Since L. pneumophila is a pathogen of freshwater protozoa which replicates in amoebae and survives in water as a free-living bacterium, it must be able to adapt to environments of high and low osmolarity and nutrition. It is of particular importance to respond to a wide range of environmental stimuli to ensure its survival in the environment. The influence of growth phase on expression of different virulence factors was shown [3]. L. pneumophila is highly motile in post-exponential growth phase as shown by phase contrast microscopy [3]. In this study, the expression of the flagellum subunit gene flaA was determined at the transcriptional level using a flaA-specific promoter fusion with the reporter gene luxAB of Vibrio harveyi [7]. The flaA promoter of L. pneumophila is recognised by the alternative σ28 factor [6]. The effect of different environmental factors on flaA expression was investigated.

2 Materials and methods

2.1 Bacterial strains and plasmids

The strain L. pneumophila Corby (serogroup 1) [8] was used in this study. L. pneumophila was grown in yeast extract broth (YEB; 1% yeast) supplemented with 1% ACES, 0.025% ferric PPi, 0.04%l-cysteine (Oxoid, Wesel, Germany) and 8 µg chloramphenicol ml−1 to the onset of stationary growth phase at 37°C if not indicated otherwise. Low copy vector pMMB207 [9] was used to construct the flaA promoter reporter gene fusion-containing plasmids.

2.2 Construction of a plasmid containing a flaA promoter fusion to luxAB

The construction of a flaA promoter fusion to the luxAB gene was recently described [6]. Briefly, plasmid pKH20 is a pUC18 derivative containing a SmaI-BamHI fragment encoding the promoterless luxAB gene. In order to fuse the flaA specific promoter to the luxAB gene, the promoter was amplified by PCR and cloned in front of the luxAB gene. This fragment was ligated into vector pUC18 and the promoter sequence was verified by DNA sequencing. The pflaA-luxAB fusion was then cloned into the low copy vector pMMB207. This fusion plasmid was designated pKH23. The construction of a plasmid containing a pflaA-lacZ fusion was done by the same cloning strategy using the 3.6-kb BamHI-SalI fragment containing the promoterless lacZ gene of plasmid pDN19lacΩ[10], resulting in plasmid pKH12.

For electrotransformation of L. pneumophila with the fusion-containing plasmids, electrocompetent cells were mixed with plasmid DNA, transferred into prechilled 0.1-cm electroporation cuvettes (Bio-Rad) and pulsed with the pulse controller set at 2.3 kV, 25 mF and 100 Ω of resistance. After electric discharge, 1 ml of prewarmed YEB medium was added to each cuvette. Phenotypic expression was allowed to occur for 6 h in 4 ml of YEB medium. The culture was then plated onto ACES (N-(2-acetamido)-2-aminoethanesulfonic acid) [Sigma]-buffered charcoal yeast extract medium (ABCYE) agar plates containing 20 mg ml−1 Cm. After another 4–5 days of incubation at 37°C, colonies were isolated and streaked again on selective ABCYE agar plates. Strains were stored at −80°C to avoid serial passage effects.

2.3 Luciferase activity assay and β-galactosidase measurement

L. pneumophila strains were grown either to the late exponential growth phase or to the transition from the late exponential growth phase to the beginning of the stationary phase or to the early stationary growth phase. Cells were adjusted to an optical density at 600 nm of 1 and kept on ice. 400 µl of reaction buffer (50 mM sodium phosphate buffer, pH 7.0; 50 mM β-mercaptoethanol, 2% BSA) was added to 10 µl of bacterial suspension in a luminometer cuvette. The cuvette was placed into a luminometer (Lumat LB 9501, Berthold, Bad Wildbad, Germany [automated]) and 300 µl of a decanal (n-decyl-aldehyde; Sigma)/reaction buffer (1:2000) solution was injected before measuring the light output. The reaction time was 5 s and luciferase activity was expressed in relative light units (RLU). During growth of L. pneumophila in YEB, β-galactosidase activity was measured by quantifying the hydrolysis of ONPG as described by Miller [11] and is given in Miller units (MU). Isolation of bacterial RNA and Northern blot analysis using an internal flaA-specific probe was carried out as described previously [4].

2.4 SDS-PAGE and immunoblotting

Total cell extracts of L. pneumophila strains were analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting. SDS-PAGE was performed by use of the discontinuous buffer system of Laemmli [12]. L. pneumophila suspensions were adjusted to an optical density at 600 nm of 1 and 300 µl of the cell suspension was pelleted by centrifugation then suspended in 50 µl of SDS sample lysis buffer and loaded onto a SDS-13% polyacrylamide gel. Western blots were carried out as described elsewhere [13] using a polyclonal monospecific antibody against L. pneumophila Corby flagellin [4].

3 Results and discussion

3.1 Reporter gene activity assays

L. pneumophila Corby harbouring plasmids pKH23 or pKH12 were used for reporter gene activity assays and luciferase or β-galactosidase activity was measured at different growth phases (see Section 2.3). Cells were adjusted to an optical density at 600 nm of 1 and luciferase activity and β-galactosidase activity were determined as described in Section 2. In control experiments, no significant reporter gene expression was obtained with the L. pneumophila strain harbouring the promoterless reporter gene (luxAB or lacZ)-containing plasmids (data not shown). Adding supplements to the YEB medium to study the influence of various factors on the flaA expression may have secondary regulatory effects, e.g. stationary growth phase may be reached at higher or lower bacterial densities. To avoid false interpretation of reporter gene expression, we determined the bacterial cell density (OD600) and the time of incubation needed to reach the late exponential/onset of stationary growth phase (Table 1). During the transition from late exponential to onset of stationary growth phase maximum flaA expression was observed (Fig. 1c). In parallel, total cell extracts of L. pneumophila Corby (pKH23) were analysed by Western blotting.

View this table:
Table 1

Influence of environmental factors on the growth of L. pneumophila Corby (pKH23) in YEB medium

Bacterial density (OD600)Time of incubation (h)
YEB medium2.4–2.518–22
supplemented with:
30 mM serine2.6023
30 mM threonine2.6323
30 mM serine/threonine2.6623
100 mM sucrose2.2523
200 mM sucrose2.1029
300 mM sucrose1.8533
1.5% PVP2.3620
3% PVP2.3422
6% PVP1.8025–28
  • Bacterial density (OD600) and time of incubation at 37°C are given for cell cultures in the late exponential/onset of stationary growth phase.

Figure 1

Effects of temperature and growth phase on flagellin expression of L. pneumophila strain Corby (pKH23, pflaA-luxAB fusion or pKH12, pflaA-lacZ fusion). To determine the effects of growth temperature on flaA expression (a), cells were grown in YEB medium at 30°C, 37°C or 42°C to the late exponential growth phase and luciferase activity (RLU) was measured. β-Galactosidase activity is given in Miller units (MU). Error bars indicate the S.E.M. obtained from three independent experiments. b: Western blot analysis of total cell extracts of bacteria grown at different temperatures (as indicated) and harvested at different growth phases using an antiflagellin antiserum: (A) exponential, (B) late exponential and (C) onset of stationary growth phase. c, d: Measurement of the OD600 for bacterial growth in YEB at 30°C (filled squares), relative light units (RLU) for the luciferase activity (open circles) and Northern blot analysis for flaA mRNA determination (D; lanes 1–4, RNA agarose gel electrophoresis; lanes 5–8, Northern blot); lanes 1, 5, 0.5 A (OD600)/10 h (h of incubation); lane 2, 6, 2.0 A/20 h; lanes 3, 7, 2.4 A/42 h; lanes 4, 8, 2.5 A/72 h.

3.2 Influence of temperature and growth phase on flaA expression

Growth temperature has recently been shown to regulate the flaA expression which is repressed at temperatures higher than 37°C [4, 5]. These results could be confirmed by determining the light production as shown in Fig. 1a (left). It could be demonstrated that the luciferase activity at 42°C was decreased to 1/13 as compared to the activity obtained at 37°C; likewise the FlaA protein was not detectable by Western blot analysis (Fig. 1b). Only a slight difference in luminescence could be measured at 30°C or 37°C. Similar data were obtained using a transcriptional flaA promoter fusion to lacZ (Fig. 1a, right), thus we can exclude instability effects of the luciferase enzyme especially at temperatures higher than 30°C. However, when cells were grown in the liquid medium at 37°C, light production and flagellin synthesis increased during the late exponential growth phase and peaked at the onset of stationary growth phase indicating that the expression of flaA is not only temperature but also growth phase regulated (Fig. 1c). These data could be confirmed at the RNA level by Northern blot analysis (Fig. 1d).

3.3 Repression of the flaA expression by amino acids

Amino acids such as serine, threonine and to a lesser extent tyrosine and glutamic acid are the primary source of carbon and energy for legionellae while carbohydrates such as sugars are not fermented (for a review, see [14]). To study if amino acids are factors directly influencing the flaA expression, cells were grown in YEB medium supplemented with the amino acids serine and threonine at a final concentration of 30 mM. Luciferase activity was determined in different growth phases (Fig. 2, I). Without addition of any amino acids, light production reached its maximum in the late exponential growth phase. Compared to the light production in the non-supplemented culture luminescence was reduced to 22% and 11% when serine or threonine were added, respectively. Only a small reduction of flaA expression was observed after adding glutamic acid to the medium (data not shown). Western blot analysis confirmed these results since there was no detectable flagellin observed in the presence of the amino acids (Fig. 3a). These results demonstrate that flaA expression is repressed by the addition of the amino acids serine or threonine while the addition of carbohydrates such as glucose (1%) had no effect (data not shown). The repression of the flaA expression observed during bacterial growth indicates that supplying Legionella with sufficient nutrients leads to the repression of flaA expression. In contrast to Legionella, flagellin genes of Escherichia coli are not expressed in the presence of glucose [15] because the expression of the flhD master operon is sensitive to catabolite repression [16].

Figure 2

Effects of amino acids, osmolarity and viscosity on the flagellin expression of L. pneumophila strain Corby (pKH23, pflaA-luxAB fusion). I: To study the influence of the carbon source on flaA expression cells were grown in YEB medium at 37°C or supplemented with 30 mM serine (Ser) or 30 mM threonine (Thr). II: The effect of osmolarity on flaA expression was determined for cells grown in YEB medium at 37°C supplemented with 100, 200 or 300 mM sucrose (Suc). III: The significance of the viscosity of the growth medium for flaA expression of L. pneumophila was tested by adding 1.5%, 3% or 6% polyvinylpyrrolidone (PVP) to the YEB medium where cells were grown at 37°C. Cells were harvested at different growth phases: (a) exponential, (b) late exponential and (c) onset of stationary growth phase. The luciferase activity is expressed in RLU. Error bars indicate the S.E.M. obtained from three independent experiments.

Figure 3

Western blot analysis of total cell extracts of bacteria grown at different environmental conditions: (a) YEB medium supplemented with 30 mM serine, 30 mM threonine or both; (b) YEB supplemented with 100, 200 or 300 mM sucrose; (c) YEB supplemented with 1.5%, 3% or 6% PVP. Cells were grown at 37°C and harvested at different growth phases: (A) exponential, (B) late exponential/onset of stationary growth phase, (C) stationary growth phase. Western blotting was performed using an antiflagellin antiserum.

3.4 Effects of osmolarity and viscosity on flagellum expression

Since Legionella must be able to adapt to different habitats, the role of osmolarity in the regulation of flaA expression was examined by adding sucrose to the medium (YEB) in a final concentration of 100, 200 or 300 mM. Light emission of the cells was measured during the different growth phases as mentioned above. As shown in Fig. 2, II, maximum light production was observed during late exponential growth phase (b). Addition of 200 mM sucrose to the growth medium led to a 2.5-fold increase in luciferase activity in comparison to the activity obtained with non-supplemented medium. The presence of a sucrose concentration of 300 mM did not further increase the luciferase activity. Simultaneously with the increased luciferase activity, we could also observe attenuated bacterial growth and retarded entry into the stationary growth phase although the cell density was significantly lower compared to growth in non-supplemented medium (Table 1). Remarkably, increased luciferase activity (Fig. 2, II) could already be detected in the exponential growth phase while bacteria grown in non-supplemented medium did not express flaA. These results were confirmed by Western blot analysis, as shown in Fig. 3b. These data indicate that high osmolarity of the medium induces flaA expression, overriding repressed expression of flaA observed during the exponential growth phase and thereby allowing the bacteria to escape from an unprofitable environment. However, in E. coli the contrary phenomenon could be observed. High osmolarity leads to the repression of flagellar expression mediated by the two-component regulatory system OmpR/EnvZ [17, 18].

Viscosity of the environment is proposed to have a pronounced effect on the function of different flagellar types [19] but nothing is known about this phenomenon in L. pneumophila. The viscosity of the YEB medium was varied by addition of 1.5%, 3% or 6% polyvinylpyrrolidone (PVP, MW 360 000; Sigma) and light emission was determined as described above (Fig. 2, III). A slight increase in luciferase activity could be observed during maximal lux expression in the late exponential growth phase after supplementing the medium with 1.5% PVP. However, this increase seems not to be significant. In contrast, light emission of cells is decreased by adding 3% or 6% PVP in comparison to the emission observed in non-supplemented medium (Fig. 2, III). This repression could also be observed in Western blot analysis (Fig. 3c) since no flagellin could be detected in cell extracts from bacteria grown in YEB medium supplemented with 6% PVP. Similar results have been reported for E. coli considering not directly flagellin expression but swimming speed, which at first slightly increased but then decreased rapidly under high viscosity conditions and no motility was observed with 12% PVP [19].

3.5 Conclusions

In this study, we have demonstrated that the flagellin is primarily regulated by the growth phase at the transcriptional level but in addition, environmental stimuli such as temperature, nutrient stage and osmolarity, as well as viscosity are able to modulate the expression. Our data corroborate at the transcriptional level data published by Byrne and Swanson [3] using by phase contrast microscopy. The authors showed that post-exponential bacteria were highly motile and that starvation favoured the flagellated form of L. pneumophila. We could show that the growth phase-dependent flaA regulation occurs at the transcriptional level and that the primary energy source (the amino acids serine and threonine) represses flaA expression. Flagellum expression and virulence seem to be correlated genetically [3] and the regulation of flaA gene expression may involve the alternative σ28 factor [4, 5]. Post-exponential phase legionellae are more infectious [3] and motility may be necessary to escape the spent host and find another one to survive in the aquatic environment. The role of the flagella in pathogenicity and fitness of L. pneumophila and the role of the σ28 factor in gene expression have to be further studied.


We would like to thank David Gally for his careful review of the manuscript. This work was supported by the ‘Graduiertenkolleg Infektiologie’ (Deutsche Forschungsgemeinschaft (DFG Lu 485/1-1), by the Bundesministerium für Forschung und Technologie (BMFT 01 KI 8829; 8812) and by the Fonds der Chemischen Industrie.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
View Abstract