© 2014 APMIS. Published by John Wiley & Sons Ltd. DOI 10.1111/apm.12349
APMIS 123: 330–341
Identification and characterization of genes, encoding the 3-hydroxybutyrate dehydrogenase and a putative lipase, in an avirulent spontaneous Legionella pneumophila serogroup 6 mutant MARIA SCATURRO,1 CRISTINA BARELLO,2 MELANIA DE GIUSTI,1 STEFANO FONTANA,1 FEDERICA PINCI,1 MARIA GABRIELLA GIUFFRIDA2 and MARIA LUISA RICCI1 1
Department of Infectious Parasitic Immune-mediated Diseases, Istituto Superiore di Sanit a, Rome and Istituto di Scienze delle Produzioni Alimentari, CNR, Sezione di Torino, Colleretto Giacosa (TO), Italy
2
Scaturro M, Barello C, De Giusti M, Fontana, S, Pinci F, Giuffrida MG, Ricci ML. Identification and characterization of genes, encoding the 3-hydroxybutyrate dehydrogenase and a putative lipase, in an avirulent spontaneous Legionella pneumophila serogroup 6 mutant. APMIS 2015; 123: 330–341. Legionella pneumophila is a pathogen widespread in aquatic environment, able to multiply both within amoebae and human macrophages. The aim of this study was to identify genes differently expressed in a spontaneous avirulent Legionella pneumophila serogroup 6 mutant, named Vir, respect the parental strain (Vir+), and to determine their role in the loss of virulence. Protein profiles revealed some differences in Vir proteomic maps, and among the identified proteins the undetectable 3-hydroxybutyrate dehydrogenase (BdhA) and a down-produced lipase. Both Legionella enzymes were studied before and were here further characterized at genetic level. A significant down-regulation of both genes was observed in Vir at the transcriptional level, but the use of defined mutants demonstrated that they did not affect the intracellular multiplication. A mutant (MS1) showed an accumulation of poly-3-hydroxybutyrate (PHB) granules suggesting a role of bdhA gene in its degradation process. The lipase deduced amino acid sequence revealed a catalytic triad, typical of the ‘lipase box’ characteristic of PHB de-polymerase enzymes, that let us suppose a possible involvement of lipase in the PHB granule degradation process. Our results revealed unexpected alterations in secondary metabolic pathways possibly linking the loss of virulence to Legionella lack of energy sources. Key words: Legionella; bdhA; lipase; proteome; ketone compounds. Maria Scaturro, Department of Infectious Parasitic Immune-mediate Diseases, Istituto Superiore di Sanit a, Rome, Italy. e-mail: [email protected]
Legionella pneumophila is a facultative intracellular pathogen causing an atypical pneumonia known as Legionnaires’ disease. Legionella is ubiquitous in natural freshwater environments and moist soil, and is capable of multiplying within a wide range of host cells, such as protozoa and human macrophages (1). After inhalation or micro-aspiration, in the lungs Legionella cells are engulfed by macrophages into a membrane bound compartment known as Legionella -containing vacuoles (LCVs). LCVs do not fuse with lysosomes, but are surrounded by smooth vesicles originating from the endoplasmic reticulum to generate a niche that Received 16 May 2014. Accepted 22 October 2014
330
supports bacterial growth (2, 3). The ability to subvert the host membrane traffic is the key of Legionella replication within eukaryotic cells. Legionella produces a Type IV secretion system known as Dot/Icm that injects > 200 bacterial proteins into the host cells promoting control of the membrane transport (4). The intracellular lifecycle consists of a replicative and a transmissive phase with the transition from the one and the other characterized by differential expression of several genes (5, 6). Most of the known virulence factors are up-regulated during the transmissive phase when the bacteria appear flagellated, highly motile and ready to exit from the current host and invade a new host cell (7). Several known molecular mechanisms
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
regulating Legionella pathogenesis have been characterized using defined mutants. However, spontaneous avirulent mutants are also powerful tools for virulence studies. Two-dimensional electrophoresis (2-DE) is a useful approach to study global changes in bacterial protein production occurring during intracellular growth (8), and to differentiate proteins secreted through vesicles from those that are naturally soluble (9, 10). Furthermore, with genomic and proteomic analysis now available for some L. pneumophila strains, many proteins have been identified, including several virulence factors (11–14). In a previous study we characterized a spontaneous avirulent L. pneumophila serogroup 6 strain, named Vir, obtained by multiple passages on agar plates (15, 16). Vir did not have a flagellum and alterations were found in flaA and dotA gene expression when compared to Vir+ (17). Although no detectable level of flaA expression was observed, no alterations were found in the nucleotide sequence of both promoter and encoding regions. In contrast, the dotA sequence revealed an in frame transition mutation of C to T at nucleotide 1006 that produced a stop codon. DotA was consequently not produced and the accumulation of lysosomal marker LAMP-1 on phagosomes containing the Vir strain was observed as previously demonstrated in L. pneumophila serogroup 1 (17, 18). In this study, Vir proteome was explored to determine further differently produced proteins in comparison to the Vir+ proteome, which might have affected the loss of Vir virulence. Out of 13 identified proteins, two produced at lower levels in the Vir strain, namely BdhA and a putative lipase, were further analyzed at gene level and the possible involvement in virulence was also verified. An alteration of metabolic pathways involving the synthesis and degradation of the PHB granules and of the ketone compounds was also observed. MATERIALS AND METHODS Bacterial strains, plasmids and growth conditions All bacterial strains and plasmids used in this study are listed in Table 1. Legionella strains were grown on Buffered Charcoal yeast Extract with a-ketoglutarate (a-BCYE) agar plates and in BYE broth (Oxoid, Hampshire, UK) supplemented with 100 lg/mL Thymidine (Sigma-Aldrich, St. Louis, MO, USA) (a-BCYE-T) when required. L. pneumophila strain Lp02 (kindly supplied by Dr Isberg) was used to construct deletion mutants. DH5a k pir and MT607 E. coli strains were used to propagate pSR47s vector (also kindly provided by Dr Isberg) and for Legionella mating, respectively (19). When required, antibiotics were added to the media at the following final concentrations: Kanamycin (IBI, Peosta, IA, USA) at
© 2014 APMIS. Published by John Wiley & Sons Ltd
100 lg/mL for E. coli and 25 lg/mL for Legionella; streptomycin (IBI) at 50 lg/mL and ampicillin at 100 lg/mL.
Protein extraction Vir+ and Vir strains were grown in 15 ml BYE-broth to post-exponential phase (OD600 = 2) prior to total protein extraction. Bacterial cells were collected by centrifugation at 4500 g for 10 min and pellets re-dissolved in ice-cold sterile distilled-H2O to determine the colony forming unit/ mL (CFU/mL) by serial dilutions on a-BCYE agar plates. Bacterial cells were then centrifuged as above and pellets re-suspended in 5 mL ice-cold Tris-EDTA solution containing 1 mM PMSF (Sigma-Aldrich), and lysed with 50 mg/ml lysozyme (Sigma-Aldrich) and 10% w/v SDS on ice for 10 min. Nucleic acids were removed by digestion with DNase (10 U/lL, Invitrogen-Life technologies, Monza, Italy) and RNase (10 mg/mL, Invitrogen-Life technologies), for 15 min on ice. Cellular debris was removed by two consecutive centrifugations, the first at 10 000 g for 10 min and the second one at 17 000 g for 1 h. Supernatant was divided into aliquots (100 lL) and proteins precipitated with 3 volumes of ice-cold acetone, for a minimum of 2 h at 20 °C. Protein pellets were obtained by centrifugation, 13 000 g for 20 min at 4 °C and stored at 20 °C until used. The extraction was repeated for 3 preparations and protein concentrations were determined using the Bradford assay (Pierce-Thermo fisher Scientific, Rockford, IL, USA).
Protein solubilization and two-dimensional electrophoresis In order to perform Isoelectric focusing (IEF) analysis, dried pellets were solubilized in 250 lL of re-hydration solution (7.0 M urea, 2.0 M Thiourea, 4% CHAPS, 1% Triton X-100, 0.02 M Tris, 0.5% IPG buffer either pH 4–7 or pH 3–10 (GE Healthcare, Pittsburg, CA, USA), 1% DTT) and applied to 13 cm IPG strips (GE Healthcare) using the in-gel re-hydration method (20). After 12 h of re-hydration, IEF was performed with a linear gradient ranging from either 3 to 10 or 4 to 7 pI, at a constant temperature of 20 °C with 58 000 V 9 h. After IEF, the IPG strips were incubated for 15 min at room temperature in 6 M urea, 30% v/v glycerol, 2% w/v SDS, 50 mM Tris-HCl pH 8.6 enriched with 2% w/v DTT. A second equilibration step was carried out for 15 min in the same solution, with the exception that DTT was replaced by 2.5% w/v iodoacetamide plus a trace of bromophenol blue as tracking dye. The IPG strips were then placed at the top of the 1.0 mm vertical second dimensional gels and sealed. SDS-PAGE for all experiments was carried out as previously described (21). Running conditions were 400 V, 50 mA/gel, for 3 h. Three gel replicates were performed for each sample. Gels were automatically stained using the Processor Plus device (GE Healthcare) with freshly prepared Neuhoff stain (22).
Proteome data analysis 2-DE gels were scanned by a GS-800 Densitometer (BioRad, Hercules, CA, USA) and analyzed with Progenesis PG 220 software (Nonlinear Dynamics, UK). The average of
331
SCATURRO et al.
Table 1. Bacterial strains and plasmids Strain or plasmid L. pneumophila Lp02 Vir+ Vir MS1 FP2 E. coli DH5a (kpir) MT607 XL1-Blue Plasmids pSR47s pSDbdhA pSDlipase pGEM-Teasy pGbdhA+ pGbdhA pBlueScript (SK+) pBlip+ pBlip
Relevant genotype
Reference or source
hsdRrpsLthyA Lp6 clinical strain Lp6 clinical strain avirulent mutant Lp02DBdhA Lp02Dlipase
(19) (17) (17) This study This study
DH5a (kpir) tet::Mu recA pRK600 F’::Tn10 proA+B+lacIqD(lacZ)M15/ recA1 endA1 gyrA96 (Nalr) thi hsd R17 (rk-mk+) supE44 relA1 lac
(19) (19) (19)
pSR47sacB pSR47sDbdhA pSR47sDlipase
(19) This study This study Promega This study This study Stratagene This study This study
pGEM-bdhAVir+ pGEM-bdhAVirpBlueScriptlipaseVir+ pBlueScriptlipaseVir-
each gel was determined by calculating the mean of the triplicates. Spot intensities were measured as normalized spot volumes. Spots were selected on the basis of a 2-fold difference in production between the two strains that was statistically significant based on a Student’s t-test p < 0.05.
Protein identification Spots were excised from the gels and subjected to ‘in-gel’ trypsin digestion (Promega, Madison, WI USA), as described by Hellmann et al. (23). For MALDI-TOF analysis, 0.5 lL of each peptide mixture was applied to a target disk, and allowed to dry. Subsequently, 0.5 lL of HCCA matrix solution (a-cyano-4-hydroxycinnamic acid) was applied to the dried sample and allowed to dry again. Spectra of protein digests were obtained using a Bruker Reflex III mass spectrometer (Bruker Daltonics, Italy). MSFit (http://prospector.ucsf.edu/prospector/cgi-bin/msform. cgi?form=msfitstandard) software package was used to interpret the MS spectra through the peptide mass fingerprinting method (PMF) (24). The software package was searched against UniProt KB database (www.uniprot.org). The following parameters were used: S-carbamidomethyl derivate on cysteine as fixed modification, oxidation on methionine as variable modification and two missed cleavage sites for trypsin digestion. Peptide mass tolerance was 30 ppm.
tide sequences were analyzed using the above-mentioned primers and those indicated by an asterisk in Table 2.
Construction of in frame DbdhA and Dlipase gene deletions and sequencing analysis In frame deletions in bdhA and lipase were introduced into the Lp02 chromosome by allelic exchange using the pSR47s suicide vector (25). Briefly, regions located upstream and downstream of the bdhA gene were amplified using primers 1.1bdhA with 1.2bdhA and 2.1bdhA with 2.2bdhA, respectively and then linked by overlapping recombinant PCR using primers 1.1bdhA and 2.2bdhA. The lipase gene deletion was similarly generated using 1.1lipase with 1.2lipase and 2.1lipase with 2.2lipase. The oligonucleotide sequences of each primer are listed in Table 2. Restriction sites (underlined in Table 2) were introduced at the 50 and 30 end of the recombinant fragments and used to clone the final products into similarly digested pSR47s. The resulting pSDbdhA and pSDlipase recombinant plasmids were introduced into Lp02 via mating for the homolog allelic exchange. Selection for plasmid integration was obtained by resistance to kanamycin on a-BCYE-T-Kan plates. Purified colonies were then grown on a-BCYE plates containing 5% sucrose to counter-select for the plasmid. Finally, colonies were screened for the presence of deletion and the absence of the targeted gene by both PCR and sequence analysis. Legionella MS1 (DbdhA) and FP2 (Dlipase) strains were thus isolated and used in further analyses.
Sequence analysis Genomic DNA from Vir+ and Vir strains was extracted using 20% Chelex 100 (Sigma-Aldrich) followed by boiling for 10 min. The bdhA and lipase genes were amplified by PCR by using the 1.1bdhA with 2.2bdhA and 1.1lipase with 2.2lipase specific primers, respectively (Table 2). Amplicons were subcloned into pGEM-Teasy, obtaining the pGbdhA+ and pGbdhA, and pBlueScript, obtaining pBlip+ and pBlip recombinant vectors and the nucleo-
332
Southern analysis Genomic DNA from L. longbeachae (ATCC 33462), L. gormanii (ATCC 33297), L. dumoffii (CDC collection), L. bozemani (ATCC 35545), L. micdadei (ATCC 33204), L. oakridgensis (ATCC 33761), L.jamestowniensis (ATCC 35298), L. quinlivanii (kindly provided by Dr Lee J.) and L. parisiensis (ATCC 35299) were digested with EcoRI © 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
Table 2. Primers used in this study Primer Sequence1 1.1bdhA ATAtctagaTAAAGTACCAGCCTCACCAGTCGT 1.2bdhA caaatgacattgAGATCAGCGATTGCTACCTTTGCC 2.1bdhA aatcgctgatctCAATGTCATTTGCCCGGGGTTTGT 2.2bdhA ATAggatccGGTGGGAACCATCCCAAAAGAATCA DbdhA.seq2 TGTTACAGGAGCAGCAAGCGG F bdhA.seq2 TTCATGGAAAACAATGACAAGGATAG R bdhA.seq2 CAATATGGCGATACTCAATCTGTC F bdhA GCAAAAGAATTAGGCATCAGTGAAG R bdhA CAATATGGCGATACTCAATCTGTC 1.1lip ATAgtcgacGTGCGCTTATCCCACATTGCTT 1.2 lip atccgcttggaCGGTTTTCCACCCGCATTTTGT 2.1 lip gtggaaaaccgTCCAAGCGGATGTTAGTCATGCT 2.2 lip ATAggatccGGAGAGGGAATTAAGGCAATTGCCA F lip.seq2 GGGCCACCACGATCTTTAGC R lip.seq2 CCACCTGTTATTGTTATCCAC F lipase probe ATCCCAGACACCATCCACCTGTTA R lipase probe TCAGCATGACTAACATCCGCTTGG F lipase ATGCGGGTGGAAAACCGATT R lipase GTGTCGGGTACTGAGCTTCT F rib16S CGTAGAGATCGGAAGGAACACC R rib16S GCGTGGACTACCAGGGTATCT 1 Underlined letters are the sequences of restriction enzymes used for cloning. 2 Primers used for sequencing. and PstI restriction enzymes, and the fragments were separated by electrophoresis on 1% agarose gel. Genome fragments were then transferred to nylon membrane (Amersham-GE Healthcare, Pittsburg, CA, USA) by capillary blotting. A DNA probe corresponding to the lipasecoding region was amplified by using the F lipase probe and R lipase probe primers shown in Table 2, then labeled and detected by using North2South Chemiluminescent Hybridization and Detection kit (Pierce- Thermo fisher Scientific).
Reverse Transcription (RT)-PCR analysis To verify the transcription of bdhA and lipase, Vir+ and Vir strains were grown to the post-exponential phase and harvested by centrifugation at 4500 g for 10 min at 4 °C. Total RNA was then extracted and purified by Nucleospin RNA II (Machery-Nagel, Duren, Germany) and 100 ng were used for retro-transcription reaction with ImProm-IITM Reverse Transcription System (Promega), according to the manufacturer’s instructions. cDNA synthesis was obtained by using specific primer pairs: bdhA cDNA was generated using the F bdhA.seq and R bdhA.seq primers, located directly upstream and downstream of the bdhA open reading frame; lipase cDNA was generated using the F lipase and R lipase primers (Table 2). cDNA was obtained also from RNA extracted from the MS1 mutant to further confirm the bdhA deletion (with the 1.1bdhA and 2.2bdhA primers) and to provide the transcription of the downstream patD (with 2.1bdhA and 2.2bdhA primers). The absence of genomic DNA was verified by amplification with the AmpliTaq polymerase (Applied Biosystems-life technologies, Monza, Italy).
Quantitative RT-PCR To determine levels of bdhA and lipase transcripts in the Vir+ and Vir strains, quantitative RT-PCR was carried © 2014 APMIS. Published by John Wiley & Sons Ltd
Restriction enzyme XbaI BamHI
SalI BamHI
out. To this aim, specific cDNAs were generated and an aliquot of 2 lL was added to 20 lL reaction mixture, which included 10 lL of SYBR green master mix (BioRad) and 0.5 lM of each primer. Primer pairs were as follow: F bdhA and R bdhA, F lipase and R lipase (Table 2). 16S cDNA was used as reference gene to normalize gene expression and the primer pair was F rib16S and R rib16S (Table 2). Reactions were performed on a Chromo-4 real-time PCR detection system (BioRad) using the following parameters: 95 °C for 3 min, 95 °C for 30 s, 57 °C for 30 s and 72 °C for 30 s for 30 cycles. For each experiment, 4 replicates were done and each of them was tested twice. Gene expression level was assessed by determining the cycle at which the amplification crossed the detection threshold (Ct). The fold difference of the two target genes between Vir+ and Vir was calculated using the 2DDCt method (26).
Intracellular growth assay A. castellanii (ATCC 30234) and Thp-1 monocytes (ATCC TIB-202), a human cell line that differentiates into macrophages-like cells upon treatment with 20 ng/ml phorbol ester (12-O_tetradecanoylphorbol-13-acetate (SigmaAldrich) for 72 h, were used to test the intracellular growth. A. castellanii were maintained in PYG (peptoneyeast extract-glucose) medium (Oxoid, Thermo Fisher Scientific) in 75 cm2 tissue culture flasks for no more than 1 week. On the day of infection, cells were transferred to 24-well tissue culture plates with a final concentration of 105 amoebae/mL. Thp-1 cells were grown in RPMI completed medium (Euroclone, Milano, Italy) and after differentiation seeded in 24-well tissue culture plates at a final concentration of 106 cells/mL. Lp02, MS1 and FP2 strains were used to infect amoeba and macrophage cells at an MOI of 1. At 0, 24, 48, and 72 h time points the of Thp-1 and A. castellanii monolayers were lysed with 1 ml of dis-
333
SCATURRO et al.
tilled sterile water and 0.2% Triton 100 (ICN Biomedicals, Irvine, CA, USA) respectively (27), then serial dilutions were plated on a-BCYE agar plates to determine Legionella CFU/mL.
Visualization of PHB granules Lp02, MS1, Vir+, and Vir strains were grown in BYEbroth overnight at 37 °C. A 10-fold dilution was inoculated in fresh medium then after 24, 48 and 72 h of incubation at 37 °C, 100 lL were removed and centrifuged to pellet bacterial cells. Pellets were then treated with 1% formalin for at least 3 h. Dry smears of formalin-treated bacteria on coverslips were heat-fixed for 10 min and finally stained with 25 mM Nile red (Sigma-Aldrich), diluted 1:250, for 30 min at 37 °C. Coverslips were then examined by fluorescence microscopy (Leica, Milano, Italy).
RESULTS Proteome analysis
Spontaneous avirulent mutants are valuable tools in bacterial pathogenesis investigations, although they have a strong limit due to the difficulty of determining promptly how many and what factors have been altered. Proteome analysis is a helpful approach because it permits to bypass this limit. In this study, proteomic analysis was performed with protein extracts obtained from bacteria grown to the post-exponential phase, in which Legionella shows a virulent phenotype. Although equivalent amounts of bacterial cells (1.3 1 9 109 CFU/ mL) were used to prepare protein extracts, a slightly different protein recovery between the two strains, namely 0.68 0.04 mg/mL for Vir+ and 0.97 0.07 mg/mL for Vir, was observed. Nevertheless, comparing Vir and Vir+ proteome maps, differently produced proteins were observed and some of them were identified. Twenty-four spots from the Vir+ maps, either undetectable or downproduced in both the 3–10 pI and 4–7 pI Vir maps, were chosen according to the statistical criteria but only 13 were successfully identified. Figure 1 shows two representative proteome maps, obtained by 2-DE of the Vir+ and Vir proteins. The 3–10 pI IEF separation shows the Vir+ and Vir total protein expression (Fig. 1A and C, respectively). The zoomed IEF 4–7 pI maps are also reported in Fig. 1B and D. The analyzed spots and the corresponding proteins are listed in Table 3 and shown by arrows in Fig. 1A and B. In Vir protein maps, spots corresponding to flagellin (spot 1), 2-malate dehydrogenase (spot 2), 4-glutatione S-transferase (spot 4), a hypothetical protein (spot 6) and 3-hydroxybutyrate dehydrogenase (spot 7) were undetectable.
334
Quantification of normalized spot volumes showed a statistically significant down-production of spot 3, 5, 8, 9, 10, 11, 12 and 13. These spots corresponded to a putative lipase (spot 3), a recombinant protein RecR involved in DNA repair (spot 5), elongation factor Tu (spots 8 and 9), an acetoacetate decarboxylase (spots 10 and 11), a hypothetical protein with homology (46%) to OsmC family protein (spot 12) and a small heat shock protein (spot 13), respectively. Among these proteins, the lipase was the most down-produced. Except for flagellin, most of the identified proteins suggested that important changes might have occurred in specific metabolic pathways, in particular involving the degradation of PHB granules and ketone metabolism, usually considered to be not essential for Legionella growth (Fig. 2). Proteins analyzed at gene level: BdhA and the putative lipase
3-hydroxybutyrate dehydrogenase (BdhA) is an enzyme known to be involved in the first step of PHB degradation and, as demonstrated for other bacteria, is encoded by the bdhA gene (28). It is known that PHB accumulates within cells of various bacterial species as well as Legionella, forming granules-like cytoplasmic inclusions that represent an important reserve of energy especially in low nutrient environments (29, 30). In HeLa cell infection experiments, it has been observed that Legionella accumulates and metabolizes PHB granules during the transmissive phase (5, 31). However, because BdhA catalyses the oxidoreduction of 3-hydroxibutyrate to acetoacetate, it also plays a role in the metabolic pathway of ketone compounds (Fig. 2). On the other hand, lipases are enzymes that can act in different ways: as membrane destructors or as inducers of host cell processes such as cell migration and membrane trafficking. Lipase activities have been often related to virulence mechanisms adopted by bacteria as Legionella (32). For all the above-mentioned reasons and because still uncharacterized, we chose to explore the bdhA and the putative lipase genes. In particular, to determine whether the undetected or lower produced proteins were due to mutations affecting gene expression, the chosen approach was to analyze the nucleotide sequence. bdhA gene sequence analysis
bdhA sequence analysis was determined using pGbdhA+ and pGbdhA vectors which contain the entire open reading frame (ORF) plus 277 nucleo© 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
MW kDa
pI 3
MW
10
pI 4
kDa
A
7
B 97
97
66 66
2 1
45
45
8
3 7
10
31
31
11
9
4
12 13
5
21
21
6
14
14
MW kDa
pI 3
MW
10
pI 4
kDa
C
7
D 97
97
66
66
45
45
31
31
21
21
14
14
Fig. 1. 2-DE maps of total protein extracts of Vir+ (Panels A, B) and Vir (Panels C, D) separated by IEF 3-10 pI (Panels A, C) and 4-7 pI (Panels B, D) and then by 12.5% SDS-PAGE. Arrows indicate the analyzed spots. Table 3. Spot identification from the L. pneumophila serogroup 6 Vir+ map Spot
UniProt KB entry
Name
Gene (ID)
Species
Function
1
Q7ATT9
Flagellin
flaA (AJ496382)
2
Q5ZRB1
maeA (lpg2971)
Q5ZUB8
4
Q5X3Z8
5
Q5ZRX0
6
Q5ZZ46
7
Q5ZT49
8
Q5WZL4
9
Q5WZL4-
10
Q5ZXQ9
11
Q5ZXQ9
12
D5T890
L. pneumophila sg6 Vir+ L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Paris L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Lens L. pneumophila sg1 strain Lens L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg 1 strain 2300/99 Alcoy
Structural protein Oxidoreductase
3
Malate dehydrogenase Lipase
13
Q5ZSM4
lipase (lpg1889)
Glutathione s-transferase Recombination protein RecR Hypothetical protein 3-hydroxybutyrate dehydrogenase Elongation factor- Tu Elongation factorTu fragment Acetoacetate decarboxylase Acetoacetate decarboxylate Uncharacterized protein
gst (lpp1883)
Small heat shock protein
hspC2 (lpg2493)
recR (lpg2756) unknown (lpg0165) bdhA (lpg2316) Tuf1 (lpl0355) Tuf1 (lpl0355) adc (lpg0672) adc (lpg0672) Unknown (lpa 03493)
L. pneumophilasg 1 strain Paris
© 2014 APMIS. Published by John Wiley & Sons Ltd
MW kDa theoretical/ experimental
pI theoretical/ experimental
Matching peptides by PMF/number of signals
Sequence coverage (%)
Localization
47.9/48.0
4.8/5.0
22/46
52
61.9/62.5
6.1/6.2
16/26
38
Outer membrane cytoplasm
Involved in metabolism Transferase
35.5/37.0
6.3/6.5
14/21
51
Cytoplasm
22.8
6.8/7.0
11/34
38
DNA repair
22.0/22.0
6.2/6.4
13/53
62
Periplasm and cytoplasm cytoplasm
Unknown
16.9/17.0
5.5/5.5
8/38
41
unknown
Oxidoreductase
27.6/32.0
6.0/6.0
15/28
61
cytoplasm
Protein synthesis
43.3/43.0
5.2/5.2
19/35
57
Cytoplasm
Protein synthesis
43.3/26.5.0
5.2/5.4
18/41
42
Cytoplasm
Oxidoreductase
28.3/28.0
5.4/5.5
12/28
53
Cytoplasm
Oxidoreductase
28.3/28.0
5.4/5.7
15/33
62
Cytoplasm
Unknown (homology to OsmC family protein) Chaperon
21.1/23
5.5/5.5
9/44
42
Unknown
18.9/22.0
6.3/6.5
13/43
65
Cytoplasm
335
SCATURRO et al.
SYNTHESIS AND DEGRADATION OF KETONE BODIES
A
Fay acid metabolism
2-malate dehydrogenase
Pyruvate metabolism
Acetoacetyl-CoA Glycolysis Acetyl-CoA Sterol biosynthesis
(S)-3-Hydroxy-3methylglutaryl-CoA
Acetoacetate
acetoacetate decarboxylase
Butanoate metabolism
Acetone
3-hydroxybutyrate dehydrogenase
(R)-3-Hydroxy-butyrate
SYNTHESIS AND DEGRADATION OF PHB
B
CH3
O
Lipase gene analysis
-O-CH-CH2-C PHB PHB synthase
PHB depolymerase
3-hydroxybutyryl CoA
3-hydroxybutyrate
Hydroxybutyrate dehydrogenase (bdhA)
Acetocetil-CoA reductase
Acetoacetate
Acetoacetil-CoA
CoA transferase
Fig. 2. Scheme of synthesis and degradation pathways of ketone compounds (www.genome.jp) and PHB in bacterial metabolism (35).
tides upstream the start codon, hence bdhA promoter region from Vir+ and Vir strains (GenBank accession number KJ123748). bdhA Vir sequence showed 100% identity to the bdhA Vir+ one in both coding and promoter regions, thus no obvious nucleotide mutation affecting gene function and expression was detected. RT-PCR provided that bdhA Vir was transcriptionally active (Fig. 3A), but quantitative RT-PCR demonstrated that it was significantly down regulated (fold = 0.37, p < 0.0001) (Fig. 3B). Microscope observation of PHB granules
As previously reported, Legionella accumulates PHB granules as an endogenous reserve of energy
336
(29). For bacteria other than Legionella, it has also been demonstrated that BdhA is one of the enzymes of the metabolism of PHB, involved in the first step of the degradation process (28). Therefore, this function was investigated in the bdhA defective MS1 mutant and compared to that of the parental Lp02 strain by microscopy and quantification of the Nile red stained granules. The analysis demonstrated a significantly higher granule content in MS1 (p < 0.05) than the parental strain, with an average content of 0.89 granules per bacterium in MS1 and of 0.59 in Lp02 (Fig. 4A and C). Granule content was also determined in Vir+ and Vir strains. In Vir+ the Nile red staining was evident at 24 h of growth (Fig. 4B), but not at 48 h, while in Vir- no granule was detected (Fig. 4B). This observation suggested both a different metabolism of the PHB granules in Vir+ and also a possible defect at level of genes of the synthesis of the granules in Vir.
Lipase sequence was determined using the pBlip+ and pBlip recombinant vectors which include the entire coding region plus 878 nucleotide upstream of the start codon (GenBank accession number KJ123749). Comparison of the sequencing results revealed 99% of identity between Vir and Vir+ lipase with 5 mismatches identified. In the promoter region no nucleotide difference between Vir+ and Vir was found. RT-PCR revealed that transcription of lipase gene was indeed present and, as expected for the lower protein production, quantitative RT-PCR demonstrated that the Vir lipase transcript was significantly down regulated (fold = 0.39, p < 0.0001) (Fig. 3B). As determined by the ‘mfold’ software available at www.mobile. pasteur.fr, prediction of RNA secondary structures provided different possible secondary structures of the Vir lipase RNA than the Vit+ RNA (data not shown). The deducted Vir lipase amino acid sequence consisted of 321-amino acids and compared with the Vir+ sequence showed 2 different amino acids D151G and G285S, due to 2 of 5 observed nucleotide mismatches. The other 3 mismatches determined 3 conservative amino acid changes, i.e., E176D, K230Q and A239S (Fig. 5). In addition, Vir lipase amino acid sequence showed 98% of identity with lipases from all the Legionella genomes sequenced, and 7 different amino acids (in addition there were T32A and T46N). No signal peptide was found by using SignalP.4.1 software (http://www.cbs.dtu.dk/), suggesting its cytoplasmic localization. The Vir+ and Vir amino acid lipase sequence showed at the
© 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
A
B
M 1 2 3 4 5 6
bdhA
16S
C
M 1
bdhA lipase
2
3
4
Fold difference bdhA expression level
1.2 1 0.8 0.6 0.4 0.2
0 Vir+
Vir-
Lipase Fold difference lipase expression level
1.2 1 0.8 0.6 0.4 0.2 0 Vir+
Vir—
Fig. 3. bdhA and lipase transcripts were detected from post-exponential phase bacteria by RT-PCR in Vir+ (lane 3, 5) and Vir (lane 4, 6) strains (A). 16S rRNA (Vir+ in lane 1, Vir in lane 2) were used as reference RNA and to normalize the quantification of the transcripts. Quantitative RT-PCR and gene-specific primers were used to assess the fold changes in bdhA and lipase expression between Vir+ and Vir strains at the post-exponential phase (B). The data are means and standard deviations obtained from quadruplicate RNA samples. In (C), RT-PCR with specific primers to amplify bdhA and patD genes from MS1 showed a fragment of 450 bp corresponding to the deletion of bdhA gene and of 900 bp corresponding to patD gene. In lanes 3 and 4, there were the RNAs amplified with Taq polymerase to ascertain the absence of genomic DNA
position 159 a unique putative conserved catalytic motif GDSVG, similar to the sequence motif known as ‘lipase box’ (GX1SX2G) (Fig. 5). This protein was annotated in the Legionella pneumophila subsp. pneumophila str. Philadelphia 1 genome as a putative hydrolase/esterase encoded by the lpg1889 gene. Southern analysis demonstrated that the lpg1889 probe hybridized also to DNA from several other Legionella non-pneumophila, consisting with a highly conserved gene (data not shown). All these results indicated that significant alterations at the transcriptional level in this highly conserved lipase could be occurred. The nucleotide sequence changes also indicated variations of the RNA secondary structure that could reflect alterations of the polypeptide folding, determining its instability. Importance of bdhA and lipase genes for growth in Thp1 and A. castellanii
Defined mutants were constructed to address the possible involvement of bdhA and lipase in the virulence loss of the Vir strain. As attempts of genetic manipulation of Vir+ strain failed, the Lp02 strain was instead used and the MS1, defective in bdhA, and FP2, defective in lipase gene, were obtained. The growth of MS1 and FP2 strains in Thp-1 © 2014 APMIS. Published by John Wiley & Sons Ltd
macrophage-like cells and A. Castellanii amoebae was compared to that of the wild type strain. As shown in Fig. 6A and 6B, the mutants and wild type strain grew at the same rate in both macrophages and in amoebae demonstrating that both mutants were not affected in their ability of multiplying into host cells. Aurass et al. (33) have recently demonstrated that bdhA was co-transcribed with the downstream patD gene and the bdhA-patD operon was required for intracellular infection, so that it could be considered a novel virulence-associated determinant. To better explain the ability of MS1 to grow in the host cell, we determined whether patD function was still present in the MS1 strain. As expected for an in frame delete mutant, bdhA transcript was deleted of its greater part but the down-stream patD gene was still transcribed (Fig. 3C).
DISCUSSION In this study, we analyzed differences in the proteome of the spontaneous avirulent L. pneumophila sg 6 Vir mutant respect the wild type Vir+ strain. Most of the identified proteins were related to the ketone compound metabolism and/or involved in the degradation of PHB granules. As previously
337
SCATURRO et al.
A
B Vir+
Lp02
C
D Vir–
MS1
E Granules/bacterium
1.5 1 0.5 0
Lp02
SM1
Fig. 4. Staining of PHB granules’ content in Lp02 (A), MS1 (B), Vir+ (C) and Vir (D). Means and standard deviations of content of granules per bacterium were determined in Lp02 and MS1 strains, by counting at least 100 bacteria for each strain in three different image fields (E).
Fig. 5. Peptide alignment of Vir and Vir+ deduced lipase sequence showing the mismatches found as boldface. The aster (*) indicates the non-conservative and colon (:) indicates the conservative changes. In gray, it is shown the putative motive similar to a ‘lipase box’ (GX1SX2G).
observed by Hayashi et al. (34), we found that at the post-exponential growth phase, when Legionella expresses the virulent phenotype and genes like those for motility are active, L. pneumophila serogroup 6 Vir+ strain expressed genes encoding proteins that could be useful in starvation condition. In particular, this study was focused on two genes encoding the 3hydroxybutyrate dehydrogenase enzyme and a putative lipase, which were respectively undetectable and down-produced in the Vir mutant. RT-PCR showed that the Vir- mutant has been impaired in the bdhA and lipase expression with transcript levels 338
significantly lower than found in Vir+. Concerning bdhA, it is present on the bacterial chromosome near some flagellar genes, and because the Vir mutant is defective for the flagellin expression (17) it might be hypothesized a defect in a regulator factor that acts both on the bdhA and flagellin genes, thus connecting genes of motility with genes of metabolism. However, this dysfunction has not contributed to the loss of virulence of Vir, as a matter of fact the deletion of bdhA did not affect the ability of intracellular multiplication. This observation was also useful to clarify the role of bdhA and the down-stream patD © 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
Molplicazione in Thp1: MOI = 1
A
9 8.5 8 Log CFU/mL
7.5 7 6.5
Lp02
6
MS1 FP2
5.5 5
4.5 4 T0
T24
T48
T72
Molplicazione in A. castellanii. MOI = 1
B
10
Log CFU/mL
9 8 7 Lp02 6 MS1
5 FP2 4 T0
T24
T48
T72
Fig. 6. Intracellular multiplication of wild type Lp02 and MS1 and FP 2 Legionella mutants in Thp-1 macrophageslike cells (A) and in A. castellanii amoebae (B) are shown. The data are means and standard deviations CFU determined at 0, 24, 48 and 72 h after infection, from three independent experiments.
gene in the infection process. Indeed, it has recently been demonstrated (33) that in L. pneumophila bdhA is co-transcribed in a unique operon with patD, encoding a protein with phospholipase and lipophospholipase activities. L. pneumophila bdhA/patD mutants showed a severe growth defect in amoebae and U937 cell line. In this study we demonstrated that patD is actively transcribed in the MS1 strain, suggesting that only the patD but not bdhA has most probably a crucial role during the amoeba infection process. It is known for other microorganisms that BdhA catalyses the first steps of PHB granule degradation (28). In this study, the MS1 strain indeed showed a higher accumulation of PHB granules than its parental strain suggesting a role bdhA in the metabolism of this important source of energy. The different PHB accumulation in Vir+ and Vir strains compared to Lp02 and MS1 suggested also differences in the granule metabolism between L. pneumo© 2014 APMIS. Published by John Wiley & Sons Ltd
phila serogroup 6 and L. pneumophila serogroup 1 that would require further investigation, also regarding Legionella pneumophila environmental survival in stress conditions. The gene encoding for a putative lipase is annotated in the Legionella pneumophila subspecies Philadelphia genome as a a/b hydrolase/esterase belonging to the peptidase_S9 superfamily. Here, we demonstrated that the reduced transcriptional activity caused a very low lipase content as found by proteome analysis. The predicted changes in the mRNA secondary structure due to the mutations we observed in the lipase-coding region, might have determined the down-production of the protein. In addition, the deduced amino acid sequence showed 5 different amino acids that could interfere with the right folding and also influence the half-life of this enzyme. Finally, the presence of a motif similar to a catalytic triad typical of de-polymerase enzymes (i.e., GX1SX2G), characterizes its function as depolymerase in the metabolism of PHB granules. As PHB de-polymerases produce 3-hydroxibutyrate, the substrate for BdhA, a possible correlation between undetectable BdhA and down-regulation of lipase should be investigated. In conclusion, this study highlighted changes in an avirulent Legionella pneumophila serogroup 6 mutant in the production of proteins mainly involved both in lipid and carbohydrate metabolism, in turn linked to the synthesis and degradation of ketone compounds. It is known that Legionella utilizes amino acids as carbon source and it is also known that the post-exponential growth phase represents a time to switch virulence genes on and move to the transmissive phase, however, the expression of gene related to the energy metabolism has never been explored before. Therefore, even though the two genes investigated did not show a link with virulence, their expression should be considered in a greater contest, in which a damage of some secondary metabolic pathways could affect the Legionella pneumophila environmental fitness. The authors thank Dr Tamara O’Connor for critical reading of the manuscript and important advice about the experiments. They are also grateful to Dr Ralph Isberg for providing Lp02 strain and pSR47s vector, and Dr Stanimir Ivanov for useful suggestions about mutant constructions. They wish also to thank Massimo Mentasti for English editing.
REFERENCES 1. Fields BS, Benson RF, Besser RE. Legionella and Legionnaires’ disease: 25 years of investigation. Clin Microbiol Rev 2002;15:506–26.
339
SCATURRO et al.
2. Kagan JC, Roy CR. Legionella phagosomes intercept vesicular traffic from endoplasmic reticulum exit sites. Nat Cell Bio 2002;4:945–54. 3. Isberg RR, O’Connor TJ, Heidtman M. The Legionella pneumophila replication vacuole: making a cosy niche inside host cells. Nat Rev Microbiol 2009;7: 13–24. 4. Ninio S, Roy CR. Effector proteins translocated by Legionella pneumophila: strength in numbers. Trends Microbiol 2007;15:372–80. 5. Richards AM, Von Dwingelo JE, Price CT, Abu Kwaik Y. Cellular microbiology and molecular ecology of Legionella-amoeba interaction. Virulence 2013;15:307–14. 6. Jules M, Buchrieser C. Legionella pneumophila adaptation to intracellular life and the host response: clues from genomics and transcriptomics. FEBS Lett 2007;581:2829–38. 7. Br€ uggemann H, Cazalet C, Buchrieser C. Adaptation of Legionella pneumophila to the host environment: role of protein secretion, effectors and eukaryotic-like proteins. Curr Opin Microbiol 2006;9:86–94. 8. Abu Kwaik Y, Harb OS. Phenotypic modulation by intracellular bacterial pathogens. Electrophoresis 1999;20:2248–58. 9. Bandyopadhyay P, Xiao H, Coleman HA, Price-Whelan A, Steinman HM. Icm/dot-independent entry of Legionella pneumophila into amoeba and macrophage hosts. Infect Immun 2004;72:4541–51. 10. De Buck E, H€ oper D, Lammertyn E, Hecker M, Anne J. Differential 2-D protein gel electrophoresis analysis of Legionella pneumophila wild type and Tat secretion mutants. Int J Med Microbiol 2008;298: 449–61. 11. Chien M, Morozova I, Shi S, Sheng H, Chen J, Gomez SM, et al. The genomic sequence of the accidental pathogen Legionella pneumophila. Science 2004;305:1966–8. 12. Lebeau I, Lammertyn E, De Buck E, Maes L, Geukens N, Van Mellaert L, et al. First proteomic analysis of Legionella pneumophila based on its developing genome sequence. Res Microbiol 2005;156:119–29. 13. Galka F, Wai SN, Kusch H, Engelmann S, Hecker M, Schmeck B, et al. Proteomic characterization of the whole secretome of Legionella pneumophila and functional analysis of outer membrane vesicles. Infect Immun 2008;76:1825–36. 14. D’Auria G, Jimenez-Hern andez N, Peris-Bondia F, Moya A, Latorre A. Legionella pneumophila pangenome reveals strain-specific virulence factors. BMC Genom 2010;11:181. 15. Pastoris MC, Proietti E, Mauri C, Chiani P, Cassone A. Suckling CD1 mice as an animal model for studies of Legionella pneumophila virulence. J Med Microbiol 1997;46:647–55. 16. Ricci ML, Torosantucci A, Scaturro M, Chiani P, Baldassarri L, Pastoris MC. Induction of protective immunity by Legionella pneumophila flagellum in an A/J mouse model. Vaccine 2005;23:4811–20. 17. Scaturro M, Meschini S, Arancia G, Fontana S, Ricci ML. Characterization of a spontaneous avirulent mutant of Legionella pneumophila serogroup 6.Evidence of DotA and flagellin involvement in the loss of virulence. J Microbiol 2009;47:768–73.
340
18. Roy CR, Berger KH, Isberg RR. Legionella pneumophila DotA protein is required for early phagosome trafficking decisions that occur within minutes of bacterial uptake. Mol Microbiol 1998;28:663–74. 19. Andrews HL, Vogel JP, Isberg RR. Identification of linked Legionella pneumophila genes essential for intracellular growth and evasion of the endocytic pathway. Infect Immun 1998;66:950–8. 20. Sanchez JC, Rouge V, Pisteur M, Ravier F, Tonella L, Moosmayer M, et al. Improved and simplified ingel sample application using reswelling of dry immobilized pH gradients. Electrophoresis 1997;18:324–7. 21. Quaranta S, Giuffrida MG, Cavaletto M, Giunta C, Godovac-Zimmermann J, Ca~ nas B, et al. Human proteome enhancement: high-recovery method and improved two-dimensional map of colostral fat globule membrane proteins. Electrophoresis 2001;22: 1810–8. 22. Neuhoff V, Arold N, Taube D, Ehrhardt W. Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 1988;9: 255–62. 23. Hellmann U, Wernstedt C, Gonez J, Heldin CH. Improvement of an in-gel digestion procedure for the micropreparation of internal protein-fragments for aminoacid sequencing. Anal Biochem 1995;224:451–5. 24. Pappin DJ, Hojrup P, Bleasby AJ. Rapid identification of proteins by peptide-mass fingerprinting. Curr Biol 1993;3:327–32. 25. VanRheenen SM, Dumenil G, Isberg RR. IcmF and DotU are required for optimal effector translocation and trafficking of the Legionella pneumophila vacuole. Infect Immun 2004;72:5972–82. 26. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;25:402–8. 27. Belyi Y, Tabakova I, Stahl M, Aktories K. Lgt: a family of cytotoxic glucosyltransferases produced by Legionella pneumophila. J Bacteriol 2008;190:3026–35. 28. Aneja P, Charles TC. Poly-3-hydroxybutyrate degradation in Rhizobium (Sinorhizobium) meliloti: isolation and characterization of a gene encoding 3hydroxybutyrate dehydrogenase. J Bacteriol 1999; 181:849–57. 29. James BW, Mauchline WS, Dennis PJ, Keevil CW, Wait R. Poly-3-hydroxybutyrate in Legionella pneumophila, an energy source for survival in low-nutrient environments. Appl Environ Microbiol 1999;65:822–7. 30. Verlinden RA, Hill DJ, Kenward MA, Williams CD, Radecka I. Bacterial synthesis of biodegradable polyhydroxyalkanoates. J Appl Microbiol 2007;102: 1437–49. 31. Gardu~ no RA, Gardu~ no E, Hiltz M, Hoffman PS. Intracellular growth of Legionella pneumophila gives rise to a differentiated form dissimilar to stationaryphase forms. Infect Immun 2002;70:6273–83. 32. Lang C, Flieger A. Characterisation of Legionella pneumophila phospholipases and their impact on host cells. Eur J Cell Biol 2011;90:903–12. 33. Aurass P, Pless B, Rydzewski K, Holland G, Bannert N, Flieger A. A novel large-scale approach for identi-
© 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
fication of Legionella pneumophila mutants defective for amoeba infection uncovers the bdhA/patD operon as a virulence determinant. Appl Environ Microbiol 2009;75:4506–15. 34. Hayashi T, Nakamichi M, Naitou H, Ohashi N, Imai Y, Miyake M. Proteomic analysis of growth phasedependent expression of Legionella pneumophila pro-
© 2014 APMIS. Published by John Wiley & Sons Ltd
teins which involves regulation of bacterial virulence traits. PLoS ONE 2010;22:11718. 35. Charles TC, Cai GQ, Aneja P. Megaplasmid and chromosomal loci for the PHB degradation pathway in Rhizobium (Sinorhizobium) meliloti. Genetics 1997;146:1211–20.
341
APMIS 123: 330–341
Identification and characterization of genes, encoding the 3-hydroxybutyrate dehydrogenase and a putative lipase, in an avirulent spontaneous Legionella pneumophila serogroup 6 mutant MARIA SCATURRO,1 CRISTINA BARELLO,2 MELANIA DE GIUSTI,1 STEFANO FONTANA,1 FEDERICA PINCI,1 MARIA GABRIELLA GIUFFRIDA2 and MARIA LUISA RICCI1 1
Department of Infectious Parasitic Immune-mediated Diseases, Istituto Superiore di Sanit a, Rome and Istituto di Scienze delle Produzioni Alimentari, CNR, Sezione di Torino, Colleretto Giacosa (TO), Italy
2
Scaturro M, Barello C, De Giusti M, Fontana, S, Pinci F, Giuffrida MG, Ricci ML. Identification and characterization of genes, encoding the 3-hydroxybutyrate dehydrogenase and a putative lipase, in an avirulent spontaneous Legionella pneumophila serogroup 6 mutant. APMIS 2015; 123: 330–341. Legionella pneumophila is a pathogen widespread in aquatic environment, able to multiply both within amoebae and human macrophages. The aim of this study was to identify genes differently expressed in a spontaneous avirulent Legionella pneumophila serogroup 6 mutant, named Vir, respect the parental strain (Vir+), and to determine their role in the loss of virulence. Protein profiles revealed some differences in Vir proteomic maps, and among the identified proteins the undetectable 3-hydroxybutyrate dehydrogenase (BdhA) and a down-produced lipase. Both Legionella enzymes were studied before and were here further characterized at genetic level. A significant down-regulation of both genes was observed in Vir at the transcriptional level, but the use of defined mutants demonstrated that they did not affect the intracellular multiplication. A mutant (MS1) showed an accumulation of poly-3-hydroxybutyrate (PHB) granules suggesting a role of bdhA gene in its degradation process. The lipase deduced amino acid sequence revealed a catalytic triad, typical of the ‘lipase box’ characteristic of PHB de-polymerase enzymes, that let us suppose a possible involvement of lipase in the PHB granule degradation process. Our results revealed unexpected alterations in secondary metabolic pathways possibly linking the loss of virulence to Legionella lack of energy sources. Key words: Legionella; bdhA; lipase; proteome; ketone compounds. Maria Scaturro, Department of Infectious Parasitic Immune-mediate Diseases, Istituto Superiore di Sanit a, Rome, Italy. e-mail: [email protected]
Legionella pneumophila is a facultative intracellular pathogen causing an atypical pneumonia known as Legionnaires’ disease. Legionella is ubiquitous in natural freshwater environments and moist soil, and is capable of multiplying within a wide range of host cells, such as protozoa and human macrophages (1). After inhalation or micro-aspiration, in the lungs Legionella cells are engulfed by macrophages into a membrane bound compartment known as Legionella -containing vacuoles (LCVs). LCVs do not fuse with lysosomes, but are surrounded by smooth vesicles originating from the endoplasmic reticulum to generate a niche that Received 16 May 2014. Accepted 22 October 2014
330
supports bacterial growth (2, 3). The ability to subvert the host membrane traffic is the key of Legionella replication within eukaryotic cells. Legionella produces a Type IV secretion system known as Dot/Icm that injects > 200 bacterial proteins into the host cells promoting control of the membrane transport (4). The intracellular lifecycle consists of a replicative and a transmissive phase with the transition from the one and the other characterized by differential expression of several genes (5, 6). Most of the known virulence factors are up-regulated during the transmissive phase when the bacteria appear flagellated, highly motile and ready to exit from the current host and invade a new host cell (7). Several known molecular mechanisms
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
regulating Legionella pathogenesis have been characterized using defined mutants. However, spontaneous avirulent mutants are also powerful tools for virulence studies. Two-dimensional electrophoresis (2-DE) is a useful approach to study global changes in bacterial protein production occurring during intracellular growth (8), and to differentiate proteins secreted through vesicles from those that are naturally soluble (9, 10). Furthermore, with genomic and proteomic analysis now available for some L. pneumophila strains, many proteins have been identified, including several virulence factors (11–14). In a previous study we characterized a spontaneous avirulent L. pneumophila serogroup 6 strain, named Vir, obtained by multiple passages on agar plates (15, 16). Vir did not have a flagellum and alterations were found in flaA and dotA gene expression when compared to Vir+ (17). Although no detectable level of flaA expression was observed, no alterations were found in the nucleotide sequence of both promoter and encoding regions. In contrast, the dotA sequence revealed an in frame transition mutation of C to T at nucleotide 1006 that produced a stop codon. DotA was consequently not produced and the accumulation of lysosomal marker LAMP-1 on phagosomes containing the Vir strain was observed as previously demonstrated in L. pneumophila serogroup 1 (17, 18). In this study, Vir proteome was explored to determine further differently produced proteins in comparison to the Vir+ proteome, which might have affected the loss of Vir virulence. Out of 13 identified proteins, two produced at lower levels in the Vir strain, namely BdhA and a putative lipase, were further analyzed at gene level and the possible involvement in virulence was also verified. An alteration of metabolic pathways involving the synthesis and degradation of the PHB granules and of the ketone compounds was also observed. MATERIALS AND METHODS Bacterial strains, plasmids and growth conditions All bacterial strains and plasmids used in this study are listed in Table 1. Legionella strains were grown on Buffered Charcoal yeast Extract with a-ketoglutarate (a-BCYE) agar plates and in BYE broth (Oxoid, Hampshire, UK) supplemented with 100 lg/mL Thymidine (Sigma-Aldrich, St. Louis, MO, USA) (a-BCYE-T) when required. L. pneumophila strain Lp02 (kindly supplied by Dr Isberg) was used to construct deletion mutants. DH5a k pir and MT607 E. coli strains were used to propagate pSR47s vector (also kindly provided by Dr Isberg) and for Legionella mating, respectively (19). When required, antibiotics were added to the media at the following final concentrations: Kanamycin (IBI, Peosta, IA, USA) at
© 2014 APMIS. Published by John Wiley & Sons Ltd
100 lg/mL for E. coli and 25 lg/mL for Legionella; streptomycin (IBI) at 50 lg/mL and ampicillin at 100 lg/mL.
Protein extraction Vir+ and Vir strains were grown in 15 ml BYE-broth to post-exponential phase (OD600 = 2) prior to total protein extraction. Bacterial cells were collected by centrifugation at 4500 g for 10 min and pellets re-dissolved in ice-cold sterile distilled-H2O to determine the colony forming unit/ mL (CFU/mL) by serial dilutions on a-BCYE agar plates. Bacterial cells were then centrifuged as above and pellets re-suspended in 5 mL ice-cold Tris-EDTA solution containing 1 mM PMSF (Sigma-Aldrich), and lysed with 50 mg/ml lysozyme (Sigma-Aldrich) and 10% w/v SDS on ice for 10 min. Nucleic acids were removed by digestion with DNase (10 U/lL, Invitrogen-Life technologies, Monza, Italy) and RNase (10 mg/mL, Invitrogen-Life technologies), for 15 min on ice. Cellular debris was removed by two consecutive centrifugations, the first at 10 000 g for 10 min and the second one at 17 000 g for 1 h. Supernatant was divided into aliquots (100 lL) and proteins precipitated with 3 volumes of ice-cold acetone, for a minimum of 2 h at 20 °C. Protein pellets were obtained by centrifugation, 13 000 g for 20 min at 4 °C and stored at 20 °C until used. The extraction was repeated for 3 preparations and protein concentrations were determined using the Bradford assay (Pierce-Thermo fisher Scientific, Rockford, IL, USA).
Protein solubilization and two-dimensional electrophoresis In order to perform Isoelectric focusing (IEF) analysis, dried pellets were solubilized in 250 lL of re-hydration solution (7.0 M urea, 2.0 M Thiourea, 4% CHAPS, 1% Triton X-100, 0.02 M Tris, 0.5% IPG buffer either pH 4–7 or pH 3–10 (GE Healthcare, Pittsburg, CA, USA), 1% DTT) and applied to 13 cm IPG strips (GE Healthcare) using the in-gel re-hydration method (20). After 12 h of re-hydration, IEF was performed with a linear gradient ranging from either 3 to 10 or 4 to 7 pI, at a constant temperature of 20 °C with 58 000 V 9 h. After IEF, the IPG strips were incubated for 15 min at room temperature in 6 M urea, 30% v/v glycerol, 2% w/v SDS, 50 mM Tris-HCl pH 8.6 enriched with 2% w/v DTT. A second equilibration step was carried out for 15 min in the same solution, with the exception that DTT was replaced by 2.5% w/v iodoacetamide plus a trace of bromophenol blue as tracking dye. The IPG strips were then placed at the top of the 1.0 mm vertical second dimensional gels and sealed. SDS-PAGE for all experiments was carried out as previously described (21). Running conditions were 400 V, 50 mA/gel, for 3 h. Three gel replicates were performed for each sample. Gels were automatically stained using the Processor Plus device (GE Healthcare) with freshly prepared Neuhoff stain (22).
Proteome data analysis 2-DE gels were scanned by a GS-800 Densitometer (BioRad, Hercules, CA, USA) and analyzed with Progenesis PG 220 software (Nonlinear Dynamics, UK). The average of
331
SCATURRO et al.
Table 1. Bacterial strains and plasmids Strain or plasmid L. pneumophila Lp02 Vir+ Vir MS1 FP2 E. coli DH5a (kpir) MT607 XL1-Blue Plasmids pSR47s pSDbdhA pSDlipase pGEM-Teasy pGbdhA+ pGbdhA pBlueScript (SK+) pBlip+ pBlip
Relevant genotype
Reference or source
hsdRrpsLthyA Lp6 clinical strain Lp6 clinical strain avirulent mutant Lp02DBdhA Lp02Dlipase
(19) (17) (17) This study This study
DH5a (kpir) tet::Mu recA pRK600 F’::Tn10 proA+B+lacIqD(lacZ)M15/ recA1 endA1 gyrA96 (Nalr) thi hsd R17 (rk-mk+) supE44 relA1 lac
(19) (19) (19)
pSR47sacB pSR47sDbdhA pSR47sDlipase
(19) This study This study Promega This study This study Stratagene This study This study
pGEM-bdhAVir+ pGEM-bdhAVirpBlueScriptlipaseVir+ pBlueScriptlipaseVir-
each gel was determined by calculating the mean of the triplicates. Spot intensities were measured as normalized spot volumes. Spots were selected on the basis of a 2-fold difference in production between the two strains that was statistically significant based on a Student’s t-test p < 0.05.
Protein identification Spots were excised from the gels and subjected to ‘in-gel’ trypsin digestion (Promega, Madison, WI USA), as described by Hellmann et al. (23). For MALDI-TOF analysis, 0.5 lL of each peptide mixture was applied to a target disk, and allowed to dry. Subsequently, 0.5 lL of HCCA matrix solution (a-cyano-4-hydroxycinnamic acid) was applied to the dried sample and allowed to dry again. Spectra of protein digests were obtained using a Bruker Reflex III mass spectrometer (Bruker Daltonics, Italy). MSFit (http://prospector.ucsf.edu/prospector/cgi-bin/msform. cgi?form=msfitstandard) software package was used to interpret the MS spectra through the peptide mass fingerprinting method (PMF) (24). The software package was searched against UniProt KB database (www.uniprot.org). The following parameters were used: S-carbamidomethyl derivate on cysteine as fixed modification, oxidation on methionine as variable modification and two missed cleavage sites for trypsin digestion. Peptide mass tolerance was 30 ppm.
tide sequences were analyzed using the above-mentioned primers and those indicated by an asterisk in Table 2.
Construction of in frame DbdhA and Dlipase gene deletions and sequencing analysis In frame deletions in bdhA and lipase were introduced into the Lp02 chromosome by allelic exchange using the pSR47s suicide vector (25). Briefly, regions located upstream and downstream of the bdhA gene were amplified using primers 1.1bdhA with 1.2bdhA and 2.1bdhA with 2.2bdhA, respectively and then linked by overlapping recombinant PCR using primers 1.1bdhA and 2.2bdhA. The lipase gene deletion was similarly generated using 1.1lipase with 1.2lipase and 2.1lipase with 2.2lipase. The oligonucleotide sequences of each primer are listed in Table 2. Restriction sites (underlined in Table 2) were introduced at the 50 and 30 end of the recombinant fragments and used to clone the final products into similarly digested pSR47s. The resulting pSDbdhA and pSDlipase recombinant plasmids were introduced into Lp02 via mating for the homolog allelic exchange. Selection for plasmid integration was obtained by resistance to kanamycin on a-BCYE-T-Kan plates. Purified colonies were then grown on a-BCYE plates containing 5% sucrose to counter-select for the plasmid. Finally, colonies were screened for the presence of deletion and the absence of the targeted gene by both PCR and sequence analysis. Legionella MS1 (DbdhA) and FP2 (Dlipase) strains were thus isolated and used in further analyses.
Sequence analysis Genomic DNA from Vir+ and Vir strains was extracted using 20% Chelex 100 (Sigma-Aldrich) followed by boiling for 10 min. The bdhA and lipase genes were amplified by PCR by using the 1.1bdhA with 2.2bdhA and 1.1lipase with 2.2lipase specific primers, respectively (Table 2). Amplicons were subcloned into pGEM-Teasy, obtaining the pGbdhA+ and pGbdhA, and pBlueScript, obtaining pBlip+ and pBlip recombinant vectors and the nucleo-
332
Southern analysis Genomic DNA from L. longbeachae (ATCC 33462), L. gormanii (ATCC 33297), L. dumoffii (CDC collection), L. bozemani (ATCC 35545), L. micdadei (ATCC 33204), L. oakridgensis (ATCC 33761), L.jamestowniensis (ATCC 35298), L. quinlivanii (kindly provided by Dr Lee J.) and L. parisiensis (ATCC 35299) were digested with EcoRI © 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
Table 2. Primers used in this study Primer Sequence1 1.1bdhA ATAtctagaTAAAGTACCAGCCTCACCAGTCGT 1.2bdhA caaatgacattgAGATCAGCGATTGCTACCTTTGCC 2.1bdhA aatcgctgatctCAATGTCATTTGCCCGGGGTTTGT 2.2bdhA ATAggatccGGTGGGAACCATCCCAAAAGAATCA DbdhA.seq2 TGTTACAGGAGCAGCAAGCGG F bdhA.seq2 TTCATGGAAAACAATGACAAGGATAG R bdhA.seq2 CAATATGGCGATACTCAATCTGTC F bdhA GCAAAAGAATTAGGCATCAGTGAAG R bdhA CAATATGGCGATACTCAATCTGTC 1.1lip ATAgtcgacGTGCGCTTATCCCACATTGCTT 1.2 lip atccgcttggaCGGTTTTCCACCCGCATTTTGT 2.1 lip gtggaaaaccgTCCAAGCGGATGTTAGTCATGCT 2.2 lip ATAggatccGGAGAGGGAATTAAGGCAATTGCCA F lip.seq2 GGGCCACCACGATCTTTAGC R lip.seq2 CCACCTGTTATTGTTATCCAC F lipase probe ATCCCAGACACCATCCACCTGTTA R lipase probe TCAGCATGACTAACATCCGCTTGG F lipase ATGCGGGTGGAAAACCGATT R lipase GTGTCGGGTACTGAGCTTCT F rib16S CGTAGAGATCGGAAGGAACACC R rib16S GCGTGGACTACCAGGGTATCT 1 Underlined letters are the sequences of restriction enzymes used for cloning. 2 Primers used for sequencing. and PstI restriction enzymes, and the fragments were separated by electrophoresis on 1% agarose gel. Genome fragments were then transferred to nylon membrane (Amersham-GE Healthcare, Pittsburg, CA, USA) by capillary blotting. A DNA probe corresponding to the lipasecoding region was amplified by using the F lipase probe and R lipase probe primers shown in Table 2, then labeled and detected by using North2South Chemiluminescent Hybridization and Detection kit (Pierce- Thermo fisher Scientific).
Reverse Transcription (RT)-PCR analysis To verify the transcription of bdhA and lipase, Vir+ and Vir strains were grown to the post-exponential phase and harvested by centrifugation at 4500 g for 10 min at 4 °C. Total RNA was then extracted and purified by Nucleospin RNA II (Machery-Nagel, Duren, Germany) and 100 ng were used for retro-transcription reaction with ImProm-IITM Reverse Transcription System (Promega), according to the manufacturer’s instructions. cDNA synthesis was obtained by using specific primer pairs: bdhA cDNA was generated using the F bdhA.seq and R bdhA.seq primers, located directly upstream and downstream of the bdhA open reading frame; lipase cDNA was generated using the F lipase and R lipase primers (Table 2). cDNA was obtained also from RNA extracted from the MS1 mutant to further confirm the bdhA deletion (with the 1.1bdhA and 2.2bdhA primers) and to provide the transcription of the downstream patD (with 2.1bdhA and 2.2bdhA primers). The absence of genomic DNA was verified by amplification with the AmpliTaq polymerase (Applied Biosystems-life technologies, Monza, Italy).
Quantitative RT-PCR To determine levels of bdhA and lipase transcripts in the Vir+ and Vir strains, quantitative RT-PCR was carried © 2014 APMIS. Published by John Wiley & Sons Ltd
Restriction enzyme XbaI BamHI
SalI BamHI
out. To this aim, specific cDNAs were generated and an aliquot of 2 lL was added to 20 lL reaction mixture, which included 10 lL of SYBR green master mix (BioRad) and 0.5 lM of each primer. Primer pairs were as follow: F bdhA and R bdhA, F lipase and R lipase (Table 2). 16S cDNA was used as reference gene to normalize gene expression and the primer pair was F rib16S and R rib16S (Table 2). Reactions were performed on a Chromo-4 real-time PCR detection system (BioRad) using the following parameters: 95 °C for 3 min, 95 °C for 30 s, 57 °C for 30 s and 72 °C for 30 s for 30 cycles. For each experiment, 4 replicates were done and each of them was tested twice. Gene expression level was assessed by determining the cycle at which the amplification crossed the detection threshold (Ct). The fold difference of the two target genes between Vir+ and Vir was calculated using the 2DDCt method (26).
Intracellular growth assay A. castellanii (ATCC 30234) and Thp-1 monocytes (ATCC TIB-202), a human cell line that differentiates into macrophages-like cells upon treatment with 20 ng/ml phorbol ester (12-O_tetradecanoylphorbol-13-acetate (SigmaAldrich) for 72 h, were used to test the intracellular growth. A. castellanii were maintained in PYG (peptoneyeast extract-glucose) medium (Oxoid, Thermo Fisher Scientific) in 75 cm2 tissue culture flasks for no more than 1 week. On the day of infection, cells were transferred to 24-well tissue culture plates with a final concentration of 105 amoebae/mL. Thp-1 cells were grown in RPMI completed medium (Euroclone, Milano, Italy) and after differentiation seeded in 24-well tissue culture plates at a final concentration of 106 cells/mL. Lp02, MS1 and FP2 strains were used to infect amoeba and macrophage cells at an MOI of 1. At 0, 24, 48, and 72 h time points the of Thp-1 and A. castellanii monolayers were lysed with 1 ml of dis-
333
SCATURRO et al.
tilled sterile water and 0.2% Triton 100 (ICN Biomedicals, Irvine, CA, USA) respectively (27), then serial dilutions were plated on a-BCYE agar plates to determine Legionella CFU/mL.
Visualization of PHB granules Lp02, MS1, Vir+, and Vir strains were grown in BYEbroth overnight at 37 °C. A 10-fold dilution was inoculated in fresh medium then after 24, 48 and 72 h of incubation at 37 °C, 100 lL were removed and centrifuged to pellet bacterial cells. Pellets were then treated with 1% formalin for at least 3 h. Dry smears of formalin-treated bacteria on coverslips were heat-fixed for 10 min and finally stained with 25 mM Nile red (Sigma-Aldrich), diluted 1:250, for 30 min at 37 °C. Coverslips were then examined by fluorescence microscopy (Leica, Milano, Italy).
RESULTS Proteome analysis
Spontaneous avirulent mutants are valuable tools in bacterial pathogenesis investigations, although they have a strong limit due to the difficulty of determining promptly how many and what factors have been altered. Proteome analysis is a helpful approach because it permits to bypass this limit. In this study, proteomic analysis was performed with protein extracts obtained from bacteria grown to the post-exponential phase, in which Legionella shows a virulent phenotype. Although equivalent amounts of bacterial cells (1.3 1 9 109 CFU/ mL) were used to prepare protein extracts, a slightly different protein recovery between the two strains, namely 0.68 0.04 mg/mL for Vir+ and 0.97 0.07 mg/mL for Vir, was observed. Nevertheless, comparing Vir and Vir+ proteome maps, differently produced proteins were observed and some of them were identified. Twenty-four spots from the Vir+ maps, either undetectable or downproduced in both the 3–10 pI and 4–7 pI Vir maps, were chosen according to the statistical criteria but only 13 were successfully identified. Figure 1 shows two representative proteome maps, obtained by 2-DE of the Vir+ and Vir proteins. The 3–10 pI IEF separation shows the Vir+ and Vir total protein expression (Fig. 1A and C, respectively). The zoomed IEF 4–7 pI maps are also reported in Fig. 1B and D. The analyzed spots and the corresponding proteins are listed in Table 3 and shown by arrows in Fig. 1A and B. In Vir protein maps, spots corresponding to flagellin (spot 1), 2-malate dehydrogenase (spot 2), 4-glutatione S-transferase (spot 4), a hypothetical protein (spot 6) and 3-hydroxybutyrate dehydrogenase (spot 7) were undetectable.
334
Quantification of normalized spot volumes showed a statistically significant down-production of spot 3, 5, 8, 9, 10, 11, 12 and 13. These spots corresponded to a putative lipase (spot 3), a recombinant protein RecR involved in DNA repair (spot 5), elongation factor Tu (spots 8 and 9), an acetoacetate decarboxylase (spots 10 and 11), a hypothetical protein with homology (46%) to OsmC family protein (spot 12) and a small heat shock protein (spot 13), respectively. Among these proteins, the lipase was the most down-produced. Except for flagellin, most of the identified proteins suggested that important changes might have occurred in specific metabolic pathways, in particular involving the degradation of PHB granules and ketone metabolism, usually considered to be not essential for Legionella growth (Fig. 2). Proteins analyzed at gene level: BdhA and the putative lipase
3-hydroxybutyrate dehydrogenase (BdhA) is an enzyme known to be involved in the first step of PHB degradation and, as demonstrated for other bacteria, is encoded by the bdhA gene (28). It is known that PHB accumulates within cells of various bacterial species as well as Legionella, forming granules-like cytoplasmic inclusions that represent an important reserve of energy especially in low nutrient environments (29, 30). In HeLa cell infection experiments, it has been observed that Legionella accumulates and metabolizes PHB granules during the transmissive phase (5, 31). However, because BdhA catalyses the oxidoreduction of 3-hydroxibutyrate to acetoacetate, it also plays a role in the metabolic pathway of ketone compounds (Fig. 2). On the other hand, lipases are enzymes that can act in different ways: as membrane destructors or as inducers of host cell processes such as cell migration and membrane trafficking. Lipase activities have been often related to virulence mechanisms adopted by bacteria as Legionella (32). For all the above-mentioned reasons and because still uncharacterized, we chose to explore the bdhA and the putative lipase genes. In particular, to determine whether the undetected or lower produced proteins were due to mutations affecting gene expression, the chosen approach was to analyze the nucleotide sequence. bdhA gene sequence analysis
bdhA sequence analysis was determined using pGbdhA+ and pGbdhA vectors which contain the entire open reading frame (ORF) plus 277 nucleo© 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
MW kDa
pI 3
MW
10
pI 4
kDa
A
7
B 97
97
66 66
2 1
45
45
8
3 7
10
31
31
11
9
4
12 13
5
21
21
6
14
14
MW kDa
pI 3
MW
10
pI 4
kDa
C
7
D 97
97
66
66
45
45
31
31
21
21
14
14
Fig. 1. 2-DE maps of total protein extracts of Vir+ (Panels A, B) and Vir (Panels C, D) separated by IEF 3-10 pI (Panels A, C) and 4-7 pI (Panels B, D) and then by 12.5% SDS-PAGE. Arrows indicate the analyzed spots. Table 3. Spot identification from the L. pneumophila serogroup 6 Vir+ map Spot
UniProt KB entry
Name
Gene (ID)
Species
Function
1
Q7ATT9
Flagellin
flaA (AJ496382)
2
Q5ZRB1
maeA (lpg2971)
Q5ZUB8
4
Q5X3Z8
5
Q5ZRX0
6
Q5ZZ46
7
Q5ZT49
8
Q5WZL4
9
Q5WZL4-
10
Q5ZXQ9
11
Q5ZXQ9
12
D5T890
L. pneumophila sg6 Vir+ L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Paris L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Lens L. pneumophila sg1 strain Lens L. pneumophila sg1 strain Philadelphia L. pneumophila sg1 strain Philadelphia L. pneumophila sg 1 strain 2300/99 Alcoy
Structural protein Oxidoreductase
3
Malate dehydrogenase Lipase
13
Q5ZSM4
lipase (lpg1889)
Glutathione s-transferase Recombination protein RecR Hypothetical protein 3-hydroxybutyrate dehydrogenase Elongation factor- Tu Elongation factorTu fragment Acetoacetate decarboxylase Acetoacetate decarboxylate Uncharacterized protein
gst (lpp1883)
Small heat shock protein
hspC2 (lpg2493)
recR (lpg2756) unknown (lpg0165) bdhA (lpg2316) Tuf1 (lpl0355) Tuf1 (lpl0355) adc (lpg0672) adc (lpg0672) Unknown (lpa 03493)
L. pneumophilasg 1 strain Paris
© 2014 APMIS. Published by John Wiley & Sons Ltd
MW kDa theoretical/ experimental
pI theoretical/ experimental
Matching peptides by PMF/number of signals
Sequence coverage (%)
Localization
47.9/48.0
4.8/5.0
22/46
52
61.9/62.5
6.1/6.2
16/26
38
Outer membrane cytoplasm
Involved in metabolism Transferase
35.5/37.0
6.3/6.5
14/21
51
Cytoplasm
22.8
6.8/7.0
11/34
38
DNA repair
22.0/22.0
6.2/6.4
13/53
62
Periplasm and cytoplasm cytoplasm
Unknown
16.9/17.0
5.5/5.5
8/38
41
unknown
Oxidoreductase
27.6/32.0
6.0/6.0
15/28
61
cytoplasm
Protein synthesis
43.3/43.0
5.2/5.2
19/35
57
Cytoplasm
Protein synthesis
43.3/26.5.0
5.2/5.4
18/41
42
Cytoplasm
Oxidoreductase
28.3/28.0
5.4/5.5
12/28
53
Cytoplasm
Oxidoreductase
28.3/28.0
5.4/5.7
15/33
62
Cytoplasm
Unknown (homology to OsmC family protein) Chaperon
21.1/23
5.5/5.5
9/44
42
Unknown
18.9/22.0
6.3/6.5
13/43
65
Cytoplasm
335
SCATURRO et al.
SYNTHESIS AND DEGRADATION OF KETONE BODIES
A
Fay acid metabolism
2-malate dehydrogenase
Pyruvate metabolism
Acetoacetyl-CoA Glycolysis Acetyl-CoA Sterol biosynthesis
(S)-3-Hydroxy-3methylglutaryl-CoA
Acetoacetate
acetoacetate decarboxylase
Butanoate metabolism
Acetone
3-hydroxybutyrate dehydrogenase
(R)-3-Hydroxy-butyrate
SYNTHESIS AND DEGRADATION OF PHB
B
CH3
O
Lipase gene analysis
-O-CH-CH2-C PHB PHB synthase
PHB depolymerase
3-hydroxybutyryl CoA
3-hydroxybutyrate
Hydroxybutyrate dehydrogenase (bdhA)
Acetocetil-CoA reductase
Acetoacetate
Acetoacetil-CoA
CoA transferase
Fig. 2. Scheme of synthesis and degradation pathways of ketone compounds (www.genome.jp) and PHB in bacterial metabolism (35).
tides upstream the start codon, hence bdhA promoter region from Vir+ and Vir strains (GenBank accession number KJ123748). bdhA Vir sequence showed 100% identity to the bdhA Vir+ one in both coding and promoter regions, thus no obvious nucleotide mutation affecting gene function and expression was detected. RT-PCR provided that bdhA Vir was transcriptionally active (Fig. 3A), but quantitative RT-PCR demonstrated that it was significantly down regulated (fold = 0.37, p < 0.0001) (Fig. 3B). Microscope observation of PHB granules
As previously reported, Legionella accumulates PHB granules as an endogenous reserve of energy
336
(29). For bacteria other than Legionella, it has also been demonstrated that BdhA is one of the enzymes of the metabolism of PHB, involved in the first step of the degradation process (28). Therefore, this function was investigated in the bdhA defective MS1 mutant and compared to that of the parental Lp02 strain by microscopy and quantification of the Nile red stained granules. The analysis demonstrated a significantly higher granule content in MS1 (p < 0.05) than the parental strain, with an average content of 0.89 granules per bacterium in MS1 and of 0.59 in Lp02 (Fig. 4A and C). Granule content was also determined in Vir+ and Vir strains. In Vir+ the Nile red staining was evident at 24 h of growth (Fig. 4B), but not at 48 h, while in Vir- no granule was detected (Fig. 4B). This observation suggested both a different metabolism of the PHB granules in Vir+ and also a possible defect at level of genes of the synthesis of the granules in Vir.
Lipase sequence was determined using the pBlip+ and pBlip recombinant vectors which include the entire coding region plus 878 nucleotide upstream of the start codon (GenBank accession number KJ123749). Comparison of the sequencing results revealed 99% of identity between Vir and Vir+ lipase with 5 mismatches identified. In the promoter region no nucleotide difference between Vir+ and Vir was found. RT-PCR revealed that transcription of lipase gene was indeed present and, as expected for the lower protein production, quantitative RT-PCR demonstrated that the Vir lipase transcript was significantly down regulated (fold = 0.39, p < 0.0001) (Fig. 3B). As determined by the ‘mfold’ software available at www.mobile. pasteur.fr, prediction of RNA secondary structures provided different possible secondary structures of the Vir lipase RNA than the Vit+ RNA (data not shown). The deducted Vir lipase amino acid sequence consisted of 321-amino acids and compared with the Vir+ sequence showed 2 different amino acids D151G and G285S, due to 2 of 5 observed nucleotide mismatches. The other 3 mismatches determined 3 conservative amino acid changes, i.e., E176D, K230Q and A239S (Fig. 5). In addition, Vir lipase amino acid sequence showed 98% of identity with lipases from all the Legionella genomes sequenced, and 7 different amino acids (in addition there were T32A and T46N). No signal peptide was found by using SignalP.4.1 software (http://www.cbs.dtu.dk/), suggesting its cytoplasmic localization. The Vir+ and Vir amino acid lipase sequence showed at the
© 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
A
B
M 1 2 3 4 5 6
bdhA
16S
C
M 1
bdhA lipase
2
3
4
Fold difference bdhA expression level
1.2 1 0.8 0.6 0.4 0.2
0 Vir+
Vir-
Lipase Fold difference lipase expression level
1.2 1 0.8 0.6 0.4 0.2 0 Vir+
Vir—
Fig. 3. bdhA and lipase transcripts were detected from post-exponential phase bacteria by RT-PCR in Vir+ (lane 3, 5) and Vir (lane 4, 6) strains (A). 16S rRNA (Vir+ in lane 1, Vir in lane 2) were used as reference RNA and to normalize the quantification of the transcripts. Quantitative RT-PCR and gene-specific primers were used to assess the fold changes in bdhA and lipase expression between Vir+ and Vir strains at the post-exponential phase (B). The data are means and standard deviations obtained from quadruplicate RNA samples. In (C), RT-PCR with specific primers to amplify bdhA and patD genes from MS1 showed a fragment of 450 bp corresponding to the deletion of bdhA gene and of 900 bp corresponding to patD gene. In lanes 3 and 4, there were the RNAs amplified with Taq polymerase to ascertain the absence of genomic DNA
position 159 a unique putative conserved catalytic motif GDSVG, similar to the sequence motif known as ‘lipase box’ (GX1SX2G) (Fig. 5). This protein was annotated in the Legionella pneumophila subsp. pneumophila str. Philadelphia 1 genome as a putative hydrolase/esterase encoded by the lpg1889 gene. Southern analysis demonstrated that the lpg1889 probe hybridized also to DNA from several other Legionella non-pneumophila, consisting with a highly conserved gene (data not shown). All these results indicated that significant alterations at the transcriptional level in this highly conserved lipase could be occurred. The nucleotide sequence changes also indicated variations of the RNA secondary structure that could reflect alterations of the polypeptide folding, determining its instability. Importance of bdhA and lipase genes for growth in Thp1 and A. castellanii
Defined mutants were constructed to address the possible involvement of bdhA and lipase in the virulence loss of the Vir strain. As attempts of genetic manipulation of Vir+ strain failed, the Lp02 strain was instead used and the MS1, defective in bdhA, and FP2, defective in lipase gene, were obtained. The growth of MS1 and FP2 strains in Thp-1 © 2014 APMIS. Published by John Wiley & Sons Ltd
macrophage-like cells and A. Castellanii amoebae was compared to that of the wild type strain. As shown in Fig. 6A and 6B, the mutants and wild type strain grew at the same rate in both macrophages and in amoebae demonstrating that both mutants were not affected in their ability of multiplying into host cells. Aurass et al. (33) have recently demonstrated that bdhA was co-transcribed with the downstream patD gene and the bdhA-patD operon was required for intracellular infection, so that it could be considered a novel virulence-associated determinant. To better explain the ability of MS1 to grow in the host cell, we determined whether patD function was still present in the MS1 strain. As expected for an in frame delete mutant, bdhA transcript was deleted of its greater part but the down-stream patD gene was still transcribed (Fig. 3C).
DISCUSSION In this study, we analyzed differences in the proteome of the spontaneous avirulent L. pneumophila sg 6 Vir mutant respect the wild type Vir+ strain. Most of the identified proteins were related to the ketone compound metabolism and/or involved in the degradation of PHB granules. As previously
337
SCATURRO et al.
A
B Vir+
Lp02
C
D Vir–
MS1
E Granules/bacterium
1.5 1 0.5 0
Lp02
SM1
Fig. 4. Staining of PHB granules’ content in Lp02 (A), MS1 (B), Vir+ (C) and Vir (D). Means and standard deviations of content of granules per bacterium were determined in Lp02 and MS1 strains, by counting at least 100 bacteria for each strain in three different image fields (E).
Fig. 5. Peptide alignment of Vir and Vir+ deduced lipase sequence showing the mismatches found as boldface. The aster (*) indicates the non-conservative and colon (:) indicates the conservative changes. In gray, it is shown the putative motive similar to a ‘lipase box’ (GX1SX2G).
observed by Hayashi et al. (34), we found that at the post-exponential growth phase, when Legionella expresses the virulent phenotype and genes like those for motility are active, L. pneumophila serogroup 6 Vir+ strain expressed genes encoding proteins that could be useful in starvation condition. In particular, this study was focused on two genes encoding the 3hydroxybutyrate dehydrogenase enzyme and a putative lipase, which were respectively undetectable and down-produced in the Vir mutant. RT-PCR showed that the Vir- mutant has been impaired in the bdhA and lipase expression with transcript levels 338
significantly lower than found in Vir+. Concerning bdhA, it is present on the bacterial chromosome near some flagellar genes, and because the Vir mutant is defective for the flagellin expression (17) it might be hypothesized a defect in a regulator factor that acts both on the bdhA and flagellin genes, thus connecting genes of motility with genes of metabolism. However, this dysfunction has not contributed to the loss of virulence of Vir, as a matter of fact the deletion of bdhA did not affect the ability of intracellular multiplication. This observation was also useful to clarify the role of bdhA and the down-stream patD © 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
Molplicazione in Thp1: MOI = 1
A
9 8.5 8 Log CFU/mL
7.5 7 6.5
Lp02
6
MS1 FP2
5.5 5
4.5 4 T0
T24
T48
T72
Molplicazione in A. castellanii. MOI = 1
B
10
Log CFU/mL
9 8 7 Lp02 6 MS1
5 FP2 4 T0
T24
T48
T72
Fig. 6. Intracellular multiplication of wild type Lp02 and MS1 and FP 2 Legionella mutants in Thp-1 macrophageslike cells (A) and in A. castellanii amoebae (B) are shown. The data are means and standard deviations CFU determined at 0, 24, 48 and 72 h after infection, from three independent experiments.
gene in the infection process. Indeed, it has recently been demonstrated (33) that in L. pneumophila bdhA is co-transcribed in a unique operon with patD, encoding a protein with phospholipase and lipophospholipase activities. L. pneumophila bdhA/patD mutants showed a severe growth defect in amoebae and U937 cell line. In this study we demonstrated that patD is actively transcribed in the MS1 strain, suggesting that only the patD but not bdhA has most probably a crucial role during the amoeba infection process. It is known for other microorganisms that BdhA catalyses the first steps of PHB granule degradation (28). In this study, the MS1 strain indeed showed a higher accumulation of PHB granules than its parental strain suggesting a role bdhA in the metabolism of this important source of energy. The different PHB accumulation in Vir+ and Vir strains compared to Lp02 and MS1 suggested also differences in the granule metabolism between L. pneumo© 2014 APMIS. Published by John Wiley & Sons Ltd
phila serogroup 6 and L. pneumophila serogroup 1 that would require further investigation, also regarding Legionella pneumophila environmental survival in stress conditions. The gene encoding for a putative lipase is annotated in the Legionella pneumophila subspecies Philadelphia genome as a a/b hydrolase/esterase belonging to the peptidase_S9 superfamily. Here, we demonstrated that the reduced transcriptional activity caused a very low lipase content as found by proteome analysis. The predicted changes in the mRNA secondary structure due to the mutations we observed in the lipase-coding region, might have determined the down-production of the protein. In addition, the deduced amino acid sequence showed 5 different amino acids that could interfere with the right folding and also influence the half-life of this enzyme. Finally, the presence of a motif similar to a catalytic triad typical of de-polymerase enzymes (i.e., GX1SX2G), characterizes its function as depolymerase in the metabolism of PHB granules. As PHB de-polymerases produce 3-hydroxibutyrate, the substrate for BdhA, a possible correlation between undetectable BdhA and down-regulation of lipase should be investigated. In conclusion, this study highlighted changes in an avirulent Legionella pneumophila serogroup 6 mutant in the production of proteins mainly involved both in lipid and carbohydrate metabolism, in turn linked to the synthesis and degradation of ketone compounds. It is known that Legionella utilizes amino acids as carbon source and it is also known that the post-exponential growth phase represents a time to switch virulence genes on and move to the transmissive phase, however, the expression of gene related to the energy metabolism has never been explored before. Therefore, even though the two genes investigated did not show a link with virulence, their expression should be considered in a greater contest, in which a damage of some secondary metabolic pathways could affect the Legionella pneumophila environmental fitness. The authors thank Dr Tamara O’Connor for critical reading of the manuscript and important advice about the experiments. They are also grateful to Dr Ralph Isberg for providing Lp02 strain and pSR47s vector, and Dr Stanimir Ivanov for useful suggestions about mutant constructions. They wish also to thank Massimo Mentasti for English editing.
REFERENCES 1. Fields BS, Benson RF, Besser RE. Legionella and Legionnaires’ disease: 25 years of investigation. Clin Microbiol Rev 2002;15:506–26.
339
SCATURRO et al.
2. Kagan JC, Roy CR. Legionella phagosomes intercept vesicular traffic from endoplasmic reticulum exit sites. Nat Cell Bio 2002;4:945–54. 3. Isberg RR, O’Connor TJ, Heidtman M. The Legionella pneumophila replication vacuole: making a cosy niche inside host cells. Nat Rev Microbiol 2009;7: 13–24. 4. Ninio S, Roy CR. Effector proteins translocated by Legionella pneumophila: strength in numbers. Trends Microbiol 2007;15:372–80. 5. Richards AM, Von Dwingelo JE, Price CT, Abu Kwaik Y. Cellular microbiology and molecular ecology of Legionella-amoeba interaction. Virulence 2013;15:307–14. 6. Jules M, Buchrieser C. Legionella pneumophila adaptation to intracellular life and the host response: clues from genomics and transcriptomics. FEBS Lett 2007;581:2829–38. 7. Br€ uggemann H, Cazalet C, Buchrieser C. Adaptation of Legionella pneumophila to the host environment: role of protein secretion, effectors and eukaryotic-like proteins. Curr Opin Microbiol 2006;9:86–94. 8. Abu Kwaik Y, Harb OS. Phenotypic modulation by intracellular bacterial pathogens. Electrophoresis 1999;20:2248–58. 9. Bandyopadhyay P, Xiao H, Coleman HA, Price-Whelan A, Steinman HM. Icm/dot-independent entry of Legionella pneumophila into amoeba and macrophage hosts. Infect Immun 2004;72:4541–51. 10. De Buck E, H€ oper D, Lammertyn E, Hecker M, Anne J. Differential 2-D protein gel electrophoresis analysis of Legionella pneumophila wild type and Tat secretion mutants. Int J Med Microbiol 2008;298: 449–61. 11. Chien M, Morozova I, Shi S, Sheng H, Chen J, Gomez SM, et al. The genomic sequence of the accidental pathogen Legionella pneumophila. Science 2004;305:1966–8. 12. Lebeau I, Lammertyn E, De Buck E, Maes L, Geukens N, Van Mellaert L, et al. First proteomic analysis of Legionella pneumophila based on its developing genome sequence. Res Microbiol 2005;156:119–29. 13. Galka F, Wai SN, Kusch H, Engelmann S, Hecker M, Schmeck B, et al. Proteomic characterization of the whole secretome of Legionella pneumophila and functional analysis of outer membrane vesicles. Infect Immun 2008;76:1825–36. 14. D’Auria G, Jimenez-Hern andez N, Peris-Bondia F, Moya A, Latorre A. Legionella pneumophila pangenome reveals strain-specific virulence factors. BMC Genom 2010;11:181. 15. Pastoris MC, Proietti E, Mauri C, Chiani P, Cassone A. Suckling CD1 mice as an animal model for studies of Legionella pneumophila virulence. J Med Microbiol 1997;46:647–55. 16. Ricci ML, Torosantucci A, Scaturro M, Chiani P, Baldassarri L, Pastoris MC. Induction of protective immunity by Legionella pneumophila flagellum in an A/J mouse model. Vaccine 2005;23:4811–20. 17. Scaturro M, Meschini S, Arancia G, Fontana S, Ricci ML. Characterization of a spontaneous avirulent mutant of Legionella pneumophila serogroup 6.Evidence of DotA and flagellin involvement in the loss of virulence. J Microbiol 2009;47:768–73.
340
18. Roy CR, Berger KH, Isberg RR. Legionella pneumophila DotA protein is required for early phagosome trafficking decisions that occur within minutes of bacterial uptake. Mol Microbiol 1998;28:663–74. 19. Andrews HL, Vogel JP, Isberg RR. Identification of linked Legionella pneumophila genes essential for intracellular growth and evasion of the endocytic pathway. Infect Immun 1998;66:950–8. 20. Sanchez JC, Rouge V, Pisteur M, Ravier F, Tonella L, Moosmayer M, et al. Improved and simplified ingel sample application using reswelling of dry immobilized pH gradients. Electrophoresis 1997;18:324–7. 21. Quaranta S, Giuffrida MG, Cavaletto M, Giunta C, Godovac-Zimmermann J, Ca~ nas B, et al. Human proteome enhancement: high-recovery method and improved two-dimensional map of colostral fat globule membrane proteins. Electrophoresis 2001;22: 1810–8. 22. Neuhoff V, Arold N, Taube D, Ehrhardt W. Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 1988;9: 255–62. 23. Hellmann U, Wernstedt C, Gonez J, Heldin CH. Improvement of an in-gel digestion procedure for the micropreparation of internal protein-fragments for aminoacid sequencing. Anal Biochem 1995;224:451–5. 24. Pappin DJ, Hojrup P, Bleasby AJ. Rapid identification of proteins by peptide-mass fingerprinting. Curr Biol 1993;3:327–32. 25. VanRheenen SM, Dumenil G, Isberg RR. IcmF and DotU are required for optimal effector translocation and trafficking of the Legionella pneumophila vacuole. Infect Immun 2004;72:5972–82. 26. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;25:402–8. 27. Belyi Y, Tabakova I, Stahl M, Aktories K. Lgt: a family of cytotoxic glucosyltransferases produced by Legionella pneumophila. J Bacteriol 2008;190:3026–35. 28. Aneja P, Charles TC. Poly-3-hydroxybutyrate degradation in Rhizobium (Sinorhizobium) meliloti: isolation and characterization of a gene encoding 3hydroxybutyrate dehydrogenase. J Bacteriol 1999; 181:849–57. 29. James BW, Mauchline WS, Dennis PJ, Keevil CW, Wait R. Poly-3-hydroxybutyrate in Legionella pneumophila, an energy source for survival in low-nutrient environments. Appl Environ Microbiol 1999;65:822–7. 30. Verlinden RA, Hill DJ, Kenward MA, Williams CD, Radecka I. Bacterial synthesis of biodegradable polyhydroxyalkanoates. J Appl Microbiol 2007;102: 1437–49. 31. Gardu~ no RA, Gardu~ no E, Hiltz M, Hoffman PS. Intracellular growth of Legionella pneumophila gives rise to a differentiated form dissimilar to stationaryphase forms. Infect Immun 2002;70:6273–83. 32. Lang C, Flieger A. Characterisation of Legionella pneumophila phospholipases and their impact on host cells. Eur J Cell Biol 2011;90:903–12. 33. Aurass P, Pless B, Rydzewski K, Holland G, Bannert N, Flieger A. A novel large-scale approach for identi-
© 2014 APMIS. Published by John Wiley & Sons Ltd
CHARACTERIZATION OF BDHA AND A PUTATIVE LIPASE
fication of Legionella pneumophila mutants defective for amoeba infection uncovers the bdhA/patD operon as a virulence determinant. Appl Environ Microbiol 2009;75:4506–15. 34. Hayashi T, Nakamichi M, Naitou H, Ohashi N, Imai Y, Miyake M. Proteomic analysis of growth phasedependent expression of Legionella pneumophila pro-
© 2014 APMIS. Published by John Wiley & Sons Ltd
teins which involves regulation of bacterial virulence traits. PLoS ONE 2010;22:11718. 35. Charles TC, Cai GQ, Aneja P. Megaplasmid and chromosomal loci for the PHB degradation pathway in Rhizobium (Sinorhizobium) meliloti. Genetics 1997;146:1211–20.
341
