If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
School of Chemistry, Sun Yat-sen University, Guangzhou, ChinaSchool of Life Sciences, Sun Yat-sen University, Guangzhou, ChinaRun Ze Laboratory for Gastrointestinal Microbiome Study, Sun Yat-sen University, Guangzhou, China
ClpP is the major determinant of biphasic life cycle–dependent protein turnover.
ClpP-dependent proteolysis monitors SpoT abundance for cellular differentiation.
ClpP-dependent regulation of life cycle and bacterial virulence is independent.
ClpP-dependent proteolysis of T4BSS and effector proteins is vital for virulence.
Legionella pneumophila, an environmental bacterium that parasitizes protozoa, causes Legionnaires’ disease in humans that is characterized by severe pneumonia. This bacterium adopts a distinct biphasic life cycle consisting of a nonvirulent replicative phase and a virulent transmissive phase in response to different environmental conditions. Hence, the timely and fine-tuned expression of growth and virulence factors in a life cycle–dependent manner is crucial for survival and replication. Here, we report that the completion of the biphasic life cycle and bacterial pathogenesis is greatly dependent on the protein homeostasis regulated by caseinolytic protease P (ClpP)-dependent proteolysis. We characterized the ClpP-dependent dynamic profiles of the regulatory and substrate proteins during the biphasic life cycle of L. pneumophila using proteomic approaches and discovered that ClpP-dependent proteolysis specifically and conditionally degraded the substrate proteins, thereby directly playing a regulatory role or indirectly controlling cellular events via the regulatory proteins. We further observed that ClpP-dependent proteolysis is required to monitor the abundance of fatty acid biosynthesis–related protein Lpg0102/Lpg0361/Lpg0362 and SpoT for the normal regulation of L. pneumophila differentiation. We also found that the control of the biphasic life cycle and bacterial virulence is independent. Furthermore, the ClpP-dependent proteolysis of Dot/Icm (defect in organelle trafficking/intracellular multiplication) type IVB secretion system and effector proteins at a specific phase of the life cycle is essential for bacterial pathogenesis. Therefore, our findings provide novel insights on ClpP-dependent proteolysis, which spans a broad physiological spectrum involving key metabolic pathways that regulate the transition of the biphasic life cycle and bacterial virulence of L. pneumophila, facilitating adaptation to aquatic and intracellular niches.
). In broth cultures and within host cells, the biphasic life cycle alternates between two distinct and reciprocal forms, replicative and transmissive. This process is called microbial differentiation, in which L. pneumophila undergoes physiological, morphogenetic, and metabolic changes (
). For example, when conditions are favorable for multiplication, virulence traits are neither required nor expressed. However, the bacteria will not replicate in adverse conditions (e.g., nutrient deprivation) (
). Strikingly, analyses of the global gene expression profiles of L. pneumophila revealed that the pathogen’s life cycle is very similar between in vivo infection models and in vitro broth cultures, as evidenced by the profound and comparable changes in gene expression during transition from the replicative phase (RP)/exponential growth phase to the transmissive phase (TP)/postexponential growth phase (
). For example, during the RP, genes related to metabolism, amino acid degradation, sugar assimilation, cell division, and biosynthetic processes are generally upregulated. In contrast, during the TP, genes associated with host entry, virulence, and survival, including those encoding Dot/Icm (defect in organelle trafficking/intracellular multiplication)-translocated effectors, motility machinery (flagellar and type IV pilus genes), enhanced entry proteins, and cyclic-di-GMP regulatory proteins, are upregulated. Hence, the transition between the RP and the TP requires a highly coordinated metabolic pathway (
). The biphasic life cycle is controlled by a multitude of regulatory elements that control gene expression, including regulatory proteins (e.g., CsrA, RpoS, FliA, IHF), two-component systems (TCSs; e.g., LetA/S, PmrA/B, LqsR/S), and stringent response metabolites (e.g., RelA, SpoT, second messenger guanosine tetraphosphate (p)ppGpp) (
). Therefore, the timely and fine-tuned expression of growth and virulence factors and adaptation traits in a life cycle–dependent manner is crucial. Several factors involved in the expression of virulence genes are also major regulators of metabolic pathways (
), indicating that cellular differentiation and key metabolic changes play a significant role in the regulation of L. pneumophila virulence. However, the mechanisms underlying the regulation of these metabolic pathways and the function of multiple regulatory factors during this dimorphic life cycle are incompletely understood.
L. pneumophila virulence is characterized by the translocation of approximately 330 effector proteins into the host cells via the Dot/Icm type IVB secretion system (T4BSS), thereby triggering the recruitment of vesicles derived from the endoplasmic reticulum or the direct manipulation of various signaling pathways in the host cells (
). Notably, not all effectors are simultaneously translocated into the host cytosol at the onset of infection, suggesting that the temporal control of effector activity is required to effectively manipulate the host cell pathways (
). Furthermore, the temporal control of some effectors in the host cell is consistent with their biological functions. For example, effectors SidM, LepB, and SidD target and antagonize Rab1 to temporally control its activation/deactivation on the Legionella-containing vacuole (LCV), whereas SidJ inhibits the ubiquitin ligase activity of the SidE family effectors (
). However, the mechanisms underlying the regulation of life stage–specific effectors are still unknown.
Bacteria use regulated proteolysis for the degradation or activation of regulatory proteins to temporally control specific physiological processes and to mediate signaling pathways, such as stress response, growth, division, cell cycle, development with cell differentiation, pathogenesis, and protein secretion (
). We previously reported that the caseinolytic protease P (ClpP), the catalytic core of the Clp proteolytic complex that is conserved in most bacterial species, plays an integral role in the expression of transmission traits and regulation of life cycle transition and virulence of L. pneumophila (
). In our previous studies, we successfully constructed a strain with a clpP deletion (ΔclpP), using a nonpolar strategy without introducing any antibiotic resistance gene, in which the clpP gene alone was successfully knocked out, as validated by whole genome resequencing (
). Notably, our findings reveal that compared with the WT strain, the deletion of clpP delays the transition of L. pneumophila from the TP to the RP in liquid culture, impairs the survival and proliferation ability in host cells, and reduces the abundance of multiple effector proteins (
). Although ClpP-dependent proteolysis was proven to be critical for L. pneumophila, the function of ClpP in regulating intracellular protein homeostasis and the role of regulatory proteins in life stage–specific expression are still unelucidated.
In this study, we aimed to investigate the dynamic profiles of global protein abundances during phase transition and to determine the function of ClpP in regulating the biphasic life cycle and pathogenesis of L. pneumophila using proteomic approaches. Such an integrative analysis has revealed the potential networks of interconnected proteins with substantial involvement in the life cycle transition of L. pneumophila. We further validated the requirement of ClpP-dependent proteolysis in regulating the abundance of the proteins for the regulation of L. pneumophila differentiation. Interestingly, we indicated that ClpP-dependent regulation of biphasic life cycle and bacterial virulence is independent. This study advances our understanding of Legionella in response to different conditions for replication and survival and provides additional evidence that the completion of the biphasic life cycle and bacterial pathogenesis is greatly dependent on protein homeostasis mediated by ClpP-dependent proteolysis.
Experimental Design and Statistical Rationale
To investigate the role of ClpP-mediated proteolysis in the regulation of life stage–specific proteins of L. pneumophila, the bacteria in the RP/TP of WT and ΔclpP were collected and analyzed by mass spectrometry (MS). To screen the substrates of ClpP during the whole life cycle, bacterial whole-cell lysates from clpPwt and clpPtrap in the TP were prepared and His-tagged proteins were purified with nickel–nitrilotriacetic acid affinity column (GE Healthcare) following the manufacturer’s instructions. Substrates captured inside the proteolytic barrel were copurified along with the His-tagged ClpP complex and identified by MS to identify substrates of ClpP in the WT background. Each sample was analyzed in biological triplicates to allow for statistical tests and to improve consistency. Database searching of all LC–MS/MS raw files were analyzed with the Q Exactive HF-Orbitrap MS (Thermo Fisher Scientific, Co). Proteome Discoverer, version 2.2 software (Thermo Fisher Scientific, Co) was used for quality control and statistical processing. Only proteins identified in each of three biological replicates were quantified. Protein identification and quality criteria were very strict throughout the study with the label-free quantification. Protein abundances greater than 55 are considered significant (
). Student's t test (p < 0.05) was performed on the normalized protein intensities and defined as significant difference in protein abundance of two groups. Ratios of quantity of significantly different proteins were log2 transformed, and only those were approved who exceeded 1 or fell below −1 (
Bacterial Strains, Sample collection, Primers and Media
All L. pneumophila strains were cultured on buffered charcoal yeast extract (BCYE) plates, or in N-(2-acetamido)-2-aminoethanesulfonic acid–buffered yeast extract (AYE) medium, supplemented with thymidine (100 μg/ml) (
) when required. E. coli DH5α used as host strains for cloning strategies was grown in LB and agar at 37 °C. For liquid culture, AYE broth was inoculated with TP bacteria grown in the previous cycle to a final absorbance of 0.2 at 600 nm and incubated at 37 °C with vigorous shaking. RP bacteria were harvested at an absorbance of 0.7 to 1.0 at 600 nm, and TP bacteria were harvested approximately 6 h after the cessation of growth, which is at an approximate absorbance of 3.0 to 3.5 at 600 nm, according to the one previously reported (
). To ascertain colony-forming units (CFU), serial dilutions of bacteria were incubated on BCYE for 4 days and resultant colonies were counted. The bacterial strains, plasmids, and primers used in this work are listed in supplemental Tables S18 and in S19 in the supplementary data, respectively.
Proteomic Analysis (LC–MS)
Proteomic analysis was performed as we previously described (
). In brief, for each sample, 100 μg of protein was reduced with 10 mM DTT at 37 °C for 45 min, and iodoacetamide was then added to a final concentration of 15 mM, with incubation at room temperature for 1 h in the dark. The samples were then diluted with 100 mM ammonium bicarbonate buffer and digested with trypsin (1:50, trypsin/lysate ratio) for 16 h at 37 °C. Digests were centrifuged through 3 kDa filter tubes so that only digested peptides can go through. The concentrations of peptides were determined with a modified Lowry Protein Assay Kit (Sangon Biotech, Co). About 20 μg of peptides were desalted on Pierce C18 Spin Columns (Thermo Fisher Scientific, Co) according to the manufacturer's instructions.
Peptides were analyzed with the Q Exactive HF-Orbitrap MS. For each sample, the same amounts of peptides from total protein were separated on the analytical column with a 70 min linear gradient at a flow rate of 400 nl/min (0–3% B in 3 min; 3–8% B in 4 min; 8–32% B in 44 min; 32–99% B in 5 min; 99% B for 4 min; and 3% B for 10 min). The spectra were acquired in the positive ionization mode by data-dependent methods consisting of a full MS scan in high mass accuracy Fourier transform–MS mode at 60,000 resolution, with the precursor ion scan recorded over the m/z range of 350 to 1500. Database searching of all LC–MS/MS raw files were done using Proteome Discoverer (version 2.2) against the UniProt L. pneumophila database (L. pneumophila subsp. pneumophila strain Philadelphia 1/American Type Culture Collection 33152/DSM 7513 proteome, last modified: October 26, 2018; 2930 proteins). For protein identification, the following options were used. Trypsin was specified as enzyme, cleaving after all lysine and arginine residues and allowing up to two missed cleavages. Carbamidomethylation of cysteine was specified as fixed modification. Protein N-terminal acetylation and oxidation of methionine were considered variable modifications. The peptide mass tolerance for precursor ions was set to 10 ppm, and the mass tolerance for fragment ions was set to 0.02 Da. “Maximum precursor mass” was set to 5000 Da, and “minimum precursor mass” was set to 350, and everything else was set to the default values, including the false discovery rate (FDR) limit of 5% on both the peptide and protein levels. The target with FDR <0.01 was defined as high confidence, and the target with 0.01 ≤ FDR <0.05 was defined as medium confidence. The threshold was based on a default FDR calculator using target/decoy strategy. Quantification abundances are normalized to the same total peptide amount per channel and scaled, so that the average abundance per protein and peptide is 100. Calculates quantification ratios for peptide-spectrum matches, peptides, and proteins based on precursor quantification. Details on identified proteins and peptides are provided in supplemental Tables S1 and S2, respectively.
In Vivo Trapping of ClpP Substrate
Proteomic analysis was performed as we previously described (
). The ClpP trapping system was constructed according to the previous report with minor modifications. Briefly, to generate the ClpPtrap, the active site (serine 110) of ClpP was replaced with an alanine (S110A). The plasmids expressing His-tagged ClpPwt and ClpPtrap were transformed into ΔclpP, respectively, to create ΔclpP/pclpPwt and ΔclpP/pclpPtrap. The ΔclpP/pclpPwt and ΔclpP/pclpPtrap strains in TP were grown in 100 ml of AYE at 37 °C to an absorbance of 0.2 at 600 nm. To screen accumulated substrates of ClpPtrap during the whole life cycle, bacterial whole-cell lysates from ΔclpP/pclpPwt and ΔclpP/pclpPtrap in the RP and TP were prepared, and His-tagged proteins were purified with nickel–nitrilotriacetic acid affinity column (GE Healthcare) following the manufacturer’s instructions. Substrates captured inside the proteolytic barrel were copurified along with the His-tagged ClpP complex and identified by MS to identify substrates of ClpP in the WT background. Details on identified proteins and peptides are provided in supplemental Tables S3 and S4, respectively.
Protein identification and label-free quantitation were performed with Proteome Discoverer, version 2.2 software using default setting. In the study, only proteins identified in each of three biological replicates were considered. Protein abundances greater than 55 are considered significant (
), and a Student’s t test p < 0.05 was applied to identify proteins of two groups for which ratios of quantity of significantly different proteins were log2 transformed and only those were approved who exceeded 1 or fell below −1 (
). Functions of differential abundant proteins were queried from LegioList (http://genolist.pasteur.fr/LegioList/) and UniProt (https://www.uniprot.org/). Functional category annotation and gene essentiality were fetched from these websites. Voronoi treemaps have been proven as a powerful tool for the visualization of large proteomic data and functional relatedness of proteins in other research (
). Thus, to visualize our complex dataset, we functionally mapped the quantified L. pneumophila proteome based on their annotation and prediction.
Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed using the OmicShare tools, a free online platform for data analysis (http://www.omicshare.com/tools), which significantly enriched pathways were retrieved by searching against KEGG databases. Protein–protein interaction network construction was performed using STRING web service (https://string-db.org/) (a database of known and predicted protein–protein interactions) with the differentially expressed proteins revealed by proteomic data used as input. The integrated network analysis results based on a required minimum interaction score of at least 0.7 by STRING database were output by Cytoscape software (https://cytoscape.org/), version 3.7.2. Random forest analysis to investigate the contribution of the putative proteins to the biphasic life cycle transition (
). Briefly, the sequences of lpg0361, lpg0362, lpg0102, and spoT gene were amplified by PCR using the primer pairs Plpg0361-F/Plpg0361-R, Plpg0362-F/Plpg0362-R, Plpg0102-F/Plpg0102-R, and PspoT-F/PspoT-R, respectively. The PCR products were digested with BamHI and XhoI and subcloned into the ectopic plasmid pJB908, creating plasmid pJB908-lpg0361, pJB908-lpg0362, pJB908-lpg0102, and pJB908-spoT. A hexahistidine tag was added to the C terminus of the protein in all the resulting recombinant plasmids pJB908-lpg0361, pJB908-lpg0362, pJB908-lpg0102, and pJB908-spoT during cloning. The resultant plasmid pJB908-lpg0361, pJB908-lpg0362, pJB908-lpg0102, and pJB908-spoT was electroporated into WT and ΔclpP to create ectopic strain WT/plpg0102, WT/plpg0361, WT/plpg0362, WT/plpgspoT, ΔclpP/plpg0102, ΔclpP/plpg0361, ΔclpP/plpg0362, and ΔclpP/pspoT, respectively.
Growth Curve Assay
Growth curve assay was performed as we previously described (
). Fresh L. pneumophila cells were inoculated into 5 ml of AYE(T) medium and were cultured to the TP at 37 °C. Then the cultures were transferred into 50 ml of AYE in flasks, incubated to the TP, then diluted into new flasks to similar absorbances at approximate absorbance of 0.2 at 600 nm at time 0. Cultures were grown at 37 °C with shaking. To measure the growth curve, 1 ml of the cells was sampled every 3 h for measurement of absorbance at 600 nm. To ensure conformity, multiple replicates on different days were examined. All experiments were performed in biological triplicate.
RNA Isolation, Complementary DNA Preparation, and Quantitative RT–PCR
Quantitative RT–PCR (qRT–PCR) was performed as we previously described (
). RNA for real-time qRT–PCR was prepared using an Eastep Super Kit following manufacturer’s protocols (Promega Co) and treated with DNase I according to the manufacturer’s instructions (Promega Co) prior to complementary DNA preparation. Complementary DNA was prepared using GoScript Reverse Transcription System as described by the manufacturer (Promega Co). qRT–PCR was performed in a 20 μl reaction volume using an Applied Biosystems StepOne Plus 96-well RT–PCR system with Power SYBR Green PCR Master Mix following the manufacturer’s instructions (Applied Biosystems Co). 16S rRNA was used as the reference sample in all comparative threshold cycle (ΔΔCT) experiments. All qRT–PCR primers were tested for amplification efficiency. qRT–PCR data were analyzed using StepOne System software (Applied Biosystems, Inc) and GraphPad Prism (GraphPad Software, Inc). Primers used in qRT–PCR experiments are shown in supplemental Table S19. All analyses were performed in biological triplicate.
Bacterial Infectivity in A. castellanii
Bacterial infectivity and intracellular growth were performed as we previously described (
). The A. castellanii cells were seeded onto a 24-well plate (5 × 105 per well) allowing to adhere for 2 h prior infection. L. pneumophila cells were grown for 20 h in AYE broth at 37 °C with shaking, diluted in HL5, and were used to infect amoebae at a multiplicity of infection of 10. About 30 min postinfection, extracellular bacteria were removed by washing three times with warm HL5 medium (
). At each time points indicated, culture supernatant was removed, and the amoebae cells were lysed with 0.04% Triton. The supernatant and lysates were combined, and serial dilutions were prepared and aliquots were plated on BCYE plates for CFU counting (
). Total cell extracts of L. pneumophila were prepared at various time points after growth at 37 °C. Briefly, bacterial cell pellets were resuspended in 1 ml of lysis buffer and sonicated for 2 min. The cells were then centrifuged for 1 h at 120,00g. The protein-containing supernatant was removed, and the protein concentration was measured using a commercial kit (Bio-Rad Co). Samples were normalized for protein loading and run on a 10% SDS-PAGE as described previously (
). The levels of Lpg0102, Lpg0361, Lpg0362, or SpoT were immunoblotted with anti-His tag antibody. Isocitrate dehydrogenase was probed as a loading control.
Growth Phase–Dependent Protein Expression Profiles of L. pneumophila
To investigate the significance of ClpP-dependent proteolysis in regulating the biphasic life cycle and virulence of L. pneumophila (Fig. 1A), we performed a proteomic analysis of whole lysates obtained from cultures of L. pneumophila WT at the RP and TP to acquire a comprehensive proteomic profile of the two phases and to identify the proteins that are differentially expressed per life stage (Fig. 1B). A total of 1927 differentially expressed proteins between the RP and TP were identified and quantified (Table 1). These proteins represent 65.48% of the annotated proteins, consisting of 2943 open reading frames, in L. pneumophila Philadelphia-1 proteome (
). The expression levels of 428 proteins (22.21%) showed significant differences between the RP and TP (supplemental Tables S5 and S6), including 220 and 208 proteins that were upregulated in the RP (RP-on and RP-up) and TP (TP-on and TP-up), respectively, which were above or below the threshold of +1 and −1 log2(RP/TP) (supplemental Fig. S1A). Both groups contained a number of on/off proteins solely identified either in the RP or in the TP, including 23 that were RP-on (synonymously TP-off) and 11 that were TP-on (synonymously RP-off).
Table 1Overview of identified protein of L. pneumophila RP and TP in the WT and ΔclpP strain, respectively
The KEGG pathway enrichment analysis was subsequently performed to functionally annotate differentially expressed proteins between the RP and TP. We discovered that during the RP, the proteins associated with the pathways for ribosome, amino sugar, nucleotide sugar metabolism, and biotin metabolism were the most significantly enriched, followed by those related to the pathways for homologous recombination, base excision repair, fatty acid biosynthesis, and mismatch repair (supplemental Fig. S1B). The RP-specific proteins were found to play vegetative functions for growth and reproduction (e.g., RplS, RpsH, RpsP, RpmA, RpsT) and cell division (e.g., FtsA, FtsQ, FtsZ) (supplemental Fig. S1C). Overall, the RP-specific pathways are mainly responsible for bacterial replication and growth. In contrast, TP-specific proteins were related to multiple microbial metabolic pathways, such as propanoate metabolism and synthesis and degradation of ketone bodies. Poly-3-hydroxybutyrate, the main carbon and energy storage for L. pneumophila that is utilized under nutrient starvation conditions (
), we also confirmed that majority of the flagellar assembly proteins (e.g., MotA, FlgH, FlgI, FlgA, FlhF, Lpg0907, FliA, FliS, PilM), signal transduction proteins (e.g., LetE, YhbH, PilR, YegE, Lpg0829, RpoS, LqsR), and oxidoreductase-related proteins (e.g., MmsA, PntB, MaeA) were upregulated in the TP. Collectively, the TP specific primarily facilitates the expression of transmissive traits.
These results provide more detail data for understanding the mechanism of how L. pneumophila adopted a reciprocal biphasic life cycle and revealing an unexpected complex picture of life cycle–dependent regulation profiles of L. pneumophila.
ClpP-Dependent Proteolysis Plays an Important Role in Maintaining the Biphasic Life Cycle
To adapt to the biphasic life cycle, L. pneumophila employs a bipartite metabolism that requires fine-tuned regulation (
). Since our previous study revealed that ClpP is necessary for the transition from the TP to RP (Fig. 1A), we then used the ΔclpP strain to investigate the differences in ClpP-dependent regulatory proteins at the RP and TP on a large scale (Fig. 1B). To avoid the unknown affection of extra plasmid (and the antibiotic resistance gene) on the physiology of Legionella and facilitate the carried out experiments, the whole genomes of WT and ΔclpP have been resequenced and compared. It showed that only the clpP gene alone was successfully knocked out in ΔclpP strain (supplemental Fig. S2). Compared with the WT, a total of 629 proteins showed significantly different levels in the ΔclpP strain, corresponding to 32.64% of the identified proteins (i.e., 21.37% of the annotated proteins in the L. pneumophila Philadelphia-1 proteome) (Table 1). These included 453 proteins with different abundances at the RP (255 were upregulated and 198 were downregulated) and 280 proteins with different levels at the TP (152 were upregulated and 128 were downregulated) (supplemental Tables S7–S10). Furthermore, Voronoi treemaps were used to comprehensively characterize the proteomic data at the RP and TP between the ΔclpP strain and WT as well as determine the role of ClpP during the transition between phases (
Based on the Voronoi treemaps, the downregulated proteins in the RP of the ΔclpP strain were found to be related to the functional categories for “DNA replication, recombination and repair,” “Transport, uptake,” “Dot/Icm apparatus,” “Cell envelope, cell division,” and “Glycan biosynthesis and metabolism,” whereas the upregulated proteins in the RP were related to the functional categories for “Carbohydrate-metabolism and energy,” “Stress response, defense, xenobiotics,” “Amino acid metabolism, other amino acids,” “Flagellar assembly, motility,” “Regulation, TCS, signal transduction,” and “Nucleotide metabolism” (Figs. 2, S3 and supplemental Table S11). Interestingly, the upregulated proteins in the RP of the ΔclpP strain were greatly enriched in the pathways for TCSs, microbial metabolism, and flagellar assembly, which were previously observed in the TP of the WT (supplemental Fig. S4A). In addition, the downregulated proteins in the RP of the ΔclpP strain were greatly enriched in the pathways for biotin metabolism, fatty acid metabolism, ABC transporters, and ribosomes, which were previously observed in the RP of the WT (supplemental Fig. S4B). The discrepancies in the metabolic pathways observed during the RP in the ΔclpP strain suggest that ClpP-dependent proteolysis is essential for the normal function of RP-specific proteins. Moreover, the interaction network analysis showed that the upregulated proteins in the RP of ΔclpP strain were associated with flagellar assembly (e.g., FliS, Lpg0907, FlgD, FlgE, FliA, FlgI, MotA, FliC, FlhF, Lpg1783, PilM), signal transduction (e.g., YegE, Rre41, Lpg0156, Lpf1490, Lpg0829, Lpg2180, LqsR, RpoS, Lpg0477, PilR, FixL), transcriptional regulators (e.g., FleQ, FleR, MraZ, SkgA, IHFA, IHFB), ATP-binding protease components (e.g., ClpA, GrpE, DnaJ, ClpB, ClpS), and microbial metabolism in diverse environments (e.g., Lpg2664, Lpg0245, AcsB, AtoB) (supplemental Fig. S4C). In addition, the downregulated proteins in the RP of the ΔclpP strain were found to interact with ribosomal proteins (e.g., RplS, RpsH, RpmC, RpsP, RpmA), cell division proteins (e.g., Lpg0934, PilF, Lpg0394, FtsB, FtsQ), lipid metabolism proteins (e.g., FabZ, Lpg0102, Lpg0361, Lpg0362), and outer membrane proteins (e.g., OstA, SurA, BamB, Lpg0840, LpxB) (supplemental Fig. S4D). These results suggest that the fine-tuned regulation of these proteins is important during the RP.
Moreover, in the TP of the ΔclpP strain, the downregulated proteins were primarily under the functional categories for “Amino acid metabolism, other amino acids,” “Stress response, defense, xenobiotics,” and “Toxin production, other pathogen functions,” whereas the upregulated proteins were under the functional categories for “Flagellar assembly, motility,” “Cofactors and vitamins, secondary metabolite,” “Fatty acid/lipid metabolism,” “DNA replication, recombination and repair,” “Cell envelope, cell division,” and “Transcription, RNA stability, translation” (Figs. 2, S3 and supplemental Table S12). The KEGG pathway analysis revealed that the upregulated proteins in the TP of the ΔclpP strain were highly enriched in the pathways for amino sugar and nucleotide sugar metabolism, homologous recombination, biotin metabolism, fatty acid metabolism, and base excision repair, which were previously observed for the RP-specific proteins in the WT (supplemental Fig. S5A). Furthermore, the downregulated proteins in the TP of the ΔclpP strain were greatly enriched in the pathways for microbial metabolism, and TCSs, which were previously observed for the TP-specific proteins in the WT (supplemental Fig. S5B). These results suggest that the functional loss of ClpP could cause a disorder in the metabolic pathways involved by the TP-specific proteins, confirming that ClpP-dependent proteolysis is also essential during the TP. In addition, the interaction network analysis demonstrated that the upregulated proteins in the TP of the ΔclpP strain were associated with ribosomal proteins (e.g., RpsL, RpmB, RpsU), cell division proteins (e.g., ParB, Lpg1446, FtsA, FtsZ), cell envelope synthesis (e.g., Lpg0840, Lpg0768, Lpg0748, YvfE), flagellar assembly (e.g., Lpg0906, FlgI, FlgD, FlhF, FleQ), and lipid metabolism (e.g., LipA, Lpg0102, Lpg0361, Lpg0362) (supplemental Fig. S5C), whereas the downregulated proteins in the TP of the ΔclpP strain were related to signal transduction (e.g., sensory box GGDEF family proteins LssE, Lpg2642), pathogen function (e.g., IcmL, EnhB, EnhA, LemA), and crosslinking of the peptidoglycan cell wall (e.g., MrdA, FtsA) (supplemental Fig. S5D). In the proteome data, we found that Lpg0279, a TCS regulator expressed at the TP and promotes both L. pneumophila cell differentiation and survival (
), was decreased during the TP in the absence of clpP, compared with WT. Moreover, the low expression of protein NrdA or LipA was required during the TP (no significant difference in the RP) in the WT, but a high expression level was detected in the TP of the ΔclpP strain. Also, the accumulation of NrdA or LipA during the TP caused a prolonged lag phase from the TP to the RP of L. pneumophila (supplemental Fig. S6). These results suggest that the temporal regulation of these proteins may be vital during the TP.
Because of the complexity of the ClpP-mediated network, specific regulatory proteins and predictable effects of their respective changes were also observed in the proteome. As shown in supplemental Fig. S7, compared with that CsrA was highly accumulated in the ΔclpP mutant than in WT during the TP, the expression of LetE and CegC4 was significantly decreased in the ΔclpP strain than in the WT, whereas the expression of Lem27 was significantly increased in the ΔclpP strain than in the WT (supplemental Fig. S7, A–D). The ΔclpP/pcsrA strain was used to further investigate whether the effect of these proteins was related to CsrA. Compared with that in the ΔclpP/pcsrA and ΔclpP/pJB908, the expression of csrA was significantly increased, and correspondingly, the expression of letE and cegC4 was indeed downregulated and the expression of lem27 observed was indeed upregulated (supplemental Fig. S7, E–H). These data were in consistent with the study by Sahr et al. (
Taken together, comparison of the bipartite metabolic proteins between the WT and ΔclpP strain using Voronoi treemaps, KEGG pathway enrichment, and interaction network analysis revealed that the regulation of the regulatory proteins via ClpP-dependent proteolysis is crucial in maintaining the biphasic life cycle of L. pneumophila.
ClpP Controls the Biphasic Life Cycle–Dependent Protein Turnover Between the RP and the TP
To further explore the function of ClpP-dependent proteolysis in regulating the transition of the biphasic life cycle, we analyzed the abundance of 428 RP- and TP-specific proteins between the ΔclpP strain and WT (supplemental Table S13). A total of 316 life stage–specific proteins (73.83%) were observed that their growth-dependent expression was linked to the regulation of ClpP-dependent proteolysis (Table 1). Notably, the Voronoi treemap demonstrated that most of the proteins were regulated via ClpP, including proteins that were classified under the following functional categories: “Regulation, TCS, signal transduction,” “Dot/Icm apparatus,” “Toxin production, other pathogen functions,” and “Cell envelope, cell division.” The remaining proteins belonged to the following categories: “Virulence effector,” “Carbohydrate metabolism and energy,” “Fatty acid/lipid metabolism,” “Amino acid metabolism, other amino acids,” “Stress response, defense, xenobiotic,” “Protein secretion/trafficking, protein fate,” “Nucleotide metabolism,” “DNA replication, recombination, and repair,” “Cofactors and vitamins, secondary metabolite,” and “Flagellar assembly, motility” (Figs. 3 and S8). These data indicate the significant contribution of ClpP-dependent proteolysis in the timely regulation of protein expression levels at different phases.
Transcriptional control is crucial for the regulation of L. pneumophila differentiation to activate the genes necessary for adapting to new conditions and to repress the genes that are no longer required (
). In this study, we discovered that the abundance of key transcriptional regulators was controlled in a ClpP-dependent manner (supplemental Table S14). The downregulation of several RP-specific regulators in the TP of WT depended on ClpP, including Fis1, Fis2, and Fis3 (regulators that repress the expression of numerous effector-encoding genes) (
). Notably, several TP-specific proteins that were downregulated in the RP of the WT were upregulated in the ΔclpP strain, including Lpg0689 (a Dps-like DNA-binding stress protein previously found abundant in starved E. coli that can be DNase resistant when complexed with DNA) (
EHEC/EAEC O104:H4 strain linked with the 2011 German outbreak of haemolytic uremic syndrome enters into the viable but non-culturable state in response to various stresses and resuscitates upon stress relief.
), were less abundant in the RP of the WT (i.e., more abundant in the TP of the WT), but become more abundant in the RP and less abundant in the TP, in the absence of clpP. These results indicate that the alteration in the abundance of ClpP-dependent transcriptional regulators at different stages is essential for normal function at a specific phase in the life cycle of L. pneumophila.
Like other bacterial pathogens, L. pneumophila employs a variety of distinct TCSs to control post-transcriptional regulation (
). Here, we confirmed that the regulation of several TCS proteins and proteins under the control of TCSs was ClpP dependent (supplemental Table S15). In the absence of clpP, several TP-specific TCSs were observed to be upregulated at the RP, including Lpg0829 (a LapD-like c-di-GMP receptor that can act as a coincidence detector) (
Characterization of the alternative sigma factor sigma54 and the transcriptional regulator FleQ of Legionella pneumophila, which are both involved in the regulation cascade of flagellar gene expression.
)) was downregulated in the TP of the WT but upregulated in the ΔclpP strain. In addition, several TP-specific proteins involved with TCSs that were less abundant in the RP of the WT had a higher level in the ΔclpP strain, such as CydA (cytochrome D ubiquinol oxidase subunit I; plays an important role in cell growth and stress resistance in E. coli (
)) and FliA (an RNA polymerase sigma factor involved in the regulation of flagellum production; also acts as a regulator of virulence genes that are required for the expression of pathways for cytotoxicity, lysosome evasion, and replication in L. pneumophila (
)). In addition, the RNA polymerase sigma factor RpoS (involved in the regulation of multiple pathways associated with motility and pathogenic functions as well as the activity of transcriptional regulators and Dot/Icm effectors) (
) observed at the RP had a lower level in the WT than in the ΔclpP strain. The transmission trait enhancer LetE (a connector protein between the CpxRA TCS and the LetAS-RsmYZ-CsrA regulatory cascade) (
) was less abundant at the RP in the WT (i.e., more abundant at the TP in the WT) but become more abundant at the RP and less abundant at the TP in the ΔclpP strain. Similarly, the RP-specific global regulator CsrA (a pivotal repressor of transmission traits and activator of replication) (
) that was less abundant in the TP of the WT had a higher level in the ΔclpP strain.
To systematically determine the role of ClpP in regulating the life stage–specific proteins, we subsequently analyzed the dynamic changes of phase-specific 316 proteins during the transition from the TP to the RP and vice versa (Fig. 3). During the TP to RP transition, ClpP-dependent proteolysis was confirmed to be essential for downregulating 154 TP-specific proteins and upregulating 68 RP-specific proteins (supplemental Fig. S9, A and B). The TP-specific proteins were observed to be highly enriched in 13 significant pathways (e.g., microbial metabolism, TCSs, flagellar assembly), whereas the RP-specific proteins were mainly enriched in the pathways for biotin metabolism, fatty acid biosynthesis, fatty acid metabolism, base excision repair, and ribosome metabolism (supplemental Fig. S9C). The integrated network analysis showed that the key proteins involved in the TP to RP transition, including Lpg2664, AcsB, MmsA, AtoB, FliA, FliS, FlhF, Frr, Lpg0358, and Lpg2228, presented decreased abundance, whereas Lpg0102, Lpg0361, Lpg0362, FabZ, RpmA, RplS, RpsP, RpsH, Lpg0394, and FtsQ increased in abundance (supplemental Fig. S9D). During the RP to TP transition, ClpP-dependent proteolysis caused the downregulation of 108 RP-specific proteins and the upregulation of 40 TP-specific proteins (supplemental Fig. S10, A and B). Furthermore, the RP-specific proteins were found to be highly enriched in the pathways for amino sugar and nucleotide sugar metabolism, homologous recombination, fructose and mannose metabolism, and biotin metabolism, whereas TP-specific proteins were mainly enriched in the pathways for degradation of aromatic compounds, aminobenzoate degradation, starch and sucrose metabolism, and benzoate degradation (supplemental Fig. S10C). The integrated network analysis also revealed that the key proteins involved in the RP to TP transition, including RecA, LipA, Lpg0361, Lpg0362, Lpg0102, RfbA, Lpg2486, Pgi, Lpg0757, YvfE, Lpg0906, HisF, HisH, NeuB, Lpg0748, Lpg0768, and Lpg0769, presented decreased abundance, whereas Lpg0420 increased in abundance (supplemental Fig. S10D).
Taken together, these results suggest that L. pneumophila employ ClpP-dependent proteolysis to either degrade or activate regulatory proteins and control their expression at a specific life stage, leading to the temporal and spatial regulation of the processes required for morphological development. The reciprocal regulation of the RP and TP mediated by ClpP may aid in the successful adaptation of Legionella to harsh environments.
Fatty Acid Biosynthesis–Related Proteins Play an Important Role During Phase Transition
The strict regulation of differentiation is critical for the biphasic life cycle of L. pneumophila (
). Thus, we also investigated whether the levels of selected life stage–specific proteins were regulated in a ClpP-dependent manner during both phases. Fifty-nine proteins (13.79%) were successfully screened and quantified (Figs. 4, S11 and supplemental Table S16), indicating their significant functions during both the RP and TP. The Voronoi treemaps showed that at either phases, the levels of the identified proteins in the WT were influenced by ClpP. Three proteins, namely Lpg0102, Lpg0361, and Lpg0362, were found to be under the “Cofactors and vitamins, secondary metabolite” and “Fatty acid/lipid metabolism” categories. These proteins also showed the highest confidence (0.900) among the interactions after the STRING web service analysis.
Since the ClpP deletion delayed the TP to RP transition ((
), Fig. 1A), we postulated that the identified differentially expressed proteins may be critical during this process. The KEGG pathway analysis revealed that the 59 life stage–specific proteins were significantly enriched in the pathways for biotin metabolism, fatty acid biosynthesis, and fatty acid metabolism (supplemental Fig. S12A). Notably, Lpg0102 (3-oxoacyl-acyl carrier protein [ACP] synthase) (FabF), Lpg0361 (3-oxoacyl-ACP synthase 2) (FabB), and Lpg0362 (FabB, N-terminal region) were also highly enriched (supplemental Fig. S12B). Random forest analysis (
) further revealed that 30 of the identified proteins potentially perform a significant role in mediating the transition process (supplemental Fig. S12C), including Lpg0361, Lpg0362, and Lpg0102. Collectively, these results suggest that specific fatty acid biosynthesis–related proteins may have essential functions during phase transition.
Lpg0102, Lpg0361, and Lpg0362 Protein Levels are Critical for Phase Transition
To explore the function of fatty acid biosynthesis–related proteins Lpg0102, Lpg0361, and Lpg0362 during phase transition, we analyzed their levels at different growth phases via proteomic analysis of whole lysates obtained from cultures of L. pneumophila WT at indicated time points. The results showed the lower levels of Lpg0102, Lpg0361, and Lpg0362 during the TP, whereas higher levels were detected upon entry into the RP (Fig. 5, A–C), which suggests that the expression of Lpg0102, Lpg0361, and Lpg0362 during the biphasic life cycle of L. pneumophila is growth phase dependent.
To determine whether the regulation of Lpg0102, Lpg0361, and Lpg0362 is associated with ClpP, we also measured their protein levels in the ΔclpP strain. During the TP, the levels of Lpg0102, Lpg0361, and Lpg0362 were significantly upregulated in the ΔclpP strain compared with the WT. In contrast, during the RP, the levels of Lpg0102, Lpg0361, and Lpg0362 were significantly downregulated in the ΔclpP strain compared with the WT. These data indicate that the temporal levels of Lpg0102, Lpg0361, and Lpg0362 are also regulated via the ClpP-dependent proteolytic pathway.
To verify the function of Lpg0102, Lpg0361, and Lpg0362 during the life cycle of L. pneumophila, ectopic expression plasmids (lpg0102, lpg0361, and lpg0362) were constructed and then transformed into the WT and ΔclpP strain to create the WT/plpg0102, WT/plpg0361, WT/plpg0362, ΔclpP/plpg0102, ΔclpP/plpg0361, and ΔclpP/plpg0362 strain, by the method we reported previously (
). Bacterial inoculum from the TP culture was used to measure the impact of Lpg0102, Lpg0361, and Lpg0362 on the whole life cycle. The results of the in vitro growth assay showed that compared with WT/pJB908 and ΔclpP/pJB908, the ectopic expression of Lpg0102, Lpg0361, and Lpg0362 did not affect the growth of the WT but significantly prolonged the lag phase (p < 0.001) and weakened the proliferation of the ΔclpP strain (Fig. 5D), indicating that the effects of Lpg0102, Lpg0361, and Lpg0362 may have been caused by the impaired ClpP-dependent proteolytic pathway.
Western blot analysis revealed that Lpg0102, Lpg0361, and Lpg0362 were more abundant at the RP than the TP in a new round of the life cycles of WT/plpg0102, WT/plpg0361, and WT/plpg0362, demonstrating that their levels were life cycle dependent (Fig. 5E). However, in ΔclpP/plpg0102, ΔclpP/plpg0361, and ΔclpP/plpg0362, the three proteins were more abundant at the TP than the RP, which was contrary to the results of ectopic expression in the WT. Liquid culture observation and quantitative analysis verified that during the prolonged lag phase of ΔclpP/plpg0102, ΔclpP/plpg0361, and ΔclpP/plpg0362, the levels of Lpg0102, Lpg0361, and Lpg0362 in the clpP mutant gradually decreased, as its growth slowly recovered (Fig. 5, D and F), suggesting that the degradation of Lpg0102, Lpg0361, and Lpg0362 is ClpP dependent during the TP.
Overall, these findings demonstrate that the protein levels of Lpg0102, Lpg0361, and Lpg0362 are both life cycle dependent and temporally regulated in a ClpP-dependent manner during phase transition. In addition, the result that the accumulation of Lpg0102, Lpg0361, and Lpg0362 delays the transition from the TP to the RP added new evidence to understand part of the intermediates/proteins in the fatty acid biosynthesis–related pathways in the regulation of L. pneumophila life cycle.
ClpP Regulates SpoT Levels via Fatty Acid Biosynthetic Proteins to Mediate the TP to RP Transition
L. pneumophila is reportedly lacking in SpoT hydrolase activity, which impairs its transition from the TP to RP in either culture media or macrophages (
). Since the ectopic expression of three fatty acid biosynthesis–related proteins significantly prolonged the lag phase of the ΔclpP strain, we aimed to determine whether SpoT expression was also affected by these proteins. Hence, transcriptional analysis of spoT levels in the ΔclpP/pJB908, ΔclpP/plpg0102, ΔclpP/plpg0361, and ΔclpP/plpg0362 strains was performed using qRT–PCR. At each indicated time point, the transcriptional levels of spoT were significantly lower in ΔclpP/plpg0102, ΔclpP/plpg0361, and ΔclpP/plpg0362 than in ΔclpP/pJB908 (Fig. 6A). However, the spoT expression in the three strains gradually increased relative to the growth progression (Fig. 6A), which contradicted the protein levels of Lpg0102, Lpg0361, and Lpg0362 (Fig. 5F). In the WT, spoT expression was upregulated during the TP to RP transition (24 h, 0 h) and the RP to TP transition (15 h, 18 h) (Fig. 6B), suggesting that SpoT is modulated in a life cycle–dependent manner and required for phase transition. These findings were consistent with previous studies reporting that SpoT expression is required throughout the L. pneumophila life cycle to mediate ppGpp turnover via its hydrolase and synthase activities (
). The Lpg0102 (FabF), Lpg0361 (FabB), and Lpg0362 (FabB) were already confirmed to be involved with the ACP metabolite (Fig. 6C). Notably, the ectopic expression of these proteins in the ΔclpP strain reduced the expression level of spoT, suggesting that the prolonged lag phase in the ΔclpP strain may have been due to the lack of SpoT. To confirm this hypothesis, we performed qRT–PCR to detect the spoT mRNA levels during the lag phase in the ΔclpP strain. Compared with the WT, the spoT expression was significantly downregulated in the ΔclpP strain at 0 h and the lag phase (6, 12, and 18 h) (Fig. 6D), suggesting that the temporal upregulation of SpoT expression via the ClpP-mediated pathway is essential for the TP to RP transition.
The spoT ectopic expression strains were also constructed for the WT (i.e., WT/pspoT) and ΔclpP strain (i.e., ΔclpP/pspoT). The in vitro growth assay showed that compared with WT/pJB908 and ΔclpP/pJB908, the ectopic expression of SpoT did not affect the growth of the WT but significantly recovered from the lag phase (>80%) and enhanced the proliferation of the ΔclpP strain (Fig. 6E). Western blot analysis of the SpoT levels at indicated time points revealed that in a new round of WT/pspoT life cycle, SpoT levels were detected during the TP and TP to RP transition but were downregulated upon the shift into RP (Fig. 6F). In ΔclpP/pspoT, SpoT levels were detected during the TP, upregulated during the TP to RP transition (0 to 3 h), and downregulated upon entry into the RP. Furthermore, the ectopic expression of RelA (another trigger of the stringent response that follows fluctuations in amino acid availability (
)) in the ΔclpP strain did not recover the lag phase (data not shown).
Overall, these results demonstrate that the functional loss of ClpP resulted in the accumulation of fatty acid biosynthesis–related proteins, and the downregulation of SpoT expression delayed the transition from the TP to RP. Thus, L. pneumophila requires ClpP-dependent proteolysis to monitor fatty acid biosynthesis–related proteins and SpoT expression for the normal regulation of microbial differentiation.
Regulation of the Biphasic Life Cycle and Bacterial Virulence is Independent
We previously observed that the deletion of clpP not only delays the entry to the RP but also impairs bacterial infectivity to the host and inhibits the proliferation ability in cells (
). To investigate whether the levels of Lpg0102, Lpg0361, Lpg0362, and SpoT influenced bacterial infectivity, L. pneumophila strains in the TP were exposed to amoebae A. castellanii for 2 h. The extracellular bacteria were subsequently cleared, and the amoebae were lysed to release L. pneumophila and calculate CFU of the infectious bacteria. The results showed that the survival capabilities of WT/plpg0102, WT/plpg0361, WT/plpg0362, and WT/pspoT were similar to that of WT after phagocytosis (Fig. 7, A and B). However, the survival capabilities of ΔclpP/plpg0102, ΔclpP/plpg0361, ΔclpP/plpg0362, and ΔclpP/pspoT were significantly lower than that of ΔclpP harboring only the empty vector. Furthermore, the proliferation rates of the WT with lpg0102, lpg0361, lpg0362, and spoT ectopic expression were identical to that of the WT, whereas the proliferation of the ΔclpP strains with and without ectopic expression was consistently inhibited (Fig. 7, C and D). Comparing the survival and proliferation abilities of ΔclpP/plpg0102, ΔclpP/plpg0361, and ΔclpP/plpg0362 with their growth curves (Figs. 5D, 7, A and C), the ectopic expression of SpoT in vitro almost completely recovered the transition of the life cycle and strengthened the proliferation of the ΔclpP strain (Fig. 6E); however, the proliferation ability of ΔclpP/pspoT in vivo did not improve, and its survival capability in vivo was further impaired (Fig. 7, B and D). These results suggest that the regulation of the biphasic life cycle and bacterial virulence is independent.
Thus, our data complemented and improved the existing model (
). Particularly, our model showed that the regulation of SpoT might be more vital than other regulator proteins for the progress of biphasic life cycle, and the regulation of more than 120 ClpP-dependent phase-specific effectors identified in this study provide new insights into the independent regulation of bacterial virulence.
Several T4BSS and Effector Proteins are Regulated in a ClpP-Dependent Manner During the Biphasic Life Cycle
During the RP, L. pneumophila cannot initiate infection to macrophages and avoid fusion with lysosomes. During the TP, the bacterial pathogen can effectively infect the macrophages, departing from the endocytic pathway shortly after internalization to establish a replicative vacuole (
). This indicates that the abundance of effectors at different stages is essential for normal function at a specific period. Notably, we observed that the protein levels of several Dot/Icm effectors at different phases were dependent on ClpP, including 93 RP-specific and 52 TP-specific effectors (Fig. 2 and supplemental Table S17).
Compared with the WT, 44 and 49 effectors were downregulated and upregulated, respectively, in the RP of the ΔclpP strain (Fig. 8, A and B). The downregulated effectors include LnaB (an L. pneumophila activator of NF-κB) (
Compared with the WT, 31 and 21 effectors were downregulated and upregulated, respectively, in the TP of the ΔclpP strain (Fig. 8, C and D). The downregulated effectors include VipD (a phospholipase A that blocks endosome fusion with LCVs) (
), SidC (accumulates during the TP; after anchoring to phosphatidylinositol-4 phosphate on the LCVs, supports efficient intracellular growth and recruitment of endoplasmic reticulum–derived vesicles to the LCV) (
We also observed that several ClpP-dependent effectors were regulated during both the RP and TP (Fig. 8E). In the absence of ClpP, the levels of various effectors varied depending on the phase: nine (Lpg1776, LegC8/Lgt2, Lpg0112, Lem28, LegC6, Lpg0301, Lpg2327, Lpg2888, and LnaB) were downregulated at both the RP and TP; two (MavD and LegA12) were downregulated at the RP but upregulated at the TP; eight (Lpg2207, LpnE, RalF, Lpg0279, LirA, SdeA, SdeC, and RavE) were upregulated at the RP but downregulated at the TP; and six (Lpg0130, Ceg35, Lem27, RavK, LegC7, and MavQ) were upregulated both at the RP and TP (Fig. 8F). Interestingly, several components of the Dot/Icm apparatus, which is critical for L. pneumophila virulence (
), were also regulated via the ClpP-dependent proteolytic pathway. Among these, six Dot/Icm apparatus core complex proteins (DotK, DotH/IcmK, DotU/IcmH, IcmL-like/Lpg0708, DotA, and IcmV) were expressed at the RP, whereas four (DotI/IcmL, DotU/IcmH, DotC, and IcmV) were expressed at the TP (Fig. 2, supplemental Tables S11 and S12). Notably, L. pneumophila loses its virulence and infectivity without DotA (
). Taken together, these findings suggest that the ClpP-dependent proteolysis of T4BSS and effector proteins during the biphasic life cycle is vital for Legionella pathogenesis.
ClpP Directly Controls Several Substrates Associated with the Biphasic Life Cycle and Bacterial Virulence
Regulated proteolysis is the specific and conditional degradation of substrate proteins that allows the downstream controlled proteins to regulate cellular adaptations and differentiations in response to extracellular or intracellular signals (
). To identify the substrates of ClpP, we performed an in vivo experiment using a proteolytic inactive form of ClpP (ClpPtrap) that retains but does not degrade the substrates that translocate into the proteolytic chamber, as we and others reported previously (
). The plasmids expressing His-tagged ClpPwt and ClpPtrap were transformed into the ΔclpP strain to create ΔclpP/pclpPwt and ΔclpP/pclpPtrap, respectively. Since the His-tagged ClpPwt in this strain has an intact active site, substrates will be degraded upon entry into the ClpP-proteolytic chamber; hence, the proteins that are copurified with the ClpPtrap but are not captured by ClpPwt represent ClpP substrates (Fig. 1C). Substrates captured inside the proteolytic barrel were copurified with the His-tagged ClpP complex and identified by MS (
), such as Pgi (glucose-6-phosphate isomerase), PanC (pantothenate synthetase), and GatB (aspartyl/glutamyl-tRNA[Asn/Gln] amidotransferase subunit B). On the other hand, a large number of substrates was not described before. Interestingly, 37 ClpP-regulated proteins were also captured by the ClpPtrap, demonstrating that the repeated capture of known ClpP substrates and unstable proteins verifies that the ClpP substrates were specifically copurified with the ClpPtrap. The identified ClpP substrates included 19 phase-specific proteins, namely Pgi, PanB, MreC, CydA, CsrA, FlhF, RavE, SdeC, ClpS, ProQm/Lpg0133, Lpg1446, Lpg0248, Lpg0773, Lppg2440, Lpg2948, Lpg1993, Lpg2724, Lpg1102, and Lpg0953, and six effectors (Lem19, LegD1, Lem22, RavE, MavQ, and SdeC). These results suggest that ClpP also is directly involved in the regulation of L. pneumophila life cycle and virulence.
Table 3The identified substrates of ClpP in L. pneumophila
Proteins with different abundances in the ΔclpP strain compared with the WT at the RP that are upregulated or downregulated, respectively, which are above or below the threshold of +1 and −1 log2(ΔclpP/WT).
Proteins with different abundances in the ΔclpP strain compared with the WT at the TP that are upregulated or downregulated, respectively, which are above or below the threshold of +1 and −1 log2(ΔclpP/WT).
Differentially expressed proteins of L. pneumophila WT at the RP and TP, which were above or below the threshold of +1 (RP-specific) and −1 (TP-specific) log2(RP/TP). (−) means no significant difference between the two groups.
ATP-dependent Clp protease adapter ClpS
Flagellar GTP-binding protein FlhF
Cytochrome D ubiquinol oxidase subunit I
Glutamate-rich protein GrpB
Hypothetical signal peptide protein
ABC transporter ElsE
Peptidase, M23/M37 family
Cell shape–determining protein MreC
Segregation and condensation protein B
RNA chaperone ProQ
Polysaccharide ABC transporter
Carbon storage regulator CsrA
Mg2+ and Co2+ transporter CorB, hemolysin
Lipid A biosynthesis acyltransferase
Long chain fatty acid transporter
30S ribosomal protein S20
Type 4 fimbrial biogenesis protein PilZ
GTP cyclohydrolase 1
Type IV pilus biogenesis protein PilQ
Phosphatase family protein
Oxaloacetate decarboxylase alpha subunit
Small protein A, tmRNA-binding
Thiol:disulfide interchange protein DsbD
Siderophore biosynthetic protein FrgA
Two-component sensor kinase
Hypothetical virulence protein
Integral membrane protein
Acetoacetate decarboxylase ADC
Copper efflux ATPase
Riboflavin synthase, alpha subunit RibE
(Type IV) pilus assembly protein PilB
Asn/Gln amidotransferase subunit B
Glycerophosphoryl diester esterase
Hydrogenase maturation factor HypA
Protein-export membrane protein SecG
Peptide transport protein, POT family
Probable Fe(2+)-trafficking protein
a Proteins with different abundances in the ΔclpP strain compared with the WT at the RP that are upregulated or downregulated, respectively, which are above or below the threshold of +1 and −1 log2(ΔclpP/WT).
b Proteins with different abundances in the ΔclpP strain compared with the WT at the TP that are upregulated or downregulated, respectively, which are above or below the threshold of +1 and −1 log2(ΔclpP/WT).
c Differentially expressed proteins of L. pneumophila WT at the RP and TP, which were above or below the threshold of +1 (RP-specific) and −1 (TP-specific) log2(RP/TP). (−) means no significant difference between the two groups.
In addition, we found that two fatty acid biosynthesis–related substrates, WaaM/Lpg0363 (lipid A biosynthesis acyltransferase) and LipB/Lpg1511, showed high confidence values in the interaction with Lpg0102, Lpg0361, and Lpg0362 (supplemental Fig. S13, A and B). A model of the hierarchical position of ClpP in the regulatory network suggests that ClpP directly regulates the abundance of WaaM and LipB and indirectly controls the protein levels of Lpg0102, Lpg0361, and Lpg0362, which then monitors the SpoT expression levels, thereby completing the regulation of L. pneumophila differentiation (supplemental Fig. S13C). Moreover, we have demonstrated that ClpP modulates the level of substrate protein CsrA via direct degradation during the TP, consequently facilitating the progress of the biphasic life cycle of L. pneumophila (
). These results exhibit that ClpP-dependent proteolysis can specifically and conditionally degrade substrate proteins, either to directly perform a regulatory role or to indirectly control cellular events by regulating the abundance of controlled proteins.
We finally compared the life cycle proteome data with the ClpPtrap data and discussed the crosstalk of the regulatory elements that govern L. pneumophila life cycle and virulence in a ClpP-dependent manner using a systems view. The comparison provides testable hypotheses and putative substrates for further determining the significance of ClpP-driven proteolysis (Fig. 9).
The fine-tuned control of biphasic life cycle and temporal delivery of approximately 330 effector proteins into host cells enables L. pneumophila to adapt to changing intracellular and extracellular environments for surviving and proliferation (
). In this study, we present a comprehensive proteomic profile on the life cycle–dependent proteins that are regulated by ClpP-mediated proteolysis (Table 1 and Fig. 3) and report the temporal regulation of effector expression via the ClpP proteolytic pathway (Fig. 8). On a proteomic scale, our data will help to further understand the underlying mechanisms that regulate the phase-specific proteins of this highly adaptive pathogen.
Comparison of the protein levels between the RP and TP revealed that 428 proteins, including 220 that were upregulated in the RP and 208 that were upregulated in the TP, were significantly differentially expressed (Table 1 and supplemental Fig. S1). This trend corroborates the results of a previous study, in which 176 proteins were identified as upregulated in the RP and 147 were upregulated in the TP via an LC–MS-based proteomic analysis (
). Similar to a previous report, cell division proteins (FtsA, FtsQ, FtsZ, and MreC) were observed to be more abundant at the RP, whereas proteins associated with flagellar assembly were more abundant at the TP (MotA, FlgH, FlgI, FlgA, FlhF, Lpg0907, FliA, FliS, and PilM), which contradict previous results (FliA, FlgD, FlgE, FlgK, FliD, and FliC) (
) reported that >20 ribosomal proteins were identified with high abundance both at the RP and TP. This discrepancy may have been due to the differences in the time of sample collection (at 6 h after the cessation of growth versus at 13.5 h when the bacteria were entering the TP and still growing) (
), including the RP-specific proteins related to amino and sugar metabolism, cell division and biosynthetic processes, and the TP-specific proteins associated with virulence and survival (e.g., Dot/Icm-translocated effectors and motility machinery [flagellar and type IV pilus genes]).
Clp proteases are powerful molecular machines that contribute to protein homeostasis during balanced growth, stress responses, and specific pathway regulation (
). In the present study, we identified the ClpP-regulated proteins that are involved in the aforementioned phenotypes, especially the proteins involved in the stringent response network for governing L. pneumophila differentiation (e.g., IHF, LqsR, FliA, PmrA, RpoS, LetE, and CsrA) and the effectors associated with virulence (e.g., RalF, LepA, LpnE, LegC7/YlfA, VipA, VipD, and SidM/DrrA), which are consistent with the previous reports for other bacteria. For instance, ClpP in Staphylococcus aureus is also involved in the regulation of bacterial growth, stress tolerance, intracellular replication, and virulence (