Trigger Factor in Association with the ClpP1P2 Heterocomplex of Leptospira Promotes Protease/Peptidase Activity.

The genomic analysis of Leptospira reveals a trigger factor (TF) encoding gene (tig) to be colocalized along with the clpP1 and clpX. The TF is a crouching dragon-like protein known to be a ribosome-associated chaperone that is involved in cotranslational protein folding in bacteria in an ATP-independent mode. In Leptospira, tig is localized upstream of the clpP1 with a short (4 bp) overlap. In the present study, we document the distinctive role of Leptospira TF (LinTF) in the caseinolytic protease (ClpP) system. The recombinant LinTF (rLinTF) was found to improve the peptidase or protease activity of the ClpP1P2 heterocomplex and ClpXP1P2 complex, respectively, on model substrates. In addition, on supplementation of rLinTF to rClpP1P2 bound to its physiological ATPase chaperone ClpX or the antibiotic analogue acyldepsipeptide (ADEP), an augmentation in the activity of ClpP1P2 was observed. These studies underscore the novel role of LinTF in aiding the caseinolytic protease activity of Leptospira. Supplementation of rLinTF to a peptidase assay of rClpP1P2 conditionally in the presence of a salt (sodium citrate) with high Hofmeister strength led us to speculate that rLinTF may have a role in the assembly of multimeric proteins. The deletion of one of the arms (arm-2) of the LinTF structure from the carboxy terminal domain indicated a reduction in its capacity to stimulate rClpP1P2 activity. Thus, the C-terminal domain of LinTF may have a role in the assembly of multimeric ClpP protein, leading to enhancement of ClpP activity.

Introduction

Leptospirosis is a zoonotic disease which is caused by the pathogenic spirochetes of the genus Leptospira. With around 1 million cases of leptospirosis reported every year globally, this disease causes 58,900 annual deaths.[1] Factors leading to gain in virulence and pathogenesis of Leptospira are subjects of interest for further analysis. Caseinolytic proteases (ClpP) are known to contribute to bacterial virulence by regulating the activity of major virulence factors.[2] Numerous pathogenic bacteria (Staphylococcus aureus, Legionella pneumophila, and Listeria monocytogenes) which were genetically deficient in ClpP protease components got impaired in their ability to inflict infections.[2−6] As a result, caseinolytic proteases are deemed valuable targets for the development of distinctive ways of treating bacterial infections. The spirochete L. interrogans harbors two ClpP isoforms—ClpP1 and ClpP2.[7] In this organism, pure ClpP1 and ClpP2, despite forming oligomeric complexes, are functionally inactive. However, ClpP1 and ClpP2, coupled to each other, can acquire a tetradecameric active complex. While assessing the genes in the vicinity of clpP1 and clpP2 in the genome of Leptospira, we found another gene tig encoding for molecular chaperone trigger factor (LinTF) located near to the serine protease clpP1 and its cognate chaperone, clpX.

In bacteria, tig encodes for the trigger factor (TF). The TF of Escherichia coli (EcoTF) was first discovered for its role in the folding of an outer membrane protein ProOmpA into translocation competent form during membrane assembly.[8] Later, EcoTF was identified as a ribosome-associated peptidyl prolyl isomerase (PPIase) (catalyzes cis/trans isomerization of proline residues) with a chaperone function.[9] It also associates with other known chaperones (DnaK and GroEL) for cotranslational folding of newly synthesized protein chains into active configurations.[10−14] Structure analysis of EcoTF has shown that it is composed of three domains such as N-terminal domain, central/PPIase (peptidyl-prolyl cis/trans isomerase activity) domain, and C-terminal domain.[15] Later, the crystal structure of EcoTF illustrates an interesting arrangement of the domains, as a “crouching dragon”, where the N-terminal domain forms the “tail”, C-terminal forms the “arms”, and the middle PPIase domain protrudes from the structure as the “head”.[16] The N-terminal domain carries a helix-loop-helix element, which is known to contain the so-called TF signature motif (“GFRxGxxP”) accountable for mediating ribosome docking.[17] The central domain, known for its prolyl isomerase activity, belongs to the FKBP (FK506-binding protein) family.[9] The FKBP family of proteins are immunophilins that contain one or several PPIase domains.[18,19] The function of the C-terminal domain of EcoTF is inadequately understood and is believed to provide the chaperone function by aiding the binding of unfolded proteins, and it probably plays a significant role in promoting the dimerization of EcoTF.[15,20−22] Anfinsen hypothesized that the native structure of a protein achieves the confirmation with minimum free energy by folding through intermediates, thus rendering the protein molecule most thermodynamically stable.[23] In the TF dimer of Thermotoga maritima (TmaTF), the C-terminal domain envelopes substrate S7 of T. maritima (TmaS7, a ribosomal protein) in a compact native-like conformation within an Anfinsen cage-like chamber, organized as the asymmetric heterotetramer.[24] The inner hydrophilic lining of the TmaTF Anfinsen cage attracts the hydrophilic groups of TmaS7, thus inciting its externalization and concomitant burial of hydrophobic amino acids in TmaS7, hence stabilizing the complex.[24] Apart from this, the C-terminal domain of TmaTF is also known to interact with ribosomal protein S11 and 16S RNA in T. maritima, which leads the authors to deduce the possible role of the TF in ribosome biogenesis.[24] The TF, in addition to its major function in ribosome-dependent cotranslational folding of nascent polypeptides, has also been contended to be involved in the cotranslational assembly of multimeric protein complexes.[24] The EcoTF purified by size exclusion chromatography was found to interact with a large repertoire of E. coli proteins, ranging in size from ∼8 to ∼120 kDa while stably associating with them, determined by mass spectrometry. A large number of these TF substrates form oligomers or macromolecular assemblages. In another study, Yu-Wei-Shieh et al. in 2015 demonstrated the significant contribution of TF in the cotranslational assembly of the heterooligomeric luciferase complex in Vibrio harveyi, which contains the subunits LuxA and LuxB.[25] The TF in Brucella melitensis, L. monocytogenes, and Streptococcus pneumoniae is also known to possess immunogenic potential, emphasizing its proficiency to serve as a vaccine candidate.[26−30]

To our interest, genomic analysis of L. interrogans serovar Copenhageni reveals the colocalization of tig, clpP1, and clpX. The organization is suggestive of an operon composed of tig, clpP1, and clpX, extending the possibility of functional cooperation of LinTF with the ClpXP degradation system in Leptospira.[31] Interestingly enough, such colocalization is not detected in other pathogenic bacteria with the two isoforms of ClpP. The presence of LinTF near the genes involved in proteolysis in Leptospira hints toward their coordinated regulation. Although the TF is widely accepted to be a polypeptide-folding chaperone, the possible role of the TF in degradation remains unexplored. Insights into the involvement of LinTF in the caseinolytic protease system might help us to gain a meticulous understanding of the functioning of ClpP proteases and to characterize their mode of action. A thorough understanding will enable us to enhance our strategies to design an effective drug to target ClpPs for preventing leptospirosis. This study uncovers the role of LinTF in the caseinolytic protease system of Leptospira by in vitro experiments.

Results and Discussions

Organization of clpP and tig Genes in the Leptospira Genome and Other Pathogenic Bacteria

The genome analysis of L. interrogans serovar Copenhageni revealed that the tig gene is located upstream of the clpP1 in chromosome-1, with a short (4 base pair) overlap between the coding sequences (CDS). The CDS of clpX is located downstream of the clpP1, where only 10 nucleotides separate clpX from clpP1. In addition, all other Leptospira species, classified recently on comparative genomics,[32] demonstrated to have the tig gene lying adjacent to clpP1 (Table S1). The location of tig, clpP1, and clpX was also analyzed in other pathogenic bacteria with two isoforms of ClpP, that is, Clostridium difficile, Mycobacterium tuberculosis, Pseudomonas aeruginosa, Chlamydia trachomatis, and L. monocytogenes (Figure ). In all the listed organisms, the tig gene was separated from the clpP1 by more than 90 nucleotides (Table ). The most extensively studied EcoTF also demonstrated an intergenic distance of 245 bp between the tig and the clpP gene. The three genes (tig, clpP1, and clpX), because of overlapping or short intergenic regions, may be regulated under a single operon in Leptospira. In agreement, in a recent study, tig, clpP1, and clpX are predicted to have an operon organization based on the position of transcription starts in L. interrogans.[31] Such an arrangement of intergenic regions raises the possibility of a functional association of LinTF with the ClpXP degradation system of Leptospira. Perhaps, such colocalization stresses the distinctive role of LinTF in the protein degradation system apart from being actively involved in nascent polypeptide chain folding. In agreement, a study on EcoTF documented it to stimulate the GroEL-dependent degradation of CRAG (CRMP5-associated GTPase);[33] nevertheless, the role of TF in the caseinolytic protease system remains unexplored. In another study, the TF was proposed to be associated with the proteolysis pathways of bacteria and direct aggregation-prone proteins to degradation.[24] In Leptospira, the ClpP1 is not active independently, and it requires to be coupled with ClpP2 to constitute the active protease complex,[7] unlike P. aeruginosa and C. difficile, where pure ClpP1 is an active protease.[34,35] Colocalization of genes encoding LinTF and ClpP1 led us to address whether rLinTF can aid ClpP1 of Leptospira in acquiring the functional 14-mer state.

Figure 1

Schematic arrangement of the genes encoding trigger factor (tig), caseinolytic proteases (clpPs), and co-chaperone ClpX (clpX) in the genome of L. interrogans and other pathogenic bacteria. The arrangement of the genes and their annotations is represented by arrow diagrams (not scaled) based on bioinformatics analysis. The intergenic regions are depicted as interrupted lines. E. coli carry one isoform of clpP, whereas the other represented bacteria harbor two clpP isoforms.

Table 1

Gene Coordinates of tig and clpP1 in Different Pathogens

organism	tig gene coordinates (locus tag)	clpP1gene coordinates (locus tag)	intergenic distance
L. interrogans	1,741,996–1,743,351 (LIC11416)	1,743,348–1,743,944 (LIC11417)	4 bp overlap
C. difficile	316,108–317,394 (BZ168_RS01485)	317,544–318,128 (BZ168_RS01490)	149 bp
M. tuberculosis	2,763,891–2,765,291 (Rv2462c)	2,763,172–2,763,774 (Rv2461c)	116 bp
P. aeruginosa	1,952,665–1,953,975 (PA1800)	1,954,069–1,954,710 (PA1801)	93 bp
L. monocytogenes	1,290,241–1,291,524 (lmo1267)	1,172,076–1,172,648 (lmo1138)	117,592 bp
C. trachomatis	813,177–814,505 (CT_707)	500,360–500,938 (CT_431)	312,238 bp
E. coli	455,133–456,431 (b0436)	456,677–457,300 (b0437)	245 bp

LinTF Influences the Peptidase Activity of the ClpP1P2 Heterocomplex of Leptospira

The open reading frame of the tig gene (LIC11416) was amplified from the genomic DNA of L. interrogans serovar Copenhageni by PCR. The amplicon was cloned in the pET-23a expression vector, and the recombinant protein was overexpressed in E. coli BL21 (DE3) cells. The overexpressed recombinant protein (rLinTF, 53.3 kDa) was purified using Ni-NTA affinity column chromatography with a net yield of 0.2 mg/L and stored at −80 °C after concentration (0.53 μg/μL). The influence of rLinTF on the activity of pure rClpP isoforms of Leptospira was evaluated by performing peptidase assays on model fluorophore-tagged peptide substrate S1 (Suc-LY-AMC, Sigma-Aldrich), as described previously.[7] The influence of rLinTF on the peptidase activity of rClpPs was studied by the rate of production of fluorescent AMC (7-amino-4-methyl coumarin) after cleavage from substrate S1. In contrast to our presumption, the rLinTF did not aid pure rClpP1 and rClpP2 in establishing peptidase activity (data not shown). The purified rLinTF alone also does not possess any peptidase activity on the model peptide substrate (Figure A). However, the rLinTF could aid rClpP1P2 heterocomplex in stimulating the peptidase activity. The functional heterocomplex of rClpP1P2 was generated by mixing pure rClpP1 and rClpP2 (1:1 stoichiometry) and incubating for 48 h at 4 °C.[7] There was a ∼4-fold stimulation in the peptidase activity of rClpP1P2 on the supplementation of rLinTF (Figure A). To address if the influence of rLinTF on the peptidase activity of ClpP1P2 is specific, we added another protein (rLIC13341) of Leptospira in the place of rLinTF in the ClpPs peptidase reaction mixture. The rLIC13341, an outer membrane protein of Leptospira,[36] was found to instead moderately inhibit the peptidase activity of rClpP1P2 on the peptide substrate (Figure A). This observation substantiates the specificity of the role of rLinTF in stimulating the peptidase activity of rClpP1P2. Furthermore, on increasing concentrations of rLinTF (0–1.13 μM) in the peptidase reaction of rClpP1P2 (1 μM), the measured rate of peptide hydrolysis by rClpP1P2 got enhanced (Figure B).

Figure 2

Peptide degradation assays using recombinant trigger factor (rLinTF) and rClpP isoforms of Leptospira. (A) Peptidase activity of the rClpP1P2 heterocomplex (1 μM) on the model fluorogenic peptide substrate, S1 (Suc-LY-AMC), in the presence of either rLinTF (0.37 μM) or rLIC13341 (0.43 μM). The concentration of rClpP1P2 heterocomplex was calculated using the molecular weight of the monomeric subunits. Peptide degradation was measured fluorometrically as a relative fluorescent unit (RFU×1000) using fluorogenic substrate S1 for 1 h. Control refers to the reaction containing buffer, nuclease-free water, and substrate S1. In the inset, peptidase activity of rClpP1P2 is represented in terms of percentage, wherein, after 1 h of the enzymatic reaction, the end-time fluorescence has been presented. The activity of the reaction containing the only rClpP1P2 has been taken as 100%, and the reading for the rest of the reactions has been calculated with respect to it. The activity of the rClpP1P2 gets stimulated significantly in the presence of rLinTF. (B) Rate of peptidase activity of rClpP1P2 (1 μM) on the model fluorogenic peptide substrate, S1 (Suc-LY-AMC) in the presence of increasing molar concentrations of rLinTF (0–1.13 μM), recorded at 1 h time point. The velocity of the reaction, represented in terms of ΔRFU/min, increases with increasing the molar concentration of rLinTF. (C) Peptidase activity of ADEP1 (15 μM)-activated rClpP1P2 (1 μM) on the substrate S1 (Suc-LY-AMC), in the presence of either rLinTF (0.37 μM) or rLIC13341 (0.43 μM). In the inset, peptidase activity of rClpP1P2 is represented in terms of percentage, wherein, after 1 h of the enzymatic reaction, the end-time fluorescence has been presented. The activity of the reaction containing the only rClpP1P2 has been taken as 100%, and the reading for the rest of the reactions has been calculated with respect to it. The activity of the ADEP1-activated rClpP1P2 gets stimulated significantly in the presence of rLinTF. Error bars indicate the standard errors of the mean (SEM) from the two independent experiments performed in duplicates. Statistical analysis has been carried out using Student’s t-test; *p < 0.05.

ADEP1 is a natural antibiotic belonging to the class of acyldepsipeptides, which acts by dysregulating the activity of ClpP protease.[37−39] In Leptospira, ADEP1-activated rClpP1P2 shows stimulated peptidase activity.[40] Next, in order to gain insights into the stimulation function of LinTF, we evaluated whether rLinTF can further unconditionally impact the ADEP1-bound rClpP1P2 peptidase activity. Thus, another peptidase assay was conducted using ADEP1-bound rClpP1P2 on model substrate S1 (Suc-LY-AMC) in the presence of rLinTF. It was observed that rLinTF unconditionally further stimulates the ADEP1-bound ClpP1P2 activity (∼3-fold) versus the control reaction without the rLinTF (Figure C). Interestingly, when rLIC13341 was used instead of rLinTF, such stimulation of the ADEP1-bound rClpP1P2 was at the basal level. This underscores the specificity of the rLinTF role in promoting the rClpP1P2-bound ADEP1 activity.

LinTF Promotes the Protease Activity of the rClpP1P2 Bound to ADEP1/ClpX on the Model Casein Substrate

In Leptospira, the ClpP1P2 requires co-chaperone ClpX for the hydrolysis of protein substrates.[7] ClpX utilizes ATP and functions as an unfoldase for the protein substrates and mediates the passage of the unfolded protein through the ClpP proteolytic chamber. The TF is an ATP-independent molecular chaperone.[41] We illustrated in this study that rLinTF facilitates peptidase activity in ADEP1-bound rClpP1P2, and therefore, it prompted to assess if LinTF could also promote activity in ClpP1P2 bound to ClpX, an ATP-dependent cognate physiological chaperone.

Supplementation of rLinTF to the protease reaction of the rClpXP1P2 complex on fluorescent-labeled substrate FITC-Casein (S2) detected a ∼2.5-fold stimulation in substrate (S2) degradation versus the control with no rLinTF (Figure A). The logical reason for such incitation in protease activity of the rClpXP1P2 complex in the presence of rLinTF may be through facilitating the structural stabilization of the multimeric protein. Moreover, the probability of rLinTF having a role in catalytic stimulation of the ClpXP1P2 complex cannot be denied, as rLinTF may bind to ClpP1 and ClpP2 protomers within the ClpP1P2 complex to maintain the correct folding/conformation of different subunits and allow the correct position of the catalytic triad.

Figure 3

Protein degradation assays using the recombinant trigger factor (rLinTF) and ClpP isoforms of Leptospira. (A) Protease activity of rClpXP1P2 on FITC-Casein in the presence of rLinTF (0.37 μM) after 2 h. Proteolysis by rClpXP1P2 gets stimulated significantly in the presence of rLinTF (0.37 μM). (B) Protease activity of ADEP1-activated rClpP1P2 on FITC-Casein after 2 h. The proteolytic activity of ADEP1 (15 μM)-activated rClpP1P2 (1 μM) gets stimulated significantly in the presence of rLinTF (0.37 μM). (C) Protease activity in the presence of rLinTF in increasing amounts of 0–3 μg (0–1.13 μM). Maximum activity was observed when 1 μg of rLinTF (0.37 μM) was used. Error bars indicate the standard errors of the mean (SEM) from the two independent experiments performed in duplicates. Statistical analysis has been done using Student’s t-test; *p < 0.05. (D) Protease activity of ADEP1-activated rClpP1P2 on β-Casein in the presence of rLinTF (0.37 μM). rLinTF itself does not act as a substrate for the proteolytic chamber, rather stimulates the degradation of β-Casein.

In Leptospira, antibiotic acyldepsipeptide (ADEP1) has been illustrated to activate the proteolytic activity of the ClpP1P2 heterocomplex to degrade protein substrates even in the absence of co-chaperone ClpX.[40] To gain insights into the novel function of stabilizing/generating the functional multimeric protein by the rLinTF, an identical protease assay was conducted with the ADEP1-bound rClpP1P2 in the presence of rLinTF. Protease activity of ADEP1-bound rClpP1P2 was found to be further enhanced (∼3-fold) on supplementation of rLinTF in the reaction mixture versus its control with no rLinTF at 2 h of the reaction period (Figure B). In a successive experiment, rLinTF was added in increasing concentrations (0–1.13 μM) to the protease reaction of ADEP1-bound rClpP1P2 and measured the activity at 2 h of the reaction. In agreement with the conclusions obtained in ClpP1P2 peptidase assays (Figure C), the protease activity of ADEP1-bound rClpP1P2 got augmented in the presence of increasing concentrations of rLinTF (Figure C). The maximum enhancement in activity of ADEP1-bound rClpP1P2 at the given time (2 h) was detected when 0.37 μM of rLinTF was supplemented to the reaction mixture, while maintaining all the other parameters of the protease reaction constant (Figure C). Notably, in the protease reactions in which a higher amount (0.75 and 1.13 μM) of rLinTF was supplemented, a decline in the stimulation was witnessed versus the reaction supplemented with 0.37 μM of rLinTF. Nevertheless, the readings of the reaction mixture fortified with a high amount (0.75 and 1.13 μM) of rLinTF were higher than the basal activity of ADEP1-bound rClpP1P2 with no rLinTF. It is speculated that this could be a fallout of the dysregulated activity of rClpP1P2 because of the overactivation in the presence of both rLinTF and ADEP1. The effect of overactivation of rClpP1P2 may result in an early substrate depletion and autodegradation of the proteolytic complex. In a sequential event, the initiation of autodegradation of ADEP1-bound rClpP1P2 subunits may lead to a reduction in measured activity in the presence of a higher amount of rLinTF.

Antibiotic ADEPs are known to switch the rClpP1P2 machinery into a dysregulated proteolytic chamber, which is capable of degrading its subunits in the absence of enough substrates.[40] Interestingly, in the presence of an adequate amount of substrate casein, such an autoproteolysis event of overactivated ADEP1-bound ClpP1P2 is reduced.[40] With this perception, it is currently critical to address if rLinTF acts as an additional substrate to the model casein for the ADEP-activated ClpP1P2 proteolytic chamber, resulting in further increase protease stimulation due to reduction of autoproteolysis. To address this, β-casein substrate degradation was analyzed by the ADEP1-bound rClpP1P2 heterocomplex activity at 30 min time intervals for 2 h in the presence and absence of rLinTF. The reaction products were resolved on denaturing SDS gel and stained with Coomassie blue. The model casein substrate got degraded faster by the ADEP1-bound rClpP1P2 in the presence of rLinTF (Figure D). On the contrary, the amount of rLinTF with respect to the casein substrate did not witness any remarkable change at numerous time points of the protease reaction by ADEP1-bound rClpP1P2. Also, intensified degraded products at the lower molecular weight (below 15 kDa) were detected (Figure D, lower panel). Thus, it is deduced that rLinTF itself does not act as a substrate for activated ClpP1P2; it rather promotes the ClpP1P2 potential to degrade the model casein substrates efficiently.

LinTF Conceivably Promotes the Assembly of Pure rClpP Isoforms into the Active Heterocomplex

The biochemical assays illustrated a novel role of rLinTF in enriching the rClpP1P2 activity. The activity of the rClpP1P2 got enhanced significantly on both the peptide and casein model substrates. In an earlier study, EcoTF has also been determined to have a role in promoting the assembly of oligomeric proteins such as LuxA and LuxB.[25] Based on this statement, we presumed that LinTF might have a role in the assembly of oligomeric ClpP isoforms to form a functionally active ClpP1P2 protease in Leptospira. In the presence of TF, more ClpP heterocomplexes are getting assembled or stabilized, as a result of which enhancement of the peptidase/protease activity was detected. In a recent study by our group, it is illustrated that rClpP1P2 exhibits enhanced peptidase activity in the presence of sodium citrate,[7] a similar trend as was detected in the current study with rLinTF. Such a citrate salt is classified to have high Hofmeister strength, and because of the “salting-out” effect, there is a stabilization of multimeric complexes.[42] To uphold our presumption about rLinTF having a role in multimeric ClpP assembly, an experiment was conducted in which the peptidase reaction mixture was supplemented with 0.1 and 0.2 M of sodium citrate, both in the presence and absence of rLinTF. When rLinTF (0.37 μM) was added to the reaction containing 0.1 M sodium citrate, additional stimulation of ClpP1P2 peptidase activity was detected (Figure ). However, when rLinTF (0.37 μM) was added to the reaction containing 0.2 M sodium citrate, no additional stimulation was detected. Instead, a decline in rClpP1P2 peptidase activity was noted versus the reaction with no rLinTF containing 0.2 M sodium citrate. From these results, it can be deduced that in the presence of 0.2 M sodium citrate, the maximum number of heterocomplexes have been formed, and hence no additional assembly of subunits takes place in the presence of rLinTF, and the presence of such free form of rLinTF rather inhibits the peptidase reaction. This assay hints toward the possible role of TF in the assembly of multimeric proteins such as ClpP, as a result of which rClpP1P2 exhibits higher activity in the presence of rLinTF.

Figure 4

Effect of rLinTF on the peptidase activity of rClpP1P2 on fluorogenic substrate S1 in the presence of sodium citrate. rLinTF (0.37 μM) showed less stimulation activity in the presence of 0.2 M of sodium citrate. Error bars indicate the standard errors of the mean (SEM) from the two independent experiments performed in duplicates. Statistical analysis has been done using Student’s t-test; *p < 0.05.

Deletion of the Partial C-Terminal End of LinTF Demonstrates a Reduction in Capacity to Promote rClpP1P2 Peptidase Activity

The TF has been documented to have three domains, N-terminal, central, and C-terminal.[16] From the multiple sequence alignment of LinTF with its known orthologues (Figure S1), it was figured that residues (1–144) constitute the N-terminal and the linker of LinTF, residues (145–219) constitute the central domain, whereas residues (220–451) constitute the C-terminal domain (Figures A, S1). In previous reports, the C-terminal domain of TF has been assigned to have the chaperone activity.[15,20−22] To understand the role of a specific segment of LinTF behind the stimulation of ClpP activity, we initially targeted at deleting 100 residues constituting the C-terminal domain (351–451 residues approximately) that encompass the arm-2 of the LinTF (Figure A). The Leptospiratig gene encoding for LinTF with the deleted 100 residues from the C-terminal end (rLinTF_CΔ100) was PCR-amplified and cloned in pET-23a vector. The truncated LinTF (rLinTF_CΔ100) was overexpressed and purified by the native method using Ni-NTA affinity chromatography. A peptidase assay was performed on substrate S1 (Suc-LY-AMC, Sigma-Aldrich) by rClpP1P2 in the presence of rLinTF_CΔ100 and compared with that of rLinTF (as control). The presence of a mutant rLinTF_CΔ100 compromised the rClpP1P2 peptidase activity stimulation (∼3-fold) relative to the full-length rLinTF (∼4-fold) (Figure B). The given assay indicates the possible role of the C-terminal of LinTF in assembling multimeric proteins. Understandably, the effect of deleting both arm-1 and arm-2 of rLinTF could be more significant than deleting only arm-2. Thus, further experiments on deleting the C-terminal of rLinTF (rLinTF_ΔC) and assessing the activity of leptospiral ClpP heterocomplexes in presence of rLinTF_ΔC are warranted. With the given shreds of experiments performed in this study, we speculate that in the presence of rLinTF, more active and stable ClpP heterocomplexes are possibly generated, which increase the peptidase/proteolytic activity, as portrayed in the model (Figure ).

Figure 5

Peptide degradation assays using recombinant arm-2 truncated trigger factor (rLinTF_CΔ100) and ClpP isoforms of Leptospira. (A) Predicted tertiary structure of rLinTF using the SWISS-MODEL server. Models have been represented in the surface form and visualized using PyMol. The modeled structure has a GMQE score of 0.61 and a QMEAN score of −2.08. The 100 amino acids covering the arm-2 of the C-terminal of LinTF were truncated to generate rLinTF_CΔ100. (B) Peptidase activity of rClpP1P2 on the model fluorogenic peptide substrate, S1 (Suc-LY-AMC) in the presence of either rLinTF (0.37 μM) or rLinTF_CΔ100 (0.5 μM). In the right panel, the concentration of the protein samples has been compared in order to verify the accuracy of the results. rLinTF_CΔ100 shows compromised stimulation when compared to rLinTF. Error bars indicate the standard errors of the mean (SEM) from the two independent experiments performed in duplicates. Statistical analysis has been done by Student’s t-test; *p < 0.05.

Figure 6

Model illustrating the possible mechanism of action of trigger factor in Leptospira. TF assists the folding of polypeptides cotranslationally. After protein folding is complete, it may remain attached to the polypeptide chains and aid in forming active oligomeric protein complexes. In the given process, TF may aid in the assembly of ClpP subunits with a higher rate (kT1, kT2 > k1, k2) including ClpX, a molecular chaperone. This may lead to the efficient assembly of complexes and the development of a more number of functional and stable ClpP1P2 and ClpX. Under the given condition, if the equilibrium of heterocomplex formation shifts toward the right, the cumulative activity of the assembled ClpP1P2 heterocomplexes can be higher in the presence of TF.

Conclusions

The biological function of natural supramolecular protein assemblies depends on its stability.[43] Moreover, there are numerous proteins related to diseases that are in equilibrium with oligomer forms and are central in holding the activity levels of supramolecular proteins in vitro and in vivo.[44] Here, as the gene encoding for ClpP and trigger factor of Leptospira are located near each other, we investigated the role of the trigger factor in the caseinolytic protease system of Leptospira by indirectly measuring the effect on the biological activity of the ClpP complex. Based on the ClpP’s activity in the presence of sodium citrate and rLinTF, we speculate that rLinTF may be involved in assembling the active ClpP tetradecamer complex. Moreover, the gain in peptidase activity of ClpP1P2 was compromised in the presence of rLinTF with deleted arm-2. To understand the novel role of LinTF in assembling multimeric proteins, our laboratory is in the process to comprehend how the C-terminal of LinTF mechanistically assists the multimeric proteins to rearrange in the active form and whether such events occur cotranslationally in the proteome of Leptospira.

Materials and Methods

Bioinformatics Analysis

The gene neighborhoods of tig, clpP isoforms, and clpX in Leptospira and other selected organisms (C. difficile, M. tuberculosis, P. aeruginosa, C. trachomatis, and L. monocytogenes) with multiple ClpP or single ClpP (E. coli) were analyzed from the gene database available at the National Centre for Biotechnology Information (NCBI).

Bacterial Strains, Primers, and Plasmids

Bacterial strains, primers, and plasmids used in the study are listed in Table . The spirochete L. interrogans serovar Copenhageni strain Fiocruz L1130 was obtained from the Indian Council of Medical Research (ICMR), Regional Medical Research Centre (RMRC), Port Blair, Andaman, and Nicobar Island, India. Spirochetes were cultured in Ellinghausen–McCullough–Johnson–Harris (EMJH) media at 28–30 °C at an interval of 5–7 days. Luria–Bertani (LB) medium was used for culturing E. coli DH5α and BL21 (DE3) (Novagen) required for cloning and overexpression of recombinant proteins.

Table 2

Bacterial Strains, Plasmids, and Oligos Used in this Study

bacterial strains, plasmids, or oligos	characteristics or sequence	source/reference
L. interrogans serovar Copenhageni strain Fiocruz L1-130	wild-type	ICMR, Port Blair
E. coli DH5α	supE44 ΔlacU169 (ϕ80 lacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1	Novagen
E. coli BL21 (DE3)	F–ompT hsdSb (rBmB–) gal (λ c I 857 ind 1 Sam7 nin5 lacUVt7gene 1) dcm (DE3)	Novagen
pET-23a(+)	bacterial vector for expression of C-terminally His6-tagged proteins	Novagen
tig (1356 bp) oligos	F(NheI): CTAGCTAGCATGGATTATAAAACAAAAAAAAATTC	this study
	R(XhoI): CCGCTCGAGTGCTTTCACTTCTTCCTTT
tigΔ303 (1053 bp) oligos	F(NheI): CTAGCTAGCATGGATTATAAAACAAAAAAAAATTC	this study
	RΔ303(XhoI): CCGCTCGAGCAATTTTTGAAAGGATTCTTGGACT

Cloning, Protein Overexpression, and Purification

The full-length tig gene (LIC11416) of L. interrogans serovar Copenhageni strain Fiocruz L1-130 was PCR-amplified using its genomic DNA as a template. The oligonucleotides for PCR were designed using the genomic sequence of L. interrogans Copenhageni strain Fiocruz L1-130 available at NCBI. The tig gene was cloned individually into the pET-23a (+) vector at the NheI and XhoI sites that can express a C-terminal His6-tagged recombinant protein. The recombinant plasmid construct was sequenced for confirmation of the clone at the outsource facility (Eurofins, India). The recombinant plasmid pET-23a-tig was transformed into competent BL21(DE3) cells, and one of the obtained clones were cultured in LB medium supplemented with 100 μg/mL ampicillin for 3.5 h at 37 °C in the presence of 3% ethanol at 180 rpm. The culture was induced with IPTG (0.75 mM), and the cells were harvested by centrifuging at 5000g for 10 min and washed with 1× phosphate buffer saline (pH 7.4, PBS; 10 mM sodium phosphate, 137 mM NaCl, and 2.7 mM KCl). The recombinant protein was purified by affinity column chromatography using nickel–nitrilotriacetic acid (Ni-NTA) resins (Invitrogen). Initially, cells were lysed with a native lysis buffer (pH 7.8, 100 mM Tris-Cl, 300 mM NaCl, 1% Triton-X, 10% glycerol) followed by brief sonication. Soluble fractions obtained after centrifugation were allowed to bind to the pre-equilibrated Ni-NTA beads for 1.25 h in the presence of MgCl2 (7 mM) and imidazole (5 mM), followed by subsequent washing of the beads with 6 column volume of native wash buffer (pH 7.8, 100 mM Tris-Cl, 300 mM NaCl) containing increasing concentrations of imidazole (50–70 mM). Post washing, the recombinant protein was eluted using 250 mM imidazole in the native elution buffer (pH 7.8, 100 mM Tris-Cl, 300 mM NaCl, 10% glycerol). Elutes were exchanged to storage buffer (pH 7.8, 100 mM Tris-Cl, 100 mM NaCl, 10% glycerol) and further concentrated using 3 kDa Amicon centrifugal units (Amicon, catalog no. UFC200324). The purified proteins were visualized on 12% sodium dodecyl sulfate–polyacrylamide gel by Coomassie Blue staining. Protein concentrations were estimated by the Bradford method with bovine serum albumin as the standard. The truncated form of rLinTF and rLinTF_CΔ100 was overexpressed in BL21(DE3) cells and was purified by the native method as described for rLinTF. The recombinant pure ClpP1, ClpP2, and ClpX were purified by affinity column chromatography as described before.[7]

Prediction of the Secondary Structure of LinTF

The crystal structure of TF from different organisms (E. coli, V. cholerae, M. tuberculosis, and T. maritima) has been reported in the previous study.[24,45−48] The amino acid sequence of LinTF was aligned pairwise with the amino acid sequences of these homologues. The TF of T. maritima (TmaTF) was found to be most similar (47.9%) to LinTF (data not shown), and hence the crystal structure of TmaTF was taken as a template for the prediction of the secondary structure of LinTF using the SWISS-MODEL webserver. Secondary structures used for comparison were retrieved from the protein data bank (PDB). The three-dimensional model generated was visualized using PyMOL.

Peptidase Assay

Peptidase activities of the pure rClpPs and rClpP1P2 in the presence of the rLinTF and their mutant variant were monitored by the rate of production of fluorescent AMC (7-amino-4-methyl coumarin) after cleavage from the model peptide substrate S1 (Suc-LY-AMC, Sigma-Aldrich), as described previously.[7] Assays were performed in black 96-well flat-bottom plates (Invitrogen) at 37 °C. Each reaction contained 1 mM of substrate (S1) in a reaction volume of 50 μL in ClpP peptidase activity buffer[7] (pH 7.6, 50 mM phosphate buffer, 100 mM KCl, 5% glycerol), along with a combination of 1 μg each of rClpP1 (1 μM), rClpP2 (1 μM), and rLIC13341 (0.43 μM) or LinTF (0.37 μM) or LinTF_CΔ100 (0.5 μM), unless otherwise mentioned. Concentrations of rClpP1, rClpP2, and LinTF were calculated using the molecular weight of the monomers. Pure rClpP isoform mixtures were incubated for 48 h at 4 °C before the initiation of the assay to stabilize the rClpP1P2 heterocomplex.[7] Before initiating the reaction, the preincubated (48 h) rClpP1P2 was again incubated with rLinTF or its variant at 37 °C for 10 min. Fluorescence was measured in the Infinite M200Pro plate reader (Tecan) at 380 and 460 nm of excitation and emission wavelength, respectively. Experiments were performed twice independently and in duplicates. Data obtained were processed either in Microsoft Excel, GraphPad Prism 8, or Origin 2020b software packages, as per requirements.

For the peptidase activity of the antibiotic ADEP1-stimulated ClpP1P2 heterocomplex in the presence of rLinTF, the given reaction volume 50 μL was fortified with 15 μM of ADEP1. Suitable controls were designed to analyze the results. The subsequent experimental procedure to detect the hydrolysis of substrate S1 was the same as described above. Experiments were performed twice independently and in duplicates.

Protease Assay

All protease assays were performed using a protease fluorescent detection kit (Sigma-Aldrich, catalog no. PF0100) as per the manufacturers’ instructions. The preincubated (48 h) rClpP1P2 along with the chaperone rClpX was assayed for protease activity in the presence of rLinTF containing a fluorescent protein substrate S2 (FITC-casein) provided in the kit, in the way as described previously.[7] The preincubated (48 h) rClpP1P2 (1 μg each of pure ClpP isoforms) (1 μM) was incubated with 2 μg of rClpX (0.85 μM) and 1 μg of rLinTF (0.37 μM) in ClpP protease activity buffer (pH 7.8, 50 mM Tris-Cl, 50 mM KCl, 1 mM DTT, and 8 mM MgCl2) at 37 °C for 10 min before initiating the reactions. Each reaction contained 20 μL of FITC-casein substrate (1.5 μg/μL) in 50 μL of the protease activity buffer. Briefly, 30 μg of the substrate was prewarmed at 37 °C for 10 min before the addition of 2 mM ATP in a total volume of 50 μL of the reaction. After incubation for 2 h at 37 °C in the dark, the reactions were terminated using 0.6 N TCA (trichloroacetic acid). The resulting fluorescence was recorded at 492 and 519 nm wavelength of excitation and emission, respectively, in the Infinite M200Pro plate reader (Tecan). The study was performed twice independently and in duplicates. For studying the protease activity of ADEP-activated rClpP1P2 in the presence of rLinTF, a similar approach was followed. Briefly, 1 μg each of the ClpP isoforms were preincubated for 48 h to form the stable rClpP1P2 heterocomplex (1 μM), followed by further incubation for 10 min at 37 °C with 1 μg of LinTF (0.37 μM) and 15 μM of ADEP1 in the protease activity buffer, unless otherwise mentioned. The subsequent experimental procedure was the same as described above. A similar study was performed using substrate β-casein. The preincubated rClpP isoform mixture (1 μM) was incubated with ADEP1 (15 μM) in the presence of LinTF (0.37 μM) for 10 min at 37 °C. The activity was studied on the model β-casein substrate (0.5 μg/μL) in a total reaction volume of 100 μL. The reactions were incubated for 2 h at 37 °C. From the total reaction volume, 20 μL of the reaction was terminated at an interval of 30 min by adding a 4×sample loading buffer (pH 6.8, 200 mM Tris-HCl, 8% SDS, 0.4% bromophenol blue, 100 mM DTT, and 40% glycerol) and boiling at 95 °C for 10 min. A control reaction containing the rClpP isoform mixture with ADEP1 was prepared for comparison. The reaction products at each time point were resolved on 12% SDS-PAGE and visualized by Coomassie Blue staining.