Characterization of Rv0888, a Novel Extracellular Nuclease from Mycobacterium tuberculosis

Bacterial extracellular nucleases play important roles in virulence, biofilm formation, utilization of extracellular DNA as a nutrient, and degradation of neutrophil DNA extracellular traps. However, there is no current data available for extracellular nucleases derived from M. tuberculosis. Herein, we have identified and characterized Rv0888, an extracellular nuclease in M. tuberculosis. The protein was overexpressed in E. coli, and the purified Rv0888 protein was found to require divalent cations for activity, with an optimal temperature and pH of 41 °C and 6.5, respectively. Further results demonstrated that Rv0888 nuclease activity could be inhibited by four Chinese medicine monomers. Based on sequence analysis, Rv0888 nuclease exhibited no homology with any known extracellular nucleases, indicating that Rv0888 is a novel nuclease. Site-directed mutagenesis studies revealed that the H353, D387, and D438 residues play catalytic roles in Rv0888. In vivo infection studies confirmed that Rv0888 is required for infection and is related to pathogenicity, as the persistent ability of recombinant Mycobacterium smegmatis (rMS) Rv0888NS/MS and Rv0888S/MS is significantly higher than pMV262/MS in the lung tissue, and the Rv0888NS/MS and Rv0888S/MS could produce pathological changes in the mice lung. These results show that Rv0888 is relevant to pathogenicity of M. tuberculosis.


Results
Expression and Purification of Rv0888. The SignalP server revealed a signal sequence (1 to 31 residues) in the Rv0888 amino acid sequence and, as this hydrophobic sequence might increase the difficulty of expression, the fragment of rv0888 from M. tuberculosis H37Rv was cloned without the predicted signal sequence. In order to purify the protein, Rv0888 was expressed as a 6 × His-tagged protein in E. coli. Though it was expressed in inclusion body form, Rv0888 was renatured and purified via a Ni 2+ affinity chromatography resin (Fig. 1A). Then, the eluted Rv0888 protein was loaded onto an anion exchange chromatography column (Fig. 1B) and subsequently passed through a gel filtration chromatography resin, subsequently resulting in greater than 98% protein purity (Fig. 1C).
Rv0888 protein identity confirmation. The purified Rv0888 protein was analyzed by mass spectrometry and peptide mass fingerprinting results were submitted to Mascot and used to search the SwissProt database. The analysis returned 10 results, with a score of 78 assigned to the expected protein Rv0888 (GeneBank Accession No. NP_215403.1) from M. tuberculosis H37Rv (Fig. 1D).
Rv0888 nuclease activity specificity. To verify the nuclease activity, purified Rv0888 was incubated with different nucleic acids, including linear dsDNA (PCR production), circular plasmid DNA (pGEX-6p-1 vector), chromosomal DNA (E. coli DNA) or RNA from baker's yeast. Surprisingly, all of the nucleic acids were degraded by the Rv0888 protein ( Fig. 2A,B). These results indicated that Rv0888 is a non-specific nuclease.

Effect of divalent cations and metal chelators on Rv0888 activity. The effect of different divalent
cations on nuclease activity of Rv0888 was evaluated. In the absence of divalent cations, nuclease activity was not detected. The enzymatic activity was optimal in the presence of 5 mM CaCl 2 and 5 mM MnCl 2 . Other divalent cations salts-CaCl 2 , MgCl 2 , BaCl 2 and NiCl 2 -were demonstrated to display different stimulation effects of Rv0888 activity, and Rv0888 activity was fully inhibited by 20 mM EDTA (Table 1; Fig. 2C).
Effect of pH and temperature on Rv0888 activity. The effects of pH and temperature on Rv0888 activity towards circular plasmid DNA are displayed in Fig. 3A,C, respectively. Rv0888 had full activity across a pH range of 6.0-8.0, with maximal activity at pH 6.5 (Fig. 3D). Measurements of Rv0888 activity across a wide temperature range indicated that the optimal temperature for the Rv0888 was 41 °C (Fig. 3B).
Kinetic studies of Rv0888 activity. The kinetic parameters of the Rv0888 nuclease were determined at 41 °C, pH 6.5 and including 5 mM divalent cation salts (CaCl 2 and MnCl 2 ) with two different substrates: calf thymus DNA and baker's yeast RNA. Michaelis-Menton kinetics assays revealed that the K m values toward DNA and RNA were 0.306 ± 0.04 mg/mL and 0.012 ± 0.01 mg/mL, respectively (Fig. 4A,B). V max values were found to be 600.56 U/mg/min and 241.11 U/mg/min for deoxyribonuclease and ribonuclease activities, respectively (Fig. 4A,B).
Site-Directed Mutagenesis. Multiple sequence alignment of Rv0888 with related homologs indicated that nine targeted residues (N131, E267, G303, H353, D387, N389, D438, D472, and H473) were highly conserved in the endonuclease/exonuclease/phosphatase family (Fig. 6A). To identify residues in Rv0888 that are necessary for catalytic activity, specific codons corresponding to the important residues were altered by site-directed mutagenesis.
Ten protein variants were produced, each with a single residue of the conserved residues (N131, E267, G303, H353, D387, N389, D438, D472, H473, and D472-H473) replaced with alanine. Alanine was chosen for substitution, as it lacks a bulky side chain and therefore would be unlikely to disrupt the main chain conformation. Comparison of the variant enzyme to wild-type Rv0888 demonstrated that the D438A substitution resulted in a 92% loss of activity, the H353A and D387A substitutions resulted in a ~50% loss of activity, and both the H473A and D472A-H473A substitutions resulted in a ~30% loss of activity. The N131A, E267A, G303A, N389A, and D472A substitutions resulted in losses in activities of 10%, 4%, 6%, 11%, and 15%, respectively (Fig. 6B).

Role of Rv0888 in M. smegmatis persistence in lung and histopathological analysis. The lungs
are a portal to infection by M. tuberculosis, which can adhere to and invade lung epithelial cells 15 . In order to determine the role of Rv0888 in mycobacterial pathogenicity, the persistence ability of M. smegmatis overexpressing Rv0888 in lung tissue was estimated. Three groups of 3 mice were infected intranasally with a dose (2 × 10 7 colony forming units) of the rMS strains pMV262/MS, Rv0888NS/MS, and Rv0888S/MS, respectively. Bacterial loads in lung tissue were measured at 4 h, 24 h, 4 d, 7 d, and 17 d after infection (Fig. 7). No significant difference was observed between the bacterial loads of Rv0888NS/MS and Rv0888S/MS groups at all time-points, whereas the bacterial loads of Rv0888NS/MS and Rv0888S/MS groups were remarkably higher, compared with that of pMV262/MS group, at 4 d, 7 d, and 17 d after infection. Importantly, in contrast to the almost total clearance of Histopathological analysis revealed that lungs from mice at 7 d after infection in the pMV262/MS group had no pathological changes, whereas mild hyperplasia was observed in alveolar epithelial cells of the mice in the Rv0888NS/MS group, and partial mild hematopedesis and hyperplasia were observed in alveolar epithelial cells in the Rv0888S/MS group (Fig. 8). Taken together, it was suggested that nuclease activity is required for mycobacteria to resist defensive clearance of lung tissue and may be related to mycobacterial pathogenicity.

Discussion
Extracellular nucleases from Gram-positive bacterium play vital roles in virulence and degrading DNA in NETs produced by neutrophils 16 or macrophages 17 . A variety of studies characterizing extracellular nucleases have focused on the proteins that derived from Streptococcus agalactiae 6 , Streptococcus pyogenes 18 and Staphylococcus aureus 6,19 . Though a previous report had described the intracellular nucleases in M. tuberculosis 20 , no data is currently available regarding extracellular nucleases from the pathogen. In our investigation, the Rv0888 protein was identified as an extracellular nuclease from M. tuberculosis for the first time.
Activity characterization revealed that Rv0888 nuclease was active at temperatures ranging from 37 °C to 51 °C, with an optimal temperature of 41 °C. This temperature range is consistent with the body temperature during fever in humans and animals after M. tuberculosis infection, suggesting that Rv0888 may be related to M. tuberculosis virulence. Similar to other DNase/nucleases, Rv0888 nuclease requires divalent cations for its

Divalent cation (5 mM) Relative activity (%)
None ND activity. Mn 2+ and Ca 2+ were found to be optimal for its nuclease activity, although other divalent cations (Mg 2+ , Ba 2+ , Ni 2+ ) could also stimulate Rv0888 activity (Table 1). These observations indicated that either the divalent cation diameter was not a factor in accessing the Rv0888 catalytic site or that these cations bind to different amino acids to maintain activity, as previously reported for the E. coli nuclease, Colicin E9 21 . Metal chelation, as assessed by EDTA treatment, had a strong inhibitory effect on the activity of the Rv0888 enzyme, suggesting that metal ions were obligatory for nuclease activity and that they might play vital roles in stabilizing the enzymatic structure. Amino acid substitution studies indicated that the D438 residue may play a key role in Rv0888 nuclease activity, while the H353A and D387A substitutions had a positive influence on Rv0888 activity. The importance of these residues indicate that Rv0888 differs from other bacterial extracellular nucleases which contain the conserved DRGH motif (as in S. pneumoniae EndA) 22 or the degenerate DKGH motif (as in S. agalactiae Gbs0661) 6 . Although the Rv0888 protein showed non-specific nuclease activity, it also shares no homology to any known nucleases, as determined by the primary amino acid sequence ( Supplementary Fig. S1). Thus, our data suggested that Rv0888 is a novel member of a non-specific nuclease family.
Tuberculosis is difficult to control due to co-infection with HIV and the emergence of virtually untreatable and/or extensively drug-resistant M. tuberculosis strains. New drugs and more effective vaccines are urgently needed, thus necessitating a better understanding of the genetic basis for M. tuberculosis virulence and pathogenesis 23 . Interestingly, Rv0888 activity was inhibited by four Chinese medicine monomers (Oleuropein, 6-Gingerol, Corylifolinin, and Acteoside). Previous studies have concentrated on small molecule inhibitors, focusing on proteins and factors related to the biosynthesis of pathogens: limiting S. aureus entry into endothelial cells by structural analogs of ML 141 24 ; blocking coronavirus and filovirus entry into host cells by the cysteine protease inhibitor K11777 25 ; blocking phosphatidylglycerol-LtaS binding and inhibiting LTA synthesis in ItaS-expressin S. aureus and in Escherichia coli with compound 1771 26 ; Inhibiting A. baumannii Lipid II (an essential precursor of cell wall biosynthesis) through the use of BAS00127538 27 . Subsequent drug studies then shifted to cell wall proteins and proteases: small molecule inhibitors of sortase that act as an anti-infective therapy against hospital-acquired S. aureus infection without the side effects of standard antibiotics 28 ; tetrahydrolipstatin that inhibits the phospholipase/thioesterase (Rv3802c) of M. tuberculosis 29 ; an inhibitor of mPTPB (a virulence factor of tuberculosis) which was acquired by organocatalysis 30 ; six small molecules that were confirmed to act as novel EndA (surface endonuclease of the S. pneumoniae) inhibitors 31 . Importantly, this manuscript report for the first time that Chinese medicine monomers are effective inhibitors of bacterial nucleases. New drug research is   nuclease localizations in bacteria: secreted in the medium (S. pyogenes), membrane-associated (S. pneumoniae) and peptidoglycan-anchored (S. pyogenes, Streptococcus suis) 18,32,33 . In M. tuberculosis, Rv0888 contains a signal peptide and is secreted into the medium 14 . This exported product could be more effective for M. tuberculosis pathogenecity than a membrane-associated nuclease like S. pneumoniae EndA, which would have limited access  to its substrate 34 . The characterization of Rv0888 nuclease in this study will facilitate the understanding of the pathogenesis of M. tuberculosis and might prompt identification of novel drugs to treat TB.

Materials and Methods
Animals, bacterial strains, and growth conditions. The mice used in this study were purchased from Vital River, Beijing, China. All experimental work involving animals was performed according to the guidelines recommended by the animal welfare and ethics of the Heilongjiang Animal Ethics Committee at the Heilongjiang science and technology government agency (Harbin, People's Republic of China) and was approved and supervised by the commissioner for animal welfare at the Harbin Veterinary Research Institute (HVRI) representing the Institutional Animal Care and Use Committee. All the experiments were designed to minimize the numbers of animals used and every effort was made to minimize both pain and distress to the animals.
Cloning and expression. The amino acid sequence of Rv0888 was checked for potential signal sequences using the SignalP 4.0 server (http://www.cbs.dtu.dk/services/SignalP/). Primers (Supplementary Table S1) were designed to amplify the rv0888, rv0888NS, and rv0888S genes from genome of M. tuberculosis strain H37Rv. PrimeSTAR Max DNA polymerase (TaKaRa Bio) was used for PCR amplification. The rv0888 amplicon was cloned into a pET22b vector (Novagen) with a 6 × His-tag at the C-terminal and the recombinant plasmid (pET-22b-Rv0888) was transformed into Rosetta 2 (DE3) E. coli. The Rv0888 protein was expressed as the inclusion-body form. The recombinant Rosetta 2 (DE3) E. coli was grown in LB medium (200 μ g/mL ampicillin and 34 μ g/mL chloromycetin) at 37 °C until OD600 = 0.6-1.0, and were then induced with 1 mM IPTG at 37 °C for 4 hours. The rv0888NS and rv0888S amplicons were cloned into a pMV262 vector (assembled in our laboratory, the corresponding sequence is shown in the supplementary data) and the recombinant plasmids (pMV262-Rv0888NS and pMV262-Rv0888S, respectively) were transformed into M. smegmatis mc 2 155 cells. The rMS was grown in 7H9 medium (containing 0.05% Tween-80, 0.2% Glycerol, 10% OADC, 50 μ g/mL kanamycin) at 37 °C until OD600 = 0.8-1.0. Protein expression was analyzed by SDS-PAGE.  Table S2) to introduce N131A, E267A, G303A, H353A, D387A, N389A, D438A, D472A, H473A, D472A-H473A substitutions into the nucleotide sequence; PCR amplification was performed using the PrimeSTAR Max DNA Polymerase (TaKaRa Bio). The PCR products were digested with DpnI to damage the methylated parental template DNA and, the mutated plasmids were transformed into DH5α E. coli. All of the substitutions were confirmed by DNA sequencing. The protein purified from affinity chromatography was centrifuged in a 30 kDa Centricon concentrator (Millipore) and concentrated to ~2 mL. The concentrated protein was loaded onto a HiLoad 16/600 Superdex 200 pg column (GE Healthcare) in buffer J (20 mM Tris-HCl, pH 8.0) and all peak fractions were analyzed by SDS-PAGE. The verified protein was concentrated again and loaded onto a Resource S 5 mL column (GE Healthcare), which was pre-equilibrated with buffer J. The target protein was eluted from the column with a linear gradient of NaCl concentration from 0 M to 2 M. Fractions were collected and analyzed by SDS-PAGE.

Purification of
Mass Spectrometry Analysis of Rv0888. The purified protein was run on a 12% SDS-PAGE gel and the lane containing the target protein was removed and cut into pieces of about 1 mm 3 . After washing three times with sterile water, the gel pieces were destained by three sequential 15-min treatments with 100 μ L of 25 mM ammonium bicarbonate (Fluka) in 50% (v/v) acetonitrile (Fisher) at pH 8.0, and the pieces where then washed once more with sterile water. The gel slices were dehydrated with 30 μ L 100% acetonitrile and completely dried with a Speed-Vac at room temperature. The samples were digested with 8 μ L of 0.1 mg/mL trypsin (Promega) overnight at 37 °C. Then, 0.3 μ L of the digested protein was mixed with 0.3 μ L of matrix solution (5 mg/mL α -cyano-4-hydroxycinnamic-acid (Fluka) in 50% (v/v) acetonitrile (Fisher) and 0.1% (w/v) trifluoroacetic acid (DIMA)) and spotted onto a sample plate for matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry analysis. Positive ion reflection mode was selected and the results were analyzed using 4000 Series software. The target protein was identified by comparing the peptide sequences against homologous proteins in the SwissProt databases using the following parameters: precursor tolerance, ± 0.2 Da; MS/MS tolerance, ± 0.8 Da; missed cleavages, 1.
Substrates for Rv0888 nuclease. To measure the substrate specificity for the recombinant Rv0888 nuclease, 0.2 μ g of linear dsDNA (PCR amplified), circular plasmid DNA (pGEX-6p-1 vector), chromosomal DNA (E. coli DNA) or RNA from baker's yeast (Sigma-Aldrich) were incubated with 3 μ g of the purified Rv0888 protein in 10 μ L of reaction buffer (20 mM Tris-HCl, 5 mM MgCl2, pH 7.5) at 37 °C. Buffer (20 mM Tris-HCl pH 7.5) served as a negative control for the Rv0888 protein. After 60 min, 10 μ L of the reaction solution was mixed with 1 μ L of 10 × loading buffer and analyzed by electrophoresis on a 1.0% agarose gel. All assays were conducted in triplicate.
Effect of temperature and pH on Rv0888 activity. The effect of pH on Rv0888 activity was determined using optimal divalent cations over a pH range of 6.0-8.0 (at 0.5 intervals) in 20 mM Tris-HCl at 37 °C for 1 h. The impact of temperature on Rv0888 activity was measured using the optimal divalent cations and the optimal pH over a temperature range of 33 °C-51 °C (at 2 °C intervals) for 1 h. All results were verified by electrophoresis on a 1.0% agarose gel. Rv0888 activity was quantified using a previously described spectrophotometric method 38 . For a 200 μ L reaction mixture, 20 μ L of 10 × reaction buffer (100 mM Tris-HCl, 50 mM CaCl 2 , 50 mM MnCl 2 , pH 7.5), 20 μ L of circular plasmid DNA (100 ng/μ L pGEX-6p-1), 10 μ L of Rv0888 protein (1 mg/mL), and 150 μ L of sterilized water were combined. Next, after incubation at 37 °C for 1 hr, the reaction was stopped by adding 8 μ L of 0.5 M EDTA. Then the reaction mixtures were diluted to 2 mL using sterilized water and were measured using a Nanophotometer TM Peal Ultramicro UV-Vis spectrophotometer set at 260 nm. One enzymatic activity unit was defined as the amount of enzyme required to increase the absorbance by 0.001 A260 mL -1 ·min -1 ·cm -1 at 41 °C.
Kinetic studies of Rv0888 nuclease activity. Michaelis-Menton kinetic parameters of Rv0888 were measured using spectrophotometric assays with calf thymus DNA and baker's yeast RNA. To achieve this, 10 μ g of purified Rv0888 protein was incubated at 41 °C with different substrates at varying concentrations (DNA: 0.00625 mg/mL, 0.0125 mg/mL, 0.025 mg/mL, 0.05 mg/mL, 0.1 mg/mL, 0.2 mg/mL, 0.4 mg/mL, 0.6 mg/mL, 0.8 mg/mL; RNA: 0.00625 mg/mL, 0.0125 mg/mL, 0.025 mg/mL, 0.05 mg/mL, 0.1 mg/mL) for 1 h. Rv0888 protein was substituted with 20 mM Tris-HCl (pH 6.5) as a negative control. After incubation, the reactions were stopped by adding an equal volume of 5% ice-cold perchloic acid, followed by dilution to 2 mL with sterilized water. The diluted solutions were measured using a Nanophotometer TM Peal Ultramicro UV-Vis spectrophotometer set at 260 nm. All assays were repeated three times.
Lung infection. In this study, 5-6 week old female BALB/c mice (Vital River, Beijing) were maintained under specific pathogen-free conditions and used for lung infection assays. The rMS strains (pMV262, Rv0888NS-pMV262 without its signal peptide, Rv0888S-pMV262 containing its signal peptide) were prepared in 7H9 medium, harvested at full log phase, and washed twice in 2 mL PBS before animal injection. Mice were randomly divided into 4 groups and anaesthetized with diethyl ether prior for bacterial administration. A total of 2 × 10 7 cfu of rMS strains (pMV262, Rv0888NS-pMV262, Rv0888S-pMV262) in 50 μ L of PBS solution were nasally administered, using PBS as a negative control. At 4 h, 24 h, 4 d, 7 d, and 17 d post infection, 3 mice from each group were killed. Lungs were removed from the mice, with the entire left lobe of the lung homogenized with LB under aseptic conditions and then spread onto LB Agar plates containing 50 μ g/mL kanamycin. The plates were incubated at 37 °C for 3-4 days for counting bacteria. About one third of the right lobe of the lung was fixated with 10% formalin for pathological analysis.
Each experiment was independently repeated three times. A two-tailed Student's t test was used to determine statistical significance, and Prism software (version 5.0; GraphPad, San Diego, CA, USA) was used for these analyses. A P < 0.05 was considered significant (***P < 0.001).