ORIGINAL RESEARCH published: 02 July 2019 doi: 10.3389/fimmu.2019.01500 Human Antimicrobial RNases Inhibit Intracellular Bacterial Growth and Induce Autophagy in Mycobacteria-Infected Macrophages Lu Lu 1 , Javier Arranz-Trullén 1,2 , Guillem Prats-Ejarque 1 , David Pulido 1† , Sanjib Bhakta 2* and Ester Boix 1* 1 Department of Biochemistry and Molecular Biology, Faculty of Biosciences, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain, 2 Mycobacteria Research Laboratory, Department of Biological Sciences, Institute of Structural and Molecular Biology, Birkbeck, University of London, London, United Kingdom Edited by: The development of novel treatment against tuberculosis is a priority global health Alexandre Corthay, challenge. Antimicrobial proteins and peptides offer a multifaceted mechanism Oslo University Hospital, Norway suitable to fight bacterial resistance. Within the RNaseA superfamily there is a group Reviewed by: Anthony George Tsolaki, of highly cationic proteins secreted by innate immune cells with anti-infective and Brunel University London, immune-regulatory properties. In this work, we have tested the human canonical United Kingdom Jianguo Liu, members of the RNase family using a spot-culture growth inhibition assay based Saint Louis University, United States mycobacteria-infected macrophage model for evaluating their anti-tubercular properties. *Correspondence: Out of the seven tested recombinant human RNases, we have identified two Sanjib Bhakta members, RNase3 and RNase6, which were highly effective against Mycobacterium s.bhakta@bbk.ac.uk; sanjib.bhakta@ucl.ac.uk aurum extra- and intracellularly and induced an autophagy process. We observed Ester Boix the proteins internalization within macrophages and their capacity to eradicate the Ester.Boix@uab.cat intracellular mycobacterial infection at a low micro-molar range. Contribution of the † Present Address: enzymatic activity was discarded by site-directed mutagenesis at the RNase catalytic David Pulido, The Jenner Institute, University of site. The protein induction of autophagy was analyzed by RT-qPCR, western blot, Oxford, Oxford, United Kingdom immunofluorescence, and electron microscopy. Specific blockage of auto-phagosome formation and maturation reduced the protein’s ability to eradicate the infection. Specialty section: In addition, we found that the M. aurum infection of human THP1 macrophages This article was submitted to Molecular Innate Immunity, modulates the expression of endogenous RNase3 and RNase6, suggesting a function a section of the journal in vivo. Overall, our data anticipate a biological role for human antimicrobial RNases Frontiers in Immunology in host response to mycobacterial infections and set the basis for the design of novel Received: 18 March 2019 Accepted: 14 June 2019 anti-tubercular drugs. Published: 02 July 2019 Keywords: antimicrobial peptides, ribonucleases, tuberculosis, macrophage, autophagy Citation: Lu L, Arranz-Trullén J, Prats-Ejarque G, Pulido D, Bhakta S INTRODUCTION and Boix E (2019) Human Antimicrobial RNases Inhibit Tuberculosis (TB) is an ancient life-threatening infectious disease currently rated among the top ten Intracellular Bacterial Growth and Induce Autophagy in causes of death worldwide. According to the World Health Organization (WHO), TB is responsible Mycobacteria-Infected Macrophages. for about 1.6 million TB deaths and 10 million (5.8 million men, 3.2 million women, and 1.0 million Front. Immunol. 10:1500. children) new cases have been detected in 2017 and more than a third of the world population is doi: 10.3389/fimmu.2019.01500 hosting Mycobacterium tuberculosis, the causative pathogen of TB, in its latent form (1, 2). In recent Frontiers in Immunology | www.frontiersin.org 1 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities years, we have seen a general decline in the TB incidence rate as Our research group works on the mechanism of action of well as in the absolute number of registered cases of tuberculosis. human antimicrobial RNases, a group of proteins that belong Nonetheless, the emergence of a growing number of new cases to the RNase-A superfamily. The family encompasses eight of multi-drug resistant TB (MDR-TB) and even more alarming functional members in humans, referred as the “canonical extensively drug-resistant TB (XDR-TB) cases have placed the RNases” (17). Among them, there are proteins secreted by eradication of TB as one of the major global challenges to epithelial and blood cell types during host-defense response that overcome in the twenty-first century. Furthermore, TB prognosis are endowed with the characteristic biophysical features of AMPs is significantly aggravated during human immunodeficiency (12, 18, 19). In particular, human RNase3, RNase6, and RNase7 virus (HIV) co-infection. A complex cellular machinery are highly cationic proteins with reported antimicrobial activity is activated during the host immune response against M. against a variety of microbes (20–23). Interestingly, secretion of tuberculosis (Mtb) bacilli; unfortunately, their underlying RNases 3 and 7 was associated to Mycobacterium bovis (M. bovis) mechanisms remain poorly understood. Hence, enlargement BCG and M. tuberculosis infection, respectively (14, 24). Indeed, of novel and effective therapies to battle mycobacteria are both RNases displayed antimycobacterial activity in vitro (25). urgently required (3, 4). Mtb is an intracellular pathogen able to In this work, we aimed to study the antimycobacterial activity survive indefinitely under unfavorable conditions inside primary of the human canonical RNases by applying an integrated host immune cells, preferably residing in human alveolar surrogate model for screening of anti-tubercular drugs against M. macrophages (5, 6). Several studies have recently shown that tuberculosis (8). The semi-solid agar-based spot-culture growth features such as its high infectivity, slow growth, and complex inhibition assay (SPOTi) has facilitated a rapid high-through cell-wall structure, make M. tuberculosis a major challenge screening of antimicrobial agents against Mycobacterium aurum to be faced, since the specific mechanisms that reinforce the (M. aurum) using a macrophage infected model (8, 26, 27). high virulence of Mtb remain largely unknown (4, 7). Despite M. aurum was chosen as a relatively fast-growing and stable the antimicrobial activity of macrophages, Mtb has been able intracellular dwelling species within infected macrophages, to establish a series of strategies to handle the host immune previously validated as an appropriate surrogate model for machinery, interfere with, and arrest the phagosome maturation, Mtb (8). RNases with positive antimycobacterial activity were counteract mycobactericidal molecules and ultimately survive further characterized by exploring their mechanism of action in a hostile intracellular environment (8–10). Therefore, the in macrophages. The present results indicate that the protein search for new anti-TB agents that are able to safely penetrate antimycobacterial action is partly mediated by the induction of the host immune cells and eradicate the pathogen intracellularly an autophagy process. is now one of the priorities of global health action programs to tackle tuberculosis. MATERIALS AND METHODS The eventual outcome of a mycobacterial infection largely depends on the readiness of the host immune system to Protein Expression and Purification counter the pathogen. During infection a large assortment of The cDNA for RNase1 was a gift from Prof. Maria Vilanova antimicrobial protein and peptides (AMPs) are released by (Universitat de Girona, Spain) and RNase5 from Prof. Demetres host immune cells to the bloodstream and nearby tissues to Leonidas (University of Thessaly, Greece). RNase4 synthetic gene the infected areas (11, 12). At the same time, neutrophil, and was purchased from Nzytech company and RNase6 was obtained eosinophil granules, loaded with lytic enzymes, and antimicrobial from DNA 2.0 (Menlo Park, CA, USA). RNase2, RNase3, and molecules, can be acquired by macrophages to counter bacterial RNase7 genes were obtained as previously described (20). Active invasion (13–15). Because AMPs exert a potent anti-infective site mutations into the RNases genes were introduced using the effect against a wide range of human pathogens and the likelihood Quick changeTM site-directed mutagenesis kit (Agilent, 200523) of microbial resistance is mostly reduced in comparison to following the manufacturers procedure (28). Escherichia coli conventional antibiotics, they are emerging as a new generation (E. coli) BL21 (DE3) (Novagen, 69450) competent cells were of natural lead candidates and its administration in combination transformed with the pET11c/RNase plasmids. Recombinant with or without other drugs is showing highly promising proteins were expressed and purified in E. coli BL21 (DE3) by results (16). an adaptation of the protocol previously described (20). Briefly, E. coli BL21(DE3) cells were induced by 1 mM isopropyl β-D-1- thiogalactopyranoside (G Bioscience, RC-063) and the inclusion Abbreviations: DOPC, 1,2-dioleoyl-sn-glycero-3-phosphocholine; DOPG, bodies enriched pellet was resuspended in 80 mL of 10 mM 1,2-dioleoyl-sn-glycero-3-phosphoglycerol); 3MA, 3 Methyadenine; MTT, 3- Tris-HCl pH 8.5, 2 mM EDTA and left incubating 30 min with (4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; AO, Acridine orange; AMPs, Antimicrobial protein and peptides; BA, Bafilomycin A; CFU, 40 µg/mL of lysozyme prior to sonication. Then, the sample was Colony forming unit; XDR, Extensively drug-resistant; FBS, Fetal bovine centrifuged at 30.000 × g for 30 min at 4◦ C and the pellet was serum; HTS, High-throughput screening; LUVs, Large unilamellar vesicles; resuspended in 25 mL of the same buffer supplemented with MB, Middlebrook; MAC, Minimum agglutination activity; MBC, minimum 1% triton X-100 and 1 M urea and was left stirring at room bactericidal concentration; MIC, minimum inhibitory concentration; MOI, temperature for 30 min and centrifuged 30 min at 22.000 × g. Multiplicity of infection; MDR, Multi-drug resistant; Mtb, Mycobacterium tuberculosis; NCTC, National Collection of Type Cultures; PMA, Phorbol Following, 200 mL of 10 mM Tris-HCl pH 8.5, 2 mM EDTA myristate acetate; SPOTi, Spot-culture growth inhibition assay; TEM, was added to the pellet, and then the sample was centrifuged Transmission electron microscopy; TB, Tuberculosis; ZN, Ziehl-Neelsen. at 22.000 × g for 30 min (4◦ C). The resulting pellet solubilized Frontiers in Immunology | www.frontiersin.org 2 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities in 6 M guanidine hydrochloride and rapidly 80-fold diluted in minimum inhibitory concentration (MIC100 ) in mycobacteria the refolding buffer was left in gentle stirring for 48–72 h at cultures. The semi-solid agar-based SPOTi assay system was 4◦ C. The folded protein was then concentrated, dialyzed against applied for fast and high throughput screening and the colony the chromatography buffer and purified by cation exchange forming unit (CFU) counting assay for the validation and chromatography on a Resource S column (GE Healthcare Life accurate MBC/MIC100 determination (8). The MBC100 and Sciences, GE17118001). The identity and homogeneity of the MIC100 of each protein was determined from two independent purified proteins were checked by SDS-PAGE, MALDI-TOF, and experiments performed in triplicate for each concentration. N-terminal sequencing. Briefly, log-phase cultures of mycobacteria (OD600 ≈1) were first checked for quality control using cold Ziehl–Neelsen (ZN) Macrophage Cell Culture staining (also called “acid fast staining”; TB-color staining Mouse RAW 264.7 cells (NCTC, #91062702) and human THP-1 kit, Merck Millipore, 116450) according to the manufacturer’s cells (NCTC, #88081201) were maintained or passaged in 25 cm2 protocol. For extracellular analysis, the mycobacterial cultures tissue culture flasks (BD Biosciences, 353108) using DMEM were then diluted to 2 × 105 CFUs/mL and directly incubated (Lonza, BE04-687F/U1) and RPMI-1640 (Lonza, BE12-702F) with proteins at 37◦ C for 4 h in PBS in 1 mL. For intracellular medium with 10% heat-inactivated fetal bovine serum (FBS, analysis, the mycobacterial cultures diluted to 2 × 105 CFUs/mL Gibco, 26140079) respectively at 37◦ C and humidified 5% CO2 were used to infect macrophages and then treated with the conditions. RAW 264.7 cells were seeded at 5 × 105 per well proteins for up to 72 h in RPMI-1640 medium with 10% FBS. and allowed to attach for 2 h before infection and treatment. MIC100 were calculated at final incubation time. Uptake of THP-1 cells were treated with 100 nM phorbol myristate acetate mycobacteria by RAW macrophages was confirmed using the (PMA, Sigma-Aldrich, P8139) for 48 h to induce differentiation ZN staining assay, as described (8). No significant reduction into macrophage-like cells and allowed to rest for 24 h before in the number of live macrophages was observed during the further treatment. The number of viable cells was counted using experiment. Macrophage infection was monitored by CFUs a Trypan blue (Invitrogen, 15250-061) exclusion assay. counting. Following the treatment, to evaluate the activity at the protein extracellular activity the cultures were centrifuged Growth of Mycobacteria Culture and and suspended in 50 µL of distilled water and 5 µL of the Macrophage Infection sample was spotted onto wells of a 24-well plate containing Mycobacterium aurum was purchased from the UK National MB7H10/OADC/agar. For evaluation of the protein intracellular Collection of Type Cultures (NCTC). Cells cultures of M. aurum activity, the macrophages were washed twice with RPMI-1640 (NCTC, 10437), Mycobacterium smegmatis mc2 155 (ATCC, and lysed in water to harvest intracellular mycobacteria for 700084) and M. bovis BCG Pasteur (ATCC, 35734) were grown in plating in 24-well plate (SPOTi) or 10 mm petri dish (CFU assay). Middlebrook (MB) 7H9 broth (BD Biosciences, 271310) enriched Initial infection was adjusted to 2 × 105 CFU/mL for all samples. with 10% (v/v) albumin/dextrose/catalase (ADC; BD Biosciences, Mycobacterial cell growth at 37◦ C was recorded after 2 days 212352) containing 0.05% Tween 80 and 0.05% Glycerol for for M. smegmatis, 4–5 days for M. aurum, and 14 days for M. liquid growth at 37◦ C for BCG, and in MB7H10 (BD Biosciences, bovis BCG. 262710) with 10% (v/v) oleic acid/albumin/dextrose/catalase (OADC; BD Biosciences, 212240) for semi-solid agar growth Minimum Agglutination Activity (MAC) at 37◦ C. Stock cultures of log-phase cells were maintained in Assay glycerol (25% final concentration of glycerol) at −80◦ C. The Mycobacterial cells were grown at 37◦ C to an OD600 of 1.0 bacteria were vortexed and sonicated using ultrasound sonication and diluted ten times. Then, cells were centrifuged at 5,000 × bath to obtain a single cell suspension, and then the bacterial g for 2 min, and resuspended in M7H9/ADC media containing concentration was determined by measuring the optical density 0.05% Tween-80 and 0.05% glycerol. An aliquot of 100 µl of (OD) of the culture at 600 nm (1 OD =109 CFU/mL). Mid- the mycobacterial suspension was treated with increasing protein log phase M. aurum cells, harvested in RPMI-1640 complete concentrations (from 0.01 to 25 µM) and incubated at 37◦ C for medium, were co-cultured with RAW264.7 or THP-1 at a 1 h. The aggregation behavior was observed by visual inspection multiplicity of infection (MOI) of 10:1 and were incubated at with a stereomicroscope at 50x, and the agglutinating activity was 37◦ C for 3 h, then were washed 3 times with PBS and replaced calculated as the minimum agglutinating concentration of the with fresh media supplied with 50 µg/mL gentamycin (Apollo protein, as previously described Pulido et al. (22). Scientific, BIG0124) to remove extracellular mycobacteria during further treatment. Cytotoxicity Assay To assay the toxicity of RNases to RAW264.7 macrophages, the High Throughput SPOTi Assay, Minimum assay was performed in 96-well cell culture flat-bottom plates (Costar; Appleton Woods) in triplicate. To each well, 100 µL Inhibitory Concentration (MIC), and of diluted macrophage cells (5 × 105 cells/mL) were dispensed Minimum Bactericidal Concentration in 96-well plates and then proteins were added serially at final (MBC) Assays concentrations ranging from 0.01 to 50 µM (100 µL/well) and Antimicrobial activity of RNases was evaluated by calculating the incubated for 48 h; cells were then washed twice with 1X PBS, 100% minimum bactericidal concentration (MBC100 ) and 100% and fresh RPMI-1640 complete medium was added. Plates were Frontiers in Immunology | www.frontiersin.org 3 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities then treated with 30 µL of a freshly prepared 0.01% resazurin the treatment was finished, cells were lysed by RIPA buffer solution and incubated overnight at 37◦ C. The following day the and the protein concentration was determined with the Pierce change in color was observed and the fluorescence intensity was BCA Protein Assay kit (Thermo Fisher Scientific, 23225). Equal measured (λem 560, λex 590 nm, FLUOstar OPTIMA microplate amounts of protein (50 µg) for each sample were loaded onto reader; BMG LABTECH GmbH). The 50% growth inhibitory a 15% SDS–PAGE gel. After the electrophoresis, proteins were concentration (IC50 ) was determined. transferred to PVDF membrane. The membranes were blocked and incubated with the primary antibody of interest. The primary Tracking of RNases Internalization Into antibodies used were rabbit polyclonal anti-LC3 (1:1,000, Abcam, RAW264.7 Macrophages by Confocal ab48394) and chicken polyclonal anti-GAPDH (1:2,000, Abcam, Microscopy ab9483). After washing, a horseradish peroxidase-conjugated RNase3, RNase6, and RNase7 were labeled with Alexa Fluor 488 antibody (Goat Anti-Rabbit IgG antibody, Sigma Aldrich, 12-348 Labeling kit (Invitrogen, A10235), following the manufacturer s ′ and Goat anti-chicken IgY H&L, Abcam, ab6877) was applied instruction as previously described (21). To 0.5 mL of 2 mg/mL for detection using an enhanced chemiluminescent detection protein solution in PBS, 50 µL of 1 M sodium bicarbonate, system (Supersignal West Pico Chemiluminescent Substrate, pH 8.3, was added. The protein was incubated for 1 h at ThermoFisher Scientific, 32209). Densitometry analysis was room temperature with the reactive dye, with stirring, and performed using Quantity One software. the labeled protein was separated from the free dye by PD- 10 desalting column (GE Healthcare, 17-0851-01). Labeled Immunofluorescence Microscopy protein distribution in cell cultures was followed by confocal RAW264.7 cells were grown, plated, and infected as above on microscopy. About 2.5 × 105 RAW cells were harvested in glass coverslip (5 × 105 cells each one). After fixation and 3 cm diameter microscopy plates (MatTek, P35G-1.5-14-C) 2– blocking, cells were incubated with the primary antibodies anti- 3 h before the assay. Macrophages were washed with RPMI and LC3 (1:200, Abcam, ab48394); the cultures were incubated for labeled with Hoescht 33342 (Thermo Fisher Scientific, 62249) 1 h at room temperature with labeled secondary antibodies (1: and Cell Mask Deep Red Plasma membrane Stain (Thermo Fisher 500, anti-rabbit IgG-Alexa Fluor 488, ThermoFisher Scientific, Scientific, C10046) at 0.5 µg/mL for 5–10 min before observation A-11008) and then washed 3 times with PBS. Cells were in Leica TCS SP5 AOBS equipped with a PL APO 63 × 1.4– incubated with Hoechst (ThermoFisher Scientific, 62249) and 0.6 CS oil immersion objective (Leica Microsystems, Mannheim, Cell Mask Deep Red (Invitrogen, C10046) to visualize the Germany). Following, Alexa Fluor labeled proteins were added at nuclei and membrane, and then were loaded with mounting 2 µM to the cultures and time lapse was recorded at intervals of media and sealed with nail oil. Images were captured by Leica 30 s for 30 min. Fluorochromes were excited by 405 nm (Hoechst TCS SP5 AOBS microscope equipped with a PL APO 63 33342), 649 nm (CellMask Deep Red), and 488 nm (Alexa Fluor × 1.4–0.6 CS oil immersion objective (Leica Microsystems, 488). Emissions were collected with a HyD detector. Mannheim, Germany). Quantification of fluorescence intensity was performed by using Imaris software. Real-Time qPCR Assays Alternatively, to quantify LC3 accumulation RAW264.7 cells Total RNA was extracted using mirVanaTM miRNA Isolation Kit were plated on glass coverslip (5 × 105 cells each one) and as described by the manufacturer (Ambion, Life Technologies, treated with RNase3 and/or Bafilomycin A. After treatment, AM1560) at each time point (4, 24, 48, 72 h). Total RNA was cells were fixed with 2% paraformaldehyde, permeabilized quantified by NanoDropTM spectrophotometer (Thermo Fisher with 0.05% triton-X100, and blocked with 5% non-fat milk. Scientific, Wilmington, DE USA), and cDNA was synthesized Cells were incubated with the primary antibodies anti- using iScriptTM cDNA Synthesis Kit (Bio-Rad, 170-8891). LC3 at 4◦ C overnight (1:200, Abcam, ab48394), and room The unique ID of the primers used (synthesized by Bio-Rad, temperature 2 h with secondary anti-rabbit IgG-Alexa 488 Hercules, CA, USA) is shown in Table S1. Transcripts of human antibody(ThermoFisher Scientific, A-11008) and then washed 3 RNase2, RNase3, RNase6, and RNase7 genes relative to the times with PBS. Cells were stained with DAPI (ThermoFisher human housekeeping GAPDH, and mouse BECN1 and ATG5 Scientific, 62248) to visualize the nuclei for 30 min at room genes relative to mouse housekeeping β-actin were measured temperature and then were loaded with mounting media and in triplicate from cDNA samples by real-time quantitative sealed with nail oil. Images were captured by Leica DM RB PCR using CFX96 Real-Time PCR detection system (Bio-Rad, equipped with a Leica DFC 500 camera under 100-fold oil Hercules, CA, USA). The results were analyzed by using the objective. LC3 puncta vesicle quantification was performed relative standard method (29). using Image J. Western-Blot Analysis Transmission Electron Microscopy Expression level of the autophagy marker LC3 in RAW264.7 For transmission electron microscopy (TEM) analysis, 106 macrophages was evaluated by western blot analysis. Autophagy infected RAW264.7 macrophages cells were seeded in 6-well was inhibited by addition of either 100 nM of Bafilomycin tissue culture plates and were treated with 10 µM RNase3 or A (BA, Sigma-Aldrich, 19-148) or 5 mM of 3 Methyladenine RNase3-H15A for 24 h. After treatment, macrophages were fixed (3MA, Sigma-Aldrich, M9281) (30, 31). Rapamycin (Sigma- with 2% (w/v) paraformaldehyde and 2.5% (v/v) glutaraldehyde Aldrich, R0395) at 100 nM was used as a positive control. When in PBS for 1 h. The samples were then dehydrated in acetone (50, Frontiers in Immunology | www.frontiersin.org 4 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities 70, 90, 95, and 100%) and were immersed in Epon resin (32). The and treated with RNases for 24 h. Rapamycin (100 nM) was ultrathin sections were examined in a JEOL JEM 2011 instrument used as a positive control and BA (100 nM) was used as an (JEOL, Ltd., Tokyo, Japan). inhibitor of acidification of vesicular content. At the end of the treatments, cells were washed with pre-warmed PBS and Acidic Vesicular Compartment stained with 5 µg/mL AO for 10 min in PBS. After washing Quantification by the Acridine Orange with PBS three times, the increase in the amount of the Assay acidic vesicular compartment was quantified by measuring the To quantify the volume/amount of the acidic vesicular red/green fluorescence intensity ratio of AO staining (AO green compartment, RAW264.7 macrophage cells were treated fluorescence λexc 485 and λem 535 nm; AO red fluorescence with the proteins and were stained with acridine orange (AO, λexc 430 and λem 590 nm). Values were normalized by cell ThermoFisher Scientific, A1301) as previously described (33). proliferation count by the Thiazolyl Blue Tetrazolium Bromide Briefly, cells were seeded in 96-well plates (104 cells/well) (MTT, Sigma-Aldrich, M5655-100MG) assay. FIGURE 1 | A representative SPOTi image comparing human RNases activity on Mycobacterium aurum infected macrophages. RAW 264.7 macrophages were infected with M. aurum at 10:1 MOI for 3 h at 37◦ C. The culture was washed with RPMI-1640 thrice and incubated with different concentrations of the proteins in RPMI-1640 complete medium for 4 h (Left) or 24 h (Right). Macrophages were washed twice with RPMI-1640 and lysed. Then, an aliquot was spotted onto wells of a 24-well plate containing MB7H10/OADC/agar and incubated at 37◦ C for 4–5 days to determine intracellular survival. TABLE 1 | Comparison of the antimycobacterial activity of RNase3, RNase6, and RNase7 and their respective active center mutants (H15A). Protein Extracellular MBC (µM) MAC (µM) Intracellular IC50 MIC (µM) (µM) M. aurum M. smegmatis M. bovis BCG M. aurum M. smegmatis M. bovis M. aurum RAW264.7 mc2 155 mc2 155 BCG RNase3 18.75 ± 0.05 9.37 ± 0.05 18.75 ± 0.05 2.34 ± 0.5 1.17 ± 0.3 4.68 ± 1.0 5 ± 0.05 > 25 RNase3-H15A 18.75 ± 0.05 18.75 ± 0.05 18.75 ± 0.05 4.68 ± 0.5 4.68 ± 1.0 4.68 ± 1.0 10 ± 0.05 > 25 RNase6 18.75 ± 0.05 18.75 ± 0.05 18.75 ± 0.05 4.68 ± 1.0 2.34 ± 0.5 4.68 ± 1.0 10 ± 0.05 > 25 RNase6-H15A 18.75 ± 0.05 18.75 ± 0.05 37.5 ± 0.05 4.68 ± 1.0 2.34 ± 0.5 4.68 ± 1.0 10 ± 0.05 > 25 RNase7 18.75 ± 0.05 18.75 ± 0.05 18.75 ± 0.05 > 10 > 10 > 25 10 ± 0.05 > 25 RNase7-H15A 18.75 ± 0.05 18.75 ± 0.05 18.75 ± 0.05 > 10 > 10 > 25 10 ± 0.05 > 25 Extracellular minimum bactericidal concentration (MBC100 ) and intracellular minimum inhibitory concentration (MIC100 ) were calculated at 72 h incubation time using the SPOTi assay. Minimum agglutination concentration (MAC) and Inhibitory Concentration (IC50 ) for RAW 264.7 mouse macrophages were assayed as described in the methodology. Results are an average of three independent repeated experiments. Mean average values ± SD are indicated. Frontiers in Immunology | www.frontiersin.org 5 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities Liposome Aggregation Assay superfamily using the SPOTi surrogate model adapted for the Large unilamellar vesicles (LUVs) were prepared as screening of novel anti-TB drug candidates (8). Previous work previously described (20). Briefly, 1,2-dioleoyl-sn-glycero- in our laboratory selected M. aurum as a fast non-pathogenic 3-phosphocholine (DOPC, Avanti, 850375) and 1,2-dioleoyl-sn- species suitable for the screening of anti-tubercular drugs at the glycero-3-phospho-(1’-rac-glycerol) (DOPG, Avanti, 840445P) extracellular (directly on mycobacterial culture) and intracellular were mixed in chloroform with 3:2 molar ratio and used (on mycobacteria residing inside infected macrophage cell-line) to prepare LUVs by Rotavapor. Then the liposomes were levels, referred in this article as in vitro and ex vivo assays, resuspended with 10 mM Tris/HCl, pH7.4 and generated respectively. Screening results indicated that RNase3, RNase6 a defined size (100 nm) by extrusion through 100 nm and RNase7 can completely inhibit the growth of M. aurum polycarbonate membranes. cultures at a low micromolar concentration range, while the Aggregation of liposomes was monitored by recording rest of the RNases did not reduce the mycobacteria population the scattering intensity with the excitation and emission even at the highest concentration tested (Figure S1). Next, wavelengths at 470 nm using Cary Eclypse spectrofluorimeter. we assayed the antimicrobial activity of the seven human Liposomes were incubated with proteins for indicated RNaseA protein family members against M. aurum within RAW time at room temperature, the signal was read 90 264.7 mouse macrophages. Results from the screening of the degree from the excitation beam with slits at 2.5 seven human secreted RNases confirmed that only RNase3, and 5 nm. RNase6 and RNase7 are active against mycobacterial infection (Figure 1). Interestingly, when comparing the reduction of Statistical Analysis CFU counts in comparison to control untreated samples Data are presented as mean ± SD and comparisons between for in vitro and ex vivo conditions we observed that the groups were analyzed by paired Student t-test and one-way effectiveness of the protein was higher at the intracellular level ANOVA for comparing more than 2 groups. p < 0.05 was (ex vivo assay) (Table 1). Besides, comparison of the macrophage considered as statistically significant. intracellular antimycobacterial activity at 4 and 24 h indicated that the protein activity is enhanced at longer incubation RESULTS times. Total inhibition of intracellular M. aurum growth was achieved at about 2-fold lower protein concentration at 24 h Screening of Human Canonical RNases incubation time in comparison to 4 h (Figure 1). Monitoring Identifies Three Members With of the protein activity within infected macrophages up to Antimycobacterial Activity at Extracellular 72 h also revealed an increase of the antimicrobial efficiency and Intracellular Levels at longer exposure times. The results suggested that other In this work we have evaluated for the first time the antimicrobial mechanisms, together with a direct killing of the bacilli, are potency of the seven human RNases from the RNase A taking place. FIGURE 2 | Tracking the translocation of Alexa Fluor 488-labeled RNase3 into RAW 264.7 mouse macrophages. (A) macrophages cells were stained with Hoechst and Deep Red to visualize the nuclei (blue) (B) and membranes (red) (C), respectively. (D) The protein location was registered after 30 min of macrophages post-treatment with Alexa Fluor 488-RNase3 (green). A panel showing merged images is shown in (E). Protein tracking assays were performed using 2.5 × 105 cells/mL and adding 2 µM of labeled protein final concentration. The images were taken using a Leica TCS SP5 AOBS microscope (see Materials and Methods section for more details). Frontiers in Immunology | www.frontiersin.org 6 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities The Enzymatic Activity Is Not Involved in species was chosen for macrophage intracellular studies, being the RNases Antimycobacterial Activity among the three tested species the only one that provides a fast Following, RNase3, RNase6, and RNase7 were chosen to further growing profile together with an efficient macrophage uptake investigate whether the enzymatic activity contributes to the (8). Overall, no significant differences were observed between antimycobacterial properties of human RNases by comparing the the native proteins and their respective active site mutants recombinant native proteins with their catalytic inactive mutants. against M. aurum, both at the extracellular and intracellular Previous work from our laboratory confirmed that the H15A levels. Therefore, we can conclude that the three RNases can substitution fully removed the catalytic activity without altering eradicate M. aurum dwelling within the macrophage at a low the protein 3D-structure (28). First, the three antimicrobial micromolar concentration range. To note, RNase3 stands out proteins (RNases 3, 6, and 7) together with their respective active as the most active, achieving mycobacterial growth inhibition at site mutants (H15A) were tested in vitro against M. smegmatis 5 µM (Table 1). mc2 155, M. aurum and M. bovis BCG. As summarized in Table 1, Next, we tested the agglutinating activity of the RNases against RNase3, RNase6, and RNase7 showed similar antimicrobial the three assayed mycobacterial species. The agglutinating activities against the three mycobacteria species. Then, M. aurum activity was calculated as the minimum agglutinating concentration (MAC). Results indicated that RNase3 and RNase6 can agglutinate mycobacteria within the same concentration range, while RNase7 is not active even at the maximum concentration tested. Besides, there were no significant differences among the MAC values of the wild type and active site mutants (Table 1). From our findings, we can infer that catalytic activity of the ribonucleases is not involved in their antimycobacterial mode of action. Tracking the Protein Internalization Within Macrophages Following, the RNases internalization within macrophages was visualized by confocal microscopy. Results confirmed that the three RNases are able to internalize within the mouse macrophages through a vesicle-mediated mechanism (Figure 2). Analysis of the profiles of selected regions of interest confirmed the protein confined location within macrophage vesicles (Figures S2, S3). In addition, time-lapse assays indicated that internalization is a fast process, where most of the protein enters into the macrophage within the first 30 min. Examples of internalization fluorescence profiles of labeled RNase3, RNase6, and RNase7 were recorded at several time intervals (Figure S4). Confocal microscopy images confirmed that no damage to the macrophage cells at the assayed conditions was taking place. Besides, the protein toxicity on the mouse macrophages in the assayed conditions was also discarded by using the resazurin assay; no significant cytotoxicity was observed at the maximum concentration tested (25 µM) (Table 1). M. aurum Infection Modulates the Expression of RNases in THP1 Macrophage Differentiated Cells Encouraged by positive results on the antimycobacterial activity of recombinant RNases we decided to test whether the FIGURE 3 | Human RNases 3 and 6 induce macrophage autophagy. RAW mycobacteria infection can induce the protein expression 264.7 macrophages were infected with M. aurum and treated with 10 µM of in human macrophages. We selected the human monocyte proteins (RNase1 to RNase7) for 24 h. Real-time qPCR measured the relative expression of BECN1 (A) and ATG5 (B) genes normalized by housekeeping THP1 cell line differentiated to macrophages and infected it gene β-actin. Results are shown from 3 independent experiments (mean ± with M. aurum. Levels of expression of human RNases were SD); (C) Immunoblot of LC3-processing and the housekeeping GAPDH evaluated by real time qPCR. Primers of the four known control. Results are shown from 3 independent experiments (mean ± SD). antimicrobial RNases expressed in blood cell type were used *Statistically significant differences compared with the control sample are indicated (p value < 0.05). to quantify their transcription levels (Table S1). Transcription levels were quantified in relation to the GAPDH housekeeping Frontiers in Immunology | www.frontiersin.org 7 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities gene. Results indicated that both control macrophage derived THP1 cells and M. aurum infected cells expressed the tested RNase genes (Figure S5). However, the protein transcription rates differed significantly among the RNases. Results of non- infected THP1 cells were in agreement with previous expression profile reports showing the predominant transcription of RNase2 (18). Comparison of the relative expression levels highlighted also the significant quantities of RNase6 followed by RNase3. On the contrary, RNase7 expression was very scarce, more than 1,000-fold lower than RNase2. Interestingly, M. aurum infection modulated the expression of RNase3 and RNase6, while RNase2 transcription levels remained unaffected (Figure S5). In addition, analysis of the expression profiles as a function of time also highlighted significant differences between early and late infection periods. Overall, M. aurum infection significantly inhibited the THP1 cells expression of RNases at the beginning of the infection (24 h), but triggered their expression at longer infection periods (48–72 h). RNases 3 and 6 Induce Autophagy in Macrophages Autophagy is one of the main mechanisms activated by infected macrophages to remove intracellular resident mycobacteria (34). Considering that some AMPs are known to participate in the host immune response against mycobacteria by the activation of autophagy (12), we decided to explore the action of human RNases on infected macrophages. RAW 264.7 mouse macrophages were chosen as a stable well-characterized cell line for the analysis of the recombinant proteins. First, we tested whether the human RNases could induce the expression of the autophagy markers BECN1 and ATG5, two essential genes involved in autophagosome formation and maturation, respectively. Analysis of qPCR results indicated that RNase3 and RNase6, out of the seven screened RNases, significantly upregulated the mRNA expression of BECN1 and ATG5 in RAW 264.7 macrophages (Figures 3A,B). These positive results were corroborated by analysis of LC3 processing and detection of the LC3II form, involved in the autophagosome formation (35). LC3 transformation was also evaluated by immunoblotting, FIGURE 4 | Both RNase3 and the RNase-H15A active site mutant induce confirming an increase in the LC3II fraction in detriment of LC3I autophagy in macrophages. RAW 264.7 mouse macrophages were infected in RNase3 and RNase6 treated samples in comparison to control with M. aurum and treated with 10 µM of protein (RNase3 and RNase3-H15A). (Figure 3C). Therefore, we conclude that both RNases activate an Real-time qPCR measured the relative expression of BECN1 (A) and ATG5 (B) genes normalized by the housekeeping gene β-actin. Results are shown from autophagy process in the treated macrophages. 3 independent experiments (mean ± SD); (C) Immunoblot of LC3-processing and the housekeeping GAPDH control. Results are shown from 3 independent experiments (mean ± SD). *Statistically significant differences compared with RNase Catalytic Activity Is Not Required the control sample are indicated (p value < 0.05). for the Protein Autophagy Activation We then analyzed in more details the protein activation of the autophagy pathway by using as a reference RNase3, In addition, LC3 transformation was monitored by the most active family member. First, the RNase3 wild-type immunofluorescence microscopy after treatment of M. aurum ability to induce autophagy was compared with the RNase3- infected RAW macrophage cells with either RNase3, RNase3- H15A enzymatically inactive variant by following the expression H15A, or rapamycin, used as a positive control of autophagy of BECN1 and ATG5 and monitoring the LC3II/LC3I ratio induction, observing in all cases a significant increase of LC3 (Figure 4). Equivalent activity of both wild-type and H15A intensity (Figures 5A,C). Complementarily, we applied TEM mutant proteins discards any contribution of the enzyme to visualize the ultrastructure of mouse macrophages infected catalytic activity in the autophagy pathway activation. with M. aurum and treated with RNase3 and RNase3-H15A. Frontiers in Immunology | www.frontiersin.org 8 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities FIGURE 5 | RNase3 and RNase3-H15A induce autophagic vacuole formation and LC3 processing. (A) Immunofluorescence microscopy analysis of LC3 processing, Mouse macrophages were infected with M. aurum and treated with 10 µM of RNase3, or 10 µM RNase3-H15A mutant, or 100 nM Rapamycin for 24 h. After treatment, the cells were fixed and stained with Hoechst to visualize the nuclei (blue), and with anti-LC3 followed by the addition of Alexa Fluor 488-conjugated anti-rabbit IgG (green). One representative immunofluorescence image out of 3 independent replicates are shown; scale bars: 10 µm; (B) Representative transmission electron microscopy images of M. aurum infected RAW264.7 cells treated with 10 µM RNase3 or 10 µM RNase3-H15A. Autophagic vacuoles are indicated by white arrow; (C) Quantitative analysis of LC3 intensity normalized with cell volume from 3 replicates. Results are shown from 3 independent experiments (mean ± SD). *Statistically significant differences compared with the control sample are indicated (p value ≤ 0.05). (D) total number of autophagic vacuoles per cell for 30 cells randomly selected per each treated sample. Autophagic vacuoles were defined as double-membrane vacuolar structures containing recognizable cytoplasmic contents. *Statistically significant differences compared with the control sample are indicated (p-value ≤ 0.05). Frontiers in Immunology | www.frontiersin.org 9 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities Inspection of electron micrographs revealed a higher number of double-membrane vacuoles in both RNase3 and RNase3- H15A treated macrophages in comparison with the control group (Figures 5B,D). Thus, the enzymatic activity is not been involved in the RNases autophagy induction process. The Protein Autophagy Induction Correlates With Vesicles Agglutination Activity Accumulation of double-membrane vacuoles, associated to the autophagosome formation, constitutes a characteristic trait of the autophagy pathway (36). Together with the initial autophagic vacuoles we could also identify abundant single-membrane degradative vacuoles, that could be ascribed to autolysosomes and are characteristic of the later steps of the autophagy process (Figure 5B). Moreover, macrophage incubation with RNase3 presented a dose dependent increase in the total volume of acidic vesicular compartments, as quantified by registering the AO acidotropic dye signal, a trait associated to autolysosome formation (Figure S6). In contrast, no significant change in AO signal was recorded for the none-autophagic RNase7, used here as a negative control. Rapamycin was used as a positive control and addition of BA, an inhibitor of the vacuolar H+ -ATPase, resulted in a significant reduction of the AO signal. To note, the two RNases that activate autophagy, RNase3 and RNase6, can agglutinate liposome vesicles, in contrast to RNase7 (Figure S7). Likewise, RNases 3 and 6, but not RNase7, have mycobacteria cell agglutination activity (Table 1). The results suggest that the protein agglutination activity might be involved in the autophagy induction process. Induction of Macrophage Autophagy Contributes to RNase Antimycobacterial Activity Next, the induction of the autophagy process was confirmed FIGURE 6 | RNase3 and RNase3-H15A induce autophagy in control and M. for wild-type and mutant proteins in both infected a non- aurum infected macrophages. Mouse macrophages were uninfected or infected macrophages. The levels of the autophagy markers infected with M. aurum and treated with 10 µM of protein (RNase3, RNase3-H15A) or 100 nM Rapamycin for 24 h. Real-time qPCR measured the in RNase3 treated samples were also compared to rapamycin, relative expression of (A) BECN1 and (B) ATG5 genes normalized by the observing in all cases comparable levels of BECN1 and ATG5 housekeeping gene β-actin. Results are shown from 3 independent expression and LC3II/LC3I ratio (Figure 6). Induction of experiments (mean ± SD); (C) Immunoblot of LC3-processing and the autophagy by RNase3 was reproduced on both control and housekeeping GAPDH control. Results are shown from 3 independent infected macrophages (Figure 6). Results also corroborated that experiments (mean ± SD). *Statistically significant differences compared with the control sample are indicated (p-value < 0.05). there were no significant differences between the effect induced by wild-type RNase3 and the H15A mutant variant on both infected and non-infected macrophage populations. Following, we evaluated whether the RNase induction of blocked the RNase3 induction of autophagy, no reduction in the autophagy mediates the protein macrophage intracellular killing expression levels of the autophagy markers were observed in the of mycobacteria. First, the levels of autophagy markers samples treated with BA (Figures 7A–C). In fact, quantification (BECN1, ATG5, and LC3-II) were quantified following of LC3 accumulation by immunofluorescence microscopy RNase3 treatment in the absence and presence of autophagy revealed a significant increase of LC3 puncta per cell when inhibitors (3methyladenine and bafilomycin A1). 3MA is macrophages were incubated with RNase3 in the presence an early inhibitor of the autophagy pathway and works by of BA (Figure 8). blocking the autophagosome formation, whether BA arrest More interestingly, both autophagy inhibitors reduced the autophagic process in a later step by impeding the significantly the protein ability to kill mycobacteria within autophagosome maturation and subsequent fusion with the macrophages (Figure 7D, Figure S8). The fact that 3MA lysosome compartment. Accordingly, while the presence of 3MA inhibits RNase3 induction of autophagy, but not BA treatment, Frontiers in Immunology | www.frontiersin.org 10 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities FIGURE 7 | Autophagy induction by RNase3 contributes to its antimycobacterial activity against M. aurum. Mouse macrophages were infected with M. aurum and treated with 5 µM of RNase3 and/or 100 nM of rapamycin, and/or 100 nM of bafilomycin A1, and/or 5 mM of 3 methyladenine for 24 h. Real-time qPCR measured the relative expression of BECN1 (A) and ATG5 (B) genes normalized by the housekeeping gene β-actin. Results are shown from 3 independent experiments (mean ± SD); (C) A representative immunoblot of LC3 and housekeeping GAPDH control; (D) Intracellular mycobacterial viability was determined based on the number of CFUs. Relative percentages to control samples are indicated. Results are shown from 3 independent experiments (mean ± SD). *Significant differences between pairs is indicated (p value < 0.05). suggests that the protein triggers the onset of the autophagy top priority of global health action programs (1). Within this process. Moreover, BA reduced the protein ability to kill context, the search for novel anti-TB agents is a priority. In the intracellular mycobacteria, indicating that the RNase anti- this work, we have taken profit of our developed screening infective activity is partly dependent on the maturation of the platform adapted for testing the drug efficacy within infected autophagosome. Likewise, rapamycin co-treatment with RNase3 macrophage cells. The SPOTi high-throughput screening (HTS) enhanced the protein antimicrobial effectivity (Figure 7D). An assay system uses a non-pathogenic fast-growing mycobacterial equivalent pattern was reproduced by the RNase3-H15A mutant species (M. aurum) able to survive within RAW 264.7 mouse (Figure S8). Therefore, the results confirm that the induction of macrophages, as a surrogate model for host TB infection. autophagy participates in the RNase3 antimycobacterial activity M. aurum is an appropriate substitute for M. tuberculosis among within infected macrophages. other non-pathogen mycobacteria candidates sharing a similar cell-wall structure, gene organization, and drug resistance pattern DISCUSSION (8, 26, 27). The SPOTi assay was applied to screen the seven main canonical human members of the RNase A superfamily. Previous studies have demonstrated that macrophages serve as a The human canonical RNases, although sharing a high sequence major protecting niche for dormant mycobacteria that can trigger identity and a conserved structural fold (Figure S9), differ a later reactivation from the disease latency. This scenario is considerably in their biological properties. Overall, RNases one of the main hindrances in developing effective eradication are secreted by different types of innate immune cells and of tuberculosis. Antimicrobial peptides participate in the host participate in a variety of endogenous processes involved in innate immunity against macrophage dwelling TB bacilli (12). the host defense (18, 28, 37, 38). One of the main ascribed Currently, the fight against emergent MDR-TB-forms is at the tasks for human secretory RNases is the safeguard of the body Frontiers in Immunology | www.frontiersin.org 11 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities FIGURE 8 | Comparative analysis by immunofluorescence microscopy of LC3 accumulation following RNase3 treatment in the absence and presence of Bafilomycin A. Mouse RAW264.7 cells were incubated with 10 µM RNase3 and/or 100 nM of Bafilomycin A1. Primary anti-LC3 antibody and Alexa-488 labeled secondary antibody were used to detect LC3 (green). Nuclei were labeled with DAPI (blue). Following immunostaining, 100 cells from each treatment were analyzed and the number of LC3 puncta per cell was calculated by Image J. *Statistically significant differences are indicated, p < 0.05. fluid sterility. A diversity of anti-infective activities have been that out of the seven tested human RNases only three were reported for some family members (18, 39–42). Among them, active against both extracellular and intracellular macrophages- the protein mechanism of action on bacterial cells is probably dwelling mycobacteria (Figures 1, 2). A high antimycobacterial one of the best studied processes (37, 38, 43). In particular, activity was observed for RNases 3, 6 and 7, while RNases 1, 2, previous works identified the involvement of RNases 3 and 7 in 4 and 5 were found inactive even at the highest concentration mycobacterial infection. Expression of RNase3 together with α- tested. The three active RNases (RNases 3, 6, and 7) correspond defensin in M. bovis BCG infection was first reported by Capron to the family members with previously reported antimicrobial and coworkers (14). Interestingly, the researchers observed the activity against Gram-negative and Gram-positive species (37, 38, specific recruitment and induction of eosinophils by lipomannan, 41). The present data corroborate that RNase3 is the most active a unique mycobacterial cell-wall component. Very recently, among the family members. Rivas-Santiago and coworkers reported the overexpression of Following, the three active RNases (RNases 3, 6, and RNase7 in airway epithelial cells infected by M. tuberculosis 7) were selected for further analysis. First, the minimum (24). In our laboratory we have confirmed the antimycobacterial bactericidal activity was determined in vitro against three distinct activity of recombinant RNase3 and RNase7 on M. vaccae Mycobacterium species: M. smegmatis mc2 155, M. aurum and M. in vitro and characterized the protein mechanism of action at the bovis BCG (Table 1). In addition, the protein cell agglutination bacterial cell-wall (25). activity was tested against the three assayed mycobacteria In this work, we have screened the antimycobacterial activity species (M. smegmatis mc2 155, M. aurum, and M. bovis BCG). of the seven main canonical RNases. Taking advantage of Interestingly, RNases 3 and 6 were able to induce the aggregation the SPOTi-HTS, the protein activities were simultaneously of all the tested mycobacteria, but not RNase7 (Table 1). The assayed in vitro and ex vivo (8, 26, 27). Results indicated results are in accordance with our previously published data Frontiers in Immunology | www.frontiersin.org 12 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities using M. vaccae cultures, where RNase3, but not RNase7, viral or parasite infection together with inflammation, sepsis showed a significant mycobacterial cell agglutination activity and tissue damage (18, 42, 54). Intriguingly, the comparative (25). Previous works from our laboratory on Gram-negative expression pattern of the distinct RNaseA members reveals a species also revealed the agglutination ability of RNases 3 and clear specialization in regard to innate cell and infection types 6 in contrast with RNase7 (21, 38). Likewise, RNases 3 and 6 (18, 55). In this context, RNase7 is mainly secreted by epithelial can agglutinate phospholipid vesicles but not RNase7 (Figure S7) tissues and conforms a protection barrier against bacterial (20, 38). Positive agglutination was correlated with the presence intruders, whereas RNases 3 and 6 are abundant in blood cell of an aggregation-prone hydrophobic patch within the primary types. In particular RNase 3 is highly expressed in eosinophils sequences of RNases 3 and 6 (37, 44–46). Comparative scanning while RNase6 predominates in neutrophils and monocytes of the three RNases profiles using the AGGRESCAN 3D software (18). In addition, free secretory granules from eosinophils and (47) also highlighted the absence of an aggregation patch neutrophils can be engulfed by macrophage cells and release their at the RNase7 structure surface (Figure S9). On the other content within (18, 56, 57). hand, no significant differences were observed between the Our present results using murine RAW macrophages infected wild-type and H15A mutants of the three RNases showing with M. aurum highlighted that the three antimicrobial RNases antimycobacterial activity (Table 1). The present results on can effectively eradicate the macrophage intracellular resident mycobacteria confirm previous studies on Gram-negative and mycobacteria at a 5 to 10 µM range (Table 1). The fact that the Gram-positive bacterial species that indicated that the RNases RNases could more easily eradicate the mycobacteria resident catalytic activity did not participate in their antimicrobial action within macrophages than free in extracellular cultures suggested (37, 38, 48–51). us that the proteins might combine a direct bacterial cell killing To evaluate the potential translational significance of our action with some other biological properties. results in an in vivo scenario, we analyzed the expression levels Immunomodulatory activities mediate the antimycobacterial of the RNases in human macrophage THP1 cells following action of many AMPs (12, 58–60). One of the main strategies M. aurum infection. To this end, we synthesized primers for undertaken by AMPs to control the proliferation of macrophage the four human family members with described anti-pathogen intracellular resident bacteria relies on the triggering of the activities and previously reported to be expressed in granulocytes autophagy pathway (12, 30, 34, 61–63). Selective autophagy (18). Our data highlighted that the human monocyte-derived targeting invading pathogens contributes to maintain the host THP1 macrophages primarily expresses RNase2 and secondarily tissue homeostasis (64). On its turn, the Mtb bacilli ensure RNase3 and RNase6, while a very scarce expression of RNase7 their intracellular survival within macrophages by arresting is detected (Figure S5). This expression pattern corroborates the autophagosome maturation (65, 66). Therefore, autophagy previous studies that highlighted a predominant expression of inducers are among the favorite drug candidates against resistant RNase2 in human monocytes. On the other hand, RNase7 TB strains (12, 34, 58, 59, 62, 67). Interestingly, out of was reported to be abundantly expressed in epithelial cells the seven screened RNases, RNase3, and RNase6 significantly (52) but poorly in macrophages (23). In addition, Becknell increased BECN1 and ATG5 gene expression and LC3 processing and co-workers reported a regulation of RNase6 expression was observed (Figure 3). Subsequent analysis using RNase3 in macrophages in response to uropathogenic E. coli infection as a reference indicated that the autophagy induction process (23). The present data indicates that M. aurum infection is is unrelated to the protein catalytic activity (Figures 4, 5). only altering the expression profile of cultured macrophages for Moreover, the autophagy induction was reproduced in both RNase3 and RNase6 genes. However, further work would be infected and non-infected macrophages (Figure 6). needed to characterize the RNases expression pattern in primary The contribution of autophagy induction on the RNase3 macrophages. The results are in accordance with the previously antimycobacterial activity was corroborated by evaluating the observed anti-pathogen properties of the three detected RNases. protein effect on M. aurum growth upon blockage of the RNase2 has a high antiviral activity but no antibacterial activity autophagy pathway. The blockage of the autophagosome in vitro, and is regulated during viral infection (40). On their formation or maturation was induced by either addition of turn, RNases 3 and 6 have a high bactericidal activity and 3MA, an inhibitor of the PI3K class III kinase (68), or BA, an their expression is regulated by bacterial infections (18, 23, ATPase inhibitor that prevents the lysosome fusion with the 42). The present time-course expression profiles in M. aurum autophagic vesicles (69). When the infected macrophages were infected THP1 macrophages indicates a downregulation of treated with RNase3 in the presence of either 3MA or BA we RNase3 and RNase6 after a short infection period followed by a observed a reduction in the protein inhibitory activity of M. significant upregulation at longer incubation times. Modulation aurum growth (Figure 7 and Figure S8). On the other hand, of antimicrobial RNases following M. aurum infection are in rapamycin enhanced the antimycobacterial activity of RNase3. agreement with previously reported results using M. tuberculosis The same pattern was reproduced when assaying the RNase3- (18, 24). In the literature there are other examples of AMPs H15A mutant action in the presence of the autophagy inhibitor downregulated upon mycobacteria infection, a process induced or activator (Figure S8). by the Mtb bacilli as a protection mechanism against the host The fact that BA does not inhibit the RNase induction of innate armory (12, 30). On its turn, upon infection, the host autophagy but reduces its antimycobacterial activity suggests immune cells activate the expression of antimicrobial proteins that the protein action is dependent on the late autophagosome and peptides (12, 53). Many stimuli have been reported to maturation process. Moreover, upon cell incubation with RNase3 regulate the expression of human RNases, such as bacterial, we observed an increase in the acidic vesicle content (Figure S6), Frontiers in Immunology | www.frontiersin.org 13 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities a process that can be associated to autolysosome formation. In summary, the present results have revealed the potentiality To note, other AMPs, such as azurodicin and calgranulin, of human secretory RNases to work as antimicrobial agents in activate autophagy by promoting the autophagosome fusion the fight against tuberculosis, extendable to other intracellular with lysosomes (12, 53, 70, 71). Overall, our results confirm infectious bacterial diseases that are difficult to eradicate (78). that autophagy contributes to the RNase antimycobacterial AMPs that display a multifaceted mechanism of action (12, mechanism of action. However, the fact that addition of the 16, 79–85), like human antimicrobial RNases, deserve a special autophagy inhibitors does not fully abolishes the antimicrobial attention for their pharmacological potential. Our ongoing inter- activity of the protein indicates that other biological properties disciplinary investigation warrants the understanding of RNases’ are also participating in its antimycobacterial activity. In antimicrobial mechanism of action and can pave the design of addition, considering that RNases 3 and 6 induce autophagy, novel anti-infective therapies. but not RNase7, and that no significant differences are observed between the wild type and the active site mutant proteins, we CONCLUSIONS can suspect that the protein aggregation-prone ability might be involved in the autophagy pathway. Indeed, our previous results Screening of the seven main human canonical members of the correlated the protein self-aggregation capacity with the proteins RNaseA superfamily identified three RNases (RNase3, RNase6, lipid vesicle agglutination and membrane disruption ability (20). and RNase7) that achieved the eradication of M. aurum infection We can speculate that the formation of protein aggregates could within macrophages at a low micromolar range. The protein promote the autophagosome formation and later fusion with antimicrobial activity was found not dependent on the RNase the lysosomal compartment. Complementarily, other protein enzymatic activity. Out of the three antimycobacterial RNases, properties, such as high cationicity, were considered to underlie two of them (RNase3 and RNase6) demonstrated their ability the autophagy process. The predicted isoelectric point (pI) for to induce macrophage autophagy and thereby inhibit M. aurum all tested seven RNases, ranging from 9 to 10.7 (Table S2), intracellular growth. The expression of both RNases in human emphasizes their prominent cationicity, a characteristic trait THP1 derived macrophage cells is regulated by the mycobacterial of AMPs (72–74). However, although RNase3 is the most infection, suggesting an in vivo physiological role. cationic of all the proteins, we cannot find a direct correlation between the pI and antimicrobial or autophagy induction abilities among the seven tested human RNases. Likewise, the protein DATA AVAILABILITY hydrophobicity was analyzed, obtaining very similar calculated The raw data supporting the conclusions of this manuscript will GRAVY values (Table S2). Further work is ongoing to unravel the be made available by the authors, without undue reservation, to structural determinants underlying the RNase antimycobacterial any qualified researcher. mechanism of action. Interestingly, two RNaseA family homologs (bovine seminal RNase and onconase) were recently reported to exert their anti- AUTHOR CONTRIBUTIONS tumoral activity by the selective induction of the autophagic cancer cell death (33, 75). Bovine seminal RNase (BS-RNase) EB and SB contributed in funding acquisition. DP, EB, and SB triggers an autophagic cell death process in pancreatic cancer conceived and designed the experiments. LL, GP-E, and JA-T cells mediated by BECN1 induction; and onconase, a RNaseA performed the experiments. EB, GP-E, DP, and SB supervised superfamily member proposed for chemotherapy, inhibits the the experimental work. LL, JA-T, EB, and SB contributed to the proliferation of cancer cells through the ROS/Akt/mTOR original draft preparation. LL, DP, EB, and SB participated in the pathway (33, 75). Surprisingly, the bovine pancreatic RNaseA, manuscript revision and final edition. displaying a high catalytic activity and sharing about a 82% sequence identity with BS-RNase, could not reproduce the FUNDING induction of autophagy. The authors associated the activation of the autophagic pathway to the protein cationicity and Experimental work was supported by the Ministerio de oligomer formation, discarding any correlation between the Economía y Competitividad (SAF2015-66007P to EB), AGAUR RNase catalytic activity and the autophagic process (75). Two (2016PROD00060) co-financed by FEDER funds and Fundació main factors, protein net charge, and stability of the oligomers La Marató de TV3 (ref. 20180310). were considered to explain the difference between RNaseA and BS-RNase in inducing autophagy. Last but not least, there is ACKNOWLEDGMENTS another human RNase, the RNase L, unrelated to the RNaseA superfamily, which can also trigger the autophagy pathway. The authors wish to thank the Laboratori d’Anàlisi i RNase L expression is induced during viral infection. The Fotodocumentació (UAB), where spectrofluorescence assays protein can initiate autophagy and suppress thereby the virus were performed. LL is a recipient of a CSC predoctoral replication (31, 76). In all cases, we can outline a convergent fellowship. JA-T and GP-E were recipients of a predoctoral mechanism where the RNases induction of autophagy is activated fellowship (Personal Investigador en Formación, Universitat by a cell stress condition: oxidative injury, presence of protein Autònoma de Barcelona). Authors thank Drs. Esther Julián, aggregation, or infection (18, 77). Estela Noguera-Ortega, Maria Goetz, José Miguel Lizcano, and Frontiers in Immunology | www.frontiersin.org 14 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities Pau Muñoz-Guardiola, UAB for their technical help. SB is a infected with M. aurum for 4, 24, 48, and 72 h. The transcriptional expression of Cipla Distinguished Fellow in Pharmaceutical Sciences. human RNase2, RNase3, RNase6, and RNase7 were detected by real-time qPCR and all samples were normalized with the GAPDH housekeeping gene. (A) relative expression of RNase2, RNase3, RNase6, and RNase7 at 0 h without infection; SUPPLEMENTARY MATERIAL (B–D) The monitoring of RNases’ expression upon infection was presented as the fold change of relative expression in infected group compared with the The Supplementary Material for this article can be found corresponding control group at each time point. Results are shown from 3 online at: https://www.frontiersin.org/articles/10.3389/fimmu. independent experiments (mean ± SD), ∗ indicates a significant difference compared with control (p value < 0.05). 2019.01500/full#supplementary-material Figure S6 | Quantification of acidic vesicular compartment measured by AO Figure S1 | A representative SPOTi image comparing human RNases activity on staining. The total amount of acidic vesicles was assessed as an estimate of M. aurum cultures. M. aurum was incubated with serially diluted recombinant autolysosome formation. Mouse RAW 264.7 macrophage cells were treated with human RNases (ranging from 2 to 32 µM) for 4 h in phosphate-buffered saline RNases 3 and 7 for 24 h. Values are presented as means ± SD of 8 replicates. (PBS). Next, an aliquot was spotted onto wells of a 24-well plate containing ∗ indicates significant difference compared with control group (p ≤ 0.05). MB7H10/OADC/agar and incubated at 37◦ C for 4–5 days to determine survival. Figure S7 | Comparison of aggregation of DOPG/DOPC liposomes by RNases. Figure S2 | Confocal microscopy analysis of RAW 264.7 cell culture (2.5 × 105 The incubation time of protein with liposomes is 30 min for all assay. The proteins cells/mL) incubated with 2 µM of RNase3 labeled with Alexa Fluor 488. Cells were were 2-fold serially diluted (from 1.6 to 0.05 µM). stained with Hoechst and Deep Red following the assay incubation conditions detailed in the experimental procedures section. After addition of Alexa Fluor 488 Figure S8 | RNase3 and RNase3-H15A antimycobacterial activity in the presence labeled protein (green), the evolution of the fluorescence signals was analyzed by of autophagy regulators. RAW 264.7 macrophages were infected with M. aurum confocal microscopy. A total of 20 cells were analyzed by regions of interest (ROIs) and treated with different concentration of the proteins, and/or 100 nM of using Leica TCS software. The images were taken using a Leica TCS SP5 rapamycin (Rapa), and/or 100 nM of bafilomycin A (BA), and/or 5 mM of 3 AOBS microscope. methyladenine (3MA) for 24 h in RPMI-1640 complete medium. Macrophages were washed twice with RPMI-1640 and lysed, and then an aliquot was spotted Figure S3 | 3D reconstruction of RAW 264.7 macrophages post-treatment with onto wells of a 24-well plate containing MB7H10/OADC/agar and incubated at RNase3. Confocal microscopy analysis of RAW 264.7 cell culture incubated with 37◦ C for 4–5 days to determine intracellular survival. 2 µM of RNase3 labeled with Alexa Fluor 488 (green). Cells were stained with Hoechst and Deep Red following the assay incubation conditions detailed in the Figure S9 | Sequence alignment and prediction of aggregation propensity of eight experimental procedures section. After 45 min of protein addition, the human RNases. (A) Alignment of the eight human canonical RNases using the fluorescence signals of Hoechst and AlexaFluor (A) and Hoechst, AlexaFluor and ESPript3 software (espript.ibcp.fr/), the N-terminal domain is highlighted in blue. Deep Red (B) were analyzed by confocal microscopy. The images were taken Sequence allocation was performed following the homology between RNases. using a Leica TCS SP5 AOBS microscope. (B) 3D structure of RNases by Aggrescan3D (A3D). A3D exploits an experimentally derived intrinsic aggregation propensity scale for natural amino Figure S4 | Confocal microscopy analysis of RAW 264.7 cell culture incubated acids. This structure-based approach identifies aggregation patches (in red) at the with human RNases labeled with Alexa Fluor 488. Cells were stained with Hoechst protein surface. (C) Aggregation profile propensity of RNase3, RNase6, and and Deep Red following the assay incubation conditions detailed in the RNase7 based on A3D score for protein residues. experimental procedures section. After labeled protein addition (green), the evolution of the fluorescence signals was analyzed by confocal microscopy for Table S1 | Primer sequences for real-time qPCR. Primers of human GAPDH, 30 min. A total of 20 cells were analyzed by regions of interest (ROIs) using Leica RNase2, RNase3, RNase6, and RNase7, and human and mouse β-actin, TCS software. The images were taken using a Leica TCS SP5 AOBS microscope. Beclin-1, and ATG5. Each panel indicates the RNase assayed. Table S2 | Comparison of physicochemical parameters of human secretory Figure S5 | Expression pattern of human RNases in THP-1 derived macrophage RNases. pI and Grand average of hydropathy (GRAVY) were predicted using the cells infected with M. aurum. Human THP-1 macrophage derived cells were ProtParam tool (https://web.expasy.org/translate/). REFERENCES 9. Pieters J. Mycobacterium tuberculosis and the macrophage: maintaining a balance. Cell Host Microbe. (2008) 3:399–407. 1. WHO. Global Tuberculosis Report 2018. Geneva (2018). doi: 10.1016/j.chom.2008.05.006 2. Accelerating efforts to end TB. Nat Microbiol. (2018) 3:257–7. 10. Welin A, Raffetseder J, Eklund D, Stendahl O, Lerm M. Importance doi: 10.1038/s41564-018-0126-6 of phagosomal functionality for growth restriction of Mycobacterium 3. Pai M, Schito M. Tuberculosis diagnostics in 2015: landscape, priorities, needs, tuberculosis in primary human macrophages. J Innate Immun. (2011) 3:508– and prospects. J Infect Dis. (2015) 211:S21–8. doi: 10.1093/infdis/jiu803 18. doi: 10.1159/000325297 4. Maitra A, Kamil TK, Shaik M, Danquah CA, Chrzastek A, Bhakta S. Early 11. Saiga H, Shimada Y, Takeda K. Innate immune effectors in mycobacterial diagnosis and effective treatment regimens are the keys to tackle antimicrobial infection. Clin Dev Immunol. (2011) 2011:1–8. doi: 10.1155/2011/347594 resistance in tuberculosis (TB): a report from Euroscicon’s international TB 12. Arranz-Trullén J, Lu L, Pulido D, Bhakta S, Boix E. Host antimicrobial Summit 2016. Virulence. (2016) 1–20. doi: 10.1080/21505594.2016.1256536 peptides: the promise of new treatment strategies against tuberculosis. Front 5. Maitra A, Danquah CA, Scotti F, Howard TK, Kamil TK, Bhakta S. Tackling Immunol. (2017) 8:1499. doi: 10.3389/fimmu.2017.01499 tuberculosis: Insights from an international TB Summit in London. Virulence. 13. Lasco TM, Turner OC, Cassone L, Sugawara I, Yamada H, McMurray DN, (2015) 6:561–72. doi: 10.1080/21505594.2015.1060396 et al. Rapid accumulation of eosinophils in lung lesions in guinea pigs 6. Eldholm V, Balloux F. Antimicrobial resistance in Mycobacterium infected with Mycobacterium tuberculosis. Infect Immun. (2004) 72:1147–9. tuberculosis: the odd one out. Trends Microbiol. (2016) 24:637–48. doi: 10.1128/IAI.72.2.1147-1149.2004 doi: 10.1016/j.tim.2016.03.007 14. Driss V, Legrand F, Hermann E, Loiseau S, Guerardel Y, Kremer L, et al. TLR2- 7. Siegrist MS, Bertozzi CR. Mycobacterial lipid logic. Cell Host Microbe. (2014) dependent eosinophil interactions with mycobacteria: role of alpha-defensins. 15:1–2. doi: 10.1016/j.chom.2013.12.005 Blood. (2009) 113:3235–44. doi: 10.1182/blood-2008-07-166595 8. Gupta A, Bhakta S. An integrated surrogate model for screening of drugs 15. Soehnlein O. Direct and alternative antimicrobial mechanisms of against Mycobacterium tuberculosis. J Antimicrob Chemother. (2012) 67:1380– neutrophil-derived granule proteins. J Mol Med. (2009) 87:1157–64. 91. doi: 10.1093/jac/dks056 doi: 10.1007/s00109-009-0508-6 Frontiers in Immunology | www.frontiersin.org 15 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities 16. Lázár V, Martins A, Spohn R, Daruka L, Grézal G, Fekete G, et al. Antibiotic- 35. Klionsky DJ. Guidelines for the use and interpretation of assays for resistant bacteria show widespread collateral sensitivity to antimicrobial monitoring autophagy (3rd edition). Autophagy. (2016) 12:1–222. peptides. Nat Microbiol. (2018) 3:718–31. doi: 10.1038/s41564-018-0164-0 doi: 10.1080/15548627.2015.1100356 17. Sorrentino S. The eight human “canonical” ribonucleases: molecular diversity, 36. Yuan K, Huang C, Fox J, Laturnus D, Carlson E, Zhang B, et al. Autophagy catalytic properties, and special biological actions of the enzyme proteins. plays an essential role in the clearance of Pseudomonas aeruginosa by alveolar FEBS Lett. (2010) 584:2194–200. doi: 10.1016/j.febslet.2010.04.018 macrophages. J Cell Sci. (2012) 125:507–15. doi: 10.1242/jcs.094573 18. Lu L, Li J, Moussaoui M, Boix E. Immune modulation by human 37. Boix E, Salazar VA, Torrent M, Pulido D, Nogués MV, Moussaoui M. secreted RNases at the extracellular space. Front Immunol. (2018) 9:1–20. Structural determinants of the eosinophil cationic protein antimicrobial doi: 10.3389/fimmu.2018.01012 activity. Biol Chem. (2012) 393:801–15. doi: 10.1515/hsz-2012-0160 19. Jordaan S, Akinrinmade O, Nachreiner T, Cremer C, Naran K, Chetty S, et al. 38. Pulido D, Arranz-Trullen J, Prats-Ejarque G, Velazquez D, Torrent Updates in the development of immunoRNases for the selective killing of M, Moussaoui M, et al. Insights into the antimicrobial mechanism tumor cells. Biomedicines. (2018) 6:28. doi: 10.3390/biomedicines6010028 of action of human RNase6: structural determinants for bacterial cell 20. Torrent M, Sánchez D, Buzón V, Nogués MV, Cladera J, Boix E. Comparison agglutination and membrane permeation. Int J Mol Sci. (2016) 17:552. of the membrane interaction mechanism of two antimicrobial RNases: doi: 10.3390/ijms17040552 RNase 3/ECP and RNase 7. Biochim Biophys Acta. (2009) 1788:1116–25. 39. Singh A, Batra JK. Role of unique basic residues in cytotoxic, antibacterial doi: 10.1016/j.bbamem.2009.01.013 and antiparasitic activities of human eosinophil cationic protein. Biol Chem. 21. Torrent M, Badia M, Moussaoui M, Sanchez D, Nogués MV, Boix E. (2011) 392:337–46. doi: 10.1515/bc.2011.037 Comparison of human RNase 3 and RNase 7 bactericidal action at the Gram- 40. Rosenberg HF. Eosinophil-derived neurotoxin (EDN/RNase 2) and the mouse negative and Gram-positive bacterial cell wall. FEBS J. (2010) 277:1713–25. eosinophil-associated RNases (mEars): expanding roles in promoting host doi: 10.1111/j.1742-4658.2010.07595.x defense. Int J Mol Sci. (2015) 16:15442–55. doi: 10.3390/ijms160715442 22. Pulido D, Moussaoui M, Andreu D, Nogués MV, Torrent M, Boix E. 41. Becknell B, Spencer JD. A Review of Ribonuclease 7’s structure, Antimicrobial action and cell agglutination by the eosinophil cationic protein regulation, and contributions to host defense. Int J Mol Sci. (2016) 17:423. are modulated by the cell wall lipopolysaccharide structure. Antimicrob Agents doi: 10.3390/ijms17030423 Chemother. (2012) 56:2378–85. doi: 10.1128/AAC.06107-11 42. Koczera P, Martin L, Marx G, Schuerholz T. The ribonuclease a superfamily 23. Becknell B, Eichler TE, Beceiro S, Li B, Easterling RS, Carpenter AR, et al. in humans: canonical RNases as the buttress of innate immunity. Int J Mol Sci. Ribonucleases 6 and 7 have antimicrobial function in the human and murine (2016) 17:E1278. doi: 10.3390/ijms17081278 urinary tract. Kidney Int. (2015) 87:151–61. doi: 10.1038/ki.2014.268 43. Torrent M, Pulido D, Nogués MV, Boix E. Exploring new biological functions 24. Torres-Juarez F, Touqui L, Leon-Contreras J, Rivas-Santiago C, Enciso- of amyloids: bacteria cell agglutination mediated by host protein aggregation. Moreno JA, Hernández-Pando R, et al. RNase 7 but not psoriasin nor sPLA2- PLoS Pathog. (2012) 8:2012–4. doi: 10.1371/journal.ppat.1003005 IIA associates with Mycobacterium tuberculosis during airway epithelial cell 44. Pulido D, Moussaoui M, Victòria Nogués M, Torrent M, Boix E. Towards infection. Pathog Dis. (2018) 76:fty005. doi: 10.1093/femspd/fty005 the rational design of antimicrobial proteins: single point mutations 25. Pulido D, Torrent M, Andreu D, Nogués MV, Boix E, Nogues MV, et al. can switch on bactericidal and agglutinating activities on the RNase Two human host defense ribonucleases against mycobacteria, the eosinophil A superfamily lineage. FEBS J. (2013) 280:5841–52. doi: 10.1111/febs. cationic protein (RNase 3) and RNase 7. Antimicrob Agents Chemother. (2013) 12506 57:3797–805. doi: 10.1128/AAC.00428-13 45. Pulido D, Prats-Ejarque G, Villalba C, Albacar M, Gonzalez-Lopez JJ, Torrent 26. Guzman JD, Evangelopoulos D, Gupta A, Birchall K, Mwaigwisya S, Saxty M, et al. A novel RNase 3/ECP peptide for Pseudomonas aeruginosa biofilm B, et al.. Antitubercular specific activity of ibuprofen and the other 2- eradication. Combining antimicrobial, lipopolysaccharide binding and cell arylpropanoic acids using the HT-SPOTi whole-cell phenotypic assay. BMJ agglutinating activities. Antimicrob Agents Chemother. (2016) 60:6313–25. Open. (2013) 3:1–13. doi: 10.1136/bmjopen-2013-002672 doi: 10.1128/AAC.00830-16 27. Danquah CA, Maitra A, Gibbons S, Faull J, Bhakta S. HT-SPOTi: A rapid drug 46. Torrent M, Pulido D, Valle J, Nogués MVV, Andreu D, Boix E. susceptibility test (DST) to evaluate antibiotic resistance profiles and novel Ribonucleases as a host-defence family: evidence of evolutionarily conserved chemicals for anti-infective drug discovery. Curr Protoc Microbiol. (2016) antimicrobial activity at the N-terminus. Biochem J. (2013) 456:99–108. 40:17.8.1–17.8.12. doi: 10.1002/9780471729259.mc1708s40 doi: 10.1042/BJ20130123 28. Salazar VA, Arranz-Trullén J, Navarro S, Blanco JA, Sánchez D, Moussaoui 47. Zambrano R, Jamroz M, Szczasiuk A, Pujols J, Kmiecik S, Ventura M, et al. Exploring the mechanisms of action of human secretory RNase 3 S. AGGRESCAN3D (A3D): server for prediction of aggregation and RNase 7 against Candida albicans. Microbiologyopen. (2016) 5:830–45. properties of protein structures. Nucleic Acids Res. (2015) 43:W306–13. doi: 10.1002/mbo3.373 doi: 10.1093/nar/gkv359 29. Zhang L, Lu L, Li S, Zhang G, Ouyang L, Robinson K, et al. 1,25- 48. Torrent M, Cuyás E, Carreras E, Navarro S, López O, De La Maza A, Dihydroxyvitamin-D3 induces avian β-defensin gene expression in chickens. et al. Topography studies on the membrane interaction mechanism PLoS ONE. (2016) 11:e0154546. doi: 10.1371/journal.pone.0154546 of the eosinophil cationic protein. Biochemistry. (2007) 46:720–33. 30. Rekha RS, Rao Muvva SSVJ, Wan M, Raqib R, Bergman P, Brighenti S, et al. doi: 10.1021/bi061190e Phenylbutyrate induces LL-37-dependent autophagy and intracellular killing 49. Huang Y-C, Lin Y-M, Chang T-W, Wu S-J, Lee Y-S, Chang MD-T, et al. The of Mycobacterium tuberculosis in human macrophages. Autophagy. (2015) flexible and clustered lysine residues of human ribonuclease 7 are critical 11:1688–99. doi: 10.1080/15548627.2015.1075110 for membrane permeability and antimicrobial activity. J Biol Chem. (2007) 31. Siddiqui MA, Malathi K. RNase L induces autophagy via c-Jun N-terminal 282:4626–33. doi: 10.1074/jbc.M607321200 kinase and double-stranded RNA-dependent protein kinase signaling 50. Boix E, Torrent M, Sánchez D, Nogués MV. The antipathogen activities pathways. J Biol Chem. (2012) 287:43651–64. doi: 10.1074/jbc.M112.399964 of eosinophil cationic protein. Curr Pharm Biotechnol. (2008) 9:141–52. 32. Brambilla C, Llorens-Fons M, Julián E, Noguera-Ortega E, Tomàs- doi: 10.2174/138920108784567353 Martínez C, Pérez-Trujillo M, et al. Mycobacteria clumping increase 51. Prats-Ejarque G, Arranz-Trullen J, Blanco JA, Pulido D, Nogues MV, their capacity to damage macrophages. Front Microbiol. (2016) 7:1562. Moussaoui M, et al. The first crystal structure of human RNase 6 reveals doi: 10.3389/fmicb.2016.01562 a novel substrate-binding and cleavage site arrangement. Biochem J. (2016) 33. Fiorini C, Cordani M, Gotte G, Picone D, Donadelli M, Fiorini C, 473:1523–36. doi: 10.1042/BCJ20160245 et al. Onconase induces autophagy sensitizing pancreatic cancer cells to 52. Harder JJ, Schröder JM, Schroder J-M. RNase 7, a novel innate immune gemcitabine and activates Akt/mTOR pathway in a ROS-dependent manner. defense antimicrobial protein of healthy human skin. J Biol Chem. (2002) Biochim Biophys Acta. (2015) 1853:549–60. doi: 10.1016/j.bbamcr.2014.12.016 277:46779–84. doi: 10.1074/jbc.M207587200 34. Muciño G, Castro-Obregón S, Hernandez-Pando R, Del Rio G. Autophagy as 53. Padhi A, Sengupta M, Sengupta S, Roehm KH, Sonawane A. Antimicrobial a target for therapeutic uses of multifunctional peptides. IUBMB Life. (2016) peptides and proteins in mycobacterial therapy: current status and future 68:259–67. doi: 10.1002/iub.1483 prospects. Tuberculosis. (2014) 94:363–73. doi: 10.1016/j.tube.2014.03.011 Frontiers in Immunology | www.frontiersin.org 16 July 2019 | Volume 10 | Article 1500 Lu et al. RNases’ Antimycobacterial and Autophagy Activities 54. Gupta SK, Haigh BJ, Griffin FJ, Wheeler TT. The mammalian secreted RNases: Mycobacterium tuberculosis by enhancing calgranulin a expression. J Infect mechanisms of action in host defence. Innate Immun. (2013) 19:86–97. Dis. (2014) 209:578–87. doi: 10.1093/infdis/jit495 doi: 10.1177/1753425912446955 72. Boix E, Nogués MV. Mammalian antimicrobial proteins and peptides: 55. Acharya KR, Ackerman SJ. Eosinophil granule proteins: Form and function. J overview on the RNase A superfamily members involved in innate host Biol Chem. (2014) 289:17406–15. doi: 10.1074/jbc.R113.546218 defence. Mol Biosyst. (2007) 3:317. doi: 10.1039/b617527a 56. Shamri R, Xenakis JJ, Spencer LA. Eosinophils in innate immunity: 73. Hancock RE, Diamond G. The role of cationic antimicrobial an evolving story Revital. Cell Tissue Res. (2011) 343:57–83. peptides in innate host defences. Trends Microbiol. (2000) 8:402–10. doi: 10.1007/s00441-010-1049-6 doi: 10.1016/S0966-842X(00)01823-0 57. Melo RCN, Weller PF. Contemporary understanding of the secretory 74. Nguyen LT, Haney EF, Vogel HJ. The expanding scope of antimicrobial granules in human eosinophils. J Leukoc Biol. (2018) 104:85–93. peptide structures and their modes of action. Trends Biotechnol. (2011) doi: 10.1002/JLB.3MR1217-476R 29:464–72. doi: 10.1016/j.tibtech.2011.05.001 58. Rivas-Santiago B, Rivas Santiago CE, Castañeda-Delgado JE, León-Contreras 75. Fiorini C, Gotte G, Donnarumma F, Picone D, Donadelli M, Fiorini C, JC, Hancock REW, Hernandez-Pando R. Activity of LL-37, CRAMP et al. Bovine seminal ribonuclease triggers Beclin1-mediated autophagic cell and antimicrobial peptide-derived compounds E2, E6 and CP26 against death in pancreatic cancer cells. Biochim Biophys Acta. (2014) 1843:976–84. Mycobacterium tuberculosis. Int J Antimicrob Agents. (2013) 41:143–8. doi: 10.1016/j.bbamcr.2014.01.025 doi: 10.1016/j.ijantimicag.2012.09.015 76. Chakrabarti A, Ghosh PK, Banerjee S, Gaughan C, Silverman RH. RNase L 59. Wallis RS, Hafner R. Advancing host-directed therapy for tuberculosis. Nat triggers autophagy in response to viral infections. J Virol. (2012) 86:11311–21. Rev Immunol. (2015) 15:255–63. doi: 10.1038/nri3813 doi: 10.1128/JVI.00270-12 60. Portugal B, Motta FN, Correa AF, Nolasco DO, de Almeida H, Magalhães 77. Thompson DM, Parker R. Stressing out over tRNA cleavage. Cell. (2009) KG, et al. Mycobacterium tuberculosis Prolyl oligopeptidase induces in vitro 138:215–9. doi: 10.1016/j.cell.2009.07.001 secretion of proinflammatory cytokines by peritoneal macrophages. Front 78. Cullinane M, Gong L, Li X, Adler NL, Tra T, Wolvetang E, et al. Stimulation of Microbiol. (2017) 8:155. doi: 10.3389/fmicb.2017.00155 autophagy suppresses the intracellular survival of Burkholderia pseudomallei 61. Wang J, Yang K, Zhou L, Minhaowu, Wu Y, Zhu M, et al. MicroRNA- in mammalian cell lines. Autophagy. (2008) 4:744–53. doi: 10.4161/auto.6246 155 promotes autophagy to eliminate intracellular mycobacteria by targeting 79. Hancock REW, Sahl H-G. Antimicrobial and host-defense peptides as Rheb. PLoS Pathog. (2013) 9:e1003697. doi: 10.1371/journal.ppat.1003697 new anti-infective therapeutic strategies. Nat Biotechnol. (2006) 24:1551–7. 62. Bradfute SB, Castillo EF, Arko-Mensah J, Chauhan S, Jiang S, Mandell M, et al. doi: 10.1038/nbt1267 Autophagy as an immune effector against tuberculosis. Curr Opin Microbiol. 80. Nicolas P. Multifunctional host defense peptides: intracellular- (2013) 16:355–65. doi: 10.1016/j.mib.2013.05.003 targeting antimicrobial peptides. FEBS J. (2009) 276:6483–96. 63. Yu X, Li C, Hong W, Pan W, Xie J. Autophagy during Mycobacterium doi: 10.1111/j.1742-4658.2009.07359.x tuberculosis infection and implications for future tuberculosis medications. 81. Haney EF, Hancock REWW. Peptide design for antimicrobial and Cell Signal. (2013) 25:1272–8. doi: 10.1016/j.cellsig.2013.02.011 immunomodulatory applications. Biopolymers. (2013) 100:572–83. 64. Randow F, Youle RJ. Self and nonself: how autophagy targets doi: 10.1002/bip.22250 mitochondria and bacteria. Cell Host Microbe. (2014) 15:403–11. 82. Ashida H, Ogawa M, Mimuro H, Kobayashi T, Sanada T, Sasakawa C. Shigella doi: 10.1016/j.chom.2014.03.012 are versatile mucosal pathogens that circumvent the host innate immune 65. Cemma M, Brumell JHH. Interactions of Pathogenic Bacteria with Autophagy system. Curr Opin Immunol. (2011) 23:448–55. doi: 10.1016/j.coi.2011.06.001 Systems. Curr Biol. (2012) 22:R540–5. doi: 10.1016/j.cub.2012.06.001 83. Walch M, Eppler E, Dumrese C, Barman H, Groscurth P, Ziegler U. Uptake 66. Gupta A, Kaul A, Tsolaki AG, Kishore U, Bhakta S. Mycobacterium of granulysin via lipid rafts leads to lysis of intracellular Listeria innocua. J tuberculosis: Immune evasion, latency and reactivation. Immunol. (2005) 174:4220–7. doi: 10.4049/jimmunol.174.7.4220 Immunobiology. (2012) 217:363–74. doi: 10.1016/j.imbio.2011. 84. Arnett E, Lehrer RI, Pratikhya P, Lu W, Seveau S. Defensins 07.008 enable macrophages to inhibit the intracellular proliferation 67. Rodriguez Plaza JG. Prospective tuberculosis treatment: peptides, of Listeria monocytogenes. Cell Microbiol. (2011) 13:635–51. immunity and autophagy. J Mol Genet Med. (2014) 8:120. doi: 10.1111/j.1462-5822.2010.01563.x doi: 10.4172/1747-0862.1000128 85. Silva T, Moreira AC, Nazmi K, Moniz T, Vale N, Rangel M, et al. 68. Wu YT, Tan HL, Shui G, Bauvy C, Huang Q, Wenk MR, et al. Dual role of Lactoferricin peptides increase macrophages’ capacity to kill Mycobacterium 3-Methyladenine in modulation of autophagy via different temporal patterns avium. mSphere. (2017) 2:e00301–17. doi: 10.1128/mSphere.00301-17 of inhibition on class I and III phosphoinositide 3-kinase. J Biol Chem. (2010) 285:10850–61. doi: 10.1074/jbc.M109.080796 Conflict of Interest Statement: The authors declare that the research was 69. Yamamoto A, Tagawa Y, Yoshimori T, Moriyama Y, Masaki R, Tashiro Y. conducted in the absence of any commercial or financial relationships that could Bafilomycin A1 prevents maturation of autophagic vacuoles by inhibiting be construed as a potential conflict of interest. fusion between autophagosomes and lysosomes in rat hepatoma cell line, H-4-II-E cells. Cell Struct Funct. (1998) 23:33–42. doi: 10.1247/csf.23.33 Copyright © 2019 Lu, Arranz-Trullén, Prats-Ejarque, Pulido, Bhakta and Boix. 70. Sonawane A, Santos JC, Mishra BB, Jena P, Progida C, Sorensen This is an open-access article distributed under the terms of the Creative Commons OE, et al. Cathelicidin is involved in the intracellular killing of Attribution License (CC BY). The use, distribution or reproduction in other forums mycobacteria in macrophages. Cell Microbiol. (2011) 13:1601–17. is permitted, provided the original author(s) and the copyright owner(s) are credited doi: 10.1111/j.1462-5822.2011.01644.x and that the original publication in this journal is cited, in accordance with accepted 71. Dhiman R, Venkatasubramanian S, Paidipally P, Barnes PF, Tvinnereim academic practice. No use, distribution or reproduction is permitted which does not A, Vankayalapati R. Interleukin 22 inhibits intracellular growth of comply with these terms. Frontiers in Immunology | www.frontiersin.org 17 July 2019 | Volume 10 | Article 1500