The equine antimicrobial peptide eCATH1 is effective against the facultative intracellular pathogen

1The equine antimicrobial peptide eCATH1 is effective against the facultative intracellular 1pathogen Rhodococcus equi in mice 23Running title: eCATH1 for treatment of Rhodococcus equi infection 45Margot Schlusselhuber 1, Riccardo Torelli 2, Cecilia Martini 2, Matthias Leippe 3, Vincent Cattoir 4,5, 6Roland Leclercq 4,5, Claire Laugier 1, Joachim Grötzinger 6, Maurizio Sanguinetti 2*, Julien 7Cauchard 1#* 891 Anses, Dozulé laboratory for equine diseases, Bacteriology & Parasitology unit, Goustranville, 10F-14430 Dozulé, France 112 Istituto di Microbiologia, Università Cattolica del Sacro Cuore, Rome, Italy 123 Department of Zoophysiology, Zoological Institute, Christian-Albrechts University of Kiel, 13Olshausenstrasse 40, 24098 Kiel, Germany 144 Caen University Hospital, Department of Clinical Microbiology, F-14033 Caen, France 155 University of Caen Basse-Normandie, Research Unit EA4655 (Team “Antibioresistance”), 16Medical School, F-14032 Caen, France 176 Institute of Biochemistry, Christian-Albrechts University of Kiel, Olshausenstrasse 40, 24098 18Kiel, Germany 1920* Co-last authors 21# Corresponding author and mailing address: 22Dr. Julien Cauchard23Copyright © 2013, American Society for Microbiology. All Rights Reserved.Antimicrob. Agents Chemother. doi:10.1128/AAC.02044-12 AAC Accepts, published online ahead of print on 1 July 20132Anses, Dozulé laboratory for equine diseases, Goustranville, F-14430 Dozulé, FRANCE. 24 Phone: +33 2 31 79 22 76 25 Fax: +33 2 31 39 21 37 26 E-mail: julien.cauchard@anses.fr273ABSTRACT 28 Rhodococcus equi, the causal agent of rhodococcosis, is a major pathogen of foals and is also 29 responsible of severe infections in immunocompromised humans. Of great concern, resistant 30 strains to currently used antibiotics have emerged. As the number of drugs that are efficient in 31 vivo is limited because of the intracellular localization of the bacterium inside macrophages, new 32 active but cell-permeant drugs will be needed in the near future. In the present study, we 33 evaluated, by in vitro and ex vivo experiments, the ability of the alpha helical equine 34 antimicrobial peptide eCATH1 to kill intracellular bacterial cells . Moreover, the therapeutic 35 potential of the peptide was assessed in an experimental rhodococcosis induced in mice while the 36 in vivo toxicity was evaluated by behavioral and histopathological analysis. The study revealed 37 that eCATH1 significantly reduced the number of bacteria inside macrophages. Furthermore, the 38 bactericidal potential of the peptide was maintained in vivo at doses that appeared to have no 39 visible deleterious effects for the mice even after 7 days of treatment. Indeed, daily subcutaneous 40 injections of 1 mg/kg body weight of eCATH1 led to a significant reduction of the bacterial load 41 in organs comparable to that obtained after treatment with 10 mg/kg body weight of rifampin. 42 Interestingly, the combination of the peptide with rifampin showed a synergistic interaction in 43 both ex vivo and in vivo experiments. These results emphasize the therapeutic potential that 44 eCATH1 represents in the treatment of rhodococcosis.454INTRODUCTION 46 Rhodococcus equi (R. equi ), previously known as Corynebacterium equi , was initially described 47 in 1923 by Magnusson in foals presenting pyogranulomatous pneumonia (1). Rhodococcosis is 48 now typically reported in foals from one to six month of age and is one of the most common 49 causes of mortality at this age (2-5). The pathogen, however, can infect a wide variety of other 50 animal species including humans. Since the first case described in 1967, the number of reports on 51 the infection have increased substantially in humans, with ~200 cases in the last three decades, 52 preferentially in persons whose immune system was compromised, either naturally, or due to 53 illness or medical treatment (6-9). 54 Many antimicrobials are effective against the extracellular pathogen in vitro . However, the 55 number of antimicrobials available for treatment of rhodococcosis is limited because of the 56 intracellular location of the bacterium. The combination of rifampin and a macrolide such as 57 erythromycin, or more recently of azithromycin and clarithromycin, is considered today as the 58 treatment of choice, as these antibiotics have the ability to penetrate abscesses and cells due their 59 lipophilic nature and to accumulate in granulocytes and macrophages to kill the bacterium (10). 60 Since the introduction of these antibiotics in rhodococcosis treatment, the survival rate of foals 61 has increased dramatically from less than 20% up to 97% (reviewed in (K. Chaffin, presented at 62 the 49th Annual Ocala Equine Conference, Ocala, FL, 21 to 24 October 2011)). Unfortunately, 63 some observations suggest that there might be a problem in the treatment of R. equi infections in 64 the near future due to the emergence of rhodococci resistant to these antibiotics and macrolides´ 65 potentially fatal side effects for horses (11-14).665With the rise of bacterial resistance to antibiotics, there is a growing interest in anti-infective 67 agents with fundamentally different mechanisms of action than those of conventional antibiotics 68 (15, 16). The concept of using antimicrobial peptides (AMPs) as therapeutic tools dates back to 69 the 1990s and they are now emerging as particularly promising candidates as new anti-infective 70 agents in the antimicrobial research area (17). In the nature, these peptides participate in the first 71 line of host defense against pathogens by combining antimicrobial activity with 72 immunomodulatory properties (18). 73 We recently reported the in vitro activity of the alpha helical equine AMP, eCATH1, against 74 antibiotic-resistant and antibiotic-susceptible strains of R. equi . The peptide proved to be 75 extracellularly active at low micromolar concentrations (0.16-1.27 µM) against the pathogen 76 independently of the antibiotic resistance profile (19) (M. Schlusselhuber, K. Guldbech, C. Sévin, 77 C. Laugier, M. Leippe, J. Grötzinger, R. Leclercq, S. Giguère and J. Cauchard, submitted for 78 publication). Aiming at extending the therapeutic potential of eCATH1 against rhodococcosis, 79 the present study was designed to analyze the efficacy of the peptide to kill intracellular R. equi 80 and to combat the pathogen in an infection-model in mice. 81 82 MATERIALS AND METHODS 83 Cell line, bacterial strain and growth conditions 84 The mouse macrophage cell line J774.2 (ECACC reference number: 85011428) purchased from 85 Sigma-Aldrich was grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Rockville, 86 MD, USA) supplemented with 10% (v/v) Fetal bovine serum (FBS; Pan Biotech GmbH, 87 Aidenbach, Germany ), 100 µg/ml streptomycin, 100 units/ml penicillin (complete medium) and 88 incubated at 37 °C in a humidified atmosphere containing 5% CO 2. Before assays, cell number896and viability of cells were determined by exclusion of the vital dye Trypan blue. Virulent, vapA -90 positive R. equi ATCC 33701 was used in this study. Pure strain freshly thawed was grown in 91 Brain Heart Infusion medium (BHI; BD Difco, Franklin Lakes, NJ, USA) before each experiment 92 to avoid the lost of the virulence plasmid. Cultures were incubated aerobically at 37 °C for 24 to 93 48 h. For cell line infection, bacteria were suspended with an appropriate volume of DMEM 94 supplemented with 5% FBS. The actual number of cells was confirmed by plating serial dilutions 95 on BHI agar plates. For mice inoculation, bacteria were harvested from a liquid culture by 96 centrifugation, washed, and resuspended in phosphate-buffered saline (PBS, Gibco, Rockville, 97 MD, USA) at the required bacterial concentration. 98 99 Antimicrobials used 100 eCATH1 was chemically synthesized at a purity grade of at least 96% by GenScript USA Inc. 101 (Piscataway, NJ, USA) and dissolved in 10 mM acetic acid at the final concentration of 1 mg/ml. 102 Rifampin stock solution was prepared by dissolving antibiotic powder (Ref R7382, Sigma-103 Aldrich, St. Louis, MO, USA) in methanol at a final concentration of 50 mg/kg. Stock solutions 104 of antimicrobials were stored at -20 °C until use. 105 106 Survival of R. equi in mouse macrophage cell line 107 Before the use of animals, the activity of eCATH1 against intracellular rhodococci was evaluated 108 in vitro . J774.2 murine monocyte/macrophage-like cells were washed with PBS, detached by 109 scraping and resuspended in complete medium at a final cell concentration of 5 x 105 cell/ml. 110 Monocytes were allowed to attach for 3 h on 8-wells Lab-tek chamber slide (200 µl of cell 111 suspension per well) at 37 °C in a humidified atmosphere containing 5% CO 2. Infection of1127mammalian cells was performed by washing cells with PBS before adding 200 µl of bacterial 113 suspension per well in a bacterium-to-monocyte ratio of 2:1. Chamber slides were incubated 45 114 min at 37 °C in a humidified atmosphere containing 5% CO 2 to allow serum opsonisation of 115 rhodococci. Infected monolayers were carefully washed three times with PBS to remove unbound 116 bacteria. After the last wash, 200 µl of DMEM supplemented either with eCATH1 (final 117 concentration: 20 µg/ml) or the same volume of peptide solvent (acetic acid, final concentration 118 of 0.2 mM) was added to infected cells before incubation at 37 °C in a humidified atmosphere 119 containing 5% CO 2. Viable intracellular bacteria were assessed after 24 h by staining with the 120 LIVE/DEAD Baclight bacterial viability kit (Molecular Probes Europe BV, Leiden, The 121 Netherlands) in saponin solution (final concentration of 0.05% (w/v)) to permeabilize 122 macrophages membranes as previously described (20). Slides were examined by fluorescence 123 microscopy (microscope Axioskop 40, Zeiss). The bacterial viability kit is composed of 124 propidium iodide that stains dead cells in red and Syto 9 that stains viable cells in green. 125 Therefore, nuclei of permeabilized macrophages appear red. Before the experiment, the effect of 126 20 µg/ml of eCATH1 on plasma membrane integrity of J774.2 cell line was assessed by the 127 lactate dehydrogenase (LDH) release assay after 24h of treatment. One hundred percent lysis was 128 achieved b using 1% (vol/vol) Triton X00. The LDH release assay was carried out according to 129 the manufacturer’s instructions using a commercially available kit (TOX-7, Sigma-Aldrich, Saint 130 Louis, MO, USA). 131 132 Animal studies 133 Pathogen-free BALB/c mice (Harlan Italy S.r.l., San Pietro al Natisone, Udine, Italy) were 134 housed with free access to water and standard mouse food at animal facilities of the Sacred Heart1358University (Rome, Italy). All animals used for experiments were females of 10 weeks of age 136 weighing approximately 20-25 g. For the needs of experiments, animals were sacrificed by 137 cervical dislocation. The animal experiments were performed under a protocol approved by the 138 Institutional Animal Use and Care Committee at Università Cattolica del S. Cuore, Rome, Italy 139 (Permit number: N21, 12/05/2010) and authorized by the Italian Ministry of Health, according to 140 Legislative Decree 116/92, which implemented the European Directive 86/609/EEC on 141 laboratory animal protection in Italy. Animal welfare was routinely checked by veterinarians of 142 the Service for Animal Welfare. 143 144 145 Survival of R. equi in mouse peritoneal macrophages 146 Survival of R. equi in mouse peritoneal macrophages after treatment with eCATH1, rifampin, or 147 eCATH1 combined with rifampin was assessed by using an ex vivo infection model as described 148 previously (21, 22). Briefly, mice were infected with 107 to 108 CFU of R. equi by intraperitoneal 149 injection. After an 8 h-infection period, the peritoneal macrophages were collected by peritoneal 150 lavage, centrifuged, and suspended in DMEM containing 10 mM HEPES, 2 mM glutamine, 10% 151 (v/v) FBS, and 1× non-essential amino acids supplemented with vancomycin (10 μg/ml) and 152 gentamicin (150 μg/ml). The cell suspension was dispensed into 24-well tissue culture plates and 153 incubated at 37 °C under 5% CO 2 for 2 h. After exposure to antibiotics to kill extracellular 154 bacteria (i.e., at 10 h post-infection), the infected macrophages were washed, and triplicate wells 155 of macrophages were lysed with detergent. After dilution with BHI broth, the lysates were plated 156 on BHI agar to quantify the viable intracellular bacteria. The remaining wells containing infected 157 macrophages were maintained in DMEM and treated with eCATH1 (20 µg/ml), rifampin (0.5, 51589and 10 µg/ml), or eCATH1 (20 µg/ml) combined with rifampin (5 µg/ml) for the duration of the 159 experiment. At 24 h post-infection, supernatant fluid was removed from each well, and 160 intracellular bacteria were quantified by counting on BHI agar plates. Before the experiment, 161 toxicity of eCATH1 and rifampin on mouse peritoneal macrophages was assessed with a range 162 concentration up to 80 µg/ml in order to determine the concentration of antimicrobials used in the 163 study. To assess cell viability, macrophages treated for 24 h were analyzed using the AlamarBlue 164 test (Invitrogen, Milan, Italy) following the manufacturer’s instructions. 165 166 Activity of eCATH1 against R. equi -infected mice 167 The activity of eCATH1 alone and in combination with rifampin was assessed on 168 rhodocococcosis induced in mice. Before therapeutic trials, comparison of the natural clearance 169 of R. equi CFU from organs was established in our experimental conditions to determine time 170 points of treatment and the sub-lethal dose. Briefly, ten mice per group were inoculated 171 intravenously with 107, 108 and 109 CFU of R. equi and two animals of each group were 172 sacrificed at days 1, 2, 6, 8, and 10 post-infection for bacterial count in liver, spleen, and lung. 173 The intravenous route was chosen based on Takai et al., who reported a higher virulence of the R. 174 equi ATCC 33701 strain by this route compared to the intraperitoneal route (23). All the mice 175 were monitored for survival throughout the experimental period. According to these preliminary 176 results, mice were intravenously challenged with 200 µl of a sub-lethal dose (108 CFU) of R. 177 equi . Therapeutic trials with BALB/c mice were initiated one day after bacterial inoculation to let 178 the infection develop and symptoms appear and before natural clearance of rhodococcoci from 179 mouse organs. Stock solutions of antimicrobials were appropriately freshly diluted in isotonic 180 sodium chloride before subcutaneous injection of 200 µl to animals. Infected mice were treated18110once a day with eCATH1 (1 mg/kg body weight, Group I), rifampin (10 mg/kg body weight, 182 Group II), a combination of both in the same concentrations as mentioned above (Group III), or 183 the same volume of isotonic sodium chloride (Control group) during 7 days. The doses, 184 periodicities and route of injections were chosen based on previous reports (24-26). Five mice per 185 group were sacrificed 1, 4, and 8 days post-infection (i.e. start-, mid- and end- treatment) and 186 their spleens and livers aseptically removed. Organs were weighed and separately homogenized 187 in PBS. The R. equi concentrations in organs (CFU/g organ) were determined by plating 10-fold 188 serial dilutions of the tissue homogenates on BHI agar. The plates were incubated at 37 °C for 24 189 to 48 h before bacterial enumeration. 190 191 In vivo toxicity of eCATH1 192 Non-infected mice were used to assess the in vivo toxicity of eCATH1 by behavioral and 193 histopathological analysis. Five mice was treated once a day with eCATH1 for 7 days (1 mg/kg 194 body weight) by subcutaneous injection while a second group of five not treated mice was used 195 as a control group. All the mice were monitored for survival and presence of any drug-related 196 adverse effect (local signs of inflammation, weight loss, diarrhea and behavioral alterations) 197 throughout the experimental period. Mice were sacrificed after 7 days of treatment, and organs 198 were aseptically removed (intestine, spleen, liver, lung, kidney, and stomach). Tissue samples 199 were fixed in 3.7% formaldehyde solution, processed for paraffin embedding, and stained with 200 Hematoxylin/eosin before analysis at the Veterinary Pathological Anatomy laboratory (Amboise, 201 France) by pathologists. 202 203 Statistics20411Ex vivo and in vivo experiments were performed in triplicate and duplicate, respectively, and the 205 results were subjected to statistical analysis by using one-way analysis of variance (ANOVA) 206 with a Bonferroni correction posttest with GraphPad Prism version 5.04 for Windows (GraphPad 207 Software, San Diego, CA). 208 209 RESULTS 210 Survival of R. equi in mouse macrophage cell line after addition of eCATH1 211 Before the assessment of the bactericidal activity of eCATH1 on R. equi inside J774.2 mouse 212 macrophages, the effects of the peptide on plasma membrane integrity of the host cells were 213 evaluated by the LDH release assay. At a concentration of 20 µg/ml, eCATH1 exerted no 214 detectable cytotoxic effect on J774.2 cells (data not shown). 215 The activity of eCATH1 against the intracellular pathogen was first assessed by fluorescent 216 microscopy. J774.2 macrophages were infected with R. equi for 45 min, and cells were 217 subsequently incubated for 24 h either with 20 µg/ml of eCATH1 or the peptide solvent (negative 218 control). Fluorescent staining of intracellular R. equi with the LIVE/DEAD baclight bacterial 219 viability kit revealed clear differences in rhodococcal survival inside macrophages between both 220 conditions. Indeed, very few macrophages were found to be infected in the presence of peptide 221 and those cells contained substantially lower numbers of viable bacteria (stained in green) as 222 observed in the negative control (Fig. 1). Moreover, the overall number of macrophages was 223 higher in the presence of eCATH1 compared to the non-treated sample. In the negative control, 224 no dead bacteria (stained in red) were observed, indicating that the virulent strain used in this 225 study is able to successfully survive inside macrophages (Fig. 1 c, d),. Dead rhodococci were also 226 not observed in the presence of eCATH1, but it is most likely the result of a rapid lysis of bacteria22712by the peptide. Extracellular bacteria were observed in the negative control, but not in the treated 228 sample. Before adding the peptide or its solvent, infected cells were carefully washed several 229 times to remove any non-internalized bacteria. Therefore, the presence of extracellular bacteria 230 might be either due to the necrotic death of macrophages after bacterial multiplication during 24 231 h, to the chemical action of saponin on the macrophage membrane that may release bacteria or to 232 a mechanical break of eukaryotic membranes by the positioning of the coverslip. 233 234 Survival of R. equi in mouse peritoneal macrophages 235 To confirm the in vitro study, complementary ex vivo experiments were performed. For this, 236 mouse macrophages were naturally infected in vivo and further collected for in vitro treatment 237 with eCATH1 and rifampin before CFU enumeration. As both antimicrobials exhibited a 238 synergistic interaction in vitro against extracellular R. equi (19), the combination 239 eCATH1/rifampin in the present ex vivo model was also assessed. Before the ex vivo experiment, 240 cytotoxicity of eCATH1 and rifampin on mouse peritoneal macrophages revealed that, similarly 241 to the J774.2 cell line, eCATH1 did not exhibit significant toxicity at 20 µg/ml, whereas rifampin 242 was found to exert a significant toxicity at 20 µg/ml (data not shown, ANOVA test p<0.05). 243 Concentrations of antimicrobials used in the ex vivo experiment were therefore chosen as 20 244 µg/ml for eCATH1 and 0.5, 5, and 10 µg/ml for rifampin. Supporting the in vitro data, it was 245 observed that eCATH1 resulted in a significant killing (p<0.001) of intracellular R. equi at a 246 concentration of 20 µg/ml after 14 h of treatment, i.e. ~60% decrease of R. equi in macrophages 247 compared to the control (Fig. 2). The same killing rate was found already at 5 µg/ml of rifampin. 248 Interestingly, the combination of both antimicrobials led to an even higher significant decrease of 249 the overall number of R. equi residing in macrophages (~90%, p<0.00001) (Fig. 2). However,25013from these data it was not possible to conclude whether this is a synergy or only an additive 251 effect. 252 253 Activity of eCATH1 against R. equi -infected mice 254 Before the therapeutic trials, comparison of the natural clearance of R. equi CFU from mice 255 organs for various bacterial inoculums (107 to 109 CFU) was assessed (data not shown). 256 Intravenous injection of 109 CFU of R. equi led to the death of mice after 2 to 4 days. In contrast, 257 an inoculum of 108 CFU led to the onset of symptoms from days 1 to 6 post-infection, the in vivo 258 multiplication of the pathogen, and a progressive clearance of bacteria from spleen, liver and 259 lungs after 4 days. Therefore, this sub-lethal dose was selected for the therapeutic trials. 260 Moreover, the lung was no longer used for bacterial counting in the therapeutic trials of our study 261 as a lower concentration of bacteria was found in this organ compared to spleen and liver. 262 For therapeutic trials, mice were infected with the sub-lethal dose of R. equi . One day later, 263 therapy started with daily injections of antimicrobials during 7 days. The results of CFU 264 enumeration in spleen and liver are presented in Figure 3. After three days of treatment, bacterial 265 loads were significantly decreased by comparison to the control in a similar manner in spleen and 266 liver. Indeed, 1mg/kg of body weight of eCATH1 as well as 10 mg/kg of body weight of rifampin 267 led to a decrease of ~2 log 10 CFU in both organs (i.e. >99%) compared to the control. eCATH1 268 combined with rifampin (1 mg/kg and 10 mg/kg body weight, respectively) showed the highest 269 antimicrobial activity in mice with a decrease of ~3 log 10 CFU in organs (i.e. >99.9%). 270 271 In vivo toxicity of eCATH127214Daily injection of 1 mg/kg body weight of eCATH1 during 7 days did not result in any 273 behavioral changes, differences in body weight or any clinical evidence of antimicrobials-related 274 adverse effects (data not shown). Moreover, the histopathology analysis performed on the 275 intestine, spleen, liver, lung, kidney, and stomach of treated and non-treated mice did not reveal 276 any histological changes between the groups (Fig. 4). 277 278 DISCUSSION 279 The data presented in this study represent the first demonstration of the intracellular 280 antirhodoccoccal properties of an antimicrobial peptide by application of in vitro , ex vivo and in 281 vivo methods. Indeed, the present in vitro and ex vivo studies proved that eCATH1 remained 282 active against bacteria inside the host cell although a higher concentration was needed to reach 283 the intracellular pathogen. These results support similar work in Mycobacterium species. Sharma 284 et al. demonstrated in vitro the direct bactericidal activity of the human neutrophil peptide (HNP-285 1) against intracellular Mycobacterium tuberculosis. The number of cells per macrophage was 286 reduced by 1 log 10 with 20 µg/ml and by 2 log 10 with 40 µg/ml of peptide after three days of 287 treatment (27). More recently, Jena et al., reported the activity of NK-2, a shortened peptide 288 derived from the porcine NK-lysin, against intracellular Mycobacterium smegmatis. This study 289 revealed that 10 µM of peptide (~30 µg/ml) decreased the number of bacteria per macrophage by 290 ~40% after 8 h of treatment (28). 291 From the mouse rhodococcosis model, the main conclusion is that eCATH1 remained effective 292 against the facultative intracellular pathogen.. Interestingly, the peptide appeared to be more 293 active in vivo than in vitro in comparison to rifampin activity. A four-fold-lower concentration of 294 rifampin led to a similar decrease of rhodococci per macrophage than eCATH1 in vitro , while, in29515vivo , a ten-fold-higher concentration of rifampin than the peptide is required to equally decrease 296 the bacterial load in organs. Previously, similar data were reported by Sharma et al., (26). As 297 little as 1-5 µg of the HNP-1 per mouse was found to decrease tuberculosis infection while 40 298 µg/ml of peptide was needed to kill intramacrophage mycobacteria in ex vivo experiments (26, 299 27). These observations might indicate that eCATH1 has a more complex effect in vivo than in 300 vitro . As previously raised by Sharma et al., we hypothesize that the in vivo activity of eCATH1 301 is a combination of i) immunomodulatory properties leading to an indirect killing of bacteria and 302 ii) a direct killing effect on intra- and extracellular bacteria. Some reports in the literature about 303 immunomodulatory properties of human and bovine cathelicidins could support our hypothesis. 304 The human cathelicidin LL-37 exerts a chemoattractive activity on immune cells through direct 305 interaction with granulocytes and mononuclear cells receptors and through induction of 306 chemokines production which hypothetically increase the numbers of neutrophils and monocytes 307 at sites of infection (29-33). Interestingly, some bovine cathelicidins were also found to be 308 chemoattractants for neutrophils and had the property of enhancement of phagocytic and 309 degranulation activities of these cells (B. W. Paget, J. L. Harper and B. J. Haigh, presented at the 310 3rd Antimicrobial peptide symposium, Lille, France, 13 to 15 June 2012). Moreover, 311 macrophages have the ability to uptake granules of apoptotic neutrophils to acquire AMPs that 312 traffic via early endosomes to intracellular pathogens-containing vacuoles where they exert their 313 bactericidal activity (34). Chemoattraction of immune cells such as neutrophils by eCATH1 314 might limit the spread of R. equi infection in the host in a similar way. Furthermore, this 315 hypothesis would be in accordance with the observation that neutrophils are critical for the 316 control of rhodocococcosis (35). Nevertheless, the immunomodulatory properties of eCATH1 317 remain to be elucidated to an better understanding of the in vivo activities of this peptide.31816The mechanism by which exogenously applied AMPs might access to intracellular bacteria is not 319 fully understood. Some clues, however, were first given by Sharma et al., in 2000 (27). The 320 studied peptide was initially found to localize at the macrophage membrane and then to be taken 321 up inside the cell (27). In a similar pathway to that of neutrophil’s AMPs uptake by macrophages 322 discussed above (34), eCATH1 might be internalized by macrophages in early endosomes that 323 would further fuse with R. equi-containing vacuoles, where the peptide could exert its 324 bactericidal activity. 325 Interestingly, the slight positive interaction between eCATH1 and rifampin against extracellular 326 rhodococci described previously, was observed against the facultative intracellular bacterium in 327 the in vivo setting as well (19). Our observations are consistent with the work of Cirioni et al., 328 who similarly reported a synergistic effect between rifampin and α-helical AMPs (magainin II 329 and cecropin A) in Pseudomonas aeruginosa infection models in rats (25). It was hypothesized 330 that AMPs allow rifampin, a RNA polymerase inhibitor, to access its intracellular target by 331 permeabilization of the bacterial membrane (19, 36). 332 In vivo degradation by proteases, inactivation by physiological salt concentration and rapid 333 clearance by kidneys and toxicity are often described in the literature as major limiting factors 334 that hamper the development of AMPs as therapeutic options. In the present study we proved that 335 the bactericidal activity of eCATH1 was conserved in vivo at doses that might be compatible with 336 clinical use, without detectable deleterious effects for the host even after 7 days of daily 337 subcutaneous injections. The results of the in vivo toxicity and activity presented here agree with 338 our previous investigations of eCATH1. Indeed, we recently reported that the in vitro activity of 339 eCATH1 was not hampered by physiological salt concentration and that the peptide was not toxic 340 for various mammalian cell types in vitro (19).341。