Combined Use of the Ab105-2φΔCI Lytic Mutant Phage and Different Antibiotics in Clinical Isolates of Multi-Resistant Acinetobacter baumannii

microorganisms Article Combined Use of the Ab105-2ϕ∆CI Lytic Mutant Phage and Different Antibiotics in Clinical Isolates of Multi-Resistant Acinetobacter baumannii Lucia Blasco 1,†, Anton Ambroa 1,†, Maria Lopez 1, Laura Fernandez-Garcia 1, Ines Bleriot 1, Rocio Trastoy 1, Jose Ramos-Vivas 2, Tom Coenye 3, Felipe Fernandez-Cuenca 4, Jordi Vila 5, Luis Martinez-Martinez 6, Jesus Rodriguez-Baño 4, Alvaro Pascual 4, Jose Miguel Cisneros 7, Jeronimo Pachon 7, German Bou 1and Maria Tomas 1,*,‡ 1 Microbiology Department-Research Institute Biomedical A Coruña (INIBIC), Hospital A Coruña (CHUAC), University of A Coruña (UDC), 15495 A Coruña, Spain; [email protected] (L.B.); [email protected] (A.A.); [email protected] (M.L.); [email protected] (L.F.-G.); [email protected] (I.B.); trastoy[email protected] (R.T.); [email protected] (G.B.) 2Microbiology Department-Research Institute Biomedical Valdecilla (IDIVAL), Hospital Marques de Valdecilla, 39008 Santander, Spain; [email protected] 3 Laboratory of Pharmaceutical Microbiology, Ghent University, 9000 Gent, Belgium; T[email protected] 4 Clinical Unit for Infectious Diseases, Microbiology and Preventive Medicine, Hospital Universitario Virgen Macarena/Department of Microbiology and Medicine, University of Seville/Biomedicine Institute of Seville (IBIS), 41009 Seville, Spain; [email protected] (F.F.-C.); [email protected] (J.R.-B.); [email protected] (A.P.) 5Institute of Global Health of Barcelona (ISGlobal), Hospital Clínic-Universitat de Barcelona, 170, 08036 Barcelona, Spain; [email protected] 6 Unit of Microbiology, University Hospital Reina Sof í a, Department of Microbiology, University of C ó rdoba, Maimonides Biomedical Research Institute of Cordoba (IMIBIC), 14004 Cordoba, Spain; [email protected] 7 Clinical Unit for Infectious Diseases, Microbiology and Preventive Medicine, Hospital Universitario Virgen del Rocío/Department of Microbiology and Medicine, University of Seville/Biomedicine Institute of Seville (IBIS), 41009 Seville, Spain; [email protected] (J.M.C.); [email protected] (J.P.) *Correspondence: ma.del.mar[email protected]; Tel.: +34-981-176-399; Fax: +34-981-178-273 †These authors equally contributed to this work. ‡On Behalf of Study Group on Mechanisms of Action and Resistance to Antimicrobials (GEMARA) of Spanish Society of Infectious Diseases and Clinical Microbiology (SEIMC)/Spanish Network for the Research in Infectious Diseases (REIPI). Received: 30 September 2019; Accepted: 9 November 2019; Published: 12 November 2019   Abstract: Phage therapy is an abandoned antimicrobial therapy that has been resumed in recent years. In this study, we mutated a lysogenic phage from Acinetobacter baumannii into a lytic phage (Ab105-2phi ∆ CI) that displayed antimicrobial activity against A. baumannii clinical strain Ab177_GEIH-2000 (isolated in the GEIH-REIPI Spanish Multicenter A. baumannii Study II 2000/2010, Umbrella Genbank Bioproject PRJNA422585, and for which meropenem and imipenem MICs of respectively, 32 µ g/mL, and 16 µ g/mL were obtained). We observed an in vitro synergistic antimicrobial effect (reduction of 4 log–7 log CFU/mL) between meropenem and the lytic phage in all combinations analyzed (Ab105-2phi ∆ CI mutant at 0.1, 1 and 10 MOI and meropenem at 1/4 and 1/8 MIC). Moreover, bacterial growth was reduced by 8 log CFU/mL for the combination of imipenem at 1/4 MIC plus lytic phage (Ab105-2phi ∆ CI mutant) and by 4 log CFU/mL for the combination of imipenem at 1/8 MIC plus lytic phage (Ab105-2phi ∆ CI mutant) at both MOI 1 and 10. These results were confirmed in an in vivo model (G. mellonella), and the combination of imipenem and mutant Ab105-2phi ∆ CI was most effective (p<0.05). This approach could help to reduce the emergence of phage resistant bacteria and restore sensitivity to antibiotics used to combat multi-resistant strains of Acinetobacter baumannii. Microorganisms 2019,7, 556; doi:10.3390/microorganisms7110556 www.mdpi.com/journal/microorganisms Microorganisms 2019,7, 556 2 of 14 Keywords: Acinetobacter baumannii; multiresistant; mutant lytic phage; phage therapy; antibiotic-phage synergy 1. Introduction Multi-drug resistant (MDR) bacteria, such as A. baumannii are considered to be a major concern by the World Health Organization (WHO), because of their ability to acquire antimicrobial resistance via intrinsic characteristics and mechanisms (e.g., presence of the outer membrane) or via mechanisms acquired by horizontal genetic transfer [1,2]. This situation has led to an urgent need to develop new antimicrobial agents and to a renewed interest in phage therapy. Phage therapy was first developed in the 1920s but was abandoned in the Western world after the discovery of antibiotics. However, the use of phage therapy continued in Eastern countries, such as Poland and USSR, where bacteriophages are used for the prophylaxis and treatment of infections, such as dysentery, ulcers, and methicillin resistant Staphylococcus aureus (MRSA) infections [3,4]. Phage therapy is now considered a real option for treating MDR bacteria, and there are some examples of its use in treating human patients [ 5 ]. Phages are bacterial viruses, and like other viruses, they are obligate parasites that enter host cells through mechanisms that are based on receptor recognition. Genetic material is then injected into the bacteria and use the bacterial machinery to produce phage proteins [ 6 , 7 ]. Phages generally undergo a lytic (virulent) or lysogenic (temperate) life cycle. Lytic phages infect, and rapidly lyse and kill host cells, releasing phage progeny into the surrounding medium. Lysogenic phages infect the host cell and integrate their nucleic acid into the host genome, or exist as plasmids in the host cells, remaining in a stable prophage state for generations. Prophages can be “induced” to exit the cell as lytic phages under some conditions, such as the presence of antibiotics [ 8 , 9 ]. The lysogenic/lytic cycle of temperate bacteriophages is controlled by Cro, CI, and CII proteins; the Cro protein induces the lytic state and the CI repressor protein inhibits the Cro protein, thereby inducing the lysogenic state [10]. Only lytic phages are used in phage therapy as lysogenic phages can transfer resistance genes or virulence factors to the host [11]. The combined use of antibiotics and phages has been tested in several studies, demonstrating strong control of the bacteria, and a reduction in the development of phage and/or antibiotic resistance [ 12 , 13 ]. Phages are good candidates for use in combination with antibiotics for various reasons, including that they have a different mechanism of action from antibiotics; hold a narrow spectrum of activity, which protects the normal microbiota; they can multiply at the infection site; they are abundant in nature and can be easily isolated; and production costs are low [14–16]. In this study, we produced a mutant lytic phage from a lysogenic phage, that is incorporated in the genome of a clinical strain of A. baumannii by deleting the CI repressor gene, and thus, preventing the entry of the phage into the lysogenic cycle [ 10 , 17 ]. We then tested the antimicrobial activity of the novel lytic phage, Ab105-2phi ∆ CI, in combination with carbapenem antibiotics (meropenem and imipenem) against a carbapenem-resistant strain of A. baumannii. The combined therapy enhanced the antimicrobial activity of both, the phage and the antibiotic; the bacterium became sensitive to the antibiotics and the emergence rate of phage resistant bacteria was reduced. 2. Material and Methods 2.1. Bacterial Strains In this study, we used 20 clinical strains isolated from Spanish hospitals during the GEIH-REIPI Spanish Multicenter Acinetobacter baumannii Study II 2000–2010, GenBank Umbrella project PRJNA422585 (https://www.ncbi.nlm.nih.gov/bioproject) (Table 1). Microorganisms 2019,7, 556 3 of 14 Table 1. Bacterial strains used in this study. Phage host range determined by spot test and efficiency of plating (EOP). Strain ST Spot EOP Spanish Hospital Where the Strain Was Isolated Ab105_GEIH-2010 2+/−1 Hospital Universitario Virgen del Rocío (Seville, Spain) Ab192_GEIH-2000 2+/−0.22 Hospital Universitario Virgen del Rocío (Seville, Spain) Ab404_GEIH-2010 80 +0.0002 Hospital Dr. Molines (Valencia, Spain) Ab166_GEIH-2000 2+/−- Hospital Universitario Virgen del Rocío (Seville, Spain) Ab177_GEIH-2000 2+1.55 Hospital Universitario Virgen del Rocío (Seville, Spain) Ab13_GEIH-2010 79 - - Hospital Santiago de Compostela (Santiago de Compostela, Spain) Ab09_GEIH-2010 297 - - Hospital Santiago de Compostela (Santiago de Compostela, Spain) Ab160_GEIH-2000 2 - - Hospital Universitario Virgen del Rocío (Seville, Spain) Ab155_GEIH-2000 2 - - Hospital Universitario Virgen del Rocío (Seville, Spain) Ab05_GEIH-2010 186 - - Hospital A Coruña (A Coruña, Spain) Ab22_GEIH-2010 52 - - Hospital Pontevedra (Pontevedra, Spain) Ab421_GEIH-2010 2 - - Hospital Insular (Gran Canaria, Spain) Ab77_GEIH-2000 2 - - Hospital Universitario Ramon y Cajal (Madrid, Spain) Ab141_GEIH-2000 179 - - Complejo Hospitalario Toledo (Toledo, Spain) Ab217_GEIH-2010 2 - - Hospital Reina Sofía (Cordoba, Spain) Ab235_GEIH-2010 2 - - Hospital Marqués de Valdecilla (Santander, Spain Ab37_GEIH-2010 2 - - Hospital Virgen del Rocío (Seville, Spain) Ab222_GEIH-2000 181 - - Hospital Bellvitge (Barcelona) Ab461_GEIH-2010 2 - - Hospital del Mar (Barcelona, Spain) Ab173_GEIH-2010 88 - - Hospital San Agustín (Avilés, Spain) ST: Sequence Type. Spot test: (+) clear spot; (+/−) turbid spot; (-) no spot. 2.2. Obtaining the Lytic Phage Mutant The bacteriophage sequence Ab105-2phi (Genbank: KT5880759) detected in clinical strain A. baumannii Ab105GEIH_2010 was analyzed and the CI gene identified as ORF 17. The CI gene was deleted by double homologous recombination with the suicide vector pMo130TelR [ 18 , 19 ]. The primers were first designed for the amplification of the flanking regions (1000 bp) of the CI gene. These regions were amplified by PCR and ligated and cloned into the pMo130telR vector (Table 2). This construction was transformed in Escherichia coli DH5 α to produce large numbers of the plasmid with the gene flanking regions. The plasmid was purified and transformed in the A. baumannii Ab105 clinical strain by electroporation, and incubated for two hours at 37 ◦ C without antibiotic, thereby producing a recombinant wild type with the mutated gene integrated in its genome. Finally, the mutants were selected in the presence of kanamycin (50 µ g/mL). In order to isolate only those mutants with the CI gene deletion in the chromosome, the plasmid loss was induced in the absence of kanamycin, and the recombinant clones were selected in the presence of 15% sucrose. In order to isolate the mutant phage Ab105-2phi ∆ CI from the bacterial clones by which the CI gene was deleted, a selected clone was incubated in LB broth, which is supplemented with mitomycin (10 ug/mL) to induce release of the phages. The supernatant was collected, treated with chloroform, and filtered (20 µ m). The filtered supernatant was used to infect the clone without the phage, and plaques were obtained by the agar Microorganisms 2019,7, 556 4 of 14 overlay method [ 20 ]. A clear plaque was isolated by PCR and sequencing was conducted to confirm the correct deletion of the CI gene. Table 2. Primers used to delete the CI gene. Primer Sequence Strain/Plasmid UPCI [NotI]Fw GGGGCGGCCGCTGAAGAATTCATCACTTG Ab105_GEIH-2010 UPCI[BamHI]Rev GGGGGATCCCGTTACTTCTATCGGAAT Ab105_GEIH-2010 DWCI[BamHI]Fw GGGGGATCCATTAAGGTTTTAGGTGAT Ab105_GEIH-2010 DWCI[SphI]Rev GGGGCATGCTAAATCATCCAAATCGAC Ab105_GEIH-2010 CIFw ATGGACAAATTTATGGCTAC Ab105_GEIH-2010 CIRev TAACTTTTTCTAACACGCT Ab105_GEIH-2010 IntCIFw AAAGCGCTGCCAACTTTT Ab105_GEIH-2010 IntCIRev CAACAGATTCATCCTCAT Ab105_GEIH-2010 pMo130TelRFw ATTCATGACCGTGCTGAC pMo130TelR pMo130TelRRev CTTGTCTGTAAGCGGATG pMo130TelR Plasmid Description Origin pMo130TelR Suicide plasmid, xylE + ,sacB + , km R , Tel R[19] Restriction enzyme sites are shown in italics. A clone of strain Ab105GEIH_2010, induced with mitomycin, was isolated and excision of the phage was confirmed by PCR of the CI gene and the flanking regions (1000 pb each region) of the gene. 2.3. Host Range and Efficiency of Plating Analysis The host range of the lytic mutant phage Ab105-2phi ∆ CI was established by applying the spot test [ 21 ] to the 20 clinical strains of A. baumannii under study. Efficiency of Plating (EOP) was established as the ratio between the test strain titre and the host strain titre [22]. 2.4. Transmission Electron Mmicroscopy (TEM) and Live-Cell Imaging A broth culture of strain Ab177_GEIH-2000 was infected with the lytic mutant phage Ab105-2phi ∆ CI. The lysates were centrifuged at 3400 × gfor 10 min and the supernatant was filtered through a 0.22 µ m filter (Merck Millipore, Ltd. Tullagreen, Carrigtwahill, Co Cork, Ireland). NaCl was added to a final concentration of 0.5 M, and the suspensions were mixed thoroughly and left on ice for 1 h. The suspensions were centrifuged at 3400 × gfor 40 min at 4 ◦ C, and the supernatants were transferred to sterile tubes. PEG 6000 (10% w/v) was added, dissolved, and incubated overnight at 4 ◦ C. Bacteriophages were then precipitated at 3400 × gfor 40 min at 4 ◦ C and resuspended in SM buffer (0.1 M NaCl, 1 mM MgSO4, 0.2 M Tris-HCl, pH 7.5) [ 23 ]. The samples were negatively stained with 1% aqueous uranyl acetate before examination by electron microscopy. Live-cell imaging was carried out by time-lapse microscopy after initial adsorption of the mutant lytic phage Ab105-2phi ∆ CI to the clinical strain Ab177_GEIH-2000 at 37 ◦ C in agar slices, which were placed directly between stainless steel O-rings. The use of extracellular DNA markers enabled the lysis of more than 300 bacteria to be monitored in real time. 2.5. Adsorption Curve, One Step Growth Curve, and Infection Curve An overnight culture of A. baumannii clinical strain Ab177_GEIH-2000 was diluted 1:100 in LB broth, and incubated at 37 ◦ C at 180 rpm, until an early logarithmic phase, i.e., at an optical density of 0.2 (OD 600nm). At this point the culture was infected with the lytic mutant phage Ab105-2phi ∆ CI at Microorganisms 2019,7, 556 5 of 14 a multiplicity of infection (MOI) of 0.1. The adsorption curve and the one step growth curve were determined after growing the phage in LB, supplemented with CaCl 2 , as previously described [ 20 , 24 ]. In the one step growth curve, the latent period was defined as the interval between adsorption of the phages to the bacterial cells and the release of phage progeny. The burst size of the phage was determined as the ratio of the final number of free phage particles to the number of infected bacterial cells during the latent period [22]. An early exponential culture of the strain Ab177_GEIH-2000 in LB, supplemented with CaCl 2 , was infected with the lysogenic phage Ab105phi2 and the mutant lytic phage Ab105phi2 ∆ CI at different MOIs (0.1, 1 and 10), and the corresponding infection curves were constructed. The phage cultures were maintained at room temperature during the adsorption period and then incubated at 37 ◦ C and 180 rpm for 6 h. The optical density was measured at intervals of one hour during this period. 2.6. Frequency of Occurrence of Phage Resistant Bacteria Phage resistant mutants were produced as previously described [ 25 ]. To determine the emergence of phage resistant mutants, an overnight culture of strain Ab177_GEIH-2000 was diluted 1:100 in LB and grown to an OD600nm of 0.6–0.7. An aliquot of 100 µ L of the culture containing 10 8 colony forming units (CFU)/mL was serially diluted, and each dilution mixed with 100 µ L of 10 9 plaque forming units (PFU)/mL, and plated by the agar overlay method [ 20 ]. The plates were incubated at 37 ◦ C for 18h and the number of CFUs was counted. The same procedure was used to produce phage resistant mutants in the presence of the antibiotics doxycycline, meropenem, or imipenem, which were added to the plates, each at 25% of the minimum inhibitory concentration (MIC). The frequency of occurrence of phage resistant mutants and phage-antibiotic resistant mutants was calculated by dividing the number of resistant bacteria by the total number of sensitive bacteria. 2.7. Antimicrobial Activity of the Mutant Lytic Phage Ab105-2phi∆CI in Biofilm An overnight culture of the A. baumannii clinical strain Ab177_GEIH-2000 was diluted 1:100 and used to inoculate 100 µ L of LB in some wells of a 96 multi-well plate. The plate was maintained at 37 ◦C in static conditions for 4 h. The medium was then discarded and the wells were washed twice with PBS before 100 µ L of fresh LB was added. After 24 h at 37 ◦ C, the medium was again discarded and the wells were washed with PBS, and filled with 90 µ L of SM buffer, then 10 µ L of phage Ab105-2phi ∆ CI (10 7 PFU/mL) was added. SM buffer (100 µ L) was added to control wells. The plates were then incubated at 37 ◦ C for 24 h. Finally, the supernatant was discarded and the wells were washed with PBS. Half of the wells were used to quantify the CFUs and the other half were used to quantify the biofilm. PBS (100 µ L) was added to the wells used to quantify the CFUs and the plates were agitated for 5 min and sonicated for another 5 min. The suspension was serially diluted and plated on LB plates. For quantification of the biofilm, 100 µ L of methanol was added to each well and discarded after 10 min . Once the methanol had completely evaporated, 100 µ L of crystal violet (0.1%) was added and discarded after 15 min. Finally, the wells were washed with PBS before the addition of 150 µL of acetic acid (30%), and the absorbance was measured at OD 595 nm. 2.8. Antimicrobial Activity in Combination with Antibiotics A bacterial killing assay was constructed to determine the synergy of phage Ab105-2phi ∆ CI in combination with meropenem, imipenem and doxycycline at 1/8 and 1/4 of the respective MICs (meropenem 32 µg/mL, imipenem 16 µg/mL and doxycycline 64 µg/mL). An overnight culture of the tested strain was diluted at 1:100 in LB broth supplemented with 10uM CaCl 2 and incubated at 37 ◦ C and 180 rpm until the culture reached an early exponential phase at 0.2 OD (600nm). At this point, antibiotic and the Ab105-2phi ∆ CI phage were added to the culture. The flasks were maintained at room temperature during the adsorption period before being incubated at 37 ◦ C and 180 rpm for 24 h . Aliquots were removed after 6 h and 24 h and were serially diluted and plated in LB plates for subsequent counting of CFU. Microorganisms 2019,7, 556 6 of 14 2.9. Galleria mellonella Survival Assay The Galleria mellonella model used was an adapted version of a previously developed model also used to study bacteriophage therapy [ 26 , 27 ]. The procedure was as follows: twelve G. mellonella larvae, acquired from TruLarvTM (Biosystems Technology, Exeter, Devon, UK), were each injected in the left proleg with 10 µ L of a suspension of A. baumannii Ab177_GEIH-2000, diluted in sterile phosphate buffer saline (PBS) containing 1 × 10 5 CFU ( ± 0. 5 log). The injection was performed with a Hamilton syringe (volume 100 µ L) (Hamilton, Shanghai, China). One hour after infection, the larvae were injected in the right proleg with 10 µ L of the lytic mutant phage Ab105-2phi ∆ CI, at MOI 10, in combination with meropenem at 1/4 MIC and imipenem at 1 4 MIC. The controls included 10 µ L of the lytic mutant phage Ab105-2phi ∆ CI at MOI 10, or meropenem at 1/4 MIC and imipenem at 1/4 MIC. The injected larvae were placed in Petri dishes and incubated in darkness at 37 ◦ C. The number of dead larvae was recorded after 72 h. The larvae were considered dead when they showed no movement in response to touch [26]. The survival curves for the in vivo G. mellonella infection model were constructed using GraphPad Prism v.6 (San Diego, CA, USA), where the data were analyzed using the Log-Rank (Mantel-Cox, City, State if USA, Country) test. The data were expressed as mean values, and the differences were considered statistically significant at p<0.05. 3. Results 3.1. Obtaining the Lytic Mutant of the Phage Ab105phi-2∆CI After deleting the CI gene from the temperate phage Ab105-phi2, as previously reported in Salmonella [ 17 ], we obtained a lytic mutant, designated Ab105-phi2 ∆ CI, which produced characteristic clear lytic plaques. This is in contrast with the turbid plaques produced by the temperate Ab105-phi2 phage (Figure 1A1). PCR of the DNA, isolated from the Ab105-2phi ∆ CI phage, confirmed the deletion of the CI gene. PCRs were conducted with the CI genes and combinations of these primers with those of the flanking regions, confirming that no amplification was obtained. PCRs with the primers (UPCI[NotI]Fw/DWCI[SphI]Rev) of the flanking regions of the gene CI were also conducted, and the expected region of 2000 pb was obtained (size without the CI gene). Finally, this amplicon was sequenced and the CI gene deletion was confirmed. Excision of the phage was also confirmed in a clone induced with mitomycin, as no positive PCR were obtained with the CI primers or with the flanking region primers. Infection curves for the temperate phage Ab105-2phi and the lytic mutant phage Ab105-2phi ∆ CI were constructed and compared, showing that the lytic mutant killed the culture at all MOI levels tested, as reflected by a large decrease in the optical density. Although, a reduction in growth was observed when the culture was infected with the lysogenic phage Ab105-2phi, the decrease was less than with the lytic mutant. In both cases, the reduction in growth was first observed at MOI 10, but regrowth was also first observed at this MOI, probably due to the emergence of resistance (Figure 1B). Microorganisms 2019,7, 556 7 of 14 Microorganisms 2019, 7, x 7 of 13 Figure 1. Graphical representation of the Ab105-2phiΔCI phage. The ORF and direction of transcription are indicated by arrows. (A1) The protein functions are indicated in different colours, and the GC content and GC skew are shown as pink and green circles respectively. (A2) TEM image of the mutant lytic phage Ab105-phi2ΔCI and mutant lytic phage Ab105-phi2ΔCI attached to the cell surface. (B1) Infection curves for the lysogenic phage Ab105-2phi and (B2) the mutant lytic phage Ab105-2phiΔCI. (C) One step growth curve of the mutant lytic phage Ab105-phi2ΔCI (L: Latent period; B: burst size). Mutant lytic phage Ab105-phi2ΔCI antibiofilm activity on the biofilm produced by the clinical strain of A. baumannii Ab177_GEIH-2000. (D1) Reduction in the biofilm and reduction in the number of CFUs present in the biofilm after treatment with the mutant lytic phage Ab105-phi2ΔCI. Figures B, C and D show the mean values +/− SD from three independent assays. Statistically significant differences (p < 0.05) were determined by t-Student test (GraphPad Prism v.6). 3.5. Determination of the Emergence Rate of Phage Resistant Mutants Strain Ab177_GEIH-2000 was resistant to meropenem, imipenem and doxycycline (MICs: Meropenem 32 μg/mL, imipenem 16 μg/mL, and doxycycline 64 μg/mL). In all cases the Figure 1. Graphical representation of the Ab105-2phi ∆ CI phage. The ORF and direction of transcription are indicated by arrows. ( A1 ) The protein functions are indicated in different colours, and the GC content and GC skew are shown as pink and green circles respectively. ( A2 ) TEM image of the mutant lytic phage Ab105-phi2 ∆ CI and mutant lytic phage Ab105-phi2 ∆ CI attached to the cell surface. (B1) Infection curves for the lysogenic phage Ab105-2phi and ( B2 ) the mutant lytic phage Ab105-2phi ∆ CI. ( C ) One step growth curve of the mutant lytic phage Ab105-phi2 ∆ CI (L: Latent period; B: burst size). Mutant lytic phage Ab105-phi2 ∆ CI antibiofilm activity on the biofilm produced by the clinical strain of A. baumannii Ab177_GEIH-2000. ( D1 ) Reduction in the biofilm and reduction in the number of CFUs present in the biofilm after treatment with the mutant lytic phage Ab105-phi2 ∆ CI ( D2 ). Figures B, C and D show the mean values +/ − SD from three independent assays. Statistically significant differences (p<0.05) were determined by t-Student test (GraphPad Prism v.6). Microorganisms 2019,7, 556 8 of 14 3.2. Morphology and Host Range of the Lytic Mutant Phage Ab105-phi2∆CI The lytic mutant Ab105-2phi ∆ CI was isolated and the virion morphology was observed by TEM, revealing that this phage has the typical structure of the Siphoviridae as the wild type phage Ab105-phi2 [28]. All plaques were transparent and about 1mm in diameter (Figure 1A2). The lytic spectrum of activity of the mutant phage Ab105-2phi ∆ CI covered 25% of the clinical strains of A. baumannii tested. The strain Ab177_GEIH-2000 yielded the highest EOP (1.55) (Table 1). This strain was thus selected for further assays. 3.3. Adsorption and One Step Growth Curve Both, the adsorption and the one step growth curve were established using host strain Ab177_GEIH-2000, as the EOP of this strain was the most appropriate and also because this strain does not have complete prophages, as previously determined [ 28 ]. The adsorption time (12 min) was determined in order to establish the one step growth curve, which revealed a latent period of 30 min and a burst size of approximately 32 ±2 PFU per infected cell (Figure 1C). 3.4. Antimicrobial Activity of the Mutant Lytic Phage Ab105-2phi∆CI on Biofilm Biofilm was produced with the clinical strain of A. baumannii Ab177_GEIH-2000 susceptible to the mutant lytic phage Ab105-phi2 ∆ CI. The treatment of the biofilm with 10 7 PFU of this lytic mutant phage caused a statistically significant reduction in the biofilm biomass. The antimicrobial activity against the biofilm forming bacteria was confirmed by a decrease in the CFU, quantified in the presence of the mutant lytic phage (Figure 1D). Finally, the lytic activity of the mutant phage can be observed in Video 1 (Supplementary Materials). 3.5. Determination of the Emergence Rate of Phage Resistant Mutants Strain Ab177_GEIH-2000 was resistant to meropenem, imipenem and doxycycline (MICs: Meropenem 32 µ g/mL, imipenem 16 µ g/mL, and doxycycline 64 µ g/mL). In all cases the combination of the phage and antibiotic reduced the rate of emergence of phage-resistant mutants, relative to the rate of resistant mutants in the presence of the phage alone (Table 3). Table 3. Frequency of phage resistant mutants. Phage resistant mutant frequency in the presence of the combination of doxycycline, meropenem and imipenem at 1 4 MIC in combination with lytic mutant phage Ab105-2phi∆CI (MOI 10) was calculated. Sample Frequency of Phage Resistant Mutants Ab105-2phi∆CI 1.70 ×10−6 Ab105-2phi∆CI +Doxycycline 1.31 ×10−7 Ab105-2phi∆CI +Meropenem 2.10 ×10−7 Ab105-2phi∆CI +Imipenem 1.90 ×10−7 3.6. Effect of the Combination of Phage and Antibiotic on the Bacterial Killing Assays Bacterial killing assays were constructed for A. baumannii clinical strain Ab177_GEIH-2000 in the presence of a combination of the lytic mutant phage Ab105-2phi ∆ CI at different MOIs ( 0.1, 1, and 10 ) and three antibiotics (at 1/4 and 1/8 MIC) to which Ab177_GEIH-2000 is resistant: Meropenem, imipenem, and doxycycline (Figure 2). Microorganisms 2019,7, 556 9 of 14 Microorganisms 2019, 7, x 9 of 13 Figure 2. Bacterial killing assays for A. baumannii clinical strain Ab177_GEIH-2000 determined using the mutant lytic phage Ab105-2phi∆CI at MOI 1 and MOI10 in combination with meropenem at (A1) 1/8 MIC and (A2) 1/4 MIC;(B1) imipenem at 1/8 MIC and (B2)1/4 MIC, and (C1) doxycycline at 1/8 MIC and (C2)1/4 MIC. Values shown in the graphs are the means +/− SD from three independent assays. The curves obtained for the lytic mutant phage controls showed that Ab177_GEIH-2000 grew at control rates after 24h, due to the acquisition of phage resistance. However, the growth was higher at MOI 10, than at MOI 1 after 6 h, probably because resistance emerges faster at this MOI than at lower MOI. 3.7. Galleria mellonella Survival Assays in the Presence of Meropenem and Imipenem in Combination with the Lytic Mutant Phage Ab105-phi2ΔCI The combinations of antibiotic and the phage Ab105-2phiΔCI, that resulted in the reduction of the CFU of Ab177_GEIH-2000 at 24h in vitro were assayed in a G. mellonella (wax moth) larvae survival model (Figure 3). When the infected larvae were treated with imipenem and mutant lytic phage Ab105-2phiΔCI, the survival rate was found to be statistically significantly higher than the larvae treated with the antibiotic or the phage alone and of untreated larvae (p < 0.05). Similar results were obtained for meropenem but in this case. Although, larval survival was higher after the combinatory treatment than after phage only or no treatment, the difference relative to meropenem alone was not statistically significant (p = 0.2183). This was probably due to the higher MIC of meropenem than of imipenem (32 μg/mL versus 16 μg/mL) for the Ab177_GEIH-2000 strain, indicating the need to administer greater amounts of mutant lytic phage Ab105-2phiΔCI. Figure 2. Bacterial killing assays for A. baumannii clinical strain Ab177_GEIH-2000 determined using the mutant lytic phage Ab105-2phi ∆ CI at MOI 1 and MOI10 in combination with meropenem at ( A1 ) 1/8 MIC and ( A2 ) 1/4 MIC;( B1 ) imipenem at 1/8 MIC and ( B2 )1/4 MIC, and ( C1 ) doxycycline at 1/8 MIC and (C2)1/4 MIC. Values shown in the graphs are the means +/−SD from three independent assays. A reduction in the number of CFU was observed with the phage at both MOI 1 (4 log) and MOI 10 (1 log) after 6 h, but no differences from the control were observed after 24 h. The reduction was even greater when the phage was combined with meropenem or imipenem (both carbapenems). For meropenem plus phage, a synergistic effect was observed after 6 h for all combinations (from 4 log to 7 log CFU/mL). The growth of the A. baumanni strain was similar to control levels after 24 h for all concentrations of meropenem plus phage at MOI 1. The synergistic effect was only maintained with the combination of meropenem at 1/4 MIC and phage Ab105-2phi ∆ CI at MOI10, yielding a difference in bacterial growth of 6 log CFU/mL, relative to that corresponding to the meropenem control (Figure 2A1,2A2). As with meropenem, the combination of different concentrations of imipenem and the lytic mutant phage had a synergistic effect after 6 h in all cases, with a reduction in bacterial growth of 8 log CFU/mL for the combination of imipenem at 1/4 MIC, plus phage, and 4 log CFU/mL for the combination of imipenem at 1/8 MIC plus phage. The synergistic effect was maintained for 24 h in the combinations of imipenen at 1/4 MIC, with phage at MOI1 and MOI10, but not in the combinations of imipenem at 1/8 MIC and both phage concentrations (Figure 2B1,2B2). No synergistic effects were observed with doxycycline, and the combination had no more effect than the phage alone at MOI 1. However, when the combinations included the phage at MOI 10,