pharmaceutics Article A Novel Generation of Tailored Antimicrobial Drugs Based on Recombinant Multidomain Proteins Adrià López-Cano 1 , Neus Ferrer-Miralles 2,3,4 , Julieta Sánchez 2,3 , Jose Vicente Carratalá 2,3 , Xavier Rodriguez Rodriguez 4,5 , Imma Ratera 4,5 , Judith Guasch 4,5,6 , Oscar Q. Pich 7 , Paula Bierge 7 , Cristina Garcia-de-la-Maria 8 , Jose M. Miro 8,9 , Elena Garcia-Fruitós 1,† and Anna Arís 1, *,†,‡ on behalf of the FUNCATH Investigators 1 Department of Ruminant Production, Institute of Agriculture and Food Research (IRTA), Caldes de Montbui, 08140 Barcelona, Spain; adrialopezcano@gmail.com (A.L.-C.); elena.garcia@irta.cat (E.G.-F.) 2 Institute for Biotechnology and Biomedicine, Autonomous University of Barcelona, Bellaterra, 08193 Barcelona, Spain; neus.ferrer@uab.cat (N.F.-M.); jsanchezqa@gmail.com (J.S.); josevicente.carratala@uab.cat (J.V.C.) 3 Department of Genetics and Microbiology, Autonomous University of Barcelona, Bellaterra, 08193 Barcelona, Spain 4 Bioengineering, Biomaterials and Nanomedicine Networking Biomedical Research Centre (CIBER-BBN), Bellaterra, 08193 Barcelona, Spain; xrodriguez2@icmab.es (X.R.R.); iratera@icmab.es (I.R.); jguasch@icmab.es (J.G.) 5 Institute of Materials Science of Barcelona (ICMAB-CSIC), Campus UAB, Bellaterra, 08193 Barcelona, Spain 6 Dynamic Biomimetics for Cancer Immunotherapy, Max Planck Partner Group, Institute of Materials Science of Barcelona (ICMAB-CSIC), Campus UAB, Bellaterra, 08193 Barcelona, Spain 7 Laboratori de Recerca en Microbiologia i Malalties Infeccioses, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí (I3PT-CERCA), Universitat Autònoma de Barcelona, 08208 Sabadell, Spain; oquijada@tauli.cat (O.Q.P.); pbierge@tauli.cat (P.B.) 8 Infectious Diseases Service, Hospital Clinic-IDIBAPS, University of Barcelona, 08007 Barcelona, Spain; cgarciad@clinic.cat (C.G.-d.-l.-M.); josemaria@miromoreno.org (J.M.M.) 9 CIBERINFEC, Instituto de Salud Carlos III, 28029 Madrid, Spain Citation: López-Cano, A.; * Correspondence: anna.aris@irta.cat; Tel.: +34-93-467-40-40 Ferrer-Miralles, N.; Sánchez, J.; † These authors contributed equally to this work. ‡ Collaborators of the FUNCATH Investigators are indicated in Acknowledgments. Carratalá, J.V.; Rodriguez, X.R.; Ratera, I.; Guasch, J.; Pich, O.Q.; Bierge, P.; Garcia-de-la-Maria, C.; Abstract: Antibiotic resistance has exponentially increased during the last years. It is necessary to et al. A Novel Generation of Tailored develop new antimicrobial drugs to prevent and treat infectious diseases caused by multidrug- or Antimicrobial Drugs Based on extensively-drug resistant (MDR/XDR)-bacteria. Host Defense Peptides (HDPs) have a versatile Recombinant Multidomain Proteins. role, acting as antimicrobial peptides and regulators of several innate immunity functions. The Pharmaceutics 2023, 15, 1068. results shown by previous studies using synthetic HDPs are only the tip of the iceberg, since the https://doi.org/10.3390/ synergistic potential of HDPs and their production as recombinant proteins are fields practically pharmaceutics15041068 unexplored. The present study aims to move a step forward through the development of a new Academic Editor: Ivana Cacciatore generation of tailored antimicrobials, using a rational design of recombinant multidomain proteins based on HDPs. This strategy is based on a two-phase process, starting with the construction Received: 7 February 2023 of the first generation molecules using single HDPs and further selecting those HDPs with higher Revised: 19 March 2023 bactericidal efficiencies to be combined in the second generation of broad-spectrum antimicrobials. As Accepted: 22 March 2023 a proof of concept, we have designed three new antimicrobials, named D5L37βD3, D5L37D5L37 and Published: 26 March 2023 D5LAL37βD3. After an in-depth exploration, we found D5L37D5L37 to be the most promising one, since it was equally effective against four relevant pathogens in healthcare-associated infections, such as methicillin-susceptible (MSSA) and methicillin-resistant (MRSA) Staphylococcus aureus, methicillin- Copyright: © 2023 by the authors. resistant Staphylococcus epidermidis (MRSE) and MDR Pseudomonas aeruginosa, being MRSA, MRSE Licensee MDPI, Basel, Switzerland. and P. aeruginosa MDR strains. The low MIC values and versatile activity against planktonic and This article is an open access article biofilm forms reinforce the use of this platform to isolate and produce unlimited HDP combinations distributed under the terms and as new antimicrobial drugs by effective means. conditions of the Creative Commons Attribution (CC BY) license (https:// Keywords: host defense peptides; antimicrobial; recombinant; nosocomial infections creativecommons.org/licenses/by/ 4.0/). Pharmaceutics 2023, 15, 1068. https://doi.org/10.3390/pharmaceutics15041068 https://www.mdpi.com/journal/pharmaceutics Pharmaceutics 2023, 15, 1068 2 of 17 1. Introduction The discovery of antibiotics led to a golden age in human healthcare, providing a wide range of therapies to cope with bacterial infections [1,2]. As a result of the preva- lent use, and sometimes misuse, of antibiotics, multidrug- or extensively-drug resistant (MDR/XDR)-bacteria have rapidly risen, generating a global health crisis affecting both human and animal health [3,4]. In this context, the search for new antimicrobial compounds has become imperative. Several approaches are under investigation, such as the use of enzymes, probiotics, antimicrobial peptides or bacteriophages, to name a few. Among these options, the Host Defense Peptides (HDPs) or antimicrobial peptides from innate immunity have stood out over others due to their natural versatility [5–7]. HDPs are short peptides, ranging from 12 to 50 amino acids, with cationic and amphiphilic features with a ubiquitous presence in nearly all biological kingdoms [8,9]. These evolutionary conserved molecules have an essential role in the innate immune system, regulating a broad range of immunological responses and modulating several biological signaling pathways related to processes such as wound healing, autoimmune disease and others [10,11]. Likewise, HDPs exhibit broad-spectrum activity against viruses, fungi and bacteria, including MDR/XDR strains [1,12] in both planktonic and biofilm forms since they can interact with the cell mem- brane through permeabilization or other antimicrobial mechanisms [12–14]. Resistance to HDPs could occur through several mechanisms, including changes in the bacterial cell wall or membrane or enzymatic degradation of the peptides, but research has shown that the risk of resistance to HDPs is lower than that of traditional antibiotics, in part because HDPs have a reduced half-life, which, combined with their variety of mechanisms of action, hamper the emergence of new resistances [15]. These peptides are broadly expressed in neutrophils and macrophages, being released during inflammatory responses, where they can act to either directly resolve infections by killing pathogenic bacteria or by coordinating immune responses [13,16]. HDP production has commonly been carried out by chemical synthesis, although recombinant production has already been proven to be an alternative that allows producing these peptides through a scalable and cost-effective process, without limits in peptide length [17,18]. However, when produced in recombinant hosts, HDPs need to be fused to a carrier protein [19] to protect the peptide from host proteases and mask their possible toxic effect on the producer cell [17]. The removal of the carrier protein involves extra steps in the downstream purification and hence there is yield reduction and additional costs [20]. In this scenario, a recent study carried out by Roca-Pinilla et al. demonstrated that the combination of different functional HDP-based domains in a single polypeptide enabled the synthesis of a potent antimicrobial protein without compromising recombinant host viability and without the need of using protein carriers [21]. The present study aims to move a step forward through the development of a new generation of tailored antimicrobials using a rational design of multidomain proteins. This strategy is founded on a two-phase process (Figure 1), starting on the first generation of molecules produced from a library of HDPs fused to the carrier fluorescent protein Green Fluorescent Protein (GFP). After their testing against planktonic and biofilm forms of target pathogens, the best performing HDPs are combined in the second generation of chimeric molecules, where GFP is removed and tactical linkers are included, obtaining highly active and synergic HDP-based multidomain antimicrobial polypeptides. Pharmaceutics2023, Pharmaceutics 15, x1068 2023,15, FOR PEER REVIEW 3 of 17 3 of 1 Figure1.1.Scheme Figure Scheme depicting depicting the the fabrication fabrication approach approach of enhanced of enhanced broad-spectrum broad-spectrum antimicrobial antimicrobials. The 1st generation of HDPs linked to a GFP carrier were evaluated in a triple assay, The 1st generation of HDPs linked to a GFP carrier were evaluated in a triple assay, allowing the allowing th selection of the most promising ones to generate in phase 2, the 2nd generation of antimicrobial selection of the most promising ones to generate in phase 2, the 2nd generation of antimicrobials, devoidofof devoid a non-functional a non-functional carrier, carrier, fullyfully tunabletunable andenhanced and with with enhanced antimicrobial antimicrobial features. features. 2.2.Methods Methods 2.1. Bacterial Strains 2.1. Bacterial Strains Escherichia coli BL21 (DE3) was used for recombinant protein expression. To evaluate Escherichia antimicrobial coli BL21 activity, (DE3) selected the strains was used for methicillin-sensitive were recombinant proteinStaphylococcus expression. To au-evaluat antimicrobial activity, the strains selected were methicillin-sensitive Staphylococcus aureu (MSSA, ATCC-3556), methicillin-resistant Staphylococcus aureus (MRSA, ATCC-33592 methicillin-resistant Staphylococcus epidermidis (MRSE, ATCC-35984) and Pseudomonas ae ruginosa (ATCC-10145). E. coli strains were grown in Luria–Bertani (LB) medium, an Pharmaceutics 2023, 15, 1068 4 of 17 reus (MSSA, ATCC-3556), methicillin-resistant Staphylococcus aureus (MRSA, ATCC-33592), methicillin-resistant Staphylococcus epidermidis (MRSE, ATCC-35984) and Pseudomonas aerug- inosa (ATCC-10145). E. coli strains were grown in Luria–Bertani (LB) medium, and MRSA, MSSA, MRSE and P. aeruginosa were grown in Brain Heart Infusion (BHI) broth (Scharlau, Barcelona, Spain). 2.2. Genetic Construct Design The 1st generation of molecules was based on the mature sequences of lingual antimi- crobial peptide (LAP, Uniprot entry Q28880, V25-K64), human β-defensin 2 (HβD2, Uniprot entry O15263, G24-P64), human β-defensin 3 (HβD3, Uniprot entry P81534, G23-K67), hu- man α-defensin 5 (HD5, Uniprot entry Q01523, A63-R94) and cathelicidin LL-37 (Uniprot entry P49913, L134-S170), fused to the GFP gene through the linker sequence GGSSRSS. The gene for the 2nd generation construct D5L37D5L37 comprised the combination of the repeated HD5 and LL-37 motifs, forming the HD5-LL-37-HD5-LL-37 construct. The gene encoding for D5LAL37βD3 consisted of the HD5, LAP, LL-37 and HβD3 sequences, and the D5L37βD3 construct was identical to D5LAL37βD3 removing the LAP domain. The same linker sequence was used to connect domain–domain sequences in 2nd generation molecules but removing the GFP gene. All constructs were C-terminally fused to a 6 histi- dine (H6)-tag for protein purification and a cysteine. The sequences were codon-optimized for the E. coli platform by GeneArt (GeneArt® , Life technologies, Regensburg, Germany), cloned into pET22b (ampR ) and transformed by heat shock in competent E. coli BL21 (DE3) cells. Sequences of proteins are included in supplementary materials (Table S1). 2.3. Protein Production and Purification Protein production was performed as previously described [22] and purification was performed from soluble fraction of recombinant cultures in all cases except proteins D5L37βD3 and D5LAL37βD3, which were obtained using inclusion bodies solubilization following established protocols [22]. 2.4. Evaluation of Antimicrobial Activity A triple antimicrobial assay was performed combining a broad screening antimi- crobial assay, minimal inhibitory concentration (MIC) analyses and biofilm eradication (determination of Minimum Biofilm Eradication Concentration (MBEC)). Antibiotic MIC determination was included as a control (Figure S2). The broad screening antimicrobial assay was based on the BacTiter-Glo TM Microbial Cell Viability assay (Promega).Briefly s, an overnight (O/N) culture of the selected strain (MRSA, MSSA, MRSE or P. aeruginosa) was reinoculated in 10 mL of fresh BHI broth and grown at 250 rpm and 37 ◦ C until an exponential growth phase was reached (OD600 = 0.4–0.6). Then, 150 µL from the bacterial diluted stock (106 –104 cfu/mL) was centrifuged at 6200× g at 4 ◦ C for 15 min. The super- natant was removed, and the bacterial pellet was resuspended with 150 µL of either 0.01% acetic acid (negative control) or 5 µM of antimicrobial protein treatment in 0.01% acetic acid and disposed in a sterile polypropylene 96-well plate (Costar). The 5 µM concentration was previosly determined with a compilation of previous exploratory experiments (Figure S1). After sample incubation for 5 h at 37 ◦ C, 100 µL were withdrawn and tested with 100 µL of the BacTiter-GloTM reagent following the manufacturer’s instructions. MIC was determined following performance standards for antimicrobial susceptibility testing with slight modifications [23]. Specifically, 10% Mueller Hinton Broth cation- adjusted medium (MHB-II, Sigma-Aldrich, St. Louis, MO, USA) was used to be more appropriate for testing cationic peptides and proteins [24]. Another relevant modifica- tion was the use of BacTiter-GloTM reagent to detect bacterial growth instead of visual observation since the growth was less evident because of the use of the diluted medium. Antibiofilm activity of each antimicrobial construct was assessed on pre-formed MRSA biofilms following the methodology described by Hancock et al. [25]. Pharmaceutics 2023, 15, 1068 5 of 17 2.5. Protein Cytotoxicity Assay Human peripheral blood mononuclear cells (PBMCs) were isolated from buffy coats of adult donors by density gradient centrifugation using Ficoll (Stemcell Technologies, Vancouver, BC, Canada). The buffy coats were provided by “Banc de Sang i Teixits de Barcelona (Spain)” under the approval No. 5099 of the “Ethics Committee on Animal and Human Experimentation” of the Autonomous University of Barcelona. The isolated PBMCs were plated in 96-well plates at a density of 2·105 in 200 µL of RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. The PBMCs were treated for 24 h with several concentrations of D5L37D5L37 in Hepes 20 mM. Then, PBMCs were stained with 0.5 µL of propidium iodide (Merck) during 3 min at room temperature before performing the flow cytometry measurements that were carried out in a CytoFLEX LX U3-V5-B3-Y5-R3-I0 Flow Cytometer (Beckman Coulter, Indianapolis, IN, USA). 2.6. SEM Imaging of Antimicrobial Effects Ultrastructural effects of 1st and 2nd generation constructs were assessed in P. aerugi- nosa and MRSA cultures. Briefly after, an O/N culture of both strains was 100-fold diluted in 10 mM KPi buffer. Then, 500 µL from the diluted bacteria was aliquoted and centrifuged at 6200× g and 4 ◦ C for 15 min. The supernatant was removed, and the bacterial pellet was resuspended with 500 µL of antimicrobial construct at 5 µM in 0.01 % acetic acid. The treat- ments were disposed over coverglasses in a sterile 24-well plate and incubated for 5 min at 37 ◦ C without agitation. After that, the supernatant was withdrawn, and the samples were fixed with 500 µL of 2.5% (v/v) glutaraldehyde (Merck, Rahway, NJ, USA) in 100 mM of phosphate buffer for 2 h at 4 ◦ C. Following this, the coverglasses were washed with 100 mM phosphate buffer and fixed with 1% (w/v) osmium tetroxide-potassium ferrocyanide for 2 h. The samples were washed with miliQ water, dehydrated in a graded series of ethanol (50, 70, 90, 96 and 100% v/v) at RT and desiccated with hexamethyldisilazane (HMDS). Before microscopy observation, samples were metal-coated and then observed in a FESEM Merlin (Zeiss, Jena, Germany) operating at 3 kV. 2.7. Dynamic Light Scattering Measurements The volume size distribution of 1st and 2nd generation molecules was determined in a Zetasizer Pro (Malvern Instruments Ltd., Malvern, UK) by dynamic light scattering (DLS). The protein concentration was adjusted to 1 mg/mL in 0.01 % acetic acid buffer, at pH 3.8. A 100 µL aliquot (stored at −80 ◦ C) was thawed and then centrifuged at 15,000× g for 15 min at 4 ◦ C to remove non-specific aggregates. Further, the supernatants were measured in triplicate, and the average size and polydispersity index (PI) were displayed as mean ± SEM. 2.8. Statistical Analysis Results are expressed as the means of non-transformed data ± standard error of the mean (SEM). Data were obtained in triplicate, and normality was checked using JMP software (SAS Institute Inc.), being transformed when required. The p-values (statistically significant when p < 0.05) and letters correspond to the ANOVA and Tukey test analyses. 3. Results 3.1. First Generation of HDP-Based Antimicrobial Proteins The first generation of antimicrobials was obtained by fusing the codifying region of five different HDPs to the GFP gene and a H6-tag (Figure 2a) and they were successfully produced in the soluble fraction of recombinant E. coli. After IMAC purification, good yields and purity were achieved for all proteins (Table 1). Protein Yield (mg/L) a Purity (%) LAP 23.4 >99 HβD2 3.48 97 Pharmaceutics 2023, 15, 1068 HβD3 1.84 89 6 of 17 HD5 5.84 >99 LL37 1.74 58 Figure2.2.Schematic Figure Schematicrepresentation representationofofboth both1st 1stand and2nd 2ndgenerations generationsofofantimicrobial antimicrobialproteins. proteins.(a)(a)The The 1st generation constructs are constituted from N- to C-terminal by a single HDP-based 1st generation constructs are constituted from N- to C-terminal by a single HDP-based domain domain (LAP, HβD2, HβD3, HD5 or LL37) fused to the GFP gene. (b) The 2nd generation constructs are (LAP, HβD2, HβD3, HD5 or LL37) fused to the GFP gene. (b) The 2nd generation constructs are multidomain proteins combining HD5, LL37 and HβD3 domains (D5L37βD3), combining the last multidomain proteins combining HD5, LL37 and HβD3 domains (D5L37βD3), combining the last three with LAP (D5LAL37βD3) and using HD5 and LL37 tandem repetitions (D5L37D5L37). All three with LAP constructs have(D5LAL37βD3) and usingfor an H6-tag at C-terminal HD5 and purification protein LL37 tandem repetitions (D5L37D5L37). All purposes. constructs have an H6-tag at C-terminal for protein purification purposes. The antimicrobial activity of first generation molecules was evaluated against both Table 1. Antimicrobial Gram-positive MSSA,protein MRSA yield and(mg L−1 culture) MRSE and theand purity (%) of soluble Gram-negative bacteriaLAP, HβD2, HβD3, P. aeruginosa. The HD5 and of LL37. a Yields calculated after protein purification. testing antimicrobial activity was carried out in three steps: (1) wide screening antimi- Protein Yield (mg/L) a Purity (%) LAP 23.4 >99 HβD2 3.48 97 HβD3 1.84 89 HD5 5.84 >99 LL37 1.74 58 The antimicrobial activity of first generation molecules was evaluated against both Gram-positive MSSA, MRSA and MRSE and the Gram-negative bacteria P. aeruginosa. Pharmaceutics 2023, 15, 1068 7 of 17 Pharmaceutics 2023, 15, x FOR PEER REVIEW 7 of 18 The testing of antimicrobial activity was carried out in three steps: (1) wide screening antimicrobial assay, crobial assay, (2) (2) determination determination of the of the MIC ofMIC HDPs of selected HDPs selected in the in the first stepfirst andstep (3) and bio- (3) biofilm eradication and MBEC determination (Figure 1). The film eradication and MBEC determination (Figure 1). The wide screening assay showed wide screening assay showed that mostthat most active active molecules molecules were those were those based onbased HβD3onand HβD3HD5, and HD5, reducing reducing at least at least 3-log in 3-log pathogens (Figure 3) and 5-log in MRSE and P. aeruginosa all bacterial pathogens (Figure 3) and 5-log in MRSE and P. aeruginosa (Figure 3c,d).3c,d). in all bacterial (Figure The The LAPLAPandand HβD2-based HβD2-based protein protein activity activity werewere strain-dependent,killing strain-dependent, killingcompletely completely MSSA MSSA 5 3 and P. and P. aeruginosa, aeruginosa, either either at at 10 105 and and 10103 cfu/mL (Figure 3b,d, cfu/mL (Figure 3b,d, respectively), respectively), but but showing showing lower lower performance against the MRSA and MRSE strains (Figure 3a,c, respectively). On On performance against the MRSA and MRSE strains (Figure 3a,c, respectively). the the other hand, the LL-37-based construct only showed mild bactericidal effects against other hand, the LL-37-based construct only showed mild bactericidal effects against MRSA (Figure 3a), MRSE (Figure 3c) and P. aeruginosa (Figure 3d) at 1055cfu/mL and was MRSA (Figure 3a), MRSE (Figure 3c) and P. aeruginosa (Figure 3d) at 10 cfu/mL and was not selected in the next step of the MIC determination. not selected in the next step of the MIC determination. Figure 3. Antimicrobial activity of 1st generation antimicrobials. Antimicrobial activity (reduction Figure 3. Antimicrobial activity of 1st generation antimicrobials. Antimicrobial activity (reduction of of log of cfu/mL) of the different 1st generation constructs against (a) MRSA, (b) MSSA, (c) MRSE log of cfu/mL) of the different 1st generation constructs against (a) MRSA, (b) MSSA,5 (c) MRSE and and (d) P. aeruginosa. The constructs were tested against an initial concentration of 10 cfu/mL (dark (d) P. aeruginosa. The (white constructs were tested against an initial 5 cfu/mL (dark bars) bars) or 103 cfu/mL bars). Data shown are the meanconcentration of a triplicateof±10 SEM. Different letters or 3 10 cfu/mL (white bars). Data shown are ± SEM.and depict statistically significant differences (p