Health-care associated infections (HAIs) are a major issue in the medical field. Indeed, according to Centers for Disease Control and Prevention (CDC), each year in the USA, 3 % patients develops nosocomial infections [1]. These infections are responsible for 72 000 deaths per year in the USA [2]. To decrease the risk of infections, antibiotics are delivered to patients. Today, the use of conventional antibiotics is the major way to prevent and to fight against bacterial infections. However, in the last decades, some limitations to their use have appeared. Some strains of bacteria developed resistance to conventional antibiotics and in 2019 a study on 25 countries and territories showed that methicillin-resistant S. aureus represent about 12% of the S. aureus strain and antibiotic resistant E. coli is about 36% of the E. coli strain [3]. The CDC estimates that infections associated with multi-antibiotic resistant bacteria caused more than 29 000 deaths in 2020 [1]. Thus, conventional antibiotics are no longer effective on some strains that are more and more resistant to them [4]. Thus, finding new strategies and new molecules which would not induce bacteria mutation into resistant strains is very important for the medical field. Different strategies to decrease the use of conventional antibiotics are studied such as dynamic therapies and the design of new antimicrobial candidates [5].

Antimicrobial peptides (AMPs) are, for example, a class of molecules studied to replace or decrease the use of conventional antibiotics. AMPs are natural small molecules from the innate immunity. Their role is to fight against first microbial invasion [6]. These peptides are most of the time cationic and amphipathic and present a broad spectrum of antimicrobial properties, namely antibacterial, antifungal and antiviral activities [7,8]. A lot of studies detailed that the antibacterial activity of AMPs was due to the interaction between positive charges of the AMPs and negative charges of the outer membranes of bacteria. Indeed, due to these electrostatic interactions, the polycation can be adsorbed to the bacterial membrane. Then, other interactions such as hydrophobic interactions lead to the deformation of the membrane and the formation of pores [[9], [10], [11]].

Among AMPs, antimicrobial peptides with branched globular structures around a core, also called dendrimers, are highly studied to fight against multidrug resistant bacteria [[12], [13], [14]]. For example, Siriwardena et al. showed the strong antimicrobial properties of an AMP dendrimer due to its higher density of functional groups in interaction with the surface of the bacteria [14]. Besides, AMP dendrimers present a 3D structure which seems to favor the access of all the functions of the molecules to the bacterial membrane. Thus, these molecules can be effective for bacterial inhibition and other properties can be matched to focus on several applications, such as biofilm prevention, cancer treatment, intracellular delivery, cell transfection … by adding active molecules on the branched structure [[15], [16], [17]]. Moreover, the use of dendrimeric structures enables the optimization of their stability by avoiding their degradation by protease for instance, leading to better antimicrobial activity [18].

We demonstrated recently that homopolypeptides as polycations are a promising tool to fight against bacteria and can be used to design antimicrobial coatings [19]. These polymers are positively charged, enabling them to interact with and destabilize negatively charged bacterial membranes. Finally, these polycations can be associated to a negatively charged polymer to develop a coating with longer release of antibacterial properties [20]. In these studies, strong antibacterial properties were conferred to the coatings by the use of a linear homopolypeptide, poly (l-arginine) (PAR). pKa of arginine is of 12.5 and thus all residues will be protonated in physiological conditions (i.e. pH from 2 to 8.5 in the body). This is one of the main advantages of the polyarginine peptides, they are always charged and thus they strongly interact with negatively charged bacterial membranes and they finally destabilize them. By using only PAR associated with hyaluronic acid in a layer-by-layer buildup process, a thin coating can be produced on all surfaces. This leads to a “multilayer” film with the ability to prevent colonization by any kind of bacteria. Moreover, we showed that the size of the PAR chain (number of arginine residues per chain) play a key role on the antibacterial properties of the coating [21]. Indeed, the antibacterial properties are related to the diffusion of the PAR chains inside the multilayer of polyelectrolytes. This mobility determines their ability to be available on the surface and to adsorb on the membrane of the incoming bacteria. Finally, small chains can better diffuse inside the film compared to longer one. However, too small chains produce thinner films. Thus, the antimicrobial activity of the coating is based on a tradeoff between mobility of the PAR chains and their ability to produce thick films. The results show that the PAR with 30 residues (PAR30) is the optimal chain length range to design efficient coatings.

From these results, the next question is if the size of the PAR chains is the unique parameter tuning the antibacterial properties or if the 3D structure of the PAR can also influence the antimicrobial properties of this molecule. For example, cyclic antimicrobial peptides seem interesting as some studies present the stronger antibacterial activity but also a higher resistance to degradation of the cyclic structure compared to a linear one [22,23]. Besides, peptide dendrimers are interesting structures and some of them have been described to possess strong antibacterial activities. Thus, we decided to study multiple antigenic peptides (MAPs) of arginine amino acids. MAPs are synthetic branched peptides that can be compared to dendrimers because of their structure. MAPs were already studied for various applications such as diagnostics, antiviral and vaccines strategies but there were only few studies on their antibacterial properties [[24], [25], [26], [27]]. Indeed, MAPs are based on a lysine tree structure which provides the core of the dendrimer structure. In some applications, these structures have shown to improve the immunogenic response and they can also improve the diagnosis performance [27]. Finally, they can be easily tuned and functionalized with various amino acid sequences. In the present study, we designed MAPs based on arginine residues with a lysine core. As lysine possesses two amine groups, two amino acids can be attached to one lysine on the amine groups. In this way, multibranched molecules can be designed through relative simple peptidic synthesis routes. It is then possible to develop MAPs with different numbers of arms depending on the number of lysine residues used in the core. Thereby we can obtain molecules comprising only arginine in the arms but distinguished by their configurations, as presented in Fig. 1. For the nomenclature, the peptide composed of 4 arms with 5 or 8 arginine residues per arm was named respectively R5MAP4 and R8MAP4 (Fig. 1B) and the peptide composed of 8 arms with 5 or 8 arginine residues in each arm was named respectively R5MAP8 and R8MAP8 (Fig. 1C). A cyclic peptide containing 10 arginine residues was also produced (cyclo (R10)). A linear peptide called R20 was added as a control and was composed of 20 arginine residues (Fig. 1A). In our previous studies [20,21], a linear peptide with 30 arginine residues (PAR30) was the most widely used arginine homopolypeptide but in the present study our aim was to compare the R5MAP4 with a linear peptide containing the same number of arginine residues, instead of PAR30, so R20 was selected.

In this study we evaluated antimicrobial activities of these MAPs towards Gram positive and Gram negative bacteria and we compared these activities with that of a linear PAR composed of 20 arginine residues (R20). Biocompatibility, pro- or anti-inflammatory properties were also evaluated. Then, to better understand the experimental results obtained and the mechanism involved, molecular simulation was performed which monitors the interactions between the different peptides and the bacterial membrane. This method enables to provide a good understanding of the mechanism involved between peptides and bacterial membranes.