Exploring the functionalization of Ti-6Al-4V alloy with the novel antimicrobial peptide JIChis-2 via plasma polymerization

Abstract This study aimed to characterize the immobilization of the novel JIChis-2 peptide on the Ti-6Al-4V alloy, widely used in the biomedical sector. The antimicrobial activity of JIChis-2 was evaluated in the Gram-negative bacterium E. coli. Its immobilization occurred by inducing the formation of covalent bonds between the N-terminus of the peptides and the surface previously submitted to acrylic acid polymerization via the PECVD technique. Coated and uncoated surfaces were characterized by FTIR, AFM, SEM and EDX. Studies of global and localized corrosion were carried out, seeking to explore the effects triggered by surface treatment in an aggressive environment. Additionally, the ability of the functionalized material to prevent E. coli biofilm formation evidenced that the strategy to immobilize JIChis-2 in the Ti-6Al-4V alloy via PECVD of acrylic acid resulted in the development of a functional material with antibiofilm properties. Graphical Abstract


Introduction
Commercially pure titanium (cp) and its alloys (e.g. Ti-6Al-4V) are the most commonly used metallic biomaterials in dental implants (Panayotov et al. 2015). The use of this alloy is justified by its good resistance to global and localized corrosive processes, mechanical properties, and biocompatibility with the human body (Zhao et al. 2013). Specifically, the biphasic alloy (a þ b) Ti-6Al-4V allows the production of devices with therapeutic properties of high mechanical strength due to its low Young's modulus (101-120 GPa) (Oliver et al. 2019;Nicholson 2020). In addition, titanium-based implantable materials are considered bioinert, where no active interaction exists between these metallic materials and the host tissue (Mutreja et al. 2020). The bioinert nature of these biomaterials can have a positive effect on implantation due to their chemical stability. However, being bioinert can hinder the growth and differentiation of bone cells in the region of the devices, which can lead to surgical failures (Chen et al. 2013).
In this sense, surface functionalization appears as a promising strategy to improve the integration of surrounding tissues and/or to impart the antimicrobial properties of metallic biomaterials (Panayotov et al. 2015). Because of this, one of the effective ways to obtain biocompatible materials is through the immobilization of biofunctional molecules, such as peptides (Chen et al. 2013), proteins (Adden et al. 2006), and liposomes loaded with specific drugs (Kastellorizios et al. 2012). Specifically, functionalization with bioactive peptides on metal surfaces can be performed via the following techniques: Self-Assembled Monolayers (SAMs) of thiol (Huang et al. 2003) or alkyl phosphates (Gawalt et al. 2003;Zorn et al. 2007), silanization (Godoy-Gallardo et al. 2015;Zhou et al. 2017), plasma polymerization (Seo et al. 2010;Vreuls et al. 2010), physical adsorption (Cho et al. 2019), dopamine (Cao et al. 2020) and electrodeposition of (poly(ethylene glycol) (Park et al. 2011).
In the biomedical field, plasma deposition methods are often used to impart specific functional groups to materials and can be used to improve animal cell adhesion to surfaces (Kastellorizios et al. 2012). Plasma polymerization is a sustainable coating method and allows the subsequent immobilization of biomolecules in metallic biomaterials. Thus, the Plasma Enhanced Chemical Vapor Deposition (PECVD) technique can be adopted for the formation of thin films on the metallic surface with specific compositions (Sardella et al. 2002). Seo et al. (2010) used plasma polymerization to deposit a thin film of polyacrylic acid (PAA) on titanium to immobilize the RGD peptide. The polymerization process was used to promote the formation of carboxyl groups (COOH) in the biomaterial, to provide the subsequent covalent bond with the amide group (NH 2 ) of the peptide.
Antimicrobial peptides (AMPs) have the function of protecting humans from various pathogens, including activity against a broad spectrum of Gram-negative and Gram-positive bacteria (Toke 2005;Godoy-Gallardo et al. 2014;Kazemzadeh-Narbat et al. 2021). In the realms of living things, there are 3,425 AMPs reported according to the Antimicrobial Peptide Database (APD), of which 31 of these were immobilized on the surface. They present a membranolytic activity against bacterial cells (Toke 2005;Bhandari et al. 2020;Kazemzadeh-Narbat et al. 2021) and the selectivity for the target membranes is attributed to the presence of acidic phospholipids on their lipid composition (Shai 1999;Kazemzadeh-Narbat et al. 2021). Electrostatic interactions are responsible for the attraction between the peptide and the lipid interface.
The cationic peptides assume a helical structure in the vicinity of the membrane and accumulate on the surface of the bacterial membrane (Seo et al. 2012;Bhandari et al. 2020;Kazemzadeh-Narbat et al. 2021). This accumulation induces imbalance, pore formation, and ultimately cell lysis (Seo et al. 2012;Bhandari et al. 2020).
Among the AMPs there are the Jelleines that make up a family of peptides, found in royal jelly, endowed with high antimicrobial action (Fontana et al. 2004). Royal jelly is an acidic secretion produced in the hypopharyngeal and mandibular glands of young worker bees (Ramanathan et al. 2018). The royal jelly has several other therapeutic properties, such as: antitumor, anti-inflammatory, anti-hypercholesterolemic, and hypotensive, among others (Cornara et al. 2017). Fontana et al. (2004) isolated four peptides from honeybee royal jelly, named Jelleine I, Jelleine II, Jelleine III and Jelleine IV. Thus, three of these peptides (Jelleines I-III) showed antimicrobial activity against a wide spectrum of Gram-positive and Gram-negative bacteria (Fontana et al. 2004). Recently, Petrin et al. (2019) presented a chitosan functionalization strategy with different short-chain amino acid peptides. Among these, the conjugation of JIChis-I, a peptide analogue of Jelleine-I, with chitosan exhibited superior antimicrobial activity to Jelleine-I for E. coli and S. aureus (Petrin et al. 2019) Here, the immobilization of a new analogue of Jelleine I named JIChis-2 (JI-2) on the Ti-6Al-4V surfaces is presented. This strategy aims to develop a material with potential antimicrobial properties and good cell compatibility for applications in dental devices. Firstly, the antimicrobial activity of JI-2 was accessed for the Gram-negative bacterium E. coli. Then, a three-step procedure was adopted to perform the peptide immobilization: (I) coating the Ti-6Al-4V alloy with a polymerized acrylic acid (PAA) via plasma polymerization to form carboxyl groups (II) activation of these groups and (III) covalent immobilization of the JI-2 on the metal surface. The PAA coating has carboxyl groups (COOH), which allows it to covalently bond with the peptide N terminal. In this way, the functionalized and non-functionalized surfaces with the peptide were characterized via Fourier-transform infrared spectroscopy (FTIR), atomic force microscopy (AFM), scanning electron microscopy (SEM), Energy-Dispersive X-ray spectroscopy (EDX). Furthermore, global immersion tests were performed in 0.1 mol L À1 sodium fluoride (NaF) which is an aggressive medium for Ti-6Al-4V alloy. The localized tests were accessed via scanning vibrating electrode technique (SVET) in 0.01 mol L À1 NaF solution at room temperature. The evolution of the corrosion tests was accompanied by morphological analyzes obtained by optical microscopy (OM) and SEM/EDX.

Peptide antimicrobial activity bioassay
The antimicrobial action of the JI-2 peptide was evaluated using the standard broth microdilution method (M€ uller-Hinton) to determine the minimum inhibitory concentration (MIC) according to Souza et al. (2005) and Martins et al. (2020). The microorganism used in this investigation was the Gram-negative bacterium E. coli (ATCC 25922). Initially, E. coli. cells were suspended in saline (0.9% NaCl) solution to a value of 0.5 on the McFarland scale, (1,5x10 8 UFC/mL). From this suspension, 50 mL was inoculated into each well of a 96-well microplate, in which each well contained the JI-2 peptide diluted in the medium in the concentration range of 0,9 -2000 mg mL À1 and 50 mL of M€ uller-Hinton broth. Cell Proliferation Kit I (MTT) (RocheV R ) was applied according to the manufacturer's instructions. After 24 h of E. coli exposure to JI-2, 10 lL of reagent 1 (MTT labelling reagent) was added to a final concentration of 0.5 gL À1 and incubated for 4 h at 37 C in an atmosphere of 5.0 ± 0.2% CO 2 . Soon after, 100 lL of reagent 2 (solubilization buffer) was added and incubated for 24 h at 37 C in an atmosphere of 5.0 ± 0.2% CO 2 . After 24 h, the reading was performed in a spectrophotometer (EnSpire-PerkinElmer) at the following wavelengths: 550 and 600 nm. Thus, the absence of bacterial growth was indicated by the absence of colour. Chloramphenicol was the standard antibiotic used as a positive control and wells containing only the culture medium were used as negative controls. The antimicrobial assay was performed in triplicate and MIC values were determined.

Preparation of the Ti-6Al-4V substrates
The Ti-6Al-4V alloy (1 cm x 1 cm) was subjected to the process of sanding, mechanical polishing and cleaning. For this, silicon carbide (SiC) sandpapers were used according to the following sequence 800, 1200, 2500 and 4000#. Then, the metallic substrates were polished with diamond paste (Buehler Metadi Diamond Suspension) in a 3 lm granulometry. Finally, the samples were washed in ultrasound with distilled water and detergent for 10 min, this process was repeated with isopropyl alcohol and later with distilled water under the same conditions.
Polymerized acrylic acid coating on Ti-6Al-4V alloy via plasma polymerization Polymerized acrylic acid films were deposited over the metallic surfaces to enrich them with COOH functional groups (Ti-6Al-4V/PAA). The polished and clean substrates were inserted into the PECVD custom design system. The deposition equipment is composed of a vacuum chamber of quartz tube coupled to the microwave source of 1000 W, as well needle valves to control gas flux, rotative pump vacuum and Pirani gauge. The residual pressure was 8 Â 10 À3 Torr. The film depositions were conducted using a gas mixture of argon and AA, at partial pressures of 3 Â 10 À1 and 5 Â 10 À1 Torr, respectively, for 6 min.
Structural, morphological and topographical characterization of functionalized and nonfunctionalized surfaces

Structural characterization
Both functionalized and non-functionalized surfaces were structurally characterized by FTIR with an Agilent Technologies Care 600 spectrometer. The spectra were obtained with the aid of an accessory of Infrared Reflection-Absorption Spectroscopy (IRRAS) purchased from PIKE Instruments in the range of 4000 cm À1 to 500 cm À1 , resolution of 4 cm À1 , 30 of incidence and 64 number of scans.

Morphological characterization
AFM was used to obtain topographic mapping and roughness analysis of the studied surfaces. The equipment used was a Shimadzu SPM 970 atomic force microscope. Thus, the AFM operated in dynamic mode using cantilevers purchased from NT-MDT Co. The analysis of the AFM images and the acquisition of surface roughness were made through the Gwyddion software.
SEM (FEG-JEOL-JSM7001F) was used to perform the morphological analysis of the surfaces. For these analyses, the samples were subjected to a metallization process with gold (Au) due to the nature of the functionalized surfaces. The voltage used in the equipment was 20 kV and the morphological images of backscattered and secondary electrons were obtained with the magnification of 1000x and 10000x, respectively. SEM images were analyzed with the aid of ImageJ software. EDX coupled with SEM was also used to analyze the samples under the same conditions as above.

Immersion test in aggressive medium
The corrosive properties of the functionalized and non-functionalized surfaces were evaluated through tests of immersion in an aggressive medium. For this, the materials were exposed to a 0.1 mol L À1 solution of sodium fluoride (NaF) at immersion times of 1 h, 3 h, 10 h, 15 h and 24 h. After the end of each immersion time, the samples were washed with distilled water for 10 min in ultrasound. Subsequently, the morphology of the materials was analyzed through Optical Microscopy (OM) and SEM/EDX. For image processing, ImageJ software was used.

Scanning vibrating electrode assay
SVET measurements were performed in the Applicable Electronics Inc. equipment. The control software was the ASETScience wares, and the maps were constructed in the software QuikGrid version 5.4. The vibrating microelectrode used was a platinum/iridium vibrating probe with a 10 lm diameter of platinum deposit on the tip of the probe, with two auxiliary platinum semi-references electrodes. These measurements were performed using a vibration frequency of 124 Hz in the Y axis, with the distance between the substrate surface and the microelectrode of 100 lm. The analysis area of the sample was delimited with beeswax. The electrolyte solution used was 0.01 mol L À1 NaF solution at pH 2, and the scans were performed every hour, for 15h.

Biofilm formation and inhibition tests
The inhibitory effect of Ti-6Al-4V and Ti-6Al-4V/PAA/JI-2 on bacterial biofilms was performed with E. coli, according to Rocha et al. (2019). E. coli has grown in BHI broth and adjusted to 0.5 McFarland standard for biofilm formation. For that, 100 mL of culture was transferred to a 24-well polystyrene microtiter containing 100 mL of fresh BHI broth (control) and BHI broth containing the metallic materials of study. After incubation for 24 h at 37 C, the culture medium was discarded and the wells were washed three times with saline solution (0.85% NaCl) to remove non-adherent cells. Biofilms were fixed with 200 mL of cold methanol for 15 min. Methanol was removed and the biofilms were dried at room temperature. 100 mL of crystal violet solution (1%) was added and, after 15 min, the plates were washed 3 times with 200 mL of saline solution. Biofilms were dried at room temperature and 200 mL of acetic acid solution (33%) was added for 20 min to solubilize crystal violet. The total biomass of biofilm was measured by absorbance at 600 nm, and the percentage of inhibition of biofilm formation was assessed as follows: 1-(OD treated /OD control ) Â 100. The experiment was performed in triplicate with three independent experiments.

Results and discussion
Antimicrobial activity of JI-2 The JI-2 amino acid sequence was developed in search to obtain an antimicrobial peptide that can be immobilized on both metallic and polymeric surfaces. For this, a spacer consisting of two alanine and one cysteine residue was included in the JIF2WR peptide (Martins et al. 2020). This strategy aimed to take advantage of the potent antimicrobial action arising from the substitutions performed in Jelleine-I to obtain the JIF2WR (Martins et al. 2020). Thus, the inclusion of the spacer at the N-terminus allows the immobilization not to involve the amino acid sequence responsible for the activity of the peptide. Furthermore, the presence of the spacer containing the C residue allows the immobilization to occur through the formation of disulfide bridges between this residue and a properly functionalized surface, increasing the possibilities of immobilization of the same (Castellanos et al. 2017;Petrin et al. 2019).
The minimum inhibitory concentration (MIC) is an in vitro study that allows obtaining the lowest concentration that an antimicrobial compound is capable of inhibiting the growth of a specific microorganism (Kowalska-Krochmal and Dudek-Wicher 2021). Here, this approach was adopted to trace the antimicrobial profile of the JI-2 against E. coli. Some strains of this bacterium are related to the appearance of several infections in implantable devices, in addition to favouring the formation of biofilms (Sharma et al. 2016). Table 1 displays the result obtained for JI-2 and the control, chloramphenicol. JI-2 exhibited a MIC of 445.4 lmol L À1 showing that the peptide has significant antimicrobial action against E. coli.

Structural analysis of functionalized and nonfunctionalized surfaces
This work aims to immobilize the JI-2 peptide, analogous to the antimicrobial peptide Jelleine in the Ti-6Al-4V alloy surfaces. This strategy was used to develop a material with good potential for applications in the dentistry and orthodontic sectors. The peptide was immobilized on the surfaces after the deposition of AA via PECVD. This method consists of using cold plasma discharge to fragment volatile precursor molecules and place them over solid substrates. Thus, polymeric thin films with a structure consisting of a high degree of crosslinking are obtained. The polymerized AA are biocompatible and the presence of carboxyl groups allows the functionalization of materials with different biomolecules (de Giglio et al. 2007), such as peptides (Lopez et al. 2005;Seo et al. 2010), proteins (Ulbricht and Riedel 1998) and heparin loaded liposomes (Kastellorizios et al. 2012). In addition, PAA coating has excellent anti-corrosion properties (de Giglio et al. 2007), acting as a protective barrier on metal surfaces.
FTIR spectroscopy was used in several stages of surface functionalization to monitor the emergence of functional groups after each process: AA deposition and peptide immobilization. Figure 1 displays the spectral set of samples, through which it is possible to observe the appearance of two intense bands at 1702 cm À1 and 2925 cm À1 for Ti-6Al-4V/PAA samples. The first is associated with the C ¼ O functional group, indicating the presence of carboxyl groups in the alloy surface (Seo et al. 2010). The second is associated with the presence of C-H groups and is characteristic of AA monomer (Topala et al. 2009). Moreover, it is possible to notice other bands at 1454 cm À1 referring to C-C and 1374 cm À1 belonging to CH 3 . The appearance of additional bands to the characteristics of the group's carboxyl is related to the deposition method since the polymerization processes using cold plasma can induce the formation of chemical structures, with supplementary vibration frequencies as a consequence of the phenomena of molecular fragmentation and rearrangement in the gas phase (Khelifa et al. 2016). These findings suggest that the polymerization of acrylic acid in the Ti-6Al-4V alloy was performed satisfactorily.
The result obtained for the alloy with the immobilized peptide, Figure 1, exhibits the absorption bands around 1700 cm À1 and 2920 cm À1 , previously observed for Ti-6Al-4V/PAA, which are no longer present on the spectrum. This result suggests that the immobilization of the peptide occurs via a covalent bond between the activated carboxyl groups on the metallic surface and the N-terminus of the peptide. This occurs due to the bond between the atoms belonging to the peptide and the functional groups (C ¼ O) detected in the spectrum of Ti-6Al-4V/PAA. Specifically, the disappearance of the band at 1700 cm À1 reflects the bonds between the amine groups of the peptide terminals with the C ¼ O present on the surface. This same behaviour was reported by Seo et al. (2010) when immobilizing the RGD peptide in pure titanium via plasma polymerization of AA. Furthermore, it is possible to identify the appearance of the band around 2350 cm À1 which is associated with vibration CO 2 (Ni and Hao 2012; Sulaiman 2019) Topography, morphology and roughness of functionalized and non-functionalized surfaces The topography of Ti-6Al-4V, Ti-6Al-4V/PAA and Ti-6Al-4V/PAA/JI-2 surfaces were evaluated by AFM. Figure 2 displays the topographic images of the surfaces before (Figure 2a-c) and after the deposition of acrylic acid (Figure 2d-f) and JI-2 immobilization (Figure 2g-i). Ti-6Al-4V surface topography exhibits some irregularities and grooves that may have been caused by the sanding and polishing process (see Figure 2a-c). After the deposition of the acrylic acid film, it is possible to observe that the surface of Ti-6Al-4V/PAA exhibits a more homogeneous aspect compared to the substrate topographies, Figure 2d-f. This shows that the PAA deposition was able to cover some of the surface irregularities and agrees with the FTIR evidence that acrylic acid deposition was achieved. Notable topographical differences are observed for Ti-6Al-4V/PAA/JI-2 in comparison to Ti-6Al-4V/PAA, Figure 2g-i. The presence of several agglomerates is observed in distinct regions of the surfaces containing the immobilized peptide. The 3D image of this sample shows the presence of some peaks, Figure 2i. These structures are similar to those reported by Hernandez-Montelongo et al. (2018) when immobilizing two peptides (Tet-124 and Tet-124-Br) in a polyethyleneimine (PEI) film. These authors observed some peaks distributed throughout the sample that can be attributed to peptide local accumulation. Roughness parameters are used to describe the characteristics of a given surface in two or three dimensions (Gadelmawla et al. 2002). Mean roughness (Sa), mean square roughness (RMS or Sq) and asymmetry (Ssk) are amplitude parameters obtained in three dimensions, in which the Sa measurement is widely used to describe the surface characteristics of dental implants (L€ oberg et al. 2010). Sa provides the arithmetic mean of the surface height variations analyzed from a midplane. RMS is equivalent to the standard deviation of the height distribution, considered even more sensitive than Sa at larger height deviations (Gadelmawla et al. 2002;Karan et al. 2010;L€ oberg et al. 2010;Sedla cek et al. 2017). The roughness of implantable biomaterials is an important factor for cell adhesion, where rough surfaces improve osseointegration (Seo et al. 2010;Hacking et al. 2012). However, the topographic characteristics due to the increase in surface roughness can be an obstacle for applications in implants and brackets, since they provide an increase in bacterial colonization (Teughels et al. 2006;Godoy-Gallardo et al. 2014;Wang et al. 2018). However, some studies show that the roughness threshold for not influencing bacterial adhesion is 200 nm (Bollen et al. 1996;1997). Here, the roughness values for the modified surfaces were obtained by AFM. The results are shown in Table 2 and are lower than that value, showing that this parameter possibly has a reduced impact on the formation of bacterial biofilms.
The asymmetry parameter quantifies the asymmetry of the height distribution on a given surface around the midplane. In this sense, Ssk is sensitive to valleys or peaks distributed on the surface of the sample (Gadelmawla et al. 2002). Table 2 shows the asymmetry parameters obtained for Ti-6Al-4V, Ti-6Al-4V/PAA and Ti-6Al-4V/PAA/JI-2. These results show that Ti-6Al-4V and Ti-6Al-4V/PAA surfaces present negative asymmetry values (Ssk or Rsk), while Ti-6Al-4V/PAA/JI-2 samples resulted in positive ones. According to Peltonen et al. (2004), the Ssk is zero in a Gaussian distribution, that is, a symmetrical distribution around its midpoint. In distributions of asymmetric character, with negative values of Ssk, the surface is considered porous where valley profiles are predominant. However, the opposite occurs with positive Ssk values, in which peak profiles are dominant (Peltonen et al. 2004). In this sense, it is possible to suggest that the surface of the substrate and the acrylic acid coating have a porous character, since Ssk < 0. Godoy-Gallardo et al. (2014) found that the asymmetry values increased in the modified surfaces, after the functionalization of a titanium surface with the peptide hLf1-11. These authors state that the result obtained is an indication that a peptide layer fills some surface pores. This same effect is verified for the Ti-6Al-4V/PAA/JI-2sample (Ssk > 0 for surface with peptide). This result reveals that the presence of JI-2 affects the structure of the material, which assumes a more heterogeneous structure where randomly distributed peaks can be observed.
The morphology of the surfaces was obtained by using SEM and the backscattered electrons (BSE) images are shown in Figure S1. Regarding the substrate ( Figure S1a and b), the surface with the polymeric coating presented different morphological characteristics. As observed in the AFM topographies, the irregularities caused by the substrate preparation process seem to be covered by AA polymerization. However, it is possible to notice that this covering also presents defective regions, Figure S1c and d. Despite the deposition by PECVD having interesting features, it can present the formation of irregular films (Dudeck et al. 2007). Figure S1e and f shows the morphology of the surface Ti-6Al-4V/PAA/JI-2, in which it is possible to perceive differences in the surface characteristics of the sample in comparison to the other surfaces. In this way, the appearance of agglomerates on the surface of the material is observed. These structures are similar to those observed by AFM in samples that contain immobilized peptides. As this effect was not observed in the images for the other surfaces, especially those containing PAA, this result reinforces that the immobilization of the peptide results in a structure in which the peptides can be agglomerated.
Semiquantitative analysis of the surfaces were performed via Energy-dispersive X-ray Spectroscopy (EDX). Figure 3a presents the EDX spectrum for the Ti-6Al-4V alloy and the presence of Titanium (Ti), Aluminum (Al) and Vanadium (V) elements in terms of % by weight. The EDX spectra for the samples after acrylic acid polymerization and peptide immobilization are shown in Figure 3b and c. As can be seen, Figure 3b allows checking the abundance of carbon (C) present on the surface of Ti-6Al-4V/PAA, which comes from the properties of the polymerized acrylic acid. Thus, the appearance of this element once again verifies the acrylic acid deposition on the metallic surface. The presence of silicon (Si) element was verified in the EDX spectra and, probably, comes from SiC sandpapers, as already discussed by this research group in previous works (Eur ıdice et al. 2020). However, an abundance of oxygen (O) was observed on the surfaces with the Table 2. Roughness parameters of functionalized and nonfunctionalized surfaces.
immobilized peptide, Figure 3c. Given these aspects, the EDX spectra allow confirming again the immobilization of the biomolecule in the Ti-6Al-4V alloy, since O is an element abundantly found in the amino acids of the peptide. Figure 4 makes it possible to highlight and compare the topographic and morphological properties, and chemical composition of the agglomerates in the peptide-containing surfaces. Figure 4 (a-c) shows the comparison between the agglomerates observed by AFM and SEM in different samples. It is worth mentioning that the agglomerates in Figure 4a and c do not have the same magnitude, but they have similar topographical characteristics. Specifically, Figure 4d shows the chemical composition of the selected agglomerate. It reveals that there is a predominance of Ti and C in the surroundings of the structure while an abundance of oxygen can be observed along this structure. This result suggests that after the immobilization process, the peptide is not evenly distributed over the surface. The peptides tend to aggregate randomly in different regions of the samples. This effect can be triggered both by peptide-peptide interactions and by the irregularity of AA deposition on the surface, giving rise to regions of greater accumulation of carboxyl groups. Especially when considering the polyanionic hydrophilic character of the polymerized acrylic acid, which may favour the interaction with the cationic peptide (Stankevich et al. 2017;Bitar et al. 2018;Trino et al. 2018).
Thus, through the results obtained by topographic and morphological analyzes associated with the roughness and asymmetry parameters of the modified and unmodified surfaces, it was verified that the coating of the surfaces with polyacrylic acid is present in the samples of Ti-6Al-4V/PAA and Ti-6Al-4V/PAA/JI-2. In addition, these analyses made it possible to observe that the JI-2 peptide tends to agglomerate on the metallic surface.

Immersion tests in a corrosive medium
Considering that the release of metal ions can cause adverse effects on the organism (Prando et al. 2017), it is worth mentioning that the biocompatibility of biomedical devices is related to the corrosion resistance of these materials. Titanium-based materials have excellent corrosion resistance properties, which occur due to the spontaneous formation of a thin layer of titanium dioxide (TiO 2 Þ on the surface of the material (Sivakumar et al. 2011;Maestro et al. 2021). This layer has a protective character and gives the material excellent biological properties (Lyon 2012;Panayotov et al. 2015). Despite this, these materials are susceptible to uniform and localized corrosion in aggressive environments such as the physiological environment (Trino et al. 2018). To explore the corrosion properties of the surfaces used in this study, immersion tests were carried out in a corrosive medium of 0.1 mol L À1 NaF solution at different immersion times (1, 3, 10, 15 and 24 h). The choice of this solution is attributed to the effect of fluoride ions, which affect the stability of the passive oxide layer of titanium-based materials through the dissolution of Ti (Sivakumar et al. 2011;Prando et al. 2017). Thus, it was possible to analyze the stability of the samples in this medium, considering the differences between the base material and the surfaces coated with PAA and with the peptide at different immersion times.
The OM analyses were used to visualize the surfaces at different immersion times and the results are shown in Figure 5. An overview of the Ti-6Al-4V alloy before immersion tests is displayed in Figures 5a  and b. It is possible to notice that the Ti-6Al-4V surface showed some corroded areas after 3 h of immersion when exposed to an aggressive medium (see Figure 5c and d). In other words, the presence of these corroded areas characterizes a localized corrosion process, such as pitting (Lyon 2012;Akpanyung and Loto 2019). After 24 h of immersion, some pits are observed all over the metal surface, as seen in Figure 5c and d.
These localized corrosion aspects of the non-functionalized surface were analyzed by SEM and EDX. Figure S2 shows the chemical composition of the surface after 3 h of immersion in an aggressive solution. The morphological analysis also reveals the presence of pits, which is in good agreement with the optical microscopy at equivalent immersion times. The behaviour of Ti-6Al-4V/PAA at different immersion times was investigated and can be seen in Figure 6. From this, it is possible to verify the differences between the corrosive process than the unmodified material. Ti-6Al-4V/PAA also exhibits corrosive effects in the aggressive medium within 3 h. The results clearly show a delamination process (Nazir 2017) for the Ti-6Al-4V alloy containing acrylic acid. This effect becomes even clearer after 24 h of immersion as displayed in Figure 6e and f. These results indicate that the AA polymerization process via PECVD enables the formation of a thin film that is not completely homogeneous over the entire surface (Bitar et al. 2018), as verified by the AFM and SEM images. In this sense, surface irregularities open gaps for ions fluoride to interact directly with the metal alloy causing localized corrosion, as presented in Figure 6d. A schematic representation that explains the localized corrosion process as well as the delamination process for the Ti-6Al-4V and Ti-6Al-4V/PAA is shown in Figure S3.
PAA delamination was also analyzed by SEM and EDX. Figure 7 shows the chemical composition of a region in the sample, which was also subjected to 24 h of immersion in 0.1 mol L À1 NaF solution. The predominance of elements present in the Ti-6Al-4V/PAA alloy is observed in the delaminated region as shown in Figure 7c-f. The EDX maps demonstrated the absence of carbon in the delaminated area. On the other hand, there is an accumulation of this element in the surroundings of the mentioned region. Finally, the absence of C reflects that in this region there are no traces of polymerized AA and, therefore, corroborates the existence of the delamination process.
Regarding the surfaces on which the peptide is immobilized, the OM images for immersion tests are shown in Figure S4. Through this, it is possible to notice that there is a corrosion process very similar to the Ti-6Al-4V/PAA samples. Therefore, regions of delamination in the film are observed along this surface after 3 h of immersion, Figure S4d. As previously emphasized, delamination in these samples occurs due to the presence of irregularities on the metal surface after PAA deposition. However, the presence of pitting is observed only in the bare material in the first hours of immersion, showing once again that the PAA coating acts as a protective barrier for the Ti-6Al-4V alloy.
Despite the good properties of the Ti-6Al-4V alloy, the release of vanadium and aluminium ions in the human body may be associated with the onset of diseases such as Alzheimer's and Parkinson's (Kawahara and Kato-Negishi 2011;Nascimento et al. 2021). Thus, the PAA coating is an alternative to help reduce the release of these aggressive ions, which can improve the biocompatibility of this material (de  Giglio et al. (2007) performed the deposition of PAA on Ti-6Al-4V alloy through electrochemical polymerization, aiming to improve the properties of this alloy. To investigate the corrosive properties, immersion tests were carried out in a solution of sodium chloride (0.17 M NaCl) and EDTA (0.0027 M). The immersion protocol adopted by the authors consisted of verifying the release of metallic ions in the aforementioned aggressive medium by inductively coupled plasma mass spectrometry. The results obtained by the authors indicated that the thin films of polyacrylic acid act as a protective barrier against corrosion in this material. Since acrylic acid coatings reduce the release of metal ions in the medium (de Giglio et al. 2007). Some studies have demonstrated the efficiency of PAA against corrosion, considering other materials and aggressive mediums. Umoren et al. (2010) investigated the corrosive characteristics of pure aluminium coated with PAA in sulfuric acid (H 2 SO 4 ) through electrochemical methods. Under these conditions, these authors verified that acrylic acid inhibits the corrosion of this metallic material in the studied aggressive environment. In this sense, in addition to PAA being a crucial element to promote the immobilization of the peptide in the Ti-6Al-4V alloy, it has a protective effect on Ti-6Al-4V surfaces.

Scanning vibrating electrode results
SVET maps of Ti-6Al-4V alloy, Ti-6Al-4V/PAA, Ti-6Al-4V/PAA/JI-2 by the beginning of exposure and after 15 h in 0.01 mol L À1 NaF solution, as well as the optical image of the respective surfaces, are displayed in Figures 8 and 9. As can be seen, the base material (see Figure 8a) showed an electrochemical activity in the first hours of immersion. As the immersion time increased, the active sites moved along the exposed surface as represented by Figure 8e for the 15 h of immersion. OM images of the mentioned immersion times can be seen in Figure 8b and f, respectively. The Ti-6Al-4V/PAA specimens demonstrated a better performance in the localized corrosion process when compared to the base material (Figure 8c) in an aggressive medium containing F-ions. This result was observed throughout the test as shown in Figure 8g. In other words, acrylic acid acts as a protective barrier against the localized corrosion process. These results are in agreement with the morphological analyses, showing that the regions most susceptible to localized corrosion are irregularities in the film from the PECVD deposition process. This information is confirmed with OM images (Figure 8d and h). In addition, the SVET results are in line with those obtained in the immersion test as discussed above. Regarding the Ti-6Al-4V/PAA/JI-2, there was a succession of events taking place and the shift between the cathodic and anodic behaviour at specific points as presented in Figure 9a and c. These findings are probably due to the presence of several agglomerates observed in distinct regions of the surfaces containing the immobilized peptide as may be observed in Figure 9b and d. Considering the cationic character of the peptides, it is possible to elucidate that the corrosion process is intensified due to the electrostatic attraction between the F-ions and the peptide richagglomerates Figure 9e. In line with this, Trino and collaborators were able to correlate good corrosion properties of Ti surfaces where osteogenic peptides are homogeneously immobilized (Trino et al. 2018). In this sense, more efforts on improving the surface treatment of the Ti-6Al-4V/PAA containing peptide is necessary, seeking to combine the good properties of acrylic acid as a physical barrier against to localized corrosion process with the possibility of immobilization of an antimicrobial molecule on metallic biomaterial.

Activity against biofilms
The ability of the functionalized material to prevent the formation of E. coli biofilms was tested to assess whether the immobilization of the antimicrobial peptide JI-2 is a promising strategy to prevent biofilmassociated infections in implants. Ti-6Al-4V/PAA/JI-2  surfaces were prepared at 500 lmol L À1 of JI-2 to ensure that the material presented a peptide concentration close to the MIC obtained for E. coli ( Table  1). The results demonstrate that the immobilization of the JI-2 significantly reduced biofilm formation, showing an inhibition efficiency of the microorganism biofilm biomass of approximately 60%. Furthermore, optical density (OD) values were obtained ( Figure  10), demonstrating that the functionalized material has a lower optical density than the control material. This effect indicates the superior anti-biofilm capacity of the coating (Cao et al. 2018). Therefore, these results demonstrate that the immobilization of the JI-2 peptide on the Ti-6Al-4V alloy via AA plasma polymerization was effective in obtaining a new material with potential biomedical applications.

Conclusions
In this work, a new antimicrobial peptide, JI-2, was immobilized on Ti-6Al-4V surfaces. For that, Ti-6Al-4V surfaces were coated with acrylic acid polymerized via PECVD to enrich them with carboxyl groups. After activation of these groups, JI-2 was covalently bound to the surface of the material. FTIR results indicate that it was possible to perform PAA thin film deposition and JI-2 immobilization on Ti-6Al-4V surfaces. AFM and SEM data showed topographical differences at all stages of the surface treatment that contributed to the confirmation of the peptide immobilization and revealed the formation of agglomerates associated with its presence. EDX of the agglomerates showed a higher percentage of carbon and oxygen than surfaces coated only with acrylic acid indicating that the peptides are more concentrated in these regions. However, global and localized corrosion tests in aggressive media indicated that the presence of these agglomerates seems to intensify the material corrosion process. Even so, the ability of the functionalized material to prevent the formation of E. coli biofilms was verified, demonstrating that this work resulted in the development of a new material with interesting functional properties.

Disclosure statement
No potential conflict of interest was reported by the authors.