Hepatotropic Peptides Grafted onto Maleimide-Decorated Nanoparticles: Preparation, Characterization and In Vitro Uptake by Human HepaRG Hepatoma Cells

We recently demonstrated the strong tropism of George Baker (GB) Virus A (GBVA10-9) and Plasmodium circumsporozoite protein (CPB) derived synthetic peptides towards hepatoma cells. In a first approach, these peptides were covalently bound to poly(benzyl malate) (PMLABe73) and poly(ethylene glycol)-block-PMLABe73 (PEG62-b-PMLABe73) (co)polymers, and corresponding peptide-decorated nanoparticles (NPs) were prepared by nanoprecipitation. We showed that peptide enhanced NPs internalization by hepatoma cells. In the present work, we set up a second strategy to functionalize NPs prepared from PMLABe73 derivates. First, maleimide-functionalized PMLABe73 (Mal-PMLABe73) and PEG62-b-PMLABe73 (Mal-PEG62-b-PMLABe73) were synthesized and corresponding NPs were prepared by nanoprecipitation. Then, peptides (GBVA10-9, CPB and their scramble controls GBVA10-9scr and CPBscr) with a thiol group were engrafted onto the NPs’ maleimide groups using the Michael addition to obtain peptide functionalized NPs by post-formulation procedure. These peptide-modified NPs varied in diameter and dispersity depending on the considered peptides and/or (co)polymers but kept their spherical shape. The peptide-functionalized NPs were more efficiently internalized by HepaRG hepatoma cells than native and maleimide-NPs with various levels relying on the peptide’s nature and the presence of PEG. We also observed important differences in internalization of NPs functionalized by the maleimide-thiol-peptide reaction compared to that of NPs prepared from peptide-functionalized PMLABe73 derivatives.


Introduction
According to the GLOBOCAN 2020 report published by the International Cancer Research Center, primary liver cancer is the sixth most commonly diagnosed cancer in the world, and the third leading cause of cancer-related death in 2020, with 906,000 new cases diagnosed and 830,000 deaths per year [1]. Primary liver cancers mainly include hepatocellular carcinomas (HCC), 75% to 85% of cases, and cholangiocarcinomas, 10% to 15% of cases [2][3][4]. Curative therapies, mainly based on surgical resection of small tumors, are only proposed for early stages of HCC, knowing that the majority of HCCs are usually detected at advanced or very advanced stages, for which only palliative treatments or monium hydroxide and 6-maleimidohexanoic acid were purchased from Sigma-Aldrich (Sigma-Aldrich, Saint-Louis, MO, USA). Peptides were provided by Eurogentec (Eurogentec, Liege, Belgium). 1,1 -Dioctadecyl-3,3,3 ,3 -tetramethylindo dicarbocyanine perchlorate (DiD Oil) was purchased from Invitrogen (Thermo Fisher Scientific, Illkirch Graffenstaden, France). Solvents were purchased from Sigma-Aldrich (Sigma-Aldrich, Saint Quentin Fallavier, France).
Nuclear Magnetic Resonance spectroscopy (NMR): The standard temperature was adjusted to 298 K. NMR spectra were recorded on a Bruker Avance III 400 spectrometer (Bruker, Wissembourg, France) operating at 400. 13 MHz for 1 H, equipped with a BBFO probe with a Z-gradient coil and a GREAT 1/10 gradient unit. The zg30 Bruker pulse program was used for 1D 1 H NMR, with a TD of 64 k, a relaxation delay d1 = 2 s and 8 scans. The spectrum width was set to 18 ppm. Fourier transform of the acquired FID was performed without any apodization in most cases.
Size Exclusion Chromatography (SEC): Weight average molar mass (Mw) and dispersity (Ð = Mw/Mn) values were measured by SEC in THF at 40 • C (flow rate = 1.0 mL/min) on a GPC2502 Viscotek apparatus equipped with a refractive index detector Viscotek VE 3580 RI, a guard column Viscotek TGuard, Org 10 × 4.6 mm, a LT5000L gel column 300 × 7.8 mm and a GPC/SEC OmniSEC Software (Malvern, Worcestershire, UK). The polymer samples were dissolved in THF (2 mg/mL). All elution curves were calibrated with polystyrene standards.
Differential scanning calorimetry (DSC): DSC measurements were performed on NET-ZSCH DSC 200 F3 instrument equipped with an intracooler (ERICH NETZSCH GMBH and CO. Holding KG, Selb, Germany). DSC traces were measured at 10 • C/min up heating (red curve) and at −5 • C/min upon cooling (purple curve). Glass transition temperature (Tg) was defined as the midpoint of the glass transition process upon the heating process.
Dynamic Light Scattering (DLS): DLS measurements were performed on a Nano-sizer ZS90 (Malvern, Worcestershire, UK) at 25 • C, with a He-Ne laser at 633 nm and a detection angle of 90 • C. Three runs of 70 scans each were performed on each NPs suspension, and average values of hydrodynamic diameter (Dh) and dispersity (PDI) were given. The size distribution reports were given by Intensity. Two examples of report were given in the SI files ( Figure S1).
Transmission electron microscopy (TEM): TEM images were recorded using a Jeol 2100 microscope (Jeol, Tokyo, Japan) equipped with a Glatan Orius 200D camera using an 80 KeV accelerating voltage on the THEMIS platform (ISCR-Rennes, France). Each sample was deposited on a Formvar-carbon film coated on a 300-mesh copper grid. After 6 min, the excess of sample was removed and a staining was realized with phosphotungstic acid (1 v%).
Flow cytometry: Cells were analyzed by flow cytometry using a LSRFortessa™ X-20 cytometer (Becton Dickinson, Becton Drive Lake, NJ, USA), using the cytometry core facility of the Biology and Health Federative research structure Biosit (Rennes, France) to quantify the fluorescence emitted by the DiD Oil-loaded NPs within cells. Cytometry data were analyzed using FACSDiva TM software (Becton Dickinson, Becton Drive Lake, NJ, USA).

NPs Formulation and Peptides' Grafting Procedure
Maleimide-decorated and native NPs encapsulating 0.1 wt% of the fluorescence probe DiD Oil were formulated using the nanoprecipitation method as previously described by Fessi et al. [16] and peptides were grafted onto maleimide-decorated NPs using Michael addition [17].
Native PMLABe 73 -Based NPs (NPs 1 and NPs 1 ) -NPs 1: 140 µL of a solution of PMLABe 73 at a concentration of 7.14 mg/mL in DMF were mixed with 10 µL of a DiD Oil solution in dimethylformamide (DMF) at a concentration of 0.1 mg/mL (0.1 wt%). This blue solution was rapidly added to 2 mL of distilled water under vigorous stirring and the resulting suspension was stirred at room temperature (RT) for 15 min. The bluish suspension was analyzed by DLS ( Figure 2).
-NPs 1 : 8.5 µL of PBS pH7.4 were added to NPs 1 suspension, and the mixture was stirred for 5 min at RT. The suspension was deposited on a Sephadex column, eluted by 0.5 mL then 3.5 mL of water, and analyzed by DLS (Table 1).
Maleimide-Decorated PMLABe 73 -Based NPs (NPs 2 and NPs 2 ) -NPs 2: 5 µL of a solution of Mal-PMLABe 73 (10 wt%) at a concentration of 20 mg/mL in DMF were mixed with 135 µL of a solution of PMLABe 73 (90 wt%) at a concentration of 6.67 mg/mL in DMF. 10 µL of a DiD Oil solution in DMF at a concentration of 0.1 mg/mL (0.1 wt%) were then added. This blue solution was rapidly added to 2 mL of distilled water under vigorous stirring and the resulting suspension was stirred at RT for 15 min. The bluish suspension was analyzed by DLS ( Figure 2).
-NPs 2 : 8.5 µL of PBS pH7.4 were added to NPs 2 suspension, and the mixture was stirred for 5 min at RT. The suspension was purified by filtration through a Sephadex column as described above, and analyzed by DLS (Table 1).
Peptide-Decorated PMLABe 73 -Based NPs (NPs 3 to NPs 6) The Mal-decorated PMLABe 73 NPs encapsulating DiD Oil were prepared as described for NPs 2, followed by: (i). for NPs 3, the addition of 8.5 µL of PBS pH7.4 solution of GBVA10-9 at a concentration of 2 mg/mL, (ii). for NPs 4, the addition of 8.5 µL of PBS pH7.4 solution of GBVA10-9scr at a concentration of 2 mg/mL, (iii). for NPs 5, the addition of 8.5 µL of PBS pH7.4 solution of CPB at a concentration of 2 mg/mL, (iv). for NPs 6, the addition of 8 µL of PBS pH7.4 solution of CPBscr at a concentration of 2 mg/mL. All the suspensions were purified by filtration through a Sephadex column as described above, and analyzed by DLS (Table 1) and TEM.
Native PEGylated PMLABe 73 -Based NPs (NPs 7 and NPs 7 ) -NPs 7: 5 µL of a solution of PEG 42 -b-PMLABe 73 (10 wt%) at a concentration of 20 mg/mL in DMF were mixed with 135 µL of a solution of PMLABe 73 (90 wt%) at a concentration of 6.67 mg/mL in DMF. 10 µL of a DiD Oil solution in DMF at a concentration of 0.1 mg/mL (0.1 wt%) were then added. This blue solution was rapidly added to 2 mL of distilled water under vigorous stirring and the resulting suspension was stirred at RT for 15 min. The resulting bluish suspension was analyzed by DLS ( Figure 2).
-NPs 7 : 8.5 µL of PBS pH7.4 were added to NPs 7 suspension, and the mixture was stirred 5 min at RT. The suspension was purified by filtration through a Sephadex column as described above, and analyzed by DLS (Table 1). were then added. This blue solution was rapidly added to 2 mL of distilled water under vigorous stirring and the resulting suspension was stirred at RT for 15 min. The resulting bluish suspension was analyzed by DLS ( Figure 2).

Maleimide-Decorated
-NPs 8 : 8.5 µL of PBS pH7.4 were added and the mixture was stirred 5 min at RT. The suspension was purified by filtration through a Sephadex column as described above, and analyzed by DLS (Table 1).
Peptide-Decorated PEGylated PMLABe 73 -Based NPs (NPs 9 to NPs 12) The Mal-decorated PEG 62 -b-PMLABe 73 NPs encapsulating DiD Oil were prepared as described for NPs 8, followed by: (i). for NPs 9, the addition of 7.5 µL of PBS pH7.4 solution of GBVA10-9 at a concentration of 2 mg/mL, (ii). for NPs 10, the addition of 7.5 µL of PBS pH7.4 solution of GBVA10-9scr at a concentration of 2 mg/mL, (iii). for NPs 11, the addition of 7.5 µL of PBS pH7.4 solution of CPB at a concentration of 2 mg/mL, (iv). for NPs 12, the addition of 6.5 µL of PBS pH7.4 solution of CPBscr at a concentration of 2 mg/mL. All the suspensions were purified by filtration through a Sephadex column as described above, and analyzed by DLS (Table 1) and TEM.

Cell Culture and In Vitro Uptake Assay of NPs
The human HepaRG hepatoma cells were grown in William's E medium supplemented with 10% Fetal Calf Serum (FCS), 2% L-glutamine, 5 mg/L insulin, hydrocortisone hemisuccinate at 5 × 10 −5 M, 100 units/mL penicillin and 100 µg/mL streptomycin. The HepaRG cells were cultured at 37 • C in a humidified atmosphere of 5% CO 2 , and the medium was renewed every 2 days.
For the NPs' uptake assay, 10 5 HepaRG cells were seeded in 24-well plates the day before the incubations with the different batches of NPs. The culture media of the cells were renewed with media containing NPs at a final concentration of (co)polymers of 25 µg/mL. The cells were incubated with NPs for 24 h, then were detached with trypsin-EDTA and resuspended in complete medium for flow cytometry analysis and evaluation of the NPs' uptake by measuring the fluorescence emitted by the DiD oil loaded NPs.

Synthesis of (co)Polymers
The four selected (co)polymers were synthesized as described previously by aROP of MLABe in presence of the corresponding tetraethylammonium carboxylate (Figure 1), either commercially available (tetraethylammonium benzoate) or synthesized by an acid/base reaction between the corresponding carboxylic acid and tetraethylammonium hydroxide [12].
The molar mass of the PMLABe block has been defined by the monomer/initiator ratio and fixed at 15,000 g/mol (n = 73). As shown by Figure 1, two non-functionalized (co)polymers, PMLABe 73 and PEG 42 -b-PMLABe 73 , and two maleimide-functionalized (co)polymers, Mal-PMLABe 73 and Mal-PEG 62 -b-PMLABe 73 , have been successfully synthesized. After purification by precipitation, the proton NMR spectra (structure, Figure S2), the chromatograms obtained by size exclusion chromatography (weight average molar mass (Mw) and dispersity (Ð), Figure 1 and Figure S3) and thermograms obtained by differential scanning calorimetry (glass transition temperature, Tg, Figure S4) were recorded. The (co)polymers had the expected structures ( Figure S2) and molar masses calculated from the proton NMR spectra ( Figure S2) were close to theoretical ones ( Figure 1). As already observed [12], weight average molar mass (Mw) measured by SEC were lower than the theoretical one and varied from 5700 to 8500 g/mol, while dispersity values were in the range of 1.45 to 1.59 ( Figure 1 and Figure S3). PMLABe 73 and Mal-PMLABe 73 showed similar glass transition temperature at 30 • C and 25 • C, respectively ( Figure S4A and S4B, respectively). On the other hand, amphiphilic block copolymers, PEG 42 -b-PMLABe 73 and Mal-PEG 62 -b-PMLABe 73 , showed also only one glass transition temperature at 13 • C and −14 • C, respectively ( Figure S4C and S4D, respectively). The molar mass of the PMLABe block has been defined by the monomer/initiator ratio and fixed at 15,000 g/mol (n = 73). As shown by Figure 1, two non-functionalized (co)polymers, PMLABe73 and PEG42-b-PMLABe73, and two maleimide-functionalized (co)polymers, Mal-PMLABe73 and Mal-PEG62-b-PMLABe73, have been successfully synthesized. After purification by precipitation, the proton NMR spectra (structure, Figure  S2), the chromatograms obtained by size exclusion chromatography (weight average molar mass (Mw) and dispersity (Ð), Figures 1 and S3) and thermograms obtained by differential scanning calorimetry (glass transition temperature, Tg, Figure S4) were recorded. The (co)polymers had the expected structures ( Figure S2) and molar masses calculated from the proton NMR spectra ( Figure S2) were close to theoretical ones ( Figure 1). As already observed [12], weight average molar mass (Mw) measured by SEC were lower than the theoretical one and varied from 5700 to 8500 g/mol, while dispersity values were in the range of 1.45 to 1.59 (Figures 1 and S3). PMLABe73 and Mal-PMLABe73 showed similar glass transition temperature at 30 °C and 25 °C, respectively ( Figure S4A and S4B, respectively). On the other hand, amphiphilic block copolymers, PEG42-b-PMLABe73 and Mal-PEG62-b-PMLABe73, showed also only one glass transition temperature at 13 °C and -14 °C, respectively ( Figure S4C and S4D, respectively).
These native and maleimide-decorated NPs were analyzed by DLS to determine their hydrodynamic diameter (Dh) and dispersity (PDI). As evidenced by results gathered in Figure 2, native and maleimide-decorated NPs have relatively similar hydrodynamic diameters, varying from 69 to 87 nm, with quite low dispersity varying from 0.13 to 0.23.
Since native and maleimide-decorated NPs had the expected characteristics in terms of diameter and dispersity, we therefore proceeded to the grafting of the four selected peptides (GBVA10-9, GBVA10-9scr, CPB and CPBscr) C-terminated by a thiol group onto the preformed NPs bearing maleimide functions at their surface ( Figure 3A(a,b)) using the Michael addition [17]. In parallel, control NPs 1, NPs 2, NPs 7 and NPs 8 were also treated under the same reaction conditions (PBS pH 7.4, Sephadex column) in absence of peptides ( Figure 3B), thus leading to NPs 1 , NPs 2 , NPs 7 and NPs 8 , to evaluate if these reaction conditions had any influence on NPs' characteristics.
These peptide-free NPs and the peptide-decorated NPs were characterized by DLS (Table 1). Their hydrodynamic diameter varied from 65 to 95 nm with dispersity ranging from 0.12 to 0.26 (Table 1).
On the other hand, peptides' grafting globally led to an increase in both hydrodynamic diameters and dispersity of the corresponding NPs (NPs 3 to 6, and NPs 9 to 12- Table 1), with Dh values varying from 82 to 232 nm and PDI between 0.21 and 0.44 for non-PEGylated NPs (NPs 3 to 6- Table 1), and Dh values varying from 74 to 212 nm and PDI between 0.21 to 0.37 for PEGylated NPs (NPs 9 to 12- Table 1). We also observed that the Dh of peptide-decorated NPs formed from a mixture of PMLABe 73 (90 wt%) and Mal-PEG 62 -b-PMLABe 73 (10 wt%) were slightly smaller than those measured for their corresponding peptide-decorated NPs prepared from a mixture of PMLABe 73 (90 wt%) and Mal-PMLABe 73 (10 wt%). These native and maleimide-decorated NPs were analyzed by DLS to determine their hydrodynamic diameter (Dh) and dispersity (PDI). As evidenced by results gathered in Figure 2, native and maleimide-decorated NPs have relatively similar hydrodynamic diameters, varying from 69 to 87 nm, with quite low dispersity varying from 0.13 to 0.23.
Since native and maleimide-decorated NPs had the expected characteristics in terms of diameter and dispersity, we therefore proceeded to the grafting of the four selected peptides (GBVA10-9, GBVA10-9scr, CPB and CPBscr) C-terminated by a thiol group onto the preformed NPs bearing maleimide functions at their surface ( Figure 3A(a,b)) using the Michael addition [17]. In parallel, control NPs 1, NPs 2, NPs 7 and NPs 8 were also treated under the same reaction conditions (PBS pH 7.4, Sephadex column) in absence of peptides ( Figure 3B), thus leading to NPs 1′, NPs 2′, NPs 7′ and NPs 8′, to evaluate if these reaction conditions had any influence on NPs' characteristics.  These peptide-free NPs and the peptide-decorated NPs were characterized by DLS (Table 1). Their hydrodynamic diameter varied from 65 to 95 nm with dispersity ranging from 0.12 to 0.26 (Table 1).   Peptides-decorated NPs were also visualized by transmission electron microscopy (TEM, Figure 4) to evaluate the impact of peptides' grafting on the morphology of resulting modified NPs.  TEM images (Figure 4) showed that all the peptide-modified NPs kept their spherical morphology, and that they had diameters and dispersity in agreement with those measured by DLS, knowing that the NP's TEM images are realized under dry conditions thus usually leading to underestimated diameters.

Nanoparticle's Uptake by HepaRG Hepatoma Cells In Vitro
The effect of peptide grafting on maleimide-decorated NPs by the post-formulation method onto cell uptake has been evaluated using HepaRG hepatoma cells ( Figure 5). This cell line, used worldwide as an alternative to human hepatocytes [18], was also chosen in order to compare the NP's uptake evaluated in previous studies [12]. The cells were incubated with NPs suspensions without maleimide functions or peptides on their surfaces (control NPs 1′ and NPs 7′, Table 1), with maleimide decorated NPs suspensions (control NPs 2′ and NPs 8′, Table 1) and with peptide-modified NPs (NPs 3 to 6, and NPs 9 to 12, Table 1). TEM images (Figure 4) showed that all the peptide-modified NPs kept their spherical morphology, and that they had diameters and dispersity in agreement with those measured by DLS, knowing that the NP's TEM images are realized under dry conditions thus usually leading to underestimated diameters.

Nanoparticle's Uptake by HepaRG Hepatoma Cells In Vitro
The effect of peptide grafting on maleimide-decorated NPs by the post-formulation method onto cell uptake has been evaluated using HepaRG hepatoma cells ( Figure 5). This cell line, used worldwide as an alternative to human hepatocytes [18], was also chosen in order to compare the NP's uptake evaluated in previous studies [12]. The cells were incubated with NPs suspensions without maleimide functions or peptides on their surfaces (control NPs 1 and NPs 7 , Table 1), with maleimide decorated NPs suspensions (control NPs 2 and NPs 8 , Table 1) and with peptide-modified NPs (NPs 3 to 6, and NPs 9 to 12, Table 1). As evidenced by Figure 5, NPs 1′ and NPs 7′ and maleimide-decorated NPs (NPs 2′ and NPs 8′) were poorly internalized by the HepaRG cells. Peptide-decorated NPs without PEG were significantly better internalized in a peptide-dependent manner, with higher uptake for GBVA10-9 and GBVA10-9scr decorated NPs. All NPs with PEG were less internalized demonstrating that presence of the PEG spacer reduced the cell uptake ( Figure 5). However, NPs decorated with GBVA10-9 (NPs 9) or CPB (NPs 11) through the PEG spacer were significantly better internalized than the same NPs decorated with GBVA10-9scr (NPs 10) or CPBscr (NPs 12) ( Figure 5).

Discussion
In the past years, our team has developed a large set of biocompatible (co)polymers derived from PMLABe, which can be used for multiple applications in the field of nanomedicine and biomaterials [19].
In this report, our first aim was to demonstrate that peptide-functionalization of NPs prepared from PMLABe derivates was possible after nanoprecipitation with a post-formulation procedure using maleimide modified PMLABe73 and PEG62-b-PMLABe73 (co)polymers. First, this study highlighted the robustness of the polymerization reaction of MLABe. Indeed, the four (co)polymers (PMLABe73, PEG42-b-PMALBe73, Mal-PMLABe73 and Mal-PEG62-b-PMLABe73) have identical structures ( Figure S2), molar masses and dispersity (Figures 1 and S3) to those obtained previously [12]. Furthermore, we have also completed the characterization of the (co)polymers by a DSC analysis. As expected, the two amorphous homopolymers, PMLABe73 and Mal-PMLABe73, showed only one glass transition temperature with values being in good agreement with those given by Guérin et al. [20] and Cammas et al. [21] for PMLABe homopolymers. Surprisingly, the amphiphilic block copolymers, PEG42-b-PMLABe73 and Mal-PEG62-b-PMLABe73, also showed As evidenced by Figure 5, NPs 1 and NPs 7 and maleimide-decorated NPs (NPs 2 and NPs 8 ) were poorly internalized by the HepaRG cells. Peptide-decorated NPs without PEG were significantly better internalized in a peptide-dependent manner, with higher uptake for GBVA10-9 and GBVA10-9scr decorated NPs. All NPs with PEG were less internalized demonstrating that presence of the PEG spacer reduced the cell uptake ( Figure 5). However, NPs decorated with GBVA10-9 (NPs 9) or CPB (NPs 11) through the PEG spacer were significantly better internalized than the same NPs decorated with GBVA10-9scr (NPs 10) or CPBscr (NPs 12) ( Figure 5).

Discussion
In the past years, our team has developed a large set of biocompatible (co)polymers derived from PMLABe, which can be used for multiple applications in the field of nanomedicine and biomaterials [19].
In this report, our first aim was to demonstrate that peptide-functionalization of NPs prepared from PMLABe derivates was possible after nanoprecipitation with a postformulation procedure using maleimide modified PMLABe 73 and PEG 62 -b-PMLABe 73 (co)polymers. First, this study highlighted the robustness of the polymerization reaction of MLABe. Indeed, the four (co)polymers (PMLABe 73 , PEG 42 -b-PMALBe 73 , Mal-PMLABe 73 and Mal-PEG 62 -b-PMLABe 73 ) have identical structures ( Figure S2), molar masses and dispersity (Figures 1 and S3) to those obtained previously [12]. Furthermore, we have also completed the characterization of the (co)polymers by a DSC analysis. As expected, the two amorphous homopolymers, PMLABe 73 and Mal-PMLABe 73 , showed only one glass transition temperature with values being in good agreement with those given by Guérin et al. [20] and Cammas et al. [21] for PMLABe homopolymers. Surprisingly, the amphiphilic block copolymers, PEG 42 -b-PMLABe 73 and Mal-PEG 62 -b-PMLABe 73 , also showed one glass transition temperature with lower values than those observed for the PMLABe 73 homopolymers, while one could expect to see two Tg values, one corresponding to the Tg of the PEG block and the other to the Tg of the PMLABe 73 block. The first hypothesis that can be formulated to explain such results was that: (i). the Tg of PEG (Tg = −60 • C [22]) was not visible on the recorded thermograms since the temperature range used was of −40 • C to 100 • C), and (ii). the measured values corresponded to the Tg of the PMLABe block having an organization disturbed by the PEG block thus leading to a sharp reduction in the Tg of this PMLABe block. The second hypothesis that can be formulated was based on a phenomenon described by Bae et al. [23] on PEG-b-poly(L-lactic acid) block copolymers. These authors observed: (i). a single large peak of Tg in a wide temperature range and an absence of Tg from the amorphous zones of PLLA due to the phenomenon of high phase mixing between the PEG and PLLA blocks, and (ii). a significant decrease in temperature crystallization of PEG due to the presence of the PLLA block [23]. Based on this work, it can be assumed that the two blocks, PEG and PMLABe, constituting the copolymers considered in the present study might have a high degree of phase mixing resulting in: (i). the presence of a single glass transition temperature, and (ii). the absence of crystallization temperature of the PEG block. Further studies will have to be carried out to evidence, or not, the phase mixing between the two blocks. Nevertheless, the synthesized (co)polymers have the required properties allowing NPs formulation. Several methods have been described in the literature for the conjugation of targeting agents, such as peptide or antibodies after formation of the NPs [14]. Among them, carbodiimide, maleimide and click chemistries are particularly prominent, each with their own advantages and drawbacks. Due to its ease of use, we chose to use maleimide-based Michael addition [17]. It also enabled us to compare the results obtained here with the ones previously obtained with the pre-formulation procedure, in which the thiol-terminated peptides were grafted on the polymer chain through Michael addition [12]. Indeed, when dealing with peptides, choice of the linker is not trivial and without consequences, since even the slightest modification can alter the peptide's behavior [24,25]. It has been notably recently highlighted with radiolabeled peptide derivatives, for which the chelate conjugated to the peptide demonstrated a significant influence [26,27]. There even is a report of a somatostatin receptor antagonist peptide that unexpectedly turned into an agonist when coupled to a DOTA chelator [28].
Unmodified NPs, without or with maleimide groups at their surface, were successfully prepared by nanoprecipitation of corresponding mixture of PMLABe 73 derivatives. Since their hydrodynamic diameters were lower than 100 nm, as already observed for PMLABe 73 -based NPs [12], with dispersity lower than 0.25, it can be concluded that quite narrow-dispersed NPs' suspensions were obtained. Finally, the presence of 10 wt% of either maleimide or Mal-PEG 62 groups on NPs' surface had no significant influence on NPs' characteristics.
Treatment of these native NPs under the conditions used to graft peptides (Michael addition conditions) showed that these conditions had no significant impact on NPs in terms of diameter and dispersity. Indeed, no drastic changes have been observed either for the diameter or dispersity values, since only non-significant diameter's decrease (NPs 1 , NPs 7 and NPs 8 - Table 1) or increase (NPs 2 - Table 1) were observed, while the dispersity values were nearly identical.
We then confirmed that peptides containing a thiol residue at their C-terminal end could be engrafted onto PMLABe-based NPs bearing maleimide groups by the Michael addition. Indeed, results highlighted that the grafting of peptides onto maleimide-decorated NPs, with or without a PEG spacer, did not destabilized the preformed polymeric NPs and that well-defined peptide-decorated nanoobjects were obtained. Those results have been compared to the ones we previously obtained [12] for peptide-decorated NPs obtained by nanoprecipitation of PMLABe 73 /Pept-PMLABe 73 and PMLABe 73 /Pept-PEG 62 -b-PMLABe 73 (90/10 wt%) mixture (Table 2). The comparison between all the results ( Figure S5) showed that native NPs (NPs 1 , NPs 2 , NPs 7 and NPs 8 Table 1; and NPs 1 and NPs 7, Figure 2) had similar characteristics in terms of diameter ( Figure S5A) and dispersity ( Figure S5B). On the other hand, NPs decorated post-formulation with GBVA10-9 and GBVA10-9scr (NPs 3, NPs 4, NPs 9 and NPs 10- Table 1) had significantly higher diameters ( Figure S5A) than corresponding NPs obtained by nanoprecipitation of peptide-functionalized (co)polymers (NPs 15, NPs 16, NPs 20 and NPs 21- Table 2). A significant increase in dispersity ( Figure S5B) has been also observed between NPs obtained by the post-formulation procedure vs. NPs obtained through the pre-formulation method. Such results most probably indicate that the GBVA10-9 and GBVA10-9scr were effectively located at the surface of NPs modified after their formulation thus increasing their diameters, while they were entrapped into the hydrophobic inner core of NPs prepared from peptide-functionalized (co)polymers. For the CPB and CPBscr-decorated NPs, we have observed a slight decrease in the values of diameters ( Figure S5A) for the NPs decorated with these peptides after their formulation (NPs 5, NPs 6, NPs 11 and NPs 12- Table 1) in comparison to the ones obtained from peptide-functionalized (co)polymers (NPs 17, NPs 18, NPs 22 and NPs 23- Table 2). An increase in the dispersity values for the CPB-and CPBscr-decorated NPs obtained by modification of the formulated NPs has been observed ( Figure S5B). The slight decrease in the diameters values between the two formulation methods might be due to the more hydrophilic character of CPB and CPBscr, leading to a more expensed conformation of these peptides in aqueous media. Finally, TEM images of peptide-modified NPs prepared by the post-formulation procedure ( Figure 4) and by the pre-formulation procedure [12] were similar, thus highlighting that the procedure used to graft peptides had no influence on the NPs morphology.
Then, we looked at the impact of the functionalization by two peptides, GBVA10-9 and CPB, previously shown to have a strong tropism for hepatoma cells [11], on the NP's cell internalization. We also used scrambled peptides for both CPB and GBVA10-9 to investigate whether their specific amino acid sequence was the main determinant to drive the hepatoma cell uptake of NPs functionalized by peptides with a post-formulation method ( Figure 5). Our data showed that the peptide-functionalized NPs with both amphipathic-helical GBVA10-9 and scrambled peptide strongly enhanced internalization by hepatoma cells. Significant differences in the cell uptake between NPs functionalized with peptides using the Michael addition, in post-formulation, and NPs prepared from peptide-modified PMLABe (co)polymers have been observed (Figures 5 and 6).
For instance, we previously reported that cell uptake of NPs prepared from GBVA10-9 functionalized PMLABe 73 homopolymer was very low (Figure 6) [12] while, in this report, we showed that addition of the same peptide onto maleimide modified NPs strongly enhanced their cell uptake ( Figure 5). Similarly, CPB-and CPBscr-modified NPs by post-formulation were internalized nearly at the same levels ( Figure 5) while the uptake of NPs prepared from CPB-modified PMLABe 73 homopolymer in pre-formulation was much higher than that of CPBscr-functionalized NPs ( Figure 6). On their side, Valetti et al. found that conjugation of their peptide prior NPs formulation led to a better cellular uptake [13].
Polymers 2022, 14, x FOR PEER REVIEW 13 of 15 Figure 6. In vitro nanoparticle's uptake assays using progenitor HepaRG cells of peptide-decorated NPs obtained by the pre-formulation method (Adapted from ref [12]).
For instance, we previously reported that cell uptake of NPs prepared from GBVA10-9 functionalized PMLABe73 homopolymer was very low ( Figure 6) [12] while, in this report, we showed that addition of the same peptide onto maleimide modified NPs strongly enhanced their cell uptake ( Figure 5). Similarly, CPB-and CPBscr-modified NPs by postformulation were internalized nearly at the same levels ( Figure 5) while the uptake of NPs prepared from CPB-modified PMLABe73 homopolymer in pre-formulation was much higher than that of CPBscr-functionalized NPs ( Figure 6). On their side, Valetti et al. found that conjugation of their peptide prior NPs formulation led to a better cellular uptake [13].
These data demonstrated that the method used to functionalize NPs with peptides, post-vs. pre-formulation, strongly influence the cell uptake of the nanovectors, and cannot be predicted a priori. Since there was no obvious relationship between neither the NPs' diameters nor the NPs' dispersity values and the uptake of such NPs, these features could not explain the differences in cell internalization. Moreover, all NPs obtained by pre-or post-formulation had spherical shape, which could not provide a clear explanation for the differences in NP's cell uptake.
It might be postulated that the peptide grafting method, post-vs. pre-formulation, had a significant influence on the peptide accessibility and cell recognition capacity. In addition, we previously demonstrated that the amphipathic-helical GBVA10-9 peptide [29] could interact with apolipoproteins present in the culture medium, which strongly affected the opsonization of the NPs functionalized with this peptide [11]. The functionalization methods could also affect the binding of plasma proteins that would explain some of the differences observed in the cell internalization of peptide-decorated NPs.
Together, our data confirmed a previous hypothesis that peptides physicochemical features, the method used to functionalize NPs and/or formation of a protein corona could be the main drivers of the NPs' cell internalization rather than the recognition of a membrane receptor mediated by a specific amino acid sequence of the peptides. Rigorous investigations on all these aspects are warranted in order to fully understand their uptake mechanism and how the different moieties interact, aiming at the most efficient tumor targeting.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: DLS size distribution report by intensity of: A. NPs 1′, B. and B. NPs 7′. Figure   In vitro nanoparticle's uptake assays using progenitor HepaRG cells of peptide-decorated NPs obtained by the pre-formulation method (Adapted from ref [12]).
These data demonstrated that the method used to functionalize NPs with peptides, post-vs. pre-formulation, strongly influence the cell uptake of the nanovectors, and cannot be predicted a priori. Since there was no obvious relationship between neither the NPs' diameters nor the NPs' dispersity values and the uptake of such NPs, these features could not explain the differences in cell internalization. Moreover, all NPs obtained by pre-or post-formulation had spherical shape, which could not provide a clear explanation for the differences in NP's cell uptake.
It might be postulated that the peptide grafting method, post-vs. pre-formulation, had a significant influence on the peptide accessibility and cell recognition capacity. In addition, we previously demonstrated that the amphipathic-helical GBVA10-9 peptide [29] could interact with apolipoproteins present in the culture medium, which strongly affected the opsonization of the NPs functionalized with this peptide [11]. The functionalization methods could also affect the binding of plasma proteins that would explain some of the differences observed in the cell internalization of peptide-decorated NPs.
Together, our data confirmed a previous hypothesis that peptides physicochemical features, the method used to functionalize NPs and/or formation of a protein corona could be the main drivers of the NPs' cell internalization rather than the recognition of a membrane receptor mediated by a specific amino acid sequence of the peptides. Rigorous investigations on all these aspects are warranted in order to fully understand their uptake mechanism and how the different moieties interact, aiming at the most efficient tumor targeting.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/polym14122447/s1, Figure S1: DLS size distribution report by intensity of: A. NPs 1 , B. and B. NPs 7 . Figure  Institutional Review Board Statement: Not applicable. The current study did not require ethical approval since we did not use animals, human samples or primary human cells.

Informed Consent Statement:
Not applicable. The current study did not involve humans.
Data Availability Statement: Data presented in the results are not publicly available but Supplementary Materials can be downloaded.