Deep Insight into PEGylation of Bioadhesive Chitosan Nanoparticles

Subscriber access provided by University of South Dakota

Biological and Medical Applications of Materials and Interfaces

Deep insight into PEGylation of bioadhesive chitosan nanoparticles: Sensitivity study for the key parameters through artificial neural network model Ugur Bozuyuk, Nihal Olcay Dogan, and Seda Kizilel ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b11178 • Publication Date (Web): 13 Sep 2018 Downloaded from http://pubs.acs.org on September 13, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Deep Insight into PEGylation of Bioadhesive Chitosan Nanoparticles: Sensitivity Study for the Key Parameters Through Artificial Neural Network Model Ugur Bozuyuk, Nihal Olcay Dogan, Seda Kizilel* Koç University, Chemical and Biological Engineering, Sariyer, Istanbul, 34450, Turkey email: [email protected] fax: +90-212-338-1548 Abstract: Ionically-crosslinked chitosan nanoparticles have great potential in nanomedicine due to their tunable properties and cationic nature. However, low solubility of chitosan severely limits their potential clinical translation. PEGylation is a well-known method to increase solubility of chitosan and chitosan nanoparticles in neutral media, however, effect of PEG chain length and chitosan/PEG ratio on particle size and zeta potential of nanoparticles are not known. This study presents a systematic analysis of the effect of PEG chain length and chitosan/PEG ratio on size and zeta potential of nanoparticles. We prepared PEGylated chitosan chains prior to the nanoparticle synthesis with different PEG chain lengths and chitosan/PEG ratios. To precisely estimate the influence of critical parameters on size and zeta potential of nanoparticles, we both developed an artificial neural network (ANN) model and performed experimental characterization using the three independent input variables: (i) PEG chain length, (ii) chitosan/PEG ratio and (iii) pH of solution. We studied the influence of PEG chain lengths of 2, 5 and 10 kDa and three different chitosan/PEG ratios (25 mg chitosan to 4, 12 and 20 µmoles of PEG) for the synthesis of chitosan nanoparticles within the pH range of 6.0-7.4. Artificial neural networks is a modelling tool used in nanomedicine to optimize and estimate inherent properties of the system. Inherent properties of a nanoparticle system such as size and zeta potential can be estimated based on previous experiment results, thus, nanoparticles with desired properties can be obtained using an ANN. With the ANN model, we were able to predict the size and zeta potential of nanoparticles under different experimental conditions and further confirmed the cellnanoparticle adhesion behavior through experiments. Nanoparticle groups that had higher zeta potentials promoted adhesion of HEK293-T cells to nanoparticle-coated surfaces in cell culture medium, which was predicted through ANN model prior to experiments. Overall, this study comprehensively presents the PEGylation of chitosan, synthesis of PEGylated chitosan 1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nanoparticles, utilizes ANN model as a tool to predict important properties such as size and zeta potential and further captures the adhesion behavior of cells on surfaces prepared with these engineered nanoparticles. Keywords: chitosan, PEGylation, ionotropic gelation, artificial neural networks, PEG chain length, PEGylated chitosan, engineered nanoparticles, chitosan nanoparticles 1. INTRODUCTION Recent advances in nanomedicine have led to the development of nanoparticle-based drugs that are promising for the treatment of various diseases.1-5 One of the main goals in nanomedicine is to develop nanoparticle-based therapeutics that can overcome the biological barriers in the body. Engineered nanoparticles with optimal surface charge and size improve the efficiency and therapeutic index of the system and hence have gained growing interest for effective treatment of various diseases.6 Bioadhesive polymers can extensively interact with cells, and therefore increase the mean residence time of the drug and therapeutic activity of the system. One such example is chitosan, which has a wide range of applications from agriculture to biopharmaceutics.7 In addition to its desirable properties such as biocompatibility8-12 and biodegradability13, chitosan is a suitable candidate for mucoadhesive drug and gene delivery systems due to its cationic nature.14 However, its solubility in only acidic medium and low colloidal stability at physiological pH severely limits its further use in biomedical applications.15 PEGylation is commonly used to overcome this challenge; however, the influences of fundamental parameters such as PEG chain length and chitosan/PEG ratio on the size and zeta potential of chitosan nanoparticles remained unknown. The conjugation of polyethylene glycol (PEG), i.e. PEGylation of chitosan, is a wellknown strategy to introduce new physiological and chemical properties to chitosan including solubility at wide range of pH values. To achieve this goal, various synthetic strategies have been developed for chemical conjugation of PEG to chitosan in previous studies. Among these approaches, amino group substitution of chitosan through n-succinimidyl ester (NHS) derivative (known as NHS-amino coupling) of PEG is desirable due to its simplicity and does not require the use of complex reaction pathways or equipments.16 Chitosan nanoparticles can be prepared by using various techniques such as emulsificiation solvent diffusion method17, reverse micellar method18, microemulsion method19, ionotropic 2 ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


gelation etc.20-21 Among these synthesis methods, ionotropic gelation seems to be the most convenient as the reaction can be carried out at room temperature with only a vortex mixer and does not require organic solvents.22 In this method, interaction between positively charged chitosan chains and negatively charged polyanion tripolyphosphate (TPP) results in the formation of chitosan nanoparticles within short time periods in acidic medium.20 However, unmodified chitosan nanoparticles aggregate at neutral conditions and hence cannot have further biomedical use. PEGylation of chitosan chains prior to nanoparticle formation can positively contribute to the water solubility and stability of nanoparticles. Both PEG chain length and chitosan/PEG ratio are important parameters for tuning the size and surface charge of nanoparticles and hence careful optimization of these parameters are crucial in getting nanoparticles with desired properties. To the best of our knowledge, the influence of these critical parameters on the inherent properties of ionically-crosslinked and PEGylated chitosan nanoparticles has not been studied in the literature. Recently, Yang et al. investigated the impact of PEG chain length on the bioactivity of chitosan/siRNA electrocomplexes where the authors used single PEG/chitosan molar ratio and compared the differences in the circulation time of nanoparticles in vivo.23 One other detail that is not addressed in the literature is the lack of information on chitosan nanoparticle colloidal stability under physiological pH conditions. Majority of the previous studies in the literature about chitosan nanoparticles report the size and zeta potential values of nanoparticles immediately after nanoparticle formation, where the nanoparticle medium pH values range between 4.0-5.5.24-26 These values do not represent the behavior of chitosan nanoparticles at physiologically relevant pH values, as chitosan nanoparticles behave differently at different pH due to their electrostatic nature and lose their colloidal stability at physiological pH.27 In our work, we addressed this urgent need through including ‘pH of the medium’ as one of the critical parameters to be optimized both through experiments and ANN model. ANN is a computational method that mimics human brain in the processing of information and recently has become a promising approach in modeling non-linear and complex systems.28 The aim of ANN model is to investigate the effect of independent parameters on systems, and predict inherent properties of nano-systems such as stability, size, encapsulation efficiency, etc.28-35 With this approach, there is an opportunity to predict inherent properties of nanoparticle systems and hence synthesis of nanoparticles with desired properties can be achieved based on previous 3 ACS Paragon Plus Environment


experimental results. Currently, there is no standard available protocol in the literature where a researcher could utilize towards the goal of obtaining desired chitosan nanoparticle properties with minimum resources and time. Since optimizing the experimental conditions would not be feasible to obtain nanoparticles with specific properties, we propose to use ANN as an approach to predict desirable chitosan NP properties without extensive experiments. Through our approach, PEGylated chitosan nanoparticle synthesis can be standardized and nanoparticles with required properties can be synthesized without extensive experiments. Here, we present a deep insight into the PEGylation of ionically-crosslinked chitosan nanoparticles, where we precisely investigated the impact of PEG chain length, chitosan/PEG ratio and pH of the medium on the size and zeta potential of nanoparticles. We used 2, 5 and 10 kDa chain lengths of PEG for PEGylation of linear chitosan monomer. Selection of PEG chain length was based on the following criteria: The results suggested that conjugation of PEG with molecular weight lower than 2 kDa to chitosan nanoparticles would compromise colloidal stability and conjugation of PEG with molecular weight higher than 10 kDa to chitosan nanoparticles would cause loss of inherent properties of chitosan (i.e. positive charge). Chitosan nanoparticles with positive charge would be desirable for adhesivity and colloidal stability. Finally, 5 kDa PEG molecular weight was selected as an intermediate value for thorough interpretation of whole system. We also used three different PEG number of moles, i.e. 4, 12 and 20 µmole, per 25 mg of chitosan for each PEG chain length condition used. Selection of PEG mole was based on our experimental observations; we were not able to obtain monodisperse distribution of nanoparticles when we used higher than 20 µmoles or lower than 4 µmoles of PEG. Once the nanoparticles were formed with TPP crosslinker, we changed the pH of nanoparticle solutions to 6.0, 7.0 and 7.4 since pH interval of 6.0-7.4 is relevant for biomedical applications, and measured alterations in size and zeta potential values for each group. Once we confirmed the validity of the model through experiments, we developed an ANN to estimate the size and zeta potential of nanoparticles to demonstrate that we can precisely engineer the nanoparticle system and predict the resultant properties under different conditions. Furthermore, we investigated the bioadhesion capability of different group of nanoparticles with HEK293-T cell line, where experimental adhesion behavior observed between nanoparticles and cells were consistent with the predicted zeta potential values of the developed model. In summary, this study is important for understanding fundamental aspects of PEGylated chitosan nanoparticle 4 ACS Paragon Plus Environment

Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


systems, useful for possible drug/gene delivery applications to increase treatment efficiency and ultimately for clinical translation of chitosan nanoparticle systems for biomedical applications. 2. EXPERIMENTAL 2.1. Materials Low-molecular-weight chitosan, anhydrous dimethyl sulfoxide (DMSO), dialysis membrane (14 kDa cut-off), sodium tripolyphosphate (TPP), phosphate buffered saline 1x, fluorescein diacetate (FDA) were purchased from Sigma Aldrich (St. Louis, MO). 2, 5 and 10 kDa methoxypoly(ethylene glycol) succinimidyl valerate (mPEG-SVA) were obtained from Laysan Bio Inc. (Arab, AL). Glacial acetic acid (99.7+%) was purchased from Alfa Aesar (Ward Hill, MA.). Sodium hydroxide (NaOH) and hydrochloric acid (HCl) and acetone were purchased from Merck. 2,4,6Trinitrobenzene Sulfonic Acid (TNBS) was purchased from Termo Fisher Scientific. DMEM (41966), fetal bovine serum (FBS), penicillin/streptomycin, DPBS, and trypsin-EDTA (0.05%) were purchased from Gibco (Grand Island, NY). Ultrapure water was used for all experiments (Milli-Q system water - 18.2 MΩ). 2.2. Preparation of PEGylated Chitosan Low molecular weight chitosan (25 mg) was dissolved in 50 mL (2 mg·mL-1) acetic acid solution and stirred for overnight. Then, pH of the solution was adjusted to 6 with 6 M NaOH. Next, mPEG-SVA (2, 5 and 10 kDa) were dissolved in DMSO at a concentration of 50 mg·mL-1. mPEG-SVA solutions were added to chitosan solution at different moles (4, 12, 20 µmoles and others). Reactions were carried out for 2 days and reaction mixtures were dialyzed against ultrapure water for 4 days. Dialyzed mixtures were lyophilized and resulting new polymers (CSPEG) were dissolved in acetic acid (1 mg·mL-1) solution (with a final of 0.5 mg·mL-1 chitosan concentration). PEGylated chitosan solutions were stored at -20 °C until further use. Degree of PEGylation for each polymer group was analyzed with TNBS assay. Briefly, 80 µL CS-PEG polymers that were prepared with 0.5 mg·mL-1 chitosan concentration were incubated separately with 40 µL of 2% (w/v) NaHCO3 and 60 µL of 0.1% (v/v) TNBS reagent at 37 °C for 2 hours. Reactions were terminated using 1 N HCl and absorbance of samples were measured with a plate reader (Biotek, Synergy H1) at 345 nm. Unmodified chitosan at 0.5 mg·mL-1

5 ACS Paragon Plus Environment


Page 6 of 30

concentration was also prepared as a control and degree of PEGylation for different samples were calculated according to the following equation: Degree of PEGylation% = 100 -

Absorbance of Sample ×100 Absorbance of Unmodified Chitosan

(1)

PEGylation of chitosan was also confirmed with FTIR analysis. The FTIR spectra of chitosan, mPEG-SVA and PEGylated chitosan were recorded in the region of 650-4000 cm-1 with Thermo Scientific iS10 FTIR. 2.3. Synthesis of PEGylated Chitosan Nanoparticles PEGylated chitosan nanoparticles (CS-PEG NPs) were synthesized with ionotropic gelation method with some modifications.20 PEGylated chitosan solutions (0.5 mg·mL-1 chitosan concentration) were equilibrated at room temperature and pH of the solutions were adjusted to 4.8 with 6 M NaOH. TPP was dissolved (0.5 mg·mL-1) in ultrapure water and pH of the solution was adjusted to 2.5 with 6 M HCl. Chitosan and TPP solutions were mixed at a chitosan to TPP ratio of 4:1 (v/v) and reactions proceeded for 45 minutes. Different nanoparticle groups are named with their PEG chain length type and 25 mg chitosan to PEG ratio (i.e. 2 kDa - 20 µmole). 2.4. Characterization of CS-PEG NPs Size and zeta potential of nanoparticles were characterized with dynamic light scattering (DLS) technique (Zetasizer, Malvern ZS). After synthesis of nanoparticles, each particle solution was sonicated with probe sonicator for short time period and pH of each solution was adjusted with 1 M NaOH to 6.0, 7.0 and 7.4. Size and zeta potentials were measured after each pH change, the zaverage hydrodynamic diameter and zeta potential values were reported with three measurements and values were presented as means ± standard deviation. Morphology and shape of nanoparticles were analyzed using scanning transmission electron microscopy (STEM) and field emission scanning electron microscopy (FE-SEM) (Zeiss Ultra Plus, Bruker, MA). For these experiments, nanoparticle solutions were dropped onto silicon wafer and dried under vacuum. Images were collected at different magnifications. 2.5. ANN Study


Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ANN model was created using MATLAB R2018a. Samples were divided into two categories; train and test. Samples were not randomly divided; train data (26 samples) contained experimental conditions with regular increments (i.e. PEG MW of 2, 5 and 10 kDa, PEG mole 412-20 µmole, pH 6-7-7.4). These values are consistent with the experimental boundaries of the system, as was mentioned in the introduction. Test data included randomly chosen conditions within feasible experimental limits (5 samples). Two separate artificial neural networks were created to estimate the size and zeta potential of nanoparticles. Three independent input parameters were selected: (i) PEG Chain length (kDa), (ii) PEG mole per 50 mg chitosan (µmole) and (iii) pH of the solution. Output parameters for individual neural networks were size and zeta potential. Different backpropagation methods were selected to test the validity of ANN, and number of layers was altered until lowest mean squared error (MSE) and highest determination of coefficient (R2) for test data were obtained. R2 was calculated according to following equation: ∑ ( )

= − ∑

( )

(2)

and are observed, predicted and mean of the dependent variables respectively. where , 2.6. Cell Culture, Cell Viability, Cell Attachment and Cell Staining Experiments HEK293-T cell line was used for cell culture experiments, since there are numerous studies in the literature where HEK293-T cell line has been used to characterize cell survival and cell interaction of various nanoparticles.36-38 HEK293-T cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin at 37 °C and 5% CO2 for cell viability and cell attachment experiments. Viability of cells was measured using ATP-based CellTiter-Glo Luminescent Cell Viability Assay (Promega). Cells were seeded (10,000 cells per well) on 96-well plates in triplicates and treated with altered concentrations of PEGylated chitosan nanoparticles (25, 50, 75, 100, 125 µg·mL-1) for 24 h and 48 h. After the defined time intervals, cells were incubated with shaking at 100 rpm using Cell Titer-Glo reagent at 25 °C, for 15 min and luminescence was measured using a plate reader (BioTek’s Synergy H1, Winooski, VT, USA).



For cell staining experiments, a stock solution of fluorescein diacetate (FDA) was prepared by dissolving 5 mg·mL-1of FDA in acetone. The FDA working solution was obtained by adding 0.008 mL of stock solution to 5 mL of DMEM. For cell attachment experiments, 80 microliters of nanoparticle solution were dropped onto 1x1 cm glass slides. Glass slides were dried in 65 °C oven. After slides were dried, 30,000 cells were seeded on nanoparticle coated glass slides. Next, cells were incubated for 3 hours, glass slides were washed with phosphate buffered saline (PBS) and 80 µL of FDA working solution was added. The cells were stained with FDA for 4 min at room temperature under dark conditions. After this step, cell culture medium was added onto cells and cells were examined under fluorescence microscopy. Images were taken from glass slides and number of attached cells was quantified using ImageJ. 2.7. Statistical Analysis All quantitative values are presented as means ± standard deviation. All measurements were performed with at least three replicates for each group. Three-way ANOVA test was used for statistical analysis (separately for size and zeta potential, independent parameters were PEG Mole, PEG molecular weight and pH). Student’s t-test was used for cell attachment experiments. p-value of less than 0.05 was considered statistically significant. 3. RESULTS AND DISCUSSION The key steps of the experimental protocol are summarized schematically in Figure 1. First, PEGylated chitosan with different PEG molecular weights (2, 5 and 10 kDa) and chitosan/PEG ratios (25 mg chitosan to 4, 12 and 20 µmole PEG) were synthesized. Next, nanoparticles were synthesized using these new polymers via ionotropic gelation method. Size and zeta potential of different nanoparticle batches were measured at different pH values. ANN model was created using this data set, where sizes and zeta potentials for different groups of nanoparticles were predicted with the created model. As a final step, adhesion behavior of cells with altered nanoparticle sets were investigated further through the comparison of zeta potential predictions of the ANN model with the total number of cells adhered to nanoparticles from experiments. 3.1. Synthesis of PEGylated Chitosan


Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PEGylated chitosan was synthesized using NHS-Amine coupling chemistry. NHS group selectively reacts with primary amino groups, and in this work, NHS group of mPEG-SVA selectively reacted with amino groups of chitosan. We adjusted pH of the chitosan solution to 6.0, since nucleophilic/deprotonated state of amino groups reacts more efficiently with NHS group at this pH value. NHS group also reacts with water (hydrolysis reaction) and resulting product does not react with amino groups.39 NHS-amino reaction and NHS-water reaction compete with each other, and NHS-water reaction interrupts amino conjugation of NHSderivative molecules. To minimize this effect, we used mPEG-SVA which has the highest hydrolysis half-life among its counterparts.40 After this reaction, dialysis and freeze-drying steps, resulting polymers could be dissolved in water. TNBS assay was used to characterize the degree of PEGylation for different groups of PEGylated chitosan. TNBS selectively reacts with amino groups and gives chromogenic product which can be quantified in UV spectrophotometer.41 Decreases in the concentration of amino groups in PEGylated chitosan samples (compared to unmodified control group) were attributed to PEG conjugation after NHS-Amine coupling. Table 1 summarizes the degree of PEGylation of synthesized polymer groups. Although we used the same PEG (or NHS) mole per mass of chitosan, we obtained different conjugation degrees. For example, 2 kDa PEG could be conjugated to chitosan at higher amounts compared to the other groups, where 10 kDa PEG had the least amount for the degree of PEG conjugation to chitosan. This effect can be explained as follows: in the beginning, reaction rates were similar for the groups with the same mole but different PEG chain lengths. However, after some time, conjugated PEG chains started to hinder the availability of chitosan monomers. PEG chains with higher molecular weight had more hindrance effect and prevented other PEG molecules to react with amino groups of chitosan (Fig. 2). We also observed that conjugation degrees were not directly proportional with the experimental chitosan/PEG ratios used (Table 1). This effect can be explained with the high initial rates of reaction. In the beginning of the reaction, some amount of PEG reacted with chitosan, and later hindrance effects played more significant role in decreasing PEG conjugation rate. In addition, hydrolysis of NHS groups in PEG-SVA also slowed down the reactions. PEGylation of chitosan was also confirmed with FTIR analysis. 5 kDa - 20 µmol PEG condition was selected as a representative group among other PEGylated chitosan groups. The IR spectrum of mPEG-SVA has a peak at 1739 cm-1, which corresponds to the free carboxylic group (C=O) 9 ACS Paragon Plus Environment


on PEG chain (Supporting Information, Figure S1). This peak completely disappeared for PEGylated chitosan, which was a further evidence of PEGylation reaction. Furthermore, amide bond formation was noted at 1562.54 cm-1 for PEGylated chitosan. PEGylated chitosan also retained characteristic broad –OH peak between 3000-3500 cm-1 region, it also retained residual C=O bonds present in residual acetyl groups on 1655.10 cm-1. These results strongly confirmed that PEG was chemically conjugated to chitosan. 3.2. Synthesis of PEGylated Chitosan Nanoparticles Positively charged chitosan and negatively charged TPP form polyelectrolyte complexes under rigorous stirring. We synthesized chitosan nanoparticles using 9 different types of PEGylated chitosan monomer. After preparation of nanoparticles, the resulting pH of the solutions were measured in between 4.6-4.7 for all samples and the pH of solutions were increased to 6.0, 7.0 and 7.4. Next, nanoparticles were characterized with respect to their particle size and zeta potential. Observed size and zeta potential values for all groups are presented in Table 2 (Raw data is presented in Supporting Information, Figure S2-S53). Fig. 3A demonstrates the size and zeta potential of nanoparticles at different pH values for all PEG molecular weights (MWs). Similar trend was observed for all PEG MWs; size of nanoparticles were increased when we increased the pH of the solutions. The most dramatic changes in nanoparticle size were observed for the least PEGylated (4 µmole) groups of 2 and 5 kDa PEG MW. 2 kDa - 4 µmole group aggregated irreversibly at pH 7.4. This suggested that 29.9% degree of PEGylation (for 2 kDa) was not sufficient to synthesize nanoparticles with colloidal stability at a pH of 7.4. However, we were able to obtain nanoparticles with colloidal stability using 2 kDa - 12 µmole group which had 32.7% degree of PEGylation. This shows that there is a threshold value for the degree of PEGylation between 29.7-32.7% with 2 kDa PEG. The group with 5 kDa - 4 µmole PEG group resulted in a dramatic increase in the size from 74.87±0.96 to 153.13±5.65 nm (Table 2). Among all groups studied at pH 6.0 and pH 7.0, 2 kDa - 4 µmole PEG group had the highest zeta potential. Commonly, nanoparticles with higher zeta potential values have been evaluated to have better stability;42-43 however, our observation suggests that inherent factors of chitosan nanoparticle system play more important role on the nanoparticle colloidal stability than the value of zeta potential. For the group with 10 kDa - 4 µmole PEG, the observed change was not as dramatic as it were with 2 and 5 kDa PEG even though it had lower zeta potential at 7.4 pH. This result could be explained by the effect of PEG chain length, which plays an important role 10 ACS Paragon Plus Environment

Page 10 of 30

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


on the colloidal stability of the system. That is, higher PEG chain lengths have more positive contribution on the general stability of the system. This phenomenon had less influence at higher PEG doses of 12 and 20 µmole where pH changes resulted in less dramatic size changes for 2, 5 and 10 kDa PEG. Although, 10 kDa contributes more to the stability of the system in general, using higher amounts of 2 or 5 kDa PEG can be a solution to obtain nanoparticles with comparable colloidal stability. It is also important to consider the clinical applicability and the need for high circulation time of nanoparticles, where chitosan nanoparticles with higher PEG chain length were found to have much higher circulation time in vivo.23 We also observed significant size alterations between pH 7.0 and 7.4 for all groups. This was probably caused by the decrease in repulsion effects, which finally resulted in increases in size of the nanoparticle system. In summary, two main factors probably play important roles on the stability on PEGylated chitosan nanoparticles: (i) hydrogen bonding between grafted PEG chains and water influences the stability of nanoparticle system, where PEG with more or longer chains on the chitosan backbone increase overall solubility or stability, (ii) stronger repulsion forces between nanoparticles contribute to the stability of the system; however, this has much less effect compared to the former one. Zeta potential of all groups decreased while pH was increased from 6.0 to 7.4. Deprotonation of amino groups resulted in decreases in zeta potential values. Higher PEG doses resulted in less zeta potential in individual PEG groups as expected (2, 5 and 10 kDa) (Fig. 3B). An interesting phenomenon was observed in zeta potentials of different groups; even though more amino groups were consumed after the synthesis of groups that have smaller chain length (Table 1), zeta potentials of these groups were much higher than groups with less conjugation with higher PEG chain length (Table 2). It was expected that more amino group containing nanoparticles would result in higher zeta potential values as protonated amino groups contribute the surface charge. However, results suggested an opposite trend (Fig. 3B). This opposite trend could be explained by the hindrance of surface charges of nanoparticles by conjugated PEG macromolecules with higher chain length (Fig. 4). This effect became more obvious for nanoparticles conjugated with 10 kDa PEG. For example, 10 kDa - 20 µmole PEG group has 21.3% degree of PEGylation, in other words, 21.3% of total amino groups were consumed during PEGylation. On the other hand, 2 kDa - 20 µmole group has 37.4% degree of PEGylation which means that this group has much less number of amino groups compared to 10 kDa-20 11 ACS Paragon Plus Environment


µmole group. When nanoparticles were formed from these two groups, zeta potentials of nanoparticles were measured as 3.28±3.04 and 11.5±4.53 for 10 kDa - 20 µmole and 2 kDa-20 µmole groups, respectively at pH 6.0. Even though 10 kDa - 20 µmole group should have more number of amino groups, it has about three times less amount of zeta potential compared to that observed for 2 kDa - 20 µmole group. Similar trends were observed for all groups, which strongly confirmed the opposite trends suggested above. Morphology and shape of nanoparticles were analyzed using FE-SEM and STEM. For FE-SEM experiments, 5 kDa - 20 µmole and 10 kDa - 20 µmole groups were selected as representative groups. FE-SEM experiments demonstrated that nanoparticles have spherical shape, and that diameter of nanoparticles were measured as consistent with DLS results (Supporting Information, Figure S54-55). For STEM experiments, 2 kDa - 20 µmole group is selected as representative group. STEM images confirmed that nanoparticles have spherical shape (Supporting Information, Figure S56). Cytotoxicity of synthesized nanoparticles was evaluated using CellTiter-Glo viability assay after 24 and 48 hours. HEK293-T cells were exposed to nanoparticles at doses of 25, 50, 75, 100 and 125 µg·mL-1. Cell survival in all groups had more than 80% viability for the highest dose of nanoparticle concentration at 24 hours (Supporting Information, Figure S57). For 48 hours, there was a similar trend, where cells had higher viability at high doses of nanoparticle condition compared to that observed at 24 hours (Supporting Information, Figure S58). These results suggest that PEGylated chitosan NPs are biocompatible at relatively high doses. 3.3. ANN Model ANNs with 9 backpropagation methods were created to test validity of the networks. MSE and R2 values of test data (for size and zeta potential separately) are presented in Table S1. Optimization of ANNs for both size and zeta potential is presented in Figure 5. Three representative backpropagation algorithms (Levenberg–Marquardt, Bayesian Regularization, Gradient descent with adaptive learning rate) were selected among 9 to illustrate the results of optimization process and the results are presented in Figure 5 (Supporting Information, Table S1). For the size of nanoparticles, Levenberg–Marquardt algorithm with 1 layer was selected, since this method resulted in the highest R2 and the lowest MSE value (Figure 5A). R2 values were calculated as 0.8734 and 0.7869 for train and test data, respectively. These values certify 12 ACS Paragon Plus Environment

Page 12 of 30

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


that the created model is competent for further prediction of other groups.44 For zeta potential, Gradient descent with adaptive learning rate backpropagation method with 4 layers was selected since it has the lowest MSE and highest R2 value (Figure 5B). R2 were found as 0.9892 and 0.8987 for train and test data, respectively and these values suggest that the model has strong predictive capability to predict zeta potential of nanoparticles. Training parameters of ANNs is presented in Table 3. Test results demonstrate that ANNs could accurately estimate the size and zeta potential of different groups (Table 4 includes the comparison of experimental observation and predictions, Supporting Info, Figure S59-68 includes the raw data of experimental observations). Thus, PEGylated chitosan nanoparticles with desired properties can be synthesized based on the neural network results. This model allows for reducing the experimental efforts significantly towards obtaining nanoparticles with desirable properties (2 days for synthesis, 4 days for dialysis and 2 days for freeze drying). As it will be shown in the following section, interaction between cells and nanoparticles is highly sensitive due to zeta potential differences, where these interactions between cells and nanoparticles can be estimated with our model. This may be further useful for other applications such as gene delivery for tuning transfection efficiency. 3.4. Cell Attachment Experiments Cationic polymers electrostatically interact with negatively charged plasma membrane of cells.4546

Thus, it is expected that nanoparticles with higher positive charge interact more with cells. To

demonstrate that surface charge has direct impact on cell-nanoparticle interaction, we performed two sets of experiments. In the first set, we fabricated nanoparticles from 2 kDa - 12 µmole PEG group and coated these nanoparticles on glass slides. We seeded the cells on the uncoated and nanoparticle coated glass slides and we used cell culture mediums with pH 6.5 and 7.4 to present different charges to nanoparticles (Fig 6A). After 3 hours, it was observed that higher number of cells attached on the nanoparticle coated glass slides compared to that of uncoated surfaces particularly at pH 6.5 (Fig 6B). When the pH was at 6.5, the number of cells attached on the surface was greater compared to the condition at pH 7.4. This can be explained by the higher surface charge of nanoparticles at low pH condition, which promotes the adhesion of cells (Fig. 6A, 6B). Also, at pH 7.4, it was observed that the number of cells attached on uncoated glass slides were greater than that of coated surface. There could be two reasons for this observation: (i) charge of nanoparticles at pH 7.4 was not strong enough to attach cells, (ii) a different 13 ACS Paragon Plus Environment


attachment mechanism between cells and glass slide were dominant at that particular pH. Once we confirmed the direct influence of pH/surface charge on cell and nanoparticle interaction, we performed second set of experiments. In the second set, we compared 2, 5 and 10 kDa - 12 µmole nanoparticle groups at a pH of 6.5. We measured zeta potential values (at pH 6.5) of these three groups and measured as 8.35, 6.87 and 4.94 mV for 2 kDa - 12 µmole, 5 kDa - 12 µmole and 10 kDa - 12 µmole groups, respectively (Supporting Information, Figure S60, 69 and 70). Then, we performed cell attachment experiments with these three groups and observed that 2 kDa group had the highest number of attached cells as expected and uncoated sample had the lowest number of attached cells (Fig. 7A, 7B). There was a linear trend between 2, 5 and 10 kDa - 12 µmole groups in terms of number of attached cells (Fig. 7C). We used this linear trend as a “calibration curve”, which made it possible to estimate the interaction between nanoparticles and cells, and hence number of attached cells on the coated glass slides (Fig. 7C). First, we used ANN model to predict zeta potential value of a different nanoparticle group at pH 6.5 and then substituted this value into equation of the fitted line. Resulting value gave the number of attached cells (relative to control group). We used 10 kDa - 4 µmole group to test the prediction of ANN model, where estimated zeta potential value was 7.34 mV at pH 6.5. Then, we substituted this zeta potential value into the equation of the fitted line and we obtained 12,968 attached cells (relative to the control group). Finally, we performed the real experiment to compare this prediction with the experimental observation and determined that 9,971 of cells were attached on the surface (relative to the control group). This confirmed further that the model had relatively good ability to predict the number of attached cells at a particular condition. CONCLUSIONS Here, we systematically investigated the impact of PEG chain length and chitosan/PEG ratio on size and zeta potential of nanoparticles at different pH values. We synthesized PEGylated chitosan with different PEG chain lengths and found that smaller chain lengths of PEG reacted more with chitosan. We also explored that nanoparticles conjugated with higher PEG chain length had smaller zeta potential value despite the presence of higher number amino groups. This can be attributed to the hindrance of the surface charge of particles by longer PEG chains. Once fundamental parameters of this experimental system were defined, we developed an ANN model based on our experimental data. We used this model to predict size and zeta potential values of nanoparticle groups at different conditions and confirmed the validity of the ANN model. This 14 ACS Paragon Plus Environment

Page 14 of 30

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


suggests that, PEGylated chitosan nanoparticle system can be precisely engineered, nanoparticles with desired size and surface charges can be obtained based on the information obtained from the ANN model. As a final step, we demonstrated alterations in the degree of attachment of cells on surfaces treated with various nanoparticle groups in vitro. We were able to estimate interaction between nanoparticles and cells based on ANN results. In conclusion, this study demonstrates fundamental aspects of PEGylated chitosan nanoparticle systems and proves that nanoparticles with the desired surface charge and size properties can be prepared in conjunction with the ANN model. This study will also be useful for optimal design conditions of various delivery systems such as nasal delivery route, for possible drug and gene delivery applications.

ASSOCIATED CONTENT Supporting Information FT-IR analysis for PEGylation of chitosan. Raw DLS results for training data. FE-SEM and STEM experiments. Cell viability tests. ANN Optimization. Raw DLS results for test data. Additional data for cell attachment experiments. AUTHOR INFORMATION Corresponding Author *email: [email protected], fax: +90-212-338-1548 ACKNOWLEDGEMENTS This study is supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under International Support Program (COST Action - European Cooperation in Science and Technology - CA15216, project number: 116M995). Authors also thank Prof. Ugur Unal and Dr. Baris Yagci for STEM and FE-SEM experiments. REFERENCES 1. Lavan, D. A.; McGuire, T.; Langer, R., Small-Scale Systems for in Vivo Drug Delivery. Nat. Biotechnol. 2003, 21, 1184-1191. 2. Ferrari, M., Cancer Nanotechnology: Opportunities and Challenges. Nat. Rev. Cancer 2005, 5, 161-171. 15 ACS Paragon Plus Environment


3. Farokhzad, O. C.; Karp, J. M.; Langer, R., Nanoparticle–Aptamer Bioconjugates for Cancer Targeting. Expert Opin. Drug Delivery 2006, 3, 311-324. 4. Zhang, L.; Gu, F.; Chan, J.; Wang, A.; Langer, R.; Farokhzad, O., Nanoparticles in Medicine: Therapeutic Applications and Developments. Clin. Pharmacol. Ther. 2008, 83, 761769. 5. Farokhzad, O. C.; Langer, R., Impact of Nanotechnology on Drug Delivery. ACS Nano 2009, 3, 16-20. 6. Sun, T.; Zhang, Y. S.; Pang, B.; Hyun, D. C.; Yang, M.; Xia, Y., Engineered Nanoparticles for Drug Delivery in Cancer Therapy. Angew. Chem., Int. Ed. 2014, 53, 1232012364. 7. Rinaudo, M., Chitin and Chitosan: Properties and Applications. Prog. Polym. Sci. 2006, 31, 603-632. 8. Rodrigues, S.; Dionísio, M.; López, C. R.; Grenha, A., Biocompatibility of Chitosan Carriers with Application in Drug Delivery. J. Funct. Biomater. 2012, 3, 615-641. 9. Garg, N. K.; Dwivedi, P.; Campbell, C.; Tyagi, R. K., Site Specific/Targeted Delivery of Gemcitabine Through Anisamide Anchored Chitosan/Poly Ethylene Glycol Nanoparticles: An Improved Understanding of Lung Cancer Therapeutic Intervention. Eur. J. Pharm. Sci. 2012, 47, 1006-1014. 10. Jiang, X.; Dai, H.; Leong, K. W.; Goh, S. H.; Mao, H. Q.; Yang, Y. Y., Chitosan‐g‐ PEG/DNA Complexes Deliver Gene to the Rat Liver via Intrabiliary and Intraportal Infusions. J. Gene Med. 2006, 8, 477-487. 11. Hirpara, M. R.; Manikkath, J.; Sivakumar, K.; Managuli, R. S.; Gourishetti, K.; Krishnadas, N.; Shenoy, R. R.; Jayaprakash, B.; Rao, C. M.; Mutalik, S., Long Circulating PEGylated-Chitosan Nanoparticles of Rosuvastatin Calcium: Development and in Vitro and in Vivo Evaluations. Int. J. Biol. Macromol. 2018, 107, 2190-2200. 12. Mohammed, M. A.; Syeda, J.; Wasan, K. M.; Wasan, E. K., An Overview of Chitosan Nanoparticles and Its Application in Non-Parenteral Drug Delivery. Pharmaceutics 2017, 9, 53. 13. Kean, T.; Thanou, M., Biodegradation, Biodistribution and Toxicity of Chitosan. Adv. Drug Delivery Rev. 2010, 62, 3-11. 14. Bernkop-Schnürch, A.; Dünnhaupt, S., Chitosan-Based Drug Delivery Systems. Eur. J. Pharm. Biopharm. 2012, 81, 463-469. 15. Sogias, I. A.; Khutoryanskiy, V. V.; Williams, A. C., Exploring the Factors Affecting the Solubility of Chitosan in Water. Macromol. Chem. Phys. 2010, 211, 426-433. 16. Casettari, L.; Vllasaliu, D.; Castagnino, E.; Stolnik, S.; Howdle, S.; Illum, L., PEGylated Chitosan Derivatives: Synthesis, Characterizations and Pharmaceutical Applications. Prog. Polym. Sci. 2012, 37, 659-685. 17. El-Shabouri, M., Positively Charged Nanoparticles for Improving the Oral Bioavailability of Cyclosporin-A. Int. J. Pharm. 2002, 249, 101-108. 18. Brunel, F.; Véron, L.; David, L.; Domard, A.; Delair, T., A Novel Synthesis of Chitosan Nanoparticles in Reverse Emulsion. Langmuir 2008, 24, 11370-11377. 19. Maitra, A.; Ghosh, P. K.; De, T. K.; Sahoo, S. K., Process for the Preparation of Highly Monodispersed Polymeric Hydrophilic Nanoparticles. Google Patents: 1999. 20. Calvo, P.; Remuñán-López, C.; Vila-Jato, J. L.; Alonso, M., Novel Hydrophilic ChitosanPolyethylene Oxide Nanoparticles as Protein Carriers. J. Appl. Polym. Sci. 1997, 63, 125-132. 21. Divya, K.; Jisha, M., Chitosan Nanoparticles Preparation and Applications. Environ. Chem. Lett. 2017, 16, 101–112. 16 ACS Paragon Plus Environment

Page 16 of 30

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


22. Grenha, A., Chitosan Nanoparticles: A Survey of Preparation Methods. J. Drug Targeting 2012, 20, 291-300. 23. Yang, C.; Gao, S.; Dagnæs-Hansen, F.; Jakobsen, M.; Kjems, J., Impact of PEG Chain Length on the Physical Properties and Bioactivity of PEGylated Chitosan/siRNA Nanoparticles in Vitro and in Vivo. ACS Appl. Mater. Interfaces 2017, 9, 12203-12216. 24. Fan, W.; Yan, W.; Xu, Z.; Ni, H., Formation Mechanism of Monodisperse, Low Molecular Weight Chitosan Nanoparticles by Ionic Gelation Technique. Colloids Surf., B 2012, 90, 21-27. 25. Makhlof, A.; Tozuka, Y.; Takeuchi, H., Design and Evaluation of Novel pH-Sensitive Chitosan Nanoparticles for Oral Insulin Delivery. Eur. J. Pharm. Sci. 2011, 42, 445-451. 26. Huang, Y.; Cai, Y.; Lapitsky, Y., Factors Affecting the Stability of Chitosan/Tripolyphosphate Micro-and Nanogels: Resolving the Opposing Findings. J. Mater. Chem. B 2015, 3, 5957-5970. 27. Gan, Q.; Wang, T.; Cochrane, C.; McCarron, P., Modulation of Surface Charge, Particle Size and Morphological Properties of Chitosan–TPP Nanoparticles Intended for Gene Delivery. Colloids Surf., B 2005, 44, 65-73. 28. Amani, A.; Mohammadyani, D., Artificial Neural Networks: Applications in Nanotechnology. In Artificial Neural Networks-Application, InTech: 2011. 29. Baharifar, H.; Amani, A., Size, Loading Efficiency, and Cytotoxicity of Albumin-Loaded Chitosan Nanoparticles: An Artificial Neural Networks Study. J. Pharm. Sci. 2017, 106, 411417. 30. Hashad, R. A.; Ishak, R. A.; Fahmy, S.; Mansour, S.; Geneidi, A. S., ChitosanTripolyphosphate Nanoparticles: Optimization of Formulation Parameters for Improving Process Yield at a Novel pH Using Artificial Neural Networks. Int. J. Biol. Macromol. 2016, 86, 50-58. 31. Baharifar, H.; Amani, A., Cytotoxicity of Chitosan/Streptokinase Nanoparticles as a Function of Size: An Artificial Neural Networks Study. Nanomedicine 2016, 12, 171-180. 32. Aslan, C.; Çelebi, N.; Değim, Đ. T.; Atak, A.; Özer, Ç., Development of Interleukin-2 Loaded Chitosan-Based Nanogels Using Artificial Neural Networks and Investigating the Effects on Wound Healing in Rats. AAPS PharmSciTech 2017, 18, 1019-1030. 33. Ali, H. S.; Blagden, N.; York, P.; Amani, A.; Brook, T., Artificial Neural Networks Modelling the Prednisolone Nanoprecipitation in Microfluidic Reactors. Eur. J. Pharm. Sci. 2009, 37, 514-522. 34. Amani, A.; York, P.; Chrystyn, H.; Clark, B. J., Factors Affecting the Stability of Nanoemulsions—Use of Artificial Neural Networks. Pharm. Res. 2010, 27, 37-45. 35. Ketabchi, N.; Naghibzadeh, M.; Adabi, M.; Esnaashari, S. S.; Faridi-Majidi, R., Preparation and Optimization of Chitosan/Polyethylene Oxide Nanofiber Diameter Using Artificial Neural Networks. Neural Comput. Appl. 2017, 28, 3131-3143. 36. Woźniak, A.; Malankowska, A.; Nowaczyk, G.; Grześkowiak, B. F.; Tuśnio, K.; Słomski, R.; Zaleska-Medynska, A.; Jurga, S., Size and Shape-Dependent Cytotoxicity Profile of Gold Nanoparticles for Biomedical Applications. J. Mater. Sci.: Mater. Med. 2017, 28, 92. 37. Jiang, X.; Lu, C.; Tang, M.; Yang, Z.; Jia, W.; Ma, Y.; Jia, P.; Pei, D.; Wang, H., Nanotoxicity of Silver Nanoparticles on HEK293T Cells: A Combined Study Using Biomechanical and Biological Techniques. ACS Omega 2018, 3, 6770-6778. 38. Wang, P.; Wang, X.; Wang, L.; Hou, X.; Liu, W.; Chen, C., Interaction of Gold Nanoparticles with Proteins and Cells. Sci. Technol. Adv. Mater. 2015, 16, 034610.



39. Cline, G. W.; Hanna, S. B., Kinetics and Mechanisms of the Aminolysis of NHydroxysuccinimide Esters in Aqueous Buffers. J. Org. Chem. 1988, 53, 3583-3586. 40. Le, Y.; Scott, M. D., Immunocamouflage: the Biophysical Basis of Immunoprotection by Grafted Methoxypoly (Ethylene Glycol)(mPEG). Acta Biomater. 2010, 6, 2631-2641. 41. Habeeb, A. S. A., Determination of Free Amino Groups in Proteins by Trinitrobenzenesulfonic Acid. Anal. Biochem. 1966, 14, 328-336. 42. Bhattacharjee, S., DLS and Zeta Potential–What They are and What They are Not? J. Controlled Release 2016, 235, 337-351. 43. Patel, V. R.; Agrawal, Y., Nanosuspension: An Approach to Enhance Solubility of Drugs. J. Adv. Pharm. Technol. Res. 2011, 2, 81-87. 44. Alexander, D.; Tropsha, A.; Winkler, D. A., Beware of R 2: Simple, Unambiguous Assessment of the Prediction Accuracy of QSAR and QSPR Models. J. Chem. Inf. Model. 2015, 55, 1316-1322. 45. Fröhlich, E., The Role of Surface Charge in Cellular Uptake and Cytotoxicity of Medical Nanoparticles. Int. J. Nanomed. 2012, 7, 5577-5591. 46. Nafee, N.; Schneider, M.; Schaefer, U. F.; Lehr, C.-M., Relevance of the Colloidal Stability of Chitosan/PLGA Nanoparticles on Their Cytotoxicity Profile. Int. J. Pharm. 2009, 381, 130-139.

FIGURES AND CAPTIONS


Page 18 of 30

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Key steps of the study.



Figure 2. Demonstration of hindrance effects caused by longer PEG chains. Longer PEG chains hinder available amino group containing chitosan monomers which results in lower conjugation efficiency.


Page 20 of 30

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Three dimensional surface plots based on experimental observations for size and zeta potential of nanoparticles with respect to different independent variables. Plots were sketched using Table 2. A) Size and zeta potentials of nanoparticles for different PEG chain lengths. B) Zeta potentials of different PEGylated nanoparticles at different pH values. Note: p

Deep Insight into PEGylation of Bioadhesive Chitosan Nanoparticles

Recommend Documents