Enzyme-encapsulating polymeric nanoparticles: A potential adjunctive therapy in Pseudomonas aeruginosa biofilm-associated infection treatment

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Enzyme-encapsulating polymeric nanoparticles: A potential adjunctive therapy in Pseudomonas aeruginosa biofilm-associated infection treatment

T

Chendong Hana, James Goodwineb,c, Nicholas Romeroa, Kyle S. Steckb, Karin Sauerb,c, ⁎ Amber Doirona,c,d, a

Department of Biomedical Engineering, Binghamton University, Binghamton, NY, 13902, United States Department of Biological Science, Binghamton University, Binghamton, NY, 13902, United States c Binghamton Biofilm Research Center, Binghamton University, Binghamton, NY, 13902, United States d Department of Electrical and Biomedical Engineering, University of Vermont, Burlington, VT, 05405, United States b

A R T I C LE I N FO

A B S T R A C T

Keywords: Drug delivery Enzyme immobilization Poly(lactic-co-glycolic) acid PLGA

Pseudomonas aeruginosa is a pathogen known to be associated with a variety of diseases and conditions such as cystic fibrosis, chronic wound infections, and burn wound infections. A novel approach was developed to combat the problem of biofilm antibiotic tolerance by reverting biofilm bacteria back to the planktonic mode of growth. This reversion was achieved through the enzymatic depletion of available pyruvate using pyruvate dehydrogenase, which induced biofilm bacteria to disperse from the surface-associated mode of growth into the surrounding environment. However, direct use of the enzyme in clinical settings is not practical as the enzyme is susceptible to denaturation under various storage conditions. We hypothesize that by encapsulating pyruvate dehydrogenase into degradable, biocompatible poly(lactic-co-glycolic) acid nanoparticles, the activity of the enzyme can be extended to deplete available pyruvate and induce dispersion of mature Pseudomonas aeruginosa biofilms. Several particle formulations were attempted in order to permit the use of the smallest dose of nanoparticles while maintaining pyruvate dehydrogenase activity for an extended time length. The nanoparticles synthesized using the optimal formulation showed an average size of 266.7 ± 1.8 nm. The encapsulation efficiency of pyruvate dehydrogenase was measured at 17.9 ± 1.4%. Most importantly, the optimal formulation dispersed biofilms and exhibited enzymatic activity after being stored at 37 °C for 6 days.

1. Introduction It is estimated that more than 60% of chronic infections are related to biofilms, which are commonly defined as communities of surfaceassociated bacteria embedded in a self-produced, hydrated polymeric matrix [1,2]. Biofilm infections have become a prominent focus of attention due to their extraordinary tolerance to traditional antimicrobial therapy. Biofilm bacteria are up to 1000-times less susceptible to traditional antimicrobial agents than their planktonic counterparts [3,4]. Antibiotics are a primary therapy option for biofilm infections, yet high doses are required, which can be problematic due to concerns over toxicity and the selection of antibiotic-resistant bacteria [5]. Surgical removal of recalcitrant biofilm infections is an option after antibiotic failure; however, complete removal is difficult and biofilm-contamination of surgical instruments is a concern. Pseudomonas aeruginosa is an opportunistic pathogen that causes infections in patients that are immunocompromised. P. aeruginosa is

⁎

known to be associated with a variety of diseases and conditions such as cystic fibrosis, chronic wound infections, and burn wound infections. P. aeruginosa possesses a genome that encodes a number of resistance genes such as efflux pumps and antibiotic-inactivating enzymes, which make it a challenging task to treat with conventional antibiotics [5]. Despite the biofilm generally constituting a protective mode of growth for P. aeruginosa, stressful environments can develop within the biofilm. For instance, oxygen availability varies throughout the P. aeruginosa biofilm architecture, with the central core of the biofilm being essentially anoxic. The cells at the core of the biofilm therefore experience an environment of reductive stress (abundant electrons/insufficient oxygen) [6]. The proliferation of P. aeruginosa in such anaerobic environments can be supported by denitrification and/or arginine fermentation in the presence of corresponding substrates [7]. In denitrification, P. aeruginosa utilizes nitrate or nitrite as an alternative electron acceptor for anaerobic respiration [8,9]. When nitrate is not available but arginine is present, available arginine can be fermented to

Corresponding author at: Department of Electrical and Biomedical Engineering, University of Vermont, Burlington, VT, 05405, United States. E-mail address: [email protected] (A. Doiron).

https://doi.org/10.1016/j.colsurfb.2019.110512 Received 31 March 2019; Received in revised form 13 September 2019; Accepted 15 September 2019 Available online 20 September 2019 0927-7765/ © 2019 Elsevier B.V. All rights reserved.

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

2.1. Materials

support anaerobic growth [10]. When nitrate/nitrite and arginine are absent but pyruvate is present, P. aeruginosa can ferment pyruvate under anoxic conditions to sustain long-term survival. A previous report analyzing the growth medium and cell viability has shown that pyruvate fermentation supports the survival of P. aeruginosa for up to 18 days compared to the control group where pyruvate was absent [7]. Pyruvate also plays an essential role in biofilm formation; the addition of exogenous pyruvate significantly promoted the development of P. aeruginosa biofilms, while conversely, the depletion of pyruvate from growth medium impaired biofilm formation [11]. Furthermore, the use of pyruvate dehydrogenase (PDH) to deplete pyruvate from growth medium coincided with the dispersion of existing P. aeruginosa biofilms [12]. Hence, enzymatic depletion of pyruvate through PDH provides a potential therapeutic means to controlling P. aeruginosa biofilm infections, where biofilm bacteria are reverted to an antibiotic-susceptible state via pyruvate-depletion induced dispersion [12]. However, like the majority of enzymes, PDH exhibits low stability in varying temperature storage and decreasing activity over time in vitro. Immobilization of an enzyme into or onto solid supports is a proven and effective strategy to overcoming such drawbacks. In comparison to free enzymes, immobilized enzymes are more stable under harsh physical and chemical conditions [13]. Upon immobilization, the handling of enzymes becomes more convenient, and it eases the separation of enzymes from their products [14]. In addition, immobilization of enzymes often enables sustained enzyme activity over longer periods of time, which is ideal from an economic point of view due to the high cost of enzymes [15]. The objective of the present study is to identify the most effective of the tested formulations in enzyme encapsulation and preservation of enzymatic activity. Specifically, PDH was entrapped in nanoparticles, where PDH was restricted in a confined polymer network, porous to substrates and products but not the enzyme [13]. A release study showed that only 0.0002% of the encapsulated PDH was detected by ELISA after a six-day incubation at 37 °C, indicating PDH was immobilized into NPs and not released within the time periods of interest here.

PLGA 50:50 (Mw 38,000-54,000), PLGA 75:25 (Mw 66,000100,000), PDH from porcine heart, thiamine pyrophosphate (TPP), βnicotinamide adenine dinucleotide sodium salt (β-NAD+), coenzyme A (CoA), and rhodamine 6 G (Rho) used in this research were purchased from Sigma-Aldrich (St. Louis, Missouri, USA). LIVE/DEAD™ stain was purchased from Thermo Fisher Scientific (Waltham, Massachusetts, USA). Lysogeny broth (LB) was purchased from BD (Franklin Lakes, New Jersey, USA). Polyvinyl alcohol (PVA, 5%) was purchased from Ward’s Science (470302, Henrietta, New York, USA). Trehalose (99% anhydrous) was purchased from Acros Organics (New Jersey, USA). 3(N-morpholino)propanesulfonic acid (MOPS) was purchased from Amresco (Solon, Ohio, USA). Precision Plus Protein™ dual color standards, 1x tris/glycine/SDS running buffer, PowerPac™ basic power supply, Mini-PROTEAN tetra electrophoresis cell, 4–15% MiniPROTEAN® TGX™ precast protein gels (12-well, 20 μl), Trans-Blot® Turbo™ transfer starter system, mini polyvinylidene difluoride (PVDF) membrane, 5x transfer buffer (25 mM Tris, 192 mM glycine, pH 8.3), tween 20 (100% nonionic detergent), 1x tris buffered saline (TBS), and Clarity Max Western ECL substrate were purchased from Biorad (Hercules, California, USA). Bovine serum albumin was purchased from VWR (Radnor, Pennsylvania, USA). Anti-E1α rabbit antibody (product number ab168379) was purchased from Abcam (Cambridge, Massachusetts, USA). Horseradish peroxidase (HRP)-anti-rabbit IgG antibody (product number 7074) was purchased from Cell Signaling Technology (Danvers, Massachusetts, USA). 2.2. Synthesis and optimization of PDH-loaded PLGA nanoparticles PLGA nanoparticles were synthesized using a W/O/W double emulsion method (Fig. 1) [23,24]. Briefly, a mixture of PDH with filtered MOPS (pH 7.4) in a protein concentrator (Pierce, City, State, molecular weight cutoff of 100 kDa) was centrifuged at 6500×g for 4 h. Five different amounts of PDH were used in order to synthesize nanoparticles with optimum encapsulation and enzyme activity. The concentrate recovered was dispensed into fresh MOPS to make the inner aqueous phase in a total volume of 400 μL. When blank PLGA nanoparticles were synthesized, the inner aqueous phase was 400 μL MOPS buffer. The organic phase was 4 mL acetone containing 100 mg PLGA 50:50 or 50 mg PLGA 75:25, respectively. The inner aqueous phase was emulsified in the oil phase under sonication and vortex for 40 s each. When rhodamine 6 G (Rho) was included to aid in vitro particle localization, it was added into the organic phase at the concentration of 0.01 mg/mL. The solution was then transferred to 7 ml of an outer aqueous phase under sonication and vortexing. When PLGA 50:50 was used, the outer aqueous phase was 0.1% PVA solution; when PLGA 75:25 was used, the outer aqueous phase was 2% PVA solution. The obtained emulsion was diluted in 50 mL of 0.03% w/v PVA, and acetone was evaporated under moderate magnetic stirring and vacuum. The solution was then centrifuged at 35,000×g for 15 min, decanted, and re-suspended in water to wash away any unencapsulated PDH; this was repeated 3 times. Particles were either used fresh or frozen overnight in the presence of 400 mg cryoprotectant trehalose and freeze dried before analysis. The mass of freeze-dried nanoparticles was measured and used to calculate the yield of the entire batch.

2. Experimental The synthetic copolymer poly(lactic-co-glycolic acid) (PLGA) was selected for this study due to its biocompatibility, degradability, and inclusion in several FDA-approved applications [16,17]. The ester linkages in PLGA go through hydrolysis when water is present, and the polymer degradation products lactic acid and glycolic acid are natural byproducts of several human metabolic pathways [18]. PLGA-based colloidal drug delivery systems have been explored extensively and are available on the clinical market for the treatment of prostate cancer, pediatric growth hormone deficiency, acromegaly, and periodontal disease [17]. Generally, PLGA nanoparticles (PLGA NPs) have been reported to have sizes ranging from 200 nm to 400 nm and exhibit a negative zeta potential without chemical surface functionalization [19]. A number of macromolecular agents, such as proteins, peptides, vaccines, growth factors, and genes, have been encapsulated into PLGANPs synthesized using different methods with varying polymer molecular weight and hydrophobicity as well as particle surface charge. Among many different PLGA NP synthesis methods, the water-in-oil-inwater (W/O/W) double emulsion method was selected in this study due to the inner aqueous phase that helps stabilize the hydrophilic PDH. PDH is a large enzyme complex composed of three subunits, with a molecular weight over 100 kDa, which is much larger than commonly used protein models such as bovine serum albumin (BSA, 68 kDa) and lysozyme (14.3 kDa). In addition, the diameter of PDH ranges from 25 nm to 40 nm [20–22]. As far as the authors are aware, no one has previously reported encapsulation of such a large macromolecule into relatively small PLGA NPs.

2.3. Characterization of PDH-loaded PLGA nanoparticles Size distribution and surface charge were measured by electrophoretic dynamic light scattering (DLS, Malvern Zetasizer Nano-ZS, Malvern, UK). The morphology of PLGA nanoparticles was evaluated by scanning electron microscopy (SEM, Zeiss Supra 55 V P, Dublin, California, USA) and transmission electron microscopy (TEM, JEM 2100 F from JEOL, Peabody, Massachusetts, USA). The amount of encapsulated PDH was determined by Western blot 2

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

Fig. 1. Schematic of PLGA nanoparticles preparation using W/O/W double emulsion method.

batch. Modified from the manufacturer’s instruction, reactions were carried out at room temperature in a total volume of 1 mL. Typical concentrations of PDH cofactors and pyruvate used for activity assays were as follows: 2 mM ß-NAD+, 2 mM CoA, 20 μM TPP, 50 μM MgSO4, and 10 mM pyruvate. A total of 100 μL of each reaction was loaded into a clear 96-well plate, and absorbance at 340 nm was measured at t = 0, 15, 30, and 60 min at 25 °C using a SpectraMax™ i3x microplate reader (Molecular Devices). Particles were centrifuged out of the suspension before reading the absorbance in order to avoid any effects of particle scattering. Since the PDH loading capacity reflects only the amount of PDH encapsulated in the nanoparticles while there is no indication of enzyme activity from that value, we introduced the term active loading capacity to describe the activity of PDH per mass of particles.

under reducing conditions where free PDH served as the positive control and blank nanoparticles served as negative control. Briefly, PDHencapsulated PLGA nanoparticles, free PDH, or blank PLGA nanoparticles were boiled with β-mercaptoethanol for 10 min at 100 °C. Samples were loaded into a 15% polyacrylamide gel and proteins were separated by SDS-PAGE at 150 V for 90 min. Proteins were transferred to a PVDF membrane using a Trans-Blot® Turbo™ transfer system (BioRad). Blots were blocked while rotating overnight at 4 °C in a 5% solution of bovine serum albumin (BSA) (Fisher) added to 0.1% Tween20® (Sigma) in tris-buffered saline (BSA-TTBS) pH 7.0 (Trizma®, Sigma). Primary anti-E1-a monoclonal antibody (Abcam) produced in rabbit was diluted at 1:5000 in fresh blocking buffer (BSA-TTBS) and incubated for two hours at 24 °C while rotating. Blots were washed three times at room temperature for 10 min with TTBS while rotating. Blots were subsequently incubated in secondary anti-rabbit horseradish peroxidase (HRP)-linked antibody (Cell Signaling) diluted at 1:2000 in fresh blocking buffer for 2 h at room temperature while rotating. Blots were again washed three times for 10 min with TTBS at room temperature while shaking. Chemiluminescent development was achieved by exposing PVDF blots to Clarity™ Western ECL substrate (Bio-Rad) for five min while shaking in the dark. The membrane was imaged digitally with the ChemiDoc MP System (Bio-rad). The result was quantified by densitometry using Image Lab software (Bio-rad). Encapsulation efficiency and loading capacity of PDH were calculated as follows:

Active loading capacity =

PDH activity detected (mU) amount of nanoparticles collected (mg) × 100%

2.5. Storage stability To determine whether immobilization of PDH in PLGA NPs improved PDH stability, aliquots of free PDH, PDH-PLGA 50:50 NPs, and PDH-PLGA 75:25 NPs in 50 mM MOPS (pH 7.4) were incubated at 4 °C, 24 °C, 30 °C, and 37 °C for 48, 96, and 144 h. More than six separate batches were synthesized with PLGA 50:50 and PLGA 75:25 separately. All samples were agitated constantly to prevent aggregation. Enzymatic activity of free PDH and PDH-PLGA NPs was measured prior to incubation and used as initial activity. Following incubation, PDH activity of free PDH and immobilized PDH was measured using the activity assay described above. The measured PDH activity was then compared to the expected activity based on the encapsulation efficiency to determine a percentage of PDH activity retained throughout the PDHPLGA NPs production process.

Encapsulation efficiency amount of PDH encapsulated in nanoparticles (mg) = × 100% amount of PDH used (mg) PDH loading capacity amount of PDH encapsulated in nanoparticles (mg) = × 100% amount of nanoparticles collected (mg)

2.4. Enzymatic activity determined with biochemical assay

2.6. Biofilm culture

PDH activity was calculated by the conversion of β-NAD+ to βNADH over 60 min (Δ340 nm/60 min) using an extinction coefficient of 6.22 mol−1 x cm−1. Only a fraction of PDH-PLGA NPs produced in each batch was needed for activity assays, yet activity assay measurements were used to calculate the total PDH activity (mU/mL) for the entire

To determine the biofilm architecture prior to and post-addition of PDH, biofilms formed by P. aeruginosa PAO1 or PAO1 constitutively expressing a chromosomally-integrated green fluorescent protein (PAO1-gfp) were grown for 4 days in 24-well plates in 5-fold diluted LB medium under shaking conditions (220 rpm), with the growth medium 3

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

which are estimated to affect approximately 6.5 million people in the United States [25]. Biofilms were observed in chronic wound specimens but not in acute wound specimens. The presence of P. aeruginosa has been shown to slow down and even prevent the wound healing process [26]. Reversal of biofilm bacteria to an antibiotic-susceptible state, which can be achieved through the depletion of pyruvate with PDH, makes it possible to circumvent the recalcitrant biofilm state and provide more traditional routes of infection treatment. Goodwine et al. reported that co-administration of 100 mU or 200 mU PDH and tobramycin on porcine burn wounds significantly increased the efficacy of tobramycin. An average reduction in the P. aeruginosa biofilm CFU/ wound of 3.5- and 4-log was observed with 100 and 200 mU PDH, respectively [12]. In this study, PDH was immobilized into PLGA NPs made of either PLGA 50:50 (Mw 38,000-54,000) or PLGA 75:25 (Mw 66,000-100,000). Since the organic solvent used in this study, acetone, is miscible in water and should cause a direct precipitation of particles in the first step of this emulsification, we hypothesize that PDH facilitated the emulsification. Proteins such as BSA have been used as emulsifiers in PLGA nanoparticle synthesis, so it is very likely that PDH facilitated the emulsification. Proteins have been reported to disperse at the surfactant film layer [27]. The amount of PDH that can be added into the reaction is limited by the volume of the inner aqueous phase in the W/O/W double emulsion method. The utilization of protein concentrators prior to the emulsion step increased the usage of PDH 10-fold, expanding the test range of PDH usage. In order to find the optimum formulation of nanoparticles with high encapsulation efficiency and retained enzyme activity, various amounts of PDH, starting from 2320 mU and increasing to 6960 mU with an increment of 1160 mU, were used during the synthesis process in the preliminary study. The physiochemical properties of PLGA 50:50 NPs were not affected drastically by the addition of increasing amounts of enzyme (Table 1). DLS measurements suggested that all five batches of PDH-PLGA 50:50 NPs fell in the 200–300 nm range. All batches exhibited a negative zeta potential, similar to blank PLGA 50:50 NPs. Conversely, the addition of various amounts of enzyme had an impact on the yield, initial PDH activity, and encapsulation efficiency of PDH-PLGA 50:50 NPs. The yield of PDH-PLGA 50:50 NPs decreased as the amount of PDH increased. Both encapsulation efficiency and initial activity of PDH reached the highest value when 4640 mU of PDH was added in the reaction and decreased as more PDH was added during synthesis. As shown in Table 2, PDH-PLGA 75:50 NPs, on the other hand, did not show this drastic variation in terms of yield. The encapsulation efficiency and initial activity remained at a relatively similar level among H1 to H4 batches. When 6960 mU of PDH was added, the encapsulation efficiency and initial activity increased drastically. It was initially expected that with more enzyme added into the

Table 1 Optimization of the PDH-PLGA 50:50 NPs. L denotes the use of lower molecular weight polymer (Mw 38,000-54,000) while numbers 1 through 5 denote the increasing concentration of PDH used to formulate each batch. Batch ID

L1

L2

L3

L4

L5

PDH (mU) Size (nm) PDI Zeta potential (mV) Yield (mg) Encapsulation efficiency PDH loading capacity (mg PDH: mg nanoparticles) Post encapsulation PDH activity (mU/mL enzyme) Active Loading capacity (mU Activity PDH: mg nanoparticles)

2320 285.7 0.103 −17.9 75.8 14.4% 0.008

3488 288.7 0.078 −9.7 72.3 16.4% 0.014

4640 271 0.018 −15.8 64.8 19.7% 0.025

5814 271.7 0.025 −17.0 41.4 10.4% 0.025

6960 270.0 0.058 −19.7 18.6 9.0% 0.059

6.391

13.087

31.672

17.355

17.307

0.169

0.362

0.978

0.838

1.861

being exchanged every 12 h. As inoculum, bacteria grown to stationary phase and diluted to an optical density of OD600 = 0.1 were used. Confocal laser scanning microscopy (CLSM) images were acquired using a Leica TCS SP5 confocal microscope (Leica Microsystems, Wetzlar, Germany). In studies tracking rhodamine-loaded nanoparticles, P. aeruginosa PAO1-gfp were used to allow visibility of microcolonies. In studies examining biofilm dispersion, PAO1 biofilms were stained prior to microscopy using the LIVE/DEAD™ BacLight Bacterial Viability Kit (Life Technologies). PAO1-gfp and LIVE/DEAD™ staining were not used concurrently. 2.7. In vitro biofilm treatment In order to visualize the interaction of nanoparticles with biofilms, P. aeruginosa PAO1-gfp biofilms were incubated with 125 μL of rhodamine 6 G-loaded PLGA (Rho-PLGA) NPs in the presence of PDH-associated cofactors and incubated for 16 h in the absence of light. Confocal laser scanning microscopy (CLSM) was utilized to image the nanoparticle-exposed biofilms. Separately, blank PLGA NPs and PDH-PLGA NPs at the same volume were used to assess the potential of PDHcontaining nanoparticles in the treatment of biofilms. P. aeruginosa biofilms grown for 4 days to the maturation-2 stage, were exposed to nanoparticles for 16 h. Maturation-2 stage is the penultimate stage in the biofilm developmental cycle. P. aeruginosa PAO1 biofilms were stained with LIVE/DEAD™ BacLight™ viability stain kit (SYTO 9 ex/em 480/500 nm, propidium iodide ex/em 490/635 nm) prior to microscopy. Stained samples were then examined under CLSM. In addition, P. aeruginosa microcolonies in CLSM images were counted and scored as dispersed or not dispersed based on the presence of a void within the microcolony for each condition. The diameters of all dispersed and not dispersed microcolonies were measured and reported using the Leica TCS SP5 confocal software.

Table 2 Optimization of the PDH-PLGA 75:25 NPs. H denotes the use of higher molecular weight polymer (Mw 66,000-100,000) while numbers 1 through 5 denote the increasing concentration of PDH used to formulate each batch.

2.8. Statistical analysis Results are expressed as mean ± standard error. The data among groups were compared using a 1-way ANOVA. Means were considered statistically significant when p-value ≤ 0.05. The effect of independent variables, namely polymer type and amount of PDH, on outcomes such as PDH activity, PDH encapsulation efficiency, and nanoparticle yield were analyzed using Minitab®18.1 (Minitab). 3. Results and discussion P. aeruginosa is among one of the leading causes of biofilm infection. The most pronounced cases are in burn wounds and chronic infections, 4

Batch ID

H1

H2

H3

H4

H5

PDH used (mU) NP Size (nm) PDI Zeta Potential (mV) Yield (mg) Encapsulation Efficiency PDH loading capacity (mg PDH: mg nanoparticles) Post encapsulation PDH activity (mU/ml enzyme) Active loading capacity (mU Activity PDH: mg nanoparticles

2320 608.8 0.348 −13.9 35.1 9.41% 0.011

3488 376.2 0.11 −9.94 36.3 8.65% 0.015

4640 487.6 0.22 −12.4 28.2 9.89% 0.029

5814 327.1 0.132 −8.78 37.5 7.47% 0.020

6960 491.3 0.288 −14.2 33.3 22.2% 0.081

7.669

1.921

9.180

8.159

30.257

0.218

0.053

0.326

0.218

0.909

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

subunits [20]. The molecular weight of PDH obtained from Escherichia coli K12 was reported to be 3.75 × 103 kDa [29]. Depending on the source of PDH, the diameter of PDH ranges from 25 nm to 45 nm [20–22], while PLGA NPs are typically 200–300 nm in size. It is not surprising that PDH behaved in a rather different way than small molecules when encapsulated. In contrast to PDH’s enormous size, the molecular weight of pyruvate is merely 87 Da. It is considerably easier for pyruvate and associated cofactors to diffuse into PLGA NPs and react with PDH than PDH diffusing out of nanoparticles. The activity of PDH-PLGA 50:50 NPs and PDH-PLGA 75:25 NPs was compared to free PDH that was stored in matched conditions in order to find out whether the immobilization of PDH in PLGA NPs rendered PDH more stable. Five or more individual batches of each type of nanoparticles were synthesized and combined together to produce enough particles for the storage biochemical assay. Fig. 3a shows that both increasing storage time and elevated temperature result in free PDH activity reduction. PDH-PLGA 50:50 NPs, in comparison to free PDH, showed far more reduced enzymatic activity after storage at all conditions for just 48 h (Fig. 3b). However, unlike free PDH, the activity of which decreased as the storage time elongated, the activity of PDHPLGA 50:50 NPs gradually recovered over time in comparison with the starting activity. Moreover, after 144 h, PDH-PLGA 50:50 NPs exhibited more retained activity than free PDH at all temperatures. To be noted, the activity of free PDH almost completely diminished at higher temperature, such as 30 °C and 37 °C, while PDH-PLGA 50:50 NPs had 28.2% and 53.7% retrained activity, respectively. When stored at 4 °C, PDH-PLGA 75:25 NPs and free PDH shared the same trend.: the activity decreased over time. At the end of the sixth day, PDH-PLGA 75:25 NPs retained 63.5% of their initial activity. However, the activity of PDH-PLGA 75:25 NPs plummeted rapidly when the temperature was raised to 24 °C. While free PDH lost just 17% of its activity after 48 h, PDH-PLGA 75:25 NPs lost about 70% of their activity. Unlike PDH-encapsulated PLGA 50:50 nanoparticles, enzymatic activity in PDH-encapsulated PDH-PLGA 75:25 NPs did not come back after longer storage. Only 3.7% of activity was detected after 144 h. When temperature was elevated to 30 °C, free PDH exhibited 50% of its initial activity while PDH-PLGA 75:25 NPs maintained just 4.5% of its initial activity. After being stored at 37 °C for 6 days, PDH activity almost completely diminished from PDH-PLGA 75:25 NPs. PDH storage stability tests suggest free PDH was relatively stable when stored at room temperature or refrigerated for several days, but the enzyme was unstable at body temperature. PDH-PLGA 50:50 NPs exhibited relatively low initial activity in the activity test, possibly due to substrate inhibition, which is commonly observed in enzymes immobilized through entrapment [13]. PLGA degradation likely gradually increased the porosity of the polymer matrix, facilitating the exchange of products and substrates with bulk medium. With more access to cofactors and less concentrated products present, PDH was able to react with the substrate more readily, apparent by the recovering activity over time. These results indicate that the immobilization of PDH into PLGA 50:50 NPs did improve the stability of PDH at clinically relevant temperatures. PDH-PLGA 75:25 NPs on the other hand, started with a high initial PDH activity and showed a similar trend as free PDH at 4 °C but lost activity quickly at elevated temperatures. It is possible that PDH was immobilized on the exterior of the PLGA 75:25 nanoparticles or PDH formed a protein/polymer complex with PLGA 75:25, which made substrate and cofactors more available to PDH-PLGA 75:25 NPs at

reaction, the loading of PDH would increase until reaching a plateau, indicating the capacity of nanoparticles to encapsulate PDH was saturated. A similar trend was anticipated for PDH activity as well because if more PDH was encapsulated, a higher activity was likely to be detected. However, in the case of PDH-PLGA 50:50 NPs, the activity decreased rapidly once the highest activity was achieved instead of plateauing. One likely explanation for this trend is that the addition of an excessive amount of PDH disrupted the formation of PLGA NPs, which is backed up by the fact that when more PDH was used in the synthesis process, fewer nanoparticles were recovered after freeze drying. Since PDH is a very large protein and it is known that proteins tend to be drawn to the W/O interface in W/O/W double emulsion method [28], it is possible that PDH ruptured the nanodroplets formed during the primary emulsion. When fewer PLGA NPs were formed, the opportunity of PDH being encapsulated into those nanoparticles became lower, leading to low encapsulation efficiency. Due to the longer polymer chains, the formation of PLGA 75:25 NPs was not affected by PDH as significantly. The selection of an optimum formulation was mainly driven by initial activity and encapsulation efficiency of PDH. Taking all aspects into consideration, PLGA 50:50 NPs synthesized with 4640 mU of PDH and PLGA 75:25 NPs synthesized with 6960 mU were selected as the optimal formulations. The average physiochemical properties of three batches each of L3 and H5 PDH-PLGA NPs are summarized in Table 3. DLS results indicated that these PDH-PLGA 50:50 NPs had an average size of 274.3 ± 5.4 nm while PDH-PLGA 75:25 NPs exhibited a larger average size at 395.5 ± 48 nm. PDH-PLGA 50:50 NPs had a relatively narrow size distribution compared to PDH-PLGA 75:25 NPs. Both PDHPLGA 50:50 NPs and PDH-PLGA 75:25 NPs had an average zeta potential around −12 mV. SEM images in Fig. 2 reveal the spherical shape of PDH-PLGA 50:50 NPs and PDH-PLGA 75:25 NPs as well as low levels of aggregation. Additionally, these SEM images confirmed that PDHPLGA 75:25 NPs were larger sized compared to PDH-PLGA 50:50 NPs. When PLGA 50:50 was used, the encapsulation efficiency of PDH reached approximately 19%, while about 17% of PDH was entrapped when PLGA 75:25 was used. To be noted, it is possible that the amount of encapsulated PDH is higher than the what was interpreted from Western blot results as some of the PDH may have been denatured during the nanoparticle synthesis process and was not recognized by primary antibodies during analysis. The activity of immobilized PDH was measured by following the reduction of β-NAD at 340 nm. Both formulations showed repeatability in terms of activity. As shown in Table 4, PDH-PLGA 75:25 NPs exhibited robust activity post-immobilization in nanoparticles. PDH-PLGA 50:50 NPs showed activity but at a lower level. When used to deliver small molecules, PLGA NPs often exhibit a triphasic release profile, which is characterized by an initial burst release, a sustained release due to the degradation of polymer and diffusion of the small molecules, and a second burst release as the nanoparticles finally collapse. Preliminary release studies performed in our work showed that 0.0002% of the encapsulated PDH released from the L3 batch was detected by ELISA after a six-day incubation in buffered saline (pH 7.4) at 37 °C (data not shown), indicating PDH was immobilized within PLGA NPs. However, since no protease inhibitors were added into the supernatant, we cannot completely rule out the possibility that PDH degradation contributed to the low absorbance reading. PDH is an enormous multienzyme complex composed of three

Table 3 Characteristics of PDH-PLGA NPs. n = 3. PLGA type

PDH usage (mU)

NP size (nm)

PDI

Zeta Potential (mV)

Yield (mg)

Encapsulation Efficiency

PLGA 50:50 PLGA 75:25

4640 6960

266.7 ± 9.8 395.5 ± 48

0.055 0.161

−13.9 ± 1 −12.3 ± 1.1

69.4 ± 2.3 32.3 ± 1.0

17.9% ± 1.4% 16.9 ± 3%

5

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

Fig. 2. (a) SEM image of PDH-PLGA 50:50 NPs (b) SEM image of PDH-PLGA 75:25 NPs. Scale bar = 200 nm.

PLGA NPs were counted as dispersed indicating that the biofilm developmental cycle is a dynamic process and natural biofilm dispersion events are always occurring. These findings suggest that firstly, PLGA 50:50 NPs are capable of depleting pyruvate in the culture environment similar to free PDH, which results in a biofilm dispersion response. Secondly, the dispersion of biofilms is indeed triggered by PDH through the action of pyruvate depletion instead of PLGA NPs themselves. The interaction between PLGA 50:50 NPs and P. aeruginosa biofilms was revealed by tracking Rho-PLGA NPs with CLSM. Rho-PLGA NPs were noted at the close periphery of microcolonies suggesting PLGA NPs to likely interact with the biofilm matrix but not to penetrate the microcolony structure (Fig. 4a). This finding is consistent with a previous study where the size limitation for nanoparticles to diffuse into P. aeruginosa biofilms was about 200 nm [36]. It is a generally accepted model that P. aeruginosa biofilm bacteria produce pyruvate at the periphery and interstitial region inside a biofilm. Since PDH is immobilized in PLGA NPs, being in the proximity of biofilms is highly advantageous as nanoparticles bring PDH closer to its substrate and create a high local concentration of PDH. This may explain why PDH-PLGA NPs were not as active as free PDH in activity tests but induced biofilm dispersion successfully. The localization of PLGA NPs at the periphery of biofilms can be attributed to the negative charge on nanoparticles surface, which facilitate the hydrophobic interaction between PDH-PLGA NPs and biofilm bacteria [37]. We furthermore determined the efficiency of PDH and PDH-PLGA 50:50 NPs in inducing dispersion, by determining the percent of dispersed microcolonies as well as the average diameter for dispersed and non-dispersed biofilms using P. aeruginosa PAO1-gfp biofilms grown under semi-batch culture conditions in five-fold diluted LB. Batch cultures are liquid cultures in which the medium is not changed during the growth of the bacterium, meaning that the nutrient supply is limited. In contrast, in semi-batch cultures, the growth medium is being replaced frequently. We used a 24-well plate system to grow the biofilms that allowed for the removal and replacement of culture media over the course of growth. Biofilms were either untreated or exposed to PDH (10 mU), heat-killed PDH (10 mU), 125 μL PLGA-NP-PDH (NP-PDH), or blank NPs in the presence of cofactors at saturating concentrations for 16 h. Biofilms left untreated were used as controls (LB control).

Table 4 Initial PDH activity and PDH active loading. PLGA type

Initial activity (mU/mL)

Activity loading capacity (mU PDH: mg NPs)

PLGA 50:50 PLGA 75:25

22.06 ± 6.74 26.85 ± 2.32

0.54 ± 0.24 0.83 ± 0.08

the beginning of the activity test. Since PDH-PLGA 50:50 NPs outperformed PDH-PLGA 75:25 NPs in exhibiting extended capability of depleting pyruvate, those nanoparticles were used to treat established P. aeruginosa biofilms in vitro. Maturation-2 stage biofilms were used for in vitro treatment because this developmental stage is required to generate anoxic conditions at the microcolony core. Under these conditions, the cells at that anoxic core rely on pyruvate produced and secreted elsewhere in the biofilm to utilize by fermentation to sustain survival via NADH/NAD + cycling (redox balancing). [30–32]. We first investigated the interaction between PLGA 50:50 NPs and P. aeruginosa PAO1 biofilms, by treating biofilms with Rho-PLGA NPs. Following incubation, visual observation of the biofilms and Rho-PLGA NPs via CLSM demonstrated that the PLGA NPs tightly adhered to the periphery of biofilms. No rhodamine-loaded nanoparticles were observed in the interior of the microcolony structure (Fig. 4a). This result is interesting since it is known that the biofilm surface and PLGA nanoparticles are both negatively charged and a repulsion between them would be expected. The existence of steric interaction between negatively charged nanoparticles and biofilm bacteria has been discussed previously [33]. A hollow center can be seen in the biofilms treated with free PDH or PDH-PLGA 50:50 NPs (Fig. 4c, d), while this was not observed in biofilms treated with blank PLGA NPs (Fig. 4e). Hollowing or void formation has previously been linked with biofilm dispersion, a process in which sessile, surface-attached organisms liberate themselves from the biofilm to return to the planktonic state [34,35]. Biofilms treated with free PDH and PDH-PLGA 50:50 NPs had significantly higher occurrence of central hollowing than LB control and blank PLGA NPs groups. It is interesting to note that about 10% of biofilms treated with LB and blank

Fig. 3. PDH storage stability test (a) Free PDH (b) PDH-PLGA 50:50 NPs (c) PDH-PLGA 75:25 NPs. 6

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

Fig. 4. (a) CLSM image of mature P. aeruginosa PAO1 consitutively expressing gfp were exposed to 125 μL Rho-PLGA 50:50 NPs for 16 h. Separate CSLM channels are shown for biofilm gfp (biomass), rhodamine (nanoparticles, ex/em = 525 nm/555 nm), and brightfield. Merged, all channels combined. CLSM images of mature P. aeruginosa PAO1 (b) left untreated, or exposed to (c) 125 μL PDH-PLGA 50:50 NPs, (d) 10 mU free PDH, or (e) blank NPs (PLGA only). White size bar, 100 μm.

As shown in Fig. 5a, approximately 40% of biofilm microcolonies treated with PDH-PLGA 50:50 NPs and 66% of biofilm microcolonies treated with free PDH were found to have dispersed, apparent by microcolonies demonstrating void formation. Conversely, while the results presented show no statistical difference, only 10% of untreated biofilms and 8% of the ones treated with blank PLGA NPs shared the same characteristic. In terms of microcolony diameter, for all groups, dispersed microcolonies had a larger diameter than non-dispersed microcolonies in untreated biofilms (Fig. 5b). Additionally, treated biofilms exhibited even larger diameters than the control group. It has been proposed that that pyruvate depletion-induced dispersion of P. aeruginosa biofilms under semi-batch culture conditions requires a threshold microcolony size or diameter allowing the development of an anoxic core such that biofilm cells within that anoxic core are reliant upon pyruvate and pyruvate fermentation for survival. Furthermore, sagging of PDH-dispersed biofilms due to loss of structural support via partial or full removal of the matrix during dispersion has been reported [12]. For biofilms treated with blank NPs, it is possible that the increased diameter of dispersed microcolonies is a result of abrasion from the nanoparticles spreading out the remaining microcolony structures. The effect of individual formulation synthesis factors and their combined effects were analyzed by generating main effect plots and interaction plots, respectively. The responses studied here were PDH initial activity, PDH active loading capacity, and PDH encapsulation efficiency. According to Fig. 6a, a high initial PDH activity was seemingly associated with PLGA 50:50 and higher theoretical PDH loading amount. A typical “cross-over” interaction is shown in Fig. 6d. In general, one cannot isolate the effect of one factor from the other when such interaction occurs. As such, when PLGA 50:50 was used, the initial PDH activity is dependent on both factors. In the case of PDH active loading (Fig. 6b), a combination of PLGA 50:50 and more PDH added into the reaction is expected to reach a higher PDH active loading capacity, while PLGA 75:25 and low amount of PDH may lead to low active loading. When two lines in an interaction plot are neither parallel to nor crossing each other, it is called an “ordinal” interaction. Fig. 6e

Fig. 5. (a) Percent dispersion of microcolonies of P. aeruginosa PAO1 biofilms exposed to heat killed PDH, PDH, nanoparticles containing PDH, and empty nanoparticles compared to control. (b) Average microcolony diameters of dispersed (black) and not dispersed (gray) microcolonies of P. aeruginosa PAO1 biofilms. Error bars are standard deviation. (HK = heat killed) N = 5. * denotes statistical significance p < 0.05.

7

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

Fig. 6. (a) Main effect plot showing the effect on initial PDH activity. (b) Main effect plot showing the effect on PDH loading capacity. (c) Main effect plot showing the effect on PDH encapsulation efficiency. (d) Interaction plot of the effects on PDH activity. (e) Interaction plot of the effects on PDH active loading capacity. (f) Interaction plot of the effects on PDH encapsulation efficiency.

4. Conclusion

indicates the amount of PDH had a predominant effect on the loading capacity of PDH. Fig. 6c shows that PDH encapsulation efficiency increases as the PLGA type is changed from PLGA 50:50 to PLGA 75:25. Increasing the amount of PDH used in the synthesis leads to higher encapsulation efficiency as well. In order to achieve a higher encapsulation efficiency, more PDH and PLGA 75:25 needed to be used. Fig. 6f shows two crossing lines, meaning it is again a “cross-over” interaction between the type of PLGA and amount of PDH used in the reaction when the dependent variable is encapsulation efficiency. These data show that it is critical to balance the amount of PDH added into the nanoparticle synthesis and the type of PLGA in order to preserve the activity of PDH for an extended period.

In this work, PDH was successfully immobilized into PLGA NPs in order to improve the stability of PDH under different storage conditions. The most effective formulation was selected from ten formulations and remained active after being incubated at clinically relevant temperature for six days. In in vitro biofilm treatments, PDH-PLGA NPs actively dispersed mature P. aeruginosa biofilms through the action of depleting pyruvate. In conjunction with recently published data showing that PDH-facilitated biofilm dispersion renders bacteria susceptible to conventional antibiotics [12], these results demonstrate the potential of this nanoparticle technology for treatment of biofilm infections in wounds and cystic fibrosis. 8

Colloids and Surfaces B: Biointerfaces 184 (2019) 110512

C. Han, et al.

Acknowledgements

[18] H.K. Makadia, S.J. Siegel, Poly lactic-co-Glycolic acid (PLGA) as biodegradable controlled drug delivery carrier, Polymers 3 (September (3)) (2011) 1377–1397. [19] R. Dinarvand, N. Sepehri, S. Manoochehri, H. Rouhani, F. Atyabi, Polylactide-coglycolide nanoparticles for controlled delivery of anticancer agents, Int. J. Nanomed. 6 (2011) 877–895. [20] G. Sutendra, et al., A nuclear pyruvate dehydrogenase complex is important for the generation of Acetyl-CoA and histone acetylation, Cell 158 (July (1)) (2014) 84–97. [21] Z.H. Zhou, D.B. McCarthy, C.M. O’Connor, L.J. Reed, J.K. Stoops, The remarkable structural and functional organization of the eukaryotic pyruvate dehydrogenase complexes, Proc. Natl. Acad. Sci. U. S. A. 98 (December (26)) (2001) 14802–14807. [22] B. Sumegi, Z. Liposits, L. Inman, W.K. Paull, P.A. Srere, Electron microscopic study on the size of pyruvate dehydrogenase complex in situ, Eur. J. Biochem. 169 (November (1)) (1987) 223–230. [23] D. Blanco, M.J. Alonso, Protein encapsulation and release from poly(lactide-coglycolide) microspheres: effect of the protein and polymer properties and of the coencapsulation of surfactants, Eur. J. Pharm. Biopharm. 45 (May (3)) (1998) 285–294. [24] T. Feczkó, J. Tóth, Gy. Dósa, J. Gyenis, Optimization of protein encapsulation in PLGA nanoparticles, Chem. Eng. Process. Process Intensif. 50 (August (8)) (2011) 757–765. [25] C.K. Sen, et al., Human skin wounds: a major and snowballing threat to public health and the economy, Wound Repair Regen. Off. Publ. Wound Heal. Soc. Eur. Tissue Repair Soc. 17 (6) (2009) 763–771. [26] S.L. Percival, S.M. McCarty, B. Lipsky, Biofilms and wounds: an overview of the evidence, Adv. Wound Care (New Rochelle) 4 (July (7)) (2015) 373–381. [27] A. Rafati, et al., Chemical and spatial analysis of protein loaded PLGA microspheres for drug delivery applications, J. Control. Release 162 (September (2)) (2012) 321–329. [28] F. Kang, G. Jiang, A. Hinderliter, P.P. DeLuca, J. Singh, Lysozyme stability in primary emulsion for PLGA microsphere preparation: effect of recovery methods and stabilizing excipients, Pharm. Res. 19 (May (5)) (2002) 629–633. [29] O. Vogel, B. Hoehn, U. Henning, Molecular structure of the pyruvate dehydrogenase complex from Escherichia coli K-12, Proc. Natl. Acad. Sci. U. S. A. 69 (June (6)) (1972) 1615–1619. [30] M. Schobert, D. Jahn, Anaerobic physiology of Pseudomonas aeruginosa in the cystic fibrosis lung, Int. J. Med. Microbiol. 300 (December (8)) (2010) 549–556. [31] A. Price-Whelan, L.E.P. Dietrich, D.K. Newman, Pyocyanin alters redox homeostasis and carbon flux through central metabolic pathways in Pseudomonas aeruginosa PA14, J. Bacteriol. 189 (September (17)) (2007) 6372–6381. [32] M. Eschbach, K. Schreiber, K. Trunk, J. Buer, D. Jahn, M. Schobert, Long-term anaerobic survival of the opportunistic pathogen Pseudomonas aeruginosa via pyruvate fermentation, J. Bacteriol. 186 (July (14)) (2004) 4596–4604. [33] K. Ikuma, A.W. Decho, B.L.T. Lau, When nanoparticles meet biofilms—interactions guiding the environmental fate and accumulation of nanoparticles, Front. Microbiol. 6 (June) (2015). [34] K. Sauer, M.C. Cullen, A.H. Rickard, L. Zeef, D.G. Davies, P. Gilbert, Characterization of nutrient-induced dispersion in Pseudomonas aeruginosa PAO1 biofilm, J. Bacteriol. 186 (November (21)) (2004) 7312–7326. [35] K. Sauer, A.K. Camper, G.D. Ehrlich, J.W. Costerton, D.G. Davies, Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm, J. Bacteriol. 184 (February (4)) (2002) 1140–1154. [36] K. Forier, et al., Probing the size limit for nanomedicine penetration into Burkholderia multivorans and Pseudomonas aeruginosa biofilms, J. Control. Release 195 (December) (2014) 21–28. [37] N. Nafee, et al., Antibiotic-free nanotherapeutics: ultra-small, mucus-penetrating solid lipid nanoparticles enhance the pulmonary delivery and anti-virulence efficacy of novel quorum sensing inhibitors, J. Control. Release Off. J. Control. Release Soc. 192 (October) (2014) 131–140.

Funding for this work was provided by the NIH (1R56AI12781501A1), Binghamton University Transdisciplinary Areas of Excellence and Interdisciplinary Collaboration Grants awards, and the Binghamton University Analytics and Diagnostics Laboratory. References [1] J.W. Costerton, P.S. Stewart, E.P. Greenberg, Bacterial biofilms: a common cause of persistent infections, Science 284 (May (5418)) (1999) 1318–1322. [2] K. Lewis, Riddle of biofilm resistance, Antimicrob. Agents Chemother. 45 (April (4)) (2001) 999–1007. [3] J.C. Nickel, I. Ruseska, J.B. Wright, J.W. Costerton, Tobramycin resistance of Pseudomonas aeruginosa cells growing as a biofilm on urinary catheter material, Antimicrob. Agents Chemother. 27 (April (4)) (1985) 619–624. [4] T.F. Mah, G.A. O’Toole, Mechanisms of biofilm resistance to antimicrobial agents, Trends Microbiol. 9 (January (1)) (2001) 34–39. [5] C. Han, N. Romero, S. Fischer, J. Dookran, A. Berger, A.L. Doiron, Recent developments in the use of nanoparticles for treatment of biofilms, Nanotechnol. Rev. (2016). [6] M. Schobert, D. Jahn, Anaerobic physiology of Pseudomonas aeruginosa in the cystic fibrosis lung, Int. J. Med. Microbiol. IJMM 300 (December (8)) (2010) 549–556. [7] M. Eschbach, K. Schreiber, K. Trunk, J. Buer, D. Jahn, M. Schobert, Long-term anaerobic survival of the opportunistic pathogen Pseudomonas aeruginosa via pyruvate fermentation, J. Bacteriol. 186 (July (14)) (2004) 4596–4604. [8] K.J. Davies, D. Lloyd, L. Boddy, The effect of oxygen on denitrification in Paracoccus denitrificans and Pseudomonas aeruginosa, J. Gen. Microbiol. 135 (September (9)) (1989) 2445–2451. [9] W.G. Zumft, Cell biology and molecular basis of denitrification, Microbiol. Mol. Biol. Rev. MMBR 61 (December (4)) (1997) 533–616. [10] C. Vander Wauven, A. Piérard, M. Kley-Raymann, D. Haas, Pseudomonas aeruginosa mutants affected in anaerobic growth on arginine: evidence for a four-gene cluster encoding the arginine deiminase pathway, J. Bacteriol. 160 (December (3)) (1984) 928–934. [11] O.E. Petrova, J.R. Schurr, M.J. Schurr, K. Sauer, Microcolony formation by the opportunistic pathogen Pseudomonas aeruginosa requires pyruvate and pyruvate fermentation, Mol. Microbiol. 86 (November (4)) (2012) 819–835. [12] J. Goodwine, et al., Pyruvate-depleting conditions induce biofilm dispersion and enhance the efficacy of antibiotics in killing biofilms in vitro and in vivo, Sci. Rep. (2019). [13] N.R. Mohamad, N.H.C. Marzuki, N.A. Buang, F. Huyop, R.A. Wahab, An overview of technologies for immobilization of enzymes and surface analysis techniques for immobilized enzymes, Biotechnol. Biotechnol. Equip. 29 (March (2)) (2015) 205–220. [14] S. Datta, L.R. Christena, Y.R.S. Rajaram, Enzyme immobilization: an overview on techniques and support materials, 3 Biotech 3 (February (1)) (2013) 1–9. [15] D.N. Tran, K.J. Balkus, Perspective of recent progress in immobilization of enzymes, ACS Catal. 1 (August (8)) (2011) 956–968. [16] R.A. Jain, The manufacturing techniques of various drug loaded biodegradable poly (lactide-co-glycolide) (PLGA) devices, Biomaterials 21 (December (23)) (2000) 2475–2490. [17] R.C. Mundargi, V.R. Babu, V. Rangaswamy, P. Patel, T.M. Aminabhavi, Nano/micro technologies for delivering macromolecular therapeutics using poly(d,l-lactide-coglycolide) and its derivatives, J. Control. Release 125 (February (3)) (2008) 193–209.

9