Investigation on hemolytic effect of poly(lactic co-glycolic) acid nanoparticles synthesized using continuous flow and batch processes

1 Introduction

Nanoparticles play an increasingly important role in drug delivery [1]. Polymeric nanoparticles in particular have garnered extraordinary interest as drug delivery systems because of their controllable physical-chemical characteristics and their ability to entrap, dissolve, or target drugs [2]. Hydrophilic drugs, hydrophobic drugs, proteins, vaccines, and biological macromolecules can be delivered with biocompatible and biodegradable polymeric nanoparticles [3].

Poly(lactic-co-glycolic acid) (PLGA) is a frequently used polymeric compound composed of glycolic acid and lactic acid, approved by the US Food and Drug Administration (FDA) for drug delivery and therapeutic encapsulation [4]. PLGA is known for its biocompatibility and biodegradability [5], as it breaks down into monomer acids metabolized through natural pathways. The size and the size distribution of the PLGA nanoparticles can be determined using various synthesis methods [6]. Batch methods such as emulsion evaporation, emulsion diffusion, and nanoprecipitation are common techniques used to synthesize nanoparticles from the preformed PLGA polymer.

Emulsion evaporation is one of the most commonly used methods to synthesize polymeric nanoparticles. In this method, the emulsification of a nonpolar organic phase in an aqueous phase containing the surfactant is followed by the evaporation of the organic solvent. Usually, the polymer and active components are dissolved in a suitable solvent to form the organic phase (e.g. ethyl acetate and dichloromethane), which is mixed with water under high pressure (e.g. microfluidization) or sonication to form an emulsion. Finally, the organic solvent is removed by evaporation under vacuum.

Nanoprecipitation is another common batch technique often used to synthesize nanoparticles. Typically, the polymer is dissolved in a polar, water-miscible solvent such as acetone, acetonitrile, ethanol, or methanol. The organic solution is then poured into an aqueous solution containing surfactants. In this process, polymeric nanoparticles are quickly formed through rapid solvent diffusion to the aqueous phase with further polymer precipitation. The method requires the extraction of the organic solvent by evaporation as well, but it uses solvents that are less toxic [6].

The nanoprecipitation technique has been converged with different continuous flow lab-on-a-chip devices for particle synthesis. Microfluidic platforms offer reaction environments that are precisely controlled at the microscale level and can be reproduced easily and also allow systematic tuning of the biophysical properties of nanoparticles [7]. Because of the increased speed of analysis, parallel screening, well-defined and repeatable conditions, and lower cost of research, microfluidic platforms are attractive in the preparation of nanoparticles and in the study of their interaction with biological systems [8]. Karnik et al. [9] demonstrated the use of the continuous flow method in nanoparticle synthesis using microfluidic channels to synthesize PLGA-PEG nanoparticles with diameters in the range of 10–50 nm. A fluidic nanoprecipitation system constructed with an inlet channel feeding the center of a dispersing channel (3/32″) was used to fabricate uniform PLGA particles in the range of 140–500 nm [10]. As stated in the work performed by Wang et al. [11], the size-controlled synthesis of PLGA nanoparticles ranging from 55 to 135 nm diameter has been accomplished using the tubing method. It was reported that nanoparticles produced at a high flow rate were smaller than those produced at a lower flow rate [11]. A parallel 3D hydrodynamic flow focusing device using a multilayer microfluidic system was used to synthesize PLGA-PEG nanoparticles within the range of sizes (13–150 nm) [7].

Alternatively, millifluidic chips provide a scaled-up version of microfluidics and offer additional advantages over microfluidics as they are less expensive, have more area for chemical reactions, and are capable of providing more precise control over fluid flow. Several research groups have used a lab-on-a-chip platform for synthesis of inorganic nanoparticles in a continuous flow configuration [12], [13], [14], [15], but few have focused on polymeric nanoparticles. In the literature, millifluidic reactors have been used to synthesize larger (100 μm to 3 mm) polymeric particles with narrow size distributions and a large variety of shapes, dimensions, and compositions [16]. This research has successfully demonstrated the use of continuous flow technique in a millifluidic platform to synthesize nanoparticles with similar surface properties of that of batch synthesized nanoparticles. The present paper has assessed the hemolytic effect of these nanoparticles on red blood cells (RBCs).

When PLGA nanoparticles are administered to humans through oral, nasal, intramuscular, or intravenous routes, some particles will make their way into the blood. How these nanoparticles associate with and affect RBCs is a health concern because RBCs are among the first cells to interact with foreign materials present in the circulatory system [17]. Only a few studies have been published on the compatibility of PLGA nanoparticles with RBCs in general, and all have studied the interaction and effect on RBCs of PLGA nanoparticles made with batch methods [5], [17], [18], [19].

Overall, the literature published indicates that PLGA nanoparticles at certain concentrations affect erythrocyte hemolysis. Most studies were conducted at low NP concentration (μg/ml levels) and found no toxicity of PLGA nanoparticles on RBCs. The effects of the synthesis process (continuous flow compared with batch processing) on the particle association and compatibility with RBCs have not been reported. The main objective of this study was to assess the concentration and size-dependent interaction of PLGA particles prepared in continuous flow mode via a millifluidic platform versus PLGA particles synthesized by a batch process with RBC. The emulsion evaporation method was used for batch processing, as the emulsification process resulted in smaller sized nanoparticles compared with the continuous flow nanoprecipitation technique. The main hypothesis of this experiment was that the hemolytic effect of PLGA nanoparticles made by two synthesis methodologies, a batch emulsion evaporation method and a continuous flow method using a millifluidic lab-on-a-chip device, is influenced by size and size distribution.

2 Materials and methods

3 Results

3.1 Nanoparticle characterization

3.1.1 DLS characterization

PLGA nanoparticles were synthesized by the conventional bulk nanoprecipitation method and by using the millifluidics chip at different flow rates of aqueous phase and varying aqueous-organic phase ratio and different flow rate ratio (10:1, 20:1, 30:1, and 40:1 v/v), respectively, with PVA as the surfactant. The size, the polydispersity, and the zeta potential of nanoparticles prepared by each method were measured and compared with optimize the nanoparticle synthesis on the millifluidic chip (Figure 1).

Figure 1:

Size of PLGA nanoparticles as a function of aqueous organic flow rate ratio and phase ratio for continuous flow and batch nanoprecipitation method. Data are presented as means±SD, n=3.

The minimum average size of nanoparticles synthesized using the continuous flow method at different flow rates was larger than the nanoparticles prepared with batch nanoprecipitation method, but in the same range of 200–250 nm. The average size of PLGA nanoparticles at lower flow rate (600 μl/min) was 248.3±3.08 nm, whereas the average size at higher flow rate (1000 μl/min) was 196.93±14.30 nm. PLGA nanoparticles synthesized at a higher flow rate (1000 μl/min of aqueous phase) and higher flow rate ratio (40:1 v/v) were close in size to nanoparticles synthesized using the batch processing method at that aqueous-organic phase ratio (Figure 1). Increasing the flow rate of the aqueous phase and flow rate ratio allowed for ~200 nm nanoparticles to be prepared. PLGA nanoparticles synthesized by continuous flow method had a zeta potential of –11.67±0.02 mV. PLGA nanoparticles synthesized with batch nanoprecipitation method had similar zeta potential of –9.71±0.75 mV.

In addition to the PLGA nanoparticles of approximately 200 nm with PDI of 0.199±0.028 synthesized by continuous flow on the millifluidic chip, smaller PLGA nanoparticles were made by batch processing via emulsion evaporation method using ethyl acetate as organic phase (120 nm, PDI 0.17±0.02) and were negatively charged (–20±1.10 mV) (Table 1).

Table 1:

Comparison of size, size distribution, PDI, and zeta potential of nanoparticles prepared using the continuous flow and batch processing methods.

Method	Solvent	Minimum size (nm)	PDI	Zeta potential (mV)
Nanoprecipitation on millifluidic chip (continuous flow)	Acetonitrile	196.93±14.30	0.199±0.028	–11.67±0.20
Nanoprecipitation (batch processing)	Acetonitrile	198.43±0.95	0.108±0.02	–9.71±0.75
Emulsion evaporation (batch processing)	Ethyl acetate	119.3±0.98	0.17±0.02	–20±1.1

3.2 Nanoparticle-RBC interaction

3.2.1 Nanoparticles-RBC association – FC analysis

Nanoparticles synthesized using batch processing with emulsion evaporation (~120 nm) and continuous flow in millifluidics chip (~200 nm) were used to study the nanoparticles-RBC association. FC analysis showed that the interaction of RBCs with PLGA for nanoparticles synthesized by continuous flow was concentration dependent, as indicated by the SSC and intensity of FITC change upon exposure of RBCs to PLGA nanoparticles (Figure 2).

Figure 2:

Representative FITC intensity histograms of the blood samples exposed to fluorescent PLGA nanoparticles at different concentrations for 200 nm nanoparticles, as follows: untreated blood (A) and blood exposed to nanoparticles at 2 mg/ml (B), 3 mg/ml (C), 5 mg/ml (D), and 10 mg/ml (E). M1 represents the population with specific signal over control.

When RBCs were treated with higher concentration of nanoparticles, more nanoparticles attached to the surface of the RBCs (Figure 3). The intensity of light emitted by RBCs shifted to the area with FITC-specific signal; the shift was more pronounced with increasing concentration of nanoparticles. The FITC intensity histogram for blood samples exposed to batch processed nanoparticle (emulsion evaporation synthesized) PLGA nanoparticles was similar (data not shown) to that for nanoparticles synthesized in continuous flow. In both cases, FITC intensity histograms suggested that at higher nanoparticle concentration, more nanoparticles associated with the RBCs.

Figure 3:

FACS analysis of nanoparticles associated with RBCs exposed to increasing concentrations of continuous flow and batch processing synthesized nanoparticles. Untreated blood sample was used as the negative control. *Significant difference between two samples at same concentration (p-value<0.05). Data are presented as means±SD, n=3.

Of 100,000 cells counted in each experiment, only 0.22% gave a fluorescence signal for control (blood without nanoparticles). By contrast, when exposed to nanoparticles synthesized by the continuous flow method at concentrations of 2, 3, 5, and 10 mg/ml, the fraction of RBCs with a fluorescence signal in the FITC-specific section was 13.95%, 17.41%, 23.32%, and 37.21%, respectively. Similarly, the blood cells treated with batch processing synthesized nanoparticles gave a fluorescence signal of 15.23%, 18.27%, 29.92%, and 34.32% for nanoparticle concentrations of 2, 3, 5, and 10 mg/ml. In both cases, the fluorescence signal increased with increasing nanoparticle concentrations.

Statistical analysis showed that the results were significantly different for all nanoparticle concentrations relative to the control (p-value<0.05). When nanoparticle-RBC association was compared for continuous flow and batch processing synthesized nanoparticles at the same concentration, it was observed that for nanoparticle concentration of 2 and 3 mg/ml, there was no significant difference (p-value>0.05) between treatments, whereas at concentrations higher than 3 mg/ml, the difference was significant. Bigger PLGA nanoparticles made by the continuous flow synthesis had, in general, a lower association with RBCs than the smaller batch-made PLGA nanoparticles (Figure 3).

3.2.2 Hemolytic assay

PLGA nanoparticles in contact with diluted human blood for 3 h 15 min induced hemolysis, detected as increased absorbance intensity with increasing concentration of nanoparticles in the PBS/blood mixture (Figure 4). Nanoparticle samples at the highest concentrations (10 mg/ml) caused no significant level of hemolysis than the negative control for both continuous flow and batch processing synthesized nanoparticles.

Figure 4:

In vitro hemolysis of rat blood exposed to different concentrations of PLGA nanoparticles. Untreated blood sample was used as the negative control (–), and Triton X-10% treated blood was used as the positive control (+). Data are presented as means±SD, n=3.

Nontreated blood samples (negative control) showed nearly 16% and 20% hemolysis because of simple handling and sample preparation when the experiment was conducted for continuous flow synthesized nanoparticles and for batch processed nanoparticles, respectively. Taking the negative control as a baseline, nanoparticles made by continuous flow caused 17%–18% hemolysis of RBCs, whereas batch processed nanoparticles caused 20%–22% hemolysis. In hemolysis at the highest nanoparticle concentrations tested, 10 mg/ml was not significantly different from the negative control in both cases (Figure 4). Nanoparticle concentration lower than 10 mg/ml did not induce any significant hemolysis in the diluted blood exposed to the nanoparticles.

Blood samples were collected on different days for the two different experiments and from different rats of the same species. This may have contributed to the difference observed between the baseline and the results at the same nanoparticle concentration for continuous flow and batch processing synthesized nanoparticles. Therefore, the% change hemolysis was compared for the two types of particles studied. On the basis of the% change hemolysis as a function of nanoparticle concentration, it was noted that for all the concentrations, the hemolytic profile was not significantly different for continuous flow and batch processing synthesized nanoparticles except for 3 mg/ml. The hemotoxic effect increased by only 10.85%±1.34% and 13.01%±1.42% at 10 mg/ml nanoparticle concentration, with batch processed samples being associated with a slightly higher hemolysis (Figure 5), most likely because of their smaller size.

Figure 5:

Hemolytic effect for continuous flow and batch processing with increase in nanoparticle concentration. *Significant difference between two samples at same concentration (p-value<0.05). Data are presented as means±SD, n=3.

According to the hemolysis criterion in ASTM E2524-08 standard, an increase in the percent hemolysis by more than 5% indicates damage to RBCs caused by the test materials [23]. Considering the baseline for the hemolysis due to sample handling and preparation, the criteria in ASTM E2524-08 standard were not exceeded at smaller PLGA nanoparticle concentration (2, 3, and 5 mg/ml) for both synthesis methods. PLGA nanoparticles at higher concentrations of 10 mg/ml exceeded the criteria.

4 Discussions

The convergence and integration of continuous flow systems with particle technologies has shown considerable progress lately for the development of microparticles and nanoparticles [9]. Using a continuous flow method, nanoparticles in the range of 200 nm were synthesized, which is a significant improvement over other published millifluidic methodologies, reporting synthesis of larger polymeric particles in the range of 100 μm to 3 mm [16]. Nanoparticles formed at the higher flow rates using continuous flow in the present study were smaller than nanoparticles synthesized at lower flow rates. Similar results were reported by Wang et al. [11], in which nanoparticles produced at a high flow rate were smaller than those produced at a lower flow rate.

Nanoparticles synthesized by the batch emulsion evaporation processing method were ~80 nm smaller than the nanoparticles synthesized by the continuous flow nanoprecipitation method. High-shear stress applied during the emulsification process in batch processing reduced the size of particles formed in the emulsion evaporation method, whereas the particles synthesized by continuous flow were formed instantaneously by rapid solvent diffusion. Nanoparticles associated with RBCs in a concentration-dependent manner in both cases. RBCs treated with a higher concentration of nanoparticles (10 mg/ml) showed 30% or more RBCs tagged with nanoparticles.

On the basis of different studies published, it has been established that although Au and platinum nanoparticles interacted with blood components without much effect, Ag nanoparticles caused cell damage [24], [25], Cu nanoparticles tended to destroy the cell membrane and caused hemolysis of RBCs [26], [27], Fe nanoparticles only influenced the physiology of RBCs [28], Si-based nanoparticles also had adverse effects on the RBCs [29], [30], [31], [32], and TiO₂ nanoparticles [33], [34] and MWCNTs had negative effect on the RBCs as well. Although the effect of inorganic nanoparticles such as those described previously on RBCs was extensively covered in the literature, only a limited number of studies were published on the interaction of polymeric nanoparticles with blood constituents (Table 2).

Polymeric nanoparticles such as polystyrene nanoparticles, polymethacrylate nanoparticles, PEGylated and non-PEGylated copolymer conjugated nanoparticles, and nanoparticles of complex polymeric structures, such as those made of poly(3-hydroxybutyrate)-poly(ethylene glycol)-poly(3-hydroxyl butyrate) (PHB-PEG-PHB) and chitosan-N-trimethylaminoethylmethacrylate chloride-PEG copolymers, have been previously reported to have no hemotoxic and cytotoxic effect on RBCs [36]. Polymethacrylate nanoparticles were found to have minimal cytotoxicity [37]. Similarly, complex polymeric structure PHB-PEG-PHB recorded no signs of cytotoxicity and hemotoxicity [38] (Table 3).

Lower nanoparticle concentrations up to 5 mg/ml did not induce significant hemolysis with either continuous flow or batch processing synthesized PLGA nanoparticles, as reported herein. The hemolytic assay suggested that when the concentration of the nanoparticles was higher than 5 mg/ml, nanoparticles induced hemotoxic effect on the RBCs for both cases. Similar to these results, published literature on the interaction of erythrocytes with nanoparticles made from PLGA conjugated with alendronate (PLGA-ALE nanoparticles) showed that PLGA-ALE nanoparticles were not cytotoxic and were suitable for intravenous administration at the lower concentrations of 560, 56, and 5.6 μg/ml [5]. Biomimetic mucin modified PLGA nanoparticles, studied for the hemocompatibility and cytocompatibility, showed that the RBCs exhibited favorable compatibility to both PLGA and modified nanoparticles; results indicated that interactions of PLGA nanoparticles with erythrocytes did not induce hemolysis under tested conditions [19] (Table 3).

According to the hemolysis criterion in ASTM E2524-08 standard, an increase in the percent hemolysis by more than 5% indicates damage to RBCs caused by the test materials [23]. Considering the baseline for the hemolysis due to sample handling and preparation, the criteria in ASTM E2524-08 standard were not exceeded at smaller PLGA nanoparticle concentration (2, 3, 5 mg/ml) for both synthesis methods.

Although RBC damage was encountered at nanoparticle concentrations greater than 10 mg/ml, according to ASTM E2524-08 standard, it is important to note that this concentration is equivalent to a human dose of 50 g nanoparticle for an average man of 70 kg. This number was calculated taking into account that a typical male rat that weighs 300 g has a blood volume of 50 ml/kg [39]. Such high concentrations are too high to be relevant for biomedical applications, but the data presented are important in understanding the dose effect of PLGA nanoparticles on RBCs.

5 Conclusions

A continuous flow method, based on rapid solvent diffusion for particle synthesis, has been explored in this work for the synthesis of nanoparticles in a millifluidic device. Flow rate and flow rate ratios played a critical role in controlling the size of nanoparticles prepared in a continuous flow synthesis. Nanoparticles with an average size of 200 nm and narrow size distribution were synthesized with the continuous method. Particles of 120 nm in diameter were made by a batch method to provide a comparison. The continuous flow method provided the advantages of flow rate control, ratio of aqueous phase to organic phase, and mixing of the solutions for the synthesis of particles of different sizes compared with batch processing. The ability to control the nanoparticle properties by simply varying the synthesis parameters in a continuous flow method, in particular flow rate and aqueous–organic phase ratio, can open a new pathway for formulation of nanoparticles.

As stated, independent of the administrative routes, nanoparticles will eventually enter the circulatory system. The interaction of these nanoparticles with human blood constituents is of extreme importance. Two different sized nanoparticles synthesized by the continuous flow method and by the batch processing method were compared with each other in terms of nanoparticle association and hemolytic effect on RBCs at different concentrations. Experimental investigation of the interaction of PLGA nanoparticles with RBCs suggested that PLGA nanoparticles associated with the surface of RBCs but had no significant hemotoxic effect at concentration lower than and equal to 10 mg/ml for both continuous flow and batch processing synthesized nanoparticles. This RBC association of nanoparticles at lower concentration without any side effect to the blood cells indicates that nanoparticles can be used as a nanocarrier platform in countless drug delivery applications.

Contrary to the hypothesis, it was observed that the size of nanoparticles did not have a significant difference in the hemolytic activity. However, the dose equivalent to the high concentration of nanoparticles exceeds doses likely used in biomedical applications by a wide margin and is not representative.