Food and Chemical Toxicology 67 (2014) 80–86

Contents lists available at ScienceDirect

Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox

Toxic effects of silver nanoparticles and nanowires on erythrocyte rheology Min Jung Kim a, Sehyun Shin a,b,⇑ a b

School of Mechanical Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-713, Republic of Korea ICT Unit, Anam Medical Center, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-713, Republic of Korea

a r t i c l e

i n f o

Article history: Received 4 August 2013 Accepted 5 February 2014 Available online 15 February 2014 Keywords: Silver nanoparticles Silver nanowires Blood Toxicology Hemorheology Deformability

a b s t r a c t Rapid developments in the food applications of silver nanomaterials (Ag-NMs) have resulted in concerns related to the risk of overexposure of human blood. We investigated the effect of size and aspect ratio of Ag-NMs on rheological characteristics of human erythrocytes, including hemolysis, deformability, aggregation, and morphological changes. Red blood cells (RBCs) were exposed to two different sizes of spherical particles (d  30 nm or 100 nm) or nanowires (d  40 nm, l–2 lm in length) at a range of concentrations and incubation times. The concentrations of Ag-NMs were carefully chosen to avoid any hemorheological alteration due to hemolysis. Rheological properties were measured using microfluidic-laser diffractometry and aggregometry. RBC deformability apparently decreased after treatment with a low concentration of Ag-NPs for a short exposure time. However, RBC aggregation was significantly altered after treatment with a low concentration of either Ag-NWs or large Ag-NPs compared to small Ag-NPs. Additional experiments with Ag ions confirmed that the observed rheological changes were mainly caused by the Ag-NMs rather than the Ag ions. These hemorheological findings provide a better understanding of the interaction between RBCs and Ag-NMs and will help in assessing the risk of nanomaterial toxicity in blood. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Nanomaterials are increasingly used in various food applications such as nanosensors in food packaging to detect pathogens or pesticides and nano-ingredients to improve the flavor and taste of foods and drinks (Powell and Kanarel, 2006; Cockburn et al., 2012). Especially, silver nanomaterials (Ag-NMs) are among the most popular nanomaterials used for food applications because of their strong antimicrobial and disinfectant activity (Asharani et al., 2009). However, expansion in the use of Ag-NM-based technology has resulted in concerns regarding the risk of overexposure in humans (Panyala et al., 2008). In fact, a recent epidemiological study showed a high correlation between exposure to nanoparticles and the incidence of life-threatening cardiovascular events (Buzea et al., 2007). Thus, it is particularly important to study the interaction between blood and Ag-NMs. Recently, a few studies on the hemorheological toxicity of nanomaterials reported interesting results. Attachment of mesoporous silica nanomaterials to RBC membranes was found to be responsible for a significant reduction in cellular deformability ⇑ Corresponding author at: School of Mechanical Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-713, Republic of Korea. Tel.: +82 2 3290 3377. E-mail address: [email protected] (S. Shin). http://dx.doi.org/10.1016/j.fct.2014.02.006 0278-6915/Ó 2014 Elsevier Ltd. All rights reserved.

(i.e., the ability of an erythrocyte to deform, thus permitting flow through the microcirculation) (Zhao et al., 2011). In addition, exposure of lead nanomaterials to blood resulted in not only hemolysis but also a decrease in RBC aggregation (Kim et al., 2013). Ag-NMs have shown both hemolytic and hemagglutinating properties. Cell lysis may be a result of membrane damage as suggested by lipid peroxidation studies (Asharani et al., 2010). Jun et al. (2011) reported that exposure to Ag-NMs in vitro at a concentration of 50–100 lg/mL caused spontaneous increase in platelet reactivity. Smock et al. (2013) reported that the in vivo serum concentration of silver nanoparticles was up to 10 lg/mL. Such hemorheological alterations associated with nanomaterials could eventually cause circulatory diseases in macro- and micro-circulation systems (Shin et al., 2007b). In fact, RBC deformability and aggregation, which are major determinants of blood viscosity, are crucial for normal microcirculation function; subjects with normal hemorheology have proper oxygen delivery and tissue perfusion as well as stable metabolic activity. Slight alterations of RBC deformability and aggregation could thus result in circulatory diseases in macro- and micro-vascular systems (Shin et al., 2007a; Baskurt et al., 2011). We therefore investigated the effects of Ag-NMs on the deformability and aggregation of RBCs using a range of concentrations and incubation times. The effects of Ag-NMs on erythrocyte

M.J. Kim, S. Shin / Food and Chemical Toxicology 67 (2014) 80–86

deformability and aggregation were examined according to varying particle size and shape. RBCs were exposed to the three most common types of Ag-NMs (particles and wires) using different concentrations and exposure times. We used scanning electron microscopy to assess Ag-NM interactions with RBC membranes, and determined the rheological properties of erythrocytes by measuring hemolysis, deformability, and aggregation of RBCs.

2. Experimental 2.1. Silver nanomaterials Ag-NMs were purchased from Nanocomposix, Incorporated (San Diego, CA, USA). The silver nanoparticles (Ag-NPs) had a hydrodynamic diameter of 36.2 nm or 117 nm. Silver nanowires (Ag-NWs) had a diameter of 40 nm and a length of 2 lm. Ag-NMs were used at a concentration of 1 mg/mL with polyvinylpyrrolidone (PVP) surfaces and were extensively purified prior to use. PVP is a polymer that binds strongly to the nanomaterial surface and provides greater stability than citrate or tannic acid-capped nanomaterials (Wang et al., 2005). Characterization of Ag-NMs was performed by the manufacturer using UV–visible spectroscopy, dynamic light scattering (DLS), and transmission electron microscopy (TEM) for quality verification. A summary of the characterization data is provided in the Supplementary information, Fig. S1.

2.2. Preparation of samples On the day of each experiment, whole blood was drawn from the antecubital vein of healthy male donors through a 21-gauge butterfly infusion set into a VacutainerÒ (BD, Franklin Lakes, NJ, USA) containing K2-EDTA as the anticoagulant. All institutional requirements for human subject protocol review and biohazard control were followed (Baskurt et al., 2009). None of the donors had taken any medication during the preceding week. Whole blood was centrifuged at 3000 rpm for 12 min, and the plasma, buffy coat, and top layer of cells were decanted. The remaining packed RBCs were washed two times with phosphate-buffered saline (PBS). After washing, 500 lL of packed RBCs were diluted to 1.5 mL with PBS (25% hematocrit). The diluted RBC suspension (0.3 mL) was then mixed with Ag-NM suspensions in PBS (1.2 mL) at various concentrations (final hematocrit, 5%). RBC suspension in PBS (1.2 mL) without AgNMs was used as a control. The final combined suspensions were gently mixed with a roll-mixer and incubated at room temperature. To study the effect of incubation time on erythrocyte properties, we incubated the samples for 4 h, and carried out measurements every hour. All measurements were performed according to the new guidelines for hemorheological measurements (Baskurt et al., 2009). The concentrations of Ag-NMs were carefully determined with considering the previous studies and the hemolytic behavior of RBCs with Ag-NMs. Jun et al. (2011) showed that exposure to silver nanoparticles in vitro at a concentration of 50– 100 lg/mL caused spontaneous increase in platelet reactivity. Smock et al. (2013) reported that the in vivo serum concentration of silver nanoparticles was up to 10 lg/mL. Thus, initially, the concentrations of Ag-NMs were widely examined with hemolytic behavior of RBCs up to 500 lg/mL. Then, the concentrations of Ag-NMs were limited within the 5% hemolysis, which could not affect apparently the hemorheological behavior of RBCs (Kim and Shin, 2013).

81

2.5. Erythrocyte deformability The deformability of RBCs exposed to Ag-NMs was determined using a microfluidic ektacytometer (Rheoscan-AnD300, RheoMeditech, Seoul, Korea) at 25 °C. After washing, 5 lL of packed RBCs were suspended in a highly viscous PVP (Sigma Aldrich, MO, USA) solution. The test chip (K-02, RheoMeditech, Seoul, Korea) consisted of a micro-channel, a sample reservoir, and a waste sample reservoir. Test suspensions were driven by differential pressure between the inlet and outlet of a disposable micro-channel (H  W  L: 0.2  4  40 mm). During this procedure, a beam (wavelength, 635 nm) from a 1.5-mW laser diode was passed through the diluted RBC suspension. The diffraction patterns of the flowing RBCs, at multiple shear stress levels, were projected on a screen and captured by a charge-coupled device (CCD) video camera. The image data were analyzed by a computerized ellipse-fitting program. Deformability was expressed in terms of the elongation index (EI) of the RBCs, which was defined as (L W)/(L + W), where L and W are the major and minor axes of the ellipse, respectively. Further details of this technique are provided elsewhere (Shin et al., 2007a). 2.6. Erythrocyte aggregation We measured RBC aggregation after treatment with Ag-NMs using transmittedlight, microchip-stirring aggregometry (Rheoscan-AnD 300, RheoMeditech, Seoul, Korea). Washed RBCs were resuspended in PBS-polymer solutions prepared using water-soluble polymer dextran (MW 70 kD, Sigma Aldrich, MO, USA) and hematocrit values of the samples were adjusted to 40%. The microchip (C-01, Rheomeditech, Seoul, Korea) consisted of a flat-cylindrical test chamber, a sample inlet, an air outlet, and a stirrer. Prior to RBC aggregation measurement, aggregates were completely dispersed by a rotating stirrer for 10 s. After stirring, followed by an abrupt halt, the intensity of the light transmitted through the microchip was measured with respect to time and analyzed. RBC aggregation was expressed as an aggregation index (AI), which is defined the ratio of the area below the syllectogram (the plot of transmitted light intensity versus time) to the total area over either a 10- or a 120-s time-period, indicating the normalized degree of accumulated aggregation. Further details of RBC aggregation measurements are provided elsewhere (Shin et al., 2009). 2.7. Scanning electron microscopy RBCs were incubated with different Ag-NMs for 4 h to monitor morphological changes before the onset of hemolysis. PBS containing the Ag-NMs was removed and the RBCs were washed twice in fresh PBS, fixed in 2.5% glutaraldehyde (Sigma Aldrich, MO, USA) for 30 min, and washed three times using PBS. RBCs were dehydrated in increasing concentrations of ethanol (25%, 50%, 70%, 90%, and 100%) for 10 min at each concentration. The RBC pellet was dropped onto a glass coverslip, dried, and sputtered with Pt before viewing under a field-emission scanning electron microscope (FE-SEM, Hitachi S-4300, Japan). 2.8. Statistical analysis Data are reported as means ± standard deviation (SD). Statistical analysis was performed using ANOVA, and the statistical significance was represented with p-values. When p < 0.01, it was considered statistically significant in the present study.

3. Results 2.3. Silver ion toxicity

3.1. Hemolysis assay To rule out the effect of excess product reaction on the toxicity assays, additional controls including Ag+ and AgNO3 (Sigma Aldrich, MO, USA) were used. RBCs were exposed to AgNO3 at low concentrations of 5, 10, 20, and 30 lg/mL. AgNO3 was processed in a similar manner to the Ag-NM suspensions.

2.4. Hemoglobin (Hb) concentration measurement A hemolysis assay was used to evaluate the hemolytic properties of Ag-NMs on human RBCs. RBCs were first isolated by centrifugation and washed two times with PBS, then diluted to 25% hematocrit before incubating with Ag-NM suspensions at various concentrations. Controls were prepared in the same manner as described above for RBCs, adding PBS instead of Ag-NM suspensions. After incubation for 2 h at room temperature, the samples were spun down for the detection of hemoglobin released from hemolyzed RBCs. The suspension was gently mixed with a roll-mixer and incubated at room temperature for 2 h, followed by centrifugation at 5000 rpm for 5 min. The absorbance of the treated RBCs was measured at 540 nm with a DR/4000 U spectrophotometer (Haach, Germany). The measurements were based on the cyanmethemoglobin method using Drabkin’s Reagent (D 5941, Sigma Aldrich, MO, USA), which calculates the cyanmethemoglobin concentration by absorption photometry.

Fig. 1 shows hemoglobin release from RBCs exposed to Ag-NPs (d  30 nm or d  100 nm), or Ag-NWs (d  40 nm, l–2 lm). They were measured with a spectrophotometer using Drabkin’s reagent after 2 h. Hemolytic activity of Ag-NPs (d  30 nm) was first observed at 200 lg/mL with about 14.2% hemolysis detected, while relatively minor hemolysis (