Supporting Information s22

Supporting Information

Static and dynamic microscopy of the chemical stability and aggregation state of silver nanowires in components of murine pulmonary surfactant

Ioannis G. Theodoroua, Danielle Botelhob, Stephan Schwanderc, Junfeng (Jim) Zhangd, Kian Fan Chunge, Teresa D. Tetleye, Milo S. P. Shafferf, Andrew Gowb, Mary P. Ryan*,a and Alexandra E. Porter*,a

aDepartment of Materials and London Centre for Nanotechnology, Imperial College London, Exhibition Road, London SW7 2AZ, United Kingdom

bDepartment of Pharmacology and Toxicology, Rutgers University, Piscataway, New Jersey 08854, United States

cRutgers School of Public Health, Department of Environmental and Occupational Health, Piscataway, New Jersey 08854, United States

dNicholas School of the Environment and Duke Global Health Institute, Duke University, Durham, NC 27708, United States

eNational Heart and Lung Institute, Imperial College London, London SW3 6LY, United Kingdom

fDepartment of Chemistry and London Centre for Nanotechnology, Imperial College London, Exhibition Road, London SW7 2AZ, United Kingdom

Corresponding Authors:

*(A.E.P.):

*(M.P.R.):

Number of pages: 14

Number of figures: 3

Number of tables: 0

Methods

AgNW Synthesis:

Pure AgNWs were synthesized using a modified polyol process, originally developed by Xia et al.1 Ethylene glycol (EG) acts as both solvent and reducing agent, whereas poly(vinyl pyrrolidone) (PVP) is used as the capping agent. The reduction of Ag+ ions by EG leads to the formation of Ag nuclei at the early stages of the reaction. Due to the stronger affinity of PVP for the (100) facets than the (111) facets of the Ag nuclei, this passivation leads to one-dimensional growth and the formation of high aspect-ratio AgNWs. The process was optimized in order to: (i) Eliminate the generation of AgNPs from the synthesis product, as the presence of different populations of particles would confound the correlation between the observed effects and the physicochemical properties of the particles. (ii) Avoid the use of other transition metal impurities, such as Cu 2 or Fe,3 previously used to control the product morphology in the polyol synthesis of Ag nanostructures, because their impact on the physicochemical properties of AgNWs is not well-understood. (iii) Eliminate the need to control the rate of injection of the reactants, making the synthesis less complicated.4

Briefly, Ethylene Glycol (EG, Sigma-Aldrich, anhydrous, 99.8%) (2.5 mL) was placed in a double-neck round-bottom flask connected to a condenser. A stock solution of sodium chloride (NaCl) (0.05 M) was prepared by dissolving NaCl in EG by bath sonication. The appropriate amount of NaCl stock solution was added to the flask so that the concentration of NaCl in the final reaction volume was 60 μM. The flask was heated in an oil bath at 160 oC for 30 minutes to remove trace amounts of water. Meanwhile, argon flow (Ar, BOC, Pure Shield Argon) and magnetic stirring were applied and maintained throughout the synthesis. Silver nitrate (AgNO3, 25 mM, Sigma-Aldrich, >99%) and poly(vinyl pyrrolidone) (PVP, Sigma-Aldrich), with an average molecular weight Mw≈360k, were dissolved in EG (3.5 mL) by magnetic stirring in the dark. The molar ratio of PVP to AgNO3 in the final reaction volume was 1.5 and the concentrations of PVP were calculated in terms of the repeating unit. To remove oxygen, the AgNO3/PVP/EG solution was purged with Ar for 30 minutes. The AgNO3/PVP/EG solution (3.5 mL) was added to the reaction flask drop-wise. After injection, the reaction mixture was refluxed at 160 oC and went through a number of color changes until the mixture became stable at approximately 90 min. The reaction was quenched by cooling the flask in a room-temperature water bath. The reaction mixture was transferred to a centrifuge tube and diluted with acetone 5 times by volume. The AgNWs were collected by centrifugation at 4500 rpm for 10 min. The washing process was repeated by repeated cycles of centrifugation with ethanol and three times with deionized water (DI-H2O), to ensure that residual EG, PVP and unreacted Ag+ ions were removed. To confirm that most Ag+ ions had been removed, their concentration in the synthesis product was measured by Inductively Coupled Plasma–Optical Emission Spectroscopy after ultrafiltration and was found to be under the ICP-OES detection limit (i.e. <0.6 ppb). Finally, the sample was dispersed in deionized water (5 mL) and stored in a sealed glass container at 4 oC in the dark, to avoid exposure to contaminations and reactions induced by the ambient atmosphere (e.g. sulfidation due to gaseous hydrogen sulfide (H2S), carbonyl sulfide (OCS) and carbon disulfide (CS2) in the atmosphere). All the following experiments were performed using a single batch of AgNWs.

Commercial AgNWs are available from at least 9 companies, with dimensions that range between 20-200 nm in diameter and 2-200 μm in length.5 Therefore, the diameter of the AgNWs synthesized for this work is at the average of this range while their length is at the lower-end of the range. Most of previous in vitro and in vivo studies on AgNWs have been performed on commercially available AgNWs.6-8 For example, the recent work by Silva et al. compared the in vivo pulmonary effects post instillation of commercial AgNWs of two different dimensions (“Short” AgNWs, length 2.0 μm, diameter 33.1 nm and “Long” AgNWs, length 20.8 μm, diameter 64.7 nm).6 Both AgNWs produced dose-dependent inflammation indicative of foreign body responses in the lung, but different inflammatory responses depending on AgNW length or higher dissolution rates by the smaller AgNWs. Details about AgNW fabrication methods are rarely disclosed by manufacturers and few characterization data of the products are provided. However, differences in the synthesis procedures, AgNW dimensions or capping agents may lead to differences in dissolution rates, agglomeration kinetics and ultimately in toxicological responses.9 In this work, we chose to produce our AgNWs in house in order to have a full control over their physicochemical properties, which were thoroughly characterized.

Animal LLF extraction:

Male rats were anesthetized by injection of a lethal dose of ketamine/xylazine, and then sacrificed by exsanguination. Bronchoalveolar lavage (BAL) was collected using buffered saline (1x-10mL wash). Cells were removed from the BAL by centrifugation (300xg, 10 minutes at 4°C). The supernatant (2mL) was utilized to fractionate small aggregate (SA, supernatant) and large aggregate (LA, pellet) portions of BAL. The LA was re-suspended in 0.9% saline. Protein content for both the small (31.5µg/mL) and large (166.1µg/mL) aggregate fractions was determined by the bicinchoninic acid assay (BCA), Thermo Scientific (Rockford, IL, USA). Phospholipid content of the large aggregate fraction was assessed by determining the concentration of organic phosphate (1.57µg/µL).10

This protocol was approved by the Rutgers University Institutional Animal Care and Use Committee (IACUC) (Protocol Number: 06-028). The study was conducted in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.

AgNW incubations:

AgNWs were incubated at a concentration of 25 μg/mL from the original stock solution (on a Ag atom basis as determined by ICP-OES) in a temperature-controlled dri-block incubator at physiological temperature (37 oC) for 1 hour up to 336 hours (2 weeks) in the dark. This dose was selected in order to provide direct comparisons with our previous work on spherical AgNPs11 and on the in vitro effects of AgNWs on human alveolar epithelial cells.12 Accurate dosimetry in laboratory evaluations of the effects of inhaled particles has been a subject of concern, but very few data on AgNMs are currently available to evaluate realistic occupational and consumer exposures.13 In one study, published nanomaterial concentrations, including AgNPs, measured in air in manufacturing and research and development laboratories were reviewed, to identify input levels for estimating the nanomaterial mass retained in the human lung using the multiple-path particle dosimetry (MPPD) model. Model results were then converted (using the surface area and volume delivered in different types of cell culture well plates) to solution mass concentrations for in vitro testing. For AgNPs, alveolar retention for a working-lifetime (45 years) exposure duration was similar to higher concentrations (~ 50-200 µg/mL), tested in in vitro studies in the literature. The alveolar retention for a 24 hour exposure duration was equivalent to lower doses (~ 0.1-1 µg/mL) previously tested. Therefore, a dose representative of the lower-end of those previously tested in in vitro studies of AgNMs was selected for this study,14 whose primary aim is to deconvolute the effects of individual components of the LLF on the physicochemistry of AgNWs.

The presence of complexing anions such as Cl- or S- is expected to lead to the precipitation of insoluble silver species15, which would confound the measurement of free Ag+ ions, therefore non-interacting perchlorate buffers were selected as the dispersion medium. AgNWs were incubated in Sodium Perchlorate (NaClO4•H20, Sigma-Aldrich, >99%) (0.1 M) and the pH of the buffers was adjusted to 7 or 5, using either Perchloric Acid (HClO4, Sigma-Aldrich, 70%, 99.999% trace metals basis) or Sodium Hydroxide (NaOH, Sigma-Aldrich, anhydrous, 99.99%). These pH values were selected to simulate characteristic environments found in the lung. The interstitial and alveolar extracellular fluids have a pH of ∼7.416 while the cell cytoplasmic pH is ∼7.2. In the endocytic pathway of the cells, the pH decreases progressively from the early endosomes (pH∼6.5) to late endosomes (pH<6.0) and ultimately lysosomes (pH<5.5).17

To study the effect of each component of the lung surfactant on the stability of AgNWs, various combinations of these components were incubated together. Human surfactant consists mostly of phospholipids, with the most abundant being DPPC, therefore DPPC was used to study the effect of phospholipids. The effect of the two hydrophobic surfactant-associated proteins, SP-B and SP-C, was investigated using Curosurf®. Curosurf® is a natural surfactant, prepared from porcine lungs and used for the treatment of endogenous pulmonary surfactant deficiencies by intratracheal administration. It contains almost exclusively phospholipids but also about 1% of SP-B and SP-C. Finally, the role of SP-A and SP-D on the stability of AgNWs was studied by adding the small aggregate (SA) fraction of lung surfactant extracted from rat lungs. SA contains the more soluble components of the lung lining fluid, such as the non-specific lung proteins IgG and albumin, but also the collectins SP-A and SP-D. Atochina et al. showed that, in the bronchoalveolar lavage of control healthy mice, over 90% of phospholipids, 100% of both SP-B and SP-C and about 60% of SP-A, could be found in the LA fraction. In contrast, the majority of SP-D and the remaining SP-A were found in the SA fraction.18

A stock solution of 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC, Sigma-Aldrich, semisynthetic, ≥99%) (10 mg/mL) in DI-H2O was prepared by magnetic stirring overnight. DPPC, Curosurf® (poractant alfa, Chiesi Farmaceutici, S.p.A.) and LA were used at amounts that correspond to a total lipid concentration of 100 μg/mL. SA was used at amounts that correspond to a total protein concentration of 100 μg/mL. These concentrations were chosen to correspond to the average values observed in the surfactant obtained from the bronchoalveolar lavage fluid of healthy persons.19 For the phospholipids that constitute most of pulmonary surfactant, the critical micelle concentrations (CMC) fall within the range of 10−10 to 10−9 M.20 For DPPC, for instance, a CMC of 10−10 M is equivalent to 7.34×10-5 μg/mL, therefore the formation of micelles is expected in our experiments. However, in essentially all of the in vitro studies, and at estimated physiological concentrations, pulmonary surfactant is well above its CMC.20

Scanning Electron Microscopy (SEM):

The morphology and size distribution of the AgNWs were characterized using a LEO 1525 Field Emission Gun Scanning Electron Microscope (FEG-SEM, Carl Zeiss Microscopy GmbH, UK). The SEM was operated in secondary electron mode at an accelerating voltage of 5 kV, using the InLens detector. Samples were prepared by drop-casting aliquots of the AgNW suspensions on a piece of silicon wafer and dried under ambient conditions in a fume cupboard and in the dark. Samples were stored under vacuum and in the dark. The size distribution of the AgNWs was characterized using SEM images and ImageJ software (http://rsb.info.nih.gov/ij/).

Transmission Electron Microscopy (TEM) Sample Preparation:

To prepare TEM samples from the incubated AgNWs, aliquots were removed at each time point and, after washing three times with DI-H2O to remove excess salts or organic surfactant, they were drop cast on 300 mesh holey carbon film TEM grids (TAAB) in the dark. The grids were blot-dried with filter paper and were immediately placed under vacuum and in the dark to avoid reactions induced by ambient atmosphere. The grids were imaged within a period of 2 weeks. For TEM samples prepared from as-synthesized AgNWs and imaged after a storage period of up to 3 months under vacuum, no changes in their physicochemical properties were detected (more than 100 wires analyzed, Figure S1). To enhance phospholipid contrast, samples were positively stained with 2% uranyl acetate in water.

Figure S1. TEM images of a sample prepared from as-synthesized AgNWs, imaged after a 3-month storage period under vacuum and in the dark.

TEM:

Bright field transmission electron microscopy (BFTEM), high resolution transmission electron microscopy (HRTEM) and high angle annular dark field scanning transmission electron microscopy (HAADF-STEM), combined with selected area electron diffraction (SAED) and energy-dispersive X-ray spectroscopy (EDX) were carried out using a JEOL JEM-2100F fitted with an EDX detector (Oxford Instruments). The scattering intensity in HAADF-STEM is proportional to Zn (n ~ 2), therefore this technique is highly sensitive to atomic number (Z) variations within the sample. An accelerating voltage of 200 kV was used for both TEM and STEM experiments. For STEM experiments, the inner and outer HAADF collection angles were 150 and 400 mrad, respectively, and the probe diameter was <1 nm.