A quantitative method for determining uptake of silica nanoparticles in macrophages by single particle inductively coupled plasma-mass spectrometry

Abstract

Engineered nanomaterials are becoming increasingly ubiquitous in our society, with numerous applications in medicine, consumer products, bioremediation, and advanced materials. As these nanomaterials increase in variety, analyzing their characteristics is of great importance. Single particle inductively coupled plasma- mass spectrometry (SP-ICP-MS) is a high throughput, sensitive, and robust instrumental analysis method used to simultaneously characterize and quantify nanoparticles in a variety of matrices. One such type of nanoparticle of interest are amorphous silica nanoparticles (SiNPs). SiNPs have widespread use in consumer products such as food and cosmetics, and are prime candidates for novel medical applications and uses in environmental bioremediation. Despite their increased use, SiNPs have been shown to have toxicological properties in vitro and in vivo, particularly with regard to the immune system. Because of the potential for increased SiNP exposure to the general public and in occupational settings, examining the relationship SiNPs have with immune cells such as macrophages, is vital to elucidating mechanisms of toxicity. To effectively determine the toxicity of nanoparticles, it is critical to examine dosimetry and the amount of nanoparticle taken up by the cell of interest. Different cell types have different uptake profiles, and varying physicochemical properties govern nanoparticle dosimetry and uptake in cells. Here, we describe a protocol using SP-ICP-MS to quantify and characterize the size, size distribution, and amount of SiNPs present in a cell and media sample. We use a single-step digestion, which allows for the digestion of biological matrices while simultaneously keeping the SiNPs intact for SP-ICP-MS analysis. Clinically, this approach has potential to be used as a method for analyzing SiNPs in other biological matrices, potentially as a way of defining SiNP uptake as a biomarker in immune-mediated diseases.

INTRODUCTION:

The advent of engineered nanomaterials has sparked a rapidly growing field, with a multitude of applications in medicine, consumer products, bioremediation, and advanced materials (Rudramurthy and Swamy, 2018). The unique applications of nanomaterials has resulted in their increased heterogeneity, with various sizes, chemical structures, shapes, and surface functionalization. Because of the heterogeneity of nanoparticles being used across applications, their toxicity, both environmental and biological, has not been well characterized (Ajdary et al., 2018).

One such class of nanoparticles with limited toxicological understanding are silica nanoparticles (SiNPs). SiNPs are a promising candidate for many applications in medicine and bioremediation, and are already used in a wide variety of cosmetics (Fytianos et al., 2020). SiNPs, particularly amorphous SiNPs, were thought to have very limited toxicity when compared to crystalline silica. Chronic exposure to crystalline silica, for instance, is well known to cause robust pulmonary fibrosis in humans, known as silicosis (Leso et al., 2019). Crystalline silica has ordered molecular shapes, which strongly interact with alveolar macrophages and produce oxidative stress in the lung. Amorphous SiNPs, on the other hand, have less ordered shapes, giving more variety to shape. As a result, characterizing general toxicity profiles in amorphous SiNPs has been difficult. In the past, crystalline silica was the major focus for silica toxicity studies due to the development of chronic pulmonary inflammation in vivo, while amorphous silica was shown to cause only transient pulmonary damage (Warheit et al., 1995). More recent studies, however, have shown that amorphous SiNPs can be immunogenic and cause oxidative stress in vitro (Wang et al., 2009), further highlighting the need for more detailed studies into these particles. While informative, these mechanistic studies often lack uptake data when defining the interface and dosimetry between the nanoparticle and the in vitro system, and one study has already shown that silica toxicity is dependent on the uptake of silica by alveolar macrophages (Hamilton et al., 2008). This dosimetry data is thus critically important to characterize toxicity and for comparison across studies, and will assist in defining new toxicity paradigms for SiNPs.

The recent toxicological concerns surrounding SiNPs necessitate the ability to analyze the interactions between SiNPs and biological matrices, in particular, the immune system. Because the toxicity of SiNPs involves the activation of the NLRP3 inflammasome, ROS generation, and cytotoxicity (Peeters et al., 2013), immune cells that mediate many of these responses, such as macrophages, are often examined as primary targets of SiNP exposure (Fytianos et al., 2020). Importantly, there are several immune cell models where literature is lacking on the toxicity of SiNP exposure, such as in mast cells (Maurer-Jones et al., 2010), Langerhans cells (Nabeshi et al., 2010), and other critical immune cell types. Interestingly, Langerhans cells and mast cells are heavily present in the skin, a location where SiNP exposure is expected to be high given their current applications in cosmetics and skincare. This gap in the literature requires the ability to properly characterize and quantitate SiNP exposure in these cells and other cell types.

SiNP characterization and quantification of uptake, particularly in in vitro systems, has previously been done using fluorescence microscopy and transmission electron microscopy (Choi et al., 2010). Other techniques include inductively coupled plasma methods, such as ICP orbital emission spectroscopy (OES) for elemental analysis of silicon. One method used to analyze elemental composition in many matrices is ICP mass spectrometry (MS). Conventional ICP-MS is used to analyze the concentration of dissolved elements in a sample, similar to ICP-OES. However, in recent years, the advent of microsecond resolution in dwell times for ICP-MS analysis has allowed for individual nanoparticles to be counted in an individual sample. This is known as single particle ICP-MS (SP-ICP-MS) (Montaño et al., 2016). SP-ICP-MS allows for the simultaneous quantification and characterization of nanoparticles in any digestible matrix.

Given the need to analyze the relationship between SiNPs and immune cells, we have developed a method to quantitatively determine the uptake of SiNPs in vitro using SP-ICP-MS. Normally, SiNP digestion requires the use of hydrofluoric acid (HF) to dissolve the very stable SiO₂ and SiNPs. However, it is well characterized that SiNPs are stable at strongly acidic pHs. These same pHs can be used to digest biological matrices, leaving SiNPs intact for analysis by SP-ICP-MS. By removing the necessity to utilize HF, digestion can be done in a single step. In this article, we describe a digestion and instrumental method to quantify SiNP uptake in a variety of biological matrices at various nanoparticle sizes. The data generated is a frequency histogram that not only quantifies the number of SiNPs in each individual sample, but also characterizes the size and polydispersity of the nanoparticle population in the sample. This method can be used in a variety of biological matrices and here, and to show the relevance of the approach and illustrate the protocol, we use an immune cell model to measure uptake of two sizes of SiNPs as a function of time. Macrophages were chosen as the exemplar immune cell model because SiNP exposure and toxicological effects are well characterized, particularly in RAW 264.7 murine macrophages (Marquardt et al., 2017; Dussert et al., 2020).

In this protocol, we define a way to treat and digest cells for SP-ICP-MS analysis to assay uptake. Specifically, we describe a novel method for the digestion of cells and media to analyze SiNPs (Basic Protocol 1). Additionally, we describe, in detail, the culturing of RAW 264.7 macrophages for the purpose of assaying SiNP uptake (Support Protocol 1). Then, and prior to analysis by SP-ICP-MS, the anticipated size and distribution of the nanoparticle population can be described using dynamic light scattering (Support Protocol 2), which is important because the determination of the anticipated size of the nanoparticles in the sample is vital to the ability to analyze the particles via SP-ICP-MS. Particles must be above a certain size for proper analysis with SP-ICP-MS and, therefore, anticipated sizes must be above this lower limit of detection. Instrument calibration and optimization for SP-ICP-MS should be performed before sample analysis, and we describe this in Support Protocol 3. Finally, in Basic Protocol 2, we describe how to run the samples and acquire the data for both characterizing and quantifying the uptake of the nanoparticles in vitro. Figure 1 provides a general outline of the workflow.

Figure 1.

Flowchart of of the protocol described here for the treatment and digestion of cells and media, as described in Basic Protocols 1 and 2. Figure created with BioRender.com.

All biological experiments should be completed in a biosafety cabinet rated for biological safety level 2 (BSL2). Nitric acid (HNO₃) is highly corrosive and oxidizing, and should be handled with caution with proper laboratory personal protective equipment (PPE), including eye protection, laboratory coats, and gloves. All samples should be added only to plastic tubes, to prevent silica contamination. Chemical reagents should be reconstituted in either Milli-Q grade H₂O or TraceMetal grade HNO₃. Because of instrument sensitivity, the calibration solution and other solutions should be capped as long as possible to prevent possible ambient contamination.

STRATEGIC PLANNING

The doubling time of the cell model to be used, particularly if the cells of interest are a cancerous cell line, should be taken into consideration prior to beginning the protocol. This will allow the user to plate cells to appropriate confluency before the treatment and digestion of the cells following SiNP treatment. Additionally, the concentration of the SiNPs for the treatment must be optimized. The highest noncytotoxic dose should be used in order to allow the user the most control over their concentration for the dilutions the user will need to do prior to analysis (dilution steps for the workflow of the experiment are illustrated in Figure 1.)

BASIC PROTOCOL 1: TREATMENT OF CELLS WITH SiNPs AND DIGESTION OF BIOLOGICAL MATRICES

Here, users will treat cells with SiNPs, for different times, and then digest them in order to subsequently analyze uptake via SP-ICP-MS. This procedure is important because SiNPs must be assayed and subsequently liberated from the biological matrix they are contained within. To assay cellular uptake in macrophages, users will digest cells with a single-step digestion. In the end, users will obtain both cell and media samples which have been digested and, therefore, are ready for SP-ICP-MS analyses.

Materials:

• Cells at passage 4-10 (see Support Protocol 1)

• Sterile 48-well cell culture plates (BioLite, Thermo Scientific, 130187)

• 200-nm pristinely synthesized SiNPs (NanoComposix) (NanoXact SISN200)

• 300-nm pristinely synthesized SiNPs (NanoComposix) (NanoXact SISN300)

• Complete cell culture media (See Reagents and Solutions)

• 0.25% Trypsin/EDTA (Gibco 25200-056)

• TraceMetal grade (or better) 70% HNO₃ (Fisher Chemical 7697-37-2)

• 1.5-mL microcentrifuge tubes

• Ultrasonic Processor Model CV188 (or equivalent) sonicator

• Eppendorf 5424R (or equivalent) microcentrifuge

• FA-45-24-11 (or equivalent) microcentrifuge rotor

• Micropipettes (P-1000, P-200, P-20, P-2.5)

• Milli-Q water (from a Millipore or equivalent system)

Protocol Steps

Day 1

• 1Culture RAW 264.7 cells as described in Support Protocol 1. When cells are approximately 85% confluent, resuspend in complete medium and count cells, as also described in Support Protocol 1.
• 2In a 48-well plate, seed cells at 7 x 10⁴ cells per well in 100 μL of complete cell medium. Plate cells in technical triplicates. Prepare enough wells to treat with both particle sizes.

  In order to allow cells to become confluent for the assay, plate cells the night before the experiment at approximately 70% confluency based on the doubling time of the cells used. In the case of terminally differentiated cells such as primary cells, plate the exact number of cells required for the assay (in the case of a 48-well plate, 1 x 10⁵ cells).

  Triplicate technical replicates should be seeded for each individual passage; each passage assayed is one biological replicate.

Day 2

• 3Make stock solutions of pristine nanoparticles in 1.5-mL Eppendorf tubes in Milli-Q water. Lightly sonicate suspensions of both 200 and 300 nm SiNPs for 1 minute. Then, using complete cell medium, dilute particles to prepare a 1 mg/mL stock from the concentration given by the certificate of analysis from the manufacturer. From each stock, prepare enough of a solution containing 100 μg/mL of particles in complete media for cell treatments. You will need 200 μL of each of the particle solutions per treatment.

  Sonication can be done using a hand sonicator or a bath sonicator for pristine SiNPs because they are monodisperse in water. With polydisperse or environmental samples, harder sonication is likely required to suspend particles in solution.

  100 μg/mL was found to be a non-cytotoxic dosage with the SiNPs and cells used in this study. Treatment dosages may need to be changed for different cell lines or SiNPs with different physicochemical properties. This can be done with any cytotoxicity assay, such as an MTS or LDH assay (Malich et al., 1997; Kumar et al., 2018).

  Prior to treatment of cells with SiNPs, verify the size of the particles using dynamic light scattering, as described in Support Protocol 2.

• 4Aspirate media from treatment wells and replace with 200 μL of complete medium supplemented with 100 μg/mL SiNPs of both sizes (prepared in Step 3).

  This works if cells are adherent. If cells are in suspension, centrifuge the plates in a microcentrifuge at 300 x g for 5 minutes prior to medium aspiration.

• 5Following the time points of interest (in this case 1, 6, and 24 hours), remove media and add to 1.5-mL Eppendorf tubes. Place the tube with media on ice. Add 200 μL of 0.25% Trypsin/EDTA to the cells and incubate at 37°C for 5 minutes. Resuspend cells in the 0.25% Trypsin/EDTA, ensure they were removed from the plate (as checked by a cell culture microscope), and place them in 1.5-mL Eppendorf tubes. Spin down the 1.5-mL Eppendorf tubes containing the cells at 300 x g for 5 minutes at 4°C. Aspirate the 0.25% trypsin/EDTA and leave the cell pellet in the tube. Freeze both the cell pellet tube and its corresponding media tube overnight at −20°C.

  For cells in suspension, 0.25% trypsin/EDTA is not required to remove cells from the media. Instead, cells and media should be moved to a 1.5-mL Eppendorf tube and centrifuged at 300 x g for 5 minutes at 4°C. The media should then be moved to its own 1.5-mL Eppendorf tube corresponding to the cell pellet that it came from. Then, both the cell pellets and media should be frozen at −20°C, as described above.

Day 3

• 6Thaw samples (media and pellets) and add 200 μL of 70% TraceMetal grade HNO₃ into the tubes containing the 200 μL of media. Additionally, add 200 μL of 70% TraceMetal grade HNO₃ to each corresponding cell pellet, and let both sets of tubes sit at room temperature overnight, or at 60°C for at least 6 hours.

Day 4

• 7Dilute digested cells and media to exactly 10 mL in 1.5% HNO₃ (made from 70% TraceMetal grade HNO₃ and Milli-Q H₂O) in separate 15-mL conical tubes. Then, optimize the amount of digested pellet or media for proper instrument signal (see Support Protocol 3 These cells and media samples in 15-mL tubes can be stored at room temperature indefinitely as long as they are covered in parafilm or capped.
• 8Analyze samples as described in Basic Protocol 2.

SUPPORT PROTOCOL 1: CULTURING RAW 264.7 CELLS FOR SiNP UPTAKE ASSAY

In this protocol, users will culture RAW 264.7 cells, a well established murine macrophage model (Taciak et al., 2018), for use in Basic Protocol 1, for assaying the uptake of SiNPs, to understand the toxicological consequences of their exposure and uptake. In the end, users will have knowledge of how to culture immune models of cells as a way to quantify SiNP uptake in vitro.

Materials:

• RAW 264.7 mouse cells obtained from ATCC (TIB-71™)

• Complete cell culture media (See Reagents and Solutions)

• Phosphate buffered saline (PBS) (Thermo Scientific, 28372)

• 10 mg/mL acridine orange (Invitrogen, A3568)

• 175-cm² cell culture flasks (Thermo Scientific, 159920)

• Cell scrapers (Thermo Scientific, 179707PK)

• Nexcelom spectrum (or equivalent cell counter)

• Nexcelom Cellometer cell counting chambers (Nexcelom, SD100)

• Sterile incubator, 37° C, 5% CO₂

• Thermo Scientific 1300 Series A2 sterile cell culture hood (or equivalent laminar flow hood)

• Tissue culture microscope (Fisher scientific MicroMaster or equivalent)

Protocol Steps

This protocol is specific to RAW 264.7 mouse macrophages, however, many immune cells are cultured in similar manners with similar reagents, and could be used for SP-ICP-MS analysis of SiNP uptake.

1. Culture 5 X 10⁶ RAW 264.7 cells in a 175-cm² cell culture flask containing 25 mL of complete media. Place the flask in a sterile incubator at 37°C with 5% CO₂.

   Cells can also be cultured in a 75-cm² flask with 15 mL of complete cell culture media.

2. Upon reaching approximately 85% confluency (approximately 2 days), passage cells by scraping. First, wash cells with 10 mL of warm PBS and aspirate the media, taking care not to aspirate the cells. Then, add 10 mL of warm complete media to scrape cells into. Scrape cells firmly into the media. Passage these cells into a new 175-cm² flask with 25 mL complete media at a ratio of 1:4-1:6.

3. After passaging cells, count cells using acridine orange stain. To do this, mix 20 μL of the suspended cells with 20 μL of a 1 mg/mL acridine orange solution. Then, count the cell concentration utilizing a Nexcelom Spectrum cell counter or similar cell counter by adding the 1:1 dilution of cells and acridine orange to the cell counter slide and then inserting it into the cell counter. Then, and using the FL channel for acridine orange, count cells. This will give the concentration of viable cells for seeding purposes. Continue growing and passaging, or proceed to Basic Protocol 1.

   For the purpose of the experiments described in this article, cells between passage 4 and 10 were utilized. Cells were passaged every third day.

SUPPORT PROTOCOL 2: DETERMINATION OF SiNP SIZE VIA DYNAMIC LIGHT SCATTERING

Dynamic light scattering (DLS) measurements allow determining the anticipated size of the SiNPs analyzed. It is important to determine the expected size of nanoparticles in any given sample in different matrices because when the expected size of the nanoparticle is determined, it assists in the analysis of the particle by ICP-MS. This is to ensure that SiNPs being assayed are able to be detected by the instrument with respect to its lower limit of detection of size. In this protocol, users will utilize DLS to determine if their SiNP sample is appropriate for SP-ICP-MS analysis. This is vital because if the SiNP sample is too small or too polydisperse, then the particle will be unable to be read by SP-ICP-MS. This will be done by using a zetasizer dynamic light scattering instrument. The sizing data that is generated in this protocol will determine if users can properly quantify and characterize SiNPs in Basic Protocol 2.

Materials:

• Complete cell culture media (see Reagents and Solutions)

• 70% TraceMetal grade (or higher purity) HNO₃

• Milli-Q water (from a Millipore or equivalent system)

• Pristinely synthesized nanoparticles from Basic Protocol 1

• Nanoparticle sample used for treatments in Basic Protocol 1

• Zetasizer Nano ZS dynamic light scattering instrument and software, version 7.12 (Malvern, UK).

• Fisher-brand semi-micro (1.5-mL) cuvettes (or other equivalent disposable sizing cuvette for DLS). (Fisher Brand, Cat. 14955127)

• Micropipettes (P-1000, P-200, P-20, P-2.5)

Protocol Steps

1. Optimize nanoparticle concentration for the DLS analysis by first making a solution of 1 mg/mL SiNPs in each assayed size in Milli-Q H₂O.

   Milli-Q H₂O is the typical vehicle used for pristinely synthesized SiNPs, however, the vehicle can be any given buffer that the 1 mg/mL SiNPs were prepared in, such as cell culture media.

2. Make a 1:1000, 1:100, and 1:10 dilution of 1 mg/mL SiNPs in separate 1.5-mL semi-micro cuvettes (for a final concentration of 1 μg/mL, 10 μg/mL, and 100 μg/mL, respectively).

3. In the zetasizer software, open a new measurement file. Under the measurement file, select the size measurement. Using the parameters section, select the proper dispersant using the correct refractive index of the solvent being analyzed. Then, insert the disposable cuvette containing the sample into the receptacle. Then, using the sizing module on the zetasizer software, analyze the size of the pristinely synthesized particles by clicking start. Examine the sizing distribution and polydispersity index of the sample analyzed for each dilution, which will be shown automatically. The sample that shows the narrowest peak and the lowest polydispersity index should be the dilution used for the remainder of DLS analyses.

   If the sample was diluted in Milli-Q H₂O, the size of the particle should be close to the expected synthesized size. In a buffered solution potentially needed for treatments in different cell lines, the size may be larger than expected, however, the polydispersity index will still be small.

   Samples should be measured in triplicate, so averages and statistics can be analyzed. Additionally, if sample is sedimenting or actively agglomerating, this can also be noted in order to change buffering conditions for treatments.

4. Using the optimized dilution, measure hydrodynamic diameter (as done is step 3) of each type of nanoparticle sample now prepared in Milli-Q water, 70% HNO₃, and complete cell culture media, using the same method as step 3. Compare this peak shape and the polydispersity index to the pristinely synthesized sample that was assayed in step 3.

   This is the expected size and polydispersity index that the nanoparticle sample will have. The information the user obtains from this protocol determines whether the sample is large enough and monodisperse enough to accurately analyze via SP-ICP-MS, given the SiNP is greater than the defined LLOD of the method (~180nm). If the polydispersity index is less than 0.200, then the sample is considered monodisperse.

SUPPORT PROTOCOL 3: OPTIMIZATION OF SAMPLE AND ICP-MS PARAMETERS FOR SP-ICP-MS ANALYSIS OF CELLS AND MEDIA

In this protocol, users will optimize the SP-ICP-MS instrument parameters to be able to accurately analyze uptake of SiNPs in Basic Protocol 2. This is important because using the ICP-MS for the purposes of SP-ICP-MS can have complicated but necessary optimization steps. Prior to each analysis with SP-ICP-MS, the instrument must be optimized. Optimizing the instrument gives the best possible sensitivity and proper temperature of plasma to maximize ionization efficiency of the SiNP analyte. This must be done for successful sample analysis in Basic Protocol 2.

Materials:

• Standard pristinely synthesized nanoparticles at sizes of 100, 200, 300, and 500 nm (in Milli-Q H₂O) at 10 mg/mL

• 70% TraceMetal grade (or better)( Fisher Chemical 7697-37-2)

• Milli-Q water (from a Millipore or equivalent system)

• 1.5% HNO₃ (see Reagents and Solutions)

• Calibration solution or 1 ppb of In, Pb, Be, Ce, and U in 1.5% HNO₃ (PerkinElmer N9300218, or see Reagents and Solutions)

• NexION 2000B (or equivalent) inductively coupled plasma-mass spectrometer with single particle capability (PerkinElmer, Waltham, MA)

• Syngistix 2.3 software with nano module or appropriate instrument specific software

Protocol Steps

1. Initiate plasma on NexION 2000B ICP-MS instrument through the Syngistix software and allow plasma to warm for 15 minutes. Activate the peristaltic pump through the pump section such that the speed is −35 rpm. Following this, verify proper sample flow into the nebulizer by introducing at least 5 mL of Milli-Q water for 5 minutes into the sample probe and examine the spray chamber to see mist forming. Then, ensure proper draining from the bottom of the spray chamber by looking for excess mist draining in the form of water droplets.

2. Introduce at least 5 mL of calibration solution (see Reagents and Solutions) into the sample probe, again ensuring proper flow in and out of the spray chamber. Once ensured, begin the performance check. In the software, utilizing the daily performance check function, optimize the torch alignment, the QID, and the nebulizer gas flow parameters. The instrument will automatically determine if the instrument’s readings are optimized for sensitivity and proper torch temperature.

   Adjustments for instrument parameters to address sensitivity and torch temperature are more thoroughly covered in the “troubleshooting” section.

   If the instrument automatically determines that the calibration and optimization has failed, review the “troubleshooting” section (see Table 1).

3. Verify under the conditions section of the Syngistix software that the optimal parameters have saved for the instrument and will be applied to the next run by saving the conditions. This ensures that Basic Protocol 2 will be run at optimal instrument conditions.

4. Proceed to Basic Protocol 2

Table 1.

Troubleshooting Guide for SP-ICP-MS analysis

Problem	Possible Cause	Solution
Sample does not appear monodisperse after DLS analysis.	Sample is polydisperse.	Check sample for debris or other contamination.
Sample appears of different size to expected.	Dilution factor is incorrect.	Change dilution factor so that sample appears to the expected size.
Cells growing more slowly than expected	Incorrect medium used.	Utilize correct medium (see Reagents and Solutions for Complete Cell Media recipe).
	Cells were overgrown previously, thus, inhibiting future growth.	Begin a new cell culture and discard present culture.
Cell viability decreases (<80%) after treatment	Treatment is reducing viability of cells.	Reduce concentration of treatment and determine cytotoxicity of treatment dose.
High variability in particle content between technical replicates found in Basic Protocol 2.	Variability in cell numbers between treatments	Ensure cell counts are the same during the plating process following subculture
	Instrument not properly optimized	Optimize instrument such that sensitivity is its highest and calibration solution has an RSD% (given by the computer program) of <3.0%. This involves changing the nebulizer gas flow and the torch alignment.
	Sample not fully digested	Digest sample for longer amount of time and apply heat (60°C)
Instrument does not pass optimization	Calibration solution is not the correct concentration.	Remake calibration solution
	Mass calibration failed	Using the Syngistix software, recalibrate the Mass calibration with full optimization
No signal or peaks during run in Basic Protocol 2.	Particle sizes are below the lower limit of detection for particle diameter	Determine the lower size limit of the instrument by defining the standard deviation of the background signal. Then, using this standard deviation, multiply by 3. Ensure particles that are being analyzed are within the capability of the instrument .
	Particle concentration is below the lower limit of quantification.	Run sample with a lower dilution factor.

BASIC PROTOCOL 2: ANALYSIS AND QUANTIFICATION OF SiNP UPTAKE IN MACROPHAGES WITH SP-ICP-MS

In this protocol, the SiNPs that have been taken up by cells or are still in the media (Basic Protocol 1) are counted and characterized. Once the samples have all been run, the given data allows for the quantitative calculation of the percent uptake in each individual sample. Compiled within groups of biological and technical replicates, trends in uptake across time, concentrations, nanoparticle physicochemical properties, and immune cell models can be elucidated. To do this, users will utilize the calibrated SP-ICP-MS system (as described in Support Protocol 3). The data obtained with this protocol will define the nanoparticle population that has been taken up by cells, and final analysis of the data will give the user data to determine the percent uptake of the particles in the cell model and the size and polydispersity of the particles in each sample.

Materials:

• Digested cell and supernatant samples from Basic Protocol 1

• 96-well ICP-MS microtiter plate (Elemental Scientific, MT-96-2mL-02)

• Standard 5.0 ppb solutions of 100 nm, 200 nm, 300 nm, and 500 nm pristinely synthesized monodisperse, spherical SiNPs (NanoComposix, SISN100, 200, 300, and 500 diluted in 1.5% HNO₃)

• NexION 2000B (or equivalent) inductively coupled plasma-mass spectrometer with single particle capability (previously optimized, per Support Protocol 3)

• Syngistix 2.3 software with nano module or appropriate instrument specific software

Protocol Steps

1. Add 2 mL of each cell and supernatant sample from Basic Protocol 1 into separate wells of a 96-well microtiter plate. Additionally, consider wells for blanks and standards.

   The standards will be used to run a standard curve.

2. First, employing the nano module within the Syngistix software, define a method by entering desired parameters in the “Method” section of the software. In this method, define the standards that will be run. In this case, you will be using the pristinely synthesized SiNPs (see materials above). Additionally, the analyte should be defined as ²⁸Si⁺, and the density used should be 2.3 g/cm³.

3. Add the solutions of the pristinely synthesized particles of 4 different sizes listed and the blanks to the plate, and generate a standard curve. Define which autosampler location the standards are in and list them in the batch section of the Syngistix software. Begin the run by loading the method, the batch list, and by saving the dataset to a new folder. Begin the run once everything is defined by clicking “analyze batch”. In the method, it should be defined that each sample and standard is run for 100 seconds, using a 50 μs dwell time analyzing the defined ion.

   For a standard curve to be considered valid for the run, the correlation coefficient (r²) for the four standards plus the blank must equal or exceed 0.995.

4. When the calibration has finished running, save and apply that calibration to the batch list containing all of the samples and the method being used, as defined in the software. Verify that the method used to run the standards is the same method used to run the samples. It will take approximately 1 mL to run each standard and sample. Run samples with the same parameters as the standards (from the conditions defined in Support Protocol 3).

   Cells and the corresponding supernatants should be run together in the same run. Samples should be from 3 individual biological replicates with 3 technical replicates per biological replicate in order to make valid statistical comparisons.

5. Export the full sample list to an excel spreadsheet.

   After the run, the data will be given in the form of frequency histograms. Additionally, quantitative and characterization data is given in the form of average diameter of the nanoparticles detected, most commonly detected nanoparticle size, concentration of the nanoparticles in particles/mL, and mean intensity of the pulses for each individual sample. This data can be exported directly to Microsoft Excel. Additionally, a Gaussian function will automatically be determined for each sample based on a regression of the frequency histogram of the size of the particle detected against the frequency at which that size nanoparticle was detected. This Gaussian function is demonstrative of the representative nanoparticle population contained within the sample. This can be exported directly to Excel for each individual sample.

   For a sample with SiNPs from an environmental sample or other amorphous types of SiNP, a log normal regression may be required to adequately describe the representative nanoparticle population in the sample.

   Frequency histograms from the data will show the representative nanoparticle populations along with the most common size. Frequency histograms of cells and media treated with 200 and 300 nm SiNPs can be seen in Figure 2.

   Then, using the particle concentration per mL, the % uptake can be found quantitatively with the following equation:

   $$\mathrm{\%}uptake=\frac{{mC}_{cp}}{{mC}_s}X100$$

   Where mC_cp = the mass concentration of SiNPs in the cell pellet and mC_s = the mass concentration of the SiNPs in the supernatant media.

   Representative data of uptake as a function of time and SiNP size in a RAW 264.7 murine macrophage cell line is shown in Figure 3.

   If the user decides to assay the number of nanoparticles taken up per cell, the following equation can be used:

   $$\#ofparticlestakenuppercell=\frac{Totalnumberofparticlesdetectedinsample}{\#ofcellsdigestedinsample}$$

Figure 2. Representative frequency histograms of SiNPs detected by SP-ICP-MS in RAW 264.7 cells and media from the same technical and biological replicate following treatment with SiNPs for 24 hours.

A) non-treated cells; B) cells treated with 200 nm SiNPs; C) cells treated with 300 nm SiNP; D) media from no treatment samples; E) media from 200 nm SiNP treatment; and F) media from 300 nm SiNP treatment. Red line denotes the representative nanoparticle population as given by the Gaussian non-linear distribution of the histogram. Dashed line represents expected SiNP size.

Figure 3. Time course of percent uptake of SiNPs in RAW 264.7 cells at 0, 1, 6, and 24 hours.

The red line represents mean uptake of 200 nm SiNPs (obtained from Nanocomposix) by cells at each time point. The blue line represents mean uptake of 300 nm SiNPs by cells at each time point. Error bars represent standard deviation. *p < 0.05 and **p < 0.01 when compared to the 200 nm uptake percentage at that timepoint by an unpaired t-test. Three individual experiments were performed with 3 technical replicates at each time point and particle treatment.

REAGENTS AND SOLUTIONS:

Complete cell culture media

• Dulbecco’s modified eagle medium (DMEM) + 4.5 g/L D-glucose, L-glutamine (Gibco, 11-965-092)

• Penicillin/streptomycin (10,000 U/mL Pen, 10,000 μg/mL Strep) (Corning, MT30002CI)

• Fetal bovine serum (FBS), heat inactivated (Gibco, 10-082-147)

  All reagents should be added together in a sterile biosafety cabinet.

  Following combination, media should be filtered sterilized using a 0.22 μm filter.

1.5% v/v HNO₃

• 70% TraceMetal grade HNO₃ (Fisher Chemical 7697-37-2)

• Milli-Q H₂O (from a Millipore or equivalent system)

Add 10.71 mL of 70% HNO₃ into 489.29 mL of Milli-Q H₂O. Store in a polycarbonate plastic bottle at room temperature for up to 12 months

• Acid should always be added to water, in that order.

• All reagents should be added slowly in a fume hood.

• Preparation should be conducted in a fume hood as the nitric acid is caustic and can burn the skin or lungs if exposed.

Calibration Solution

• 10 ppm solution of dissolved indium metal in 1.5% HNO₃

• 10 ppm solution of dissolved beryllium metal in 1.5% HNO₃

• 10 ppm solution of dissolved cerium metal in 1.5% HNO₃

• 10 ppm solution of dissolved uranium metal in 1.5% HNO₃

• 10 ppm solution of dissolved lead metal in 1.5% HNO₃

  Dilute to 1 ppb in 1.5% HNO₃ solution, to at least 50 mL in Milli-Q water.

  All reagents should be added slowly in a fume hood.

COMMENTARY:

Background Information:

Silica nanoparticles have become ubiquitous in our society, as they are used in a multitude of consumer products. These include cosmetics, food, advanced materials, and biomedical products (Niculescu, 2020; Rudramurthy and Swamy, 2018). Because of the potential and expanding exposure that humans face when it comes to SiNPs, the interactions between these nanomaterials and human systems must be examined, particularly through a toxicological lens; evaluating the toxicity of SiNPs is important to determining future uses for them. The protocol described above will advance the field of SiNP toxicity by allowing for a quantitative method in order to characterize different SiNP interactions with various cell types to determine if their toxicities are dependent upon uptake. Compared to well defined toxicities observed with crystalline silica, SiNPs are less well characterized (Warheit et al., 1995). This is partially because SiNPs are heterogenous, and their heterogeneity makes the general toxicological analysis of SiNPs difficult. For instance, SiNPs of different sizes may interact with different cellular pathways (Hsu et al., 2021). Additionally, morphology of SiNPs can govern different modes of observed toxicity (Kersting et al., 2020). In general, each individual study in the literature only tests one kind of SiNP and in one cell model, and often makes generalizations about the behavior of SiNPs in vitro. The heterogeneity of cell types and SiNP types combine to make generalizations about SiNP toxicity impossible. However, it is important to have the ability to characterize SiNPs and quantify uptake as a method to define the interactions different SiNPs have with different cell models. Additionally, it is important for this method to be applicable for comparisons across different cell models with different SiNPs.

Previous characterization methods to describe the uptake of SiNPs in cell models are generally qualitative forms of microscopy. Recently, inductively coupled plasma methods such as ICP-MS and ICP-OES have allowed for less time intensive and more quantitative analyses of SiNPs in biological matrices; however, conventionally, they lack characterization data of the SiNPs.

ICP-MS is an excellent tool for the sensitive detection of elemental analytes (Cubadda, 2007). Prior to the usage of ICP-MS for the analysis of elemental analysis, other plasma-mediated methods, such as ICP-OES, have been used (Comparison of ICP-OES and ICP-MS for Trace Element Analysis - US). ICP-OES instrumental methods are prone to error by having a lower linear dynamic range (LDR) and lower sensitivity than that of ICP-MS (Comparison of ICP-OES and ICP-MS for Trace Element Analysis- US). One major advantage of ICP-OES is its ability to analyze samples with more total dissolved solid than conventional ICP-MS, however, by utilizing SP-ICP-MS, this shortcoming is overcome (Single Particle ICP-MS and Its Advantages in Analyzing Nanoparticles in Environmental Matrices). The analysis of nanoparticles utilizing SP-ICP-MS has generally been done with metal nanoparticles (Donovan et al., 2016; Yang et al., 2016). This is largely because, in the past, technical limitations have hindered ICP-MS approaches to non-metal analyses. These include high background, a variety of interferences, and high lower limits of detection. With our optimization of digestion and analytical parameters, non-metal nanoparticles, in particular SiNPs and their demonstrated toxicological effects in a variety of cell models, can now be analyzed with this method.

The analysis method described in this paper will advance the usage of SP-ICP-MS with respect to SiNP detection for improving characterization of toxicity profiles of SiNPs in vitro. There are several advantages of this method when compared to conventional ICP-MS methods for SiNP detection. With this method, for instance, trace levels of SiNPs are able to be detected. Additionally, the single-step digestion method presented here is compatible with a variety of complex matrices, reduces the need for harmful solvent usage, reduces the total number of potentially contaminating steps with respect to conventional SiNP digestion methods for ICP-MS, and characterizes the size distribution and concentration of the SiNPs simultaneously (Geiss et al., 2019; Laborda et al., 2014). However, a major limitation of this method is the limit of detection in relation to the size of the nanoparticle able to be accurately measured. Despite the fact that a majority of SiNPs observed in most applications are too small to be detected by this method, secondary particles and potential aggregates can be used as a proxy to analyze those particles in conjunction with a qualitative method such as fluorescent, darkfield, and transmission electron microscopy. The minimum size for this type of analysis is around 180 nm, due to the background signal that is observed with general silica analysis (Lee et al., 2014; Aureli et al., 2020). Because the size of the nanoparticles needs to be quite large with respect to general particle sizes seen in biological matrices, it lowers the total number of applications for the method. Additionally, with a largely polydisperse sample, there may be a bimodal model that fits the data better than that of a Gaussian or lognormal distribution. This is a complication for analysis, because the representative nanoparticle population will not be accurately determined due to the method’s assumption of monodispersity.

The main difference between our method and previous SiNP analysis methods is the removal of the necessity for a multitude of other reagents or instruments. Previous studies have reported a method to remove some of the harmful reagents from the ICP analysis of silica, however, this is prone to a strong matrix effect, which makes the quantitative analysis of silica, and SiNPs in particular, difficult (Bossert et al., 2019). Additionally, quantifying uptake of SiNPs in biological matrices is novel. In most literature, uptake is assayed using fluorescently-tagged nanoparticles, and biodistribution is often analyzed using total silicon, which does not provide any nanoparticle characterization data. Silica nanoparticles have been assayed by SP-ICP-MS, however, it requires more specialized equipment (Bolea-Fernandez et al., 2017). Nevertheless, it is of particular importance to examine the relationship between SiNP uptake in immune cells in a high throughput manner given that previous studies have demonstrated a robust immune response as a function of SiNP exposure (Chen et al., 2018).

The quantitative uptake of SiNPs in immune cells has many applications due to the heterogeneity of SiNPs and immune cell types that need to be evaluated. By describing, quantifying, and characterizing SiNPs in different immune cells, further investigation can be done into the immunotoxicity of SiNPs. This method can distinguish which cells are taking up more SiNPs, and if there are size and physicochemical effects on the uptake of SiNPs in immune cells. To determine if toxicities related to SiNPs are uptake-mediated, functional studies and uptake studies can be used as complimentary methods to more fully characterize toxicological effects of SiNPs. Furthermore, SiNP uptake in different immune cells may be identified as a potential biomarker for immune-mediated diseases of unknown etiologies.

In summary, we present a single step digestion method and analysis of SiNPs by SP-ICP-MS as a way to quantitatively analyze and characterize SiNP uptake in immune cells. This can be used as a replacement to other SiNP characterization techniques in vitro which do not provide quantitative results nor characterize the entire SiNP population in the sample. In the future, this will shed light into the toxicity of SiNPs in immune cells and further characterize their toxicity across multiple in vitro systems.

Critical Parameters:

There are two important aspects to the digestion of the biological matrix to which special attention should be given. First, the digestion must be complete. In order to ensure this, it is encouraged that tubes are either left overnight at room temperature or at 60°C for at least 6 hours. Additionally, cell counts should be as close to each other as possible between treatments. This allows normalization of counted particles per cell or total cells. Although absolute cell count may vary slightly between wells in an immortalized cell model, the growth between wells should be comparable. Cell number should be within ~5-10% variability within technical replicates when plated. To verify, extra replicate wells can be plated, and at the time of the experiment, cells can be trypsinized and counted (as described in Support Protocol 1).

One major critical parameter is to determine if the SiNP being assayed is appropriate for SP-ICP-MS analysis. The limit of detection of the method is around 180 nm. While this excludes a majority of SiNPs used for consumer purposes, this protocol can also be used to analyze aggregated particles or secondary SiNPs formed from the primary particles, as discussed above. Support Protocol 2 is vital to determining if there is a particle population that is able to be assayed via SP-ICP-MS.

One of the most critical parameters requiring attention to in this experiment is the optimization of the instrument prior to experimentation. If the instrument is not properly optimized for sensitivity and precision, then the variability between counts in samples increases. The optimization prior to every run should be a torch alignment for maximum intensity and sensitivity, the optimization of the nebulizer flow to optimize sample introduction and ionization, and the optimization of the deflector voltage to maximize sensitivity of the instrument. Running these optimizations is described in Support Protocol 3.

Troubleshooting:

Please see Table 1 for a list of common problems with the protocols, their causes, and potential solutions.

Understanding Results:

SP-ICP-MS is a highly sensitive, robust tool for the determination, quantification, and characterization of SiNP uptake in immune cells, amongst other biological matrices. This method shows that for any biological matrix, particularly that of immune cells, a single step digestion can be used to characterize the SiNP uptake profile of any immune cell. Culturing of the immune cells as shown in Support Protocol 1 leads to the ability to treat and digest cells for SP-ICP-MS (Basic Protocol 1 and shown in Figure 1). Following a run in Basic Protocol 2, the sample will have the number of nanoparticles in the sample quantified, as well as the size and the size distribution of the nanoparticles (Figure 1). Then, the representative nanoparticle population, which is governed by the quantity, the mean size, and the most detected particle size, is obtained. This population will be conveyed to the user with a nonlinear distribution (as above, for a normally distributed sample, a Gaussian distribution, and for a non-normally distributed sample, a log normal distribution). This lets the user compare nanoparticle populations across samples.

By following Support Protocol 2, and in an experiment to exemplify the workflow, the nanoparticle suspensions generated were found to be monodisperse in Milli-Q H₂O, using dynamic light scattering (Table 2). These results show that the nanoparticle sample is appropriate for analyses with SP-ICP-MS due to their size being larger than 180 nm and monodispersed (PDI < 0.200). These particles were synthesized commercially, so the monodispersion that is observed is to be expected. Additionally, the particle sizes are close to what the manufacturer finds as well.

Table 2.

Dynamic light scattering analysis of 200 and 300 nm SiNPs in Milli-Q H₂O shows accurate sizing of SiNP with a small standard deviation and a small polydispersity index, indicating monodisperse samples approximately the manufacturer-certified size.

Manufacturer Size	Hydrodynamic diameter in Milli-Q H2O (± SD)	PDI
200 nm	211.53 nm ± 1.45 nm	0.027
300 nm	320.03 nm ± 9.11 nm	0.077

For the example protocol described in this manuscript, the instrument was first optimized and the parameters for running the instrument were found (Table 2) as described in Support Protocol 3. Based on previous literature, SiNP and silicon analysis is typically done for the ²⁸Si⁺ isotope using ammonia dynamic reaction cell (NH₃ DRC) collision prior to analysis.

Using the methods in Support Protocol 2 and Support Protocol 3, we found the anticipated size of the 200 and 300 nm of the SiNPs we were assaying (Table 2). This means they are appropriate for analysis with the instrument that we have optimized according to parameters shown in Table 3. This ensures SP-ICP-MS analysis is appropriate for these samples.

Table 3.

Optimal SiNP analysis parameters for SP-ICP-MS analysis described here based on optimization steps given in Support Protocol 3.

SP-ICP-MS Parameters	Operating Conditions
RF power	1,600 W
Pulse voltage	1,250 V
Analog voltage	−1,750 V
Plasma Argon Flow	15.0 L/min
Nebulizer gas flow	0.72 L/min
Auxillary gas flow	1.2 L/min
Analyte monitored	28Si
Scan time	100 s
Dwell time	50 μs
Sample flow rate	0.441 mL/min
Reaction gas (NH3) flow	0.30 mL/min
RPq	0.6
Detector mode	Dual
Oxide ratio	<3.0%
Double charged ion ratio	<2.5%
Background counts	<0.1 cps

When the instrument gives the final result in a frequency histogram, the software will automatically regress the data from the histogram in a nonlinear distribution. The user should note whether the sample appears normally distributed or not. For a normally distributed nanoparticle population, as would be given by a pristine nanoparticle sample, then using a Gaussian distribution is advised. For a more polydisperse nanoparticle population, consider selecting a log normal distribution in the Syngistix software.

In analysis of both cells and media by SP-ICP-MS, we observed that the SiNPs present in the samples were both accurate and precise to the anticipated size (Figure 2). Figure 2 also shows that at 24 hours, we observed the same population of SiNPs in both cells (Figure 2A, 2B, 2C) and media (Figure 2D, 2E, 2F). This means that the digestion method used removes matrix effect, as there is no significant difference between the detected sizes in cell pellets and media. We also observed a very low level of background in the control samples, both from the cells and the media. The described protocol overcomes previous issues with Si analyses with high background signals. Additionally, the representative nanoparticle population given by the Gaussian function each histogram was fitted to shows very similar shapes across matrices. This protocol functions as a proper analytical technique for the sizing of SiNPs because the % error from the true value of the samples was low (1.2%-3.0%) across sizes. It should be noted that there is a lower total number of 300 nm SiNPs detected in the cell pellet and media than in the 200 nm SiNP treated cells and media. This is because the cells were treated with SiNPs on a mass basis. In this case, 100 μg/mL of 300 nm SiNPs contains a smaller number of nanoparticles than 100 μg/mL 200 nm SiNPs. Despite this, however, because individual SiNPs are counted by the instrument, percent uptake can still be accurately determined.

The uptake of SiNPs in RAW 264.7 cells is governed by their size (Figure 3). Figure 3 shows that 300 nm SiNPs are taken up in significantly greater amounts and significantly more quickly than 200 nm SiNPs. This could be due to 300 nm SiNPs sedimenting more rapidly than 200 nm SiNPs, which would increase their exposure to RAW 264.7 cells. Additionally, the increased surface area of the 300 nm SiNPs likely make it easier for the cells to interact with and, thus, take up the SiNPs. This data, when coupled with mechanistic data, can be used to describe if there are differences in SiNP toxicity in this cell model between sizes. Given that 300 nm SiNPs are taken up more rapidly and in greater amounts than 200 nm SiNPs, then mechanistic toxicological assays will allow for the elucidation of more information pertaining to SiNP toxicity. If 300 nm SiNPs cause more robust toxicological effects in cells than cells treated with 200 nm SiNPs, then it can be inferred that SiNP toxicity in that cell model is governed by their uptake.

Time Considerations:

Basic Protocol 1 will have ^~3 hours of active time, broken up by a 6-24 hour incubation. The first hour is dedicated to removing and separating cells, pelleting cells, and digesting media and cells in 70% HNO₃. Then, digesting samples will take between 6 and 24 hours. Finally, the last 2 hours are to create dilutions for the samples. Basic Protocol 1 ends when samples are frozen.

Support Protocol 1 can be done indefinitely. However, when a new passage of cells begins growing, they will only be able to be used for experiments from passages 4-10. If cells are passaged every 2-3 days, then cells should be used only for 12-18 days.

Support Protocol 2 will take approximately 1 hour to both optimize the dilution for the DLS sample and determine the hydrodynamic diameter and PDI.

Support Protocol 3 will take approximately 30 minutes for the instrument to automatically calibrate and optimize. However, if the calibration fails and the user must troubleshoot the instrument, this protocol can take up to 2 hours.

Finally, Basic Protocol 2 requires approximately 1 hour to plate the samples in the 96 well microtiter plate, and the ICP-MS run will utilize approximately 4 minutes per sample. This means a run with 3 timepoints, with 3 technical replicates for 3 biological replicates with a negative control and 2 treatment groups will have 81 samples. This run will take approximately 5 hours and 30 minutes.

DATA AVAILABILITY STATEMENT:

Data available on request from the authors